Bronze and the Bronze Age

Published by Oxbow Books, Park End Place, Oxford OXl IHN © Oxbow Books and the individual authors, 2000 ISBN 1 84217019 8 A CIP record for this book is available from The British Library. This book is available direct from Oxbow Books, Park End Place, Oxford OXl IHN (Phone: 01865-241249; Fax: 01865-794449) and The David Brown Book Company PO Box 511, Oakville, CT 06779, USA (Phone: 860-945-9329; Fax: 860-945-9468) or from our website www.oxbowbooks.com Cover: Three Bronze Age daggers from (left to right) Myrsinochorion (Aegean), Lyon (France) and Fossombrone (Italy). These daggers can befound in previous volumes of Priihistorische Bronzejunde (Vl/ 11, VII5 and Vl/lO). The world map is from Mountain High maps® copyright © 1993 Digital Wisdom, Inc. Printed in Great Britain at The Short Run Press Exeter 1. Bronze and the Bronze Age Christopher Pare The term 'Bronze Age' has been in use since the birth of modem archaeology, and one would expect the concept to be well understood. Strangely, this is not the case, and there is no consensus on how to use the term. This is surely because the Three Age System is thought not to be a profitable subject for modem research. But if the Three Age System is obselete, why is it so widely used? Is there, after all, something which makes the Bronze Age fundamentally different from other'Ages'? Andrew Sherratt (1993; 1994) is the most convincing contemporary exponent of the 'Bronze Age Hypothesis', and his work, together with studies by Kristian Kristiansen (e.g. 1987), provides the best introduction to the questions discussed below. To approach these questions, the first step must be a discussion of the definition of the Bronze Age, and in particular its start and finish . Peter Northover once used the following definition (1988:44): "Bronze Age is a loaded terminology with a conventional meaning that varies from region to region. Here it defines that period when coppers and copper alloys were predominant for all metal products save those of precious metals ." Northover should be commended for making his use of the term explicit; however, his definition is surely too broad for general use, and could include any period before the Iron Age using copper - smelted or unsmelted, native, 'pure' or intentionally alloyed. It is surely advisable, in archaeological usage, to reserve the term bronze for intentional alloys of copper with tin ; this would include ternary alloys such as Cu-Sn-As (arsenical bronze) or Cu-Sn-Pb (lead bronze) . With this terminology, the Bronze Age is easier to define: simply by the predominant use of bronze in the production of tools, weapons and other important artefacts. Indeed, this is the method generally used to define the transition to the Iron Age, for example in the well-known developmental stages described by A. Snodgrass (1980: 336 f.), based on 'working iron': "The criterion used ... is that of 'working iron', that is, iron used to make the functional parts of the real cutting and piercing implements that form the basis of early technology.... Using this criterion of working iron, we can discern three broad stages in the development of an iron technology; they are I think applicable to every area of Eurasia ... In stage 1, iron may be employed with some frequency, but it is not true working iron ... In the main, its employment is for ornament, as is appropriate for the expensive commodity which we know it to have been in many cases. ... In stage 2, working iron is present but is used less than bronze for implements of practical use. In stage 3, iron predominates over bronze as the working metal, although it need not, and usually does not, completely displace bronze even in this role. ... Simple proportion alone is used to distinguish between stages 2 and 3. It might be thought that such an abstract criterion could have had little economic or industrial significance for the period in question. Yet study of many ancient cultures shows a fairly abrupt change, at a certain point, from a predominant use of bronze to a predominant use of iron, within the strict field of working metal." The crucial feature of this process of technological development, which makes it widely - perhaps universally - useful as an indicator of cultural change, is that the transition to an iron-based technology (stage 3) is normally abrupt, as Snodgrass noted. This is quite different to the adoption of tin bronze which can either be abrupt or gradual, depending on the region involved. This difference in the take-up of bronze and iron can be explained, at least in part, by the availability of workable iron, copper and tin ores. Whereas iron ores are common in many parts of the ancient 2 CHRISTOPHER PARE world, bronze does not occur naturally. Tin deposits are rare - indeed absent in many parts of the world. Although much more common than tin, copper ores are unevenly distributed in Europe and the Near East. They are also quite varied, the type of ore affecting both the ease of metal extraction and the quality of the copper produced. Some ores, for example, contain copper with quite high levels of associated elements (e.g. arsenic or antimony) and, when smelted, these can produce so-called 'unin ten tional' alloys with properties which match low-tin bronzes (see for example Northover 1989). The contrast to the Bronze/Iron transition is clear. Iron is much easier to come by than copper and tin, and has technological qualities which differ markedly from bronze. Adopting a bronze technology, on the other hand, requ ires access to reliable supplies of copper and tin, which are liable to come from distant sources; and, in some cases, the properties of tin bronze did not represent a dramatic improvement on available arsenical or antimonal coppers. So it is no surprise that the Copper /Bronze transition does not have the 'universal' abrupt nature of Bronze/ Iron . We might, for example, predict that a region with easy access to tin, and only relatively pure copper, would adopt bronze with alacrity. If, on the other hand, tin is hard to come by, then the transition to bronze might proceed more slowly, especially if there is a plentiful supply of a good alternative raw material such as arsenical copper. Despite these adverse factors, bronze did come to be adopted as the dominant metal for a wide range of products (tools , weapons, metal vessels, ornaments) all over Europe. For me, this is the essence of the 'Bronze Age', and for that reason I recommend a simple definition of the term: the span of time in which bronze was the predominant material in metallurgical production. 'Predominant' could, for example, be defined as >75% of metal artefacts, and 'bronze' could be defined as any intentional copper alloy with >4% Sn, but the parameters used are not of crucial importance - in Europe, at least, much higher proportions of objects, with much higher concentrations of tin, became standard. However, it does seem advisable to differentiate between high and low tin alloys: in some cases very small amounts of tin (e.g. 0.5-1.0% Sn) could be added, probably to facilitate the processing of copper, for example to lower the melting point and to increase the fluidity for casting. For example a text of the mid 3rd millennium BC from Ebla records the production of a copper alloy with 0.79% Sn (Miiller-Karpe 1989: 183). As Cleuziou and Berthoud(1982: 15) explained, a use of tin for this kind of alloying is not very different from the use of As, Sb or Pb; high tin alloying (e.g. 6-14% Sn) produced a very different kind of metal. Even from these preliminary com- ments, it is obvious that the Copper/Bronze transition is not a simple matter, and specialists in archaeometallurgy have become quite circumspect in their interpretations. Despi te the complexity of the subject, a diffusionist view of the start of the Bronze Age remains deeply rooted, even in the specialist literature. This is encouraged by maps such as Fig. 1.1, published in 1976 by A. Gallay and M.-N. Lahouze, or Fig. 1.2, a diagram purporting to show the spread of tin bronze from south-east to north-west Europe between ca. 2500 BC and ca. 1600 BC, published by A. Sherratt in 1993. The diffusionist view is further encouraged by conventional chronological terminology: in southeast Europe the Early Bronze Age begins at the end of the 4th millennium BC, and in north-west Europe at the end of the 3rd millennium BC: Aegean: ca. 3100 BC (e.g. Manning 1995; Bulgaria: Carpathian Basin: C and NW Europe: ea. 3100 BC (e.g. Weninger 1992) ea. 2500 BC (e.g. Forenbaher 1993) ea. 2300/2200 BC (e.g. Needham Maran 1998) 1996; Rassmann 1996) This gives the impression of a cultural gradient down which influences can gradually diffuse from the Near East, to south-east, central and finally north-west Europe. However, people often forget that the traditional terminology for the Early Bronze Age is purely a matter of convention and largely arbitary definition. In Central and western Europe the Early Bronze Age is generally held to start after the Bell Beaker phenomenon. In south-east Europe, in the absence of Beakers, the Early Bronze Age is simply an extension of the west Anatolian and Aegean Early Bronze phases. Finally, in the Carpathian Basin, we find that cultures previously thought to be contemporary with Reinecke Br A in Central Europe are today, as a consequence of radiocarbon calibration, known to begin considerably earlier. A good example of this crucial change is illustrated by a chronological table published by 1. Bona (1992: 40 f.), in which EBA I (Vucedol C/Zok, Mako, early Nyfrseg) is dated to the 19th century BC, even though the same publication includes a summary of calibrated HC dates clearly indicating that these cultural groups reach back to the mid 3rd millennium BC (Raczky et al. 1992: 47, table 2; for the radiocarbon chronology of the Early Bronze Age in the Carpathian Basin see now O'Shea 1992; Forenbaher 1993). A similar development has taken place in Romania: 30 years ago, A. Vulpe assigned the 'Transition Period' to the three or four centuries before or around 2000 BC; then followed the Early Bronze Age ea. 2000-1700 BC and the Middle Bronze Age ea. 1700-1300 BC (1970: 6). A few years later, he raised the start of the BRONZE AND THE BRONZE ACE 3 -'. " CV \W ......ｾ .................. Fig. 1.1. The spread of tin bronze technology from the Near East to Europe, according to A. Gal/ay and M.-N. Lahouze 0976: 157,fig. 4: 'siade 5, maftrisedu bronze'). - The radiocarbon dates are uncalibrated. - The technology was first discovered in Mesopotamia (3000 bc,ca.37th century BC), then spread to Anatolia and the Aegean (2500 bc,ca.31st century BC), south-east Europe (2000 be, ca. 25th century BC) and central and western Europe (1700 be, ca. 2000 BC). - Gal/ay and Lahouze's dates have been calibrated using the OxCal v.2.01 programme. - Gal/ay and Lahouze 0976: 158) also note two areas with early evidence for bronze: the British Isles (towards 2100 be) and Macedonia (before 2000 be), which could represent autonomous centres of innovation. 'Transition Period' to ea. 2700 BC, but the start of the Early Bronze Age remained at ea. 2000 BC or a little before (1976). Today, according to calibrated 14C, the start of the 'Transition Period' is dated even earlier ( ea. 3500 BC), and the start of the Early Bronze Age (Glina III-Schneckenberg B)now seems to have taken place around the mid 3rd millennium BC. The conventional structures and terminologies for the Early Bronze Age were created before the scientific revolution in archaeology which led to the assembly of large quantities of chemical analyses, and multiple high quality 14C dates. But even today, in possession of this hugely improved empirical foundation, it is not necessarily easy to interpret the earliest stages in the adoption of copper and its alloys. The study of copper and early bronze metallurgy in Europe was revolutionised by the work in the 1950s and 1960s of the SAM (Studien zu den Anfangen der Metallurgie) team, based in Stuttgart (Iunghans et al. 1960; 1968; 1974). This massive project, involving the analysis of about 22,000 metal objects, is without doubt the single most important research contribution. But the interpretations and conclusions drawn by the SAM team, and other 4 CHRISTOPHER PARE __ｾｎｏｉｾｮ ｾ［ＺＭ ［Ｚ ｉ ｾｉ｜ｬｊｌｅ 1500 ﾷｉｾ Ｍ ｜ ＧＺﾷｅｓ ｾｲ SOUI'll-EAST ｉＭ Ｍ Ｍ Ｍ ｌｾ ｾ ｾｴＭ Ｍ ＭＬ IIRONI.E ACE (1'11111111115 CIIIIII,e) long d istance exchange chm-iot , EARLY em-ly hillf0l1/i OTOMANI IiIW NZE 2000 ACE z-pl cce moulds 2500 C O IUl E O WAR E JOOO use of wool horses PIT -GRA vrs BAOEN JSOO F1n ST ss, EUROPE AN ｾｉｅｃａｊＮｔｓ TRIl 4000 CO PPER AGE 4500 NEOLlTlIIC SOOO SSOO Fig. 1.2. Illustration published by A. Sherratt (1993: 16, fig. 4) showing the spread of tin-bronze from south-east to north-west Europe between ea. 2400 and 1600 BC. scholars in the following decades, have often been rendered obselete by the radiocarbon revolution and the arrival of dendrochronological dates, happening in parallel with the take-off in production of metal analyses . This means that the corpus of metal analyses has been subjected to a continuous process of reinterpretation in the last decades, as chronological sequences have been reshuffled and refined. Considerable care must be taken when reading earlier publications, in which it is often not immediately apparent if relative or absolute dates are based on traditional historical methodology (cross­dating), uncalibrated or calibrated 14c. Unfortunately, this problem also applies to the first systematic study of our subject, 'On the production of tin bronze in the early metallurgy of Europe', by K. Spindler (1971), based mainly on the 21,170 SAM analyses available at that time. The mass of available analytical data is still far from being fully digested and synthesised. The question of the reliability of analytical results requires a brief comment. Projects comparing ana­ lyses from different laboratories, using different analytical methods, have shown that reliability has improved over the past decades (see for example Northover & Rychner 1998). Nevertheless, even today it is difficult to interpret the conflicting analytical results which are sometimes published, for example the widely varying results of Optical Emission Spectography, Electron Microprobe Analysis, Neutron Activation Analysis and X­Ray Fluorescence on metal artefacts from Kastri, Syros (Muhly 1991: 362). Apart from variations in the results of different analytical procedures, we must also bear in mind the non­uniform distribution of elements, including tin, in copper alloy artefacts. An even more important source of inaccuracy is the analysis of strongly mineralised or oxidised metal samples, especially when this factor is not clearly described by analysts. I have mentioned these archaeometallurgical problems in order to make clear that isolated analyses of poorly preserved objects are generally difficult to interpret. Obviously, this is much more . BRONZE AND THE BRONZE AGE important for the earliest stages of metallurgical innovation. In the case of tin bronze, for example, the earliest artefacts will probably always be somewhat controversial; on the other hand, the question of the adoption of a fully bronze-using technology, when we often have hundreds or even thousands of analyses at our disposition, is much less susceptible to the problems of analytical inaccuracy. In the following pages, after a brief introduction to the development of early metallurgy'", two main subjects will be discussed: the earliest introduction of tin bronze alloys, and the transition to metal production based on the predominant use of tin bronze. The latter subject will be reviewed in more detail, in the light of our improved analytical and chronological data, to address the question of the nature of the European Bronze Age. THE COPPER AGE BACKGROUND Research on the earliest copper alloys (mainly with arsenic, antimony and/or tin) is at the same time one of the most crucial and one of the most difficult fields in Chalcolithic and Bronze Age studies. The past two decades have seen dramatic advances in our knowledge, and models have been put forward which have profound implications - particularly for the 3rd millennium BC. Artefacts made from native copper appear on archaeological sites from the late 8th millennium BC in south-east Turkey (e.g. Cayoni; Tepesi, Muhly 1989), and from the 7th millennium BC in Mesopotamia (e.g. Tell Maghzaliyeh, Ryndina & Yakhontova 1985). The mace-head from Can Hasan lIB in southern Anatolia demonstrates casting in the early 6th millennium BC (French 1962), but good evidence for the intentional smelting of copper ores (furnaces and slags), appears in the archaeological record much later, towards the end of the 5th millennium BC, at sites such as Norsuntepe, Degirmentepe, Tali-Iblis, Seh Gabi and Tepe Ghabristan. The increased occurrence in copper artefacts of arsenic and other impurities such as iron, likewise indicating copper ore smelting, is well documented in the Near East in the late 5th millennium BC (late Ubaid) at sites such as Mersin XVI-XVII, Norsuntepe, Susa I and Tepe Yahya V (Pemicka 1990: 45 ff.), but much earlier evidence from the 6th millennium BC at Yarim Tepe has also been mentioned (Merpert & Munchaev 1987: 17; Muller-Karpe 1989: 181; Gale et al. 1991: 50 f.). At the same time, there is a marked increase in the number and size of copper artefacts being produced, for example the 55 copper axes dating from the late 5th millennium BC from Susa (Talion 1987: 311 H.; Muhly 1988:8). True alloys (mainly Cu-As and more rarely Cu-Ag, Cu-Pb, Cu-Sb and Cu-As-Pb), in 5 which the added elements markedly change the properties of the copper, first appear in the Near East in the 4th millennium BC, for example at Nahal Mishmar in Palestine (Bar-Adon 1980) and Ilipmar IV in north-west Anatolia (Begemann et al. 1994). Arsenic is relatively common in copper ores and, according to most authors, the appearance of arsenical copper can be explained by preferentially obtaining copper from ores which have higher concentrations of arsenic. In the case of finds like Nahal Mishmar, with high levels of arsenic or antimony, specialist opinions differ, some authors believing that the alloys were produced by smelting copper ores (e.g. Pemicka 1990:48 ff.), others arguing that alloys with more than 4% As were made by eosmelting with arsenic-containing minerals (e.g. Tylecote 1991). In south-east Europe, artefacts made of 'pure' copper appear in the late 6th millennium BC, considerably later than in the Near East. However, after a preliminary horizon with copper ornaments and light implements, the following millennium saw the swift growth of copper production, most notably of heavy implements (Vinca-Plocnik lIB), which culminated in a veritable boom in the Late and Final Chalcolithic at the turn of the 5th/4th millennium BC (KodZadermen-Gumelnita-Karanovo VI), and the spread of the 'pure' copper heavy implement complex to the north and north-east, for example to the Tripolye, Tiszapolgar, Bodrogkeresztur and Balaton cultures (see for example Strahm 1994: 10 ff.; Pernicka 1990: 49 ff.; Pernicka et al. 1997). It seems reasonable to assume that the horizon of heavy copper implements corresponds with the start of extraction at mines such as Ai Bunar and Rudna Glava around the second quarter or middle of the 5th millennium BC (for a review of the evidence, see [ovanovic 1988);as in the Near East, the inception of smelting would go hand-in-hand with increased production of copper artefacts. However, it is often claimed that the vast majority of heavy implements is made of native copper (but note the difficulty of distinguishing native copper from pure smelted oxide or carbonate ores, see Maddin et al. 1980; Muller-Karpe 1989: 181; Gale et al. 1991: 54 f.), and recently it has even been argued that this earliest mining activity was aimed at malachite, for use as a semi-precious stone in jewellery, not at ores for metal production (Pemicka et al. 1993; but see the discussion in Gale et al. 1991: 53 ff.). It remains to be seen how this controversy will be resolved; it seems likely, however, that some of these early artefacts, at least, were made from smelted copper (ibid .: 51 f.). In south-east Europe the Final Chalcolithic and Proto Bronze Age (first half of the 4th millennium BC) saw a marked change in the organisation of metal production: in E. N. Chernykh's 6 CHRISTOPHER PARE terminology the replacement of the Carpatho­Balkan by the Circum­Pontic Metallurgical Province (Ch ernykh 1992; Pernicka et al. 1997: 54 H.). After the 'boom' in copper production in south-east Europe, some areas (e.g. the Varna and KodzaderrnenCumelnita-Karanovo VI groups) seem to have experienced a collapse of production (ibid.). The new metallurgical tradition, beginning in the early 4th millennium BC, was based on arsenical copper, perhaps earlier in south-east Europe, but quickly copied north of the Alps, for example in the Mondsee, Altheim and Pfyn cultures (Pernicka 1990: 51; Strahm 1994; Vajsov 1993). These changes in metallurgy have been incorporated into more general developmental schemes, for example by J. D. Muhly (1988: 9 f.): "The intensive mining activity ... resulted in the depletion of the oxide (and carbonate) copper ores of the Balkans by Late Eneolithic times, resulting in a great drop in metal production. With this metal shortage came a period of experimentation and innovation resulting in the first production of arsenical copper. The search for new sources of copper eventually led to the exploitation of the massive deposits of sulfide ores and a shift in the main centers of metallurgical development from the Danube Basin and the Carpathians to the Alps and the ore mountains of Czechoslovakia, both areas rich in copper sulfide ore deposits." Christian Strahm, too, sees an important distinction between the 'transitional' arsenical copper technology and the so-called 'A ufbauphase' (Foundation Phase) of the 3rd millennium BC, the latter based on the exploitation of complex sulphide ores, especially Fahlerze. According to Strahm, the technology for smelting complex copper ores spread from the Carpathian Basin not only to the Corded Ware and Bell Beaker cultures north of the Alps, but also to central Italy (Rinaldone), presumbaly reaching southern France (Fontbuxien) by the early 3rd millennium BC (Strahm 1994). It is significant that both Christian Strahm and Barbara Ottaway have recognised a 'hiatus' between the early arsenical copper production in the first half of the 4th millennium BC (Mondsee-Altheim-Pfyn north of the Alps, TRB C on the north European plain) and the more developed metallurgy (Strahm's 'A ufbauphase') of the Bell Beaker and Corded Ware cultures (Ottaway 1989; Strahm 1994). The 'Aufbauphase', with its mining and smelting of complex sulphide ores, is the context in which tin alloying was introduced. A very important general scheme for the historical development of metallurgy has been presented in a number of publications by E. N. Chernykh (most recently: 1992). In Chernykh's scheme, the Copper Age Carpatho-Balkan Metallurgical Province was replaced in the Early and Middle Bronze Age by the Circum-Pontic Metallurgical Province (ca . mid 4th to mid 2nd millennium BC). This was only eclipsed in the Late Bronze Age, by the emergence of regional metallurgical traditions: the European, the Caucasian and the Eurasian Metallurgical Provinces. Chernykh's work, involving the reconstruction of Metallurgical Provinces, Metallurgical Zones, and Metallurgical and Metalworking Focal Areas, represents a crucial advance in our understanding of the subject. However, within the broadly convincing picture of metallurgical development, one aspect surely requires revision. A European Metallurgical Province is certainly already apparent by the early 2nd millennium BC, at the time of the widespread adoption of tin bronze, and probably even in the 3rd millennium BC, at the time of Strahm's 'Aufbauphase'. We will return to this question later in the article. THE EMERGENCE OF TIN Pernicka (1998) has recently summarised his thoughts on the introduction of tin bronze, basing his conclusions on an impressive series of detailed studies in south-east Europe, the Aegean and the Near East. According to Pernicka (ibid .: 137 f.) metallic tin was discovered at the start of the Bronze Age. Tin was probably first smelted from tin-stone, an oxide ore (Sn0 2) , perhaps discovered as a by-product of panning for alluvial gold. In contrast to other early alloys, such as arsenical or antimonal copper, from the start Cu-Sn alloys were produced by melting together metallic copper and tin; this is thought to be much more likely than the smelting of copper/tin ores or the addition of tin ores (e.g. tin-stone) to molten copper (Pernicka 1998; but see Charles 1980: 174 f.; Gale et al. 1985: 155). Following the early appearance of copper-tin alloys at Mundigak, Afghanistan, in the second half of the 4th millennium BC (Stech & Pigott 1986: 47; see also Cleuziou & Berthoud 1982), tin bronze first appeared in the Near East at around 3000 BC or the start of the 3rd millennium BC in Anatolia and northern Mesopotamia (e.g. Tell al-Judaidah, Braidwood & Braidwood 1960: 300 H.; Tepe Gawra layer VIII, Waetzoldt 1981: 374; Muhly 1985: 281; Moorey 1994: 297 H.). A few bronze objects are known from the early 3rd millennium BC (e.g. Kish, Miiller-Karpe 1989: 184, fig. 5), but regular use starts in the middle of the millennium, as shown most clearly by the 'Royal' graves of Ur (Early Dynastic IIIa, ca. 26th century BC) and the hoards of Troy IIg. There is a scatter of contemporary mid 3rd millennium finds of tin bronze reaching from the Aegean in the west to Susa in the east'", suggesting that this technology was a common cultural phenomenon, involving BRONZE AND THE BRONZE ACE intensive contacts and exchange between the individual regions (Pernicka 1998: 138 H.). Pernicka summarises his conclusions as follows (1998: 140 f.): "Die Ausbreitung erfolgte nicht zufallig - bald hier, bald da -, sondern nach einem klaren Muster mit einer relativ grofsen Ursprungsregion. Zumindest im Westen der Alten Welt hatten die sich entwickelnden Regionen Bertihrung mit anderen, in denen Zinnbronze schon langer bekannt war. Es ist deshalb sinnvoll, die Ausbreitung der Zinnbronzetechnologie als einheitlichen Prozef zu betrachten, der die Umwandlung der menschlichen Gesellschaft von einem einfachen zu einem hoheren Organisationsgrad begleitet." Pernicka emphasises that this view is opposed to the model developed by C. Renfrew, which posited the autonomous invention of tin bronze in the north-east Aegean as one of the primary factors causing profound social change. Renfrew's view, according to Pernicka, is contradicted by the results of Lead Isotope analysis, which shows that the great majority of copper and bronze objects from sites like Troy, Poliochni and Kastri could not have been made from local ores. Therefore the metallurgical boom in the north-east Aegean was caused by 'stimulation' from the Near East (Muhly & Pernicka 1992: 312 ff.): importation to the Troad of copper alloyed with tin - probably as finished artefacts - from the Near East (Pernicka 1987: 703). He concludes as follows (1987: 705): "Wichtigstes Ergebnis der Artefaktenanalysen ist der Nachweis, daf die EinfUhrung der Zinnbronze im trojanischen Kulturkreis nicht auf eine lokale Entwicklung zurtickgefiihrt werden kann, sondern daf zumindest das zu deren Herstellung notwendige Zinn uber sehr weite Entfernungen, moglicherweise aus Zentralasien herantransportiert werden mufite." As for the reasons behind the introduction of bronze in the Near East, Pernicka (1998: 135 f.) notes that arsenical copper can match the properties of tin bronze. However it has crucial disadvantages, mainly the difficulty in controlling the amount of arsenic in an alloy: it was impossible to measure precisely the arsenic content of an ore, and the volatility of arsenic makes it difficult to produce objects with more than 5% As. Indeed, 97.1% of the objects analysed by the SAM project have less than 3% As, so arsenical copper rarely reached the hardness of a typical 10% tin bronze. He also draws attention to the adoption of tin bronze mainly in 'wealthy' cultural contexts (in Anatolia for example at settlements like Troy IIg and Poliochni 'giallo', and the 'princely graves' from Horoztepe, Alaca Huyuk, Ahlathbel, Kayapmar and Mahmatlar), often in the form of prestige objects made using advanced casting techniques (for tin bronze in Anatolia, see Yener et al. 1996). So the introduction 7 of tin bronze was not just a diffuse transfer of raw material and knowledge, it was the result of trade over long distances (idem 1990: 53; see also Stech & Pigott 1986: 52 H.). An international trade in tin (or tin bronze), controlled by large city-states, began by the mid 3rd millennium BC. Before this horizon there are only a few isolated finds of tin bronze objects in south-east Europe, such as the knife from Velika Gruda (Primas 1996: 94, fig . 7.1, M2 with 7.6% Sn). Objects like this are often interpreted as evidence for an experimental phase in the history of alloying technology. However, Pernicka believes that experimentation is made unlikely by the rarity of tin ores, and their infrequent association with copper ores, suggesting that isolated finds like Velika Gruda can probably be interpreted as deriving from the international trade in the Near East (1990: 53). He adds that the spread of tin bronze into south-east Europe is impossible to follow at present, owing to the imprecise chronology of the region, but he entertains the possibility that bronze was introduced in south-east Europe at roughly the same time as in the Aegean. Finally, he notes that tin bronze spread to the rest of Europe about 500 years after its adoption in the Near East and the Aegean; the tin bronze alloying technology not only spread to the west, but also to the east, to the Indus Valley, via the Iranian highlands and Central Asia (1998: 138 H.). J. D. Muhly and E. Pernicka agree w ith H. McKerrel (1978: 19) that " ... there can be no question of any major Near Eastern source [of tin] which was exploited in the Bronze Age and yet remains still to be discovered", and they note that the "sources of tin remain the great enigma of Bronze Age archaeology" (Muhly & Pernicka 1992: 315). In the Aegean (including Crete), the East Mediterranean (including Cyprus) and Western Asia (including the Caucasus) there are no workable sources of tin ore (Muhly 1985; Muhly & Pernicka 1992: 314 f.; Pernicka 1998: 137; 142 f.). Among the various claims for Old World tin sources, Pernicka (e.g. 1998: 142 f.) argues strongly against Sogukpmar (north-west Anatolia), Sulucadere and Kestel (both in the Taurus mountains); the case of Kestel is most controversial (see, for example Hall & Steadman 1991; Pernicka et al. 1992; Muhly 1993; Yener & Goodway 1992; Yener & Vandiver 1993a .b; Willies 1992; 1993). The situation in Europe, where both tin-stone (Sn0 2) and stannite (CuleSnS4) occur in some quantity, is quite different. The most prolific tin sources in Europe are in Cornwall and the Ore Mountains (Erzgebirge, on the border between Saxony and Bohemia); important deposits are also known from Brittany and the Massif Central in France, and the north-west Iberian peninsula (Pernicka 1998: 137; 142 f.). Less well documented sources in Tuscany (Monte Valerio) and southern Sardinia 10 CHRISTOPHER P ARE Early Helladic (n = 139) 30 125 124 123 25 20 ｾ '" ｾ "c: 15 .....0" 0 z la 5 4 3 5 2 0 2 3 4 5 6 7 . 8 9 10 1I 12 13 14 14+ I 0.5 0 2 4 3 5 6 7 3 4 4+ 8 8+ As (%) Sn (%) Middle Helladic (n = 34) 15 "j "., "- "., 10 "- coc: .....'" coc: ....." 0 0 0 0 z la >. >. z 5 o 2 3 4 5 6 7 8 9 10 JI 12 Sn (%) 13 14 15 16 17 5 o 0.5 1 2 As (%) Fig. 1.4. Histograms of the tin and arsenic contents of copper alloy objects in Early Helladic and Middle Helladic mainland Greece. For references to the metal analyses included in the histograms, see Table 1.1. objects have over 5% Sn, two have 3­5% Sn and no less than seven contain below 0.5% Sn (Varoufakis 1973) . Crete also seems to have used only small amounts of tin during the Early and Middle Bronze Age. Apart from the 28 EM and MM analyses published recently by Mangou & Ioannou (1998), at least 90 further analyses have been published from Archanes, Charnaizi, Fortetsa, Hagia Triadha, Katsambas, Kalathiana, Koumasa, Krasi, Lebena, Marathokephalon, Mochlos, Myrtos, Phaistos, Platanos, Porti, Pyrgos, Salarne, Tekes and Traostalos (Slater 1972 [1 object] ; Branigan 1974: 150 H. [82 objects]; Varoufakis 1995 [7 statuettes]). Only 7% of these analyses indicate BRONZE AND THE BRONZE ACE Early Helladic Macedonia 1 Mandalo 23 Petralona hoard PetraIona district 4 1 Seratse 2 Servia McGeehan­Liritzis 1996 Mangou & Ioannou 1999 Ibid. Heurtley 1930: 144; 1939: 253 f. [ones 1979 Thessaly Petromagoula SeskIo 9 1 McGeehan­Liritzis & Gale 1988 Ibid.; Maran 1998: 264, note 1069 Phocis Ay. Marina 2 Dickinson 1977: 114 Euboea Tharounia Cave 'Euboea' Manika 5 1 23 Mangou & Ioannou 1999 Phelps et al. 1979 Sampson 1985: 306; Stos­Cale et al. 1996 Boeotia Eutresis Lithares 5 10 Goldman 1931: 285 Kayafa et al., this volume Attica Aghios Kosmas Rouf Mylonas 1959: 78 Petrikaki 1980: 173 Peloponnese Corinth Lema Ill­IV Tsoungiza Voidokoilia 'Peloponnese' 1 25 7 5 1 Caley 1949: 60 H. Kayafa et al., this volume Ibid. Kayafa 1999: table 3 Phelps et al. 1979 Ionian Sea Levkas 11 McGeehan­Liritzis 1996: 365 Middle Helladic Attica Eleusis Peloponnese Argos Ayios Stephanos Lema V Malthi Nichoria Voidokoilia Mylonas 1932: 146 f.; Dickinson 1977: 114 1 5 10 2 12 3 VollgraH 1906: 40; Dickinson 1977: 114 Kayafa 1999: table 8; R. E. [ones (pers. comm.) Kayafa et al., this volume Mangou & Ioannou 1999 Kayafa 1999: tables 33­34; Stos­Gale et al., in press Mangou & Ioannou 1999; Kayafa 1999: table 3 Table 1.1. Metal analyses of Early and Middle Helladic copper-based objects. The numbers refer to the number of samples analysed. more than 5% Sn, compared to 86% with less than 2% Sn. However, J. D. Muhly (1991) has mentioned nine further tin bronze artefacts analysed by the SAM project'", including two daggers from Krasi which might date to EM I. Even if some or all of them can be assigned with confidence to the EarlyMiddle Minoan period, they will not significantly alter our general conclusion, based on ea. 120 published analyses, that tin bronze was only used rarely in Crete before the start of Late Minoan. A change in 11 alloying practice clearly took place during the 16th and 15th centuries BC. In the Unexplored Mansion at Knossos (LM 11) 60% of the analysed objects contain more than 5% tin(6); and in Sellopoulo, tomb 4 (LM II­I1IA), all the copper alloys contained over 5% tin (Catling & [ones 1976). In both Crete and mainland Greece, tin bronze was the dominant metal used from the mid 15th century BC onwards (LM I1IA/ LH IlIA). Whereas the use of tin only seems to have started to increase in Crete in the Late Minoan period, from around the 17th century BC onwards, on the mainland bronze already seems to have played a significant role from the start of the Middle Helladic period. In the course of our discussion of Aegean metals, we have come across two different models of supply. On the one hand, there was a limited, and perhaps short­lived, influx of so­called 'Fremdmetalle', probably entailing the exchange of bronze (copper alloyed with tin) over long distances, presumably organised as a form of sea­borne or caravan trade. On the other hand, the regular and predominant production of tin bronze, at least by LH/LM IlIA, indicates the existence of a reliable supply of tin, alloyed with local sources of copper. The earliest indication of the tin trade is the famous tin bangle from Thermi IV (Begemann et al. 1992: 224 ff.), and according to the Lead Isotope data imported tin was probably alloyed with local sources of copper at Manika in EH III (Stos­Gale et al. 1996: 56, table 3, "Cycladic copper") and at Lema V in Middle Helladic (Kayafa et al., this volume, "Rhodopi, Lavrion?"). It is interesting to compare the situation in Cyprus, illustrated by the following quotations: In Middle Cypriot 11 (ca. 1800­1725 BC), "practically all copper and arsenical copper objects are made from Cypriot copper. Only a few MC 11 tin bronzes occur, but those appear to be made of non­Cypriot copper" (Stos­Gale et al. 1991: 344) ... "The transition from Middle to Late Cypriot times is marked clearly by the increase and dramatic improvement of Cypriot metallurgy. There is a move from the import of small amounts of foreign bronze in Middle Cypriot times and the first, halting, steps in local manufacture to the full blown Late Cypriot manufacture of tin bronze in Cyprus, using foreign tin but Cypriot copper" (Gale & Stos­Gale 1989: 254). It seems, then, that a reliable tin supply was first established in Cyprus around the start of the Late Bronze Age (ca. 1600 BC); interestingly, deliberate alloying with tin also becomes 'universal' around this time in Palestine (Northover 1988: 50; see also Philip 1991; Rosenfeld et al. 1997). The long­distance tin trade seems to have been able to supply Mesopotamia and central and western Anatolia in the 3rd millennium BC (Frangipane 1985: 221, fig. 3; 226 'period 3'). In the first half of the 2nd 12 CHRISTOPHER PARE millennium BC, the tin supply was still very uneven in the Near East and East Mediterranean (ibid.: 222, fig. 4), but an important change does seem to have occurred around the start of the Late Bronze Age in Cyprus and the Levant, when much larger quantities of tin must have been obtained on a regular basis, possibly indicating the start of trade with new trading partners or new sources of metallic tin . Romania, Bulgaria and Yugoslavia A convenient link between the metallurgy of the Aegean, the Balkans and the Carpathian Basin is provided by the shaft­hole axes from Petralona, Poliochni ' rosso: and Thebes (Maran 1989: 131, fig. 1,2.6.7). These come from contexts of Early Helladic Il/III or Ill, and belong to a large family of similar axes, studied in detail by A. Vulpe (1970); Pol iochni and Petralona may be linked to Vulpe's Izvoarele series and the Veselinovo Il type, the Thebes axe to Vulpe's Patulele type. As J. Maran (1989) has shown, the axes are important for linking chronological systems between the Carpathian Basin and the Aegean, and suggest the rough contemporaneity of Early Helladic Ill, the first part of the Romanian Middle Bronze Age (Monteoru IC 3-IC 2) and Reinecke Br AI. The shaft-hole axes are important for two further reasons: they represent a relatively large proportion of metal objects known from the Early and Middle Bronze Age in south-east Europe, and their metallurgical compositions have been intensively analysed - especially the Romanian (SAM project) and Bulgarian (E. N. Chernykh) examples. According to A. Vulpe, the early shaft-hole axes cast in open bivalve moulds (e. g. the Baniabic, Fajsz, Corbasca, Durnbravioara and Veselinovo I types) are all made of either pure or arsenical copper. The only exception is the Dumbravioara axe from the Early Bronze Age (Schneckenberg phase) settlement of Sfintu-Cheorghe, with 1."45% tin. Pure or arsenical copper is also predominant in the first part of the Middle Bronze Age/Monteoru IC3-IC2 (Veselinovo Il, Izvoarele, Patulele, Monteoru I and Padureni I types); the only exceptions are the Padureni I axe from Halchiu, and two Patulele axes with up to 0.4% tin. It is only in Monteoru IC 2-IA/Br A2 (Padureni Il, Monteoru Il, Hajdusamson, Balsa. Apa-Nehoiu types) that use of tin bronze becomes more general, but even now there are hoards like Sinaia (26 axes with 1.15-3.5% Sn) and Borlesti (five axes with 0.04%, 0.63%, 5.1%, 5.8% and 7.8% Sn), which indicate that alloying was by no means standardised. However, in the second part of the Romanian Middle Bronze Age (Monteoru Il/Br BC) almost all the axes are alloyed with tin . For the Romanian axes, it is clear that the use of tin increased markedly during the Middle Bronze Age, and became predominant in the Apa-Hajdusamson horizon (Monteoru lA/Br A2b). Indeed, Vulpe notes that the earlier Middle Bronze Age bronze axes only contained between 0.9% and 4% Sn, with the later examples reaching up to 7% Sn. South of the arc of the Carpathians, the adoption of tin bronze may have happened slightly later: this is suggested by the cemeteries of Sarata Monteoru, where graves of phase lA (Apa-Hajdusamson horizon) contain bronze objects with 2.7-5.7% Sn, whereas those of phase IlA (Br B) have 5.8- ea. 10% Sn (Vulpe 1976: 155). Tin bronze was clearly extremely rare in Romania before the Middle Bronze Age. In the hoard of 10 neckrings from Deva, dated to the transition from the Early to the Middle Bronze Age, only two contain tin (0.31/0.34% Sn, 0.26/0.67% As); the other eight contain 1.3-1.7% As. According to Vulpe, apart from the Early Bronze Age axe from Sfintu-Cheorghe. mentioned above, there is only one earlier find: the ochre-grave burial from Clavanesti (tumulus 1, grave 11) with two bronze buttons (3.4% Sn) and a spiral ring (1.55% Sn), which presumably dates before the start of the Early Bronze Age . In the case of Bulgaria, we are able to base our discussion on the important research results published by E. N . Chernykh (1978). In general, tin bronze seems to have been quite rare in the earlier parts of the Bulgarian Bronze Age: in the Early and Middle Bronze Age only 10% of 144 analysed objects were of tin bronze, compared to 57% of arsenical copper (Greeves 1982). This contrasts with the Late Bronze Age, when tin bronze was practically the only alloy used, although 29 out of the total 549 analysed objects were found to be of 'p ure' copper (ibid.). However, even in the Late Bronze Age, Pernicka et al. (1997: 138) note that the wide range of tin contents, between 1.1% and 12.3%, indicates there was no strict control over the alloy composition. For the Early Bronze Age, the most important site is Ezero, with up to 4 m of stratified deposits (for 14C dating evidence, see Weninger 1992: 420 H.). In Ezero A (layers 13-9, ca. 3100-3000 BC) there are 14 rather simple metal objects, four made from 'pure' copper, the rest from copper with low concentrations of arsenic, in Ezero B (layers 6-1 , ea. 2900-2500 BC) the 19 metal objects are still made of either 'pure' (five) or arsenical copper, but now with rather higher concentrations of arsenic. The only tin bronze artefact is an unstratified pin (4.5% Sn) from the surface of the settlement (Chernykh 1978: pl. 28,43). The most important site from the end of the Early Bronze Age (EBA 3), post-dating Ezero, is Novozagora, where the analysed metal objects were again of arsenical copper. Ninety-nine analyses are available from the Early Bronze Age lake -side settlement of Ezerovo Il (Chemykh 1978:analyses 11883-11982), five of which 13 BRONZE AND THE BRONZE AGE have :2:2% Sn, including one object with 4% Sn and another with 6% Sn; but the reliability of the results is somewhat questionable, as the metal is reported to be affected by saline contamination (Greeves 1982: 540; Pernicka et al. 1997: 126). For the Middle Bronze Age, the Emenska Pest cave provides the largest collection of metal objects: the 12 analyses all show the use of Cu­Sn­As; the average tin content is 5.3% Sn, with values ranging widely between 0.8% and 15% Sn (Chernykh 1978: analyses 10912­10925). Apart from the two objects from Ezerovo with 4% and 6% Sn, the unstratified pin from Ezero, a flat axe with 8% Sn from Cradesnitsa (Chernykh 1978: pl. 27, 13), and the shaft­hole axes discussed below, the Emenska Pest cave is the only Bulgarian site of the Early and Middle Bronze Age with tin bronzes (ibid.: pls 27, 3.5; 28, 3.5.8.12.13.39.40; 29, 1.22). All the flat axes, knives, awls, chisels etc. from other sites analysed by Chernykh were made of copper or arsenical copper. This suggests that tin bronzes were still relatively rare in Bulgaria in the Middle Bronze Age; metallurgy changed markedly in the Late Bronze Age, when large numbers of tin bronzes are known. Shaft-hole axes can again be taken as an example for alloying practices. Ezero B shows that simple open bivalve moulds (Chernykh's type 1) were used in the first half of the 3rd millennium BC for producing axes of Veselinovo I type. Closed bivalve moulds (Chernykh's type 4) came into use in the Middle Bronze Age, producing tools like the Padureni axe from Emenska Pest, similar to the axe from Poliochni 'rosso' mentioned above (Chernykh 1978: pl. 25, 5). A summary of Chernykh's results is illustrated on Fig. 1.5. It is immediately clear that the axes made from open bivalve moulds (types T.2, T.4, T.6, T.8) are made from 'pure' or arsenical copper (all <0.15% Sn). Among the axes made in closed bivalve moulds, 'pure' and arsenical coppers remain in use (T.12, T.16, T.20: all <0.5% Sn), but arsenical copper with tin and tin bronze appear for the first time. Tin bronze first seems to be used in the early Middle Bronze Age, for example for axes of types T.10 and T.14, related to the Veselinovo Il type axes in finds like Ostrovul Corbului hoard Il. Nevertheless, it is clear that tin bronze was by no means the predominant alloy used for shaft-hole axes, with the exception of type T.22 (Apa-Nehoiu) of the Apa-Hajdusarnson horizon?'. Chernykh (1977) has also provided a wider view of alloying in shaft-hole axes, in a series of four maps (Fig. 1.6; see also Fig. 1.7). Fig. 1.6, a-b shows the alloying practices at the time of the Early Bronze Age axes produced in open bivalve moulds (Chernykh's mould types 1 and 2; axes of types Baniabic, Veselinovo I, Fajsz, Corbasca etc.). Arsenical copper is predominant in Bulgaria and the Caucasus; Cu As SnJAs Sn As ｾ 0.4% As ｾ 0.5% As ｾ 0.5% As ｾｏＮＴＥ Sn:O; 0.4% Sn ｾ 0.4% Sn ｾ 0.5% Sn ｾ 0.4% G-: U7, (1: 0: G:o IT: IT· 1 4 0 0 1 5 0 0 I I 0 0 I I 0 0 0 I 2 1 I? 3 0 0 I I I 5 I 7 2 0 0 0 8 I? EBA 2 EBA 3 MBA I TI4 Ｎｾ ｾＢ TI8 MBA2 [-rro r 0 4 0 0 0 0 0 2+1 ? 1'22 Fig. 1.5. Summary of metal analyses of shaft-hole axes from Bulgaria. - T2-T22 refer to the types as defined by E. N. Chernykh . - Cu = 'pure' copper, As = arsenical copper, Sn/As =copper with both arsenic and tin contents greater than 0.5%, Sn = tin bronze. - After Chernykh 1978: 146 i, table [[[.6. further north 'pure' copper seems to have been the main material used. Fig. 1.6, d shows that arsenical copper was still widely used in Bulgaria and the Caucasus around the time of the Apa-Hajdusamson C=J I D • ｉｾ :J 3 1 10 CJ ｾ｣］Ｚｊ 12 - ｾ D 20 ｯ 16 c;:J ｏ ｾ ｾ c::::=:=J ｉ｣ 24 0 ｬｾ ｾ ＶＮ I i 8J c:;J c;::J ｾ 27 28 :r: ' '' ' - '-" :N 38 ｾ ｾ I ｾＴ g ｾ ｾｯ o ｾ Ｒｾ ｾ 'l ｾ ｾ ｾ ｾ ｾ d ""0 > :N m 48 I c=::::J ｏ 51 "0 :r: m :N 46 <::;:j 25 ｾ n Ｑｾ 0 ｄ ｾ ｏ 18 23 ｾ - 'v 14 8 ｾ 13 r - -- c:::::J 0 7 ｾ D - ｾ 6 • • ｾ 11 5 • c::J 6. 6. G=J ...... b c::=t 2 1 ( c;;:J la ｾ I ｾｃｩｌ｝ l e' ｉ ｉｾ ＶＮｃＷｊｾ Fig. 1.6. Summary of metal analyses of shaf t-hole axes from the Caucasus, the north Pontic steppes, the Volga-Ural region and the Carpatho-Balkan region. ­ Empty symbols: not analysed. ­ Vertical line: 'pure' copper. ­ Cross: arsenical copper. ­ Black symbols: tin bronze. ­ After Chernykh 1977. 15 BRONZE AND THE BRONZE ACE N Pontic Caucasus 100 90 Volga­Ural 0=0 0= 35 Balkans Carpathians 0=27 0=3 0=8 80 70 Axe types 1­8 60 50 (%) 40 30 20 10 0­l..L_...LlLL..L.L­_­­l100 90 ­­I.l....­­­I. ­­­­JL..l....­­­­­J..l...L...:.....:..l_ _ 0=0 0= 8 0=17 Ｎ ｵＮ Ｎ ｟Ｎ ＮｌＮｌＮ ＮＺｾＺｬＮ Ｎ｟ Ｎ Ｎｊ 0=12 0=7 80 70 Axe types 9­18 60 50 (%) 40 30 20 10 0...L.­_­­­'­L.L....LL_ _'­- ­­­­'­l..­­­­l ...J..l._­­u..:.....:;.­"'­'­_ _­L.L..._....L.J..."­"­"­'­_­­­­J 100 90 0=78 80 70 Axe types 19­37 60 50 (%) 40 30 20 10 0...L.­_­­­'­L.L....LL_­JL..l...­_.J..J....<:....<....t."'­­_­­­'­l..­­­­l..L....:..""­"J... 100 90 0=76 0= 3 80 70 Axe types 38­62 60 50 (%) 40 30 20 10 ｏＭＫｌＮＢＧｙＬｦｲｬｾ Cu CuAs CuSn Cu CuAs CuSn Cu CuAs CuSn Cu CuAs CuSn Cu CuAs CuSn Fig. 1.7. Histograms of metal analyses of shaft-hole axes from the Caucasus, the north Pontic steppes, the VolgaUral region and the Carpaiho-Balkan region (cf. Fig. 1.6). ­ Cu = 'pure' copper, CuAs = arsenical copper, CuSn = tin bronze. ­ After Chernykh 1977: 36, table 2; 49, table 4. horizon; however, tin bronze predominates in the Carpathian Basin, and makes an appearance not only in Bulgaria, but also in the Caucasus and to the north, between the Volga and the Urals. Fig. 1.6, c is rather less easy to in te rp ret, as La te Chalcolithic, Early Bronze Age and early Middle Bronze Age axe types have all been included on the map . For example, the north­west Balkan copper 16 CHRISTOPHER PARE axes of Kozarac type are shown, even though they can be linked to the Vucedol and Ljubljana cultures'". Nevertheless, Fig. 1.6, c repeats the general distinction between areas with arsenical copper axes (eastern Balkans, Caucasus, now also the steppes) and areas with 'pure' copper axes (western Balkans, western Carpathian Basin, steppes and Volga) . The seven tin bronze analyses in the Carpatho-Balkan region indicate the adoption of this alloy for producing some axes in the early part of the Middle Bronze Age. According to Chernykh's results for the axes shown on Fig. 1.6, d, 80% of the axes in the Carpathian Basin were made of tin bronze, compared to 35% in the Balkans, 26% in the Volga-Ural region and only 3% in the Caucasus (Fig. 1.7 - axe types 3862). This scarcity of tin in the Early and Middle Bronze Age Balkans is also indicated by Chernykh's summary of alloying practices in Bulgaria, shown on Fig. 1.8, indicating that tin bronze was less common than arsenical copper and arsenical copper with tin in the Middle Bronze Age, a situation which was reversed in the Late Bronze Age. Apart from the shaft-hole axe evidence reviewed above, there are a few other more or less reliable finds of tin bronze from the Vucedol and Baden cultures (Vinkovci, Velika Gruda, Brekinjska, Okukalj), the Proto Bronze Age (Kacica, Velika Humska Cuka) and even the Late Chalcolithic (Smjadovo, Zaminec) (Pernicka 1990: 52 f.; Pernicka et al. 1993; 1997; Primas 1996: 104 f.; Durman 1997: 11 f.). A marked increase in the use of this alloy, however, is first evident around the last quarter of the 3rd millennium BC, at the start of the Romanian and Bulgarian Middle Bronze Age, and the Cetina culture in the western Balkans. Another major development takes place around the time of the Apa-Hajdusarnson horizon (ca. 17th-16th centuries BC), with regular use of tin bronze in the area between the Tisza and Prut (Fig. 1.6, d). The Carpathian Basin David Liversage (1994) has contributed a very useful review of early alloying practices in the Carpathian Basin, based on ca. 2,500 SAM analyses. His results were summarised in a series of histograms, showing changing tin content from the start of the Early Bronze Age to the Late Bronze Age '" (Fig. 1.9). The first Early Bronze Age horizon is marked by cemeteries of the Nitra group in southwest Slovakia, corresponding roughly to Br Ala (Fig. 1.9, a). It is clear that bronze was hardly used at all; 98% of the analyses contained less than 1% tin . The next histogram (Fig. 1.9, b) is mainly based on finds from Br A1b, from later cemeteries of the Nitra group and Gemeinlebarn phases 1-2. Again Copper type VIII Pb ;>: 0.03% Bi ;>: 0.002% Copper type VI Pb < 0.03% Cu Cu As MBA 2 MBA I EBA3 EBA2 EBA I Fig. 1.8. Summary of E. N. Chernykh 's results showing the change of copper types and alloys in the Bulgarian Early and Middle Bronze Age. - Cu = 'pure' copper, As = arsenical copper, Sn/As = copper with both arsenic and tin contents greater than 0.5%, Sn = tin bronze. After Chernykh 1978: 168, fig. 86. the great majority of samples (89%) has less than 1% tin; however, there is now a scatter of analyses reaching up to a small peak at 10% Sn. Fig. 1.9, c shows the use of tin during the 'classic' Unetice phase, or Br A2a, with analyses again coming from south-west Slovakia and from Gemeinlebarn (phase 3). The tin distribution is now clearly bimodal, with almost 30% 'unalloyed' and the rest climbing to a clear peak around 10% Sn. The following pair of histograms derives from hoards of Br A2b in Transdanubia (Tolnanemedi series, Fig. 1.9, d) and northeast Hungary and Transylvania (Hajdusamson series, Fig. 1.9, e). Both show a small minority of unalloyed objects, and the mass containing 4-10% Sn . The last histogram illustrates tin use in the Middle Bronze Age, or Br B-C (Fig. 1.9, f), with an almost perfectly normal unimodal tin distribution around a peak at 6-7% Sn. In view of the variety of data utilised by Liversage, including cemeteries and hoards from a wide area, it is worth looking at one well-studied site in more detail: the chronology of the cemetery of Gemeinlebarn has been worked out by F. Bertemes (1989) and the analytical data summarised by Liversage (1994: 80 f.; 81, table xvii) , The Gemeinlebarn phases can roughly be paralleled with the Reinecke /Ruckdeschel system, as follows 1 (Br Ala), 2 (Br A1b), 3 (Br A2a), 4 (Br A2b). Phase <1% Sn ｾＱ Ｅ 1 2 3 5 (100%) 44 (83%) 11 (23%) 0(0%) 0(0%) 9 (17%) 8 (15%) 36 (77%) 35 (75%) 0.101% 1.370% 7.257% 4 0(0 %) 6 (100%) 5 (83%) 8.483% Sn ｾＴ Ｅ Sn Average Sn Analyses n=5 n = 53 n = 47 n=6 The Gemeinlebarn cemetery follows the trends illustrated by Liversage's more general histograms. An increase in tin, however, happens rather earlier 17 BRONZE AND THE BRONZE ACE Ｚｾ Ｑ Ｑ a ｾ 40i 20 n = 197 d ｾ s-, • I! JU !I 5- I! 35 . J: i': l . . """"" , o 60 b n = 157 <\ \ 1 2 3 4 5 6 7 8 9 0 <I I 2 3 4 5 6 7 8 9 10 1I 12 13 14 8 9 10 I1 12 13 14 8 9 10 I I 12 13 14 Sn (% ) 10 II 12 13 14 Sn( %) 25 e n = 126 n = 37 20 ｾ [ [ )5 [ u, 5 0 ｾＺ ｾ Ｎ ｾ 15 15 10 u, 30 ＺＮ o <I I ｩＺｾ ｩＡｩ､ 2 3 4 5 6 7 8 9 Ｂ Ｂ 0 0 <I I 2 J 4 5 6 7 Sn (%) 10 I 1 12 13 14 Sn(% ) c 25 25 20 20 n = 332 z ｾ ! ｾ "'" 15 u, 10 [ 10 o c-, u 15 o <l I 2 3 4 5 6 7 8 9 10 11 12 13 14 Sn( %) o <\ I 2 3 4 5 6 7 Sn(%) Fig. 1.9. Histograms showing the tin content of copper and copper alloy objects in the Carpathian Basin. - a) earlier part of Br Al. - b) later part of Br Al. - c) earlier part of Br A2. - d) hoards of the Tolnanemedi group. - e) hoards of the Hajdusamson group. - j) hoards of the Middle Bronze Age. - After Liversage 1994: 76 [; figs 2-4; 6-8 . than in south-west Slovakia; in Gemeinlebarn 3, for example, 75% of the samples contain over 4% Sn, compared to 55% in the cemeteries of Hurbanovo, Nesvady and Matuskovo, which are supposed to be roughly conternporary''?'. A marked rise in tin is already apparent in Gemeinlebarn 2 (Br A1b), with eight of the 44 samples containing more than 4% tin . Tin certainly seems to be more strongly represented at Gemeinlebarn at this time than in contemporary finds further to the southeast, at the Maros-Tisza confluence. The cemetery of Mokrin, from the middle phase of the Maros culture, for example, contains a number of 'Cypriot' knot-headed pins, characteristic for Br A1b in the Unetice and related cultures, and has 14C dates clustered around 2000 BC (1824, 1953, 2034, 2034, 2035,2074 BC, see O'Shea 1992). According to the 44 analyses from Mokrin, tin was hardly used: the average tin content for objects in the cemetery was 0.113%, and 95% of the objects have less than 1% Sn (Liversage 1994: 81, table xvii). This comparison between Mokrin and Gemeinlebarn would seem to argue against a spread of tin bronze alloying from the south-east. Indeed, our previous discussion of the metal analyses in Crete, mainland Greece, Romania, Bulgaria and Yugoslavia showed that tin bronze was adopted rather haltingly in south-east Europe. It is interesting to note David Liversage's comment that there was more tin in Bell Beaker material than in either the Vucedol-Zok culture in the Carpathian Basin or in the Nitra group (1994: 96). For objects from Bell Beaker graves in southern Germany, Austria and Czechoslovakia, the average tin content is 0.354% (ibid. 1994: 80), including a handful of objects with surprisingly large amounts of tin: Bohdalice (awl 2.5%- Sn), Ledce (dagger 2.6% Sn) and Smolfn (dagger 5.1% Sn) in Moravia, Bylany (dagger 6.1% Sn) in Bohemia, and 18 CHRISTOPHER P ARE Pfutzthal (awl 10.5% Sn) in Sachsen­Anhalt (SAM 3238, 3248, 19935, 19936; Junghans et al. 1960: 194; Schickler 1981: 437; see also Kuna & Matousek 1978: 79, fig. 9, Il, crosses) . According to Spindler (1971: 207 H.) 43% of analysed objects from Bell Beaker contexts contain more than a trace of tin, compared with only 7% from Corded Ware contexts. Nevertheless, even in the Corded Ware culture objects with relatively high quantities of tin do seem to be represented, even though the reliability of association is not always beyond question: Altenburg (axe 5% Sn) and Ranis (bead 3% Sn) in Thuringia, HalleHeide (spiral armlet 1.3% Sn) and KirchscheidungenKloBholz (pin 11.5% Sn) in Sachsen-Anhalt, and Niederkaina, grave 7 (spiral 2.2% Sn) in Saxony (Otto & Witter 1952: anal ysis 211, 212, 692; Otto 1953: analysis B; Schickler 1981: 436). Even though it would appear to speak against his belief in a spread of tin bronze alloying to Europe from the south-east (Ana tolia, Aegean), Pernicka admitted that early (Copper Age) tin bronzes seem to be concentrated in the Corded Ware and Bell Beaker cultures of Central Europe, but not in southeast Europe?" (Pernicka et al. 1997: 125). a 60 40 20 11••• Fig. 1.10 (right). Histograms showing the tin content of copper and copper alloy objects in the area north of the Alps. - Only objects with at least a trace of tin are included on the histograms. - a) Br Al . - b) earlier part of Br A2. - c) later part of Br A2 . - d) Middle Bronze Age. - After Spindler 1971: 209, Diagram 1. _ b 60 11 = 799 40 --- 20 North of the Alps Turning to the area north of the Alps, we are able to draw on the important study by K. Spindler (1971). As he was interested in the earliest appearance of bronze, especially in small quantities, he organised his data in a rather unfamiliar way; on his histograms, for example, he uses a logarithmic scale, providing more information on low concentrations of tin than high tin allo ys (Fig. 1.10). In contrast to the Nitra group, the graves of the earliest Early Bronze Age horizon (Br Ala) north of the Alps contain hardly any metal objects, their place being taken by artefacts made of stone, bone or shell. Significant quantities of copper and its allo ys appear first in Br A1b, particularly in the graves of the Adlerberg, Singen and Straubing groups (Cerneinlebarn is also included in Spindlers analysis: Fig. 1.10, a). About 520 analyses were available for this horizon, roughly 200 of which had no tin at all, only 8% had more than 1% Sn, and less than 5% had more than 4% Sn. A recently discovered grave of the early Straubing culture from Buxheim, Upper Bavaria, =320 11 11 c 60 11 =553 40 --_. 20 d 60 11 =229 511 (%) 19 BRONZE AND THE BRONZE AGE contained 47 tin beads with a segmented shape resembling Early Bronze Age faience beads (Moslem & Rieder 1997). S. Moslem and K. H. Rieder pointed out the similarity with the segmented tin beads from Exloo, Prov. Drenthe, and Sutton Veney, Wiltshire (Penhallurick 1986: frontispiece; 67, fig. 24; for other tin objects in Europe, see Primas 1985) ­ suggesting a north­west European origin for the Buxheim beads. In Br A2a, metalwork becomes more widespread, and is now also well represented in graves from Moravia, Bohemia and central Germany. Spindler's histogram for this phase (Fig. 1.10, b) is slightly less easy to interpret, because he did not state how many samples analysed contained no tin . However, the 799 samples with at least a trace of tin indicate a major change in alloying: 71% of the objects contain more than 1% Sn and 50% have more than 4% Sn. Furthermore, Spindler notes that the tinpoor objects are mainly difficult to date, and the securely dated objects are generally alloyed with tin. In Br A2b (Fig . 1.10, c) all the samples have at least a trace of tin, and only 3% of the analyses contained less than 1% Sn. 88% of the objects have more than 4% Sn . Finally, in the Middle Bronze Age (Br B-C) 92% of the 229 analyses had more than 4% Sn (Fig . 1.10, d) . According to both Liversage and Spindler, it is clear that for the triangle reaching from central Germany in the north, to southern Germany and south-west Slovakia in the south, the transition phase to a full bronze-using metallurgy happened around Br A2a, some time between the 20th and 18th centuries BC. At this time, the distribution of tin was generally bimodal, with roughly equal numbers of artefacts containing above and below 4% Sn. It is interesting to compare a pair of histograms published by Helle Vandkilde (Fig. 1.11), showing tin distributions in the classic Unetice phase (Br A2a). Whereas in the 'central' area with 'princely' graves and rich hoards (middle Saale-Unstrut in Thuringia, southern Sachsen-Anhalt; mapped on Schmidt & Nitzschke 1980: 183, fig. 3) there are roughly equal numbers of artefacts containing above and below 2% Sn (Fig. 1.11, a), in 'peripheral' regions (north Bohemia, Spree-Neisse, Riesa-Dresden-Bautzen, Berlin-Brandenburg and Mecklenburg-Pomerania) the sampled objects are poorer in tin, with only 21% containing more than 2% Sn (Fig. 1.11, b). Vandkilde's important research on the transition 'From Stone to Bronze' (1996) has shown that in Denmark the situation in Late Neolithic rr (roughly comparable with Br A2a) was similar to that in the 'peripheral' Unetice regions (compare Figs 1.11, b; 1.12), although, with 34%, Denmark apparently has slightly more artefacts with over 2% Sn. From Per. lA (roughly comparable with Br A2b) onwards, almost all copper is alloyed with at least 4% Sn. 50 n = 194 a =625 b 25 ｾ C <:: "g" 0 50 Lt n 25 o o Tr-0.126 0.127-2.0 2.01-4.00 4.01-7.95 > 7.95 Sn( %) Fig. 1.11 . Histograms showing the tin content of copper and copper alloy objects in the I1netice culture. - a) the classic I1netice culture centre around the Unsirut-Saale in Thuringia. - b) the periphery of the classic I1netice culture centre (north Bohemia, Spree-Neisse, RiesaDresden-Bauizen, Berlin-Brandenburg, andMecklenburgPomerania) . - After Vandkilde 1990: 125, fig . 10. David Liversage (1994: 77, with fig . 5), discussing the Late Neolithic metalwork from Denmark, drew attention to the fact that the tin distribution was not bimodal, most of the tin-containing objects having tin concentrations between 1% and 7% Sn. He concludes as follows: "This must mean either that the Danish smiths were not interested in concentrating their tin in a full bronze, or more probably that objects of copper and bronze were being imported as separate commodities but were being mixed locally or on the way northwards in the recycling process. The northern metallurgists were thus obviously less advanced than those of central Europe. As copper and bronze were mixed the bipolarity disappeared from the tin distribution." 20 CHRISTOPHER PARE In Per la metalworking practices obviously changed markedly, and for the first time standard bronzes were used ­ either imported or alloyed locally from imported copper and tin. 75 50 The British Isles For the British Isles, we can base our review on an important study of southern British Early Bronze Age metallurgy by Needham et al. (1989), and a new chronological summary by Needham (1996; see also Gerloff 1996). Stuart Needham has divided early British metalwork into a series of chronological horizons, one for the Copper Age (Metalwork Assemblages I­Il), dating to the mid­late 3rd millennium BC, and 11 for the Bronze Age. The Early Bronze Age Metalwork Assemblages (MAs), and date­ranges, where possible based on modern precision 14C dates, are as follows: 25 o 75 LN 11 (n = 169 ) 50 III = ca. 2300­2050 BC (Butterwick daggers, Migdale axes etc .) 25 IV = ca. 2050­1900 BC (Aylesford hoards, Parwich grave etc.) V = ca. 1900­1700 BC (Wessex I grave series, Armorico­British daggers etc.) O---L..I----,- VI = ca. 1700 BC onwards (Wesse x II grave series, Carnerton­Snowshill daggers etc .) 75 Per. LA (n =61) 50 25 o­ ­ ­ ' ­ ­ ­ ­ ­ ­ ­ , ­ ­ ­ ­ ­ ­ r ­ ­ ­ , 75 Per. LB (n = 194) 50 25 0­+­­­­.­­­­..,­­­­.­­­­, o Tr­0 .126 $n (% ) Fig. 1.12. Histograms showing the tin content of copper and copper alloy objects in Late Neolithic and Early Bronze Age Denmark. ­ After Vandkilde 1996. Metalwork Assemblages I­Il are characterised by copper, arsenical copper and occasional bronzes (Fig. 1.13). The following horizon, however, shows a marked change: now over half of the metal objects contain 8­14% Sn, and the great majority (more than 93%) have over 5% Sn . Needham & Kinnes (1981: 133) have even argued for 'tinning' of undecorated flat axes at this time, a technique which is apparently most common among axes of the Dunnottar and Migdale groups (however, see also Close­Brooks & Coles 1980; Kinnes et al. 1979). According to Needham et al. (1989: 392, fig . 3), the transition from copper to bronze took place quite rapidly during the life of MA Ill. In the subsequent phases, there is only gradual change in alloying practices, involving increasing amounts of tin: 812% in MA IV, 8­14% in MA V, and 10­16% in MA VI. The early adoption of tin bronze in Britain was already noticed in a remarkable article by Hugh McKerrel (1978). He drew atten tion to the fact tha t the vast majority of thin­butted ' type B' axes, both in Scotland (95%) and in Ireland (90%), contain over 5% Sn . He also pointed to a similar development in metalwork from Bell Beaker graves, characterised by arsenical copper in steps 1­4 (21 objects of arsenical copper, three bronzes), and bronze in steps 5­7 (1 object of arsenical copper, 21 of bronze). And, using available 14C dates, he suggested a date BR O NZ E AND TH E BRONZE ACE MA I 20 (n = 5) 10 0 : MA n 20 : (n = 18) la 0 : MA III 20 V'J : (n = 59) 10 v V'J ｾ >-. c: ..... 0 0 Z 20 ｾ 0 : MA IV : (n = 26) la 0 :MAV 20 : (n =48) la 0 : MA VI 20 : (n = 120) 10 0 5 la 15 So (0/0) Fig. 1.13. Histograms showing the tin content of copper and copperalloyobjects in southern Britain in Metalwork Assemblages (MA) i- VI . - After Needham et al. 1989 : 391 , fig. 2. 21 of ea. 2200 BC for this important change in metalworking practice. McKerrel 's arguments have, therefore, been confirmed by the systematic modem research of Needham et al. Although there is a degree of fluidity about th e date-r anges of the individual phases, there has for a long time been consensus that the typical finds of MA III date well before Wessex I (Needham 1996: 130). This is important, in view of the fact that the Wessex I series of graves can be paralleled fairly reliably with Br A2a on the continent, for example with the aid of finds in graves and hoards of the classic Unetice phase in central Germany (Gerloff 1993; 1996). McKerrel identified the most important consequence of the early and regular use of tin bronze alloys, the question of the supply of Cornish tin to distant parts of the British Isles (1978: 11): " ... the distances involved are considerable; from Cornwall to Aberdeenshire some eight hundred miles by sea and from Cornwall to Northern Ireland perhaps half this distance. Yet, to judge by the consistency of high tin levels in the thin-butted Scottish axes, this was not an occasional or intermittent activity. Clearl y, within Britain, we do seem to have a consistent, well organized, long-distance tin movement dating to around 2200 BC" The characteristic feature which demonstrates a "consistent, well organi zed, long-distance tin movement" is a unimodal, tight and normal distribution of tin in copper objects. As we have seen, this is encountered in Britain already in MA Ill-IV (Fig. 1.13), roughly corresponding with Br A1. On the continent, the development sets in several centuries later, chiefly in Br A2b, for example in Denmark (Fig. 1.12), the area north of the Alps (Fig. 1.10, c), and in the Carpathian Basin (Figs 1.6, d; 1.9, d-e). This, too, was noticed by McKerrel; in view of the long distances involved in supplying Scotland with Cornish tin , he suggested, not unreasonably, that Cornish tin may equally have been taken across the English Channel to supply parts of continental Europe (1978: 14): "After 2200 BC, there is a remarkable change in the European situation and, as has been noted above, the transition to total bronze use in Britain takes place ap paren tly very rapidly; for Scotland nearly all copper alloys after 2200 BC are sound tin bronze. For central Europe and Italy from 2200 to 1800 BC, about one-third of all copperbased metal is good bronze. Thereafter, the proportion is very much higher. It is of course not yet possible to clarify the origins of the continental tin component, but, in view of the certain and extensive British use of the metal and the distances involved even within Britain, it is entirely conceivable that it was British tin being used at this time on the Continent." Although, as we have seen, there does seem to be 22 CHRISTOPHER PARE a consensus that tin bronze alloying became predominant in Britain well before Wessex I, it should be realised that the radiocarbon evidence is by no means as clear as one would wish. This has been pointed out by Fernandez-Miranda et al. (1995: 62): for example the Migdale hoard (wooden bear core: 3665 ± 75 BP) is not necessarily earlier than 2000 BC, and Manor Farm burial 1 (human and animal bone: 3450 ± 70 BP, 3270 ± 80 BP) is almost certainly later (for the 14C dates, see Needham 1996: 129). Owing to wiggles in the calibration curve, radiocarbon dating around 2000 BC will always be problematical, and there is clearly a risk that the earliness of the introduction of British tin alloying will be exaggerated. Indeed, there are problems around the same time in other parts of Europe. The five 14C dates for the halberds from Melz, Kr. Robel, Mecklenburg, hoard 11, conventionally dated to Per. lA/classic Unetice'!", are earier than expected, with a date before 2000 BC being most likely (Rassmann 1993: pis 26-27; 1996: 205, fig. 7). However the Melz dates are interpreted, they do not have much effect on our study of the introduction of tin bronze alloying: in Mecklenburg and Brandenburg true bronzes are hardly represented at this time - the main exception being halberds, which have an average tin content of 7.56% in the 14 analysed examples from Mecklenburg (ibid. 1993: 41; 246, table 7). The precise absolute chronology of the introduction of bronze to the British Isles must remain somewhat uncertain. However, it is surely reasonable to relate the transition from arsenical copper to tin bronze with roughly contemporary changes around 2000 BC - in the organisation of copper mining and metals supply. According to the results of recent research (see for example Craddock 1993; 1994; Ixer & Budd 1998; O'Brien 1996; 1999), during the second half of the 3rd millennium BC Ireland and much of western Britain was supplied with arsenical copper from the mines at Ross Island in County Kerry. This type of arsenical copper ('type A') was still used for the production of the earliest tin bronzes of the Killaha-Migdale horizon (see Northover 1982; the suggestions of Budd et al. 1992 have not been followed by other specialists). Around 2000 BC the Ross Island mines fell into disuse, and at the same time a large number of mining operations began in south-west Ireland, north Wales and north-west England . According to mineralogical analysis, these new 'Bronze Age ' mines produced relatively pure copper, which was then alloyed with tin, presumably from Cornwall. Both strands of study, the artefactual analyses of Needham et al. (1989) and the mining research described above, point to the same profound change in metallurgy in the British Isles at the end of the 3rd millennium BC. The Iberian Peninsula Evidence for the early use of tin bronze has recently been published from Late Chalcolithic Bell Beaker contexts in the northern Iberian peninsula (Alcalde et al. 1998; Comendador Rey 1999). From the extreme north-east, the rock shelter at Bauma del Serrat del Pont (Gerona) has tin bronze from three Chalcolithic layers (111 .1, 11.5, 11.4) as well as the Early Bronze Age layer 11.3; the 14C and analytical evidence is shown below: Layer " C (2 0) nu n.5 2783-2280 BC 2915-2583 BC 1 4 n.4 n .3 2889-2457 BC 2495-2030 BC 1 2 Samples Tin 7.096% Sn 0%, 0.547%, 0.606%, 7.687% Sn 4.642% Sn 23.51%, 41.37% Sn On the other side of the peninsula, at the site of Guidoiro Areoso on the Illa de Arousa (Pontevedra) in the extreme north-west of Spain, two awls with 21.52% and 21.86% Sn come from contexts dated to 2618-2458 BC (2 c). Most of the other Beaker bronzes, including tanged daggers and Palmela points, are not directly dateable; in France, however, similar high dates (2873-2489 BC, 1 c) are associated with a bronze dagger from a Beaker context at the Abri du Capitaine (Basses Alpes). It is quite clear, particularly from the Bauma del Serrat del Pont, that alloying with tin was mastered before the middle of the 3rd millennium BC in northern Spain. On present evidence, this is earlier than anywhere else in Europe, and it lends support to the view that the Bell Beaker cultures played a crucial role in the introduction of this metallurgical technique. It is less easy to demonstrate when bronze became predominant in the Iberian peninsula'!", although there is a consensus that tin alloying was adopted earlier in the north than the south (Fernandez-Miranda et al. 1995: 68): "Una vez en la Peninsula, la tecnologia del bronce continua el avance lentamente de norte a sur, como rumiada en el seno de grupos poco estructurados espacialmente, requiriendo rnas de dos siglos recorrer el camino que separa Navarra del Sureste." The authors are referring to the important Bronze Age site of Monte Aguilar in the Bardenas Reales (Navarra), where 13 metal objects have been found in layers A/V and B/VII, dated by 14C to ea. 18901750 BC, of which six are low-tin bronzes, and seven are copper or arsenical copper. According to Pernandez-Miranda et al., whereas tin bronze appears in the second half of the Early Bronze Age in the north, it is only introduced about two hundred years later further south, during the Middle Bronze Age (ca. 1700-1500 BC) - at roughly the same time throughout La Mancha, the Bronce 23 BRONZE AND THE BRONZE ACE Valenciano, the El Argar culture and the Balearic islands. Although, as we have seen from the Beaker evidence, there must have been some knowledge of bronze from the start of the Early Bronze Age , it did not dominate production in the north till the Middle Bronze Age (ca. 1700­1500 BC). In the northern Meseta, for example, the site of Tolmos de Caracena (Soria), with five HC dates between 1850 and 1520 BC, has produced eight metal objects, all of which copper or arsenical copper. On the other hand, in the central part of the peninsula, the site of Loma del Lomo (Cogolludo, Guadalajara), dated by HC to 1670­1390 BC, has analyses from eight objects, six of which are bronze with on average 11.7% Sn, the other two being copper or arsenical copper (Fernandez-Miranda et al. 1995: 64). Further south, Margarita Diaz-Andreu and Ignacio Montero (this volume) have shown that in Prov. Cuenca bronze starts to increase in importance during the Middle Bronze Age (17th-16th century BC), but only becomes dominant in the Late Bronze Age (15th14th century BC; see Fig. 8.3). The same authors state that bronze use started even later in La Mancha and the area of the El Argar culture. Tin bronze was absent in Mancha phases 1-11, and only two objects are known from Mancha phase III (ca. 1700-1500 BC); bronze use presumably increased in the subsequent Cogotas I culture (Martin et al. 1993: 33 H.; Fernandez-Miranda et al. 1996). Fernandez-Miranda et al. (1995) note tha t the earliest bronze objects from HC-dated contexts of the El Argar culture are from the site of Cerro de la Campana (Murcia), dating to ea. 1700-1500 BC; here there are equal numbers of tin bronzes and copper objects. Ignacio Montero Ruiz writes that 20% of Argaric tools and weapons were made of a coppertin alloy, he notes that it is at present impossible to determine precisely the start of alloying, but continues (1993: 53): "The available evidence from other areas of the Peninsula suggests that the appearance of bronze may not be much earlier than the middle of the 2nd millennium BC." The study by Hook et al. (cl 1987: 171, fig . 5) also indicates that the big change in metallurgy, with the transition to tin bronze, took place after the El Argar culture. Italy In the absence of systematic modern studies, the manner of the adoption of tin in Italy is impossible to chart with any accuracy, and we can merely summarise general trends. E. R. Eaton (1991) has published evidence for a rather significant use of tin bronze in the Copper Age (Remedello, Rinaldone and Gaudo cultures, beginning in the second half of the 4th millennium BC, and Bell Beaker culture, beginning in the first half of the following millennium). In a sample of 110 analyses, no less than seven contain more than 1% Sn (3 samples with over 5% Sn, 4 with 1-5% Sn, 3 with ea. 0.5% Sn) . In the sepulcral cave or Grotticella of Val Frascarese, Prov. Genova, for example, there is a tin bronze bead (ca . 16.4% Sn according to XRF analysis) from a context with five I4C dates ranging from 4340 ± 60 BP (3103-2872 BC, 2 0) to 3920 ± 60 BP (2501-2201 BC, 2 0) (Maggi & Formicola 1978; Campana et al. 1996). At the start of the Bronze Age there is a marked change in metalwork in northern Italy where, in the Polada culture, a wide variety of artefact types was adopted from the region north of the Alps (for a review of the types involved, see Hundt 1974). This introduces an important division in metalworking in the Italian peninsula during the Early and Middle Bronze Age: compared to the rich finds of the lake-side and Terramare settlements of northern Italy, southern Italy and Sicily remain extremely poor in copper and bronze. In Sicily, for example, the first object known by analysis to be of tin bronze is from the 14th century BC, a sword of the Thapsos culture at Caltagirone (Giardino 1997: 408; and pers. comm.). Konrad Spindler (1971: 230 H.) assembled the available SAM analyses of objects from Italy and Sardinia, and assigned them to chronological phases corresponding roughly to the following phases north of the Alps: Br Alb-A2a (EBA 2-3), Br A2b (EBA 4) and Br B-e (MBA): >7.95% Sn 4.01-7.95% Sn 2.01-4 .00% Sn <2.01% Sn EBA 2-3 EBA 4 MBA 29% 38% 4% 29% n = 52 37% 28% 15% 20% n = 68 53% 35% 5% 6% n = 94 According to this data, the development in alloying is similar to that north of the Alps: tin bronze production increased markedly at the time of the earliest northern 'imports' from the north Alpine Blechkreis (Br Alb), the tin distribution is bimodal in EBA 2-3, became predominant in EBA 4, and finally in the Middle Bronze Age there was standard tin bronze production. However, in the 30 years following Spindler's publication, the chronology of Italian Early Bronze Age metalwork has been the subject of some controversy. On the one hand, the dating of the early axes must be revised, following the surprisingly early HC date of the Otztal glacier mummy (L. H. Barfield, pers. comm.); on the other hand, in modem chronological schemes much of Spindler's EBA 4 material (e.g. hoards from Cascina Ranza, Costa di Monticelli etc.) is now assigned to 24 CHRISTOPHER PARE Br medio 1, which is associated in central and southern Italy with early Mycenean pottery. It is very likely that a new, systematic study of all the available analytical evidence will provide a less simplistic view of changes in metallurgy (eJ. Fernandez­Miranda et al. 1995: 61). Peter Northover, for example, gives a tantalising glimpse of regional variation in alloying practices (1988: 50): "The Rinaldone industry [in Tuscany] with its As/ Sb composition ... is replaced by an industry using 5­8% tin bronzes allied to a copper type with up to 2% or more lead and zinc . The association of finished axes and copper ingots shows that the alloy was produced by alloying tin, perhaps from Tuscany, with copper from a Cu/Pb/Zn source. The Remedello industry [in northern Italy] on the other hand is succeeded by two alloy types: in some areas, particularly Lombardy, there is a low tin bronze with 3­5% Sn, while to the east in Veneto and Friuli, an imported As/Sb/ Ag/Ni composition is used, a composition that can have attractive properties ... Only at a later date is this material replaced by a high tin bronze." The suggestion that tin deposits in the eolline metallifere may have supported a relatively precocious bronze production in Tuscany gains even more importance in the light of evidence for an early trade in metals from the the island of Vivara, off the bay of Naples. Excavations at the Punta di Mezzogiorno and the Punta d'Alaca have brought to light a large quantity of bronze objects, and metalworking debris including crucible fragments with tin bronze residues, associated with early ProtoApennine B pottery and Mycenean pottery reaching back to the MH/LH I transition (Giardino 1998). As Claudio Giardino writes (ibid .: 158­161): "Since the Campania area is wholly without metal resources, the raw material must have come from other areas, which could reasonably be assumed to be in the areas on the Tyrrhenian Sea; the Gulf of Naples was an important stopping point in the Tyrrhenian routes. ... The Vivara settlement may have been the most westerly point of a network of maritime traffic ensuring the Eastern Mediterranean with a regular metal supply. The island was the point of arrival for a complex system of short and medium distance 'local' trade for collecting and transporting metal from the mining areas, probably through various intermediate stages." Giardino's conclusion that Vivara articulated a trade in metals with the Aegean, lasting from the mid 17th till the end of the 15th century BC (LH I­IIIA1), seems difficult to avoid. As the Vivara finds include many tin bronze objects, but also crucibles, slags, copper alloy prills and a piece of copper casting residue, the commodities involved were presumably copper, tin and/ or bronze, most likely from Tuscany. RESULTS AND DISCUSSION The definition of the Bronze Age and the 'Bronze Age Hypothesis' In the absence of an agreed definition, general usage of the term Bronze Age is ­ not surprisingly imprecise and often misleading. In my opinion, there can only be one viable definition for the start of the Bronze Age : an adaptation of Snodgrass' criterion ('working iron') for the start of the Iron Age, quoted at the beginning of this article. The Bronze Age is the period in which bronze was the material predominantly used to make the functional parts of those major categories of cutting and piercing implements which were fundamental for early technology. Whereas the transition to the Iron Age seems to have been 'abrupt' in most parts of the Old World, this is certainly not universally true for the transition to the Bronze Age. In many parts of the Near East, for example, arsenical copper was still widely used in the 2nd millennium BC. But in other areas, such as Central and north­west Europe, there was a relatively abrupt adoption of bronze as the main working metal, and it remained predominant for many hundreds of years. This suggests that whereas the term Bronze Age may be almost irrelevant in some areas, it may be both applicable and useful in others. When archaeologists use the term Bronze Age, it can have merely a conventional meaning, for example for classificatory or chronological purposes; on the other hand there are more interesting uses, chiefly as a stage in technological evolution or culture­history (the 'Three Age System' combines elements of all these uses and meanings). However, evolutionary schemes are today decidedly out of fashion, and the most widely (if not explicitly) held view ­ in Europe at least ­ is probably what I call the 'Bronze Age Hypothesis' (an interesting version in Kristiansen 1987: especially 46 f.). This proposes that in the Bronze Age: • • • • • Bronze was fundamental both in economic production and social reproduction. It was therefore essential for societies to obtain bronze (or copper and tin). As the vast majority of societies did not have local supplies of copper and tin, they were obliged to participate in exchange networks which linked them, directly or indirectly, with distant sources of metal. Consequently, and compared with earlier periods, the Bronze Age was characterised by a massive increase in exchange. The exchange system lent itself to control by emerging elites, which tend to be more noticeable in the Bronze Age than in previous periods. BRONZE AND THE BRONZE ACE If this hypothesis is correct, for those cultures and periods in which 'bronze was fundamental both in economic production and social reproduction' the Bronze Age was a time with a distinct character, significantly different from anything before or after. Europe and the Near East Bronze was adopted very differently in Europe and the Near East. In the Near East tin bronze was introduced and disseminated in the context of an elite long­distance exchange network, organised by the cities of Mesopotamia. However, after the sudden inception of bronze use in the mid 3rd millennium BC, bronze was only taken up haltingly in the Near East, as the histograms published by M. Frangipane clearly demonstrate (1985: 219/ fig. 2/ b; 221/ fig. 3; 222/ fig. 4). In Europe, by contrast, there were no cities to administer long­distance caravan trade, and exchange was obviously organised in completely different ways. Nevertheless, in quite large areas of Central and north­west Europe there was a relatively abrupt transition to total bronze use : particularly in the British Isles (Fig . 1.13)/ Denmark (Fig. 1.12)/ and the region between central Germany, Transylvania, Tuscany and the area northwest of the Alps (e.g. Figs 1.9; 1.10). One of the reasons for the difference between Europe and the Near East is the occurrence of tin: workable tin deposits were, according to the present state of research, either absent or extremely rare in the Near East. Although there is general agreement that workable sources of tin were rare throughout the ancient world, there is considerable controversy about individual cases (for good reviews of European tin, see Roden 1985; Penhallurick 1986). It is interesting that the most authoritative specialists in archaeometallurgy are also the most sceptical about tin sources. J. D. Muhly and E. Pemicka, for example, argue strongly against Kestel and other mining regions in Turkey, as significant sources of tin. However, in Europe there are persistent claims for minor tin workings, for example in Serbia (McGeehan-Liritzis & Taylor 1987; Taylor 1989: 84; Durman 1997)/ the Slovakian Ore Mountains (Schalk 1992: 158; 1998) and south-east Spain (Harrison 1974: 79'near Cartagena). Muhly's earlier doubts about prehistoric tin mining in the Bohemian/Saxon Erzgebirge have been dispelled by the results of recent excavations (Bartelheim et al. 1998; see also Taylor 1983)/ and it is certainly not inconceivable that future research will find evidence for Bronze Age activity at previously unknown tin mines and placer deposits. Nevertheless, whether five (Cornwall/ Brittany, north-west Iberia, Erzgebirge, Tuscany) or ten sources of tin were in use in Bronze Age Europe, it remains true that for the vast majority of 25 cultural groups the metal was not locally available, and had to be obtained by exchange. Many regions, such as southern Scandinavia, were equally reliant on external exchange for their supplies of copper. In view of the difficulties involved in organising and maintaining an exchange network for the supply of copper and tin, even in Europe, the abrupt adoption in large parts of Europe of a metallurgy dominated by bronze is something of a mystery which deserves serious discussion. It is essential to understand that there was no techno-evolutionary imperative which made this change unavoidable. The Minoan palaces, for example, flourished with very little bronze, and there is no reason why other European cultures could not have survived quite happily in the 2nd millennium BC without bronze, for example using a variety of work-hardened coppers with small quantities of arsenic, antimony or lead . For this reason, the adoption of 'total bronze' in Europe should be understood as a cultural choice, not a product of technological determinism, even though in many respects bronze may represent a technological advance compared to stone or copper. The natural consequence of this suggestion is that the term Bronze Age has quite different meanings when applied to Europe and the Near East. In each area, bronze was adopted for distinct and unique cultural reasons. The European Metallurgical Province We have already suggested that the European Metallurgical Province, in E. N. Chernykh's terminology/ emerged alongside the Circum-Pontic Metallurgical Province. The roots of this new European metallurgical tradition can probably be traced back to the early 3rd millennium BC, when evidence for the mining of complex sulphide copper ores begins in many parts of Europe; this is the 'Aufbauphase' (Foundation Phase) in C. Strahm's developmental scheme. During the late Chalcolithic there was also a marked increase in the use of tin : this is most clear for the Bell Beaker culture, but also seems to have been true for the Corded Ware, Remedello and Rinaldone cultures. Presumably there was some knowledge (and use) of local tin deposits, for example in Cornwall, north-west Iberia, Tuscany or the Erzgebirge. The adoption of bronze took place against this background, in a development which I have attempted to illustrate on Fig. 1.14. For each area where published data is available, I have noted the approximate date of the transition to the full Bronze Age, according to the above definition. The arrows on the map are included to emphasise the chronological gradient between the adoption of bronze in various areas; they do not necessarily show the direction of N 0\ n :r: ;<l ｾ ­e :r: tT1 o ;<l t:() "'l '\:I )- f;; <S ' ", o o \ 1" " ,," , o oI"""" I 500 m iles I 500 kilometres Fig. 1.14. Map showing probable dates for the transition to full bronze use in Europe. ­ The arrows indicate the chronological gradient between the adoption of bronze in some areas. BRONZE AND THE BRONZE AGE 'influences' involved in the expanding distribution of bronze. The evidence used for Fig. 1.14 has been described in the preceding pages, and it is clear that for many areas we are lacking either good analytical data or good chronological definition or both: for quite large areas the available analytical data is difficult to assess, owing to the lack of a clear chronology (e.g. France), and even in the areas where data is available, it is by no means uniformly reliable. Nevertheless, it is possible to reconstruct the broad outlines of the adoption of bronze in Europe, and there is every reason to believe that future research will be able to provide a much more detailed picture of the process. The earliest evidence for a full Bronze Age in Europe is in the British Isles, where, according to the work of Hugh McKerrel, Stuart Needham and others, bronze swiftly replaced copper at some time during the last quarter of the 3rd millennium BC, or at least prior to 2000 BC (cf Fig. 1.13). Shortly after this, at around 2000 BC, there seems to have been a radical reorganisation of copper mining operations in the British Isles . Obviously, to supply metalworkers in Britain, tin was already being obtained in considerable quantities from Cornwall; it is uncertain to what extent Cornish tin was reaching the continent, although the segmented tin beads from Exloo and Buxheim might hint at exchange over long distances even at the end of the 3rd millennium BC. Tin was probably being obtained from sources in Brittany by the late 3rd millennium BC, but it is uncertain at present exactly when bronze became predominant in north-west France . Alloying with tin was increasing rapidly in Central Europe, north Italy and southern Scandinavia in the first quarter of the 2nd millennium BC, at the time of the phase Reinecke Br A2a (cf Figs 1.9; 1.10; 1.11; 1.12). In the subsequent chronological phase (by ea. 1800/1700 BC), a vast area between Denmark, Transylvania, Tuscany and the north-west Alps was using bronze for almost all non-precious metal products (note that it is unclear when full bronze was adopted in France and the Low Countries) . It would be logical to assume that the transition to bronze occurred earlier near putative tin-producing regions (e.g. Tuscany or the Erzgebirge), but the present state of chronology is insufficiently precise to draw such detailed conclusions (but cf Fig. 1.11). This was a crucial stage in the transition to the Bronze Age in Europe: in a large area full 'Bronze Age' production commenced around 1800/1700 BC, in what seems to have been an inter-related development. Within this zone, an exchange network was clearly functioning well enough to ensure reliable supplies of copper and tin. By the mid 17th century BC, the southernmost part of this 'netw ork' was linked via Vivara to the early Mycenaean Aegean. 27 Owing to the lack of detailed information on France and the Low Countries, it is uncertain how the adoption of full bronze use spread from northwest to Central Europe. Our information is likewise somewhat sketchy when we turn to southern and eastern Europe; nevertheless, according to the present consensus, it seems that full bronze use was only adopted in the southern Iberian and Italic peninsulae after the 15th century, and the same seems to be true for Bulgaria and the north Pontic steppes. Owing to its developed social and economic organisation, and wide-ranging trading activity, the Aegean must obviously be treated as an exceptional case, but even here full bronze production only seems to have occurred from LH/ LM IIIA onwards. The general picture which emerges is an expansion of full bronze use from 1) north-west, to 2) Central, and 3) southern and eastern Europe. On one level, this development clearly indicates a growth, both in volume and range, of the exchange networks supplying copper and tin for bronze production. But the expansion of full bronze use also required the adoption of a cultural norm in which bronze was fundamental both in economic production and social reproduction (cf the 'Bronze Age Hypothesis'). It is this cultural norm which lends the European Bronze Age its unique character. Metallic currencies The special feature of the European Metallurgical Province was the relatively abrupt and complete transition to full bronze production. In our review of the evidence, particularly for Central and northwest Europe, we noted a Widespread change to full bronze production between the last quarter of the 3rd and the first quarter of the 2nd millennium BC. In view of the problems involved in obtaining tin, it is surprising that tin bronze was used not only for cutting and piercing implements such as axes, daggers and awls, but also for decorative items like jewellery. In fact, in the full Bronze Age , practically all metal artefacts were made of bronze; exceptions include most ingots, and objects made of precious metals such as gold and silver. This exclusive use of bronze suggests that the reasons behind its adoption were not simply its technical advantages; if that were the case one would expect it to have been used preferentially for implements requiring the improved hardness and castability of bronze. It is more likely that bronze quickly came to have a standard value, which led to the rapid decline in use of other kinds of copper and copper alloy . The golden brown colour of bronze was not only attractive and well suited for making ornaments, but could 28 CHRISTOPHER PARE also have been useful in assaying the quality of the metal, in a way which is impossible for other copper alloys, for example those containing small quantities of arsenic, antimony or lead. Using this bronze standard, artefacts would have a 'convertible' material value, and could be used as a kind of currency. This idea has been suggested by Andrew Sherratt and Step hen Shennan: "The diversification of exchange systems required an element of liquidity: a standard of exchange that was convertible between different kinds of goods, and which could balance the flows between different areas - in other words a kind of protocurrency. To be acceptable between different cultures and value systems, this had to be a material which was generally desirable, but which could change its form - easily and without loss between different forms of local expression. It must be neither too rare nor too common. Copper on its own might potentially have come to play such a role; but it was inhibited both by the limited range of exchangeable goods, and by its own limited range of finished forms. As Step hen Shennan has argued ... it was only in the alloyed form, as bronze, that it came to be such a medium. ... Bronze was more than a 'primitive valuable' for use only in ceremonial prestations within restricted spheres of exchange ... Instead, it was an internally produced and valued material which had both an inherent attractiveness and a variety of useful forms ... it was capable of being reshaped into other forms of object; but as a major display material it carried its own connotations both as adornment and through its association with bright weapons and cutting tools. In conjunction with livestock and textiles, it could circulate in local exchange cycles; but melted and recast, it could be transferred between local typologies and the regional ideologies they expressed. It was the large-scale use of bronze which linked these potentialities into actuality, and made bronze into the primary material of the margin." (Sherratt 1993: 17) "In the core areas [of Central Europe] the key to the change was that copper and bronze underwent a process of commoditisation - they became important as unit quantities of metal rather than as restricted prestige items for social transactions, and may indeed have functioned in some respects as a proto-currency - as a means of exchange and store of value.... In summary then, copper and bronze represented wealth in the core area convertible, it is suggested, for a range of other goods and materials, as well as being themselves convertible into tools and weapons." (Shennan 1993: 62 f.) "The argument that copper and bronze became commodities measured in standardised quantities clearly has implications for the nature of Bronze-Age transactions.... We may ... suggest the development of commensurability between a variety of different kinds of goods in terms of metal ..." (ibid.: 65). Important contributions to this debate have also been published recently by M. Lenerz-de Wilde (1995) and M. Primas (1997; Primas & Pernicka 1998). Whereas Primas : most important results concern Middle and Late Bronze Age scrap hoards, Lenerz-de Wilde concentrated mainly on a careful analysis and interpretation of Osenring ingots; she summarised her conclusions as follows: "Die hier skizzierte Entwicklung vom Osenhalsring als Schrnuck zur Barren- und gewichtsgenormten Geldform, die begleitet oder abgelost wird durch immer shirker fragmentierte Metalle, stellt eine der klassischen Entstehungsformen von Geld dar: Ein Gegenstand ist als Schrnuck weit verbreitet und sehr beliebt, entwickelt sich so zum store of value, zum begehrten Tauschobjekt und schliefslich zum Tauschmittel. Anders ausgedruckt ist dies die Kette: Schmuck - Ringgeld Kummerforrn ('Zeichengeld')." (Lenerz-de Wilde 1995: 319) Broadly speaking, it seems that in Europe a new relationship to metals emerged in the late 3rd millennium BC. The most obvious signs of this are 1) the thousands of copper Osenring ingots in Central Europe, which seem to have been made to standard weights for use as a sort of currency (see also Pare 1999: 478 f.), and 2) the adoption of standard bronze, which gave artefacts a convertible material value. The coincidence of these two developments is surely not fortuitous . What use was this mass ive increase in the supply of copper, in the form of Osenring ingots, without alloying? In Britain, it seems that following the earliest horizon of tin bronzes using 'type A' copper (Migdale-Killaha phase), a new system of copper and tin exchange developed around 2000 BC. 'Pure' copper from a series of new mines came into circulation, which could be alloyed with tin from Cornwall and Devon. Owing to the restricted occurrence of tin, there must have been an element of long-distance trading, which may now have been controlled by elites of the sort buried in the tumuli of the Wessex culture. According to 14 C dates from the copper mines, this new system of metals circulation probably functioned throughout the Early and Middle Bronze Age, roughly corresponding to the time covered by the Wessex I and 11 graves. In Central Europe, the mass prod uction of Osen- BRONZE AND THE BRONZE AGE ring ingots seems to have commenced around the time of the early Unetice culture (Br A1b). Copper and tin production in the Saxon/Bohemian Erzgebirge must have started by the time of the classic Unetice culture (Br A2a), and it is to this phase that the rich Unetice hoards, and the 'princely' burials such as Helmsdorf, Leubingen, Dieskau (Schmidt & Nitzschke 1980) and Leki Mate are assigned. In this case, too, the emergence of elites is generally explained by their control of long-distance trade in copper and tin (e.g. Gerloff 1996: 14). At a slightly later date, the organisation of metals exchange could have been administered by a series of concentrated settlements, which are found over large parts of Central Europe in Br A2: defended settlements in the northern part of the Carpathian Basin (Veterov, Mad'arovce, Fuzesabony. Otomani cultures), hillforts in the area north of the Alps, and lake-side settlements north-west of the Alps. Around the transition from the Early to the Middle Bronze Age, a radical change in the organisation of metal circulation is indicated by the appearence of hoards of scrap bronze in Central Europe (BuhlAckenbach horizon) . With the increased quantity of copper, tin and bronze which was now in circulation, it seems that the earlier high-level long-distance (directional) trade in copper and tin was undermined by a low-level short-distance (down-the-line) exchange in scrap bronze'!", This has been described in a number of articles by M. Primas, for example: "Mit dem Einsetzen der Zirkulation und Deponierung von Bronzebruchstucken und Rohmetall veranderte sich die Quellenlage fur beides deutlich. Die gegen Ende der Fruhbronzezeit aufkommende Horte mit diesen Materialklassen setzen sich von den alteren Deponierungsmustern in verschiedener Hinsicht klar ab. DaB wir hier den Aufbau einer neuen Realwertordnung und den Untergang des vorher erfolgreichen Systems vor uns sehen, ist ziemlich evident." (Primas & Pernicka 1998: 56 f.) "The centuries following the Early Bronze Age are now seen to be a period of fundamental change. The pattern of hoard contents was modified drastically between 1500-1300 BC ... Scrap metal and ingots of various shapes began to circulate regularly on an interregional scale . Gold was now available in considerable quantities. These features speak for the Widespread use of weighed metal as a means of payment. At the same time , the functional use of different alloys, specialised tools ... and adequate working techniques became common practice and attest a more professional handling of metal. ... raw material shaped to flat or planoconvex ingots and the circulation of scrap bronzes started simultan- 29 eously at a time rightly designated an innovation horizon." (Primas 1997: 123) From this time onwards, there is increasing evidence that fragmented bronze was hoarded and exchanged, in the context of a 'Weighed Currency Economy' (Pare 1999: 510 ff.). This radical change in the organisation of the exchange of copper, tin and bronze corresponds with equally fundamental changes in other aspects of life: the rise of the Tumulus culture, the end of Tell settlements in the Carpathian Basin (David 1998) and the end of concentrated settlements (lakeside settlements and hillforts) in the area north of the Alps (Pare 1996). Somewhat later, there seems to have been a similar 'collapse' of the copper-tin exchange system in Britain, with the decline in operational activity in most copper mines and the end of the Wessex culture, both happenning around 1400 BC. At about the same time, there is increasing evidence for trade in scrap metal, both from wrecks found in the English Channel (Muckelroy 1980; 1981) and from analytical research (e.g. Northover 1982). Apparently, it was possible for local elites to control copper and tin supplies during the earlier part of the Bronze Age, but with the steadily increasing amount of standard bronze in circulation, the system of control was undermined and collapsed. The exchange system which replaced it was based on weighed fragmented bronze, which was collected in hoards; the amount of bronze which could be collected is illustrated by the hoard from Uioara de Sus in Transylvania, which contained nearly 6,000 pieces with a total weight of 1,300 kg (Coles & Harding 1979: 409). The first evidence of the system of bronze exchange based on hoarded scrap metal occurred in the Biihl-Ackenbach horizon at the transition from the Early to the Middle Bronze Age, in the 16th century BC (variously called ' Br A2c', ' Br A3' or ' Br A /B') . Over the ne xt two or three centuries this system seems to have been adopted over most of the European Metallurgical Province, and in the Urnfield period we find a vast number of hoards in an area reaching between the Carpathians and the Atlantic. By the 13th century BC, the expanding European Metallurgical Province had evidently reached the Aegean, in a process described in the stimulating article by Sue Sherratt later in this volume. Diffusionism and the tin trade Early studies of the start of the Bronze Age in Europe were dominated by diffusionist models, often involv ing the agency of 'prospectors' travelling from 30 CHRISTOPHER PARE the Near East in search of copper and tin. Indeed, the start of the Central European Bronze Age was originally defined by the appearance of new bronze types introduced from the East Mediterranean: "Zeitbestimmend waren fur den Beginn der europaischen Bronzezeit Metallformen, deren Herkunft aus dem Mittelmeerraum uneingeschrankt angenommen wurde: Osenhalsring, triangulare Dolchklinge, Stabdolch, Schleifennadel und die Nadel mit gelochter Halsschewellung. Mit ihrer Hilfe wurde der Beginn der friihen Bronzezeit in Mitteldeutschland in das 19. [ahrhundert v. Chr. angesetzt." (Schickler 1981: 433) Variations on this theme can be found, for example, in the works of Montelius, Dechelette and Childe, but one of the most important milestones was doubtless the detailed discussion by Claude F.A. Schaeffer, published in 1949, which focused on the Qsenring ingots found at Ras Shamra: "ll semble done qu'au troisierne millenaire les porteurs de torques avaient leur centre de diffusion initial dans les pays miniers par excellence bordant au nord le Croissant Fertile. Vers la fin du troisierne millenaire et au debut du second, au cours ou a la suite d'une periode critique ayant cause des destructions dans de nombreux centres urbains de l' Asie Occidentale, nous les voyons s'installer en Syrie septentrionale. lls contribuent puissamment a l'essor de l'industrie du metal et du bronze, notamment aRas Shamra et a Byblos. A la meme periode, en Chypre aussi s'ouvre un ere de prospection et de developpernent des gisements cupriques. A la fin du troisierne millenaire et au debut du second, les porteurs de torques s'installent aussi en Palestine OU ils repandent la technique du bronze. De Palestine, ils ont penetre dans la vallee du Nil ..." (Schaeffer 1949: 109 H.) ... [after discussing how the porteurs de torques reached Europe, ShaeHer continues as follows] ... "Force est done d'admettre que les porteurs de torques de Syrie, prospecteurs et artisans en metal, ont pris la mer. En suivant les cotes sud de l' Anatolie, les iles de l'Egee et les rives de l' Adriatique, ils semblent s'etre avances directement vers l'Europe centrale ou ils ont du provoquer le prodigieux developpement des mines et de l'industrie metallurgique de Boherne et de Hongrie qui marque le debut du Bronze." (ibid.: 115) Although diffusionism, as a catch-all explanatory model, has declined in popularity, it is still often argued that tin bronze alloying spread from the Near East to south-east, Central and northern Europe: as we have seen, this is postulated by E. Pernicka, and the idea is also reflected on an illustration published by A. Sherratt (Fig. 1.2). Indeed, there is nothing inherently wrong with this suggestion. However, the diffusionist position was seriously weakened by a later article by SchaeHer himself, in which he reversed his earlier argument for Near Eastern prospectors: "Aujourd'hui, ou nous possedons des releves topographiques et typologiques precis des plus anciens bronzes detain, et ou leur datation absolue est mieux assuree, le problems est a reprendre. Nous avons vu ... que les centres de fabrication des premiers objets, armes, outils et parures, en bronze riche en etain des Porteurs de torques se trouvaient non en Anatolie, dans le Proche-Orient ou en Egypte, mais indiscutablement en Europe centrale. A la lumiere des recherches recentes, la these de l'anteriorite des torques provenant du Proche-Orient, ainsi que des armes, outils et parures auxquels ils etaient associes, par rapport aux trouvailles analogues de l'Europe centrale, doit etre abandonnee." (SchaeHer 1978: 486) ... "C'est aux Porteurs de torques de l'Europe centrale, par l'intermediaire de leurs artisans les plus hardis venus s'installer dans les grands centres de commerce sur les cotes orientales de la Mediterranee, a Byblos, a Ras Shamra-U gari t et ailleurs en Proche-Orient, qu'etait done devolue la mission de propager la technique d'une metallurgie avancee qui a marque l'ouverture d'une nouvelle ere." (ibid.: 490) In this new version of events, SchaeHer suggests that specialist artisans from Central Europe travelled to the Fertile Crescent, bringing with them a more advanced technology of tin bronze production. Central European specialists also play an active role in a slightly earlier article by Konrad Spindler, published in 1971; in this case, the tin sources of Cornwall and the Iberian peninsula were discovered by prospectors from Central Europe, Switzerland or southern France: "Die Zinnseifenlagerstatten der Pyrenaenhalbinsel werden ... von m6glicherweise schweizerisch-sudfranzosischen Prospektoren entdeckt, die beginnen, diese abzubauen und das Zinn in den Rhoneraum zu verhandeln." (Spindler 1971: 240) ... "Mit der Betrachtung der metallurgischen Erscheinungen auf den Britischen Inseln wird deutlich, daB hier die Zinnbronze oHenbar eh er bekannt wird als auf der Iberischen Halbinsel. Vornehmlich werden wohl mi tteleuropaische Prospektoren die kornischen Zinnfelder ... entdeckt und fur eine weitreichende Belieferung des Kontinentes mit Zinn gesorgt haben." (ibid.: 242) This complex subject has been reviewed in an important recent article by Sabine GerloH (1993), in BRONZE AND THE BRONZE ACE which she studied artefacts in Central Europe which seem to suggest contacts with the East. ｍ･､ｾｴ ｲﾭ ranean in the late 3rd and early 2nd millennium BC and arrived at conclusions comparable to those of SChaeffer (1978). The crucial difference in Gerloff's account, however, is that she explains the 'contact finds' in the East Mediterranean and Central Europe as the result of trade, not the movement of specialist artisans such as the 'porteurs de torques '. GerloH argued that the optimal bronze alloy with ea. 10% tin was first adopted in Central and especially western Europe (1993: 83; 84, fig . 11). She summarised her arguments as follows: " Der Vertrieb von atlantischem und auch erzgebirgischem Zinn darf wohl auch in Zukunft mit den oben genannten Kontaktfunden zwischen dem 'barba rischen' Europa und den bronzezeitlichen Hochkulturen des Mittelmeerraumes in Verb indung gebrachtwerden." ... "Wir mussen wohl davon ausgehen, daf diese Kontakte nicht, wie in jiingster Zeit vielfach propagiert. ｷｵｲｾ･Ｌ auf den gelegentlichen Austausch und die ｺｾｦ｡ｬﾭ lige Dbertragung von ｐｲ･ｳｴｩｧ･ ｩｴ･ｾｮ ｺｵｲ ｣ｾﾭ zufuhren sind, sondern daf es zu diesern Zeitpunkt schon bekannte und f:st etablie:te Handelswege gab, die vor allem mit dem Vertneb von Rohmaterialien in Verbindung gebracht werden miissen." (GerloH 1993: 84 f.) ... "Folgt man der Meinung, daf der Zinnhandel das ausschlaggebende Moment fur die Fernverbindungen zwischen dem zinnreichen Mittel- und Westeuropa einerseits und den zinnarmen bronzezeitlichen Hochkulturen des Mittelmeerraumes andererseits war, konnen wir davon ausgehen, daf zu Beginn der mitteleuropaischen Bronzezeit wohl vornehmlich Zinn aus dem Erz- und wahrscheinlich auch aus dem Fichtelgebirge am Ende des 3. und Anfang des 2. vorchristlichen [ahrtausends in mehreren Schiiben uber die Donau an die Westkiiste Vorderasiens gelangte. Im zweiten Viertel des 2. Jahrtausends scheint sich der Zinnhandel mit den mittelmeerlandischen Hochkulturen mehr nach Westen verlagert zu haben. Spatestens zu Beginn des 16. vorchristlichen [ahrhunderts, wenn nicht noch friiher, wurde zum Zeitpunkt der friihesten m ykenischen Schachtgraber vor allem atlantisches Zinn in ?en Mittelmeerraum verhandelt. Dies gelangte nicht iiber die Donau in den Vorderen Orient, sondern, wie auch spater in klassischer Zeit, iiber den westlichen Mittelmeerraum vor allem in den fruhspatbronzezeitlichen griechischen Raum und wohl auch noeh nach Zypern." (ibid.: 85 f.) The idea of a tin trade supplying the Mycenaean Aegean with tin from Cornwall and Brittany is far 31 from new, and is supported by influential archaeometallurgists such as J. D. Muhly: "l have long argued for the possibility that, from the late Middle Helladic period onward, beginning with the period of the Shaft Graves at Mycenae, the Aegean world was making use of northwest European sources of tin, especially those in southwest England ... and Brittany ... To argue for the use of Cornish tin at Late Bronze Age Mycenae is not to have the Mycenaeans as builders of Stonehenge. It is most unlikely that anyone from the Aegean ever reached ｳｾｵｴｨＺｭ England during the Late Bronze Age. Tin, like amber, made its way across Europe through a series of middlemen, perhaps as Diodorus Siculus describes, albeit for a much later period ..." (Muhly 1985: 287 f.) The article by Gerloff represents the clearest and most recent statement of what might be called the ma xima list view of relations between Early Bronze Age Europe, the Near East and the Aegean. Similar ideas, more subtly expressed, are also to be found in the fascinating World System approach of Andrew Sherratt (1993: 22 H.). Basically, these authors entertain the possibility that developments in the Central and western European Early Bronze Age were profoundly influenced by contact with the N:ar Eastern and Mediterranean worlds . Two major phases of influence are recognised: the first ea. 23001700 BC dominated by contact with the Near East ('Troy'), and the second, from the 17th century BC onwards, dominated by long-distance trade with the Aegean ('Mycenae'). CONCLUSIONS The purpose of this article was to explore the meaning of the term ' Bron ze Age'. During my research on the Bronze Age/Iron Age transition (for the first in a series of planned publications, see Pare 1998), the simple question arose: how important was bronze for Bronze Age societies? Looking at the rich variety of bronze objects made for functional use and for display, for example in Central Europe between Reinecke Bronze A and Hallstatt B, one could easily believe that bronze was 'fund amen tal both for economic production and social reproduction'; if so, obtaining, displaying and exchanging bronze became a vitally serious 'game' which had to be played, at least by aspiring socio-economic elites, This belief has been reinforced by my work on this article, which suggested that tin bronze was adopted in Europe in a special way. The abrupt and complete transition to full bronze use, first in north- 32 CHRJSTOPHER PARE west and then Central Europe, indicates that bronze use and exchange became a cultural norm, part of a value system in which 'standard bronze' came to be used as a kind of currency. As this seems to be a cultural phenomenon typical for Europe, it is apposite to use Chernykh's term 'European Metallurgical Province' to describe it. Fig. 1.14 illustrates in a schematic fashion the expansion of full bronze use within the European Metallurgical Province from the last quarter of the 3rd millennium to the third quarter of the 2nd millennium BC. Over this long period, the bronze exchange network expanded, and reached its greatest extent by the 13th century BC. The Late Bronze Age hoards are clear evidence of the continuing validity of the bronze currency until its collapse at the Bronze Age/Iron Age transition, which first occurred in the Aegean in the 11th century BC, and can then be traced in other parts of southern and south­eastern Europe around the second half of the 10th century BC, and in Central Europe at ca. 800 BC. Some approaches to the question of the start of the Bronze Age have been outlined above. It is difficult to believe that prospectors could have contributed anything to this process: specialist tin miners almost certainly did not exist in the Near East or the Aegean, and there was already limited knowledge of tin ores and copper­tin alloys in Europe before the mid 3rd millennium BC. This seems to have been most prevalent in Central and western Europe, especially in the Bell Beaker culture. Long­distance trade in tin is extremely unlikely in the late 3rd millennium BC: before supplying the Near East with tin, there must have been a functioning supply network somewhere in Europe, and this was only just being established in the British Isles at this time. An 'intern ational' tin trade becomes more likely after the first quarter of the 2nd millennium BC, when a functioning supply network seems to have existed, serving a vast area between the British Isles, southern Scandinavia, the Carpathian Basin and Tuscany (Fig. 1.14). Indeed, good evidence for a long­distance trade in tin, with the early Mycenean Aegean, is first encountered in the 17th century BC on the island of Vivara. As we have suggested, the crucial change in the late 3rd mill ennium BC involved a new relationship to metals, particularly noticeable in Central and north­west Europe. This involved both a massive increase in the supply of copper (and its commodification as Osenring ingots), and the beginning of a long­distance trade in tin, for the production of bronze. It is not clear to me why this should have happened precisely at this time. However, Sabine Gerloff and Andrew Sherratt have recently drawn attention to the signs of contact between the East Mediterranean and Central Europe (faience beads, the slitted spearhead from Kyhna, and Schleifennadeln in Europe, Osenring ingots from Ugarit, Byblos etc.), and the same horizon of 'influences' has traditionally been held responsible for changes at the start of the Bronze Age (cf. Montelius, Reinecke, Dechelette, Childe etc.) . It is not inconceivable that new ideas about metallic currencies spread to Europe at this time from the Near East (for tin and silver, see Stech & Pigott 1986). Acknowledgements As a novice in the field of early bronze metallurgy, I have relied heavily on the generous support of a number of colleagues. I am particularly grateful for information and advice from the following: Lawrence Barfield (Birmingham), Paul Garwood (Birmingham), Claudio Giardino (Rome), Richard [ones (Glasgow), Maria Kayafa (Birmingham), Veronica Liritzis­Maxwell (Glasgow), Iosef Maran (Heidelberg), Ignacio Montero (Madrid), Michael MullerKarpe (Mainz), Peter Northover (Oxford), Mark Pearce (Nottingham), Graham Philip (Durham) and Sophie Stos­Gale (Oxford). Notes 1. In writing these introductory paragraphs, I have drawn heavily on publications by Ernst Pernicka: see for example Pernicka 1987; 1990; 1998; Pernicka et al. 1993; 1997 . In a slightly earlier publication, Pernicka (1990: 52) suggested that a true tin bronze technology was adopted in Mesopotamia about 100­200 years earlier than in north­west Anatolia, but according to the latest 14C dating evidence (Korfmann & Kromer 1993: 168, fig. 23), Troy IIg should be roughly contemporary with the start of regular use of tin bronze in Mesopotamia . 3. Veronica Liritzis­Maxwell has kindly drawn my attention to a forthcoming study of important discoveries from Giali (Nissyros) : V. M. Liritzis Maxwell & A. Sampson, forthcoming. 'Tin at the end of the Neolithic: the case of Giali', in P. Northover & C. Slater (eds), Founders, Smiths and Platers (in press). 4. Apart from the examples from Aghios Kosmas, Levkas (R7), Lithares and Manika, the other tin bronzes included in Table 1.1 and Fig. 1.4 are somewhat unreliable: According to Veronica Liritzis­Maxwell (pers. comm .) the chr onology of the axe from Levkas grave R9 (7.8% Sn) is somewhat uncertain. For Levkas, see the forthcoming article by V. M. Liritzis Maxwell & N. H. Gale, 'The Ionian enigma: the metalwork from Levkas reconsidered', Annual of the British School at Athens, in press (2000). ­ According to J. Maran (1998: 264, note 1069) the chisel from Sesklo with 2.24% Sn does not definitely belong to EH Ill; the chronology of the tweezers from Aghios Kosmas is also uncertain (ibid. 268). ­ The chisel from the Tharounia Cave (13.09% Sn) is presumably heavily mineralised, as the Atomic Absorption analysis recorded only 50.28% Cu. ­ As for the chisel from the Petr alona District, with 7.85% Sn (Mangou & Ioannou 2. BRONZE AND THE BRONZE AGE 5. 6. 7. 8. 9. 10. 11. 1999: 89, table 2), it is not clear from the publication whether its Early Helladic date is secure. ­ Early tin bronzes have also been reported from two other sites in Macedonia, but the early analyses were non­quantitative: Axiochori (Vardar6phtsa) : Davies 1927; Perivolaki (Saratse) : Heurtley 1930: 144. The nine analysed objects come from Krasi, Mochlos, Palaikastro and Platanos. These were not included in the 'corpus' of analyses published by K. Branigan (1974: 147 H., Appendix I), and have not been included in my summary of tin concentrations. 68% of the 22 objects analysed by Gale and Stos­Gale (1985: 61) and 58% of the 106 objects (not including waste and copper wire) analysed by Catling and [ones (1977). Chernykh (1978) has published the following data for shaft­hole axes of types T.l0, T.14, T.18 and T.22. ­ no: four analyses, including one with more than 5% Sn. T.14: eight analyses, including three above 5% Sn. T.18: eight analyses, including three above 5% Sn , T.22: three analyses, all between 9% and 13% Sn . For a discussion of the analyses of the hoards of Crica, Kozarac, Vinkovci, Vranovici, Donja Lohinja and Dunakornlod (average 0.001% Sn), see Liversage 1994: 80; 81, table xv ii; 95 f.; 96, fig . 27. Ibid. 80: "If they are representative these 67 analyses show that at the time of the Vuc edol z Zok culture bronze was essentially unknown in the Carpathian Basin, and even the slightest contamination with tin was very rare." ­ Only one tin bronze Kozarac axe is known to me, from Staro Selo near [erinin Grad in Serbia, with 1.82% Sn (Pernicka et al. 1993: 48, pl , 11,4; Primas 1996: 104). Interestingly, the axe is almost identical to the silver shaft­hole axe from the rich tumulus grave at Mala Gruda (Vucedol culture). In the western Balkans, tin bronze shaft­hole axes first appear in the Early Bronze Age Cetina culture (Citluk, 'Vel ike Gromile' and Vedrine, 'Rareva Comila', see Zeravica 1993: 26, nos 82­83; 30; 128). Fig . 1.9 only shows the Early and Middle Bronze Age histograms. ­ The chronological terminology used here is based on the 'pin chronology' devised by W. Ruckdeschel (1978). Br Alb corresponds to early Unetice. Br A2a to classic Unetice. Br A2b to the Apa­Hajdiisamson horizon; hoards of Buhl­Niederosterwitz / Ackenbach­Buhl type are assigned to the the end of the Early Bronze Age or the Early /MiddJe Bronze Age transition (Br A /B, sometimes referred to as 'Br A2c' or 'Br A3'). E. Schalk (1998: 203 H.) argues for a dynamic local metallurgy in the Hernad valley, using metals, including tin, from the Slovakian Ore Mountains. In the earliest phase of the cemetery of Hernadkak (phase 1, roughly Br Ala), there is one analysis demonstrating the use of copper alloyed with 11.4% Sn . However, a standard tin bronze was never achieved during the life of this cemetery: in phase 3, the average tin content is 3.82% Sn, with only three of the 13 analyses containing more than 4% Sn . For Hernadkak, see idem 1992. ­ Note, however, that the conventional chronology of the Nitra group, and Schalk's early chronology for the Kost'any group, has recently been questioned by Slawomir Kadrow (1997). Using "C dates, Kadrow suggests a date for earl y Nitra 11 and Mokrin, a horizon characterised by widespread cultural contacts (e.g. faience beads), in the first half of the 20th century BC; early Kost'any (Kosice I) and later Nitra 11 would then follow between ea. 1950/1900 and 1800 BC. See also Spindler (1971: 217 f.): " Auffa llig ist aber, daB 33 bei relativ gut belegter Gleichzeitigkeit von Glockenbechern und Schnurkeramik einerseits und von Vucedol und Laibach andererseits die mitteleuropaischen Cruppen mehr Zinnbronzen aufweisen als di e sudosteuropaischen." 12. The terminology for these phases is somewhat confusing: Rassmann's Per . lA (north­east Germany) is contemporary with Vandkilde's Late Neolithic II (Denmark); Rassmann's Per. lE corresponds to Vandkilde's Per. lA. ­ See Rassmann 1996: 200. 13. Unfortunately, the author was not able to obtain a copy of an important publication of research results on early metallurgy in the Iberian Peninsula: Delibes, G. & Montero, I. (eds) 1999. Las primeras etapas meialurgicas en la Peninsula lberi ca. H. Estudios regionales. Madrid, Instituto Unversitario Ortega y Gasset. 14. The following paragraphs emphasising the importance of the scrap bronze trade owe much to the stimulating article by Sue Sherratt (this volume). Bibliography Alcalde, G., Molist, M., Montero, I., Planaguma, L., Safia, M. & Toledo, A. 1998. 'Producciones metalurgicas en el nordeste de la Peninsula Iberica durante el III milenio cal. AC: el taller de la Bauma del Serrat del Pont (Tortella, Cirona)', Trabajos de Prehistoria 55 (1), 81­100. Bar­Aden, P. 1980. The Cave of the Treasure. Thefind s from the caves in Nahal Mishmar. Jerusalem, Israel Exploration Society. Bartelheim, M., Niederschlag. E. & Rehren, T. 1998. ' Research into prehistoric metallurgy in the Bohemian/Saxon Erzgebirge', in B. Hansel (ed.), Mensch und Umwelt in der Bronzezeit Europas. Kiel, Oetker­Voges Verlag, 225­229. Begemann, F., Pernicka, E. & Schrnitt­Strecker, S. 1994. 'Metal finds from Ihpinar and the advent of arsenical copper', Anatolica 20, 203­219. Begemann, F., Schmitt­Strecker, S. & Pernicka, E. 1992. 'The metal finds from Thermi Ill­V: a chemical and leadisotope study', Studia Troica 2, 219­239. Bertemes, F. 1989 . Das [r iihbronzezeitliche Grdberfeld von Gemeinlebarn (Saarbrucker Beitrage zur Altertumskunde 45). Bonn . Bona, I. 1992. 'Bronzezeitliche Tell­Kulturen in Ungarn'. in W. Maier­Arendt (ed.), Bronzezeit in Ungarn. Forschungen in Tell-Siedlungen an Donau und Theiss. Frankfurt a. M., Museum fUr Vor­ und Frtihgeschichte, 9­41. Braidwood, R. J. & Braidwood, L. S. 1960. Excavations in the plain of Antioch. I The earlier assemblages, phases A-J (The University of Chicago Oriental Institute Publications 61). Chicago, University of Chicago Press. Branigan, K. 1974. Aegean metalwork of the Early and Middle Bronze Age. Oxford, Clarendon Press. Budd, P., Gale, D., Pollard, A. M., Thomas, R. G. & WiIliams, P. A. 1992. ' The early development of metallurgy in the British Isles ', Antiquity 66, 677­686. Caley, E. R. 1949. 'On the prehistoric use of arsenical copper in the Aegean region', in Commemorative Studies in Honor of Theodore Leslie Shear (Hesperia. Supplement 8), 60­63. Campana, N ., Franceschi, E., Maggi, R. & Stos­Cale, Z. 1996. 'Crotticella sepolcrale di Val Frascarese (Genova) : nuove analisi dei reperti metallici', in D. Cocchi Genick (ed .), L'antica eta del bronzo. Atti del Congresso di Viareggio, 1995. Florence, OCTAVO Franco Cantini Editore, 556­557. Catling, H. W. & [ones, R. E. 1976. 'Sellop ou lo tomb 4: some analyses', Annual of the British School at Athens 71, 21­23. 34 CHRISTOPHER PARE Catling, H . W. & [ones, R. E. 1977. 'Analyses of copper and bronze artefacts from the Unexplored Mansion, Knossos', Archaeometry 19, 57­66. Charles, J. A. 1980. 'The coming of copper and copper­based alloys and iron: a metallurgical sequence', in T. A. Wertime & J. D. Muhly (eds), The coming of the Age of Iron. New Haven­London, Yale University Press, 151­183. Chernykh, E. N. 1977. 'Ob evr opeyskoy zone circumpontiyskoy metallurgicheskoy provintsii (CMP)', Acta Archaeologica Carpathica 17, 29­53. Chernykh, E. N. 1978 . Gornoe delo i metallurgiya v drevneyshe1j Bolgarii. Sofia. Chernykh, E. N. 1992. Ancient metallurgy in the USSR. Cambridge, Cambridge University Press. Cleuziou, S. & Berthoud, T. 1982. 'Early tin in the Near East. A reassessment in the light of new evidence from western Afghanistan', Expedition 25 (1), 14­19. Close­Brooks, J. & Coles, J. M. 1980. 'Tinn ed axes', Antiquity 54, 228­229. Coles, J. M. & Herding, A. F. 1979. The Bronze Age in Europe. London, Methuen & Co . Comendador Rey, B. 1999. 'The early development of metallurgy in the north­west of the Iberian peninsula', in S. M. M. Young, A. M. Pollard, P. Budd & R. A. Ixer (eds), Metals in Antiquity (British Archaeological Reports, International Ser. 792), 63­67. David, W. 1998. 'Zu m Ende der bronzezeitlichen Tellsiedlungen im Karpatenbecken'. in H . Kuster, A. Lang & P. Schauer (eds), Archiiologische Forschungen in urgeschicht- lichen Siedlungslandschaften. Festschrift fur Georg Kossack zum 75. Geburtstag (Regensburger Beitrage zur pr ahistorischen Archaologie 5). Bonn , Dr . Rudolf Habelt GmbH, 231 H. Davies, 0 .1927. 'Chemical analyses of metal remains', Annual of the British School at Ath ens 28, 1926­27, 195­198. Dickinson, O. T. P. K. 1977. The origins of Mycenaean civilisation (Studies in Mediterranean Archaeology 49). Coteborg, Paul Astr6ms F6rlag. Durrnan, A. 1997. 'Tin in southeastern Europe?', Opuscula Archaeologica 21, 7­13. Eaten. E. R. 1991. 'Early Metallurgy in Italy', in W. A. Oddy (ed.), Aspects of Early Metallurgy (British Museum Occasional Paper 17). London, British Museum, 159­168. Eaton, E. R. & McKerrel, H. 1976. 'Near Eastern alloying and some textual ev id en ce for the early use of arsen ical copper', World Archaeology 8 (2), 169­191. Fernandez­Miranda, M., Fernandez­Posse, M. D., Martin, C, Montero, I. & Rovira, S. 1996. 'Cha nges in Bronze Age metallurgy as depicted by laboratory analysis: th e "La Mancha" model, Spain' , in S, Dernirci, A. M. Ozer & G. D. Summers (eds), Archaeometry 94. The Proceedings of the 29th International Symposium on Archaeometry, Ankara 914 May 1994. Ankara, Tiibitak, 23­34. Fernandez­Miranda, M., Montero Ruiz, I. & Rovira Llorens, S. 1995. ' Los primos objetos de bronce en el occidente de Europa, Trabajos de Prehistoria 52 (1), 57­69 . Forenbaher. S. 1993. 'Radiocarbon dates and absolute chronology of the central European Early Bronze Age', Antiquity 67, 218­220; 235­256. Frangipane, M. 1985. 'Early developments of metallurgy in the Near East', in M. Liverani, A. Palmieri & R. Peroni (eds), Studi di Pa/tenologia in onore di Salvatore M. Puglisi. Rome, Universita di Roma " La Sap ienza", 215­228 . French, D. H. 1962. 'Excavations at Can Hasan. First preliminary report, 1961', Anatolian Studies 12,27­40. Gale, N. & Stos ­Gale, S. 1985. 'Cyprus and the Bronze Age metals trade', in T. Papadopoullou & S. A. Hatzistulli (eds), Praktika tou deuierou Diethnous Kupriologikou Sunedriou, Tomos A, 51­66. Gale , N . H. & Stos­Cale, Z. A. 1989. 'Some aspects ofCypriote metallurgy in the Middle and Late Bronze Age', in R. Laffineur (ed.), Transition. Le monde egeen du Bronze moyen au Bronze recent. Actes de la deuxieme rencontre egeenne interntionale, Liege 1988 (Aegaeum 3), 251­256. Gale, N. H ., Stos­Cale, Z. A. & Gilmore, G . R. 1985. 'Alloy types and copper sources of Anatolian copper alloy artifacts', AnatoIian Studies 35, 143­173. Gale, N . H ., Stos­Cale, Z . A., Lilov, P., Dirnitrov, M. & Todorov, T. 1991. 'Recent studies of en eolithic copper ores and artefacts in Bulgaria', in J.­P. Mohen & C. Eluere (eds) , Decouverte du Metal (Amis du Musee des Antiquites Nationales, Millenaires dossier 2). Paris, Picard Editeur, 49­75. Gallay, A. & Lahouze, M.­N. 1976. 'Pour une prehistoire de la metallurgie (Europe, Proche­Orient)', Archives suisses d'anthropologie generate 40 (2), 137­200. Cerloff, S. 1993. 'Zu Fragen mittelrneerland ischer Kontakte und absoluter Chronologie der Friihbronzezeit in Mittelund Westeuropa', Priihistorische Zeitschrift 68, 58­102. GerloH, S. 1996. ' Wessex, Mycenae and related matters: the chronology of the British Bronze Age in its European setting', in C. Belardelli & R. Peroni (eds), The BronzeAge in Europe and the Mediterranean (International Union of Prehistoric and Protohistoric Sciences, Forli 1996, Colloquia, vol. 11). Forli, ABACO Edizioni, 11­19. Giardino, C. 1997. ' La metallotecnica nella Sicilia preprotostorica', in S. Tusa (ed .), Prima Sicilia, alle origini dell societa siciliana. Palerrno, Ediprint, 405­414 . Giardino, C. 1998. 'Tyrrhen ian Italy and Sicily in the protohistoric metal trad e across the Mediterranean: an archaeometallurgical outline', in C. Mordant, M. Pemot & V. Rychner (eds), l /Atelier du bronzier en Europe du XX' au VIII' siecle avant notre ere. Actes du colloque international Bronze '96, Neuchtuel et Dijon. 1I: Du minerai au metal, du metal ii l'obiet. Paris, CTHS, 157­167. Coldman, H . 1931. Excavations at Eutresis in Boeotia. Cambridge Mass., Harvard University Press. Creeves, T. A. P. 1982: 'Review of E. N. Chernykh, Gomoe delo i metallurgiya v drevneyshey Bolgarii'. Proceedings of the Prehistoric Society 48, 538­542. Hall, M. E. & Steadrnan, S. R. 1991. 'Tin and Anatolia : Another Look', Journal of Mediterranean Archaeology 4 (1),217­234. Harrison, R. J. 1974 . 'A recons ideration of the Iberian background to Beaker metallurgy', Palaeohistoria 16, 63105. Heurtley, W. A. 1930. ' Rep o rt on excavations at the Toumba of Saratse, Macedonia, 1929', Ann ual of the British School at Athens 30, 1928­30, 113­150. Heurtley, W. A. 1939. Prehistoric Macedonia. Cambridge, Cambridge University Press. Hook, D. R., Arribas Palau, A., Craddock, P. T., Molina, F. & Rothenberg, B. 1987. 'Copper and silver in Bronze Age Spain', in W. H. Waldren & R. C. Kennard (eds); BellBeakers of the Western Mediterranean (British Archaeological Reports, International Ser . 331). Oxford, BAR, 147­172. Hundt, H . J. 1974. 'Donaulandische Einflusse in der friihen Bronzezeit Norditaliens', Preistoria Alpina 10, 143­178. [ones, R. E. 1979. ' Ap p en d ix 6: Two metal objects from Servia', in C. Ridley & K. A. Wardle, 'Rescue Excavations at Servia 1971­1973: a preliminary report', Annual of the British School at Athens 74, 229­230. [ovanovic, B. 1988 . ' Ea rly metallurgy in Yugoslavia', in R. Maddin (ed.), The Beginning of the Use of Metals and Alloys. Papers from the Second International Conference on the Beginning of the Use of Metals and Alloys, Zhengzhou, China, BRONZE AND THE BRONZE ACE 21-26 October 1986 . Cambridge, Mass .­ London, MIT Press, 69­79. [unghans, S., Sangmeister, E. & Schroder, M. 1960. Metal- lanalysen kupferzeitlicher undjrunbronzezeitticner Bodenfunde aus Europa (Studien zu den Anfangen der Metallurgie I). Berlin, Mann. [unghans, S., Sangmeister, E. & Schroder, M. 1968. Kupfer und Bronze in der jruhen Metallzeit Europas: die Materialgruppen beim Stand von 12000 Analysen (Studien zu den Anfangen der Metallurgie II/l­3). Berlin, Mann. [unghans, S., Sangmeister, E. & Schroder, M. 1974. Kupfer und Bronze in der [riihen Metallzeit Europas: Katalog der Analysen Nr. 10041-22000, mit Nachuntersuchungen der Analysen Nr. 1-10040 (Studien zu den Anfangen der Metallurgie II/4) . Berlin, Mann. Kadrow, S. 1997. ' Bem erku ngen zur Chronologie der Friihbronzezeit in Kleinpolen und in der Slowakei', in e. Becker, M.-L. Dunkelmann, e. Metzner-Nebelsick, H. Peter-Rocher, M. Roeder & B. Terzan (eds), Xpovoc, Beitriige zur prd- historischen Archiiologie zwischen Nord- und Siidosieuropa (Internationale Archaologie. Studia honoraria 1). Espelkarnp, Verlag Marie Leidorf, 229-235. Kayafa, M. 1999: Bronze Age metallurgy in the Peloponnese, Greece (Unpublished PhD thesis, Dept. of Ancient History and Archaeology, University of Birmingham). Kinnes.T, A., Craddock, P. T., Needharn, S. & Lang, J. 1979. 'Tin-plating in the Early Bronze Age : the Barton Stacey axe', Antiquity 53, 141-143. Tepe, Korfrnann, M. & Krorner, B. 1993. 'Dernircihuyuk, ｂ･ｾｩｫ Troia - eine Zwischenbilanz zur Chronologie dreier Orte in Westanatolien', Studia Troica 3, 135-171. Kristiansen, K. 1987. 'From stone to bronze - the evolution of social complexity in Northern Europe, 2300-1200 BC, in E. M. Brumfiel & T. K. Earle (eds), Specialization, Exchange, and Complex Societies. Cambridge, Cambridge University Press, 30-51. Kuna, M. & Matousek, V. 1978. 'Medena industrie kultury zvoncovitych poharu ve stredni Evrope', in J. Filip (ed.), Varia Archaeologica 1 (Praehistorica VII) . Prague, Univerzita Karlova, 65-89. Lenerz-de Wilde, M. 1995. 'Prarnonetare Zahlungsmittel in der Kupfer- und Bronzezeit Mittelleuropas', Fundberichte aus Baden-wurttemberg 20, 229-327. Liversage, D. 1994. 'Interpreting composition patterns in ancient bronze: the Carpathian Basin', Acta Archaeologica (Kobenhavn) 65,57-134. Maddin, R, Muhly, J. D. & Wheeler, T. 1980. 'Distinguishing artefacts of native copper', Journal of Archaeological Science 7,211-225. Maggi, R & Forrnicola, V. 1978. 'Una grotticella sepoIcrale dell'inizio dell'eta del bronzo in Val Frascarese (Cenova)', Preistoria Alpina 14, 87-113. Mangou, H. & Ioannou, P. V. 1997. 'On the chemical composition of prehistoric Greek copper-based artefacts from the Aegean region', Annual of the British School at Athens 92, 59-72. Mangou, H. & Ioannou, P. V. 1998. 'On the chemical composition of prehistoric Greek copper-based artefacts from Crete', Annual of the British School at Athens 93, 91-102. Mangou, H . & Ioannou, P. V. 1999. 'On the chemical composition of prehistoric Greek copper-based artefacts from mainland Greece', Annual of the British School at Athens 94, 81-100. Manning, S. W. 1995. The Absolute Chrnology of the Aegean Early Bronze Age: archaeology, history and radiocarbon (Monographs in Mediterranean Archaeology 1). Sheffield, Sheffield Academic Press. 35 Maran, J. 1989. 'Die Schaftlochaxt aus dem Depotfund von Theben (Mittelgriechenland) und ihre Stellung im Rahmen der bronzezeitlichen Axte Sudosteuropas'. Archiiologisches Korrespondenzblatt 19, 129-136. Maran, J. 1998. Kulturwandel auf dem griechischen Festland und den Kykladen im spdten 3. Jahrtausend v. Chr. (Universitatsforschungen zur prahistorischen Archaologie 53). Bonn, Rudolf Habelt Verlag. Martin, C; Fernandez-Miranda, M., Fernandez-Posse, M. D. & Gilman, A. 1993. ' The Bronze Age of La Mancha', Antiquity 67, 23-45. McGeehan-Liritzis, V. 1996. The role and development of metallurgyin the Late Neolithic and Early Bronze AgeofGreece (Studies in Mediterranean Archaeology and Literature, Pocket-Book 122). [onsered, Paul Astroms Forlag. McGeehan-Liritzis, V. & Gale, N. H. 1988. 'Chemical and Lead Isotope analyses of Greek Late Neolithic and Early Bronze Age metals', Archaeometry 30 (2), 199-225. McGeehan-Liritzis, V. & Taylor, J. W. 1987. ' Yugoslavian tin deposits and the Early Bronze Age industries of the Aegean region', Oxford Journal of Archaeology 6, 287-300. McKerrel, H. 1978. 'The Use of Tin-Bronze in Britain and the Comparative Relationship with the Near East', in A. D. Franklin, J. S. Olin & T. A. Wertime (eds), The Search for Ancient Tin. Washington D.e., Smithsonian Institute Press, 7-24. Merpert, N . Y. & Munchaev, R. M. 1987. 'The earliest levels at Yarim Tepe I and Yarim Tepe n in northern Iraq', Iraq 49, 1-36. Montero Ruiz, I. 1993. 'Bronze Age metallurgy in southeast Spain', Antiquity 67, 46-57. Moorey, P. R. S. 1994. Ancient Mesopotamian Materials and Industries. Oxford. Moslem. S. & Rieder, K. H. 1997. 'Zinnperlen aus einem fruhbronzezeitlichen Grab von Buxheim, Landkreis Eichstatt, Oberbayern', Das archiiologische [ahr in Bayern 1997, 68-70. Muckelroy, K. 1980. 'Two bronze age cargoes in British waters', Antiquity 54, 100-109. Muckelroy, K. 1981. 'Middle Bronze Age trade between Britain and Europe: a maritime perspective', Proceedings of the Prehistoric Society 47, 275-297. Muhly, J. D. 1985. 'Sources of tin and the beginnings of bronze metallurgy', American Journal of Archaeology 89, 275-291 . Muhly, J. D. 1988. 'The Beginnings of Metallurgy in the Old World', in R. Maddin (ed.), The Beginning of the Use of Metals and Alloys. Papers from the Second International Conference on the Beginning of the Use of Metals and Alloys, Zhengzhou, China, 21-26 October 1986. Cambridge, Mass.London, MIT Press, 2-20. Muhly, J. D. 1989. 'Cayonu Tepesi and the beginnings of metallurgy in the Ancient World', in A. Hauptrnann, E. Pernicka & G. A. Wagner (eds), Old World Archaeo- metallurgy. Proceedings of the International Symposium, Heidelberg 1987 (Der Anschnitt, Beiheft 7). Bochurn, Deutsches Bergbau-Museum, 1-11. Muhly, J. D. 1991. 'Copper in Cyprus: the earliest phase', in J.-P . Mohen & e. Eluere (eds), Decouuerte du Metal (Amis du Musee des Antiquites Nationales, Millenaires 2). Paris, Picard, 357-374. Muhly, J. D. 1993. ' Ea rly Bronze Age Tin and the Taurus', American Journal of Archaeology 97, 239-253. Muhly,J. D. & Pernicka, E. 1992. 'Early Trojan Metallurgy and Metals Trade', in J. Herrmann (ed.), Heinrich Schliemann . Grundlagen und Ergebnisse moderner Archiiologie 100 [ahre nach Schliemanns Tod. Berlin, Akademie Verlag, 309-318. Milller-Karpe, M. 1989. 'Neue Forschungen zur fruhen 36 CHRISTOPHER PARE Metallverarbeitung in Mesopotamien', [ahrbuch des RbmischGermanischen Zeniralmuseums, Mainz 36, 179­192. Mylonas, G. E. 1932. 'Preistoriki Eleusis', in K. Kourouniotis (ed.), Eleusiniaka, 146-147. Mylonas, G. E. 1959. Aghios Kosmas, an Early Bronze Age settlement and cemetery in Attica. Princeton N . J., Princeton University Press. Needham, S. 1996. 'Chronology and periodisation in the British Bronze Age', in K. Randsborg (ed.), Absolut e Chronology. Archaeological Europe 2500-500 BC (Acta Archaeologica Supplementa 1), 121­140. Needham, S. &0Kinnes, I. 1981. 'Tinned axes again', Antiquity 55, 133­­134. Needham, S. P., Leese, M. N ., Hook, D. R. & Hughes, M. J. 1989. 'Developments in the Early Bronze Age metallurgy of southern Britain', World Archaeology 20 (3),383­­402 . Northover, J. P. 1982. 'Th e exploration and long distance movement of bronze in Bronze Age and Early Iron Age Europe', Bulletin of the Institute of Archaeology 19,45­72. Northover, J. P. 1988. 'Alloy design in the Bronze Age', in J. Ellis Evans (ed .), Aspects of ancient mining and metallurgy. Acta of a British School at Athens conference at Bangor, 1986. Bangor, Univ. College of North Wales, 44­54. Northover, J. P. 1989. 'Properties and use of arsenic­copper alloys', in A. Hauptmann, E. Pernicka & G. A Wagner (eds), Old World Archaeometallurgy. Proceedings of the International Symposium, Heidelberg 1987 (Der Anschnitt, Beiheft 7). Bochurn, Deutsches Bergbau ­Museum, 111­118 . Northover, J. P. & Rychner, V. 1998. ' Bron ze analysis: experience of a comparative programme', in e. Mordant, M. Pernot & V. Rychner (eds), L'Atelier du bronzier en Europe du XX' au VIII' siecle avant notreere. Actes du colloque international Bronze '96, Neuchdtel et Dijon. I: Les analyses de composition du metal: leur apport a I'Archiologie de rAge du Bronze. Paris, CTHS, 19­40. O'Shea, J. 1992. ' A radiocarbon­based chronology for the Maros Group of southeast Hungary', Antiquity 66, 97­102. Ottaway, B. S. 1989. 'Interactions of some of the earliest copperusing cultures in Central Europe', in A. Hauptmann, E. Pernicka & G. A. Wagner (eds), Old World Archaeo- metallurgy. Proceedings of the International Symposium, Heidelberg 1987 (Der Anschnitt, Beiheft 7). Bochurn, Deutsches Bergbau­Museurn, 19­23. Otto, H. 1953. 'Analysen der Metallbeigaben aus schnurkeramischen Crabern von Niederkaina, Kreis Bautzen', Arbeits- und Forschungsberichte zur siichsischen Bodendenkmalpflege 3, 40­47. Otto, H. & Witter, W. 1952. Handbuch deriiltesten vorgeschichtlichen Metallurgie in Mitteleuropa. Leipzig, [ohann Arnbrosius Barth Verlag. Pare, e. F. E. 1996. 'Settlemen t systems and social organisation in Bronze Age Central Europe', in e. Belardelli, J.W. Neugebauer, M. Novotna, B. Novotny, e. F. E. Pare & R. Peroni (eds), The Bronze Age in Europe and the Medi- terranean. Proceedings of the 13th Int. Congr. of Prehist. and Protohist. Sciences, Forli 1996. Forll, A.B.Ae.O. Edizioni, 223­­232. Pare, e. F. E. 1998. 'Beitrage zum Ubergang von der Bronzezur Eisenzeit in Mitteleuropa. Teil I: Grundzilge der Chronologie im ostlichen Mitteleuropa (11.­­8. [ahrhundert v. Chr.)', [ahrbuch des Ramisch-Germanischen Zeniralmuseums, Mainz 45, 293­­443. Pare, e. F. E. 1999. 'Weigh ts and weighing in Bronze Age Central Europe', in Eliten in der Bronzezeit. Ergebnisse zweier Kolloquien in Mainz und Athen (Monographien des Rornisch­Gerrnanischen Zentralmuseums 43). Mainz, RGZM , 421­514. Penhallurick, R. D. 1986. Tin in Antiquity, its mining and trade throughout the ancient world with particular reference to Cornwall . London, The Institute of Metals. Pernicka, E. 1987. 'Erzlagerstatten in der Agais und ihre Ausbeutung im Altertum: geochemische Untersuchungen zur Herkunftsbestimmung archaologischer Metallobjekte', [ahrbucb des Riimisch-Germanischen Zentralmuseums, Mainz 34, 607­714. Pemicka, E. 1990. 'Gewinnung und Verbreitung der Metalle in prahistorischer Zeit', [ahrbuch des Riimisch-Germaniscnen Zeniralmuseums, Mainz 37, 21­129. Pernicka, E. 1998. 'Die Ausbreitung der Zinnbronze im 3. [ahrtausend. In B. Hansel (ed.), Mensch und Umwelt in der Bronzezeit Europas. Kiel, Oetker­Voges Verlag, 135­147. Pernicka, E., Begemann, F., Schmitt­Strecker, S. & Grimanis, A. P. 1990. 'On the composition and provenance of metal artefacts from Poliochni on Lemnos', Oxford Journal of Archaeology 9(3), 263­­298. Pernicka, E., Begemann, F., Schmitt­Strecker, S., Todorova, H. & Kuleff, I. 1997. 'Prehistoric copper in Bulgaria, its composition and provenance', Eurasia Antiquo 3, 41­178. Pernicka, E., Begemann, F., Schmitt­Strecker, S. & Wagner, G. A. 1993. 'Eneolithic and Early Bronze Age copper artefacts from the Balkans and their relation to Serbian copper ores', Praehistorische Zeitschrift 68, 1­54. Pernicka, E., Seeliger, T. c, Wagner, G. A, Begemann, F., Schmitt­Strecker, S., Eibner, e., Oztunali, O. & Baranyi, I. 1984. 'Archaometallurgische Untersuchungen in Nordwestanatolien', [ahrbuch des Riimisch-Cermanischen Zentralmuseums, Mainz 31, 533­599. Pernicka, E., Wagner, G. A, Muhly, J. D. & Oztunali, O. 1992. 'Comment on the Discussion of Ancient Tin Sources in Anatolia', Journal of Mediterranean Archaeology 5 (1), 91­98. Petrikaki, M. 1980. 'Leipsana protoelladikou oiki smou sto Rouf", Archaiologikon Deltion 35 (A), 147-174. Phelps, W. W., Varoufakis, G. J. & [ones, R. E. 1979. ' Five copper axes from Greece', Annual of the British School at Athens 74, 175­184. Philip, G. 1991. 'Tin, Arsenic, Lead : Alloying Practices in Syria­Palestine around 2000 B.C.', Levant 23, 93­­104. Primas, M. 1985. 'Tin objects in Bronze Age Europe', in M. Liverani, A Palmieri & R. Peroni (eds), Studi di Paletnologia in onore di Salvatore M. Puglisi. Rome, Universita di Roma "La Sapienza", 555­562. Primas, M. 1996. Velika Gruda I (Universitatsforschungen zur prahistorischen Archaologie 32). Bonn, Rudolf Habelt Verlag. Primas, M. 1997. 'Bronze Age economy and ideology: Central Europe in focus', Journal of European Archaeology 5 (1), 115­130. Prim as, M. & Pernicka, E. 1998. ' Der Depotfund von Oberwilflingen. Neue Ergebnisse zur Zirkulation von Metallbarren', Germania 76, 25­65. Raczky, P., Hertelendi, E. & Horvath, F. 1992. 'Zur absoluten Datierung der bronzezeitlichen Tell­Kulturen in Ungarn', in W. Maier­A rend t (ed .), Bronzezeit in Ungarn. Forschungen in Tell-Siedlungen an Donau und Theiss. Frankfurt a. M., Museum filr Vor­ und Frtihgeschichte, 42­47. Rassmann, K. 1993. Spdineolithikum und friihe Bronzezeit im Flachland zwischen Elbe und Oder (Beitrage zur Ur­ und Frilhgeschichte Mecklenburg­Vorpommerns 28). Rassmann, K. 1996. 'Zu m Forschungsstand der absoluten Chronologie der frilhen Bronzezeit in Mitteleuropa auf der Grundlage von Radiokarbondaten', in K. Randsborg (ed.), Absolute Chronology. Archaeological Europe 2500-500 BC (Acta Archaeologica Supplementa 1), 199­209. BRONZE AND THE BRONZE AGE Roden, C. 1985. 'Montanarchaologische Quellen des ur­ und friihgeschichtlichen Zinnbergbaus in Europa ­ Ein Uberblick', Der Ans chnill 37, 5G-80. Rosenfeld, A., Ilani , S. & Dvorachek, M. 1997. ' Bronze Alloys from Canaan During the Middle Bronze Age', Journal of Archaeological Science 24, 857-864. Ruckdeschel, W. 1978. Die frubbronzezeitlicnen Griiber Sudbayerns. Ein Beitrag zur Kenninis der Straubinger Kuliur (Antiquitas 2/11). Bonn, Rudolf Habelt Verlag. Ryndina, N. V. & Yakhontova, L. K. 1985. 'The earliest copper artifact from Mesopotamia', Sovetskaya Archeologiya 1985 (2), 155-165. Sarnpson, A. 1985. Manika, an Early Helladic town in Chalkis. Volume 1. Schaeffer, C. F.-A. 1949. 'Porteurs de torques', Ugaritica 2, 49-120. Schaeffe r, C. F.-A. 1978. 'Ex occidente ars', Ugaritica 7, 475552. Schalk , E. 1992. Das Griiberfeld von Herruidkak (Universitatsforschungen zur prahistorischen Archaologie 9). Bonn, Rudolf Habelt Verlag. Schalk , E. 1998. The development in metal composition and typological forms in the Carpathian Basin during the Early Bronze Age: a case study', in C. Mordant, M. Pernot & V. Rychner (eds). L'Atelier du bronzier en Europe du XX' au Vll!' siecle auant notre ere. Actes du colloque international Bronze '96, Neuchtuel et Dijon. 1: Les analyses de composition du metal: leur apport al'Archeologie de l'Age du Bronze. Paris, CTHS, 203-214. Schickle r, H. 1981. "Neolithische' Zinnbronzen', in H. Lorenz (ed .), Studien zur Bronzezeit. Festschrift fur Wilhelm Albert v. Brunn. Mainz, Philipp von Zabern, 419-445. Schmidt, B. & Nitzschke, W. 1980. 'Ein friihbronzezeitlicher "Furstenhugel" bei Dieskau im Saalkreis. Vorbericht', Au sgrabungen und Funde 25,179-183. Shennan, S. 1993. 'Commodities, transactions, and grow th in the central-European Early Bronze Age ', Journal of European Archaeology 1 (2), 59-72. Sherratt, A. 1993. ' What would a Bronze-Age World System look like? Relations between temperate Europe and the Mediterranean in Later Prehistory', Journal of European Archaeology 1 (2), 1-57. Sherratt, A. 1994. 'Core, Periphery and Margin: Perspectives on the Bronze Age', in C. Mathers & S. Stoddart (eds), Development and Decline in the Mediterranean Bronze Age (Sheffield Archaeological Monographs 8). Sheffield, J. R. Collis Publications, 335-345. Slater, E. A. 1972. 'Appendix IX. Electron Microprobe metal analysis', in P. Warren, Myrtos, an Early Bronze Age settlement in Crete. London, Thames and Hudson, 329. Snodgrass, A. 1980. 'Iron and Early Metallurgy in the Mediterranean', in T. A. Wertime & J. D. Muhly (eds), The Coming of the Age of Iron. New Haven-London, Yale University Press, 335-374 . Spindler, K. 1971. 'Zur Herstellung der Zinnbronze in der fruhen Metallurgie Europas', Acta Praehistorica et Archaeologica 2, 199'-253. Stech, T. & Pigott, V. C. 1986. 'The metals trade in southwest Asia in the third millennium B.C.', Iraq 48, 39-64. Stos-Cale, Z. 1989. 'Cyclad ic copper metallurgy', in A. Hauptrnann, E. Pernicka & G. A. Wagner (eds), Old World Archaeometallurgy. Proceedings of the International Symposium, Heidelberg 1987 (Der Anschnitt, Beiheft 7). Bochum Deutsches Bergbau-Museum, 279-291. Stos-Cale, Z. 1992. 'The origin of metal objects from the Early Bronze Age site of Thermi on the island of Lesbos', Oxford Journal of Archaeology 11 (2), 155-177. 37 Stos-Gale, Z. A., Gale, N. H . & Gilmore, G. R. 1984. 'Early Bronze Age Trojan metal sources and Anatolians in the Cyclades', Oxford Journal of Archaeology 3 (3), 23-43. Stos-Cale, S., Kayafa, M. & Gale, N. in press. 'The origin of metals from the Bronze Age site of Nichoria', Opuscula Atheniensia (forthcoming) . Stos-Gale, Z., Macdonald, C. F. & Gale , N. 1991. 'New studies of final Chalcolithic and early Bronze Age copper metallurgy in Cyprus', in J.-P . Mohen & C. Eluere (eds), Decouoerte du Metal (Amis du Musee des Antiquites Nationales, Millenaires 2). Paris, Picard, 341-356 . Stos-Gale, Z., Sarnpson, A. & Mangou, E. 1996. 'Analyses of metal artefacts from the Early Helladic cemetery of Manika on Euboea', Aegean Archaeology 3, 49-62 . Strahm, C. 1994.'Die Anfange der Metallurgie in Mitteleuropa', Helvetia Archaeologica 25, 2-39. Talion, F. 1987. Metallurgie susienne 1. De lafondation de Suse au XVlll' avant J. -c. Paris, Editions de la Reunion des musees nationaux. Taylor, J. W. 1983. 'Erzgebirge tin: a closer look', Oxford Journal of Archaeology 2 (3), 295-298 . Taylor, T. 1989. 'Iron and Iron Age in the Carpatho-Balkan region: aspects of social and technological change 170G400 B.C.', in M. L. Stig-Serensen & R. Thomas (eds), The Bronze Age-IronAge transitionin Europe (British Archaeological Reports, International Ser . 483). Oxford, BAR, 68-92 . Tylecote, R. F. 1991. 'Early copper base alloys; natural or man-made?', in J.-P. Mohen & C. Eluere (eds), Decouoerte du Metal (Amis du Musee des Antiquites Nationales, Millenaires dossier 2). Paris, Picard Editeur, 213-221. Vajsov, I. 1993. 'Die frtihesten Metalldolche Stidost- und Mitteleuropas', Praehistorische Zeitschrift 68, 103-145. Vandkilde, H. 1990. 'Metal analyses of the Skeldal hoard and aspects of early Danish metal use', Journal of Danish Archaeology 9, 114-132. Vandkilde, H. 1996. From Stone to Bronze. The metalworkof the Late Neolithic and Earliest Bronze Age in Denmark (jutland Archaeological Society Publications 32). Aarhus . Varoufakis, G. 1973. 'Metallourgiki ere una antikeimenon ek chalkou ek tou taphikou peribolou B ton Mykinon', in G. Mylonas, 0 Taphikos Kyklos B ton Mykinon, 363-378 . Varoufakis, G. 1995. 'Chernische und metallurgische Untersuchung von 45 minoischen Statuetten', in E. SapounaSakellarakis, Die bronzenen Menschenfiguren auf Kreia und in der Agiiis (Prahistorische Bronzefunde liS), 154-166. Vollgraff, W. 1906. 'Fouilles de Argos, Bulletin de Correspondance Hellenique 30, 5-45. Vulpe, A. 1970. DieAxte und Beilein Rumdnien I (Prahistorische Bronzefunde IX/2). Munich, C. H. Beck. Vulpe, A. 1976. 'Zu den Anfangen der Kupfer- und Bronzemetallurgie in Rumanien', in H. Miiller-Karpe (ed .), Les Debuts de la MetaUurgie. Union Internationale des Sciences Prehisioriques et Protohistoriques, IX' Congres, Nice, Colloque XXIll, 134-175 . Waetzoldt, H. 1981. 'Zur Terminologie der Metalle in den Texten aus Ebla', in L. Cagni (ed.), LaLingua di Ebla. Alii del convegno internazionale, Napoli 1980 (Istituto Universitario Orientale, Seminario di Studi Asiatici, Ser. Minor 14). Naples, 363-378 Weeks, L. 1999. 'Lead isotope analyses from Tell Abraq, United Arab Emirates: new data regarding the 'tin problem' in Western Asia', Antiquity 73, 49-64. Weninger, B. 1992. 'Fallanalysen zur I4C-Chronologie in Bulgarien', Studia Praehistorica 11-12, 407-422. Willies, L. 1992. 'Reply to Pernicka et al.: Comment on the Discussion of Ancient Tin Sources in Anatolia', Journal of Mediterranean Archaeology 5 (1), 99-103. 38 CHRISTOPHER PARE Willies, L. 1993. 'Appendix: Early Bronze Age Tin Working at Kestel', American Journal of Archaeology 97, 262­264 . Yener, K. A., Ceckinli, E. & Ozbal, H . 1996. 'A brief survey of Anatolian metallurgy prior to 500 BC', in S. Dernirci, A. M. . Ozer & C. D. Summers (eds), Archaeometry 94. Proceedings of the 29th International Symposium on Archaeometry, Ankara 9-14 May 1994. Ankara, Tubitak, 375­391. Yener, K. A. & Coodway, M. 1992. 'Response to Mark E. Hail and Sharon R. Steadman 'Tin and Anatolia: Another Look", Journal of Mediterranean Archaeology 5 (1), 77­90. Yener, K. A. & Vandiver, P. B. 1993a. 'Tin Processing at Goltepe, an Early Bronze Age Site in Anatolia', American Journal of Archaeology 97, 207­238. Yener, K. A. & Vandiver, P. B. 1993b. 'Reply to]. D. Muhly, "Early Bronze Age Tin and the Taurus:". American Journal of Archaeology 97, 255­262. Zeravica. Z. 1993. Axte und Beile aus Dalmatien und aus anderen Teilen Kroaiiens, Monienegro, Bosnien und Herzegoioina (Prahistonsche Bronzefunde IX/18).