What we know of Archimedes’ last days are sketchy but indicative of how one worldview was replacing another. Plutarch tells us it all began when Archimedes developed the compound pulley, not for practical purposes but as a way of explaining a mathematical problem to his king. He then demonstrated its efficacy by pulling a laden ship from its dock single-handed, using a complex series of ropes and pulleys.
As the Roman siege tightened, it was these mechanical skills that Hiero called on Archimedes to use against his enemy. Plutarch tells us that Archimedes turned his mind to creating wonderful siege engines which employed the mathematics he so loved to perform similar apparently superhuman acts. Great catapults were designed to rain stones upon the legions who foolishly believed themselves to be out of range. When Roman ships approached the harbor walls, cranes were swung out which either dropped huge rocks on their ships to sink them or hooked them out of the water, dumping their crews in the sea before overturning them and consigning them to the depths. Plutarch provides a vivid description of the result on the Roman soldiers’ psyches:
Such terror had seized upon the Romans, that, if they did but see a little rope or a piece of wood from the wall, instantly crying out that there it was again, Archimedes was about to let fly some engine at them, they turned their backs and fled.
Plutarch, Life of Marcellus, in Parallel Lives, 17
Of course, these tales had probably been greatly embellished over the centuries between the siege of Syracuse and Plutarch’s day, but they contain at least an echo of what was clearly a gargantuan struggle and a glimpse into the difference in mind-set between Greeks and Romans. Romans could terrify Greeks with numbers and brutality—their strength was their army; Greeks, however, could panic Romans with their minds. Their strength was in books—their arsenal, a library in Alexandria.
In the end, however, the relentless might of the Roman army triumphed, and thought gave way to force. As the Roman army flooded into the town, the Roman general Marcellus sent out orders to fetch Archimedes for him. But somehow things went wrong and Archimedes was killed. But Plutarch cannot leave it so baldly writ and gives us instead one last glimpse of the great man, still the disconnected theoretician, still more in love with ethereal mathematical ideas than reality, still more concerned with universal constants than the value of his own life:
Archimedes . . . was . . . as fate would have it, intent upon working out some problem by a diagram, and having fixed his mind alike and his eyes upon the subject of his speculation, he never noticed the incursion of the Romans, nor that the city was taken. In this transport of study and contemplation, a soldier, unexpectedly coming up to him, commanded him to follow him to Marcellus; which he declining to do before he had worked out his problem to a demonstration, the soldier, enraged, drew his sword and ran him through.
Plutarch, Life of Marcellus, in Parallel Lives, 19
This is probably Plutarch’s own romantic notion of the great man’s death. He also records two other versions, the last of which may perhaps hold more of a glimmer of the truth:
As Archimedes was carrying to Marcellus mathematical instruments, dials, spheres, and angles, by which the magnitude of the sun might be measured to the sight, some soldiers seeing him, and thinking that he carried gold in a vessel, slew him.
Plutarch, Life of Marcellus, in Parallel Lives, 19
Death at the hands of a looting soldier searching for gold would not be a surprising end for a Syracusan captured by the Romans. But what is intriguing is the mention of what Archimedes was actually carrying, which was likely worth more than gold. These strange devices created by Archimedes for predicting the movement of the heavens, or marking time, or estimating location may have been what Marcellus was really after. They were certainly within Archimedes’ scope. We know he wrote a book on the construction of planetaria, although tragically this is lost, and in his long friendship with Eratosthenes he must have spoken of, and had probably seen, the armillary spheres constructed by his friend in the great porticoes of the museum. Furthermore, his creation of mechanical planetaria is hinted at by the Roman statesman Cicero in his Tusculan Disputations (book 1, chapter 25). We also have another tantalizing quote by the fourth-century AD poet Claudian, suggesting that Archimedes took the theoretical models of the universe he had read of in Alexandria and, through his engineering genius, made them into real devices:
Look! By his skill an old man of Syracuse has copied the laws of the heavens. . . . An animate force within attends the different stars and moves along the living mechanism with its regulated motions. A pretend zodiac runs through its own private year and the simulated moon waxes with the new month.
Claudian, Archimedes’ Sphere, in Epigrams
But even more extraordinary than this, we may actually have an ancient example of one of Archimedes’ machines. Thanks to a chance discovery in 1900 and the work of another Alexandrian, we might even be able to reconstruct one of these devices, perhaps the most exotic mechanism from the ancient world: not a clock or a planetarium, but a computer.
The story of Archimedes’ computer begins, appropriately enough, in Egypt. Since its invention there sometime in the fourteenth century BC, a device known as the clepsydra, or “water thief ,” had been used for marking out time.
Improved on by the Greeks, it was a simple but very effective timer consisting of a metal sphere with a tube sticking out of the top and small holes drilled in the bottom. The container was filled with water and the tube was stopped up. When you wanted to begin timing something, you’d take the stopper out and let the water slowly drain out of the tiny holes in the bottom. Every time it was filled with the same amount of water it took the same time to empty.
The clepsydra was an ideal timer, often used in courts to time the speeches for the defense and prosecution so lawyers and clients had equal time and didn’t drone on too long. When the bung was taken out, you started pleading your case. When the water ran out, you stopped—time up! The system had an added advantage. Should your great speech be interrupted, you could put the bung back in and put everything on hold. And so in Greek courts began our obsession with, and indeed enslavement to, time. Plato became one of the earliest writers to complain about the relentless pace of life when he said, “Lawyers are driven by the clepsydra—never at leisure” (Plato, Theaetetus, Introduction and Analysis, part 2).
But the problem with this device is that it wasn’t a clock, just a timer. When it was full, water rushed out quickly, slowing as the head pressure reduced. This made it a very uneven way to measure time even if it was refilled as soon at it emptied. The people of Alexandria were used to uneven time. Each day was divided into twelve “hours” not of a fixed length but simply one-twelfth of the time between dawn and dusk. Confusingly, this meant that an hour in summer was a lot longer than an hour in winter.
Then around 270 BC a barber in Alexandria had a simple but brilliant idea that would change time forever. Ctesibius worked in his father’s barbershop but was something of an amateur inventor as well. Visitors to the shop could have seen his ingenuity in the angle-poise mirror he invented to assist his father, but it was when he turned his mind to time measurement that he made his greatest breakthrough. Ctesibius realized that if the clepsydra was always full, then the water pressure would always be the same and the water would flow out at the same speed. So he added another water tank above the clepsydra. This poured water into the top faster than it could flow out the bottom. That meant the clepsydra was always full and any excess water just overflowed. Then he put another tank below the outflow to catch the water coming out of the clepsydra. That tank now filled up at a constant speed, so if a scale was put on the tank and a pointer was floated in it, then he could measure time constantly. Ctesibius’s water clock made him justly famous. In 270 BC he had created not only the first mechanical clock but one so accurate it wouldn’t be bettered for another 1,800 years.
Using the knowledge he had gained from his days of experimenting in his father’s barbers
hop, Ctesibius was now inventing a whole new subject: hydraulics. He realized that the constant dripping of water could do more than tell the time. Soon Ctesibius’s clocks were smothered in stop-cocks and valves, controlling a host of devices from bells and puppets to mechanical doves that sang to mark the passing of each hour—the very first cuckoo clock! Having mastered the hours, Ctesibius now found time to invent the organ and build singing statues for the Ptolemaic pharaohs.
Mechanical marvels like these must have come to Archimedes’ attention when he first arrived in Alexandria, since they’d first been constructed only twenty years or so before. Indeed, Ctesibius may still have been alive when the young philosopher first stepped ashore at the Great Harbor.
Just where these devices were set up is unclear from what we can recover of the city plan of ancient Alexandria, but across the Mediterranean in Athens a nearly contemporary device does survive which may fill in a missing link in the story of Archimedes’ “spheres.”
If you walk through the Plaka district of Athens that skirts the foot of the Acropolis, you come eventually to the Roman marketplace, or agora. Here amid the ruins a strange little octagonal building, constructed in white Pentelic marble, still stands. Known today as the Tower of the Winds, after the weather-beaten figures of the winds carved onto its faces, it has survived only because it was thought to be the tomb of Socrates and Plato and was later converted into a small Christian church or baptistery. Long before that, however, it was known as the Horologion of Andronicus and was the municipal clock of ancient Athens, built by the astronomer Andronicus of Cyrrhus in the first century BC. On the sides facing the sun were sundials for telling the time on sunny days, while on the top a weather vane indicated the wind direction.
But inside, through one of the two Corinthian doorways, was Ctesibius’s legacy: a mechanical device for telling the time on cloudy days or at night when a sundial wouldn’t work—a mechanical water clock. The sockets and scars that line the walls of this peculiar marble tower can still be made out today. These once held the mechanism for one of the water clocks invented by Ctesibius, and inside the drip, drip, drip of his clepsydra beat out the rhythm of the classical world.
But the Tower of the Winds may have held a device even more extraordinary than a water clock—indeed, Ctesibius’s device may simply have provided the steady power for a yet more groundbreaking ancient discovery, one that contemporary sources hint may have been invented by Archimedes, inspired by the astronomical work of his friend Eratosthenes and the hydraulic genius of Ctesibius. And we can conjecture this because we have one of these machines, thanks to a storm in the Aegean Sea.
Spring weather on the seas off Greece is notoriously changeable, and it was with some annoyance but little surprise that Captain Dimitrios Kondos and his group of Greek sponge divers found themselves sheltering from a storm, miles off course, by an isolated island known as Antikythera, between Kythera and Crete in the Ionian Sea, just before Easter in AD 1900.
As the weather cleared, the men decided to make the most of their unplanned trip by diving in the deep, clear waters off the island. After all, even the great Greek philosopher Aristotle had said that the best, softest sponges lie in the deepest waters. The first man down was Elias Stadiatos. It was a dangerous dive of some two hundred feet to the seafloor: There was always the chance that the divers might return to the surface with nitrogen narcosis—the notorious “rapture of the deep.”
When Elias broke the surface again that was exactly what his crew-mates thought had happened to him. The man was shouting incoherently. As they dragged him out of the water and unscrewed his heavy helmet, he clutched at Captain Kondos and began jabbering about a pile of dead women on the seafloor—everywhere dead women.
His friends tried to calm him, thinking nitrogen narcosis had driven him mad. But Elias Stadiatos wasn’t mad, and, disturbed but intrigued by his diver’s words, the captain ordered another dive later that day to discover exactly what it was he had seen. Better prepared than his comrade for what lay beneath, this diver returned to the surface with a very different story. There were indeed human bodies littering the seafloor, but they were not corpses—they were ancient statues from a long-lost wreck, surrounded by a treasure the likes of which had never been seen.
Overnight the wreck became one of the most celebrated finds from ancient history, and when Captain Kondos returned the following year, this time with Greek archaeologists, it was to salvage Elias’s “dead women” from the greatest ancient treasure ship ever found. Over the summer of 1901 thousands of priceless items were winched from the Antikythera wreck. Huge boulders scattered across the site proved to be rare bronze statues, covered with two thousand years of encrustation. Marble statues also reappeared, along with coins, beautifully decorated Greek vases, jewelry, and lavish tableware. Everything fine and rare from the ancient world seemed to be there.
But there was something else besides the obvious treasure. Divers that summer had recovered a lump of thick bronze corrosion with traces of wooden panels clinging to the outside. It certainly wasn’t a bronze statue, so no one took much interest. But it would prove to be the most extraordinary object to have survived from the ancient world. By 1902 this corroded lump had still not been properly examined, but a lot more was known about the Antikythera wreck. Pottery on board had been identified and dated. The wreck was of a merchant ship that had sailed sometime in the first century BC from the Greek islands of Rhodes and Cos to Rome with a priceless cargo of Greek art no doubt destined for the home of a wealthy Roman.
Like the sponge divers some two thousand years later, this ship too seems to have been caught out by a storm and driven toward the remote little island of Antikythera. But its captain hadn’t been as lucky as Captain Kondos. The wealthy buyer who sat in Rome all those centuries ago waited in vain for his shipment of Greek wonders, as the ship carrying them had disappeared beneath the waves.
It was sometime in 1902 that an archaeologist at the National Museum of Greece, Valerios Staïs, decided to have another look at the mysterious lump. When he looked carefully he saw there was something among all the corrosion—metal plates covered with writing. He got a specialist in Greek inscriptions to check it out, and sure enough it was ancient writing—from the first century BC. All the words on it that were legible also seemed to refer to astronomical or zodiacal terms.
But there was more to come. Since its removal from the water the wooden panels that had encased the object had begun to dry and crumble away. As Staïs gently removed them he made another discovery—layers of carefully interlaced cogs and wheels. Whatever was in this lump was some kind of ancient machine.
Initially there were two schools of thought. One said that it was clearly too complicated to be ancient and must have been dropped overboard onto the wreck centuries later. The other said it was just an astrolabe, a type of navigational device known from the seventh century BC onward. But it was in fact neither. The writing on the plates was clearly ancient Greek of the same period as the wreck, so it had to be contemporary, and it was obviously much more complicated than any ancient astrolabe. It would take more than another fifty years for the answer to be found, and then not by an archaeologist but by an English physicist.
Derek de Solla Price was in a unique position to tackle the problem of what was by then known as the “Antikythera mechanism.” Holding PhDs in both experimental physics and the history of science, he had been trying to understand the mechanism since the early 1950s, when the first X-ray images of the still only partially inspected device had been taken. In 1959 he published his initial findings in Scientific American, suggesting that the device was far more complex than previously thought, but this paper was met with disbelief by classicists, who considered such a thing impossible for the date.
It was only in the early 1970s that gamma ray images of the machine, taken by the Greek atomic energy authority, came into his hands and he could finally announce to the world what lay behind the corrosion and concretions: a
computer.
Price’s meticulous study of the cogs, gear ratios, and inscriptions enabled him to put together a model of how the Antikythera mechanism worked and what it did. The mechanism was a hugely sophisticated analog computer for calculating the movements of the planets, the rising and setting times of stars and constellations, and the phases and movements of the moon—a complete mechanical calendar and model solar system in a box. By turning a crank handle that would have been on the outside of the wooden box it was possible to calculate the time, day, month, season, and year. It even corrected for errors in the old Egyptian calendar, which, without leap years, lost a quarter of a day each year. The Antikythera mechanism had a special “slip dial” that could be adjusted for that. By looking at how this dial was set, modern computers have calculated when the mechanism was last set and hence the date of the wreck: 80 BC.
So sometime in 80 BC the proud owners must have set the correction dial on this incredible device for the last time. Perhaps they had just sold it to a wealthy Roman and were shipping it off, along with numerous other Greek treasures, from Rhodes or Cos. The new owner must have been looking forward to its arrival—perhaps more so than all the other treasures on board. With this machine he could calculate exact dates and times, make corrections for the notoriously inaccurate official calendars—in short, be master of time itself.
This, then, was the successor to the spheres and planetaria that we know Archimedes wrote about theoretically and that Plutarch places in his arms at the moment of his death. But where had this device come from? And who built it? The answer perhaps lies in the one surviving book by a Greek called Geminus. In this almost unknown work he describes a mechanism he says was built in 87 BC. There are three extraordinary things about this description. First, he appears to be describing a machine like the Antikythera mechanism. Second, the wording he uses is repeated almost exactly on the inscribed plates of the mechanism itself. Finally, Geminus was from Rhodes, where another great philosopher, possibly even his tutor, lived; Poseidonius, who according to Cicero, had made a “sphere . . . the regular revolutions of which show the course of the sun, moon, and five wandering stars, as it is every day and night performed” (Cicero, On the Nature of the Gods, book 2, chapter 34). Rhodes was also of course probably the starting point of the mechanism’s fateful last journey, some five years after Geminus had written.
The Rise and Fall of Alexandria Page 17