The Michelson-Morley experiment is taught as the experiment that killed the ether. The narrative is simple: they looked for it, they did not find it, therefore it does not exist.
This is a textbook example of a logical error: absence of evidence for one model is not evidence of absence for all models.
What They Tested
Albert Michelson and Edward Morley built an optical interferometer in 1887 to detect the "ether wind" — the expected difference in the speed of light measured along versus perpendicular to Earth's orbital motion through a stationary ether.
The prediction: if Earth moves at 30 km/s through a rigid, stationary medium, then light travelling along the direction of motion would be slightly slower than light travelling perpendicular to it. The difference would produce a fringe shift of about 0.4 fringes in their apparatus.
They observed no shift above about 0.02 fringes. Null result.
What They Proved
The experiment proved that light does not travel through a rigid, stationary medium in which the Earth moves at its orbital velocity without disturbance.
This rules out exactly one model: the "ether wind" model, in which the ether is a fixed background through which the Earth plows like a ship through water.
What They Did Not Prove
The experiment says nothing about:
1. A Fully Dragged Ether
If the ether is dragged along with the Earth (as Stokes proposed in 1845), there is no ether wind and no expected fringe shift. The null result is predicted.
2. A Lorentz-Covariant Ether
Lorentz himself, in 1904, showed that if the ether contracts lengths and dilates time in the moving frame — exactly as special relativity predicts — then the null result follows mathematically. Lorentz Ether Theory (LET) makes identical predictions to special relativity. This is Section 1 of the monograph: the empirical content of LET, Poincare's group-theoretic formulation, and Einstein's SR are provably identical.
3. A Superfluid Ether
A superfluid has zero viscosity. An object moving through it at velocities below the critical velocity experiences no drag whatsoever — this is the Landau criterion. A superfluid ether would produce exactly zero ether wind for any subluminal motion. The Michelson-Morley null result is not merely consistent with a superfluid ether; it is predicted by it.
4. Einstein's GR Ether
Einstein's own "ether of general relativity" (1920) is a spacetime metric with physical properties, not a substance with a state of motion. You cannot measure your velocity relative to a metric tensor. The null result is automatic.
The Logical Error
The standard narrative commits the fallacy of affirming the consequent:
- If there is a rigid stationary ether, then Michelson-Morley will find a fringe shift.
- Michelson-Morley found no fringe shift.
- Therefore, there is no ether. (Invalid)
The valid conclusion is only: therefore, there is no rigid stationary ether. The invalidity becomes obvious when you note that the same null result is predicted by at least four other ether models, plus by the no-ether hypothesis. The experiment cannot distinguish between them.
After Michelson-Morley
The history did not stop in 1887. Dayton Miller conducted extensive repetitions between 1902 and 1926, reporting small positive signals that he attributed to ether drift. His results remain contested but were never definitively explained. Modern laser interferometers (Brillet & Hall 1979, Muller et al. 2003) have pushed the null to extraordinary precision (), but all test for the same thing: anisotropy of the speed of light, which a Lorentz-covariant or superfluid ether does not predict.
The Ether That Michelson-Morley Could Not Detect
The ether physics monograph proposes a specific model: a superfluid Bose-Einstein condensate with zero viscosity, whose acoustic metric is the Painleve-Gullstrand form of the Schwarzschild solution (Section 3).
This ether has no preferred rest frame detectable by electromagnetic experiments, because its field equation is the Einstein equation (Section 3), which is manifestly Lorentz-covariant. The PPN parameters (Section 3) — no preferred-frame effects in gravitational experiments, despite the ether having a physical existence.
Michelson and Morley's experiment was a triumph of precision measurement. But the conclusion drawn from it — that space has no physical properties — was an overreach. It ruled out one model and left infinitely many others standing.
The question was never whether the ether exists. The question is what kind of ether it is.