For decades physicists have been convinced that most of our universe is invisible, but how do we know that if we can’t see it? I want to explain the thought process that leads one to believe in a theory via indirect evidence. For those who want to see a nice summary of the evidence, check this out. So this post isn’t 3000 words, I will simply say that either our theories of gravity are wrong, or the vast majority of the matter in our universe is invisible. That most of the matter in the universe is invisible, or “dark”, is actually well supported. Dark matter as a theory fits the data much better than modifications to gravity (with a couple of possible exceptions like mimetic dark matter). This isn’t necessarily surprising; frankly it would be a bit arrogant to assume that only matter similar to us exists. Particle physicists have known for a long time that not all particles are affected by all the fundamental forces. For example, the neutrino is invisible as it doesn’t interact with the electromagnetic force (or strong force, for that matter). So the neutrino is actually a form of dark matter, though it is much too quick and light to make up most of what we see.
The standard cosmological model, the ΛCDM model, has had tremendous success explaining the evolution of our universe. This is what most people refer to when they think of dark matter: the CDM stands for “cold dark matter”, and it is this consistency that allows us to explain observations from almost every cosmological epoch that is so compelling about dark matter. We see the effect of dark matter across the sky in the CMB, in the helium formed in primordial nucleosynthesis, in the very structure of the galaxies. We see dark matter a minute after the big bang, a million years, a billion years, and even today. Simply put, when you add in dark matter (and dark energy) almost the entirety of cosmological history makes sense. While there some elements that seem to be lacking in the ΛCDM model (small scale structure formation, core vs cusp, etc), these are all relatively small details that seem to have solutions in either simulating normal matter more accurately, or small changes to the exact nature of dark matter.
Dark matter is essentially like a bank robber: the money is gone, but no-one saw the theft. Not knowing exactly who stole the money doesn’t mean that someone isn’t living it up in the Bahamas right now. The ΛCDM model doesn’t really care about the fine details of dark matter: things like its mass, exact interactions and formation are mostly irrelevant. To the astrophysicist, there are really two features that they require: dark matter cannot have strong interactions with normal matter (electromagnetic or strong forces), and dark matter must be moving relatively slowly (or “cold”). Anything that has these properties is called a dark matter “candidate” as it could potentially be the main constituent of dark matter. Particle physicists try to come up with these candidates, and hopefully find ways to test them. Ruling out a candidate is not the same as ruling out the idea of dark matter itself, it is just removing one of a hundred suspects.
Being hard to find is a crucial property of dark matter. We know dark matter must be a slippery bastard, as it doesn’t interact via the electromagnetic or strong forces. In one sense, assuming we can discover dark matter in our lifetime is presumptuous: we are assuming that it has interactions beyond gravity. This is one of a cosmologist’s fondest hopes as without additional interactions we are screwed. This is because gravity is by far the weakest force. You can test this yourself – go to the fridge, and get a magnet. With a simple fridge magnet, weighing only a few grams, you can pick up a paperclip, overpowering the 6*10^24 kg of gravitational mass the earth possesses. Trying to get a single particle, weighing about the same as an atom, to show an appreciable effect only through gravity is ludicrous. That being said, the vast quantities of dark matter strewn throughout our universe have had a huge and very detectable gravitational impact. This gravitational impact has led to very successful and accurate predictions. As there are so many possibilities for dark matter, we try to focus on the theories that link into other unsolved problems in physics to kill two birds with one stone. While this would be great, and is well motivated, nature doesn’t have to take pity on us.
So what do we look for in indirect evidence? Essentially, you want an observation that is predicted by your theory, but is very hard to explain without it. If you see an elephant shaped hole in your wall, and elephant shaped foot prints leading outside, and all your peanuts gone, you are pretty well justified in thinking that an elephant ate your peanuts. A great example of this is the acoustic oscillations in the CMB. These are huge sound waves, the echo of the big bang in the primordial plasma. The exact frequency of this is related to the amount of matter in the universe, and how this matter interacts. Dark matter makes very specific predictions about these frequencies, which have been confirmed by measurements of the CMB. This is a key observation that modified gravity theories tend to have trouble explaining.
The combination of the strong indirect evidence for dark matter, the relative simplicity of the theory and the lack of serious alternatives means that research into dark matter theories is the most logical path. That is not to say that alternatives should not be looked into, but to disregard the successes of dark matter is simply foolish. Any alternative must match the predictive power and observational success of dark matter, and preferably have a compelling reason for being ‘simpler’ or philosophically nicer then dark matter. While I spoke about dark matter, this is actually something that occurs all the time in science: natural selection, atomic theory and the quark model are all theories that have all been in the same position at one time or another. A direct discovery of dark matter would be fantastic, but is not necessary to form a serious scientific consensus. Dark matter is certainly mysterious, but ultimately not a particularly strange idea.
Disclaimer: In writing this for a general audience, of course I have to make sacrifices. Technical details like the model dependent nature of cosmological observations are important, but really require an entire blog post to themselves to answer fully.