Does anyone else oftentimes feel like dark matter is an unfalsifiable fudge factor? The article kind of says this, without really saying it: Dark matter just always feels like a moving target to...
Does anyone else oftentimes feel like dark matter is an unfalsifiable fudge factor? The article kind of says this, without really saying it:
Grabowska is careful to note that it’s difficult to completely throw away a theory of dark matter.
Dark matter just always feels like a moving target to me, whose properties have basically always been specifically chosen in order to make our current theories of gravity applicable to large cosmological structures.
Why can't it be that we just have an incomplete theory of gravity? When we noticed epicycles in the motion of planets, we resolved it by replacing the geocentric model with a heliocentric one. When we noticed that the procession of Mercury's orbit violates Newtonian gravity, we developed general relativity. Yet, when our current understanding of gravity fails to explain things like the rotation curves of galaxies, we desperately cling to what we already know and try to fix it by adding the precise amount of invisible matter we need to make it work? And then we use that as evidence to justify the supposed existence of this invisible matter? Doesn't it all just feel extremely presumptuous and a bit circular?
For the record, none of what I said should be taken as an endorsement of any of the existing alternative theories of gravity. They all have their own serious problems, but other people failing is not evidence that everyone else is right.
GR is very well supported by evidence. Modifying/extending GR so that it (1) still fits observations, (2) explains rotation curves, (3) explains the bullet cluster, (4) explains BAO, (5) explains...
GR is very well supported by evidence. Modifying/extending GR so that it (1) still fits observations, (2) explains rotation curves, (3) explains the bullet cluster, (4) explains BAO, (5) explains the CMB, and (6) explains structure formation, is extremely difficult. Extra, weakly interacting, mass is a simpler explanation.
On the other hand, extending the standard model to include weakly interacting particles, or other exotic particles, is ok.
I think things start getting very technical if you want to get into why one is difficult and the other ok.
Well of course it's difficult. But it took literally hundreds of years since Newton to develop the mathematics required for general relativity to even exist, so it's not like it would be a strange...
Well of course it's difficult. But it took literally hundreds of years since Newton to develop the mathematics required for general relativity to even exist, so it's not like it would be a strange thing for it to be difficult.
And I don't understand how "GR is very well supported by evidence" means anything when any theory could potentially be considered to be a special case of a more general one. Newtonian gravity is "well supported by evidence", too.
I think the point being made is that theories have regimes of validity, and while you are certainly correct that GR could be a special or limiting case of a larger theory (and it most certainly...
And I don't understand how "GR is very well supported by evidence" means anything when any theory could potentially be considered to be a special case of a more general one. Newtonian gravity is "well supported by evidence", too.
I think the point being made is that theories have regimes of validity, and while you are certainly correct that GR could be a special or limiting case of a larger theory (and it most certainly is), we shouldn’t expect to see modifications in regimes that the theory self-consistently explains well. This is the same reason that relativity took so long to figure out - in most cases, Newtonian gravity works perfectly well to explain observations and the discrepancies that motivated relativity were hard to discern at first.
In the case of galactic rotation curves for example, one of the primary pieces of evidence for either dark matter or modified gravity, general relativity should work well based on how we understand where the theory is valid. In fact, galactic rotation curves are well described by Newtonian gravity since the speeds are far from relativistic. Indeed the basic calculations needed to see the problem are no more than computing a circular rotation velocity assuming the central force is gravity, something you could do in an introductory physics class. Obviously it can get more complicated than that once you include more realistic effects, but the point remains that these computations are mostly non-relativistic. That one of the leading modified gravity theories is called “Modified Newtonian Dynamics” is a reflection of this fact.
All of this is to say that we think we understand gravity well in the regimes where these discrepancies appear. This is a different situation than, say, gravitational effects in a highly curved region of spacetime (e.g. near a small black hole) where we might reasonably expect the theory to break down and GR to be subsumed by a more complete theory. So the situation is different than the Newtonian-GR transition of the early 20th century. It would be as though we had observed then that occasionally an apple falling would be deflected from its trajectory - a more reasonable assumption would be that something was deflecting it rather than that gravity was wrong (weak example I know).
I think the key lesson here is that very often physical theories not only contain descriptions of nature, but they also encode information about their regimes of validity. We expect GR to break down in highly curved regions because the equations spit out infinities and divide by zero, and we expect Newtonian mechanics to fail at high energies because the equations would imply a violation of causality. In the case of rotation curves and the associated scales and energies, neither the equations of GR nor Newtonian dynamics break down.
Of course, modified gravity is a possibility and a very active area of research. But given our history in the past century of discovering new particles as well as dark matter’s ability to explain nearly all anomalous observations in a consistent way, as well as the multiple mechanisms from particle physics that could give rise to a new particle with the appropriate mass and abundance as the hypothetical DM particle, it is not nearly as lazy as it might first seem.
The detection of gravitational waves ruled out a lot of the alternatives. The astonishing accuracy and predictive value of general relativity on cosmological scales is hard to explain if it's...
The detection of gravitational waves ruled out a lot of the alternatives. The astonishing accuracy and predictive value of general relativity on cosmological scales is hard to explain if it's fundamentally wrong. It's not the ultimate theory of gravity, because it doesn't work on quantum scales, but on the balance of evidence, it seems to be a very good description of gravity at large scales. While invisible matter may seem presumptuous, throwing away the successes of GR also seems presumptuous. Especially because we've detected near-invisible particles like neutrinos. It's estimated that billions pass through your body every second, yet we have to build 50,000 ton detectors to catch a few of them.
It's one thing to be skeptical of a theory that has never predicted anything novel to any degree of accuracy. It's another thing to discount a theory that has been validated again and again for 100 years. At that point, invisible matter starts to look very attractive.
Okay, but nobody ever said "general relativity is fundamentally wrong". I only suggested that it might be incomplete. Newtonian mechanics is also "astonishingly accurate" whenever you are not...
Okay, but nobody ever said "general relativity is fundamentally wrong". I only suggested that it might be incomplete. Newtonian mechanics is also "astonishingly accurate" whenever you are not close to regions of spacetime with high density, and even today we continue to rely on its predictive power all the time. We didn't have to "throw away" the successes of Newton following the development of GR. GR didn't prove that Newton "was wrong". In fact, it's quite important that at small scales, GR can be reduced to Newtonian mechanics.
Similarly, coming up with a more comprehensive theory of spacetime that eliminates the need for dark matter would absolutely not mean GR is "wrong", only that, like Newtonian mechanics, a few puzzle pieces were missing.
I'm not "skeptical" of GR any more than I am "skeptical" of Newtonian gravity. They are both clearly "on the right track", so to speak, and any decent alternative theory of gravity that eliminates the need for dark matter should similarly be reducible to GR on the scales where GR works without a hitch, which would necessarily imply that GR, like Newtonian mechanics, would still be an extremely useful theory.
The idea that gravity just works differently than we thought and explains dark matter is called a MOND theory (modified Newtonian dynamics) and it's mentioned in the article that those theories...
The idea that gravity just works differently than we thought and explains dark matter is called a MOND theory (modified Newtonian dynamics) and it's mentioned in the article that those theories can't explain dark matter. No one figured out a way it could exactly explain things like the galactic rotation speeds, but the really big issue is that can't explain the Bullet Cluster, where the regular matter and dark matter have visibly separated. It's really hard to imagine how a theory that there's no dark matter, just weird gravity around regular matter, could explain that situation.
Everything we've seen is consistent with the idea that most matter in the universe is made of weakly-interacting particles probably like neutrinos. We know that there exist weakly interacting particles, and we don't have reason to think we know all particle types, so this theory is pretty mundane. (I've seen dark matter skepticism around online, and I wonder if the skepticism is fueled in part by thinking dark matter is stranger than it is.) Every time we've come up with a new test, it's matched up with what we'd expect if this were true. Theorizing otherwise is like living with roommates, finding your food disappearing from the fridge, and theorizing that food just has a tendency to disappear rather than it being that your roommates are eating the food, and then trying to come up with a separate explanation for the appearance of dirty dishes.
Does anyone else oftentimes feel like dark matter is an unfalsifiable fudge factor? The article kind of says this, without really saying it:
Dark matter just always feels like a moving target to me, whose properties have basically always been specifically chosen in order to make our current theories of gravity applicable to large cosmological structures.
Why can't it be that we just have an incomplete theory of gravity? When we noticed epicycles in the motion of planets, we resolved it by replacing the geocentric model with a heliocentric one. When we noticed that the procession of Mercury's orbit violates Newtonian gravity, we developed general relativity. Yet, when our current understanding of gravity fails to explain things like the rotation curves of galaxies, we desperately cling to what we already know and try to fix it by adding the precise amount of invisible matter we need to make it work? And then we use that as evidence to justify the supposed existence of this invisible matter? Doesn't it all just feel extremely presumptuous and a bit circular?
For the record, none of what I said should be taken as an endorsement of any of the existing alternative theories of gravity. They all have their own serious problems, but other people failing is not evidence that everyone else is right.
GR is very well supported by evidence. Modifying/extending GR so that it (1) still fits observations, (2) explains rotation curves, (3) explains the bullet cluster, (4) explains BAO, (5) explains the CMB, and (6) explains structure formation, is extremely difficult. Extra, weakly interacting, mass is a simpler explanation.
On the other hand, extending the standard model to include weakly interacting particles, or other exotic particles, is ok.
I think things start getting very technical if you want to get into why one is difficult and the other ok.
Well of course it's difficult. But it took literally hundreds of years since Newton to develop the mathematics required for general relativity to even exist, so it's not like it would be a strange thing for it to be difficult.
And I don't understand how "GR is very well supported by evidence" means anything when any theory could potentially be considered to be a special case of a more general one. Newtonian gravity is "well supported by evidence", too.
I think the point being made is that theories have regimes of validity, and while you are certainly correct that GR could be a special or limiting case of a larger theory (and it most certainly is), we shouldn’t expect to see modifications in regimes that the theory self-consistently explains well. This is the same reason that relativity took so long to figure out - in most cases, Newtonian gravity works perfectly well to explain observations and the discrepancies that motivated relativity were hard to discern at first.
In the case of galactic rotation curves for example, one of the primary pieces of evidence for either dark matter or modified gravity, general relativity should work well based on how we understand where the theory is valid. In fact, galactic rotation curves are well described by Newtonian gravity since the speeds are far from relativistic. Indeed the basic calculations needed to see the problem are no more than computing a circular rotation velocity assuming the central force is gravity, something you could do in an introductory physics class. Obviously it can get more complicated than that once you include more realistic effects, but the point remains that these computations are mostly non-relativistic. That one of the leading modified gravity theories is called “Modified Newtonian Dynamics” is a reflection of this fact.
All of this is to say that we think we understand gravity well in the regimes where these discrepancies appear. This is a different situation than, say, gravitational effects in a highly curved region of spacetime (e.g. near a small black hole) where we might reasonably expect the theory to break down and GR to be subsumed by a more complete theory. So the situation is different than the Newtonian-GR transition of the early 20th century. It would be as though we had observed then that occasionally an apple falling would be deflected from its trajectory - a more reasonable assumption would be that something was deflecting it rather than that gravity was wrong (weak example I know).
I think the key lesson here is that very often physical theories not only contain descriptions of nature, but they also encode information about their regimes of validity. We expect GR to break down in highly curved regions because the equations spit out infinities and divide by zero, and we expect Newtonian mechanics to fail at high energies because the equations would imply a violation of causality. In the case of rotation curves and the associated scales and energies, neither the equations of GR nor Newtonian dynamics break down.
Of course, modified gravity is a possibility and a very active area of research. But given our history in the past century of discovering new particles as well as dark matter’s ability to explain nearly all anomalous observations in a consistent way, as well as the multiple mechanisms from particle physics that could give rise to a new particle with the appropriate mass and abundance as the hypothetical DM particle, it is not nearly as lazy as it might first seem.
The detection of gravitational waves ruled out a lot of the alternatives. The astonishing accuracy and predictive value of general relativity on cosmological scales is hard to explain if it's fundamentally wrong. It's not the ultimate theory of gravity, because it doesn't work on quantum scales, but on the balance of evidence, it seems to be a very good description of gravity at large scales. While invisible matter may seem presumptuous, throwing away the successes of GR also seems presumptuous. Especially because we've detected near-invisible particles like neutrinos. It's estimated that billions pass through your body every second, yet we have to build 50,000 ton detectors to catch a few of them.
It's one thing to be skeptical of a theory that has never predicted anything novel to any degree of accuracy. It's another thing to discount a theory that has been validated again and again for 100 years. At that point, invisible matter starts to look very attractive.
Okay, but nobody ever said "general relativity is fundamentally wrong". I only suggested that it might be incomplete. Newtonian mechanics is also "astonishingly accurate" whenever you are not close to regions of spacetime with high density, and even today we continue to rely on its predictive power all the time. We didn't have to "throw away" the successes of Newton following the development of GR. GR didn't prove that Newton "was wrong". In fact, it's quite important that at small scales, GR can be reduced to Newtonian mechanics.
Similarly, coming up with a more comprehensive theory of spacetime that eliminates the need for dark matter would absolutely not mean GR is "wrong", only that, like Newtonian mechanics, a few puzzle pieces were missing.
I'm not "skeptical" of GR any more than I am "skeptical" of Newtonian gravity. They are both clearly "on the right track", so to speak, and any decent alternative theory of gravity that eliminates the need for dark matter should similarly be reducible to GR on the scales where GR works without a hitch, which would necessarily imply that GR, like Newtonian mechanics, would still be an extremely useful theory.
The idea that gravity just works differently than we thought and explains dark matter is called a MOND theory (modified Newtonian dynamics) and it's mentioned in the article that those theories can't explain dark matter. No one figured out a way it could exactly explain things like the galactic rotation speeds, but the really big issue is that can't explain the Bullet Cluster, where the regular matter and dark matter have visibly separated. It's really hard to imagine how a theory that there's no dark matter, just weird gravity around regular matter, could explain that situation.
This list about the evidence for dark matter is really good and concise: https://www.reddit.com/r/space/comments/6488wb/comment/dg05wx4.
Everything we've seen is consistent with the idea that most matter in the universe is made of weakly-interacting particles probably like neutrinos. We know that there exist weakly interacting particles, and we don't have reason to think we know all particle types, so this theory is pretty mundane. (I've seen dark matter skepticism around online, and I wonder if the skepticism is fueled in part by thinking dark matter is stranger than it is.) Every time we've come up with a new test, it's matched up with what we'd expect if this were true. Theorizing otherwise is like living with roommates, finding your food disappearing from the fridge, and theorizing that food just has a tendency to disappear rather than it being that your roommates are eating the food, and then trying to come up with a separate explanation for the appearance of dirty dishes.