Over at my new blog, Twisted Physics, I wrote about the Telectroscope, a whimsical art installation linking New York City and London, ostensibly via a giant tunnel under the Atlantic Ocean connected on either end via a telescopic lens. In reality, the connection is made via fiber optic cable and a gigantic Webcam, but either way, you've got information traveling across the pond at the speed of light, enabling Londoners and New Yorkers to wave and hold up signs in greeting, pretty much in real time. It just so happens that I've also been reading George Johnson's excellent new book, The Ten Most Beautiful Experiments, this week while recovering from some minor bug. (I swear my body breaks down on a regular basis just to force me to get some sleep.) Chapter 8 describes historical attempts to measure the speed of light, culminating with the famed Michelson-Morley Experiment in 1887 to detect something called the "luminiferous aether, a medium thought to pervade all of space to enable light waves to propagate. People might have their quibbles with Johnson's choice of experiments to include in his top ten, but I'd have to agree that Michaelson and Morley deserve to be in the pantheon -- if only because it's a prime example of how illuminating (even revolutionary) a null result can be.
Galileo first suggested a scheme for measuring the speed of light in the early 1600s: stand on a hilltop at night and flash a bright light toward a distant hill, where one's assistant would see the flash and respond by flashing back. Galileo lacked an accurate timepiece, alas -- this was the man who used his own pulse when measuring the speed of balls rolling down an incline plane, after all -- so the best he could conclude was that "if not instantaneous, light is very swift." Even back then, "really, really fast" fell a bit short of prevailing scientific standards for precision, but it wasn't until the 1670s that Danish astronomer Ole Roemer used his observations of Jupiter and its moon, Io, to arrive at a light speed of about 140,000 miles per second. (The critical observation: Io seemed to be slowing in its orbit at certain times of year, and Roemer rightly surmised that this was because as it moved farther from Earth, its light took longer to reach Earth.)
Fifty years later, English astronomer James Bradley confirmed Roemer's estimate while tracking a star called Gamma Draconis. It seemed to wander from its expected position, and eventually Bradley figured out it was because by the time the starlight reached his telescope, the Earth had shifted position. Bradley refined Roemer's original estimate to 183,000 miles per second.
Next on the scene were a pair of French physicists, Hippolyte Fizeau and Leon Foucault (of Foucault's pendulum fame). In 1849, Fizeau used two fixed mirrors to measure the speed of light, one partially obscured by a rotating cogwheel. Light projected between the spinning cogwheel's teeth would reflect off the mirror and be sent back through the wheel, and from the length of the light path and the speed of the wheel, Fizeau was able to estimate the speed of light. It was a noble effort, a more sophisticated version of Galileo's earlier approach with two lanterns, but the resulting estimate was about 5% too high (196,000 miles per second). About 13 years later, Foucault improved on Fizeau's original design and devised an apparatus that reflected light off one rotating mirror, toward a stationary mirror some 20 miles away. He refined Fizeau's estimate to about 185,000 miles per second.
And this is where Albert Michelson made his grand entrance onto the speed-of-light stage. He knew all about his predecessors' experiments. Per Johnson, I learned that Foucault's experiment produced a displacement less than a single millimeter, which was very difficult to measure accurately. Michelson figured he'd just project the beam down a much longer path than Foucault's 20 meters, resulting in a greater lag time. "The returning beam would hit the mirror later in its cycle, resulting in a larger deflection and, he hoped, a better value for the speed of light," Johnson writes. And it did! Michelson's experiment yielded the best measurement of the speed of light yet: 186,350 miles per second. (Today's accepted value is 186,282.397 miles per second, so Michelson was impressively accurate.)
Ah, but this is not the famed experiment honored in Johnson's Top Ten, impressive though it was. After successfully measuring the speed of light, Michelson turned his attention to the question of the luminiferous aether. This was a mysterious substance believed to pervade the universe, serving as a transport medium for light. The assumption was that light, like sound, needed a medium through which to propagate.
Johnson's poetic take is that Michelson yearned for an unmoving constant, something fixed against which all things could be measured. The luminiferous aether served such a purpose, and Michelson conceived of a method of measuring the motion of the Earth against the aether by sending a light beam in the same direction that the Earth was moving around the sun. The light beam should be slowed a bit by the "aether wind" that everyone assumed would be produced as the aether flowed over the Earth as our blue planet moved through it. (It was assumed at the time that light did not travel at the same velocity in all directions.)
To make this delicate measurement, Michelson invented a truly ingenious device, later called an interferometer (scientists still use such instruments today). He sent a beam of light through a half-silvered mirror, thereby splitting it into two beams traveling at right angles to each other. The beams traveled out to the ends of long arms and bounced off small mirrors, causing them both to return to the center and recombine in an eyepiece, producing an interference pattern. Even the slightest change in the time spent to make the trip should be observed as a shift in the positions of the interference fringes.
He teamed up with a chemist named Edward Morley to refine his basic prototype apparatus. In the seminal experiment, the light as reflected back and forth along the arms repeatedly. The entire apparatus was housed in a closed room in the basement of a stone building to shut out vibrations and variations in temperature (which might cause the brass arms to expand or contract), and they placed the experiment on a large block of marble floating in a pool of mercury to further reduce vibrations. To their surprise, there were no interference effects at all. In essence, the speed of the earth through the aether was, for all intents and purposes, zero. So there was no need for an aether at all, and furthermore, the speed of light in a vacuum appeared to be independent of the speed of the observer. This experiment was refined and repeated many times up until 1929, always with the same results and conclusions.
Technically, the Michelson-Morley experiment was a failure, in that it did not measure the effect the men expected to find; quite the opposite, it disproved their basic premise, and the very existence of the aether. Albert Einstein explained the results when he published his theory of special relativity in 1905. One of the central tenets is that there is no such thing as a fixed frame of reference. (In physics, a frame of reference simply denotes where a person or object happens to be standing relative to the rest of the universe.) So there is no absolute reference frame for time against which all motion can be measured. This was a pretty radical idea at the time, although modern scientists take it as a given.
Einstein explained that this is because everything (and everyone) is constantly in motion through both space and time, and therefore has its own unique frame of reference. This has some pretty bizarre implications. For instance, two people who are moving relative to each other, wearing identical watches, will measure time differently. Time will slow down or speed up depending on how fast each is moving. This usually isn't noticeable because your average wristwatch isn't sensitive enough to measure the tiny discrepancies that appear at slower speeds. Time dilation only becomes significant at speeds approaching the speed of light, when the effects are greatly magnified.
Einstein also asserted that space and time are one. Our three-dimensional existence -- the "where" of an event -- evolves along the fourth dimension of time -- the "when" of an event -- so we live in a four-dimensional space time. Ergo, what happens to time must also happen to space. So as time dilates for an object in motion, the object's length contracts along its horizontal axis by a corresponding amount.
If one wanted to be silly -- and one might, here at the cocktail party -- one might say that a good way to look noticeably thinner is to travel at faster speeds relative to the rest of the world. After all, the faster an object moves, the more its horizontal length contracts, at least from the perspective of an outside observer. Should one approach the speed of light, one would become so thin as to appear almost two-dimensional to an outside observer -- all without giving up a single calorie. Alternatively, one could claim that one wasn't so much fat, as really, really slow. (Here's a bit more risque take: Jen-Luc Piquant once asked a male acquaintance how he felt about the possibility relativistic "shrinkage"; he took it in stride, shrugged and replied, "That's why I try to move as slowly as possible.")
Silliness aside, while it's true that space and time are in the eyes of the beholder, the speed of light is constant, regardless of frame of reference. That's the true significance of the Michelson-Morley experiment. Scientists before Einstein had assumed that motion through space was the same as motion through time, but if space and time are one, then the two types of motion are linked. Time dilation and length contraction occur because space and time adjust with motion to ensure that two people moving relative to each other will always measure the same speed for light. Light is the link between space and time.
It also sets the cosmic speed limit. According to Einstein, nothing can travel faster than the speed of light. And there's no such thing as an event happening at the same exact moment for two observers in different frames of reference because none of us ever sees the world as it is right "now." We can't tell something has happened until the information about that event reaches us, and there is always a delay of at least the speed of light before that happens. How much it is delayed depends upon the relative speed of whoever is observing an event.
So far, Einstein's theory is holding strong, despite the deployment of ever-more-sensitive instruments and ingenious approaches to find the tiniest hint of violation of the laws of special relativity (as well as general relativity). And the null results continue. Thus far, it seems, special (and general) relativity holds true (at least until you get down to the quantum scale). I'll give Johnson the last word:
[Michelson] died ... in 1931, just months after meeting Einstein, whose special theory of relativity had explained the true significance of Michelson and Morley's beautiful experiment: they had proved, contrary to their expectations, that there is no fixed backdrop of space, or even of time. As we move through the universe, our measuring sticks shrink and stretch, our clocks run slower and faster -- all to preserve the one true standard. Not aether, but the speed of light.