The Straits from Troy to Constantinople

The Ancient History of the Dardanelles, Sea of Marmara and Bosporos

John D Grainger

• Publisher: Pen and Sword Military
• Year: 2022
• ISBN:
  9781399013253
• Topic: History
• Subtopic: Ancient History
• Index: History

In ancient times, the series of waterways now known as the Turkish Straits, comprising the Dardanelles (or Hellespont), Sea of Marmara and the Bosporus, formed both a divide and a bridge between Europe and Asia. Its western and eastern entrances were guarded, at different times, by two of the most fabled cities of all time: respectively Troy (in Asia) and Byzantion (or Byzantium, on the European coast). The narrow crossing points at the Hellespont and Bosporus were strategically important invasion routes while the waters themselves were vital routes of travel and commerce, particularly the supply of grain from the hinterland of the Black Sea to the Greek cities. This made them sought after prizes and sources of friction between successive empires, Persians, Macedonians and Romans among them, and ensured they were associated with some of the great names of history, from Odysseus to Xerxes, Alexander to Constantine the Great. John D Grainger relates the fascinating history of this pivotal region from the Trojan War to Byzantion's refounding as the new capital of the Roman Empire. Renamed Constantinople it dominated the straits for a thousand years.

Chapter 1

The Early Settlers (before C.3000 BC)

The initial human occupation of the Straits region came after the ending of the Ice Age, during which the form and geography of the land and waters had been altered and changed repeatedly. Since this affected the human process of occupation it is necessary to delve into that geological and aquatic history first, somewhat awkward and technical though it is.

The crooked form of both the Bosporos and the Hellespont shows that both originated as rivers, a factor emphasised by the powerful currents flowing through them, which generally erode their valleys by swinging from one bank to another. This original form the Greeks appreciated,1 but other forces were also involved in their formation. Geologically both sides of both straits are similar, and it seems very probable that the rivers eroded their way along lines of least resistance. The banks of both straits show considerable and frequent fault lines,2 and it seems likely that the erosion process was further assisted by the shattering effect of earthquakes – both straits are along or close to the major North Anatolian fault, a very active earthquake zone. But the major continuous erosion was caused by seawater.

The Black Sea, originally a smaller, lower lake with no outlet to the south, was filled up by meltwater from the northern icecap at the end of the Ice Age, until it overflowed to the south, possibly at first by way of the wide valley now occupied by the lower course of the Sakarya River, to the east of the Bosporos, flowing along that valley and into the Gulf of Izmit. There is also the later possibility that it was also drained by the present Bosporos, which, as its course shows, clearly began as a river. But this supply of eroding water ceased soon after the Ice Age, when the meltwater was redirected westwards into the pre-Baltic Yoldia Sea instead of south along the great Russian rivers, and later for a second time when a dry climatic period reduced the flow of the great rivers. The drainage pattern thereby repeatedly changed.

In each of these periods the freshwater Black Sea shrank by evaporation, or as a result of its supply from the north being held up in the ice cap from about 10,000 BC, and again probably about 3000 BC. The level of the sea was lowered until its surface was considerably below the present level; it was also lower than the level of the Propontis and any outflow along the Sakarya River and the Bosporos stopped. There are remains of human habitation in what are now water-covered parts of the Black Sea, notably along the coast of modern Bulgaria. The two outlets from the Black Sea therefore ceased to flow, and when the Black Sea filled up again, only the Bosporos was used, perhaps because earth movements and earthquakes had opened it up, and blocked the Sakarya route.

The Propontis was similarly at a low level in these periods, lower than today, and it formed a lake unconnected to either the Black or the Aegean Seas, and with little water inflowing from the land; it was probably half its present size. As the level of the Mediterranean rose with the melting of the ice and the general rise in the world ocean levels, however, the Propontis was reinforced by saltwater flowing upstream through the Hellespont from the Aegean. This caused a deepening and widening of the channel by erosion. The force imposed was considerable, and this flow of saline water eventually brought the Propontis to a level above that of the Black Sea, with only the narrow trans-Bosporos mountain range separating the two seas. There, a small river (the ‘Bosporos River’) had begun eroding a valley, no doubt assisted by the freezing action of the ice on the damaged rock during the preceding Ice Age, and perhaps also by the effect of earthquakes, until the saline waters broke through, and the flow of water from the Propontis to the Black Sea began – that is, in the reverse direction of today. The force of the overflowing seawater powerfully worked to erode further the size and depth of the minor river valley, in much the same way as the Hellespont had been eroded. In addition, the sea water flowing into the formerly freshwater Black Sea changed that sea’s nature, killing off much of the freshwater aquatic life, which is the source of the toxic lower levels of the sea, and possibly driving the shoreline-living human inhabitants away. The force of the water increased the erosion effect at the ‘Bosporos River’, assisted by the heavily faulted land, and very quickly, at least in geological terms, the inflowing water, from both the Mediterranean and the Russian rivers, released by the melting of the ice, ‘filled’ the Black Sea.3

One of the effects of the high level of the Mediterranean waters had been to flood the plains surrounding the old, much reduced and smaller, Propontis, and this now happened also in the Black Sea, though to a much greater degree. (It has been theorised that this set off migrations of people displaced by the floods, in many directions; something like this is quite possible, though it seems unlikely that the effect was as rapid and widespread as has been supposed, or even that it necessarily happened – the population was hardly very large.) In time, of course, the sea levels on both sides equalised. The Black Sea continued to receive freshwater from the many rivers flowing into it from central and Eastern Europe – the Danube, the Bug, the Dniester, the Dnieper, and others, and overflowed through the Bosporos – but it also continued to receive seawater from the Mediterranean. Seawater being denser than fresh, the seawater flows northwards beneath the southward-flowing surface freshwater current, the two flowing in opposite directions, a singularly curious effect, which was always known to the local fishermen, who exploited it in their methods, but also was utilised by submarines in the Great War.4

Locally, in the Straits area, the effects of these geological and aquatic changes were no doubt dramatic, but were essentially marginal for any inhabitants, as the rise in the level drove them away from the old shores to the present dry land. Archaeological surveys have suggested that the region was essentially uninhabited until after all these major events had run their courses,5 and when the Bosporos ceased to flow the inhabitants left. Hunters of the Ice Age had, of course, moved through the area, and a string of sites has been located, identified by the finds of Palaeolithic and Mesolithic flints, along the present coast of the Black Sea and on either side of the Bosporos, and they spread into the Propontis shores.6 A substantial population has been suggested (unless, of course, this was the remains of a small population which moved about often). This would not be a particularly useful series of sites for hunters while the Black Sea level was lower than at present; perhaps these are the refugees that may have been displaced by the floods.

It was not until perhaps 5000 BC or later that any permanent settlement by human beings was established on the shores, and this was some centuries after the overflow of the Propontis into the Black Sea began, when the disturbances of the waters had ended, and the relative sea levels had probably stabilised. By this time, the melting of the Russian ice cap had increased the flow of the rivers, so that the flow of the sea’s surface water had reversed, and it now ran from the Black Sea into the Propontis, and from the Propontis into the Mediterranean, which is the process as it is now.

That the earliest known permanent settlements on the Straits date from the time following the decline of the obvious turbulence of the waters is hardly a coincidence. For one thing the changes in the sea levels in both the Propontis and the Black Sea involved substantial rises; if any settlement existed on the old shores, whenever and wherever they had been fixed, it quickly became drowned, and the remains are largely buried by the sediment which now covers the old sea floor. Those submerged off the Bulgarian coast were probably replicated in the Marmara.

The earliest known permanent settlement in any of the shores of the Straits is at Fikirtepe, on the north shore of the Gulf of Izmit in Kadikoy, the suburb of Istanbul east of the Bosporos, with another similar settlement of the same group of people at Pendik, a few miles to the east.7 The culture was at first identified at Fikirtepe, provoking considerable puzzlement as to its origins and affiliations, but it is now clear that it was actually derived from a longer-lasting set of settlements further east, of which the most important site is probably Ilipinar, on the shore of Lake Isnik, between that lake and the long Gednik Bay.8 This in turn is derived from, or at least influenced by, earlier settlements which had developed in the south centre of Anatolia, notably the well-known site at Hacilar. That is, the people of Fikirtepe – and Pendik – were migrants, or descendants of migrants, originating from the centre of Anatolia.

The economy of these settlements was still largely that of hunter-gatherers. In fairly typical Mesolithic style, they settled at places which were particularly rich in foods which could be collected without too much effort, particularly on sea bays and lakes, where shellfish could be collected, supplemented by fruits and vegetables acquired on the land. The earliest strata at Ilipinar are dated to about 5200 BC, and the latest are of about 4800; at Fikirtepe the settlement is dated to about 5000 BC. Across the Propontis on the western side there are two or three other sites: the cave at Yarimburgaz just west of Istanbul, Toptepe halfway along the Marmara coast, and Hoca Cesme near the mouth of the Maritza River west of the Thracian Chersonese on the Aegean coast. Other evidence has emerged in the course of the excavations connected with the extensive building activity in Istanbul in recent years.9 All of these sites from Ilipinar to Hoca Cesme are situated close to the sea, or on a bay or at a river mouth, or, at Ilipinar, between the sea and the Iznik Lake. They show much the same characteristics of lifestyle as Fikirtepe and Ilipinar, with the exception of the architecture of the houses. These western examples rather extend the timescale originally determined, with the site at Ilipinar commencing about 5200 BC and that at Toptepe lasting to about 4200 BC. (The cave at Yarimburgaz had also been used earlier in the Palaeolithic period.) There is some indication that the culture penetrated south into the Thracian Chersonese, at Kaynarca Mevkii; this is well inland, but it is suggested that the stream the site overlooks had originally been an inlet of the sea; and nowhere on that peninsula is far from the sea.10

These discoveries all show a slow progression from hunter-gatherers through to pastoral farming, to which was eventually added some arable farming, and the last phase at Ilipinar is on the verge of the Chalcolithic. And all of these sites come to an end well before 4000 BC.11 There were connections with Balkan cultures further inland, particularly at the Thracian sites, but only in the later phases. On the other hand, the ‘Fikirtepe’-type pottery which appears in later phases is found in southern Thrace and as far west as Thessaly; the people were mobile still and liable to shift to new sites; this may be the explanation of the desertion of their original sites, from Thrace on into Greece and north into the Balkans.12

This implication of mobility and wide trading connections is emphasised by the presence of obsidian being used in the place of, or as a supplement to, flint at several sites. The obsidian must have arrived from some distance, perhaps – again – from central Anatolia (where a centre of production sent its goods even further afield into Syria), though the island of Melos in the Aegean is perhaps as likely a source – which would imply seagoing capability by the miners or their customers. This was a continuation of the connection which had already existed before the development of the settled society, for obsidian is also found in some of the Palaeolithic pre-Fikirtepe sites along the Black Sea coast.13

In the southern part of the Straits region, on either side of the Hellespont, the story is much the same, but begins later. The fluctuations of the sea levels in the Hellespont area were even more dramatic than in the Bosporos region, and it is probable that this accounts for the rather later settlement of this southern area. The key site here is Kumtepe on the Asian coast, overlooking the entrance to the Hellespont, though it was not actually the first place a permanent settlement was made. The first village on the site had a similar economy to that which operated in much the same way at the Fikirtepe coastal site, where the inhabitants gathered shellfish to provide much of their diet – oysters were the preferred diet here – or perhaps only the easiest to gather. But this early settlement is only dated to about 3200 BC, a good millennium after the end of the Fikirtepe sites to the north.

A slightly earlier settlement is suggested at two other sites, Bezik Tepe, overlooking Bezik Bay on the Troad coast facing the Aegean, and Hanay Tepe, close to the valley of the Scamander River. The earliest layer at Kumtepe is labelled ‘Ia’, and it is thought these two sites show signs that their pottery was even earlier than Kumtepe Ia. However, the differences appear slight, and it is perhaps best to regard the settlements of the three places as originating at much the same time.14 The date of these settlements is well before 3000 BC, perhaps as early as 3300 BC.15 The economy of the earliest inhabitants was similar to that of the Fikirtepe inhabitants: some pastoral herding, hunting, and a strong bias towards shellfish, but they also rapidly adopted, or applied, agriculture, and in Kumtepe Ib the site had access to copper and the wealth to acquire small pieces of it.

The site of Kumtepe is a prime one: close to the sea, with freshwater in the river, at a fairly low level but clear of the sea and occupying a small fertile plain on a long narrow peninsula between the sea and the bay. That plain has since been formed by alluvium filling the bay, brought down by the Scamander River, partly as a result of deforestation in the pursuit of agriculture. Yet its first inhabitants settled there two millennia after the development of a similar society at Ilipinar, and the Fikirtepe sites had been abandoned for a millennium before the Kumtepe sites were settled.

Across the Hellespont, on the peninsula of the Thracian Chersonesos, the earliest site, apart from Kynarca Mevkii, was Karaagactepe (also known as ‘the grave of Protesilaos’), more or less contemporary with a settlement at Kumtepe, from which it was separated only by the mouth of the Hellespont. It is near the southern tip of the Thracian Chersonese, a substantial mound (‘tepe’ in Turkish is equivalent to the Arabic ‘tell’) and was first excavated in his usual brutal fashion by Schliemann and then by a team from the French military occupation forces in 1921–1922; a clear and accurate result can hardly be expected under the circumstances. The occupation layers are up to eleven metres thick, so not a great deal of the earliest material was reached – and neither excavation reached bedrock. The lowest finds were contemporary with the second level at Kumtepe (Kumtepe Ib), and there was still lower, that is, earlier, material.16

Occupation of this period appears at several sites in the peninsula, while on the opposite shore it is found only at Kumtepe; the early sites at Bezik Tepe and Hanay Tepe were abandoned after their early occupation and ‘Kumtepe Ib’ appears only at that site.17 But across the Hellespont there are at least seven sites, besides Karaagactepe, where pottery of that date and style has been found during surface surveys conducted by Ozdogan in the 1990s. These are all along the coast of the Strait, with none apparently either inland or on the western, Aegean, coast. It looks as though the early inhabitants of Kumtepe were particularly interested in this side of the Strait.18 We thus have clear evidence of contacts acro...

Table of contents

• Cover
• Title
• Copyright
• Contents
• Maps
• Introduction
• Chapter 1 The Early Settlers (before C.3000 BC)
• Chapter 2 Troy and its War (C.3000–C.1000 BC)
• Chapter 3 Thracians and Greeks (C.1000–546 BC)
• Chapter 4 Persians and Greeks (546–478 BC)
• Chapter 5 Under Another Empire – Athens (478–405 BC)
• Chapter 6 Precarious Independence (405–336 BC)
• Chapter 7 After Alexander (336–301 BC)
• Chapter 8 In the Seleukid Empire (301–223 BC)
• Chapter 9 Seleukids, Romans, Attalids (228–133 BC)
• Chapter 10 Into the Roman Empire (146–30 BC)
• Chapter 11 In the Roman Empire (30 BC–AD 180)
• Chapter 12 The New City (AD 180–330)
• Conclusion
• Notes
• Abbreviations
• Bibliography
• Plate section