Measuring the Speed of Light: From Galileo’s Experiments to Modern Physics

In September 1676, Danish astronomer Ole Rømer made a bold prediction before the Royal Academy of Sciences in Paris: that the next eclipse of Io, one of Jupiter’s moons, would arrive exactly ten minutes later than predicted.^[1] His colleagues were skeptical. On November 9th, Io disappeared into Jupiter’s shadow precisely ten minutes behind schedule.

What Rømer had predicted was not merely an astronomical timing. It was a claim that light travels at a finite speed — that the delay in Io’s eclipse was caused by the extra time light needed to cross the greater distance between Earth and Jupiter. It had taken humanity over two thousand years of debate to reach this moment.

“Light Arrives Instantaneously” — Ancient and Medieval Debates

The first to ponder how fast light travels were philosophers. The Greek philosopher Empedocles, in the fifth century BCE, argued that light moves at a finite speed. He viewed light as a kind of traveling substance, reasoning that anything in motion must take time to do so.^[2]

Aristotle, however, disagreed. He argued that “light is not the result of any motion, but a change in the state of a medium.”^[2] Just as a drop of ink falling into a bowl of water changes the state of the water rather than moving through it, Aristotle believed that light does not travel across space but that the entire medium changes all at once. Under Aristotle’s authority, the view that light arrives instantaneously became dominant for centuries.

The debate continued in the Islamic world. The eleventh-century Arab scholar Ibn al-Haytham (known in the West as Alhazen), in his Book of Optics (Kitāb al-Manāẓir, 1021), argued for the “intromission theory” — that light enters the eye from objects rather than emanating from the eye.^[2] This theory presupposes that light travels, and Alhazen reasoned that light’s speed must be finite. His contemporary al-Bīrūnī also held that the speed of light was finite and estimated it to be far greater than the speed of sound.^[2] Yet these discussions remained in the realm of philosophical reasoning; no methodology existed for experimental measurement.

Galileo’s Lantern Experiment — The First Attempt and Its Limits

The first person to attempt an experimental measurement of the speed of light was Galileo Galilei. In his 1638 work Two New Sciences (Discorsi e Dimostrazioni Matematiche, intorno a due nuove scienze), Galileo proposed an experimental method.^[3] Two people would stand on hilltops, each holding a lantern. One would uncover their lantern, and the other would uncover theirs the instant they saw the light. By measuring the distance between the two lanterns and the round-trip travel time of the light, one could calculate its speed.

The experiment failed. Galileo himself admitted that the measured values were indistinguishable from human reaction times. If the two hilltops were about 1.5 kilometers apart, the round-trip travel time of light would be a mere 0.00001 seconds — something no seventeenth-century clock, and no human reflex, could capture.^[3]

What Galileo’s experiment left behind was not a measurement but a way of framing the problem. He transformed the philosophical question — “Is the speed of light infinite or finite?” — into a question for experimental science: “Is it a measurable physical quantity?” This transformation became the foundation for all subsequent experiments.

Ole Rømer’s Discovery — The Cosmic Clock Provided by Jupiter

About forty years after Galileo held his lantern, Rømer, working at the Paris Observatory, took an entirely different approach. He observed Io, the moon of Jupiter that Galileo himself had discovered. Io orbits Jupiter with a period of approximately 42.5 hours, and in doing so it periodically disappears into and emerges from Jupiter’s shadow. These “eclipses of Io” were so regular that they could serve as a kind of cosmic clock.

Between 1671 and 1673, Rømer observed more than thirty eclipses of Io.^[1] A strange pattern emerged. When Earth was approaching Jupiter, the eclipses occurred ahead of schedule; when Earth was moving away, they ran late. At first, this was suspected to reflect irregularities in Io’s orbit, but Rømer offered a different interpretation.

If light travels at a finite speed, then when Earth is close to Jupiter, the light from Io has a shorter distance to cover; when Earth is farther away, the light must travel farther. Covering that difference in distance takes time, and that time shows up as a difference in the timing of the eclipses.^[1] According to Rømer’s calculations, it took approximately 22 minutes for light to cross the diameter of Earth’s orbit (the modern value is about 16.5 minutes). Combining this with the known size of Earth’s orbit yielded a speed of approximately 220,000 kilometers per second — about 25% below the modern value of 299,792 km/s, a discrepancy attributable to inaccurate measurements of Earth’s orbital diameter at the time.^[1]

Rømer’s discovery immediately sparked controversy. His superior, Giovanni Cassini, initially supported the interpretation but later tried to explain it as an irregularity in Io’s orbit.^[1] Much of the European scientific community remained skeptical. It would take another 51 years before the finite speed of light was independently confirmed.

James Bradley’s Stellar Aberration — Independent Confirmation

In 1727 (published in 1728), British astronomer James Bradley confirmed the finite speed of light by a completely different method.^[4] He had set out to measure stellar parallax but discovered something unexpected: the apparent position of stars shifted periodically according to the direction of Earth’s orbit. Bradley named this phenomenon the “aberration of light.”

To grasp aberration intuitively, imagine walking in the rain. Even if the rain falls straight down, when you walk, the drops seem to come at you from an angle in front. Similarly, when viewed from an orbiting Earth, the direction from which starlight appears to come is slightly tilted. The magnitude of this tilt depends on the ratio of the speed of light to the speed of Earth’s orbit.^[4]

By measuring this ratio, Bradley calculated that it takes about 8 minutes and 12 seconds for light to travel from the Sun to the Earth — only 1.3% below the modern value of approximately 8 minutes and 20 seconds.^[4] This independently confirmed Rømer’s claim. Light was unambiguously traveling at a finite speed.

Yet astronomical methods had inherent limits. The precision of the underlying data — the distance between Earth and Jupiter, the size of Earth’s orbit — directly affected the results, and improving that precision was not easy. A new approach was needed.

Hippolyte Fizeau’s Toothed Wheel — The First Ground-Based Measurement

In 1849, French physicist Hippolyte Fizeau succeeded in measuring the speed of light on the ground, without any astronomical assistance.^[5] His apparatus was simple yet ingenious.

Fizeau placed a powerful light source in the bell tower of his father’s country house in Suresnes, on the outskirts of Paris, and erected a mirror on the hill of Montmartre, approximately 8,633 meters away.^[5] Between them he placed a rotating wheel with 720 teeth. Light passed through a gap between the teeth, reflected off the mirror, and had to pass through a gap again on the return journey to reach the observer.

When the wheel spun slowly, the gap through which the light had passed was still open when the reflected light returned. But as the wheel’s speed increased, the next tooth had moved into position to block the returning light. Fizeau found that at approximately 12.6 rotations per second, the light was completely blocked. Combining this rotational speed with the round-trip distance allowed him to calculate the speed of light.^[5]

Fizeau’s result was 313,274 kilometers per second — about 4.5% higher than the modern value, but it was the first measurement of the speed of light ever conducted entirely on Earth, without recourse to cosmic scales.^[5] The true significance of the experiment lay not in its precision but in its methodology. Fizeau demonstrated that a physics laboratory could be an instrument for measuring the universe.

Léon Foucault’s Rotating Mirror — Light Slows Down in Water

Fizeau’s contemporary Léon Foucault took a different approach. In 1850, he built an apparatus using a rapidly rotating mirror.^[6] Light reflected off the rotating mirror traveled to a distant fixed mirror and returned; in the time it took to make that journey, the rotating mirror had turned slightly. Measuring this tiny angular displacement revealed the travel time of the light, and combining that with the distance yielded the speed.

In his first experiment in 1850, Foucault obtained a result more significant than any absolute measurement of the speed of light. He demonstrated that light travels more slowly in water than in air.^[6] This was more than a simple measurement: it resolved a long-standing theoretical debate. Two competing theories of the nature of light had made opposing predictions. The “corpuscular theory” held that light is made of particles and should therefore travel faster in a denser medium like water; the “wave theory” predicted the opposite — that light should slow down in a denser medium. Foucault’s measurement was decisive evidence in favor of the wave theory.

In a second experiment in 1862, Foucault obtained a value of 298,000 kilometers per second.^[6] The deviation from the modern value was only 0.6%.

Albert Michelson — Perfecting the Art of Measurement

American physicist Albert A. Michelson obtained a value of 299,940 kilometers per second in his first light-speed measurement in 1879 — a deviation of only 0.05% from the modern value.^[7] The key to Michelson’s improvement on Foucault’s rotating mirror method was the length of the measurement path. Where Foucault had used a path of about 20 meters, Michelson employed a path of over 600 meters. The longer the path, the greater the angular displacement of the mirror, and thus the more precise the measurement.^[7]

Michelson’s most decisive contribution, however, came in the mountain experiments conducted between 1924 and 1926.^[7] He installed an octagonal rotating mirror between Mount Wilson and Mount San Antonio in California, a distance of approximately 35 kilometers. The reflected light reached the telescope only when the rotating mirror was aligned at precisely the right angle. From this experiment, Michelson obtained a value of 299,796 kilometers per second, with an error of approximately 0.001%.^[7]

Michelson received the Nobel Prize in Physics in 1907 for his work on light-speed measurement — the first Nobel Prize in science awarded to an American.

The Michelson–Morley Experiment — A “Failure” That Changed Physics

Michelson’s name is also linked to another experiment. In 1887, Michelson and Edward Morley set out to test one of the central assumptions of nineteenth-century physics.^[8] Physicists of that era believed that light propagated through an invisible medium called the “ether,” just as sound propagates through air. If a cosmic ether filled the universe, light was thought to travel through it.

If the ether existed, the speed of light should differ depending on whether it was measured in the direction of Earth’s motion through the ether or perpendicular to it. Michelson and Morley built a precision interferometer to detect this difference. By splitting a beam of light into two perpendicular paths and then recombining them, any difference in speed — however small — would appear as a shift in the interference pattern.^[8]

The result was: nothing. No matter which direction the measurement was taken, the speed of light did not change.^[8] This “null result” became one of the most famous failures in the history of physics. The ether theory was dealt a decisive blow, and the result became a crucial foundation for Albert Einstein’s special theory of relativity, published in 1905.

Einstein’s special theory of relativity rendered the concept of the ether unnecessary. His theory rested on two postulates: first, that the laws of physics are the same in all inertial frames of reference; and second, that the speed of light is always constant, regardless of the motion of the observer.^[8] The result of the Michelson–Morley experiment provided the experimental basis for this second postulate.

Lasers and the 1983 Resolution — Speed Defines Length

The development of laser technology in the mid-twentieth century elevated the measurement of the speed of light to an entirely new level. In 1972, the United States National Institute of Standards and Technology (NIST, then known as NBS) used laser interferometry to obtain a value of 299,792,456.2 meters per second, with an uncertainty of only ±1.1 meters.^[9] This was a hundredfold improvement in precision over previously accepted values.

Laser interferometry works by splitting a laser beam into two paths and then recombining them to observe interference patterns. By precisely adjusting the path length while analyzing how the pattern changes, the wavelength of the laser light can be measured with extraordinary precision; combining this with the known frequency yields the speed. The precision of this method was incomparable to anything achieved by rotating mirrors or toothed wheels.

In 1983, the 17th General Conference on Weights and Measures (CGPM) made an unprecedented decision. It fixed the speed of light at exactly 299,792,458 meters per second and used this value to redefine the meter.^[9] Today, one meter is defined as “the distance traveled by light in a vacuum in 1/299,792,458 of a second.”

The philosophical implications of this decision are profound. Originally, the meter was defined as one forty-millionth of Earth’s meridian; later, it was redefined as the length of a platinum-iridium alloy bar. Now it is defined by a physical constant. Rather than measuring speed, humanity began using speed to define length itself. Compared to Rømer’s attempts to gauge the speed of light from Io’s eclipses, the distance humanity has traveled in three hundred years is staggering.

The Meaning of the Act of Measurement Itself

Looking back on the history of measuring the speed of light, we can see that each era’s “limits” became the next era’s “methodology.” Galileo’s lantern experiment targeted something too fast for human reflexes; Rømer instead exploited the cosmic difference in distances. Fizeau sliced time with a rotating wheel; Michelson turned the gap between mountains into a laboratory.

What the Michelson–Morley experiment revealed is even more paradoxical. An experiment designed to find the ether instead proved its absence, and that “failure” became a cornerstone of special relativity. Few episodes in the history of science illustrate so clearly that a negative result can carry more significance than a positive one.

The 1983 redefinition of the meter was the culmination of that journey. Humanity, which had begun by measuring a physical constant, had arrived at the point of using that constant to redefine length itself. The number 299,792,458 meters per second is no longer a measurement. It is a definition.