In this video I want to tell you where I stand on the question of dark matter or modified gravity.
As I said in my two earlier videos, we have a lot of
observational evidence that we are misunderstanding something about
the way the universe works. Either there must be a new type of matter,
the so called dark matter,
or gravity works differently than Einstein taught us, and this is called
modified gravity. If you ask an astrophysicist, they will tell you that
dark matter can explain all the observations,
while modified gravity cannot,
and therefore dark matter is the better explanation.
For a long time I thought that this was a reasonable argument, but I no longer think so.
Before I tell you why I changed my mind, I want you to do a little thought experiment.
Imagine it's the 18th century and you're Daniel Bernoulli and you have
just discovered the key equations of fluid dynamics.
You've tested them back and forth and they work great.
Now you want all your colleagues to check these equations for you, so you send letters to them
and you ask them to please repeat your experiment and to report back if the equations work.
One after the other, they all write back: yes the equations work great, you're a genius.
Except for the guy in Alaska who says
You're an idiot.
Now you can go and add all kinds of parameters to your equations and try to
explain why water sometimes remains in
the shape of a glass, but that would be entirely insane. What you want to do
instead is to go and look for environmental parameters that explain
why water sometimes behaves differently.
Let us now come back to dark matter.
Astrophysicists normally describe dark matter as a fluid. And I know this can
be confusing, but a fluid can be a gas.
Describing dark matter as a fluid works very well for the early universe and for
the Cosmic Microwave Background and for the galaxy clusters.
It does not work so well for the galaxies.
For the galaxies modified gravity is the much simpler explanation.
So should you go now and add all kinds of parameters to the equations for dark matter
to make them fit the galaxies?
I think that's as insane as trying to explain ice by adding all kinds of parameters to the equations of hydrodynamics.
What we should do instead is look for environmental parameters that separate one behavior from the other.
I think what the data are trying to tell us is that matter comes in two
different phases, and we have to use different equations for each phase.
The first time I came across this idea was in a 2015 paper by Berezhiani and Khoury.
They pointed out that the mathematical structure of modified gravity looks very much
like the mathematical structure of a superfluid, so the two phases of dark matter are not,
as in the case of water, a liquid in a solid, but it's a fluid and a superfluid.
What's a super fluid?
A superfluid is a fluid that has no internal friction and that has quantum correlations
that span over very long distances. In the case of dark matter,
these correlations can span through whole galaxies.
Now the idea that Dark Matter may form a superfluid is not new. What is new here is that the superfluid
gives rise to a force which acts on the normal matter.
This new force is the reason why the effects of superfluid dark matter can look like modified gravity.
If you remember what I told you earlier, you can use this force to
calculate the relation between the velocity of a star and the distance from
the galactic center. If you draw this on a graph, then the force that you get from
the normal gravitational pull of the normal matter is not enough to explain
the observations. If you now add the force that comes from the superfluid,
it works nicely. So let me be clear that this additional force is not the
gravitational pull from the superfluid. It's actually a new force that comes
from the interaction of the super fluid with the normal matter.
But if you want dark matter to condense to a superfluid, you need it
to be cold, and you need a gravitational potential that is deep enough to create
sufficient pressure. These conditions are not fulfilled in the early universe and
they are not fulfilled in the galaxy clusters. They are fulfilled in the
galaxies, and this explains why modified gravity sometimes works and sometimes doesn't.
Sounds good, huh? But it's not so simple.
Really what I just told you is merely words and the equations to back it up are not quite there yet.
The biggest problem is that it's not very well understood under exactly which
conditions Dark Matter forms a superfluid. There are also different
kinds of particles that can form a superfluid and it's not clear which of
those fits the data best. Another problem is that it's really not well understood
how a fluid condenses to a superfluid in a curved space-time. That's because the
people who normally study superfluids don't have to think about gravity all that much.
If they take it into account at all, it's a vertical gradient in the laboratory,
but you can't describe galaxy formation with that.
Can you experimentally distinguish this
type of superfluid dark matter from the normal dark matter?
I believe you can.
This is something that I have been working on with a PhD student in the past year.
His name is Tobias Mistele.
Maybe I will tell you something about this some other time.
The message that I want to get across here
is that I think it's a mistake to regard dark matter and modified gravity as two
competing theories, each of which has to be made to fit all of the data.
To me the data say the answer is a combination of both.