The Strategic Purpose of the Dema Wall.

The Dema wall was built to enable an army to prevent an enemy force in the plain of Eleusis from entering the plain of Athens. Since all other passes between these plains are considerably narrower than the Aigaleos-Parnes gap, only this pass needed to be fortified in order to assure the defenders a decisive advantage over the attacking enemy. Such a defensive stance clearly implies that the enemy was known to be numerically superior and therefore too powerful to meet in pitched battle in the plain. In order for the Dema to have been useful in any comprehensive plan of defense, there must have been no obvious routes toward Athens circumventing the Dema that were left open to the enemy. This

[3] 3. DEMA 176.

[4] McCredie 1966, 114-15. On the date of the Chremonidean War, see chapter l, note 29.

Map 5.
Routes and passes across the Attic-Boiotian-Megarian frontiers

implies that the enemy—did not have decisive control of the sea and that no major overland routes other than those crossing the plain of Eleusis were directly accessible to the enemy.

The only routes not crossing the plain of Eleusis that could give an invading army access to Athens and southern Attica are the routes by way of Phyle, Dekeleia, and Aphidna in the north of Attica.^[5] The Phyle route involves long stretches up and down steep slopes and through narrows within Mount Parnes and therefore would never have needed a fieldwork like the Dema wall to make it defensible. Likewise, the Dekeleia road, though not as long as the mountain way past Phyle, is sufficiently steep and rugged that it could be regarded as naturally defensible. The passes through Aphidna, however, on the route of the modern National Road, provide an open way into Attica that would be at least as easy for an army as the way through the Aigaleos-Parnes gap. Here, if the northern frontier were as exposed to danger as the western, one would expect that a barrier similar to the Dema would have been built. The pass on the north side of the Aphidna basin is the most defensible gap on the Aphidna route, and here the rocky slopes within the pass afford both the same sort of defensible terrain and the same durable building materials that were utilized in the Dema. The absence of any such fieldwork here, or anywhere else along this route, indicates that this northern quarter was not considered to be subject to the same threat of invasion as was the plain of Eleusis.

The outlook of the Dema tower lends support to this conclusion. From the tower, several outposts toward the western frontiers can be seen, but none toward the north. The outposts that do exist along the northern frontiers bear no resemblance to the Dema tower, are not intervisible with it, and almost certainly were not built at the same time as the system to which the Dema tower belongs.^[6] If, as has been argued, the Dema was the base for an Athenian army that was prepared to react to signals indicating the approach of an enemy force, then the absence of any signal connection with the northern frontier indicates that the enemy was not anticipated in that quarter. Like the wall itself, the Dema tower and its visual contacts indicate that the crisis which led to the con-

[5] For descriptions of these routes, see (for Phyle): Gell 1819, 48-57; Milchhoefer 1895, 10-11, and 1900, 31-32; Wrede 1924, 157-64; Chandler 1926, 4-6; Munn 1983, 110-16; Ober 1985a, 116-17, 185. For Dekeleia, see Milchhoefer 1895, 1-4; Chandler 1926, 15-16; Philippson 1951, 540-44, 546; Munn 1983, 107-10; Ober 1985a, 115, 184. For Aphidna, see Milchhoefer 1900, 14-15, 26; Chandler 1926, 16-17; Philippson 1952, 784-85; Munn 1983, 99-101,105-7; Ober 1985a, 114-15, 183-84.

[6] The square tower on Katsimidi and the enclosure on Beletsi, both north of Dekeleia, are the chief watchposts above the northern passes, and they are quite unlike the round towers associated with the Dema tower; see Munn 1983, 410-12; Ober 1985a, 142-45; Ober 1987a, 203-5.

struction of the Dema involved the possibility that Attica would be invaded specifically by way of the plain of Eleusis.

The main routes into the Eleusinian plain are the road from the Megarid and the Peloponnese along the coast, and the road from Boiotia across Kithairon by way of Eleutherai and Oinoe.^[7] Any army that was able to invade Attica from Boiotia by crossing Kithairon would likely have been able just as easily to enter Attica via Aphidna.^[8] Since no defenses seem to have been prepared along the latter route, it seems unlikely that the enemy was Boiotian or any power able to move at will through Boiotia. That being the case, the source of danger most likely lay in the Peloponnese.^[9]

If Attica in fact lay under threat of invasion from the Peloponnese, it may be asked why so considerable a fortification was built in the Aigaleos-Parnes gap rather than closer to the Megarian frontier. The answer emerges when we consider the nature and complexity of the routes through the western frontiers of Attica. An army coming from the Peloponnese could have used either of the two major routes entering the plain of Eleusis, the coastal route from Megara or the route through Kithairon, which was followed by Archidamos in 431. Moreover, the mountain terrain west of the plain of Eleusis is complex, and several minor routes might have proven useful to an invader seeking to outflank a defensive position.^[10] To build permanent fieldworks across any one of these routes would thus have been futile, while to fortify all of them would have been wastefully expensive. On the other hand, all routes through these western mountains pass through terrain where the ways are so constricted that practically no artificial barriers would have been

[7] On the main road from the Megarid, see Milchhoefer 1895, 19, and 1900, 40-41; Wrede 1934, 20; Munn 1983, 129-41; Ober 1985a, 128. On the main road from Kithairon to the Eleusinian plain, see Vanderpool 1978, 227-31.

[8] Archidamos, who led his army into Attica in 431 B.C. from the direction of Kithairon (Thucydides 2.18-19), evidently departed by the route through Aphidna, for we are told that on his way he ravaged the demes between Parnes and Brilessos (Pentele, Thucydides 2.23.1), but left Dekeleia unplundered (Herodotos 9.73.3). The Aphidna route is not known with certainty to have been used by any other ancient army, though either it or the Dekeleia pass must have been used by the Chalkidians in 506 (Herodotos 5.74.2), by Polemaios in 313 (Diodoros 19.78.:3-4), and by Philip V in 200 B.C. (Livy 31.24.1-8). On the eve of the liberation of Greece from Turkey, Reshid Pasha used the Aphidna route in his march from Chalkis to Athens in 1826 (Gordon 1832, 331-32).

[9] This was the conclusion also reached by Bow 1942, 208: "The enemy's route is so well determined, or he is so close at hand, that the Athenians are sure he will not turn and come down through Phyle or round Parnes past Dekeleia. Probably, therefore, the enemy is from Peloponnesos."

[10] On the variety of potential routes through the mountains of the Megarid and northwestern Attica, see Munn 1983, 97-98, 116-43; Ober 1985a, 121-29, 186-88; and Spence 1990, 94-96.

needed to add strength to a defending force's position. If a stand on this frontier seemed desirable, strong positions could be held without fortifications.^[11]

The construction of the Dema wall in the Aigaleos-Parnes gap, therefore, is intelligible in terms of a strategy for the defense of Attica against an invasion from the Peloponnese that might have provided for the positioning of advanced forces at key points on the western frontiers but that did not commit Athenian forces to a decisive stand either in the mountains of those frontiers, where they might have been outmaneuvered, or in the plain of Eleusis, where they might have been outnumbered. The decisive battle, if it came to that, would be fought as the enemy attempted to pass Mount Aigaleos, in which case the action would certainly center on the Aigaleos-Parnes gap, which was the most obvious way for a large army to enter the plain of Athens from the plain of Eleusis. It was through this pass that the army of Archidamos entered the plain of Athens in 431, and very likely it remained the route for invading Peloponnesian armies throughout the Peloponnesian War.^[12] Closing this gap would mean that no army from the Peloponnese could occupy and plunder the whole Attic countryside as the Spartans had in the Peloponnesian War.

The connection between Spartan strategy in the Peloponnesian War and the strategic concerns that led to the construction of the Dema wall is more than fortuitous. When we consider what has been deduced from the nature and placement of the Dema wall, together with its probable date as indicated by purely archaeological evidence, everything points to Spartan power as the threat to Attica that the Dema was designed to counteract. The only power from the Peloponnese whose forces on land were known to be numerically superior to those of Athens was Sparta, at the head of its Peloponnesian alliance. From the end of the Peloponnesian War until the third decade of the fourth century, Sparta was at the height of its power, and twice during that period, Sparta and Athens were at war.

In the first of these conflicts, the Corinthian War of 395-386, the Athenians chose to fight the Spartans because they calculated that the war would be fought well beyond the confines of Attica, as in fact it was. With their resources committed to other strategies that, for defensive purposes, were consistently successful, it is extremely unlikely that the

[11] Defensive positions were held along the Kithairon frontier on several occasions during and after the Boiotian War; see Xenophon Hell . 5.4.14, 36-37, 47, 59; Pausanias 9.13.7. See also Xenophon Mem . 3.5.25-27 on the natural defensive advantages of the mountains on the frontiers of Attica.

[12] Thucydides 2.19.2, 20.4; cf. Plutarch Alk . 34.3.

Athenians would have diverted manpower and money to the creation of a defensive network within Attica itself.^[13] The second conflict with Sparta, the Boiotian War, was thrust upon the Athenians by a chain of circumstances that culminated in an actual invasion of Attica by the Spartan harmost Sphodrias. Thereafter, the course of the war was predictable. It was expected that Spartan armies crossing the Isthmus would march against Thebes, but they might also invade Attica, since the Athenians were now fighting together with the Thebans against Sparta. With Theban and Athenian forces planning to resist the Spartans in the vicinity of Thebes, the Athenians could feel reasonably confident that a Spartan army would not easily march through Boiotia to enter Attica from the north. The threat to Athens lay in the possibility that Agesilaos would repeat the strategy of his father, Archidamos, and invade Attica through the plain of Eleusis.^[14]

Here are circumstances that exactly match the evident strategic purpose of the Dema wall and the tower system associated with it. The Boiotian War began in 378, a date that tallies with the archaeological evidence favoring a date in the first quarter of the fourth century for the Dema. The tactics implied by the form of the wall, as discussed in chapter 2, are known to have been common at the time of the Boiotian War. The network of towers and outposts west of the Dema tower would precisely suit the defensive concerns of the Athenians after the lesson of Sphodrias' raid was learned. Finally, the closest known parallel to the Dema is the wall built around Thebes in the first year of this war, and the Athenian commander most closely associated with fieldworks of this sort was Chabrias, who distinguished himself for his tactical sagacity at Thebes at the outset of this war. The circumstantial case is therefore extremely strong that the Dema wall was built after the raid of Sphodrias to safeguard Athens during the Boiotian War and that Ghabrias had a significant role in the design of this fieldwork.

Two general points still require demonstration before this strong circumstantial case can be regarded without reservation as the explanation of the Dema wall. The first is that it be shown that the Dema wall and tower system accord well with what is known of the tactical and strategic

[13] See the further discussion below, pp. 115-17.

[14] On Athenian fears that involvement in the Boiotian War threatened Attica with invasion, see Xenophon Hell. 5.4.19; Plutarch Pel . 14.1; cf. Demosthenes 2.24, 4.3, 9.47, 20.76. On the raid of Sphodrias and preparations made by the Athenians as a consequence, see Xenophon Hell . 5.4.20-34; Polybios 2.62.6-7; Diodoros 15.29.5-8; cf. 15.26.1-4, and Deinarchos 1.39; Plutarch Ages . 24.4-26.1 and Pel . 14-15. On the conservative and predictable manner of Spartan warfare at this time, see Demosthenes 9.48 and Munn 1987, 133-38.

principles of territorial defense at the time of the Boiotian War, and the second is that it be shown that no other event within the archaeologically permissible dates has an equal or stronger claim on likelihood.