Wednesday, September 21, 2016

We understand gravity just fine, thank you.

Yesterday I came across a Q&A on the website of Discover magazine, titled “The Root of Gravity - Does recent research bring us any closer to understanding it?” Jeff Lepler from Michigan has the following question:
Q: Are we any closer to understanding the root cause of gravity between objects with mass? Can we use our newly discovered knowledge of the Higgs boson or gravitational waves to perhaps negate mass or create/negate gravity?”
A person by name Bill Andrews (unknown to me) gives the following answer:
A: Sorry, Jeff, but scientists still don’t really know why gravity works. In a way, they’ve just barely figured out how it works.”
The answer continues, but let’s stop right there where the nonsense begins. What’s that even mean scientists don’t know “why” gravity works? And did the Bill person really think he could get away with swapping “why” for a “how” and nobody would notice?

The purpose of science is to explain observations. We have a theory by name General Relativity that explains literally all data of gravitational effects. Indeed, that General Relativity is so dramatically successful is a great frustration for all those people who would like to revolutionize science a la Einstein. So in which sense, please, do scientists barely know how it works?

For all we can presently tell gravity is a fundamental force, which means we have no evidence for an underlying theory from which gravity could be derived. Sure, theoretical physicists are investigating whether there is such an underlying theory that would give rise to gravity as well as the other interactions, a “theory of everything”. (Please submit nomenclature complaints to your local language police, not to me.) Would such a theory of everything explain “why” gravity works? No, because that’s not a meaningful scientific question. A theory of everything could potentially explain how gravity can arise from more fundamental principles similar to, say, the ideal gas law can arise from statistical properties of many atoms in motion. But that still wouldn’t explain why there should be something like gravity, or anything, in the first place.

Either way, even if gravity arises within a larger framework like, say, string theory, the effects of what we call gravity today would still come about because energy-densities (and related quantities like pressure and momentum flux and so on) curve space-time, and fields move in that space-time. Just that these quantities might no longer be fundamental. We’ve known since 101 years how this works.

After a few words on Newtonian gravity, the answer continues:
“Because the other forces use “force carrier particles” to impart the force onto other particles, for gravity to fit the model, all matter must emit gravitons, which physically embody gravity. Note, however, that gravitons are still theoretical. Trying to reconcile these different interpretations of gravity, and understand its true nature, are among the biggest unsolved problems of physics.”
Reconciling which different interpretations of gravity? These are all the same “interpretation.” It is correct that we don’t know how to quantize gravity so that the resulting theory remains viable also when gravity becomes strong. It’s also correct that the force-carrying particle associated to the quantization – the graviton – hasn’t been detected. But the question was about gravity, not quantum gravity. Reconciling the graviton with unquantized gravity is straight-forward – it’s called perturbative quantum gravity –  and exactly the reason most theoretical physicists are convinced the graviton exists. It’s just that this reconciliation breaks down when gravity becomes strong, which means it’s only an approximation.
“But, alas, what we do know does suggest antigravity is impossible.”
That’s correct on a superficial level, but it depends on what you mean by antigravity. If you mean by antigravity that you can let any of the matter which surrounds us “fall up” it’s correct. But there are modifications of general relativity that have effects one can plausibly call anti-gravitational. That’s a longer story though and shall be told another time.

A sensible answer to this question would have been:
“Dear Jeff,

The recent detection of gravitational waves has been another confirmation of Einstein’s theory of General Relativity, which still explains all the gravitational effects that physicists know of. According to General Relativity the root cause of gravity is that all types of energy curve space-time and all matter moves in this curved space-time. Near planets, such as our own, this can be approximated to good accuracy by Newtonian gravity.

There isn’t presently any observation which suggests that gravity itself emergens from another theory, though it is certainly a speculation that many theoretical physicists have pursued. There thus isn’t any deeper root for gravity because it’s presently part of the foundations of physics. The foundations are the roots of everything else.

The discovery of the Higgs boson doesn’t tell us anything about the gravitational interaction. The Higgs boson is merely there to make sure particles have mass in addition to energy, but gravity works the same either way. The detection of gravitational waves is exciting because it allows us to learn a lot about the astrophysical sources of these waves. But the waves themselves have proved to be as expected from General Relativity, so from the perspective of fundamental physics they didn’t bring news.

Within the incredibly well confirmed framework of General Relativity, you cannot negate mass or its gravitational pull. 
You might also enjoy hearing what Richard Feynman had to say when he was asked a similar question about the origin of the magnetic force:

This answer really annoyed me because it’s a lost opportunity to explain how well physicists understand the fundamental laws of nature.


Paul Titze said...

Hi Bee,

Thanks for the post. If gravity is curved space-time maybe one day we'll have a model that explains how mass/energy bends space-time with or without gravitons. Is it a reasonable question to ask: is it possible for mass/energy NOT to tell space-time how to curve?

Cheers, Paul.

CGT said...

I, for one, would like to read an exposition on anti-gravity when you have the time.
The only paper I can remember reading on 'anti-gravity' was actually about gravitational shielding. In the case I recall, it was a discussion about coupling gravity (with torsion) to extended objects (2+1 dimensional, i.e., surfaces) carrying macroscopic angular momentum 'currents' on the 2d surface that coupled to torsion. The solution of the coupled equations led to a gravitational Meisner effect at the surface of the extended object in the sense that the gravitational field did not penetrate (was 'expelled') from the surface of the object. It was never clear to me though that the interior of the closed object (as embedded in 3+1) was also free of gravitation or if the effect was confined only to the surface where the angular momentum density was present. (I cannot find the reference to this paper -- published in Physics Letters a decade or so ago. The only other references I can see to gravitational shielding are on the wikipedia page on this subject. No comment on those... I whimsically liked the above idea as it made 'more concrete' the possibility of the 'spin-dizzy' of Blish's in 'Cities in Flight" ( ) ).
Less whimsically, I'd like to hear how anti-gravity is being discussed these days, whenever that's possible... thanks.

Rob van Son (Not a physicist, just an amateur) said...

What I like most about this debate is that, like you write, both GR and QM are able to predict every measurement to the last decimal. However, whenever you try to fit them together you end up hammering a square peg into a round hole. I do understand why theorist love this.

An example is the graviton. If it exists, we must be able to detect it, mustn't we? I once read a discussion about this question and the conclusion was that a detector that had a chance of registering a graviton would be so heavy that it would collapse into a black hole and could not communicate the detection to us. Other discussion of this matter likewise resulted in conclusions that looked like there was a cosmic conspiracy to make quantum gravity undetectable.

What about the entropic/entanglement view of space-time? It does allow theorists to go a long way without bothering about experiments.

naivetheorist said...

you say that "we don’t know how to quantize gravity ". an interesting alternative approach has recently been proposed by Sean Carroll is his blog entitled "Space Emerging from Quantum Mechanics" in which he says:

"The other day I was amused to find a quote from Einstein, in 1936, about how hard it would be to quantize gravity: “like an attempt to breathe in empty space.” Eight decades later, I think we can still agree that it’s hard. So here is a possibility worth considering: rather than quantizing gravity, maybe we should try to gravitize quantum mechanics. Or, more accurately but less evocatively, “find gravity inside quantum mechanics.” Rather than starting with some essentially classical view of gravity and “quantizing” it, we might imagine starting with a quantum view of reality from the start, and find the ordinary three-dimensional space in which we live somehow emerging from quantum information."

As for your statement that "The purpose of science is to explain observations" some relevant comments by theoretical physicists/comedians are:

"All any model is supposed to do - whether it is a cellular automaton, a differential equation, or anything else - is to provide an abstract representation of effects that are important in determining the behavior of a system. And below the level of these effects there is no reason that the model should actually operate like the system itself." - Stephen Wolfram

"Everything we know is only some kind of approximation - Richard P. Feynman

"A physical theory is just a mathematical model and it is meaningless to ask whether it corresponds to reality. All that one can ask is that its predictions should be in agreement with observation." - Stephen Hawking

Reality... what a concept." - Robin Williams

eternalinquisitor said...

Seems like this Bill is one of those who are well-read in 'scientific literature' available in best-sellers and in online platforms, but who never have worked out anything in GTR or in QFT. It is amusing how people can get to make big statements now-a-days, without even going to the equivalent rigor of a grad school.

Sabine Hossenfelder said...


Yes, indeed, that's the impression I also had.

Sabine Hossenfelder said...


I don't really know what you mean with 'is it possible'. Is it possible to write down a theory in which space-time doesn't curve? Yes, all theories before Einstein were of that type. Are these theories correct? No they aren't.

eternalinquisitor said...


The problem regarding quantization starting from an essentially classical theory takes me back to Dirac's Lectures in QFT in Yeshiva, where he frankly admits that this is the only way we know to arrive at a quantum theory for a system already known classically. I think an example where people tried to arrive at the same goal completely axiomatically is perhaps the string theory. But then again, the 'Classical' element just got relocated to the level of strings. The standard mechanisms of Path-integral or defining a quantum brackets have always to be added.

As for gravitizing QM, sounds attractive. But their also, problems faced in the older alternative might re-surface in different ways, as it is essentially marrying continuum with discreteness (or converting into each-other!). One has to be prepared for long and tedious efforts.

Finally, stating quotes from people, in analytic sciences, are signs that we are desperate yet unsuccessful (it includes all of us).

Jayarava Attwood said...

Could we describe dark energy as an anti-gravity force? In the sense that it does the opposite of gravity - it makes space expand and masses accelerate away from each other.

Koenraad Van Spaendonck said...

Hello Sabine,

"According to General Relativity the root cause of gravity is that all types of energy curve space-time and all matter moves in this curved space-time."

Yes, and a great insight we've been given.
Except that the radial (time) curvature is a mathematical curve. Spacetime being hyperbolic has essentially nothing to do with a curve or curvature, but with an exponential relation between cause and effect : With every increment of distance you get closer to the earth (or more effectively to heavy bodies), you get an exponentially stronger effect, as the graph shows (gravitational time dilatation and related effects).

The apple however, falls along a straight line in reality.
So describing what happens along that straight line in therms of that exponentiality, would be more fundamental than describing it along the mathematical curve. Wouldn't it ?

Thus generating a more fundamental understanding of gravity. (never THE fundamental level, a concept which may not exist at all)

Best, Koen

akidbelle said...

Hi Sabine,

I can only agree with your post. However, the purpose of science is firstly to kill the need for ghosts (or fairy tales in your words). One must recognize that science now uses a number of those - to begin with dark matter and dark energy which, in the most terrible case, are just ad-hoc patches to the theory.

In your response to Paul, the old theories are incorrect. OK, but then why would GR be correct if it needs 95% ghostry? (quantitatively speaking).


akidbelle said...


you can account for total energy (the SM spectrum), dark energy and dark matter as the sole effect of the universe expansion. But the theoretical physicists do not care about free parameters (SM or cosmology) - even though this is actually the only remaining data to analyze where coherence can be found.


Uncle Al said...

"General Relativity that explains literally all data of gravitational effects" given the Equivalence Principle (achiral spacetime curvature). Einstein-Cartan-Kibble-Sciama gravitation allows EP = false (chiral spacetime torsion), an Eötvös experiment contrasting alpha-quartz enantiomorphs, sourcing baryogenesis and Milgrom acceleration. Look.

" The detection of gravitational waves is exciting because" it was 0.2 seconds prompt with no information, quantum, or angular momentum (singularities dancing) anomalies. Consider a (2 + epsilon) dimensional event horizon, then curvature plus surface tension (binding energy) defining the externally-viewed 2-sphere containing no internal volume, no singularity.
Fold each four-area triangle into a tetrahedral entropy unit, then
Half up, half down close-packed tessellation - soap bubble bilayer (reduced event horizon area).

емиванов фф said...

Speaking of gravity, it seems that MOND has just received a great boost:
Stacy McGaugh arXiv:1609.05917
Perhaps it's time to stop reframing problems as 'dark substances' .

Bill said...

If one peruses the innumerable arXiv articles of the past twenty years one sees attempts to unify gravity through all manner of mathematical approaches, including scalar, tensor and spinor fields, all of which seem to be a waste of time and effort. Attempts to quantize gravity have also been met with efforts to gravitize quantum mechanics, which have also been in vain. In doing so, every one of these attempts have resulted in an avalanche of abstruse mathematical arguments that only experts in various rarified, academic fields can follow. Enough!

Perhaps we need a new approach to mathematics, which until fairly recently has been sufficiently adequate to explain the physical world. Do the super-intelligent residents of the Andromeda star system (I'm speaking hypothetically) utilize Newton's infinitesimal calculus? I tend to believe not. The only alternative I can think of is that the universe is a slightly flawed computer simulation (a la the suppositions of Oxford's Nick Bostrom), whose mathematics and/or physical laws obviate the unification of quantum mechanics and gravitation at any fundamental level.

Louis Tagliaferro said...

In spite of it not being perfect or complete Newtonian mechanics gave us a deeper understanding of how gravity worked; yet over time there were signs those mechanics needed improvement for things like Mercury’s precession, or compatibility with Maxwell electromagnetism. Einstein further refined our understanding how gravity worked in a vastly different mathematical framework, yet GR also shows signs of needing improvement. Two of which are, it’s not incompatible with quantum mechanics, and (in my opinion) the Elephant in the room is dark energy.
I think it’s important for physicists working on these problems to continually be aware their own human behavior will influence them to favor working within the mathematical framework of what is already accepted. Yet it is reasonable to consider the framework needed is going to be vastly different than GR or Newton, in explaining what we’ve learned from these giants and what we still need to reconcile.
P.S. That is one of my favorite Feynman clips

Thierry Periat said...

Thank you for the debate and for centering the debate. Even an amateur like me can easily understand and appreciate the necessity to periodically re-read and recall the basics and the foundations.

This being said, the point of view which has been adopted in the article is a little bit surprising. Theoretical physics is a human effort which is not exclusively devoted to the explanation of observations.

De facto, it seems to me that it is also and always related to the discovery of correlations between diverse observations (causes and effects; logical relationships – observing the object, its behavior and the context where it happens). For example, Newton’s apple falling down from a tree has been put into relation with the see level variations and the rotation of the Moon around the Earth. Was it trivial to do it? No.

I think that the desperate efforts invested into the discovery of correlations between gravity and the others interactions belong to a similar quest. But as long as we shall not be able to find a realistic and plausible link between what appear today to be only unrelated phenomenon, the quest will stay open and the question about its legitimacy will be understood as a question of religion.

I am also surprised that you don’t consider Einstein’s efforts to rewrite his own theory in terms of vierbein (1928) as the premises of a more fundamental theory where the metric is emerging from a kind of connection (Einstein's vierbein field theory of curved space; arXiv: 1106.2037v1 [gr-qc], 10 June 2011).

Peter Byrne said...

Bill Andrews is an editor at Discover.

Tam Hunt said...

It's not at all clear to me that GR explains gravity very well. Dark matter has been proposed specifically because GR without modification did NOT explain galactic rotation anomalies, for example. Ditto with dark energy: it was proposed specifically because GR couldn't explain the apparent accelerating expansion of the universe. And ditto with inflation: it was proposed specifically because GR without inflation couldn't explain the composition of our current universe from the Big Bang. So I think it's pretty fair to say that we don't understand gravity very well, particularly when we consider that every effort over 40 years to find dark matter has failed, and we still have no idea what dark energy is and we haven't found inflatons...

Tman said...

Hi Sabine, after reading this and the comments it made me curious as to what your thoughts were in regards to the current hypotheses surrounding gravity acting on antimatter, is there one you are more attracted to?

Sabine Hossenfelder said...


What you say is incorrect. Dark matter and dark energy does NOT require a modification of general relativity. It can, and has, both easily been added as sources of gravity. The puzzle with dark matter and dark energy is that there is no microscopic description of what it's made of.

Sabine Hossenfelder said...


The brief answer is no. Antigravity is the repulsive interaction between masses (or, in general, energies) of different sign. We never see antigravitational effects because there aren't any negative masses here. Maybe that's not so surprising because if they existed, they wouldn't stay anywhere near us. Dark energy does not have a negative energy. It is extremely misleading to refer to it as antigravity, it's really a different thing. The easiest way to see this is maybe that to include antigravity into general relativity you need to modify the theory. To include dark energy you don't have to modify it. Best,


Rob van Son (Not a physicist, just an amateur) said...

With respect to dark matter. Eric Verlinde (and his brother) are developing an entropic account of gravity that surprisingly also explains dark matter. I have also seen explanations of dark energy as a direct result of entropic&holographic gravity.

Entropic gravity does still not explain the underlying microscopic mechanisms, but it does seem to be internally consistent and consistent with observations, including dark matter and energy. And by way of entanglement entropy it also seems to give a handle on a theoretical description of space itself.

Fedantos said...

Sabine (or others), please enlighten me on the perturbative approach to gravity:

Isn't it the case that the perturbative quantization of gravity is normally done in a Minkowski (flat) space-time? The resulting gravitons are then force-carriers similar to, e.g., photons, as treated in the QFT framework. Gravitons then carry momentum which modifies the momenta of involved particles such that these follow curved paths in the Minkowski space. Isn't it the case that this curved path behavior can be *alternatively* described by letting the space-time itself be curved, where the particles now follow geodesics. I guess one could start out with a space-time which is already curved, but then we would also play around with a different total set of gravitons, right?

So, there seems to be an antagonism between two perspectives on gravity here: Either, the effect of energy is to curve space, *or* particles exchange gravitons (in a way which matches energy), but not both simultaneously (for the same energy).

Please tell me if this is right, wrong, or confused, and how.

Sabine Hossenfelder said...


You can quantize the perturbations around any background. Indeed, in cosmology you don't want to quantize around Minkowski, but around FRW. Well, yes, you could describe the motion of a particle in the background by using the geodesic and ignoring the gravitons, but that would just be GR. Perturbatively quantized gravity goes beyond GR by adding quantum fluctuations to the background which are then described by graviton scattering. No, you cannot describe this in GR because it's not quantized.

No, there's no antagonism. Lots of graviton exchange can be effectively described by a curved space like lots of photon exchange can be effectively described by electromagnetic fields (in suitable approximations). Best,


Sabine Hossenfelder said...


Last time I looked enthropic gravity was very far from explaining dark matter.

George Musser said...

Very well put! It's important to avoid a mode of explanation that says everything's a mystery. I've been guilty of this sometimes, too.

Feynman's video is great, but he's incorrect about the slipperiness of ice: it's not primarily a pressure effect. The late Robert Rosenberg had a nice article about this in the December 2005 issue of Physics Today. Scott Sandford at NASA Ames had a similar discussion in "Why You Can't Have a Snowball Fight on Mars" in the January/February 1998 issue of Mercury magazine (

Unknown said...

Regarding Dark Matter, it is important to point out that by now it's not that physics is tied to the paradigm of GR, but that GR + DM *fits the data far better than any proposed other model*. This includes getting structure formation (galaxy formation & clumping etc) reasonably right (there are discrepancies involving numbers of dwarf galaxies, which may in future be measured and simulated well enough to constrain the Dark Matter model beyond 'one species of identically-massive point particles', which is the current assumption in many structure formation simulations).

Note that dark matter being 'matter' means that you expect it to go along geodesics (straight lines), hence why the evidence from the Bullet cluster is so compelling (two clusters of galaxies colliding, the dark matter / anomalous source of gravity can be imaged by gravitational lensing, and it passes through on straight lines, and isn't tied to where the visible matter is)

Regarding direct detection, in the simplest case, of Dark Matter *only* coupling to the rest of the standard model through gravitational interactions, one would not expect to see any direct detection (except in the strong gravity limit, where graviton self-interactions could couple to the dark fields, and this is well beyond the LHC energy range or the ability earth-based detectors to observe).

Check out this article for a detailed analysis of how well GR matches experiment.

As for dark energy, again, modifications to GR to explain the observed effect wind up being far more involved than just positing a field with a given potential (subject to the constraint that the field's properties prevent us from having seen it directly at the LHC, which isn't hard at all).

Also, one can excise inflation/early-universe cosmology from our knowledge of (non-quantum) gravity, by pointing out that GR + Dark Energy + Dark Matter + Standard Model predicts the universe nigh on perfectly from a given set of initial conditions (hypothesized to be the conditions after an epoch of inflation), if one is prepared to take a set of very flat initial conditions as an assumption -- gravity tells that apples fall from trees, not why apples happen to be up trees in the first place.

(Full disclosure: professional physicist, but not in GR/QFT/Quantum Gravity)

Uncle Al said...

@Unknown GR postulates the Equivalence Principle (EP). GR is prediction versus observation perfect in all venues at all scales. Physics contains no EP-falsifying experiment as no Euclidean construction obtains a 179° or a 181° triangle's internal angle summation. Chemistry has an EP-falsifying footnote; cartography dings Euclid.
János Bolyai was ridiculous but not incorrect.
Geometric Eötvös experiments are ridiculous but not incorrect. (Terrible site typeface and formatting.)
Baryonic matter explains the Tully-Fisher relationship.

akidbelle said...

Hi Sabine,

the discussion often comes back to MOND. I would like to know if there is a simple intuitive basis to this theory. It is my impression that it is mostly (if not only) a mathematical modification of Newton's equations; I mean the equations are tricked to fit the curve where Newton's theory is based on "natural" flux dispersion 1/r^2 which is intuitive, and MOND just adds new ad-hoc terms and coefficients.

Is that correct?

Uncle Al, thanks for the reference (


Sabine Hossenfelder said...


MOND is incompatible with general relativity. It was never a plausible model, if one could call it a model at all. There are now theories of modified gravity (MOG) that are compatible with gr which reproduce MOND in non-relativistic limits. Best,


Tam Hunt said...

Hi Sabine, I disagree with your interpretation of dark matter and dark energy. This is a common interpretation, of course, but I think it ignores the key point that the observations of galactic rotation and accelerating expansion could NOT be explained with GR unmodified. While the maths didn't change much with DM and DE added, they very much did change the nature of the theory, again particularly considering that we have no idea with DM and DE actually are. They're simply patches for "something weird is going on here that our current theories can't explain." So to say that GR explains gravity is in my view not at all accurate. At best, it explains the interactions of 4% of the universe.

Phillip Helbig said...

"MOND is incompatible with general relativity. It was never a plausible model, if one could call it a model at all. There are now theories of modified gravity (MOG) that are compatible with gr which reproduce MOND in non-relativistic limits."

True. However, one should distinguish between MOND phenomenology and MOND theories. By the former, I mean various phenomena (there are a dozen or so) which have no obvious explanation in the conventional paradigm, but fall naturally out of the MOND idea, which has just one free parameter (and the same value fits all phenomena). There is no debate about these observational facts, but many non-MONDians assume that they will eventually be explained within the standard framework.

Justin Khoury has done some interesting work on an idea which looks like dark matter in the regimes where this seems to work and like MOND in the regimes where this seems to work. This might be a more promising approach than trying to get MOND phenomenology (which is very simple) from very complicated processes in the standard framework, or developing some sort of relativistic version of MOND.

Uncle Al said...

@Sabine Hossenfelder With due respect, falsify an incomplete founding postulate not a derivation.

"MOND is incompatible with general relativity." ECKS gravitation allows trace non-conservation of angular momentum (Tully-Fisher via Noetherian leakage given spacetime torsion chiral anisotropy) and satisfies baryogenesis’ parity-violating Sakharov condition. Run an otherwise exhaustively nulled Eötvös experiment contrasting left-handed versus right-handed single crystal alpha-quartz test masses. All observables except geometry exactly cancel; 99.97% active mass (ignore electrons) is 420× the largest composition contrast.
Geometric Eötvös experiment.

This experiment does not exist within physics. If it arose through derivation, it could not falsify an incomplete founding postulate. Einstein’s inertial elevator has a footnote that does not naturally occur in isolation. Look.

Phillip Helbig said...

I mean the equations are tricked to fit the curve where Newton's theory is based on "natural" flux dispersion 1/r^2 which is intuitive, and MOND just adds new ad-hoc terms and coefficients.

Yes, the 1/r^2 is natural. The rest of your claim is a caricature. Fit the curve? Well, science is supposed to produce theories which explain observations, and Newtonian gravity doesn't (GR is not relevant here since the differences to Newtonian gravity are negligible in MOND phenomena). Newtonian theory famously does not fit the curve for spiral galaxies nor for about a dozen other phenomena. Yes, one can introduce dark matter to explain some of these effects, but it is here where the fitting takes place: the dark matter has to be distributed in a particular way---with no apparent explanation---in order to explain the observations. It doesn't add "terms and coefficients", but rather has one additional free parameter. Its value, like the value of most (or all, depending on the definition) constants of nature, is determined by observation. All observations of MOND phenomena lead to the same value for this parameter.

Does this mean that MOND is right? No, for several reasons. First, outside of the MOND regime, the evidence for dark matter is sound. Even MOND people accept this. Second, MOND, in its original form, violates a huge number of principles which it is better not to violate. Third, relativistic theories which reproduce MOND phenomenology are ugly and ad-hoc.

But it does mean that those who think that MOND or something like it is wrong still have to explain the MOND phenomenology. One shouldn't criticize MOND people for saying that MOND is just a heuristic ansatz and a proper theory might be more understandable and at the same time say that "introducing more complex physics, feedback, etc will enable our simulations to reproduce all MOND phenomenology". I'll believe it when I see it, and when it falls naturally out of the framework, rather than being a fudge factor. Read up on the baryonic Tully-Fisher relation for something which is very straightforward in MOND and a mess in the more conventional framework.

Phillip Helbig said...

"Speaking of gravity, it seems that MOND has just received a great boost:
Stacy McGaugh arXiv:1609.05917"

One of many MOND papers. Worth reading. Although a supporter of MOND, note that McGaugh is careful to distinguish between "MOND phenomenology" and "MOND theory". Any theory must explain the data. CDM does worse on galactic scales, MOND does worse on larger scales.

Perhaps it's time to stop reframing problems as 'dark substances' ."

Whoever coined the term "dark energy" did a great disservice. It makes cosmologists seem like clueless jerks who, as you say, call something "dark" if they don't understand it. Why not call it the cosmological constant? Yes, maybe it varies with time and/or has a different equation of state. So? Maybe the gravitational constant changes with time. The fact is, we have no evidence for the latter, and no evidence that "dark energy" is anything but the cosmological constant. We don't understand what it is? Big deal. Does anyone understand the value of the gravitational constant, i.e. why a body with a certain inertial mass produces a gravitational field of a certain strength? GR can explain the equivalence of inertial mass and passive gravitational mass, but not the equivalence of inertial mass and active gravitational mass. Why isn't G zero? No-one knows. But nevertheless gravity exists, just as the cosmological constant exists.

andrew said...

"There isn’t presently any observation which suggests that gravity itself emerges from another theory, though it is certainly a speculation that many theoretical physicists have pursued."

I'm inclined to think that this is a bit too dismissive, because it omits the biggest points that are driving all of those theoretical physicists to pursue research. In my mind, it would be appropriate to provide some of the motivation along the lines of the following:

"There are a few reasons that theoretical physics thing that there may be more to the story.

"First, almost everything about the physical world that does not involve gravity is described, at least in principle (we can't always do the mathematics necessary to solve the equations that should predict how nature acts) by a theory call the Standard Model. But, the way that the Standard Model is formulated mathematically is irreconcilable in a number of technical respects, with General Relativity, although, the vast majority of the time, only one of the two core theories of fundamental physics is necessary to explain the physical world, so this isn't a practical problem. In the circumstances of understanding the time period immediately around the Big Bang and some of the detailed properties of real world black holes, however, there is a practical need to be able to employ both theories at the same time. These theoretical inconsistency keeps hundreds of theoretical physicists up at night considering every possible way that the foundational achievements of modern physics can be reconciled.

"Second, some theorists think that astronomy observations such as those attributed to "dark matter", "dark energy", and "cosmological inflation" can be explained by modifications to General Relativity beyond that inclusion of a "cosmological constant" in the equation of General Relativity, which are sufficient to fit all "dark energy" observations to date. Cutting edge astronomy observations are helping physicists determine if there is any merit to these theories, or if more conventional explanations that do not modify General Relativity better explain what we observe.

"And, third, some efforts to unify all fundamental physics favor describing gravity as a force transmitted by a particle called a graviton rather than as a bending of space-time which is assumed to be the mechanism in General Relativity. There is good reason to believe that the conclusions of the two approaches may be virtually indistinguishable from each other with the experiments that we have done so far. Because there is one than one possible mechanism for the gravitational phenomena we observe, we need to continue to pursue all possibilities.

"Even accepting General Relativity as a perfect description of Nature, there is also lots of work for physicists to do in order to understand these deceptively simple equations better, and to measure the two physical constants of that theory, Newton's constant, G, and the cosmological constant, more accurately. While physicists have worked out exact solutions to these equations in highly simplified situations that are mathematically manageable, no one has done so for a host of more complex circumstances. And, the precision with which we know Newton's constant and the cosmological constant is much lower than the precision with which we know many other fundamental physical constants."

While it is certainly appropriate to establish the remarkable amount of knowledge that we do have, it isn't inappropriate to be equally clear about what we are still trying to find out.

Sabine Hossenfelder said...


You say you disagree but then you just confirm what I said. You can explain the gravitational effects of dark matter and dark energy just fine with GR. As I said above, what their microscopic structure is we don't know, sure. Esp for what dark energy is concerned one can question whether that's even a reasonable thing to ask.

It is wrong what you say that "the observations of galactic rotation and accelerating expansion could NOT be explained with GR unmodified." You can explain all the observations with GR. You are mistakenly thinking that because we don't know the origin of a source term for gravity, we don't understand gravity, but these are two separate things.

And, yes, sure, there are people who argue that maybe we should be modifying GR instead. That might be the solution. But at least presently this isn't necessary. Best,


Sabine Hossenfelder said...


I don't disagree with your comment, but this really wasn't the question asked. If you re-read the sentence from me which you quote you will note that it was very carefully formulated in saying that we don't have any observation. Sure we have good reasons to think that there must be a more fundamental theory, and I have written about this extensively elsewhere. But fact is there isn't presently any data to support such a theory. Best,


Koenraad Van Spaendonck said...

Concerning MOND

Erik Verlinde is currently working on scenario's that go beyond MOND.

He hasn't written a paper on it yet (or did he ?), but did a presentation at the Perimeter Institute about it in july 2015.

At 40:45 min, he argues that at larger distances, the volume law takes over from the area law etc., thus modifying gravity.

Best, Koen

Rob van Son (Not a physicist, just an amateur) said...

Here is a link to a talk by Erik Verlinde where he uses entropic gravity to create a Mond like explanation of dark matter:

He has given more talks like this. I have not seen a paper where these ideas were fully developed or worked out. There was a joined paper with his brother, but that I did not understand ;-)

akidbelle said...

Hi Sabine,

you state that there's no theoretical reason to modify GR. OK, I understand your perspective; but as far as I know, theory imposes no limit to dark matter search. Then a fruitless search of this kind can last endlessly (btw, it is similar to SUSY now "asymptotically dying").

Hence the question: assuming no dark matter exists, can this process die? How?


Sabine Hossenfelder said...


I don't know what you mean. Assuming no dark matter exists is just wrong, why would you do that?