Not too long ago, I was fortunate enough to spend two years in Greece studying Greek fortifications of the Early Iron Age and Archaic period, say between 1000 and 500 BC. One of the things that I was struck by after studying the secondary literature on Greek fortifications, is the often a priori assumption that fear of attack was a prime reason for communities to construct fortifications walls.
It is perhaps most strongly worded by Anthony Snodgrass in his Archaic Greece (1980) when he suggests that “Security was clearly a prime motive behind them.” He considers the construction of an enceinte around Miletus in Mycenaean times as “significant”, and adds that the main concern for sites on the islands, like Zagora, was piracy (pp. 32–33).
Most scholars follow Snodgrass in arguing that fear of attack was the main reason for communities in Early Greece to construct walls around at least part of their settlements. For example, Louis Rawlings, in his book The Ancient Greeks at War (2007), begins his chapter on fortifications and siege warfare with the unambiguous statement that “The construction of defences in stone implies the perception of significant threat (real or imagined)” (p. 128).
Naturally, it would be foolish to deny that military concerns (“fear”) were a motive behind the construction of fortifications. My aim here is rather to qualify that statement and add some necessary nuance.
Fear from natives and pirates
The distribution of the earliest post-Mycenaean fortifications in the Aegean seems to confirm the view that threats posed by hostile natives and seaborne raiders were the main impetus behind the constructing of defensive works. As regards the former, it has been suggested that the settlers of Old Smyrna built their circuit wall out of fear from the natives whose country they had just invaded (see, for example, Fredrick Winter’s Ancient Greek Fortifications (1971), p. 20). However, Smyrna is exceptional at this early date, being completely fortified as soon as it was founded (even though some deny the earliest walls actually served a military purpose).
Most Greek towns in Asia Minor did not, in fact, construct walls until sometime after they had been founded, which suggests that there was no immediate pressing need for them. A similar pattern emerges when we considers the Greek colonies in Magna Graecia (Southern Italy and Sicily): here, too, not all cities were fortified, and most walls were built only when the settlements had already been firmly established. In addition, a few sites had been continuously inhabited from the Late Bronze Age onwards (such as Ephesus and Miletus). In other cases, Greek traders and other visitors may have already frequented a particular area before it was colonized, so that Greeks may not have been considered as wholly foreign invaders in the first place.
Fear from seaborne raiders seems like a plausible reason for coastal sites to construct fortifications, especially in the period down to around 700 BC, when piracy seems to have been very common based on, for example, vase-paintings from Attic Late Geometric pots that often show battles on and around (beached) ships. In the centuries following the destruction of the Mycenaean palaces, many coastal settlements on Crete, but also on the Cyclades, were abandoned in favour of more defensible sites located inland.
However, a survey of all known fortified sites of the Early Iron Age and Archaic period reveals that most settlements were located in places that were relatively easy to defend, especially hilltops. The number of fortified settlements that are sited entirely on level ground is so small as to be virtually negligible, such as Halae in East Locris and Abdera in Thrace. The bulk of the known fortified places, some eighty sites or so, were located on hilltops or plateaus on the coast, with an additional fifteen sites located on hilltops at a distance of several kilometres from the coast, such as Argos and Agios Andreas on Siphnos. For a convenient catalogue of sites, refer now to Rune Frederiksen’s Greek City Walls of the Archaic Period (2010).
Continuity from the Bronze Age
Curiously, the supposed distribution of fortified sites as canonized by Anthony Snodgrass wholly ignores the continued use of fortifications originally constructed during the Mycenaean era. (Admittedly, he’s not the only one to do so; Frederiksen, cited above, also doesn’t engage much with the Bronze Age sites.)
After the fall of the Mycenaean palaces around 1200 BC, not only did many Bronze-Age fortification walls remain visible, some continued in use, although no new walls of the characteristically Mycenaean “Cyclopean” type – characterized by the use of massive boulders – were constructed from scratch. These sites include Salamis and several Cycladic islands; some continued in use until Protogeometric and even Geometric times.
The continued use of these sites suggests that it is perhaps unfair to draw general historical conclusions about the nature and aims of fortifications in Early Greece without taking local and regional considerations into account. Furthermore, continued excavations have revealed new fortifications in the Peloponnese and Central Greece that date to before 700 BC, of which Asine is a good example.
Other reasons for building fortifications
Furthermore, the notion that fortifications were primarily constructed as a result of some kind of external threat also ignores or downplays another possibility. A study of the distribution patterns of fortified sites reveals that they generally appear in clusters. Especially notable are the relatively large number of fortified sites in Ionia – an important Greek region on the western coast of Asia Minor – and on the islands. This seems to suggest that fortifications were constructed as the result of contacts with neighbouring polities, more specifically the result of what is referred to as peer-polity interaction.
Frequent interaction in the form of diplomatic exchanges, trade and warfare between more or less autonomous communities of comparable might and influence may have led to status rivalry; fortifications are one of the archaeologically visible traces of such competitive behaviour on the level of entire settlements. For example, it has plausibly been suggested by Jan Paul Crielaard in his contribution to Kurt Raaflaub and Hans van Wees’s Companion to Archaic Greece (2009) that Greek cities in Asia Minor not only competed among themselves, but may also have built their walls in competition with fortified Lydian towns further inland (p. 365).
Excavations at the Lydian capital of Sardis have revealed a section of wall, dated to either the seventh or sixth century BC, that is very similar to the fortifications at Old Smyrna, consisting as it does of a large stone sockle executed in polygonal blocks on the west side and ashlar blocks on the east, some 30m in length and a staggering 18.5m in width, topped by a mudbrick superstructure.