This article is a plain-language companion to Paper 11 of the Windstorm Institute. The technical version, with the math and the references, is at github.com/Windstorm-Institute/gravitational-entropy-escrow.

Pick up an apple. Drop it. It falls. Of course it falls. Newton sorted that one out a few centuries ago.

Now ask: why does it fall?

You can answer with equations — force equals mass times the second derivative of position, the Earth has a big M, separation matters as one over r squared, and there you go. Or you can answer with curvature — spacetime is bent around the Earth and the apple is just rolling down the hill. Both answers are correct. Neither answer tells you why it pulls and never pushes. Neither tells you why you can’t shield gravity the way you can shield a magnet with a mu-metal cup. Neither explains the very strange fact that black holes are simultaneously the gravitational pinnacle of the universe and the entropic pinnacle of the universe — that the place with the strongest gravity also happens to be the place with the most disorder packed into the least volume.

This paper proposes that all of these mysteries have the same answer. Gravity, it suggests, isn’t a force. It’s the universe’s collection agency for an entropy debt — and bound systems are simply piles of entropy that the cosmos is patiently trying to balance.

The bookkeeping idea

Imagine the universe runs a giant ledger. Every particle, every star, every photon — all of them contribute to a single number, the entropy of everything. The second law of thermodynamics says this number can only ever go up. It is the most reliable rule in physics. You can break almost any other law if you try hard enough; you cannot break this one.

Now consider what happens when two masses get bound together — the Earth and the Moon, say, or two stars in a binary, or you and the planet under your feet. They form a system that is, in some sense, more ordered than two unbound masses drifting freely. Less entropy, not more. So where did the missing entropy go?

The proposal of Paper 11 is that the missing entropy was placed into escrow. It hasn’t vanished. The universe is just holding it off the books, the way a real estate escrow account holds money in trust between parties. The bound system “owes” that entropy. And the universe never forgets a debt.

Gravity is what it looks like when the universe collects on an entropy escrow.

That single picture — once you let yourself take it seriously — turns out to explain a startling number of things gravity does that nobody really had a good story for.

Why gravity always pulls and never pushes

Every other force in nature has an opposite. Push a magnet against another magnet the right way and they repel. Charges come in plus and minus. The strong nuclear force binds quarks but doesn’t care about the difference between a proton and an antiproton.

Gravity is alone in not having a partner. Mass attracts mass. There is no anti-mass. There is no “gravity shield” you can put around a building to make it weigh less. People have searched for one for centuries. They have always come back empty-handed.

The escrow picture explains why this had to be the case. Gravity is the second law dressed up in different clothes — and the second law has no opposite either. Entropy goes up. It does not go down. There is no “anti-entropy” that would let a force point the other way. There is no operation that lets you shield a system from the universe’s bookkeeping. If gravity really is the universe collecting on an entropy debt, then of course it only pulls. Of course you can’t hide from it. Those aren’t accidents of how gravity happens to work; they are inevitable consequences of what it actually is.

Why a falling elevator feels like nothing

Here is a fact about gravity that is so familiar people forget how strange it is. Imagine you’re in an elevator and the cable snaps. (You’re fine; this is a thought experiment.) For the few seconds before things end badly, you and the elevator are in free fall. And during those seconds, you don’t feel any gravity at all. You float. A pen released from your hand floats next to you. If the windows were painted black, you wouldn’t be able to tell whether you were in deep space or falling toward Earth.

This is the equivalence principle. Einstein built general relativity on it. But take a second to actually appreciate how weird it is. The Earth, all six trillion trillion kilograms of it, is pulling you with enormous force. And you can’t feel it. You feel weightless.

Even stranger: the Earth’s gravity isn’t the only thing pulling on you. The Sun pulls on you. The Galaxy pulls on you. Sagittarius A*, the four-million-solar-mass black hole at the center of our Galaxy, pulls on you. We can compute the force. None of these forces are trivial. And you can’t feel any of them either. The Earth-Moon system tumbles around the Sun, falling endlessly toward Andromeda, and from inside it everything just feels like a closed, peaceful system. Why?

The escrow picture answers cleanly. The thing you feel as gravitational force is a difference in the entropy debt across your body. If the debt is uniform from your head to your toes — if the universe’s books look the same on top of you and underneath you — there’s no difference, and you feel nothing. When you’re in free fall, the elevator and your body and the pen all carry the same entropy debt; the books are perfectly balanced inside the falling frame. The pull from the distant black hole is uniform across your body too — you fall toward Sagittarius A* with exactly the same acceleration as everything around you. So you feel nothing.

What you do feel are tidal forces — the bits of the entropy debt that aren’t uniform. The Moon pulls slightly harder on the side of Earth facing it than the side facing away. That difference is what makes the oceans bulge. If you fell into a black hole, the difference between the pull on your feet and the pull on your head would eventually become enormous — that’s spaghettification. Tidal forces are the entropy debt you can’t balance away by changing reference frames. They are what’s left over after you fall freely.

The escrow picture predicts this without needing general relativity to motivate it. Tidal forces are the leftovers in the entropy ledger.

Why black holes are entropy factories

In the 1970s, Jacob Bekenstein and Stephen Hawking discovered something that physicists are still digesting. Black holes have entropy — not a little, but the maximum amount of entropy you can pack into a region of that size. And the entropy is proportional to the surface area of the event horizon, not the volume inside. A black hole the size of a basketball has more entropy than every grain of sand on every beach on Earth, multiplied by every star in the observable universe, by a factor of trillions of trillions.

This is one of the strangest results in modern physics. It says the universe stores its information on surfaces, not in volumes. And it says that a black hole, the gravitational pinnacle of the cosmos, is also the entropic pinnacle. Maximum gravity coincides with maximum entropy. Why should two completely different things hit their maximum in the same place?

The escrow picture says they aren’t two different things at all. As a star collapses, the entropy held in escrow by its self-binding grows and grows. At some point the debt becomes so large that no amount of external entropy production could ever pay it off. The universe’s only remaining option is to write the debt directly onto the horizon, in plain view, where it becomes the area entropy Bekenstein and Hawking wrote down. The black hole is the moment the universe finally stops pretending and just records the bill in public.

Plug the escrow formula into the conditions at a black hole horizon and you get back the Bekenstein-Hawking entropy exactly — the same numerical coefficients, the same scaling with mass squared, the same appearance of Planck’s constant. You don’t fit anything. It just falls out.

Why galaxies don’t obey Newton

Now for the part that’s genuinely contentious. For 50 years, astronomers have known that galaxies rotate too fast. The outer stars in a spiral galaxy whip around the center at speeds that, by Newton’s law, would have flung them into intergalactic space billions of years ago. The standard explanation is dark matter — a vast invisible halo of unknown stuff that provides the extra gravity to hold galaxies together.

An alternative explanation, going back to Mordecai Milgrom in 1983, is that gravity itself behaves differently when it gets weak enough. Below an acceleration of about 10⁻¹⁰ meters per second per second — ten billion times weaker than Earth gravity — the law subtly changes shape, in a way that exactly matches what we see in galaxy rotation curves and a relationship called the Tully-Fisher law. This is MOND (Modified Newtonian Dynamics), and it works embarrassingly well in galaxies. The reason it’s contentious is that nobody has had a good physical reason for why nature would have a special acceleration scale at exactly that value.

The escrow picture supplies one. Empty space, even completely empty space, has a tiny irreducible “chill” to it — a temperature set by the cosmological constant Λ, the same dark energy that’s accelerating the expansion of the universe. This temperature is fantastically small, around 10⁻³⁰ degrees Kelvin, but it is not zero, and it is the same everywhere and at every moment in cosmic time.

It also acts as a thermodynamic floor — a Carnot floor, if you remember thermodynamics class. The universe can’t balance entropy debts more efficiently than this floor allows. When local gravity gets weak enough that its “temperature” falls toward this cosmic floor, the bookkeeping changes. The escrow can’t be redeemed at the same rate. Plug this into the math and you get back exactly the deep-MOND behavior that astronomers see in galaxy rotation curves — including the special acceleration scale, which falls out as a number you can compute from Λ alone.

The high-redshift test

Here’s where the framework actually puts itself on the line. If the special MOND acceleration is set by Λ — which is constant across cosmic time — then it shouldn’t change as you look at galaxies at higher redshifts, billions of years in the past. If it’s set by something that was changing back then, like the matter density of the universe (which was much higher) or the Hubble expansion rate (which was much faster), then it should look different in old galaxies.

In 2017, an astronomer named Mordehai Milgrom looked at six big disk galaxies that the Genzel team had observed at extreme distances — we see them as they were 8 to 10 billion years ago, when the universe was a third of its current age. He found that the “acceleration scale times changes with cosmic time” version of MOND was, in his words, “all but excluded” by the data. Constant works. Evolving doesn’t.

For Paper 11 we ran a slightly bigger version of that test. Five different prescriptions for how the MOND scale might or might not evolve with redshift were checked against the same six high-redshift galaxies, with two different mathematical interpolation forms, for a total of ten checks. The three constant prescriptions all pass cleanly. The two evolving prescriptions both fail by margins three to ten times larger than the acceptance threshold. The data prefer a constant cosmic chill — exactly what the escrow picture predicts.

We also reanalyzed the standard SPARC database of 175 nearby galaxies and got an empirical MOND scale of 1.24 × 10⁻¹⁰ m/s², matching Milgrom’s canonical value to within four percent.

The honest part

This is not a finished theory of gravity. The picture is a way of looking, not a derivation. The math underneath comes from work by Bekenstein, Hawking, Unruh, Jacobson, Padmanabhan, Verlinde, and van Putten, going back fifty years. The contribution of this paper is the framing — the claim that all these results are facets of one phenomenon — not the equations themselves.

And the picture has a real problem we can’t make go away. Galaxy clusters. The framework predicts that the inner cores of clusters — where gravity is strongest — should look perfectly Newtonian, with no MOND-style enhancement. But observations show the largest gravitational anomalies appear precisely in those cores, with mass discrepancies a factor of ten in the inner few hundred kiloparsecs. That’s the wrong way around for the framework as it stands. Either there’s an extra matter ingredient in cluster cores (sterile neutrinos are a candidate), or there’s something missing from the picture. We flag this honestly rather than try to dress it up. It is a real difficulty.

The framework also doesn’t derive the precise value of the MOND scale from first principles. It gets the right answer up to an order-one fudge factor (about 1.39); calculating that factor from the underlying physics is the central remaining technical task.

The picture, in one sentence

If the picture is right, gravity is the bookkeeping department of the second law.

And the rest follows. The universal attractiveness of gravity? That’s the second law having no opposite. The impossibility of shielding gravity? That’s the impossibility of shielding the second law. The slowing of clocks deep in gravity wells? That’s the entropy debt running thicker. The area law of black hole entropy? That’s the universe finally cashing out a debt it could no longer hide. The strange acceleration scale where galaxies stop obeying Newton? That’s the cosmic chill imposing a floor on how fast the books can balance.

None of these dots is new. The contribution of Paper 11 is the dotted line connecting them — the suggestion that they are obviously related rather than coincidentally related. Whether that turns out to be right or wrong, it’s the kind of suggestion that should be argued out in plain view, with the limitations stated honestly and the open problems flagged clearly. That’s what the paper tries to do. That’s what this article tries to do too.

Where this fits in the Institute’s work

Paper 11 is the second paper in the Institute’s second research track — Entropic Bounds in Analog Systems. The first track (Papers 1–9) was about the throughput basin in serial decoders, from ribosomes to AI transformers. Different topic, same lens: the hidden mathematical constraints that govern what physical systems can do.

Paper 10, the first Track 2 paper, was a narrow falsifiable laboratory prediction — a 17% efficiency suppression in cold-atom analog gravity that an experimental group could plausibly test in the next several years. Paper 11 is the broader, more philosophical sibling: an interpretive synthesis covering universal attraction, the equivalence principle, black holes, and the MOND scale. Together they map out what Track 2 is trying to do — ask, on every scale where gravity-adjacent physics meets thermodynamics, what entropy is really doing under the hood.

Gravitational Entropy Escrow is Paper 11 of the Windstorm Institute — the second paper in the Entropic Bounds in Analog Systems track.
Zenodo: 10.5281/zenodo.20032023 · Code & data: github.com/Windstorm-Institute/gravitational-entropy-escrow
Download the full paper (PDF) · Markdown source