One of our previous postings, Of Sundials, suggested that scientific discovery was a great generator of anachronism within antiquity. Here we turn to a related area of ancient enquiry, geography – and to the challenges posed to Greco-Roman mappings of the world by new information about a strange group of islands set in the wide expanse of sea north of the European mainland.
One of the most striking features of the image of the world presented in Homer is a great encircling river known as Okeanos – the Ocean. By the time of Herodotus, however, that Homeric picture no longer seemed adequate: ‘It is impossible to argue against the person who spoke about Ocean, transporting the story (muthos) into the realm of uncertainty. I do not know the existence of any River Ocean, and I think that Homer or one of the other poets from past time invented the name and introduced it into his poetry.’ The encircling river is not, however, an anachronism for Herodotus: he simply does not have the knowledge to disprove the story. What he finds objectionable about the old poets is that they did not have any proper geographical knowledge either, but simply invented an encircling Ocean.
Change in geographical knowledge, as well as change in geographical features, plays an important role in discussions of anachronism in ancient criticism. Among the numerous complaints that Polybius brought against the Sicilian historian Timaeus was his out-of-date information on the geography of Africa. Diodorus noted that new geographical knowledge refuted the view that the Argonauts sailed on their return journey along the River Ister (the Danube) from the Black Sea to the Adriatic (which received the waters of a different River Ister). And within the wide-ranging field of Homeric scholarship, geographers such as Strabo were interested in the question of whether Homer’s narrative reflected the geographical realities known to the poet or those known to his characters.
What about the Ocean itself? By the time of the Roman empire there was rather more knowledge about areas outside the Mediterranean than in Herodotus’ lifetime. An interesting discussion is contained in the Byzantine encyclopedia, the Suda, in an entry on the Roman empire. I here modify slightly the translation offered in the invaluable online edition of the Suda:
This empire surpassed by far the empire of the Assyrians and Persians and Macedonians, the previous ones. In the East it is bounded by the Indians and the Red Sea and the Nile and Cataracts and lake Maeotis. As regards the west, it is bounded by Ocean itself, which was shown by the Romans’ accomplishments to be no myth; nor did the poets falsely sing its name for entertainment, since in fact the land of the Britons, which is an island surrounded by Ocean, has now been discovered and is considered part of the Roman empire.
The author of this piece is probably Dexippus, an Athenian statesman and historian of the third century AD, who wrote a world chronicle covering 1,000 years, an account of the events after the death of Alexander, and a contemporary history, the Scythica. I say ‘probably Dexippus’ because that name is preserved in the Suda after the citation, and so it is possible that what Dexippus actually wrote are the words that follow rather than precede the name.
Whoever the author, one of the implications of this Suda entry seems to be that Herodotus’ dismissal of the poets’ account of the Ocean is anachronistic. The advance of the Roman empire has replaced uncertainty with the clarity of knowledge and proved the poets right. But the fragment is not so straightforward as that. There is uncertainty over the text as well as over the authorship. Editors often assume that some words have dropped out after ‘as regards the west’, perhaps mentioning the great rivers normally seen as the bounds of the Roman empire, the Danube and Rhine. More to the point here is that the translation offered above translates an emendation in the crucial phrase about the Ocean’s status as myth: hon mê muthon for hon dê muthon.
Felix Jacoby in his great edition of fragments of the Greek historians offered a different interpretation of the passage (FGrH 100 F 12). He preserved the manuscript reading in that clause while including a different emendation (the addition of a single Greek letter, an eta meaning ‘than’) later in the same sentence. According to the text he prints, the Ocean ‘was shown by the Romans’ accomplishments to be a myth, its name nothing other than something sung by the poets for entertainment’. With this text, Herodotus’ geographical judgement is vindicated – though Herodotus himself is still an anachronism: the Ocean is a myth, but Dexippus (unlike Herodotus) can prove it.
The argument of the passage as a whole supports Jacoby’s interpretation. To show that the Ocean was no myth did not require the conquest and circumnavigation of Britain. A trip to the northern coast of Gaul or the western coast of the Iberian peninsula would have done just as well – or just as badly. After all, Herodotus could rightly object that the mere discovery that the Ocean flowed around Britain scarcely justifies the poet’s account of the Ocean as a whole. What the conquest of Britain does prove is that the poets were wrong – because they posited a River Oceanus that was undifferentiated and uninhabited.
Debate over this Suda entry is perhaps appropriate at a time when the British face the long-term consequences of an anachronistic commitment to a particular myth of British insularity. It is also a valuable reminder that our sense of the difference of classical antiquity is partly based on the fragility of our own knowledge of it. Much of what we say about the ancient world is nothing but wild surmise about fragments laden with interpretative problems. And yet – as we shall see in another blog – the historian Dexippus himself is a wonderful example of the unexpected leaps in our knowledge that can occur thanks to the eagle eyes of modern technology.
http://www.stoa.org/sol/ (Adler number: rho 246 = Jacoby, FGrH 100 F 12)
Herodotus 2.23; Polybius 12.3.1-3; Diodorus 4.56.7-8; Strabo 1.2.23, 1.2.31, 12.3.23.