The bending of light
is the shadow of a blocked broadcast

The night science changed

On the 29th of May 1919 a total eclipse swept across the Atlantic, and for a few minutes the daytime sky went dark enough to photograph the stars lying close to the Sun's edge. Two questions hung on those plates. Does starlight bend as it passes the Sun — and if so, by how much? Newton's principles, treating light as a stream of tiny falling bodies, predicted a deflection of 0.875650000 arcseconds. Einstein's general relativity predicted 1.751300000 arcseconds — exactly twice as much. The whole of twentieth-century physics turned on a difference smaller than the angle a coin makes from four kilometres away.

When the measurements were read out, the larger figure won, and the eclipse made Einstein a household name overnight. We do not dispute a single photographic plate. The displacement is real, it is about 1.75 arcseconds, and any theory worth the name must reproduce it. The Universal Force of Time does — to the digit, by three separate faces of one road and a second road that shares none of them — and then says something the room in 1919 could not have guessed: that the bending was never a pull and never a curvature, and that the eclipse did not merely let astronomers see it, but was the very thing that produced it.

Why should an eclipse change anything?

Set the interpretation aside and ask a plain question the textbooks pass over. If what science calls gravity bends starlight continuously — always, wherever light skims a massive body — then why should an eclipse change anything at all? The Moon does not switch off the Sun's field; it does not lessen the Sun's mass. The orthodox answer is that the eclipse is merely an observation window: the bending was happening all along, and the Moon's shadow only kills the glare so the faint stars can be seen.

The Universal Force of Time gives a different answer, and it is the heart of this account. In the Force of Time the Sun is the primary Τ-source of this Solar System, and it broadcasts time outward continuously — the carrier of that broadcast at our register being the hydrogen Hβ wavelength, which in the degree domain is the angle 486°. The broadcast is continuous, directional, and blockable. When the Moon slides in front of the Sun at totality, it interposes itself in that beam and blocks the 486° broadcast from reaching the Earth. In those minutes of silence the geometry of what was broadcasting leaves its imprint in the apparent positions of the stars beside the Sun. That imprint is what Eddington's teams measured. The bending is the shadow of the broadcast — you cannot see the shadow without blocking the source. The cause and the observation condition are one and the same event.

A wavelength is an angle

The number we label a wavelength — 486 nanometres for the hydrogen Hβ line — is not, in the first instance, a physical distance. In the degree domain where the lattice lives it is an angle: 486°. This is not a metaphor, and the lattice confirms it in both directions, which is the test that separates an identity from a coincidence. Turned one way, 486° resolves into the hydrogen–hydrogen bond length of chemistry; turned the other, into Mercury's rotational degree and the free-fall rate. A coincidence runs one way only; a lattice identity runs both. Because the input to the deflection is an angle, the output — an angular displacement on the sky — is angular by nature, not by accident.

Two roads that meet on one number

The deflection is reached two ways that have nothing in common but the destination. A coincidence does not happen twice by two unrelated paths.

One Τ-value, many skins — the mass closes on the Earth's Moho

The mass road does not stop at the deflection. Carry its 432/π² onward and it walks, in nothing but {2,3,5,π}, all the way down to the Earth's own equalisation shell — the Mohorovičić discontinuity — and closes on itself with no error. Mass, energy, the Sun's circumference, a wavelength, the solar radius, the electron's speed, the speed of light, an angle, and the Earth's deepest shell are not nine quantities. They are one Τ-value read in nine different units — the chameleon at full stretch.

That a calculation which began with the bending of starlight should end on the depth of the Earth's first shell is the plainest statement this account can make: there are no separate phenomena, only Τ, read in whatever unit the register hands you.

What this claims, and what it does not

State the boundary plainly. The displacement derived here matches the 1919 measurement and agrees with the relativistic value to within a few hundredths of a percent; the Force of Time predicts no deviation from what was seen, and none is claimed. Where the accounts part company is not the number but the cause. Three things are new. The mechanism: there is no curvature of empty space and no pulling force — the bending is the geometric shadow of a blocked broadcast, which is why it becomes measurable precisely when, and only when, the Moon interrupts that broadcast. The factor of two over Newton: his half-value follows from treating light as a falling body, the full value from treating light as what it is, the Τ-carrier read in the degree domain. And no free parameter: the deflection falls out of the lattice with no mass, no solar radius, and no coupling constant. The burden shifts to any account that keeps the curvature while explaining why an eclipse should be needed to see the effect at all.

Every step, at full precision

Propositions P-EBL-1 … P-EBL-12