Empirical Earth · does the Earth turn?

How do we know the Earth turns? Instruments that share no physics keep returning the same number.

In 1851 Foucault hung a weight on a wire in Paris and watched the plane of its swing creep round. In 1913 Compton flipped a sealed ring of water and measured the drift. In 1925 Michelson, Gale and Pearson piped light around a mile-long rectangle of buried pipe in Illinois. Today a laser sits under a kilometer of rock at Gran Sasso, and a Vienna group has done the same measurement with entangled photons. A pendulum, a tube of water, a beam of light and a pair of correlated photons have no physics in common, and not one of them needs a view of the sky. They agree on how fast the ground is turning, and on which way the axis points.

What this page does not do. None of it proves the Earth is round. Rotation and shape are separate questions, and an instrument that measures a spin cannot tell you the shape of the thing spinning. What the measurements below settle is narrower and harder to argue with: the ground under the instrument turns once a sidereal day, and the angle of that turn changes with latitude in the way a sphere requires and a flat disc does not.

The number every one of them returns

Rotation rate7.292115 × 10⁻⁵ rad/s, about 15 degrees an hour
One turn23 h 56 m 04 s, the sidereal day, not the 24 h solar day
Latitude dependencethe measured component scales as sin(latitude): full at the poles, zero at the equator, reversed in the south
On a flat discevery horizontal instrument would read the same value everywhere, with no latitude term and no sign change
Four kinds of instrument, then the consequences and the objections
  1. Mechanical
  2. Optical and quantum
  3. Astronomical
  4. Relativistic
  5. Consequences you can see
  6. The objections that backfire

01Mechanical

1851 onward

The oldest measurements, and the ones you can watch with your eyes. Nothing here needs a laser, a satellite or anybody else’s data. A weight on a wire, a ring of water, a spinning wheel.

44
“If Earth spun we’d feel it or be flung off; nothing actually detects rotation.”
Foucault’s pendulum precesses at 15°/hr × sin(latitude) — a direct, repeatable rotation signature.
Foucault's pendulum & the Coriolis effect →
46
“Nobody has ever measured the Earth spinning — it’s just an assumption.”
It’s been measured many ways, from Compton’s flipped water ring (1913) to entangled photons (2024).
Measuring Earth's spin — from a flipped tube of water to entangled photons →
47
“A gyroscope shows no rotation — its ‘drift’ is just error, proving the Earth is still.”
A spinning gyroscope holds its axis fixed in space, so a turning Earth makes it appear to drift, faster or slower depending on your latitude. That drift is not error. It is the rotation itself, and a gyrocompass finds true north by measuring it.
The gyroscope — rigidity, precession, and the spin it can’t hide →

02Optical and quantum

1925 onward

Light sent both ways around a closed loop comes back out of step if the loop is turning. That one fact runs from the Michelson-Gale-Pearson survey of 1925, through the ring lasers buried at Gran Sasso, to the entangled photons measured in Vienna in 2024, and it works in a sealed room with no view of the sky.

45
“Michelson-Morley’s null result proves a stationary Earth.”
It found no aether wind (which relativity expects), not a still Earth; Michelson-Gale-Pearson (1925) measured the rotation itself.
Michelson — aether, light-speed & Earth's rotation →
48
“Ring lasers see some rotation and calling it the Earth’s is just an assumption.”
The signal scales with latitude, reverses when the instrument is inverted, and three rings at right angles recover one common axis. A flat disc would read the same everywhere.
The rotating optical resonator — a sealed box that finds the spin axis →

03Astronomical

the sky keeps the same time

The sky is not the strongest evidence, because a determined objector can always reinterpret what the sky is doing. It is included because it agrees, to the second, with instruments that never look at it.

85
“Star trails are just stars wheeling over a flat disc.”
Trails center on two opposite celestial poles and run dead straight at the equator — only a sphere does that.
Star trails →
76
“Polaris never moves — proof of a fixed Earth under a dome.”
Polaris circles ~1.3° nightly and precession changes the pole star over ~25,800 years; near-fixedness is what a spin axis pointing near it predicts.
Polaris is not nailed in place →
52
“If Earth were spinning and hurtling through space we’d feel it — so it must be motionless.”
You feel acceleration, not velocity. Earth’s motions are steady or free-fall, so they can’t be felt — only the spin leaves a trace (~0.3% lighter at the equator), and every motion is independently measured.
The Earth’s many motions — and why we feel none of them →

04Relativistic

the spin drags spacetime

A rotating mass twists the spacetime around it. That effect has been measured by bouncing lasers off satellites, and the clocks in orbit have to be corrected for it or navigation fails.

116
“Satellites aren’t real, and you can’t prove Earth’s shape or spin from the ground.”
Observatories laser-range passive reflector spheres to the millimeter — measuring Earth’s oblateness, spin, and even frame-dragging.
LAGEOS — bouncing lasers off a satellite to weigh the spinning Earth →
115
“GPS is just triangulation; relativity is irrelevant.”
Satellite clocks need a +38 µs/day relativistic correction or fixes drift ~10 km/day — computed for clocks orbiting a round, rotating Earth.
GPS only works with relativity →

05Consequences you can see

what a spinning planet does to itself

A planet that turns is not the same shape as one that does not, and the air, water and weather on it do not behave the same either. These are not measurements of rotation so much as the fingerprints it leaves everywhere.

33
“They keep changing the shape — sphere, spheroid, geoid — so it’s made up.”
Each refinement is a confirmed prediction: Newton derived the equatorial bulge from spin in 1687, and expeditions measured the ~21 km flattening by the 1740s. Refining a sphere to an ellipsoid is measuring it, not inventing it.
Earth is an oblate spheroid — the equatorial bulge →
30
“Water can’t curve around or cling to a spinning ball.”
Oceans are a film ~0.02% of Earth’s mass and 0.06% of its radius; gravity outpulls the equatorial spin ~290 to 1.
The “spinning water ball”: oceans are a thin film →
49
“The Coriolis effect is a fudge factor, and the draining-sink demo proves it’s fake.”
Storms spin opposite ways by hemisphere, artillery corrects for it, and dropped objects land east — the spin, measured. (The sink demo is a strawman.)
The Coriolis effect — the spin written into storms and gunfire →
51
“If the Earth were spinning, wouldn’t there be some measurable mechanical sign of it?”
There is: gravimeters read lighter heading east and heavier heading west (the Eötvös effect), confirmed by ship trials in 1908 and corrected for in gravity surveys today.
Travel east and you weigh less — the Eötvös effect →
50
“On a spinning ball rivers would flow uphill, or reverse direction several times a day.”
Neither: water runs downhill in the gravity field while the bed and banks co-rotate with it, so a river flows one steady way — the only daily reversals are the Moon’s tidal bores near the sea.
Rivers on a spinning globe — they don’t flow uphill, and they don’t reverse →
41
“Tides aren’t from the Moon — its gravity is too weak — and gravity isn’t real anyway.”
Tides come from the difference in the Moon’s pull across Earth’s width, not from the pull itself. That difference gives two bulges and two high tides a day, tracking the Moon and Sun.
Tides — two bulges a day from a force that falls off as 1/r³ →

06The objections that backfire

stated at full strength, then followed through

Each of these is a serious-sounding reason the Earth cannot be turning. Each one, followed to the end, produces a number that only works on a rotating globe.

42
“If Earth spun at 1,000 mph we’d feel it, and the oceans would fly off.”
You never feel constant velocity, only change; the residual spin acceleration (0.034 m/s²) is below the vestibular threshold, while gravity beats the outward pull ~289:1. You feel quakes, not the steady spin.
Why we don’t feel the spin (and the oceans don’t fly off) →
43
“A helicopter could hover while the Earth turns beneath it; planes couldn’t land on a 1,000 mph runway.”
Everything shares Earth’s eastward motion by inertia, so a hover drifts nowhere and a landing sees only airspeed — like pouring coffee on a cruising jet.
Why a hovering helicopter doesn’t land somewhere else →
54
“Pilots follow barometric pressure and never dip the nose, so the Earth can’t be a ball.”
The curve needs only ~0.0022°/s of pitch (~8°/hour), held automatically; a constant-pressure altitude is a shell wrapped around the globe.
Planes never “dip the nose” — and why that’s expected →
152
“Flight mechanics destroy the heliocentric hypothesis.”
The plane and the air share the Earth’s motion, so a steady spin cannot be felt — and the aircraft’s own inertial navigation finds north by sensing that very rotation.
Flight mechanics do not sink heliocentrism — they help confirm it →
153
“Motion is relative, so you cannot prove the Earth moves. A still, flat Earth is as valid as any other frame.”
Uniform motion is relative, but rotation and acceleration are not, and no change of frame turns a globe into a disc — the shape is the same in every frame.
Reference frames — what is relative, and what is not →
140
“NASA’s own documents say ‘flat, non-rotating Earth’ — they’re admitting it.”
It’s a simplifying assumption for short-range problems, stated with its limits; the same documents and agencies use the round, rotating Earth for long-range work — some patents switch to the sphere in the same filing.
When NASA writes “flat, non-rotating Earth” — idealizations quoted out of context →

07The arithmetic, if you want it

Both formulas below live in the reference’s equation appendix with copy and LaTeX buttons, alongside the plain-language walkthrough of what each symbol means.

Ring resonator f = (4*A / (lambda*P)) * (Omega . n_hat)
Held horizontal f = (4*A / (lambda*P)) * Omega_E * sin(phi)

A four-meter square ring lit at 633 nm has a scale factor of about 6.3 × 10⁶. Multiply by the Earth rate and, with the ring’s plane square to the axis as it would be at the pole, the beat note lands near 460 Hz. That is an audible tone, and it is the point worth holding on to: this is not an inference drawn from a model. It is a frequency somebody counts.

08Do it yourself

The pendulum is the one most people can manage: a long wire, a heavy bob, a still room and an afternoon. The swing plane turns 15 degrees an hour multiplied by the sine of your latitude, so at 45 degrees it comes to about 10.6 degrees an hour, or three and a half degrees in twenty minutes. The longer the wire and the heavier the bob, the longer the swing survives to show it.

169
Tests anyone can run, with what is already in the house.
The full list, with what each one costs, what it needs, and what it honestly does and does not settle.
Do-it-yourself — tests anyone can run →
X Reddit
Companion pages Rotation · Radio · Aviation · Aviation (plain) · Weather radar · Eratosthenes · Eclipse Aug 12 · Resources · Academy · What’s new