regions flat roofs were preferred.
Unlike pitched field battle, urban warfare took place in three dimen-
sions. In city fights, house roofs provided a vital height advantage. Roof
tiles, which could weigh from 10 to 30 kg, provided ready-made projec-
tiles for defenders to hurl down upon invaders. Even women and slaves
ascended the roofs of their homes to assail advancing enemies with
such missiles.56 Sometimes other structures gave a height advantage. A
Theban attempt on Corinth in 369 BC was repelled by light troops who
mounted burial monuments and grave markers to hurl stones and jav-
elins.57 Attackers too used roofs as firing platforms, as the Thebans did
when taking Oeum from the Spartans in 370–369.58 Rooftop positions
were not invulnerable. Boeotian troops defending Corinth in 393, for
example, climbed to the roofs of ship sheds and warehouses, only to be
trapped and killed there.59
Urban Warfare 147
With their narrow doorways and solid construction, private homes
could become fortresses of last resort for a defending population. A
city might be declared secure once the agora and public buildings were
taken, but inhabitants determined to resist could still force an attacker
to root them out house by house. If neighbors cut through shared
walls to link up with each other, they could convert an entire block
into a final redoubt. House-to-house combat was dangerous and dif-
ficult. Beyond every darkened doorway and around every blind corner
might lurk a desperate enemy ready to fight to the last. At Olynthos,
the distribution of excavated sling bullets and arrowheads suggests that
Macedonian attackers had to force their way into house courtyards,
only to be shot at from the surrounding rooms. The Macedonians ap-
parently responded with volleys of their own missiles before moving in
to clear each room.60 Thirteen years later, the Macedonians probably
faced a similar situation at Thebes. After Alexander’s forces seized the
city’s key points, some Theban infantry fled to their houses, where they
and their families fought and died.61
In addition to posing tactical difficulties, combat in houses threat-
ened an invading army’s discipline and cohesion. Soldiers who turned
aside from the fight to loot, rape, and pillage were useless for further
combat. Worse, they might be surprised by a counterattack. At Syra-
cuse in 355 BC, for instance, Dion and the Syracusans caught enemy
mercenaries in the act of plundering and utterly routed them.62
Houses had such defensive potential that they were sometimes in-
corporated into fortification architecture. Motya in Sicily, for example,
featured multiple-story houses near its northern gate. During the
Greek capture of the city in 397 BC, Carthaginian defenders employed
these houses as a second line of defense.63 When Philip of Macedon
tried to force his way into Perinthos in 341–340, the defenders turned
their homes into impromptu fortresses, blocking streets and alleys to
stymie the Macedonian advance.64 Plato advocated that houses “be ar-
ranged in such a way that the whole city will form a single wall; all the
houses must have good walls . . . facing the roads so that the whole city
will have the form of a single house, which will render its appearance
not unpleasing, besides being far and away the best plan for ensuring
safety and ease of defense.”65 Plato’s suggestion is reflected at Olynthos,
148 Lee
where the backs of the first row of houses along the west edge of the
North Hill are built into the northwestern fortifications of the city.66
Sparta’s unusual topography turned the two Theban attacks on it
into hybrids of open battle and city fight. Classical Sparta was unwalled
and spread out along the banks of the Eurotas River. At its outer neigh-
borhoods, houses were interspersed with groves and fields. The central
area of town, where the Spartiates or full Spartan citizens lived, seems
to have been densely built up, without a regular plan. Even so, the cen-
ter of town contained walls, fences, and open spaces. Around the town
center were a number of religious sanctuaries and public buildings.67
In 370–369, the Thebans under Epaminondas initially confined them-
selves to plundering the suburbs. There they felled trees to build field
fortifications wherever they camped, just as they did in rural terrain.
Eventually the Thebans took a stab at the heart of the city, advancing
toward the open racecourse in the sanctuary of Poseidon. In response,
the Spartans used the urban topography to their advantage, by setting
an ambush in the Temple of the Tyndaridae.68 The ambush, combined
with a conventional cavalry charge across the racecourse, halted the
Thebans. In 362 BC, fearing a direct assault, the Spartans prepared by
demolishing houses in the central part of town and using the rubble to
block up entrances, alleys, and open spaces. Some even alleged that the
Spartans used large bronze tripods taken from religious sanctuaries to
build barricades.69 Epaminondas, however, did not attempt a head-on
attack, fearing that his troops would be exposed to missile attack from
rooftops.70 Instead, he took an indirect approach, as if maneuvering on
an open battlefield, which allowed his forces to advance on the inhab-
ited area without coming under missile fire. Only a desperate counter-
attack of fewer than a hundred Spartans under King Archidamus threw
the Thebans back.
The Combatants
The equipment, formations, and command structures of classical Greek
armies were ill-suited for built-up terrain. Hoplites, the mainstay of all
polis armies, were armored militia infantry who carried large round
shields and long thrusting spears. Although the hoplite shield has been
Urban Warfare 149
judged heavy and unwieldy, there are some indications it could be han-
dled effectively in individual combat, even in cities; fourth- century tomb
reliefs from Asia Minor even show shield-carrying hoplites climbing as-
sault ladders. Although there is no certain evidence, possibly hoplites
fighting house to house discarded their shields for greater maneuver-
ability. The more serious problem for hoplites in cities was weaponry.
While they carried swords as secondary weapons, hoplites were primar-
ily spearmen; their 2.5-meter-long spears would have been awkward at
close quarters or inside houses. Matters would have been even worse
for Macedonian troopers equipped with the 12- to 16-foot-long sarissa,
or pike. Some Greeks did study swordsmanship, but systematic weap-
ons training remained the province of a wealthy few. Indeed, outside of
Sparta, most hoplites underwent no formal training until the end of the
classical period.
A greater challenge for hoplites in cities was battle formation. Hop-
lites typically employed a deep infantry array called the phalanx. The
ideal phalanx, an unbroken mass eight ranks deep, could extend a mile
or more across an open battlefield. Needless to say, a phalanx could
not be maintained on city str
eets or in houses. Only in an agora could
hoplites employ their customary formation. Splitting a phalanx into
smaller detachments to cope with urban topography was complicated
by the general lack of subordinate units and officers. With the excep-
tion of the Spartans, who had a complex tactical hierarchy and an al-
most religious devotion to good order, most Greek armies had a very
low proportion of officers and no tactical units below the company
level. What officers there were could be rendered impotent by the lax-
ness of classical military discipline.71
The amateur ethos of most polis armies had other important conse-
quences for urban war. For one thing, the Greeks never developed units
of specialists such as pioneers, sappers, or combat engineers. Hoplites
could and did build improvised field works, but their engineering skills
and equipment never came close to matching those of the Roman le-
gions. At the same time, because hoplite militias equipped themselves,
a wide proportion of citizens owned arms. Fighting in a city, whether
as a result of invasion or civil war, typically involved the whole popu-
lace, not just regular armed forces.
150 Lee
Light infantry, including archers, slingers, and javelineers, was much
more effective in urban fighting. Light troops could hurl missiles from
rooftops or sweep the streets with volleys of projectiles.72 The archaeo-
logical evidence from Olynthos indicates that slingers and archers could
wield their weapons even inside the confines of houses.73 Light troops
proved their value during the fighting at Piraeus in 404–403. The oligar-
chic forces, with hoplites enough to mass fifty shields deep, advanced
up the Mounichia hill toward the democrats, who were able to muster
only ten ranks of hoplites. Behind these ten ranks, though, were the
democratic light infantry. The hilly topography of Mounichia gave the
defenders a height advantage, allowing the light infantry to shoot over
the heads of their own hoplites. With their opponents packed fifty deep
into the street below them, the light troops could hardly miss.74
The role of cavalry in urban battle is difficult to determine. The
Athenian Thirty Tyrants apparently brought a sizable cavalry force
to Piraeus in 404–403, but the horsemen played no role in the battle.75
They may have deployed in Piraeus’s agora to guard the rear of the
oligarchic hoplite force. Theban cavalry participated in the fighting at
Thebes in 335, although they were hampered by the narrow streets, and
quickly fled once the Macedonians captured the agora.76 The Roman
writer Pausanias saw a trophy near the Painted Stoa at Athens, just out-
side the agora, that commemorated an Athenian cavalry victory there
against Macedonian cavalry, probably in 304 BC.77 At least the classical
Greeks did not deploy elephants in urban fighting. Pyrrhus of Epiros
would later try to do so at Argos in 272, only to discover that his troops
had to remove the fighting platforms from the elephants in order to
pass through the city gates.78
Urban Warfare and Classical
Military Thought
Assessing the place of urban warfare in classical military thought re-
quires understanding the centrality of walls in the polis mind-set.
Building a circuit wall was the largest and most expensive communal
task that the citizens of most poleis would ever undertake.79 Once built,
walls marked polis identity and autonomy. Plato might advocate “walls
Urban Warfare 151
of bronze and iron” rather than earth, but when it came to defending
their cities, Greeks never ignored the practical value of fortifications.80
Indeed, while classical warfare has been portrayed as an agonal affair
that valorized open battle over sieges and stratagems, by the mid-fifth
century BC the idea of a defensive strategy based on impregnable city
walls was well established at Athens.81 Athens was exceptionally well
prepared for such a strategy because it could draw supplies from its
overseas empire. Nonetheless, the citizens of smaller poleis considered
staying behind the walls a perfectly normal defensive move, especially
when faced with a numerically superior invader. They would choose
open battle only if the numbers were even. In fact, close analysis of the
Peloponnesian War reveals that sieges and city assaults were twice as
common as pitched battles.82
The stock that classical Greeks placed in their walls is reflected in the
absolute panic that sometimes overcame defenders when they realized
enemy forces had penetrated the bounds of their city. Not even the
Spartans, who prided themselves on their lack of city walls, were im-
mune to this reaction: in 370–369, both men and women panicked at the
appearance of Thebans in the suburbs.83 Given the expense involved
in building a city wall, and the psychological value attached to main-
taining its integrity, it is no surprise that fighting inside city walls was
almost always undertaken out of necessity rather than as a strategic
choice. Tellingly, our ancient sources preserve only a single reference
to troops deliberately abandoning their walls in order to fight inside
their city. This was at Pharcedon in Thessaly during the mid-fourth cen-
tury BC, where the defenders unsuccessfully attempted to draw Philip’s
Macedonians into an urban ambush.84
Greek military thinkers were probably also disinclined to adopt ur-
ban battle as a preferred mode of warfare, because it upset accepted
gender and status hierarchies. The classical citizen ideal emphasized
warfare as the exclusive realm of free males. Women and slaves were
supposed to stay indoors, secure within the walls of the household.
Combat inside cities, however, upset the masculine dominance over
war, not to mention the notion of the household as an inviolable pri-
vate space. It is notable that accounts of urban battle prominently men-
tion the active participation of women and slaves.85 As well, city fights
152 Lee
favored the poor and unarmored over middle-class hoplites, challeng-
ing hoplite dominance of the battlefield.
Furthermore, Greek commanders understood that urban warfare
was particularly vicious and uncertain, even by ancient standards.
Women and children, along with combatants, were fair game. Treach-
ery, massacres, and fights to the death were commonplace. Urban to-
pography made battle more desperate, as troops confined in streets and
houses could not easily flee. Even soldiers inclined to grant quarter to
surrendering foes might hesitate to do so if they feared surprise attack
from another direction. The lack of communications and control in ur-
ban fighting meant that commanders had less opportunity to appeal to
the limiting rituals, such as truces, that moderated field battles. Fight-
ing at night or in bad weather exacerbated the effects of topography
and poor control. The nature of the combatants as much as the nature
of the terrain contributed to the brutality of city fighting. Opposing
factions in civil strife were implacably ho
stile; at Corcyra, infamously,
citizens set fire to their own city in an attempt to drive out rivals.86
Troops defending a city against external invasion, too, knew they were
fighting not just for their own lives but also for their families and for the
very existence of their home. For their part, attackers who gained ac-
cess to a city after a long siege or a bloody assault were primed to inflict
as much revenge or “payback” as possible on any inhabitant, armed or
not.87 All these factors made Greeks wary of fighting in cities.
There are nonetheless a few indications that Greek commanders
understood how to conduct urban warfare when they had to. The
Plataeans certainly were quick to take advantage of their city’s topogra-
phy to entrap the Theban invaders. In Piraeus during 404–403, since the
democrats did not have enough men to hold the entire circuit wal sur-
rounding the harbor, they deliberately concentrated on the Mounichia
hil , a strong point that could be approached only through the town’s
street grid. By deploying on Mounichia’s slopes, the democratic leader
Thrasyboulos maximized the defensive potential of the urban landscape
and exploited his preponderance in light troops to offset the oligarchs’
strength in hoplites.88 Epaminondas, one of the masterminds of the
Theban uprising of 379, was likewise wel aware of the complexities of
street fighting. Recognizing that the urban terrain of central Sparta was
Urban Warfare 153
not good for pitched battle between phalanxes, he avoided direct assaults
on the city center in both 370–369 and 362.89 At Syracuse in the 350s, the
general Dion made attempts to overcome the fragmentation of urban
fighting. He divided his troops into separate commands and grouped
them in columns, so that he could attack at several places at once.90
In the later classical era, urban warfare did receive some attention in
the writings of Aeneas Tacticus. Aeneas, who perhaps hailed from the
town of Stymphalos in the Peloponnesus and possibly served as a gen-
eral of the Arcadian League, was active in the first half of the fourth
century BC. Although today his work is largely unknown outside the
specialist circles of Greek history, Aeneas might be called the world’s
first strategist of urban war.91
Aeneas penned several treatises, of which only one has survived,
Makers of Ancient Strategy: From the Persian Wars to the Fall of Rome Page 24