got to work on one, and in 1632 he published it: Dialogue Concerning the Two Chief World Systems. In arguing on behalf of a Copernican view of the universe, Galileo knew he was going to have to explain certain phenomena that in the Aristotelian view of the universe needed no further explanation. Actually, he was going to have to explain the absence of certain phenomena: If Earth were turning and if this turning Earth were orbiting the sun, as Copernicus contended, then wouldn’t birds be rent asunder, cannonballs be sent off course, and even simple stones, dropped from a modest height, be flung far from their points of departure, all according to the several motions of the planet?
No, Galileo said. And here’s why. He asked you, his reader, to imagine yourself on a dock, observing a ship anchored in a harbor. If someone at the top of the ship’s mast were to drop a stone, where would it land? Simple: at the base of the mast. Now imagine instead that the ship is moving in the water at a steady rate across your field of vision as you observe from the dock. If the person at the top of the ship’s mast were to drop another stone, where would it land now? At the base of the mast? Or some small distance back, behind the mast—a measurement corresponding to the distance on the water that the ship would have covered in the time between the release of the stone at the top of the mast and its arrival on the deck of the ship?
The intuitive, Aristotelian answer: some small distance back. The correct—and, Galileo argued, Copernican—answer: the base of the mast, because the movement of the ship and the movement of the stone together constitute a single motion. From the point of view of the person at the top of the mast, the motion of the stone alone might indeed seem a perpendicular drop—the kind that Aristotle argued a stone would make in seeking to return to its natural state in the universe. Fair enough. That’s what it would have to seem to someone standing on the steadily moving ship whose only knowledge of the motion of the Earth was that it stood still. That person would feel neither the motion of the Earth nor the motion of the ship and so would take into account only the motion of the stone. But for you, observing from the dock, the stone would be moving and the ship would be moving, and together those movements would make up a single system in motion. To you, the motion of the stone falling toward the ship would seem not a perpendicular drop—not at all an Aristotelian return to its natural state—but an angle. If you could trace the trajectory of the stone from the dock, it would just be geometry.
And vice versa. If, instead, you the observer standing on the dock were the one dropping a stone, then to you the motion of that stone relative to the Earth would appear perpendicular, because all you would be taking into account was the motion of the stone alone. That’s all Aristotle did—take into account only the motion of the stone. But from the point of view of the person at the top of the mast on the ship in the harbor, looking at you on the dock and taking into account the motion of the stone and the apparent motion of the dock together, the trajectory of the falling stone would describe an angle.
And there it is: a principle of relativity. Neither observer would have the right to claim to be absolutely at rest. The onboard observer would have as much right to claim that the ship was leaving the dock as that the dock was leaving the ship. Rather than standing still at the center of the cosmos, our position in the new physics was just the opposite: never at rest. After Galileo, everything in the universe was in motion relative to something else—ships to docks, moons to planets, planets to sun, sun (as astronomers would come to discover by the end of the eighteenth century) to the so-called fixed stars, those socalled fixed stars (as astronomers would come to discover by the middle of the nineteenth century) to one another, and, conceivably, our entire vast system of stars (as astronomers were trying to determine at the turn of the twentieth century) to other vast systems of stars.
Unless you counted the ether. For this reason alone, the ether was—as Einstein had first recognized as a teenager—at least somewhat objectionable. Not long after he’d written the ether paper that he’d sent to his uncle, Einstein found himself wandering the grounds at his school in Aarau, Switzerland, wondering what the presence of an absolute space would do to Galileo’s idea of relativity. If you were on board Galileo’s ship but belowdecks, in an enclosed compartment, you shouldn’t be able to detect whether you were moving or standing still, relative to the dock or anything else in the universe that wasn’t moving along with you. But if the ship were traveling at the speed of light through the ether, that’s just what you would be able to detect. You’d know you were the one traveling at the speed of light—rather than someone on the dock, for instance—because you’d see the light around you standing still.
By the early years of the twentieth century, Einstein had done only what other physicists of his era had done. He’d thought about ways to define the ether through mathematics. He’d thought about ways to detect the ether through experiments. He’d even begun to think about whether physics really needed an ether. But then, one night in May 1905, Einstein did what no other physicist of his era had done. He thought of a new way of thinking about the problem.
Einstein had been spending the evening with a longtime friend both from his student years and at the patent office, Michele Besso, the two of them talking, as they often did in their off-hours, about physics. In the preceding three years, Einstein had moved to Bern, gotten married, and fathered two children (one illegitimate, whom he and Mileva gave up for adoption). Yet all the while he’d been applying himself to the most pressing issues of contemporary physics, often in the company of his patent-office sounding board, Besso. On this particular occasion, Einstein had approached Besso for the express purpose of doing “battle” with a problem that had been plaguing him on and off for the past decade. After a lively discussion, Einstein returned home, where, all at once, he understood what he and everyone else who had been studying the situation had been overlooking all along.
“Thank you!” he greeted Besso the following day. “I have completely solved the problem.” The trouble with the current conception of the universe, he explained, wasn’t absolute space—or at least wasn’t only absolute space. It was absolute time.
“If, for example, I say that ‘the train arrives here at 7 o’clock,’ that means, more or less, ‘the pointing of the small hand of my watch to seven and the arrival of the train are simultaneous events.‘“ This sentence comes early in “Zur Elektrodynamik bewegter Körper” (“On the Electrodynamics of Moving Bodies”), the paper that Einstein completed and mailed to the Annalen der Physik six weeks later. In its audacious simplicity, even borderline simplemindedness, this sentence is deceptive, for with this description of one of the most mundane of human observations—one that just about any eight-year-old can make—Einstein pinpointed precisely what everyone else who had been studying the problem had missed: “time” is not universal or absolute; it is not sometimes universal and sometimes local or relative; it is only local.
The key was the speed of light. The fact that the speed of light is not infinite, as Aristotle and Descartes and so many other investigators of nature over the millennia had supposed, had been common knowledge since the late seventeenth century. So had its approximate value. In 1676, the Danish astronomer Ole Rømer used the data from years of observations at the Paris Observatory to determine that the timing of the eclipses of Jupiter’s innermost moon depended on where Jupiter was in its orbit relative to Earth. The eclipses came earlier when Earth was nearest Jupiter, later when Earth was farthest from Jupiter, suggesting that the eclipses didn’t happen at the very same moment we saw them happen. That, in fact, when we saw them depended on where they happened, nearer or farther. “This can only mean that light takes time for transmission through space,” Rømer concluded—140,000 miles per second, by the best estimates of the day.
But the combination of these two factors—that the speed of light is incomprehensibly fast; that the speed of light is inarguably finite—didn’t begin to assume a literally astronomical dimension for another hundred years. Beginning in the 1770s, William Herschel (the same observer who proved that the sun is in motion relative to the fixed stars) began systematically exploring the so-called celestial vault—the ceiling
