3 ms·

Dark matter is not a “recursive property of gravity” because it’s not consistent like that, as your own initial objection suggests. The concept of a particle t

Dark matter is not a “recursive property of gravity” because it’s not consistent like that, as your own initial objection suggests. The concept of a particle that only interacts via gravity is straightforward. Consider neutrinos, which only interact via gravity and the weak force. Trillions of them pass through your body every second. Take away the weak force interaction and add some mass, and you have a dark matter candidate particle. > I fail to see how dark matter could ever be falsified. You’d have to explain what should be done about that. It could nevertheless be a correct theory, and we might one day verify it more directly. Naive falsificationism is an almost useless principle at the frontiers of modern physics. We’re long past the days when you could run some simple experiments and come up with a clear yea or nay. That’s living in the 19th century, basically - which admittedly, many people still seem to want to do.

choeger 4y ago
> You’d have to explain what should be done about that. Well, there should be something that follows from the assumption that there are huge amounts of particles that only interact via gravity. What's the prediction that this model makes? If it doesn't make any useful prediction, it's pretty much worthless. > That’s living in the 19th century, basically - which admittedly, many people still seem to want to do. I think this attitude is inappropriate. The scientific progress made with 19th century methods, and that includes most of the stuff discovered until the 1950s, is massive. Maybe scientists should focus more on a clear nay or yea? I mean, the LHC is certainly a significant achievement, but compared to the detection of gravitational waves or even just exoplanets it doesn't seem to deliver as much bang for the buck.
- YawningAngel 4y ago
  Indulge the counterfactual for a second. Suppose that dark matter really does exist, and we can't ascertain that fact with any certainty. What do you want Physics as a whole to do about that? I think everybody agrees a simple answer would be great, we're just unable to provide one in this context
  - naasking 4y ago
    > Suppose that dark matter really does exist, and we can't ascertain that fact with any certainty. What do you want Physics as a whole to do about that? Be more bold in exploring alternative gravity theories and conceiving of tests we can perform to check how gravity behaves in regimes that are not yet well tested.
- cauch 4y ago
  > I mean, the LHC is certainly a significant achievement, but compared to the detection of gravitational waves or even just exoplanets it doesn't seem to deliver as much bang for the buck. Well, first, LHC was "unlucky": they could have observed a lot of new particles if they would have been there. It feels unfair to say "this experiment that looked at X and had 50% chance of seeing something was a waste of money because they haven't found anything, this experiment that looked at Y and had 50% chance of seeing something was not because they saw something": both experiments are as worth a try. LHC still have observed the Higgs boson, which is still quite a big thing. LHC has also constrained a lot of theories (seeing that a particle predicted by a given theory is not there is very useful scientifically). But more importantly, I have the feeling that the notion of "bang for the buck" is pretty different from a scientific point of view and a layman point of view. For example, you mention exoplanets. Scientifically, those are really not a big deal. We already know that planets can exist, there are 8 around the Sun, so finding others around other stars is really not a big deal: everybody was expecting them to exist and knew the only reason they were not seen yet was the limits of the technology. One could even argue that using money to detect things that are pretty trivial is in fact the worst delivery of bang for the buck. Don't get me wrong, this is still nice to see them, but it's the same level as pentaquark (discovered at LHC). It is sometimes surprising to see the public/media reactions to discoveries. I'm still not sure why layman people were so excited about gravitational waves but not much about the tau neutrino.
onli 4y ago
If it is not falsifiable, it is not science nor physics. It is meta-physics, ersatz-religion. Which is exactly the core of the complaint against concepts like this. Of course same thinking can't easily be applied to things that are not always true, like social behaviour. Part of the reason the replication crisis hit so hard in the social sciences. And of course this is not really as absolute as I wrote above. No field works like this all the time. But it is a good guideline. If falsifiability is completely ignored, then a concept really is nothing more than a religious belief and should be considered as such. BTW, Popper did not publish in the 19th century.
- pydry 4y ago
  As far as I understood it if somebody could craft an experiment that would demonstrate, quite clearly that it was something other than a special kind of matter causing this effect then the theory would be falsified.
donatj 4y ago
> Naive falsificationism is an almost useless principle at the frontiers of modern physics. We’re long past the days when you could run some simple experiments and come up with a clear yea or nay. That’s living in the 19th century, basically - which admittedly, many people still seem to want to do. To a layman like myself this implies both that the things being studied then are both unprovable and not potentially of practical use or significance. I certainly am not saying that is the case, but rather am looking for more informed opinions to the contrary. The cynical side of me feels like study of something unfalsifiable is as the study of religion. A poor investment for public funds. As I mentioned however, I am seeking opinions here to the contrary of my gut reaction as my gut is often dumb and reactionary.
- mytailorisrich 4y ago
  The comment you replied to does not suggest that modern physics is unprovable... no simple experiment to prove something does not mean no experiment at all. The other, related point is that modern physics' frontier is in the quantum world and thus way outside layman's intuition, common sense, or reasoning. Even experts are struggling.
- oneoneonetwo 4y ago
  In science nothing is provable. Things can be falsified but if a hypothesis is true it can actually never be proven. This is because at any point in time and additional observation can be added to your sample that definitively disproves your hypothesis. All zebras have stripes. is my hypothesis. I observe 100 zebras and I see they have stripes. I see 10 million zebras and I see they have stripes. There is no amount of observations I can make to prove my hypothesis. However if at any point in time I see a zebra with no stripes, I have disproven my hypothesis. So anything that is in actuality true, is not falsifiable and also un-provable. In short, some statement about our universe can be true, but we can never absolutely know for sure whether that statement is actually true. Basically your comment isn't logical because you're saying we shouldn't study anything that is true because it's unfalsifiable by definition. In addition to falsification, Science can make statements that are vague and UN-definitive. Un-definitive claims include causal claims and correlative claims. Correlation is A and B are casually linked directly or through some other source. A maybe causes B or B maybe causes A or maybe a C is causing A and B. Causation is A and B are directly linked causally. A causes B. Science can make the two claims above but such claims are made based off of a sample. If the sample is bad or biased the claim is also bad and biased. Given that we live in an unknown universe of which any sample we extract can only be of finite size out of a potentially infinite size it makes both correlative and causative claims un-definitive. The sample can never fully encompass everything. This is the reason why in science, nothing can be proven. So in short science can definitively falsify things, and in a vague sense it can make statements about correlations and causation. The later two, though undefinitive can be used to say something about some topic of the universe that is true. But it can't say anything definitive and we can never know for sure. Because Falsification is limited to things that are false, so much of science doesn't actually use it even though it is a definitive metric. >As I mentioned however, I am seeking opinions here to the contrary of my gut reaction as my gut is often dumb and reactionary. The GP is kind of the one that's not too bright. You're definitely a layman but you ask the right questions. GP is not a layman and he's suppose to know certain things and not say what he said. He's very opinionated and makes broad statements without addressing nuance. So don't worry about it! You're not dumb! GP is utterly wrong. Falsification is still used on the frontiers of modern physics, though much of it nowadays is just correlative results, but if one such correlation is shown not to hold at all, it becomes a falsification. In fact, EVERY single experiment in modern physics is an attempt at falsification. And when the falsification fails... a correlative result is presented as the final answer. But note, even a falsification is not clear in certain cases. This is because our scientific observation tools have limited accuracy. Most of the time we just assume the observation tools are 100% accurate. But at the frontiers of modern physics the observation tools are so bad that the tools themselves yield probabilistic answers. So while in science falsification is still very much used to clearly discard theories, in modern physics, a negative result is not indicative of a clear cut falsification.
oneoneonetwo 4y ago
>which admittedly, many people still seem to want to do. This occurs because most science is not the "the frontiers of modern physics". But even then, falsification in the frontiers of modern physics is still a useful tool.

choeger 4y ago

> You’d have to explain what should be done about that. Well, there should be something that follows from the assumption that there are huge amounts of particles that only interact via gravity. What's the prediction that this model makes? If it doesn't make any useful prediction, it's pretty much worthless. > That’s living in the 19th century, basically - which admittedly, many people still seem to want to do. I think this attitude is inappropriate. The scientific progress made with 19th century methods, and that includes most of the stuff discovered until the 1950s, is massive. Maybe scientists should focus more on a clear nay or yea? I mean, the LHC is certainly a significant achievement, but compared to the detection of gravitational waves or even just exoplanets it doesn't seem to deliver as much bang for the buck.

YawningAngel 4y ago
Indulge the counterfactual for a second. Suppose that dark matter really does exist, and we can't ascertain that fact with any certainty. What do you want Physics as a whole to do about that? I think everybody agrees a simple answer would be great, we're just unable to provide one in this context
- naasking 4y ago
  > Suppose that dark matter really does exist, and we can't ascertain that fact with any certainty. What do you want Physics as a whole to do about that? Be more bold in exploring alternative gravity theories and conceiving of tests we can perform to check how gravity behaves in regimes that are not yet well tested.
cauch 4y ago
> I mean, the LHC is certainly a significant achievement, but compared to the detection of gravitational waves or even just exoplanets it doesn't seem to deliver as much bang for the buck. Well, first, LHC was "unlucky": they could have observed a lot of new particles if they would have been there. It feels unfair to say "this experiment that looked at X and had 50% chance of seeing something was a waste of money because they haven't found anything, this experiment that looked at Y and had 50% chance of seeing something was not because they saw something": both experiments are as worth a try. LHC still have observed the Higgs boson, which is still quite a big thing. LHC has also constrained a lot of theories (seeing that a particle predicted by a given theory is not there is very useful scientifically). But more importantly, I have the feeling that the notion of "bang for the buck" is pretty different from a scientific point of view and a layman point of view. For example, you mention exoplanets. Scientifically, those are really not a big deal. We already know that planets can exist, there are 8 around the Sun, so finding others around other stars is really not a big deal: everybody was expecting them to exist and knew the only reason they were not seen yet was the limits of the technology. One could even argue that using money to detect things that are pretty trivial is in fact the worst delivery of bang for the buck. Don't get me wrong, this is still nice to see them, but it's the same level as pentaquark (discovered at LHC). It is sometimes surprising to see the public/media reactions to discoveries. I'm still not sure why layman people were so excited about gravitational waves but not much about the tau neutrino.

onli 4y ago

If it is not falsifiable, it is not science nor physics. It is meta-physics, ersatz-religion. Which is exactly the core of the complaint against concepts like this. Of course same thinking can't easily be applied to things that are not always true, like social behaviour. Part of the reason the replication crisis hit so hard in the social sciences. And of course this is not really as absolute as I wrote above. No field works like this all the time. But it is a good guideline. If falsifiability is completely ignored, then a concept really is nothing more than a religious belief and should be considered as such. BTW, Popper did not publish in the 19th century.

pydry 4y ago
As far as I understood it if somebody could craft an experiment that would demonstrate, quite clearly that it was something other than a special kind of matter causing this effect then the theory would be falsified.

donatj 4y ago

> Naive falsificationism is an almost useless principle at the frontiers of modern physics. We’re long past the days when you could run some simple experiments and come up with a clear yea or nay. That’s living in the 19th century, basically - which admittedly, many people still seem to want to do. To a layman like myself this implies both that the things being studied then are both unprovable and not potentially of practical use or significance. I certainly am not saying that is the case, but rather am looking for more informed opinions to the contrary. The cynical side of me feels like study of something unfalsifiable is as the study of religion. A poor investment for public funds. As I mentioned however, I am seeking opinions here to the contrary of my gut reaction as my gut is often dumb and reactionary.

mytailorisrich 4y ago
The comment you replied to does not suggest that modern physics is unprovable... no simple experiment to prove something does not mean no experiment at all. The other, related point is that modern physics' frontier is in the quantum world and thus way outside layman's intuition, common sense, or reasoning. Even experts are struggling.
oneoneonetwo 4y ago
In science nothing is provable. Things can be falsified but if a hypothesis is true it can actually never be proven. This is because at any point in time and additional observation can be added to your sample that definitively disproves your hypothesis. All zebras have stripes. is my hypothesis. I observe 100 zebras and I see they have stripes. I see 10 million zebras and I see they have stripes. There is no amount of observations I can make to prove my hypothesis. However if at any point in time I see a zebra with no stripes, I have disproven my hypothesis. So anything that is in actuality true, is not falsifiable and also un-provable. In short, some statement about our universe can be true, but we can never absolutely know for sure whether that statement is actually true. Basically your comment isn't logical because you're saying we shouldn't study anything that is true because it's unfalsifiable by definition. In addition to falsification, Science can make statements that are vague and UN-definitive. Un-definitive claims include causal claims and correlative claims. Correlation is A and B are casually linked directly or through some other source. A maybe causes B or B maybe causes A or maybe a C is causing A and B. Causation is A and B are directly linked causally. A causes B. Science can make the two claims above but such claims are made based off of a sample. If the sample is bad or biased the claim is also bad and biased. Given that we live in an unknown universe of which any sample we extract can only be of finite size out of a potentially infinite size it makes both correlative and causative claims un-definitive. The sample can never fully encompass everything. This is the reason why in science, nothing can be proven. So in short science can definitively falsify things, and in a vague sense it can make statements about correlations and causation. The later two, though undefinitive can be used to say something about some topic of the universe that is true. But it can't say anything definitive and we can never know for sure. Because Falsification is limited to things that are false, so much of science doesn't actually use it even though it is a definitive metric. >As I mentioned however, I am seeking opinions here to the contrary of my gut reaction as my gut is often dumb and reactionary. The GP is kind of the one that's not too bright. You're definitely a layman but you ask the right questions. GP is not a layman and he's suppose to know certain things and not say what he said. He's very opinionated and makes broad statements without addressing nuance. So don't worry about it! You're not dumb! GP is utterly wrong. Falsification is still used on the frontiers of modern physics, though much of it nowadays is just correlative results, but if one such correlation is shown not to hold at all, it becomes a falsification. In fact, EVERY single experiment in modern physics is an attempt at falsification. And when the falsification fails... a correlative result is presented as the final answer. But note, even a falsification is not clear in certain cases. This is because our scientific observation tools have limited accuracy. Most of the time we just assume the observation tools are 100% accurate. But at the frontiers of modern physics the observation tools are so bad that the tools themselves yield probabilistic answers. So while in science falsification is still very much used to clearly discard theories, in modern physics, a negative result is not indicative of a clear cut falsification.

oneoneonetwo 4y ago

>which admittedly, many people still seem to want to do. This occurs because most science is not the "the frontiers of modern physics". But even then, falsification in the frontiers of modern physics is still a useful tool.