The purpose of this brief note is to draw the attention of historians interested in the ancient Erythraean Sea world to a short excerpt from Agatharchides of Cnidus’ treatise entitled On the Erythraean Sea1. To my knowledge, this text has not elicited any particular comments from commentators on Agatharchides, with the exception of D. Woelk (see below). In this passage, Agatharchides recounts in a few lines a myth that was common among the Ikhthyophagoi (“Fish-Eaters”), a people living on the African coast of the Red Sea (an area also known as Trôgodytikê). There is little doubt, as we shall see, that this ancestral myth refers to a tsunami that struck the coastline inhabited by some of these Ikhthyophagoi.
1. Introductory remarks
a) What is a tsunami ?
A tsunami is a maritime phenomenon whose true power is visible on the coastline but whose causes often originate in the open sea. To summarize what a tsunami is in a few lines, I reproduce below the essential elements of the article from Britannica Academic2.
“Tsunami, (…) also called seismic sea wave or tidal wave: catastrophic ocean wave, usually caused by a submarine earthquake, an underwater or coastal landslide, or a volcanic eruption. (…). After an earthquake or other generating impulse occurs, a train of simple, progressive oscillatory waves is propagated great distances over the ocean surface in ever-widening circles, much like the waves produced by a pebble falling into a shallow pool. In deep water a tsunami can travel as fast as 800 km (500 miles) per hour. The wavelengths are enormous, sometimes exceeding 500 km (about 310 miles), but the wave amplitudes (heights) are very small, only about 30 to 60 cm (1 to 2 feet). The waves’ periods (the lengths of time for successive crests or troughs to pass a single point) are very long, varying from five minutes to more than an hour. (…). As the waves approach the coast of a continent, however, friction with the rising sea bottom reduces the velocity of the waves. As the velocity lessens, the wavelengths become shortened and the wave amplitudes (heights) increase. Coastal waters may rise as high as 30 metres (about 100 feet) above normal sea level in 10 to 15 minutes. The continental shelf waters begin to oscillate after the rise in sea level. Between three and five major oscillations generate most of the damage, frequently appearing as powerful “run-ups” of rushing water that uproot trees, pull buildings off their foundations, carry boats far inshore, and wash away entire beaches, peninsulas, and other low-lying coastal formations. (…) Much like any other water waves, tsunamis are reflected and refracted by the topography of the seafloor near shore and by the configuration of a coastline. As a result, their effects vary widely from place to place. Occasionally, the first arrival of a tsunami at a coast may be the trough of the wave, in which case the water recedes and exposes the shallow seafloor . Such an occurrence took place in the bay of Lisbon, Portugal, on November 1, 1755, after a large earthquake; many curious people were attracted to the bay floor, and a large number of them were drowned by the wave crest that followed the trough only minutes later.”
Regarding the phenomenon of recessing seawater (in the case of tsunamis caused by the displacement of an underwater fault), the geographer B. Bousquet provides the following interesting details:
“Dans le cas d’un jeu de failles, les vagues se propagent différemment de part et d’autre de l’accident. Sur un côté, le soulèvement entraîne à la côte d’abord une forte houle, à laquelle succèdent des reflux correspondant aux creux entre deux ondes ; suivent des déferlements et des afflux d’eau qui peuvent être destructeurs si les conditions locales s’y prêtent. Du côté effondré, le phénomène s’inverse, commençant par un mouvement de reflux. L’intervalle qui sépare le déferlement des vagues est régulier et peu dépasser l’heure”3.
b) Descriptions of Mediterranean tsunamis in Greek and Roman sources
Several tsunamis affecting the coasts of the Mediterranean Sea in ancient times are known to us from written sources4. Some were just hinted at, with the authors providing almost no descriptive details . In contrast, four tsunamis were described more satisfactorily. Not only did the authors offered readers a more descriptive account, but they also suggested possible causes to explain these coastal phenomena. Some put forward physical explanations, while others regarded them as a kind of divine punishment. In addition, three texts converge, pointing to a very specific aspect of this phenomenon: their authors explicitly report seawater flowing back before the arrival of the destructive wave that submerged the coastline with different degrees of gravity.
i) The Potidaea (Peninsula of Pallene) tsunami (479 BC)- Herodotus, 8.129
Ἀρταβάζῳ δὲ ἐπειδὴ πολιορκέοντι ἐγεγόνεσαν τρεῖς μῆνες, γίνεται ἄμπωτις τῆς θαλάσσης μεγάλη καὶ χρόνον ἐπὶ πολλόν. ἰδόντες δὲ οἱ βάρβαροι τέναγος γενόμενον παρήισαν ἐς τὴν Παλλήνην. ὡς δὲ τὰς δύο μὲν μοίρας διοδοιπορήκεσαν, ἔτι δὲ τρεῖς ὑπόλοιποι ἦσαν, τὰς διελθόντας χρῆν εἶναι ἔσω ἐν τῇ Παλλήνῃ, ἐπῆλθε πλημμυρὶς τῆς θαλάσσης μεγάλη, ὅση οὐδαμά κω, ὡς οἱ ἐπιχώριοι λέγουσι, πολλάκις γινομένη. οἱ μὲν δὴ νέειν αὐτῶν οὐκ ἐπιστάμενοι διεφθείροντο, τοὺς δὲ ἐπισταμένους οἱ Ποτιδαιῆται ἐπιπλώσαντες πλοίοισι ἀπώλεσαν. αἴτιον δὲ λέγουσι Ποτιδαιῆται τῆς τε ῥηχίης καὶ τῆς πλημμυρίδος καὶ τοῦ Περσικοῦ πάθεος γενέσθαι τόδε, ὅτι τοῦ Ποσειδέωνος ἐς τὸν νηὸν καὶ τὸ ἄγαλμα τὸ ἐν τῷ προαστείῳ ἠσέβησαν οὗτοι τῶν Περσέων οἵ περ καὶ διεφθάρησαν ὑπὸ τῆς θαλάσσης· αἴτιον δὲ τοῦτο λέγοντες εὖ λέγειν ἔμοιγε δοκέουσι.
But when Artabazus5 had besieged Potidaea for three months , there was a great ebb-tide in the sea, lasting for a long while, and when the foreigners saw that the sea was turned to a marsh they made to pass over it into Pallene. But when they had made their way over two fifths of it and three yet remained to cross before they could be in Pallene, there came a great flood-tide, higher, as the people of the place say, than any one of the many that had been before; and some of them that knew not how to swim were drowned, and those that knew were slain by the Potidaeans, who came among them in boats. The Potidaeans say that the cause of the high sea and flood and the Persian disaster lay herein, that those same Persians who now perished in the sea had profaned the temple and the image of Poseidon that was in the suburb of the city; and I think that in saying that this was the cause they say rightly.” (translated by A. D. Godley)
ii) A tsunami in Euboea and other Aegean places (424 BC) – Thucydides, 3.89.2-5
Kαὶ περὶ τούτους τοὺς χρόνους, τῶν σεισμῶν κατεχόντων, τῆς Εὐβοίας ἐν Ὀροβίαις ἡ θάλασσα ἐπανελθοῦσα ἀπὸ τῆς τότε οὔσης γῆς καὶ κυματωθεῖσα ἐπῆλθε τῆς πόλεως μέρος τι, καὶ τὸ μὲν κατέκλυσε, τὸ δ᾿ ὑπενόστησε, καὶ θάλασσα νῦν ἐστι πρότερον οὖσα γῆ· καὶ ἀνθρώπους διέφθειρεν ὅσοι μὴ ἐδύναντο φθῆναι πρὸς τὰ μετέωρα ἀναδραμόντες. καὶ περὶ Ἀταλάντην τὴν ἐπὶ Λοκροῖς τοῖς Ὀπουντίοις νῆσον παραπλησία γίγνεται ἐπίκλυσις, καὶ τοῦ τε φρουρίου τῶν Ἀθηναίων παρεῖλε καὶ δύο νεῶν ἀνειλκυσμένων τὴν ἑτέραν κατέαξεν. ἐγένετο δὲ καὶ ἐν Πεπαρήθῳ κύματος ἐπαναχώρησίς τις, οὐ μέντοι ἐπέκλυσέ γε· καὶ σεισμὸς τοῦ τείχους τι κατέβαλε καὶ τὸ πρυτανεῖον καὶ ἄλλας οἰκίας ὀλίγας. αἴτιον δ᾿ ἔγωγε νομίζω τοῦ τοιούτου, ᾗ ἰσχυρότατος ὁ σεισμὸς ἐγένετο, κατὰ τοῦτο ἀποστέλλειν τὲ τὴν θάλασσαν καὶ ἐξαπίνης πάλιν ἐπισπωμένην βιαιότερον τὴν ἐπίκλυσιν ποιεῖν· ἄνευ δὲ σεισμοῦ οὐκ ἄν μοι δοκεῖ τὸ τοιοῦτο ξυμβῆναι γενέσθαι.
At about the same time, while the earthquakes prevailed, the sea at Orobiae in Euboea receded from what was then the shore-line, and then coming on in a great wave overran a portion of the city. One part of the flood subsided, but another engulfed the shore, so that what was land before is now sea; and it destroyed of the people as many as could not run up to the high ground in time. In the neighbourhood also of the island of Atalante, which lies off the coast of Opuntian Locris, there was a similar inundation, which carried away a part of the Athenian fort there, and wrecked one of two ships which had been drawn up on the shore. At Peparethos6 likewise there was a recession of the waters, but no inundation; and there was an earthquake, which threw down a part of the wall as well as the prytaneum and a few other houses. And the cause of such a phenomenon, in my own opinion, was this: at that point where the shock of the earthquake was greatest the sea was driven back, then, suddenly returning with increased violence, made the inundation; but without an earthquake, it seems to me, such a thing would not have happened. (translated by C. F. Smith)
iii) Various tsunamis in Euboea and Central Greece – Strabo, 1.3.20
Kῦμά τε ἐξαρθὲν τριχῆ τὸ μὲν πρὸς Σκάρφην ἐνεχθῆναι καὶ Θρόνιον, τὸ δὲ πρὸς Θερμοπύλας, ἄλλο δὲ εἰς τὸ πεδίον ἕως τοῦ Φωκικοῦ Δαφνοῦντος, πηγάς τε ποταμῶν ξηρανθῆναι πρὸς ἡμέρας τινάς· τὸν δὲ Σπερχειὸν ἀλλάξαι τὸ ῥεῖθρον καὶ ποιῆσαι πλωτὰς τὰς ὁδούς, τὸν δὲ Βοάγριον κατ’ ἄλλης ἐνεχθῆναι φάραγγος, καὶ Ἀλόπης δὲ καὶ Κύνου καὶ Ὀποῦντος πολλὰ καταβλαβῆναι μέρη, οἶον δὲ τὸ ὑπερκείμενον φρούριον πᾶν ἀνατραπῆναι, Ἐλατείας δὲ τοῦ τείχους καταρραγῆναι μέρος, περὶ δὲ Ἄλπωνον θεσμοφορίων ὄντων πέντε καὶ εἴκοσι παρθένους ἀναδραμούσας εἰς πύργον τῶν ἐλλιμενίων κατὰ θέαν, πεσόντος τοῦ πύργου, πεσεῖν καὶ αὐτὰς εἰς τὴν θάλατταν. Λέγουσι δὲ καὶ τῆς Ἀταλάντης τῆς πρὸς εὐβοίᾳ τὰ μέσα ῥήγματος γενομένου διάπλουν δέξασθαι, μεταξὺ καὶ τῶν πεδίων ἔνια καὶ μέχρι εἴκοσι σταδίων ἐπικλυσθῆναι, καὶ τριήρη τινὰ ἐκ τῶν νεωρίων ἐξαιρεθεῖσαν ὑπερπεσεῖν τοῦ τείχους.
A wave rose up in three parts, one of which was carried toward Skarphe and Thronion, another toward Thermopylai, and the remainder into the plain as far as Phokic Daphnous. The springs of rivers dried up for several days, the Spercheios changed its course and made the roads navigable, the Boagrios was carried into another ravine, and parts of Alope, Kynos, and Opous were gready damaged, with Oion (the fortress lying above the latter) overturned. Part of the walls of Elateia collapsed, and when Alponos was holding the Thesmophoria, twenty five girls ran up to a tower on the harbor for a view, and the tower fell in, and they fell into the sea. They also say that the central portions of Atalante near Euboia, because they had been torn apart, received a channel through it, some of the plains were inundated for twenty stadia, and a trireme was lifted out of the shipyards and thrown over the wall. (translated by D. W. Roller7)
iv) A tsunami in Eastern Mediterranean (365 AD)- Ammianus Marcellinus, 26.10.15-19
Hoc novatore adhuc superstite, cuius actus multiplices docuimus et interitum, diem duodecimum Kalendas Augustas, consule Valentiniano primum cum fratre, horrendi terrores per omnem orbis ambitum grassati sunt subito, qualis nec fabulae nec veridicae nobis antiquitates exponunt. Paulo enim post lucis exortum, densitate praevia fulgurum acrius vibratorum, tremefacta concutitur omnis terreni stabilitas ponderis, mareque dispulsu, retro fluctibus evolutis abscessit, ut retecta voragine profundorum, species natantium multiformes limo cernerentur haerentes, valliumque vastitates et montium tunc (ut opinari dabatur), suspicerent radios solis, quos primigenia rerum natura sub immensis gurgitibus amendavit. Multis igitur navibus velut arida humo conexis, et licenter per exiguas undarum reliquias palantibus plurimis, ut pisces manibus colligerent et similia: marini fremitus velut gravati repulsam, versa vice consurgunt, perque vada ferventia insulis et continentis terrae porrectis spatiis violenter illisi, innumera quaedam in civitatibus, et ubi reperta sunt aedificia, complanarunt: proinde ut elementorum furente discordia, involuta facies mundi, miraculorum species ostendebat. Relapsa enim aequorum magnitudo cum minime speraretur, milia multa necavit hominum et submersit, recurrentiumque aestuum incitata vertigine, quaedam naves, postquam umentis substantiae consenuit tumor, pessumdatae visae sunt, exanimataque naufragiis corpora supina iacebant aut prona. Ingentes aliae naves, extrusae rabidis flatibus, culminibus insedere tectorum (ut Alexandriae contigit) et ad secundum lapidem fere procul a litore contortae sunt aliquae, ut Laconicam prope Mothonen oppidum nos transeundo conspeximus, diuturna carie fatiscentem.
While that usurper of whose many deeds and his death we have told, still survived, on the twenty-first of July in the first consulship of Valentinian with his brother, horrible phenomena suddenly spread through the entire extent of the world, such as are related to us neither in fable nor in truthful history. For a little after daybreak, preceded by heavy and repeated thunder and lightning, the whole of the firm and solid earth was shaken and trembled, the sea with its rolling waves was driven back and withdrew from the land, so that in the abyss of the deep thus revealed men saw many kinds of sea-creatures stuck fast in the slime; and vast mountains and deep valleys, which Nature, the creator, had hidden in the unplumbed depths, then, as one might well believe, first saw the beams of the sun. altered face of the earth revealed marvellous sights8. For the great mass of waters, returning when it was least expected, killed many thousands of men by drowning; and by the swift recoil of the eddying tides a number of ships, after the swelling of the wet element subsided, were seen to have foundered, and the lifeless bodies of shipwrecked persons lay floating on their backs or on their faces. Other great ships, driven by the mad blasts, landed on the tops of buildings (as happened at Alexandria), and some were driven almost two miles inland, like a Laconian ship which I myself in passing that way saw near the town of Mothone, yawning apart through long decay. Hence, many ships were stranded as if on dry land, and since many men roamed about without fear in the little that remained of the waters, to gather fish and similar things with their hands, the roaring sea, resenting, as it were, this forced retreat, rose in its turn; and over the boiling shoals it dashed mightily upon islands and broad stretches of the mainland, and levelled innumerable buildings in the cities and wherever else they were found; so that amid the mad discord of the elements the the altered face of the earth revealed marvellous sights. For the great mass of waters, returning when it was least expected, killed many thousands of men by drowning; and by the swift recoil of the eddying tides a number of ships, after the swelling of the wet element subsided, were seen to have foundered, and lifeless bodies of shipwrecked persons lay floating on their backs or on their faces. Other great ships, driven by the mad blasts, landed on the tops of buildings (as happened at Alexandria﴿, and some were driven almost two miles inland, like a Laconian ship which I myself in passing that way saw near the town of Mothone, yawning apart through long decay. (translated by J. C. Force)
v) Additional testimony
To add one last piece of evidence, it is worth mentioning the probable tsunami witnessed by Strabo during his stay in Egypt (which supports his hypothesis that at some point in the future, following an exceptional geological phenomenon, the isthmus separating the Red Sea and the Mediterranean Sea may disappear). (Strabo, 1.3.17)
Ἡμῶν δ’ ἐπιδημούντων ἐν Ἀλεξανδρείᾳ τῇ πρὸς Αἰγύπτῳ, περὶ Πηλούσιον καὶ τὸ Κάσιον ὄρος μετεωρισθὲν τὸ πέλαγος ἐπέκλυσε τὴν γῆν καὶ νῆσον ἐποίησε τὸ ὄρος, ὥστε πλωτὴν γενέσθαι τὴν παρὰ τὸ Κάσιον ὁδὸν τὴν ἐς Φοινίκην. Οὐδὲν οὖν θαυμαστὸν οὐδ’ εἴ ποτε διαστὰς ὁ ἰσθμὸς ἢ ἵζημα λα βὼν ὁ διείργων τὸ Αἰγύπτιον πέλαγος ἀπὸ τῆς Ἐρυθρᾶς θαλάττης ἀποφανεῖ πορθμὸν καὶ σύρρουν ποιήσει τὴν ἐκτὸς θάλατταν τῇ ἐντός, καθάπερ ἐπὶ τοῦ κατὰ τὰς Ἡρακλέους στήλας πορθμοῦ συνέβη.)
When I was living in Alexandria next to Egypt, the sea around Pelousion and Mount Kasion rose and inundated the land, making an island of the mountain, so that the road from Kasion into Phoenicia became navigable. Thus it would be nothing marvellous if at some time, the isthmus (the one that divides the Egyptian Sea from the Erythraean Sea) should be separated or undergo subsidence, revealing a strait and making the outer sea confluent with the inner, just as happened with the strait at the Pillars of Herakles. (translated by D. W. Roller)
2) Contextualizing the excerpt from Agatharchides’ treaty
a) Agatharchides’ sources of information (brief overview)
The passage discussed in this short note is taken from Book 5 of Agatharchides of Cnidus’ Treatise on the Erythraean Sea. Agatharchides was active in Alexandria in the 2nd century during the reigns of Ptolemy VI and Ptolemy VIII9. This treatise is known through two important paraphrases: the first one by Diodorus of Sicily and and the second by the byzantine scholar Photius. We must also take into account the paraphrase made by Artemidorus of Ephesus in Book 11 of his Geographoumena, of which an abridged version survives in Book 16 of Strabo’s Geography10.
The question of the sources used by Agatharchides has been studied for a long time, and it suffices to recall the main conclusions that have been drawn11. On the one hand, Agatharchides certainly used the achievements of recent geographical science (e.g., Eratosthenes), even though he does not mention them. On the other hand, he specifically mentions two sources, which are the basilika hupomnêmata (royal archives) and eyewitnesses. The former were the official reports and accounts written by all those who sailed the Red Sea on behalf of the Ptolemaic kings and were subsequently archived in Alexandria; these documents recorded what had been seen and what had been heard from local informants. As for the latter, when they were not those who were employed by the Ptolemies, they must have been either merchants traveling in the Red Sea or individual travelers. Alternatively some scholars suggested that Agatharchides used this term to refer to unofficial works written by Greek travelers and thus were not preserved in the royal archives. At any rate, it is clear, as Desanges argued, that most of Agatharchides’ historia describes a situation that generally does not date from later than the time of Ptolemy III – a time when Greek explorators had already crossed the Bab al-Mandab strait12. In other words, Agatharchides had at his disposal geographical information on the whole of Trôgodytikê, even though his account was, as J. Desanges has shown, “largement périmé à l’époque où il écrit” (largely out of date at the time of writing)13 . It should be noted in passing that this body of knowledge was not updated by those who could have done so, Diodorus and Strabo; Artemidorus however enhanced Agatharchides’ description of Trôgodytikê with minor additions.
b) In which part of Trôgodytikê did this tsunami occur?
Book 5 of the Treatise on the Erythraean Sea (the only book of which substantial fragments have been passed down to us) contains a lengthy ethnographic account of the Ikhthyophagoi, immediately followed by another one devoted to nomad tribes collectively called Trogodytes. After these comes a detailed description of the fauna of the Erythraean area. Then Agatharchides takes the reader on a journey (periplous) around the Red Sea (Arabian Gulf), describing Trogodytikê and Arabia (including the hinterland) as if he was sailing along the seashore. The periplous begins with the western (African) coast, from the Gulf of Suez to the Bab al-Mandab area (the excerpt under investigation takes place in this section), before describing the eastern (Arabian) shore, according to the same principles.
Unfortunately, the remaining spatial indications are not precise enough to locate what part of Trogodytikê was struck by this tsunami. There are only two kinds of evidence that can help us approximately identify the area : place names on the one hand (1), and landscape descriptions on the other (2)14.
(1) Place names. The region in question is located (sailing along the coastline, from the Gulf of Suez to the Strait of Bab el-Mandeb), between the port called by the Greeks Sôteria, or Sôteira, and Ptolemais of the Hunts. It is not easy to identify Sôteriâ exactly. We only know for certain is that it lay beyond Berenice, the Gulf Akathartos (Foul Bay), and the island of Ophiodês (Gezirad Zabarjad). According to Agatharchides, Sôteria was located before a point where the coastline began to curve eastward, causing the Red Sea to narrow in this section. D. Woelk, adopting Müller’s view, believes that Sôteria must be identified with Suakin15. As for Ptolemais of the Hunts, this place must be sought somewhere in the region of Aqiq.
(2) Landscape features. A characteristic detail of the area referred to by Agatharchides was the absence of mountains near the coast (unlike the preceding coastline). It was also characterized by the shallow depth of the sea along the shore. In addition, the Ptolemaic observers were struck by the spectacular green color of sea water and the abundance of fish. We read in Strabo’s version that mangroves trees grew there, more precisely in a channel (poros). Unlike the mangroves, which are not of great help, the “channel” represents an interesting landmark. The word poros must be understood as “maritime passage”16 . In this part of the Red Sea, the word poros is very likely to point to a passage formed by the coastline on one side and reefs or islets on the other. In this area the Shubuk Channel is a good candidate17.
Putting together these scattered pieces of evidence, we can surmise that the tsunami occurred within the area shown in the maps below.

c) The Fish Eaters as seen by Agatharchides
The Ikhthyophagoi by whom the Ptolemaic explorers were told this extraordinary story, and whose ancestors were victims of the tsunami, were certainly settled on the Suakin-Aqiq coastline, where they would exploit the sea’s resources. Here it is worth recalling that the numerous and various inhabitants of Trogodytikê were divided by the Greeks into two main groups : the “Fish eaters” on the one hand, and on the other many pastoral peoples collectively called Trogodytes. Agatharchides provides many details concerning the lifestyle (Greek : bios) of the Ikhthyophagoi, emphasizing their very simple needs.
- Food was exclusively provided by the sea : small and large fish, sometimes cooked in the sun and processed to give a kind of “bread”; shellfish; cetacean bones and fish bones when stormy weather prevented them from collecting fish. Some groups of fish-eaters fed on sea turtles.
- The level of technical skill was very low: no tools were used, except ropes, goat horns – to catch large fish – and stones; having no boats they could not go fishing at sea.
- Housing was very rudimentary and varied according to the environment: galleries dug into coastal dunes, huts made of fish bones, caves, shelters built with mangrove trees.
- The social organization was accordingly very simple: no chief ; social life limited to the collective satisfaction of basic needs (food, drinking, reproduction) etc.
- Funeral rites were almost nonexistent: the dead were carried away by the sea and eaten by fish.
in Agatharchides’ point of view the Ikhthyophagoi were concerned only with satisfying the essential needs that ensured the survival of both individuals and group. He regarded them as peoples without history, as they led a kind of “cyclical life” (Greek : anastrophê). This ethnographic account was the starting point of various political and ethical thoughts : the Fish Easters represented, in Agatharchides’ view, the paradigm of societies where life was governed by phusis (“nature,” i.e., basic sensations and needs) and not by doxa (“opinion,” i.e., what is socially conventional: perception of good and evil, compassion, enrichment…), hence the comparison he sometimes makes between the lifestyle of the Fish Eaters and the life of animals. On the other hand, he emphasizes that such destitute societies limiting themselves to the satisfaction of natural needs were safe from the disorders experienced by more complex societies where, for example, social inequalities created serious inequalities and much trouble.
3) The story told by the Fish-Eaters
a) The Greek text with translation
Having described the area lying between Berenice and Ptolemais of the Hunts (see above), Agatharchides embarks on a long digression, describing the dramatic consequences of shipwrecks that can befall elephant-carrying vessels (elephantêgoi). Immediately after this account, the author provides his audience with this short story (Greek: λόγος) – actually a mythical story that was passed down from generation to generation among the Ikhthyophagoi inhabiting this part of the coastline. This story is found only in Diodorus (3.40.9 = Agatharchides, 83 Müller [85b Burstein]); neither Photius nor Artemidorus/Strabo included it in their paraphrases.
Παρὰ δὲ τοῖς πλησίον κατοικοῦσιν Ἰχθυοφάγοις παραδέδοται λόγος, ἐκ προγόνων ἔχων φυλαττομένην τὴν φήμην, ὅτι μεγάλης τινὸς γενομένης ἀμπώτεως ἐγεώθη τοῦ κόλπου πᾶς ὁ τόπος ὁ τὴν χλωρὰν ἔχων τοῦ τύπου πρόσοψιν, μεταπεσούσης τῆς θαλάττης εἰς τἀναντία μέρη, καὶ φανείσης τῆς ἐπὶ τῷ βυθῷ χέρσου πάλιν ἐπελθοῦσαν ἐξαίσιον πλήμην ἀποκαταστῆσαι τὸν πόρον εἰς τὴν προϋπάρχουσαν τάξιν.
And among the Fish Eaters who dwell near by has been handed down a tale which has preserved the account received from their forefathers, that once, when there was a great receding of the sea, the entire area of the gulf which has what may be roughly described as the green appearance [Burstein : “the whole portion of the gulf that appears green”] became land and the solid ground in the depths of it had emerged to view; a mighty flood came back upon it again and returned the body of water [wrong translation of the Greek word τὸν πόρον, “channel”] to its former place. (translated by C. H. Olfather)
b) Commentaries
As mentioned above, among modern commentators on the Treaty of the Erythraean Sea, only D. Woelk paid attention to this particular piece of information, which was ignored by K. Müller and S. M. Burstein. Here is his commentary:
“Sehr interessant ist die Niedergabe einer alten Ichthyophagensage, einst sei der gesamte Busen durch eine geweltige Ebbe ausgetrocknet worden. Bezeichnenderweise hat sich diese Nachricht gerade bei Bewohnern der Sudkuste, einem Gebiet mit wesentlich starkerer vulkanischer Tatigkeit als im Norden erhalten. Wie bereits Gosselin (Bredow) S.128 erkannt hat, darf an einen vulkanischen Vorgang gedacht werden, verbunden, so mochte ich hinzufugen, wahrscheinlich mit starker Hebung und nachfolgender Senkung eines bestimmten Abschnitts. Zur Geologie des Rotenmeergebiets und die dortige vulkanische Tatigkeit cf. das Kapitel Geology and Physical Geography in Handbuch der Naval Intelligence Division S.11 ff; auch Weber S. 50 ff.”18
Regretably D. Woelk only highlights the volcanic activity, ignoring others valuable details provided by the text. Indeed, in this account collected by the Greeks (undoubtedly the explorers in the service of Ptolemy II who sailed beyond Berenikê to survey the Trôgodytikê), two characteristic features of a tsunami appear: a powerful withdrawal of the sea, followed by the return of the tide in the form of a violent wave. The vocabulary and descriptive details (provided by the Ikhthyophagoi themselves?) are explicit:
- To describe the retreat of the sea, Agatharchides uses the term μεγάλη ἄμπωτις: “great ebb.” The word μεγάλη points to an exceptional ebb, not a common one. To emphasize the exceptional nature of this phenomenon and make this clearer to his audience, he adds that the sea moved “in the opposite direction”, in other words not towards the shore (μεταπεσούσης τῆς θαλάττης εἰς τἀναντία μέρη).
- Subsequently the normally submerged seabed became exposed : the entire part of the sea whose green water was a characteristic feature (see above) “became land” (ἐγεώθη).
- Some time after, the sea moved back to the shore in the form of an exceptionally powerful wave, as shown by the expression ἐξαίσιος πλήμη.
We can therefore state, without risk of error, that Agatharchides describes a tsunami probably resulting from an underwater earthquake in the Red Sea. This phenomenon occurred in the poros (“channel”) area, the location of which has been suggested below. The retreat of the sea was followed by a violent wave, which may have penetrated inland, since this part of the coastline was not bordered by mountains. Agatharchides’ description bears similarities to the four descriptions reproduced above. Tsunamis were not an unknown phenomenon to the Greeks, and Agatharchides (and before him the Greeks explorers) undoubtedly recognized what happened to this (these) tribe(s) of Fish-Eaters, even though he makes no explicit connection. It is certain that this tsunami did not cause the same damage as in the Mediterranean. There were probably no other structures on the coastline besides the shelters of the Fish-Eaters, which were light structures, easy to rebuild. However, we will never know it caused many casualties.
The mention of a past tsunami that struck the Aqiq region is interesting information in itself. But the story of the Fish-Eaters has another significance that should not be downplayed, since it considerably nuances the image of the Fish-Eaters that Agatharchides wanted to convey . In his general account of the Fish-Eaters, as seen above, he depicts them as primitive and destitute peoples, whose life was limited to satisfying their immediate needs; from Agatharchides’ point of view, they represented nothing more that an object of ethnographic curiosity. However, the tsunami story reveals that the Ptolemaic travelers entered into contact (and perhaps interaction) with these local populations: they undoubtedly questioned them about the environment in which they lived; it was during these conversations that the Fish-Eaters told them this ancestral myth. This anecdote attests to the fact that the Fish-Eaters possessed a collective memory of an ancient catastrophe. The memory of a past event was necessarily the basis of a collective identity. This very simple fact seriously contradicts the philosophical-theoretical ideas carefully put forward by Agatharchides in his ethnographic account19.
References
Bousquet 2006 = B. Bousquet, “Les séismes de l’Antiquité, entre nature et société”, in J. Jouanna, J. Leclant, M. Zink (edd.), L‘homme face aux calamités naturelles dans l’Antiquité et au Moyen Âge. Actes du 16ème colloque de la Villa Kérylos à Beaulieu-sur-Mer les 14 & 15 octobre 2005 (Cahiers de la Villa Kérylos, 17), Paris, 2006. p. 33-59.
Burstein 1989 = Agatharchides of Cnidus, On the Erythraean Sea, translated and edited by S. M. Burstein, London.
Desanges 1998 = J. Desanges, “Du bon usage d’Agatharchide ou de la nécessité de la Quellenforschung“, in P. Arnaud and P. Counillon (ed.), Geographica historica, Bordeaux, p. 69-82.
Desanges & Reddé 1994 = J. Desanges & M. Reddé, “La côte africaine du Bab el-Mandeb dans l’Antiquité”, in C. Berger, G. Clerc , N. Grimal (ed.), Hommages à Jean Leclant, vol. 3 (Études isiaques), Le Caire, p. 161-194.
Naco Del Hoyo & Nappo 2013 = T. Naco Del Hoyo & D. Nappo, “When the waters recede: the economic impact of tsunamis in the Graeco-Roman world”, Revue belge de philologie et d’histoire, 91/1, 2013, p. 45-68.
Freitag & Reicherter 2019 = K. Freitag & K. Reicherter, “The earthquake and tsunami of 426 BC in Greece: observations by Thucydides and contextual interpretations”, Zeitschrift für Geomorphologie, 62, Suppl. 2, p. 47–62.
Goukowsky 2006 = P. Goukowsky, “Flottes antiques et marées subites dans l’Antiquité grecque”, in J. Jouanna, J. Leclant, M. Zink (edd.), L’homme face aux calamités naturelles dans l’Antiquité et au Moyen Âge. Actes du 16ème colloque de la Villa Kérylos à Beaulieu-sur-Mer les 14 & 15 octobre 2005 (Cahiers de la Villa Kérylos, 17), Paris, p. 73-117.
Jacques & Bousquet 1984 = F. Jacques & B. Bousquet, “Le raz de marée du 21 juillet 365, Mélanges de l’École française de Rome. Antiquité, tome 96/1, p. 423-461.
Kelly 2004 = G. Kelly, “Ammianus and the Great Tsunami”, Journal of Roman Studies, 94, p.141-167.
Mathes-Schmidt & al. 2019 = M. Mathes-Schmidt, I. Papanikolaou, K. Reicherter, A. Pallikarakis, “Event deposits in the Eastern Thermaikos Gulf and Kassandra Peninsula (Northern Greece): evidence of the 479 BC ‘Herodotus tsunami’”, Zeitschrift für Geomorphologie, 62, Suppl. 2, p.101–125.
Morgan & Davies 2002 = E. Morgan & S. Davies, Red Sea Pilot (second edition), St Ives Huntington.
Müller 1855 = K. Müller (ed.),Geographi Graeci Minores, I, Paris (reprint. 1965, Hildesheim).
Nalesini 2012 = O. Nalesini, “Sacred Places and Beings of the Red Sea littoral Societies”, in D. A. Agius et al. (eds), Navigated Spaces, Connected Places. Proceedings of Red Sea Project V held at the University of Exeter, 16–19 September 2010, Oxford, 2012, p. 77-84.
Nenna 2018 = Nenna M.-D., « Les changements de niveau de la Méditerranée et l’archéologie », in J. Jouanna, Chr. Robin, M. Zink, Vie et climat d’Hésiode à Montesquieu (Cahiers de la Villa Kérylos, 29), Paris, p. 185-203.
Schneider 2011 = P. Schneider, “La connaissance des mangroves tropicales dans l’Antiquité : éléments complémentaires”, Topoi. Orient-Occident, 17/2, p. 353-402.
Schneider 2014 = P. Schneider, « Fauces Rubri maris : the Greco-Roman Bab al-Mandab », Orbis terrarum [Stuttgart], 12, p. 193-272.
Schneider 2019 = P. Schneider, « Ptolémaïs des chasses : une fondation urbaine grecque dans l’océan Indien (IIIe siècle a.C. – Ier siècle p.C) », Carnets de recherches de l’océan Indien, 3, 2019, p. 5-33
Stefanakis 2006 = M. I. Stefanakis, « Natural catastrophes in the Greek and Roman world: loss or gain ? Four cases of seaquake-generated tsunamis”, Mediterranaean archaeology and archaeometry, 6/1, p. 61-88.
Smid 1970, T. C. Smid, “‘Tsunamis’ in Greek Literature”, Greece & Rome, 17/1, p. 100–104.
Woelk 1966 = D. Woelk, Agatharchides von Knidos, Bamberg.
- Agatharchides, On the Erythraean Sea, 83 Müller [85b Burstein] = Diodorus of Sicily, 3.40.9 ↩︎
- Tsunami 2025. Britannica Academic. Retrieved 26 August 2025, from https://academic-eb-com.ezproxy.univ-artois.fr/levels/collegiate/article/tsunami/73633 ↩︎
- Jacques & Bousquet 1984, p. 440-441. ↩︎
- On these tsunamis, see, in particular, Smid 1970; Jacques & Bousquet 1984; Kelly 2004 ; Stefanakis 2006; Goukowsky 2006 ; Naco Del Hoyo & Nappo 2013; Nenna 201; Mathes-Schmidt & al. 2019; Freitag & Reicherter 2019 . ↩︎
- Artabazus was a lieutenant of Mardonios during the first Persian invasion of Greece. The passage leading to the Pallene Peninsula was blocked by the fortifications of Potidaea. ↩︎
- Today Skopelos. ↩︎
- Strabo draws on Demetrius of Kallatis: “Greek historian and geographer; at the end of the 3rd cent. BC he composed a work ‘On Europe and Asia’ in 20 bks (Diog. Laert. 5,83), a ‘a general history (…) arranged geographically’ [1. 2807]; its finishing-point is unknown, the death of Hieron II of Syracuse in 215 is mentioned in F 3 (5). He was used by, inter alia, Agatharchides (FGrH 85 T 3) and perhaps influenced the latter’s principle of representation (universal history according to geographical centres of influence). FGrH 85 with comm. Demetrius [30].” Brill’s New Pauly Online. Brill. https://doi-org.ezproxy.univ-artois.fr/10.1163/1574-9347_bnp_e314200 ↩︎
- This description, however emphatic it may sound, does not detract from the value of Ammianus’s testimony. See Jacques & Bousquet 1984, p. 440-441 : «Il présente comme initial le reflux découvrant largement les fonds sous-marins, avec une évidente exagération ; son information correspond donc au mouvement des flots dû à l’effondrement du sol sous-marin. Il n’y a pas lieu de mettre a priori en cause les textes qui décrivent ce reflux initial à Alexandrie, ville qu’Ammien évoque ensuite et d’où lui viennent probablement ses renseignements.» ↩︎
- For a biography of Agatharchides, see, for example, Burstein 1989, p. 12-18. ↩︎
- The Treatise on the Erythraean Sea was commented by Müller 1855, p. 111-195 ; Woelk 1966; Burstein 1989. Also see Desanges 1998. ↩︎
- See Woelk 1966, p. 255-266; Burstein 1989, p. 29-33; Desanges 1998. ↩︎
- Schneider 2014, p. 196-201. ↩︎
- Desanges 1998, p. 80. ↩︎
- Agatharchides, On the Erythraean Sea, 5, 83 Müller [Burstein 85a; 85b] = Diodorus Siculus, 3.40.1-4 & Photius 456b-457a ; Strabo, 16.4.7. On this topic, see Woelk 1966, p. 201-203; Müller 1855, p. 172-173; Burstein 1989, p. 140-144; Desanges & Reddé 1994, p. 185-186; Schneider 2011, p. 358-361 ; Schneider 2019, p. 15-20. ↩︎
- Woelk 1966, p. 201. ↩︎
- Liddell-Scott-Jones: πόρος: “a narrow part of the sea”; “strait”. Lat. Fretum. ↩︎
- “The channel provides a well sheltered alternative to the Inner Channel, which is used by heavier traffic. The reefs are marked by many small, low, sandy islets and are intersected by narrow channels. The water is usually clear but it can sometimes be muddy (…) Passage through the Shubuk Channel requires a sharp lookout though we recently traversed it S’bound under sail (…) with no major alarm.” (Morgan & Davies 2002, p. 133). ↩︎
- Woelk 1966, p. 204. There is also a short comment in Nalesini 2012, p. 80 (Nalesini does not think of a tsunami). ↩︎
- For further examples of local myths collected from the peoples of the Indian Ocean coast, see Arrianus, Ind. 22.5 (= Nearchus, FGrH 133 F1 ii); 31.1-2 (= Nearchus, FGrH F1 v); 31.6 (= Nearchus, FGrH 1 v). ↩︎
OpenEdition vous propose de citer ce billet de la manière suivante :
Pierre Schneider (7 décembre 2025). A tsunami on the African coast of the Red Sea in Antiquity. Les sociétés méditerranéennes antiques et les mondes de l'océan Indien. Consulté le 20 avril 2026 à l’adresse https://doi.org/10.58079/15ajh

