Wednesday, October 31, 2018

Dark matter. Or what? [Video]

Fixed some problems, created some new ones. Sorry for the glassy stare at some places; I had to redo the recording but had a gigantic headache. I promise not to do that again.


Click on “CC” to get English subtitles.

I encountered the same problem with YouTube as with the previous video, that the automatic transcription falsely identifies the language as German. (This time I left it there for your amusement.) I have no idea why this is. I have all my language settings in English and really I don't think my pronunciation is so bad you’d mistake my English for German. The only thing I learned by Googling the issue is that others have it too. In case you know a fix, please let me know.

Oh, and Happy Halloween :)

Update Nov 2nd: Now with Italian and German subtitles. Click “CC” to turn on captions, click on settings/gear icon to change language.

47 comments:

Phillip Helbig said...

" I don't think my pronunciation is so bad you’d mistake my English for German"

Your written English, based on your book and posts here, is very good, but your pronunciation gives you away as German right off the bat. Still, it should be possible to distinguish between German and English with a German accent. Feed, say, Chris Howland or Bill Ramsay into, speaking German, and see if it thinks that they are speaking English.

Liralen said...

I don't understand your concerns about the video, because it looked just fine to me. I do know that Google Assistant does a good job of interpreting foreign accents, relative to English, and hence is why I bought Google Home for my Mom (because my guess is that it will be the first to provide medical alerts to family members, and I want to get her used to it before she needs it), who is Japanese, although for some strange reason, it did better with Japanese than German: https://www.youtube.com/watch?v=gNx0huL9qsQ.

Interesting data point: My Mom's first interaction with Google Home was "OK Google, play music". It played Japanese music.

So much for form over substance.

With respect to the substance of your post, my feeling (i.e., not based on empirical evidence at all) is that both dark matter and MOND are both just placeholders for the fact that observations don't match theory.

I am bothered by the acceleration, which could be explained by either mass or force.

I like MOND because I think it's prettier because gravity involves both mass and force, despite the ugly math to calculate it. And because Newton’s law of universal gravitation lacks charges, despite its similarity in form to Coulombs Law which does, and that fact just seems ugly to me. And because if there is an anti-grav charge, it would appear to us as both a mass and a force.

I know you despise that way of thinking, and I respect you lots for doing so. But we must make some guesses in order to focus research. I think the research has focused on the wrong direction, which you have very brilliantly explained why, and it needs to change because it's not productive.


So much for form over substance.

With respect to the substance of your post, my feeling (i.e., not based on empirical evidence at all) is that both dark matter and MOND are both just placeholders for the fact that observations don't match theory.

I am bothered by the acceleration, which could be explained by either mass or force.

I like MOND because I think it's prettier because gravity involves both mass and force.

I know you despise that way of thinking, and I respect you lots for doing so. But we must make some guesses in order to focus research. I think the research has focused on the wrong direction, which you have very brilliantly explained why, and it needs to change because it's not productive.

I'm just not sure any reason other than that is needed.

Sabine Hossenfelder said...

Phillip,

I know that I have a Germany accent, but I don't think that explains it. I have done this a few times previously and it used to work just fine. It worked even for the more quiet pieces of my music videos.

Sabine Hossenfelder said...

Liralen,

The English captions that are on YouTube now don't come from the automatic transcription. I had to write those by hand. (Next time I'll use a third-party software for this; it's extremely cumbersome.)

Yes, of course dark matter and modified gravity are "placeholders" for something we don't know. But that's normal procedure: You quantify what you observe and then you look for the best theory to explain it. The problem is that we presently have this community-split where each side has a piece of what's probably correct but each side thinks they have the full picture. Best,

B.

George said...

I really, absolutely love the content of your video - from a science standpoint, it is the best and most complete explanation I have seen so far. But you could improve your overall presentation. In this regard, PBS spacetime is the gold standard. They have really nicely looking animations, graphs and speaker integration. Please also set up a Patreon so that fans can support you.
Please make more videos!

Sabine Hossenfelder said...

George,

To say the obvious, I'm not PBS. I'm a theoretical physicist who knows next to nothing about filming, or cameras, or, for that matter, about make-up.

I do not have a Patreon page because Patreon expects you to produce content at a regular rate. I have a job to do and kids to raise and cannot make any such commitment. If you want to support me, please use the "Donate" button in the top right corner. Or buy my book. Or both ;)

David Schroeder said...

Oh goody, a real treat for Halloween day, a 14 minute video on the puzzle of Dark Matter, or whatever it is. Just began watching it on this frosty -4 C. morning here in the northeast USA. This is much more interesting than tricky discussions of philosophy.

David Schroeder said...

Oops, I think I put my previous comment in the wrong post. Meant to put it here with the latest video.

David Schroeder said...

Oh wow, that was an outstanding video, completely objective, not favoring either approach exclusively, but suggesting a combo of both might be the way forward to solving this mystery. Thank you, Sabine. Looking forward to the next video.

trebonian said...

Re: German transcription
Perhaps Youtube is using some parameters from your GoogleID (used to upload the video). Is your default language or location set to German (or Germany)?

Sabine Hossenfelder said...

trobinian,

I have all my default languages on English, on my computer, on my browser, in all my apps. I checked the location setting on my YouTube account and it was still set to Sweden, so make of this what you wish. The default language on my YouTube videos is English. Indeed the video language was correctly set to English (though I double-checked). YouTube knows of course that the upload comes from a German IP address, that's the only explanation that makes sense to me. As I said, a year ago this worked just fine, even with heavy background "noise".

Now the issue is that there doesn't seem to be a way to ask YouTube to redo the automatic transcription in a different language.

Perry E. Metzger said...

Great video! I enjoyed watching it. I already more or less knew the content, but the explanation is very crisp.

By the way, what setup and software are you using for producing your videos?

neo said...

Sabine,

what do you think and call theories of gravity that try to reproduce MOND but don't introduce new fields and particles, but modify the einstein equation, such as f(R) gravity and verlidne's entropic gravity?

Sabine Hossenfelder said...

neo,

Verlinde does not modify Einstein's equations. He works not only in a non-relativistic limit but also only looks at the spherically symmetric, static case. He explicitly introduces a new field, it's called the "displacement vector" in his paper.

As I say in the video, there are modifications of gravity that do not introduce new fields, but those are not the ones that explain the observations normally attributed to dark matter in galaxies.

Uncle Al said...

Modeling a net matter universe of galaxies and clusters may be contained in disciplines lacking physics’ default omissions. Cheap, fast, and facile heresy? Look.

"simulation" 3900 ARGO gps floaters in the oceans (increasing toward 5000) plus multiple satellite telemetries plus… Simulations are dense over space and time, plus multispectral consistency.

http://www.argo.ucsd.edu/
https://earth.nullschool.net/
… "real time" Drag to rotate, mouse wheel for scale. Click "earth" for options
http://www.aoml.noaa.gov/phod/docs/jgrc20210.pdf
… DOI:10.1002/jgrc.20210
http://www.adrift.org.au/
… Play the simulation.

https://phys.org/news/2018-07-reveals-foreign-kelp-surfed-antarctica.htm
… Simulations were crashed by kelp.

JimV said...

About donations and Patreon, when I use PayPal to make a donation here, it asks me if I want to make the contribution monthly (instead of a single contribution). So it seems to me the same mechanism as Patreon is available, via PayPal. (I haven't used the monthly option though so I can't verify it works, but my confidence in PayPal is good.)

I prefer text to video explanations personally, but videos probably reach people that text doesn't, so it is a valuable service and I will make a donation for it. (It only takes a minute.)

Sabine Hossenfelder said...

JimV,

Thanks, I appreciate the support.

Personally, I also prefer text to video as I tend to find videos both contain too much superfluous information (all this visual input!) and they take up too much time. But there are some topics where visualizations are really helpful (take, eg, the CMB peaks or the rotation curves). And, as you, point out, it reaches some people who are hard to reach by text alone.

neo said...

Bee

thanks for clarification.

how difficult would it be to propose theory of gravity modeled on quantum phase transition similar to condense matter physics where the cc is the transition regime, where above the cc gravity behaves classical and below cc it behaves quantum mechanical, and reproduces MOND

has any condense matter physicist attempted this challenge?

Sabine Hossenfelder said...

Neo,

The cosmological constant is an energy density and it's not the relevant property for a phase transition.

I can tell you that there is pretty much no work on how superfluids condense in gravitational potentials. The mathematical treatment is generally involved, as it requires you to deal with two types of fluids (the normal and the superfluids) that are generally both present. Justin Khoury and his group are working on an ansatz for this (I wrote about this here), but I have my reservations about this. There was a new paper on this on the arXiv just last week, but it's a long paper and I have not read it.

The brief summary is it's difficult and it's something that condensed matter physicists never look at because for them the gravitational potential, if it exists at all, is a vertical gradient. A particularly difficult issue is the question what a superfluid with a long-range interaction would do to structure formation as this is something which usually can't be done analytically.

Yours is a good question to ask because a lot of people tend to think, erroneously, either (a) it's all trivial or (b) it's all been done, but neither is the case. What is correct is that there is basically no work on the topic because all the funding is eaten up by people who work on fancy computer simulations for particle dark matter.

neo said...

Bee,

thanks. unlike Khoury et al, treatment, I am thinking gravity is itself behaves like a condense matter fluid.

"The cosmological constant is an energy density and it's not the relevant property for a phase transition."

something empirical then, since empirically cc seems to be related to the MOND scale of acceleration ao. it could be a numerological coincidence, or it could be something important.

just create a quantum phase transition diagram that empirically matches MOND and GR, worry about how this all works later.

above cc, GR represents one quantum phase transition of gravity, below cc, MOND represents the other quantum phase transition of gravity. just get it to match observation and also consistent with quantum physics.


I think Wen and Volvolik offered similar ideas, that gravity is a quantum fluid, though they don't connect it with MOND, and both are condense matter.

Sabine Hossenfelder said...

neo,

yes, maybe, or maybe not. What I am saying is that words are not sufficient. Words are enough to make videos, not to make theories. Theories take time and effort which means in plain English, they take money. No money, no people, no theory.

SRP said...

Another terrific video. The level of the visuals is just right--no need to jazz it up--and your crisp manner and objective approach enhance the credibility of your judgments.

Lawrence Crowell said...

Neo wrote: how difficult would it be to propose theory of gravity modeled on quantum phase transition similar to condense matter physics where the cc is the transition regime, where above the cc gravity behaves classical and below cc it behaves quantum mechanical, and reproduces MOND

Bee responded with:The cosmological constant is an energy density and it's not the relevant property for a phase transition.

The cosmological constant is an energy density, which except for in quintessence models is constant through space. With inflationary cosmology we think it is not constant in time, and the big rip proposes a future where the vacuum energy density increases and even asymptotes to “infinity” some finite time in the future. The idea sounds mad, but I keep big rip and phantom energy in an “open file.”

Inflationary cosmology the vacuum evolved with time, where the value was enormously larger for a time interval Δt_i = 10^{-36}sec some 10^{-33}sec into the early evolution of the observable cosmos. The cosmological constant was Λ ~ 10^{-12}/ℓ_p^2. Now multiply Δt_i×Λ ~ 10^{18}s/m^2 and by multiplying by c gives ~ 10^{26}m^{-1}. This when multiplied by a meter gives a multiplication factor by which that meter is expanded ~ 10^{26} ~ e^{60}. This is a back of the envelope way of thinking about what is called the 60 e-folds. This means the causal universe with Λ ~ 10^{-12}/ℓ_p^2 ~ 10^{54}m^{-2} and a horizon length d = sqrt{3/Λ) ~ 10^{-26}m was expanded into a region about one meter in radius. This happened in 10^{-36}sec, which means the expansion of space had a velocity v ~ 10^{36}m/s ~ 10^{27}c!

This ended because this extremely large cosmological constant transitioned into the small cc we have today. This means a high energy false vacuum collapsed into the physical vacuum with a tiny energy and the mass-energy gap was converted into radiation and matter.

Now suppose I were interested in finding a solid state physics analogue of this. I think one start would be to look at mass renormalization in a Brillouin zone. An electron can have a very large mass, or even a negative mass! There might then be different phases for the momentum of an electron p = ½mv^2, such as with graphene where m ~ 0 and v ~ c and things are actually relativistic, or there may be a large mass induced on the electron so the velocity is very slow. This sort of thing does happen with edge or boundary states that can have a different conduction or insulating phase compared to the bulk interior. This is what happens with Mott insulating phases that are topological states, or states with topological protected symmetries. So where might we have something like that with cosmology? Clearly AdS/CFT is a case and if the the vacuum can locally exhibit symmetry breaking there can be boundaries of causal wedges with topological edge states. These would be holographic, and in AdS_5 this might define a dS_4 spacetime.

I could go into considerable depth on this, and in fact I transmitted a very long email on this. I have generated some numerical graphs as well, which this blog does not support. Further I would not transmit that without Bee's permission for this might be too much self-promoting.

Sheever said...

Your facial expressions are very rigid and controlled. Relax :)

Sheever said...

Good comment. Condensed matter physics constantly give more hint to the topic you speak here.its relevance increasing as technology improving

Uli Thomsen said...

Maybe the youtube language settings identify your upload originated from Germany, so it 'must' be German lanuage... Next time you could try to upload from France, Bavaria ;-) or any other foreign country to check this :-)

Uli Thomsen said...

And I would prefer text to video...!

Videos are so booooring... https://www.youtube.com/watch?v=ZWmHrjWhCKU :-)

Unknown said...

Cracking waffle Sabine! I have to watch a second time.
Even though I am only a late beginner and do not know most of the terms yet; I like the philosophy of "know yourself well first". So why not start at the simple question of Why does the earth have an elliptic orbit? A ellipse being a complex construction with many components. Know this well first and then cast to the galaxy ; universe ...

Goran SAVIC said...

A correct overview of the subject, at least.
In my opinion, we have to abandon an idealized picture of four-dimensional spacetime -- it's a so virtual concept, applicable only at some scales.

Marco Parigi said...

Hi Sabine,
You mention Modified Gravity in the context of MoND. However, there are peer reviewed papers that consider modified inertia instead. That is, a reduction in inertial mass at low accelerations is equivalent to a greater Force or more Gravity. It would have been nice if this alternative was mentioned for completeness, if for no other reason.

I like your videos, and they really add to the written word.

George Herold said...

Sabine, I really loved the content! I hope you won't take this wrong, but you had rather a stark demeanor. (maybe it was the headache.) Some smiles and such would be welcome.
(But maybe that's just me. :^)

abyss said...

As a good German speaker, you should know the difference between "Schomberg" and "Schombert"

Unknown said...

Dear Sabine. These videos are excellent! Please stop criticizing yourself in the introduction to the videos and just let the viewer decide for himself if you have a glassy stare or not. That, and a smile every now and then, and your videos are on par with Don Lincoln.

David Brown said...

I predict that Milgrom will be a Nobel prize winner by the end of the year 2023. Google "kroupa milgrom", "mcgaugh milgrom", "sanders milgrom", "scarpa milgrom", and "finzi milgrom".

Sabine Hossenfelder said...

David,

Oh, I would love to see that! But I think 2023 is way too optimistic. It would take a decade at least, I suspect, to sort out the situation.

Phillip Helbig said...

"I predict that Milgrom will be a Nobel prize winner by the end of the year 2023."

Probably not, for the same reason that Vera Rubin didn't get one. Both put forward an explanation for an observed anomaly, but with neither independent confirmation of the explanation nor with a viable theory to back it up. I don't think that there will be a relativistic theory of MOND by 2023. I think it rather unlikely that anyone will get a Nobel Prize for work in this area in the next 20 years, but if I had to bet, I would bet on Justin Khoury.

Phillip Helbig said...

"You mention Modified Gravity in the context of MoND. However, there are peer reviewed papers that consider modified inertia instead. That is, a reduction in inertial mass at low accelerations is equivalent to a greater Force or more Gravity. It would have been nice if this alternative was mentioned for completeness, if for no other reason."

Milgrom actually played with this idea. I think I have an email from Bob Sanders saying why it isn't a viable concept. If I can find it I'll post a summary.

Phillip Helbig said...

Milgrom's paper on modified inertia is in Annals of Physics, 229, pp. 384--415 (1994), also at arXiv:astro-ph/9303012. Bob's executive summary is that such a theory, if viable, would have to be non-local.

Marco Parigi said...

Philip,

Thank you for that. I shall look at Milgrom's take on modified inertia. I note also that Mike McCulloch has some peer reviewed papers on modified inertia. At face value, these ideas appear to be viable, and Mike's ideas relate to a Hubble Scale Casimir effect, which is maximally non-local. So, if Bob's idea is correct, that modified inertia has to be non-local, we are talking about a peer reviewed paper that includes this necessary aspect. I suggest that you look up these papers yourself, to be convinced that Bob Sanders is correct in that modified inertia is non-viable.

Joe said...

Hi, The Video in this post is the String Theory video not the Dark Matter video.

Anon

opamanfred said...

@David Brown
I have indeed Googled "finzi milgrom" and the like. What I found is that a certain D.B. posts the same comment over and over again on blogs about cosmology...

Phillip Helbig said...

One remark on pronunciation. While the German accent is OK (I wish that I could get rid of my accents when speaking other languages!), one thing jumps out, not really related to a German accent: the "u" in "Bullet cluster" is pronounced essentially exactly like a German short "u", e.g. as in "bunt" (which definitely doesn't rhyme with the English word "hunt" nor with the English word "bunt" (which does rhyme with "hunt")).

Yehonatan Knoll said...

Sabine, your popular-level introduction to the subject had to omit (perhaps) the most important aspect of the dark-matter problem: How does one say "matter" in mathematics, viz., what to put on the r.h.s. of Einstein's equations. Einstein himself was not sure of the answer (calling the r.h.s. "the wing .... made of cheap wood" as oppose to the l.h.s. "marble wing").

That was before the quantum revolution. Nowadays - so the story goes - matter is said to be represented by QM, but what does this mean? Should one represent matter by a single-particle Dirac field? By a quantum field? In either case the l.h.s. is real valued, whereas the r.h.s. is either complex or operator valued. Or should one take the vacuum expectation value (VEV) of those? But why? The VEV, in its usual usage, simply refers to the expectation value of a large number of measurements preformed on identically prepared systems, and there is one universe only. Or, perhaps, a `sufficiently matterish', calssical energy-momentum tensor should be used? But this matter is ultimately made of atoms, subject to QM, no? Why should we trust this phenomenological representation anymore than Einstein did?

If everything would have worked perfectly for GR - as was the general impression until the 1970's -then Einstein's discomfort with his own creation could, perhaps, have been ignored. But with the current crisis of dark-matter, such mathematical vagueness can't be tolerated.

I believe it is the source of the dark-matter mystery.

Sabine Hossenfelder said...

Yehonatan,

In the limits that are relevant for the problem under consideration the use of the classical source term is appropriate. You can convince yourself of by estimating how large the quantum contributions are. Note that this does not mean that quantum effects do not play a role for the equation of state, it merely means that you can effectively describe dark matter as a fluid and use the semi-classical limit.

Yehonatan Knoll said...

Sabine,

You give the "canonical" answer; We know how to represent matter in all sorts of other applications (hydrodynamics etc.) and it works perfectly well. Or does it...? When we use those classical representations in domains very different from those intended by their nineteenth century inventors, and at small scales in particular, they completely fail. QM takes over, with the emergence of classical physics out of QM being as mysterious today as a century ago (I mean the full emergence, measurement problem etc., not some formal hbar-->0 limit). Isn't it possible that, also on the vast scales of astrophysics, that nineteenth century representation of matter fails, hence the dark-matter problem?

I have worked out in details this conjecture, and it seems very plausible to me.

DG said...

Many thanks for a clear explanation that enabled me as a layman to understand.

Dovid Gottlieb

Unknown said...

Hi Sabine,

I like your video. Some random thoughts: (N=1 SUGRA) GR can be derived by gauging the (N=1 Super-)Poincare algebra, and Newtonian gravity (in the form of Newton-Cartan) by gauging the Bargmann algebra. How about modified gravity? In particular, can MOND be derived from a deformation of the Bargmann algebra?

In contrast, N=2 SUGRA cannot be obtained by a simple gauging of the N=2 Super Poincare algebra (in the relativistic, D=4 case) because the additional of an extra U(1) field on top of the graviton and gravitino; would the extra fields of modified gravity prevent a derivation of a simple gauging procedure similarly?

Just wondering ;)