The wheels of Greek astronomical science

Over the past fifty years, the Antikythera Device has gone from being the most anomalous and controversial artefact to one of the most renowned pieces of evidence of the scientific genius of our ancestors – a millennium ahead of its time.

Philip Coppens

In 1900, a Greek sponge diver called Elias Stadiatos, working off the small Greek island of Antikythera, found the remains of a Greek ship at the bottom of the sea. The wreck was 50 metres long, located 15-25 metres off Point Glyphadia, lying in 43 metres of water. At the time, diving had to be done without the aid of any modern technology currently available to the diving community. It meant the work was highly dangerous. In fact, when the authorities began to remove objects from the wreck, out of the ten divers, one was accidentally killed, while two other divers became permanently disabled. Conditions had vastly improved when Cousteau visited the wreck in 1953, but by that time, the Greek government had long removed everything from the boat.
The rewards of the initial team’s work were marble and bronze statues, gold jewellery, amphorae and other artefacts, all dating from the first century BC, when the ship was believed to have sunk, on what is believed to have been a delivery from Rhodes to Rome. In early 1902, Valerio Stais began sorting through the recovered material, all donated to the Museum of Athens. On May 17, 1902, Stais noticed a calcified lump of bronze that did not fit anywhere, and which looked like a big watch. He guessed it was an astronomical clock and wrote a paper on the artefact. But when it was published, he was labelled ridiculed for even daring to suggest such a thing.
His critics argued that sundials were used to tell the time. A dial mechanism was unknown, even though it was described on what must thus have been a purely theoretical basis. The status quo was that “many of the Greek scientific devices known to us from written descriptions show much mathematical ingenuity, but in all cases the purely mechanical part of the design seems relatively crude. Gearing was clearly known to the Greeks, but it was used only in relatively simple applications.”
So they could do it, but they did not do it. So: had Stais rightfully identified what some called the “most complicated piece of scientific machinery known from antiquity”, or was it too good to be true? The future would tell, but it was for the moment definitely too good to be believed.

In 1958, Yale science historian Derek J. de Solla Price stumbled upon the object and decided to make it the subject of a scientific study, which was published the following year in Scientific American. Part of the problem, he felt, was its uniqueness. De Solla stated: “Nothing like this instrument is preserved elsewhere. Nothing comparable to it is known from any ancient scientific text or literary allusion. On the contrary, from all that we know of science and technology in the Hellenistic Age we should have felt that such a device could not exist.” He likened the discovery to finding a jet plan in Tutankhamen’s tomb and at first believed the machine was made in 1575 – a date of the first century BC remained hard to accept – let alone defend.
Still, Price must have realised that whereas its age was a dangerous subject to discuss, it was safe to explore the mechanism and function of the instrument. He thus concluded that the object was a box with dials on the outside and a series of gear wheels inside.
At least 20 gear wheels were preserved, including a sophisticated assembly of gears that were mounted eccentrically on a turntable. The device also contained a differential gear, permitting two shafts to rotate at different speeds. Doors were hinged to the box to protect the dials inside. As to its purpose: the mechanism appeared to have been a device for calculating the motions of stars and planets: a working model of the solar system.
This was not just speculation on his part. Price noted that the front dial was just clean enough to read its function: “It has two scales, one of which is fixed and displays the names of the signs of the zodiac; the other is on a movable slip ring and shows the months of the year. Both scales are carefully marked off in degrees. […] Clearly this dial showed the annual motion of the sun in the zodiac. By means of key letters inscribed on the zodiac scale, corresponding to other letters on the parapegma calendar plate, it also showed the main risings and settings of bright stars and constellations throughout the year.”

Price knew he had merely postponed the inevitable and would have to tackle its age. Evidence of its ancient origin could be found in the device itself: the Greek inscriptions. Price was helped in this work by George Stamires, a Greek epigrapher. To quote Price: “Some of the plates were marked with barely recognisable inscriptions, written in Greek characters of the first century BC, and just enough could be made of the sense to tell that the subject matter was undoubtedly astronomical.” There was no way back and scientists could only pretend the device and Price’s analysis did not exist – or accept the undeniable truth: it was ancient, it was Greek… embedded belief systems of what the ancients were, could and did would have to be adjusted.
There was also circumstantial evidence, which created a historical framework into which the device fit nicely: similar mechanisms were described by Cicero and Ovid. Cicero, writing in the first century BC – the right timeframe –, mentioned an instrument “recently constructed by our friend Poseidonius, which at each revolution reproduces the same motions of the sun, the moon and the five planets.” He also wrote of a similar mechanism that was said to have been built by Archimedes and which was purportedly stolen in 212 BC by the Roman general Marcellus when Archimedes was killed in the sacking of the Sicilian city of Syracuse. The device was kept as an heirloom in Marcellus' family.

Despite these literary references, scientists were doubtful and Price summed up their thinking: “Even the most complex mechanical devices described by the ancient writers Hero of Alexandria and Vitruvius contained only simple gearing. For example, the taximeter used by the Greeks to measure the distance travelled by the wheels of a carriage employed only pairs of gears (or gears and worms) to achieve the necessary ratio of movement. It could be argued that if the Greeks knew the principle of gearing, they should have had no difficulty in constructing mechanisms as complex as epicyclic gears.”
Still, it was clear that someone had obviously applied the theory and had come up with a practical tool. But who had created the machine? The likely suspect may have been the Greek astronomer, mathematician and philosopher Geminus, a student or late follower of Poseidonius. The latter, of course, was the one whom Cicerco credited with inventing exactly what the device was.
Geminus was a Stoic, from a school founded by Zeno, and lived from 135 to 51 BC, teaching on Rhodes. Rhodes was the centre of astronomical research. Geminus himself not only is known to have defended the Stoic view of the universe, but in particular to defend mathematics from attacks by Sceptic and Epicurean philosophers. The Antikythera device would have been right up his street, as it combined astronomy and proved the powers and the excellence to which applied mathematics could excel: science and mathematics could mimic the motions of the universe.
Most importantly, he lived in the right timeframe. Furthermore, the date for which this calculator was set was the year 86 BC, which some researchers have argued can be seen by the positions of the dials and pointers. 86 BC was an important astronomical year, as five conjunctions of planets in four zodiacal signs occurred that year, an ideal time to set an astronomical calendar. This date has also influenced the dating of the ship wreck, as many believe it will not have been much later – as otherwise the clock would have been reset to an astronomical event at a later date. Many thus argue for a date of 83-81 BC, though others posit dates such as 71 BC, adding that there is no guarantee the device was not idle for a number of years before being transported to Rome.

All of this understanding is intriguing, but for one researcher, Maurice Chatelain, one important ingredient was missing: logic. Chatelain argued that “if someone wants to construct an astronomical calculator by using intermeshing gears, the first condition is to find the number of cycles necessary to obtain an exact number of whole days. Some of these cycles are easily found but many are nearly impossible.”
Each gear is a cycle; this is how any mechanical clock works: seconds turn to minutes, to hours, and in some clocks to days, if not larger cycles. To make such clocks work, not only the cycles need to be known, but also the ratios between the cycles: how seconds relate to minutes (60:1), minutes to hours (60:1), hours to days (24:1), etc. It is difficult enough to construct such a device for the solar year, but the Antikythera device also incorporated the cycles of the moon and five of the nearest planets. No wonder scientists were sceptical that the device was… a device.
To make the system work, the system would have to be based on days, and thus the cycles would be expressed in full, whole days, with the ratios between the various cycles based upon the day counts of the cycles too. The genius that created the artefact would thus have to be aware of the cycles of the heavenly bodies. This in itself was within the remit of the Greek scientific community – and many generations and civilisations older than that. But a key question was what system was used, as each country had its own. The Greeks used the so-called Metonic cycle of 19 tropical years, but this, Chatelain felt, had no real value in creating a gear calculator.

According to Chatelain, only the Egyptian calendar system is suited for being used as a calculator – and he also found it was the one at the basis of the Antikythera machine: “The seemingly complicated Egyptian calendar, based on Sirius, the Sun, and also the Moon, actually works like a charm. Every four years represents exactly 1,461 days which in turn represent 49.474 synodical moon months. This last number has to be multiplied only 19 times to give a number of whole days – 27,759 – equal to 940 months, or 76 Sothic years, which is the cycle of the Rhodes calculator!”
Still, some do not share Chatelain’s enthusiasm for an Egyptian origin. One inscription on the device itself significantly reads “76 years, 19 years”. This refers to the Calippic cycle of 76 years, which is four times the Metonic cycle of 19 years, or 235 synodic (lunar) months. The next line includes the number “223”, which refers to the eclipse cycle of 223 lunar months. Price himself reasoned that “using the [Metonic] cycles, one could easily design gearing that would operate from one dial having a wheel that revolved annually, and turn by this gearing a series of other wheels which would move pointers indicating the sidereal, synodic and draconitic months. Similar cycles were known for the planetary phenomena; in fact, this type of arithmetical theory is the central theme of Seleucid Babylonian astronomy, which was transmitted to the Hellenistic world in the last few centuries BC.” Though it was quite clear that all of this knowledge was not Greek in origin, the question remained whether it was Babylonian or Egyptian.

Price had injected a new life fluid into the device and major breakthroughs occurred in the last decade of the 20th century. With the arrival of powerful computers, those machines were used to reminisce about what many considered to be the oldest computer – and the latest generation was used to shed light on what some considered to be the “Adam” of the line.
First, a partial reconstruction was built by Australian computer scientist Allan George Bromley (1947–2002) of the University of Sydney, working together with the Sydney clockmaker Frank Percival. This project led Bromley to review Price's X-ray analysis made in 1973 and to make new, more accurate X-ray images that were studied by Bromley's student, Bernard Gardner, in 1993.
Later, John Gleave constructed a working replica of the mechanism. According to his reconstruction, the front dial shows the annual progress of the sun and moon through the zodiac… against the Egyptian calendar. But, as if to remain neutral in the Egyptian or Greek debate, he stated that the upper rear dial displays a four-year period and has associated dials showing the Metonic cycle of 235 synodic months (19 solar years). The lower rear dial plots the cycle of a single synodic month, with a secondary dial showing the lunar year of 12 synodic months.
Another reconstruction was made in 2002 by Michael Wright, mechanical engineering curator for the Science Museum in London, working with the above mentioned Allan Bromley. On November 30, 2006, the journal Nature published an article on Wright’s and his team’s analysis of the Antikythera device. It confirmed that the instrument had been used to predict solar and lunar eclipses. The article credited Derek Solla Price, but equally stated that “although Solla Price's work did much to push forward the state of knowledge about the device's functions, his interpretation of the mechanics is now largely dismissed.”
The new analysis confirmed that the major structure had a single, centrally placed dial on the front plate that showed the Greek zodiac and an Egyptian calendar on concentric scales. On the back, two further dials displayed information about the timing of lunar cycles and eclipse patterns. Previously, the idea that the mechanism could predict eclipses had only been a hypothesis. The study also revealed some of the complexity of the engineering that had gone into this device. The Moon sometimes moves slightly faster in the sky than at others because of the satellite's elliptic orbit. To overcome this, the designer of the calculator used a "pin-and-slot" mechanism to connect two gear-wheels that introduced the necessary variations.
The team was also able to decipher more of the text on the mechanism, doubling the amount of text that can now be read. Some of the inscriptions mention the word “Venus” and “stationary”, suggesting that the tool could look at retrogressions of planets.
Wright also believes the device was not a one-off. "The designer and maker of the device knew what they wanted to achieve and they did it expertly; they made no mistakes. To do this, it can't have been very far from their every day stock work." So it was probably “mass produced” at the time and must have been the product of previous, less fancy clocks. That those earlier models have been lost in the mists of time is understandable, but the big question by which everyone is baffled, is why such clocks did not continue to be build in the centuries that followed… indeed, why it took more than a millennium before a clock of the same technological expertise appeared again.

Derek Price © Jeffrey Price

Despite acceptance that this is a 1st century BC planetarium, some questions remain. Price pointed out that he himself did not know whether it was operated manually, by turning, or automatically. He said: “I feel it is more likely that it was permanently mounted, perhaps set in a statue, and displayed as an exhibition piece. In that case it might well have been turned by the power from a water clock or some other device. Perhaps it is just such a wondrous device that was mounted inside the famous Tower of Winds in Athens. It is certainly very similar to the great astronomical cathedral clocks that were built all over Europe during the Renaissance.” – 1500 years later. Wright’s team argue that it was manually operated, but this would somewhat work against a mass produced item, for it would require the most work from those people buying it; care for the device would be labour intensive. So perhaps Price’s hypothesis that it was to be used within a religious setting is more appealing – though every hypothesis is currently guesswork.

The discovery of the Antikythera Device led to one gigantic realisation: that our everyday clock started as an astronomical showpiece that happened also to indicate the time – and not vice versa, as most believed half a century ago. Gradually, the timekeeping functions of the clocks became more important and the device that showed the cycles of heaven became subsidiary – only to be forgotten, and then reinvented all over again – all wheels inclusive.
Today, the device is worshipped by many as it is seen as the first calculator – computer. Price labelled the Antikythera Device “in a way, the venerable progenitor of all our present plethora of scientific hardware.” It should not come as a surprise then that whereas the original mechanism is displayed in the Bronze collection of the National Archaeological Museum in Athens, accompanied by a replica, another replica is on display at the American Computer Museum in Bozeman, Montana. In substance, it is bronze; intellectually, it is a computer.