Walls and War

Around the time Khirokitia reached its greatest size, the way into the village did not go through the village wall. Someone coming up from the river on the west faced a stone wall standing more than 2.8 m above the slope, and in front of it a rough stone structure, preserved almost 2.5 m high, that blocked any straight approach [1, p. 23; 2]. Round its end, a stair only 0.80 m wide climbed into a separate block of masonry, 10 m long and 1.60 m deep, built against the wall's outer face and plastered with care. The stair rose towards the wall, turned left, climbed a flight hidden inside the block, turned right, and came out on top of the wall, where the path ran south along the wall top and down into the village [1, pp. 23, 25]. Alain Le Brun, who excavated it, found its one earlier Near Eastern parallel, at Beidha in Jordan, a plain flight set in the village wall; here a monumental block was built for the stair alone [1, p. 25].

Follow the climb on the plan, number by number, and look at the grey wedge: it is the ground a walker on the slope could not see. The profile beside it unrolls the route at its true length and height.

The entrance looks defensive, and nothing at Khirokitia shows that anyone ever attacked it. Cypriots built walls in a handful of episodes over seven thousand years, and only one of them, at the Northeast Gate of Palaepaphos, still holds the fight it was built for. Everywhere else the purpose of a wall has to be read from its shape, its site and what lay against it: keeping people or flocks out, holding up a slope, marking who belonged inside, showing off the labour that built it, or guarding copper. Nearly every wall in this lesson was a socle, a stone footing, carrying mudbrick that has since melted or gone (Lesson 1).

The first village walls: Khirokitia and Tenta

Khirokitia, a village of the Late Aceramic Neolithic (c. 7000 to 5200 BC), was walled more than once. The first village's boundary wall on the east was later doubled by a massive wall of pisé (rammed earth) faced with stone outside, which the French team numbers Mur 100 [1, p. 23; 2]. Porphyrios Dikaios, digging from 1936, exposed about 185 m of it and called it the "Main Road", admitting that some stretches gave "the impression of a town wall" [3].

When the settlement outgrew that wall, the villagers built over a strip about 10 m wide to the west and closed it with a new wall, number 284, up to 2.50 m thick [27]. It is preserved more than 3 m high and has been traced for more than 60 m [2]. The houses along its inner side were built with it, their walls part of its fabric [27], and the stair entrance, access 2, belongs to the same phase. The Khirokitia household reconstruction stands against this wall. Raise the wall in the model, then step on: the stair goes up as a separate plastered block against the outer face, with no opening cut through the wall, and in the step after that the largest house is built into its inner face.

Open the 3D model

Opens on the build sequence, step 2 of 12: raise wall 284.

Access 1, in the first wall, was a gap about 1.20 m wide whose plastered floors ran out over a stone platform built against the outer face, only to bridge the drop [1, pp. 23, 25]. Access 2 was too narrow and too carefully plastered for flocks, and since no animal pens have been found inside, the animals were kept outside. Le Brun saw the paradox: a defensive work built so that its defence was never put to use [1, p. 25].

Louise Steel notes that such walls are "assumed" to be defensive, and asks what threat they met: there is no evidence for violent destruction or deaths, no weapons, and the danger "might equally have come from the wild landscape beyond" [4]. Bernard Knapp reads the walls as marking out the community, and at the least as proof that someone could organise the labour to build them [5]. On a hill this steep a wall also holds up the terraces behind it. What access 2 certainly did was control movement.

Walk in as a visitor would have done, from the slope below: round the end of the blocking structure, up the first flight, 0.80 m wide, along the middle flight screened from the slope, and out along the wall top. The tour marks the third flight, still unexcavated, as reconstruction.

Open the 3D model

Opens on the guided tour, stop 1 of 9: the way in.

Kalavasos-Tenta, a Cypro-PPNB hilltop village in the Vasilikos valley, was walled earlier still, "by a substantial encircling wall and a ditch cut in the havara", the soft natural limestone, immediately outside it [6]. The wall survives up to about 1.2 m and probably carried mudbrick. The ditch, about 1.65 m deep, is clear only on the south and elsewhere runs in disconnected stretches, so Ian Todd, the excavator, wondered whether it was ever finished [7]. The villages that followed were open [4], and as far as we know no village was ringed by a stone wall like Khirokitia's for some three and a half thousand years after it.

Forts on the plateau edges

The walls came back at the turn from Middle Cypriot (MC) III to Late Cypriot (LC) I, around 1650 BC on this course's dates. Michel Fortin counted about twenty fortified sites of this horizon in three belts: the Karpas peninsula, the passes and southern slopes of the Kyrenia range, and the plateaux of the central Mesaoria facing the copper-bearing foothills of the Troodos [8]. Most lie well inland. Their walls are plain: two parallel rows of undressed or roughly cut limestone with rubble between, 1 to 1.8 m wide, carrying mudbrick, set along plateau edges so that the cliffs did the rest. At Ayios Sozomenos, south of Nicosia, three enclosures stand on one plateau whose cliffs, 40 to 50 m high, did much of the defending [8].

One fort breaks the pattern. At Korovia-Nitovikla, above the south coast of the Karpas, Erik Sjöqvist of the Swedish Cyprus Expedition exposed in 1929 a square building about 40 by 36 m round a courtyard about 20 by 24 m, in one corner of a much larger walled plateau [25; 9]. Rooms lined the inner face of its curtain walls, which were about 2 m thick. These are casemates, rooms built against or inside a defensive wall; the excavators assumed their flat roofs were fighting platforms, though no roof survives [25]. The gate faced north-east between two square towers, its passage lined with orthostats, upright slabs of conglomerate up to about 1.95 m long with drafted margins (a smooth band chiselled round the edge of the face). Inside, a visitor met a blank wall and had to turn, past a cistern [25; 9].

Its date has moved. Sjöqvist placed it in MC III; Gunnel Hult's restudy of the pottery put its building in LC IB, roughly 1550 to 1450 BC, with a rebuilding in LC IIA [9], a century or more after the "troubled end of the Middle Bronze Age" it used to illustrate.

Every generation has explained the forts differently. Sjöqvist in 1940 tied them to invaders in the age of the Hyksos; Hector Catling answered that forts so far from the sea pointed to trouble inside the island [4]. Fortin argued that they guarded copper and its routes to the coast [8], and Edgar Peltenburg made them the work of an emerging state centred on Enkomi [23], whose own "Fortress" Fortin read as a copper workshop secured against theft [12]. Lindy Crewe sees regional responses with no single centre [11]. Yet the copper sources, and Politiko-Phorades, the only excavated smelting site of the period, had no walls at all [10]. Whatever the forts were for, the need passed quickly: Crewe finds them out of use by LC IIA, in the 15th century BC [11], with Nitovikla's rebuilding of that phase the one late exception [9].

Cyclopean walls around 1200 BC

Towards the end of the 13th century BC, at the close of LC IIC, walls went up round towns near the coast in cyclopean masonry: very large unhewn or roughly shaped blocks, named after the giants said to have built Mycenae. At Enkomi the wall is two parallel rows of unhewn blocks about 1.5 m high, the outer row larger, with a rubble core and mudbrick above, and rectangular towers 4 to 5 m long at irregular intervals round a grid town about 400 by 350 m [13]. At Kition a destroyed mudbrick wall was rebuilt in cyclopean masonry about 2.4 m wide, with a stone anchor built into its face [13].

Two new settlements of the same moment show how far a "fortification" can vary. Maa-Palaeokastro occupies a narrow headland on the west coast. Across its neck runs a landward wall about 70 m long and 3.5 m thick, two faces of boulders with rubble between and mudbrick above, its overlapping ends forming a dog-leg gate; a second wall, 3.5 to 4 m thick, faces the sea [24; 14]. Pyla-Kokkinokremos, on a steep plateau above Larnaca Bay, had no separate wall. Its houses stood against one another along the edge, and their thickened outer walls made a casemate ring, "established to define and envelope the settlement rather than to firmly protect it", in Artemis Georgiou's words [14, p. 70]. Vassos Karageorghis and Martha Demas first read both as footholds of Aegean newcomers; their pottery and buildings are local, and Georgiou reads them as outposts of coastal centres, Palaepaphos for Maa and Kition for Pyla, guarding copper and sea trade [14].

Meanwhile Kalavasos-Ayios Dhimitrios, Maroni-Vournes and Alassa-Paliotaverna in the south put up the finest masonry of the age in ashlar, squared and dressed blocks (Lesson 5), and were never walled. In this period a town wall records a decision.

On a map the three kinds of place separate. The forts of about 1650 BC sit inland or on the Karpas, the walls of about 1200 BC ring towns at or near the coast, and the unwalled towns with the finest masonry, with the copper-smelting site at Phorades, lie in the south and the foothills.

Map of Cyprus

• Forts, c. 1650 to 1450 BC
• Town walls, c. 1200 BC
• Never walled
• Ground above 200, 500, 1,000 m

N

50 km

Sites on this map

1. 1Ayios Sozomenos
2. 2Korovia-Nitovikla
3. 3Enkomi
4. 4Kition
5. 5Maa-Palaeokastro
6. 6Pyla-Kokkinokremos
7. 7Kalavasos-Ayios Dhimitrios
8. 8Maroni
9. 9Alassa
10. 10Politiko-Phorades (approx.)

City walls of the kings

Claire Balandier sees the Iron Age kingdoms keeping Bronze Age habits, stone footings with mudbrick above: at Idalion the Cypro-Archaic I wall follows the line of a Cypro-Geometric III enclosure [15]. Early in the 5th century BC the whole city of Idalion was enclosed in a wall of limestone rubble faced with ashlar, probably with mudbrick above. At its best-preserved point it still stands 7.80 m high and 10.75 m wide, and it enclosed about 40 ha [16]. A bronze tablet from the site records a siege of Idalion by "the Medes and the Kitians", and a grant of land to a doctor who treated the wounded free of charge [18]. At Amathus, perimeter walls with fortified gates went up at the end of the Archaic period and grew in the 5th century; Catherine Kearns calls the walled lower town "arguably ideological, and not defensive" [17].

Palaepaphos, 498 BC: a siege left in place

In 498 BC every Cypriot kingdom except Amathus joined the Ionian revolt against Persia under Onesilos of Salamis; after their defeat in the plain of Salamis, the Persians took the cities one by one, Soli last, in the fifth month, by undermining its wall [26]. At Kouklia, the village on the site of Palaepaphos, the Marchello hill preserves what happened at one of them.

The wall there, first built in the second half of the 8th century BC, was a massive mudbrick structure with rough stone facings on both sides, 4.80 to 6.30 m wide [19]. Shortly before the siege the Northeast Gate had been remodelled into a narrow passage with a sharp double bend, covered by crossfire from bastions (solid towers projecting from the wall) on both sides, and closed at its city end by heavy wooden doors in an opening only 2.90 m wide [19, p. 194]. Outside lay a fosse, a dry ditch, beyond a berm, the level strip at the wall's foot.

The Persians demolished a Late Archaic sanctuary outside the walls and, with its debris, loads of soil and tree trunks, filled the ditch and raised a ramp up to the wall [19, p. 202]. What survives of it stands 4.50 m above the outer lip of the ditch and more than 7 m above its bottom [19, p. 197]. The fill holds the sanctuary it consumed: more than 1,000 fragments of limestone statues of people and animals, and hundreds of syllabic inscriptions [21, p. 284]. It also held more than 500 iron and bronze arrowheads, javelin points and spearheads, and a Greek bronze helmet of Corinthian type, which Franz Georg Maier, the excavator, called the only Greek bronze helmet then known from a battle rather than a grave or a sanctuary [19, p. 198].

A limestone head of the same years from Golgoi wears the same type. More than 450 rounded limestones with one flattened side, weighing 2.7 to 21.8 kg, lay in the ramp too. Maier first suggested they were ammunition for a stone-throwing engine, which would put the catapult a century before Dionysios I's siege of Motya in 399/398 BC [19, p. 200]. In his later guide to the site he settled for missiles thrown by hand from the parapet [20, p. 68].

The defenders fought the ramp from underneath. They broke through the lower part of their own wall to dig a Passage across the berm and cut four tunnels in the rock beneath the wall. Tunnel 1, running from a trench inside the wall to the bottom of the ditch, is 1.20 to 1.70 m wide and 1.70 to 2.30 m high, with Archaic lamps in niches cut in its sides [19, p. 200]. Through the tunnels they dragged stone from the ramp into the city, propping the hollow with timber and mudbrick. Tunnel 2 was abandoned when part of its roof fell in, and Tunnel 4 was never finished: "apparently the fighting ended while it was still under construction" [19, p. 200].

At the ends of the Passage and of Tunnels 1 and 3 stood the same arrangement: a large bronze cauldron, badly burnt, on two or three mudbricks, filled with charred wood and lime to fire the props [19, p. 200]. The tunnels drew like furnaces, parts of the ramp collapsed, and above the cauldrons the fill fused into a hard block of burnt stone and lime [19, p. 202]. Maier inferred that the Persians meant to roll wooden siege towers up the ramp, the only load worth bringing down by a sudden collapse; no tower was found [19, p. 202]. The mining did not save the gate: its doors burnt, and Maier concludes that the attackers forced the passage [19]. He called Marchello, apart from Masada, "the only archaeologically fully documented ancient siege operation" [19, p. 202].

That reading has held for half a century. Maria Iacovou notes that a sounding in 1973 found no ditch north-west of the gate, and that the rampart's south face is the one reinforced with larger boulders. She suggests, tentatively, that the wall faced south and set off a special zone on the plateau, so the sculpture need not have come from outside the walls [21, p. 284]. The fight itself, which the weapons attest, she does not deny [21].

When the walls were rebuilt, more than a century later, the Paphians, unable to shift the mound, faced it with stone and made it a horseshoe-shaped bastion [19].

After the kings: Nea Paphos and Pyla-Vigla

Nikokles, the last king of Paphos, moved his capital to a new harbour town, Nea Paphos, late in the 4th century BC. An inscription from the old sanctuary records that he ringed a "wide-spaced city" with towers, and scholars now almost all identify that city with Nea Paphos [21, p. 287]. At its Northwest Gate a rock-cut ramp about 35 m long climbed between two towers cut from the rock and faced with masonry [19]. Beside it Balandier recorded eight sortie tunnels through the rock base of the wall, linked in pairs, and asks whether their model was the tunnels dug at Palaepaphos in 498 or newer Hellenistic tactics; they are not yet dated [15].

The last fort here, Pyla-Vigla, stands on a plateau east of Larnaca, on the road between Salamis and Kition. A fosse cut in the bedrock guards its north side, and a wall 1.7 to 2.8 m thick runs round the plateau, built like the houses inside it: rough fieldstone faces on bedrock, a rubble core, and mudbrick above [22]. Its earliest floors and its final collapse hold finds of one date: a garrison held it briefly, around 300 BC, in the wars of Alexander's successors, for Ptolemy or for Antigonos; its excavators weigh both [22]. Its commonest weapon is the lead sling bullet. Among its bronze projectiles are tanged bolts for catapults [22], the machine Maier first proposed for Palaepaphos two centuries earlier, and then withdrew.

Quiz

Check yourself

Question

How wide was the stair of Khirokitia's access 2, in metres?

Your answer, in m

m

Check

Question

Khirokitia's entrance looks defensive. Why do scholars hesitate to call its walls defences?

• AThe walls were too low to stop anyone
• BThe village had no way in through its wall
• CNothing shows an attack: no violent destruction, no deaths, no weapons

Question

Fortin argued that the forts of about 1650 BC guarded copper and its routes to the coast. What weighs against that?

• AThe forts lie far from the sea
• BThe copper sources, and the smelting site at Politiko-Phorades, had no walls at all
• CNitovikla was built a century or more later than first thought

Question

Pyla-Kokkinokremos, above Larnaca Bay, is counted among the fortified sites of about 1200 BC. What formed its defence?

• AA wall 3.5 m thick across the neck of a headland, with a dog-leg gate
• BThe thickened outer walls of houses built against one another along the plateau edge
• CTwo rows of unhewn blocks with towers at intervals round a grid town

Question

Put the story of the Northeast Gate at Palaepaphos in order.

Tap an item, then tap its place. Or use the arrows

First

1. 11st: Rebuilding the walls, the Paphians face the mound with stone as a horseshoe-shaped bastion
2. 22nd: The Persians demolish a sanctuary and build a ramp of its debris, soil and tree trunks across the ditch
3. 33rd: The defenders break through their own wall and dig a Passage and tunnels under the ramp
4. 44th: The gate's doors burn and the attackers force the passage
5. 55th: Cauldrons of charred wood and lime fire the props, and parts of the ramp collapse
6. 66th: The gate is remodelled into a narrow passage with a double bend between bastions

Last

Check the order

Question

More than 450 rounded limestones lay in the Palaepaphos ramp. What did the excavator finally make of them?

• ASling bullets
• BAmmunition for a stone-throwing engine
• CMissiles thrown by hand from the parapet

Question

How sure can we be of each of these?

Tap a statement, then the bin it belongs in

To sort (5)

• The Persians meant to roll wooden siege towers up the ramp
• The Palaepaphos wall faced south and set off a special zone on the plateau
• A ramp of sanctuary debris, soil and tree trunks was raised against the wall at Palaepaphos
• Burnt bronze cauldrons stood at the ends of the defenders' tunnels
• The rounded stones in the ramp were shot from a catapult

1. 1ExcavatedMeasured on excavated remains
2. 2Strong inferenceIndirect but good evidence
3. 3SpeculativeNo direct evidence
4. 4RejectedRuled out by the evidence; shown for comparison

Check the sort5 still to sort

What to look for

• Read the socle first. A stone footing 1 to 2 m wide is the base of a mudbrick wall whose upper part has gone; two faces of big boulders with rubble between, as at Enkomi or Maa, is the cyclopean manner of about 1200 BC.
• Note where the wall runs. A wall along a plateau edge lets the cliff do the work (the MC III to LC I forts); a wall across the neck of a headland, as at Maa, closes the only way in.
• Count the turns at the gate. A bent or dog-leg entry, towers either side and a narrow inner opening (Nitovikla, Palaepaphos, Maa) were built to control who came in; a plain gap, like Khirokitia's access 1, served the walker.
• Look for rooms against the inside of the wall. These are casemates: a fort's own rooms at Nitovikla, the backs of houses at Pyla-Kokkinokremos.
• Check what an unwalled neighbour did. Kalavasos and Maroni built in fine ashlar without a wall, so a wall marks a choice, and its absence is evidence too.
• Look in the debris for burning, weapons in the fill, a ramp or tunnels. Most Cypriot walls have none of these, and the case for "defence" then rests on the plan alone.
• Separate old from new. Khirokitia's stair at the foot of the site is a full-size modern replica, Tunnel 1 at Kouklia stands on modern concrete supports, and at Nea Paphos most of what shows is the rock-cut bed of a vanished wall.

Go deeper

Walk Khirokitia's entrance in the Khirokitia household reconstruction (its report section "Up, left, right and right again" sets out every measurement), then read Fortifications in Ancient Cyprus for the full site table and the debates, and Iron Age arms and armour for the weapons in the Palaepaphos ramp. The site pages for Khirokitia, Kalavasos-Tenta, Enkomi, Kition, Idalion, Amathus and Paphos carry the places further; the articles on the excavations at Idalion, the city of Amathus and Maria Iacovou's kingdoms take up the kingdoms behind the walls.