previous chapter
One The Study of Attic Fortifications
next sub-section

The Nature of The Evidence and the Nature of the Problem

In a lesson on generalship, Sokrates quizzes the younger Perikles on his knowledge of subjects that should be familiar to him: "Well, have you considered this, Perikles, that great mountains reaching Boiotia protect our country, through which the passes are narrow and steep, and that the interior of our country is divided by sheer mountains?" The point is conceded by Perikles, and Sokrates goes on: "Don't you think, then, that young Athenians armed with light weapons and occupying the mountains that protect our country could do injury to our enemies while providing a strong bulwark of defense to our citizens in the countryside?" (Xenophon Memorabilia 3.5.25-27). Elsewhere in the same dialogue, the young Glaukon, hopeful of becoming a leading statesman, is embarrassed when Sokrates shows him to have no real understanding of the importance of watchposts and the relative strengths of their garrisons in the countryside (3.6.10-11). Clearly, according to Sokrates (or, more properly to Xenophon, the author of these passages), such subjects should be thoroughly familiar to generals and statesmen alike.

Similar advice is given by Aristotle. Ideally, the territory of a state should be formed so that it is difficult for enemies to invade yet easy for its inhabitants to set forth from. It should also be easy to keep under surveillance (

), for a territory that is easily watched is easily defended. With all of these objects in mind, the advice of professional military men () should be consulted (Politics 1326b-1327a). Yet not just generals, but anyone who would take an active role as an orator in directing the affairs of state, must be familiar with the lay of the land and the positions and respective strengths of watchposts in the countryside (Rhetoric 1360a). Aristotle's advice, and


that of Xenophon as well, is not merely expository but derives from contemporary experience; for in their days the Athenians charged one of their ten annually elected generals with the defense of the countryside (

) and required that issues concerned with the defense of the countryside () be introduced for discussion in each of the ten annual mandatory meetings of the assembly ().[1]

From the middle decades of the fourth century onward, when our evidence (including the foregoing passages) becomes abundant, territorial defense emerges as an institutionalized concern of the highest order among the Athenians. Its importance in the

ranked along with discussion of the grain supply.[2] Athenian youths, the ephebes, upon enrollment and verification of their citizenship following their eighteenth birthday, entered a two-year course of military training that included garrison duty in Peiraieus and in the fortresses of the countryside.[3] The defense of Attica was, in a literal sense, the first duty of all Athenian citizens, and it was a recurrent issue in public debates.

At a certain level, such responsibility of citizens for the defense of the territory of their state was, and is, axiomatic and is therefore unremarkable. But when viewed in their historical context, the institutions of territorial defense attested among the Athenians of the fourth century do seem remarkably developed, especially by contrast with earlier practices as they may be deduced from the experiences of the fifth century.[4] Reviewing the conditions of warfare in Greece on the eve of the Peloponnesian War, when Attica would experience the fire and ax of Peloponnesian invaders, A. W. Gomme was compelled to ask, "Why were not the


strategy and the tactics of mountain warfare by light-armed troops developed in order to prevent the invasion reaching the plains?"[5] Xenophon, in the passage from the Memorabilia quoted above, seems to have anticipated Gomme's querying observation, and the Athenians of his day appear to have adopted measures to render Attica less vulnerable than it had been in the days of Perikles. But what, we must ask, was the full nature of those measures, and more important, what concrete effects did they have on the conduct of war and diplomacy by Athens in the fourth century?

Such questions should be easily resolved by a review of the abundant literary and epigraphic sources for Athens in the fourth century. Yet the answers are surprisingly elusive. Despite a great many texts that describe aspects of, or refer in passing to, the institutions of territorial defense, we have no ancient account that explicitly and comprehensively presents the methods and goals or general effects of the Athenian

Our sources take much for granted, and the gulf between us and them should not be underestimated. What we lack, most of all, are not facts about the deployment and armaments of men but an appreciation of what men and their armaments were meant to do.

Despite the absence of precise statistics, we are reasonably well informed about the normal deployment and armaments of men in defense of the countryside. Armaments, discussed in chapters 2, 4, and 5, are standard and, broadly speaking, predictable. Deployment, on the other hand, depends entirely upon local conditions and circumstances. Here the principal source of evidence is the archaeological remains of fortifications found throughout Attica. Defense is an art of waiting and watching and of making preparations in advance of the enemy. Since an attacker generally commands greater numbers and chooses the moment, an essential feature of the defender's routine preparation is the construction of fortifications to assure that those watching and waiting will be secure when the attack occurs. The mass, extent, and elaboration of various works of fortification are gross indicators of their relative importance in terms of what they protected and roughly determine the numbers of men committed to them under normal circumstances. By these indicators, the fortifications of the Athens-Peiraieus complex, some 29.5 kilometers of walls, demonstrate the vastly preponderant defensive importance of this urban complex over that of the garrison forts, none of which is even a twentieth the size of the Athens-Peiraieus complex.[6]

Outside of Athens and Peiraieus, the garrison forts of Attica dominated the defensive priorities of the Athenians. The locations of these


Map 1.
Attica, classical and Hellenistic forts and garrisons

forts establish the pattern of routine deployment of men engaged in the defense of the countryside. Patrols, the peripoloi , regularly made their way cross-country between these garrisons.[7] Smaller outposts and lookout stations were sometimes manned in addition to the major garrisons, but the garrisons of the permanent forts of Attica represented the chief commitment of manpower and resources in the

. The remains of these forts are therefore primary evidence in the process of recovering the defensive priorities and activities of the Athenians. For an appraisal of the issues of concern here, a brief survey of the classical garrison forts of Attica is in order.


Eleusis was the largest and clearly the most important of the Attic garrison forts. Lying close to the Megarian frontier, Eleusis was already substantially fortified in the late sixth century, and its circuit was rebuilt and enlarged in both the fifth and the fourth centuries, when it reached a perimeter of 1.35 kilometers (see figure 1).[8] Eleusis was a regular assembly point for expeditionary forces bound both for Boiotia and the Peloponnese (e.g., Thucydides 4.68.5, Xenophon Hellenika 7.5.15, Demosthenes On the Crown 177, 184). When, in the fourth century, war threatened Attica by land, Eleusis was likely to have been the regular headquarters of the general in charge of the countryside (

). By the third century this was certainly the case, for the dudes of the were then divided between a coastal command () and a frontier command, designated as .[9]

Dependent upon Eleusis in the command structure of the third century, Panakton (with a circuit of 480 meters) was the most important garrison fort on the northwestern frontier. This fortress overlooking the Skourta plain, a mountain-bound plateau between Parnes and Kithairon, was built in the mid fifth century, and although partially destroyed during the Peloponnesian War (Thucydides 5.3.5, 39.3), it was rebuilt and garrisoned more or less continuously for almost two centuries (figure 2). When, in the 340s, tensions were high along the much-disputed frontier with Boiotia, Panakton was the headquarters of a general, certainly the

, and the base for an expanded citizen levy called out to guard the frontier (Demosthenes On the Embassy 326, Against Konon 3-5).[10]

Near to Panakton, and possibly even older, the fortress at the deme of Oinoe (with a circuit of approximately 560 meters) was an important garrison post also on the northwestern frontier (Thucydides 2.18.2; cf.


Herodotos 5.74.2). Oinoe lay in another upland plain below Kithairon, along the main ancient road from Athens and Eleusis to Thebes (figure 3). The fortress was garrisoned by the Athenians during the Peloponnesian War, but like Panakton, it fell to the Boiotians (Thucydides 8.98). Literary and epigraphic sources are strangely silent about Oinoe thereafter, but it is likely to have had a history very much like that of Panakton. It certainly returned to Athenian hands not long after the Peloponnesian War, and its walls show clear evidence of substantial rebuilding, probably within the fourth century.[11]

Further to the west, near, or even within, Boiotia itself was Eleutherai. An Athenian dependency since the late sixth century, Eleutherai was not fortified until the fourth century.[12] The impressive and well-preserved walls of this fortress (a circuit of 860 meters), standing above the ancient (and modern) road to Thebes as it enters the Kithairon pass, have seemed to many to be the perfect embodiment of the defensive planning of the fourth-century Athenians (figure 4). Ironically, as with Oinoe, we know nothing of the history of this fortress in the fourth century. The obscurity of the fortress at Eleutherai has led some to doubt the identity of these remains (sometimes referred to by their modern name, Gyphtokastro) as Eleutherai and caused them to place here the name of one or another of the better-known forts of Attica. In the early nineteenth century, when the fourth-century date of its walls was not yet clearly established, Leake championed the view that this was Oinoe, besieged by Archidamos at the beginning of the Peloponnesian War.[13] By the early twentieth century, Beloch developed the view, still held by some, that this fortress was Panakton, whose long use by the Athenians is well attested.[14] When the urge to identify imposing walls with a well-known


fortress is set aside, the evidence unambiguously indicates that this fortress bore the name of Eleutherai.[15] Pausanias, who traveled this road from Attica to Boiotia in the second century A.D ., explicitly refers to the fortress of Eleutherai standing a little above the plain, in Kithairon, on the road to Boiotia (1.38.9; cf. 9.2.3). The walls of abandoned Eleutherai clearly impressed him as they have modern travelers, for they are the only fortifications he mentions in the Attic countryside.

Southeast of Panakton, within the folds of Mount Parries, lay the fortress at Phyle (figure 5). The way through Phyle from Thebes to Athens was made famous by the march of Thrasyboulos at the end of the fifth century. The natural stronghold (

, Xenophon Hellenika 2.4.2) occupied by Thrasyboulos and his men might have been the site of the later fortress, although it could just as well have been another of the many naturally defensible eminences in the area. It is certain, at any rate, that the small but well-built fortress at Phyle (perimeter of 260 meters) did not yet exist in his day but was a product of Athenian concerns for territorial defense sometime in the fourth century.[16]

Between Parnes and the coast of the Euripos facing Euboia, the town of Oropos provided a stronghold for the Athenians whenever Oropos was controlled by Athens.[17] Otherwise, by contrast with the northwest, the fortifications of this portion of the frontier with Boiotia seem slight. Only at Aphidna, where the prominent hill now known as Kotroni preserves slight remains of an ancient circuit wall (perimeter approximately 300 meters), is there evidence of an Athenian garrison post, although


good evidence is lacking for the date of this fort (figure 6).[18] Remains at Dekeleia, Katsimidi, and Ayia Paraskevi have been identified as Athenian forts, but for various reasons none of these identifications is plausible.[19]

The only other regular garrison post on this frontier lay well to the east at Rhamnous, on the northeastern coast of Attica (figure 7). The acropolis of this deme was fortified at least by the time of the Peloponnesian War, when it must have been one of the coastal garrisons main-rained by the Athenians during the period of the Spartan occupation of Dekeleia. The fortress at Rhamnous was substantially enlarged in the fourth century (to a perimeter of 940 meters), and its importance as a garrison post endured well into the Hellenistic era, as attested by the numerous garrison decrees that have been found there.[20]

The coastal fortress at Sounion (perimeter 790 meters) had a history similar to that at Rhamnous. Its establishment in the time of the Dekeleian War is attested by Thucydides (8.4), and like Rhamnous, it remained an important garrison post well into the Hellenistic era (figure 8). In both cases, the close connection between these fortresses and vital roadsteads on the sea lanes serving Athens accounts for the importance of these places.[21] Similarly, a maritime fort was established on the


Ayios Nikolaos peninsula at Thorikos (perimeter 850 meters) a little to the northeast of Sounion during the Dekeleian War (Xenophon Hellenika 1.2.1).[22] The use of this and another fort at Anaphlystos northwest of Sounion is noted briefly by Xenophon in the middle of the fourth century (Poroi 4.43). While perhaps not all of these maritime fortresses were continuously garrisoned in the fourth century, they would have been manned when war threatened the Attic seaboard. Likewise in the Hellenistic era, the maritime forts at Koroni, Vouliagmeni, and Kynosoura at Marathon played important roles in particular episodes, although these and certain other fortifications in Attica were more ephemeral in nature and not part of the garrison system of the fourth century.[23]

Other works of fortification, both enduring and ephemeral, were part of the defenses of Attica in the fourth century. The most remarkable of these is the barrier wall in the Aigaleos-Parnes gap, which is known by its modern Greek name, the Dema (

, "the link"), since it joins Aigaleos to Parnes (see figure 9). The ancient name of this wall (if it was not also , or ) is unknown to us, and the occasion of its construction is unrecorded in any extant sources. Although study within the past generation has considerably narrowed the range of speculation about its date, its uniqueness has given rise to wide-ranging speculation on the subject. Robert Scranton described it as "the most ancient known extant example of the art of fortification as practiced by the Classical Greeks."[24] His advocacy of a date in the eighth or seventh century, in the belief that it was a frontier defense of Athens against an independent Eleusis, was consonant with the opinions of many scholars in the first half of this century who felt that this peculiar work must predate the era of the Attic garrison forts of the fifth and fourth centuries.[25] Yet, for various reasons, the high antiquity of this wall did not seem supportable to others familiar with the arts of classical Greek fortification, and consequently, a great variety of dates and occasions have been suggested for the wall, ranging from Kleomenes' invasion of Attica in 506 to the time


of the Gallic invasion of Greece in the early third century.[26] The wall has also captured the imagination of those who live near it, who speak of it as a work of Theodoros Kolokotronis during the modern Greek War of Independence. As will emerge later in this work, there is more to this unstudied claim than patriotic boasting.

For those who have seen it, the Dema wall demands an explanation. It is a monumental work, "as ambitious a project, in its way, as the Long Walls," according to Scranton.[27] After walking its length (4.36 kilometers overall) and coming to appreciate the care in planning and workmanship that went into the wall, one is inclined to agree with Scranton on that point. Yet since no ancient reference to it survives, its place in history remains unknown, and we are unable to appreciate the conditions, the motives, and the means that brought it into existence, nor can we appreciate its effectiveness once it was built. Historians of Greek military architecture have noticed the Dema for the many peculiarities in its design, but otherwise it has remained historically insignificant by force of its obscurity.

That the Dema wall belongs to the era of the classical garrison forts of Attica was established by the fundamental study by J. E. Jones, L. H. Sackett, and C. W. J. Eliot, published in 1957. After a careful survey of the archaeological and historical evidence, they concluded that it was built within the fourth century, and they advanced 337 as the most probable date for its construction.[28] James McCredie, in his study of military camps in Attica published in 1966, suggested that the Dema wall could instead be associated with the Chremonidean War of 268-262.[29] These


two studies have provided the foundations for the common view of the Dema wall, which is summarized by C. W. J. Eliot in his article "Dema Pass," in The Princeton Encyclopedia of Classical Sites (1976):

The date of the Dema's construction cannot as yet be determined with any precision. What little evidence there is might seem to favor a date in the second half of the 4th c., but a date in the first half of the 3d must also be considered a possibility . . .. Without new evidence a choice between this or that event is probably unjustified.

Since the evidence indicates that the Dema wall belongs to a well-documented period of Greek history, there is no reason to believe that the event which occasioned this monumental undertaking is unknown to us. Rather, as Eliot observes, the evidence has seemed ambiguous as to which event it was. The arguments that have so far been advanced in favor of one event or another have by no means exhausted the historical possibilities, so it is possible to introduce a good deal of circumstantial evidence not previously considered, as well as new evidence of a more tangible sort. The site of the Dema tower, adjacent to the wall and demonstrably part of the same defensive scheme, has now been excavated, and finds there provide significant evidence for the date of the Dema wall. There is reason, therefore, to believe that a thorough review of the available evidence, both new and old, can in fact establish the place of the Dema wall in Athenian history.

Why is it important to do so? The Dema wall is a barrier wall, designed to close a pass against an invader. Unlike the circuit walls of garrison forts, it was not a regular post for a garrison. It was not, in other words, part of the routine defensive establishment of Attica represented by the garrison forts. It had some other purpose in the scheme of the

. Like the proverbial exception that proves the rule, the Dema wall holds the promise of illustrating, when it is properly understood, the purpose and functional limitations of its more numerous contemporary works of fortification, the garrison forts of Attica. Only then can we comment on the relationship between territorial defense and the conduct of war and diplomacy by the fourth-century Athenians.


The fundamental difference between a defensive barrier like the Dema wall and circuit walls, whether of the city or of forts in the countryside, is emphasized in a comment by Plato, writing, in the Laws , close to the middle of the fourth century. Speaking of his ideally constituted state, which shares certain features with Sparta but many more with Plato's own Athens, Plato's Athenian remarks:

Concerning [city] walls, Megillos, I am of the same mind as Sparta. I would let walls sleep in the ground and not wake them, for these reasons. First of all, as the poet's verse so aptly puts it, walls ought to be of bronze and iron, and not of stone. Secondly, our practice would be justly ridiculed when each year we sent out our young men into the countryside to block an enemy's path by ditches, entrenchments, and various constructions, all in order to keep the foe from crossing our borders, while at the same time we surrounded ourselves with a wall, which  . . . invites the inhabitants to seek refuge within it, and not to ward off the enemy.[30]

By "various constructions" (

), Plato refers to barrier walls like the Dema, although he carefully avoids calling them "walls" () because this is the usual term for the circuit wall of the city, whose employment he eschews. The important point here is the opposition clearly drawn between the effects of barriers in the countryside and circuit walls. To Plato, barrier walls and entrenchments are emblems of a laudable determination to resist an enemy in the open, while circuit walls pander to the craven instinct to fly for shelter in the face of an enemy. Plato is specifically discussing the circuit wall of the city, but here, as in every other instance in Greek literature where the defense of the city is contrasted with the defense of the countryside, the circuit walls of garrison forts do not form a separate, third category in the operations of the . Forts are commonly referred to as ,"walls," and as such they are clearly classified with city circuits in this dichotomy.[31]


Here we have a paradox. For the evidence we have considered indicates that the garrison forts of Attica were the most important elements of the institutions of territorial defense, yet at the same time they are functionally never distinguished from the walls of the city; and reliance on city walls means, in some sense, that territorial defense has been abandoned. This was in fact the case in Attica during the Peloponnesian War. For then the Athenians, under the leadership of Perikles, evacuated the Attic countryside and withdrew to the walls of the Athens-Peiraieus complex during enemy invasions and occupations of Attica, while keeping their fortresses in the countryside fully garrisoned. The Athenian cavalry, meanwhile, bravely skirmished with the overwhelming forces of the Peloponnesian army in an effort to limit their depredations, but no one could say that the Athenians were fully committed to the defense of their countryside.[32] What was

and did the Athenians of the fourth century construe it differently from Perikles?

The Dema wall is one clue leading toward the resolution of this paradox. But before we turn to a detailed examination of that wall and its function, we require a fuller understanding of the use of the garrison forts of Attica than that provided by the brief survey of forts and their chief testimonia above. We may usefully begin by considering modern views on the subject. Not that we will thereby find the matter readily clarified, for as will emerge, most modern treatments of the subject have introduced suppositions about the functions of fortifications that are not reflected in our ancient sources. We must inquire, therefore, as we follow the evidence, whether such modern interpretations are justified or whether our sources indicate some other interpretation that has not yet been generally apprehended.

previous chapter
One The Study of Attic Fortifications
next sub-section