physical barriers, cultural connections: prehistoric metallurgy across the alpine region

Physical Barriers, Cultural Connections:Prehistoric Metallurgy across the AlpineRegion

LAURA PERUCCHETTI1, PETER BRAY1, ANDREA DOLFINI2 AND A. MARK

POLLARD1

1Research Laboratory for Archaeology and the History of Art, University of Oxford, UK2School of History, Classics and Archaeology, Newcastle University, UK

This paper considers the early copper and copper-alloy metallurgy of the entire Alpine region. It intro-duces a new approach to the interpretation of chemical composition data sets, which has been applied toa comprehensive regional database for the first time. The Alpine Chalcolithic and Early Bronze Ageeach have distinctive patterns of metal use, which can be interpreted through changes in mining, socialchoice, and major landscape features such as watersheds and river systems. Interestingly, the Alpinerange does not act as a north-south barrier, as major differences in composition tend to appear on aneast-west axis. Central among these is the prevalence of tin-bronze in the western Alps compared to theeast. This ‘tin-line’ is discussed in terms of metal flow through the region and evidence for a deeplyrooted geographical division that runs through much of Alpine prehistory.

Keywords: Copper Age, Bronze Age, Alps, GIS, metal artefacts, early metallurgy, elementalcomposition

INTRODUCTION

Archaeometallurgical research in theEuropean Alps has tended to engage witha narrow range of traditional questions.Although this has been crucial for devel-oping our field of research and for creatinga firm basis for future work, it is possiblethat such questions now hinder more thanhelp. Instead of focusing on the proveni-ence of raw material, we would like tofocus our attention on flows of metal and,of course, of people, and on how thismovement was related to the physicalworld. Our analysis aims to place the met-allurgical evidence in a three-dimensionalworld where topographical elements play a

key role alongside the cultural context formaterial culture production and use. TheAlpine region is a perfect test area for thisapproach, because of its dramatic topogra-phy and the number of different cultures itcontains, the distribution of which oftenreflects the location of the main riversystems. In this paper, we explore theshifting perceptions and use of metalduring the period 3800/3600 to 2000 BC

using a novel approach to the modelling ofthe chemical data in association with Geo-graphical Information Systems (GIS). Onekey, and indeed surprising, result of ourresearch is that the circulation of metalwas apparently not dictated by the geo-morphology of the Alps, but was instead

European Journal of Archaeology 18 (4) 2015, 599–632

© European Association of Archaeologists 2015 DOI 10.1179/1461957115Y.0000000001Manuscript received 25 September 2014,accepted 27 January 2015, revised 23 December 2014

influenced by a number of strong culturalbarriers that seem more dependent on theriver systems, and possibly on prehistoricethnic boundaries, than on the mountainrange itself.

DATASETS AND METHODOLOGY

Designing the database

This research encompasses the entireAlpine region, roughly corresponding totoday’s southeast France (the Rhônevalley), Switzerland, southern Germany,northern Italy, Austria, and Slovenia(Figure 1). The first step of the researchwas to create a database of all the pub-lished chemical analyses of copper-alloyartefacts dating to 3800/3600–2000 BC, towhich a further set of 35 unpublisheddeterminations (kindly provided by PeterNorthover) was added (Table 1). Thelatter artefacts were dated with referenceto Pedrotti’s (2001) scheme, alongsidefurther personal communication with Ped-rotti (2014). For further details on thedata set, see note 1.1

The database (to be published as part ofthe doctoral thesis of the first author)records, for each object, information aboutthe source (author) of the chemical

analysis and its original identifyingnumber, typology, find site, chronology,and alternative dates (if applicable),chemical composition in weight percen-tage, the museum where it is held,inventory number, and bibliography. Thedatabase comprises 4760 entries. Most ofthe data are based on Krause (2003), plussome new analyses (e.g. Angelini, 2004,2007; Angelini & Artioli, 2007; Cattinet al., 2011; Pernicka, 2011). Given thatsome objects have been analysed morethan once, our statistical modelling wasbased on the mean value of the set ofavailable data. The nature of the findcontext was also noted (i.e. settlement,hoard, burial, or single find) as well as itsgeographical coordinates. The coordinateswere mostly obtained from Krause (2003)and were validated through cross-referencing with other sources (Bosio &Calzavara, 1990; Rossi & Bishop, 1991;Keller, 1992; Casini, 1994), includingGoogle Earth and the reference databasefor the Alpine stilt houses (http://www.palafittes.org/).The creation of such a broad database

poses key methodological questions,including: (a) the challenge of combiningdata produced by different analyticalmethods; (b) how best to express chemicaldata with different inherent accuracies andprecisions, due to the range of analyticalmethods employed; and (c) how to recon-cile the different chronologies andseriation sequences proposed by differentauthors.One may question the practicality and

wisdom of including incomplete and olddata within a single database, since ana-lyses undertaken with older techniquesmay not be as accurate and precise asthose carried out with modern methods.To compound the problem, calibrationand secondary standard data are usuallynot given in journal publications. Pernicka(1986) considered this question in the

1Chemical data were acquired using Krause’s database (Krause,

2003) and relative bibliography, which includes (Otto &

Witter, 1952; Junghans et al., 1960, 1968a, 1968b, 1974;

Barker, 1971; Ottaway, 1982; Rychner, 1995). Further infor-

mation were acquired from (Marchesetti, 1889; Colini, 1907;

Barfield & Broglio, 1966; Matteoli & Storti, 1974; Budd,

1991; Barfield, 1995; Artioli et al., 2003; Hook, 2003; Angel-

ini, 2004, 2007; De Marinis, 2005: 200, 2006a, 2006b;

Höppner et al., 2005; Angelini & Artioli, 2007; Pearce, 2007;

Kienlin, 2008; Angelini et al., 2010; Cattin et al., 2011; Per-

nicka & Salzani, 2011). The compositional data published by

Pittioni and co-workers (Preuschen & Pittioni, 1939; Pittioni,

1957; Neuninger & Pittioni, 1963) were not included in the

database because their determinations are semi-quantitative,

and the presence of elements is solely indicated by symbols (e.g.

Sn: +++). The quality of these data is insufficient for the kind

of geostatistical analysis employed in our research, which math-

ematically assesses the influence of elevation, distance and other

topographical metrics on the dataset (see below).

600 European Journal of Archaeology 18 (4) 2015

http://www.palafittes.org/





context of the feasibility of using theStudien zu den Anfängen der Metallurgie(SAM) data (Junghans et al., 1960, 1968a,1968b, 1974) and concluded that they arecomparable with data obtained by moderntechniques. Rychner & Northover (1998)also undertook a comparative study of theseveral techniques used to analyse ancientmetal artefacts. The results obtained withoptical emission spectroscopy (OES),inductively coupled plasma atomic emis-sion spectroscopy (ICP-AES), atomicabsorption spectroscopy (AAS), X-ray flu-orescence (XRF), electron probemicro-analyzer (EPMA), and neutronactivation analysis (NAA) were comparedand the authors concluded that ‘Themodern analytical techniques are capableof producing accurate, reproducible datathat can, with thorough standardisation,be used interchangeably with other dataand behave similarly in cluster and classifi-cation’ (Rychner & Northover, 1998: 31).According to Rychner and Northover, theaccuracy of XRF was more problematic,since the inter-laboratory comparisonsshowed quite a wide range of results. Lutz& Pernicka (1996) compared dataobtained with a portable XRF and those

obtained with NAA and demonstrated ageneral comparability of the two sets ofresults. It seems likely that the XRF issuesfound by Rychner and Northover weredue more to human errors in different lab-oratories rather than an intrinsic deficiencyof the technique. Finally, Merkl (2011:89, figures 8.4 and 8.5) undertook Princi-pal Components Analysis (PCA) on dataobtained by Otto & Witter (1952), theSAM group (1960, 1968a, 1968b, 1974),and Krause (2003). He observed that ana-lyses on the same objects by differentlaboratories are usually clustered togetherwhen plotted. The few outliers were ident-ified as individual errors of measurement.We may conclude that, with caution, dataobtained with different methods, by differ-ent laboratories, and at different times, canbe compared.The second point, namely consistency

of expression, is fundamental, given theway in which we wish to treat the data. Inorder to undertake geostatistical analysis,the data must be expressed purely numeri-cally, i.e., without symbols intended toconvey semi-quantitative observations,such as +, <, > or �. For the data pub-lished by Otto & Witter (1952) and

Figure 1. Map showing the research area.

Perucchetti – Physical Barriers, Cultural Connections 601

http://www.maneyonline.com/action/showImage?doi=10.1179/1461957115Y.0000000001&iName=master.img-000.jpg&w=359&h=179

Table 1. Previously unpublished composition data courtesy of Peter Northover

Analysis Item Site Context sf Period Cu Sn Pb As Sb Ag Ni Bi Au Zn Co Fe S

MC10 Droplet Monte Covolo S1E1 M5 CA 99.74 0.01 0.02 0.28 2.56 1.24 1.51 0.05 0 0.01 0.02 0.03

MC11 Droplet Monte Covolo S3E1 M5 CA 96.04 0.02 0.03 0.16 0.79 1.25 1.57 0.02 0 0.08 0.01 0.03

MC12 Waste Monte Covolo S2W2 CA 99.72 0 0.26 0 0 0 0.02 0 0 0 0 0

MC13 ?Bead Monte Covolo CA 99.82 0 0.04 0 0 0 0.08 0.05 0 0 0.01 0

MC6 Awl Monte Covolo N3W2 M2 CA 95.31 0.02 0.03 0.22 1.89 1.08 1.42 0.02 0 0.01 0.01 0

MC8 Droplet Monte Covolo S2E4 M6 CA 88.86 0.04 0 0.78 5.71 1.86 1.68 0 0 0.03 0.02 0.02

MC9 Wrought frag. Monte Covolo M4 CA 99.38 0 0.12 0.27 0 0.18 0.01 0.02 0.02 0.01 0 0

MC95/1 Copper strip orawl

Monte Covolo MC92 F 46 CA 96.72 0.01 0.02 2.67 0.05 0.19 0.25 0.01 0.05 0 0.01 0 0

MC95/10

Tubular beadfrag.

Monte Covolo MC93 RF60

2325 CA 94.28 0.01 0.05 0.1 0.01 0.03 0.07 0.03 0.03 0 0.01 0 0

MC95/11

Sheet fragments Monte Covolo MC93RF53

1979 CA 51.12 0.04 0 0.27 0 0.01 0.21 0.04 0.03 0.08 0 1.07 0.01

MC95/12

Tubular beadfrags.


3469 CA 99.72 0 0.14 1.93 0.02 0.04 0.05 0.02 0.03 0.13 0 0.32 0.05

MC95/2 Tubular bead Monte Covolo MC93 RF67

2511 CA 99.25 0 0.01 0.31 0.01 0 0.42 0 0 0 0 0 0

MC95/3 Awl/wire Monte Covolo MC93 RF78

3091 CA 91 6.89 0.5 0.58 0.26 0.06 0.21 0.19 0.01 0.03 0.02 0.24 0.94

MC95/4 Tube fragments Monte Covolo MC92 RF47

1421 CA 99.79 0 0.02 0.1 0.02 0 0.09 0.01 0 0.01 0.01 0 0.01

MC95/6 Tubular bead Monte Covolo MC92 RF29

1459 CA 98.59 0 0.02 1.29 0 0.02 0.05 0 0.01 0 0 0 0

MC95/7 Fragment Monte Covolo MC94 RF94

3597 CA 99.84 0 0 0.1 0 0.03 0.09 0 0.02 0 0.01 0 0.01

602

European

Journal

ofArchaeology

18(4)2015

MC95/8 Large tubularbead


3509 CA 98.55 0.01 0.21 1.03 0.02 0.08 0.04 0.06 0 0 0.01 0 0.01

MC95/9 Cast fragment Monte Covolo MC94 104 CA 99.57 0 0.01 0.25 0 0.02 0.09 0.01 0.02 0.01 0.01 0 0

TBC1 bar ingot Valle di Non A1 tr 0 0.92 0.01 0.02 0.28 0 0 0.02 0.13

TBC2 shaft-hole axe Valle di Non A1 3.21 0 0.06 0.02 0.21 0 0 tr tr 0.01

TBC3 bar ingot Valle di Non A1 0.03 0 9.72 0.21 0.06 0.6 0 0 0.48 1.28

TNS1 wire fragm. Moletta Pattone CA 0 0.82 0.12 0.02 0.07 tr 0.05 0.02 0.01 tr

TNS10 wire, fragm. Moletta Pattone CA 0 0.19 tr 0.03 0.07 0.02 0.02 tr 0.005 tr

TNS11 spiral bead Vela Vabusa CA 0 0.03 3.2 0.03 0.04 0.01 tr tr tr 0.01

TNS12 armlet Bersaglio deiMori

CA tr 4.27 0.78 0.03 0.28 tr 0.03 0.01 0 tr

TNS15 needle Riparo Gaban CA 0 0.69 0.2 0.02 0.2 0.01 0 0 tr 0

TNS16 riveted strip Dos della Forca CA 0 0.07 0.05 0 0.06 tr 0 0 0 tr

TNS2 large bead Moletta Pattone CA 0.02 0.13 0.11 tr 0.18 0.02 tr 0 tr tr

TNS3 spriral Moletta Pattone CA 0.01 0.34 0 tr 0.27 tr 0 0 tr tr

TNS4 strip Moletta Pattone CA 0 0.59 0 0 0.28 0.02 0 0 0 0.005

TNS5 strip Moletta Pattone CA 0 0.39 0.1 0 0.36 0 0 0 0 0.02

TNS6 strip Moletta Pattone CA 0 0.07 0.05 tr 0.19 0.01 0 0.02 tr 0.02

TNS7 wire Moletta Pattone CA 0 0.36 0.06 tr 0.04 0.02 0 0 tr tr

TNS8 wire Moletta Pattone CA tr 0 0.07 0 0.35 tr 0.03 0 tr tr

TNS9 wire Moletta Pattone CA 0 0.21 0.07 0 0.09 0.01 tr 0 0 0.01

Perucchetti

–Physical

Barriers,

Cultu

ralCon

nection

s603

Junghans et al. (1960, 1968a, 1968b,1974), the symbols have been convertedinto numerical values using the table ofconversion proposed by Ottaway (1982,section XXIII) (Figure 2). In other cases,we have made a number of simpleassumptions. Where the symbol indicatesthe presence of an element as ‘trace’ or‘less than detection limits’, we haveassigned a value of half the detection limit.So, if for any element the detection limitis 0.01 per cent, we have assigned 0.005per cent to ‘trace’ or ‘<0.01’.

CHRONOLOGICAL SYSTEMS

Another issue to be considered is thechronology of the artefacts. Working withsuch a wide geographical area forces us tomerge, simplify and often struggle withdifferent chronological schemes. The firstmajor challenge lies in the terminology of

the period in question, variously known asLate/Final Neolithic in northern Alpineterminology and as Copper Age (or Eneo-lithic) in the southern Alps (c. 3800/3600–2200 BC); it is followed in bothregions by the initial Early Bronze Age (c.2200–2000 BC), which is also included inour research. Scholarly tradition goingback to the nineteenth century has it thatthe broadly equivalent terms Chalcolithic/Eneolithic/Copper Age were solely intro-duced in those regions in which asubstantial metallurgical phase predatingthe Bronze Age was identified. This hasgenerated an unhelpful terminologicalsplit across Europe whereby mostsouthern and eastern countries (with thenotable exception of Greece) now recog-nise a Copper Age phase wedged betweenthe Neolithic and Bronze Age, while mostnorthern and western countries do not(Lichardus, 1991; Kienlin, 2010: 3–20;Heyd, 2013: 34, fig. 10). The fault line

Figure 2. Table used to convert symbols into numbers from Ottaway (1982: Appendix XXIII). Inorder to allow numerical calculations, Ottaway devised a table to convert the semi-quantitative symbolsused by Otto & Witter (1952) and Junghans et al. (1960, 1968a, 1970) into estimated values for eachelement, based on the detection limits of the instrument used by the authors (OES). The field 0.01 percent stands for <0.01 per cent.



slices neatly through the Alpine region,for the French, Swiss, German, and Aus-trian scholarships define the earliestmetallurgical phases as Late/Final Neo-lithic, while in Italy and Slovenia these arecalled Copper Age or Eneolithic (not-withstanding the recent attempts toidentify a Chalcolithic phase in centraland western Europe: Heyd, 2013). Theterminological inconsistency is best exem-plified by the metal-equipped Iceman,whose mummy was found on the borderbetween Austria and Italy; following theirrespective scholarly traditions, the Aus-trians see him as ‘Stone Age man’ and theItalians as ‘Copper Age man’ (De Marinis& Brillante, 1998: 45–46).A further, more severe, problem is

encountered when one tries to define a setof parameters that could unambiguouslyidentify the earliest metallurgical period inthe entire region. Purely technologicalfactors based on the presence/absence oramount of metal artefacts are plainlyinadequate. For example, the north-alpinePfyn, Altheim and Mondsee groups,which show precocious florescence ofmetal-using beginning c. 3800 BC, werefollowed by the Horgen and Chamgroups, which were significantly poorer inmetal objects (Strahm, 1994, 2005;Kienlin, 2010: 13–16); conversely, the

initial stages of the Italian Copper Agewere long thought to be poor in metalartefacts, although this reading has beentempered by recent research (Pearce, 2007:51–52; Kienlin & Stöllner, 2009; Dolfini,2013). Until recently, a seemingly usefuldistinction was made between metal use,which would occur in securely Neolithiccontexts (c. 4500–3800/3600 BC), andmetal production, which would define aFinal Neolithic/Copper Age phase (c.3800/3600–2200 BC; Skeates, 1993;Strahm, 2005). However, this readingmust now be rejected given that the ear-liest evidence of smelting andmetalworking both north and south of theAlps has been pushed back to the latefifth millennium BC (Höppner et al.,2005; Mazzieri & Dal Santo, 2007; seeDolfini, 2013, for discussion). Likewise,cultural parameters that seek to correlatethe emergence of metallurgy with sweep-ing changes in the social organisation ofsociety are not without their problems.Leaving aside the growing scepticismregarding the role played by early metal-work in triggering social inequality, onemust note that, perhaps influenced bytheir differing terminologies and evidence,French and German-speaking archaeolo-gists tend to be much more cautious thantheir Italian colleagues in postulating

Table 2. Chronological phases used in this work and correspondence between the authors

Phases Dates BC Krause David-Elbiali De Marinis

Copper Age ≈3600–2200 2, 20, 22, 25 _ CA

Bz A1 2200–2000 3 A1 A1a

Bz A2a 2000–1800 4 A2a A1b/A1c

Bz A2b 1800–1600 5 A2b A2

The database provided by Krause (2003) comprises a date field, in which different phases of the CopperAge are marked by numbers 2, 20, 22, 25. 2 refers to the Copper Age in general, 20 to the first phaseof the Copper Age and 22 to the second, while 25 refers to the Bell Beaker ’culture’, which straddles thelate Copper Age and initial Early Bronze Age. The numbers 3-5 mark the three main phases of theEarly Bronze Age based on Reinecke’s chronology. This is the scheme followed by David-Elbiali(2000), while De Marinis (2005, 2006a, 2006b) uses a different terminology to define these phases


structural links between pre-Bronze Agemetallurgy and the emergence of socialcomplexity [compare, for example, Kienlin(2010: 80–117) and Mille & Carozza(2009) with Guidi (2000: chap. 4)].A final problem is presented by the

absolute chronology of the earliest metal-lurgical phases. Here, disagreementbetween scholars is aggravated by theexistence of national sequences in allthe countries concerned. Moreover, therespective phase boundaries are oftenplaced at different times in the sequence,thus making correlations difficult(Table 2). The problem becomes intract-able when one tries to discern sub-phaseswithin the major phases. For example,early copper artefacts from northern Italyhave been divided into several sub-phasesby Carancini (2001) and De Marinis(1997, 2013), but they disagree on boththe absolute chronology of the sub-phasesand on which objects should be assignedto each of them. To make things worse,both chronological schemes have been cri-ticised by Barfield (2007), Barfield &Kuniholm (2007), and Genick (2012:556–81). Furthermore, Dolfini (2010,2013, 2014a) has recently argued that theentire sequence of early Italian metalworkmust be revised considering the flawedmethodological premises in which it isgrounded, as well as its manifest mismatchwith the radiocarbon dates available.Trying to synchronise any of thesesequences with the no less controversialschemes proposed by Krause (2003),David-Elbiali (2000) and others for thenorthern and western Alps (Hafner, 1993;Rageth, 1993; Neugebauer, 1998; Strahm,2005; Vital, 2005; Hafner & Suter, 2007)is an unachievable task.Given the difficulties with disentangling

such a veritable Gordian knot, we havedecided to cut it clean through. In prac-tice, this means that all the objectsassigned to the Late Neolithic/Copper

Age in the entire Alpine region have beengrouped together within anall-encompassing category, which for thesake of consistency we call ‘Copper Age’.This phase begins with the first sustainedproduction and use of metalwork c. 3800/3600 BC and ends with the inception ofthe Bronze Age c. 2200 BC. It excludes theearliest experiments with copper metal-lurgy in the late fifth and initial fourthmillennia BC on the account that theobjects belonging to this phase are, bothnorth and south of the Alps, relativelyfew, poorly dated, and not alwaysanalysed.Although rather crude, our grouping

has the distinct advantage of beinggrounded in conventional (and widelyaccepted) chronological and culturalboundaries for the period in question; italso makes our work less susceptible tolater revisions, which would be all themore likely given the volatile artefact seria-tion sequences currently available for theregion. By choosing to group together allmetal artefacts which can be labelled LateNeolithic or Copper Age, we consciouslyreject the current possibility of arriving ata finer-grained chronological understand-ing of metal circulation and exchange inthe long period of time from 3800/3600–2200 BC – an exercise that demands a con-certed and extensive research effort.Similar problems, albeit on a less dra-

matic scale, are raised by the chronologyof the initial Early Bronze Age. In par-ticular, technological or culturalparameters are just as inadequate tocapture the emergence of tin-bronzemetallurgy and the parallel development ofBronze Age Alpine society, as they are todefine the Copper Age. From a metallur-gical viewpoint, it has long beenrecognised that the adoption of tin-bronzeprobably occurred during the advancedEarly Bronze Age, and that the new alloywould have coexisted for some time with


fahlerz copper and other Copper Agecompositional groups (Pare, 2000; Krause,2003; De Marinis, 2006a). Instead ofeither seeing an alloy sequence, or parallelexistence of alloys, below we will highlightthe practice of adding tin to arsenical andantimonial composition groups. From acultural viewpoint, the frantic search forelites that characterised so many prehisto-ric studies in the twentieth century hasrecently been questioned by new concep-tual approaches, which stress the smallscale of mining and smelting operations aswell as the scarcity of Early Bronze Age‘central places’ in the Alpine region(Kienlin & Stöllner, 2009; Kienlin, 2013).As with the Copper Age, a definition ofthe Alpine Early Bronze Age must begrounded in broadly accepted chronologi-cal parameters to be of any use. Luckily,the exercise is this time made easier by thenear-synchronous appearance of BronzeAge cultural groups in the entire region c.2200 BC (Krause, 1989; Pearce, 1998: 57;David-Elbiali, 2000: 267–270, fig. 14;Della Casa, 2013: 712, fig. 39.3;Nicolis, 2013: 694; Roberts et al., 2013:18–19, fig. 2.1), with a possible slightdelay in south-eastern France (Strahm,2005: 33), but this is now denied by Gui-laine et al. (2001) and Mille & Carozza(2009).Another aspect of the chronological

knot lies in the clear-cut subdivision of theEarly Bronze Age into three major phasesbased on Reinecke’s (1924) chronology,which has long provided a solid frameworkfor dating central European prehistory(Table 2; see also, among others,David-Elbiali, 2000: 267–270, fig. 14;Krause, 2003: 84, fig. 34; Roberts et al.,2013). This subdivision has good corre-spondence to the chronological frameworkdrawn by De Marinis (2005, 2006a,2006b) for northern Italy (Table 2; seealso Nicolis, 2013: 694), bearing in mindthat our research is solely concerned with

the beginning of the sequence in Rein-ecke’s (1924) phase Bz A1. The mainissue here is that, even when agreeingupon the same overall chronology, differ-ent authors propose different absolutedates for the same objects. This is the casewith De Marinis (2005, 2006b) vis-à-visCarancini (1996; see also Carancini &Peroni, 1999) for northern Italy, and withKrause (2003) vis-à-vis David-Elbiali(2000) with regard to western Switzerland.Considering that Carancini’s frameworkhas been questioned repeatedly (DeMarinis, 2005, 2006a; Dolfini, in press),we have decided to disregard it for thepurpose of this research. Moreover, DeMarinis’ chronology holds the distinctadvantage of being comparable withKrause’s, with the exception of certainobjects from the Italian north-west. In thiswork, we have tried to take into consider-ation the different dates provided by thesescholars, and to explain how our sub-sequent interpretations differ according totheir systems.

INTERPRETING THE CHEMICAL DATA

In this work, we have not analysed thechemical composition data using clusteranalysis, which, since Ottaway (1982), hasbeen the most common method used onmetal chemical datasets. Cluster analysisresults in the creation of a number ofstatic metal compositional groups, each ofwhich is commonly considered as comingfrom a specific source. We think that thisobscures useful archaeological structures inthe data associated with recycling, oxi-dative loss linked with use, mixing,alloying, and so forth (Bray & Pollard,2012). These changes have been demon-strated experimentally (McKerrell &Tylecote, 1972), and have long beenrecognised in industrial metallurgy(Hampton et al., 1965; Charles, 1980;


Beeley, 2001). The chemical shifts overthe course of a unit of metal’s ‘life-time’means that it could easily move betweenthe very fine divisions defined in someclassification schemes – see, for example,the mathematically defined categories ofthe SAM project: Junghans et al. (1960,1968a). Clustering places artificial bound-aries on interpreting material flows overtime and space. Instead, to aid interpret-ation, we place the data into simplepre-defined metal categories by consider-ing the presence/absence of arsenic,antimony, silver, and nickel (see Table 3).These can be thought of as heuristic cat-egories intended to broadly capturevariation within the dataset, which is thenfurther investigated in a number of ways(see below). To determine presence andabsence, we first subtract tin from thetotal, and then normalise the data to 100per cent. The aim of this procedure is toidentify the main chemical signature ofthe copper-alloy base, free from the dilut-ing effect of alloying. Characterising thelevel and distribution of tin is of coursecrucial to understanding metal flow, andoccurs in parallel with investigating thecopper-base composition.A threshold value of 0.1 per cent for

‘present’ was used for As, Sb, Ag, and Ni,after correcting for alloying (Bray &Pollard, 2012). These four elements werechosen firstly as all past chemical compo-sition projects have analysed for them. Wealso need elements that have been consist-ently recognised as being diagnostic ofcopper identity in some way. As, Sb, Ag,and Ni, with their different chemicalproperties, mechanical properties andnatural abundances fulfil this requirement,and they have been commonly used inprevious chemical-typological work (suchas the various SAM schemes, and North-over, 1980). These four elements capturemuch of the variation seen in coppermineral deposits, pass into the metal

during the smelt, and then behave differ-ently during technological processes – forexample, different oxidation rates uponmelting. Possible issues over the hetero-geneity of metal, for example segregationof arsenic, may be resolved with a propersampling strategy, which Pernicka (1986)describes as taking at least 3 mg of metal.This is a threshold widely passed, at leastfor the SAM project (Pernicka, 1986: 25),which is the core of our database. The useof these four elements gives us a reliable, ifbroad, picture of the early history of metalflow. Of course, other elements couldprofitably be investigated, and will be infuture research. However, our currentapproach has been tested in detail for theBritish Isles (Bray & Pollard, 2012), andhas the distinctive advantage of makingdata processing and grouping manageabledue to the number of elements involved.Bismuth was not taken in consideration,even if it is claimed to be a diagnosticelement for the mineral deposits (Per-nicka, 1999, 2011, 2014), because, whenpresent, it is often found in very smallquantities, well below our threshold (0.1per cent). This is often close to the detec-tion limit of analytical instruments, so thatwe may have too many false negatives inthe dataset. More importantly, it was notanalysed by all the authors and, had wedecided to include it, the consistency ofour database would have been affected.Similarly, lead was not considered in thisphase of the research because it is comple-tely insoluble in copper, thus leading tohigh segregation rates in the objects. Inany case, less than 10 per cent of objectsin our database pass the threshold of 0.1per cent lead.The resulting sixteen compositional

groups (all possible combinations of pres-ence/absence for the four trace elements)are outlined in Table 3. This approachaims to objectively capture variation withinthe whole assemblage – each


compositional group is not seen as deriv-ing from a single source, and each sourceis not necessarily expected to producecopper of just one compositional group.Also, through recycling, mixing and smi-thing, a unit of copper may pass through anumber of groups in its ‘lifetime’ (Bray &

Pollard, 2012). This kind of classificationallows the metal chemistry to be con-sidered from an archaeological rather thana purely statistical perspective, in order tocharacterise the flow of metal, rather thanjust as clusters of maximum difference inthe final composition. In other words, thesixteen groups are a starting point tounderstand the process of metal flow, aprocess that encompasses the whole ‘life-time’ of a unit of copper: from theextraction of the ore from a specific mine,through possible alloying, manipulation,mixing and recycling. Our approach aimsto tease apart the complete sequence offactors that affect the final chemical com-position of a copper-alloy assemblage,rather than conflate them into one step ofstatistical analysis. The compositional dataare linked with the coordinates of the find-spots using ArcGIS 10.2, which allows usto investigate the relationship betweenmetal groups and topographical elements, inparticular rivers and watersheds. We do notsimply consider the raw number of objectswith a certain composition per zone, butinstead analyse the ‘percentage presence’ ofthe chemical groups. This approach is also

Table 3. Definition of copper groups used in thispaper

Elements present at 0.1 per cent Group

Cu 1

Cu, As 2

Cu, Sb 3

Cu, Ag 4

Cu, Ni 5

Cu, As, Sb 6

Cu, Sb, Ag 7

Cu, Ag, Ni 8

Cu, As, Ag 9

Cu, Sb, Ni 10

Cu, As, Ni 11

Cu, As, Sb, Ag 12

Cu, Sb, Ag, Ni 13

Cu, As, Sb, Ni 14

Cu, As, Ag, Ni 15

Cu, As, Sb, Ag, Ni 16

Figure 3. Distribution map of Copper Group 1, copper without trace elements in the Copper Age. Thesize of the dots indicates the number of objects. The colour scale represents the percentage of each geo-graphical square’s metal assemblage that belongs to Copper Group 1.



known as Ubiquity Analysis, and has beenfruitfully applied to datasets from a rangeof disciplines including archaeobotany andpalaeoclimate reconstruction (Bray et al.,2006).

RESULTS: A HISTORY OF ALPINE METAL

Copper Age: Copper compositionalgroups

In the Copper Age (c. 3800/3600–2200BC), roughly 55 per cent of all the analysedmetal artefacts were made of either purecopper (Group 1; Figure 3) or copper withtraces of arsenic (Group 2; Figure 4).Artefacts with these compositions are dis-tributed all over the Alps, but the mapshows slight east-west differences that maybe significant: whereas arsenical copper(Group 2) was slightly more dominant inthe north-western part (33 per cent of allobjects were made from arsenical copperversus 29 per cent of pure copper), purecopper (Group 1) predominated in thenorth-east (36 per cent versus 23 per centof arsenical copper). In northern Italy, the

prevailing composition was arsenicalcopper (Copper Group 2: 26.5 per cent).Interestingly, the second most commoncomposition was not pure copper, butcopper with arsenic and silver, in particularin the eastern Alps (Group 9; Figure 5).The well-known relationship betweencomposition and artefact classes in theCopper Age should be remembered,namely that daggers tend to be made ofarsenical copper and axes of pure copper(De Marinis, 2006b; Pearce, 2007;Dolfini, 2014b). As has been notedbefore, in copper chemical groups thatcontain both axes and daggers, forexample, Group 9 (As and Ag), daggershave a higher average arsenic level thanaxes. Our new model can offer an expla-nation for this, as arsenic is lost from axesas they are heated, re-smithed andrecycled. Meanwhile, daggers tend to besharpened through abrasion at roomtemperature, with no associated chemicalchanges (see Bray & Pollard, 2012, forsimilar case studies).Overall, however, what this brief survey

underlines is the importance of a largenumber of different copper groups during

Figure 4. Distribution map of copper Group 2, arsenic copper in the Copper Age. The size of the dotsindicates the number of objects. The colour scale represents the percentage of each geographical square’smetal assemblage that belongs to Copper Group 2.



the Copper Age across the Alpine region.Considering the entire zone, arsenicalcopper and pure copper together representapproximately 50 per cent of all artefacts:the remaining 50 per cent is made ofseveral different copper groups, manifestedas various combinations of arsenic, anti-mony, silver, and nickel. In the AlpineCopper Age, at least seven differentcopper groups, apart from pure copper andarsenic copper, are represented, in a rangeof 8 per cent (Group 4) to 2 per cent(Group 12). These seven groups oftenhave clearly defined, restricted distri-butions. For example, group 4 is more‘eastern’ (Figure 6), whereas groups 5 and7 are more ‘western’ (Figures 7 and 8).Group 4, defined as copper with low

levels of silver, has a clear eastern distri-bution. In the north-east of Italy, it isparticularly prevalent and is more commonthere than pure copper and arsenicalcopper, contrary to the general pattern ofthe alpine region. In particular, this coppercomposition is closely associated with theBocca Lorenza type axes. These arecommon in the northern-west part of

Italy, and are represented, beside the axesfrom the Bocca Lorenza burial cave, byone axe from Merendole (PD) and onefrom Tormiceva cave in San Canziano(TS) (Pearce, 2007: 45). In the northernzone of the Alps, grip-tongue daggers alsoshow the copper group 4 signature (copperwith low levels of silver). These artefactsare often related to Bell Beaker groups, forexample the daggers found in Kircheim(Krause, 2003), Wolfarshauser, andMoosinning.Copper Group 9 (silver and arsenic as

minor elements) also has a predominantlyeastern distribution. Most of the artefactsfrom the Remedello cemetery have thiscomposition signature, in particular thedaggers of Remedello type, whereas theBeaker-style dagger found in the cemeteryis made of arsenical copper (Group 2).Many objects from Trentino are made ofcopper belonging to Group 9. This is thecase for the beads and an awl from LaRocca di Manerba, Frana del Bersaglio,Riparo Gaban, Arco, and the axe fromCol del Buson. In the northern zone ofthe Alps, this composition is mainly found

Figure 5. Distribution map of Copper Group 9, copper with arsenic and silver in the Copper Age. Thesize of the dots indicates the number of objects. The colour scale represents the percentage of each geo-graphical square’s metal assemblage that belongs to Copper Group 9.



in flat axes, as in the cases of Lieferind,Attersee, Rainberg, Mondsee.Group 5, nickel as the only trace

element, is typical of Switzerland, broadlycorresponding to the SAM copper groupFC (Junghans et al., 1960, 1968b, 1974).Its geographically focussed distribution wasrecognised by SAM, and also, morerecently, by Matuschik (2004) and Cattin(2008). Strahm (1994: 29, 2005: 32)hypothesised a west-Alpine source andnotes its presence in France. This coppergroup is also found in the Southern part ofthe Alps, for example in the Copper Agecemetery at Sabbione (Pearce, 2007: 85).The antimony-silver pattern of Group 7

is commonly linked with the early minesexcavated near Cabrières in southernFrance (Ambert, 1995; Bourgarit & Mille,2005; Prange & Ambert, 2005). As withmany of the other copper patterns thistype has clear regional hotspots and associ-ations with particular object classes. Insouth-western France ornaments oftenshow the Sb-Ag signature, in particularbeads found in burials in dolmens (such asDolmen des Cudières) and caves (for

example Le Baume de Lan). In Switzer-land, on the other hand, this compositionis more often found in axes, such as thosefrom La Graviers, Vallamand, Treytel, andVinelz. Similarly in Italy, the axe fromFiesse is group 7, which was found in aBeaker Culture grave (Krause, 2003).Overall, the picture that builds up is

one of a diverse range of subtly differentcopper types in use across the Alpineregion, many of which have clear andlimited geographical ranges and associ-ations with particular object classes. Thisseems to indicate that societies in theCopper Age mainly extracted and smeltedcopper locally, and did not trade the metalwidely outside of their territory. It isimportant to note that for this period,whether we classify a certain group ofobjects based on Krause’s (2003), De Mar-inis’s (2005, 2006a, 2006b), orDavid-Elbiali’s (2000) chronology, theoverall result is the same. In other words,even considering slight differences in thegeneral chronology of the objects, thesame metal groups and distributional pat-terns consistently stand out as significant.

Figure 6. Distribution map of Copper Group 4, copper with silver in the Copper Age. The size of thedots indicates the number of objects. The colour scale represents the percentage of each geographicalsquare’s metal assemblage that belongs to Copper Group 4.



The Early Bronze Age 1: Coppercompositional groups

The transition between the Copper Ageand the Early Bronze Age (c. 2200–2000BC) is accompanied by a clear change inthe trace element composition of the arte-facts. Compared to the Copper Age, thechemical groups are far more consistentacross the entire study area, suggesting thegrowth of fewer but larger-scale exchangenetworks. Almost 80 per cent of all theartefacts dating to this period fall intoeither Groups 12 (Figure 9), 16(Figure 10), or 1 (Figure 11). Almost halfof all the artefacts dated by Krause as BzA1 belong to Group 12 (copper with As,Sb, and Ag), and are mostly found in thenortheast Alps. This group represents thefamous Ösenringe chemical composition,and indeed 80 per cent of the Group 12objects in the Alpine region are Ösenr-ginge from hoards. Many of these hoardsare extremely large, containing more thana hundred artefacts. Axes and daggers inthe north-east of our study zone have asimilar composition.

Group 16 (copper with low levels ofarsenic, antimony, silver, and nickel) is alsoa very significant composition pattern, com-prising roughly 22 per cent of the artefactsdated to Bz A1. This group seems to havebeen more common in the western andcentral regions, with a particular focus onthe Rhine valley and the northern parts ofthe Rhône, Oglio and Adda rivers. Outsideof this principal distribution, there areoccasional deposits of objects made fromGroup 16 copper. This is the case, forexample, of the Wolnzach axe hoard fromthe north-east of our study area. This collec-tion of local, north-east Alpine style ‘Saxon’axes (Kienlin et al., 2006: 462; Kienlin,2010: 137) stands out in an area dominatedby the use of Group 12 copper. Hence, wehave a case of artefacts made to a localdesign using copper typical of anotherregion. This might be interpreted as evi-dence for the movement of a single coherentbatch of metal, which was then recast.If we consider the Alps as a whole

region, Group 16 copper was used in theproduction of all categories of artefact, andwas less closely tied to a specific type, as

Figure 7. Distribution map of Copper Group 5, copper with nickel in the Copper Age. The size of thedots indicates the number of objects. The colour scale represents the percentage of each geographicalsquare’s metal assemblage that belongs to Copper Group 5.



with Group 12 and the Ösenringe. It hasto be noted, though, that as we movefrom east to west, the number of hoardsmade of Ösenringe decreases and, at thesame time, the number of hoards com-posed by axes goes up. These axes mainlybelong to Group 16.The third most important composition

is pure copper (Group 1, approximately 15per cent of the initial Early Bronze Ageassemblage), which is found most com-monly at the western end of our studyarea, roughly corresponding to the Rhônevalley. As in the Copper Age, the use ofdifferent chronologies does not change thegeneral patterns of copper distribution (orits composition) across the Alpine range.Awls and pins in French burials com-monly have this copper signature,particularly assemblages found in caves,such as Grotte de la Carrière. In westernSwitzerland, again, awls, pins and lunulaefrom burials (e.g. in Les Places) are oftenGroup 1 copper, through these are notcave depositions. In the eastern part of theAlps, away from the group’s main distri-bution, there are occasional finds of very

pure copper Ösenringe deposited inhoards, for example from Sirndorf, Stock-erau, Bergen, and Eiselfing. Here again,we have an example of metal with a com-position typical of one region (the west)used to make objects characteristic ofanother (the eastern Alps).In northern Italy, the pure copper,

Group 1 composition was related to axe-hoards deposits, which are reminiscent ofwestern alpine behaviour, such as theRemedello Sotto and Serravalle assem-blages (Perucchetti, 2008). The typologyof these axes, similar to the Type Neyruz(Tecchiati, 1992), is also related to thewestern zone of the Alps.

The appearance of tin in the EarlyBronze Age

The east-west pattern observable in theCopper Age copper groupings is echoed atthe beginning of the Early Bronze Age (c.2200–2000 BC), when copper-tin alloysmake their first appearance. Figure 12shows a map of the mean percentage of

Figure 8. Distribution map of Copper Group 7, copper with antimony and silver in the Copper Age.The size of the dots indicates the number of objects. The colour scale represents the percentage of eachgeographical square’s metal assemblage that belongs to Copper Group 7.



tin per site based on Krause’s (2003)chronology. Objects classified archaeologi-cally as Ösenringenbarren andSpangenbarren, and which tend to containonly copper, were excluded from thisanalysis, because, if they were used ascopper ingots (De Marinis, 2006b), thentheir overwhelming presence in the easternregion would have significantly underesti-mated the presence of tin. We arefocussing here on finished objects, ratherthan copper in a putative trade form.It is clear from Figure 12 that the

mountains were not a barrier to the north-south distribution of tin, but there was astrong cultural barrier, a ‘tin line’, whichdivided the western part of the Alps,where tin was frequently used to alloycopper, from the east, where it was not. Itis also important to note that the alloyingprocess of adding tin to copper was notlinked to any specific copper compo-sitional group mentioned above. Thisstrongly suggests that tin was movingindependently of the copper.If the chronologies of David-Elbiali

(2000) and De Marinis (2005, 2006a,

2006b) are integrated in western Switzer-land and northern Italy, the tin line is notevident in the first phase of the EarlyBronze Age. As mentioned previously,David-Elbiali and De Marinis tend to post-date objects; hence, with their chronologicalframework, the tin line only appears in asecond phase of the Early Bronze Age.Overall, if we take a broad approach andconsider the entire Early Bronze Age(encompassing the different attributions ofobjects to one or another sub-phase of theEarly Bronze Age), use the chronology pro-posed by De Marinis and David-Elbialiwhen possible, and exclude hoards (whichwere made of hundreds of objects thatmight be considered as copper ingots), westill have a picture of a higher percentage ofobjects made of bronze in the western partrather than the eastern.Figure 13 shows the percentage of

objects alloyed with tin in each zone in asecond phase of the Early Bronze Ageaccording to the chronology proposed byDe Marinis (2005, 2006a, 2006b) andDavid-Elbiali (2000). The blue-tintedsquares indicate the zones where there is a

Figure 9. Distribution map of copper with Copper Group 12, arsenic, antimony and silver in the firstphase of the Early Bronze Age (A1). The size of the dots indicates the number of objects. The colourscale represents the percentage of each geographical square’s metal assemblage that belongs to CopperGroup 12.



greater percentage of objects containingtin. To the west of the tin line, there wasa much higher percentage presence ofbronzes: over 97 per cent in most places.Moreover, in particular in Switzerland, ina relatively small region there is a concen-tration of sites whose assemblage had asignificantly high level of tin (e.g. ceme-teries at Hubel, Scloss, Thun, andEcublens). On the other hand, in the eastthere was a zone of lower tin presence,with most areas having only 60–85 percent of their metal assemblage as bronze.A particularly interesting result appears ifwe divide our chemical dataset by artefactcategory. Ornaments show a particularlystrong division between coherent areas oftin-bronze use compared to the continu-ation of copper use (Figure 14). Tinaddition would produce a brighter, morelustrous object, therefore the pattern of itsuse in ornaments perhaps supports theconcept that display and colour, ratherthan mechanical properties, influenced theadoption and use of tin-bronze.Overall, whichever chronology one

adopts, we can conclude that in the central

zone of the Alps, east of our tin line, therewas a higher continuity of copper ratherthan bronze use. It is important to add tothis that there is an area of high bronzeuse on the eastern edge of our study area.This was probably influenced by CentralEurope and may be related to the alloyingpatterns of groups such as the UneticeCulture rather than the Alpine region.Finally, it is important to note that theAlps were not a north/south barrier to thepresence or absence of tin, but that thesame scenario appeared on both sides ofthe mountains.

TOPOGRAPHY: A KEY ELEMENT

As described above, the distribution ofdifferent groups of metal is not random,but shows a series of clear geographicalpatterns over time. Can we push this lineof reasoning beyond the concept of two-dimensional geography and deal withtopography? When we speak about metalflows, we are, ultimately, referring to themovement of people, which, of course, is

Figure 10. Distribution map of Copper Group 16, copper with arsenic, antimony, silver and nickel inthe first phase of the Early Bronze Age (A1). The size of the dots indicates the number of objects. Thecolour scale represents the percentage of each geographical square’s metal assemblage that belongs toCopper Group 16.



influenced by the topography of the terri-tory. One individual need not accompanyobjects through their entire journey, butaccess points and mutual meeting placesmust have been essential. In the Alps,attention naturally turns to the passesbetween the mountains, but it is importantto connect these with their associatedwater and valley system. Rivers mayreasonably be described as the ‘highwaysof the continent’ (Cunliffe, 2008: 38–47)in prehistory. Large rivers, such as theDanube, Rhine, Rhône, and Po wouldhave been easy to navigate, even upstream,especially with the help of sails or humanand animal traction (see Cunliffe’s (2008:45) considerations). As Van de Noort(2013: 390) claims, ‘There is no doubtingthat the earliest boats in Europe were builtfor use on rivers and lakes, and that thetypes of craft that enable riverine trafficinclude hide- and skin-covered boats andlogboats’. The oldest logboat from Italy isfrom Lake Bracciano, Central Italy, and isdated to the sixth millennium BC (Fugaz-zola & Mineo, 1995). However, most ofthe logboats found in Italy are from thenorth, from both lakes and rivers, and

some of them have been dated to theEarly Bronze Age. Unfortunately, there isa general lack of research on the chronol-ogy of these crafts as well as on theirsocial role and technological developmentthroughout the Bronze Age (Ravasi &Barbaglio, 2008). But even if they werenot navigated, the importance of rivers ishardly diminished: for terrestrial travel,rivers work perfectly as reference pointsand as a secure marker of the path, and asa reliable source of potable water. The keyrole of rivers increases in the Alpineregion, where glacial river valleys work ascorridors between mountains, offering inenergy terms the least costly transalpinepaths.Using GIS we tested the hypothesis

that different metal groups were likely tofollow the flow of different rivers. UsingArcGIS there is the facility to draw thewatershed of each river, and since eachartefact is linked to a site and, ultimately,to its coordinates, it was therefore possibleto create a spatial join between each arte-fact and the watershed to which itbelongs. Thus, for each artefact, two kindsof information are provided: the metal

Figure 11. Distribution map of Copper Group 1, copper without traces elements in the first phase ofthe Early Bronze Age (A1). The size of the dots indicates the number of objects. The colour scale rep-resents the percentage of each geographical square’s metal assemblage that belongs to Copper Group 1.



compositional group and the watershed towhich it belongs. Therefore, it waspossible to undertake a statistical analysisto verify the correlation between water-sheds and compositional groups. The χ2

test suggests that the distribution of differ-ent metal groups was not random withrespect to watershed, neither in theCopper Age nor in the Bronze Age A1,with a p-value of less than 0.005 in bothcases.As summarised by Figure 15, in the

Copper Age, the Rhône watershed wascharacterised by a dominance of arsenicalcopper (Group 2); the Adige and Po rivershave similar levels of pure copper (Group1), but they differ in the percentage ofGroups 5, 7 and 9 present. The Rhine,Danube, and Piave watersheds have sig-nificantly higher levels of pure copper thanthose of the Rhône, Adige, and Po(Figure 15). Two points should be high-lighted: the specific presence of Group 5(Ni) in the Rhine and the high values ofGroup 9 (As, Sb) in the Piave, which,combined with the percentages in Po andAdige, reflect the previously mentioned

importance of this group in the north-eastof Italy.In the initial Early Bronze Age

(Figure 16), the specific presence of differ-ent copper groups per watershed becomesmore evident, with the exception of thePo and Rhine, which have extremelysimilar patterns. The importance of copperwithout impurities (Group 1) in theRhône region is confirmed, whereas theDanube has a predominance of Group 12(As, Sb, Ag), while Group 16 (As, Sb,Ag, Ni) dominates the Rhine and Powatersheds.It is interesting to note that the Rhône

and Rhine had different patterns of coppergroup presence, because the tin line runsbetween these two rivers. In the Rhône,tin is present in all three dominant coppergroups, 1, 12, and 16, while tin is gener-ally absent from the Rhine’s metalartefacts made of the same three coppergroups. Copper was flowing across thisline, Group 1 from the west, and Group16 from the east into the west. However,tin does not accompany this copper move-ment, remaining prominent only in the

Figure 12. Map of the mean percentage of tin per site in the first phase of the Early Bronze Age (A1),using Krause (2003, appendix) chronology.



western Alps. Moreover, we should notethat in the initial Early Bronze Age thehigher similarity of the patterns of theRhine and Po compared to the CopperAge allows us to hypothesise close transal-pine contacts and exchange of metalbetween these two areas.

DISCUSSION

The transition from the Copper Age tothe Early Bronze Age marked a change inAlpine metal production: from local pro-duction with a great variety of coppertypes in circulation, to a picture of largeflows of metal of specific types. In particu-lar, three types of copper were verycommon in the first phase of the EarlyBronze Age: copper with As, Sb, and Ag(Group 12), copper with As, Sb, Ag, andNi (Group 16), and copper with no traceelements (Group 1). This concurs withthe conclusions of Krause (2002). Apartfrom Group 1, these are new dominantgroups, indicating that there was a new

way of procuring metal from the CopperAge to the Early Bronze Age, with thepossibility of new sources being exploited.Though each of these groups tend to havea zone where they dominate the artefactassemblage, it is interesting to note thatthere are several examples, especially inhoards, of objects that have a local typol-ogy but which are made from copper withthe signature typical of another part of theAlps. These deserve more detailed studyas they may provide the clearest casestudies of how metal was moving throughand around the Alpine zone.The Early Bronze Age may also have

seen the establishment of an important tinline running north-south across the Alps,from Oglio and Adda in northern Italy toReuss and Rhine in Switzerland, with tin-bronze far more prevalent in the west thanin the east. Although this line comes froma fresh analysis of the metal data, itmirrors well-known cultural boundariesand contact networks: in the north alpineregion the tin line corresponds well to theboundary between the Early Bronze Age

Figure 13. Distribution map of tin-bronze presence, and the mean percentage of tin per site, in thesecond phase of the Early Bronze Age (A2a), using David-Elbiali (2000: 389–519) and De Marinis’s(2005: 249–64; 2006a: 212–72; 2006b: 1289–317) chronology. The colour scale represents the percen-tage of each geographical square’s metal assemblage that is bronze.



Rhône culture in the west and the Blechk-reis culture in the east (Merkl, 2011:fig. 4.8). But it is also a witness to theimportance of river systems: the westernpart was connected with the Rhône valley,whereas the eastern part communicatedpreferentially with the Rhine and, ulti-mately, the Danube watersheds. Thequestion of the source of this tin is stillopen. Based on the analytical work of theSAM group (Junghans et al., 1960, 1968a,1968b), we may postulate that there wasan early movement of tin as an alloyingmaterial from France up the Rhône, but toidentify the source of this flow requiresfurther research.Even more interesting is the fact that

this tin line continues to the south of theAlps. In contrast with the river systemmodel for the north of the Alps, the lineneatly cuts the River Po, which is navig-able, into two discrete areas. However, asin the northern Alps, the tin line marks theexpression of cultural differences. In 1989,Peroni stated: ‘It may seem that, regardlessof the Alpine barrier, and more importantthan it, a cultural barrier [...] cut northern

Italy transversely from North to South.’(Peroni, 1989: 364–65). He was referringto the Middle and Late Bronze Age, andin particular to metallurgy from a typologi-cal perspective. Since then, the idea of acultural barrier has appeared constantly inItalian studies discussing the MiddleBronze Age and later periods. Forexample, two separate spheres of exchange,the first linking the south-western with thenorth-western Alps and the second linkingthe south-eastern with the north-easternAlps, have been recognised by De Marinis(2006a) based on the find-spots of certaintypes of swords, and similar observationswere put forward by Gambari (1998) andBaioni (2008). Two points are of specialinterest here: firstly, that the east-west sep-aration of northern Italy was extremelylong-standing, since it remained visiblefrom the mid-second millennium to themid-first millennium BC, and perhaps forlonger; secondly, that it was initiallymarked by the manufacture and circulationof metalwork, but quickly became visible inother domains of social action, includingpottery production and use.

Figure 14. Map of the mean percentage of tin per site for only ornaments. Second phase of the EarlyBronze Age (A2a), using David-Elbiali (2000: 389–519) and De Marinis’s (2005: 249–64; 2006a:212–72; 2006b: 1289–317) chronology.



The results of our research raise thequestion as to whether this ‘culturalbarrier’ emerged in the Middle BronzeAge following the patterns of tin exchangehighlighted above, or whether metalmovement in the EBA was dependent onalready existing barriers. Significantly, aneast-west division first emerged in north-ern Italy during the Late and FinalNeolithic (c. 4500–3600 BC), when theappearance of the Chassey-Lagozzapottery style in the north-west pushed theboundaries of the Square-MouthedPottery style, previously found all over theregion, further east. Whereas the formerstyle shows strong connections withsouthern France and the western Alps, thelatter is linked to cultural development inthe north-eastern Alps and Slovenia; thefrontier between the two areas, which isnot clear-cut, runs from the westernshores of Lake Garda to the northernApennines, slightly to the east of the cul-tural boundaries and circulation zonesvisible in the Bronze Age (Barfield, 1996:67; Mottes & Nicolis, 2002; Pessina &

Tiné, 2008: 99, fig. 3c–d). It is alsoworth noting that, in the same timeperiod, the area encompassing the Rhônewatershed, Piedmont, and Liguria, lay atthe core of the Europe-wide,westward-looking network of polishedstone exchange (Barfield & Broglio, 1966:63–65; Bouard, 1993; Bouard & Fedele,1993; Gambari & Aimar, 1996; Pétrequinet al., 2005, 2012).The picture is less clear in the northern

Italian Copper Age (c. 3600–2200 BC),which is characterised by the break-downof the extensive pottery styles of the LateNeolithic. At domestic sites, these werereplaced by a plethora of ceramic tra-ditions and decorative motifs, which wereoften re-elaborated and re-combined inoriginal ways even at sites lying at shortdistance from one another (Genick, 2012:498). In contrast, funerary sites featuresignificant super-regional similarities inburial behaviour and grave goods, whicharguably reflect new, widespread ideasconcerning the treatment of the dead andthe importance of the ancestors (Dolfini,

Figure 15. Percentage presence of each metal group per watershed in the Copper Age.



2004, in press; Barfield, 2007; Fedele,2013). Yet, even in the absence of clearmarkers of long-ranging exchange, thesurvival of the previous east-west divisionis hinted at by the directional movementof raw materials, goods and ideas acrossthe Alps. In the western quadrant, certainshapes and decoration styles are found inpottery from Piedmont, eastern Franceand Switzerland (Gambari, 1998: 52–56);in the eastern quadrant, the trans-alpinenetworks emerging in the Final Neolithicfor the exchange of the earliest metalobjects were further developed in theCopper Age to include north-east Italy,Austria and Slovenia (Carancini, 2001;Visentini, 2009; Klassen, 2010; Dolfini,2013). It is thus suggested that the circula-tion patterns of copper and tin highlightedin this study may reflect enduring culturaldemarcations, which date back to the latefifth millennium BC.We have shown here that the alloy of

copper with tin seems to have been firstadopted in the western part of the Alps,

or, at least, contemporaneously with thefar eastern edge of the study area, leavinga gap in the central part of the Alps, thussuggesting some level of continuity fromthe Neolithic through to the Iron Age andbeyond. Remarkably, in the Alps tin wasconsistently added to all of the coppergroups that were in circulation, includingpure copper and copper with small percen-tages of arsenic, antimony silver andnickel. This is very important, because onthe one hand it suggests that tin wasmoving separately to the copper and wasbeing alloyed locally (as opposed tocopper-tin alloys arriving pre-made from asingle remote source), and on the otherhand it may say something about thereason why people decided to add tin. Thepresence of trace elements, especially whencombined together, increases the ductility,malleability and also hardness of the fin-ished products. In addition, it lowers themelting point of copper (Northover, 1989;Kienlin et al., 2006; Merkl, 2011: 77–78;64, fig. 7.1). Nevertheless, the addition

Figure 16. Percentage presence of each metal group per watershed in the first phase of the EarlyBronze Age (A1).



of tin seemed not to take account of this,since it was added both to pure copperand to copper with significant levels oftrace elements. It therefore seems likelythat the addition of tin was made, at leastin part, for non-functional reasons (a‘recipe’, a ‘ritual’, or ‘a good way to makemetal’), implying that perhaps colour anddisplay were influencing factors. Onecould speculate that it would have made apowerful statement to be able to obtainand use such a rare material, particularlyone that could be displayed so strikingly.This reading is also supported by the factthat the most clear-cut difference in theadoption of tin between the western andeastern areas is seen in ornaments(Figure 14). Identifying these socialchoices, in part through artefact chemistry,underlines the importance of linkingdifferent datasets together. As Pearce(1998, 2007) laments, collections ofarchaeometallurgical data tend to beunderused, often being limited to consul-tation when considering provenance ortechnology. In this case, however, wemanage to have a hint of ‘what actuallyhappened’ (Pearce, 1998: 53).The idea that the use of copper-tin

alloy spread throughout Europe from theNear East and Anatolia has long been theaccepted model (e.g. Pare, 2000; Robertset al., 2009). However, the early use of tinin the western part of the Alps perhapsreflects more detailed local patterns ofmetal movement, which may occasionallygo against a simplistic radiating patternfrom a single area or region. Withoutdenying the overall westward-spreadingmodel, Primas (2003) highlighted that inthe south-east of Europe tin alloying wascontemporary with lead alloying, whereasin central Europe tin alloying was absol-utely predominant and lead was not usedregularly until the Late Bronze Age,despite there being accessible sources oflead. According to her model, in central

Europe there was a first phase in which‘selective use’ of tin (Primas, 2003: 88) mayhave come from the south-east andinvolved a small number of objects. In alater phase, however, tin alloying becamecommon practice in the British Isles,where abundant sources of the metal wereavailable in Cornwall and Devon; thencethe new technology would have spread toCentral Europe, perhaps taking advantageof the old communication channels pro-vided by the Bell-Beaker phenomenon.Primas’ (2003) model provides a plausibleexplanation for the patterns highlighted inthis research, and in particular for the earlyappearance of tin in the western Alps.

CONCLUSION

In conclusion, through linking the chemi-cal composition, location, and typology ofearly metal artefacts in the Alpine region,we have shown that the Alps did notprovide an insurmountable north-southbarrier to the circulation of metalwork, ascopper with similar impurity patterns arefound on both sides of the range. Simi-larly, the spread of tin was not affected bythe topography of the mountain range.Compositional copper groups were notrandomly distributed, but reflected ore dis-tribution, river flows and their watersheds.Hence, it seems that river systems deeplyinfluenced the movement of raw materialsand ideas, and, probably, of people.Occasionally though cultural barriers canbe demonstrated to be a more powerfulinfluence, as is the case with the tin linecross-cutting the Po river.If the concept of the Alps as a physical

barrier is rejected, we can recogniseinstead an important cultural barrier thatcreated an east-west division in the flow ofmetalwork. In the north of the Alps, thisreflects the river system, whereas in north-ern Italy, it cuts the Po valley in two,


approximately between the Oglio andAdda rivers. This is important because thePo, which is a navigable river, becomes aremarkable exception to the river systemtransport network hypothesised by Cun-liffe (2008) and seen north of the Alps.This barrier mirrors perfectly patterns seenin the typological study of Middle andLate Bronze Age metal artefacts, but thenew synthesis presented here allows us topush its origin back in time to the EarlyBronze Age and perhaps earlier. It mayeven be speculated that its origin isgrounded in surprisingly long-standingNeolithic exchange networks and thus pre-dates the emergence of metallurgy in theregion.A final observation must be made about

the existence of different chronologies forthe region in question. A unified chrono-logical framework for the Alps is still workin progress: the present situation is notideal for a synthetic study such as the onecarried out here. Further research isrequired, whose ideal outcome is an inte-grated chronology for the entire Alpineregion, underpinned by new radiocarbondates directly associated with metalobjects. Nevertheless, we are confidentthat the major patterns described here areso fundamental to the flow of metals inthe Alpine Copper and Early Bronze Agesthat they will probably survive any futurerevision of the local sequences. It is onlywhen a new, finer-grained chronologybecomes available that the scientific,archaeological and geographical datasetsdiscussed in this article can be cross-referenced to give a more nuanced insightinto people’s relationship with metal andtheir environment in the Alpine region.

ACKNOWLEDGMENTS

The authors are greatly indebted to JohnPouncett for his help with ArcGIS, and to

Peter Northover for sharing his data andknowledge about ancient metallurgy.Thanks also to Annaluisa Pedrotti for hervaluable help with defining the chronologyof the objects analysed by Peter Northover.We are also grateful to Denis Ramseyer,the director of the Museum of Neuchatel,for his assistance and advice on theproject. This study has been supported bythe John Fell Fund (University of Oxford)and the Leverhulme Trust (Grant F/08622/D). Laura Perucchetti gratefullyacknowledges the support given bySt. Peter’s College, Oxford, the SidneyPerry foundation, and the HistoricalMetallurgy Society. Peter Bray is amember of the Atlantic Europe in theMetal Ages project (AHRC grant AH/K002600/1).

REFERENCES

Ambert, P. 1995. Les mines préhistoriques deCabrières (Hérault) : quinze ans derecherches. état de la question. Bulletin dela Société Préhistorique Française, 92:499–508.

Angelini, I. 2004. Studio archeometallurgicodi un ornamento a spirale dal Riparo diPeri (Verona). Padusa, 40:39–50.

Angelini, I. 2007. Indagini archeometrichedell’ascia di Caprino Veronese. QuaderniCulturali Caprinesi, 2:34–41.

Angelini, I. & Artioli, G. 2007. Le armi eneo-litiche di Spessa (Cologna Veneta) e diGambarella (Povegliano): indagini chi-miche e tessiturali. Bollettino Del MuseoCivico di Storia Naturale di Verona, 31:51–61.

Angelini, I., Polla, A., Perucchetti, L., deMarinis, R.C. & Artioli, G. 2010. Analisiarcheometallurgiche di asce e pugbnali delbronzo antico-bronzo medio provenientidall’Italia Settentrionale. presented at theVI Congresso nazionale di Archeometria,Scienza e Beni Culturali, Pavia, unpub-lished paper.

Artioli, G., Dugnani, G., Angelini, I.,Lutterotti, L., Pedrotti, A. & Fleckinger,A. 2003. Early Copper Alpine Metallurgy.


In: Proceedings of the InternationalConference on Archaeometallurgy inEurope. pp. 24–26.

Baioni, M. 2008. Tra Oglio e Adda. Terra dicommerci e di confine nella Tarda età delBronzo. In: M. Baioni & C. Fredella,eds., Archaeotrade. Antichi Commerci inLombardia Orientale. Milano: ET, pp. 63–74.

Barfield, L. 1995. Monte Covolo : una mon-tagna e il suo passato : ricerche archeologiche1972–1994. Birmingham: FieldArchaeology Unit, University ofBirmingham.

Barfield, L. 1996. The Chalcolithic in Italy:Considerations of Metal Typology andCultural Interaction. In: B. Bagolini & F.Lo Schiavo, eds. The Copper Age in theNear East and Europe. 13th Congress ofthe International Union of Prehistoric andProtohistoric Sciences, Colloquium XIX.Forlì: Abaco, pp. 65–74.

Barfield, L. 2007. Copper Age Pottery. In: L.Barfield, ed. Excavations in the RiparoValtenesi, Manerba, 1976–1994. Firenze:Istituto italiano di Preistoria e Protostoria,pp. 159–216.

Barfield, L. & Broglio, A. 1966. Materiali perlo Studio del Neolitico del TerritorioVicentino. Bullettino Paletnologia Italiana,75:51–95.

Barfield, L. & Kuniholm, P.I. 2007.Radiocarbon Dating and the AbsoluteChronology of the Cemetery. In: L.Barfield, ed. Excavations in the RiparoValtenesi, Manerba, 1976–1994. Firenze:Istituto italiano di Preistoria e Protostoria,pp. 419–27.

Barker, G. 1971. Analyses of Samples fromthe Pigorini Museum. Bullettino diPlaetnologia Italiana, 80:183–212.

Beeley, P. 2001. Foundry Technology. Oxford:Butterworth-Heinemann.

Bosio, L. & Calzavara, C. L. 1990. Veneto(Italy). Segretaria regionale per il territorio.Carta archeologica del Veneto : carta d’ItaliaIGM 1:100.000. Venezia: F.C. Panini.

Bouard, M. 1993. Trade in Neolithic JadeiteAxes from the Alps: New Data. In: C.Scarre & F. Healy, eds. Trade andExchange in Prehistoric Europe : Proceedingsof a Conference Held at the University ofBristol, April 1992. Oxford: Oxbow Booksin Association with the Prehistoric Societyand the Société Préhistorique Française,pp. 61–67.

Bouard, M. & Fedele, F.G. 1993. NeolithicRock Resources across the Western Alps:Circulation Data and Models.Geoarchaeology, 8:1–22.

Bourgarit, D. & Mille, B. 2005. Les NouvellesDonnées de l’Atelier MétallurgiqueChalcolithique de La Capitelle du Broumdans le District de Cabrières (Hérault): LaTransformation des Minerais de Cuivre àBase de Sulfures se Précise. In: P. Ambert& J. Vaquer, eds. La Première MétallurgieEn France et Dans Les Pays Limitrophes.Toulouse: Société Préhistorique Française,pp. 97–108.

Bray, P.J., Blockley, S.P.E., Coope, G.R.,Dadswell, L.F., Elias, S.A., Lowe, J.J. &Pollard, A.M. 2006. Refining MutualClimatic Range (MCR) QuantitativeEstimates of Palaeotemperature UsingUbiquity Analysis. Quaternary ScienceReviews, Quaternary Beetle Research: TheState of the Art, 25:1865–76.

Bray, P. & Pollard, A.M. 2012. A NewInterpretative Approach to the Chemistryof Copper-Alloy Objects: Source,Recycling and Technology. Antiquity,86:853–67.

Budd, P. 1991. A MetallographicInvestigation of Eneolithic ArsenicalCopper. Unpublished PhD thesis,University of Bradford.

Carancini, G.L. 1996. La metallurgia dell’an-tica Età del Bronzo. In: D.C. Genick, ed.L’antica età del Bronzo : atti del congresso diviareggio, 9–12 gennaio 1995. Octavo,F. Firenze: Cantini editore, pp. 33–56.

Carancini, G.L. 2001. Origini e Sviluppi dellametallurgia in Toscana nell’ambito dellefasi più antiche della protostoria. In: Attidella XXXIV Riunione Scientificadell’Istituto Italiano di Preistoria eProtostoria. Firenze: Istituto italiano dipreistoria e protostoria, pp. 235–49.

Carancini, G.L. & Peroni, R. 1999. L’Età delBronzo in Italia: Per una Cronologia dellaProduzione Metallurgica. Perugia: Alieno.

Casini, S. 1994. Carta archeologica dellaLombardia. 4, La provincia di Lecco.Modena: F.C. Panini.

Cattin, F. 2008. Modalités d’approvisionne-ment et modalités de consommation ducuivre dans les Alpes au 3e millénaire avantnotre ère: apport des analyses métalliques àla connaissance des peupleuments duNéolitique final, du Campaniforme et du


http://www.maneyonline.com/action/showLinks?crossref=10.1002%2Fgea.3340080102

http://www.maneyonline.com/action/showLinks?crossref=10.1002%2Fgea.3340080102

http://www.maneyonline.com/action/showLinks?crossref=10.1017%2FS0003598X00047967&isi=000308514600018

Bronze ancient. Unpublished PhD thesis,Université Genève.

Cattin, F., Guénette-Beck, B., Curdy, P.,Meisser, N., Ansermet, S., Hofmann, B.,Kündig, R., Hubert, V., Wörle, M.,Hametner, K., Günther, D., Wichser, A.,Ulrich, A., Villa, I.M. & Besse, M. 2011.Provenance of Early Bronze Age MetalArtefacts in Western Switzerland UsingElemental and Lead IsotopicCompositions and their Possible Relationwith Copper Minerals of the nearbyValais. Journal of Archaeological Science,38:1221–33.

Charles, J.A. 1980. Recycling Effects on theComposition of Non-Ferrous Metals.Philosophical Transactions of the RoyalSociety of London A, 295:57–68.

Colini, G.A. 1907. Il Sepolcreto di RemedelloSotto nel Bresciano e il Periodo Eneoliticoin Italia. Bullettino di Paletnologia Italiana,26:203–67.

Cunliffe, B.W. 2008. Europe Between theOceans : Themes and Variations : 9000 BC–AD 1000. London: Yale University Press.

David-Elbiali, M. 2000. La Suisse Occidentaleau IIe millénaire av. J.-C. : Chronologie,Culture, Intégration européenne. Lausanne:Cahiers d’archéologie romande.

Della Casa, P. 2013. Switzerland and theCentral Alps. In: A.F. Harding & H.Fokkens, eds. The Oxford Handbook of theEuropean Bronze Age. Oxford: OxfordUniversity Press, pp. 706–22.

De Marinis, R.C. 1997. The EneolithicCemetery of Remedello Sotto (BS) andthe Relative and Absolute Chronology ofthe Copper Age in Northern Italy. NotizieArcheologiche Bergomensi, 5:33–51.

De Marinis, R.C. 2005. Évolution et variationde la composition chimique des objets enmétal pendant les âges du Cuivre et duBronze Ancien dans l’Italie septentrionale.In: P. Ambert & J. Vaquer, eds. Lapremière métallurgie en France et dans lespays limitrophes : actes du colloque inter-national, Carcassonne, 28–30 septembre2002. Toulouse: Société préhistoriquefrançaise, pp. 249–64.

De Marinis, R.C. ed. 2006a. Aspetti dellametallurgia dell’età del Rame dell’anticaetà del Bronzo nella penisola italiana.Rivista di Scienze Preistoriche, 56:212–72.

De Marinis, R.C. 2006b. Circolazione delmetallo e dei manufatti nell’età del Bronzo

dell’Italia settentrionale. In: Atti dellaXXXIX Riunione Scientifica dell’IstitutoItaliano di Preistoria e Protostoria, Materieprime e scambi nella preistoria italiana,Firenze, 25–27 novembre 2004. Firenze:Istituto italiano di preistoria e protostoria,pp. 1289–317.

De Marinis, R.C. 2013. La necropoli diRemedello Sotto e l’età del Rame nellapianura padana a nord del Po. In: R.C.De Marinis, ed. L’età del rame. La pianurapadana e le Alpi al tempo di Otzi. Brescia:Compagnia della Stampa, pp. 301–52.

De Marinis, R.C. & Brillante, G. 1998. Ötzi,l’Uomo Venuto dal Ghiaccio. Padova:Marsilio.

Dolfini, A. 2004. La Necropoli di Rinaldone(Montefiascone, Viterbo): Rituale Funerarioe Dinamiche Sociali di una ComunitàEneolitica in Italia Centrale. Bullettino diPaletnologia Italiana, 95:127–278.

Dolfini, A. 2010. The Origins of Metallurgyin Central Italy — New RadiometricEvidence. Antiquity, 84:707–23.

Dolfini, A. 2013. The Emergence ofMetallurgy in the Central MediterraneanRegion: A New Model. European Journalof Archaeology, 16:21–62.

Dolfini, A. 2014a. Early Metallurgy in theCentral Mediterranean: Goals for the NextDecade. Historical Metallurgy, 47:33–50.

Dolfini, A. 2014b. Early Metallurgy in theCentral Mediterranean. In: B.W. Roberts& C.P. Thornton, eds. Archaeometallurgyin Global Perspective. New York: Springer,pp. 473–506.

Dolfini, A. in press. Neolithic andChalcolithic Mortuary Practices in theItalian Peninsula: Change of Meaning orChange of Medium? In: J.R. Brandt, H.Ingvaldsen & M. Prusac, eds. Death andChanging Rituals: Function and Meaningin Ancient Funerary Practices. Oxford:Oxbow Books.

Fedele, F.G. 2013. La società dell’età delRame nell’area alpina e prealpina. In: R.C.De Marinis, ed. L’età del rame. La pianurapadana e le Alpi al tempo di Ötzi. Brescia:Massetti-Rodella, pp. 45–59.

Fugazzola, M.A. & Mineo, M. 1995. Lapiroga neolitica del lago di Bracciano (‘LaMarmotta 1’). Bullettino di PaletnologiaItaliana, 86:197–266.

Guidi, A. 2000. Preistoria della complessitàsociale. Rome: Laterza.



http://www.maneyonline.com/action/showLinks?crossref=10.1098%2Frsta.1980.0075&isi=A1980JG82600005

http://www.maneyonline.com/action/showLinks?crossref=10.1098%2Frsta.1980.0075&isi=A1980JG82600005

http://www.maneyonline.com/action/showLinks?system=10.1179%2F1461957112Y.0000000023&isi=000316004200003

http://www.maneyonline.com/action/showLinks?system=10.1179%2F1461957112Y.0000000023&isi=000316004200003

http://www.maneyonline.com/action/showLinks?crossref=10.1016%2Fj.jas.2010.12.016&isi=000290420300007

Guilaine, J., Claustre, F., Lemercier, O. &Sebatier, P. 2001. Campaniforme etenvironment culturel en FranceMéditeranéenne. In: F. Nicolis, ed. BellBeakers Today: Pottery, People, Culture,Symbols in Prehistoric Europe:Proceedings of the InternationalColloquium, Riva del Garda (Trento, Italy)11–16 May 1998. Trento: ProvinciaAutonoma di Trento, Servizio BeniCulturali, Ufficio Beni Archeologici, pp.229–75.

Hafner, A. 1993. Suisse occidentale et Valais.In: M. Flüeler-Grauwiler, ed. Die Schweizvom Paläolithikum bis zum frühenMittelalter: vom Neandertaler zu Karl demGrossen = La Suisse du paléolithique à l’aubedu Moyen-Age : de l’homme de Néandertal àCharlemagne = La Svizzera dal paleoliticoall’Alto Medio Evo: dall’uomo di Neandertala Carlo Magno. Basel: VerlagSchweizerische Gesellschaft für Ur- undFrühgeschichte, pp. 20–31.

Hafner, A. & Suter, P. 2007. VomEndneolithikum zur Frübronzezeit wandelund kontinuität zwischen 2400 un 1500v. Chr in der Schweiz. In: H. Richard, M.Magny & C. Mordant, eds.Environnements et cultures à l’âge dubronze en Europe occidentale : [actes desCongrès nationaux des sociétés historiques etscientifiques, 129e, Besançon, 2004].Presented at the Congrès national des sociétéshistoriques et scientifiques (129th : 2004 :Besançon). Paris: Editions du Comité destravaux historiques et scientifiques, pp.179–195.

Hampton, D.F.G., Bennet, P., Brown, D.,Lancaster, R., Rice, J.L., Sharp, A.L.,Stephens, H.A. & Bidwell, H.T. 1965.Metal Losses in Copper-Base Alloys. TheBritish Foundryman, 58:225–40.

Heyd, V. 2013. L’Europa nell’età del Rame: la‘calcolitizzazione’ di un continente. In: R.C. De Marinis, ed. L’Età del Rame: LaPianura Padana e le Alpi al Tempo di Ötzi.Brescia: Compagnia della Stampa, pp. 23–38.

Hook, D.R. 2003. The Composition andTechnology of Early Copper-AlloyMetalwork from Italy. In: InternationalConference: Archaeometallurgy in Europe,24–25–26 September 2003, Milan, Italy :Proceedings. Milano: Associazione italianametallurgia, pp. 57–66.

Höppner, B., Bartelheim, M., Huijsmans, M.,Krause, R., Martinek, K.-P., Pernicka, E.& Schwab, R. 2005. Prehistoric CopperProduction in the Inn Valley (Austria),and the Earliest Copper in CentralEurope. Archaeometry, 47:293–315.

Junghans, S., Sangmeister, E. & Schröder, M.1960. Metallanalysen kupferzeitlicher undfrühbronzezeitlicher Bodenfunde aus Europa,Studien zu den Anfängen der Metallurgie.Gebr. Berlin: Mann.

Junghans, S., Sangmeister, E. & Schröder, M.1968a. Kupfer und Bronze in der frühenMetallzeit Europas, Studien zu denAnfängen der Metallurgie. Gebr. V 1Berlin: Mann.

Junghans, S., Sangmeister, E. & Schröder, M.1968b. Kupfer und Bronze in der frühenMetallzeit Europas, Studien zu denAnfängen der Metallurgie. Gebr. V 2Berlin: Mann.

Junghans, S., Sangmeister, E. & Schröder, M.1974. Kupfer und Bronze in der frühenMetallzeit Europas, Studien zu denAnfängen der Metallurgie. Gebr. V 3Berlin: Mann.

Kienlin, T.L. 2008. Frühes Metall im nordal-pien Raum : eine Untersuchung zutechnologischen und kognitiven Aspektenfrüher Metallurgie anhand der Gefügefrühbronzezeitlicher Beile. Bonn: RudolfHabelt.

Kienlin, T.L. 2010. Traditions andTransformations: Approaches to Eneolithic(Copper Age) and Bronze Age Metalworkingand Society in Eastern Central Europe andthe Carpathian Basin. BritishArchaeological Reports InternationalSeries 2184. Oxford: Archaeopress.

Kienlin, T.L. 2013. Copper and Bronze:Bronze Age Metalworking in Context. In:A.F. Harding & H. Fokkens, eds. TheOxford Handbook of the European BronzeAge. Oxford: Oxford University Press, pp.414–36.

Kienlin, T.L., Bischoff, E. & Opielka, H.2006. Copper and Bronze during theEneolithic and Early Bronze Age: AMetallographic Examination of Axes fromthe Northalpine Region. Archaeometry,48:453–68.

Kienlin, T.L. & Stöllner, T. 2009. SingenCopper, Alpine Settlement and EarlyBronze Age Mining: Is There a Need forElites and Strongholds?. In: B.W. Roberts


http://www.maneyonline.com/action/showLinks?crossref=10.1111%2Fj.1475-4754.2006.00266.x&isi=000239255900005

http://www.maneyonline.com/action/showLinks?crossref=10.1111%2Fj.1475-4754.2005.00203.x&isi=000229975400008

& T.L. Kienlin, eds. Metals and Societies:Studies in Honour of Barbara S. Ottaway.Bonn: Rudolf Habelt, pp. 67–104.

Klassen, L. 2010. Karpaten oder Alpen? ZurHerkunft der Kupferscheibe ausHornstaad (Lkr. Konstanz). ArchäologischesKorrespondenzblatt, 40:29–48.

Krause, R. 1989. Early Tin and CopperMetallurgy in South-Western Germany atthe Beginning of the Early Bronze Age.In: A. Hauptmann, E. Pernicka & G.A.Wagner, eds. Archäometallurgie der AltenWelt : Beiträge zum InternationalenSymposium ‘Old World Archaeometallurgy’,Heidelberg 1987. Bochum: Selbstverlag desDeutschen Bergbau-Museums, pp. 25–32.

Krause, R. 2002. The Cultural andChronological Context of the Developmentof the Early Bronze Age Metallurgy inCentral Europe. In: B. Ottaway & E.C.Wager, eds. Metals and Society: Papers froma Session Held at the European Association ofArchaeologists Sixth Annual Meeting inLisbon 2000. British Archaeological ReportsInternational Series 1061. Oxford:Archaeopress, pp. 33–44.

Krause, R. 2003. Studien zur kupfer- undfrühbronzezeitlichen Metallurgie zwischenKarpatenbecken und Ostsee.Vorgeschichtliche Forschungen 24.Leidorf: Rahden.

Lichardus, J. 1991. Kupferzeit als historischeEpoche: Versuch einer Deutung. In: J.Lichardus, ed. Die Kupferzeit Als HistorischeEpoche. Bonn: Habelt, pp. 763–800.

Lutz, J. & Pernicka, E. 1996. EnergyDispersive X-Ray Fluorescence Analysis ofAncient Copper Alloys: Empirical Valuesfor Precision and Accuracy. Archaeometry,38:313–23.

Marchesetti, C. 1889. Ricerche preistorichenelle caverne di S. Canziano pressoTrieste. Bollettino della Società Adriatica diScienze Naturali in Trieste, 11:1–19.

Matteoli, L. & Storti, C. 1974.Metallographic Research on Four PureCopper Flat Axes and One RelatedMetallic Block from an Eneolithic ItalianCave. Historical Metallurgy, 16:65–69.

Matuschik, I. 2004. Kupfer derGlockenbecherkultur im Nordalpengebiet:zur Endneolitischen Kupfergewinnun inden nordöstlichen Alpen. In: G.Weisgerber & G. Goldenberg, eds.

Alpenkupfer — Rame delle Alpi. Bochum:Deutsches Bergbau-Museum, pp. 285–302.

Mazzieri, P. & Dal Santo, N. 2007. Il Sitodel Neolitico Recente di Botteghino(Parma). Rivista di Scienze Preistoriche,57:113–38.

McKerrell, H. & Tylecote, R.F. 1972.Working of Copper-Arsenic Alloys in theEarly Bronze Age and the Effect on theDetermination of Provenance. Proceedingsof the Prehistoric Society, 38:209–18.

Merkl, M.B. 2011. Bell Beaker Copper Use inCentral Europe: A Distinctive Tradition?British Archaeological ReportsInternational Series. 2267. Oxford:Archaeopress.

Mille, B. & Carozza, L. 2009. Moving intothe Metal Ages: The Social Importance ofMetal at the End of the Neolithic Periodin France. In: T.L. Kienlin & B.W.Roberts, eds. Metals and Societies: Studiesin Honour of Barbara S. Ottaway. Bonn:Rudolf Habelt, pp. 143–171.

Mottes, E. & Nicolis, F. 2002. RapportiCulturali Tra i Territori a Nord e a Suddelle Alpi Centrali Durante il Neolitico el’Età del Rame. In: G. Schnekenburger,ed. Attraverso Le Alpi: Uomini, Vie eScambi nell’Antichità. Stoccarda:Archäologisches LandesmuseumBaden-Württemberg, pp. 119–35.

Neugebauer, J.W. 1998. Bronzezeit inOstösterreich. Niederöst. St. Pölten-Wien:Presse, M.

Neuninger, H. & Pittioni, R. 1963.Frühmetallzeitlicher Kupferhandel imVoralpenland : zur Frage der Herkunft desRingbarren-Kupfers. Wien: F. Deuticke.

Nicolis, F. 2013. Northern Italy. In: A.F.Harding & H. Fokkens, eds. The OxfordHandbook of the European Bronze Age.Oxford: Oxford University Press, pp. 693–705.

Northover, P. 1980. The Analysis of WelshBronze Age Metalwork. In: H. Savory, ed.Guide Catalogue of the Bronze AgeCollections. Cardiff: National Museum ofWales, pp. 229–43.

Northover, P. 1989. Properties and Use ofArsenic-Copper Alloys. In: A. Hauptmann,E. Pernicka & G.A. Wagner, eds.Archäometallurgie Der Alten Welt : BeiträgeZum Internationalen Symposium ‘Old WorldArchaeometallurgy’, Heidelberg 1987.


http://www.maneyonline.com/action/showLinks?crossref=10.1017%2FS0079497X00012111

http://www.maneyonline.com/action/showLinks?crossref=10.1017%2FS0079497X00012111

http://www.maneyonline.com/action/showLinks?crossref=10.1111%2Fj.1475-4754.1996.tb00779.x

Bochum: Selbstverlag des DeutschenBergbau-Museums, pp. 111–18.

Ottaway, B. 1982. Earliest Copper Artifacts ofthe Northalpine Region : Their Analysis andEvaluation. Bern: Seminar fürUrgeschichte.

Otto, H. & Witter, W. 1952. Handbuch derältesten vorgeschichtlichen Metallurgie inMitteleuropa. Leipzig: J.A. Barth.

Pare, C.F.E. 2000. Bronze and the BronzeAge. In: C.F.E. Pare, ed. Metals Make theWorld Go Round: The Supply andCirculation of Metals in Bronze Age Europe.Proceedings of a Conference Held at theUniversity of Birmingham in June 1997.Oxford: Oxbow Books, pp. 1–38.

Pearce, M. 1998. Reconstructing PrehistoricMetallurgical Knowledge: The NorthernItalian Copper and Bronze Ages.European Journal of Archaeology, 1:51–70.

Pearce, M. 2007. Bright Blades and Red Metal :Essays on North Italian PrehistoricMetalwork. London: Accordia ResearchInstitute, University of London.

Pearce, M. 2009. How Much Metal WasThere in Circulation in Copper Age Italy?In: B.W. Roberts & T.L. Kienlin, eds.Metals and Societies: Studies in Honour ofBarbara S. Ottaway. Bonn: Rudolf Habelt,pp. 277–284.

Pedrotti, A. 2001. L’età del Rame. In: M.Lanzinger, F. Marzatico & A. Pedrotti,eds. La Preistoria e La Protostoria. StoriaDel Trentino. Bologna: Il Mulino-IstitutoTrentino di Cultura.

Pernicka, E. 1986. Provenance Determinationof Metal Artifacts: MethodologicalConsiderations. Nuclear Instruments andMethods in Physics Research Section B:Beam Interactions with Materials andAtoms, 14:24–29.

Pernicka, E. 1999. Trace ElementFingerprinting of Ancient Copper: AGuide to Technology or Provenance? In:S.M.M. Young, A.M. Pollard, P. Budd &R.A. Ixer, eds. Metals in Antiquity. BritishArchaeological Reports InternationalSeries. 792. Oxford: Archaeopress, pp.163–71.

Pernicka, E. 2011. A Short History ofProvenance Analysis of ArchaeologicalMetal Objects. In: A. Aspes, ed. I Bronzidel Garda: Valorizzazione delle Collezioni diBronzi Preistorici di Uno dei Più Importanti

Centri Metallurgici dell’Europa del IIMillennio a.C., Memorie del Museo Civicodi Storia Naturale di Verona / SezioneScienze dell’Uomo. Verona: Museo Civicodi Storia Naturale, pp. 27–37.

Pernicka, E. 2014. Provenance Determinationof Archaeological Metal Objects. In: B.W.Roberts & C.P. Thornton, eds.Archaeometallurgy in Global Perspective.London: Springer, pp. 239–68.

Pernicka, E. & Salzani, P. 2011. Remarks onthe Analyses and Future Prospects. In: Ibronzi del Garda: valorizzazione delle colle-zioni di bronzi preistorici di uno dei piùimportanti centri metallurgici dell’Europa delII millennio a.C., Memorie del Museo Civicodi Storia Naturale di Verona. Verona:Museo Civico di Storia Naturale, pp. 89–98.

Peroni, R. 1989. Protostoria dell’Italia continen-tale: la penisola Italiana nelle eta del bronzoe del ferro. Roma: Biblioteca de StoriaPatria.

Perucchetti, L. 2008. Studio archeometallur-gico di armi del Bronzo Anticoprovenienti dall’Italia Settentrionale.Unpublished MSc thesis, Università deglistudi di Milano.

Pessina, A. & Tiné, V. 2008. Archeologia delneolitico: l’Italia tra il VI e IV millennio a.C. Roma: Carocci.

Pétrequin, P., Cassen, S., Klassen, L. &Sheridan, A. 2012. Jade, grandes hachesalpines du néolithique européen, Ve et IVemillénaires avant J C : Tome 1 et 2.Besançon: Presses Universitaires deFranche-Comté.

Pétrequin, P., Errera, M., Cassen, S., Billand,G., Colas, C., Marechal, D. & Prodéo, F.2005. Des Alpes italiennes à l’Atlantiqueau Vè millénaire, les quatre grandes hachespolies de Vendeuil et Maizy (Aisne),Brenouille (Oise). In: G. Auxiette & F.Malrain, eds. Hommage À C. Pommepuy.Revue Archéologique de Picardie SpecialEdition 22. Amiens: Scociétés desAntiquités Historiques de Picardie, pp.75–104.

Pittioni, R. 1957. Urzeitlicher Bergbau aufKupfererz und Spurenanalyse: Beiträge zumProblem der Relation Lagerstätte-Fertigobjekt. Wien: F. Deuticke.

Poggiani Keller, R. 1992. Carta archeologicadella Lombardia. 2, La provincia diBergamo. Modena: F.C. Panini.


http://www.maneyonline.com/action/showLinks?crossref=10.1016%2F0168-583X%2886%2990418-0&isi=A1986AZE2700006




http://www.maneyonline.com/action/showLinks?system=10.1179%2Feja.1998.1.1.51

Prange, M. & Ambert, P. 2005. Caractérisationgéochimique et isotopique des minerais etdes métaux base cuivre de Cabrières(Hérault). In: P. Ambert & J. Vaquer, eds.La Première Métallurgie En France et DansLes Pays Limitrophes. Toulouse: SociétéPréhistorique Française, pp. 71–81.

Preuschen, E. & Pittioni, R. 1939.Untersuchungen im Bergbaugebiete Kelchalpebei Kitzbühel. Mitteilungen derPrähistorischen Kommission 3. Wien:R.M. Rohrer.

Primas, M. 2003. The Use of Tin and Lead inBronze Age Metallurgy. In: A.R.Giumlia-Mair & F. Lo Schiavo, eds. LeProblème de l’étain à l’origine de laMétallurgie. British ArchaeologicalReports International Series 1199. Oxford:Archaeopress, pp. 87–91.

Rageth, J. 1993. Grisons. In: M.Flüeler-Grauwiler ed. Die Schweiz vomPaläolithikum bis zum frühen Mittelalter:vom Neandertaler zu Karl dem Grossen = LaSuisse du paléolithique à l’aube duMoyen-Age : de l’homme de Néandertal àCharlemagne = La Svizzera dal paleoliticoall’Alto Medio Evo: dall’uomo di Neandertala Carlo Magno. Basel: VerlagSchweizerische Gesellschaft für Ur- undFrühgeschichte, pp. 42–45.

Ravasi, T. & Barbaglio, F. 2008. Merci ePersone sui Fiumi. Le imbarcazionimonossili conservate presso il MuseoCivico di Crema e del Cremasco. In: M.Baioni & C. Fredella, eds. Archaeotrade.Antichi Commerci in Lombardia Orientale.Milano: ET, pp. 37–61.

Reinecke, P. 1924. Zur chronologischenGliederung der süddeutschen Bronzezeit,Germania. Bamberg.

Roberts, B.W., Thornton, C.P. & Pigott,V.C. 2009. Development of Metallurgy inEurasia. Antiquity, 83:1012–22.

Roberts, B.W., Uckelmann, M. & Brandherm,D. 2013. Old Father Time: The BronzeAge Chronology of Western Europe. In:A.F. Harding & H. Fokkens, eds. TheOxford Handbook of the European BronzeAge. Oxford: Oxford University Press, pp.17–46.

Rossi, F. & Bishop, J. 1991. Carta archeologicadella Lombardia. 1: La Provincia di Brescia.Modena: F.C. Panini.

Rychner, V. 1995. Arsenic, nickel et antimoine:une approche de la métallurgie du bronze

moyen et final en Suisse, par l’analysespectrométrique. Lausanne: Cahiersd’archéologie romande.

Rychner, V. & Northover, P. 1998. BronzeAnalysis: Experience of a ComparativeProgramme. In: C. Mordant, M. Pernot& V. Rychner, eds. L’atelier du BronzierEn Europe Du XXe Au VIIIe Siècle AvantNotre Ère: Actes Du Colloque International‘Bronze ‘96’, Neuchâtel et Dijon, 1996.Paris: Comité des Travaux historiques &scientifiques, pp. 19–38.

Skeates, R. 1993. Early Metal-Use in theCentral Mediterranean Region. AccordiaResearch Papers, 4:5–48.

Strahm, C. 1994. Die Anfänge derMetallurgie in Mitteleuropa. HelvetiaArchaeologica, 25:2–39.

Strahm, C. 2005. L’introduction et ladiffusion de la métallurgie en France. In:P. Ambert & J. Vaquer, eds. La premièremétallurgie en France et dans les pays limit-rophes: actes du colloque international,Carcassonne, 28–30 septembre 2002.Toulouse: Société préhistorique française,pp. 27–36.

Tecchiati, U. 1992. ‘PrähistoricheBrunzefunde’ conservati al museo civico diRovereto (Trento): le asce. Annuali delMuseo Civico di Rovereto, 7:3–36.

Van de Noort, R. 2013. Seafaring andRiverine Navigation in the Bronze Age ofEurope. In: A.F. Harding & H. Fokkens,eds. The Oxford Handbook of the EuropeanBronze Age. Oxford: Oxford UniversityPress, pp. 382–97.

Venturino Gambari, M. 1998. Il neolitico eeneolitico in Piemonte. In: Atti della XXXIIRiunione Scientifica dell’Istituto Italiano diPreistoria e Protostoria: Preistoria e protostoriadel Piemonte. Firenze: Istituto italiano di pre-istoria e protostoria, pp. 10–63.

Venturino Gambari, M. & Aimar, A. 1996.Le vie della pietra verde: l’industria liticalevigata nella preistoria dell’Italia settentrio-nale: Torino, Museo di antichità; Alba,Palazzo mostre e congressi, settembre-dicembre 1996. Torino: Omega edizioni.

Visentini, P. 2009. Asce dell’Età del Rame inFriuli Venezia Giulia (ItaliaNord-Orientale). Gortania, 31:165–72.

Vital, J. 2005. Composantes culturelles eteconomiques du premier bronze ancienrhodanien. In: C.R., J. Driessen, & E.Warmenbol, eds. Actes du XIVème



Congrès UISPP. Section 11, L’âge dubronze en Europe et en Méditerranée =the Bronze age in Europe and theMediterranean: Sessions générales et posters =General sessions and posters. Presented at theInternational Congress of Prehistoric andProtohistoric Sciences (14th: 2001:Université de Liège), BAR InternationalSeries 1337. Oxford: Archaeopress, pp.137–46.

BIOGRAPHICAL NOTE

Laura Perucchetti is a D.Phil student atthe Research Laboratory for Archaeologyand the History of Art (RLAHA), Uni-versity of Oxford under the supervision ofMark Pollard. Her interests are focused onancient metallurgy and its analytical tech-niques and on the application of GIS inthe study of material culture.

Address: Research Laboratory for Archae-ology and the History of Art, Universityof Oxford, Dyson Perrins Building, SouthParks Road, Oxford OX1 3QY, UK.[email: [email protected]]

Peter Bray is a postdoctoral researcher atthe Research Laboratory for Archaeologyand the History of Art (RLAHA), Uni-versity of Oxford, and is also the Fellows’Junior Research Fellow at Linacre College.His research focuses on early technology,particularly metallurgy, and aims tocombine scientific, archaeological andtheoretical approaches. He is currently amember of the Atlantic Europe in theMetal Ages project, supported by AHRCgrant AH/K002600/1.

Address: Research Laboratory for Archae-ology and the History of Art, University

of Oxford, Dyson Perrins Building, SouthParks Road, Oxford OX1 3QY, UK.[email: [email protected]]

Andrea Dolfini is a Lecturer in Later Pre-history at Newcastle University (UK),where he also coordinates the Centre forInterdisciplinary Artefact Studies (CIAS).He is interested in material culture andsociety in the later prehistoric centralMediterranean. His current researchexplores the emergence of metal technol-ogy as well as the function of Bronze Agetools and weapons. He collaborates in theCase Bastione fieldwork project, Sicily.

Address: School of History, Classics andArchaeology, Armstrong Building, New-castle University, Newcastle upon Tyne,NE1 7RU, UK. [email: [email protected]]

Mark Pollard is the Edward Hall Pro-fessor of Archaeological Science at theResearch Laboratory for Archaeology andthe History of Art (RLAHA), Universityof Oxford and he is also a Member of theRoyal Society of Chemistry, a Fellow ofthe Society of Antiquaries and a Memberof the Oriental Ceramic Society. Hisresearch over the past 35 years has encom-passed the application of the physicalsciences, particularly chemistry, withinarchaeology, and has included a widerange of topics.



mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

mailto:

Mark Pollard is the Edward Hall Pro-fessor of Archaeological Science at theResearch Laboratory for Archaeology andthe History of Art (RLAHA), Universityof Oxford and he is also a Member of theRoyal Society of Chemistry, a Fellow ofthe Society of Antiquaries and a Memberof the Oriental Ceramic Society. Hisresearch over the past 35 years has encom-passed the application of the physical

sciences, particularly chemistry, withinarchaeology, and has included a widerange of topics.


Barrières physiques, liens culturels: métallurgie préhistorique dans la région alpine

Cet article porte sur la première métallurgie du cuivre et des alliages cuivreux dans l’ensemble de lazone alpine. Nous introduisons une nouvelle approche quant à l’interprétation des ensembles de donnéesportant sur la composition chimique, qui a été appliquée pour la première fois à une base de donnéesexhaustive régionale. Le Chalcolithique et le début de l’Âge du Bronze alpins présentent chacun desschémas distincts quant à l’utilisation du métal, qu’on peut interpréter comme découlant de changementsdans l’exploitation minière, de choix sociaux et de caractéristiques majeures du paysage, comme bassins etsystèmes fluviaux. Il est intéressant de noter que les Alpes ne figurent pas comme une barrière nord-sud,car des différences majeures dans la composition ont tendance à apparaître sur un axe est-ouest. Parmicelles-ci figurent la prévalence du bronze-étain dans les Alpes occidentales par rapport aux Alpes orien-tales. Cette ‘frontière de l’étain’ est examinée par rapport à la circulation des métaux dans la région eten tant que preuve d’une division géographique profondément enracinée présente pendant la plus grandepartie de la préhistoire alpine.

Mots-clés: Âge du Cuivre, Âge du Bronze, Alpes, GIS, artefacts en métal, première métallurgie,composition élémentaire

Physikalische Grenzen, kulturelle Verbindungen: Transalpine vorgeschichtlicheMetallurgie

Dieser Beitrag behandelt die frühe Metallurgie von Kupfer und Kupferlegierungen des gesamten Alpen-raumes. Er stellt einen neuen Ansatz der Interpretation von Datensätzen chemischerZusammensetzungen vor, der erstmalig auf eine umfassende regionale Datenbank angewendet wurde.Die alpine Kupfer- und Frühbronzezeit weisen bestimmte Merkmale der Metallnutzung auf, die durchÄnderungen im Bergbau, in sozialer Auswahl und größeren landschaftlichen Elementen, wiez. B. Wassergrenzen und Flusssystemen, interpretiert werden können. Interessanterweise wirken dieAlpen sich nicht als Grenze zwischen Nord und Süd aus, denn größere Unterschiede in den Zusammen-setzungen neigen dazu, in einer Ost-West-Achse aufzutreten. Dabei ist das verstärkte Auftreten vonZinnbronzen in den westlichen Alpen im Gegensatz zum östlichen Raum von zentraler Bedeutung.Diese, Zinnlinie‘ wird im Sinne eines Metallflusses durch die Region und als Beweis einer tiefverwur-zelten geographischen Unterteilung, die durch den Großteil der alpinen Vorgeschichte läuft, diskutiert.

Stichworte: Kupferzeit, Bronzezeit, Alpen, GIS, Metallartefakte, frühe Metallurgie, elementareZusammensetzung


mailto:

mailto:

mailto:

mailto:

mailto:

physical barriers, cultural connections: prehistoric metallurgy across the alpine region

Documents