4 ms·

I mean, the name is literally 'the stuff we can't see'. Dark matter is very much just a question mark looking for an explanation.

iainmerrick 5y ago
The word “stuff” there is exactly the point. It’s a lot more specific than “just a question mark” -- the hypothesis is that it’s some exotic form of matter. If something like MOND gains popularity and there’s no longer a need to postulate invisible matter, it would be fair to say that dark matter is incorrect and outdated, just like ether or phlogiston.
- generalizations 5y ago
  That would be fair. However, from what I've read MOND is relatively unlikely. If its proponents can address the biggest difficulties (e.g. not every galaxy appears to have "dark matter") and make it somewhat more feasible, then I'd agree the name is insufficiently broad.
- dnautics 5y ago
  Not really, MOND and the like falls into the category of "dark matter theory", and that's the problem with the name.
  - pdonis 5y ago
    > MOND and the like falls into the category of "dark matter theory" No, it doesn't. See my response to your other post upthread about the two hypotheses. "Dark matter" is one hypothesis; "MOND" and friends are the other.
  - raattgift 5y ago
    No, it is unfair to call MOND (and at least much of "the like") as a dark matter theory, because it manifestly generates a solution (or solutions) in the absence of any matter at all. MOND is characterized by an interpolating function between its fundamental constant a_0 and acceleration, such that in the "deep-MOND" regime, F = m \frac{a^2}{a_0} rather than Newton's F = ma. Both MOND and Newton work in everywhere-vacuum, and in everywhere-vacuum perturbed by a point mass. Neither the vacuum case nor the point mass case qualifies as a "dark matter theory" any more than the vacuum Schwarzschild solution of General Relativity does. These are model universes where one can calculate a Gauss law for gravity. MOND would behave differently for deep deep extragalactic space with small icy body moving around a large gassy body than Newton would, or than we would get from e.g. a roughly Einstein-de Sitter solution. Milgrom has pursued several relativistic corrections to the simple MOND formula, and they generally allow for solutions where the stress-energy tensor is everywhere 0, and everywhere 0 except for one point. Milgrom originated MOND. He has collected numerous thoughts and many links at http://www.scholarpedia.org/article/The_MOND_paradigm_of_modified_dynamics http://www.scholarpedia.org/article/The_MOND_paradigm_of_mod... When applied to large systems like galaxies and especially galaxy clusters, Milgrom's corrections may be insufficient, and other corrections (catalogued well in https://arxiv.org/abs/1112.3960 https://arxiv.org/abs/1112.3960 section 7). Several of these are very hard to make work with the stress-energy tensor everywhere zero, and those (certainly not all of them) one might ungraciously call "just another dark matter theory". On the other hand, several of these theories admit plausible gravitational radiation in the total absence of matter, just as does General Relativity. (Gravitational radiation is not a feature of Newton or the simple original Milgrom formula.) McGaugh runs tritonstation and is a well known sceptic of particle dark matter. Relativistic dark matter theories deliberately introduce nonzeros into the stress-energy tensor. You need a distribution of dark matter and trace out how that generates curvature, or you don't have a (relativistic) dark matter theory. You can also of course set out a distribution of cold dark matter and see how it behaves under F = ma or F = m \frac{a^2}{a_0}. Neither MOND nor Newton requires dark matter; likewise neither MOND nor Newton is incompatible with dark matter. Likewise, General Relativity doesn't require dark matter. But one gets more physically realistic solutions at a wide range of scales when one adds dark matter to a defined large-scale distribution of matter and radiation. MOND or the various attempts at generally covariant Milgrom gravitation would also behave differently when one adds dark matter to a large-scale distribution of matter and radiation. That is unlikely to produce more physically realistic solutions to a wide range of scales, however. Indeed, Milgrom et al's most concrete argument is that their formula for gravitation is worse if one adds in a distribution of dark matter, and that their formula is better than e.g. a Lemaître-Tolman-Bondi solution equipped with a distribution of cold dark matter. That's as much a non-dark-matter theory as I think it is possible to have!
    - dnautics 5y ago
      Forgive me, but I was under the impression the phrase "dark matter theory" can confusingly refer to something that generally explains galactic rotation curves whether or not it's a theory that involves dark matter. What would you call the general category of theories which explain galactic rotation curve deviations?
      - raattgift 5y ago
        Maybe just solutions for non-Keplerian redshifts, see http://hyperphysics.phy-astr.gsu.edu/hbase/Astro/darmat.html http://hyperphysics.phy-astr.gsu.edu/hbase/Astro/darmat.html second image and bracketing paragraphs. As in computing, naming things is hard. The redshifting things are mostly molecular gas clouds which we would expect would orbit in a Keplerian fashion given the overall distribution of luminous matter and occluding/extinguishing dust visible in galaxies. Astrophysicists love hydrogen gas clouds because they ionize easily from background emitters (stars, quasars) producing very sharp spectral lines, and you can pick up their changes by looking on different sides of an edge-on spiral or are at nearly perpendicular angles to the rotational axis of spheroidals. For giant ellipticals, which have practically zero axial rotation, an instrument that gets a sweeping view of them will see a more chaotic red-and-blue shifting of the Lyman-alpha hydrogen line, rather than an obvious gradient between advancing and retreating edges. (Face-on spirals often also have gas clouds on excursions back and forth in front of the disk plane, where "forth" means "towards us" means "relatively blueshifted" and that's hard to explain without a halo rather than a ring of dark matter). So really it's a question explaining the non-Keplerian https://en.wikipedia.org/wiki/Peculiar_velocity https://en.wikipedia.org/wiki/Peculiar_velocity of structures from ionized hydrogen to molecular clouds to supernovae with highly predictable light curves. While basic MOND can explain edge-on spirals and axis-perpendicular spheroidals and lenticulars, it does a bad job with face-on spirals, ellipticals, and lots of other unusual shapes in the galaxy zoo. It also doesn't do well with entire clusters, where there are enormous gas clouds between the member galaxies, and often a bright quasar (or at least central galaxy) really lighting them up. Corrections to basic MOND patch some of these up, mostly starting by adding extra terms to the initial basic Milgrom formula which modified F = ma. A ball-like distribution of dark matter around these very large structures (galaxies, clusters) on the other hand is a decent explanation, with the major drawback that the only sufficiently transparent stuff we've produced in labs is simply too light to stay constrained within these structures (electron neutrinos and their antis would fly away to infinity, for example, rather than stick around in a hierarchy of halo/shell/ball structures; free neutrons aren't stable). We also don't have a really good explanation about why these proposed dark matter structures don't have a strong density gradient leading to grossly non-Keplerian orbits of stars in the cores of galaxies (including our own). That's why you get wildly speculative ideas like phase transitions, "dark chemistry", dark-matter/dark-matter annihilation, and so on. There are similar questions about why the outer reaches of dark matter halos don't blow off into infinity rather than lingering around. ( See the excellent III A and III E of Adams & Laughlin https://arxiv.org/abs/astro-ph/9701131 https://arxiv.org/abs/astro-ph/9701131 or the Binney & Tremaine textbook https://press.princeton.edu/books/paperback/9780691130279/galactic-dynamics https://press.princeton.edu/books/paperback/9780691130279/ga... ) Good explanations should cover the entire lifetime of a galaxy including the things that lead up to it and its ultimate decay. This also means it should be good for the oldest galaxies that we can see. It also means that as we get better at simulating initial value problems we can take a large number of observed quantities of a highly-redshifted galaxy and evolve that initial snapshot step by step such that after trillions of steps we end up with something that looks remarkably like a galaxy with low redshift. We're not really there yet. Work harder, supercomputer builders! :D :D

iainmerrick 5y ago

The word “stuff” there is exactly the point. It’s a lot more specific than “just a question mark” -- the hypothesis is that it’s some exotic form of matter. If something like MOND gains popularity and there’s no longer a need to postulate invisible matter, it would be fair to say that dark matter is incorrect and outdated, just like ether or phlogiston.

generalizations 5y ago
That would be fair. However, from what I've read MOND is relatively unlikely. If its proponents can address the biggest difficulties (e.g. not every galaxy appears to have "dark matter") and make it somewhat more feasible, then I'd agree the name is insufficiently broad.
dnautics 5y ago
Not really, MOND and the like falls into the category of "dark matter theory", and that's the problem with the name.
- pdonis 5y ago
  > MOND and the like falls into the category of "dark matter theory" No, it doesn't. See my response to your other post upthread about the two hypotheses. "Dark matter" is one hypothesis; "MOND" and friends are the other.
- raattgift 5y ago
  No, it is unfair to call MOND (and at least much of "the like") as a dark matter theory, because it manifestly generates a solution (or solutions) in the absence of any matter at all. MOND is characterized by an interpolating function between its fundamental constant a_0 and acceleration, such that in the "deep-MOND" regime, F = m \frac{a^2}{a_0} rather than Newton's F = ma. Both MOND and Newton work in everywhere-vacuum, and in everywhere-vacuum perturbed by a point mass. Neither the vacuum case nor the point mass case qualifies as a "dark matter theory" any more than the vacuum Schwarzschild solution of General Relativity does. These are model universes where one can calculate a Gauss law for gravity. MOND would behave differently for deep deep extragalactic space with small icy body moving around a large gassy body than Newton would, or than we would get from e.g. a roughly Einstein-de Sitter solution. Milgrom has pursued several relativistic corrections to the simple MOND formula, and they generally allow for solutions where the stress-energy tensor is everywhere 0, and everywhere 0 except for one point. Milgrom originated MOND. He has collected numerous thoughts and many links at http://www.scholarpedia.org/article/The_MOND_paradigm_of_modified_dynamics http://www.scholarpedia.org/article/The_MOND_paradigm_of_mod... When applied to large systems like galaxies and especially galaxy clusters, Milgrom's corrections may be insufficient, and other corrections (catalogued well in https://arxiv.org/abs/1112.3960 https://arxiv.org/abs/1112.3960 section 7). Several of these are very hard to make work with the stress-energy tensor everywhere zero, and those (certainly not all of them) one might ungraciously call "just another dark matter theory". On the other hand, several of these theories admit plausible gravitational radiation in the total absence of matter, just as does General Relativity. (Gravitational radiation is not a feature of Newton or the simple original Milgrom formula.) McGaugh runs tritonstation and is a well known sceptic of particle dark matter. Relativistic dark matter theories deliberately introduce nonzeros into the stress-energy tensor. You need a distribution of dark matter and trace out how that generates curvature, or you don't have a (relativistic) dark matter theory. You can also of course set out a distribution of cold dark matter and see how it behaves under F = ma or F = m \frac{a^2}{a_0}. Neither MOND nor Newton requires dark matter; likewise neither MOND nor Newton is incompatible with dark matter. Likewise, General Relativity doesn't require dark matter. But one gets more physically realistic solutions at a wide range of scales when one adds dark matter to a defined large-scale distribution of matter and radiation. MOND or the various attempts at generally covariant Milgrom gravitation would also behave differently when one adds dark matter to a large-scale distribution of matter and radiation. That is unlikely to produce more physically realistic solutions to a wide range of scales, however. Indeed, Milgrom et al's most concrete argument is that their formula for gravitation is worse if one adds in a distribution of dark matter, and that their formula is better than e.g. a Lemaître-Tolman-Bondi solution equipped with a distribution of cold dark matter. That's as much a non-dark-matter theory as I think it is possible to have!
  - dnautics 5y ago
    Forgive me, but I was under the impression the phrase "dark matter theory" can confusingly refer to something that generally explains galactic rotation curves whether or not it's a theory that involves dark matter. What would you call the general category of theories which explain galactic rotation curve deviations?
    - raattgift 5y ago
      Maybe just solutions for non-Keplerian redshifts, see http://hyperphysics.phy-astr.gsu.edu/hbase/Astro/darmat.html http://hyperphysics.phy-astr.gsu.edu/hbase/Astro/darmat.html second image and bracketing paragraphs. As in computing, naming things is hard. The redshifting things are mostly molecular gas clouds which we would expect would orbit in a Keplerian fashion given the overall distribution of luminous matter and occluding/extinguishing dust visible in galaxies. Astrophysicists love hydrogen gas clouds because they ionize easily from background emitters (stars, quasars) producing very sharp spectral lines, and you can pick up their changes by looking on different sides of an edge-on spiral or are at nearly perpendicular angles to the rotational axis of spheroidals. For giant ellipticals, which have practically zero axial rotation, an instrument that gets a sweeping view of them will see a more chaotic red-and-blue shifting of the Lyman-alpha hydrogen line, rather than an obvious gradient between advancing and retreating edges. (Face-on spirals often also have gas clouds on excursions back and forth in front of the disk plane, where "forth" means "towards us" means "relatively blueshifted" and that's hard to explain without a halo rather than a ring of dark matter). So really it's a question explaining the non-Keplerian https://en.wikipedia.org/wiki/Peculiar_velocity https://en.wikipedia.org/wiki/Peculiar_velocity of structures from ionized hydrogen to molecular clouds to supernovae with highly predictable light curves. While basic MOND can explain edge-on spirals and axis-perpendicular spheroidals and lenticulars, it does a bad job with face-on spirals, ellipticals, and lots of other unusual shapes in the galaxy zoo. It also doesn't do well with entire clusters, where there are enormous gas clouds between the member galaxies, and often a bright quasar (or at least central galaxy) really lighting them up. Corrections to basic MOND patch some of these up, mostly starting by adding extra terms to the initial basic Milgrom formula which modified F = ma. A ball-like distribution of dark matter around these very large structures (galaxies, clusters) on the other hand is a decent explanation, with the major drawback that the only sufficiently transparent stuff we've produced in labs is simply too light to stay constrained within these structures (electron neutrinos and their antis would fly away to infinity, for example, rather than stick around in a hierarchy of halo/shell/ball structures; free neutrons aren't stable). We also don't have a really good explanation about why these proposed dark matter structures don't have a strong density gradient leading to grossly non-Keplerian orbits of stars in the cores of galaxies (including our own). That's why you get wildly speculative ideas like phase transitions, "dark chemistry", dark-matter/dark-matter annihilation, and so on. There are similar questions about why the outer reaches of dark matter halos don't blow off into infinity rather than lingering around. ( See the excellent III A and III E of Adams & Laughlin https://arxiv.org/abs/astro-ph/9701131 https://arxiv.org/abs/astro-ph/9701131 or the Binney & Tremaine textbook https://press.princeton.edu/books/paperback/9780691130279/galactic-dynamics https://press.princeton.edu/books/paperback/9780691130279/ga... ) Good explanations should cover the entire lifetime of a galaxy including the things that lead up to it and its ultimate decay. This also means it should be good for the oldest galaxies that we can see. It also means that as we get better at simulating initial value problems we can take a large number of observed quantities of a highly-redshifted galaxy and evolve that initial snapshot step by step such that after trillions of steps we end up with something that looks remarkably like a galaxy with low redshift. We're not really there yet. Work harder, supercomputer builders! :D :D