troops together will permit us to understand the nature of ancient
city fighting and to assess the place of urban warfare in classical Greek
military thought. Finally, we put the classical experience into broader
historical context to see what lessons it may hold for today’s strategists
and battlefield commanders.
140 Lee
Types of Urban Combat
Classical literary sources preserve numerous episodes of sieges and as-
saults on city walls. They also describe assassinations, riots, and low-
level gang warfare inside cities. These phenomena merit study in their
own right, but here we will focus on large-scale armed clashes inside
city walls, where the combatants’ behavior was shaped by settlement
topography, not by fortifications. Within these limits, ancient texts fur-
nish dozens of accounts of city combat. Many of these accounts are
quite brief, but they permit us to distinguish several basic patterns of
urban struggle.
First, an attacking army might breach a city’s walls by assault, siege
engineering, or treachery, only to face continued resistance in streets,
houses, and public spaces. These were among the bitterest sorts of
city fight, often resulting in the complete annihilation of the defend-
ing force. Plataea in 431 and Thebes in 335 are just two examples of
this pattern. Not every successful siege or assault progressed to intra-
urban fighting. Sometimes, especially when surprised, defending forces
simply collapsed.8 Even so, urban combat during the capture of cities
likely occurred more often than classical texts might suggest. The city
of Olynthos in northern Greece, taken by Philip II of Macedon in sum-
mer 348, provides an instructive case in point. Although the literary
sources record only that certain wealthy Olynthians betrayed their fel-
low citizens to Philip, excavations in the ruins of Olynthos have uncov-
ered hundreds of lead sling bullets, arrowheads, and other weapons.
The distribution and context of these artifacts indicate that the Mace-
donians had to subdue Olynthos house by house.9 Future archaeologi-
cal investigation may someday reveal further instances of otherwise
unrecorded urban fighting of the classical period.
A second cause of urban combat was stasis, or civil strife, between
factions in a city.10 Such strife could spring from competition between
powerful families, from class-based hostility, or from the involvement
of outside interests. During the Peloponnesian War, antagonism be-
tween pro-Athenian and pro-Spartan factions was responsible for in-
ternecine bloodshed in cities throughout the Greek world. Corcyra
Urban Warfare 141
in northwestern Greece, site of the most notorious of these staseis,
underwent two years of civil strife that began with intensive urban
combat and culminated in the total annihilation of the losers and their
families.11 In other cities, factional clashes began with massacres in the
marketplace.12 Defeated factions that managed to escape often returned
to try their luck again, leading to renewed urban war.
Urban combat could also result when rebels or insurgents attempted
to eject foreign occupiers from their city. In 335 BC, for example, the
Thebans rose up against a Macedonian garrison stationed in their city.13
At other times the presence of a foreign garrison in support of a city’s
ruling faction could lead to an urban revolt intended to expel both the
foreigners and those who collaborated with them. The Athenian revo-
lution of 508–507 and the Theban uprising of 379 exemplify this sort of
situation. In both these cases, victorious insurgents allowed enemy gar-
risons to leave under a truce. Urban insurrections of this sort, although
not unknown in the classical world, would become more common in
the Hellenistic period (323–30 BC), when foreign garrisons were more
widely employed.
Invasion or civil strife occasional y resulted in opposing armies or
factions, neither in complete control of a city, confronting each other
within its bounds. So it was in the opening stages of the civil war at
Corcyra, where oligarchs and democrats held separate districts of town
and spent several days engaged in running street battles.14 While the
length of most urban clashes could be measured in hours or days, this
sort of struggle could devolve into chronic conflict, with a city semi-
permanently divided between warring sides, which might even construct
internal fortifications against each other. One such division occurred at
Notium in Asia Minor during the early years of the Peloponnesian War,
when hostile pro-Athenian and pro-Persian factions entrenched them-
selves in separate quarters of the city.15 Likewise, Syracuse in the late
460s was split between native-born citizens and rebel ious foreign mer-
cenaries, who for several years battled in and around the city.16
These rough categories by no means exactly describe every single
classical urban clash. Indeed, some urban battles featured combina-
tions of situations. At Sparta in 369, for example, King Agesilaus had
simultaneously to defend against a Theban assault and to squelch an
142 Lee
uprising by a group of disaffected Spartans.17 The Thebans in 335 had
just succeeded in regaining their city from its Macedonian garrison
when they faced an external attack from Alexander’s main army. No
matter how they began, though, all urban clashes were shaped by the
characteristics of the ancient Greek city.
The Urban Battleground
The polis, sometimes translated as “city-state,” was the characteristic
political form of classical Greece.18 In physical terms the typical polis
consisted of a walled urban settlement surrounded by a rural hinter-
land. The urban center was built around its acropolis, a defensible high
point. Within the city walls could be found temples, public buildings,
a marketplace, and private dwellings. In the fourth century, Mantin-
eia, Megalopolis, and Messene incorporated open fields and croplands
within their fortifications, but such vast circuits were exceptional. Else-
where, suburbs sometimes extended beyond the walls.19 Larger poleis
featured small towns or villages in their hinterlands; poleis near but not
directly on the sea often developed harbor towns. With the exception
of Piraeus, which grew into a sizable town, none of these subordinate
settlements ever came close to approaching the urban center in size or
significance.
By modern standards, most poleis were tiny. The acropolis of Halai
in central Greece, for example, measured a mere 160 by 70 meters, and
the city’s entire walled area may have been only 0.85 ha (2.1 acres).20
Classical Halai probably had a total population of perhaps a few thou-
sand. Athens, with its hundreds of thousands of native Athenians plus
foreigners and slaves all living within several miles of walls, was excep-
tional. Whether their polis was large or small, most Greeks lived in the
countryside, not in the city.
The circuit wall of a city defined its urban space.21 Gr
eeks had begun
constructing fortified enceintes in earnest during the sixth century BC,
and by the end of classical times only a handful of major sites, notably
Sparta, remained unfortified. Most walls were built of massive stone
blocks, though bricks, clay, and rubble were also employed. Gates with
flanking towers and sometimes with elaborate entryways regulated
Urban Warfare 143
access to the city. Additional towers and bastions along the walls pro-
vided positions for defenders.
The urban battleground proper began just inside a city’s walls, but
that did not render walls superfluous. Even if they could not forestall
entry into a city, walls in urban combat could become inverse barri-
ers, as at Plataea, where the circuit wall kept numerous Theban at-
tackers from escaping.22 The inside edges of city walls also furnished a
secure backstop against which troops involved in urban combat could
regroup. City gates, too, remained important as access points for rein-
forcements. At Tegea in 370–369 BC, for example, the factions contest-
ing the city retreated to opposite sides of town after their initial clash.
The pan-Arcadian faction fell back under the city wall, near the gates
leading east to Mantineia, whence they expected to receive additional
troops. Their opponents clustered on the other side of town, near the
gates leading to Pallantion. When the pan-Arcadians were reinforced,
their opponents quickly fled west out the gates.23
Fortified citadels inside cities could also shape the course of urban
battle. Most cities had only one acropolis, but larger ones could con-
tain multiple strong points. Athens, for instance, had the Mouseion hill
near the Acropolis and the Mounichia hill in Piraeus, in addition to
its famous Acropolis.24 Defenders who retained an acropolis or other
fortress could use it as a base for counterattacks. At Syracuse in the
350s, for example, the mercenaries of Dionysus II launched assaults
from the fortified island of Ortygia against the rest of the city.25 Hold-
ing the acropolis, however, did not guarantee control of a city. The
popular revolutionaries at Athens in 508–507 BC successfully trapped
the oligarchs and their Spartan supporters on the Acropolis.26 At Sardis
in 499, the Persians held the acropolis, but could not prevent the Athe-
nians and Ionians from ravaging the town below.27 When the Thebans
in 335 BC regained control of their city, they left the Macedonian gar-
rison bottled up on the Cadmea, Thebes’s acropolis.28 In chronic intra-
urban conflict, as we have seen, rival factions or communities might
rely on internal cross walls to bolster their positions.29 Such walls could
pen urban combatants into a narrow slaughter pen with no room for
maneuver, as happened at Syracuse in 357–356.30
144 Lee
The real nerve center of the classical city was the marketplace or
agora. Located at the junction of major streets and often containing
major administrative buildings, the agora was the largest open area
within the city walls. Foreign attackers entering a city usually headed
straight for the agora, and defenders typically fell back toward it.31 If
the defenders could hold on to the agora and reform their troops, they
stood a chance of pushing the attackers out of the city. The Athenians
and Ionians at Sardis in 499 BC, for example, were forced to fall back af-
ter they encountered Persian troops massed in the agora.32 Conversely,
the loss of the agora could be the final blow that crushed defenders’
morale.33 Even so, overconfident or outnumbered forces, like the The-
bans at Plataea, might find that taking the agora alone was insufficient.
Many civil wars began with coups or massacres in the agora.34 Again,
winning the agora did not guarantee victory, as the oligarchic party dis-
covered at Elis in 397 BC. Having seized the agora, the oligarchs declared
victory, only to find that Thrasydaios, leader of the popular faction, was
not dead but just at home sleeping off his midday wine. Shaking off a
hangover, Thrasydaios led a counterattack that routed the oligarchs.35
In addition to being communication centers and rallying points,
marketplaces could contain vital arms supplies for urban combatants.36
The conspirators who allegedly sought to seize power at Sparta in 400–
399 BC, for example, had planned to use Sparta’s tool market, with its
abundance of axes, hatchets, and sickles, as their arsenal.37 At least one
other city was taken by insurgents employing weapons that had been
smuggled into the agora inside baskets of fruit and boxes of clothing.38
Forgetting the dangers of an urban armed mass, the Spartan officer
commanding the defense of Mytilene in 427 made the mistake of arm-
ing the city’s populace, which promptly rebelled against him.39
Beyond the agora, any spacious and defensible location where com-
batants could form up or find refuge was tactical y important. These
places included theaters, temples, gymnasia, and other large buildings.40
During the Athenian civil war of 404–403, oligarchic horsemen used the
Odeion of Pericles, a meeting hal just below the Acropolis, as their base,
while the democratic light infantry gathered at the theater in Piraeus.41
Like marketplaces, temples and public buildings could furnish arsenals
Urban Warfare 145
for urban combat. At Thebes in 379, the anti-Spartan forces equipped
themselves with weapons, probably religious dedications, taken from a
portico.42 Given enough time, defenders might dig trenches across open
areas, or sow them with obstacles to impede an enemy advance.43
Large buildings promised security but could become death traps.
During the final stages of the Corcyrean civil war, members of the oli-
garchic faction, knowing they were about to be executed, tried to hold
out in what may have been a warehouse. Their enemies climbed atop
the building, broke open the roof, and rained down tiles and arrows; the
defenders who survived the barrage kil ed themselves rather than sur-
render.44 Something similar happened at Tegea in 370–369, when mem-
bers of a defeated faction took refuge in the temple of Artemis. Their
opponents surrounded the temple, climbed up, dismantled its roof, and
hurled down tiles. The men inside gave up, only to be put to death.45
Urban war also meant street fighting. The oldest Greek towns had
grown up organically over the centuries and so did not have regular
layouts. The irregular web of narrow streets and alleys that crisscrossed
these cities could confuse and disorient foreign invaders—think again
of the Thebans at Plataea—while defenders who knew the shortcuts
could move quickly from one neighborhood to another. Irregular
street networks forced commanders to split forces into small detach-
ments, making communications and mutual support nigh impossible.
With attackers and defenders split into uncoordinated small groups, a
street battle could last all night, with troops slaying each other at ran-
dom in the darkness, as happened at Syracuse in 355.46
By the mid-fi
fth century, regular street grids became popular for
new cities and for expansions to old ones.47 Street widths in these grids
could range from 3–5 meters (9.8–16.4 feet) for residential byways to 13–
15 meters (42.6–49.2 feet) for main thoroughfares.48 As Aristotle noted,
cities built in this new “Hippodamian” style made for more convenient
and pleasant living but for less security in war.49 To keep a city defen-
sible, Aristotle recommended that planners use grids only in certain
neighborhoods, or lay out blocks with a few wide avenues connecting
to smaller streets.50 A regular street plan made matters easier for at-
tackers, who could send mutually supporting detachments up parallel
avenues with less risk of getting lost. In response, defenders could dig
146 Lee
pits or trenches in streets and set up barricades. They might also bur-
row through house walls to outflank enemy forces.51
Even in grid-planned cities, narrow streets compelled command-
ers to draw up troops in unwieldy formations. In Piraeus in 404–403,
for example, the oligarchs had to form their hoplites fifty ranks deep.52
A regular street grid also provided missile troops better fields of fire.
The oligarchs at Piraeus were able to take the agora, but when they
advanced up a main avenue toward the Mounichia hill, the democrats
threw them back with a volley of stones, javelins, and arrows.53
In grid-planned cities, houses were built in blocks sharing common
walls, sometimes with a narrow alley running down the center of the
block. As in modern subdivisions, houses in each block often shared a
similar design. Houses in planned districts could be spacious. Plots on
the North Hill of Olynthos, for example, average about 17 meters on a
side.54 In older cities, houses were often smaller and house layouts less
regular. New or old, houses were perhaps the most difficult of classical
Greek urban terrain. From Sicily to Ionia, the typical house was of mud
brick on stone foundations.55 It faced inward, with a narrow entry giv-
ing access to a central courtyard around which rooms were arranged.
Exterior-facing windows were high off the ground and generally in-
accessible. Some houses had second stories, often used for women’s
quarters. Houses generally had pitched, tiled roofs, although in some
Makers of Ancient Strategy: From the Persian Wars to the Fall of Rome Page 23