How effective were Greek fortifications of the period between ca. 1000 and 500 BC at protecting the people behind the walls? Were they supposed to stand up to a prolonged siege or was the presence of a fortification wall, regardless of quality, enough to deter all but the most ardent of would-be attackers?
Before we turn to a consideration of these questions, it will be useful to give a brief overview of the construction of early fortifications. By far the most common type of wall used by the ancient Greeks consisted of a stone sockle with a mudbrick superstructure, at least sometimes with timber-framing, and frequently coated with plaster. This type of wall, made predominantly of mudbrick, is not peculiar to the Early Iron Age and Archaic period, but was also used before and since.
Furthermore, we know from written sources that wooden palisades were used (e.g. Hdt. 8.51), but of these no traces have survived at all in Greece: the evidence is therefore firmly biased toward the use defences built at least partially of stone. Moats were not used, but we know that in some cases the walls were reinforced with trenches, such as at Vroulia on Rhodes and at Samos (cf. Hdt. 3.39).
Overview of Early Greek fortifications
Warfare in the period between 1000 and 700 BC presumably consisted, for the most part, of relatively small-scale raids. What were fortifications in this period like? Let us focus on sites that have been reasonably well investigated. Some of these are relatively slight, such as at Vathy Limenari on the island of Donousa and Melia in Ionia.
However, most of the other early fortifications are more impressive, such as at Agios Athanasios in Phocis, Asine in the Argolid, Hypsili and Zagora on Andros, Iasus in Caria, Minoa on Amorgos, and Xobourgo on Tenos. Exceptional are the well-known massive fortifications of Old Smyrna. Of note, too, are the hilltop fortifications at Emporio on Chios; this acropolis was probably only used for refuge by the people living in the settlement further down the slopes. Similarly, Phaestus on Crete may also have been a refuge site.
All of these sites were located on or near the coast and sited on hilltops or (peninsular) plateaus. The manner of construction of these fortifications, while perhaps not always as impressive as later examples, does not seem to bear out Anthony Snodgrass’ suggestion that these were in general constructed solely as a temporary measure (see Snodgrass’s paper on the significance of fortifications, published in 1986, p. 130), at least not in all cases.
From the seventh century BC onwards, the number of fortified sites increases steadily, including new fortifications located further inland. Fortifications down to around 600 BC tend to be simple as far as their overall structure is concerned, with notable increases in sophistication especially from 550 BC onwards. Typically, early walls only feature a single tower. Many fortified sites have the tower close to the main gate, and in a few instances both tower and gate may have been fortified to create a gatehouse or bastion.
An example of such a bastion is found at Zagora, where the gate is located on one end of the long stretch of wall built across the neck of the promontory on which the eighth-century settlement was located. A similar structure has recently been unearthed at the site of Palaioskiathos on the island of Skiathos and was also located at the far end of a stretch of wall. I was fortunate to be shown around the site by Prof. Mazarakis Ainian in July of 2011.
This heavy emphasis on reinforcing a single weak spot, rather than distributing defence evenly along the wall by constructing multiple towers, suggests that the architects behind these early fortifications assumed that the capabilities of possible attackers, and probably also the total number of combatants, were limited. In myth, such as the Trojan War and the story of the Seven against Thebes, armies tended to focus their assaults on the gates, rather than the walls. The relative lack of towers seems to suggest that archers were relatively rare in the period under examination.
A foray into the Homeric world
At this point, it may be illuminating to consider the evidence supplied by the Iliad when it comes to assaulting towns. In the epic world, walls are apparently a feature of most major towns (e.g., Il. 2.646 and 691). Walled towns, when attacked, are invariably taken by storm (Il. 4.239, 7.164). Ares, the war-god, is frequently referred to as a teichesipleta, “stormer of walls” (e.g. Il. 5.31), as well as a ptoliporthon, “sacker of cities” (e.g. Il. 20.152).
In order to capture a fortified settlement, Homeric troops needed to scale the walls and either surprise the enemy or overwhelm them with superior numbers. This suggests that most walls in the epic world were not that high. One portion of the walls of Troy are specifically singled out as being easier to climb (Il. 6.433–434), perhaps because it was made largely of mud brick or rough stones, featured a more gentle slope, or maybe because it was not as tall as the other sections.
Siege engines of any sort seem to be unknown in the Homeric world, which echoes history. There is no evidence for the use of battering rams in Greece until the fifth century BC. Instead, some heroes use boulders to smash gates or walls (Il. 12.445–466), and stakes are employed as levers to topple enemy battlements (Il. 12.257–261).
As a brief aside, we may note that siege ladders are mentioned in a fragment attributed to Archilochus (fr. 98.16 West), ca. 650 BC. Siege ladders are also depicted on the Amathus bowl, a silver bowl of Phoenician make that depicts a siege that includes what are probably Ionian mercenaries.
But despite the lack of any apparent siege apparatus, siege warfare or, more specifically, the storming and sacking of cities, appear to have been among the most common military activities in the Homeric world (e.g. Il. 1.19, 2.728, 9.327–328). It is certainly no coincidence that war is represented on the new shield of Achilleus by a city under siege (Il. 18.509–512).
According to the epics, a siege could end in one of three ways. Firstly, the attackers could succeed in storming the walls and capturing it. Secondly, the defenders might be able to hold out long enough for the attackers to give up. Thirdly, a city under siege could also offer a bribe to their attackers. Sensing his impending doom, Hector laments that he did try to find a peaceful solution. Why, he wonders, did he not go and meet with Achilleus? In Lattimore’s translation, he adds,
“and promise to give back Helen, and with her all her possessions, all those things that once in the hollow ships Alexandros brought back to Troy (…); to give these to the Atreus’ sons to take away, and for the Achaians also to divide up all that is hidden within the city, and take an oath thereafter for the Trojans in conclave not to hide anything away, but distribute all of it” (Il. 22.111–122; see also 509–512).
It was also possible for the besiegers to demand a bribe from their victims, which is what the army on the shield of Achilles was contemplating (Il. 18.509–512). Perhaps another option to break a siege was for the defenders to call for help: smoke signals during the day and signal-fires at night were used to alert neighbours to their plight (Il. 18.207–213).
The threat of Lydia and Persia
The number of sieges predating the Persian Wars and attested in texts or through archaeological inquiry is small, which may suggest that most fortifications provided adequate defence, either in an actual siege or, perhaps more likely, as a deterrent to attackers. For example, we know that Argos captured the town of Asine toward the end of the eighth century BC, as attested by evidence of burning and destruction. The town of Abdera in Thrace was settled in two waves, firstly by people from Clazomenae and then by colonists from Teos. Remains of an early wall show signs of destruction dated to the late seventh or early sixth century BC, which is thought to coincide with the arrival of settlers from Teos, who then rebuilt the wall in the late sixth century.
Most other known sieges of the Archaic period appear to have been the work of the Lydians and, subsequently, the Persians. For example, Ephesus was besieged shortly after 560 BC by the Lydians (Hdt. 1.26). Earlier, in 600 BC, the Lydian King Alyattes besieged Old Smyrna and managed to capture the city by erecting a huge siege mound in order to scale the walls.
The growing threat of the Persian Empire has been regarded as a prime reason for the increased sophistication of Greek fortifications from the second half of the sixth century BC onwards. Early fortifications typically possess few towers (usually only one), but toward the latter stages of the period under examination there is a noted increase in the number of towers attached to walls.
This development can be traced, for example, at Eleusis. The town is located on a limestone hill that may have been fortified already in Mycenaean times. However, only part of the town was fortified. The Geometric wall, with a sockle of crude polygonal stones, was replaced by a more monumental fortification in the sixth century BC, around the time of the Athenian tyrant Pisistratus. This wall had a foundation of flat limestone slabs on top of which a sockle of trapezoidal blocks was laid with a height of up to 1.40 metres. The superstructure consisted of mudbrick, portions of which have been exceedingly well preserved. This wall featured a number of gates and at least eight rectangular towers.
Herodotus’ Histories features a number of unsuccessful sieges in the period shortly before and during the Persian Wars; in these cases, the presumably improved Greek fortifications proved sufficient to withstand attack. For example, the Athenians once besieged the fortified cities of Callipolis, Leontini, Naxos, and Zankle (all located in Magna Graecia), but failed to capture any of them (Hdt. 7.154). In 489 BC, the unlucky Athenian commander Miltiades besieged the city of Paros, telling the islanders that he would not leave until they were dead or had paid him a hundred talents. A short while later, Miltiades hurt his leg while leaping over a fence. He abandoned the siege and ultimately died of the wound on his leg (Hdt. 6.133–135).
Herodotus’ account also reveals that inhabitants of presumably undefended towns would abandon their homes when faced by invasion. For example, the Byzantines and Chalcedonians left their cities and fled inland, where they founded a new city (Hdt. 6.33). When the Persians invaded Naxos, many of the people there had fled to the mountains. The few that remained were enslaved, after which the city and its sacred places (ta hiera) were burnt (Hdt. 6.96). Likewise, the Phocians tried to evade the Persians by fleeing to the slopes of Mount Parnassus, to the city of Neon, and to Amphissa (Hdt. 8.32); others fled to the islands, or left the Aegean altogether. Famously, most Athenians abandoned their city and fled to the island of Salamis in 480 BC (Hdt. 8.41), while a few barricaded themselves on the Acropolis.
So how effective were Early Greek fortifications? It is clear that many of them were monumental works of architecture. Their massive scale in many cases may have served as a deterrent to would-be attackers. But the evidence from Homer – if we are allowed to use it for the period in question – suggests that towns were frequently taken by storm and that therefore the defences could not hold against a prolonged and determined siege.
Siege techniques were simple: an attacker relied on superior numbers to storm the walls and overwhelm the defenders. The defences themselves were comparatively simple: there was no need, and possibly also not enough manpower, to build more intricate defences. Only when faced by the superior armies of Near-Eastern Empires did the Greek start to build more sophisticated defences and engage in more elaborate forms of siege warfare.