One past or two? Ancient History and Archaeology in archaic Sicily.

I’ve been looking at the colonisation of Sicily by the Greeks in the early 1st millennium BC. Some time around the mid-eighth century it seems as though parts of Greece exploded leaving Greek cities around the Mediterranean. It’s not that distance so much that interests me it’s the distance between Ancient History and Archaeology in explaining how this happened. A lot of ancient historians see archaeology as a method of filling in the gaps in the historical record. I think there’s much more to be said for using them as two independent approaches to a common past.

Blue pins: Greece, Chalkis and Eretria Euboea - Italy, Pithekoussai and Naxos
Yellow Pins: Greece, Corinth - Italy, Syracuse
Green Pins: Greece, Megara Nicaea - Italy, Megara Hybalea
Red Pin: Italy, Himera
View Greek Colonies in a larger map

Ancient Historians have Greek colonisation pretty much worked out, barring the finer details. In the case of the western Mediterranean the earliest vessels set sail from Euboea, the first colony not being on Sicily at all but Pitheokoussai in the Bay of Naples around 750 BC. Dates for Sicily in the magisterial An inventory of archaic and classical poleis, which I’m about to deeply disagree with, 735/4 BC for Naxos from Euboea, 733/2 for Syracuse from Corinth and 728 for Megara Hyblaea from Megara Nicaea in mainland Greece. The prevarication on the dates is because Greek years did not start and end at the same time as ours. The dates and origins are derived from Thucydides and there’s good reason to assume they’re accurate. In the case of Pithkoussai, the earliest layers do have Euboean pottery. The same goes for Naxos. In Syracuse, the earliest pottery is indeed Corinthian, if you ignore Euboean and Athenian pottery below the Corinthian (Boardman 1999:163-4) and the indigenous pottery that is found before settlement that continues during through to 650 BC at the site (Frasca 1983:597-8). Megara Hyblaea in contrast has Corinthian pottery in its early layers. The solution is to conclude that pottery is diagnostic, unless the answer is wrong in which case it’s merely evidence of trade. In this case the same is true for the east Greek wares (Boardman 1999:174).

Megara Hyblaea poses more of a problem. The earliest burials at the site are not typical of Megara Nicaea. In fact the homeland styles don’t start appearing till around 650 BC (Shepherd 1995:51-82). I’m kicking myself for not making a note where I read that the earliest letter forms in Megara Hyblaea were also not similar to Megara Nicaea. Finally temples in Megara Hyblaea do not appear till after about 650 BC. None of this proves that Megara Hyblaea was not founded from Megara Nicaea, and the historical record is wrong. However, if the historical record is accurate then is there likely to be a stronger archaeological trace of ethnic and economic links?

If these colonies didn’t come from Greek settlers arriving en masse then where did they come from? Syracuse is traditionally thought to be an excellent example of Greek settlement in action because a native settlement is clearly removed with a destruction layer and a Greek layer over the top. In fact the archaeological record is more of destruction lenses, with some native houses continuing in use and with some continued use of native pottery. Is this more indicative of Greek arrival in native settlements? Himera, founded in the mid-seventh century BC, was surrounded by indigenous settlements (Vassallo 1996). The shock of the arrival of this new city on the native settlements was negligible. The closest comparison I can think of is the Islamisation of Swahili towns on the East African coast. Wynne-Jones (2007, and other papers by other authors) note two possible origins for the self-identified Omani towns. Either they were settled by Omanis, or else √©lites attracted Omanis in via exchange and inter-marriage. The lack of settlement shock means that places like Kilwa Kisiwani are assumed to have been native developments that pull settlers in, rather than sites of settlers pushed out. The traditional model for Greek colonisation is a push model from a limited number of sites. Is a pull model feasible?

Ancient Historians are happy that cities could pull in trade, to account for inconvenient pottery at sites, so the possibility for pull colonisation is not in doubt. What is lacking is evidence this happened, but it is possible that the evidence is not in Sicily, but rather Greece. Olympia and Delphi are both home to treasuries from cities around the Greek world. These were both from homeland cities, and from colonies in the Mediterranean who were placing themselves in the heart of the Greek world. At Olympia we can see this started happening. Gela, a city in southern Sicily, put down one of the early treasuries (Gardiner, 1925 dates it to the second half of the seventh century BC, but most recent books I’ve read while fact-checking this give an uncited date in the sixth century BC. If you know where this date comes from I’d be delighted if you let me know in the comment box.) and many other cities from the west followed. None of this conclusively settles the argument in favour of a pull model, but it does raise the question as to why Thucydides is so uncritically followed by ancient historians who would normally pull out all sorts of overlooked detail with forensic skill. Ancient Historians in turn could ask, if the push model is flawed, why did Thucydides write about these pushes out to settle colonies – and this is where I’m most puzzled of all.

Thucydides was an Athenian general and his history is a History of the Peloponnesian War. The foundations of Sicilian colonies are mentioned, but they’re only mentioned as they are relevant to Thucydides’ aim, which is recording the conflict between Athens and Sparta. Thucydides wrote at the end of the fifth century BC. This means that he was not witness to the foundation events. Instead of recording the mid-eighth century and the settlement of Sicily, he is recording what people in the fifth century thought happened three hundred years earlier. This solves a lot of problems. For example, in the mid-eighth century BC a lot of Greek cities did not exist. Corinth only came into being as a polis around 750-ish BC, it organised quickly enough to send out fully fledged cities within twenty years. We have no date for the cities of Euboea forming, but if history is followed then they were establishing a colony at Pithekoussai before Corinth was a polis. Is perhaps more likely that Thucydides saw cities in Sicily and described the foundation of cities because what he knew where there were fifth-century cities that needed to be explained? Possibly, but then why the detail about the order of settlement?

Megara Hyblaea and petrochemical plants
Megara Hyblaea is surrounded by petrochemical plants, giving at a pervasive stench of death. The consequent lack of tourists and facilities gives it the impression of somewhere that was built to be an abandoned city.
There is a less manipulated version of the photo at Flickr.

The dates do not only give an account of when something happened, but also of precedence. The major sea routes to Sicily arrived first at Naxos, possibly because this was the polis closest to the striking landmark Mount Etna. It’s geographical location as the first place you arrive at in Sicily suggests that it should have chronological primacy too. Syracuse, with its power and harbour is clearly the next most prestigious city. Fifth-century Greeks would not have cared that eighth-century sailors beached their vessels rather than use harbour. The history then becomes not what happened, but a tale to explain why things are the way they are now. Puzzlingly, this is not new. It’s the first lesson on any ancient history course, so the emphasis on Thucydides as a reliable source is odd.

This doesn’t simply mean that archaeology is good and history bad. It does mean that using history to analyse archaeology and vice versa is a very poor substitute to using archaeology to analyse archaeology and history to analyse history. The archaeological record gives a very different story to late prehistoric Sicily than the history recorded by Thucydides. Yet at the same time, the historical record gives a much richer account of the local ethnicity and allegiances of the fifth century BC than the archaeology. Corinthian pottery might get everywhere, but the history clearly shows that does not make everyone Corinthian. It also opens another possibility that both archaeological and historical attempts to explain Greek colonisation in Sicily could be flawed.

The notion of Greek ethnicity is based in history. I believe that the history is anachronistic, but Hall (2004) goes further. He argues that the idea of a Greek ethnicity is, in this period, possibly anachronistic. Did a Greek identity arise as a response to increased interaction across the Mediterranean? It’s common for historians to talk about ancient Greeks as though they are one thing over the course of several hundred years. We see a process of becoming Greek, and by the fifth century there is a difference between the Greeks and the barbaroi. While the archaeological record shows Greek pottery getting everywhere, the ethnic information – that some people still thought of themselves as Sikels – is purely from the historical data, which is late fifth century. This is after the invasions from Persia in Greece, and the battles with Punic forces in Sicily. Is a hunt for Greek cities taking a recently developed sense of common identity and anachronistically searching for it into the past? Hall’s proposal raises the possibility that much work, including my own, is excessively teleological.

There was one past for archaic Sicily. Instinctively I can’t help but feel that approaches that pull archaeology and ancient history together should be a good thing. However, I wonder if there’s a danger that when you do try that it becomes effectively one discipline judged by the approach of another. By keeping a distance between the two approaches you get the advantage of two independent viewpoints.