ANKARA UNIVERSITY RESEARCH CENTER FOR MARITIME ARCHAEOLOGY (ANKÜSAM) Publication No: 1

Proceedings of the International Symposium

The Aegean in the Neolithic, Chalcolithic and the Early Bronze Age

October 13th – 19th 1997, Urla - İzmir (Turkey)

Edited by

Hayat Erkanal, Harald Hauptmann, Vasıf Şahoğlu, Rıza Tuncel

Ankara • 2008 ANKARA ÜNİVERSİTESİ / ANKARA UNIVERSITY SUALTI ARKEOLOJİK ARAŞTIRMA ve UYGULAMA MERKEZİ (ANKÜSAM) RESEARCH CENTER FOR MARITIME ARCHAEOLOGY (ANKÜSAM) Yayın No / Publication No: 1

Ön kapak: İzmir - Höyücek’de ele geçmiş insan yüzü tasvirli bir stel. M.Ö. 3. Bin. Front cover: A stelae depicting a human face from İzmir - Höyücek . 3rd Millennium BC. Arka kapak: Liman Tepe Erken Tunç Çağı II, Atnalı Biçimli Bastiyon. Back cover: Early Bronze Age II horse-shoe shaped bastion at Liman Tepe.

Kapak Tasarımı / Cover Design : Vasıf Şahoğlu

ISBN: 978-975-482-767-5

Ankara Üniversitesi Basımevi / Ankara University Press İncitaşı Sokak No:10 06510 Beşevler / ANKARA Tel: 0 (312) 213 66 55 Basım Tarihi: 31 / 03 / 2008

CONTENTS

Abbreviations …………………………………………………………………………………...... xi Preface by the Editors ………………………………………………………………………………… xiii Opening speech by the Mayor, Bülent BARATALI …...……………………………………………...... xxiii Opening speech by Prof. Dr. Ekrem AKURGAL ……………………………………...... xxv Opening speech by Prof. Dr. Christos DOUMAS……………………………………………………….. xxvii

LILIAN ACHEILARA Myrina in Prehistoric Times …..……………………………………………………………. 1

VASSILIKI ADRIMI – SISMANI Données Récentes Concernant Le Site Prehistorique De Dimini: La Continuité de l’Habitation Littorale depuis le Début du Néolithique Récent jusqu’à la Fin du Bronze Ancien ……………………………………………………………………………… 9

IOANNIS ASLANIS Frühe Fortifikationssysteme in Griechenland ………………………………………………. 35

PANAGIOTA AYGERINOU A Flaked-Stone Industry from : A Preliminary Report …………………………… 45

ANTHI BATZIOU – EFSTATHIOU Kastraki: A New Bronze Age Settlement in …………………………….. 73

MARIO BENZI A Forgotten Island: in the Late Neolithic Period ……………………………….. 85

ÖNDER BİLGİ Relations between İkiztepe by the Black Sea Coast and the Aegean World before Iron Age ……………………………………………………………………………... 109

TRISTAN CARTER Cinnabar and the : Body modification and Political Structure in the Late EB I Southern Cyclades ………………………………………………………...... 119

CHRISTOS DOUMAS The Aegean Islands and their Role in the Developement of Civilisation …………...... 131

ANTHI DOVA Prehistoric Topography of : The Early Bronze Age ………………………………. 141

NIKOS EFSTRATIOU The Neolithic of the Aegean Islands: A New Picture Emerging …………………...... 159

HAYAT ERKANAL Die Neue Forschungen in Bakla Tepe bei İzmir ..…………………………………………. 165

HAYAT ERKANAL Liman Tepe: A New Light on the Prehistoric Aegean Cultures …………………………… 179

JEANNETTE FORSÉN The Asea Valley from the Neolithic Period to the Early Bronze Age ……………...... 191

DAVID H. FRENCH Chalcolithic and Early Bronze Age Pottery of Southwest Anatolia ………………...... 197 viii Contents

NOEL GALE Metal Sources for Early Bronze Age Troy and the Aegean ………………………...... 203

BARTHEL HROUDA Zur Chronologie Südwestkleinasiens in der 2. Hälfte des 3. Jahrtausends v. Chr ...... 223

HALİME HÜRYILMAZ 1996 Rettungsgrabungen auf dem Yenibademli Höyük, Gökçeada / Imbros …………….. 229

ERGUN KAPTAN Metallurgical Residues from Late Chalcolithic and Early Bronze Age Liman Tepe …………………………………………………………………………...... 243

ANNA KARABATSOLI and LIA KARIMALI Etude Comparative Des Industries Lithiques Taillées Du Néolithique Final Et Du Bronze Ancien Egéen : Le Cas De Pefkakia ………………………………………….. 251

NECMİ KARUL Flechtwerkgabäude aus Osttrakien ……………………………………………………….. 263

SİNAN KILIÇ The Early Bronze Age Pottery from Northwest Turkey in Light of Results of a Survey around the Marmara Sea ………………………………………………………….. 275

OURANIA KOUKA Zur Struktur der frühbronzezeitlichen insularen Gesellschaften der Nord- und Ostägäis: Ein neues Bild der sogenannten “Trojanischen Kultur”…………….. 285

NINA KYPARISSI – APOSTOLIKA Some Finds of Balkan (or Anatolian) Type in the Neolithic Deposit of Theopetra Cave, …………………………………………………………………. 301

LAURA LABRIOLA First Impressions: A Preliminary Account of Matt Impressed Pottery in the Prehistoric Aegean ………………………………………………………………………… 309

ROBERT LAFFINEUR Aspects of Early Bronze Age Jewellery in the Aegean …………………………………… 323

KYRIAKOS LAMBRIANIDES and NIGEL SPENCER The Early Bronze Age Sites of and the Madra Çay Delta: New Light on a Discrete Regional Centre of Prehistoric Settlement and Society in the Northeast Aegean ……………………………………………………...... 333

YUNUS LENGERANLI Metallic Mineral Deposits and Occurences of the Izmir District, Turkey ………………… 355

EFTALIA MAKRI – SKOTINIOTI and VASSILIKI ADRIMI – SISMANI Les Sites Du Neolithique Recent Dans Le Golfe Pagasetique : La Transformation Des Sites De L’age De Bronze En Sites Urbains (Le Cas De Dimini) ……………………. 369

ELSA NIKOLAOU, VASSO RONDIRI and LIA KARIMALI Magoula Orgozinos: A Neolithic Site in Western Thessaly, ………………………. 387

EMEL OYBAK and CAHİT DOĞAN Plant Remains from Liman Tepe and Bakla Tepe in the İzmir Region ……………………. 399

Contents ix

DEMETRA PAPACONSTANTINOU Looking for ‘Texts’ in the Neolithic Aegean: Space, Place and the Study of Domestic Architecture (Poster summary) …………………………………...... 407

ATHANASSIOS J. PAPADOPOULOS and SPYRIDOULA KONTORLI – PAPADOPOULOU Some thoughts on the Problem of Relations between the Aegean and in the Early Bronze Age …………………………………………………. 411

STRATIS PAPADOPOULOS and DIMITRA MALAMIDOU Limenaria: A Neolithic and Early Bronze Age Settlement at ……………………… 427

DANIEL J. PULLEN Connecting the Early Bronze I and II Periods in the Aegean ……………………………….. 447

JEREMY B. RUTTER Anatolian Roots of Early Helladic III Drinking Behaviour …………………………………. 461

VASIF ŞAHOĞLU New Evidence for the Relations Between the Izmir Region, the Cyclades and the Greek Mainland during the Third Millennium BC …………………………………. 483

ADAMANTIOS SAMPSON From the Mesolithic to the Neolithic: New Data on Aegean Prehistory ……………………. 503

EVANGELIA SKAFIDA Symbols from the Aegean World: The Case of Late Neolithic Figurines and House Models from Thessaly …………………………………………………………... 517

PANAGIOTA SOTIRAKOPOULOU The Cyclades, The East Aegean Islands and the Western Asia Minor: Their Relations in the Aegean Late Neolithic and Early Bronze Age …………………….. 533

GEORGIA STRATOULI Soziale une ökonomische Aspekte des Chalkolithikums (spätneolithikum II) in der Ägäis aufgrund alter und neuer Angaben …………………………………………….. 559

GEORGE TOUFEXIS Recent Neolithic Research in the Eastern Thessalian Plain, Greece: A Preliminary Report ……………………………………………………………………….. 569

RIZA TUNCEL IRERP Survey Program: New Prehistoric Settlements in the Izmir Region ……………….. 581

HANNELORE VANHAVERBEKE, PIERRE M. VERMEERSCH, INGRID BEULS, BEA de CUPERE and MARC WAELKENS People of the Höyüks versus People of the Mountains ? …………………………………… 593

KOSTAS VOUZAXAKIS An Alternative Suggestion in Archaeological Data Presentations: Neolithic Culture Through the Finds from Archaeological Museum ……………….. 607

Closing Remarks by Prof. Dr Machteld J. MELLINK ………………………………………………. 611

Symposium Programme ……………………………………………………………………………… 615

Memories from the Symposium……………………………………………………………………… 623

Metal Sources for Early Bronze Age Troy and the Aegean

Noel GALE

ABSTRACT: The setting of this conference at Urla on the Aegean coast of Turkey, felicitously bringing together as it does Greek, Turkish and other colleagues, marks a new era in Aegean Bronze Age archaeology, with the important new excavations at Liman Tepe, Panaztepe and Bakla Tepe. Amongst many other aspects these excavations are destined to throw new light on our knowledge of early metallurgy in the eastern Aegean, in the study of which it is planned my Laboratory will take part in collaboration with Professor Erkanal and his colleagues. This paper reviews briefly what is already known from studies based on metal artefacts from Troy, Poliochni, Lemnos and Kastri on .

The study of early metallurgy in the Where metals are concerned, in the Aegean has a long history. Schliemann himself Aegean, Branigan2 exemplifies the traditional, not only brought his Aegean preoccupations to typological, approach, studying metal as a his excavation and study of Troy, but was also chronological indicator and an index of inter- intensely interested in the development of regional contact. Archaeologists and bronze technology in the Northeast Aegean. archaeometallurgist alike have attempted at one Schliemann had some of the first chemical level to establish where such new technologies analyses of prehistoric metals ever made carried as the use of tin bronze first developed, and out on material excavated by him at Troy, whether such a technology diffused out from especially from what we now know as Treasure where it was initially discovered or whether it A from Troy IIg. Prominent amongst the was independently discovered in different experts to whom Schliemann turned was societies. At another level scholars such as Chandler Roberts of the Royal Mint and Royal Renfrew3, dealing with Aegean prehistory, School of Mines in London1. As a graduate of discussed metallurgy not only in terms of the the sister college, the Royal College of Science diffusionist/autonomist divide but also in an in London, it has been a pleasure for me to take attempt to investigate the impact of the such studies a little further and in new development of metallurgy on the evolution of directions made possible by the advance of societies, in the context of indigenous state science over the past 115 years. formation, and as a technological advance enabling the imposition of power through gift

exchange and through warfare, and as a store of Preamble wealth reflecting the status of emerging leaders Exchange is a central concept in and allowing the development of hierarchical archaeology. Exchange of material goods is of societies. It should be mentioned here that course much the same as trade, though one Renfrew4, in discussing the rise of state must not neglect the possibilities of gift societies in the Aegean, did not rely only on exchange and reciprocity (exchange between discussions of the impact of metallurgy; he equals), or the varieties of trade such as emphasised that the essential drive towards redistribution or market exchange. Exchange in growth and change came from the interaction of a broader sociological sense includes all a number of subsystems through the multiplier interpersonal contacts, including the exchange effect of systems theory. While it was possible of information. In recent archaeological to stress the importance of certain processes, programmes an attempt has been made to tackle such as the increasing skill in metallurgy and the whole question of the production and the development of the vine and the olive, distribution of traded goods. Followed, where Renfrew’s emphasis was placed on the possible, by attempts to reconstruct the aggregate of the interactions. operation of a trading system as a whole. 2 Branigan 1974. 3 Renfrew, 1967, 1972. 1 Schliemann 1881 4 Renfrew 1972. 204 Noel GALE

Contra recent rather facile suggestions of archaeological scientists, of the lead isotope the Bradford school5, the whole question of compositions of hundreds of metal ore deposits prehistoric trade, inter-regional contacts and the have shown that, in the vast majority of cases, movement of ideas remains at the forefront of the lead isotope composition of a particular ore current archaeological interest6 and, where deposit varies only over a small range, usually movement of or trade in material goods is about 0.4% in total. Note that this does not, as concerned, this relies on the ability to identify erroneously claimed by Pollard and Heron11, the sources of traded goods. Conventional depend on the assumption that these deposits typology is not always a safe guide here7; are stratiform or of any particular style of artefact forms can be imitated, or may resemble metallogenesis; it is an empirical finding based each other by chance. Much more reliable solely on actual lead isotope data measured for evidence for trade can be provided if the raw a large number of ore deposits.. Relatively rare material of which an object is made can reliably deposits which contain significant U/Pb ratios be shown to have originated elsewhere than the do range much more widely in lead isotope region of the site in which the artefact is found. composition, but even then conclusions about Of course this sort of evidence is about where provenance may often be drawn from the raw materials came from; it cannot comparative lead isotope data for ore deposits necessarily tell us where was located the and artefacts, as for example has been shown by workshop in which the object was made. It can Pernicka et al. 199312. In contrast, major, minor however tell us a great deal about trade or the or trace element chemical compositions vary exchange of goods since, as the evidence from through an ore body and are changed by the Bronze Age shipwrecks of Cape Gelidonya8 smelting in the final metal relative to the and Uluburun show abundantly9, there was composition in the ore deposit; nevertheless great traffic in raw materials, including metal certain chemical elements and element ratios ingots of copper, tin and lead, in the Bronze are useful in providing provenancing evidence Age Mediterranean additional to that coming from lead isotope analysis13. For metals we thus need to identify some characteristic property of the metal which is present also in the metal ore source from which Introduction the metal was smelted, a property moreover The topic of this paper is a brief which is not significantly altered in the discussion of what natural science has processes of smelting, refining, casting, discovered about the metal sources which hammering, etc. In addition this property must supplied Early Bronze Age Troy and some be sufficiently different from one metal ore relevant parts of the Aegean, together with a source to the next so that each relevant metal limited discussion of its archaeological ore source can be distinguished from the others. implications. Prior to carrying out such work, it The property which comes closest to satisfying was clear that success would be these requirements is the isotopic composition archaeologically important. At the least it of the metal lead present, in major or minor would allow examination of the conflict amount, in the ore sources, and which is known between the earlier diffusionist hypothesis, to be carried over isotopically unchanged, which would have the primary development of within the errors of measurement, from the ore metallurgy in the East followed by transmission source to the final metal object, be this of lead, to the Aegean etc., and the later theories of silver or copper based metal10. Thousands of Branigan or Renfrew. Branigan, basing himself measurements, by isotope geochemists and largely on typological studies, essentially saw the early development of metallurgy in the 5 Pollard & Heron 1996. Aegean as due to “regional schools” of 6 Renfrew 1983; Renfrew & Bahn 1993. 7 Caskey 1986, 22. 8 Bass 1967. 11 Pollard & Heron 1996. 9 Pulak 1988. 12 Pernicka et al. 1993. 10 Gale 1989; Gale & Stos-Gale 1996. 13 Gale & Stos-Gale 1995.

Metal Sources for Early Bronze Age Troy and the Aegean 205 metallurgy. Renfrew was at pains to discard the Cycladic islands of Keos, Seriphos and Thera. notion of Ex Oriente Lux in favour of Clearly, an ore deposit which basically contains determinedly autonomous and independent only copper minerals cannot, in the Bronze developments of metallurgy in different parts of Age, have provided silver for a silver artefact, the Mediterranean, owing little to what had etc. Second, all three possible lead isotope happened in the East. Underlying both ratios must be used. Only if all the lead isotope approaches was the idea, for which little ratios of an artefact fall within the fields for an detailed evidence was produced, that ore source in both diagrams, can that ore source independent developments of metallurgy in the have provided the metal for the artefact in Mediterranean were made possible by the question. Thus the lead isotope ratios for lead presence in that region of easily accessible, metal artefact B, which fall in the upper abundant, sources of copper and some sources diagram both into the Lavrion and the Keos of tin. It will be seen that the studies discussed fields, but in the lower diagram falls only into here gave us some surprises which were not the Keos field, is consistent only with envisaged when the work began, and which go derivation of its lead from the Keos galena beyond merely providing evidence for or deposits. On the other hand artefact A has a against the theories just mentioned. lead isotope composition which falls into none of these fields, so that its metal cannot have I should begin by recording that almost derived form any of these ore deposits. all the work bearing on my topic has been done either in Oxford or by the Max Planck Institute for Chemistry in Mainz and the Max Planck Cypriot Copper Oxhide Ingots Institute for Nuclear Physics in Heidelberg; My illustration of the application of this sometimes by all. My talk will attempt briefly principle is the question of the source of copper to review just a little of what has come out from for the Late Bronze Age copper oxhide ingots, the work done in all three laboratories. It will or talenta, which have been excavated or found however be concentrated on Troy and the in Cyprus, Sardinia and Sicily, Mainland Northeast Aegean. I do not have time to cover Greece, Cycladic islands such as Keos, in , the rest of the Aegean, except briefly where in , in Ugarit, Hattuša, the Black Sea coast there seems to be a strong link with the eastern of Bulgaria, with recent fragments being region. excavated at the Turkish site of Tekirdağ on the However, since the discussion of the northern coast of the Sea of Marmara and at the Northeast Aegean can be complex, I am going Egyptian 1300 BC site of Piramesse Quantir in to start with something simpler. This is intended the Eastern Nile Delta. This is a question which just to illustrate the application of lead isotope has received much attention for more than 15 provenancing to an archaeologically important years by the team of Muhly, Maddin and problem, where the application seems to be Stech14, who have approached the problem, simple and demonstrably to succeed. with no real success, through a combination of chemical and metallographic analyses with

archaeological discussion. More recently the Principles of lead isotope analysis application of lead isotope provenancing has Figure 1 illustrates the quite simple bases proved much more successful15. of lead isotope analyses as used for The whole basis of the investigation of provenancing the metal of which an ancient this problem has been altered significantly, and artefact was made. First we must build up a much strengthened, by recent work at Oxford, number of lead isotope analyses of ores, so as to in which we have added about 200 good quality define the fields characteristic of the ore lead isotope data for over 25 different Cypriot deposits which could in principle have supplied ore deposits. This lead isotope data has proved the particular metal under consideration. As an to be in good accord with the latest geological illustration figure 1 shows the lead isotope fields characteristic of four Aegean ore 14 Muhly et al. 1988. deposits; those of Lavrion in and of the 15 Cf. Gale 1989.

206 Noel GALE and geophysical work in Cyprus16, which It seems clear that this brief glance at relates the formation of the ore bodies in lead isotope analyses applied to copper oxhide Cyprus to sub-oceanic processes occurring near ingots demonstrates a clear success story. a number of palaeo crustal spreading centres. It Together with further evidence presented should however be emphasised that lead isotope elsewhere19 it seems that the lead isotope data based conclusions about the origin of copper for shows that, from all over the Mediterranean, the oxhide ingots do not at all depend on the those post 1250 BC oxhide ingots which have mode of metallogenesis of any ore bodies. been analysed derive from copper smelted from one mining region in Cyprus, that of Apliki. It turns out that, for comparison with the oxhide ingots, only the ore deposits related to The Northeast Aegean one ancient crustal spreading axis in Cyprus, called by geologists the Solea axis17, are of The first lead isotope analyses of copper interest. The relevant mines are those of based metals from Troy were presented in a Mavrovouni, Apliki, Skouriotissa, Alestos and paper which I gave at the First Conference on Memi, near the Morphou Bay in the region of Southern European Archaeometry, held in the north west of the Troodos mountains. Figure Delphi in November 198420. A fuller 2 shows the lead isotope compositions in publication of this data appeared in November relation to our old “overall Cypriot field” 198421; this paper contained full archaeological (which was based on only about 30 ore information about the Troy IIg material which samples), and shows that these mines tend to had been analysed at Oxford, including have somewhat different isotopic compositions references to drawings of the objects22. This from each other, in an approximate progression material came from Schliemann’s excavations through Memi, Alestos, Skouriotissa, of Troy, included axes, chisels, daggers, knives, Mavrovouni, Apliki. In confirmation of earlier loop rings and vessel parts, and had been indications based on much less data18, the new sampled by the Stuttgart programme of metal lead isotope data for the Apliki deposit tend to analyses23 of the 1960’s-1970’s. Information fall below the extent of the ‘old Cypriot lead about the chemical composition of these isotope field’ in the 208Pb/206Pb versus artefacts had previously been published, within 207Pb/206Pb diagram. a comprehensive study of Anatolian copper alloy objects, by Esin24. Comparison, in figure 3, with the lead isotope compositions of copper oxhide ingots Eight objects25 from the EB II-EB IIIA found in Cyprus itself shows a very good match Yortan cemetery26 were also included in this with some of these Cypriot ores on the Solea early work, as were a number of objects from axis, principally with the ores from the mine the “Troas” hoard published, and analysed (not the archaeological site) of Apliki. The typologically, by Bittel27 who, arguing from the oxhide ingots whose isotopic analyses are typological standpoint, classified them as of the shown here vary widely in find spot on Cyprus, Troy IIg period, and pointed in some cases to from Enkomi on the East coast to Maroni on the links with the Pontids region. (Subsequently central Southern coast; they are all consistent and parallel to our work Pernicka and his with their copper having come from the Apliki colleagues28 and Seeliger et al.29 also published region copper ore deposits on Cyprus. Finally, figure 4 shows data for some of 19 the copper oxhide ingots found on Sardinia, Stos-Gale et al. 1997. 20 Gale & Stos-Gale 1986. which again fit closely with the Apliki ores, 21 Stos-Gale et al. 1984. with two fragments straying up into the region 22 Schmidt 1902. of overlap with the Skouriotissa ores. 23 Junghans et al. 1974. 24 Esin 1969. 25 Gale et al. 1985. 26 Kamil 1982. 16 Maliotis 1995. 27 Bittel 1959. 17 See Stos-Gale et al. 1997. 28 Pernicka et al. 1984. 18 Stos-Gale & Gale 1994. 29 Seeliger et al. 1985.

Metal Sources for Early Bronze Age Troy and the Aegean 207 work on the artefacts from Troy, Yortan and the suggested the regions indicated on the map of Troas). figure 8 as those where there were relatively nearby copper deposits, which might have The lead isotope analyses for Troy IIg supplied the copper used in Troy, Yortan, and copper based objects appear in Figure 5. There for the “Troas Hoard”. Apart from the copper was an immediate surprise, in that the lead ore deposits in Anatolia itself, there were not isotope data for Trojan artefacts spreads far far away the very large Strandža deposits in beyond, towards the upper right hand side of Bulgaria near the coast around Burgas; these the diagram, the normal confines of data for were not identified as a possible source by de most Aegean copper based artefacts, which are Jesus, but clearly should be considered since usually positioned in the lower half of this Chernykh33 discovered evidence for ancient diagram, with 208Pb/206Pb ratios below 2.09. working of these deposits, period largely Figure 6 shows this is not so for the objects unknown. In our publications of 1984 and 1985 from Yortan, which cluster into one group in we made cautious attempts to relate the lead the usual part of the Pb isotope diagram, but isotope data for Troy, Yortan and the “Troas figure 7 shows that the objects from the so- Hoard” to some of these possible source called “Troas Hoard” again have objects falling regions, based on the very scanty geological into the upper part of the lead isotope diagram. evidence for the probable age of these often Moreover the lead isotope data is not scattered complex deposits. randomly across the diagram, as it would be if we had a situation where the artefacts were These preliminary assessments were of made of random mixtures of copper coming course made redundant by the excellent from different copper ore deposits, but occupies programme of fieldwork in Turkey led by a number of discrete isotopic groups30. Wagner from Heidelberg, in the years from about 1983 to 198634. For the Taurus Isotope geochronology shows us that the Mountains more evidence was provided by lead isotope compositions of metal ore deposits Aslıhan Yener and her colleagues35. For the are related to first order to the geological age of first time we were provided with hard evidence the ore deposit31. The isotopic composition of from the point of view of ancient mining and, lead in the ore deposit passes sensibly via slag heaps, about ancient metal production, unchanged into the metal objects made from though there was in most cases no sound metal smelted from it. Consequently we can evidence to assign a date to the slag heaps. Fig. relate the isotopic compositions of the various 9 summarises some of the Heidelberg isotopic groups, into which the Troy IIg objects conclusions about the existence of ancient fall, to the geological age of the copper deposits copper mines in NW Anatolia relatively near to from which they were made. Figure 5 shows Troy and Yortan. Incidentally, their work these inferred ages in millions of years. For essentially discounted previous suggestions that instance, the uppermost group in figure 5 tin occurs in usable amount in this region, contains copper based objects which must have though there is evidence for an EB tin mine at been made of copper coming from copper Kestel in the Taurus Mountains36. The copper deposits of Precambrian age, whilst those mine, TG192 at Serçeörenköy, had definite falling in the lowermost group came from much evidence for prehistoric working. Wagner’s younger copper deposits of about Miocene age. work confirmed that there are no Cambrian or At the time when we were first trying to Precambrian copper deposits in Turkey, as is understand these results, the work of de Jesus32 true also for the whole Mediterranean region. Consequently, the not inconsiderable number of 30 The lead isotope diagrams given in this paper for Troy, copper based objects from Troy IIg and the the Troas ‘hoard’ and Yortan show only one of the two “Troas Hoard”, which have exotically high lead possible isotope diagrams which were illustrated in Figure 1; both diagrams are necessary for valid interpretation. The other diagrams for these artefacts 33 Chernykh 1978. are given in Stos-Gale et al. (1984), Gale et al. (1985). 34 Wagner et al. 1989. 31 See Faure 1986, 309-340. 35 Yener et al. 1989; 1991. 32 Jesus 1980. 36 Yener et al. 1989, 1993; Willies 1990.

208 Noel GALE isotope ratios, are made of metal which, it towns II, III and V, and the very high lead seems, cannot have been mined in Anatolia or isotope ratios appear in a few arsenical copper the Mediterranean regions, something which and tin bronze objects in towns III-V. It has was not known before lead isotope analyses been suggested that Thermi V was over by the were made of these objects. time of Poliochni Yellow. It may be wrong, as has sometimes been done, to equate these Our work on the copper-based objects sudden developments in Poliochni and Troy IIg from Troy IIg and the “Troas Hoard” was with the sudden appearance in the Royal confirmed by independent lead isotope analyses Cemetery at Ur, and at Ebla, of gold, tin bronze of the same objects by Begemann37. This is and lapis lazuli. The Royal Cemetery seems to brought out by fig. 31 in their paper38, where it date to ED IIIA, or about 2600 BC, at any rate is shown that a large number of analysed some hundreds of years earlier than Troy IIg. objects are far outside the region where fall the We may well be forced back to Ex Oriente Lux lead isotope compositions of ores in the Aegean in this case. On the other hand the view was or Anatolia. Note also from this diagram that expressed at a recent conference41 that the most of the copper based objects with exotic relative dating of the Royal Cemetery at Ur and lead isotope ratios in this diagram are of tin Troy IIg is sufficiently imprecise that there bronze, identified as filled black symbols. might be overlap between these two periods. Tin bronze is an alloy which, in the NE It is usual to look to Afghanistan for the Aegean, becomes dominant in Troy IIg times, ancient source of lapis lazuli42, and both tin and though some is now known from true Troy I gold also occur there, leading Muhly to suggest horizons at Beşik-Tepe on the western shore of that all three came to Mesopotamia (and to the Trojan plain. This is well brought out in Northern Syria) via an overland route across figure 9, based on the data of Muhly and Iran, though Berthoud has suggested a route via colleagues39 for copper alloy artefacts from the the Persian Gulf43. site of Poliochni on the Northeast Aegean island of Lemnos. It shows the absence of tin A feature of the high lead isotope ratios, bronzes or high lead isotope ratios in the Blue linking a high proportion of Troy IIg artefacts or Green periods, little in the Red period, and a with very old ore deposits, is that they are not sudden explosion of tin bronze and high lead confined to tin bronzes. Some of the objects isotope ratios in the Yellow period. which plot in the ‘exotic’ range of the lead isotope ratio diagram are of copper. This Thus a combination of chemical analysis suggests that the high lead isotope ratios are not and lead isotope analyses has shown us for the tin metal in a tin bronze, but rather are something of importance for archaeology. In diagnostic of the copper used to make these tin the Troy IIg period, the Anatolian EB III of bronzes. This is further supported by the fact about 2300-2200 BC, artefacts from Troy itself, that the concentrations in the bronzes of Sn and Poliochni and the “Troas Hoard”, show a Pb are not positively correlated, so the Sn/Pb sudden increase of true tin bronzes, and the ratio in the bronzes is not constant. Also, sudden appearance of high lead isotope ratios, objects with the same Sn/Pb ratios have very indicating the use of metal from geologically diverse Pb isotope ratios. On present evidence old metal ore deposits which do not exist in the this would be expected, since what analyses Aegean or in Anatolia, though they do exist in there are for the lead content of cassiterite tin Central Asia. At the same time gold appears in ores44 show low lead contents in comparison the Treasures which Schliemann found at Troy, with most copper ores. in the gold jewellery in Poliochni Yellow, and in the princely burials at Alaca Hüyük. At All evidence suggests that the high lead Thermi on Lesbos40, a few tin bronzes appear in isotope ratios are characteristic of the copper

37 Reported in Pernicka et al. 1984. 41 Moscow 1996. 38 Pernicka et al. 1984. 42 von Rosen 1988. 39 Muhly et al. 1991. 43 von Rosen 1990 discusses possible trade routes. 40 Stos-Gale 1992. 44 e.g. Dulski 1980.

Metal Sources for Early Bronze Age Troy and the Aegean 209 metal used to make the tin bronzes. Since no Islands. Note that this settlement site is to be Anatolian or Aegean copper deposits exhibit distinguished from the nearby cemetery site of these ‘exotic’ isotope ratios, it follows that Chalandriani on Syros, which was excavated by these bronzes were not made by alloying local Tsountas49. copper with imported tin metal. It is possible The material analysed in Oxford that bronze was made at, or near, the comprises the copper based artefacts coming geologically old copper ore deposits, coming to from the excavation of EC IIIA Kastri by Troy and Poliochni as the tin/bronze alloy. Bossert50. These comprise an Early Bronze Age Since there are at Poliochni and Troy a few toolkit of nine chisels, a saw and two awls with, copper objects with high lead isotope in other parts of the site, two small daggers, two compositions, it cannot quite be excluded that flat axes and a spearhead, illustrated by tin and copper came separately and were made Bossert51 and Renfrew52. Branigan53 showed into the alloy at Troy etc. Coupling this with the that some of these copper based metal objects fieldwork, which seems to show that there are have good parallels in Anatolian objects no tin ore deposits in NW Anatolia, we can no excavated at Troy and at Thermi on Lesbos. longer accept as probable the hypotheses of Branigan, Renfrew and others that there was an It was found already by Junghans and his independent discovery and development of tin colleagues54, with further analyses in Oxford55 bronze technology in Western Anatolia, based that Kastri stood apart from other Cycladic on local tin and copper sources. The existence sites, in having a large number of tin bronzes, of an EB tin ore source at Kestel45 in the Taurus similar to Troy IIg, whereas other Cycladic Mountains does of course seem to leave it open islands had arsenical copper as their typical that, at Troy, tin from further away in Anatolia metal56. This contrast was emphasised in might conceivably have been utilised. diagrams published in 1984 and 198557, which also revealed striking similarities between the To repeat, the high lead isotope ratios for alloy compositions of metal artefacts from Troy IIg bronzes point to the exploitation of Kastri and Troy IIg. geologically old, Palaeozoic, copper deposits which do not occur in Anatolia or the Aegean, These striking similarities extend also to but do occur in Central Asia. In Afghanistan the lead isotope data, shown in figure 10. This there are not only tin deposits, but also sources diagram compares the Kastri copper based of lapis lazuli and gold which could have objects to those from Troy IIg, the “Troas supplied the other precious commodities Hoard”, Yortan and Thermi I & II58. appearing at Troy and related sites towards the Similarities between alloy compositions alone end of the 3rd. Millennium BC, perhaps via a might have been due to chance. Similarities in trade route across the Black Sea. There is the lead isotope composition also seems to indicate apparent difficulty that the copper, and tin, that the people of Kastri had their metals, or at deposits in Afghanistan so far described in the least the tin bronzes with high lead isotope archaeometallurgical literature46 are ratios, ultimately from the same ore deposits geologically young; but geological reports on which supplied the NE Aegean sites of the mineral deposits of Afghanistan47 reveal Poliochni, Lemnos, the Troas Hoard and Troy. that the complex geology of Afghanistan When in 1984 we first noticed these isotopic embraces also Precambrian/Cambrian deposits similarities, we had only the Troy IIg and of copper, though most deposits of tin are connected with the geologically much younger 49 Tsountas 1899. Eocene/Oligocene magmatism. 50 Bossert 1967. 51 Bossert 1967. I turn now to my other major topic - the 52 48 Renfrew 1972, 343. hill fort site of Kastri on Syros in the Cycladic 53 Branigan 1974. 54 Junghans et al. 1974. 55 Stos-Gale et al. 1984. 45 Yener et al. 1989. 56 Gale & Stos-Gale 1989. 46 Berthoud et al. 1977. 57 Stos-Gale et al. 1984, 32, Fig. 4; Gale et al. 1985, 156, 47 e.g. Afghanistan 1973. Fig. 4. 48 Bossert 1967. 58 Stos-Gale 1992.

210 Noel GALE

“Troas Hoard” objects as comparanda. The between EH II and EH III.... I still believe, as addition of Thermi increases confidence in the formerly, that many of the pots do indicate a groupings, whilst addition also of the objects distinct change in perception, having been made from Poliochni Yellow59 does so yet further. In by people of different traditions and with 1984 we felt that the metallurgical evidence, different techniques”, and also: “Within the plus the evidence that Kastri was a short period, present limits of our knowledge I look for hill top fortress with Anatolian pottery, and in certain events: the advent of new people (not uneasy relations with its neighbours, was necessarily hostile invaders) coming quite sufficient to put forward the hypothesis that the possibly from Asia Minor, with new knowledge people at Kastri were invaders or refugees from and new energies, to the central and the Norteast Aegean. Southeastern Greek mainland at the beginning Since then Christos Doumas60 has made of EH II; the arrival of other vigorous people, a much wider review of the evidence for the again quite probably from the East, some Northeast Aegean islands, which controlled the centuries later bringing elements of Anatolian entrance to the Dardanelles, and the Cycladic material culture to the Aegean islands, while a islanders, who may have controlled traffic related but independent group, more warlike or between Crete and the Greek mainland. meeting stronger resistance, came westward and Doumas demonstrates that contacts between the took many of the EH II settlements by force, two maritime communities were established at ...... ”. least in the Early Cycladic II period. His review It is in that light that we must interpret seems to be in accord with the interpretation we the Kastri group people’s use of tin bronze, as made more narrowly for Kastri on Syros. something foreign to the contemporary Doumas shows that the EC IIIA hill fort sites in Cycladic people65, but pointing to the origin of the Cyclades, on Syros, , Delos, and the Kastri group people from the Northern Christiana on Thera, have produced exclusively Aegean around Troy, Poliochni and Thermi, Kastri group material. These sites share not, as has been interpreted by some via the common features unknown in other EC Kastri group pottery, from Southwestern settlements: location on a hill top near the sea, Anatolia66. It seems a powerful argument small size, and short life. In at least two of the against linking the Kastri group with hill-forts, life ended violently. Between main Southwestern Anatolia rather than the 61 and outer walls at Kastri, Tsountas found a Northeastern Aegean and Troy that there are so quantity of pebbles, interpreted as sling-stones, far no sites contemporary with Troy IIg or EC and a spearhead. At Panormos on Naxos masses IIIA in Southwestern Anatolia which have of similar pebbles were recovered from the two produced tin bronzes or which have exhibited bastions protecting the entrance to the citadel. A the exotic lead isotope ratios. Apart from Troy spearhead was found, just outside the gate; IIg, Poliochni and Thermi, tin bronzes in inside it, in a layer of heavy burnt remains, Anatolian EB II levels occur only at the central were recovered the few vessels yielded by the Anatolian sites of Alaca Hüyük, Alişar and excavation. It seems that both settlements Ahlatlibel, which may be a further indication suffered attacks from outside, and that that tin was coming to Troy from further east. 62 Panormos was abandoned on fire . The people Even should we admit connections between residing in the hill forts seem to have been Troy IIg ceramics and contemporary ceramics intruders, not local Cycladic people. There is a in Cilicia, contemporary strata in sites in Cilicia 63 suggestion of this idea also by MacGillivray : such as Mersin67 do not contain high tin ‘newcomers settle in the islands...’, whilst bronzes and do not have copper based metal Caskey64 writes: “How great is the difference

59 Pernicka et al. 1990. 65 Gale & Stos-Gale 1989; Stos-Gale et al. 1984; Bossert 60 Doumas 1988. 1967. 61 Tsountas 1899. 66 E.g. Davis 1992, but on the dangers of relying too 62 Doumas 1963. much on comparisons of ceramic typology alone see 63 MacGillivray 1984. Caskey 1986. 64 Caskey 1986. 67 Gale et al 1985; Esin 1969; Berthoud et al. 1977.

Metal Sources for Early Bronze Age Troy and the Aegean 211 artefacts with the exotic lead isotope ratios As Doumas72 has suggested, from the which link Troy IIg, Poliochni, Thermi and occurrence of Kastri group material at Pefkakia Kastri on Syros. Davis’ argument68 against near Volos, at Manika, at Raphina in Attica, at linking the Kastri group with the presence of , the Kastri group intruders probably Anatolian people at these sites seems weak, came south through the route via , being based almost solely on his estimate that at following a route known to them since EC II, Ayia Irini the Anatolian features in the Kastri when they had been used to coming south on Group ceramics constitute only about 10% of peaceful missions. The Kastri group people the pottery in use, whilst at Kastri on Syros and probably fled from the destructions, such as are Panormos on Naxos the ceramics are not fully recorded at the end of Poliochni Yellow, Troy published so that a similar estimate cannot be IIg, and Thermi V. Although Troy was rebuilt made. Apart from being somewhat of an after the destructions, Poliochni and Thermi argument from silence, this does not take into were not; Lesbos and Lemnos were abandoned. account possible differences between sites and As Doumas writes “Their inhabitants, seems to place altogether too much emphasis on experienced mariners, must have left by sea. ceramic evidence alone. Davis seems not to And their safest course was to follow their give sufficient weight to the cumulative known seaway via the straits of Euboia and go evidence adduced69 not only for ceramics, to places with which they had relations and where Doumas has emphasised that the where continuation of their maritime activities contemporary ‘Dark-faced’ and Kastri group could be guaranteed. Such places, as we have pottery do not co-exist on the same site, whilst seen, included Manika and Aigina73. Having the hill-fort sites in the Cyclades ‘..have established at least two strongholds in the south, produced exclusively Kastri group material70. and in need for further territory for their Beyond ceramics, Doumas also emphasised that activities, I suggest that these refugee-pirates “These sites all share some common features, tried to expand their authority into the Cyclades unknown in other Early Cycladic settlements: where they hastily settled on some remote situation on a hill top near the sea, small size, coastal hill-tops which they fortified in their short life. Moreover, at at least two out of the own way. ... Finally the intruders were expelled three excavated hill-forts, life ended violently”, from the Cyclades as is suggested by the short as was explained above. “Their inhabitants life of the hill forts and by their violent were buried in isolated graves near other abandonment”74. cemeteries”71. To all this is to be added the evidence of the metals, where the use of tin Conclusions bronze and the exotic lead isotope compositions link Kastri on Syros with Troy IIg, Poliochni In summary, the leading discoveries and Thermi. Doumas surveys a range of which scientific investigations of metal evidence which suggests that, prior to EC III, provenance have made in relation to the the mariners maintained contacts Northeast Aegean seem to be: with Euboia and Aigina, perhaps with the tacit 1) In the Troy IIg-Poliochni Yellow agreement of the Cycladic people, using the periods there began the penetration into the NE Northern Sporades as stepping stones and Aegean of copper from geologically old, passing through the straits of Euboia, rather Precambrian, copper sources which are not to than risking the perilous crossing across the be found in Anatolia or the Aegean, coupled central Aegean, whose winds and waters are with the common use of tin bronze. Much of the often extremely rough. This status quo seems to isotopically ‘exotic’ copper is in the form of tin- have been kept until the appearance of the hill- bronze. Both tin and the ‘exotic’ copper may forts in the Cyclades in EC III. have come from Afghanistan; cf. lapis lazuli. Although most of both the tin and the copper

68 Davis 1992. 69 Stos-Gale et al. 1984; Doumas 1988. 72 Doumas 1988. 70 Doumas 1988, 25. 73 Cf. Walter & Felten 1981, 28. 71 Doumas 1988, 26. 74 Doumas 1988, 28, Walter & Felten 1981.

212 Noel GALE deposits visited in Afghanistan by the French Acknowledgements 75 expedition are geologically too young, We thank Dr H. Schickler who kindly provided our Oligocene to Jurassic, they attributed two samples of artefacts from Troy, the Troas and Yortan from copper deposits near Ainak to the Proterozoic, samples remaining from the Arbeitsgemeinschaft für Metallurgie des Altertums, Stuttgart. Thanks are also due and later Soviet geological work has identified to the organisations which have provided financial support other Precambrian/Cambrian Afghan copper to the Isotrace laboratory over the past 20 years: the deposits. We should not lose sight of the fact Volkswagen-Stifftung, the Science Research Council, the Leverhulme Trust, the British Academy, the Leventis that there are also tin deposits (Uzbekistan, Foundation, the Institute of Aegean Prehistory and the Tadjikistan, Kazakstan, Kirgizstan) and University of Oxford. Finally, acknowledgement is made Precambrian copper deposits in central Asia, to Professor W. Gentner and Professor Lord Renfrew of some of which Chernykh’s work76 shows were Kaimsthorn for instigating this avenue of research in Oxford, and to my colleague Dr Z. A. Stos-Gale, who has exploited in the Bronze Age. participated fully in all this research. 2) Close similarities between the isotopic and alloy compositions of artefacts from Troy IIg, Poliochni Yellow, the “Troas Hoard”, Yortan and Kastri/Syros suggest time NOEL GALE synchronism between Troy IIg, EC IIIA Kastri, Isotrace Laboratory, Nuclear Physics Building, Keble Road, Poliochni Yellow and the “Troas Hoard”. Oxford, OX1 3RH. UK 3) A combination of other archaeological evidence with the scientific evidence for metals from Kastri suggests that the Kastri Group people were not native Cycladic islanders but Anatolian refugees from the Northeast Aegean, fleeing from the end Troy IIg and Poliochni Yellow troubles in the north east Aegean. 4) It becomes difficult to argue for independent introduction of tin bronze technology in the Cyclades or Northeast Aegean, if the alloy components came from so far away, and bearing in mind the prior adoption of tin bronzes in Mesopotamia, who may also have had tin from Afghanistan or one of the former Soviet Central Asian republics. 5) A large number of Troy IIg, Poliochni- Yellow, etc. objects do not have the ‘exotic’, high, lead isotope ratios, but have ratios consistent with the copper having come, for different artefacts, from Aegean/Anatolian ores, including ores from the Troad and from the NE Pontids, and from the Cyclades. 6) Lead isotope analyses show that no artefacts from Kastri, Troy, Poliochni, etc. have copper coming from the Strandzha copper ore deposits on the Black Sea coast of Bulgaria.

75 Berthoud et al. 1977. 76 Chernykh 1989.

Metal Sources for Early Bronze Age Troy and the Aegean 213

Bibliography: Afghanistan, 1973, Geology and Mineral Resources of Afghanistan, Department of Geological Survey, G. Besulov & E.N. Kondrashkini. Kabul. Bass, G.F. 1967, Cape Gelidonya, a Bronze Age Shipwreck, Transactions American Philosophical Society, new series 57 (8). Philadelphia. Berthoud, Th., R. Besenval, J.P. Carbonnel, F. Cesbron & J. Liszak-Hours 1977, “Les Anciennes Mines d’Afghanistan”, Rapport Preliminaire, Commisariat a l’Energie Atomique. Fontenay-aux-Roses, 30pp. Bittel, K. 1959, “Beitrage zur Kenntnis anatolischer Metalgefässe der zweiten Hälfte des dritten Jahrtausends v. Chr.”, JdI LXXIV, 1-34. Bossert, E.-M. 1967, “Kastri auf Syros”, ADelt 22, 53-76. Branigan, K. 1974, Aegean Metalwork of the Early and Middle Bronze Age. Oxford. Caskey, J.L. 1986, “Did the Early Bronze Age end?”, in: Cadogan, G. (ed.) 1986, The End of the Early Bronze Age in the Aegean, Cincinnati Classical Studies, New Series, Vol. VI. Leiden, 1-30. Chernykh, E. N. 1978, Gornoe delo i metallurgia v drevneishei Boulgarii. Sofia. Chernykh, E. N. 1989, Drevnia Metallurgia Severnon Evrasii, Akademia NAUK CCCP. Moscow. Davis, J.L. 1992, “Review of Aegean Prehistory I: The Islands of the Aegean”, AJA 96, 699-756., esp. 754-755. de Jesus, P.S. 1980, The Development of Prehistoric Mining and Metallurgy in Anatolia, BAR Int. Series 174. Oxford. Doumas, C. 1963, “Archaiotitetes kai Mnemeia Kykladon”, ADelt 19B, 411-412. Doumas, C. 1988, “EBA in the Cyclades: Continuity or Discontinuity?”, in: French, E.B. & K.A. Wardle (eds.) 1988, Problems in Greek Prehistory. Bristol, 21-30. Dulski, P. 1980, “Spurenanalyse als Hilfsmittel bei der geochemischen Untersuchung der Genese bolivianischer Sn- Lagerstätten”, Berliner Geowissenschaftliche Abhandlungen, Reihe A/Band 28, 79. Esin, U. 1969, Kuantitatif Spektral Analiz Yardımıyla Anadolu’da Baslangıcından Asur Kolonileri Çağına Kadar Bakır ve Tunç Madenciliği. İstanbul. Faure, G. 1986, Principles of Isotope Geology. New York, 309-340. Gale, N.H., Z.A., Stos-Gale & G.R. Gilmore 1985, “Alloy Types and Copper Sources of Anatolian Copper Alloy Artifacts”, AnatSt 35, 143-173. Gale, N. H. & Z. A. Stos-Gale 1986, “Anatolian and Cycladic Metal Sources”, in: Proceedings of the First Conference on Southern European Archaeometry, Delphi, 1984, PACT Journal 15, 13-32. Gale, N.H. 1989, “Lead Isotope Studies Applied to Provenance Studies - a Brief Review”, in: Maniatis, Y. (ed.) 1989, Archaeometry. Amsterdam, 469-502. Gale, N.H. & Z.A. Stos-Gale 1989, “Some Aspects of Early Cycladic Copper Metallurgy”, in: Domergue, C. (ed.) 1989, Mineria y Metalurgia en las Antiguas Civilizaciones Mediterraneas y Europeas, Tomo I. Madrid, 21-37. Gale, N.H. 1991, “Copper Oxhide Ingots: Their Origin and Their Place in the Bronze Age Metals Trade in the Mediterranean”, in: Gale, N.H. (ed.) 1991, Bronze Age Trade in the Mediterranean, SIMA XC. Jonsered, 197-239. Gale, N.H. & Z.A. Stos-Gale 1995, “Comments on ‘Oxhide Ingots, Recycling, and the Mediterranean Metals Trade”, JMA 8.1, 33-41. Gale N.H., & Z.A. Stos-Gale 1996, “Lead Isotope Methodology: The Possible Fractionation of Lead Isotope Compositions during Metallurgical Processes”, in: Demirci, S., A.M. Özer & G.D. Summers (eds.) 1996, Archaeometry ‘94, Proceedings of the 29th International Symposium on Archaeometry, TÜBITAK, Ankara, 287-299. Junghans, S., E. Sangmeister & M. Schröder 1974, Kupfer und Bronze in der frühen Metallzeit Europas. Katalog der Analysen Nr. 10,041-22,000 (mit Nachuntersuchungen der Analysen). Berlin. Kâmil, T. 1982, Yortan Cemetery in the Early Bronze Age of Western Anatolia, BAR Int. Series I45. Oxford. MacGillivray, J.A. 1984, “The Relative Chronology of Early Cycladic III”, in: MacGillivray, J.A. & R.L.N. Barber (eds.) 1984, The Prehistoric Cyclades. Edinburgh, 70-77. Maliotis, G. 1995, “The Structural Control of the Mineralisation of North-north-east Troodos, Cyprus”, in: Fifteenth Congress of the Carpatho-Balcan Geological Association, Geol. Soc. Greece Spec. Publ. 4/2, 767-770. Moscow 1996, Troy and its Treasures, State Pushkin Museum. Moscow. Muhly, J.D., F. Begemann, Ö. Öztunali, E. Pernicka, S. Schmitt-Strecker & G.A. Wagner 1991, “The Bronze Age Metallurgy of Anatolia and the Question of Local Tin Sources”, in: Pernicka, E. & G.A. Wagner (eds.) 1991, Archaeometry ‘90, Birkhauer. Basel, 209-220. Muhly, J.D., R. Maddin & T. Stech 1988, “Cyprus, Crete and Sardinia: Copper Oxhide Ingots and the Bronze Age Metals Trade”, RDAC, 281-298.

214 Noel GALE

Pernicka, E. , T.C. Seeliger, G.A. Wagner, F. Begemann, S. Schmitt-Strecker, C. Eibner, Ö. Öztunali & I. Baranyi 1984, “Archäometallurgische Untersuchungen in Nordwestanatolien”, JbZMusMainz 31, 533-599. Pernicka, E., T.C. Seeliger, G.A. Wagner, F. Begemann 1984, “Archäometallurgische Untersuchungen in Nordwestanatolien”, JbZMusMainz 31, 533-599. Pernicka, E., F. Begemann, S. Schmitt-Strecker & A.P. Grimanis 1990, “On the Composition and Provenance of Metal Artefacts from Poliochni on Lemnos”, OJA 9, 263-297. Pernicka, E., F. Begemann, S. Schmitt-Strecker & G.A. Wagner 1993, “Eneolithic and Early Bronze Age Copper Artefacts from the Balkans and their Relation to Serbian Copper Ores”, PZ 68, 1-54. Pollard, A.M. & C. Heron 1996, Archaeological Chemistry, The Royal Society of Chemistry. Cambridge. Pulak, C. 1988, “The Bronze Age Shipwreck at Uluburun, Turkey: 1985 Campaign”, AJA 92, 1-37. Renfrew, A.C. 1967, “Cycladic Metallurgy and the Aegean Early Bronze Age”, AJA 71, 1-20. Renfrew, A.C. 1972, The Emergence of Civilization: the Cyclades and the Aegean in the Third Millennium B.C. London. Renfrew, C. 1983, “Divided we Stand: Aspects of Archaeology and Information”, American Antiquity 43, 3-16. Renfrew, C. & P. Bahn 1993, Archaeology. London, 307-370. Schliemann, H. 1881, Ilios, the City and Country of the Trojans. New York (reprinted 1976). Schmidt, H. 1902, Heinrich Schliemann’s Sammlung Trojanischer Altertümer. Berlin. Seeliger, T.C., E. Pernicka, G.A. Wagner, F. Begemann, S. Schmitt-Strecker, C. Eibner, O. Oztunali & I. Baranyi 1985, “Archäometallurgische Untersuchungen in Nord- und Ostanatolien”, JbZMusMainz 32, 597-659. Stos-Gale, Z.A., N.H. Gale & G.R. Gilmore 1984, “Early Bronze Age Trojan Metal Sources and Anatolians in the Cyclades”, OJA 3.3, 23-43. Stos-Gale, Z. A. 1992, “The Origin of Metal Objects from the Early Bronze age site of Thermi on the island of Lesbos”, OJA 11.2, 155-177. Stos-Gale, Z.A. & N.H. Gale 1994, “Metals”, in: Knapp, A.B. & J.F. Cherry, Provenience Studies and Bronze Age Cyprus. Madison, 92-121. Stos-Gale, Z.A., G. Maliotis, N.H. Gale & N. Annetts 1997, “Lead isotope Characteristics of the Cyprus Copper Ore Deposits Applied to Provenance Studies of Copper Oxhide Ingots”, Archaeometry 39.1, 83-123. Tsountas, C. 1899, “Kykladika II”, AEphem, 73-134. von Rosen, L. 1988, Lapis Lazuli in Geological Contexts and in Ancient Written Sources, SIMA Pocket Book 65. Partille. von Rosen, L. 1990, Lapis Lazuli in Archaeological Contexts, SIMA Pocket Book 93. Jonsered. Wagner, G.A., Ö. Öztunali & C. Eibner 1989, “Early Copper in Anatolia: Archaeometallurgical Field Evidence”, in: Hauptmann, A., E. Pernicka & G.A. Wagner (eds.) 1989, Archäometallurgie der Alten Welt. Bochum, 299-306. Walter, H. & F. Felten 1981, Alt-Ägina III.1: Die vorgeschichtliche Stadt: Befestigungen, Häuser, Funde. Mainz. Willies, L. 1990, “An Early Bronze Age Tin Mine in Anatolia, Turkey”, Bull. Peak District Mines Historical Society 11.2, 91-96. Yener, K.A., H. Özbal, A. Minzoni-Deroche & B. Aksoy 1989, “Bolkardag: Archaeometallurgy Surveys in the Taurus Mountains, Turkey”, National Geographic Research 5.4, 477-494. Yener, K.A., H. Özbal, E. Kaptan, A.N. Pehlivan & M. Goodway 1989, “Kestel: an Early Bronze Age Source of Tin Ore in the Taurus Mountains, Turkey”, Science 244, 200-203. Yener, K.A., E.V. Sayre, E.C. Joel, H. Özbal & I.L. Barnes 1991, “Stable Isotope Studies of Central Taurus Ore Sources and Related Artefacts from Eastern Mediterranean Chalcolithic and Bronze Age Sites”, Journal of Archaeological Science 18, 541-77. Yener, K.A. & P. Vandiver 1993, “Reply to J.D. Muhly, “Early Bronze Age Tin and the Taurus”, AJA 97, 255-264.

Metal Sources for Early Bronze Age Troy and the Aegean 215

List of Illustrations: Fig. 1: Illustration of the principles of provenancing metals using lead isotope analyses. Points A and B represent the lead isotopic compositions of two metal artefacts. Lead artefact A cannot have been made of ores from any of the ore deposits Lavrion, Keos, Seriphos and Thera whose lead isotope fields are shown here. Lead artefact B is fully consistent with having been made of lead ores from Keos, but not from any of the other ore deposits. Fig. 2: Lead isotope analyses of Cypriot ores from mines associated with the Solea palaeo spreading axis, compared with the old ‘Cypriot field’, which was based on an inadequate number of ore samples. Fig. 3: Lead isotope analyses of Late Bronze Age copper oxhide ingots found in Cyprus, compared with the lead isotope compositions of ores from mines associated with the Solea axis. Fig. 4: Lead isotope analyses of some Late Bronze Age copper oxhide ingots found in Sardinia, compared with the lead isotope compositions of ores from mines associated with the Solea axis. Fig. 5: Lead isotope analyses of copper alloy artefacts from Troy IIg, showing how they fall into discrete groups of isotopic composition. Model lead isotope geological ages are shown for the average lead isotopic composition of each group, showing that two groups of artefacts have been made of copper ores coming most probably from Precambrian ore deposits. Fig. 6: Lead isotope analyses of copper alloy artefacts from Yortan, showing that they fall into but one isotopic group, and do not have the exotic isotopic compositions exhibited by some Troy IIg objects. Fig. 7: Lead isotopic compositions of copper alloy artefacts from the Troas hoard, which again fall into discrete groups and exhibit the exotic compositions characteristic of some Troy IIg objects. Fig. 8: Map of Anatolia, after de Jesus 1980, showing metal ore deposits. Also shown is the position of the important copper ore deposits of Strandzha in Bulgaria. Fig. 9: Map of Anatolia, after Wagner and colleagues, showing probable prehistoric copper ore deposits and metal processing sites as identified by the Heidelberg fieldwork. The previously suggested tin ore deposit at Soğukpınar has now been discounted by Wagner. Fig. 10: Lead isotope compositions of copper alloy artefacts from the various phases of Poliochni on Lemnos, showing the sudden appearance of the exotic lead isotope compositions in Poliochni Yellow (After Muhly et al. 1991). The dashed field encloses the region wherein Aegean ores plot, whilst the solid ellipse encloses Troad ores. The hatched area is for objects from Poliochni with less than 1% tin, whilst black dots indicate objects with greater than 1% tin. Fig. 11: Lead isotope compositions of copper alloy artefacts from EC IIIA Kastri on the Cycladic island of Syros, compared with the lead isotope compositions of copper alloy artefacts from Troy IIg, the ‘Troas hoard’, Yortan and Thermi I & II, Lesbos.

216 Noel GALE

Metal Sources for Early Bronze Age Troy and the Aegean 217

218 Noel GALE

Metal Sources for Early Bronze Age Troy and the Aegean 219

220 Noel GALE

Metal Sources for Early Bronze Age Troy and the Aegean 221

222 Noel GALE