The fascinating story of the Antikythera Mechanism
“In around 60 BC, a large merchant ship, approximately 40 metres in length, was wrecked off the northeastern coast of the tiny island of Aegila, located between Crete and the Peloponnese. This ship was loaded with the usual amphorae of wine and other goods, but also held bronze and marble statues, glassware, and other valuables. The bronze statues were old, some crafted over a century prior. Among the passengers, we know there was at least one woman, as evidenced by the two pairs of elegant golden earrings discovered.”
Alexander Jones, a historian specialising in ancient Greek astronomy and a member of the Antikythera Mechanism Study Group since 2006, details the extraordinary discovery of the wreck and the Antikythera Mechanism it contained in his book, A Portable World – Bringing to Light the Antikythera Mechanism, a Scientific Wonder of the Ancient World. The book delves into the intricate workings of the Mechanism, often regarded as the oldest analogue computer, how it was constructed and its purpose.
The fragments of the Mechanism were discovered in 1902, towards the end of the salvage operations. They were clearly cogwheels, but no one could associate them with any mechanism or conceive that they might be related to the inscriptions found. The primary body of the Mechanism was recovered from the sea as an aggregate, with the mechanical components concealed within.
“When the wreck was initially discovered in 1900-1901 and its remains were partially salvaged, it was believed that the sculptures retrieved from the seabed were significantly older than the ship that transported them,” Jones writes in his book. “This led to the popular theory that the ship was carrying treasures looted by the Romans from a Greek city in the early 1st century BC. While this theory still has its proponents, closer examination of the salvaged items suggests that it was more likely a merchant ship carrying goods from various origins, possibly bound for different destinations.”
The exact origin of the objects carried by the ship cannot be definitively established. However, experts can make educated guesses based on the types of amphorae found: they could have been crafted in Rhodes, on Kos, near Ephesus in Asia Minor, or possibly even along the Italian coast of the Adriatic.
The marble statues were crafted from Parian marble, suggesting their origin from an Aegean workshop, possibly Delos or Pergamum. The glass vessels originated from workshops in Syria, Palestine, or Egypt. One of the passengers aboard the ship had in his possession 32 silver coins from Pergamum and Ephesus, his life savings. Another passenger carried smaller copper coins from Ephesus, along with older ones from Catania in Sicily and Knidos in Asia Minor. The most recent of the silver coins were minted between 76 and 67 BC, which means the ship must have sunk after 76 BC, perhaps a decade or two later.
The ship’s exact route remains unknown. It’s not necessary that it stopped at all the aforementioned locations during its final voyage. It’s more plausible that parts of the cargo were transported by smaller ships to a major transit port like Delos or one on the coast of Asia Minor. These ports were large enough to accommodate ships of a similar size to the wrecked vessel. The wreck’s location suggests that the ship was en route to the western Mediterranean, likely intending to deliver its cargo to a port on the Adriatic or even further west.”
Among the items carried by the ship, one stood out as the most enigmatic – a shoebox-sized and shaped box, comprised of wood and metal. Initially, it didn’t garner much interest upon retrieval from the seabed, overshadowed by the more visually striking statues and glassware. However, this rusty box, despite being severely damaged by sea water and time, was a delicate and valuable structure. It must have been securely packed in a small box or carton to protect it from damage and the elements. It’s also unlikely that such a valuable object would have been sent to its destination without an escort. We can speculate that a skilled craftsman may have accompanied this mechanical object throughout its journey, from the workshop where it was crafted to its intended recipient.
The Discovery of the Wreck and the Recovery of the Antiquities
The wreck was discovered in 1900, thanks to adverse weather conditions that brought Symi sponge divers to the location. These divers, who had been diving since 1870, utilized a dependent diving system that enabled them to reach greater depths. The bronze scuba gear and full-body waterproof suits they wore allowed them to take air from the surface via a flexible hose, thereby extending their underwater time. The wreck was discovered by a group of these divers who happened to be at the site of the sunken ship when they sought shelter in Antikythera from the weather. The two vessels that comprised the team were the 15-meter-long “mother” ship, Efterpi, which transported supplies and the sponges they had harvested, and a diving boat, Kalliopi. Both were owned by Photios Lindiacos and captained by his brother-in-law, Dimitris Kontos. In addition to the crews of the two boats, Kontos was accompanied by eight divers. The discovery was made by diver Elias Stadiatis, who spotted fragments of bronze statues at a depth exceeding 60 meters. A life-sized bronze arm was retrieved from the seabed, later authenticated as part of a statue now known as the “philosopher”. Short and his team handed the arm over to the authorities and negotiated an agreement with the government to be the ones to explore the wreck.
Spyridon Stais, the then Minister of Education and a former professor with scientific training, consulted state agency archaeologists. They quickly deduced that the remains of a Greco-Roman shipwreck had been found. Without delay, Stais began organizing the operation. He offered the divers a generous reward if they could successfully retrieve objects from the wreck site and secured a loan of a transport ship from the Navy. The divers were then transported back to the wreck site on the shipwrecked vessel Mykalis, accompanied by Professor Antonios Economou of the University of Athens’ Archaeology Department, who was to supervise their work.
Kontos only disclosed the exact location of the wreck as the expedition set sail. Windy conditions and rough seas limited the divers to working no more than three hours a day, and each dive to the depth of the ship could not exceed five minutes. Despite these challenging conditions, the initial dives yielded impressive results, including several fragments of bronze and marble statues. The most notable of these was a bronze head initially believed to depict a boxer, but after cleaning, it was revealed to be an unidentified philosopher. Work continued for several months but was frequently interrupted due to accusations in the Greek press that the divers were recklessly handling the artifacts or intentionally damaging them. Representatives of the Ministry dismissed these claims, asserting that the divers’ sole task was to retrieve the finds under the supervision of the Ministry and its archaeologists. No systematic records of the recovered objects seem to have been maintained. Interestingly, in 1960, archaeologist Peter Throckmorton was informed by descendants of the Kontos crew in Symi that the first dive had recovered some bronze statuettes, which were later sold to dealers in Alexandria.
For the average observer, the most awe-inspiring discoveries were the bronze statues.
Despite being shattered, the fragments were well-preserved and some appeared to date back to the late antiquity or early Hellenistic periods, specifically the 4th and 3rd centuries BC. Bronze statues from antiquity are rare, as the metal was often repurposed when the statues fell out of favor. Thus, the bronze fragments from the Antikythera Shipwreck significantly contributed to our understanding of the era. Many of the statues were recovered in the initial weeks of the operation, due to their high interest and accessibility for divers. The largest fragments of the most famous bronze statue from the wreck, the ‘Antikythera Adolescent’, were recovered in December 1900, and it was the only statue for which nearly all pieces were eventually found.
The sheer quantity of marble statue fragments, representing humans, gods, and larger-than-life horses, made the divers’ work laborious and time-consuming. Some fragments were located at depths too great to reach. The statues’ appearances were disappointing, with badly eroded surfaces, and it quickly became clear to archaeologists that these were not original masterpieces from the classical period, but rather late Hellenistic copies and imitations. The final notable sculpture recovered from the seabed was a beautiful bronze statuette of a boy, found in late June. A month later, under the order of Stai, the operation was halted as the most recent findings (fragments of pottery and wood) held no archaeological value.
The Discovery of the Antikythera Mechanism
.
From the outset of the recovery, newspapers reported the discovery of an “inscribed plate”, but offered no further details. The fragments of the Mechanism were found in 1902, towards the end of the salvage operation. They were clearly toothed wheels, but no one could link them to any mechanism or imagine they might relate to the previously found inscription. The main body of the Mechanism emerged from the sea as a conglomerate, with the mechanical parts concealed within.
When the salvage operation ended in late July, Stais loaded the recovered antiquities onto the Mycale for transport to Athens. It was then that an incident occurred, which only became public knowledge many years later and thus cannot be verified. According to Admiral Ioannis Theofanidis, a fellow Royal Navy officer, Pericles Rediadis, who was onboard the Mykaly during several visits to the wreck site, stopped a crew member from discarding a seemingly worthless calcified mass – the Mechanism – back into the sea. He noticed a piece of metal protruding from a crack in its surface. Redias never mentions it in his writings, but he is unofficially credited with the preservation of the Mechanism.
Upon their arrival in Athens, some artifacts were temporarily stored and put on display at the Ministry of Education. However, they were soon relocated to the Archaeological Museum. The caretakers were faced with the enormous and challenging task of cleaning, preserving, and in some instances, reconstructing and restoring them. Among the recovered treasures, the Teenager was regarded as the most exquisite and intact piece. While plans were being made for its restoration, assigning the legendary French conservator – and forger – Alfred André, to preserve and rebuild it, Spyridon Stais, a former minister, visited the museum out of personal interest in the progress of the Antikythera artifacts. Museum staff allowed him into the room where the bronze fragments were stored. There, those working on the reconstruction of Ephibos had sorted out fragments that looked like parts of a human torso or resembled pieces of a harness, setting the rest aside. Stais noticed three fragments, laid side by side, among the set-aside pieces, speculating that they might be parts of the same slab-like object. One of these fragments bore a Greek inscription that was difficult to decipher in the dim afternoon light, while the other had a system of toothed wheels visible on its surface. Although he was not the first to see them, he was the first to notice the mechanical parts of the mechanism and ensure that the fragments would receive the attention of archaeologists.
Generally, the contemporary history of the Antikythera Mechanism can be split into two periods. The first, from 1902 to the 1960s, involved studying the remains through direct observation or from photographs. The second period, from the 1970s to the early twenty-first century, saw scholars using techniques that allowed them to see through or inside the fragments. From this point forward, any advancement in our understanding of the Mechanism fragments was tied to the development of new imaging techniques.
In the initial period, significant progress was made due to changes in the objects themselves. The efforts of skilled conservators in 1905 and 1953, coupled with some accidental damage in between these years, unveiled aspects that had previously been hidden behind other parts of the fragments or buried under layers of sediment. Unfortunately, some information was lost during this process, and the older photographs and descriptions are the only resources we have to reconstruct the earlier state of the fragments.
This groundbreaking engineering project was one of the earliest portable devices designed to aid in teaching. The incorporated know-how is among the most advanced that modern scholars have discovered from classical history. The insights it offers about Greco-Roman astronomy, scientific technology, and their role in ancient Greek society are truly remarkable.
Deciphering Fragment G of the Mechanism
.
Agamemnon Tselikas, a philologist-paleographer, who deciphered the most significant fragment of the Mechanism, Fragment G, spoke exclusively to LiFO about the meticulous process of decoding it:
“Fragment G, as it has been traditionally named by the late curator of the Athens Archaeological Museum’s bronze collection, Maria Zafeiropoulou, is a key component in the study of the mechanism. It is the part that is depicted in the mechanism’s models as the ‘front cover’. Thanks to the application of new photographic technologies (for 2006-2008), we believe that we have mostly read those parts of its surface that still bear traces of writing. However, this does not mean that the identification of letters or the reading of words is always clear and definitive. There are several points of ambiguity, especially in the fragment sections and where there are overlaps with other elements.
More than 2,000 photographic images, along with CT scan sections from three different angles, were used for the reading. The surviving section contained, based on the available photographic material, thirty-seven stanzas and semi-stanzas with a maximum of fifty-seven letters per stanza. If we were to confine ourselves to the theoretical square in which the surviving fragment is inscribed, our inscription would contain at least 2,052 letters, although in reality this number could be much higher.”
The Initial Attempt at Decoding
“During the transcription process, a grid was created with horizontal and vertical columns, each cell designed to contain a letter. Then, based on the coordinates formed in the photographic shots, I began to place the identified letters in their correct positions. To accommodate our inscriptions to the irregularities of the inscription surface, I only inserted a placeholder for a space between the letters or words in a few places, by convention. The sizes of the distances in the tomographic images, depending on the angles of acquisition, appeared vast and gave the illusion of vacant spaces. For letters that were not clearly legible, yet I assumed to be correct, I temporarily affixed a question mark next to them. For letters assumed to be in worn areas of the surface, I enclosed them in square brackets. Lastly, for spots on the surviving surface where no letters are visible, I marked them with a dot inside square brackets.
.
In total, 1,380 letters were deciphered from the fragment, forming 156 complete words, excluding articles, conjunctions, prepositions, and numerical symbols. This makes up 81.33% of the surviving surface. The maximum dimensions of the surviving surface are 11 cm wide by 9 cm high, equating to 99 square centimetres. Based on these measurements, each centimetre of full verse contains 5.18 letters. Vertically, each centimetre contains 4 verses. And each square centimetre contains approximately 21 letters, meaning that each letter occupies an area of just under 2 square millimetres. It’s important to note that for each millimetre of physical surface, there were 10 horizontal and vertical sections. Thus, for every two millimetres of each letter’s area, there were 20 sections, and 400 for a square.
Multi-letter antennas end in a vertical notch of infinitesimal length under the serif type of modern lettering. No writing guide engraving has been used. An initial observation is that all the letters have the same style and form, although not always the same size. This is justifiable given the purpose and functionality of each inscription on the various parts of the mechanism.
The Inscription’s Message and Assumptions About the Mechanism’s Use
.
With regard to meaning, it can be said that this inscription, like others, must be studied and interpreted in relation to the others to reveal their essential and functional relationship. However, a problem arises due to their fragmentary nature, especially the inscription of Fragment G that we are examining, where neither the beginning nor the end of the verses is preserved. It appears, nonetheless, that we do not have a continuous text but rather instructions, suggestions, or statements in a list-like order.
It’s evident that the inscription is considered a manual of suggestions or observations about planetary phenomena, based on the data of that time. It relates to the movements of the five planets and other constellations in relation to the sun and each other, always with the Earth as the basis of observation, but not as a point of reference.
The nature of the numerous axial sections and the constant shift in the visual image of the surface create an impression of a perpetually moving image. Letters morph in shape and substance every tenth of a millimetre. Similar letters like A, D, L, O, TH, G, E, H, P necessitated the creation of many versions for identification in relation to the preceding and following letters, all contributing to the formation of a feasible word. In total, as previously mentioned, 156 words were recorded – not including articles, prepositions, and conjunctions – amounting to 1,380 letter characters. Given the fragmentary nature of the Mechanism, deriving a specific meaning was challenging without the controlled use of imagination or invention. If we consider one of the potential uses of the Mechanism, beyond its calendar of eclipses, national games, and the sidebar indicating the sunrises and sunsets of key constellations, we cannot dismiss the highly prevalent prognostic horoscope. This is based solely on the distances and approaches of planets to each other or their unique appearances, as St. Gregory of Nyssa describes and critiques in his discourse on “Eimarmenis”.
Simultaneously, a parallel reference to existing literary sources was deemed necessary for the validation and acceptance of any lexical selection. What strikes me as extremely notable is the prominence of both the user manual and other inscriptions on the Mechanism. The manufacturer did not hide the usage instructions for personal retention, but made them openly available to all. This, I believe, represents the ultimate manifestation of the humanistic and liberating Greek spirit that advocates for the dissemination of knowledge throughout the world, rather than its concealment.
In this manner, I have presented a text derived from a literary treatment, supplemented with parallel passages from earlier and later astronomical texts. I leave any potential commentary, observation or revision to specialized historians of astronomy and mathematicians, fully aware that I was navigating unstable ground. This interpretation is published in the volume of the general presentation of the findings from the Shipwreck that formed the corresponding exhibition at the National Archaeological Museum (2012).
The initial impressions were indeed stirring: a comprehensive manual-guide to the interpretation of planetary motion was plainly visible to anyone who held this astronomical instrument. Its owner would read about the movements between Mars and Venus, for instance, about how brightly Venus glows in a clear atmosphere without a moon to cast a shadow, and about the relative approaches and distances of planets to the sun.
Many significant astronomical terms are emphasized, including both conventional and unique ones such as apokatastasis, apostasis, reverses, dodecadimorion, epaegei, epochon, epetemni, eoa, backward, stasis, props, conjunction, turn, coincide, addition, ascension, planetary order, and names of planets like Mars, Mercury, Venus.
Here are some representative phrases:
Mars returning, minor restoration, minor support is lurking, Venus offering to the sun, Mercury offering to the sun as X is attacking, falling short until the eos, returning from the sun, bringing Mercury upon the sun.
Some verbs are in the second person, such as:
Bring Mercury upon the sun, extend the solar distance.
This suggests that the overall structure of the text was akin to a user’s manual. Similar terms and phrases, even word-for-word, can be found in works of earlier and later ancient writers related to the Mechanism. For instance, Theon of Alexandria, Aristotle, Alexander Aphrodisias, Siblicus, Claudius Ptolemy’s “Mathematical Composition”, Geminus, Proclus, Eudoxus, John Philopon, and Hephaestion.
In reference to Claudius Ptolemy, it’s worth noting that in the exceptional Greek codex of the Marcian Library of Venice 388, titled “Geographical Narrative”, Ptolemy is depicted in western medieval attire (actually representing Bessarion) in a study filled with astronomical instruments. Among these, on a casket, are two instruments that don’t resemble any known ones and can’t be identified with any other. They certainly don’t resemble the supposed form of the “Mechanism,” but they seem to be a kind of portable planetarium with the zodiac, upon closer inspection. Here, I quote an epigram dedicated to Ptolemy by the Byzantine scholar Maximos Planoudis (late 13th century), translated by myself:
I know that I was born mortal and ephemeral, but when
I examine the dense and twisting paths of the stars,
I no longer touch the earth with my feet, but close to Jupiter himself, I feast on the food of the gods, the ambrosia.
From this observation, a discussion began on whether the front door, i.e., Fragment G, had rotating markers corresponding to the planets or a design of concentric circles, where each circumference corresponded to a planet, including the sun. This matter remains unresolved, both mechanically and figuratively. However, the manuscript tradition of antiquity suggests that it is most likely that there were drawn concentric circles. These can be found in several Greek and Western medieval manuscripts.
A compelling example is provided by a medical codex from 1822, the so-called ‘Yatrosophy of the Humble Chian Pantoleon’.
A recently published Greek translation of Alexander Jones’ engaging book “A Portable World” offers a comprehensive examination of the Mechanism. Drawing on previous interpretations and readings of the inscriptions, Jones presents his own perspectives in many instances. He acknowledges that Fragment G contains many obscure and contested points, resulting in gaps in our complete understanding of the inscription. The astronomical interpretation of the inscription often leads to the invention or assumption of words that alter its meaning. Nonetheless, it is widely accepted that ongoing research, even with advanced and precise technical means and devoid of conjectural interference, could supplement, challenge, and propose many new interpretations of the Mechanism over the course of several years.”
