[fuhrysteve]

> I've always felt this way about dark matter, but as a lay person with no background in physics, it's an opinion I'm hesitant to express.

I have always had the same inclination. It seems like a classic candidate for using Occam's razor. Would love to hear someone who knows more about physics weigh in with a more nuanced viewpoint.

[pja]

(I have a physics degree, but it’s very rusty at this point)

Particle physics has a long history of noticing gaps in the experimental data that don’t match up with current theories and positing that maybe a new particle would solve the problem. This goes right back to things like the neutrino (which was posited to preserve momentum in certain particle interactions IIRC) and even the neutron.

Sometimes the suggestion pans out, sometimes it doesn’t. You can see this kind of thing in action - when there was that anomaly in the CERN data a year or two ago there were a flurry of papers suggesting possible particles, none of which panned out because the anomaly turned out to be statistical noise. The various particulate explanations for the observed discrepancy between the spiral galaxy rotation curves and the observed distribution of matter are very similar: Take a gap in the data & speculate about what kind of particle would fill the hole.

When you combine the fact that, historically, speculating about whether new particles could explain anomalies in the observed data has been very productive for C20th physics with the reality that physicists are also loath to toss aside general relativity (which has been spectacularly successful) just because of one observed anomaly that has yet to be explained, it’s entirely unsurprising that physicists are keen on new particles as explanations for the galactic rotation curves.

At the same time, every physicist will happily admit that there’s a great big hole in the standard model - i.e. that it doesn’t include gravity at all - and that any unifying theory might also explain away the galaxy rotation curves in a way that doesn’t require new particles. Indeed, if a unifying theory could pull off that trick it would be evidence that the theory has a good chance of being true. Sadly, to date, no one has managed to come up with such a unification, and we have no real idea what such a theory would look like beyond the obvious: that it should explain the observed facts & match QFT & GR in the regimes where those theories are accurate.

[pka]

My impression was that QFT includes gravity and "only" breaks down when things get extreme, i.e. black holes/big bang/etc.

[pja]

No, there is no quantum theory of gravity & QFT doesn’t include gravity at all. The best you can do is add gravitational effects as an after-thought, but the things you might analyse with the full QFT are generally far too small & short lived for gravitational effects to matter anyway.

[raattgift]

> No, there is no quantum theory of gravity

Here's a couple of examples of quantum theories of gravity.

http://www.phys.lsu.edu/faculty/pullin/talks/pire1.pdf

http://www.staff.science.uu.nl/~hooft101/lectures/erice02.pd...

http://einrichtungen.ph.tum.de/T31/seminars-past/seminar-tal...

http://www.damtp.cam.ac.uk/research/gr/public/qg_ss.html

The extent to which these are complete, consistent, natural (in the fine-tuning sense), and so forth is debatable but these are certainly existing examples of quantum theories of gravity, and indeed the first is a perfectly reasonable Effective Field Theory that people work in regularly.

> QFT doesn’t include gravity at all

I think you mean "The Standard Model of Particle Physics", which is a quantum field theory (as is e.g. perturbative quantum gravity).

> the things you might analyse with the full QFT

Atoms and molecules have gravitational fields; when you send one through a double slit, which way does their gravitational field go?

http://www.nature.com/nnano/journal/v7/n5/abs/nnano.2012.34....

A quantum theory of gravity is needed to answer that.

Assemble a huge number of particles in superposition with 1/(sqrt 2) (|M @ a> + |M @ b>), with a & b separated. A quantum theory of gravity is needed to describe the gravitational influence of M on a small test object (General Relativity's answer is just wrong :( ).

[pja]

I’m not sure I believe that a mathematical formalism that is unable to make useful real world predictions deserves the “theory” moniker. Hence your list of quantum theories of gravity aren’t.

But perhaps that’s me being picky :)

> I think you mean "The Standard Model of Particle Physics", which is a quantum field theory (as is e.g. perturbative quantum gravity).

Sure: I was just quoting the parent comment & using the term informally to stand in for the mouthful that is TSMoPP.

(I’d love to see an experimental setup that was capable of detecting the gravitational field of a single molecule: that would be impressive!)

[raattgift]

> I’m not sure I believe that a mathematical formalism that is unable to make useful real world predictions deserves the “theory” moniker. Hence your list of quantum theories of gravity aren’t.

On the contrary, the ones I listed are all completely in accord with General Relativity up to strong gravity and absent superposed sources, which is found from studying the renormalization group flow of perturbative quantum gravity and is four loops of gravitons in a 3+1 dimensional spacetime. Strong gravity can only be found very close to the singularity of black holes (and well inside event horizons, except at the final evaporation), or in the very early universe. So we're good for neutron stars, and have no problems studying things around the event horizons of astrophysical black holes.

The only new mathematical formalism in perturbative quantum gravity is renormalization, and that goes back to the 1980s. Perturbative quantum gravity itself comes from the 1990s.

Sean Carroll has a good explanation of renormalization and effective field theory here:

http://www.preposterousuniverse.com/blog/2013/06/20/how-quan...

Asymptotically safe gravity posits an ultraviolet fixed point at which one can take a finite number of measurements, producing a strong gravity completion that perturbative renormalzation cannot; this is prompted by asymptotic safety in QCD. Below that limit, ASG completely matches perturbative quantum gravity, and so in the EFT limit it's the same as General Relativity.

There are five or six of other viable families of quantum theories of gravity, where viability means they accord exactly with perturbatively quantized General Relativity in its effective field limit, and thus agree completely with GR in the classical limit and weak gravity, and additionally are candidates as fundamental theories because they do not rely on perturbative renormalization by power counting and thus are expected to be useful to arbitrarily high energies.

Additionally it is not wildly irresponsible to think that mathematical research (perhaps not driven by physics!) will produce a tractable renormalization that does not require nature to select a convenient effect to suppress the explosion of parameters at high energies.

> I’d love to see an experimental setup that was capable of detecting the gravitational field of a single molecule: that would be impressive!

Everyone would. We're down below milligrams and yoctoNewtons:

https://arxiv.org/abs/1602.07539

http://newscenter.lbl.gov/2014/06/26/smallest-force-ever-mea...

I'm not as au fait about how the other side of the tunnel is approaching the ultimate meeting point, but it's not unreasonable to think of nanogram masses in superposition. Experiments were only at thousands of atomic mass units a few years ago, though: https://arxiv.org/abs/1310.8343

Unfortunately General Relativity can only have the whole gravitational influence of these molecules go through one or the other slits. However all of the quantum theories in my previous message have the distribution of the gravitational influence follow distribution of the matter, as one would expect.

[dahart]

What does Occam's razor say about a contradiction?

If two things that are observably exactly the same behave differently, then what should one conclude is the simplest explanation?

If you look in a telescope for a year, and notice that an apparently free-floating star out in the galaxy with no neighbors is taking a sharply curved path and everything else in the sky is going basically straight, you have two choices: 1) either something you can't see yet is there and acting on the star, 2) conservation of momentum, and the law of equal and opposite reactions need to be thrown out because they're wrong, and we need to start from scratch with physics.

Which of those two would Occam say is more likely?

[mikeash]

If you're looking at the sky and there appears to be a lot more stuff there than just the bits that emit light, wouldn't Occam's Razor suggest that the presence of more stuff is the most natural explanation, rather than rejiggering the fundamental laws of physics?

[roryisok]

It's more like you're looking at the sky and what you can see doesn't weigh as much as it should. Does Occam's Razor suggest that there's invisible stuff? or that you're weighing stuff wrong?

[mikeash]

Except it's not, because the "missing" mass is not distributed in the same way as the stuff that emits light. Some places have a lot more "missing" than others.

[cozzyd]

If it was just galactic rotation curves that would be explained by adding a (mostly) non-interacting dark matter particle, then MOND would probably have many more proponents. It's harder to understand how to replace non-interacting particles in explaining the cosmic-microwave-background power spectrum, galaxy clusters, and structure formation. Maybe it's possible to make a complicated theory do those things, but at that point, I'm not sure it wins on simplicity anymore.

[roywiggins]

Adding in a single extra particle is a lot simpler than rejiggering the entirety of our theories of gravitation. It's also a lot easier to test, just go out and find the particle. This worked for neutrinos, antimatter, the Higgs boson...

The Higgs boson took a long time to be discovered, but people thought it would be there because the alternative was the Standard Model being wrong. And so they looked, and there it was. Perfectly compatible with the Standard Model.

[olavk]

Occam's Razor would only apply if the theories made the same prediction with and without the dark matter. You cannot use Occam's Razor to explain away a discrepancy between theory and observation.