Space Is the Void
Five different things get called space. Two of them have no properties at all, and neither can be curved, because there is nothing there to act on.
Proven in this memo
- Coordinates carry no physical content. A description replaceable without altering any measurement is not the thing described — a standard consequence of general covariance.
- Chromatic lensing is unmeasurable at optical wavelengths for every permitted lattice spacing, so the earlier prediction of EHT-detectable anomalies is withdrawn.
The central claim — that the measurements read a substance rather than a geometry — is interpretation, not proof. Section 10 states why: for light propagation the two accounts are mathematically interchangeable, and neither lensing nor the Shapiro delay nor GPS can separate them.
Draft. A working draft, published for scrutiny rather than as a settled result. The ledger at the foot records what is derived, what is measured, what is assumed and what remains unanswered. Where a figure is carried from a secondary source and not verified against the original, the sources section says so.
Clock behaviour near mass is a separate question and is handled in T.2. Its absence here is deliberate.
In plain terms
Two different things get called space, and running them together causes most of the confusion.
The first is the container — the bare fact that there is somewhere for things to be. That has no properties at all. It cannot be bent, stretched or curved, because there is nothing there to do anything to.
The second is what fills the container. That is a substance. It has density and stiffness, it can be squeezed and stretched, and it carries light across a room.
When physics says space is curved it is describing something real and measured. Light does bend near the Sun, clocks do run slow near mass, and GPS would drift eleven kilometres a day if it ignored either. What is proposed here is that those measurements read the substance, not the container. A mass makes a dip in the substance around it, like a ball on a stretched mat. The container is still nothing.
1. The numbers
- Five distinct things travel under the word “space”. Two of them have no properties at all.
- A 158 kg lead sphere distorts its surroundings by one part in 1024 — a radius of curvature of about two astronomical units.
- Two plates 10 nm apart in vacuum attract with 1.3 × 105 Pa, more than atmospheric pressure, in what is supposed to be nothing.
- Published determinations of G disagree by 552 ppm against quoted errors of 12 to 150 ppm.
- The Aether density excess at the Sun’s surface: 8.49 ppm.
2. The question, stated precisely
The claim under examination is not whether gravity bends light. It does, and the amount is measured. The claim is about what is doing the bending.
For space itself to curve, space itself would have to have a property that can take different values from place to place. A thing with no properties cannot have one of them vary. So the question reduces to: is space a something or a nothing? If it is a something, it can curve and the question is what it is made of. If it is a nothing, the curvature belongs to whatever fills it.
This memo argues the second. It does not dispute a single measurement.
3. Breaking space down
Five distinct things travel under the one word. Separating them dissolves most of the argument, because the disputes are usually between people meaning different ones.
| Component | What it is | Properties? | Can be curved? |
|---|---|---|---|
| The void | The bare container | None | No — nothing to act on |
| The lattice | What fills it. The Aether | Density, stiffness, state | Compressed, stretched, sheared |
| The coordinate grid | Our labelling of positions | None | Arbitrarily, and it means nothing |
| The metric | Measured intervals between events | Yes, and measured | Yes — this is what is measured |
| Position | Where a thing is | Relational | Not applicable |
The properties column is the test. Two of the five have none, and neither of those two can be bent. The other three can, and are.
Where the confusion sits
When a physicist says spacetime is curved, the thing described is the fourth row — the metric, meaning measured relationships between events. That is real and measured to many decimal places, and nothing here disputes it.
When a popular account draws a rubber sheet, the thing pictured is the first row: the container, bending. That is not what the theory says, and it is what Tesla objected to.
This memo agrees with the physicist about the metric and with Tesla about the container. Its proposal concerns the second row.
4. Tesla’s objection
The distinction in section 3 is not new. Tesla made it, in almost the same order, in a written statement of 15 April 1932 now held in the Nikola Tesla Papers at Columbia University.[1]
He held that space cannot be curved, for the simple reason that it can have no properties
. He then
drew the same split: properties belong to the matter filling space, not to space itself. And he concluded
that to say large bodies curve space is to say that something acts upon nothing — a view he declined
to accept.
He repeated it on 10 July 1937, announcing his dynamic theory of gravity and arguing that only the
existence of a field of force can account for
the motions, which removes any need for curvature. At
eighty-two he described Einstein’s work as mathematical dress that dazzles and blinds readers to the
errors beneath.
Three cautions. His dynamic theory of gravity was announced repeatedly from 1925 and never published. No manuscript has surfaced. He can be cited for the objection; there is no alternative theory of his to cite.
The objection has a standard answer: modern relativists hold that the geometry describes the gravitational field rather than asserting that literal nothing bends — which is the point proved in section 5. On that reading Tesla was arguing against the popular imagery rather than the theory.
A 2025 paper on the geometric meaning of general relativity cites this passage directly, naming Tesla as a distinguished non-specialist misled by taking the picture literally. That citation establishes the quotation is genuine and reads it unfavourably. Both halves are carried here.
5. Coordinates carry no physical content
ProofThe grid is not the thing
General relativity is generally covariant: its equations take the same form in every coordinate system, and every physical prediction is invariant under a change of coordinates.
A quantity that can be altered arbitrarily without altering any measurable prediction cannot itself be a measurable feature of the world. Coordinates are exactly such a quantity.
Therefore whatever is physically real about a gravitational field is not the coordinate grid. It is whatever the grid describes. This is not an argument against relativity — it is a standard feature of it, and it is what leaves the question of the underlying cause open.
Chain: general covariance → coordinate freedom → coordinates are not observables. Standard result, not original here.
6. The lead and the steel
Cavendish, 1798. Two lead spheres of 158 kg, two small balls on a suspended rod, a torsion fibre. The small balls swing toward the large ones. Anyone can watch it happen.
| Quantity | Value |
|---|---|
| Force on each small ball | 1.46 × 10−7 N |
| Equivalent weight | 15 micrograms — a grain of fine sand |
| Metric distortion at 23 cm | 1.0 × 10−24 |
| Radius of curvature implied | 3.2 × 1011 m = 2.2 AU |
A lead ball on a bench, curving space with a radius of curvature larger than the orbit of Mars.
Why the smallness is not the argument
It is tempting to say the effect is far too small for anything so dramatic to be happening. That is not a refutation. General relativity predicts exactly this force and exactly this curvature, and both are measured. Predicting a tiny effect correctly is a success, not a weakness.
Tesla’s version survives and does not depend on size at all: what is the lead bending? If space has no properties, there is nothing there to bend, whether the amount is large or small.
And the experiment cannot settle the question. It measures a force between two masses. Newton, Einstein and a medium account all predict the same force, to far beyond any achievable precision. It is a Newtonian measurement and it distinguishes nothing.
It is nonetheless where G comes from, and after two centuries the determinations still disagree. Seven published values since 1982 span 552 ppm, against quoted uncertainties between 12 and 150 ppm. They disagree by more than they claim to, and CODATA inflates the published uncertainty to cover it.[2]
7. The equations, read as a medium
The translation is exact rather than analogical, and it has been in the literature for a century.
The weak-field metric is a refractive index. The coefficient 1 + 2GM/rc² is the factor by which the coordinate speed of light differs from c. Written as an index and squared, it gives the local density:
The metric potential and the Aether density are the same quantity written twice. At the Sun’s surface the excess is 8.49 parts per million; at Earth’s, 2.8 parts per billion.
The field equation is a statement about compression. Substituting Φ = −(c²/2)(δρ/ρ0) into Poisson’s equation gives:
with a coupling of 1.87 × 10−26 m/kg. Read plainly: matter is a source of compression in the lattice.
And Schwarzschild is already a flowing medium. In Painlevé–Gullstrand coordinates, published in 1921, the Schwarzschild solution is flat space with an inflow at exactly the escape velocity — 11.2 km/s at Earth’s surface, 618 km/s at the Sun’s, reaching c at the horizon. That is not an alternative to Schwarzschild. It is Schwarzschild.
What this means
None of this is a rival set of equations. It is the same equations, read as describing a substance rather than a geometry. Every number is identical because the mathematics is identical.
Which is the point: if the description can be read either way without changing a single prediction, then the geometry was never the content. It was the language.
8. Where G came from
Newton never wrote a gravitational constant. He worked in ratios and never needed one. Cavendish in 1798 published a paper titled Experiments to Determine the Density of the Earth; he obtained 5.48 g/cm³ and did not compute G. The constant was first written explicitly in the 1870s and extracted from Cavendish’s data.
Einstein, in 1915, fixed the coefficient 8πG/c⁴ by requiring his field equations to reduce to Newton’s in the weak field. He did not discover, derive or measure G. He inherited it and inserted it as the number that makes the limit come out.
It remains the worst-measured constant in physics, with a relative uncertainty of 2.2 × 10−5 — about a hundred thousand times worse than the electron mass.
9. The mechanism is open
Mass produces a drop in the Aether’s potential. The drop falls as 1/r and the field as 1/r², and that reproduces every gravitational measurement ever made. What is not established is how mass produces the drop.
Three candidates sit elsewhere in this series, and none has been written as a source term in equation (2): matter draws Aether inward, as the neutron’s shell gathers itself from the surrounding medium; matter holds Aether, since a material contains n² times the void density permanently; and matter excludes free lattice at short range through the repulsion in the force law.
Why leaving it open is the right position
Newton gave the inverse square law and refused to explain its cause — hypotheses non fingo, I frame no hypotheses. Gravity was a described effect with no mechanism for two centuries, and it was the most successful theory in physics throughout.
General relativity describes the same effect more accurately and still does not explain why G has the value it has. This framework is at the same stage on this one point, and says so.
10. Why light cannot settle it
Gordon showed in 1923 that a gravitational field is exactly equivalent to a refractive medium for the propagation of light.[3] Plebanski gave the full constitutive form in 1960. The two descriptions are mathematically interchangeable, so lensing, the Shapiro delay and GPS timing give identical predictions under either.
That cuts both ways. The medium account cannot be refuted by pointing at lensing, and it cannot be established by pointing at it either. Anyone claiming that gravitational lensing proves space is a substance has not understood the equivalence.
11. Where they do diverge
General relativity’s lensing is exactly achromatic: photons follow null geodesics regardless of energy, so every wavelength bends by the same amount. A discrete lattice is dispersive. Any measured chromatic lensing would therefore falsify pure curvature.
The size of the effect decides whether the test is available:
| Lattice spacing | at 500 nm | at 1 GeV | at 1 PeV |
|---|---|---|---|
| 0.95 fm | 5.9 × 10−18 | 0.96 | blocked |
| 6.2 × 10−22 m | 2.5 × 10−30 | 4.1 × 10−13 | 0.41 |
| Planck length | 1.7 × 10−57 | 2.8 × 10−40 | 2.8 × 10−28 |
At optical wavelengths it is unmeasurable by any instrument that will ever exist. At PeV energies it is of order one — which is precisely why the observed PeV photons already bound the lattice spacing. The test has been run. It returned a bound, not a detection, and it lives in gamma-ray astronomy rather than in lensing images.
A prediction is therefore withdrawn. The earlier Aether white paper predicted wavelength-dependent lensing anomalies detectable by upgrades to the Event Horizon Telescope. At optical wavelengths the effect is 5.9 × 10−18 with a 1 fm lattice, and the 1 fm spacing is itself excluded by the PeV photons. That prediction does not stand.
12. What would falsify this
- Finding the rest frame would confirm the medium; not finding it is the standing problem. Rotating optical resonators bound Earth’s motion relative to any such frame below about 0.95 m/s, against an orbital speed of 29,800 m/s. This is the most serious unanswered objection in the series — memo 5.7.
- A demonstration that the vacuum has no structure would refute the content half of the claim. The Casimir force runs the other way: two plates 10 nm apart attract with more than atmospheric pressure, measured since 1997.
- A measurement distinguishing curvature from a graded medium for light would settle it. By Gordon’s equivalence, none exists for light propagation alone. Dispersion at gamma-ray energies is the only route, and it has so far given a bound.
- A mechanism producing 1/r² from the medium, with the right coefficient, would move section 9 from open to derived. Failure to find one indefinitely would leave this framework describing gravity without explaining it — the position general relativity is already in.
13. Sources
Reference codes read source.work.passage and resolve on the Master Source Register, which carries every source used across this series.
- N. Tesla. Statement relating to force and matter, to Einstein’s theories, and Tesla’s own theory of gravitation, 15 April 1932, Nikola Tesla Papers, Rare Book and Manuscript Library, Columbia University; and the Dynamic Theory of Gravity statement of 10 July 1937. The objection in section 4. 93
- E. Tiesinga, P. J. Mohr, D. B. Newell & B. N. Taylor. CODATA recommended values of the fundamental physical constants: 2018, Reviews of Modern Physics 93 (2021) 025010. The value and uncertainty of G, and the inflation applied to cover the spread between experiments. 70
- W. Gordon. Zur Lichtfortpflanzung nach der Relativitätstheorie, Annalen der Physik 377 (1923) 421–456. The exact equivalence of a gravitational field and a refractive medium for light. Volume and pages carried from secondary sources; not verified.
- J. Plebanski. Electromagnetic waves in gravitational fields, Physical Review 118 (1960) 1396. Carried as reported; not verified.
- P. Painlevé (1921) and A. Gullstrand (1922). The coordinates in which Schwarzschild appears as a flowing medium. Carried as reported; not verified.
- H. Cavendish. Experiments to Determine the Density of the Earth, Philosophical Transactions of the Royal Society 88 (1798) 469–526.
- S. G. Lamoreaux. Demonstration of the Casimir Force in the 0.6 to 6 µm Range, Physical Review Letters 78 (1997) 5. Carried as reported; not verified.
- Bureau International des Poids et Mesures. The International System of Units (SI), 9th edition (2019). 72
- H. G. Alexander (ed.). The Leibniz–Clarke Correspondence (1715–16).
- S. Herrmann et al. Rotating optical cavity experiment testing Lorentz invariance at the 10−17 level, Physical Review D 80 (2009) 105011. Carried as reported; not verified.
Verification register: references 3, 4, 5, 7 and 10 are carried as reported from secondary sources and have not been checked against the originals. The Cavendish figures in section 6 are computed from his published apparatus dimensions rather than quoted.
Ledger — memo T.1
- Derived
- That coordinates carry no physical content, from general covariance — standard, restated because the argument depends on it
- The Cavendish force, metric distortion and radius of curvature, computed from published apparatus dimensions
- That the weak-field metric, Poisson’s equation and the Schwarzschild solution all translate exactly into statements about a medium
- That chromatic lensing is unmeasurable at optical wavelengths for any permitted lattice spacing, and that the earlier prediction to the contrary must be withdrawn
- Measured
- Light deflection at the solar limb, 1.75 arcseconds; Shapiro delay to 1 part in 105
- Casimir force, 1.3 × 105 Pa at 10 nm
- G, with a relative uncertainty of 2.2 × 10−5 and a 552 ppm spread between determinations
- Bound on motion relative to a preferred frame, ~0.95 m/s
- PeV photons from galactic sources, bounding the lattice spacing
- Assumed
- That the container and the content are genuinely distinct. This is the memo’s central claim and it is a commitment about what exists rather than a measured result. Standard physics does not draw the line here and nothing in the measurements forces it.
- That the measured effects read a substance rather than a geometry. By Gordon’s equivalence the two are indistinguishable for light, so this is an interpretation. Section 10 says so directly.
- That mass produces a potential drop falling as 1/r. Taken from observation, as Newton took the inverse square law. The mechanism is not supplied (section 9).
- Open
- How mass produces the drop. Three candidates are named in section 9 and none is derived. Until one is, this framework describes gravity without explaining it
- The value of G. Not derived here, and not derived by general relativity either
- The rest frame. A real medium has one; it has not been found, to 0.95 m/s. Memo 5.7
- Nothing in this memo distinguishes the medium account from the geometric one. If distinguishing evidence exists it is elsewhere in the series
- Prior art
- N. Tesla, 1932–38 — the container/content distinction is his, made in almost the same order. Cited for the objection only; his dynamic theory of gravity was never published
- Newton, 1687 — absolute space, and hypotheses non fingo: the inverse square law stated without a mechanism, which is the position section 9 adopts
- Leibniz, 1715–16 — space as nothing but relations between bodies
- Gordon, 1923, and Plebanski, 1960 — the medium formulation of general relativity. Reading curvature as a refractive medium is theirs and is not novel here
- Painlevé, 1921, and Gullstrand, 1922 — Schwarzschild as a flowing medium