How Scientists Discovered The Speed Of Light

Think Science
73 followers
AI Generated·May contain errors
In 1676, a Danish astronomer working in Paris noticed that a moon of Jupiter was running late,
not randomly late, not occasionally late, laid in a pattern so specific that it could only mean
one thing, and that one thing was something most of the scientific world flatly refused to believe.
The moon was low. It orbits Jupiter like clockwork, eclipsed by the planet at intervals so regular
you could set a calendar by them. Ole Ermer had been timing those eclipses for months.
But as Earth moved to the far side of its orbit, pulling away from Jupiter, the eclipses kept
arriving behind schedule. Then, as Earth swung back closer, they arrived early. The total drift
across the year was close to 22 minutes. Ermer looked at that drift and said,
that is the time it takes light to cross the diameter of Earth's orbit. Nobody believed him.
To understand why that matters, you have to understand what people thought about light
before Ermer put a number on it. The dominant position was that light was instantaneous.
Not fast. Not extremely fast. Instantaneous, meaning it covered any distance in zero time.
Aristotle had said it. Descartes had defended it mathematically. Even Kepler, who rebuilt the
architecture of the solar system, wrote that light required no time to travel. When Galileo
tried to test this in 1638 by having two men uncover lanterns at a distance and timing the
response, the result told him only that if light had a speed, it was far too fast for any human
reflex to catch. That was about as far as the experiment could go. So when Ermer arrived at
the Paris Observatory in the early 1670s, nobody had assigned a number to the speed of light.
Ermer wasn't trying to. He had been tasked with something practical, refining the orbital tables
of Jupiter's moons so sailors could use them to determine longitude at sea. The moons of Jupiter
were supposed to function as a universal clock, visible from anywhere on Earth. To make that work,
the eclipse times had to be predictable with precision. That's when the pattern in the data
became impossible to ignore. Every time Earth was moving away from Jupiter,
Io's eclipses fell behind schedule. Every time Earth moved toward Jupiter, the eclipses arrived
early. The drift wasn't random. It correlated perfectly with Earth's position in its orbit,
with the changing distance between the two planets. Ermer's conclusion was direct. The delay
is the time light takes to travel the extra distance. When Earth is farther from Jupiter,
light from Io has more space to cross, so the signal arrives late. When Earth is closer,
less distance, earlier arrival. The 22-minute total discrepancy represented the time light
needed to cross Earth's entire orbital diameter. Using the best available estimate of the Earth-Sun
distance, Ermer calculated a speed of roughly 220,000 kilometers per second.
He published this in 1676 in the Journal des Savants, and his own supervisor, Giovanni Cassini,
publicly rejected it. Cassini's skepticism wasn't entirely unreasonable. The Earth-Sun distance
Ermer used was uncertain. The eclipse timing data was imprecise enough that critics could argue the
drift was observational error. There was no independent way to check the result.
Christian Huygens accepted Ermer's reasoning and used the figures to calculate his own estimate
of light speed. But most astronomers either stayed silent or remained unconvinced.
The idea sat in the literature, neither confirmed nor refuted, for decades.
Then, in 1725, an English astronomer named James Bradley set out to do something completely
unrelated. He wanted to prove heliocentrism geometrically by detecting stellar parallax.
The logic was straightforward. If Earth orbits the Sun, nearby stars should appear to shift
slightly against the background of more distant stars as Earth moves from one side of its orbit
to the other. Bradley targeted a star called Gamma Draconis using a precise telescope at Q
and started tracking its position across the year. He did detect a shift. But it was entirely the
shape. Parallax would produce a shift that peaks when Earth is at the widest points of its orbit,
perpendicular to the direction of the target star. What Bradley saw peaked when Earth was moving
directly toward or directly away from the star. The displacement was 90 degrees out of phase
with what parallax required. Bradley spent months working through possible explanations.
The one that finally fit was deeply strange. The apparent position of a star is not where the star
actually is. It's where the star appears to be, given the combined motion of light traveling
The same thing happens with rain if you're running. Rain that falls perfectly vertically
appears to come at an angle from ahead. Light from a star behaves the same way.
Because Earth is orbiting the Sun, moving at roughly 30 kilometers per second,
starlight appears to come from a slightly different direction than its true source.
The amount of that shift depends on the ratio of Earth's orbital speed to the speed of light.
Bradley called it aberration. And when he used his measurements to calculate the speed of light,
he arrived at approximately 301,000 kilometers per second. He published this in 1728 in the
Philosophical Transactions of the Royal Society. That number derived from a completely different
phenomenon using a completely different method. Landed within a few percent of what Armour's
drifting eclipses had implied. Two independent anomalies. Two separate paths through the
mathematics. Nearly the same answer. That convergence was the first real evidence
that Armour had been right all along. But both measurements shared a common dependency.
They relied on astronomical distances, specifically on knowing how far Earth is
from the Sun. If that distance was slightly wrong, both answers would be slightly wrong
in the same direction. What nobody had yet done was measure the speed of light using only equipment
you could hold in a laboratory, with distances you could walk yourself and times you could
calculate from pure mechanics. The reason nobody had done it was straightforward.
Light covers a kilometer in roughly three microseconds. In 1800, in 1820, even in 1840,
there was no mechanical device fast enough or precise enough to time that interval.
You'd need a shutter that could open and close millions of times per second. Or a wheel that
spun fast enough to meaningfully change position during the few microseconds a light pulse was in
transit. In 1849, a French physicist named Hippolyte Fizeau built exactly that. He set up a
light source in the Montmartre district of Paris, aimed it at a toothed cogwheel with 720 teeth,
and through one of the gaps sent a beam of light toward a mirror on a hill in Montmorency,
about 8.6 kilometers away. The mirror reflected the beam straight back. At the right rotation speed,
the returning light would pass back through the same gap it left through.
At slightly different speeds, a tooth would block it. Fizeau adjusted the wheel. He found the exact
speeds at which the returning beam was blocked and the speeds at which it passed cleanly,
from the geometry of the wheel and the known distance to the mirror. He could calculate
how long the light took to make the round trip. His result was approximately 313,000 kilometers
per second. Not perfect by later standards, but it was the first speed of light measurement
taken entirely on Earth, with physical apparatus you could examine and distances you could survey
with a measuring tape. A year later, Lon Foucault refined the approach. He replaced the cogwheel
with a rotating mirror. In Foucault's setup, a beam of light hit a spinning mirror, traveled to a distant
fixed mirror, and came back. By the time it returned, the spinning mirror had rotated by a
small measurable angle. That angle encoded the travel time. The smaller and more precisely
measurable the angle, the better the result. Foucault got approximately 298,000 kilometers
per second, considerably closer to the modern value. But Foucault's apparatus produced something
that had nothing to do with his original goal. He submerged part of the light path in water
and measured how fast light traveled through it. This was not an idle curiosity. There were two
competing theories of what light actually was, and they made opposite predictions about this exact
situation. Newton's crepuscular theory, which treated light as a stream of particles, required
light to speed up when entering a denser medium, like water, in order to explain refraction.
The wave theory developed by Huygens, and later extended by Young and Fresnel,
predicted light should slow down in water. Foucault measured it. Light in water travels
slower than light in air. Newton's particle model made a clean, specific, testable prediction,
and the measurement contradicted it directly. The wave theory survived. The particle theory
of light did not. That result from Foucault's rotating mirror experiment in 1850 didn't just
refine the speed of light. It settled a debate about the fundamental nature of light that had
been running for two centuries, which left one remaining question. If light is a wave,
what is it waving in? Every wave anyone had ever studied needed a medium. Ocean waves need water.
Sound waves need air. The assumption was that light waves needed something too, a substance
called the luminiferous ether that filled all of space and served as the medium for light's
oscillations. This wasn't considered radical. It was considered obvious. In 1887, Albert
Mickelson and Edward Morley built an instrument in Cleveland to detect that ether. The logic
was elegant. If the ether fills space and earth is moving through it, then light traveling in
the direction of earth's motion should be measurably slower than light traveling perpendicular
to it, because it would be fighting the ether current one way and riding it the other.
The difference in speed would be small, but Mickelson's interferometer was extraordinarily
sensitive. It split a beam of light into two perpendicular paths, sent them to mirrors,
and brought them back together. If one path took longer than the other, the returning beams would
be slightly out of phase, producing a visible interference pattern. They rotated the apparatus.
They measured. They looked for the phase shift that the ether theory required. There was no
shift. Within the sensitivity of the instrument, light arrived in the same time regardless of
which direction it traveled. They repeated the experiment. They rebuilt it larger.
They tried again. Nothing. The ether they were looking for didn't seem to be there.
Light wasn't faster or slower depending on the direction of earth's motion through space.
Its speed appeared completely constant. That result destroyed the ether theory.
But for 18 years, nobody had a satisfying replacement. Some physicists proposed that
objects physically contract in the direction of motion through the ether, which would
mathematically cancel the expected effect. The equations describing this were worked out by
Hendrik Lorentz and George Fitzgerald. They were mathematically useful. They were also physically
strange because they required the contractions to be exactly the right size to make the ether
permanently undetectable, which started to sound less like a theory and more like a workaround.
In 1905, Albert Einstein published a paper that abandoned the ether entirely.
He didn't propose it was undetectable. He proposed it was unnecessary. His starting
point was a postulate. The speed of light in a vacuum is the same for all observers,
regardless of how fast they or the light source are moving. Not approximately the same.
Exactly the same. Always, this sounds reasonable until you think about what it requires.
If you are on a train moving at a thousand kilometers per hour, and you throw a ball
forward at a hundred kilometers per hour, an observer standing still sees the ball moving
at 1100. That's basic addition. But if you shine a light forward from that train,
the stationary observer does not see it moving at the speed of light plus the train speed. They
see it moving at exactly the speed of light, and so do you, from inside the train. The speed is
identical for both of you, even though you're moving relative to each other. The only way that
can be true is if space and time themselves adjust to keep the speed of light constant.
Time runs differently depending on how fast you're moving. Distances change. Mass and energy
are related through the speed of light in the equation E equals mc squared. Everything in that
framework flows from the constancy of a number first approximated by a Danish astronomer
staring at the moons of Jupiter in 1676. Armour's original estimate was off by about 27 percent,
mostly because the earth-sun distance he used was too small. Bradley's aberration measurement
tightened it. Fizzo's cogwheel turned it into something measurable on earth. Foucault's rotating
mirror brought the number closer and eliminated Newton's light theory along the way. Mickelson,
who had built the aether detector in Cleveland, spent much of the rest of his career building
progressively better instruments to measure the speed of light more precisely. In the 1920s,
using a rotating octagonal mirror and a carefully surveyed baseline between two California mountains,
he pushed the measurement to within a few kilometers per second of the modern accepted
value. The modern definition of the meter is now derived from the speed of light. Since 1983,
one meter is officially defined as the distance light travels in a vacuum in exactly 1,299,792,458
of a second. The speed of light is no longer measured against the meter. The meter is defined
against the speed of light. The quantity that began as a mysterious drift in eclipse timings
became so precisely established that it now defines our basic unit of distance.
Go back to Io for a moment. A small rocky moon about the size of Earth's moon completing an
orbit of Jupiter every 42 hours. Ermer was watching it vanish into Jupiter's shadow
and timed it again and again. What he had actually detected was something almost absurdly indirect
light leaving Io, traveling across hundreds of millions of kilometers of empty space,
reaching a telescope in Paris, and arriving at intervals that revealed,
buried in the accumulated minutes of delay, the finite speed at which the universe communicates
with itself. He didn't know he was finding a universal constant. He thought he was improving
an astronomical table. The universe had other plans.
