After Cyprus and southern Cilicia the ‘triangle’ of Greek contact in this east end of the Mediterranean finds its third point on the coast of the Levant. In antiquity, as nowadays, this coastline was politically and culturally very varied too, the beginnings of our Near or Middle East, but less than 50 miles by sea from the north end of Cyprus. It fell into three main sections, more diverse in their contacts with Greeks and Greek objects than we would guess from the horizons of Homeric epic, for which the one point of contact with Greeks was Sidon, a rare and random port of call.
The northern section of the three was the north Syrian plain (now in Turkey), a flat, well-watered enclave which lies between two major mountain-barriers. On the north it is blocked out by the ascending hills of the Amanus range, the boundary between the north Syrian plain and the settlements of Cilicia to the north-west. From the Syrian plain, these mountains, the Cedar or Box-tree Mountains of Assyrian texts, were crossed north-westwards by the road from ‘Pahri’ in Syria at their foot (the site which Greeks called Pagras).1 This road went on up through the Syrian Gates to a height of some 6,000 feet before descending to the bay of Iskenderun near the Greek Posideion, a site chosen partly for this inland route. On a map it seems a long trail across a major mountain-barrier, but we know from early modern travellers that the journey up and over was only six to seven hours on horseback, four of them from the Syrian Gates down to Iskenderun.2
Later in antiquity, as nowadays, this north Syrian plain was to be best known for the great city of Antioch, founded by a Successor of Alexander the Great c. 300 BC. Because of Antioch, above all, we think of the plain as ‘married to the Mediterranean’ into which Antioch’s Orontes river runs.3 In the ninth to mid-eighth centuries there was no royal town as close to the coast as Antioch would later be on the Orontes’ course westwards. The plain’s ‘marriage to the Mediterranean’ was thus less evident than its long-standing engagement to two land-routes, one running south down the Orontes’ course into Syria, the other the route north-west to the Syrian Gates and the bay of Iskenderun beyond them.4
4. Cyprus and the Levantine coast
At its southern end this plain is blocked by the hills and main peak of the Jebel Aqra, the great mountain-beacon of the region. Here a second section of the coast began. Here too the settlements were no more politically coherent. The best-known were the royal city-states of the Phoenicians, which ran from the island site of Arwad (Arados, to the Greeks) as far south as Acco near Mount Carmel. Further south beyond the Phoenicians lay a third section, the settlements on the coast of ancient Palestine, those which edged the northern kingdom of Israel and those further south like Ascalon or Ekron, which were centres of Philistine culture. Here the prevailing currents would help ships to travel down from Cyprus, but so far Greek finds from the ninth and eighth centuries are minimal. However, they include Euboean cups and raise the pleasant possibility of a Euboean and a Philistine or an Israelite drinking together and using and bestowing Greek cups as mementoes.5 Fragments of two Euboean cups survive from the coastal site of Tell abu Hawam and are arguably connected with a destruction on the site, probably in the mid- to late ninth century BC.6 Tell abu Hawam lay near a protected bay off Mount Carmel, close to modern Haifa, which was linked in antiquity to an easy inland route to Megiddo and the Jezreel valley. It was perhaps through this port that more Greek items reached Megiddo (c. 830 BC) and then Samaria, the capital of the Israelite kings, where the few Greek pieces known include at least one tall mixing-bowl of Athenian origin, a substantial piece for an Israelite drinking-party among King Ahab’s descendants, c. 780–760.7 Further south, in the Jordan valley, two cups and two mixing-bowls of Euboean origin reached Tel Rehov, as did an Attic storage-jar which Euboeans probably brought to the Levant too, c. 850–830. The items travelled in from the coast.8
A Euboean presence on these three sections of the Levantine coast can be inferred in various ways and the strongest evidence of it is particularly important for the next step, their encounter with specific non-Greek, localized myths. It needs, then, to be followed piecemeal, although there is still the uncertainty of whether a Greek actually lived where a Greek object is found. Can the style of their objects help us? All down this coast the non-Greek societies already had their own style of partying, as we can see from local sculptures and above all from the Hebrew Scriptures’ book of the prophet Amos. He deplored the extravagant habits of Israelites in the kingdom of Samaria c. 760 BC, who ‘lie on couches of ivory and are draped on their beds…who howl to the sound of the lute…who drink wine from big basins and anoint themselves with the finest oils’. Amos uses strong words to intensify the excess of their noisy singing, their ‘lounging’ on couches, their drinking from huge bowls. It sounds riotous, but Amos predicts with a fine play on words that the ‘shouting spree of these sprawlers’ will be the first to go into exile.9 The big basins have been identified as ‘the beautiful ring-burnished multiple-handled ceramic bowls of Iron Age II…the finest of all products made by the Israelite potters’.10 We might, then, wonder why Levantine drinkers needed Greek cups, and thus infer that finds of such cups in a settlement imply that a Greek was indeed present on the site, using his own distinctive pottery.
Habits of drinking differed too. Like others in the Levant, Israelites drank their wine neat but flavoured with spices, whereas Greeks mixed it with water, a practice which the prophet Isaiah regarded as most inferior. Greeks therefore developed types of cups with handles, long stems and feet which they could hold between their fingers and move from side to side so as to mix their water and wine.11 The earliest Greek imports, however, do not have these distinctive Greek shapes: they are thick wide cups, or skyphoi, with a base, not a tall foot, and a pair of small handles by the rim. They would fit quite easily into the local style of drinking from bowls which Amos exaggerates. The tall Athenian-made mixing-bowls on their pedestals and feet were a novelty in the Levant, but they too could have been used locally for wine only, not for watered wine as in Greece. They might have been regarded as a curious centrepiece at a non-Greek party. From the shapes of these Greek cups and bowls, therefore, we cannot argue that they were fit only for Greek use and thus prove a presence of actual Greek residents where we find them in a settlement’s buildings. Greek pottery certainly passed into non-Greek hands. At Old Tyre, or at Khaldeh, just south of modern Beirut, a few Greek pieces ended up in the sites’ non-Greek cemeteries during the eighth century. Similarly, a big Greek jar, or pyxis, of probable Attic origin was found intact, dating to c. 850–820 BC, in a tomb about 4 miles south-west of Sidon: it was found together with five big Phoenician urns, each used for a cremation, one of which had a clear Phoenician inscription.12 These Greek pieces were placed in non-Greek tombs near examples of non-Greek writing and plainly they could only have been put there by non-Greeks. At most they imply contact with Greek visitors who brought them east to the Levantine coast and passed them on.
Even so, any such Greek presence was a tiny part of a much bigger local pattern of trade and travel. Up and down this coastline Phoenician ships were carrying anything from silver to cedarwood and were continuing to trade with Egypt quite separately from any Greek traffic. Our understanding of them has been transformed by the recovery of two Phoenician trading ships which had sunk with their cargoes on the open sea about 30 miles west of Ascalon. Located in 1997, they have now been analysed and dated to the later eighth century BC.13 They are not unduly big, having a capacity of about 30 tons and a crew of about half a dozen, Phoenicians to judge from their accompanying cooking-pots. Each sailing ship, however, held nearly 400 big jars, evidently filled with wine, each jar holding as much as 4 gallons. At Tyre and Sarepta near Sidon, big pottery kilns have been found on the mainland, places where such transport-jars were mass produced. The wine in the two Phoenician ships was probably in this category. It was being exported down to Egypt, a land which had no grapevines, but the ships were swamped on the sea-journey and sank, preserving evidence of a big bulk trade which continued on this route into Egypt for centuries. It is quite different from the trading of trinkets and amber necklaces, the trade which Homer credits to Phoenicians.
Apart from merely visiting this coast, as the arrival of Greek cups and bowls implies, did Greeks ever reside for long or develop close social contacts? Archaeologists are rightly wary of assuming from a find of imported pottery that the makers and carriers of it lived and settled on its site for any length of time. They look for local imitations of it as stronger evidence that its carriers actually resided there for a while, but on the central and southern sections of the Levantine coast we have no such imitations in the ninth and eighth centuries. Instead, historians can only cite facts from a better-documented age and leave the earlier possibility open. During the classical period of the early fourth century BC we know, but only from texts, that there was a resident Greek community at Phoenician Carthage in the west.14 Without these texts, archaeologists would not easily accept the existence of Greek residents from the sole evidence of the few Greek objects excavated on the site: there were no obvious Greek burials or local imitations of the Greek goods. What happened later at Carthage may have happened already in the mother-city of Tyre in the ninth and eighth centuries BC. A few Greeks, using Greek goods, might indeed have resided in Phoenician city-states. We cannot rule them out, even at this early date: our archaeology is not decisive.
What we can infer is some close contact: we can then best explain a practice which Euboeans imitated from Phoenicians. In Phoenician cities, as elsewhere in the Near East, goods were bought and sold for silver and precious metals. The metals were cut from bars and small objects and were weighed, not coined, because coinage was not yet in use. One unit of weight was the ‘shekel’, a word which simply meant ‘weight’. In Euboean settlements we then find a unit of weight called the ‘stater’, also the Greek word for ‘weight’. Its basic weight is almost exactly that of a shekel and like the shekel it was a fiftieth part of a bigger unit, the ‘mina’ which Greeks also imported from the Near East. Euboeans, therefore, had copied the Phoenician unit.15 It became their official unit of weight and eventually of coined money, and it is known as early as the late eighth century in one of their western settlements. It was not borrowed from western Carthage. It was borrowed directly during commercial exchanges with Phoenicians so as to match their ‘weight’ of exchange. It was surely imitated during Euboeans’ direct contact with the practices and units of Phoenicians in the Levant on something more than a casual visit.
It is consistent, at least, with this contact that, from the period between c. 850 and 750 BC, about a hundred pieces of Greek pottery have so far been recovered from Tyre, the one major Phoenician site to reveal such a quantity.16 They are still a tiny minority among the total non-Greek pottery found there, so they are not conclusive proof of a small Greek presence. They are merely explicable by one, however short it might have been in each instance. If it existed, it is not a presence we would ever guess from Homer’s epic poems. He never mentions Tyre at all.
If we move northwards from the Phoenician city-states to the next stretch of the Levantine coast, we pass out of Homeric horizons altogether. Beyond the Phoenician island-state of Arwad we enter an important chain of harbours and anchorages which Phoenicians also frequented and which Greeks and Greek objects also reached. They are not all strictly ‘Phoenician’ foundations and their place names and sequence need close attention, because their ancient names and significance have been confused with crucial points of Greek contact just beyond them. Exact places and landscapes are essential if we are to trace the myths which Greeks then developed, so we have to be sure of their ground. To appreciate the sites and their ancient place names we can draw on texts written later for Greek sailors here in the Greek age after Alexander. Even if they were compiled no earlier than c. 100 BC-AD 300 they are valuable evidence because the coastal conditions changed so little.
Beyond the island of Arwad the pattern remained one of anchorages by river-mouths or in bays formed by the few promontories on an otherwise exposed coast.17 Heading north, travellers would be able to stop at Paltos (now Arab-al-mulk) at the mouth of the Nahr-as Siren river.18 It was a very short trip from here up to Suksu (nowadays Tell Sukas), a well-excavated coastal site which Cypriot pottery had been reaching already in the eleventh and tenth centuries when Suksu was a settled town. It had a long history before the arrival of the three Greek cups, probably all Euboean, which are the only Greek ones found in the settlement between c. 850 and 750 BC. Another seven pieces, also in the settlement, date from c. 750–700.19 Nonetheless the excavators suggested that these contacts grew into a resident Greek enclave. When they found a building for cult in the centre of the town which was datable c. 650 BC their first thoughts were that it was a Greek temple serving Greek residents and Greek visitors in the seventh century, a major discovery, therefore, for our trail. The building had roof-tiles (a Greek feature) but renewed study of its plan and shape has identified it, more plausibly, as non-Greek.20 There are no early Greek burials anywhere in the settlement and given the rarity of early Greek pottery eighth-century Suksu was nothing more than an occasional port-of-call for a few Greeks.
If we go on north past a good coastal plain and the then-modest site at modern Lattakieh (which was only developed after Alexander), the next significant marker after 5 miles is the coastal promontory of Ras ibn Hani. Here, too, lay settlements with a history stretching far back into the Bronze Age.21 The cape lies only 3 miles to the south-west of the big site of Ras Shamra (ancient Ugarit), particularly important in the years c. 1400–1200 BC, when invaluable finds of clay tablets have given us fragments of the ancient Canaanites’ myths about their gods and confirmed the multi-lingual culture of their scribes. Cypriot pottery was already reaching the site in that period, as was pottery from Mycenaean Greece: Ugarit has given us about a third of all Mycenaean Greek pottery known in the Levant.22 At the end of the Bronze Age (c. 1180 BC) Ugarit’s inhabitants dispersed, but no crisis could neutralize their invaluable asset, the coast’s best natural harbour on the promontory of Ras ibn Hani. It became known from its low white cliff as the ‘White Harbour’ in later Greek coastal guidebooks, a name which persists in modern Arabic as Minet el-Beida.23 Cypriot pottery continued to reach Ras ibn Hani after the Bronze Age and as the previous Mycenaean Greek contacts had been important, we might expect that the excellence of the site and its harbour’s long history would quickly attract other Greek travellers as they returned to the Levantine coast. However, only two cups from Greece have been found there, both of which date between c. 850 and c. 750 BC. They have the familiar Euboean decoration, but they imply visitors, not settlers.
What, then, did Greeks call the settlement which they found by this harbour? At Ras Shamra, texts and statues show the honours accorded to El, father of the gods. Ras Shamra’s later small settlement and the nearby Ras ibn Hani were surely the site which Greeks later knew as ‘Betyllion’ (their version of the Semitic phrase ‘bait-El’ or ‘House of El’). They noted Betyllion for its natural harbour, surely the ‘White Harbour’ which we can still admire and which had been the port since Ugarit’s Bronze Age. We can therefore understand why Betyllion’s ‘natural’ harbour is described as the first staging-post for the Roman emperor Trajan when he set out from Rome to join his troops in Syria for their fateful campaign into Mesopotamia (modern Iraq) in AD 114–17.24 He had landed at ancient Ugarit. The correct location of Betyllion is important for different reasons, not because it was a Greek settlement but because it has become involved in a major topographical puzzle about Greek contact further north and also because the name (‘house of El’) confirms that Canaanite–Phoenician culture never entirely died at the site.
Moving north from the White Harbour, Assyrian texts of the eighth century BC refer to a separate cape called ‘re’si-surri’, a landmark for them. It is probably to be translated as the ‘Cape of the Rocks’.25 The natural candidate for this cape is the modern Ras el-Bassit, about 15 miles to the north of Betyllion. It, too, has an attractive natural harbour and an excellent view up to the defining chain of the Jebel Aqra mountain, the next major landmark to the north. Fortunately it has been well excavated, showing that it is yet another settlement on this coast whose history extends continuously back into the Bronze Age and whose culture owes next to nothing to Greeks. Phoenician pottery was imported to it from the south in considerable quantity during the ninth and eighth centuries, while the burial customs of its well-studied cemetery connect with those of the Syrian settlements inland.26 Once again a problem lies in its name, the source of a mistake about its eighth-century significance for Greeks. We happen to know from Greek texts of the late fourth century that Greek sailors called this promontory Posideion, that favoured name on many coastlines. In Alexander the Great’s lifetime, Greek influence was extending on this coast and as a ‘Posideion’, Ras el-Bassit was striking coins with the god Poseidon’s figure on them.27 The modern name ‘Bassit’ (from the Arabic for ‘flat’) has even been traced, somewhat fancifully, back to the Greeks’ ‘Posideion’ name.28 In the ninth to eighth centuries Ras el-Bassit was not a Greek settlement or indebted to a Greek initiative. Fragments of imported Greek pottery have been found on the site, especially pieces of Euboean-style cups which date from the eighth century, but there are not more than fifty pieces in all.29 They are never concentrated in any particular tombs and there are no local imitations. The implications of Ras el-Bassit for eighth-century Greek contact are that Greeks visited with their goods and perhaps resided briefly but never settled there in the ninth to eighth centuries. Only later was it called ‘Posideion’ and it was never connected with a legendary Greek hero as its founder. Ras el Bassit was not the Posideion which Herodotus credited to the travelling hero Amphilochus. He sited that one much further north.
So far, these coastal places are all sites with long non-Greek histories which extend for at least four hundred years before any Euboean cups and plates came near them. Most of them also had overland access to major royal settlements inland, Tell abu Hawam to Samaria, Ras el-Bassit by a hill-route to Syrian Hamath. Important routes ran even further into the interior, including the Jordan valley. Above all, a major route from Ras ibn Hani ran up quickly and directly to Aleppo in the north Syrian plain, linking this significant site to the southern ‘Canaanite’ sea-coast in a way which our local studies of the plain would never suspect.30 When Greek pottery turns up inland it is most likely to have been carried, still, by non-Greeks. On the coast, meanwhile, the main seafarers and visitors were Phoenicians.
Beyond Ras el-Bassit a Greek maritime guidebook of the fourth century AD is explicit about the difficulties.31 Ships had to pass the range of the towering Jebel Aqra (the modern frontier between Syria and Turkey) where they encountered severe winds from the west threatening to drive them into the mountain’s base. There were islands just off it, rocky ones with names like ‘Long Island’ and ‘Pigeon Rocks’, and there was also a small bay, later ‘sacred to the nymphs’, immediately on the Jebel Aqra’s far side.32 Northbound ships for the south Cilician coast would be wise to stay further out to sea and miss this awkward stretch of coast. They could make one long day’s journey straight up to the mouth of the Pyramus river and the edge of the Cilician plain. If necessary they could put in after about 20 miles at the last north Syrian promontory (now in Turkey) and supply themselves from the river-waters which gushed naturally from its hill, as they still do. Greeks knew this place as ‘River of Waters’ before Alexander’s Successor Seleucus developed it into the major harbour-city of Seleuceia. It became Seleucus’ first capital, housed his tomb and endured for nearly a thousand years. Significantly, his builders had to construct its harbour artificially at the end of a canal which protected it from the storms on the turbulent sea beyond.33
A direct sea-route from Ras el-Bassit up to Cilicia thus had its attractions and Phoenicians were the most likely users of it because of their well-attested presence in the Cilician plain. As well as the metal-resources of the mountains beyond Karatepe, there was a resource in the plain which has left no archaeological hoofprint: horses. The specific evidence of the biblical book of Kings is to be taken seriously here: horses were brought from ‘Que’, the Cilician plain, and taken south to Solomon’s Jerusalem.34 Some of this horse-trade may have been by land but the problems of access, horse-thieving and delays made sea-transport the obvious option. The art of shipping horses is amply attested in later Greek history, but it may already have been practised down the Levantine coast in the tenth century BC. In the later sixth century Ezekiel even credits Tyre with importing horses from central Anatolia, also, no doubt, through Cilicia and perhaps on southwards by boat.35
Ships which took this direct route to Cilicia missed the northern section of the coastline, the north Syrian plain which runs beyond the Jebel Aqra mountain for about 20 miles until the Amanus mountains intervene and cut it off. Like the Cilician plain, it was the fertile creation of three rivers, of which the most important is the Orontes, which snakes and twists onto it from modern Syria and then bends sharply west before running out onto the coastline. Here, the Orontes created a delta at its mouth which was about 10 miles wide.36
Naturally and culturally this plain was an attractive asset. Its northern limit, the Amanus mountains, had the box- and cedar-trees which were prized for buildings by the Assyrian kings. The plain had an area of wetland, valuable as always for flax (its local kings’ linen robes were notable), and a significant lake. It had once been a home for elephants, as we know from an elephant’s shoulder-bone found in the plain itself: there are grounds for supposing that elephants survived there into the first millennium BC.37 There were also hippopotamuses, whose teeth were another source of ivory for the region’s important ivory-carvers.38 There were precious metals too, including copper and silver, in the Jebel Aqra mountain which formed the plain’s southern boundary.39 We can gain an idea of the riches of the royal town which was nearest to the Orontes delta from the records of the conquering Assyrian kings. From inland ‘Kinalua’ (the modern Tell Tayinat) on the westward bend of the Orontes river, the Assyrians’ first conquest in the 870s BC took silver, gold, ‘100 talents of tin and 100 talents of iron, 1,000 oxen, 10,000 sheep’, fine linen robes, decorated couches and beds of box-wood, ‘10 female singers, the king’s brother’s daughter with a rich dowry, a large female monkey and ducks’.40
Kinalua was only one royal town in a patchwork in the plain. Even more than its Cilician counterpart, this Syrian river-plain was a multi-cultural, multi-lingual tapestry which is important for the debts which Greeks will turn out to owe to it.41 Since the ninth century BC Aramaean tribes had conquered and settled in several of north Syria’s most important sites, especially Hamath (modern Hama) and Arpad (north-west of Aleppo, at Tell Rifa‘at). These Aramaeans had a West Semitic speech and script (our Aramaic) and brought with them their own religious culture.42 At particular sites inland, the Phoenicians’ script and artistic styles are also visible, especially in evidence from the ninth century. Above all, as in Cilicia, the legacy of the former Hittite empire had lived on prominently, giving rise to a clear ‘neo-Hittite’ style.43
These various cultures coexisted and inter-related to an extent which we cannot unravel into neat zones and decades. At Hamath in the ninth century kings with neo-Hittite names put up hieroglyphic inscriptions in Luwian, but already worshipped a well-established non-Luwian goddess, ‘Our Lady’ or Ba‘alati, as they knew her by her Semitic name.44 During the later ninth century the rulers at Zincirli at the foot of the Amanus mountains changed from putting up inscriptions in Phoenician and used Aramaic instead, but continued to have neo-Hittite Luwian personal names.45 Royal inscriptions are not even proof of a town’s everyday spoken language: they may simply be statements of power and display, while ordinary people spoke something else. The patterns of speech may have been different, another cultural complication.
The neo-Hittite legacy is particularly important for discoveries which the Greeks made here and the evidence for it is strong, though individual items are not easy to date closely within the years from c. 900 to 750. Inscriptions, not only royal ones, were composed in the Luwian language and written in ‘neo-Hittite’ hieroglyphic script: styles of sculpture were distinctively neo-Hittite, whether for the big block-sculptures in the doorways of local palaces or the smaller funerary reliefs which commemorated members of private families. This culture is most prominent in the tenth and ninth centuries, especially in centres like Hamath before c. 800 BC.46 At Aleppo there is excellent evidence of a temple and important cult of the neo-Hittite Storm God;47 at Kinalua the rulers continued to have historic Hittite royal names in the eighth century; at Ain Dara, by the Afrin river (Chalos, to the Greeks), there was a palace and a temple platform much of whose style, sculptures and cults remained neo-Hittite too.48 Even in the eighth century BC this cultural imprint persisted, as Luwian continued to be written in neo-Hittite hieroglyphic script and the Storm God and other neo-Hittite deities in the area were still worshipped.49 In the 730s the king of Zincirli by the foot of the Amanus mountains kept a signet-ring with his name in hieroglyphic Luwian characters as well as having another which was inscribed in Aramaic, the language in which his family put up royal inscriptions during the eighth century BC.50
This multi-cultural vigour went with a history of changing settlement. Like the towns up the Levantine coast, the towns in north Syria had not been eclipsed by the ending of the Bronze Age. Instead, between the tenth and the mid-eighth century many of them had emerged as new settlements, clustering round a fort or palace and showing no sign of a reduced population.51 For coastal visitors the important zone was the plain nearest the sea, part of the land which Assyrians knew either as ‘Patina’ (a neo-Hittite name) or as ‘Unqi’ from the Aramaic word for its low-lying plain.52 In Patina–Unqi the royal town of Kinalua lay inland on the far eastern bank of the Orontes river where the river had wound north from Syria and was about to bend sharply west towards the sea. Kinalua is the Calneh of the Hebrew Scriptures, and its culture was still strongly marked by a neo-Hittite style when the Assyrians in 740 BC forced its king Tutammu to submit.53
Despite its position beside the Orontes’ bend to the west, Kinalua should not be assumed automatically to have had constant links with the Orontes river-delta and the coast: its connections by land-routes to the south along the river-valley were also very important. For travellers going north and south on the sea, meanwhile, the Orontes delta was not a self-evident attraction. Its coastline has advanced nowadays through river-silting but even today the open shore is not a safe haven for boats brought onto it.54 Strong winds from the west interrupt the boats at anchor, while modern naval handbooks recognize the hazards of winds and currents and also the sandbanks at the river’s mouth. In antiquity, as until recently, the Orontes remained navigable, but the winds, currents and probably some of the silting were present in the ninth century BC too. They may help to explain the muted presence, so far, of explicitly Phoenician goods and culture in the early eighth-century levels of sites in the delta and the coastal sector of the plain of Unqi. With the entire Syrian coast to choose from and south Cilicia ahead of them, it seems that Phoenicians may often have preferred to give this beach a miss.
The delta is crucial, however, for the Greeks’ trail in the eighth century because of a much-discussed mound which has been excavated on the north-west bank of the Orontes as it approaches the sea. Far more Greek pottery has been found here than in any eighth-century level of the settlements on the Levantine coast or the Cilician plain. In 858 BC the conquering Assyrian king Shalmaneser III had recorded how ‘in the wide area of the seashore I marched by right of victory. An image of my lordship, which establishes my name for all time to come, I made and set up by the sea.’55 This monument has long disappeared from the coastline and the honours have passed instead to a site with Greek contacts, founded within a hundred years of his departure.