by Rob S. Rice
USNA Eleventh Naval History
Paper for Collected Volume
At some time around 80 B.C. a heavily
laden merchant ship sank to the bottom of the sea off the Southern
coast of Greece. After two millennia, materials from that vessels
cargo have combined with the work of several scholars to allow wider
speculation on the subject of seafaring in Greek and Roman
antiquity. The objective of this treatment of the chain of events
involved is to provide a useful survey of early and modern
underwater archaeology and the mechanics of artifact
preservation and interpretation as well as to offer conclusions
drawn from the data presented here concerning ancient celestial
navigation and the island of Rhodes. The united efforts of a
wealthy Roman, a frightened Greek sponge diver, an English
physicist, and an American naval historian have combined to allow
some further inquiry into civilian and military seafaring in the
first century before Christ.
Sailing further south past the island of Antikythera off the
southernmost coast of Greece offers an alternative to, as a very
ancient proverb says, "rounding Malea and forgetting home." Whether
he sought to avoid the pirates or the storms clustered around the
infamous cape, the skipper of what apparently was a good-sized Roman
merchant vessel of around 300 tons made a wrong decision. His ship
crashed into and sank off the island’s coastal cliffs, and what was
probably a wealthy Roman buyer eventually learned that his treasure
ship’s cargo had gone down in nearly two hundred feet of very cold,
"Treasure ship" is a legitimate label. The corbita had held
everything from original bronze life-size statues, to marble
reproductions of older works, jewelry, wine, other bronzes, and at
least one immensely-complicated scientific instrument. It was the
statues that frightened a Greek sponge diver named Elias
Stadiatos nearly out of his wits in 1900, when his captain
winched him back over the side, removed his helmet and breathing
hose, and found him babbling about a "heap of dead naked women."
Rumors from around that time show a resulting pattern of events all
too familiar to the modern underwater archaeologist. The local
divers had found the ship first. The villagers of Simi, near the
site, speak of many small bronze statues sold in Alexandria soon
after the wreck was found, and when later archaeologists surveyed
her, the vessel was missing all her heavy lead anchor stocks. The
ship was big enough to have had five anchors, in water too deep to
have used any of them, and divers needed lead weights to find their
sponges and rare black coral, just as they needed money to support
Still, Captain Kondos of the sponge vessel in 1901
told the Greek government of Stadiatos’s discovery, and agreed
to hire his ship and divers for the salvage. He pushed his equipment
and his men to the limit, but he recovered one of the most amazing
troves ever winched from the bottom of the sea. Statues, jewelry,
transport jars, utensils, and tableware of all kinds came to the
surface. "Huge boulders" obscuring the cargo and hauled up to the
salvaging vessel with difficulty turned out to be statues covered
with marine growth, their marble eaten away by the chemical action
of centuries of sea-water and animals. The divers suffered from all
the hazards of their trade, one fatally.
When the winter storms came up, the
divers and the Greek government were ready to quit. The bronze
statues went into galleries, the jewelry into display cases, and a
great deal of material went into museum storage, waiting for careful
analysis to determine the significance of, among other things,
clumps of marine growth and corrosion surrounding what looked like
some kind of gearing. What wood was brought up resembled wet
cardboard in more ways than one as it dried out and shriveled away.
It would be unfair to call this proto-excavation "unscientific," for
there were trained archaeologists from the Greek antiquities service
waiting to process the material once Kondos’s divers had
brought it to the surface. A modern excavation would, for all that,
hopefully progress a great deal differently, using techniques
pioneered by Peter Throckmorton and George Bass over
the course of research beginning in 1959. Archaeologists
themselves would descend to investigate the wreck. The hoses and
helmets that had hampered the sponge divers of 1901 would be
replaced by self-contained apparatus designed to bleed off the
carbon dioxide that had exhausted and dazed the original divers.
Modern compressors would be filling air tanks and pumping air down
to the wreck level, and that air rising up again inside a tube would
lift silt and small items up to the surface for sifting and removal.
Inside plastic bags rising bubbles would lift statues and jars.
A decompression chamber would stand
ready in the event of nitrogen narcosis, with atmospheric pressure
within carefully regulated to let the nitrogen built up by the
compressed air breathed underwater leave the diver’s arteries slowly
enough to avoid damage. A grid over the wreck made of plastic
plumbing pipe would direct drawings and photography for
stratigraphic records of the objects discovered. Drawings and
recorded measurements would possibly be combined with stereoscopic
photography, the whole allowing graphic reconstruction of the
original ship and its cargo. There
might be a diving bell with a telephone to talk to the surface, or a
midget submarine to help with the photography. An underwater metal
detector would be useful, and an "air probe" to jet into the sea
bottom with compressed air to prod for things under the mud.
Computers would store information topside, and potentially
underwater as well, since one of the things that suffers with
exposure to water is a diver’s short-term memory.
Funding, as well as the physical difficulties of such intricate
underwater activity can act to limit such exploitation of first-hand
ancient material. The additional hazard of post-recovery destruction
of recovered material is not always countered by techniques of
modern artifact conservation. Shifting during the descent of the
original ship’s hull to bottom had already inflicted extensive
damage on her cargo before the first diver approached the wreckage.
The ubiquitous Mediterranean teredo worm employed the intervening
centuries to destroy the integrity of the hull and larger wooden
artifacts, while marine bacteria left only the hollow cell walls of
the remaining timber. Marine shellfish devoured the limestone of the
statues, while the sea’s own electrolytic bath wrought havoc on all
metallic artifacts unprotected by bottom mud. Unauthorized
"pot-hunting" before the official excavation undoubtedly also
further damaged the available material left behind.
The bronze gearing retrieved from the Antikythera wreck, with
its own chemical and animal accretions, broke into several pieces
soon after its return to the surface. The ship’s wooden planks and
what appears to have been a case for the mechanism shriveled soon
after retrieval. The marble statues were eaten away and disfigured
wherever they had been exposed to the sea. As usual in terrestrial
archaeological sites around the Mediterranean, ceramic material in
some form survived, except for the damage inflicted by the heavier
cargo and defacement by marine growths. The chemical composition of
the glassware retrieved in 1901 was fortunate. Phoenician beads
George Bass recovered off Cape Gelidonya exploded into
dust once they began to dry.
Modern conservators would place everything but the pottery into a
tank of fresh water until preliminary analysis was possible. Marine
conservators are a rare combination of archaeologists and chemists,
employed on occasion, and on occasion, in vain. The wood can be
preserved, as was the Swedish 17th-century galleon
Vasa, in polyethylene glycol, which fills the empty cell walls
with a waxy material over a great deal of time. Metal artifacts
receive their own immersion in chemical solutions with the goal of
stabilizing each piece and hopefully removing accumulated corrosion,
an expensive and not always successful procedure. Cleaning off what
lived and died on all materials submerged for any length of time can
be difficult as well, particularly when the person so doing is
uncertain of what lies under the accreted material and how much
cleaning the object can withstand before disintegrating or losing
In the case of the Antikythera fragments, the four large
pieces and a box of much smaller fragments were momentarily
overshadowed by the staggering other results of the first directed
retrieval of archaeological evidence from the sea. The original
excavators had their hands full reassembling the bronze statues,
sorting and identifying coins, and cataloguing all the items for
museum storage at Athens. Eventually, other scholars found time to
consider the fragments of original artifact. The initial belief
was that the bronze object was an astrolabe--a type of
navigational instrument first attested in 625 A.D. Correctly, one
Konstantin Rados in the earliest debate insisted that what was
visible on the lump’s surface was too complicated for such a device,
intricate as in fact were some medieval examples. At the same time
other scholars argued that the Greek artisans who had fabricated the
wreck’s statues could not have built even an astrolabe.
In 1951, a British physicist and historian of science named Derek
De Solla Price went to the Athens Museum for his own
analysis of the fragments taken from the Antikythera wreck.
Price himself was familiar with construction of medieval astrolabes,
and the complexity of the device and the astronomical inscriptions
visible on the surface led him to eight years of informed study. In
1959 Price published his own conclusion that the fragments
represented some form of intricate clockwork.
The idea was sufficiently unthinkable to the experts of the time for
one professor to claim in responding that someone in the Middle Ages
had dropped a machine of that era into the sea coincidentally over
the same current-swept spot off Antikythera’s rocky coast.
Price remained undiscouraged and maintained his conclusions.
In 1971 the Oak Ridge national laboratory published an article on
the use of high-energy gamma radiation to examine the interiors of
metallic objects. Price soon secured the assistance of the
Greek Atomic Energy Commission in shooting gamma rays into the
clumps of corroded bronze. He was able to produce photographic
plates that not only allowed him to reconstruct the device but to
ascertain its date of construction.
The Antikythera mechanism was an arrangement of calibrated
differential gears inscribed and configured to produce solar and
lunar positions in synchronization with the calendar year. By
rotating a shaft protruding from its now-disintegrated wooden case,
its owner could read on its front and back dials the progressions of
the lunar and synodic months over four-year cycles. He
could predict the movement of heavenly bodies regardless of his
local government’s erratic calendar.
From the accumulated inscriptions and the position of the gears and
year-ring, Price deduced that the device was linked closely
to Geminus of Rhodes, and had been built on that island off
the southern coast of Asia Minor circa 87 B.C. Besides the
inscriptions’ near-identity to Geminus’s surviving book, the
presence of distinctive Rhodian amphorae among other items from the
wreck supported Price’s deduction and date once Virginia
Grace had re-examined the pottery recovered in 1901.
Price’s straightforward and viable analysis came despite a
host of ideas the device’s discovery should have dispelled. He was
too concerned with what was before his eyes to realize that
prevailing beliefs among historians of the period would lead others
to slight or ignore what physics and archaeology had combined to
discover. Price correctly noted that Rhodes was a center
for astronomical thought. He mentioned Poseidonius,
Cicero’s friend and teacher, who built a much more complicated
astronomical computer than the one recovered.
He was unaware of the widespread belief that continues to maintain
that Rhodes in the first century B.C. was little more than a fading
ghost of past glory, crippled economically by the competition of the
Roman free port of Delos after 166 B.C.
It is neither facile nor uninstructive to remark that the
Antikythera mechanism dropped and sank--twice. The second
submersion came after Price’s publication of Gears from
the Greeks in 1975. Since that time little attention has
been paid to our most exciting relic of advanced ancient technology.
It was in the course of research into the navy of Rhodes that the
mechanism first came to this author’s attention, and it was that
research and knowledge of extant flaws in earlier scholarship that
allows this assessment of the significance of the device and
Scholars before and after Price ignored and continue to
ignore the length of Rhodes’ enduring reputation among the ancients
themselves as a center for intricate military and naval technology.
Rhodes had resisted the largest and most advanced
weapons systems produced by the Macedonian warlord-inventor
Demetrius. In 305 "the Besieger" sent a siege tower nine stories
tall, pushed by two thousand men against the Rhodians’ walls. Rhodes
was a center for the construction and use of antiquity’s heaviest
and most intricate catapults. The historian Diodorus of
Sicily would record how Demetrius’s helepolis, or
city-taker, had to retreat from one of the most intense artillery
barrages of antiquity, burning from several direct hits with
The tradition of advanced technology
on Rhodes continues to appear for centuries in the surviving
historical records of the Hellenistic Age. Mithridates V of
Pontus fared no better than the Macedonian attacker in his own
onslaught of 88 B.C., in which he encountered what F.E. Winter
considers to be one of the most formidable protected catapult
batteries in antiquity. Polybius,
Strabo, and Aristides in later years attest to the
legendary speed and surpassing deadliness of the ships and weapons
built behind the wall of Rhodes neorion.
The pirates of the Mediterranean feared and fled before the war
fleet of a single small island, and the last of the Greek
democracies successfully warded off even Roman domination until 43
B.C. Years afterward, the finest
ships in the Mediterranean world could still be found in her
In the light of the ancient literary evidence and the physical
existence of the Antikythera mechanism, it is necessary for
scholars of the period to discard the idea that Rhodes and
her economy were ruined by the Roman actions concerning Delos.
An impoverished, decaying backwater could not have provided impetus
for such a mechanism, much less supported the minds that conceived
it. Among other advances, the apparatus found among Rhodian coins
and amphora contained a differential gearing system more complex to
design than to build, and its presence among original bronzes, gold
jewelry, and marble statues clearly attests to the buyer’s
recognition of its value. The Roman
Cicero reports that the general Marcellus prized an
orrery, or analog planetarium, of Archimedes’ more than any
other booty from captured Syracuse.
The Rhodians could apparently build similar devices for export to
such wealthy Roman buyers--including, possibly, Cicero, who
knew Rhodes well and was governor of a neighboring province shortly
before the ship was lost.
Further research into the island’s history reveals additional
nourishment for the speculation the Antikythera mechanism’s
existence prompts and should have prompted about Rhodes, ancient
technology, and our study of the past in general. On Rhodes,
Philo of Byzantium encountered and described the polybolos,
a "machine gun" catapult that could fire again and again without a
need to reload. Philo left a
detailed description of the gears that powered its chain drive and
that placed bolt after bolt into its firing slot. Philo and
scholars since have believed that the polybolos was useless
because the Rhodians had convinced him that it was close range only
and couldn’t traverse from side to side.
The perspective of a naval historian can provide a kind of warfare
where a fixed weapon at close range could be useful--in an era when
ships routinely rammed each other. Anyone could have wondered why
the Rhodians built and refined something so complicated if they had
no idea of using it. Again, they conceived and built the
Antikythera device, and someone else had thought enough of it to
send it overseas.
The proof the mechanism offers of Rhodes’ enduring technological
expertise poses a question the device also helps to answer: What
could have led to the construction of such an expensive and
intricate device? Certainly the mechanical expertise that built
the polybolos indicates the physical ability to build the
mechanism. But what inspired the intricate theories and substantial
body of astronomical knowledge that lay behind the mechanism? Rhodes
even in its supposed "glory days" was chiefly famous for the
abilities of its seafarers--and therein lies the answer.
Very little indeed, is known about ancient celestial navigation,
besides indisputable proof that it did, in fact, occur.
It is worth noting, however, that the man who invented trigonometry
and first scientifically catalogued the stars’ positions was
Hipparchus of Rhodes; that in more than one ancient system of
latitude and longitude the meridians crossed at Rhodes, and that a
man Strabo rated second only to Aristotle--Poseidonius--found
support for his travels and devices on the same island where
Geminus did his writings, and inspired or built the
There is a evidence for a clear tradition of scientific research on
Rhodes, just as there is an anecdote preserved in by the Roman
architectural authority Vitruvius concerning two engineers’
competition for a city stipend.
Geminus’s surviving book shows him making a determined effort to
bring the transmitted data of the Babylonian astronomers to
the attention of his Greek readers in the first century B.C. In the
preceding century Hipparchus had laid the groundwork for
Geminus’s efforts to "popularize" Babylonian astronomy by
working their surviving eclipse data into his own astronomical
writings. Modern scholars of scientific history have yet to pay
Hipparchus his due honor for his failure to construct a
planetary system of his own even as he catalogued the observable
Although he had used observed parallax
to make an extremely close estimate of the moon’s distance from the
earth, Hipparchus had the scientific honesty to state that
there was insufficient data in his time to understand the true
arrangement of the solar system.
The refusal of others to admit that hobbled scientific thought until
well after Galileo’s death. Geminus’s contemporary
Poseidonius did much more than build complicated astronomical
devices of his own. One of the journeys celebrated and preserved by
his friend and pupil Cicero took him beyond Gibraltar to the
Bay of Biscay, where he was the first to note the connection between
the tides and the moon phases Hipparchus had measured. He
also possessed the novel theory that all the world’s oceans formed a
single body of water.
Hipparchus, Geminus, Poseidonius--we must still
search out details of what may well have been an analogue to our own
and Britain’s naval observatory, in competition and parallel with
the state-funded research at Alexandria’s museum. The Rhodians’
immunity to the pirates of the Mediterranean continued long after
their supposed post-Delian decline. The island could not feed
itself, but the grain ships continued to arrive--possibly steering
by starlight through the deep sea while the frustrated pirates
hugged the coast. The Rhodian navy displayed in a long and
distinguished operational history an almost uncanny ability to
function and maintain unit cohesion at night. In 198 B.C. a Roman
fleet eluded a Syrian squadron sent to intercept it by what seems to
have been a difficult nocturnal cruise--shortly before two of its
Rhodian escorts openly made a night voyage to locate an arriving
In 88 B.C., directly before Price’s
date for the device’s construction, the Rhodian admiral Damagoras
set the world an unforgettable example of Rhodian courage and naval
expertise. After eluding a Pontic blockade of the city’s
harbor, Damagoras led a force four times the size of his own
on a day-long chase, pausing only before sunset to turn and sink two
of the larger enemy vessels and disable two more. With the rest of
the enemy fleet alert and positioned to intercept his return,
Damagoras kept his command integrated and functional for an
entire night on the high seas, and returned safely to blockaded
Rhodes in the morning.
The discovery of the Antikythera mechanism has much to offer
besides tantalizing hints concerning state-funded research and
technological expertise on Rhodes. The very existence of such a
complicated gear train should also prompt fundamental change in
the way the ancient sources are read. We have found the tracks
for the emperor Nero’s revolving ceiling, and the Tower of
the Winds still stands in Athens, its clock faces empty, but its
functioning success materially and textually preserved.
When Cicero, Ovid,
Plutarch and others speak of "celestial spheres" going
back to the time of Archimedes, and describe their use,
the Antikythera device’s very existence should prompt us to
something besides unthinking skepticism.
Perhaps we should take a look at the
device and believe a little more of what we have been told. Wooden
ships have been set on fire with sunlight,
and John Morrison’s efforts to reconstruct the trireme
demonstrate that the full complexities of ancient ship construction
continue to elude us. When all the implications of Price’s
discovery are understood and acted upon, it will then be
possible to say that we have begun to understand the Antikythera
"Suppose a traveller carried into
Scythia or Britain the orrery recently constructed by
our friend Poseidonius, which at each revolution
reproduces the same motions of the sun, the moon, and the five
planets that take place in the heavens every day and night,
would any single native doubt that this orrery was the work of a
With the evidence before our faces, do
we continue to believe that Rhodes declined, the ancients
were technologically inept, and that our sources can be
easily discarded? Or do we accept the existence of ancient
advanced technology, study its implications, and look for deeper
meaning in what we have difficulty understanding? Much has been
learned about ancient technology and ancient seafaring. With the
right set of mind and purpose, it is clearly possible to learn a
great deal more.
1. Peter Throckmorton, "The Road to
Gelidonya," in The Sea Remembers: Shipwrecks and Archaeology
from Homer’s Greece to the Rediscovery of the Titanic, ed. Peter
Throckmorton (New York: Smithmark Publishers, 1987), p. 20.
2. Throckmorton, pp. 14-16.
3. Throckmorton, p. 16.
4. Throckmorton, 16-18.
5. v. Throckmorton, p. 29 illus.
6. Throckmorton, p. 16; Victoria Jenssen, "Archaeology and
Conservation," in The Sea Remembers: Shipwrecks and Archaeology
from Homer’s Greece to the Rediscovery of the Titanic, ed. Peter
Throckmorton (New York: Smithmark Publishers, 1987), pp.
7. Jenssen, p. 102.
8. Jenssen, pp. 102-105.
9. Throckmorton, p. 18; Derek J. de Solla Price, Gears from the
Greeks : the Antikythera Mechanism: a Calendar Computer from ca.
80 B.C. (New York: Science History Publications, 1975), p. 10.
10. Derek J. de Solla Price, "An Ancient Greek Computer,"
Scientific American Vol. 200 No. 6 (June, 1959): 60-67, with
some detailed reconstructions of the device’s original
11. Price, Gears, p. 10.
12. Price, Gears, pp. 10-13, Throckmorton, pp. 18-20.
13. The Oxford Classical Dictionary, second edition, (OCD2)
entry s.v. "Calendar" only begins to describe pre-Julian
chronological chaos between the competing regional states.
14. Price, Gears, pp. 8-9.
15. Price, Gears, pp. 56-9; Cic. Nat de. 2.34-35.
16. Dio Chrys. 31.104.
17. D.S. 20.96.3-97.3.
18. App. B.C. 4.66-7, Winter, p. 199-201.
19. Plb. 5.88.5, Str. 14.2.5 (653), Aristid. 25.4.
20. Str. 14.2.5 (653), Plb. 21.7.1-4.
21. Price, Gears, pp. 60-61.
22. Cic. De re pub, 1.14.21.
23. Cic. ad Att. 5.12.1, Brut. 1; Plu. Cic. 36.
24. Philo. Bel. 73. For reconstructions of the device, cf.
Vernard Foley and Werner Soedel, "Ancient Catapults," Scientific
American 241 (April, 1979): 155-6; J. G. Landels, Engineering in
the Ancient World (Berkeley: University of California Press,
1978), pp. 123-7.
25. V. P. M. Ptolemaic Alexandria, 3 vols. (Oxford, Clarendon
Press, 1972), 2.431.
26. Homer, Od. 5.233-40, Libanus, Progymnasmata, Sententiae
27. Dicaearchus Fr. 33, Strabo Str. 2.1.1 (67), 5.7 (114),
2.5.19 (122-3), 2.5.39 (134).
28. Vitr. 10.46-48.
29. Pappus, Comm. in Alm. 4.11.66 f., ed. Rome, Almagest, 9.2,
30. Str. 16.2.10.
31. Liv. 36.43.8; 37.14.3. App. Syr. 22, Johannes Hendrik Thiel,
Studies on the History of Roman Sea-power in Republican Times
(Amsterdam: Noord-hollandsche uitgevers mij., 1946), p. 301.
32. App. Mith. 25; FGrH 434 F 22.13-15.
33. V. Joseph Noble and Derek Di Solla Price, "The Water Clock
in the Tower of the Winds," American Journal of Archaeology 72
34. Ov. Fast. 6.263-283.
35. C.A. Kinkaid, Successors of Alexander the Great (Chicago:
Ares Publishing Company, 1980) p. 143.
36. Cic. De Nat. Deo. 2.34-5 (87-88), Rackham’s translation.