Logo IMG
HOME > PAST ISSUE > March-April 2006 > Article Detail


Lost in Einstein's Shadow

Einstein gets the glory, but others were paving the way

Tony Rothman

Tell Me It Ain't So

The central fixture of Einstein lore is that the lowly patent clerk conjured from pure thought not only his theories but also the questions they answered. Not quite: Einstein himself helped foster this myth (more through carelessness than design, one suspects) by being less than fastidious about providing references in his papers, and since then credulous scientists have equated absence of evidence with evidence of absence. Physicists are notorious for taking history on faith, but none is required to prove this point—the evidence is in plain sight if one cares to look. The papers of Einstein and his contemporaries, as well as Einstein's letters, are published. Anyone who reads them quickly realizes that Einstein had a very good sense of the currents of science swirling about him and once or twice relied on the insights of colleagues.

Take the Australian William Sutherland, for example. One of Einstein's great 1905 papers explained "Brownian motion," the random jiggling of microscopic pollen grains suspended in water. Einstein proposed that the movements were due to collisions between the pollen and invisible water molecules and inferred from this the size of the molecules themselves. In doing so, he provided one of the final "proofs" of the existence of molecules, which in 1905 was still debated.

Sutherland was also interested in the motion of small particles suspended in liquids. In 1904, he proposed to the Australian Association for the Advancement of Science a method to calculate their mass. In March 1905—at exactly the time Einstein was working on his own paper—Sutherland submitted an improved version of his idea to the Philosophical Magazine, the leading English-language scientific journal of the day. It was published in June. In this paper, Sutherland derives the "diffusion coefficient," a number giving the rate at which particles move through a liquid. Moreover, he does so by exactly the same argument Einstein gives and arrives at exactly the same answer. Einstein goes on to use this number to derive the well known "diffusion equation." It tells him how far the particles will move in a given time, depending on the size of the surrounding molecules. A stopwatch and a microscope then allow him to measure molecular dimensions.

So, yes, Einstein went further than Sutherland, but Sutherland got one of the two crucial steps first. From letters to Einstein from his best friend, engineer Michele Besso, we know that the two men showed keen interest in Sutherland's work through 1903. After that, the discoveries were certainly independent.

In any case, a Frenchman, Louis Bachelier, scooped both Sutherland and Einstein. Bachelier was not actually interested in the twitching of suspended pollen grains. He was interested in the motion of prices on French stock market.  But prices on the Bourse bounce around like pollen in water and their "random walk" can be treated mathematically like the diffusion of pollen in a liquid, which is exactly what Bachelier did in his remarkable 1900 doctoral thesis, "The Theory of Speculation." His paper is full of the jargon of economists, but Bachelier's equation giving the drift of prices with time is identical to Einstein's for pollen. Bachelier anticipated the Black-Scholes approach to options trading (which garnered its authors the 1997 Nobel Prize in Economics) and has been crowned by modern economists the "father of mathematical finance." Back then, though, academia ignored him, and it would be surprising if Sutherland or Einstein had heard of him.

Of course, 1905 is remembered above all for relativity. As a result of a famously vague statement in Einstein's own paper about "unsuccessful attempts to detect the motion of the Earth relative to the 'light medium'," pundits have long held that he was only dimly aware of the celebrated experiments that failed to reveal the mysterious medium—the ether—whose existence was synonymous with the "absolute space" implicit in classical physics. That is, sound waves travel in air, water waves travel in water; physicists naturally assumed that light required a medium in which to travel—the ether. The trouble was, a whole series of experiments designed to detect it turned up empty handed. It was these negative results that eventually led to relativity.

Did Einstein know about the vigorous search for the ether? Well, in an 1899 letter to his fiancée, Mileva Maric, Einstein mentions that he's written to renowned physicist Wilhelm Wein about Wein's review of the ether experiments, and that he's anxiously awaiting a reply. Einstein also read an 1895 paper in which Hendrik Lorentz (independently of two others) postulated his famous "Lorentz contraction"—that objects moving at high speeds actually shrink. That paper was all about the ether experiments, and Lorentz introduced the contraction precisely to explain their failure.

Lorentz was no amateur. The Dutchman was considered the leading physicist of his generation, and soon he and his colleagues were waging a well-published attack on the hypothetical ether and all its difficulties. In 1904 Lorentz tried to fix everything with his celebrated "transformations" that mixed up space and time in a way that—if true—would leave Maxwellian electromagnetic theory intact but shake the foundations of Newtonian physics.

Lorentz didn't know why his transformations should be correct. With relativity Einstein provided the explanation, but shortly before his death, he claimed to have known only about Lorentz's 1895 paper, not the later one containing the transformations. Memory is often too good to be true. In Einstein's very paper of 1905 he says, "we have thus shown that ... the electrodynamic foundation of Lorentz's theory ... agrees with the principle of relativity." This appears to be a direct reference to Lorentz's 1904 work.

comments powered by Disqus


Of Possible Interest

Feature Article: Restoring Depth to Leonardo’s Mona Lisa

Spotlight: Making the Cut

Spotlight: Briefings

Subscribe to American Scientist