The Library and Lighthouse of Alexandria

In earlier chapters, we met the Alexandria of mathematicians and scientists, of astronomers and geographers, of anatomists and physiologists, and of poets and playwrights. Now we turn to another facet of the city’s endlessly multi-faceted personality, one which was there almost from the beginning but which arguably reached full flower only after the coming of the Romans: the Alexandria of engineers and inventors.

The tradition apparently began with a man named Ctesibius, who comes to us only as a figure of legend from the glory days of the first three Ptolemies. One tale of his origins, so delightful that one has to hope it’s true, says that he was born a humble worker, the son of an Alexandrian barber. But, like a Henry Ford or a Steve Wozniak of another age, he was an inveterate tinkerer to whom invention came as naturally as breathing. His father’s barber shop benefited enormously from his talents when he was still a teenager; he rebalanced the swiveling mirrors the customers peered into, for example, so that they stood still when turned to any position. And then, broadening his horizons beyond barbering technology, he invented the first clock that we would recognize by that name today.

We must understand that the Alexandrians’ conception of time at this point in their history was very different from our own. Their only form of clock was the natural one of the sun and moon and stars wheeling overhead, or at best sundials that reckoned time from light and shadow. (Fortunately, Alexandria was blessed with near-constant sunshine.) Your day began when the sun came up and ended shortly after it went down, and if someone told you to meet him at a given place at noon, you understood that to mean nothing more specific than “around the time that the sun is directly overhead”; by necessity, schedules didn’t get any more precise than that. Luckily, there were no railroad timetables to keep.

The closest thing to a mechanical clock to exist before Ctesibius was a gadget known as a clepsydra — literally translated, a “water thief.” It was a receptacle with a spout at the top and a stoppered hole at the bottom. To use it, one filled it up with water, then removed the stopper. When the clepsydra was empty, an arbitrary amount of time — exactly how much was uncertain — was known to have passed. Rather than being used to count time in the abstract, the clepsydra was used to parcel out equal quantities of it. It might be used to ensure equal time during a political debate, or used in court to ensure that both the defense and the prosecution had the same amount of time to present their arguments: “Lawyers are driven by the clepsydra, never at leisure,” wrote Plato. Which is not to say that it didn’t have its applications for leisure-time activities: brothels allegedly used it to ensure that customers all got equal time with the girls or boys.

One might presume that it would be possible to measure more granular if still arbitrary units of time by waiting for the clepsydra to be half emptied, or a quarter so, etc. Certainly it would be simple enough to mark different water levels on the surface of a glass clepsydra. But it wasn’t quite that easy, due to a frustrating reality of physics: the more water in the clepsydra, the more quickly it flowed out of the hole at its bottom, thanks to the greater weight of water pushing on the whole from above. Thus the real halfway mark in terms of time would come at some indeterminate point well after the clepsydra was half empty of water. This meant that, while the clepsydra could be used to ensure equal time, it couldn’t be used to measure time in any real sense whatsoever — until Ctesibius.

His ingenious solution to the conundrum began with a second, much larger clepsydra mounted above the the first one, dispensing water into it at such a rate that the smaller clepsydra below it always remained full; this ensured that water would always flow out of the hole in its bottom at a constant rate. Another receptacle was then placed below this smaller clepsydra to capture this water. Markings on its surface allowed it to measure granular units of time with impressive accuracy.

From this start, Ctesibius’s water clocks evolved into gadgets of amazing sophistication, with feedback mechanisms that allowed them to run as long as the topmost clepsydra was kept full of water; some were even capable of chiming the hour. It would be some 1800 years before a more accurate method of measuring time was devised.

The story goes that, having demonstrated his genius with his water clock, Ctesibius the humble barber’s son was invited to join the Museum of Alexandria — and, indeed, may have become its head in time. He spent the rest of his life beavering away in his workshop, inventing more of the sorts of everyday conveniences for ordinary people which caused the rarefied likes of Archimedes to turn up their noses in disgust: pipe organs, lawn sprinklers, etc.

Alexandria itself soon became known as the technological center of the Mediterranean world, the place to come to buy gadgets both frivolous and practical. Alongside all of its other identities, it was a city of cogs and gears, of blowing air and running water — for, what with the absence of electricity, pneumatics and hydraulics had to serve in its stead. The docks and wharves of this eminently commercial city featured the most advanced cranes and hoists to be found anywhere in the classical world — “Give me a place to stand and a lever long enough, and I will move the world,” Archimedes had said — even as Alexandria boasted the most sophisticated indoor plumbing of any city of the era. Small wonder that, some 200 years after Ctesibius, the besieged Julius Caesar couldn’t help but admire the many intricate mechanisms his enemies rolled into place to oppose him. And yet the most astonishing days of Alexandrian engineering were still to come.

We know shockingly little today about the man with the auspicious name of Hero, who was widely regarded during his own time and long afterward as the most prolific inventor of complex, often semi-autonomous machinery that the world had ever known. We can identify even the rough era in which he lived thanks only to a lunar eclipse that we know to have taken place in AD 62, and which we find mentioned in one of his surviving writings; thus we know that he was alive and active during that year. But there is, alas, little else we can say about Hero the man with any degree of certainty whatsoever. So, his writings about the things he made will have to speak for him. Many of them too have been lost to us, but a fair number have reached us, either in the original Greek or in later translations into Latin or Arabic. They speak eloquently of his extraordinary ingenuity.

The most lengthy and complete of Hero’s texts that has come down to us is known as the Pneumatica. In it, he describes no less than 75 separate devices. Some of them have, as he puts it, “useful everyday applications,” while others produce only “quite remarkable effects”; these were designed simply for the pleasure of making them and watching them go rather than for any more practical purpose.

Even the allegedly practical inventions tend to have more than a whiff of frivolity about them. Among the most amusing of them is the world’s first vending machine, which was used to dispense holy water to penitents in Alexandria’s temples. When one dropped a coin into a slot, it fell onto a pan that lay at one end of a tiny beam balance inside the machine. The weight of the coin forced the opposite end of the balance up, where it pushed open a valve, causing holy water to flow out of the machine and into the hands or cup of the eager worshiper. But the pan on which the coin lay was formed in such a way that the coin would slowly slide down and off of it, to fall into a repository of its brethren below. When this happened, the beam balance returned to its normal position and the holy-water valve closed again until the next coin was deposited into the slot. The priests of the temple needed only fill up the machine with holy water from time to time, and of course collect the money that was constantly rolling in.

It would seem that Hero enjoyed a fruitful and presumably lucrative relationship with the priests of Alexandria. In addition to his holy-water vending machines, he provided them with the world’s first automatic doors, which they used to dramatic effect during their religious ceremonies. A hollow tube conducted heat from a fire burning atop one of their altars to a cauldron of water hidden in the basement of the temple. As its temperature grew hotter, the water expanded and was forced through a pipe to a heretofore empty bucket, which stood at one end of a beam balance whose other side was connected to a door via ropes and pulleys. When the combined weight of the bucket and the water it contained reached a tipping point, the beam balance moved and pulled the door open. This event, of course, marked the climax of the ceremony above: the audience of worshipers were meant to assume that the door had been opened by a god who had come to join them in incorporeal form.

Hero also invented a fortune-telling machine to automate the tedious work of priestly prophecy. It was a sort of Magic 8-Ball of the ancient world: pay a small fee, walk into a booth, ask the god a yes-or-no question, then turn a wheel to receive a (random) answer.

His inventions could be found in Alexandria’s theaters as well. He was a master of special effects, able to make curtains go up and down and props propel themselves about the stage, their movements “programmed” using a bewildering array of hourglass-style timers. When the actors complained about the difficulty of staying in sync with all of the other action happening onstage — once Hero started his mechanisms, there was no way to stop or pause them, meaning the actors had to adapt the performance to their inexorable progress — Hero resolved to cut real humans out of the show entirely. He thus began to develop his most astounding creations of all: intricate moving dioramas, in which life-size mechanical men played out scenes for the delight of crowds. We might go so far as to call him the world’s first roboteer.

Four centuries before Hero, Aristotle had dreamed presciently of a world where “every tool [is] able to complete its own task when ordered — or even anticipate the need.” He wrote of “shuttles that could pass through the loom by themselves, or plectra play the harp, master craftsmen [with] no need of assistants, and masters [with] no need of slaves.” Hero dedicated himself to bringing this vision to fruition, at least in the context of entertainment, and appears to have had lots and lots of fun of his own in the process; his joy of creation leaps from his descriptions of his inventions, a child’s delight in things that move and spin and whistle of their own accord. Small wonder that his contemporaries called him the machine man — a description that could apply equally well to some of the things he built.

In his book Automata, he describes one of his grand public dioramas, featuring the god of wine Dionysus, who was always a great favorite with the masses for obvious reasons. Let us picture the scene:

A roofed and elevated stage has been erected in the middle of a busy city square. Dionysus himself stands six feet (1.8 meters) tall at its center, holding a staff in his left hand and a cup of wine in his right, surrounded by dancing figures of Maenads, his libertine female followers. Altars stand some distance in front of him and behind him on the stage, a tamed panther lies at his feet, and the winged Nike, the goddess of victory, hovers above him. In response to a clockwork mechanism, run by ropes and pulleys and hourglass timers hidden beneath the stage, the god lumbers forward on a hidden track until he stands directly before the front altar, and Nike glides in the same direction to remain above him. He leans down over the altar, and fire flares up. Water, milk, or, if the crowd is lucky, wine shoots out of the end of his staff into the audience, fed via a hose from a reservoir hidden below him. At the same time, he casually turns the cup in his right hand, and a stream of wine flies in the direction of the thirsty panther. Meanwhile the circle of Maenads begins to spin around him, accompanied by gear-driven kettledrums and cymbals. After a moment, the fire in the altar dies down, the music and dancing stop, and Dionysus and Nike rotate a neat 180 degrees, then return to the center of the stage and thence onward to the altar at the other side, in order to delight the crowd standing over there with the same pantomime. After the robot performers have returned to the center of the stage from the rear altar, a harried human attendant hidden in the bowels of the mechanism hastily resets everything, then sets the show in motion again.

Hero made smaller dioramas as well, on the scale of a puppet show, which made up for their lack of size in their even more intricate, even more finely honed complexity. One of them that is described by its maker in Automata tells a complete story that takes place over five separate scenes. It involves Nauplius, the king of the Greek island of Euboea during the Trojan War. Nauplius’s son went to join the fray in Troy like most young Greeks of noble birth, only to meet a tragic end there. In the version of the tale whose sequel is presented here, the son crossed the hero Ajax in one of the Greeks’ many internal disputes over strategy and lost his life for his trouble.

As the first scene of Hero’s mechanical play begins, the Greeks have sacked Troy at long last and are preparing to return home in triumph. Nauplius, however, has asked the goddess Athena to avenge the murder of his son. Here is how Hero himself describes the action that follows:

At the outset, when the box opened, twelve painted figurines appeared: these were divided into three rows; they were made to represent some of the Greeks refitting their ships and busy launching them.

These figurines moved, some sawing, some working with axes, some with hammers, some others using bow-drills and augers, and they made a lot of noise, just like in real life. After sufficient time elapsed, the door closed and opened again, and there was another arrangement; the ships, in fact, were shown being launched by the Greeks. After [the box] closed and opened again, nothing appeared in the box except painted sky and sea.

Not long after, the ships sailed in line ahead, and some were out of sight, some in view. Often dolphins swam alongside too, sometimes plunging into the sea, sometimes visible, just like in real life. The sea gradually grew stormy, and the ships ran uninterruptedly. After [the box] closed again and opened, none of the sailing ships was seen, but Nauplius holding up the torch and Athena standing beside him were seen.

Fire blazed up above the box, as if a flame appeared on high from the torch. After [the box] closed and opened again, the wreck of the ships appeared, and Ajax swimming; and a machine was raised above the box, and as thunder rumbled in the box itself a bolt of lightning fell on Ajax, and his figure vanished. Thus, when the box closed, the story came to an end.

The invention of Hero that is the most hotly debated of all today was apparently created as just another form of entertaining spectacle. Yet it carried within it the seed of something infinitely more useful. Hero created nothing less than the first documented example of an engine powered by steam — also the first example of a reaction turbine of the sort used in a modern jet airplane.

The device that has become known as Hero’s engine starts with a closed cauldron of water mounted just above a fire pit. The lid on top of the cauldron has two pipes running up to a hollow sphere which is mounted such that it can rotate in place along a single axis. A pair of narrower, L-shaped exhaust tubes are affixed to the surface of the sphere, their outlets reciprocal to its axis of rotation.

When one kindles a fire below the cauldron, it heats the water inside, producing steam which runs up through the pipes into the sphere, then out through the exhaust tubes. This causes the sphere to spin of its own accord. The hotter the fire becomes, the more quickly the sphere will spin, in a haze of steam and high-pitched whistling. It must have been a very impressive sight for people unaccustomed to seeing non-living objects of any sort moving of their own accord.

But we have no evidence that Hero or any of the Alexandrians who followed him ever even thought about turning this parlor trick into a practical machine. Was this down to a colossal lack of vision? Could Hero’s engine have been made to do real, useful work for the people of Alexandria? John G. Landels, a historian of ancient engineering, is decidedly skeptical of the notion.

Could this form of steam engine ever have been used as a practical power source? The answer is, almost certainly not. It operates best at a high speed, and would have to be geared down in a high ratio. Hero could have managed that, since the worm gear was familiar to him, but not without friction loss. Inadequate heat transfer from the burning fuel to the cauldron would keep the efficiency low. It is in the realm of possibility that, given the technology of Hero’s age, overall efficiency might have been as low as one percent. If so, then even if a large-scale model could have been built, to deliver .1 horsepower and do the work of one man, its fuel consumption would have been enormous, about 25,000 B.T.U. per hour. The labour required to procure and transport the fuel, stoke the fire and maintain the apparatus would have been much more expensive than that of the one man it might replace, and the machine would be much less versatile.

Still, one can easily enough imagine Hero’s engine as a stepping stone to a far more useful form of steam engine. Those which powered the Industrial Revolution of the late eighteenth and early nineteenth centuries used steam to drive pistons inside sealed cylinders rather than venting it to the open air for the amusement of spectators. Most of the parts necessary to build just such a contraption were very familiar to Hero. A type of hand-driven pump called a force pump, long in use in Alexandria and elsewhere — in fact, Ctesibius was sometimes claimed to be its inventor — utilized pistons and cylinders and rocker arms uncannily similar to those of an Industrial Revolution-era steam engine. And whilst experimenting with the use of hydraulics to drive fountains, Hero himself designed and made valves adequate enough for this type of steam engine. Could the Industrial Revolution have arrived 1700 years early if this one man had but had a different set of priorities? It’s a tempting thought to contemplate.

In fact, some have been tempted by that thought into making disparaging judgments of Hero the man, portraying him as a natural genius who wasted his gifts on trivialities delivered for personal financial gain. In his authoritative two-volume study Greek Science, Benjamin Farrington writes with something close to sarcasm of how Alexandrian science,

when it lost its ambition to transform the material life of man by being applied to industry, quickly acquired fresh application. It became the handmaid of religion and was applied to the production of miracles in the Serapeum and other temples. To the conscience of the age, these scientific aids to devotion hardly differed in principle from the use of improved lighting effects or the introduction of organ music, which were also achievements of this age. They were intended to create a pious public, to make religion attractive and impressive, and seem to have done so. When science began to flourish again in the modern world, it had another purpose than to deceive.

But was the purpose of Hero’s many inventions really to “deceive?” And did the people who witnessed his “miracles” really believe that they were the products of gods? We can plainly see in many of the texts of the early first century AD that religiosity was in marked decline among the intellectual classes of that period. Many, many authors treated the gods more as metaphors than as living entities, or chose to ignore them altogether. It’s of course possible and even likely that sincere religious belief was more prevalent on the streets of Alexandria than inside the city’s museum and library, but did even these people really believe that the gods were the engines behind Hero’s clockwork miracles? I suspect from the tone of his surviving texts that he saw himself above all as an entertainer giving the people a good time with an accompanying wink and nudge, like an ancient P.T. Barnum, and that the people he supposedly duped probably saw his productions in the same light, and willingly suspended their disbelief in the same way that we do when we go to see a stage magician today. The religiosity that would eventually overwhelm the daily life of the Alexandrian streets would be, as we’ll see in later chapters, of a very different character from Hero’s showy pagan spectacles.

The issue of whether Hero should have been doing something “better” — something more serious — with his undeniable genius is a thornier one. On the one hand, it’s true that there are no equivalents of Archimedes’s screw pump or Ctesibius’s water clock in his catalog of inventions, only relatively frivolous tools for commerce or entertainment. But on the other hand, we should not be too quick to judge him, given that we know literally nothing of the man’s circumstances, nothing about what combination of compulsions and opportunities might have led him down the path he followed. And then, simply providing joy and entertainment to others is a worthy end in itself, one which our modern culture values enormously.

These known unknowns haven’t kept Hero’s apologists from defending the man just as spiritedly as his detractors have condemned him. All sorts of wild possibilities have been mooted by way of justifying his failure to build upon his proto-steam — and proto-jet — engine in particular. Perhaps he actually did keep working on it, only to have it blow up in his face and kill him. Or perhaps it exploded and merely frightened and/or injured him badly enough that he left off further experimentation. Since we know nothing concrete of him beyond our record of his inventions, we must acknowledge both as possibilities at least — although one does have to suspect that an exploded Hero would be a remarkable enough story that some ancient scribe whose writings have reached us would have mentioned it.

In the end, debates like these are moot because the fate of steam power in Alexandria never really depended on Hero alone at all. Had steam engines been obviously useful in first-century Alexandria, someone else if not the man himself would have built upon Hero’s work. The real stumbling block to a steampunk Alexandria wasn’t a lack of knowledge of pistons and cylinders, rocker arms and valves, nor even the considerable limitations of ancient metallurgy. It rather came down to the vagaries of economics and culture.

The Industrial Revolution of our actual history ran on coal, a substance which was almost unheard of in ancient Alexandria. The vastly less efficient fuel of wood was more familiar, but even it was much more expensive than in other cities, what with Egypt being such a timber-poor land. Meanwhile the wealthy elite of Alexandria had thousands and thousands of laborers at their disposal. They had no motivation to invest in steam technology as long as human capital was so cheap. Purchasing enough fuel to feed any conceivable ancient steam engine would have cost far more than any value such an engine could add for its owners over simply ordering others to do its work by hand.

Lest we be tempted to judge these ancient Alexandrian elites too harshly, we should remember that they had never seen an industrial revolution, and had no idea what such a thing might look like or, indeed, that it might come to exist at all. Likewise, the caste systems that arbitrarily made people of leisure and intellect of a few, poor laborers of most, and slaves of a substantial minority was as deeply intertwined with their society’s conception of itself as the egalitarian ethos is with so many of our own.

We can, however, take some solace in noting that some of the groundwork of our modern conception of society, which I as a product of my own times naturally find to be a fairer, juster way to live, would be laid in the Alexandria of the centuries immediately after Hero. Unfortunately, much of the empirical practicality that made his gadgets go would be retired during the same period — retired not only in Alexandria but virtually everywhere else in the Western world as well, and for many, many centuries to come. So, the Industrial Revolution would just have to wait until all of the pieces were finally in place at the same time.

Did you enjoy this chapter? If so, please think about pitching in to help me make many more like it. Pledge any amount you like on Patreon.

(A full listing of print and online sources used will follow the final article in this series.)