Preliminary experimental observations on a particular class of bifacial lithic artifact from Misiones province, northeastern Argentina.

Hugo Nami

In the wide field of archaeology, stone tool remains are one of the main pieces of evidence used for assessing knowledge and understanding of the archaeological record. To cope with its analysis and interpretation as a branch of experimental archaeology, one field of research that has become more notable is experimental lithic technology. Based on experience and development of this discipline in the southern cone of South America, and with the aim of contributing to the growing theoretical perspectives in this field, this paper addresses the theoretical and epistemological issues that deal with theoretical, variability, classification, and deontological concerns.

Interactions between archaeology and philosophy are traced, from the ‘New Archaeology’s’ use of ideas from logical empiricism, the subsequent loss of confidence in such ideas, the falsificationist alternative, the rise of ‘scientific realism’, and the influence of the ‘new’ philosophies of science of the 1960s on post-processual archaeology. Some recent ideas from philosophy of science are introduced, and that discipline’s recent trajectory, featuring debate between realists and anti-realists, as well as a return to ‘classic’ concerns about explanation, causation, and laws of nature, is described. Many interactions between philosophy of science and archaeology have been based on a misplaced quest for a single ‘off-the-peg’ methodology or other philosophical framework for archaeology. Historical conditions have fostered the damaging idea that archaeologists have to choose between ‘positivism’ and subjectivism. I conclude by suggesting what kinds of contemporary philosophical work might interest archaeologists, and argue that philosophers should recognize the distinctive heterogeneity of archaeology.

EMPIRISME HUMIEN DANS L'OEUVRE DE JEUNESSE DU PHILOSOPHE JAPONNAIS ÔMORI SHÔZÔ ? Le but de cette communication est d'abord de présenter la logique et la philosophie telles que le philosophe japonais Ômori Shôzô la perçoit. Deuxièmement, nous nous attellerons à la tâche d’ordonner cette vision et par-là dégager les points importants qui constituent sa pensée de jeunesse. Troisièmement, nous comparerons cette pensée avec celle de David Hume. Ômori Shôzô est empiriste et pour lui le défi est d’expliquer qu'en référence avec la nécessité des phrases logiquement vraie, cette nécessité doit être fondée sur l'expérience. Nous voulons préciser ici que Ômori Shôzô est armé de perspicacité par rapport à cette question. Pour nous préparer à ce problème, nous allons décrire des observations banales au sujet du mot 論理学 (ronrigaku – trad. logique, science de la logique). Car il s'agit bien de nous situer dans le langage vernaculaire qui accueille la pensée de Ômori, le japonais. Dans ces limites nous pouvons dégager deux concepts celui de 論理 (ronri – trad. logique) et de 学 (gaku – trad. science). Comme l'idéogramme 学signifie le fait de faire une étude, nous pouvons comprendre que les termes de 論理学font référence à l'étude de la logique. Cette dernière différenciation est au regard de la pensée Ômoricienne extrêmement importante. C'est à partir de cette séparation que nous aborderons la philosophie de la logique telle que présentée par Ômori. Fort de cette analyse nous parlerons donc §1 du sens de la logique pour voir ensuite §2 son sens compris en tant que science. Dans une troisième partie, à partir de la logique comprise en tant que science nous discuterons §3 la provenance de la nécessité que comportent en elles les phrases logiques. Pour finir, §4 nous comparerons ces résultats avec ceux de l' « harmonie préétablie » humienne synonyme de notre tendance machinale à uniformiser le cours de la nature des choses par la nécessité.

Skilled Production and Social Reproduction Aspects of Traditional Stone-Tool Technologies Proceedings of a Symposium in Uppsala, August 20–24, 2003 Societas Archaeologica Upsaliensis & The Department of Archaeology and Ancient History, Uppsala University Editors Jan Apel & Kjel Knutsson SAU Stone Studies 2 Uppsala 2006 ISBN 91-973740-6-7 ISSN 1404-8493 © Authors of the articles Skilled Production and Social Reproduction. Aspects on Traditional Stone-Tool Technologies. SAU Stone Studies 2 Editors: Jan Apel & Kjel Knutsson Layout: Lars Sundström Cover: Jan Apel Cover photo: Markus Andersson Revision of English: Elisabet Green & Suzanne Nash Revision of manuscript: Martin Högvall & Suzanne Nash Published and distributed by: Societas Archaeologica Upsaliensis Villavägen 6G SE-752 36 Uppsala Sweden www.sau se Printed by åtta.45, 2006 Contents Acknowledgements ............................................................................. 6 Authors .................................................................................................. 7 Introduction Jan Apel and Kjel Knutsson Skilled Production and Social Reproduction – an introduction to the subject ....................................................................................... 11 Marcia-Anne Dobres Skilled Production and Social Reproduction in prehistory and contemporary archaeology: a personal exegesis on dominant WKHPHVDQGWKHLUSV\FKRVRFLDOLQÁXHQFHV........................................ 25 Chapter 1: Experiments and Experience Jacques Pelegrin Long blade technology in the Old World: an experimental approach and some archaeological results ..................................... 37 Hugo Nami ([SHULPHQWVWRH[SORUHWKH3DOHRLQGLDQÁDNHFRUHWHFKQRORJ\ in southern Patagonia ....................................................................... 69 Greg R. Nunn Using the Jutland Type IC Neolithic Danish Dagger as a PRGHOWRUHSOLFDWHSDUDOOHOHGJHWRHGJHSUHVVXUHÁDNLQJ ......... 81 Errett Callahan Neolithic Danish Daggers: an experimental peek ...................... 115 Hugo Nami Preliminary experimental observations on a particular class of bifacial lithic artifact from Misiones Province, northeastern Argentina .................................................................. 130 Chapter 2: Theoretical Aspects Kjel Knutsson $JHQHDORJ\RIUHÁH[LYLW\7KHVNLOOHGOLWKLFFUDIWVPDQDV “scientist” .......................................................................................... 153 Anders Högberg Continuity of place: actions and narratives ................................. 187 Jan Apel Skill and experimental archaeology .............................................. 207 Leslie Harlacker .QRZOHGJHDQGNQRZKRZLQWKH2OGRZDQDQH[SHULPHQWDO approach ............................................................................................ 219 Tuija Rankama, Mikael A. Manninen, Esa Hertell and Miikka Tallavaara Simple production and social strategies: do they meet? Social dimensions in Eastern Fennoscandian quartz technologies ...................................................................................... 245 Bradford Andrews Skill and the question of blade crafting intensity at Classic Period Teotihuacan ......................................................................... 263 Mikkel Sørensen 5HWKLQNLQJWKHOLWKLFEODGHGHÀQLWLRQWRZDUGVDG\QDPLF understanding .................................................................................. 277 Chapter 3: From Experience to Interpretation Nyree Finlay Manifesting microliths: insights and strategies from experimental replication ................................................................. 299 0DUFLQ:Ċs 6RPHUHPDUNVRQFRQWDFWVEHWZHHQ/DWH0HVROLWKLFKXQWHU JDWKHUHUVRFLHWLHVDVUHÁHFWHGLQWKHLUÁLQWWHFKQRORJ\DFDVH study from Central Poland ............................................................. 315 Kim Akerman High tech–low tech: lithic technology in the Kimberley Region of Western Australia .......................................................... 323 Per Falkenström A matter of choice: social implications of raw material variability .......................................................................................... 347 Per Lekberg Ground stone hammer axes in Sweden: production, life cycles and value perspectives, c. 2350–1700 cal. BC. ................ 361 Witold Migal 7KHPDFUROLWKLFÁLQWEODGHVRIWKHQHROLWKLFWLPHVLQ3RODQG ... 387 Kim Darmark Flaked rhyolite from Jettböle: attempts at an experimental explanation ....................................................................................... 399 Reference List ................................................................................... 409 Acknowledgements We would like to thank the following persons for contributing to the sympo sium and making this book possible: Helena Knutsson and Britta Wallsten helped us to plan and execute the symposium, Elisabet Green revised the language, and Lars Sundström did the digital layout of the book, Martin +|JYDOODQG6X]DQQH1DVKPDGHWKHÀQDOUHYLVLRQRIWKHPDQXVFULSWDQG the Berit Wallenberg Foundation generously supported the symposium. Some of the symposium participants outside SAU in Uppsala, Sweden. Top row: Nyree Finlay, Anthony Sinclair, Witold Migal. Second row from top: Kjel Knutsson, Leslie HarODFNHU*UHJ1XQQ0DUFLQ:üV7KLUGURZIURPWRS0LND7DOODYDDUD7XLMD5DQNDPD Hugo Nami, Dietz Stout. Front row: Jan Apel, Esa Hertell, Errett Callahan, Kim Akerman, Mikael Manninen. 6 Authors Kim Akerman 4 Dorset St Moonah. Tasmania 7009 Australia kimakerman@tastel.net.au Nyree Finlay Dept of Archaeology University of Glasgow Glasgow G12 8QQ Scotland QÀQOD\#DUFKDHRORJ\JODDFXN Bradford Andrews 3DFLÀF/XWKHUDQ8QLYHUVLW\ 4034 E.B. St Tacoma, WA 98404 U.S.A. bdand101@yahoo.com Leslie Harlacker CRAFT Research Center 419 N. Indiana Ave Bloomington, IN 47405 USA Lharlack@indiana.edu Jan Apel Societas Archaeologica Upsaliensis Villavägen 6G 752 36 Uppsala Sweden jan.apel@sau.se Esa Hertell Lithic Studies Group Institute of Cultural Studies Department of Archaeology P.O. Box 59 ),18QLYHUVLW\RI+HOVLQNL Finland KHUWHOO#PDSSLKHOVLQNLÀ Kim Darmark Societas Archaeologica Upsaliensis Villavägen 6G 752 36 Uppsala Sweden kim.darmark@sau.se Anders Högberg Malmö Kulturmiljö Den arkeologiska verksamheten Box 406 6(0DOP| Sweden anders.hogberg@malmo.se 0DUFLD$QQ'REUHV Faculty Associate Department of Anthropology University of Maine (Orono) USA madobres@maine.edu Kjel Knutsson Department of Archaeology and Ancient History Uppsala University Box 626 6(8SSVDOD Sweden kjel.knutsson@arkeologi.uu.se Errett Callahan 2 Fredonia Avenue Lynchburg, VA 24503 USA Per Falkenström Societas Archaeologica Upsaliensis Villavägen 6G 6(8SSVDOD Sweden per.falkenström@sau.se Per Lekberg Riksantikvarieämbetet (The National Heritage Board) Box 5405 114 84 Stockholm Sweden per.lekberg@raa.se 7 Mikael Manninen Lithic Studies Group Institute of Cultural Studies Department of Archaeology P.O. Box 59 ),18QLYHUVLW\RI+HOVLQNL Finland PLNDHOPDQQLQHQ#KHOVLQNLÀ Tuija Rankama Lithic Studies Group Institute of Cultural Studies Department of Archaeology University of Helsinki P.O. Box 59 00014 Helsinki Finland WXLMDUDQNDPD#KHOVLQNLÀ Witold Migal State Archaeological museum in Warzaw Poland neolit@pma.it.pl Mikkel Sørensen SILA – The Greenland Research Centre The National Museum Frederiksholms Kanal 12 1220 Copenhagen Denmark mikkel.soerensen@natmus.dk Hugo Nami Larrea 2033 Los del Mirador Buenos Aires Argentina nami@gl.fcen.uba.ar Miikka Tallavaara Lithic Studies Group Institute of Cultural Studies Department of Archaeology University of Helsinki P.O. Box 59 00014 Helsinki Finland 0LLNNDWDOODYDDUD#KHOVLQNLÀ Gregg Nunn HC 64 Box 2107 Castle Valley, Utah 84532 USA gregn@citlink.net Jacques Pelegrin “Prehistoire et Technologie” ERA 28 CNRS 1 Place A Briand, 92120 Meudon France SHOHJULQ#PDHXSDULVIU 0DUFLQ:ĊV Department of Archaeology University of Gdansk Poland was.marcin@wp.pl 8 Introduction Jan Apel and Kjel Knutsson Skilled Production and Social Reproduction – an introduction to the subject 'XULQJDÀYHGD\V\PSRVLXPLQODWH$XJXVWDJURXSRIDUFKDHRORJLVWV HWKQRDUFKDHRORJLVWVDQGÁLQWNQDSSHUVPHWLQ8SSVDODWRGLVFXVVVNLOOLQUH ODWLRQWRWUDGLWLRQDOVWRQHWRROWHFKQRORJLHVDQGVRFLDOUHSURGXFWLRQ,WVRRQ became apparent that we, as the organizers of the symposium, should have steered the ship with more authority than we did, at least if our sole purpose was to cover the subject of the title of the present volume. As the reader no doubt will notice, not all of the papers in this volume strictly honour WKHFKRVHQVXEMHFW7KXVWKH-DPLH5HLGLQVSLUHGFRYHURIWKLVERRNLVLQ WHQGHGWRUHÁHFW´DVLWXDWLRQµFUHDWHGZKHQUHVHDUFKHUVIURPGLIIHUHQWHSLV temological positions gathered to discuss the study of traditional stone tools. We wanted to acknowledge our conviction that studies of material culture must involve outside as well as inside perspectives in order to produce both convincing and interesting archaeological interpretations. Unfortunately, or fortunately, depending on how you look at it, this was not achieved during the symposium. Participants disagreed severely over these issues and some RIWKHVHFRQÁLFWVDUHWRXFKHGXSRQLQ'REUHV·SDSHUWKDWIROORZVWKLVLQWUR GXFWLRQ(YHQLIRXULQLWLDODLPVZHUHQRWIXOÀOOHGZHDUHFRQYLQFHGWKDW this book is a step in the direction of merging practice and theory in stone technological studies. Inside and outside – realism and rationalism Behind the choice of subject lies, among other things, a fundamental prob lem related to an epistemological project introduced to the philosophy of science by Gaston Bachelard in the 1930s (see for instance Bachelard 1984), which we believe is pertinent to archaeology. Bachelard was as critical to wards orthodox empiricism as he was of logical positivism (Broady 1991). While orthodox empiricism aims to attain knowledge of the surroundings, and the primary mode of access to the surroundings is observation (Ad ams & Adams 1991:314 f), logical positivism adds to these propositions by demanding that the knowledge of the surroundings that science aspires to 11 Jan Apel and Kjel Knutsson command is of a general character, and therefore permits us to explain phe nomena not yet observed. This is achieved through a rigorous methodology and through the construction of general laws (for example the construc WLRQRIPLGGOHUDQJHWKHRU\jOD%LQIRUG :KLOHWKHVHUHVHDUFKSHUVSHFWLYHV were traditionally regarded as two ends of a continuum between realism and rationalism, Bachelard suggested that they are equally important in “the VFLHQWLVW·VHYHU\GD\ZRUNµ6LPLODULGHDVZHUHODWHULQWURGXFHGWRDQ(QJOLVK speaking audience by Thomas Kuhn and were established in the social sci HQFHV E\ OHDGLQJ SRVWVWUXFWXUDOLVWV VXFK DV )RXFDXOW DQG %RXUGLHX LQ WKH late 1960s. When we read Bachelard and Bourdieu it occurred to us that H[SHULPHQWDODUFKDHRORJ\DQGÁLQWNQDSSLQJH[SHULPHQWVLQSDUWLFXODUIRO ORZHGYHU\FORVHO\WKHZD\LQZKLFK%DFKHODUGVXJJHVWHGWKDWVFLHQWLÀFZRUN proceeds: as a continuous motion between a sensually based description and WKHRUHWLFDODQDO\VLVDQGUHÁHFWLRQ Closely connected to our view of skill is the notion that skill can be a means of making social distinctions. Valuable artefacts and social institu tions that guarantee the reproduction of the technologies producing such artefacts will be used in social strategies. The main entry to this book is WKXVWKHDFNQRZOHGJHPHQWWKDWWKHNQRZOHGJHDQGNQRZKRZDVZHOODVWKH recipes for action involved in the production of stone tools, can be used as valuable assets in different kinds of cultural and social strategies. Knowledge of the reproduction of technologies is therefore essential for social interpre tations. The symposium strived towards combining papers dealing on the one hand with theoretical issues such as the social aspects of craftsmanship and skill in traditional societies, and on the other with practical sessions on the actual making of stone tools. While research carried out in order to solve only practical, technological aspects often tends to be sterile, in the sense that researchers focus on technical procedures, research on the social aspects of stone tools often tends to be naïve and formally orientated if the researcher lacks a comprehensive knowledge of the technical aspects involved. We had hoped that by combining scholars with different backgrounds and focus, the symposium participants might be able to bridge the gap between practice and theory. This meant that participants from very different traditions were invited to present papers and make practical demonstrations and this, in turn, proved to be problematic. However, even if it was obvious that there were disagreements as to what is considered to be interesting or meaningful research, we hope that the content of this publication, or rather, the indi vidual papers, will speak for themselves. The book contains 20 papers that have at least one thing in common: they revolve around different aspects RIWUDGLWLRQDOVWRQHWRROSURGXFWLRQ$VZHZULWHWKLVLQWURGXFWLRQWKHIDFW that to some degree we experienced the symposium and its subject as a bit 12 Skilled Production and Social Reproduction – an introduction to the subject problematic will not discourage us from digging deeper into the theoretical problems that arose during the symposium. The epistemological problems involving cultural and social interpretations of past technological traditions are far too interesting not to investigate further. Stone studies in Uppsala It was not a coincidence that the “Skilled Production” symposium was held in Uppsala. During the early 1970s, the archaeological research process in Uppsala was enriched with knowledge produced during practical experi PHQWVLQFOXGLQJÁLQWNQDSSLQJ7KHODWH7KRPDV-RKDQVVRQZKRIRUPHG the Institute for Prehistoric Technology in Östersund and the MNT teaching programme at the Bäckedal school in Härjedalen, was one of the originators. He wrote his BA thesis in Uppsala on experiments with bows and arrows, and thus introduced the investigation of theories of “the middle range”. The experimental research tradition in tool technology and tool function was lat er expanded on by Noel Broadbent and Kjel Knutsson at the Department of Archaeology, Uppsala University (Broadbent & Knutsson 1975, 1980). This was an example of how archaeologists in that particular research climate recognised the behavioural interest of processual archaeologists and thus regarded the development of theories concerning the relationship between prehistoric activities and archaeological material patterns as important in a research strategy inspired by the natural sciences, that could be applied in Stone Age archaeology. During the 1980s, this tradition was further devel oped and several undergraduate and PhD theses were produced at the De partment of Archaeology in Uppsala. Already during this period, researchers and lithic technologists were invited to participate in research projects and education. Harm Poulsen (Fig. 1) introduced the processual technological YLHZRIVWRQHWRROSURGXFWLRQLQGLIIHUHQWVHPLQDUVLQ8SSVDODEHWZHHQ and 1980. Errett Callahan (Fig. 2) later became an important participant as he was involved in several experimental projects at the Department of Ar chaeology in Uppsala and the Historical Museum in Stockholm from 1981 onwards (see for instance Callahan 1987, Callahan et al. 1992) and Bo Mad sen also visited the seminar in Uppsala (Fig. 3). Several undergraduate and PhD dissertations that based important aspects of their argumentation on experimental observations were thus produced during the 1980s and 1990s ..QXWVVRQ+.QXWVVRQ7DIÀQGHU$SHO/HNEHUJ 2002 & Sundström 2003). 13 Jan Apel and Kjel Knutsson Fig. 1. Harm Poulsen in Schloss-Gottorf, Germany 1975. Fig. 2. Errett Callahan in Uppsala 1980. )LJ%R0DGVHQPDNLQJDÁLQWEODGHLQ Uppsala. 14 Skilled Production and Social Reproduction – an introduction to the subject At the end of the 1980s, the experimental research tradition in Uppsala fad ed out. In part, this was explained by the fact that certain key persons, for LQVWDQFH.MHO.QXWVVRQÀQLVKHGWKHLUGLVVHUWDWLRQVDWWKLVSRLQWDQGPRYHG on to new adventures. However, it is also fair to say that the severe critique RIVFLHQWLÀFDUFKDHRORJ\WKDWZDVEURXJKWWREHDUE\SURSRQHQWVRIDQDU chaeology that denied the value of experimental and ethnographic analogies in favour of historical and phenomenological approaches also was to be held UHVSRQVLEOH7KHVHSRVWSURFHVVXDODUFKDHRORJLVWVUHMHFWHGWKHQRWLRQRI´WKH ideal generalisation” that, from a somewhat shallow point of view, lies at the KHDUWRIWKHH[SHULPHQWDVDQDUFKDHRORJLFDOPHWKRG+RZHYHUIRUZLVHQRQ dogmatic experimentalists, this critique was aimed in the wrong direction. The technological reconstruction that is the result of carefully conducted experiments is an interpretative process that continuously moves from small to large issues and back. This is due to the fact that the experimentalists do not necessarily produce an understanding that makes objective knowledge RISUHKLVWRULFHYHQWVSRVVLEOHHYHQLIWKH\EDVHVRPHRIWKHLUFODVVLÀFDWLRQV on natural laws. From an archaeological point of view, it might even be more appropriate to talk about experience rather than controlled experiments (in WKHVFLHQWLÀFVHQVH +DQGVRQSUDFWLFHIRUPRQHLPSRUWDQWZD\IRUDUFKDH ologists interested in technology to widen their perspective. Ideally, this is VRPHWKLQJWKDWJURZVIURPDQRQJRLQJGLDORJXHEHWZHHQJHQHUDONQRZO edge and individual practical skill and thus mimics an hermeneutical circle. It is at the crossroads of practical mastering and understanding of the craft, RQWKHRQHKDQGDQGWKHGLVWDQFHGVFLHQWLÀFDQDO\VLVDQGFODVVLÀFDWLRQRQ the other, that history, cultural conventions and the general way of life is negotiated. In fact, in one of the articles in this book, it is argued that prehis WRULFÁLQWNQDSSHUVWKHPVHOYHVDOVRWRRNDGYDQWDJHRIWKHLQVLGHYHUVXVWKH outside perspectives when actively trying to recapitulate older “forgotten” industries in cultural reproduction (see Knutsson in this volume). Thus, we are of the opinion that it is important to maintain and develop the experimental tradition. Since the merging of the particular and the general is fundamental in cultural reproduction, it is only logical that it also is present in VWXGLHVRIVWRQHWRROWHFKQRORJLHV$V'DQLHO0LOOHUSRLQWHGRXW\HDUVDJRLW is striking that research on material culture, and the ways in which it affects us, diminished during the 20th century at the very same time that the amount of ar tefacts that we are surrounded by in our everyday life increased considerably. It is surprising that archaeology is one of the few subjects that actually study the complex relationship between material culture and people. As a consequence we consider it meaningful, from a general point of view, to investigate material FXOWXUHDQGWUDGLWLRQDOVWRQHWRROWHFKQRORJLHVPDNHXSDVLJQLÀFDQWSDUWRI the archaeological remains of craftsmanship. 15 Jan Apel and Kjel Knutsson Fig 4. Flint knapping session at the Skilled Production Symposium. Knappers l-r: Witold Migal, Errett Callahan and Hugo Nami. Photo: Per Falkenström, SAU. It was important for us to bring people in from different research tradi tions, since we felt this might be one way of reaching a deeper understanding of the technologies and their role in society. Researchers, who had studied WUDGLWLRQDO VWRQHWRRO WHFKQRORJLHV IURP GLIIHUHQW DVSHFWV ZHUH LQYLWHG WR Uppsala from all around the world. Thus, as participants we wanted stone smiths who had learned traditional technologies and who had knowledge of WKHFRPSOH[LW\RIWKHFUDIWV:HDOVRZDQWHGDUFKDHRORJLVWVDQGHWKQRDU chaeologists who studied societies where stone tools are still being made and used, because this would give important insights into the social framework in which the crafts were embedded. We also wanted theoretically orientated archaeologists who had worked with questions concerning technology and its role in society on a more general and theoretical perspective. We cannot do without any of these different perspectives in the discussion and inves tigation of cultural reproduction and change as it is represented in lithics DQGWKHLUFRQWH[WRYHUWKHORQJWLPHVSDQRIKXPDQKLVWRU\7KH\PXVWEH used concurrently. Skill is not just a technological activity; skill is related to the understanding of the whole cultural setting and world view in which a technology is embedded. In this book “social reproduction” refers to the cultural knowledge of stone tool technologies, and the social use of this cultural knowledge, that is reproduced between generations. This informa tion includes the recipies of action that can be described in a châine opératoire analysis as well as knowledge of raw material sources and qualities and the DELOLW\IRUHDFKLQGLYLGXDOWROHDUQWKHSUDFWLFDONQRZKRZWKDWFDQQRWEH theoretically described. 7KHVFLHQWLÀFZRUOGZLWKLWVGLIIHUHQWHSLVWHPRORJLHVDQGLQVWLWXWLRQDOL zed subjects thus create constructed borders in what is a constant interplay 16 Skilled Production and Social Reproduction – an introduction to the subject LQFXOWXUDOUHSURGXFWLRQ&RQVHTXHQWO\WKHFRQÁLFWDWWKLVPHHWLQJKDVLWV historical reasons in science itself. The papers in this book are divided into three sections that correspond WRWKHWKUHHWKHPHVGLVFXVVHGDWWKHFRQIHUHQFH7KHÀUVWVHFWLRQ´([SHUL ments and experience” contains papers dealing with careful reconstruction and description of different stone tool technologies. Experiments and experience The papers in chapter one are mainly devoted to the description and discus sions of results from practical experimentation and focussed on details of single artefact categories and stigmata related to variable method/technique concepts. The distinction between “method” and “technique” put forward LQ3HOHJULQ·VSDSHULVLPSRUWDQWHVSHFLDOO\IRUQRQ)UHQFKUHDGHUV7KHFDVH VWXG\1HROLWKLF0DFUR%ODGHSURGXFWLRQLQ(XURSHDQGWKH1HDU(DVWIXU ther illuminates the importance of experiments in the process of understand ing technology through a dialectic and, as we see it, truly relational research process that takes its departure in a description of the method (realism) and then proceeds through careful production experiments (rationalism) towards the interpretation of the techniques used. Papers by Nunn, Callahan and Nami contain detailed descriptions, reci SHVRQHPLJKWVD\RISURGXFWLRQSURFHVVHVLQFOXGLQJGHWDLOVRQÁDNLQJDQ gles, holding positions, tools etc. This valuable information is interesting for many reasons, not least since the problem of cultural transmission is obvious. To know recipes and details about production processes of complex technol ogies, where you need nimble skills to accomplish a task, does not help much if you want to replicate them. However interesting the information may be, WKHUHDGHUZLOOÀQGLWLPSRVVLEOHWRUHSOLFDWHDQ\RIWKHGLVFXVVHGSURGXFWLRQ VHTXHQFHVVLQFHWKH\UHTXLUHDODUJHGHJUHHRISUDFWLFDONQRZKRZLHVNLOOV WKDWFDQRQO\EHDFTXLUHGE\SUDFWLFH7KHSDSHUVLQWKLVVHFWLRQFRYHUÁLQW WHFKQRORJLHVLQWKH6RXWK$PHULFDQ8SSHU3DODHROLWKLF 1DPL·VWZRSDSHUV the Neolithic period in Europe and the Near East (Pelegrin, Nunn and Cal ODKDQ &DOODKDQ·V SDSHU RQ KLV H[SHULPHQWV ZLWK WKH SUHVWLJLRXV 7\SH,9 Flint Daggers has an unconventional style and his purpose was to convey the visual aspects of the paper presented at the symposium. Originally, we intended to attach a DVD of the actual presentation with this book, but unfortunately the technical quality of the tape was not good enough. We are proud to be able to present preliminary results from his Dagger Project in this book. ,QVFLHQFHVNLOOFRXSOHGZLWKSUDFWLFDONQRZKRZLVWKXVPRUHUHODWHGWR 17 Jan Apel and Kjel Knutsson how this knowledge can be transformed into the skill to read stones from prehistoric assemblages. This is a skill that to some degree can be learned and used in archaeological analysis by people not capable of actually making the artefacts. However, this would probably not have been a meaningful option in the past, unless the skills of reading the material environment included stones from variable cultural and time contexts were part of a ne cessary cultural knowledge. Examples of the need to or ability to read stones in the past as part of cultural reproduction are discussed by Knutsson and Högberg. We do not know whether this was only a cultural skill related to NQRZOHGJHUDWKHUWKDQNQRZKRZ,QDSUHKLVWRULFVHWWLQJWKHDELOLW\WRGLV FXVVGHWDLOVRIDUHGXFWLRQVHTXHQFHE\SRLQWLQJWRGHWDLOVLQWRROVDQGÁDNHV might have been a skill that was valued, for example in discussing relics rela ted to ancestral events at sacred places or on a more mundane scale, related to the general ability to track friends and strangers in the cultural landscape covered with lithic debris from different times and places. The cultural skill of using knowledge of technology and material culture to communicate important aspects of the world to members of your group must thus be understood as one aspect of skill that does not necessarily re ODWHWRWKHVNLOORISUDFWLFDONQRZKRZ7KHSDSHUVLQWKLVVHFWLRQLOOXVWUDWH this in relation to the present situation. The skill of the lithic craftsman per se is not valid for the reproduction of individuals in the culture of science. The lithic craftsmen are mainly reproducing themselves outside the acad emy. This practical knowledge has to be transformed into usable assets in a cultural value system, in this case the culture of science. This transforma tion may take different paths from sheer theft of symbolic capital to a more humble use of references. Theoretical aspects The seven papers in this section may represent another form of skill that is effective in a different setting. Here, the cultural skill of knowing how WR WUDQVIRUPH WKH NQRZKRZ LQWR NQRZOHGJH DSSHDUV ZLWK WKD DLP RI XV ing this in social reproduction within academia. Theoretical skills related to WKHVFLHQWLÀFFXOWXUHWKXVGRQRWVWDQGIRUDEHWWHURUOHVVYDOXDEOHW\SHRI knowledge; it is just different and less concrete. ,QKLVSDSHU.QXWVVRQWULHVWRVKRZKRZDUHÁH[LYHFXOWXUDOSUDFWLFHRI science and modernity in general, play an essential part in all human cultural UHSURGXFWLRQ,QDQH[DPSOHIURPWKH/DWH*ODFLDO(DUO\+RORFHQHWUDQVL WLRQLQ6FDQGLQDYLDLWLVDUJXHGWKDWDQDFWLYHUHUHDGLQJRIROGOLWKLFWHFK nologies was a decisive element in the implementation of cultural change in 18 Skilled Production and Social Reproduction – an introduction to the subject this period. The active copying of old technologies and designs thus indi FDWHVWKDWUHÁH[LYLW\VRW\SLFDORIWKHPRGHUQFRQGLWLRQZDVDQLPSRUWDQW part of cultural reproduction in the Late Glacial period. The skill needed to FRSHZLWKWKLVLVHYLGHQFHGLQWKHGHWDLOHGUHFRQVWUXFWLRQRIEODGHPDNLQJ strategies from Hamburg to Ahrensburg. Here, the ability of the prehistoric knappers to read ancient stone tools made it possible to produce similar LWHPV E\ LPSOHPHQWLQJ NQRZKRZ LQWR SUDFWLFDO DFWLRQ DQG WKLV PXVW EH UHJDUGHG DV D VLJQ RI UHÁH[LYLW\ 7KH FXOWXUDO NQRZOHGJH RU FXOWXUDO VNLOO necessary, relating to what this return might have meant to these groups, can of only be speculated on. +|JEHUJ·VSDSHUWRXFKHVRQWKHVDPHVXEMHFWDV.QXWVVRQGHDOLQJZLWK the dynamic relation between structure and agency in cultural reproduction. He uses the concept of “conspatiality” to describe the historicity of place, i.e. the repeated use of one and the same place over centuries. Repeated WHFKQRORJLFDODFWLRQVWKXVVHHPWRFUHDWHSODFHVRIVSHFLDOVLJQLÀFDQFH)R cusing on the production of square sectioned axes from the Neolithic to the Bronze Age, Högberg can demonstrate how the places of procurement and initial reduction of axe preforms are characterized by a large number of fully usable axe preforms. The material manifestation of lithic production proc HVVHVJLYHVV\PEROLFVLJQLÀFDQFHWRWKHVLWHVDQGWKH\EHFRPHHVVHQWLDOLQWKH collective memory of the community that are using and reusing them. The UHWXUQWRWKHSODFHRYHUDQGRYHUDJDLQPXVWKDYHUHVXOWHGLQWKHUHUHDGLQJ of the material from earlier periods, Högberg continues. Since references to the past build a strong argument in the creation of legitimacy and constitute DIXWXUHZDUUDQWIRUDXWKRULWLHVSRZHUVDQGULJKWVWKHVNLOOHGÁLQWNQDSSHU ZKRXQGHUVWRRGÁLQWWHFKQRORJ\ZDVWKXVDSHUVRQZKRSRVVHVVHGNQRZO edge of the past. The paper is important insofar as it shows that we have to take the impact of history seriously in the discussion of cultural repro duction, as well as that practical skills were important in the ritual sphere, being part of the conceptual skill necessary to interpret the world and thus constitute society. Inspired by a French epistemological tradition, Apel points out that a conscious relational research strategy is imperative if we aim at social or cultural interpretations. In a case study consisting of an experimental and ar FKDHRORJLFDOVWXG\RI/DWH1HROLWKLF'DQLVKÁLQWGDJJHUVKHVXJJHVWVWKDWD VRFLDOUROHRIVNLOOFRQQHFWHGWRDWHFKQRORJ\FDQRQO\EHUHODWLRQDOO\GHÀQHG through the use of personal experience, on the one hand, and objectifying techniques, on the other. Objectifying techniques, such as statistics or the XVHRIVFLHQWLÀFFDWHJRULHVDVRSSRVHGWRIRONFDWHJRULHVGHYHORSDQHFHVVDU\ UHVLVWDQFH WRZDUGV WKH VXEMHFWLYH H[SHULHQFH RI WKH ÁLQW NQDSSHU RU OLWKLF analysts. On the other hand, a research strategy that denies the subjective 19 Jan Apel and Kjel Knutsson H[SHULHQFHRIWKHH[SHULPHQWDOÁLQWNQDSSHUFDQQHYHUXQGHUVWDQGWKHVRFLDO dimension. By means of experimental strategy, Leslie Harlacker wants to clarify the aspects of skill involved in the successful production of Oldowan lithic tech nology using both technological and biomechanical information. More spe FLÀFDOO\LWUHODWHVWRWKHLQYHVWLJDWLRQRI0RGH,WHFKQRORJ\DQGWKHTXHVWLRQ RIWKHUHODWLYHFRQWULEXWLRQRINQRZOHGJHDQGNQRZKRZ VNLOO WR2OGRZDQ ÁDNLQJ%DVHGRQDVHWRIFRQWUROOHGH[SHULPHQWVFDUULHGRXWE\QRYLFHWR skilled knappers and debitage analysis, a breakdown of performance into NQRZOHGJHDQGNQRZKRZOHDGVWRWKHIROORZLQJK\SRWKHVLV,WLVWKHDFTXL VLWLRQRINQRZKRZUDWKHUWKDQNQRZOHGJHWKDWUHVXOWVLQIXUWKHUJDLQVLQ SHUIRUPDQFHHIÀFLHQF\DQGFRQVLVWHQF\7KLVFRQFOXVLRQKDVLPSOLFDWLRQVIRU the study of hominid technological evolution in general and it implies that 2OGRZDQKRPLQLGVZRXOGKDYHEHQHÀWHGIURPÀQGLQJWKHWLPHWRSUDFWLFH tool making skills. Since notions of skill normally are intuitive and more re lated to the eye of the beholder (with variable experiences of knapping) than WRH[SOLFLWDUJXPHQWDWLRQ+DUODFNHU·VFRQWULEXWLRQVKRZVWKHLPSRUWDQFHRI IRUPDOL]LQJWKHGHÀQLWLRQRIVNLOO6LQFHLQWKLVYHUVLRQVNLOOLVUHODWHGIRUH most to technical skill, the amalgamation of technical and conceptual skills discussed for example by Högberg, is interesting. How was the Oldowan tra dition reproduced as consciousness expanded and thus the past increasingly became the vehicle of the reproduction of the present and the future? Here technology as a durable and solid manifestation of practice, such as memory, has an important role in the study of early hominid cognitive development. Rankama et al. discuss quartz technologies in Fennoscandia and the prob lem of relating them to the general discussion on gender, sociality and cul tural reproduction that is currently ongoing within lithic studies in general. The fact that quartz does not so easily lend itself to the production of com plicated tools, the use of quartz knapping in social strategies is less probable. Even if more elaborate technologies may be correlated to quartz, the degree of fragmentation is an obstacle to reading the material in terms of the nec essary chaîne opératoire analyses. Recent work in Sweden (fracture analysis), however, has made it possible to partly overcome some of these obstacles. A few case studies in Finland thus give hope that more examples of different operational schemes can be detected in future analyses and interpreted as cultural or social markers, making chaîne opératoireDQDO\VLVWKHORQJDZDLWHG substitute for formal typologies in vein quartz studies, and thus aid in discus VLRQVRIVRFLDODQGFXOWXUDOVLJQLÀFDQFH3HUKDSVDVNLOOHGTXDUW]XVHUNQHZ how to make the best use of the properties of that particular raw material in VSHFLÀFFXOWXUDOVRFLDODQGHQYLURQPHQWDOFLUFXPVWDQFHV3HUKDSVWKHIRFXV towards the skill needed to make quartz tools should be changed towards the 20 Skilled Production and Social Reproduction – an introduction to the subject skill to make use of this raw material. We know from recent lithic use wear analyses that both the tools made, and those used, show the same variation DVÁLQWDVVHPEODJHVDQGDVLVVWDWHGLQWKHSDSHUE\5DQNDPDHWDOWKHVR cial and cultural context in which the quartz material was situated may have EHHQLPSRUWDQW7KXVLQDZD\ZHÀQGLWPRUHREYLRXVWREUHDNGRZQWKH old dichotomy between the ritual and the mundane, the technical and the conceptual, in dealing with quartz assemblages. This is of course as true for ÁLQWRUÁLQWOLNHPDWHULDOVEXWLWLVMXVWWRRHDV\WRIRUJHWWKDWVLQFHÁLQWKDV an overt signal of technical complexity, this approach might be obstructive to the more important issue of its cultural meaning. $QGUHZV·DQG6¡UHQVHQ·VSDSHUVERWKGHDOZLWKWKHRUHWLFDOLVVXHVDVZHOO DVFXOWXUHKLVWRULFDOLQWHUSUHWDWLRQVLQFRQQHFWLRQZLWKEODGHLQGXVWULHV$Q drews discusses the scale and organization of the Classic period Teotihua can obsidian blade production. By studying skill through surface collections from production sites, Andrews is able to infer that the obsidian craftsmen ZHUHSUREDEO\QRWIXOOWLPHVSHFLDOLVWV,QWKHSDSHUVNLOOLVGHÀQHGWKURXJK DQHVWLPDWLRQRIWKHGHJUHHRINQRZKRZ 3HOHJULQ DQGWKHUHE\$Q GUHZVFRQVLGHUV´DUWLVLQDOµYHUVXV´HIÀFLHQF\µVNLOODVWZRHQGVRIDNQRZ how continuum. Sørensen contributes to the discussion of the differences between typologies based on formal or metrical attributes, and technologi cal attributes on the other hand; the formal and metrical typologies are not suited to form a basis for answering the questions that archaeologists are interested in today. To illustrate his point, Sørensen introduces a technologi FDOEODGHGHÀQLWLRQWKDWLVXVHGLQDWHFKQRORJLFDODQGH[SHULPHQWDOVWXG\RI different blade traditions within the early Mesolithic Maglemosian tradition in Denmark . From Experience to Interpretation In this section of the book, archaeological case studies involving discussions on how lithic technology and skill is related to other social phenomena are presented. )LQOD\·VSDSHUWDNHVLWVGHSDUWXUHLQWKHIDFWWKDWWKHSURGXFWLRQRIPLFUR liths in the Mesolithic has been placed in the “functional” sphere of inquiry, and as such has not encouraged interest in the discussion of how cultures reproduce themselves through socialisation and the transmission of cultural NQRZOHGJH+DYLQJEHHQVHHQDVUHÁHFWLQJFKURQRORJLFDOFKDQJHDQG´FXO tures” within the cultural historical tradition and a measure of environmen tal and functional change by processual archaeologists, Finlay discusses the role of microlithic production in the construction and negotiation of identi 21 Jan Apel and Kjel Knutsson ties. Experiments show characteristic idiosyncrasies in microlith production. %DVHGRQWKHFRQFHSWRISHUIRUPDQFHDWHUPERUURZHGIURP-XGLWK%XWOHU·V writings, a discussion is pursued where the tension between the secluded DQGWKHRSHQWKHH[SHFWHGDQGWKHXQH[SHFWHGFUHDWHVVRFLDO G\QDPLFV The focus on the social meaning of a reasonable simple production makes skill more of a conceptual ability (to know the social game) than a techni cal ability to produce insets in weapons. The similarity in form and design allows these microliths to be “socialized” in one and the same arrowhead. Similarly to the known SanKDELWRIH[FKDQJLQJKXQWLQJDUURZVDQGWKXVWR downplay individuality, the microliths act as a materialization of the social. )LQOD\·VSDSHUXQGRXEWHGO\H[SORUHVQHZJURXQGVWKDWDUHLPSRUWDQWIRUWKH topic of this workshop. Technologies are part of social communication and thus generative in societal reproduction. Skills are embodied understandings of how things are supposed to be done according to the rule book but, once learned and made explicit, they might as well become an arena for competi tion and tension and thus lead to social change. Was initiates a discussion of exchange networks, cultural transmission and learning in the Janislavice culture in Poland based on an analysis of blade FRUHVDQGGHELWDJHIURPWZRVLWHV7KHSUHVHQFHRIEODGHSUHFRUHVNP DZD\IURPWKHÁLQWVRXUFH FKRFRODWHÁLQW ZKHUHLWZDVSURGXFHGLQGLFDWHV functioning social networks in territories of this size. The fact that the skill WRDFWXDOO\SURGXFHEODGHVIURPVXFKFRUHVZDVIRXQGIDUDZD\IURPWKHÁLQW source must have something to say about cultural transmission in this time period. How was the skill of blade making transferred between the genera WLRQVLQDQDUHDZLWKOHVVÁLQW",WVHHPVUHDVRQDEOHWKDWWKHQHWZRUNZHVHH H[SUHVVHGLQWKHVSDWLDOGLVWULEXWLRQRIFKRFRODWHÁLQWDOVRPHDQWWKHPRYH PHQW RI SHRSOH ZKHUH DW OHDVW EHFRPLQJ D ÁLQW NQDSSHU PXVW KDYH PHDQW SHULRGVLQWKHÁLQWULFKDUHDVIRUSUDFWLFH,WLVLQWHUHVWLQJFRPSDUHGWRWKH Neolithic setting discussed by Migal, that ideas of exotics and value may have been related to the actual skill of making blades. In his paper, Akerman describes the complex and varied use of lithics, their production and use in the social reproduction of Aboriginal groups in the recent past in the Kimberley region in NW Australia. It can be noted how similar technologies are used in more mundane functional settings as well as in decidedly reproductive rituals such as initiations. Such tech QRORJLHVDUHLQYROYHGLQDUDQJHRIODUJHVFDOHDQGIDUUHDFKLQJH[FKDQJH networks where tools in one area used as ordinary spear heads and knives in another are related to more ceremonial situations. It is quite clear that in the case studies presented by Akerman, functional tools are always used in different cultural settings, transgressing the modernist border between the profane and the sacred. The technologies, tools, and social settings are 22 Skilled Production and Social Reproduction – an introduction to the subject intertwined, and there seems to be an always present relation to ancestral beings and cultural heroes, where lithics many times act as representations. This importance of lithics and lithic production in variable cultural settings may account for the cultural conservatism related to technologies that is observed by Akerman in this area of Australia. Cultural transmission is thus related to the construction of identities that despite, or perhaps because of the networks of exchanges in the area, are stable over time. Lithic techno logics were embedded in and logic to local cosmologies in these societies, and thus generative of the experience of a local identity and the understand ing of the world. Falkenström looks at lithic raw materials and access to lithic raw materials as assets that can be invested in social strategies. A combination of ethnohis torical and ethnoarchaeological data together with experimentation is regarded as a fruitful avenue towards an understanding of social processes in the past. He discusses how different raw materials are related to myths and sacred ritu als and thus form part of societal reproduction as representations of a mean ingful history. Quarry sites are often related to, and guarded by, creatures and dramatis personae in culture bearing myths. To procure raw materials and to use them in the production and use of artefacts must thus be understood as deeply involved in the mythical sphere and therefore structured by cultural values and world view. Exotic raw materials in his own research area display a variability, in terms of quality and relation to source, that exemplify the complex wider cultural context that archaeologist has to grapple with to make sense of lithic technologies. The reproduction of social values and norms are seen as control led not only by rituals but also by everyday behaviour. Per Lekberg discusses the social implications of the production and con sumption of Ground Stone Hammer Axes in Sweden during the Late Neo OLWKLFSHULRG7UDGLWLRQDOO\LWKDVEHHQGLIÀFXOWWRGLVFHUQGLVWLQFWW\SHVZLWKLQ this large archaeological material, mainly because they display very little for PDOYDULDWLRQDQGWRDODUJHSDUWDUHPDGHXSRIVWUD\ÀQGVZLWKRXWFRQWH[WXDO information. By conducting a technological study of the axes, Lekberg is able to argue convincingly that the axes were originally manufactures in relatively large sizes and then consumed in sequences. A study of the axes that has been IRXQGLQFRQWH[WVVKRZVWKDWD[HVIURPGLIIHUHQWSDUWVRIWKLVOLIHF\FOHZHUH deposited in different contexts, large axes in hoards, small axes in graves and broken axes on settlements. Finally, a “social topography” of the Late Neo lithic landscape emerges when distribution maps of the large material of stray IRXQG D[HV FODVVLÀHG LQWR WKHVH FRQWH[WXDO FDWHJRULHV GLVSOD\ GLIIHUHQFHV LQ wealth and social status between groups. Apart from the fact that Lekberg has VKRZQWKDWWKLVODUJHPDWHULDORIQHJOHFWHGVWUD\ÀQGD[HVWKURXJKDQH[SOLFLW technological approach, can produce social meaning, the paper also includes a 23 Jan Apel and Kjel Knutsson valuable cultural historical interpretation of the social complexity of the Late Neolithic period in Scandinavia. 0LJDO·V SDSHU GHDOV ZLWK WKH SUHVHQFH RI UHJXODU SUHVVXUH EODGHV LQ WKH Polish Neolithic. Although this appears in the middle period of the Linear Pottery Culture, it is foremost connected to the development of the TRB tra dition, a tradition that also enters Scandinavia at this time. Migal aims to show a connection between wine production and blade making by an anticipated use of a grape pressure device to make blades. The blade technology is inter HVWLQJLQWKLVFRQWH[WEHFDXVHWKHEODGHVDFFRUGLQJWRÀQGFRQWH[WVLQGHSRWV and graves, seem to have been important not so much as a practical tool dur ing this time but as part of societal ritual reproduction (Knutsson, H. 2003). The possible metaphorical connection between wine and blades are thus a topic that could be fruitful to investigate. ,Q6FDQGLQDYLDWKHEODGHPDNLQJWUDGLWLRQFKDQJHVGUDVWLFDOO\LQWKH75% period. From being embedded in everyday activities on sites during the Meso lithic, it moves over to the sphere of the sacred where the sites of production are hidden. No doubt the special technical skills needed to produce regular pressure blades and punch blades must have been connected to and related to another skill, the conceptual skills and knowledge related to social and/or cul tural reproduction. We see a continuation in this type of ritual technology in the Middle Neolithic Battle Axe Culture in Scandinavia (Knutsson, H. 1999). )LQDOO\'DUPDUN·VSDSHUEULQJVXSRQHRIWKHSUREOHPVWKDW5DQNDPDet al. H[DPLQHLQWKHLUSDSHULHWKHGLIÀFXOW\RIGLVFXVVLQJWKHLPSRUWDQFHRI skill in relation to technologies that appear to have been carried out in an RSSRUWXQLVWLF DQG VWUDLJKWIRUZDUGV PDQQHU 7KLV WLPH WKH REMHFW RI VWXG\ HPDQDWHV IURP VSHFLÀF DUFKDHRORJLFDO FRQWH[WV WKH UK\ROLWH WHFKQRORJ\ RQ WZRDGMDFHQWVLWHVRQD\HDUROGKXQWLQJÀVKLQJVLWHLQWKHcODQGLFDU chipelago. The technology consists of a rudimentary platform technique con ducted with a hard hammer, but there are noticeable differences between the sites, and production experiments are conducted in order to explain these dif IHUHQFHV'DUPDUN·VFRQFOXVLRQLVWKDWLWLVLPSODXVLEOHWRUHJDUGWKHFKRLFHRI a simple technology as depending on a lack of skill. Rather, this choice must be understood in social terms. The preservation of the remains of lithic technologies makes them espe cially appropriate for the study of the development of skill, crossing the bor der between the long sweep of evolution and history. In the book we meet research covering the history of choppers from the Acheulean to the Late Neolithic. This is an opening up for an exciting discussion of the evolutionary history of hominids and the equally interesting development of cognition and “the historical mind” in cultural reproduction. 24 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Marcia-Anne Dobres Skilled Production and Social Reproduction in prehistory and contemporary archaeology: a personal exegesis on dominant themes and their psychosocial LQÁXHQFHV Introduction For reasons worth investigating, archaeologists are loath to lay bare the hermeneutic relationship between their own interpersonal dynamics, such as those which run rampant at professional meetings, and intellectual trends shaping the discipline. By refusing to acknowledge the personal degree of in vestment we have in our research, and by denying just how much the personal LVSROLWLFDODQGLQÁXHQFHVUHVHDUFKRQWKHSDVWZHKDYHFRPHWREHOLHYHWKDW we are successful at keeping public discourse to impersonal issues of episte mology, methodology, and unbiased interpretation. While I am no champion of “big men” theories of cultural evolution, there is no doubt that dominant personalities (of variously gendered persuasions) have indeed shaped the dis cipline both theoretically and methodologically. In the next few pages I dare WRUHÁHFWRQWKHLQWHUSHUVRQDOSV\FKRVRFLDOG\QDPLFVSHUYDGLQJWKH´6NLOOHG Production and Social Reproduction” conference, on which this volume is EDVHG,DPVSHFLÀFDOO\LQWHUHVWHGLQFRQVLGHULQJZKHWKHURUQRW DQGKRZ WKH\VLPXOWDQHRXVO\UHÁHFWDQGLQÁXHQFHGRPLQDQWWUHQGVQRZSRSXODULQ the study of ancient technology. I realize I tread on shaky ground, not only by breaking the taboo on keeping our dirty linen in the closet, but also by suggesting that there is a directional relationship between contentious inter personal dynamics and how we ply our trade. 25 Marcia-Anne Dobres Fig. 1. Dominant themes discussed during the Skilled Production symposium. (Drawing by Kim Darmark). 26 Skilled Production and Social Reproduction in prehistory and contemporary archaeology The irony, of course, is that many archaeologists are similarly loath to en tertain the possibility that psychosocial interactions and esoteric beliefs of prehistoric technicians directly shaped not only their material practices but also their social reproduction writ large. I no longer believe this parallel is a FRLQFLGHQFH$VDSDUWLFLSDQWREVHUYHULQ6ZHGHQ,OHDUQHGDJUHDWGHDOIURP interacting with my colleagues – but mostly from those with whom I disa greed on substantive issues regarding prehistoric technology. What struck PH DW WKH WLPH RI WKH FRQIHUHQFH DQG ZKDW , KDYH EHHQ DVNHG WR UHÁHFW upon here, was the contentious nature of the aggregation itself and how our LQWHUSHUVRQDOLQWHUDFWLRQVGLUHFWO\²DQGQHJDWLYHO\²LQÁXHQFHGVXEVWDQWLYH GLVFXVVLRQ$VZLWKDQ\DJJUHJDWLRQRIVWURQJZLOOHGDQGDUWLFXODWHLQGLYLGX DOV ZH VRRQ GLYLGHG LQWR WZR VHOIVHOHFWHG LQWHOOHFWXDO ´FDPSVµ %XW ZKDW ZDV PRVW FXULRXV ZDV KRZ ZH LQWHUDFWHG DFURVV WKLV VHOILPSRVHG ERUGHU ]RQHQRWRQO\GXULQJRIÀFLDOGLVFXVVLRQVEXWHYHQFROOHFWLYHPHDOVDQGOLED tion breaks. What this essay explores is the disturbing resemblance I sensed between the discursive strategies of the dominant few (in controlling the terms of debate) and how we talked about the role of skill and knowledge in ancient technical practice. Interestingly, while as a group we never quite got around to dis cussing the “social reproduction” theme of the conference, we nonetheless promoted our own social reproduction at every turn. And this is why I believe VXFKFULWLFDOUHÁHFWLRQVDUHQHFHVVDU\ Dominant topics – domineering themes I sketched Figure 1 (“Dominant and Domineering Themes re: “Skilled Pro GXFWLRQDQG6RFLDO5HSURGXFWLRQµ GXULQJWKHFRQIHUHQFHDVDVDQLW\VDYLQJ strategy to keep me from jumping out a window during some of our more fractious discussions. As I began to perceive (and wince at) the formation of antagonistic intellectual camps and the crystallization of distinctly opposite viewpoints (which drew pleasure from misrepresenting each other), I was struck by the parallel I was witnessing to larger trends currently pervading the discipline. But the speedZLWKZKLFKWKLVKDSSHQHG LQWKHÀUVWKRXURQ WKHÀUVWGD\ DQGWKHGHJUHHRIZLOOIXOHQWUHQFKPHQWRQDOOVLGHVZDVVWULNLQJ and troubling to the degree that I felt the need to chronicle the key points of contention (while fully engaged in debating them). By day #2, I had identi ÀHGVL[GRPLQDQWWKHPHVUHJDUGLQJ´VNLOOHGSURGXFWLRQµDURXQGZKLFKDOO subsequent discussion hovered: 27 Marcia-Anne Dobres • knowledge and know how • (gestural) skill and performance • identity (achieved through skillful performance) and the concept of the Individual • the social value of skilled production • mind/body dualities • theoretical cause and effect of skilled production Methodologically, linear and materialist approaches for chronicling techno logical life histories (or chaînes opératoires) dominated, and everyone, myself included, worried about or proposed concrete methodologies for studying the empirical remains of skilled practice in order to identify interpretable empirical patterns. The difference was that some of us latched on to such concerns on general anthropological (“theoretical”) grounds, while others were led to such considerations based on their personal experiences rep licating all sorts of prehistoric lithics. While I would suggest that, in fact, theory and practice merged into a single whole directing us to similar concerns with analytic methodologies, at every turn our explicit debates pitted theory against practice. Descartes would have been pleased with the pervasive mind/body split dominating the conference. For practically every topic subject to heated dis cussion, those controlling the terms of debate formulated the question of technological skill (which quickly became a proxy for “skilled production”) as either a matter of mind or a matter of physical bodies. In most instances, ERGLHVZHUHFRQVLGHUHGWKHSK\VLFDO´SODFHµZKHUHVNLOODQGNQRZKRZUH sides and is performed, while minds are the locus of knowledge (until mate rially expressed by the hand). But because the gestural skill of bodies in mo tion (“practice”) was distinguished from and privileged over mind, thought, and aesthetic considerations (“theory”), most of our conversations (about skill, knowledge, identity, the body, social values, or cause and effect) ended with a few individuals bickering about how to see and measure skilled pro duction. It is not all that surprising that measurable and empirical aspects of ancient skill (viz. practice) were continually privileged “over” issues of knowledge (viz. theory), since practice (typically expressed of in terms of rep lication experience) was the dominant claim of authority employed by those controlling discussion (see below). 28 Skilled Production and Social Reproduction in prehistory and contemporary archaeology But it seemed to me that as the dominant concern of the weekend, material ism topped practice (explaining why “theory” was relegated to third banana). For example, whether talking about skilled production or its seemingly more intangible dimension, knowledge, the conversation always came round to the problem of identifying quantitative attributes. Over and over again we grappled with how to see and measure knowledge, the proper way to quan WLI\ JHVWXUDO VNLOODQGZKLFKVSHFLÀFPDWHULDOWUDLWV LQH[HFXWLRQIRUPRU IXQFWLRQ ´VLJQLÀHGµDSDUWLFXODUOHYHORIVNLOORUVRFLDOVWDWXV EHLW´QRYLFHµ “apprentice,” or “specialist”). While no doubt important, these materialist GHEDWHVHIIHFWLYHO\FORVHGGRZQÁHGJOLQJGLVFXVVLRQVRIOHVVHPSLULFDOFRQ siderations with which many of us were (also) interested. Often the majority of the participants (including the conference organizers) were left sitting on the sidelines while an intrepid two or three individuals became mired in endless but heated debates over some proposed trait list of physical attributes of skill. To my recollection, our most contentious debates involved the social value likely attributed (in the past) to different levels of skilled production or the tools themselves. Too often, however, we uncritically projected into the past our own (“capitalist”) value judgements (concerning time management or the PRVWHIÀFLHQWXVHRIUHVRXUFHV %XWLQVSLWHRIWKLVTXDQWLWDWLYHEHQWWKHFRQ versation always came back to a visceral “Wow! Oooh ahhh” appreciation of the more remarkable displays of gestural virtuosity found in the (lithic) archaeo logical record (and ably replicated by several participants). Such judgmental conversations about social value typically slid into untutored speculations about the identity and status of the most skilled technicians. But with the vo FDOO\GRPLQDQWJURXS·VRYHUDUFKLQJFRQFHUQZLWKPHDVXULQJVXFKG\QDPLFV we most often lapsed into discussion of the (empirical) “cut off” between novice, apprentice, and specialist – while presuming that perfect execution ZDVDOZD\VDQGHYHU\ZKHUHWKHDQFLHQWWHFKQLFLDQ·VGHVLUHGJRDO6DGO\ZH never openly discussed the possibility that in at least some ancient contexts, overt (even gratuitous) displays of skilled performance might not have been culturally sanctioned or deemed socially acceptable. Without explicit discus sion, the ontological premise which held sway over the weekend was that in all times and places technicians “naturally” aspire to grandiose displays RIVNLOO²ZKLOHWKRVHZKRFDQ·WZLOOQHFHVVDULO\PDUYHODW DQGKHQFHKLJKO\ value) those who can. Ironically, this is precisely the interpersonal dynamic that shaped our own interactions and value system throughout the weekend. Expert replications were marveled at and were given (or took) precedence LQPRVWGLVFXVVLRQV1RZRQGHUZHWRRNLWIRUJUDQWHGWKDWVHOISURPRWLQJ displays of gestural skill are a universal means of social climbing. 29 Marcia-Anne Dobres Because materialist concerns with identity (trait lists) dominated most dis cussions, many of us readily (albeit tacitly) agreed that chaîne opératoire and life history methodologies are useful analytic tools. However, because we were À[DWHGRQGLVWLQJXLVKLQJGLVFUHWHVWDJHVRIVNLOOHG WHFKQRORJLFDO SHUIRUP ance and by extension identifying categories of social identity, in my view too many participants allowed their methodology to become the goal of analysis. It was as if a correct percentage or numeric degree of skill could adequately stand “for” the actual process of negotiating social identity. Underlying psychosocial dynamics shaping substantive discussion As a thoroughly “embedded” participant simultaneously engaging in these debates and observing body language, tone of voice, and style of personal interaction, I noticed a far more troubling dynamic lurking below the surface of every discussion, not only in the seminar room but even when we ate and UHOD[HG EHIRUH DQG DIWHU WKH GD\·V SODQQHG DFWLYLWLHV $V WKH LQWURGXFWLRQ to this volume mentions, it became clear to many of us that as a group of thoughtful and dedicated researchers we never directly addressed the confer HQFH·VLQWHQGHGSXUSRVHWRXQGHUVWDQGhow skilled (technological) produc WLRQLQÁXHQFHGRUFRQWULEXWHGWRDQFLHQWVRFLDOUHSURGXFWLRQ$V,VXJJHVW above and try to show in Figure 1, our inability to ever directly confront the question of social reproduction was because we were never able to get past materialist discussions of skilled production. Curiously, our own strategies of skilled production (performed in and outside the seminar room) were all DERXWVRFLDOUHSURGXFWLRQ²VSHFLÀFDOO\DERXWZKRFRXOGFRQWUROWKHGLVFXV sion and who would have the last word. These strategies varied from overt, VHOIDJJUDQGL]LQJGLVSOD\VRIJHVWXUDOVNLOOE\UHSOLFDWLQJH[WUDRUGLQDU\OLWKLF DUWLIDFWVWRDJDJJOHRIJDSLQJRQORRNHUVWRWKHVHOIVHUYLQJEDQG\LQJDERXW RIWKUHHIRXUDQGVRPHWLPHVHYHQÀYHV\OODEOHZRUGVDQGUHIHUHQFHVWR obscure (dead) philosophers. That we employed a host of interpersonal strategies of skilled production in order to socially reproduce ourselves is, of course, not all that surprising. $GPLWWHGO\RQHFDQQHYHUDJJUHJDWHDJURXSRIDUFKDHRORJLVWVZLWKRXWVHOI VHOHFWHGVXEJURXSVWU\LQJWRDGYDQFHDP\ULDGRIFRPSHWLQJVRFLRSROLWLFDO agendas that have little to do with the explicit purpose of the aggregation! :KDW·VGLVWXUELQJLQDOOWKLVDQGZKLFKZDVSDUWLFXODUO\HYLGHQWLQ6ZHGHQ is the parallel I observed: between our unwillingness to recognize or admit the degree to which our own (physical and cerebral) skilled production dur 30 Skilled Production and Social Reproduction in prehistory and contemporary archaeology ing seemingly mundane and quotidian interactions shaped our social repro duction, and our general unwillingness to appreciate how such a dynamics may have operated in prehistory. This exegesis is not intended to explicate or defend (through proper use of citations and the like) what I see as disturbing trends in archaeological re search on technology (though many of the most common were in play at this conference and are summarized in Figure 1). Nonetheless, I could not help but sense yet another “duality” pervading the conference: something of a SDUDGLJPZDUSLWWLQJZKDW,FDOO´EOXHFROODUµDQG´ZKLWHFROODUµDUFKDHROR gists against each other – between those who “do” archaeology (in the case of technology researchers, these are typically replicators or ethnoarchaeolo JLVWV DQGWKRVHZKR DFFRUGLQJWRWKLVÀUVWJURXS ´PHUHO\µWKLQNDQGWKHR rize (you know – academics). I neither defend nor critique such stereotypes. There is no doubt they exist in our folk consciousness, litter the pages of SHHUUHYLHZHGMRXUQDOV²DQGSXWDGDPSHURQWKHFRQIHUHQFHLQ6ZHGHQ /LNHLWRUQRWWKLVSLWWLQJRIWKHKDQGVRQUHVHDUFKHUVDJDLQVWLYRU\WRZ er academics pervades both anthropology and archaeology; that is, it is not unique to students of technology. Nor would I care, if it were not for how such intellectual squabbles and overt attempts at social climbing GLUHFWO\LQÁXHQFH our research and our models of the past – and this is especially true in the study of prehistoric technology. What is important here is how the mind/body and knowledge/skill dualities which pervaded [del.] substantive discussions both UHSOLFDWHGDQGZHUHGLUHFWO\VKDSHGE\WKHVHOILPSRVHGEOXHDQGZKLWHFRO ODU´FDPSVµLQWRZKLFKZHUHDGLO\SODFHGRXUVHOYHVDQGSLJHRQKROHGRWKHUV As with the privileging of the measurable aspects of body, skill, identity, and social value which dominated our conversations about skilled production in prehistory, we similarly afforded different degrees of respect and value to each other, depending on which “collar” the speaker wore. Curiously, but especially troubling to me, were the repeated claims of au WKRULW\EDVHGRQSHUVRQDO KDQGVRQ H[SHULHQFHZKLFKHIIHFWLYHO\VLOHQFHG all controversies and dissenting viewpoints (based in theory). Claims of per VRQDO DXWKRULW\ W\SLFDOO\ ZHUH H[SUHVVHG LQ VWDWHPHQWV VXFK DV ´,·YH GRQH WKLVP\VHOIWUXVWPH,NQRZZKDW,·PWDONLQJDERXWµRU´,·YHVHHQWKLV ZLWKP\RZQH\HVLQWKHÀHOGKRZFDQ\RXTXHVWLRQP\REVHUYDWLRQV"µ/HVW the reader think I am too sensitive about such tactics, I was not the only one who noticed the surprisingly visceral attempts to discredit and silence FRQWUDU\RSLQLRQV H[SUHVVHGE\WKHZKLWHFROODUFURZG WKURXJKVXFKWDFWLFV [del.]. In many instances, [del.] references to personal experience were quickly followed by someone proposing a general principle or supposedly universal theory concerning technological skill, identity, or value for all of prehistory! I KRSHWKHLURQ\KHUHGRHVQRWHVFDSHWKHUHDGHU·VDWWHQWLRQ 31 Marcia-Anne Dobres 'LVWXUELQJO\ QRW RQFH GXULQJ WKH HQWLUH FRQIHUHQFH ZHUH ZHOOIRXQGHG “theoretical” issues with epistemology ever mentioned. Most notably missing ZDV DQ\ GLVFXVVLRQ RI WKH PDQ\ ZHOOUHKHDUVHG GLOHPPDV RI HWKQRJUDSKLF analogy, the inherent biases of presentism, or the manifold problems with HPSLULFLVWEDVHGLQGXFWLRQDVDPHDQVRIJHQHUDOL]DWLRQ7KLVEODWDQWGLVUH gard for topics which have been central to anthropological archaeology since the 60s attests to just how effectively personal claims of authority silenced controversy and directed discussion. -XVW DV ZH VHOIGLYLGHG DQG FDWHJRUL]HG HDFK RWKHU DV HLWKHU EOXH RU ZKLWHFROODU DUFKDHRORJLVWV VR WRR ZH GLYLGHG WKH TXHVWLRQ RI WHFKQRORJL cal practice (skill) from theory (mindful and embodied knowledge). In each instance, we privileged the former (as more measurable, more empirically “knowable,” and probably more deterministic). Those of us who (theoreti cally) questioned the premise that throughout the past overt displays of ag grandizing technological skill and experience were universal paths to social status, were discounted out of hand. If, in the present, such aggrandizing suc ceeds in achieving social status, value, and identity, why not through all of prehistory as well?! Less overt attempts to gain social status in this conference were summar LO\GURZQHGRXWE\WKHÁH[LQJRIKDQGVDQGPXVFOHVWKHSDVVLQJDURXQGRI exquisitely replicated blades (“just made it this morning ...”) or photo albums RIWKHVDPH(YHQZKHQZKLWHFROODUSDUWLFLSDQWVDWWHPSWHGWRGHVFULEHWKHLU ownKDQGVRQH[SHULHQFHVPDNLQJSUHKLVWRULFWRROVLQRUGHUWRPDNHDPRUH “theoretical” point (thereby legitimizing this strategy as the only acceptable PHDQVRIEHLQJKHDUG LWEDFNÀUHG%\YLUWXHRIbeingZKLWHFROODUWKHLUDU JXPHQWVZHUHGLVFRXQWHGHYHQZKHQVXSSRUWHGE\KDQGVRQH[SHULHQFH2Q WKHRWKHUKDQGPRVWWKHRUHWLFDOVXJJHVWLRQVRIIHUHGE\EOXHFROODUUHVHDUFK ers were typically discounted (by the other side) as being impossibly naive. Conclusions Are there larger lessons to extract from this contextualized and ethnographic analysis, this untutored psychoanalysis, of the interpersonal dynamics played out in Sweden in late August 2003. Indeed I think there are, and they concern WKHGLVWXUELQJUHODWLRQVKLSEHWZHHQRXURZQVNLOOHGSURGXFWLRQ DVEOXHDQG ZKLWHFROODUDUFKDHRORJLVWV DQGKRZZHVWXG\DQGWU\WRPRGHOVNLOOHGSUR duction in the past. But unlike the problematic split between mind and body and between esoteric knowledge and practical skill which prevailed in our discussions, this analysis explicitly integrates a somewhat detached intellec tual analysis (informed by social theory and a few dead philosophers) with 32 Skilled Production and Social Reproduction in prehistory and contemporary archaeology personal observation and embodied experience (what I observed, what I did, DQGKRZ,IHOWDWWKHFRQIHUHQFH ,KDYHWULHGWRLQWHOOHFWXDOL]HDQGUHÁHFW critically on both substance and style in the way the conference unfolded – but to allow my analysis to be informed by personal sensibilities as a par ticipant who had her own agendas to promote that weekend. Admittedly, my observations [del.] are “merely” based on my sensing (rather than measuring and thus empirically verifying) a discernable pattern in our own means of social reproduction (ironically based on skilled production). Importantly, I do not think what happened at the Skilled Production, Social Reproduction conference in Uppsala was in any way unique. I think it is the norm in how we currently study ancient technologies and interact with each other while doing so. Perhaps the dynamics were more striking than what similarly happens at larger conferences because there were so few peo ple involved – we simply could not avoid each other and only interact with “our” kind. Nonetheless, to watch how our psychosocial interactions shaped (a lack of) tolerance for diverse ideas and analytic strategies, to see how strat egies of careerism and aggrandizement impacted what we decided skilled production was all about in prehistory, and to see which particular aspects of skilled production were privileged – these struck me as a likely explanation for why, in the study of ancient technology, we cannot seem to agree on the IXQGDPHQWDOV :KLOH PDQ\ RI XV WDON DERXW ÀQGLQJ ZD\V WR transcend VHOI imposed and skillfully performed boundaries and intellectual borders, and conferences such as this are designed to further that worthy goal – in practice our habitus revels in maintaining such distinctions. $JDLQLWZRXOGQRWEHVREDGLIDOOWKLVSOD\DFWLQJZDVFRQÀQHGWRWKH present. But the obvious impact our problematic discourse, our pet peeves, and our lack of tolerance for alternative views has on our understandings of skilled production and social reproduction in the past – that is something HOVHDOWRJHWKHU,IGLUHFWREVHUYDWLRQDQGKDQGVRQNQRZKRZLVWREHHSLV temologically privileged over “mere intellectualizing” in our research and model building, then perhaps we should be more circumspect in the gener alities we propose from such inductive reasoning. 33 Chapter 1 Experiments and Experience Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Jacques Pelegrin Long blade technology in the Old World: an experimental approach and some archaeological results Abstract With reference to an extensive body of production experiments, the author GHVFULEHVDQGGLVFXVVHVWKHORQJEODGHSURGXFWLRQLQVHYHQDUHDVIURP3RU tugal/France in the west to Bulgaria/Syria in the east. Two techniques for GHWDFKLQJ WKH EODGHV DUH GHÀQHG LQGLUHFW SHUFXVVLRQ DQG SUHVVXUH UHLQIRUFHGE\DOHYHU7KHDXWKRULVDEOHWRLGHQWLI\ÀYHWHFKQRORJLFDOWUDGL tion and is thereby, among other things, able to suggest the movement of a few specialised craftsmen over large areas. Introduction Long and regular blades, excavated in Europe and in the Near East, and dating from the Late Neolithic or Chalcolithic (4th and 3rd millenniums BC) have long been discussed. Over this vast and diverse area, we are now aware WKDWPRUHWKDQUHJLRQVULFKLQÁDNDEOHDQGKRPRJHQHRXVVWRQH PRVWRI WKHPÁLQWEXWDOVRPHWDPRUSKLFFRQWDFWURFNLQVRXWKHUQ,EHULD ZHUHH[ ploited in an extensive blade production, many of them for several centuries, but not necessarily by a large number of craftsmen. Some of these workshops have been known for more than a century, VXFKDVWKRVHRI/H*UDQG3UHVVLJQ\ ZHVWHUQFHQWUDO)UDQFH DQG6SLHQQHV (Belgium), but are still little documented. Other workshops were discovered – or rediscovered – more recently, and/or are presently under study (e.g. the )RUFDOTXLHUEDVLQLQVRXWKHDVWHUQ)UDQFHWKH´KRQH\µÁLQWZRUNVKRSVIURP northern Bulgaria). Others remain to be discovered, being suspected only IURP WKHLU EODGHSURGXFWV IRXQG LQ VHWWOHPHQWV RU JUDYH FRQWH[WV EODGHV LQWUDQVOXFHQWÁLQWLQQRUWKHUQ*UHHFHEODGHVLQWHUWLDU\ÁLQWLQQRUWKHUQ )UDQFHDQG%HOJLXPEODGHVLQEDQGHGÁLQWLQQRUWKHUQ6SDLQHWF Very few studies were conducted on these blade productions from the Neolithic or Chalcolithic. A few years ago, nothing was known about their detachment technique, and the relevant criteria were even not documented EXWW DVSHFW ULSSOHPDUNV RQ WKH EXOE FUDFNV HWF 5HJDUGLQJ WKH UHGXF tion process (or “chaîne opératoire”), i.e. the core geometry, the position of crests, the platform preparation, the rhythm of the blade detachment, the 37 Jacques Pelegrin RQO\PHWKRGGHVFULEHGZLWKVRPHSUHFLVLRQZDVWKDWRIWKHIDPRXV´OLYUH GHEHXUUHµIURP/H*UDQG3UHVVLJQ\ *HVOLQ et al. 1975, Kelterborn 1980, 0DOOHW 3HOHJULQ0LOOHW5LFKDUG &RQVHTXHQWO\ WKH RQO\DQGLQVXIÀFLHQWDUJXPHQWIRUWKHVSHFLDOL]HGQDWXUHRIVXFKDSURGXF tion lay in its relative concentration and in the diffusion of the products over large areas. Likewise, our general ignorance of the knapping techniques and production methods prevented us from any attempt to group these different ZRUNVKRSVRUSURGXFWLRQVZLWKLQZHOOGHÀQHGWHFKQLFDOWUDGLWLRQV In the hope of answering some of these questions, I conducted a long series of experiments on the matter from 1988 to 1995, most of them in the Archaeological Centre of Lejre (Denmark), while studying archaeological samples from a dozen of blade production workshops (this experimental program started in fact in 1986 and 1987 with the collaboration of Bo Mad sen on a somewhat different topic). Presently, the experimental database in cludes more than 60 series (1 serie = the 15 to 40 blades from 1 blade core) produced by indirect percussion, and 25 series produced by lever pressure WKHWZRWHFKQLFDOPRGHVLGHQWLÀHGLQWKHUHOHYDQWDUFKDHRORJLFDOPDWHULDO 7KH PRVW VLJQLÀFDQW VHULHV KDYH EHHQ V\VWHPDWLFDOO\ GRFXPHQWHG GH scription, stigmata counts, photos). Two techniques were used for the detachment of large blades: indirect percussion in a few of these workshops, and pressure reinforced by a lever in the most of them, using a copper point or an antler tool. The total charac teristics that I could consider suggests that these different workshops can be UHJURXSHGLQÀYHJURXSVRUWHFKQLFDOSK\OOD0RUHRYHUWKHGHWDLOHGDQDO\VLV of the production features within similar but distant workshops allows the assumption that in some cases, it is the movement of one or a few craftsmen that resulted in the start of a new blade production workshop (for instance, IURP /H *UDQG3UHVVLJQ\ WRZDUGV WKH WKUHH NQRZQ 3UHVVLJQLDQ VSRWV LQ VRXWKZHVWHUQ)UDQFHDQGWRZDUGVWKH9HUFRUVNPWRWKHHDVW ,QWKLVDUWLFOH,ZLOOEULHÁ\GHVFULEHWKHGLDJQRVWLFFKDUDFWHUVRIWKHWZR techniques – indirect percussion and pressure – without developing the whole of my experimental documentation. I will then present a selection of different archaeological case studies. /HWXVÀUVWUHFDOOVRPHJHQHUDOSULQFLSOHVDERXWWKHLGHQWLÀFDWLRQRIDU chaeological techniques. 38 Long blade technology in the Old World Methodological principles of method and WHFKQLTXHLGHQWLÀFDWLRQ We follow the basic distinction introduced by J. Tixier between method and technique (Tixier, 1967). The method, as an intentional process more or less systematized, refers to the organization in space and time of the removals (reduction process). The technique refers to the execution modalities of these removals, includ ing three parameters: x the mode of force (Newcomer 1975), i.e. direct percussion, indirect percussion, pressure; x the nature and morphology of the tools (stones, billets, punches, pressure sticks armed with wood, antler, or copper); x the gesture and body position, the holding of the piece, etc. Several techniques can thus be used within one single method of knapping, ZKLFKH[SODLQVZK\WKHPDLQVHTXHQFHVRIWKHSURFHVVVKRXOGEHLGHQWLÀHG SULRUWRWKHLGHQWLÀFDWLRQRIWKHWHFKQLTXH V XVHG Such a distinction between method and technique is basically relevant, because the methods on the one hand and the techniques on the other must be deduced from the archaeological material through very different proce dures. The method from an archaeological production must be recognized pri marily through a technological reading – inspection – of the whole of the collection, piece by piece, with special attention to the direction and or ganization of the negatives it bears on the dorsal surface, which provide information about the preceding sequence of the knapping. The synthesis of WKHZKROHRIWKHVHREVHUYDWLRQVWKURXJKD´PHQWDOUHÀWWLQJµ IROORZLQJ- Tixier), helps to reconstruct the method of knapping which can be expressed with diacritic schemes (Inizan et al. 1999). Only when a knapping method is precisely understood can it be reproduced employing the genuine techniques and raw material, with the aim of providing quantitative references (rate of products and waste, time, etc). 2QWKHRWKHUKDQGWKHLGHQWLÀFDWLRQRIWHFKQLTXHVVWULFWO\UHOLHVRQDQ experimental reference base, as complete as possible, including at least the WZRÀUVWSDUDPHWHUV PRGHRIIRUFHDQGWRROV DQGWKHUHOHYDQWUDZPDWHULDO Indeed, merely the observation of the archaeological material does not allow for a direct recognition of techniques. The diagnostic should be established 39 Jacques Pelegrin from a comparative study of the morphological characteristics and technical stigmata from the reference collection and from the archaeological mate rial. In this matter of techniques, one can only recognize what one already knows. Thus, while the recognition of methods is a simple inductive approach, as it consists of a synthesis of the reading of the archaeological pieces, the iden WLÀFDWLRQRIWHFKQLTXHVUHTXLUHVDQDQDORJLFDOGHGXFWLYHSURFHGXUHVLPLODU to that of medical diagnosis. Our analogy can thus be extended to the nature RIWKHSURFHGXUH7KHPHGLFDOGLDJQRVLVGRHVQRWRQO\FRQVLVWRIUHÀWWLQJ frequencies of symptoms (organized in syndromes) of a cause; it is primarily EDVHGRQDQXQGHUVWDQGLQJRISK\VLRSDWKRORJLFPHFKDQLVPV IRULQVWDQFH you can hardly diagnose and treat diabetes without understanding the func tion of insulin). The same goes, I believe, for the diagnosis of knapping techniques: a technical understanding of fracture based on systematic ex periments is more effective than complex statistics. In this way we mechani cally bind the morphological characters and technical stigmata to the initial technical parameters. Presentation of the techniques Ideally, we should here discuss the characteristics of all the techniques that can produce long blades, thus taking into consideration the direct soft per cussion and soft stone percussion techniques, but this would be of little inter est for the topic of large elegant blades from the Neolithic and Chalcolithic. We will therefore limit our discussion to the distinction between indirect percussion and lever pressure. Indirect percussion This technique appears rather late during Prehistory. A few scholars, after Bordes (1968, 1969), believed they had recognized it since the Early Upper 3DOHROLWKLF LQ (XURSH EXW UHFHQW H[SHULPHQWV DQG WKH UHH[DPLQDWLRQ RI archaeological collections, do not support this assumption. At the moment, indirect percussion seems to appear and quickly spread around 7800 BP in ZHVWHUQ(XURSH QRWLGHQWLÀHGDVVXFKEXWGLVWLQJXLVKHGE\5R]R\DV ´0RQWEDQLµVW\OH DVDVSHFLÀFIHDWXUHRIWKHUHFHQW0HVROLWKLFIRUWKHSUR duction of regular bladelets for a part used as blanks for the fabrication of trapezes. It was then generally used during the Neolithic for the production RIVPDOOWRPHGLXPVL]HEODGHVDVZHOODVWRSUHVKDSHWKHEODGHFRUHPRVW 40 Long blade technology in the Old World of the time with two or three crests. In the Middle Neolithic of Belgium, at Spiennes (Michelsberg culture), large blade cores and blades were found together with an antler punch (Cels & De Pauw 1886). During the Late Neo lithic (3rdPLOOHQQLXP H[WUDODUJHEODGHVZHUHSURGXFHGLQ/H*UDQG3UHV VLJQ\ ZHVWHUQFHQWUDO)UDQFH DQGVSUHDGRYHUPRVWRI)UDQFHDQGDEURDG (Switzerland, Belgium). The indirect percussion technique consists in using an intermediary tool (punch; chasse-lame in French) to deliver the impact provoked by a mallet (a stone, wooden or antler billet) after placing the point of the punch near or at WKHHGJHRIWKHFRUHSODWIRUPLWVHOIÁDWRUGLYHUVHO\SUHSDUHG$IHZDUFKDHR logical punches (Poplin 1976, 1979, 1980) and numerous experiments helped GHÀQHWKHPRVWHIIHFWLYHSXQFKHVDQGWKHLUPHFKDQLFDOSURSHUWLHVDVZHZLOO VHHODWHU%ULHÁ\E\YDU\LQJWKHVL]HDQGFXUYDWXUHRIWKHSXQFKDQGWKHPDVV of the mallet, indirect percussion – or punch technique – can detach a very ZLGHUDQJHRISURGXFWVIURPEODGHOHWVDQGPLQXWHÁDNHVWRODUJHEODGHVDQG ÁDNHVDVODUJHDVDKDQG,WFDQLPLWDWHYHU\ZHOOWKHGLUHFWVRIWSHUFXVVLRQ technique, and attains a regularity close to that achieved by pressure. Moreo ver, experiments proved that varying the holding positions of the core allows IRUDFRQWURORYHUWKHSURÀOHRIWKHEODGHV 3HOHJULQ Pressure The pressure technique is known in the Old World since the Upper Paleo OLWKLFIRUWKHUHWRXFKRIEDFNHGSLHFHVIURPWKH*UDYHWWLDQIRUWKHÀQLVKLQJ of small and medium size Solutrean laurel leaves, and for the production of PLFUREODGHVLQODUJHDUHDVRI1RUWK(DVWHUQ$VLD )OHQQLNHQ,QL]DQ et al. 1992) and in Europe (Alix et al. 1995). During the Neolithic or a little before (Callahan 1985), pressure is used to produce bladelets and blades in ÁLQWXSWRPPZLGHZKLFKVXSSRVHVDVWDQGLQJSRVLWLRQDOORYHUWKH 0HGLWHUUDQHDQEDVLQ 7L[LHU DQGWKH0LGGOH(DVW 3HOHJULQ From the 7th millennium on, and culminating during the 3rd millennium, the use of a lever to multiply human strength is demonstrated or suspected in a dozen different production areas, extending from the Near East to Por tugal, and from Denmark to Algeria. From my experiments, I could recognize that in some cases the pressure stick was probably made out of an antler tine, while in other cases it was armed with a copper point. I could also show that the blade core had to be completely immobilized, at best in a grooved tree trunk in which the lever ZDVDOVRÀ[HG 3HOHJULQ 41 Jacques Pelegrin Morphological characteristics Morphological characteristics provide essential information for the distinc tion of indirect percussion and lever pressure. The association of three char acteristics is strongly indicative of pressure (Fig. 1): x the extreme regularity of the edges and ridges, x WKHDOPRVWVWUDLJKWQHVVRIWKHSURÀOHH[FHSWIRUDFXUYHGGLVWDOSRU tion, x the “lightness” of the section, meaning that the blades are (or can be, for some of them) relatively thin. On the other hand, the punch technique can produce light and regular blades, but with some curve. It can also produce rather straight blades, but less regular and shorter. In any case, the most regular blades produced by indirect percussion present some undulations of the edges and ridges, as well as discrete to obvious “bellies” on the ventral side (Fig. 2). This is probably because the shock of the percussion generates vibrations in the core during the detachment, when pressure is transmitted on a strictly immobile core. 7KHGLVWLQFWLRQLVPRUHGLIÀFXOWIRUUHODWLYHO\VKRUWHUEODGHVIURPWR 20 cm long. A very accurate indirect percussion can produce almost straight EODGHV LQ WKHLU ÀUVW WZR WKLUGV ZLWK D FOHDU FXUYH LQ WKHLU GLVWDO WKLUG $ ÀQDOFKDUDFWHULVWLFFDQKHOSWRLGHQWLI\WKHSXQFKWHFKQLTXHLQVXFKDFDVH discrete undulations or a slight mesial belly, which can be found by running DÀQJHUWLSDORQJWKHYHQWUDOVLGH Technical stigmata Technical stigmata can also provide important information for a diagnosis. By these technical stigmata, I refer to the character of the butt determined by the platform preparation (dimensions, aspect, edge angle) and to the discrete details determined by the detachment itself (cracks, lip, ripple on the bulb, aspect of the bulb). Using indirect percussion with an antler punch, the tests show that the DSSURSULDWHHGJHDQJOH SODWIRUPÁDNLQJVXUIDFH VKRXOGEHYHU\QHDUWRD right angle, that is from 80 to 95°, and that the contact area of the round tip of the punch should be rather large. Indeed, if the surface of the punch in contact with the platform is too small, the punch will become crushed in a few strokes. This contact surface, for blades 20 cm long, must reach a dozen VTXDUH PLOOLPHWUHV ZKHWKHU ÁDW RU VOLJKWO\ FRQYH[IDFHWWHG 7KLV PHDQV 42 Long blade technology in the Old World Fig. 1. The 37 blades (or attempts) from test LP9, detached by lever pressure in succession from a WKUHHFUHVWHGFRUHDQGSURÀOHVIURPVL[RIWKHP&RPSDUHGWRLQGLUHFWSHUFXVVLRQWKH\DUHPRUH UHJXODUZLWKDOHVVDUFKHGSURÀOH VHH)LJ 3KRWRV/HMUH5HVHDUFK&HQWUH -3HOHJULQ 43 Jacques Pelegrin Fig. 2. Top: 4 representative blades from test L1, detached by indirect percussion with red-deer antler punch. Bottom; 6 representative blades from test Cu8, detached by indirect percussion with a copper-tipped punch. The use of copper-tipped punches did not prove more effective than antler punches. Photos: Lejre Research Centre & J. Pelegrin. 44 Long blade technology in the Old World WKDWWKHEXWWZLOOEHUDWKHUWKLFN PLQLPXPPP DQGZLGH PPRU more, according to the lateral isolation of the contact area). Meanwhile with antler material, being rather elastic and somehow “spreading” on the plat form, the butt rarely presents a crack (if so, the crack remains incomplete, sketching the large diameter of the contact). For the same reason, as the fracture initiation starts a little behind the actual contact, the butt is lipped (Fig. 3a). 8VLQJDFRSSHUWLSSHGSXQFKOHDYHVFOHDUWHFKQLFDOPDUNV&RSSHULVKDUG enough to open the fracture right at the contact; thus, there is no lip at WKHFRQWDFWSRLQW )LJE $OVRFRSSHUDOZD\VSURYRNHVDFLUFXODURUVXE circular crack with a small diameter (about 3 mm) on a plane orthogonal platform, or a regular crack across the most prominent arris of a facetted or dihedral butt on which the tool was set. On large blades detached by lever pressure, the technical stigmata also help to recognize the nature of the pressure stick. With an antler pressure VWLFNRQDÁDWRUWKRJRQDOSODWIRUP RUIDFHW ZHFDQDOVRH[SHFWDFRQWDFW surface (butt) a dozen of square mm large, with an edge angle near 90°; such a butt will bear no crack but a little lip, indicating a rather diffuse (soft) con tact (Fig. 3c). However, an antler pressure stick can also be used on a facet ted or dihedral platform, with a more acute edge angle, giving a butt thick at OHDVWPP,QWKLVFDVHDVDQWOHULVPXFKOHVVKDUGWKDQFRSSHUWKHEXWW bears no crack and a clear lip is visible at the back of it, even just behind the fracture initiation point (Fig. 3d). If the pressure stick is armed with a copper point, two cases can be seen. 2QDÁDWRUFRQYH[IDFHWWHGSODWIRUPWKHFRSSHUSRLQWEHLQJSODFHGDOLWWOH behind the front edge will provoke a crack on the butt (the crack is circular RQDÁDWSODWIRUPFRUUHVSRQGLQJWRWKHFRQXVKHDGRIWKHIUDFWXUHLQLWLDWLRQ There is no lip at the contact point (Fig. 3e). In the second case, the contact surface is prepared as an acute edge parallel to the front edge of the core, or as a triedric point. When the copper point is placed on this edge or point, the EXWWZLOOEHYHU\VPDOO )LJI DQGQRFUDFNLVYLVLEOHEHFDXVHWKHEXWW blends with the contact area. There is then usually no lip behind the pressure point (because copper is hard enough to initiate the fracture at the contact), but a lip can develop laterally (Fig. 3f–4), the more with the acuteness of the global edge angle. In some cases, the fracture starts well behind the contact SRLQWDQGVXFKDWHDURXWGHWHUPLQHVDQREYLRXVOLS )LJI ,WFDQDOVR happen that the butt crushes or splits (Fig. 3f–2), seemingly after the detach ment, since this accident does not impair the fracture propagation (when the blade is just detached and starts rotating in the air, the pressure stick is still compressing the butt and can damage it). The aspect of the bulb can also help to distinguish pressure from indi 45 Jacques Pelegrin 46 Long blade technology in the Old World rect percussion. A thick bulb, high and short, is more indicative of pressure (Fig. 3e). Bulbs by indirect percussion are often more extensive in length, or diffuse (Fig. 3a). For some of the lever pressure tests, the rate of bulb scar dropped to less than 50%, compared to the indirect percussion for which this rate is generally over 50%. Looking very carefully at the tests, I could detect a few other discrete stigmata indicative of pressure, especially lever pressure: some of the blades KDYHRQHRUDIHZPLFURULSSOHVRQWKHEXOE7KH\PLJKWEHGXH,EHOLHYH WRWKHHIIHFWRIPLFURFUXVKLQJRIWKHZRRGHQVXSSRUWRQZKLFKWKHFRUH LVUHVWLQJGXULQJWKHEXLOGXSRISUHVVXUHPLFURFUXVKLQJZKLFKSURYRNHV D[LDOPLFURPRYHPHQWVRIWKHFRUH,QDIHZFDVHVDVWUDQJHULSSOHSRVLWLYH RQ RQH VLGH DQG QHJDWLYH RQ WKH RWKHU LQGLFDWHV D VXGGHQ PLFURURWDWLRQ of the core during the course of the fracture pushed by the pressure. Such LQGLFHV RI PLFURPRYHPHQWV RI WKH FRUH GXULQJ WKH GHWDFKPHQW D[LDO RU URWDWLYHFHUWDLQO\LQGLFDWHWKDWLWZDVSODFHGRUVTXHH]HGLQDÀUPGHYLFH Finally, a rare and amazing accident can prove that the core was held using D ÀUP IURQWDO VXSSRUW ,I D EODGH FRPHV RXW ZLGHU WKDQ H[SHFWHG VR WKDW its edge reaches the contact of such a frontal support, the fracture front is highly disturbed; the edge of the blade looks as though it has been torn and a lateral crack might expand in a reverse direction, i.e. towards the butt. I will show below that one archaeological blade from Syria, certainly made by lever pressure, shows such an accident. On the other hand, on the numerous *UDQG3UHVVLJQ\EODGHV,FRXOGH[DPLQH,FRXOGVHHVHYHUDOGLVFUHWHODWHUDO overpassings – taking off a little portion of the lateral crest – at a proximal, mesial or distal situation; this proves that the edges of the core were not VTXHH]HGLQDÀUPGHYLFH Fig. 3. Back views of blades detached by different techniques. a) indirect percussion using an antler punch (same as test L1 Fig. 2 top) b) indirect percussion using a copper-tipped punch (same test as Fig. 2 bottom) F OHYHUSUHVVXUHXVLQJDQDQWOHUWLQHRQDÁDWSODWIRUP WHVW/3 d) lever pressure using an antler tine on a facetted or dihedral platform (test LP23) e) lever pressure using a copper point on a facetted (test LP9; Fig. 1) and a dihedral platform (test LP13) f) lever pressure using a copper point on an acute platform edge (test LP25). Photos: Lejre Research Centre & J. Pelegrin. 47 Jacques Pelegrin Archaeological results Given this experimental understanding of these techniques, several archaeo logical cases have been examined in the past few years, some of which have been published, others demanding further documentation. In this section, we will consider a selection of cases that I could study and that have been published with at least a minimal iconographic documenta tion and discussion. The early Neolithic blades made out of EORQGÁLQWIRXQGLQ*UHHFH (C. Perlès 2004:29pp) 7KDQNV WR & 3HUOqV , FRXOG H[DPLQH WKH 1HROLWKLF ÁLQW PDWHULDO IURP Franchthi, a very large cave open near the seashore in eastern Peloponne sian, with occupations from the late Pleistocene until the Bronze Age. In the Early Neolithic dated to the 7thPLOOHQQLXP IURPDFRKHUHQWGDWH 80 BP, that is 6624 to 6378 BC cal. at 2 SD), were found a few fragments of YHU\UHJXODUEODGHVPDGHRXWRIDEORQGÁLQWRIH[TXLVLWHTXDOLW\$FFRUGLQJ to C. Perlès, such large blades are known even a little earlier from Argissa (Thessalia), in the Preceramic Neolithic, while the pressure production of VPDOOEODGHVLQGLIIHUHQWYDULHWLHVRIÁLQWDQGREVLGLDQLVDOUHDG\FRPPRQLQ the whole of Greece (Perlès 1990:131pp, Perlès 2001:89p; 202p). These large blades found in Franchthi until the Late Neolithic are all fragmented, and quite often shined and reworked (Fig. 4). Fortunately, a few well preserved proximal fragments including one surely attributed to the Early Neolithic allow for a technical diagnosis (Fig. 4; top left). The lever pressure was certainly used to produce most if not all of them, considering the striking regularity of most of the fragments, together with the very slight curvature and thinness of several pieces. The prominent bulbs, high and short, followed by clear ripples reinforce the diagnosis. The two well preserved butts are thick and safe of cracks: they indicate without doubt the use of a relatively tender material, obviously antler (copper was unknown in Greece at the time). Considering the scarcity of such blades, Franchthi certainly lies at the margin of their distribution area. As they are lacking in the Neolithic sites of 0DFHGRQLDWKHLURULJLQVKRXOGEHH[SHFWHGLQQRUWKZHVWHUQ*UHHFHRUDE road, in Albania or in the former Yugoslavia, without excluding Cephalonia DQGWKHVRXWKHDVWHUQSDUWRI,WDO\ 48 Long blade technology in the Old World )LJ'LIIHUHQWIUDJPHQWVRIEORQGÁLQWEODGHVIURP)UDQFKWKL *UHHFH 7KHWRSOHIWSUR[LPDO fragment, undoubtedly attributed to the Early Neolithic (I1), is very informative with its SURPLQHQWDQGVKRUWEXOEIROORZHGE\ULSSOHVDQGLWVÁDWDQGWKLFNEXWWFOHDUIURPDQ\FUDFN Photos: J. Pelegrin. The Varna large blades (Bulgaria; L. Manolakakis 1994, 1996, 2004, in prep.) 7KH&KDOFROLWKLF1HFURSROLVRI9DUQD %&FDO WKHOLWKLFPDWH rial of which was studied by L. Manolakakis (1994), contains a remarkable VHULHVRIODUJHEODGHVPRVWRIWKHPSURGXFHGIURPDVRFDOOHG´ZD[µSUHE DONDQLFÁLQW QRUWKHDVWHUQ%XOJDULD 7KLVSURGXFWLRQRIODUJHEODGHV )LJ FRPSOHWHGWKDWRIVKRUWHUDQGVOHQGHUEODGHVGHWDFKHGE\VWDQGLQJ SUHVVXUH )LJ DQGDWUDGLWLRQDOGRPHVWLFSURGXFWLRQRIRUGLQDU\EODGHV E\LQGLUHFWSHUFXVVLRQ )LJ ZKLFKDUHDOVRUHSUHVHQWHGLQVRPHRIWKH Varna graves. Conversely, large blades are found as retouched fragments in domestic settlements. 49 Jacques Pelegrin In one of the richest grave of the Varna Necropolis, that contained 225 JROG REMHFWV ZDV DOVR IRXQG WKH ORQJHVW ÁLQW EODGH NQRZQ VR IDU ZLWK D OHQJWKRIFP )LJ 2QWKHZKROHWKHVHODUJHVHOHFWHGEODGHV QHL WKHU FRUWH[ QRU FUHVWQHJDWLYHV DUH WR FP ORQJ WR PP ZLGH and 3 to 8 mm thick only. They are very regular, with the exception of some WKDW DUH UHJXODUO\ ZDY\ LQ WKH SURÀOH 7KHLU FXUYDWXUH LV VOLJKW WR PRGHU ate, increasing in the distal portion. The diagnosis of lever pressure stands without doubt, as well as the material of the pressure point: a small circular FUDFNFOHDUO\YLVLEOHRQWKHÁDWDQGUHODWLYHO\ODUJHEXWWVGHPRQVWUDWHVWKHXVH of copper. Similar observations could be made on similar blades from early French excavations in Bulgaria. Such a production can result from a local development, as standing pres sure was already known in Bulgaria. However, it could also be rooted in WKHEORQGÁLQWWHFKQRORJLFDOWUDGLWLRQLIZHUHPHPEHUWKDWWKHODWHVWEORQG ÁLQWEODGHVIRXQGLQ*UHHFHH[WHQGXQWLOWKH*UHHN5HFHQW1HROLWKLFZKLFK overlaps the Karanovo VI period of the Varna Necropolis. (princeps contex tual and documental ref. from L. Manolokakis; 1996, 2005.) The blades from Tell’Atij and Tell Gudeda (Syria, Early Bronze Age; J. Chabot 2002) /DUJHÁLQWEODGHVRUWKHLUIUDJPHQWVFDOOHG´&DQDDQHDQEODGHVµKDYHORQJ been described from the Levant and Mesopotamia, although very little is NQRZQDERXWWKHLURULJLQDQGSURGXFWLRQ'LIIHUHQWYDULHWLHVRIÁLQWVHHP to be represented in the whole, and there might have been several produc tion centres of these “Canaaean blades” – and possibly different techniques – over time (4th millennium and early 3rd millennium) and within a large DUHD IURPVRXWKHDVW7XUNH\WR/HEDQRQ5RVHQ&KDERW Our colleague J. Chabot has been in charge of the study of two settlement collections carefully excavated in northern Syria by M. Fortin from Laval University (Canada). Most of the lithic material consists of locally produced ÁDNHVEXWERWKVLWHVFRQWDLQDERXWODUJHÁLQWEODGHVWKHUDZPDWHULDO of which is not local. Indeed, these blades have been imported as long frag ments from one or several workshops (ibid: 20). A study of their function indicates that they have been mainly used as elements of threshing sledges. Using my experimental references, J. Chabot and I have conducted a careful examination of the two large blade collections which are extremely VLPLODU2IWKHEODGHVLQ7HOO·$WLMFDQSRVLWLYHO\EHDWWULEXWHGWROHYHU SUHVVXUHIURPWKHLUH[FHOOHQWUHJXODULW\DOPRVWVWUDLJKWSURÀOHDQGUHODWLYHO\ “light” section (most of them are 22 to 33 mm wide and 6 to 10 mm thick) 50 Long blade technology in the Old World Fig. 5. Blades from different graves of the Varna Necropolis (Bulgaria); blades 1 and 4 from grave no.1, blade 3 from grave no. 26, blade 2 from grave no. 40. Drawings after L. Manolakakis 1994, photos: P. Kelterborn. 51 Jacques Pelegrin Fig. 6. Blades from Tell’Atij (Syria). Photos: H. Plisson (CNRS), J. Chabot & J. Pelegrin. Fig. 7. a) Blade from Tell Gudeda (Syria), showing a “reverse” scratching of the edge; b) crested blade from Tell’Atij. Photos: H. Plisson (CNRS), J. Chabot & J. Pelegrin. 52 Long blade technology in the Old World (Fig. 6). Indirect percussion was recognized for 15 other blades, most of them being obvious shaping blades saved during the preparation or opening of the EODGHFRUHRUVKRUWHUEODGHVFRPLQJIURPWKHÀQDOUHGXFWLRQRIFRUHV7KLV diagnosis was rather conservative, as 202 blades remained undetermined, 44 of which were very regular but too short for an estimation of the overall UHJXODULW\DQGSURÀOHDQGEODGHVVKRZLQJDPELJXRXVRULQGLVFULPLQDWH features (remembering that pressure can produce rather irregular blades, and that indirect percussion can give blades that are only slightly different from the best lever pressure). In Tell Gudeda, the total of 62 blades gave 35 lever pressure, 5 indirect percussion and 22 undetermined. $ YHU\ VSHFLÀF GHWDLO FRXOG EH REVHUYHG RQ D EODGH IURP 7HOO *XGHGD (n° S35): one of its sides shows a strange “reverse” scratching (Fig. 7a). The same happened during our experiments, when the fracture front of a blade FRPHVLQWRFRQWDFWZLWKWKHGHYLFHLQZKLFKWKHEODGHFRUHLVLPPRELOL]HG This is one more argument, should one be needed, for the diagnosis of pres sure, and a valuable indication that the device used by the original producers was similar to our experimental model. The butt of the lever pressure blades is very typical, with a pronounced arris. The thickness of the butt is usually about 4 to 5 mm, and the bulb is generally high and short, many of them bearing ripples (Fig. 6). Indicative of the metallic nature of the pressure tool, the impact point is well marked, occasionally with a clear crack on the butt arris, and only a very few butts showing a lip (which is a minor variation, when the initiation of the fracture starts behind the effective contact). Indeed, the use of metallic knapping tools is also demonstrated in each of the collection by a lateral crested blade with very small and clear circular cracks from indirect percussion shaping ÁDNHV )LJE 2QWKHRWKHUKDQGVHYHUDORIWKHOHVVUHJXODUEODGHVLGHQWL ÀHGDVLQGLUHFWSHUFXVVLRQEHDUDWKLFNDQGODUJHEXWWZLWKQRFRQFHQWUDWHG impact marks (showing that “soft” indirect percussion was also used for the shaping or reshaping of the cores). The long blades from Pauilhac, produced in ´EDQGHGµ2OLJRFHQHÁLQWIURPWKHVRXWK east of France ,QÀYHEODGHVZHUHGLVFRYHUHGWRJHWKHUZLWKRWKHUUHPDUNDEOHPDWHULDO such as human bones, axes and gold during the building of a railway 80 km WNW of Toulouse (Cantet 1991, Bischoff & Canéto 1865). At the time, the largest of them was the longest blade known in the south of France, with a 53 Jacques Pelegrin length of 34.5 cm. This was exhibited during the “Exposition Universelle de Paris” in 1867 before being deposited in the Musée des Antiquités Nationales DW6DLQW*HUPDLQHQ/D\H7KDQNVWR&/RXERXWLQIURPWKH0$1DQG$ 5RXVVRW IURP WKH 0XVpH G·$TXLWDLQH ZKHUH WKH RWKHU EODGHV DUH VWRUHG , could study and draw these pieces (Fig. 8). 7KHVHÀYHEODGHVDUHH[WUHPHO\UHJXODUDQGUHODWLYHO\WKLQWKH\ZHUHRE viously detached by lever pressure. The butt of them is very tiny: the pressure point was set on the very edge of the platform (angle 65° to 90°, with a lateral slope of 10° for blades 2 and 4) after the overhang from previous blades was removed. Note that the butts of blades 1 and 3 are discreetly split, as hap SHQHGWRWKHH[SHULPHQWDOEODGH/3GHWDFKHGZLWKDFRSSHUSRLQWDIWHU a similar preparation of the platform (Fig. 3f). Indeed, an antler point would certainly spread on the platform and determine a thicker butt. Blades 2 and 4 also bear a discrete transversal crack, indicative of a metallic point. Blade ZLWKLWVÁDWDFXWH EXWWZLWKRXWFUDFNDQGDVOLJKWOLSDOVRUHVHPEOHV H[SHULPHQWDOEODGH/3 )LJI 7KHVHÀYHEODGHVDUHPDGHRIWKHVDPHYDULHW\RIÁLQWDEDQGHGEURZQ EHLJHÁLQWWKDWSDWLQDWHVLQWRDFUHDP\FRORXUZKLFKSUHVHUYHVWKHEDQGVSD UDOOHOWRWKHORQJD[LVRIWKHEODGHV$IHZIUHVKQRWFKHVDQGEUHDNVFRQÀUP the quality of the raw material, which is a little “dry” (dull). This is almost FHUWDLQO\DWHUWLDU\ÁLQWRFFXUULQJLQYHU\ODUJHÁDWURXQGHGQRGXOHV7KH only known source in the south of France of such raw material, regarding its quality and size, are the tertiary limestone (Oligocene, Aquitanian) outcrops RIWKH)RUFDOTXLHUDUHDLQWKHVRXWKHDVWRI)UDQFH $OSHVGH+DXWH3URYHQ ce, vallées du Largue, du Cavalon et du Lez), where large workshops of blade production have long been recognized (Courtin 1974). Very common in the Chalcolithic collective graves from southeast France (Sauzade 1983), these blades are still frequent in the Languedoc (Vaquer 1990) and are even expec ted in the Spanish Catalunya (pers. observation). In the north, they reached the Saône valley, the French Jura and Switzerland (Honegger 2001). In a preliminary synthesis of his doctoral research, S. Renault (1998) presents such blades produced in the Forcalquier region during the Late Neolithic and Chalcolithic (about 3 500 to 2 500 BC cal.). He also describes VKRUWHU EODGHV IURP D GLIIHUHQW UHJLRQDO ÁLQW WKDW EHDU LGHQWLFDO WHFKQLFDO stigmata, on which traces of copper have been recently analyzed (Renault et al. in prep.). Apart from these particular blades, others in the same variety of )RUFDOTXLHUÁLQWDUHNQRZQZKLFKDUHWKLFNHUDQGOHVVUHJXODUXVXDOO\ZLWK a thick facetted butt (cf. some blades from La Couronne in the Martigues museum, see also Renault 1998: Fig. 4). They seem to have been detached by indirect percussion. 54 Long blade technology in the Old World )LJ7KHÀYHORQJEODGHV from Pauilhac (France). Drawings: J. Pelegrin & M. Reduron. 55 Jacques Pelegrin The large blades from Perfugas (Sardinia; L.Costa et J. Pelegrin, 2004) Ongoing excavations at Contraguda (Perfugas, northern Sardinia) revealed D ODUJH VHWWOHPHQW LQFOXGLQJ D ZRUNVKRS IRU WKH SURGXFWLRQ RI ODUJH ÁLQW blades, roughly dating to the Chalcolithic (end of the 4th millennium – early 3rd millennium BC cal.) (Boschian et al. $ WDEXODU EHGGHG ÁLQW DW tributed to the Oligocene occurs in the vicinity. Slabs 4 to 5 cm thick with a light brown colour, including lighter or darker parallel bands that might FRPHIURPDVSHFLÀFEHGZHUHVHOHFWHGIRUWKHSURGXFWLRQRIODUJHEODGHV 7KHFRUHURXJKRXWVZHUHSUREDEO\VKDSHGRQRUQHDUWRWKHH[WUDFWLRQVSRW DV ODUJH VKDSLQJ ÁDNHV DUH H[WUHPHO\ VFDUFH LQ WKH ZRUNVKRS PRVW RI WKH SUHFRUHVKDGWREHVKDSHGE\WUDQVYHUVDOÁDNHVWRFRUUHFWWKHSURÀOHRIWKH blade detachment surface). Two types of blades have been observed in the workshop: x Fragments of rather thin and regular blades with a symmetrical trap ezoidal section, most of them central blades (without cortex or trans YHUVDOVFDUV WKHEXWWRIWKHVHEODGHVLVYHU\WKLQ )LJWR x Complete blades or fragments with a thicker section and a less regu ODUSURÀOHZLWKDFRUWLFDOVLGH )LJWR WKHEXWWRIWKHVHEODGHV is thick, forming an edge angle about 90°, with a little lip. The thin and regular blades are certainly detached by lever pressure and DFRSSHUWLSSHGSUHVVXUHWRROEHFDXVHRIWKHLUYHU\WLQ\EXWW7KHVWULNLQJ similarity with the blades from Pauilhac should be noted (cf. Fig. 8). Again, such a minute contact point is not compatible with an antler tool, be it by lever pressure or by indirect percussion. The large butt of the thicker lateral blades indicates another technique, which is obviously indirect percussion. At the moment, the distribution area of these blades from Sardinia re mains to be investigated. Some of them which were collected in different sites of southern Sardinia are exhibited in and documented by the University of Cagliari (Atzeni 2000), where I could also examine them, thanks to our colleague Carlo Luglie. 56 Long blade technology in the Old World Fig. 9. Blades from Contraguda (Perfugas, Sardinia), after L. Costa & J. Pelegrin 2004 (redrawing: G. Monthel). 57 Jacques Pelegrin Large blades found in Portugal In the collections of the National Museum of Portugal (Museu Nacional de Arqueologia, near Lisbon), I could document large blades made out of different W\SHVRIÁLQWWKHRULJLQVRIZKLFKUHPDLQVWREHFRQÀUPHGRULQYHVWLJDWHG )LYHEODGHVRUIUDJPHQWVWRJHWKHUZLWKVWRQHD[HVÁLQWEODGHVDQGSRLQWV amber, copper, etc. come from the passage grave Alcalar III, one of seven PHJDOLWKLFPRQXPHQWVNPIURPVRXWKZHVWHUQ3RUWXJDOSXEOLVKHGODVW FHQWXU\E\(VWjFLRGD9HLJD DQGDWWULEXWHGWRWKH&KDOFROLWKLFSHULRG (V.S. Gonçalves, 1989). According to the artistic drawings from A.J.N. da Gloria in the original publication, there were seven large blades in “chert”, but two of them and other fragments are currently missing. What is tradi tionally called “chert” by our Portuguese colleagues is in fact a variety of OLJKW JUH\ ÁLQW WKDW WXUQV LQWR D FUHDP\ FRORXU ZKHQ SDWLQDWHG DQG WKDW has a rather coarse grain. The surface of the cortex is rough, but the shape RIWKHQRGXOHVFDQEHYHU\UHJXODUZLWKWZRSDUDOOHOÁDWRUVOLJKWO\FRQYH[ sides and a rounded peripheral edge. Indeed, discrete structural bedding ap pears on some blades, the material being coarser towards the centre of the nodule. Most of these blades are massive. Judging from the original drawings, four of them are very long, from 27 cm up to almost 40 cm. They are also wide WRPP DQGWKLFN WRPP ZLWKDVOLJKWO\FXUYHGSURÀOHEHIRUHD clear distal curvature. The regularity is rather good, considering that four of WKHPDUHFUHVWHG )LJ RU´XQGHUFUHVWµ )LJ EODGHV,QSDUWLFXODU WKHODVWRIWKHÀYH )LJ ZLWKDWUDSH]RLGDOVHFWLRQGHWHUPLQHGE\WKH scars of three very regular preceding blade, is especially informative. After DSURQRXQFHGSUR[LPDOWRUVLRQLWIROORZVDYHU\VWUDLJKWSURÀOHDQGDWKLQ section (8 mm), and reaches a length of 37 cm after a distal curvature. Three of these blades have a preserved proximal end. The butt of a heav LO\FUHVWHGEODGHWKDWKDVEHHQSUHVKDSHGE\LQGLUHFWSHUFXVVLRQ )LJ no.10027) is slightly damaged by the removal of a minute splinter, during or after the detachment. However, the original shape of the butt can be evalu ated as symmetrically dihedral and 2 mm thick, with a possible margin, 1 mm wide. The angle and the lip aspect cannot be evaluated. The two other butts are also dihedral but very asymmetrical. However, the tool was not identically placed for these two blades. On no. 10028 (Fig. WKHFRQWDFWDUHD PPWKLFN ZDVWKHSURWUXGLQJGLKHGUDO DUULV of the butt, slightly smoothed prior to the detachment and forming an edge angle of 80°. The contact area was thus very small and potentially aggres VLYH2QQR )LJ WKHFRQWDFWDUHDZDVDFWXDOO\MXVWEHVLGHWKH protruding dihedral arris of the butt. The effective contact was produced on 58 Long blade technology in the Old World Fig. 10. Different blades from Portugal; blades 1, 2 & 3 from Alcalar III, 4 & 5 from Aljezur, 6 from Vila Cha. Drawings: J. Pelegrin & G. Monthel. 59 Jacques Pelegrin a slightly concave facet, discretely smoothed towards the platform, forming a 65° edge angle and laterally 20° inclined. On both blades, there is no lip EHKLQGWKHFRQWDFWDUHDEXWDOLSDSSHDUVODWHUDOO\EHKLQGWKHEXWW·VIDFHWV which form an acute edge angle (for instance on the left of the contact area of blade no. 10026). Note that blade 10027 bears a very clear positive ripple at the lower part of its bulb, as well as the previous blade detachment visible on 10026. $WÀUVWVLJKWRQHFRXOGWKLQNWKDWVXFKPDVVLYHEODGHVZHUHSURGXFHGE\ powerful indirect percussion, but not necessarily. Lever pressure can also detach very heavy blades, up to 6 cm wide in some of my experiments. The morphology of blade no. 10028, 37 cm long combined with a very moder ate curvature, a very good regularity of its edges and arrises and a rather thin section (8 mm) indicate that it was not detached by indirect percussion. Indeed, the technical stigmata provide several arguments in favour of lever pressure: the very small contact area, the two clear ripples visible on these three blades, as well as the torsion of blade no. 10028. Moreover, the tiny dimensions of the contact area and the absence of lip behind the contact area certainly implied a rather hard material, that is to say copper. With an edge angle of respectively 65° and 80° (blades nos 10026 and 10028), one would expect a clear lip behind the contact area, if antler had been used. I now turn to a few blades from Aljezur, a collective grave 25 km from VRXWKZHVWHUQ 3RUWXJDO H[FDYDWHG YHU\ HDUO\ GXULQJ WKH WK FHQWXU\ 9 Gonçalvès considers Aljezur a Chalcolithic “fosse funéraire” (1992:172). 7KHHOHJDQWEODGHQR )LJ LVPDGHIURPDGLVFUHWHO\EDQGHG OLJKWJUH\ÁLQWYHU\KRPRJHQHRXV7KHYHU\JRRGUHJXODULW\UHGXFHGFXUYD ture, light section (25 mm wide, 5 mm thick in the mesial portion) together with a length that would reach 24 or 25 cm but for a fresh distal break clearly indicate lever pressure. The bulb is rather diffuse, with its lower limit marked E\DVRIWULSSOH QRQGLDJQRVWLF DQGLVIROORZHGE\WZRRWKHUVRIWULSSOHV 3 cm from the butt. The butt is dihedral symmetric and a little smoothed, forming a 70° edge angle. Its thinness (2 mm) indicates a copper tipped pressure tool, even though there is a little lip (due to the rather acute edge angle). 7KHUDZPDWHULDORIEODGHQR )LJ LVDYHU\KRPRJHQHRXVÀQH JUDLQÁLQWLQDOLJKWJUH\FRORXU7KHEODGHLVUHPDUNDEO\VOHQGHU WR mm) and thin (4 to less than 3 mm) compared to its length (186 mm long) and reduced curvature. The medium regularity and the triangular section suggest that it comes from an early stage of the blade core reduction. Despite the medium regularity, the morphological features are enough to strongly suggest a pressure technique, and possibly “standing pressure” in regard to 60 Long blade technology in the Old World its narrow section. The butt and bulb features further support the diagnostic of pressure. The bulb is well rounded, short and high, with a clear ripple a few millimetres just below the butt, which probably indicates that the blade ZDVÀQDOO\GHWDFKHGDIWHUDVHFRQGSXVK7KHEXWWLVYHU\WKLQ PP GL hedral symmetric, with a 90° edge angle and a little lip. This is not enough to determine the material of the pressure stick, but we are inclined to think that it was copper. Attributed to the same collection of Aljezur, blade no. 8986 is retouched E\SUHVVXUHDVDGRXEOHHQGVFUDSHU$OWKRXJKWKHEXWWLVPLVVLQJWKHPRU phological characters (length 17 + 3 cm, width 27 mm, thickness 4 mm, slight curvature, excellent regularity) indicate a lever pressure detachment. 7KHÁLQWLVYHU\ÀQHO\JUDLQHGZLWKDPDUEOHGOLJKWEURZQFRORXU$QRWKHU blade no. 9003 is very retouched and broken, but can be estimated to have originally been 20 to 25 cm long, about 3 cm wide and 9 mm thick. The small dihedral butt and the regularity of the ventral face suggest a lever pressure detachment. The greenish and slightly grainy material is unusual; a preliminary investigation of T. Aubry in the University of Coïmbra indicates a “contact silexite”, i.e. a local metamorphic formation due to the contact of extrusive melted rhyolite (provisional name “green rhyolite”, or “green silexite”). Still in Portugal, but about 135 km north of Lisbon and 10 km south of 3RPEDOWKH9LOD&KDJUDYHVKHOWHU FDOOHG´$EULJRGH6RXWRµ JDYHDODUJH collection of archaeological objects to the “Grupo de Arqueologia e Espe leologia de Pombal”. Together with numerous human bones, the collection LQFOXGHVÁLQWEODGHVDUURZSRLQWVZKLFKFDQEHDWWULEXWHGWRWKH/DWH Neolithic or Chalcolithic, 2 fragments of ground axes and a few potsherds. Thanks to the president of the group Mario Sacramento, I could study this FROOHFWLRQGHSRVLWHGDWWKHWRZQKDOORI3RXVDGDV9HGUDV $JUHDWPDMRULW\ RIWKHEODGHVLVPDGHRI´&D[DULDVÁLQWµZKLFK7 Aubry and I could sample on a vast area north of Caxarias (16 km south from Vila Cha); also known for its numerous knapping scatters (Zilhao 1994). Their variety of colour and grain indicate that these 114 blades come from PDQ\GLIIHUHQWEODGHFRUHVDQGLWZDVQRWSRVVLEOHWRUHÀWDQ\RIWKHEODGHV They seem to represent the result of occasional selection from a living stock RIIUHVKWRROVPDQ\RIWKHPEHDUDOLPLWHGUHWRXFKRUPDFURXVHZHDU EXOE removed by a thin inverse retouch, direct truncation at one end or at both ends, awkward inverse retouch at the end in an attempt to reduce the distal curvature). The general calibre of these 114 blades ranges from 18 to 12 cm in length, WRPPLQZLGWKDQGWRPPLQWKLFNQHVV7KHSURÀOHRIWKHEODGHV is moderately curved in the proximal and mesial portions, the curvature ex 61 Jacques Pelegrin panding in the distal third. Their regularity is generally good but not ex tremely so; waves are noticeable on the arrises and edges, especially for the longest specimen, more discrete or even absent on the medium long and lighter blades, some of which are also slightly straighter than the longest EODGHV)LJXUHLOOXVWUDWHVDW\SLFDOVSHFLPHQRIWKHVHEODGHV From the 75 blades from which the proximal end is preserved, the butts DUH YHU\ UHSHDWHGO\ WKLFN DQG ZLGH SUHSDUHG E\ D ÁDW RU VOLJKWO\ FRQYH[ facetting of the platform, without removing the overhang from the previous blade detachment. Indeed, the preservation of the concavity of the previous EXOEQHJDWLYHKHOSHGWRGHWDFKWKLQÁDNHVWRZDUGVWKHSODWIRUPLQRUGHUWR facet the contact zone for the detachment of the coming blade. The global edge angle formed by the facetting regularly averages 90°. Only a few partial cracks can be observed on the butts, in various positions, indicating a rather ODUJHFRQWDFWDUHD7KHEDFNOLQHRIWKHEXWWVLVUDWKHUUHJXODUZLWKDGLVFUHWH but clear lip. The bulb of these blades is variable in prominence and size. No really clear ripple could be observed in the proximal portion of the blades, but discrete and soft waves. The morphological and technical features contribute to the same diag nostic: indirect percussion. The limited length is an argument in itself, to which could be added that that the longest blades are more curved and less regular. The ideal range for the best systematic control stays within the limit RI FP D OLPLW WKDW ZH FRXOG FOHDUO\ REVHUYH IURP RXU H[SHULPHQWV ZLWKLQGLUHFWSHUFXVVLRQXVLQJZHOODGDSWHGSXQFKHVDQGSUHIHUDEO\DVKRUW elastic support. The technical stigmata correspond very well with such an interpretation (a direct soft percussion can be immediately eliminated because of the butt aspect and edge angle, as well as soft or hard stone percussion). Such thick butts could by themselves be compatible with a copper point acted by lever SUHVVXUH EXW WKLV ZRXOG OHDYH PDQ\ PRUH DQG VXEFLUFXODU FUDFNV RQ WKH butts, and no lip at the contact point. Lever pressure with an antler tine would provoke much more prominent bulbs on such thick butts. )LQDOO\DQGORJLFDOO\LWLVGLIÀFXOWWRXQGHUVWDQGZK\WKHSUHKLVWRULFNQDS pers would have used the lever pressure to produce blades that would be as HDV\WRSURGXFHZLWKDZHOODGDSWHGLQGLUHFWSHUFXVVLRQ Other similar blades, detached by indirect percussion on a thick facetted butt, could be observed in different collections from the central and south HUQUHJLRQVRI3RUWXJDO0RVWRIWKHPDUHVHHPLQJO\PDGHRI&D[DULDVÁLQW FRQÀUPLQJWKHLPSRUWDQFHRIWKHRULJLQDOZRUNVKRS+RZHYHUWKHUHDUH DOVRVRPHRIWKHPLQGLIIHUHQWYDULHWLHVRIÁLQWWKHRULJLQRIZKLFKLVSUHV ently unknown (according to T. Aubry). Conversely, I could not recognize (lever) pressure blades from Caxarias 62 Long blade technology in the Old World ÁLQW6LPLODUO\WKHOHYHUSUHVVXUHEODGHVPDGHRXWRI´FKHUWµJUHHQVLOH[LWH ÀQH JUH\ ÁLQW DQG PDUEOHG OLJKW EURZQ ÁLQW VHHP WR FRPH IURP VSHFLÀF workshops, the location of which remains unknown. A survey done together with T. Aubry in the South of Portugal combined with further information suggest that the “chert” and the green silexite may well correspond to the ´YROFDQLFVHGLPHQWDU\µ IRUPDWLRQ IRXQG QRUWK RI +XHOYD VRXWKZHVWHUQ Spain), where large blade cores are documented (Linares et al. 1998). The YDULHWLHVRIJUH\DQGEURZQÁLQWUHVHPEOHVDPSOHVIURP$QGDOXVLDZKHUH VHYHUDO ODUJH RXWFURSV DQG ZRUNVKRSV KDYH EHHQ LGHQWLÀHG LQFOXGLQJ WKH SURGXFWLRQRIUHJXODUEODGHVZLWKDFOHDUGLKHGUDOEXWW 0DUWLQH])HUQDQGH] 1997, Morgado 2002). The Grand-Pressigny long blades detached from the “livre-de-beurre” cores (western France; J. Pelegrin 2002) 7KH*UDQG3UHVVLJQ\ZRUNVKRSVDERXWWRNPVRXWKRI7RXUVKDYH EHHQLGHQWLÀHGVLQFHWKHWKFHQWXU\DVDODUJHSURGXFWLRQFHQWUHRIORQJ blades, further diffused as “daggers” in the whole of France and even up to 6ZLW]HUODQG DQG WKH 1HWKHUODQGV 0DOOHW 'HOFRXUW9ODHPLQFN Honegger 2001, Ihuel 2004). In 1970, a cache of 134 to 138 fresh blades was discovered at “La Creusette” and carefully excavated (Geslin et al. 1975). I FRXOGODWHUGRFXPHQWWKDWWKH\UHSUHVHQWDQRQVHOHFWHGIUDFWLRQRIDQRULJL nal production of 500 to 800 blades, detached from about 50 to 80 cores (Pe OHJULQ 7KH\FDQWKXVEHFRQVLGHUHGDQH[FHOOHQWVDPSOHRIWKH*UDQG Pressigny production. These blades are massive: 25 to 38 cm long, 4 to 6 cm wide and about 1 to 1.5 cm thick (Geslin et al. 1975, Kelterborn 1980, Pelegrin 1997). They VKRZWZRW\SHVRISURÀOHWKHHDUO\EODGHVKDYHDUHJXODUDQGSURQRXQFHG curvature, and the later blades are typically more curved in their proximal KDOI )LJWR DQGOHVVFXUYHGLIQRWSXUHO\VWUDLJKWLQWKHLUGLVWDOKDOI )LJWR 7KH\DUHQRWYHU\UHJXODUDOZD\VEHDULQJPRGHUDWHWRRE vious undulations of their edges and arrises (together with corresponding ZDYHVRULQÁH[LRQVRIWKHLUYHQWUDOVLGH 7KXVWKHPRUSKRORJLFDOFKDUDFWHUV RI*UDQG3UHVVLJQ\EODGHVDUHPXFKPRUHLQGLFDWLYHRILQGLUHFWSHUFXVVLRQ than pressure. The technical stigmata are very repetitive. The platform preparation con sists of shaping a clear arris in the line of the blade to be detached, and gently SHFNLQJWKLVDUULV ZLWKWKHHGJHRIDVSHFLÀFÁLQWWRROZKLFKKDVODWHO\EHHQ 63 Jacques Pelegrin LGHQWLÀHGFDOOHG´piqueteur”; Geslin et al. 1982). This pecking progressively crushes the arris into a rectangular or trapezoidal surface, about 7mm long and 2 mm wide, which will serve as the contact surface to the knapping tool DQGEHFRPHWKHEXWWRIWKHEODGH )LJSKRWR ,WKDVORQJEHHQVDLGWKDW this systematic preparation helps to adjust the geometry of the contact area UHJXODUO\LQFOLQHGZLWKDDQJOHWRWKHÁDNLQJVXUIDFH DQGWKDWLWFDQ provide a helpful grip for the knapping tool. However, more importantly, it also facilitates the initiation of the fracture. The bulb of the blades is re duced or diffused, and never bears any ripple or ridge (this has been checked systematically). Also, different blades with a side overpassing in proximal, mesial or distal position prove that the core was not held or squeezed in a ZRRGHQYLFHRUGHYLFHZKLFKDOVRVSHDNVDJDLQVWOHYHUSUHVVXUH7KH*UDQG Pressigny blades had to be detached by indirect percussion. Reproducing the blades proved to be a great challenge. All the early tests IDLOHGZLWKXQGXODWLQJGHWDFKPHQWVDQGRUDÀQDOKLQJHDWFPIURP the platform, until I understood that the problem was the recoil and shak ing of the core under the percussion (or very shortly after, in the way of a gun). Assuming that fracture speed is constant, detaching a 30 cm long blade would take twice the time than a 15 cm blade. The solution of resting the core standing on an elastic support (a piece of wood acting like a spring) proved effective, together with a subtle adaptation of the punch and ham mer, so that regular blades up to 35 cm long could be obtained. In addition, WKHVSHFLÀFSUHSDUDWLRQRIWKHSODWIRUPVKRXOGEHVHHQDVDYHU\FOHYHUZD\RI shortening the delay between the impact and the fracture initiation. In this complicated mechanical problem. that can only be approached intuitively, WKH*UDQG3UHVVLJQ\EODGHPDNHUVFHUWDLQO\SURYHGDUHPDUNDEOHHPSLULFDO LQWHOOLJHQFH,QGHHGWRRXUNQRZOHGJHWKH*UDQG3UHVVLJQ\EODGHSURGXF tion constitutes the best performance ever produced by indirect percussion. 6LQFH WKH WK FHQWXU\ VHYHUDO VFDWWHUV RI W\SLFDO OLYUHGHEHXUUH FRUHV KDYH EHHQ LGHQWLÀHG LQ VRXWKZHVW )UDQFH XS WR NP VRXWK IURP /H *UDQG3UHVVLJQ\ 3HOHJULQ :LWKDERXWDKXQGUHGFRUHVRUOHVVLQHDFK they seem to correspond to a season of activity by one individual. The “Pres VLJQLDQµZRUNVKRSIURP9DVVLHX[HQ9HUFRUV 0DOHQIDQWet al. 1971), about NPHDVWRI/H*UDQG3UHVVLJQ\LVPXFKPRUHFRQVLVWHQWZLWKDQHVWL mated 5 to 10 thousands of cores (more than 1160 are already recovered). $QRWKHU VSRW RI OLYUHGHEHXUUH FRUHV LV DOVR NQRZQ LQ QRUWKHUQ 9HUFRUV DQGVXFKDZRUNVKRSLVDOVRDVVXPHGLQQRUWKHDVW)UDQFHQHDU5HLPV&XUL ously, the Pressignian way of shaping the core was not the best method for ZRUNLQJWKH9HUFRUVWKLFNIUDJPHQWVRIQRGXODUÁLQW,WZDVWKXVLQWURGXFHG there at some time between about 2800 and 2400 BC (cal.), which is the pe ULRGRIPDWXUHW\SLFDOSURGXFWLRQDW/H*UDQG3UHVVLJQ\ZKHUHSUHFXUVRU\ 64 Long blade technology in the Old World Fig. 11. Seven blades from the “La Creusette” cache (Barrou near Le Grand-Pressigny, France), selected according to their respective order in the core reduction, and upper view of the typical butt of such blades. Drawings from G. Bastien after Geslin et al. 1972 & 1975, photo: M. Geslin (et al. 1975). 65 Jacques Pelegrin VKDSLQJPHWKRGVGDWHGWRDERXW%&DUHQRZLGHQWLÀHG 3HOHJULQ in prep.). It is unrealistic to imagine that the Pressignian method and technique, which are so particular and precise down to their details, may have travelled over hundreds of km by mouth to ear. Such distribution implies that some of WKHFUDIWVPHQRUGLQDULO\ZRUNLQJDW/H*UDQG3UHVVLJQ\KDSSHQHGWRWUDYHO very far away from it, and were probably informed on their way about the DYDLODELOLW\RIODUJHÁLQWQRGXOHV&RQVLGHULQJWKDWVKRUWDQGPHGLXPGLV tance distribution obviously follows other mechanisms (Mallet 1992), this evidence of travelling craftsmen may account for a direct transportation of a portion of the production by the producers themselves (whether ordered or not, and/or further controlled by other individuals). Conclusion At a methodological level, I would like to stress that the experimental refer ences should not be considered complete. At different steps of the research, it appeared that complementary tests had to be made when facing new types of platform preparation or particular varieties of archaeological raw material WKDWFRXOGLQÁXHQFHWKHGLDJQRVWLFVWLJPDWDDVSHFWRURFFXUUHQFH'LDJQRV tic guesses based on allusive or preliminary knowledge can be uncertain to the point of being later contradicted by further more reliable experiments consisting of minute details of platform preparation and knapping tool ad aptation. $QRWKHUGLIÀFXOW\FRPHVIURPWKHIDFWWKDWLQGLUHFWSHUFXVVLRQZDVIUH quently associated with lever pressure, as can be seen in several cases such as LQ%XOJDULDVRXWKHDVWHUQ)UDQFH6DUGLQLDDQG6\ULD DQGSUREDEO\LQRWKHU productions under study such as Spiennes in Belgium and northern Spain). $VKRUWFXWUHDVRQLQJVXFKDVDVVXPLQJWKHWHFKQLTXHIURPWKHUDZPDWH ULDOLGHQWLÀFDWLRQZRXOGEHVFLHQWLÀFDOO\LQFRUUHFWDQGSUDFWLFDOO\XQFHUWDLQ Moreover, it would compromise possibly interesting interpretations if punch blades were not accorded the same value as pressure blades, whether prima rily (in the workshop) or secondarily (during distribution or use). It may also be possible in some cases that the detachment technique retains a chronolo JLFDOVLJQLÀFDQFHLQWKHHYROXWLRQRISURGXFWLRQWKURXJKWLPH In this line, the recognition of techniques on archaeological material re mains as much a delicate matter as an interesting one. At an archaeological level, the few cases presented here allow for some brief considerations. 66 Long blade technology in the Old World 7KHEORQGWUDQVOXFHQWÁLQWEODGHVIRXQGLQ*UHHFHLQGLFDWHDVXUSULVLQJO\ early production (mid 7th millennium BC cal.) using the lever pressure tech nique and an antler point. It is certainly a genuine innovation rooted in the practice of standing pressure blades, and seems to last over the Middle and Late Neolithic without being imitated elsewhere. 7KHQH[WSURGXFWLRQDSSHDUVLQQRUWKHDVW%XOJDULDLQWKHPLGGOHRIWKH 5th millennium (BC cal.), also using lever pressure but a copper point. The SODWIRUPFRQFHSWLRQLV´FODVVLFµLHRUWKRJRQDOWRWKHÁDNLQJVXUIDFHZLWK the pressure point placed “inside” the platform. Experiments prove the dif ÀFXOWFRQWURORIWKLVPHWKRGZKLFKGHPDQGVDYHU\DFFXUDWHVHWWLQJRIWKH point and pressure incidence in order to avoid a short detachment or an excessive bulb and blade width. In Bulgaria, standing pressure (and indirect percussion) was practiced before lever pressure, but I do not know whether it was with or without copper points. 3. The Early Bronze Age north Syrian production is a late stage in this frame work, and large blades were certainly produced earlier in the Near East, pos sibly without copper and with different platform preparations or techniques (a long mesial fragment of an obsidian blade with perfect regularity, a trap ezoidal section and a width of 3.1 cm, probably detached by lever pressure was found at Cayönü in the Late PPNB, dated to about 7200 BC; examined at the Istanbul University). Experiments have proven that preparing the plat form, thus giving a dihedral or convex facetted butt to the blades, eases the fracture initiation and the control of the blade width. 7KH EODGHV SURGXFHG LQ VRXWKHDVWHUQ )UDQFH DQG LQ QRUWKHUQ 6DUGLQLD engaged a similar technique and method (lever pressure, copper and plat form treatment). These two production centres, at least partly synchronic according to their current estimated dates, might have generic relations. In both workshops, the placement of the pressure tool on a small portion of the acute platform edge is also a clever way to improve the control of the detach ment. Both workshops also used indirect percussion to produce blades with a thick orthogonal butt. 5. At least two different traditions coexisted for some time in southern Iberia, so that their respective productions ended sometimes in the same Chalcolithic graves. An extensive investigation about the distribution of the central Portugal punch blades (which might be rooted in an older regional WUDGLWLRQ ZRXOGEHUHOHYDQWDVZHOODVVWXGLHVDERXWWKHLGHQWLÀFDWLRQDQG distribution of the at least three different types of raw material used for lever 67 Jacques Pelegrin pressure blades prepared with the same dihedral butt, which probably belong to the same “south Iberic” technical tradition or group. $SDUWIURPWKHVH0HGLWHUUDQHDQZRUNVKRSVWKH*UDQG3UHVVLJQ\WUDGL tion in western central France remains original and rather continental, with a distribution much more turned to the west (Brittany cf. Ihuel 2004), east 0DOOHW DQGQRUWK 'HOFRXUW9ODHPLQFN WKDQWRWKHVRXWKZKHUH it rarely overlaps with the lever pressure blades produced in the southeast of France. By the way, the remarkable inventiveness of knapping craftsmen, to whom this research is dedicated, opens a little window both in the history of tech niques – that reaches one of its highlights with the production of large blades ²DQGLQWKHVRFLRHFRQRP\RI1HROLWKLFDQG3URWRKLVWRULFVRFLHWLHVDÀHOG of research where so much remains to be done. 1RWHÀJXUHFRPSXWHUL]LQJDQGOD\RXWE\*0RQWKHO805&156)UDQFH 68 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Hugo Nami Experiments to explore the 3DOHRLQGLDQÁDNHFRUHWHFKQRORJ\ in southern Patagonia Abstract :LWKH[FHSWLRQRIWKHIDPRXV´ÀVKWDLOµRU)HOOSURMHFWLOHSRLQWV3DOHRLQGLDQ lithic assemblages from southern Patagonia on the southern tip of South America have been little known. However, during the last few years, archae ological research has provided many new insights into this topic. It has been DFNQRZOHGJHGWKDWXQLIDFLDOVWRQHWRROVZHUHPDGHRQGLYHUVHÁDNHEODQNV many of them with very delicate morphologies. This fact suggests some kind of preparation of the cores not yet found in the archaeological record. Based on detailed experiments and documentation, this paper shows ex DPSOHVRIPRUSKRORJ\DQGYDULDELOLW\WKDWH[LVWGXULQJWKHÁDNLQJRIVLPLODU 3DOHRLQGLDQ ÁDNHV 7KXV H[SHULPHQWV DUH XVHIXO IRU H[SORULQJ QHZO\ GLV FRYHUHGWHFKQRORJLHVDQGLQWKLVFDVHIRUVXJJHVWLQJWKDWDSDUWIURPQRQ SDWWHUQHGÁDNLQJVRPHNLQGRIFRUHSUHSDUDWLRQZDVXVHGWRREWDLQÁDNH blanks by the Paleoindians in Patagonia. Introduction Patagonia in the southern part of South America is a region of about 1000 000 km² shared by the Republics of Argentina and Chile. At its southern tip, near the Magellan Strait on the Chilean side of the border, Junius Bird excavated the Fell and the Pali Aike cave in the volcanic region of Pali Aike LQ WKH V %RWK VLWHV EHFDPH ZRUOGIDPRXV EHFDXVH RI WKH DVVRFLDWLRQ RI3OHLVWRFHQHIDXQDZLWKWKHFRPPRQO\FDOOHG´ÀVKWDLOµRU)HOOSURMHFWLOH points and other stone and bone tools (Bird & Bird 1937, Bird 1988). How ever, until the new discoveries and studies were made during the 1980s, WKH/DWH3OHLVWRFHQHKXQWHUJDWKHUHUVZHUHDOLWWOHNQRZQDQGFRQWURYHUVLDO research topic (e.g. Borrero 1986, Borrero et al. 1988, etc.). Fortunately, the DUFKDHRORJLFDOÀQGVPDGHLQQHZO\GLVFRYHUHGDUFKDHRORJLFDOGHSRVLWVDQG the subsequent detailed studies made of them today allow us to understand the diverse technological issues related to the earliest human populations liv ing in southern South America during the time span lasting between 13,200 and 11,900 BP (Rubinos Perez 2003:24). From a technological perspective, based on detailed experiments and documentation, this paper describes examples of morphological variability 69 Hugo Nami H[LVWLQJDVWKHUHVXOWRIÁDNHGHWDFKPHQWRI3DOHRLQGLDQÁDNHEODQNV7KXV WKH H[SHULPHQWV VXJJHVW WKDW EHVLGHV QRQSDWWHUQHG ÁDNLQJ VRPH NLQG RI core preparation was used by the Paleoindians in Patagonia. General archaeological considerations Apart from the ground discoid stones, for a long time the only know Paleoin GLDQÁDNHGDUWLIDFWZDVWKH)HOOSURMHFWLOHSRLQWZKLFKZDVZLGHVSUHDGLQ 6RXWK$PHULFD %LUG0D\HU2DNHV3ROLWLV +RZHYHU during the past few decades, a number of archaeological excavations have been carried out in Patagonia and nearby regions that have expanded the knowledge of early human occupation in the southern zone (Massone 1987, 1DPL0LRWWL1XxH]HWDO3DXQHUR0D]]D nti 1997; Flegenheimer et al. in press). Moreover, recent technological studies on the Fell and Pali Aike´s collections curated at the American Museum of Natural History (New York) have produced new insights in the Fell lithic assemblages (Nami 1998). For example, it was observed that there are lateral VFUDSHUVDQGNQLYHVPDGHRQÁDNHEODQNVWKDWREYLRXVO\ZHUHREWDLQHGIURP QRQSDWWHUQHGÁDNLQJ+RZHYHUVRPHVKRZHGDYHU\GHOLFDWHPRUSKRORJ\ striking platforms carefully isolated and abraded, suggesting some kind of FRUHSUHSDUDWLRQ7KHUHDUHDOVRODUJHUÁDNHEODQNVUHVHPEOLQJWKRVHUHVXOW ing from bifacial thinning (pers. obs. 1999; Cattaneo 2002; Paunero 2000). 7KHVHVXJJHVWWKDWWKHÁDNHVZHUHREWDLQHGIURPELIDFLDODQGRUGLFRLGFRUHV ZLWKDFRQYH[VXUIDFHSUHSDUDWLRQIRUÁDNHGHWDFKPHQW+RZHYHUWKHFRUHV have not yet been found in the archaeological record. Experiments General remarks Experimentation in science has different objectives; on one hand, it is ben HÀFLDOWRWHVWDK\SRWKHVLVRQWKHRWKHULWLVXVHIXOWRPDNHGLVFRYHULHVDQG in these cases it plays a heuristic role as a guide in hypothesis formulation (Hempel 1989). Experimental archaeology and lithic technology are both part of such investigations. These inquiries are framed in the contemporary ÁRZRIDUFKDHRORJLFDOWKLQNLQJWKDWLVDLPHGDWWKHSUHGLFWLRQUHWURGLFWLRQ and explanation of a number of aspects related to archaeological remains (Nami 2003a; n.d.a, n.d.b). 70 Experiments to explore the Paleoindian ﬂake-core technology in southern Patagonia During the last decade, I have been carrying out experimental and archaeo logical research on different core preparation technologies from North and South America (i.e. Nami 1992; 1995; etc.). To understand the diverse aspects of those technologies, I have made over a hundred experimental specimens LH 1DPL )LJ )LJ $ D )LJ $& E D 2003b, Nami et al. 1996; etc.). In the previous section, I pointed out that in the lithic assemblages from VRXWKHUQ3DWDJRQLDWKHUHDUHGHOLFDWHÁDNHEODQNVWKDWLQP\RSLQLRQZHUH detached from prepared cores. Thus, to explore this technological hypothe sis I made several experimental, partially and totally prepared cores to obtain VLPLODUÁDNHEODQNVVRPHRIWKHPDUHVKRZQLQÀJXUHVDQG+RZHYHU to illustrate my ideas and observations in detail, I recorded the entire reduc tion process, from the initial nodule to the exhausted core and including the GHWDFKHG ÁDNHV , DOVR FDUHIXOO\ GRFXPHQWHG WKH W\SRORJLFDO DQG PHWULFDO DWWULEXWHVRIWKHUHVXOWLQJFRUHDVZHOODVWKHUHVXOWLQJÁDNHVDIWHUHDFKVHULHV RIÁDNHGHWDFKPHQW Materials, techniques and results The raw materials employed were different kinds of rocks with diverse lithic grade, varying from 1 to 5 on the Callahan scale (Callahan 1979:16). Bifa cially prepared cores were made out of obsidian from Mono Lake (Califor nia, U.S.A.), Glass Buttes from Idaho (U.S.A.), industrial glass from Buenos Aires (Argentina) and chert from the Río Negro area (Republic of Uruguay). In the experimental piece described here I used an obsidian nodule from *ODVV %XWWHV &RQFHUQLQJ ÁDNLQJ LPSOHPHQWV , HPSOR\HG DQ DQWOHU ELOOHW of 1 kg from Alaska (U.S.A.), a small soft calcareous hammerstone of 125 g from the loessid deposit existing in the subsoil of Del Viso in the Buenos Aires province and a larger one of sandstone (790 g) from the Vinchina River, La Rioja province, Republic of Argentina (Fig. 1). Finally, I employed a neoprene piece as a pad and a soft granitic stone as an abrasive. 'XULQJWKHÁDNLQJDFWLYLW\,VDWRQDFKDLUZUDSSLQJWKHFRUHZLWKWKH QHRSUHQH SDG )LJ E , GLG WKLV WR SUHYHQW GDPDJH WR WKH ÁDNH LI LW IHOO GRZQRQWKHÁRRUDQGWRDYRLGLQMXU\IURPWKHVKDUSURFN)RUWKHÁDNH GHWDFKPHQW,XVHGGLUHFWSHUFXVVLRQÁDNLQJZLWKVRIWKDPPHUVWRQHDQGDQW ler billet with the core pressed against the outside of the thigh (cf. Waldorf )LJ :KLWWDNHU )LJ , FKDQJHG SRVLWLRQV DF FRUGLQJWRWKHVL]HRIWKHFRUHDQGWKHEHVWSRVLWLRQIRUWKHÁDNHGHWDFKPHQW (Fig. 3c, 4b). During the core and platform preparation, I employed anvil SHUFXVVLRQÁDNLQJZLWKWKHVPDOOVRIWKDPPHUVWRQHZKLOHUHVWLQJWKHFRUH 71 Hugo Nami on the leg (Fig. 3a, 4a) or holding it in the left hand and left wrist steadied on the padded upper leg (cf. Whittaker 1994:183, Fig. 8.5). Figure 1. Flaking implements used in this experiment. a) Antler billet, b-c) soft stone hammerstones, d) hard hammerstone (photo by the author). Figure 2. Holding positions PRVWO\XVHGGXULQJWKHÁDNH detachments (photo by María de las Mercedes Cuadrado Woroszylo). Figure 3. Flake detachment from a partially prepare core. a) Surface and platform SUHSDUDWLRQZLWKVRIWKDPPHUVWRQHE ODWHUDOYLHZRIERWKSUHSDUDWLRQVF ÁDNHGHtachment with the antler billet, d-e) anverse and reverse of the resulting core with the ÁDNHVGHWDFKHGRQERWKVXUIDFHV 72 Experiments to explore the Paleoindian ﬂake-core technology in southern Patagonia Figure 4. Bifacial preparation of the core DQGÁDNHGHWDFKPHQWD &RUHSUHSDUDtion with the small soft hammerstone, b-c) GHWDFKPHQWDQGWKHUHVXOWLQJÁDNHE\DQYLO GLUHFWSHUFXVVLRQÁDNLQJ SKRWRE\0DUtD de las Mercedes Cuadrado Woroszylo). ,UHPRYHGWKHÁDNHVIROORZLQJGLIIHUHQWSDWWHUQV7KH\ZHUHUHJXODUO\GH tached around the margin on one face or irregularly, by alternate detach ment on both faces (see Figs. 4 to 9). However, in the experiment described LQ GHWDLO LQ WKLV SDSHU DOO WKH VHULHV RI ÁDNH GHWDFKPHQW 6)' KHUHDIWHU UHIHUULQJWRDVHULHVRIÁDNHVUHPRYHGRQRQHIDFHDURXQGWKHPDUJLQRI WKHFRUHZHUHPDGHIROORZLQJWKHVDPHSDWWHUQUHVSHFWLYHO\SHUIRUPHG on one surface and then the opposite. Sometimes, due to the natural form RIWKHQRGXOHIRUWKHÀUVWDQGVHFRQG6)'WKHLQLWLDOSUHSDUDWLRQZDVQRW necessary. Furthermore, by isolating selected points of impact and following an adequate and carefully prepared platform with an angle of 80°, I abraded the material with the granite soft stone and afterwards, in order to detach WKHEODGHV,PRVWO\XVHGDQYLOSHUFXVVLRQÁDNLQJLQWKHZD\GHVFULEHGDERYH When the platform angle was a little higher (~ 85°) I had to use the heavy VRIWKDPPHUVWRQHWRGHWDFKÁDNHQXPEHU%\XVLQJWKLVNLQGRISUHSDUHG FRUHZLWKDQDGHTXDWHSODWIRUPSUHSDUDWLRQDQGDFRQWUROOHGEORZSUHFLVH EXWQRWVWURQJ,FDQGHWDFKWKHSUHGHWHUPLQHGÁDNHV7KLVREVHUYDWLRQLV also valid for high graded lithic raw materials, such as quartzite and rhyolites 1DPLD)LJ$& 73 Hugo Nami )LJXUH%LIDFLDOFRUHDQGÁDNHVUHVXOWLQJ IURPWKHÁDNLQJDFWLYLW\VKRZHGLQSUHYLRXV ÀJXUH SKRWRE\WKHDXWKRU )LJXUHD 7DEXODUQRGXOHE &RUHDQGÁDkes resulting in the First SFD. The numbers show the order of the removals in clockwise sense (drawings by the author). $VDUHVXOWRIWKHÀUVWDQGVHFRQG6)'WKHÁDNHVZHUHPRVWO\SULPDU\DQG VHFRQGDU\DQGGLGQRWVKRZHYLGHQFHRISUHYLRXVÁDNLQJ )LJDQG 7KH\ were detached both after weak and more thorough isolation of platforms. $IWHUZDUGVNHHSLQJLQPLQGVRPHRIWKHDUFKDHRORJLFDOÁDNHVDQGP\K\ pothesis about their origin, I began a true core preparation with the small soft hammerstone. In this way, by carefully eliminating the bulbar scars re VXOWLQJIURPWKHSUHYLRXVÁDNHUHPRYDOV,PDGHDXQLIRUPFRQYH[VXUIDFH and also an adequate angle to the striking platform for detachment of the QH[WVHULHVRIÁDNHV7KLVUHVXOWHGLQDVRUWRIGLVFRLGELIDFHFRUHWKDWDOORZHG further SFD. I made this type of preparation twice, after the second and IRXUWK6)' )LJDDQGD $IWHUWKLVÀUVWFRUHSUHSDUDWLRQWKHGHWDFKHG SURGXFWVUHVHPEOHGELIDFLDOWKLQQLQJÁDNHV )LJEE 2QWKHRWKHU KDQGWKHDQYLOSHUFXVVLRQÁDNLQJIRUSODWIRUPSUHSDUDWLRQSURGXFHGVPDOO DQGVKRUWÁDNHV 74 Experiments to explore the Paleoindian ﬂake-core technology in southern Patagonia )LJXUH&RUHDQGÁDNHV resulting in the second SFD made in the reverse face of the core illusWUDWHGLQÀJXUH7KH numbers show the order of the removals counterclockwise. Figure 8. a) Bifacial prepared core arranged to continue with the next SFD, b) Bifacial FRUHDQGÁDNHVUHVXOWLQJIURPWKHVHFRQGDQG third SFD. 75 Figure 9. a) Bifacial prepared core ready to continue with the next 6)'E ([KDXVWHGFRUHDQGÁDNHV UHVXOWLQJIURPÀIWKDQGVL[WK6)' Note that this biface core might continue its development as an early stage of manufacture for a bifacial stone tool. Hugo Nami During the entire core reduction process I made six SFD and obtained twelve ÁDNHVZLWKPXFKYDULDELOLW\UHVHPEOLQJWKRVHIRXQGLQWKHHDUOLHVWDUFKDHR ORJLFDO OLWKLF DVVHPEODJHV IURP VRXWKHUQ 6RXWK $PHULFD 0RVW ÁDNHV ZHUH obtained with the antler billet. However, when the platform had a little higher angle (~85°) than the optimum value, I used the heavier soft hammerstone weighing 790 g. In my opinion, the discoid morphology of the core is optimal, because it allows platform preparation around the entire perimeter and also SURYLGHVDXQLIRUPVXUIDFHIRUÁDNHGHWDFKPHQWV7KHFRUH·VPRUSKRORJLFDO YDULDWLRQVDQGGLPHQVLRQVUHVXOWLQJIURP6)'DQGDUHLOOXVWUDWHG LQÀJXUHVWRUHVSHFWLYHO\DQGGHSLFWHGLQWDEOHVDQG After the sixth SFD, the problem is that the core is very thin. Hence, the ÁDNHV EHFRPH WKLQQHU DQG WHQG WR UHVHPEOH WKRVH IURP ELIDFLDO WKLQQLQJ producing many fractures (see Fig. 9b2); however, it is still possible to detach XVDEOHÁDNHEODQNV,QWKLVFDVHWKHH[KDXVWHGFRUHPLJKWEHGLVFDUGHGUH cycled for another tool, used as a blank or as an early stage of manufacture for a bifacial tool (cf. Nami 1997, Fig. 18A). SFD Flake’s form Length Width Thickness 1 Primary 80 115 20 6b1 1 Primary 58 43 6 6b2 1 Secondary 96 70 13 6b3 2 Primary 80 112 17 7.1 2 Angular 79 124 10 7.2 2 Angular 112 58 7 7.3 3 Angular 70 55 13 8 3 Angular 62 63 9 8 4 Angular 58 67 8 8 4 Angular 54 57 9 8 6 Angular 46 54 8 9b1 5 Angular 80 40 8 9b3 Length Width Morphological transformation of the nodule/core Thickness Table 1. Description of the useful ÁDNHVREWDLQHGLQ the experimental specimen discussed in this paper according to the SFD. Measurements are given in mm. Figure Observation Figure Natural nodule 147 152 55 -- 6a Unifacial discoidal core 145 142 54 First SFD 6b Bifacial discoidal core 145 140 32 Second SFD 7a Prepared bifacial core 112 105 32 -- 8a Bifacial core 112 105 25 Third and fourth SFD 8b Prepared bifacial core 105 92 18 -- 9a Exhausted bifacial core 102 91 15 Fifth and sixth SFD 9b Table 2. Description of morphological change in the experimental cores resulting after each preparation and SFD. Measurements are given in mm. 76 Experiments to explore the Paleoindian ﬂake-core technology in southern Patagonia Final and archaeological considerations The experiments performed with these cores, particularly the piece described in this paper, was useful to understand diverse aspects of bifacial prepared ÁDNHFRUHWHFKQRORJ\EXWDOVRWRGLVFXVVVHYHUDOWRSLFVUHODWHGWR3DOHRLQ dian technology from both regional and continental perspectives. They are as follows: From a regional perspective Experimental $FFRUGLQJ WR WKLV H[SHULPHQW WKH 3DOHRLQGLDQ ÁDNHV PLJKW EH REWDLQHG from diverse kinds of cores, some of them by using some sort of partial RUWRWDOSUHSDUDWLRQ,WDOVRVKRZHGWKHPRUSKRORJLFDOPRGLÀFDWLRQRIWKH core after each SFD. In this sense, its reduction emerges as a dynamic proc HVVZLWKPXFKPRUSKRORJLFDOYDULDELOLW\LQWKHFRUHVDQGWKHÁDNHVDVZHOO UDWKHUWKDQDVLQÁH[LEOHSUHVFULEHGDFWLRQVWKDWLQHYLWDEO\OHDGWRWKHVDPH result (cf. Kuhn 1995). 7R FUHDWH WKHVH W\SHV RI FRUHV VRPH NQRZOHGJH RI ELIDFLDO ÁDNLQJ DQG core preparation strategies is required, especially how to prepare the surface and striking platforms. By knowing the method and from a simple tabular QRGXOHZLWKWKHDSSURSULDWHÁDNLQJWRROVWHFKQLTXHVDQGKROGLQJSRVLWLRQV it is easy to maximize the raw material by extracting a good number of useful ÁDNHEODQNVSRWHQWLDOO\XVHGLQWKHPDQXIDFWXUHRIXQLIDFLDODQGRUELIDFLDO stone tools. The goal is to obtain a tool of predetermined shape with very little retouch. Regarding core preparations, there is a broad range of possibilities. How ever, in the Patagonian case, one of the probabilities might be a core similar to the piece presented here and other variations in outline, which might be ovoid, circular, semicircular, quadrangular and other varieties (cf. Anderson /HURL*RXUKDQ)LJ/RFKW 6ZLQQHQ &RQFHUQLQJWHFKQLTXHV,VXJJHVWWKDW3DOHRLQGLDQKXQWHUJDWKHUHUVXVHG VRPH YDULHW\ RI VRIW RU VHPLVRIW GLUHFW DQYLO SHUFXVVLRQ ÁDNLQJ E\ XVLQJ KDPPHUVWRQHVDQGRURVVHRXVWLVVXHV LHERQHDQWOHUHWF ZLWKKLJKGHQ VLW\YDOXHVUDQJLQJEHWZHHQ 1DPL (ONLQ Archaeological ,QWKHDUFKDHRORJLFDOUHPDUNV,SRLQWHGRXWWKDWVRPHDUFKDHRORJLFDOÁDNH EODQNV UHVHPEOH ELIDFLDO WKLQQLQJ ÁDNHV FI :KLWWDNHU SS )RU 77 Hugo Nami this reason, some archaeologists suggest that Paleoindians used “bifaces as cores” (Cattaneo 2002). In the Americas, a biface is a knapped stone with WZRÁDNHGIDFHVDQGDVLQJOHFRQWLQXRXVHGJH$OWKRXJKWKLVWHUPPD\UHIHU to any artifact, it is normally used in reference to those that are interpret HG DV XQÀQLVKHG WRROV LQ HDUO\ VWDJHV RI PDQXIDFWXUH )ULVRQ DQG %UDGOH\ 'HVSLWHWKDWLWVGHELWDJHPD\EHXVHGDVÁDNHEODQNVWKLVGRHV QRWPHDQWKDWWKH\DUH´ELIDFHDVFRUHVµ7KHÁDNHVGHWDFKHGIURPELIDFLDO cores, differs in sizes from the bifacial thinning waste observed in several 3DWDJRQLDQVLWHV 1DPLD DQGRWKHUSODFHVLQWKHZRUOG LH Bradley 1982: 203 pp; etc.). As demonstrated in this and previous experi PHQWVÁDNHVREWDLQHGIURPSUHSDUHGELIDFLDOFRUHVDQGGHWDFKHGZLWKVRIW RUVHPLVRIWÁDNLQJLPSOHPHQWVVKRZVRPHDWWULEXWHVH[LVWLQJLQWKHELID cial thinning, but with larger dimensions than those small Patagonian size bifaces (Nami 1986; 1991; 2003a). Furthermore, based on previous studies of thousands of experimental and archaeological bifaces and their debitage along the Americas (e.g. Nami 1988; 1993/94b, 1999c; 2003a, etc.), in my RSLQLRQWKLVW\SHRIÁDNHVPLJKWEHSDUWRIWKHÁDNHVYDULDELOLW\GHWDFKHG from some kind of partially or totally prepared core, particularly from unifa FLDORUELIDFLDOGLVFRLGFRUHVDQGE\XVLQJVRIWSHUFXVVLRQÁDNLQJ'HVSLWHWKH IDFWWKDWVRPHZDVWHÁDNHVIURPELIDFLDOWKLQQLQJPLJKWEHXVHGDVEODQNV for making tools, bifacial thinning produces a particular kind of debitage DFFRUGLQJWRWKHLUGLIIHUHQWVWDJHV LH&DOODKDQ)LJV 54, etc.; Nami 2003b). Consequently, it is important to differentiate between ÁDNHVREWDLQHGIURPWUXHFRUHVDQGWKRVHWKDWDUHZDVWHIURPELIDFLDOWKLQ ning, since, from a technological perspective, these are different strategies. Flakes with similar dimensions of those detached from bifaces and/or dis FRLGDOÁDNHVWDNHQIURPODUJHUHDUO\VWDJHELIDFHVZHUHXVHGDVEODQNVIRU manufacturing many projectile points, particularly in Patagonia (cf. Callahan 1DPLDE As a method, core preparation technology involves a particular and dif ferent concept from other techniques (Pelegrin in Callahan 1981: 65 p). The PDLQJRDOLQWKHÁDNHSURGXFWLRQLVWRREWDLQWKHODUJHVWXVDEOHÁDNHSRVVLEOH from a bifacial core. The aim of bifacial thinning is to thin the bifaces. De spite what might be some morphological similarities in the bifacial thinning ÁDNHVDQGWKRVHREWDLQHGIURPELIDFLDOFRUHVWKH\DUHWKHUHVXOWRIGLIIHU ent recipes for action, technical decisions and process (cf. Schiffer & Skibo 1987; Nami 1994; Apel 2001a). If so, from a technological perspective “the biface as cores” concept in this case might be fallacious. Incidentally, the WUDQVSRUWDWLRQRIELIDFHVWRGHWDFKXVDEOHÁDNHVPLJKWEHDULVN\VLWXDWLRQ ,QIDFWGLYHUVHNLQGRIIDLOXUHVPLJKWRFFXUGXULQJWKHÁDNHGHWDFKPHQWLQ WKHÁDNHVDQGRUWKHELIDFHV FI&DOODKDQ 78 Experiments to explore the Paleoindian ﬂake-core technology in southern Patagonia From a continental perspective )ODNHEODQNV IRXQG LQ WKH 3DOHRLQGLDQ DVVHPEODJHV IURP VRXWKHUQ 6RXWK $PHULFDPLJKWEHFRPSDUHGZLWKWKHKLJKHVWVWDQGDUGVRIÁDNHGHWDFKPHQW in the Old World Upper Paleolithic and North American Paleoindian lithic assemblages. Thus, in technological evolution, the technology existing in VRXWKHUQ6RXWK$PHULFDLVFRPSDUDEOHWRWKHKLJKO\GHYHORSHGÁDNHGWHFK nologies existing during the terminal Pleistocene in the world, with very well developed bifacial reduction and prepared core strategies. Particularly in the New World, several authors reported the presence of /HYDOORLVOLNHGLVFRLGELIDFLDOFRUHVDQGÁDNHVGHWDFKHGIURPWKHPLQWKH early lithic assemblages along the Americas, from Alaska (Anderson 1970:18 SS )LJV 3ODWH WR VRXWKHUQ 6RXWK $PHULFD &DUGLFK HW DO 82:197, Fig. 8). ,Q1RUWK$PHULFD3DOHRLQGLDQKXQWHUJDWKHUHUVDOVRXVHGVLPLODUGHOLFDWH ÁDNHEODQNVIRUPDNLQJXQLIDFLDOÁDNHGVWRQHWRROV7KH\VXJJHVWWKDWWKH\ ZHUHGHWDFKHGIURPSUHSDUHG/HYDOORLVOLNHDQGELIDFLDOFRUHV,QIDFW/HYDO ORLVOLNHÁDNHVDQGFRUHVZHUHUHSRUWHGLQVHYHUDO&ORYLVDVVHPEODJHV 1DPL HWDO&ROOLQV 'HVSLWHWKHIDFWWKDWQRWDOO&ORYLVODUJHVL]HGEL faces are cores, some of them might be (i.e. Parson & Pearson 2001:51, Fig. D F )ROORZLQJ&ORYLVKXQWHUJDWKHUHUV/HYDOORLVOLNHÁDNHVDQGWKRVH SUREDEO\GHWDFKHGIURPELIDFLDOFRUHVZHUHLGHQWLÀHGLQ3ODLQYLHZOLWKLFDV VHPEODJHV .QXGVRQSS)LJDQG DOVRLQWKH)ROVRP collections from the Hansom site (Frison & Bradley 1999 Fig. 6), Linden meier (Wilmsen & Roberts 1978; Nami 1999c), Lake Ilo (Root 1993: Fig. 52a & c) and Rio Grande valley area (Judge 1973:89). Although not all Folsom bifaces are cores (LeTourneau 2001), Stanford and Broilo (1981) reported a large biface from Texas that in fact might be a core. During the latest Pleistocene across South America, prepared cores DQGÁDNHVGHWDFKHGIURPWKHPZHUHLGHQWLÀHGLQWKHPLGGOHYDOOH\RIWKH Magdalena river in Colombia (López Castaño 1999), at the Cubilán site in Ecuador (Temme 1982) and Cerro Los Burros archaeological locality in the Republic of Uruguay (Nami 2001). 6XPPDU\DQGÀQDOVWDWHPHQWV In summary, as a result of this experiment, my technological hypothesis RQWKHSRVVLEOHRULJLQRIWKH3DOHRLQGLDQÁDNHEODQNVZDVWHVWHG7KXVLW is possible to suggest a derived archaeological hypothesis: the latest Pleis WRFHQHKXQWHUJDWKHUHUVIURP3DWDJRQLDPLJKWKDYHXVHGVRPHVRUWRIFRUH preparation. Based on both archaeological and experimental observations I 79 Hugo Nami VXJJHVWWKDWDSDUWIURPQRQSDWWHUQHGÁDNLQJVRPHNLQGRISDUWLDORUWR WDOFRUHSUHSDUDWLRQZDVXVHGIRUÁDNHEODQNVGHWDFKPHQWDFFRUGLQJWRWKH nodule morphology, material of the cortex and other circumstances. The SUHSDUDWLRQPLJKWYDU\IURP´WXUWOHEDFNµXQLIDFLDODQGRUELIDFLDOGLVFRLG cores and/or other variations of Levallois technology (cf. Böeda 1993; Dib EOH %DU<RVHI ,WVHHPVOLNHO\WKDWVRPHVRUWRIWRWDOO\RUSDUWLDOO\ SUHSDUHGELIDFLDOFRUHZLWKV\PPHWULFDODQGRUDV\PPHWULFDOFURVVVHFWLRQV PLJKWDOVREHFRQVLGHUHGDSDUWRIWKH/HYDOORLVFRUHWHFKQRORJ\ 7UXIIURZ 1995:413; Yalçinkaya 1995:319; Wengler 1995:315). I am aware that the South American Paleoindians might have used dif ferent kinds of prepared cores. However, one of the alternatives to consider is the bifacial one, such as it has been discussed here, which might be part of the variations occurring during the core preparation strategies. There fore, neither the bifacial prepared cores with asymmetrical or symmetrical FURVVVHFWLRQVQRUWKHSUHVHQFHRIÁDNHVUHVHPEOLQJELIDFLDOWKLQQLQJÁDNHV SURYHWKDWLQWKHLUWHFKQRORJLFDORUJDQL]DWLRQ3DOHRLQGLDQKXQWHUJDWKHUHUV used “bifaces as cores”. Like some Paleoindian groups of North America, LQ WKH HDUO\ OLWKLF DVVHPEODJHV IURP WKH IDU 6RXWK QRW DOO ELIDFLDO ÁDNLQJ might have resulted in projectile point production (cf. Collins 1999). Be yond adaptive and social differences, this fact suggests that certain technical similarities that existed among Paleoindians from North and South America hint of shared technical knowledge (Nami 1997; n.d.a). Additionally, the use of prepared cores strategies also implies that during the early times in the New World archaeology, by diverse historical and social mechanism, human populations would have shared some kind of technological information. Acknowledgements My special thanks to: Jan Apel and Kjel Knutsson for the invitation to par ticipate in the symposium; the Societas Archaeologica Upsaliensis (SAU) that supported my participation in the event; Maria Louise Sidoroff for her invaluable collaboration during the edition of this paper; Britta Wallsten for her help; E. Callahan for his discussion about this subject; María de las Mer FHGHV&XDGUDGR:RURV]\ORIRUKHUFRQWLQXRXVKHOSDQGVXSSRUWÀQDOO\WR Jan Apel for his invaluable help and kindness during my stay in Uppsala. 80 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Greg R. Nunn Using the Jutland Type IC Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge SUHVVXUHÁDNLQJ Abstract 7KH-XWODQG7\SH&1HROLWKLF'DQLVK'DJJHULVRQHRIWKHÀUVWRIDORQJ series of Late Neolithic Scandinavian Flint Dagger manufacturing tradi tions dating from approximately 2350 BC to 1750 BC. The Danish Dag ger sequence included six main types and numerous subtypes. The Jutland 7\SH&LVUHFRJQL]HGIRULWVHOHJDQWZHOOFRQWUROOHGSDUDOOHOHGJHWRHGJH ÀQDOSUHVVXUHÁDNLQJVHULHV1HROLWKLF'DQLVKÁLQWNQDSSHUVZHUHWKHRQO\ FUDIWVPHQWRFUHDWHDQGSHUIHFWWKHW\SHRIÁDNLQJUHTXLUHGIRUWKH7\SH& 'DJJHU7KLVVW\OHRISUHVVXUHÁDNLQJLVFRQVLGHUHGRQHRIWKHPRVWGLIÀFXOW DQGOHDVWXQGHUVWRRGWHFKQLTXHVE\PRGHUQÁLQWNQDSSHUV8VLQJWKH7\SH 1C Dagger as a model, the author will describe the manufacturing stages and WHFKQLTXHVUHTXLUHGWRUHSOLFDWHWKLVXQLTXHVW\OHRISUHVVXUHÁDNLQJ History Type 1C Danish Daggers (Fig. 1) were made in the early part of the late 1HROLWKLF/1FDO%& 9DQGNLOGHDVSHU$SHOa:10). According to Lomborg (1973; as per Apel 2001a:10) ”…Late Neolithic was VHWLQWKHSHULRGZKHQÁLQWGDJJHUVUHSODFHGEDWWOHD[HVDVWKHPDOHSUHVWLJH JLIWVLQWKHXSSHUJUDYHVRIWKH6LQJOH*UDYH&XOWXUHLQ-XWODQGµ)XUWKHU ´'DJJHUVRIW\SH$&DUHDVVLJQHGWRDGHOLPLWHGSHULRGLQ1RUWKHUQ-XW ODQGZKHQWKLVSDUWRI6FDQGLQDYLDZDVLQÁXHQFHGE\WKH:HVWHUQ(XURSHDQ %HOO%HDNHUDQG%HDNHUWUDGLWLRQVµ´7KHDUHDDURXQGWKH/LPIMRUGZDV«WKH centre of the dagger production during the early phase” (Rasmussen 1990; as per Apel 2001a:268). 81 Greg R. Nunn Figure 1.Original Jutland style Type 1C dagger. (National Museum of Denmark, Copenhagen, Museum number A 10198, illustrated by E. Callahan). Archaeological evidence Archaeological evidence of beginning and ending points are essential when replicating an artifact. With the Type 1C Danish Dagger, evidence of earlier stage preforms is limited, while completed daggers are abundant. However, DUFKDHRORJLVWVDQGÁLQWNQDSSHUVDUHIRUWXQDWHWRKDYHHQRXJKYHULÀHGDU FKDHRORJLFDOHYLGHQFHWRDVVHPEOHWKHSULPDU\VHFRQGDU\ÀQDOSUHIRUPDQG JURXQG SUHIRUP W\SRORJ\ )LJV DQG ,Q DGGLWLRQ $UQROG·V VNLOOIXOUHÀWWLQJRIGHELWDJHIURPDODQFHWGDJJHUUHFRYHUHGIURPDODWH1HR lithic settlement provides valuable information. With this archaeological evi dence one can understand the morphology and accurately replicate the Type 1C Danish Dagger. Figure 2. (a) Original stage 2 – initial edging, length 32 cm, width 13.5 cm (National Museum of Antiquities, Stockholm, Museum number 2549 SKJEGRIE 37A; 2(b) Original stage WKUHHSULPDU\SUHIRUP²VWUD\ÀQG length 31.5 cm, width 9cm (Danish National Museum, Copenhagen); 2(c) Original stage four secondary preform, length 32 cm, width 9 cm (Danish National Museum, Copenhagen); photographed by E. Callahan. 82 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking )LJXUH2ULJLQDOVWDJHÀYHÀQDOSUHIRUP with traces of early grinding, from unprovenienced grave site, length 38.7 cm, width 2.31 cm, thickness 1.7 cm (Moesgaard Museum, Arhauss, Denmark); illustrated by E. Callahan. Figure 4. Original stage six - ground preform; length 37.1 cm, width 7.53 cm, thickness 1.1cm; (Moesgaard Museum, Arhauss, Denmark, Museum number A 27691); illustrated by E. Callahan. Morphology There are six main dagger types with numerous subtypes (Lomborg 1973; as per Apel 2001a 7KHUHDUHÀYHVXEW\SHVIRUWKH7\SHVHULHV7\SH LV GHÀQHG E\ ´>O@DQFHW VKDSHG GDJJHUV ZLWKRXW KDQGOHV RU ZLWK PDUJLQ ally marked handles that display a lenticular cross section.” Further, “The type is distinguished from other oblong and thin, bifacial objects, such as ODQFHDQGVSHDUKHDGVE\WKHIDFWWKDWWKHHGJHVRIWKHKDQGOHWDSHUWRZDUGV the base and that the base tends to be convex or sometimes even pointed” (Lomborg 1973; as per Apel 2001a:235). 'HÀQLWLRQ 'HÀQLWLRQRI7\SH&'DQLVK'DJJHULVSDUDOOHOSUHVVXUHÁDNHG´«W\SH daggers with symmetrically curved and concaved shaped handle edges.” “In Denmark this subtype is almost exclusively concentrated to the Limfjord area in Northern Jutland” (Lomborg 1973; as per Apel 2001a:236). All other VXEW\SHVDUHSHUFXVVLRQÀQLVKHG 83 Greg R. Nunn Prior research Prior research on manufacturing Type 1C Daggers is contained in Errett &DOODKDQ·VXQSXEOLVKHGPDQXVFULSWA Successful Test Model of the Type 1 Dagger &DOODKDQ $OVRVHH´'DQLVK'DJJHU$µ Flintknappers Exchange &DOODKDQ (UUHWW&DOODKDQPDGHQXPHURXV7\SH&GDJJHUVÁDNHG HGJHWRHGJHRYHUVHYHQSOXV\HDUVDQGZRUNHGRXWWKHV\VWHPVHHQKHUH+H KDVGLVSOD\HGKDIWHGNQLYHVVRÁDNHGLQKLV3LOWGRZQ3URGXFWLRQVFDWDORJ VLQFH S DQGKDVSDVVHGRQWKLVNQRZOHGJHWRWKHDXWKRU ,Q WHUPVRIJULQGLQJÁLQWGDJJHUVWKHDXWKRU KDQG JURXQG QXPHURXV ÁLQWGDJJHUVRYHUDWZHOYH\HDUVSDQ8QUHFRUGHGJULQGLQJH[SHULHQFHYHUL ÀHVWKDWDVLPLODUDSSUR[LPDWLRQRIODERUZDVUHTXLUHGIRUHDFKÁLQWGDJJHU Preface Data in this paper was recorded from the manufacturing process of ten Type 1C daggers. Four completed daggers are made of Glass Buttes obsidian from 2UHJRQ 7DEV DQG IRXU FRPSOHWHG GDJJHUV DUH PDGH RI KHDWWUHDWHG 7H[DVÁLQWIURPWKH(GZDUGV3ODWHDX 7DEV 2EVLGLDQZDVXVHGÀUVWWR refresh and practice techniques. Data was collected for comparison analy VLV2IWKHUHPDLQLQJWZRÁLQWGDJJHUVERWKZHUHFRPSOHWHGWKURXJKVWDJH WKUHHWRFROOHFWSUHOLPLQDU\GDWD 7DEV 'DQLVKÁLQWGDJJHUVZHUHQRWKHDWWUHDWHGLQ1HROLWKLFWLPHV (UUHWW&DO ODKDQ SHUVRQDO FRPPXQLFDWLRQ /DUVVRQ 2ODXVVRQ 8QKHDWHG 'DQLVK ÁLQW PHDVXUHV RQ WKH OLWKLF VFDOH &DOODKDQ 2000:16), which is considerably easier to work when compared to raw Texas ÁLQW7KH7H[DVÁLQWIURPWKH(GZDUGV3ODWHDXLVPHGLXPTXDOLW\ÁLQWZKLFK PHDVXUHVWRRQWKHOLWKLFVFDOH &DOODKDQ 7H[DVÁLQW was heated to 163 Celsius, which approximates 3.0 on the lithic scale. At 163 &HOVLXVWKHÁLQWLVEDUOH\DOWHUHG:KHQKHDWHGWR&HOVLXVWKHÁLQW is very glassy and measures 2.5 on the lithic scale (Callahan 1979 & 2000:16). The later is too glassy for dagger work. After heat treatment of 163 Celsius, (GZDUGV3ODWHDXÁLQWLVFRPSDUDEOHWRXQKHDWHG'DQLVKÁLQW7KHDXWKRU GRHVQRWKDYHHQRXJKSK\VLFDOVWUHQJWKWRFRPSOHWHHGJHWRHGJHSUHVVXUH ÁDNLQJRQODUJHÁLQWGDJJHUVZLWK(GZDUGV3ODWHDXÁLQWZLWKRXWÀUVWKHDW treating to 163 Celsius. 84 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Type 1C Attribute Table Table 1. Dagger Number 1 Material: Dacite Obsidian Spall Stage Attributes 1 2 3 4 5 6 7 8 34,5 27,5 25,5 24,5 23,3 23,3 23 23 Width at widest span - cm 22 19 11,5 8,3 4,8 4,5 4,3 3,8 Thickness at widest span - cm 4 4 2,5 1,8 1 0,7 0,55 0,55 1: 4.5 1: 4.5 1: 4.5 1: 6 1: 7 Length - cm Width to thickness ratio Weight (g) 1: 7 64,3 Total grinding strokes 4600 Width lost from grinding - cm 0,3 Thickness lost from grinding - cm 0,15 Weight (g) lost from grinding :LGWKORVWIURPSUHVVXUHÁDNLQJFP 0,2 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP 0,15 :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ 0,5 :LGWKORVWIURPÀQDOUHWRXFKFP :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. 5 min. 15 min. 80 min. 120 min. 180 min. 162 min. 35 min. 25 min. 10 HRS. 21 MIN. Total Time Remarks: Between stages 2-4; 3.0 cm was lost in length from constant breakage of distal end. %ODQNER[HVUHÁHFWVGDWDQRWFROOHFWHG Type 1C Attribute Table Table 2. Dagger Number 2. Material: Dacite Obsidian Spall; core blank Stage Attributes 1 2 3 4 5 6 7 8 22,3 Length - cm 37,5 31 28 26,7 22,9 22,9 22,7 Width at widest span - cm 28,5 26 17 9 5 4,5 4,2 4,1 6 6 3,2 1,8 1 0,82 0,63 0,63 1: 5 1: 5 1: 5 1:5 1: 6.6 1: 6.5 Thickness at widest span - cm Width to thickness ratio Weight (g) 83,9 Total grinding strokes 4800 Width lost from grinding - cm 0,5 Thickness lost from grinding - cm 0,18 Weight (g) lost from grinding :LGWKORVWIURPSUHVVXUHÁDNLQJFP 0,3 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP 0,19 :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ 0,1 :LGWKORVWIURPÀQDOUHWRXFKFP :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. 5 min. 20 min. 130 min. 150 min. 180 min. 172 min. 40 min. 35 min. 12 HRS. 12 MIN. Total Time Remarks: Spall had a long crack beginning at distal end upper left lateral margin running inwards 6 cm at 80 degrees. Spall should have been rejected. Lost 10 cm in length because of crack. %ODQNER[HVUHÁHFWVGDWDQRWFROOHFWHG 85 Greg R. Nunn Type 1C Attribute Table Table 3. Dagger Number 3. Material: Gray Banded Obsidian from Glass Buttes Oregon; core blank Stage Attributes 1 2 3 4 5 6 7 8 Length - cm 41 33,5 33,5 33,2 31,7 31,5 31,4 31,4 22,3 17,5 13,2 9,2 5,3 4,8 4,65 4,5 7,7 2,8 2,4 1,9 1,15 0,95 0,79 0,79 1: 5.5 1: 4.8 1: 4.8 1:5 1: 5.8 Width at widest span - cm Thickness at widest span - cm Width to thickness ratio Weight (g) 1: 5.7 134,2 Total grinding strokes 7300 Width lost from grinding - cm 0,5 Thickness lost from grinding - cm 0,2 Weight (g) lost from grinding :LGWKORVWIURPSUHVVXUHÁDNLQJFP 0,15 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP 0,16 :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ 0,15 :LGWKORVWIURPÀQDOUHWRXFKFP :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. 0 min. 25 min. 75 min. 170 min. 205 min. 270 min. 35 min. 15 min. 13 HRS. 15 MIN. Total Time Remarks: The latter part of stage 2 resulted in a massive overshot which created a major disadvantage. The biface instantly advanced to the PLGGOHRIVWDJH%ODQNER[HVUHÁHFWVGDWDQRWFROOHFWHG Type 1C Attribute Table Table 4. Dagger Number 4. Material: Gray Banded Obsidian from Glass Buttes Oregon; Spall Stage Attributes 1 2 3 4 5 6 Length - cm 7 8 35,3 Width at widest span - cm 5,7 Thickness at widest span - cm 0,85 Width to thickness ratio 1: 6.7 Weight (g) 227,5 Total grinding strokes 8600 Width lost from grinding - cm Thickness lost from grinding - cm Weight (g) lost from grinding :LGWKORVWIURPSUHVVXUHÁDNLQJFP 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ :LGWKORVWIURPÀQDOUHWRXFKFP :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. 303 min. Total Time %ODQNER[HVUHÁHFWVQRGDWDFROOHFWHG 86 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Type 1C Attribute Table Table 5. Dagger Number 5. Material: Texas Flint Edwards Plateau; Tabular Nodule Stage Attributes 1 2 3 4 5 6 7 8 Length - cm 33,8 32,5 32,5 31,5 31,2 31,2 31 31 Width at widest span - cm 19,5 16,5 11,5 7,6 4,5 4,3 4,1 4 Thickness at widest span - cm 2,8 2,8 2,6 1,8 0,9 0,85 0,75 0,75 1: 4 1: 4 1: 5 1: 5 1: 5 1: 5 Width to thickness ratio Weight (g) 131 Total grinding strokes 28000 Width lost from grinding - cm 0,2 Thickness lost from grinding - cm 0,1 Weight (g) lost from grinding :LGWKORVWIURPSUHVVXUHÁDNLQJFP 0,2 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP 0,1 :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ 0,1 :LGWKORVWIURPÀQDOUHWRXFKFP :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. 0 min. 17 min. 85 min. 300 min. 315 min. 988 min. 28 min. 8 29 HRS. 1 MIN. Total Time 7RWDOJULQGLQJWLPHPRUHFORVHO\UHÁHFWVDXWKHQWLFUHSOLFDWLRQEHFDXVHQRPHFKDQLFDOSURFHVVZDVXVHG%ODQNER[HVUHÁHFWVQRGDWDFROOHFWHG Type 1C Attribute Table Table 6. Dagger Number 6. Material: Texas Flint Edwards Plauteau; Tabular Nodule Stage Attributes 1 2 3 4 5 6 7 8 35,5 35,5 33 31,8 31,7 31,5 31,5 31,3 Width at widest span - cm 19 17 11,5 7,7 4,9 4,6 4,45 4,3 Thickness at widest span - cm 3 3 2,7 1,8 1 0,9 0,75 0,75 1: 4 1: 4 Length - cm Width to thickness ratio Weight (g) 1: 5 1: 5 1: 6 1: 5.7 201 180,2 153,8 144,8 Total grinding strokes *28,000 Width lost from grinding - cm 0,3 Thickness lost from grinding - cm 0,1 Weight (g) lost from grinding 20,8 0,15 :LGWKORVWIURPSUHVVXUHÁDNLQJFP 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP 0,15 :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ 26,4 0,15 :LGWKORVWIURPÀQDOUHWRXFKFP 9 :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. 0 min. 20 min. 140 min. 265 min. 285 min. 360/120 35 min. 30 min. 20 HRS. 15 MIN. Total Time Remarks: *Regarding grinding, each hour of mechanical grinding approximates 8,000 hand strokes. Dagger 6 was ground 12,000 strokes by hand, the rest was completed machanically (2 hrs.), adding an equivelant to 16,000 hand strokes.& Regarding time per stage, 360 minutes by KDQGSOXVDQDGGLWLRQDOPHFKDQLFDOPLQXWHVWRWDOVPLQXWHV%ODQNER[HVUHÁHFWVQRGDWDFROOHFWHG 87 Greg R. Nunn Type 1C Attribute Table Table 7. Dagger Number 7 Material: Texas Flint Edwards Plateau; Tabular Nodule Stage Attributes 1 2 3 4 5 *6 7 8 Length - cm 40,5 38 34 33,5 32,8 32,8 32,5 32,5 Width at widest span - cm 19,5 17,6 14 9,2 5,1 4,8 4,5 4,4 Thickness at widest span - cm 2,7 2,7 2,5 1,7 1 0,9 0,8 0,8 1: 5 1: 5 1: 5 1 :5 1 :5.5 1 :5.5 212,6 180,3 153,8 146,7 Width to thickness ratio Weight (g) Total grinding strokes *22,466 . Width lost from grinding - cm 0,3 Thickness lost from grinding - cm 0,1 Weight (g) lost from grinding 32 0,3 :LGWKORVWIURPSUHVVXUHÁDNLQJFP 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP 0,1 :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ 26,5 0,1 :LGWKORVWIURPÀQDOUHWRXFKFP 7,1 :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. 0 min. 30 min. 115 min. 287 min. 300 min. 263 min. 55 min. Total Time 30 min 18 hrs. Remarks: * Regarding grinding, each hour of mechanical grinding approximates 8,000 hand strokes. Dagger 7 was ground 3,400 strokes by hand, the rest was completed machanically (143 min.), adding an equivalent to 19,066 hand strokes. **Regarding time per stage, 120 minutes E\KDQGSOXVDQDGGLWLRQDOPHFKDQLFDO%ODQNER[HVUHÁHFWVQRGDWDFROOHFWHG Type 1C Attribute Table Table 8. Dagger Number 8. Material: Texas Flint Edwards Plateau; Tabular Nodule Stage Attributes 1 2 3 4 5 *6 7 8 Length - cm 32,5 31,1 29,5 27,7 27,3 27,3 27,3 27,2 Width at widest span - cm 16,5 15,6 11,5 8,2 5,3 5,1 4,7 4,6 Thickness at widest span - cm 2,5 2,5 2,2 1,6 0,9 0,8 0,7 0,7 1 :5 1 :5 1 :6 1: 6 1: 6.5 1: 6.5 1087 466 180 155,3 126,6 121 Width to thickness ratio Weight (g) 2412 1958 Total grinding strokes *18,105. Width lost from grinding - cm 0,2 Thickness lost from grinding - cm 0,1 Weight (g) lost from grinding 24,7 0,4 :LGWKORVWIURPSUHVVXUHÁDNLQJFP 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP 0,1 :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ 28,7 :LGWKORVWIURPÀQDOUHWRXFKFP 0,1 :HLJKWORVW J IURPÀQDOUHWRXFK 5,6 Time per stage - min. 0 min. 25 min. 90 min. Total Time 270 min. 310 min. 325 min. 34 min. 5 min. 17 HRS. 39 MIN. Remarks: * Regarding grinding, each hour of mechanical grinding approximates 8,000 hand strokes. Dagger 8 was ground 6,800 strokes by hand, the rest was completed machanically (85 min.), adding an equivalent to 11,305 hand strokes.**Regarding time per stage, 240 minutes by KDQGSOXVDQDGGLWLRQDOPHFKDQLFDO'DJJHUEURNHGXULQJÀQDOUHWRXFK%ODQNER[HVUHÁHFWVQRGDWDFROOHFWHG 88 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Type 1C Attribute Table Table 9. Dagger Number 9. Material: Texas Flint Edwards Plateau; Tabular Nodule Stage Attributes 1 2 3 Length - cm 26 26 26 Width at widest span - cm 22 17,5 11,5 Thickness at widest span - cm 4 4 3 Width to thickness ratio 2298 2071 1: 3.8 Weight (g) 2298 2071 1164 0 min. 15 min. 85 min. 4 5 6 7 8 5 6 7 8 Total grinding strokes Width lost from grinding - cm Thickness lost from grinding - cm Weight (g) lost from grinding :LGWKORVWIURPSUHVVXUHÁDNLQJFP 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ :LGWKORVWIURPÀQDOUHWRXFKFP :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. Total Time 5HPDUNV'DJJHULQFRPSOHWH%ODQNER[HVUHÁHFWVQRGDWDFROOHFWHG Type 1C Attribute Table Table 10. Dagger Number 10. Material: Texas Flint Edwards Plateau; Tabular Nodule Stage Attributes 1 2 3 Length - cm 29,5 29,5 29,5 Width at widest span - cm 21,5 17 11,2 Thickness at widest span - cm 3,1 3,1 2,9 Width to thickness ratio Weight (g) 4 1: 3.8 2951 2262 1299 20 min. 92 min. Total grinding strokes Width lost from grinding - cm Thickness lost from grinding - cm Weight (g) lost from grinding :LGWKORVWIURPSUHVVXUHÁDNLQJFP 7KLFNQHVVORVWIURPSUHVVXUHÁDNLQJFP :HLJKW J ORVWIURPSUHVVXUHÁDNLQJ :LGWKORVWIURPÀQDOUHWRXFKFP :HLJKWORVW J IURPÀQDOUHWRXFK Time per stage - min. Total Time Remarks: Dagger terminated after encountering pre-existing crack at the proximalend, extending longitudinally for 9 cm towards distal end DQGVWRSSLQJLQWKHPLGGOH]RQH%ODQNER[HVUHÁHFWVQRGDWDFROOHFWHG 89 Greg R. Nunn Figure 5. Author grinGLQJDÁLQWGDJJHURQ sandstone slab and Douglas Fir pole table. Figure 6. Holding position used to make edge-to-edge pressure ÁDNHV 90 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking The focus of information is on the later stages of manufacture; grinding DQGSUHVVXUHÁDNLQJWHFKQLTXHV )LJV *ULQGLQJDQGSUHVVXUHÁDN LQJWHFKQLTXHVDUHUHTXLUHGWRH[HFXWHHGJHWRHGJHSDUDOOHOSUHVVXUHÁDN ing. The application of direct and indirect percussion thinning strategies on earlier stages one through three will for the most part be omitted. However, empirical data regarding weights and morphology is included to provide the UHDGHU ZLWK D FRQWLQXXP RI WKH SURFHVV VHH 7DEOHV ,Q DGGLWLRQ IRU DFRPSOHWHUDQJHRIWRROVXVHGLQDOOVWDJHVUHIHUWRÀJXUHVDQG)RU LQIRUPDWLRQRQUHSOLFDWLQJVWDJHVRQHWKURXJKIRXUVHH&DOODKDQ·VThe Basics of Biface Knapping in the Eastern Fluted Point Tradition (1979 & 2000). For an ex DPSOHRIHDUO\UHGXFWLRQVHTXHQFHVWDJHVRQHDQGWZRVHHÀJXUHVDQG It is critical one have a thorough understanding of the morphology of stages WKUHHWKURXJKÀYH )LJV 7KHSURSHUH[HFXWLRQRISUHFLVHGLUHFWSHU FXVVLRQELIDFHWKLQQLQJLQVWDJHVIRXUDQGÀYHZLOOGHWHUPLQHLIWKHELIDFH can be taken to stage six (grinding stage) (Fig. 15). Figure 7. (a) Medium hard heavy sandstone hammerstone, 450 grams; (b) Soft sandstone hammerstone, 409 grams; (c) Soft sandstone hammerstone, 134 grams; (d) Sandstone abrader, 90 grams. Figure 8. (a) Heavy moose antler billet, 810 grams; (b) Heavy moose antler billet, 545 grams; (c) Medium heavy moose antler billet, 410 grams; (d) Moose antler punch for indirect percussion, 96 grams. 91 Greg R. Nunn Figure 9. (a) Ishi stick SUHVVXUHÁDNHUZLWKFRSper tip used for edge to HGJHSUHVVXUHÁDNLQJ E 6PDOOSUHVVXUHÁDNHU with copper tip used for retouch; (c) Sandstone baton used with punch for indirect percussion, 383 grams; (d) Leather thimble worn on index ÀQJHU Figure 10. Stage one ²XQPRGLÀHG7H[DV ÁLQWQRGXOHLOOXVWUDWHG by Greg Nunn and Joe Pachak. Figure 11. Stage two – initial edging, Texas ÁLQWLOOXVWUDWHGE\*UHJ Nunn and Joe Pachak. 92 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Figure 12. Stage three – primary preform, 7H[DVÁLQWLOOXVWUDWHGE\-RH3DFKDN Figure 13. Stage four - secondary preform, GDJJHUQXPEHUHLJKW7H[DVÁLQWLOOXVWUDWHG by Joe Pachak. )LJXUH6WDJHÀYH²ÀQDOSUHIRUPGDJJHU QXPEHUVHYHQ7H[DVÁLQWLOOXVWUDWHGE\-RH Pachak. 93 Figure 15. Stage six – ground preform, dagger QXPEHUVL[7H[DVÁLQWLOOXVWUDWHGE\*UHJ Nunn and Joe Pachak. Greg R. Nunn Overview The knapper must take absolute command of thinning strategies in stages IRXUDQGÀYH)RUH[DPSOHLIGLYLQJÁDNHVDUHFUHDWHGLQWRWKHPLGGOH]RQH (Callahan, Cliffside Workshop, 1999: Fig. 16) one has produced a concavity QHJDWLYHÁDNHVFDU WKDWZLOOFUHDWHDODERULRXVKDQGJULQGLQJSURFHVV)RU PRUHLQIRUPDWLRQRQ]RQHVVHH&DOODKDQ$SHODQG2ODXVVRQ·VXSFRPLQJ book Neolithic Danish Daggers: An Experimental and Analytical Study. The biface PXVWEHJURXQGWRWKHORZHVWSODQHRIWKHGHHSHVWQHJDWLYHÁDNHVFDU,IWKH biface is not properly thinned and left too thick, it will cause excessive grind ing which will take many hours and several thousand extra grinding strokes. In cases such as these, the biface may lead to rejection before grinding is initiated. It is easier to grind out high spots than to grind out low spots. What LVXOWLPDWHO\GHVLUDEOHLVWRFUHDWHDVWDJHÀYHÀQDOSUHIRUPZLWKDOHQWLFXODU cross section and an average width to thickness ratio of 1:5 to 1:6 (Fig. 14). Figure 16. Outline of biface illustrating zones as indicated by E. Callahan:1999; illustrated by Greg Nunn. 94 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Methodology – stage four A valuable strategy from Errett Callahan in gauging biface contours and SURSHUSHUFXVVLRQÁDNHUHPRYDOLVDWHFKQLTXHWKDWXWLOL]HVDVWUDLJKWZRRGHQ dowel (Callahan, Cliffside Workshop, 1991). The author prefers the dowel to EHDWOHDVWDVORQJDVWKHELIDFH7KHGRZHOLVSODFHGÁDWRQWKHELIDFHVXUIDFH WLSWREDVHZKLFKUHYHDOVKLJKVSRWVDQGORZVSRWSUREOHPVIRUIXUWKHUÁDNH UHPRYDO'XULQJWKLVVWDJHRIGLUHFWSHUFXVVLRQÁDNHUHPRYDOXVLQJDODUJH WRPHGLXPODUJHPRRVHDQWOHUELOOHW )LJDDQGF ÁDNHUHPRYDOVKRXOGEH well controlled, widely spaced and systematic. To achieve proper contours, WKHSULQFLSDOSHUFXVVLRQWKLQQLQJÁDNHVWREHUHPRYHGVKRXOGWUDYHOIURP margin to margin with feathered terminations (Fig. 17). The result prepares WKHELIDFHIRUDQLGHDOFURVVVHFWLRQ(GJHDQJOHVKRXOGUDQJHIURPWR degrees (Callahan 1979 & 2000:30). Stage four widths to thickness ratios range from 1:4 to 1:5. Figure 17. Stage four - three complete direct percussion ELIDFHWKLQQLQJÁDNHVIURPGDJJHUQXPEHUÀYHLOOXVWUDWHG by Greg Nunn. 95 Greg R. Nunn 0HWKRGRORJ\²VWDJHÀYH 7KHEHJLQQLQJRIVWDJHÀYHZLOOXWLOL]HWKHVDPHÁDNHUHPRYDOWHFKQLTXHDV explained in stage four. There will be high spots on the surface of the biface LQWKHPLGGOHRIVWDJHÀYH8VLQJWKHZRRGHQGRZHODVDFRQWRXUJDXJHDQG direct percussion using a small soft hammerstone or medium moose antler billet (Figs. 7c & 8c), high spots will be systematically removed. In addition, WKHVHÁDNHVZLOOEHVPDOOHUWKDQWKHSUHYLRXVWKLQQLQJÁDNHV $WWKLVWLPHWKHNQDSSHUPXVWEHH[WUHPHO\FDXWLRXVWRSUHYHQWÁDNHV from diving into the middle zone (Fig. 16) of the biface. Ultimately, feath HUHGÁDNHWHUPLQDWLRQVDUHUHTXLUHG $IWHUÀQDOGLUHFWSHUFXVVLRQWKLQQLQJKDVEHHQFRPSOHWHGLWZLOOEHWLPH IRUWKHÀQDODOLJQPHQWRIWKHODWHUDOPDUJLQ6RPHSUHVVXUHÁDNLQJZLOOEH required for the removal of deltas, ridges, overhangs, and concavities that have occurred in the outer zone. The outer zone is the surface area from the ODWHUDOPDUJLQLQZDUGRQHWKLUGWKHZLGWK )LJ $VVXPLQJSURSHUH[HFX tion of percussion thinning has taken place, there will be very little pressure ÁDNLQJIRUÀQDOFOHDQXSRIWKHÀQDOSUHIRUP Next, one creates opposing bevels on both lateral margins (Callahan 2001: )LJ E\UHPRYLQJVPDOOSUHVVXUHUHWRXFKHGÁDNHVXSWRPPLQOHQJWK LQVXFFHVVLRQDORQJWKHHQWLUHPDUJLQ3ODFHWKHSUHVVXUHÁDNLQJWRROPP – 1 mm above the lateral margin and push downwards at a steep angle into the palm of the padded hand (Fig. 6). Creating opposing bevels will bring the margin line slightly below the centerline of the biface (Callahan Basics 1979 & 2000:34). )LQDOZLGWKWRWKLFNQHVVUDWLRVUDQJHIURPWRDQGWKHÀQDOHGJHDQ JOHEHIRUHJULQGLQJVKRXOGEHGHJUHHV7KHKDQGOHDQGEODGHVKRXOG EHWKHVDPHWKLFNQHVVDWWKHHQGRIVWDJHÀYH )LJ Methodology – stage six: grinding 2QO\LQWKH1HROLWKLFSHULRGGLGWKHWHFKQLTXHRIGHOLEHUDWHÁDNHRYHUJULQG ing take place. Evidence is found in Neolithic Egypt with the Gerzean knives (Kelterborn 1984) and in Denmark with various types of Danish Daggers (Callahan, unpublished paper A Successful Test Model of the Type 1 Danish Dagger 1985 and personal communication 1991). Flake over grinding gives a clean and precise appearance, which has an aesthetic appeal. The grinding stage is time consuming and laborious. One must hand grind both faces of the dag ger, excluding the handle. The author used sandstone in order to prepare the VXUIDFHRIWKHSUHIRUPIRUSUHVVXUHÁDNLQJ 96 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking It is notable that the use of sandstone slabs for grinding daggers has not been found in Denmark, or if so, has not been recorded as such (Apel, personal communication 2002). 7KDWYHULÀHGGDJJHUJULQGVWRQHVODEVKDYHQRWEHHQ located is a missing link in the Dagger making process. However, archaeologi FDOHYLGHQFHFRQÀUPVWKDW7\SH&'DQLVK'DJJHUVZHUHJURXQG )LJ $KLJKTXDOLW\JULQGLQJVWRQHLVGLIÀFXOWWRREWDLQ2QHZDQWVWRJHWWKH most out of the implement, especially if it has been transported or traded over long distances. The author traveled 187 miles (300 kilometers) to acquire the ultimate grindstone from the Chinle formation (from the Upper Triassic) LQ *DUÀHOG &RXQW\ VRXWKHDVWHUQ 8WDK 7KH VDQGVWRQH VKRXOG EH PRGHU ately bonded with silicate and composed mainly of quartz granules. During grinding, quartz granules should be continually released over the surface area to enhance cutting ability. The continual release of quartz granules will rejuvenate the surface and prevent it from glazing over. If harder sandstone is used, it will glaze over and require pecking to expose new cutting surfaces. Glazing over forms when high silicate content bonds the quartz granules to gether within the sandstone. The proper sandstone has a lower percentage of bonding silicate. The Chinle sandstone used by the author is approximately 50 grit, as compared to modern sandpaper. By adding quartz sand to the grindstone, grinding time is reduced. Quartz sand acts as an extra cutting agent and rejuvenates fresh granules from the grindstone. In addition, it is necessary to continually add water to the grind stone while grinding the preform. If the additional quartz granules are too big they will cut the sandstone rapidly, reducing the life of the grindstone (Dagger Research Project 2002). Contrarily, if the grindstone is too hard, the quartz will shatter into smaller particles and not cut effectively. One needs to experiment in order to realize optimum grinding results. Quartz sand was collected at the base of the Cretaceous Dakota sandstone horizontal cap that is a conglomerate composed mainly of quartz sand. The quartz sand particles are rounded from ancient geologic activities, which is a disadvantage in this application. Angular quartz granules cut the best. In addition, commercially crushed quartz masonry sand was used which is angular in structure and is available in 10 to 20 grit. Both Cretaceous quartz sand, and commercially crushed quartz masonry sand was used for all dag gers. Errett Callahan used crushed burned quartz chunks and screened to size by hand (Dagger Research Workshop 2002). Two sandstone slabs were used for this project. One slab measured 40 cm long, 23 cm wide and 6.5 cm thick. The slab was used solely for grinding four obsidian Type 1C Danish Daggers. A larger slab measured 50 cm long, 36 cm wide, and 6 cm thick, with a beginning weight of 19 kilograms and was XVHGVROHO\IRUÁLQWGDJJHUV 97 Greg R. Nunn Sandstone slabs were placed horizontally on a table made of Douglas Fir poles (Fig. 5). Table legs were set in postholes 46 cm deep. The height of the table measured 75 cm, 80 cm long and 97 cm wide, and was assembled without the use of nails. The tabletop was lashed together with jute cordage. $OOLQLWLDOFXWVDQGWUHHIHOOLQJZHUHGRQHZLWKDQHLJKWSRXQGPHWDOVSOLWWLQJ PDOO7KHÀQDOVTXDULQJRIWKHWDEOHWRSZDVFXWZLWKDKDQGKHOGERZVDZ to save time. A similar table could have been constructed in Neolithic times except for the use of metal cutting tools that could have been converted to ÁLQWD[HVDQGDG]HV$UXGLPHQWDU\WDEOHVXFKDVDWUHHVWXPSFRXOGDOVR have been used (Madsen 1984). 1RZWKDWWKHÀQDOSUHIRUPKDVEHHQSURSHUO\SUHSDUHGLWLVWLPHWRJULQG WKHSUHYLRXVO\SUHSDUHGGHJUHHEHYHOHGODWHUDOPDUJLQWRDGHJUHH margin using a handheld sandstone abrader (Fig. 7d). One can also grind the margin directly on the grinding slab, but there are disadvantages. Groves will be cut below the surface, which causes irregularities with long and deep striations (Dagger Research Project 2002). In addition, the preform cuts TXLFNO\LQWRWKHVODEDQGH[KDXVWVWKHVDQGVWRQHIDVWHUHVSHFLDOO\ZLWKÁLQW Phases of grinding Four phases of grinding will be described. A summary of each phase is pro vided, followed by detailed descriptions of each phase. x Phase 1: Grind or abrade lateral margin. x Phase 2: Moderately grind predominantly high spots. x Phase 3: Grind from the margin edge to the outer zone (approxi mately 1 cm) removing deltas, ridges, and concavities. x 3KDVH0DMRUDQGÀQDOJULQGLQJRIHQWLUHIDFHH[FHSWKDQGOHÀQDO control of contours, cross section irregularities (high spots and neg DWLYHÁDNHVFDUHV DQGJULQGLQJIDFHWVWKDWPD\KDYHEHHQFUHDWHG Phase 1 8VLQJDKDQGKHOGDEUDGHU )LJG WKHODWHUDOPDUJLQLVDWÀUVWJURXQGWR degrees and 1 mm thick. The purpose of grinding the lateral margin is to strengthen it so that when grinding in the outer zone and along the margin 98 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking of the dagger, one will not chip the margin. If the margin is too thin, it be comes fragile and when the edge makes contact with the grindstone it will chip away and cause undesirable concavities along the lateral margin. During phases 3 and 4 there will be considerable grinding from the outer zone and up to the edge of the margin. Because of this grinding, the margin will become thinner. It will again be necessary to regrind the margin to 90 GHJUHHV*ULQGLQJWKHGHJUHHPDUJLQZLOORFFXUVHYHUDOWLPHVGXULQJWKH grinding process of each face. Phase 2 Begin by grinding predominate high spots in the middle zone to a lower plane. This will take some effort. Initial grinding of high spots will range IURPWRVWURNHVGHSHQGLQJRQWKHPDWHULDO REVLGLDQRUÁLQWUH spectively). Phase 3 Grind the outer zone approximately 1 cm from the margin edge and up to the margin edge. Grinding will focus on minute deltas, ridges, and concavi WLHVOHIWRYHUIURPHDUOLHUSUHVVXUHUHWRXFKLQVWDJHÀYH$VJULQGLQJFRQWLQ ues in this area the lateral margin will become thinner. Once again, it will be time to thicken the lateral margin by regrinding it with a handheld abrader. As previously mentioned, thickening the lateral margin prevents chipping of the margin in future grinding. By this time one has easily accumulated an DGGLWLRQDO²VWURNHV XVLQJÁLQW ,QDGGLWLRQJULQGLQJWLPHZLOO YDU\GHSHQGLQJRQSUREOHPDUHDVDQGPDWHULDOXVHG REVLGLDQYHUVHVÁLQW Grinding will continue up to the margin edge and to the outer zone for the ÀQLVKLQJWRXFKHVODWHU Phase 4 The major grinding will now take place. Grinding may be ¼ completed. 3KDVHZLOOWDNHÀQDOFRQWURORIWKHFRQWRXUVDQGFURVVVHFWLRQLUUHJXODUL ties. Grind the remaining high spots and negative concavities. When grinding the preform, it is important not to focus on one locale for a long period of time. If grinding continues for an excessive amount of WLPHLQRQHDUHDDÁDWIDFHWZLOOUHVXOW$ÁDWIDFHWLVXQGHVLUDEOHDQGZLOOWDNH considerable labor to recontour (Madsen 1984). Any ridge remaining from WKHIDFHWPD\FDXVHWKHÀQDOVHULHVRIHGJHWRHGJHÁDNHUHPRYDOH[WUHPHO\ 99 Greg R. Nunn GLIÀFXOW7KHULGJHFDQDFWDVDZDOODQGVWRSWKHÁDNHVKRUW6RPHDUHDVRI the preform will need more grinding attention than others. As with creating IDFHWVLW·VLPSRUWDQWQRWWRFUHDWHORZVSRWVRUVKDOORZFRQFDYLWLHV,IRQH JULQGVWRRPXFKRIDFRQFDYLW\WKLVPD\DOVRVWRSWKHÀQDOSUHVVXUHÁDNH short. Grinding the handle of the preform is not necessary, nor was it elaborate O\SUHVVXUHÁDNHG+RZHYHUWKHRULJLQDO7\SH&'DQLVK'DJJHU·VKDQGOH (Fig. 1) has occasional grinding striations, due to haphazard contact during WKH JULQGLQJ SURFHVV &DOODKDQ·V 7\SH & 'DJJHU 5HSOLFDWLRQ &ODVV Grinding contact on the handle did not pose a problem for the Neolithic knappers because the completed daggers were sometimes covered with a wooden handle or other coverings (Jorgensen 1992; as per Apel 2001:254). Detail of grinding process When grinding, one takes long strokes in a back and forth motion along WKHORQJHVWD[LVRIWKHJULQGVWRQH 0DGVHQ $VWURNHLVGHÀQHGE\ beginning at the lower end closest to the person grinding, pushing forward to the opposite end of the grindstone then pulling backwards to the person grinding. Grinding in this manner will result in contours comprised of con sistent lenticular cross sections. In addition, facets and concavities will be prevented. Hold the dagger handle in one hand and the distal portion of the blade in WKHRWKHUKDQG )LJ %HJLQWKHÀUVWVWURNHLQWKHRXWHU]RQHXQGHUQHDWK the lateral margin edge farthest from you (Fig. 5). Apply pressure and push the preform away toward the opposite end of the grindstone. When pushing away, gradually rotate the biface in an arching manner so that when the biface reaches the opposite end of the grindstone, the margin closest to you will be making contact with the grindstone. Next reverse the stroke by pulling the biface backward toward yourself. Once again, rotate the biface gradually in an arching manner, so that when the biface returns to the starting position, the lateral margin farthest away will be making grinding contact with the grindstone. Most all the surfaces will be making contact during each stoke. $VSUHYLRXVO\VWDWHGWKHJULQGVWRQHWKHDXWKRUXVHGIRUÁLQWGDJJHUVLV cm long. A stroke is the total of 100 cm. In addition, it will take a consider DEOHDPRXQWRIWLPHWRFRPSOHWHWKHÀUVWIDFH$VJULQGLQJFRQWLQXHVWKH lateral margin will again become thinner. Toward the completion of grind LQJWKHÀUVWIDFHDQGDVWKHPDUJLQEHFRPHVLQFUHDVLQJO\WKLQQHULWLVDGYLV able to decrease grinding pressure when grinding around the margin. One can stop using quartz sand and grind in a gentler fashion so as not to chip 100 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking WKHPDUJLQ$IWHUDOOWKHKLJKVSRWVDQGPRVWRIWKHQHJDWLYHÁDNHVFDUVDUH UHPRYHGWKHÀUVWIDFHLVGRQH,GHDOO\RQHVKRXOGHQGXSZLWKERWKPDUJLQV UHODWLYHO\VKDUSDWWKHHQGRIJULQGLQJWKHÀUVWIDFH ,WLVLPSRUWDQWWRUHDOL]HWKDWHOLPLQDWLQJDOOQHJDWLYHÁDNHVFDUVPD\EH XQDWWDLQDEOHHVSHFLDOO\ZLWKÁLQWRQWKHÀQLVKHGJURXQGSUHIRUP,WLVOLNHO\ UHPQDQWVRIQHJDWLYHÁDNHVFDUVZLOOEHYLVLEOH9LVLEOHQHJDWLYHÁDNHVFDUV are to a certain extent allowable (Fig. 4). If negative scars are shallow, the ÀQDOVHULHVRISUHVVXUHÁDNHVPD\XQGHUFXWEH\RQGPLQRUORZVSRWV,IQHJD WLYHVFDUVDUHWRRGHHSWKHÁDNHZLOOQRWWUDYHOEH\RQGFRQFDYLWLHV.QRZLQJ the tolerance allowed will come with experience. There is a great deal of hard labor involved with grinding, therefore it is worth going the extra distance for the ultimate result. 7KHRSSRVLQJIDFHZLOOEHJLQWKHVDPHDVWKHÀUVWIDFHE\JULQGLQJWKH lateral margin to 90 degrees. Once the opposite face is near completion, one PXVWQRWUHJULQGWKHODWHUDOPDUJLQWRGHJUHHV7KHÀQDOJULQGLQJVKRXOG terminate with both lateral margins ground relatively sharp. The grinding stage is now complete. Width to thickness ratios should be an average of 1:5 to 1:6. 6WDJHVHYHQÀQDOSUHVVXUHÁDNLQJ Stage seven begins by hand abrading opposing beveled margins to approxi PDWHO\GHJUHHVDQGDSSUR[LPDWHO\PPWKLFNVRDQDFFXUDWHSODWIRUP FDQ EH FUHDWHG IRU UHPRYLQJ WKH ÀQDO VHULHV RI SUHVVXUH ÁDNHV 2SSRVLQJ beveled margins reduces the risk of overshot (Callahan 2001: Fig. 18). With the knapper sitting down (Fig. 6), the preform is placed on a leather SDGKHOGLQWKHOHIWKDQG7KHDUHDRIÁDNHUHPRYDOVKRXOGUHVWÀUPO\RQWKH leather protected thumb muscle (Thenar Eminence). Contact with the Th HQDU(PLQHQFHDQGELIDFHKHOSH[WHQGÁDNHWUDYHO7KHEDFNRIWKHOHIWKDQG LVSODFHGÀUPO\DJDLQVWWKHLQVLGHRIWKHOHIWWKLJKDQGWKHEDFNRIWKHULJKW KDQGLVSODFHGÀUPO\DJDLQVWWKHLQVLGHRIWKHULJKWWKLJK 7KHULJKWKDQGFRQWDLQVDKDQGKHOG,VKLVWLFNSUHVVXUHÁDNLQJWRROZLWK a copper tipped insert or antler insert (Fig. 9a). For this project, a copper tip was used. With the preform held in the left hand against the left thigh, begin DWWKHULJKWKDQGPDUJLQ3ODFHWKHWLSRIWKHSUHVVXUHÁDNLQJWRRODJDLQVWWKH JURXQGODWHUDOPDUJLQDQGEHJLQWKHSUHVVXUHÁDNLQJVHTXHQFHDWWKHGLVWDO end of the preform. Flakes will be removed from tip to base. Apply pres VXUHDQGUHPRYHWKHÀUVWSUHVVXUHÁDNH5HSHDWWKHSURFHGXUHIRUWZRPRUH ÁDNHVVSDFLQJWKHPDSSUR[LPDWHO\PPIURPWKHODVWÁDNHUHPRYHGIURP WKHPDUJLQ7KHÀUVWWKUHHÁDNHVZLOOJHQHUDOO\QRWUHDFKWKHRSSRVLWHPDU 101 Greg R. Nunn JLQEXWWKH\VKRXOGFRPHFORVH ZLWKLQPP 'XULQJHDUO\ÁDNHUHPRYDO if one pushes too hard it is entirely possible to shear the tip away. %HIRUHUHPRYLQJWKHIRXUWKÁDNHRQHPXVWSODFHWKHWLSRIWKHSDGGHG OHIW LQGH[ ÀQJHU ÀUPO\ DJDLQVW WKH RSSRVLWH OHIWKDQG ODWHUDO PDUJLQ IURP ZKHUHWKHSUHVVXUHWRROLVSODFHGWRUHPRYHÁDNHV7KHDXWKRUXVHVDQLQ GXVWULDO OHDWKHU WKLPEOH ZKLFK VWUDSV RYHU WKH ÀQJHU )LJV DQG 7KH WKLPEOHKDVDSURWHFWLYHVWUDSHQFDVHGZLWKDÁH[LEOHQRQYLVLEOHOHDGVKLHOG WKDWSURWHFWVWKHÀQJHUWLS &OLIIVLGH:RUNVKRS 7KHOHIWLQGH[ÀQJHU DFWVDVDQDQYLOIRUJXLGLQJÁDNHVWRWUDYHOLQWKHSURSHUGLUHFWLRQ,WLVHV VHQWLDOWKDWWKHOHIWLQGH[ÀQJHUEHSDGGHG3DGGLQJSUHYHQWVWKHSUHVVXUH ÁDNHIURPSHQHWUDWLQJWKHÀQJHUDQGKHOSVDEVRUEWKHSUHVVXUHWKDWLVEHLQJ DSSOLHGWRWKHÀQJHUIURPWKHSUHVVXUHÁDNLQJWRRO 7RUHPRYHWKHIRXUWKÁDNH ZHDULQJWKHÀQJHUJXDUG SODFHWKHOHIWLQGH[ ÀQJHU DJDLQVWWKHRSSRVLWHODWHUDO PDUJLQ DW WKH YHU\ WLS RI WKH GLVWDO HQG )LJ $SSO\WKHWLSRIWKHSUHVVXUHÁDNLQJWRROWRWKHODWHUDOPDUJLQRS SRVLWHWKHLQGH[ÀQJHUDWDSSUR[LPDWHO\DGHJUHHDQJOH &DOODKDQSHUVRQDO communication 1991; Fig. 1). Start by applying pressure from the pressure ÁDNLQJWRRODQGGLUHFWLQJWKLVSUHVVXUHSUHFLVHO\WRZDUGVWKHWLSRIWKHSDG GHG LQGH[ ÀQJHU 6ORZO\ LQFUHDVH WKH SUHVVXUH E\ OHDQLQJ VOLJKWO\ LQWR WKH ÁDNLQJWRRODQGDGGLQJXSSHUWRUVRZHLJKW2QHDSSURDFKHVWKHXSSHUSUHV sure load limit by pushing downwards from the torso and squeezing thighs together, Crabtree style (1973:24), directing outward force, which detaches WKHÁDNH0RVWRIWHQWKHSUHVVXUHÁDNHVKDWWHUVLQWRWZRRUWKUHHVHJPHQWV during detachment. On occasion they will stay complete (Fig. 18). 5HPRYHWKHQH[WÁDNHE\SODFLQJWKHWLSRIWKHSUHVVXUHÁDNLQJWRRODS SUR[LPDWHO\PPIURPWKHODVWÁDNH&RQWLQXHWKLVSURFHVVWRWKHEODGH and handle junction. For the most part the handle is complete, except for ÀQDO UHWRXFK DQG PLQLPDO VKDSLQJ ZKLFK ZLOO EH DFFRPSOLVKHG LQ VWDJH eight. $IWHU WKH ÀUVW VHULHV RI SUHVVXUH ÁDNHV KDYH EHHQ FRPSOHWHG WKH ULJKW hand lateral margin is somewhat irregular with sharp edges and deltas (Fig. 19). It is important to slightly retouch and straighten the right hand lateral PDUJLQ%\XVLQJDVPDOOSUHVVXUHÁDNLQJWRRO )LJE UHPRYHWKHUHPDLQ LQJSURWUXGLQJGHOWDVDQGPDUJLQLUUHJXODULWLHV5HPRYHVKRUWUHWRXFKÁDNHV at a steep angle as not to intrude into the outer zone more than 1 mm. After minimal retouch and edge straightening is complete, the margin should be somewhat sharp with minimal opposing bevels that will still help reduce the chance of overshot. If the margin is left irregular, it can lead to unsuccess IXOHGJHWRHGJHÁDNLQJIRUWKHVHFRQGIDFH6LPLODUO\PDUJLQLUUHJXODULWLHV FDQOHDGWRLPSURSHUSODFHPHQWRIWKHSDGGHGOHIWLQGH[ÀQJHUJLYLQJWKH SUHVVXUHÁDNHEHLQJGHWDFKHGWKHZURQJPHVVDJH7KHSDGGHGLQGH[ÀQJHU 102 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking DFWVDVDQDQYLODQGÁDNHGLUHFWLRQJXLGH,PSURSHUSODFHPHQWFDQOHDGWR PDVVLYH RYHUVKRW DQG SUHPDWXUH ÁDNH GHWDFKPHQW ,Q DGGLWLRQ LPSURSHU SODFHPHQW FDQ FKDQJH WKH DQJOH RI ÁDNH WUDYHO DQGRU FROODSVH WKH VKDUS HGJHZKHUHSUHVVXUHLVEHLQJDSSOLHGZLWKWKHLQGH[ÀQJHU 7KHVHFRQGIDFHLVDUHSHWLWLRQRIWKHÀUVW$IWHUWKHVHFRQGIDFHLVFRP plete, remove deltas and straighten lateral margin (same as above). Next cre ate slightly opposing bevels on both right hand lateral margins for prepara WLRQRIÀQDOUHWRXFK VWDJH :LGWKWRWKLFNQHVVUDWLRVFDQYDU\IURP to 1:7. 6WDJHHLJKWÀQDOUHWRXFK ,ISURSHUH[HFXWLRQRIHGJHWRHGJHSUHVVXUHÁDNHVZHUHDFFRPSOLVKHGWKHUH ZLOOEHYHU\OLWWOHÀQDOUHWRXFK,IHGJHWRHGJHÁDNHVZHUHQRWDFFRPSOLVKHG it will take a little more work and time to remove any left over grinding striations in the outer and middle zones that were not removed during stage seven. Abrade both opposing bevels along the left hand lateral margins to DERXWPPWKLFN+ROGLQJWKHGDJJHUÁDWRQWKHSDGGHGOHIWKDQGÀQDO UHWRXFKZLOOEHJLQE\UHPRYLQJVPDOOWRPHGLXPSUHVVXUHÁDNHV7RUHPRYH VPDOOSUHVVXUHÁDNHVEHJLQDWWKHEODGHWRKDQGOHWUDQVLWLRQDUHDDORQJWKH left hand margin and proceed left to right to the distal end. Using an Ishi 6WLFNRUVPDOOSUHVVXUHÁDNLQJWRROZLWKDVKDUSWLSUHPRYHDQ\JULQGLQJ VWULDWLRQV WKDW ZHUHQ·W VXFFHVVIXOO\ UHPRYHG GXULQJ VWDJH VHYHQ 7KLV ZLOO LQLWLDWHWKHVKDUSHQLQJSURFHVV$IWHUWKHUHWRXFKÁDNHVKDYHEHHQUHPRYHG from both left hand margins there will be some remaining deltas. Remove WKH GHOWDV XVLQJ D VPDOO DQG VKDUS SUHVVXUH ÁDNLQJ WRRO DW D VWHHS DQJOH :KHQGHOWDVDUHUHPRYHGWKHGDJJHUZLOOEHQHDUO\ÀQLVKHGDQGUHODWLYHO\ VKDUS/DVWO\FRPHVVRPHPLQLPDOÀQDOVKDSLQJRIWKHKDQGOHDORQJZLWK dulling the handle margins by abrasion for hafting or use. The handle will be VOLJKWO\WKLFNHUWKDQWKHEODGH )LJV )LQDOZLGWKWRWKLFNQHVVUDWLRV FDQYDU\IURPWR1RWH)LJ GDJJHU EURNHGXULQJÀQDOUHWRXFK The dagger broke because the margin was slightly above center and the pres VXUHÁDNLQJWRROZDVEOXQW%UHDNDJHRFFXUUHGGXULQJÁDNHGHWDFKPHQW)LJ 22a and b are completed daggers (numbers 3 and 5). 103 Greg R. Nunn Figure 18. Stage seven three complete edge-toHGJHSUHVVXUHÁDNHVIURP dagger number eight; illustrated by Greg Nunn. )LJXUH6WDJHVHYHQ²DIWHUÀUVW VHULHVRISUHVVXUHÁDNHVGDJJHU number eight; illustrated by Greg Nunn and Joe Pachak. Figure 20. Stage eight – dagger number eight; an excellent examSOHRIHGJHWRHGJHÁDNLQJZKLFK EURNHGXULQJÀQDOUHWRXFKLOOXVtrated by Greg Nunn. 104 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Figure 22(a) Completed obsidian dagger number three; (b) completed ÁLQWGDJJHUQXPEHUÀYHLOOXVWUDWHG by Greg Nunn and Joe Pachak. Figure 21. Photograph of dagger number eight. Presentation of data Grinding The number of grinding strokes to complete each dagger was recorded. An average of 1,700 strokes per hour was accomplished. It is possible to com plete more strokes per hour, but one is continuously evaluating and record ing the progress. Evaluating the progress reduces grinding time. All obsidian daggers were ground on Chinle sandstone enhanced with ap SUR[LPDWHO\JULWHTXLYDOHQWTXDUW]VDQG%HIRUHJULQGLQJWKHJULQGVWRQH·V greatest length was 40 cm, greatest width 23 cm, and greatest thickness 6.5 cm. Upon completion of grinding four obsidian daggers, a total of 4 cm in thickness was lost from the grindstone. Weight data was not retrieved. 105 Greg R. Nunn $VPDOOREVLGLDQÀQDOSUHIRUP GDJJHU FPLQOHQJWKWKHJUHDWHVWZLGWK of 5 cm, and the greatest thickness of 1.1 cm was ground. This dagger re quired a total of 4,800 grinding strokes, equaling the distance of 3,840 me ters or 3.84 kilometers. An average of 1.6 mm in thickness was lost from grinding (Tab. 2). $PRGHUDWHO\VL]HGREVLGLDQÀQDOSUHIRUP )LJDGDJJHU WKHOHQJWK of 31.5 cm, greatest width of 5.3 cm, greatest thickness of 1.1 cm, required 7,300 strokes, equaling the distance of 5,840 meters or 5.84 kilometers over the grindstone (Tab. 3). An average of 2.0 mm in thickness was lost from grinding. $ODUJHREVLGLDQÀQDOSUHIRUP GDJJHU WKHOHQJWKRIFPJUHDWHVW width of 6.5 cm, and greatest thickness of 1.3 cm, required 8,700 stokes, equaling the distance of 6,960 meters or 6.96 kilometers over the grindstone (Tab. 4). 7KHÁLQWGDJJHUVZHUHJURXQGRQ&KLQOHVDQGVWRQH IURPWKH8SSHU7UL assic) enhanced with approximately 20 grit equivalent quartz sand. Before JULQGLQJWKHJULQGVWRQH·VJUHDWHVWOHQJWKZDVFPJUHDWHVWZLGWKFP and greatest thickness 6 cm. Weight before use was approximately 19 kilo JUDPV8SRQFRPSOHWLRQRIJULQGLQJRQHÁLQWGDJJHUDWRWDORIFPLQ WKLFNQHVVZDVORVWIURPWKHJULQGVWRQH7KHÀQDOZHLJKWRIWKHJULQGVWRQHLV approximately 8 kilograms. $PRGHUDWHO\VL]HGÀQDOSUHIRUP )LJEGDJJHU ZDVPDGHRIKHDW WUHDWHG7H[DVÁLQW7H[DVÁLQWZDVKHDWWUHDWHGWRDOLWKLFJUDGHRI &DO lahan 1979 & 2000:16). The length of the preform was 31.2 cm, greatest width of 4.5 cm, and greatest thickness of 1cm. The preform required a total of 28,000 grinding strokes, equaling the distance of 28,000 meters or 28.0 kilometers over the grindstone. An average of 1.4 mm in thickness was lost IURPJULQGLQJ 7DE 1RWH7DEOHVDQGDUHSURYLGHGIRUFRPSDUD tive analysis. Summary 'DWDSURYLGHGLQWKLVSDSHULQGLFDWHJULQGLQJÁLQWLVODERULQWHQVLYHFRP pared to obsidian. The obsidian dagger preforms ground with quartz sand developed deep striations in the dagger. Additionally, the biface was pitted and very course in appearance. Grinding the face to a smooth surface on bare sandstone (i.e., no quartz sand added) alleviates major striations and pit WLQJWKDWPD\GHIDXOWH[WHQGHGÁDNHWUDYHO([WUDJULQGLQJRIDSSUR[LPDWHO\ VWURNHVZLWKRXWVDQGJDYHWKHREVLGLDQDVHPLSROLVKHGDSSHDUDQFH 106 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking 7KHÁLQWGDJJHUSUHIRUPVWKDWZHUHJURXQGZLWKTXDUW]VDQGKDGVRPH pitting and crushing, but striations did not cut as deep, and took on a more SROLVKHGDSSHDUDQFH+RZHYHUWKHÀQDOJULQGLQJFOHDQXSRIWKHGHHSVWULD WLRQVRQWKHÁLQWSUHIRUPZDVQRWQHFHVVDU\DQGLWÁDNHGTXLWHZHOOZLWKRXW extra grinding. 8VLQJWKHTXDUW]VDQGSDUWLFOHVGHÀQLWHO\GHFUHDVHVWKHORQJHYLW\RIWKH JULQGVWRQH7ZRÁLQWGDJJHUVFRXOGEHJURXQGRQWKHODUJHJULQGVWRQHZLWK out using the quartz sand. However, sandstones are abundant in the south eastern Utah region, so quartz sand was used to expedite the grinding proc ess. If the archaeologist assumes grinding occurred in one location over a con tinuous period throughout Neolithic manufacturing sites, it may be possible to isolate horizontal stratigraphy of grinding episodes. An indication will be debris that was created from slurry with sand particles from disintegrating grindstones. A broader stratigraphy will be present if an extra grinding agent VXFKDVTXDUW]VDQG ZDVDGGHG7KHDXWKRU·VGHEULVSLOH XQGHUWKHWDEOH IURPJULQGLQJÀYHGDJJHUVZDVFPORQJFPZLGHDQGFPLQWKH deepest portion of the debris pile. $OWKRXJKÁLQWLVODERULQWHQVLYHWRZRUNZLWKLWSURYLGHVDEHWWHUPHGLXP IRUDFFXUDWHUHSOLFDWLRQ7KHDXWKRUKDVIRXQGWKH7H[DVÁLQWXVHGLQWKLV VWXG\PRVWFORVHO\UHVHPEOHV'DQLVKÁLQW'DJJHUTXDOLW\'DQLVKÁLQWLVQRW UHDGLO\DYDLODEOHWRWKLVDXWKRUKRZHYHUDVSDOORI'DQLVKÁLQWIURP)DOVWHU Denmark was acquired (complements of Errett Callahan) for a comparison DQDO\VLVZLWKWKHKHDWWUHDWHG7H[DVÁLQW$PLQLDWXUH7\SH&GDJJHUUHSOLFD ZDVPDGHXVLQJWKH'DQLVKÁLQW$OWKRXJKGDWDZDVQRWFROOHFWHGWKH'DQ LVKÁLQWLVDVXSHULRUNQDSSLQJVWRQH Physical disadvantages to grinding $IWHUKDQGJULQGLQJÀYHGDJJHUVRYHUDWKUHHZHHNSHULRGWKHDFFXPXODW LQJHIIHFWVHVVHQWLDOO\OHIWWKHDXWKRU·VOHIWKDQGXVHOHVVIRUVL[ZHHNV/RVVRI dexterity, severe pain, and weakness in the metacarpophalangeal joints, (MP MRLQWV DVVRFLDWHGZLWKWKHPLGGOHDQGLQGH[ÀQJHURIWKHOHIWKDQGUHVXOWHG in barely being able to lift a glass of water. Four months later the author was still far from healing. Damage was mainly caused from grasping daggers ZLWKWKHÀQJHUWLSVZKLOHJULQGLQJ$VDUHVXOWGDJJHUVVL[WKURXJKHLJKWZHUH JURXQGFRPSOHWHPHFKDQLFDOO\DQGE\KDQGIRUÀQLVKLQJWRXFKHV &XUUHQW WKRXJKW VXJJHVWV GDJJHU SURGXFWLRQ ZDV FRQÀJXUHG DV DQ DS prentice system (Apel 2001a:45). Therefore it is possible to speculate that master knappers did not grind daggers, and that grinding was assigned to 107 Greg R. Nunn apprentices. In addition, the number of apprentices and rates of dagger pro duction are not known. )OLQW GDJJHU QXPEHU ÀYH WRRN KRXUV WR JULQG ,I WKH DXWKRU ZHUH WRJULQGWZRODUJHÁLQWGDJJHUVSHUZHHNRYHUDEULHIWLPHKHZRXOGOLNHO\ become seriously or permanently disabled. It is natural to speculate whether another method of grinding was used. Rather than on a slab by hand, could there have been another apparatus or semi – mechanical process? Could there have been a wheel involved? By observing grinding striations on the original Type 1C Danish Daggers, quantitative data may lead one in an ap propriate direction for further research on this subject. 3K\VLFDOGLVDGYDQWDJHVRISUHVVXUHÁDNLQJ $OWKRXJKWKHOHIWLQGH[ÀQJHULVKLJKO\SDGGHGLQWHQVHSUHVVXUHIURPWKH ÁDNLQJSURFHVVUHVXOWVLQVHYHUHQHUYHGDPDJHWRWKHGLVWDOSRUWLRQRIWKH LQGH[ ÀQJHU $IWHU PDNLQJ VHYHUDO GDJJHUV WKH NQDSSHU ZLOO H[SHULHQFH D FRPSOHWHORVVRIIHHOLQJLQWKHWLSRIWKHLQGH[ÀQJHUIRUVHYHUDOZHHNV([ perimentation in knapping holding positions is indicated in order to alleviate this side effect. Conclusion 7KH DXWKRU H[SODLQHG LQ GHWDLO WKH UHSOLFDWLRQ SURFHVV RI SDUDOOHO HGJHWR HGJHSUHVVXUHÁDNLQJXVLQJWKH-XWODQG7\SH&1HROLWKLF'DQLVK'DJJHU as a model. Included is detailed information on stages four through eight, ending with the completion of the replication process. Empirical data is the basis for the interpretation of the replication process (Fig. 23). Although replication is possible given current knowledge, further research is indicated to consider other possible techniques. For example, did they have a wheel for grinding? What other holding techniques could have been XVHGIRUSDUDOOHOHGJHWRHGJHÁDNLQJ"&RQWLQXHGVWXGLHVDQGH[SHULPHQWD tion will answer these and other questions, bringing professional knappers and archaeologists closer to a complete understanding of the manufacturing mysteries contained in these exquisite daggers. 108 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Figure 23. Group photo of all completed daggers excluding dagger number three; photographed by Jody Bierschied. Addendum After completing this research project, the author had the opportunity to analyze the original Type 1C Daggers at the Danish National Museum in &RSHQKDJHQ'HQPDUN7KHDQDO\VLVUHYHDOHGVLJQLÀFDQWLQVLJKWVLQWRWKH grinding aspects of the grinding stage. The grinding stage is basic to techno logical aspects of dagger production, and vital for completing a prestigious Type 1C dagger. The original ground preforms show dominate vertical grinding striations ÀJ WKDWDUHSHUSHQGLFXODUZLWKQRUWKDQGVRXWKSROHVRIWKHJURXQGSUH IRUP7KHDQJOHVRIWKHVWULDWLRQVYDU\IURPWRGHJUHHV ÀJ ,QDG dition, the lateral margins display considerable grinding damage due to this WHFKQLTXH ÀJ PRUHVRWKDQWKHKRUL]RQWDOJULQGLQJWHFKQLTXH )LJ 109 Greg R. Nunn Figure 25. Close-up of original Type 1C ground preform showing grinding striations. (National Museum of Denmark, Copenhagen, Museum number A 27691); photographed by John Lee. Figure 24. Original Type 1C ground preform. (National Museum of Denmark, Copenhagen, Museum number A 27691); photographed by John Lee. Figure 26. Close-up of original Type 1C ground preform showing grinding damage along the lateral margin. (National Museum of Denmark, Copenhagen, Museum number A 27691); photographed by John Lee. To begin accurate replication of original grinding evidence, one must address how to hold the dagger during this stage. As previously indicated, JUDVSLQJWKHWKLQSUHIRUPZLWKWKHÀQJHUWLSVFDXVHVGDPDJHWRWKHKDQGV $KROGLQJGHYLVHWKDWZLOOÀWZLWKLQWKH1HROLWKLFWLPHSHULRGZDVFUHDWHG (Fig. 27a and b). This device encompasses a two part system (Fig. 28a and b). Part A consists of a handle which is comprised of a straight pine branch, 25 cm long by 6 cm in diameter. One quarter of the branch is split down the en WLUHOHQJWKZKLFKFUHDWHVDFRQWLQXRXVVHPLÁDWVXUIDFH7KHVXUIDFHLVWKHQ VDQGHGÁDWWRUHFHLYHSDUW% UHFHLYHU 7KHUHFHLYHULVFRPSULVHGRIDOHQJWK of the heart section of a straight pine branch, 29 cm long by 1.5 cm thick and approximately 3 cm wide, which has been sanded smooth on both faces. 110 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking The receiver must be narrower than the width of the preform and somewhat resembles the preform in outline. Figure 27(a). Front view of holding devise, (b) side view of holding device; illustrated by Joe Pachak. Figure 28(a) and (b). Side view of two part handle system; illustrated by Joe Pachak. Two holes are drilled and matched into both the handle and receiver (Fig. 28a and b) .75 cm in depth. Next, two wooden posts (Fig. 28c) 1.5 cm long DUHLQVHUWHGZLWKDWLJKWÀWLQWRWKHGULOOKROHVRIWKHKDQGOH Hot bees wax or pine pitch is applied to the face of the handle and one face of the receiver. While the glue is still hot, both the handle and the receiver are joined together with the handle post inserted into the receiver KROHV8VLQJWKHSRVWDQGRUJDQLFJOXHLVLQVXUDQFHIRUDVHFXUHÀWDQGHOLPL nates any movement. 111 Greg R. Nunn After the handle and receiver are secure, the organic glue is applied to the VHFRQGIDFHRIWKHUHFHLYHU,QWKHPHDQWLPHWKHÀQDOSUHIRUPKDVEHHQSUH ZDUPHG DSSUR[LPDWHO\GHJUHHVFHQWLJUDGH QH[WWRDFDPSÀUHRULQDQ oven. While the preform is still warm, it is applied to the second face of the receiver with the coating of organic glue. The face of the warm preform will then melt into the glue coating. At this time there will be voids between the SUHIRUPDQGWKHUHFHLYHU$OOYRLGVZLOOWKHQEHÀOOHGZLWKDÀQDODSSOLFDWLRQ of the organic glue and then set aside to cool and harden. The cooling takes about 10 minutes. After cooling, the preform is ready to grind. The holding device (Fig. 29) works well and alleviates physical disad vantages to the hands. In addition, grinding time is slightly decreased. The DXWKRU·VVWDJHVL[JURXQGSUHIRUPVPLPLFWKHRULJLQDOVWDJHVL[JURXQGSUH forms in all attributes. Figure 29. Holding devise in use; illustrated by Joe Pachak. It is probable, most if not all original Neolithic Type 1C ground preforms de veloped grinding damage along both lateral margins (Fig. 26). This damage was to be expected, and therefore would destroy the time consuming oppos LQJEHYHOHGODWHUDOPDUJLQVWKDWWKHDXWKRUSUHSDUHGHDUOLHULQVWDJHÀYH )LJ 14). Therefore it is not necessary to prepare the opposing beveled platforms DORQJERWKODWHUDOPDUJLQVLQVWDJHÀYH5DWKHULWLVDWWKHEHJLQQLQJRIVWDJH seven when the edge damage is eliminated by the removal of small pressure ÁDNHV DSSUR[LPDWHO\WZRWRWKUHHPPORQJ IURPERWKOHIWKDQGODWHUDOPDU gins. Concurrently, two continuous opposing beveled platforms are created using pressure retouch (Fig. 30). 112 Neolithic Danish Dagger as a model to replicate parallel, edge-to-edge pressure ﬂaking Figure 30. Stage seven ground preform with opposing beveled platIRUPVFUHDWHGZLWKSUHVVXUHÁDNLQJ illustrated by Greg Nunn and Joe Pachak. The insights regarding grinding discrepancies and opposing beveled plat forms gained from analyzing the original daggers brought important new information into the production processes in stages six and seven. Acknowledgements ,·GOLNHWRWKDQN.HLWKDQG-DFNLH0RQWJRPHU\RI0RQWJRPHU\$UFKDHRORJL cal Consultants (MOAC), for their lab equipment and initial proof reading of the manuscript. In addition, Jan Apel, Errett Callahan, Jackie Montgomery, Jody Patterson (MOAC), and Dan Stueber provided technical assistance. 7KDQNVWR-RH3DFKDNDQG(UUHWW&DOODKDQIRULOOXVWUDWLQJPDQ\RIWKHÀJ ures. Jody Bierschied supplied digital photos. Theresa Breznau and Jenny Carlson provided computer imaging and scanning assistance regarding the ÀJXUHV6SHFLDOWKDQNVJRWR3HWHU9DQJ3HWHUVHQRIWKH1DWLRQDO0XVHXP RI 'HQPDUN &RSHQKDJHQ IRU DFFHVV WR WKH PXVHXP·V GDJJHU FROOHFWLRQ Furthermore, I would like to acknowledge Errett Callahan for providing illustrations and photos of original artifacts, providing a good deal of in formation which formed the basis of the manuscript, and proof reading the manuscript near its completion. Most of all, a heart felt thanks goes to my beloved wife, Karen Clark for her editing and typing skills, and her absolute support throughout the project. 113 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Errett Callahan Neolithic Danish Daggers: an experimental peek Introduction This report is a visual summary of 25 years of experimental research into the production of prestigious, Type IV Neolithic Danish daggers. Though only ÀJXUHVDUHVKRZQKHUHWKHRQJRLQJUHVHDUFKKDVHQWDLOHGWKHH[DPLQD tion of many hundreds of original daggers, the detailed analysis of 49 origi nals, the production, to date, of 242 replicas, and the detailed analysis of 88 of these examples. This experimental approach was done so as to devise a hypothetical “Production Model” which could yield accurate replicas and so explore some means by which the originals may have been produced. In conjunction with the present report, a broader study has been imple mented which entails examination of the full technology hypothesized to have been involved in dagger production in the past and a detailed analysis of the resultant debitage, executed by Jan Apel. This research is intended to lead the way to a prediction of the archaeological evidence that should be found at a Type IV production site, none of which has yet been located. It is RXUFRQWHQWLRQWKDWXQOHVVWKHUHSOLFDSHUIRUPVDQGÀQLVKHGGDJJHUVPDWFK WKH RULJLQDO GDWD DQG TXDOLW\QRI WKH RULJLQDOV DQG VR UHÁHFW WKH DUFKDHR logical reality, then the experimental production debitage has little meaning. Thus this study required decades of skill improvement before the present conclusion could be presented (Apel 2001a). 7KLV UHSRUW YLVXDOO\ GHSLFWV WKH DSSHDUDQFH RI VRPH ÀQLVKHG RULJLQDOV VRPHNQRZQDUFKDHRORJLFDOSHUIRUPVDQGVRPHUHSOLFDSHUIRUPVDQGÀQ ished daggers, stage by stage. Emphasis is given to several procedural mod els for realizing the complex transition between Stages 4 and 5 (Secondary 3UHIRUPWR)LQDO3UHIRUP 7KHÀQDOVWDJHVRISUHFLVLRQSUHVVXUHÁDNLQJDQG edge treatment of blade and handle are also presented. 115 Errett Callahan Originals and practice pieces Figure 1. Photo of dagger replica #120, by author, 1992. The author´s best, full-sized, Hindsgavl replication. ObsiGLDQFP$OVRVHHÀJXUH 64. Peter Kelterborn collection (study #31). Figure 2. The Basic Six dagger types. D IV, from Alnarp, Sweden, dubbed here as “elegance”, is considered by the author as the third best Type IV dagger known, after the Hindsgavl and Skatelöv. It is with Type IV that Neolithic knappers sterted copying bronze daggers (from Forssander 1936). Note orientation. (see Figure 60) 116 Neolithic Danish Daggers: an experimental peek Figure 3. Replica of Type I-C dagger mande by author in 1994, featuring edge to edge pressure ÁDNLQJIURPULJKWVLGHVDQGFKDUDFteristic retouch from left sides, as needed. Such replication of other dagger types provided critical practice for the author for Type IV work (See Nunn this volume & Callahan 2001). (Dagger #103; 32,1 cm). Figure 4. Close up of another Type I-C replica made by the author, showing detail RIXQLGLUHFWLRQDOHGJHWRHGJHÁDNLQJ1RW XQWLOWKLVWHFKQLTXHZDVXQGHUÀUPFRQWURO did author feel he was ready for the much PRUHGLIÀFXOW7\SH,9ÁDNLQJ1DWXUDOO\ such was not the case in the past due to the time gap between the types. (Dagger #122). Figure 5. Sketch of the six Type IV Sub-Types. Our concern was with Types C, D, and E. (From Lomborg 1973). 117 Errett Callahan Figure 6. Three examples of what the author considers are “ordinary”, as opposed to “prestigious”, Type IV daggers. Most type IV daggers are under 20 cm in length (see Apel 2001) and of limited surface control, while virtually all prestigious daggers are over 20 cm and with well-controlled surfaces. These daggers are, from left to right: 19,3; 18,9 and 18 cm. (DNM #A272, Z474, and A10169). Figure 8. The famous Hindsgavl dagger in the Danish National Museum, Copenhagen. Designated as a National Treasure. (DNM#A33093). Photo: The Danish National Museum, Copenhagen. 118 Figure 7. A bronze dagger of the type imitated by the Type IV and 9ÀVKWDLOV7KHPRGHUQVWLWFKHG leather handle wrap shown here is one hypothetical explanation of VWLWFKLQJRQÁLQWRULJLQDOV7KLVLV a modern casting in bronze of an original, found in England, made by Simon Fearnham of England, in the collection of the author (28,0 cm). Neolithic Danish Daggers: an experimental peek Figure 9. The infamous Skatelöv dagger in the State Historic Museum, Stockholm, Sweden. It exceeds the Hindsgavl (at 29,6 cm) by 6,5 cm, totalling 36,1 cm and is the longest Type IV known. It is also wider and thinner. Both were probably made by the same knapper. For undetermined reasons, this national Swedish treasure has been removed from public display since 1989. The author strongly urges that it be put back in the limelight so that Swedes may enjoy their National Treasure, which is no less elegant than Denmark’s Hindsgavl dagger. (SHM #12750) (S-1). Photo: The State Historic Museum, Stockholm. Figure 10. The Köinge dagger from halland, Sweden. In outline and in length-width ratios of blade and handle, this dagger is almost exactly the shape of the “average” prestigious dagger, as analysed in this study. It is thus more typical than the Hindsgavl in outline. It seems identical to the Skatelöv in style of surface treatment. (SHM #8325:38). (S-1) 31,2 cm. 119 Errett Callahan Original Stages Figure 11. Geological example of an unPRGLÀHG6WDJH%ODQN7KLVLVDQRGXODU SLHFHRIÁLQWLGHDOIRUGDJJHUSURGXFWLRQ 30,6 cm. Figure 12. Original Stage 2 Rough Out preform such as might be used for any type of dagger. This piece is knapped on outer zones and incompletely worked. The wide base here is especially suitable IRUÀVKWDLOW\SHV 0DOP|0XVHXP00 16), 18,5 cm. Figure 13. Original Stage 3 Primary Preform such as would be ideal for Type IV or V daggers. Note handaxe-like FRQÀJXUDWLRQSDUDOOHOKDQGOHRXWOLQH ÁLNLQJWRPLGGOH]RQHRQEODGHDQG lack of notable thickness distinction between blade and handle. This preform has been on display as a preform at the Danish National Museum opposite the Hindsgavl for years. (DNM #A12294) 25,9 cm. 120 Neolithic Danish Daggers: an experimental peek Figure 14. Original Stage 4 Secondary Preform such as would be ideal for Type IV or V daggers. Note that, Whereas the outline may be little changed from Stage 3, the thinness of the blade and thickness of handle distinguishes it from that stage. (See Figs 28 & 30 for examples of this. LHM #25582). )LJXUH2ULJLQDOQHDUO\ÀQLVKHG6WDJH 5 Final Preform of modest size and width. Suitable for Type IV or V. (It is the author’s contention that both types were probably made at the same workshops, the Vs being the “economy” model.) (LHM #24583). Figure 16. Original Stage 5 Final Preform ideal for a prestigious Type IV D or E dagger. (DNM #27966) 23,3 cm. 121 Errett Callahan Experimental setup, debitage, and tools )LJXUH9LHZRIÁLQWNQDSSLQJDUHQD by the Old Historic Smithy at the Lejre Experimental Center, Denmark 2004. Forge has now been converted LQWRDVWRUDJHRIÀFHVSDFHIRUYLVLWLQJ ÁLQWNQDSSLQJVFKRODUV)HQFHGLQDUHD is where the author, shown here with legendary Danish knapper Søren Moses and family, knapped in recent years. The author has conducted dagger knapping tests at Lejre numerous times since 1978. Ljre funded the author’s recent visit there in 1993 and 2003–2005, as public observed. The Pottery beyond houses researchers, as well as exhibits and pottery demonstration areas. Figure 18. Layout of daggers and preforms produced by the author and the Dagger Research Team during the workshop at the author’s home, 2002. Figure 19. Layout of dagger production debitage, organized by stage, as produced and analysed by the Dagger Research Team, as directed by jan Apel, at the author’s home, 2002. 122 Neolithic Danish Daggers: an experimental peek Figure 21. Pressure tools used by the author for dagger production in this study. Ishi stick to left has antler tip insert. OthFigure 20. Hammerstones, antler billets, ers have copper tips, which the author and abraders used by the author for dagger IHHOVVLPSOLÀHVWKHZRUNWHQIROG&RSSHU production in this study. was available during the Late Neolithic period in Scandinavia. Figure 22. The three forms of Danish SenoQLDQÁLQWEODQNV 6WDJH VXLWDEOHIRUGDJger production. From left to right: nodule, block core, and spall. Figure 23. Ideal Stage 2 Rough Out replica as produced from a thin blank. Emphasis here is upon FUHDWLRQRIRXWHU]RQHÁDNHVZLWK a relatively centered edge. Shape is irrelevant at this stage. (Dagger 228, #82). Figure 24. Core Blank produced from large block of irregularly-shaped material. This is now ready for Stage 2 reduction. (w/o #). 123 Errett Callahan Figure 25. Ideal, extra-wide, Stage 3 Primary Preform replica, as produced by author LQRI/ROODQGÁLQWGRQDWHGE\0DVWHU Danish knapper, Thorbjørn Petersen. (#05EC36) 28,8 cm. Figure 26. Sketch of typical Stage 3 3ULPDU\3UHIRUPUHSOLFDLQÁLQWVKRZing cross-sections. Note handaxeOLNHSURÀOHDQGWKLFNQHVVEXWZLWK parallel-sided handle area. This is the ÀUVWVWDJHLQZKLFKDFHUWDLQVKDSHLV required. Up to this stage, the preforms for Types IV and V are identical. Differences become evident in the forthcoming stages (though lastminute switching from IV to V is possible). 27,2 cm. (Compare with Fig. 13) Dagger 154 (#50), 99% soft hammer percussion. Figure 27. Near-ideal Stege 4 Secondary Preform UHSOLFDSURGXFHGLQFRDUVH'DQLVKÁLQW 6WHYQV This is the stage where direct percussion is pushed to the limit, prior to the onset of punching of handle margins. On some pieces of thick material, punching may enter earlier. Dagger 129. (#35). 124 Neolithic Danish Daggers: an experimental peek Figure 28. Side view of same preform showing how blade is knapped thin while handle is left thick. Also see #60 (#35). Ignore clay lunp on base. Figure 29. Sketch of ideal Stage 4 Secondary Preform replica, showing cross-sections and apSHDUDQFHRIODVWGLUHFWSHUFXVVLRQÁDNHVRQKDQdle (handle is a bit thin). Compare with Figure 14, dagger 178. (#58) 26,4 cm. )LJXUH3KRWRDWREOLTXHDQJOHRILGHDO6WDJH6HFRQGDU\3UHIRUPRI/ROODQGÁLQW vividly depicting contrast between thinness of blade and thickness of handle. Note that opposite things are happening with blade and handle. The blade is made to lose thickness while keeping width, while the handle is made to keep thickness while losing width, with a characteristic abrupt transition between the two. See Figures 34, 60 and 61. Dagger 240. (#88) 28,2 cm. 125 Errett Callahan Replication Stages 4-5 Figure 31. First of series of four sketches Figure 32. Same dagger showing second showing transition of Stages 4 to 5 by reVHWRIQLQHSXQFKHGÁDNHUHPRYDOV peated punching of lateral margins (as sug- 25,3 cm. gested by knapper Thorbjørn Petersen). Note changing outlines and cross-sections as well as characteristic debitage. This is presented here as a likely reduction model IRUWKHSDVW6KRZQKHUHDIWHUÀUVWVHWRI ÁDNHUHPRYDOVFP&RPSDUHZLWK Fig. 15. dagger 182 (#62). Figure 33. Same dagger showing third set RISXQFKHGÁDNHUHPRYDOVFDUVDQG VHYHQFKDUDFWHULVWLFSXQFKHGÁDNHV%ODGH has also been extensively reworked by direct percussion. Compare with Fig. 16. (#62) 25,2 cm. 126 Neolithic Danish Daggers: an experimental peek Figure 34. Finished Stage 5 Final Preform UHSOLFDDIWHUÀQDOGHOLFDWHSHUFXVVLRQDQG pressure trimming of outer zones. Note characteristic loss of outline width and thickness in cross-sections (compare with Fig. 33). This model is predicted to be the most typical path of action for Stage 5 development when no preliminary seam is punched. In general, Stage 5 is the most GLIÀFXOWVWDJHWRUHSURGXFHFDOOLQJIRUD near Master level of skill. Previous stages could probably be executed by knappers of Journeymen level of skill – if under the watchful eye of a Master. (#62) 24,2 cm. Figure 35. Four original daggers. Two at left illustrate preliminary seam punching. Two at right are Type III daggers such as might result from preforms at left. Such might also result in Type IV-A or B daggers. One at left is a reject. (SHM #2549; 8970:54SK; 2548; 33:97, 1393:33). Figure 36. Rough sketch of original preIRUPVKRZQLQODVWÀJXUH7KLVFOHDUO\ shows that the handle was punched with a preliminary seam. Though possibly a Type III perform, it shows a path of action that would have been known to Type IV knappers. The author has observed that the YDVWPDMRULW\RIÀQLVKHG7\SH,9GDJJHUV bear evidence of such preliminary seam treatment (other modern dagger knappers don’t seem to realize this and so directly press the seam without prior punch work). (SHM #8970:54SK) 23,0 cm. 127 Errett Callahan )LJXUH7ZRDEHUUDQWÀQLVKHG7\SH,9 daggers showing well-controlled punching of all four handle seams. Though probably local, “inexpensive” variants imitating the classic model, they illustrate the appearance of preliminary punched seams. The author doubts if the classic daggers were VRSXFKHGRQÀQDOVHULHVRIODWHUDOPDUJLQV ZKHUHSUHVVXUHÀQLVKLQJSUHGRPLQDWHV 1RWHWKDWWKHVHSXQFKHGÁDNHVZKLFK were removed from left to right, are backwards oriented, as are the pressure seams. (DNM w/o #s). )LJXUH6DPHGDJJHUVKRZLQJÀUVWVHW RIODWHUDOSXQFKÁDNHUHPRYDOVDQGFKDUacteristic debitage. Such alternating between top seam and laterial sides is the predicted path of action. (#56) 25,8 cm. Figure 38. First of a series of three sketches showing repeated punching of preliminary top seam during Stage 4-5 transition. Note characteristic debitage and crosssection evolution. Also note how bulbar dips below the ridge serve to catch laterial SXQFKHGÁDNHVLQWKHQH[WVNHWFKDQGWR DOORZIRUWKFRPLQJSUHVVXUHVHDPÁDNHVWR dig into prior bulbar concavities. The author thinks this is a likely reduction model for explaining the archaeological evidence seen on the original daggers and preforms. Daggers 170. (#56) 25,8 cm. 128 Neolithic Danish Daggers: an experimental peek Figure 40. Same dagger showing how both punching and pressure may be used to regulate the preliminary seam. The author hypothesizes that the preliminary seam will ODWHUEHH[DFWO\VWUDLJKWHQHGEHIRUHWKHÀ- Figure 41. First series of three photos nal pressed seam is applied. (#56) 25,2 cm. showing how the author punched the seam of dagger replica made in 1989. Preform was cushioned with a bed of clay as dumbhead of shaving horse, pressed downward with foot, held preform in place. The author later developed a much simpler press using two boards and a rope. View from front. (#23). Figure 42. Same perform viewed from left. Puch is copper-tipped antler. (#23). )LJXUH6DPHSUHIRUPÀQLVKHG7KLVLV the Final Preform Stage, Stage 5. Preliminary seam punching met the author’s expectations of the original model. (#23). )LJXUH)LQDO3UHIRUPRIGDJJHUPDGHLQRI/ROODQGÁLQWZLWKSUHliminary puched seam successfully executed. Note the two stitching practice pieces predicted to be found at a Type IV dagger production site. One doesn’t let beginners SUDFWLFHRQQHDUO\ÀQLVKHGGDJJHUVPDGHE\WKH0DVWHU7KH\PXVWKDYHSUDFWLFHGRQ scraps. (#88) 27,9 cm. 129 Errett Callahan Replication Stages 6-8, replicas and originals Figure 45. Plastic cast (by Bostrom) of original Type IV-D/E dagger, reversed face, showing clear evidence of grinding on blade. A resharpened, ordinary dagger. Handles of daggers were never ground (but handle edges were always ground quite smooth). Author has observed that about 50% of the Type IV prestigious daggers were ground on the blade. Grinding would allow more time to be expended and thus, perhaps, creater a higher value. Grinding also facilitates a higher quality surface beauty. Figure 46. Replica of a Stage 6 Ground Preform after 3000 strokes of hand grinding on sandstone slab. Note direction of grinding striations, which match originals. See Figs. 48 and 49 for holding positions (see Fig. 49 caption for number of strokes required of ÁLQW 'DFLWHREVLGLDQLVVKRZQKHUH'DJJHU 206. (#66) 25,3 cm. Figure 47. Sketch of Ground Preform #182. See earlier stage of this dagger in Fig. 34. Note that two stages are evident on the blade. Such evidence on the originals allowed the author to reconstruct the missing stages years before he discovered archaeological samples (i.e. Figs. 12-16). (#62) 24,2 cm. 130 Neolithic Danish Daggers: an experimental peek Figure 48. Author grinding dagger replica #240 at Lejre Experimental Center, Denmark 2004. Figure 50. Dagger replica #170 during stage 7 Flaked Implement reduction. Note the nature of straight-in parallel pressure ÁDNHV ´ÁDWÁDNLQJµ RQUHYHUVHG left side. Such procedures, moving from left to right along either edge, is characteristic of Type IV daggers. Three stages are now evident – Stage 5 on right side of handle, Stage 6 on right side of blade, and Stage 7 on left sides of blade an handle. (#56) 24,3 cm. Figure 49. Close-up of the author grinding dagger #240 on the side of an old farmer’s sharpening stone. It took 10 000 strokes on each side to grind the blade about 80%. By comparing this with the data in Fig. 46, it is HYLGHQWWKDWÁLQWLVmuch harder than obsidian. This underscores the importance of ÀQDOL]LQJRQH·VWHVWXVLQJRULJLQDOPDWHULDOV though non-original materials may be useful for pocedural practices. Note the use of ÁLQWVDQGLQJULQGLQJWKLVGDJJHU7KLVVDYHG QRWLPHZKDWVRHYHUZLWKJULQGLQJÁLQW though it helped noticeably with obsidian. (#88). Figure 51. Same dagger DVÁDNHGRQULJKWVLGH The Flake Removal Sequence (Callahan in this volume) shown in WKHVHWZRÀJXUHVZDV practiced quite consitently by Neolithic knappers, and hints at strict regimentation of production practices by some superior authority. (The above procedure is missed by most modern wouldbe dagger knappers). (#56) 24,3 cm. 131 Errett Callahan Figure 52. Blade of dacite obsidian dagJHUDIWHUSDUDOOHOSUHVVXUHÁDNLQJ of blade using an antler-tipped Ishi stick (see this tool in Fig. 21, far left). CorUHFW´&µÁDNLQJUHVXOWHGHVSHFLDOO\RQ right margin (Kelterborn 1984, which illustrates reverse´&µÁDNLQJ (GJH retouch has yet to be done. Stage 7 Flaked Implement. (#66). Figure 53. Blade of dagger replica #181, showing retouched margin (Stage 8) with GHOWDÁDNLQJRQOHIWVLGHDQGXQUHWRXFKHG margin (Stage 7) on right side. Clearly, “retouching” (as used in the Western VHQVHRIÀQHHGJHDOLJQPHQWÁDNLQJRQWKH outer zone cutting edge rather than in the Continental sense of referring to the long SDUDOOHOÁDNHVLQWRWKHPLGGOH]RQH LVDQ important production stage needed before the edge becomes “functional” (note also the remnant of ground surface at the bottom center of the blade. Such grinding facets did not prevent the circulation of the Neolithic daggers for such evidence is common archaeologically.) (#61). Figure 54. Blade of dagger repica #182, showing charDFWHULVWLFÁDNHWHUPLQDWLRQVLQWKHFHQWHURIWKHEODGH VKDGHG FDXVHGE\ODVWVHULHVRISDUDOOHOÁDNHUHPRYDOV IURPWKHULJKWVLGH$VWKHODVWUHPRYHGÁDNHVWKH\ XQGHUFXWWKHLUSULRUÁDNHVIURPOHIWVLGHDVZLWKPRVW originals. Stage 8 Retouched/Finished Implement. (#62). 132 Neolithic Danish Daggers: an experimental peek Figure 55. Sketch of dagger replica #181, VKRZLQJ6WDJHUHWRXFKÁDNLQJRIEODGHDQG handle (shaded areas). Blade features delta retouch (Kelterborn 1984) while handle features distinctive stitched zig-zag presVXUHVHDPÁDNLQJRIULGJHODWHUDOPDUJLQV and base. Handle edges will next be heavily abraded. See Fig. 7 (#61), 26,7 cm. Figure 56. Dagger replica #178 showing LQLWLDWLRQRIVHDPÁDNLQJDWEXOEDUGLSDSproached from base. Note sharp median ULGJHDQGÁDWSODQHVRIODWHUDOVLGHV Figure 58. Same dagger replica with top seam, to right, half –stitched, side view. Note backward-oriented ÁDNHVFDUVZKLFK DUHPRUHGLIÀFXOW to obtain than the zig-zag zipper effect of the plan view. Also note thickness loss caused by stitching. Handles must be preformed overly-thick to allow for this loss. (The same may be said for the preceeding stages and steps). (#61). Figure 57. Dagger replica #181 with handle seam half stitched by pressure. Note absence of an earlier, punched seam and presence of sharp ridge forward of the “pinched” seam. Unlike most originals the path of action shown here shows no evidence of an earlier punched seam, as in Figs. 35-42 (most modern knappers work as shown here). (#61). 133 Errett Callahan Figure 61. Handle detail of dagger replica #120, showing ÀQLVKHGVHDPVWLWFKing on handle and base and blade-handle transition (2,64 seam ÁDNHVFP &RPSDUH this, the author’s best, with the work of a far superior Neolithic knapper in Fig. 62 (#31) (Notice that transition from blade to handle is abrupt, not gradual, as with most originals). Figure 59. Dagger replica #178 showing reverse, right lateral seam, half stitched. Note backward-oriented ÁDNHVFDUVDQGORVVRIKDQGOHZLGWK ,WLVWKHDXWKRU·VFRQFOXVLRQWKDWÀQH seams of this nature may not be done without copper, though preliminary seams might be. (#58). Figure 60. Same dagger half-stitched on left ODWHUDOVLGH$OVRQRWHWKHÀQLVKHGWRSVHDP to right, and abrupt, rather than gradual, dip where handle joins blade. (#58). 134 Figure 62. Handle detail of an original PDVWHUSLHFHVKRZLQJÀQHVWVHDP stitching author has ever seen (4,63 VHDPÁDNHVFP7KHDYHUDJHLV Hindsgavl is 2,28). This is the Alnarp dagger, aka “Elegance”, at LHM, #11874. See it in our Figure 2, (L-1). Neolithic Danish Daggers: an experimental peek Figure 63. Partial sketch of the Møn dagger. At 33,4 cm, it is the longest Type IV dagger in Denmark (it is second only to the Skateløv in Sweden). Detail of some NH\EODGHDQGÀQHVWLWFKLQJVFDUV '10 #4871), (D-11). Figure 64. Three daggers. Left is author’s #120, of obsidian, made in 1992 (#31); middle LVDXWKRU·V)LQDO3UHIRUPRI)DOVWHUÁLQWPDGHLQ ULJKWLVSODVWLFFDVW of Hindsgavl, donated to author bt Søren Moses. 135 Errett Callahan Figure 65. Outlines of Stages 1-8 of dagger replica #180. Notice dramatic loss of width and thickness in early stages and minimum loss in later stages. The nature of this loss should be evident in the debitage. (#60). Figure 66. Replica dagger #123, made in 1993, as depicted for knife show exhibitions and web site. Such sales of replicas to customers/patrons were, as in the past (?) the DXWKRU·VSULQFLSDOPHDQVRIÀQDQFLDOVXSSRUWIRUWKLVUHVHDUFKSURMHFWRYHUWKHGHFades. All are indelibly signed and numbered. In the collection of Bob Verrey. (#32). 136 Neolithic Danish Daggers: an experimental peek Notes 1. 2. 3. 4. 5. 7. 8. All sketches and photos are by Errett Callahan unless otherwise noted. Replicas are of obsidian unless noted otherwise. Daggers are depicted point upwards, as originally shown in Müller 1902, Montelius 1917, and Forssander 1936, and in keeping with current professional conventions. This is noteworthy because of designations of left and right sides, etc. Original daggers have been given names according to where found. This is usually the Parrish. Replica daggers are usually named after the patron who supported its creation, but daggers in the author’s collection are named otherwise (i.e. #88 was dubbed the 0XVWDQJDVLWJDYHPHVXFKDÀJKW )LJV 0RVWRIWKHRULJLQDOGDJJHUVZHUHVWUD\ÀQGVIURPWKHVDQGHDUO\V7KH\ were probably originally placed in the graves or hoards which were uprooted by subsequent farming activities. Original daggers in this study were assigned a short code number – D, S, or L – to VLPSOLI\LGHQWLÀFDWLRQ7KJLVQXPEHUDOVRLQGLFDWHVWKHFROOHFWLRQ7KXV' 'DQLVK National Museum in Copenhagen, Denmark, About those numbers (a search for simplicity): Each dagger replica in this study was assigned several numbers according to the set of which it was a part. For instance, Dagger #240 (the last one to be embraced by the present “Production Model”) received the following numbers: 04EC19 – (The Working Number). The 19 th stone tool of any kind made by EC (Errett Callahan) produced in 2004. (He also made other stone tools than daggers). #240 – (The Dagger Number). The 240 th dagger replica of any kind produced in this study, in chronological order. Not all were selected for analysis for they included preforms, miniatures, and daggers of other types than type IV. This is usually given at the beginning of the caption. #88 – (The Study Number). The number of the set of 88 daggers, selcted from daggers 8-240, which were analysed to create the “Production Model”. These 88 were scrutinized for over 9000 bits of information. This number is usually given at the end of the caption. Daggers made subsequent to #88 are given thair regular dagger order number, once assigned (i.e. #242, which came after #88). Usually the reader only has to deal with the dagger number and the study number. 137 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Hugo Nami Preliminary experimental observations on a particular class of bifacial lithic artifact from Misiones Province, northeastern Argentina Abstract In the south of the South American forest, in the province of Misiones in northeastern Argentina, there is a particular class of bifacial stone tool with D ERRPHUDQJOLNH IRUP FRPPRQO\ FDOOHG ´FXUYHG FOHDYHUVµ 7KH VLFNOH gloss observed on some pieces suggests that they are probably related to the processing of vegetal matter. This kind of artifacts characterized the VRFDOOHG+XPDLWiWUDGLWLRQWKDWODVWHGEHWZHHQF%3%DVHGRQ experiments, the preliminary observations of the manufacture of this arti fact are reported. Stages of manufacture, the probable techniques using for making them and other useful technical observations are considered. Introduction In northeastern Argentina, at the border of Paraguay and Brazil, there is a rich and unexplored lithic archaeological record. Among the many stone WRROV D SDUWLFXODU FODVV RI ELIDFLDOÁDNHG DUWLIDFW ZLWK D ERRPHUDQJOLNH form, commonly called “curved cleavers”, is noticeable. Until now, no de tailed typological, functional or technological investigation has focused on this artifact. Hence there are no previous studies devoted to its reduction se TXHQFHSDUWLFXODUO\WKHLGHQWLÀFDWLRQRIHDUO\VWDJHVSUHIRUPVDQGÀQLVKHG products. In order to create a realistic baseline that allows us to understand some aspects of the reduction sequence of this particular artifact, I here report the preliminary experimental observations considering stages of manufacture, the probable techniques used for making the artifact and other useful tech nical observations. 139 Hugo Nami Archaeological background In southeastern South America in the area covered by south Brazil, north east Argentina, west Uruguay and east Paraguay, most archaeological sites without projectile points have, from a normative perspective, been enclosed in the so called Humaitá tradition which is divided into eighteen phases. 5DGLRFDUERQGDWHVLQGLFDWHWKDWWKLVWUDGLWLRQODVWHGEHWZHHQ%3 (Schmitz 1987: Tab. 7). In temperate zones with dense vegetation and in areas with tropical and subtropical forest, the typical lithic remains characterizing the “Caaguaçu” phase of the Humaitá tradition are rough bifaces. They are called “curved cleavers”, “cleavers” and “picks”, choppers, chopping tools, scrapers and knives, and are mostly made of chunks and tabular nodules of local volcanic rocks with red and brown tonalities. There are also ground stone tools, such as axes, bolas and metates (Rodriguez 1992:183 p). Curved cleavers are also known as a fossil type of the Altoparanaense industry (Menghin 1955/56). Since the beginning of 1990s I have carried out several research activities LQ3XHUWR(VSHUDQ]DQRUWKZHVWRIWKH0LVLRQHVSURYLQFHLQWKH5HSXEOLF of Argentina. The area is characterized by an abundance of archaeological sites, which are being impacted by several human and natural processes: the construction of houses, streets and roads and the erosive action by active watercourses, mainly the Paraná River and its tributaries, respectively. Primary and secondary sources of volcanic rocks are present in this area. Quarry workshops sites are widespread along the Paraná River with an abundance of workshop activities. There are thousands of lithic artifacts and among them it was possible to see that prehistoric human populations used EODGHDQGPLFUREODGHWHFKQRORJLHVDVZHOODV/HYDOORLVOLNHFRUHVWRREWDLQ SUHGHWHUPLQHGÁDNHV 1DPLa). Figure 1. Archaeological specimens of curved cleavers. 140 Preliminary experimental observations on a particular class of bifacial lithic artifact Figure 2. Three views of an archaeological specimen of a cuved cleaver. Scale is the same DVLQWKHSUHYLRXVÀJXUH From a technological point of view, the regional lithic technology is very LQWHUHVWLQJ2QHRIWKHPRUHUHPDUNDEOHÁDNHGDUWLIDFWVLVDSDUWLFXODUDQG XQXVXDO FODVV RI ELIDFLDO VWRQH WRRO ZLWK ERRPHUDQJOLNH IRUP FRPPRQO\ called “curved cleavers” (Figs. 1 and 2). They were made of local red/brown volcanic rocks and have diverse sizes ranging from 8 to 25 cm in length and 3 to 5 cm in thickness. Of the few unbroken archaeological pieces observed, the distal part of this particular artifact is pointed and the lower part has di verse forms; some of them have probably been formed to aid prehension. In JHQHUDOWUDQVYHUVDOFURVVVHFWLRQVYDU\EXWPRVWRIWKHPDUHELFRQYH[WUL DQJXODURUUKRPERLGDO/RQJLWXGLQDOFURVVVHFWLRQVDUHYHU\LUUHJXODU7KH ÁDNHVFDUVDUHDOVRYHU\LUUHJXODUZLWKVWHSDQGKLQJHIUDFWXUHVVXJJHVWLQJ that some sort of percussion was normally used in their manufacture. How HYHUWKHUHDUHVSHFLPHQVZLWKDPRUHHODERUDWHÁDNLQJSDWWHUQDQGVRPH RI WKHP ² HVSHFLDOO\ WKRVH ZLWK UKRPERLGDO FURVVVHFWLRQV ² KDYH VTXDUH DQGVLQXRXVHGJHV7KLVW\SHRISLHFHVRPHWLPHVVKRZVQDUURZÁDNHVFDUV VXJJHVWLQJ WKH XVH RI SUHVVXUH LQ WKH ÀQDO VWHSV RI PDQXIDFWXUH ,W VHHPV that this technique was only used to regularize the edges by eliminating the UHPDLQLQJ EXOEDU VFDUV IURP SUHYLRXV ÁDNLQJ )XUWKHUPRUH DEUDVLRQ ZDV carried out in the latter part of the production sequence. $ IHZ FOHDUO\ ÀQLVKHG DQG XVHG SLHFHV IRXQG IDU IURP WKH TXDUU\ VLWHV display sickle gloss, suggesting that they were probably employed to process some sort of vegetal material (cf. Semenov 1964:113 pp; Kamminga 1979). Currently, some old Guayaki Indians use a similar wooden artifact to dig in search of vegetal roots (Gherardi, pers. comm. 1995). 141 Hugo Nami Experimental study 0DWHULDOÁDNLQJLPSOHPHQWVPHWKRGVDQGWHFKQLTXHV Most of the actual research for this paper was carried out during the period 1995 to 1997 and in 2003. So far, I have made a limited number of experi mental specimens (n = 30). Concerning raw material, I used different stones UDQJLQJEHWZHHQWROLWKLFJUDGHVRQ&DOODKDQ·VVFDOH &DOODKDQ belonging to several sources located in Argentina. They are as follows: red/ brown local volcanic rocks of diverse characteristics, probably basalts, rhyo lites and porphyries (Puerto Esperanza, Misiones province); unheated green chert from Piedra Parada area (Chubut province), black dacite – commonly called “basalt” – (Paso Limay, Río Negro province), volcanic tuff and basalt IURP$UUR\R6DxLFyDQG=DSDOD 1HXTXpQSURYLQFH DQGÀQDOO\LQGXVWULDO glass made in Buenos Aires. &RQFHUQLQJÁDNLQJLPSOHPHQWV,HPSOR\HGWZRVPDOOGLVFRLGKDPPHU stones of granite rock weighing 120 and 125 g respectively, and a heavier broken ovoid hard hammerstone of 650 g (Fig. 3). Figure 3. Flaking implements used during the experiment. a) Broken ovoid hammerstone, b-c) discoid hammerstones. 'XULQJWKHÁDNLQJDFWLYLW\,XVHGGLUHFWIUHHKDQGDQGDQYLOGLUHFWSHUFXV VLRQÁDNLQJKROGLQJWKHSLHFHZLWKWKHOHIWKDQGDQGWKHSDGGHGXSSHUOHJ for support. I changed varieties according to the morphology of the artifact LQWKHUHGXFWLRQVHTXHQFH$VLOOXVWUDWHGLQÀJXUH,XVHGGLYHUVHKROGLQJ SRVLWLRQVGXULQJWKHÁDNLQJDFWLYLW\DFFRUGLQJWRWKHGLIIHUHQWDFWLYLWLHVSHU formed during the manufacturing process. As showed in Figure 4k, pressure ÁDNLQJZDVDSSOLHGZLWKWKH´,VKLVWLFNµ :KLWWDNHU 142 Preliminary experimental observations on a particular class of bifacial lithic artifact Figure 4. Photographs showing some portions of an experimental reduction sequence. a) Stage ÁDNHGIURPWKHEODQNLOOXVWUDWHGLQÀJXUH b-d) Different perspectives of direct percussion ÁDNLQJXVHGEHWZHHQVWDJHVHI ODWHUDODQG frontal views of the stage 3 (photographs by María de las Mercedes Cuadrado). Figure 4 (continuation). g-h) +ROGLQJSRVLWLRQXVHGWRÁDNH the distal part of the cleave curve by hand held direct perFXVVLRQÁDNLQJ1RWHWKHXVH RIWKHÀQJHUVXQGHUWKHÁDNHV UHPRYHGLM ÁDNHGVWDJHN 3UHVVXUHÁDNLQJWHFKQLTXHXVHG to regularize the edges (photographs by María de las Mercedes Cuadrado) Suggested stages of manufacture Based on this preliminary experiment it is possible to suggest the reduction sequence involved in making a “curved cleaver”. For analytical purposes, I GLYLGHGLWLQWRÀYHVWDJHVRIPDQXIDFWXUH,PDGHWKHIROORZLQJVHJPHQWD tion by considering some morphological attributes towards the development WRWKHÀQDOSURGXFW7KHDLPLVWREXLOGDFODVVLÀFDWLRQWKDWFDQEHXVHGWR predict and retrodict similar material found in archaeological records. To FDUU\RXWWKLVWDVN&DOODKDQ·VVHPLQDOZRUNZDVYHU\KHOSIXOLQGHÀQLQJDQG describing the stages of manufacture (Callahan 1979). They are as follows: 143 Hugo Nami Stage 1. Obtaining the blank 7KLVVWDJHPLJKWHQWDLOWKHGHWDFKPHQWRIODUJHÁDNHVE\VSDOOLQJODUJHFRUHV RUWKHVHOHFWLRQRIDGHTXDWHWDEXODUQRGXOHVRUFKXQNVWREHÁDNHGLQWKH QH[WVWDJH7KHWKLFNQHVVPXVWEHWKHVDPHRUDOLWWOHWKLFNHUWKDQWKHÀQ ished product (Figs. 5 and 6). Figure 5. Tabular nodule of dacite from Paso Limay used in the experimental piece shown in the previous ÀJXUH Figure 6. Flake blanks of volcanic rocks from Puerto Esperanza (a) and Zapala (Neuquén). 144 Preliminary experimental observations on a particular class of bifacial lithic artifact Stage 2. Initial edging +HUHWKHEODQNLVÁDNHGWRJLYHDQHGJHZKHUHWKHUHLVQRQHRUZKHUHWKH HGJH LV WRR VKDUS DQG ORZDQJOHG IRU IXUWKHU ÁDNLQJ E\ XVLQJ KDQGKHOG DQGRUDQYLOSHUFXVVLRQÁDNLQJZLWKKDUGKDPPHUVWRQHVDFFRUGLQJWRWKH PRUSKRORJ\RIWKHEODQN)RUWKRVHSLHFHVVWDUWLQJIURPÁDNHVVPDOOKDP merstones are useful to eliminate the sharp edges. However, to reach this goal with a tabular nodule, it was necessary to use the heavier hammerstone. At this stage, the piece has diverse forms, ranging from rough bifaces, nu cleiform artifacts, partially curved bifaces and diverse irregular artifacts. No SODWIRUPSUHSDUDWLRQLVQHHGHGLQWKLVVWHSDQGWKHÁDNHVFDUVPD\FRYHUOHVV than half of the width of the artifact. Edges are very sinuous and irregular (Figs. 7 and 8). )LJXUH6SHFLPHQVFRUUHVSRQGLQJWRVWDJHPDGHIURPWKHÁDNHVLOOXVWUDWHG LQWKHSUHYLRXVÀJXUH 145 Hugo Nami )LJXUH6WDJHDQGVRPHZDVWHÁDkes made of green chert from Piedra Parada area. Figure 8. Stage 2 made on a tabular nodule from Arroyo Sañicó, Neuquén. 6WDJH%LIDFLDOÁDNLQJ By continuing with a smaller stone hard hammer (particularly one of a dis coid form), the previously edged piece is then shaped into a rough biface, an HORQJDWHGRYDOELIDFLDOIRUP7KXVWKHÀQDOIRUPLVRXWOLQHGDQGVKDSHG into the initial preform of the piece. Generally, the forms are curved and the ÁDNHUHPRYDOVHTXHQFHIROORZVDQLUUHJXODUSDWWHUQ,QWKLVVWDJHWKHSLHFHV KDYH GLYHUVH FURVVVHFWLRQV 7KLFN ELFRQYH[ UHFWDQJXODU DQG WULDQJXODU forms. Edges are slightly sinuous and more regular (Fig. 9). Stage 4. Initial shaping :LWKWKHDLPWRUHDFKWKHÀQDOSURGXFWDWWKLVVWHSDPRUHFDUHIXOELIDFLDO ÁDNLQJFRQWLQXHVE\HPSOR\LQJWKHVDPHVWUDWHJ\DVLQWKHSUHYLRXVVWDJH%\ continuing with a smaller stone hammer (particularly one of a discoid form), WKHSUHYLRXVO\ELIDFLDOÁDNHGSLHFHLVWUDQVIRUPHGLQWRDIRUPUHPLQLVFHQWRI WKHÀQLVKHGSURGXFW1RZWKHSLHFHLVDOPRVWVKDSHGLQWRWKHUHÀQHGSUH IRUP )LJLM +RZHYHUYDULDWLRQVLQÁDNLQJLPSOHPHQWVPLJKWEHXVHGLQ WKLVVWHSDQGWKHSRVVLELOLW\RIVRIWSHUFXVVLRQÁDNLQJPXVWEHLQYHVWLJDWHG in future research. 146 Preliminary experimental observations on a particular class of bifacial lithic artifact Stage 5. Final shaping $IWHUSHUFXVVLRQÁDNLQJWKHSURGXFWLVÀQLVKHGE\UHJXODUL]LQJWKHHGJHV E\SUHVVXUHÁDNLQJ )LJN 7KLVWHFKQLTXHZDVRQO\XVHGWRHOLPLQDWHWKH UHPDLQLQJEXOEDUVFDUVLQWKHHGJHVUHVXOWLQJIURPWKHSHUFXVVLRQÁDNLQJ Furthermore, abrasion was used in the latter part of the production. Figure 10 illustrates some pieces corresponding with this stage. Figure 10. Finished products (stage 5) made of dacite (a), red volcanic rock (b) and industrial glass (c). Figures 4 to 9 show a few experimental examples of the stages of manufac ture previously described. Some preliminary observations and results The red/brown volcanic rocks from Misiones showed different textures, hardness and brittleness. The more brittle ones were red basalts with very ÀQHJUDLQZKLFKDUHFRPSDUDEOHZLWKRWKHUYROFDQLFVWRQHVXVHGLQWKLVH[ SHULPHQWVXFKDVWKHGDFLWHIURP3DVR/LPD\7KHÁLQWOLNHPDWHULDOVWKDW, used proved to be a little harder and resistant than the volcanic rocks. Thus, the best available local raw materials for making “curved cleavers” are the YROFDQLFURFNVWKH\PLJKWEHFODVVLÀHGZLWKLQWROLWKLFJUDGHVVXFKDV VRPHÀQHJUDLQUHGEDVDOW0RUHUHVLVWDQWURFNV²VXFKDVWKHÁLQWOLNHPDWH ULDOV²SUHVHQWHGVRPHSUREOHPVGXULQJWKHÁDNLQJHVSHFLDOO\ZKHQPDNLQJ the concavity in the curve (see below). The most useful hammerstones vary in size and form, particularly related WRWKHGHYHORSPHQWRIWKHDUWLIDFWWRZDUGVWKHÀQLVKHGSURGXFW$ODUJHVL]HG hammerstone of 650 g was employed in stage 2 and the smaller ones – espe 147 Hugo Nami FLDOO\WKRVHRIGLVFRLGIRUPV²GXULQJVWDJHVWR3UHVVXUHÁDNLQJZDVXVHG RQO\GXULQJWKHÀQDOVKDSLQJVWDJH,RFFDVLRQDOO\XVHGDQDEUDGHUWRFUHDWH the platforms by forming a certain degree of dullness on the edges. Specimens resulting from the initial edging showed too much variability and some of them might be confused with different kind of cores. In the early stage of manufacture, especially during the initial edging and ELIDFLDOÁDNing, transversal fractures are common causes of unsuccessful executions (Fig. 11). However, in the ELIDFLDOÁDNLQJ, initial shaping and shaping when the specimen becomes narrow, there is a risk of excessively thick edges form ing. In this sense, one of the more problematic areas is the manufacture of the concave part of the curve, particularly during stages 3 and 4. Here the excessively thick edges impeded the continuation of the work. In this case, the piece becomes too narrow and a squared edge is formed, which is a very ULVN\RSHUDWLRQGXULQJWKHELIDFLDOÁDNLQJ FI&DOODKDQ :KHQWKH ÀQLVKHGSURGXFWKDVUKRPERLGDOFURVVVHFWLRQVZLWKVTXDUHHGJHVLQWKHÀQDO shaping, another common problem is the formation of steep and deep hinge terminations. Figure 11. Rejected stage 3 broNHQGXULQJWKHELIDFLDOÁDNLQJ There are some differences in the debitage according to the stages and main O\WKLVYDULHVLQVL]HDQGIRUP$VVKRZQLQÀJXUHVWRGHELWDJHIURP VWDJHVWRWHQGWREHSULPDU\DQGVHFRQGDU\ÁDNHVRIODUJHVL]HVWKRVH IURP VWDJHV WR DUH LQQHU DQJXODU ÁDNHV RI PHGLXP VL]H WKH GHELWDJH ÁDNHVIURPVWDJHWRDUHVLPLODUEXWVPDOOHUWKDQWKHSUHYLRXVRQHVDQG ÀQDOO\WKHSUHVVXUHÁDNHVDUHVPDOOHUDQGUHVHPEOHWKHRQHVXVXDOO\UHVXOW ing from this technique. However, some overlapping might exist among the debitage between stages. 148 Preliminary experimental observations on a particular class of bifacial lithic artifact Figure 12. Some H[DPSOHVRIÁDNHV waste resulting from stages 1 to 2 (dacite from Paso Limay). )LJXUH([DPSOHVRIÁDNHVZDVWH resulting from stages 2 to 3 obtained with a discoid hammer of 125 g (green chert from Piedra Parada). Figure 14. Flakes-waste obtained with the VPDOOGLVFRLGKDPPHUVWRQHGXULQJWKHÁDNLQJ the stages 3 to 5 (reddish volcanic rock from Puerto Esperanza). 149 Hugo Nami Final considerations I consider this report as an initial attempt to reproduce the “curved cleavers”. Continuing with more experimental research in order to explore and docu ment the variability existing in its reduction sequence is crucial, especially in the early stages of manufacture, as is the study of the techniques employed, SDUWLFXODUO\GXULQJWKHÀQDOVWDJHV,QDGGLWLRQGHWDLOHGLQWHUDFWLYHDQDO\ ses on both the archaeological artifacts and the experimental baseline are necessary. The early stages of manufacture and the discrimination between SUHIRUPVDQGÀQLVKHGSURGXFWVDUHFUXFLDOWRXQGHUVWDQG Acknowledgements My special thanks to José L. and A. Gherardi for their help, support and information during the study of the artifacts examined here, and their help GXULQJWKHÀHOGZRUNWR-DQ$SHODQG.MHO.QXWVVRQIRUWKHLQYLWDWLRQWRSDU ticipate at the symposium; to the Societas Archaeologica Upsaliensis (SAU) that supported my participation in the event; to Britta Wallsten for her help; to E. Callahan for his discussion about this subject; to María de las Mercedes &XDGUDGR:RURV]\ORIRUKHUFRQWLQXRXVKHOSDQGVXSSRUWDQGÀQDOO\WR Jan Apel for his invaluable help and kindness during my stay in Uppsala. 150 Chapter 2 Theoretical Aspects Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Kjel Knutsson $JHQHDORJ\RIUHÁH[LYLW\7KH skilled lithic craftsman as “scientist” Abstract In this contribution to the workshop proceedings I propose that cultural change in the Scandinavian north in the Early Holocene may be understood as related to a crisis resulting in an activated relationship to the past. The UHPDLQVRIVHWWOHPHQWVIURPWKH/DWH*ODFLDOKXQWHUJDWKHUHUVRIWKH1RUWK European plain and the remnants of the pioneering settlement around 9000 cal BC found scattered in the barren hills of northern coastal Norway can, in the early Stone Age and according to this view of cultural change, have been experienced as the sacred times of origins. The argument relates to how culture, reproduced through socialization and embodied habits, in times of GLVVRQDQFHEHWZHHQWKHOLYHGH[SHULHQFHRIGD\WRGD\DFWLRQDQGLWVFXOWXUDO “structure”, surfaces and is thus made discursive. Relics from the period of ´VDFUHGRULJLQVµDUHLQWKHVHVLWXDWLRQVUHXVHGWRIRUPXODWHDQHZFXOWXUH bearing narrative. I use this notion of cultural reproduction in a prehistoric setting to dis FXVVWKHPRUHXUJHQWWRSLFRIPRGHUQLW\·VFULVLVZKHUHE\WKHSURIHVVLRQRI DUFKDHRORJ\LVXVHGDVDPHWDSKRUIRUUHÁH[LYLW\DQGWKXVREMHFWLÀFDWLRQDQG its psychology. I show that the period we call modernity may be the manifes tation of a human propensity to go from embedded to disembedded repro duction, thereby activating deeper levels of cognition and thus looking at the ZRUOGGLVFXUVLYHO\,QWLPHVRIVRFLHWDOFULVHVWKLVQHDUDKLVWRULFDOTXDOLW\LQ WKHKXPDQFRQVWLWXWLRQZLOOIRUPWKHEDVHIRUZKDWZHFDOOREMHFWLÀFDWLRQ and thus “science”. As such, the scientist as a disembedded person must always have been there. Skill is the technical and conceptual ability to relate to the old material world, and the lithic analyst in the present represents the distanced technical aspect of such ability. The conceptual knowledge resides in the present and in mythical narratives/history. Thus the text is not so much a discussion of the expression of moder nity, as its process of becoming. Skill in this setting is the technical and conceptual ability to “read” the relics on which the new representations are moulded. Keywords: Modernity, cultural reproduction, cultural transmission, skill, materiality, history. 153 Kjel Knutsson “Landscapes contain traces of past activities, and people select the stories they tell, the memories and histories they evoke, the interpretative narratives that they weave, to further their activities in the present-future”. Barbara Bender 2001 “But…at some point in time it just stopped being fun conceiving everything as a text WKDWZULWHVLWVHOIWKHSDVWDVDQHYHUHQGLQJQDUUDWLYHDQHQGOHVVSOD\RIVLJQLÀHUV ZLWKRXWVLJQLÀHGHVµ%M|UQDU2OVHQ “having replaced the ethnographic present as a convention for describing the life of other peoples, we have to think again about the ways in which we understand the past in our own lives”. Tim Ingold 1996 Introduction I started my talk at the workshop by showing a picture of an archaeologist in a familiar setting, crouching under a parasol in the middle of a barren ÀHOG )LJ ,ZLOOLQWKLVSDSHUSURSRVHWKDWWKHDUFKDHRORJLVW DVFXOWXUDOO\ GLVHPEHGGHG DQG WKXV ´DVVFLHQWLVWµ KDV DOZD\V EHHQ WKHUH DQG WKDW WKLV way of relating to the world and therefore to the past is one of the important mechanisms that change cultural codes. I am here consciously taking my starting point from a radical and historical unfolding now, relating to the effects of the crumbling or at least shaking tower of modernity. It is thus the background of what in western philosophy has been called modernity and its strife for epistemological grounding that is discussed, but through a KLVWRULFDOUHÁHFWLRQ$UHÁHFWLRQWKDWQRGRXEWLVIRUPHGLQWKHFRQWH[WRI an ontological and thus existential worry, in the aftermaths of the “science wars”. I thus want to study the past, not as separate from our present, but from the viewpoint of the “the worry”, the standpoint from which insecurity LVFUHDWHGDVDZD\RIGHVFULELQJDFXOWXUDOO\FRQGLWLRQHGVFLHQWLÀFVWDQFH It is thus an attempt to get an historical or genealogical perspective on the RQJRLQJSUHVHQWDSUHVHQWWKDWLV´XQIXOÀOOHGVRPHWKLQJLQFRPSOHWHWKDW always led itself to cultivation, socialization and change” (my translation) (Beronius 1991:7). A contribution to the history of a new late modern present HQJDJHGLQFRQWHPSODWLRQRYHUUHÁH[LYLW\WKURXJKDQDUFKDHRORJLFDORG\VVH\ marked by long term perspective. The topic may thus have some general value to a sociological understan ding of cultural reproduction, and archaeology as a subject with a possible methodological value because it deals with this historically situated but still general human existential drive in the process of “cultivation” over the long 154 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” Figure 1. The archaeologist under a parasol revisiting and contemplating a sacred site is the representation of PRGHUQLW\·VREMHFWLÀFDtion; the strive for meaQLQJWKURXJKUHÁH[LYLW\ term. It deals with the idea that the modern human is helplessly split, a split in the past, present, ego chain caused by the disembedding tendencies in the SRVW(QOLJKWPHQWZRUOG,DPZHOODZDUHRIWKHIDFWWKDWP\SURSRVDORILWV “eternal” presence is contra much thinking today. Giddens, for one, explicitly DVNVKLPVHOIZKHWKHUWKHUHÁH[LYHNQRZOHGJHV\VWHPVVRFKDUDFWHULVWLFIRU PRGHUQLW\DQGEDVLFWRLWVG\QDPLFFKDUDFWHULVVSHFLÀFDOO\ZHVWHUQ *LG dens 1996: 185). He goes on to say that the radical turn away from tradition WKDWDUHLQKHUHQWLQWKHUHÁH[LYLW\RIPRGHUQLW\QRWRQO\EUHDNVZLWKHDUOLHU epochs but also with other cultures. We see in this version of modernity the central theme of disembedding as a cultural process and a view of humans that constantly asks for meaning in the unfolding of life Although many people living within in modernity would not recognise themselves in this description, I am convinced that the origins of the active engagement in the past we have decided to institutionalize as archaeology and as illustrated by the person making a peephole in the ground on our “sacred places”, historically are a result of that existential “drive” for mea ning. This drive, as Elisabeth Rudebeck proposes, seems to have universal qualities: “...studying the past…is an existential need, an ethical need and practical need; without pasts we can neither feel, think nor act” (Rudebeck 2000:282). Identity and the need for the past The process of fragmentation of fundamentals in life mentioned above th reatens the core of personal identity. Oliver Sachs, the American sociologist (in Küchler 1993:48), nots how people become disoriented, helpless, aliena ted without an identity. That is, when they have lost important parts of their 155 Kjel Knutsson cultural code, the coherence between the structures of social reality and cul turally conditioned behaviour (Bauman 1990:185). Our consciousness then, the idea of a separate Self and thus an identity, is a dynamic synthesis of the past, the present and Ego. A break in that chain, when the past (as remembered) and the present (as experienced) come in FRQÁLFW FUHDWHV DPQHVLD DQG SUREOHPV RI QDYLJDWLQJ LQ WKH ZRUOG ,Q WKH present nostalgia boom we see one sign of such a break and crisis whereby the past is actively returned to and used as an instrument in constructing values in a shaky present. Our present representations of the structures of social reality are just not good any more, they have to be replaced or renego tiated and as such become materialized. As a matter of fact, this notion of a QHHGIRUDFWLYHUHÁHFWLRQDQGWKXVRI´KLVWRU\µLVRQHRIWKHNH\IRXQGDWLRQV of modernity (Delanty 1999); a split that created a longstanding separation between culture and nature, where humans became foreigners in their own DWRPL]HGZRUOGDQGGLVWDQFHGIURPWKHPVHOYHVWKHLUOLIHSDWKDQGJRDOV This way of actively relating to the past has been explained by social sci entists and anthropologists over and over again (Höiriis 1997) as the result of a need to get away from a “survival without meaning and spiritual values” (Arwidsson 2003). That is, the Janus faces of the modern project. It is a way RIÀQGLQJFRPIRUWLQD ODWHPRGHUQ ZRUOGWKDWLVLQFUHDVLQJO\EHFRPLQJ PRUHDQGPRUHFRPSOH[DQGGLIÀFXOWWRSUHGLFWDQGRULHQWLQ,WLVDIUDJPHQ WDWLRQRIRXUOLIHZRUOGWKDWVHWVWKHDJHQGD Being asked why for example retro designs are chosen, people today tend to answer: “because they look nice”. Something being “nice” then makes you think about aesthetics in a new way. I am sure that this way of referring to the outer world relates to fundamental qualities in life such as feeling safe, to feel at home because it relates to a known collective memory of successful practice. As I will show through my archaeological example at the end of the text, aesthetics in connection with social memory and as an encompassing term for a state of mind may very well be an important force in cultural reproduction. According to Ola Jensen (2002:18), referring to a discussion some 20 years ago, among others, the Swedish philosopher Svante Nordin (1981:69), “All historical interpretation is based on contemporary problems on dif ferent levels” and to the ethnologist Orvar Löfgren (1983:34) stating that: “The questions asked of history mirrors existential problems in our own VRFLHW\ZKLFKGHPDQGVVHOIUHÁHFWLRQZKDWGRZHUHDOO\ZDQWZLWKKLVWRU\"µ 7KHRQWRORJLFDOSUREOHPVVROYHGE\WKLVVHOIUHÁHFWLRQDUHLWLVVWDWHGUHODWHG to an identity crisis in our own society. Retro and nostalgia is thus part and parcel of an identity crisis that as an overall structure characterizes modernity. This type of crisis may of course 156 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” QRWRQO\EHDODUJHVFDOHVRFLDOSKHQRPHQRQEXWDOVREHH[SHULHQFHGRQD more intimate and personal level, as the example below will show. Thus, in an article in the Swedish newspaper Svenska Dagbladet, Agneta Lagercranz discusses such a situation with reference to the Swedish politician and medi FDOGRFWRU&KULVWLQD'RFWDUH·VERRNRQKRZLWIHHOVWRORVHRQH·VLPDJLQHGRU FRQVWUXFWHGSDVW,QKHUÀIWLHV'RFWDUHZDVLQIRUPHGWKDWKHU´IDWKHUµZDV QRWKHUELRORJLFDO´IDWKHUµ'XHWRWKLVKHUOLIHZRUOGFROODSVHV´If I am not the one I think I am, nothing else becomes valid”. Showing that this one important insight changes everything else, life has to be rearranged or rebuilt according to this new notion. This is in essence an illustration of the force of historicity LQRXUOLIHFRQWLQXRXVO\OHDGLQJXVWRH[LVWHQWLDOUHÁHFWLRQVRIWKHSDVW$V Doctare amply states ten years after the collapse resulting in a loss of her his tory as it was understood then, causing a painful loss of ontological security: “It has taken me all these years to build my identity according to a new life-history” (Svd 10 October 2002). The need for the construction of a personal narrative that is congruent with the historical point at which we stand, like Doctare, has been eloquently discussed by Pierre Bourdieu (1996). The “biographical strategy” of the type Doctare uses, he says, is the result of the need to formulate an orderly nar rative of one´s life in a situation of insecurity. These narratives show a conse quence and stability through understandable correlations often organized as a necessary (teleological) development. The personal narrative has features such as important events, logical connections, and where single events are given the character of causation. As such, these personal narratives echo the VWUXFWXUHRIPRGHUQLW\·V´*UDQG1DUUDWLYHµHYROXWLRQ /DQGDX ,QWKH drive for ontological security in a shaky present then, individuals reuse the past to create personal or collective narratives of similar structure and con tent and also use the aesthetic qualities of material culture as representations for this story. ,WLVW\SLFDOIRUKXPDQVZLWKLQPRGHUQLW\WREHIRUFHGWRVHOIUHÁHFWLRQ because, as related to a cultural idea of the Enlightment (Delanty 1999), we VHHDVWURQJHOHPHQWRIUHÁH[LYLW\LQPRGHUQLW\DFHUWDLQSURSHQVLW\IRUD FXOWXUDO FULWLTXH PRUDO OHDUQLQJ DQG VHOIWUDQVIRUPDWLRQ 'HODQW\ 0RGHUQLW\LVWKXVGHÀQHGE\UHIHUHQFHWRWKHFULWLTXHRIWUDGLWLRQ$FFRU GLQJWR*LGGHQVWKHXQGLVVROYDEOHUHODWLRQEHWZHHQPRGHUQLW\DQGUDGLFDO doubt is a question that – once seen – becomes problematic not only for philosophers of science but, as I have discussed above, existentially worrying for ordinary people (Giddens 1991:31) including archaeologists like myself. This theme, existential problems and tradition causing radical doubt, I will slightly expand on here. 157 Kjel Knutsson Modernity as a retro movement Pierre Bourdieu has said,: “… in situations in which the habitus in funda mental ways does not coincide with the structure” whereby “it can contribute to change the social structure” (Broady 1991:225). This rupture and loss of identity makes the taken for granted institutions through which we live, YLVLEOH DQG« TXHVWLRQHG 2U DV 0DXULFH%ORFK DUJXHV DIWHU ÀHOGZRUN LQ Madagascar: “But perhaps the Má Betisek example shows something quite different. It may show that, when in real trouble, we are able to analyze and criticise the very basis of our ideologies, to begin to demystify ourselves and to search for fundamentally different VROXWLRQVµ %ORFK In the same vein, after studying the history of the Soviets in the 20th century, 9LFWRU 6KQLHUO·PDQ WKH 5XVVLDQ DQWKURSRORJLVW VD\V ´7KH UHLQVFULSWLRQ of the past in the present becoming more important in times of instability and change in the political system, such as in the 1920s and the late 1980s” 6KQLHUO·PDQ :ULWLQJRQWKHERRNDERXWWKHGHDWKRI&DSWDLQ&RRN Marshall Sahlins says that reproduction of a cultural and social system praxis is changed in front of our eyes – especially if it is put in relation to something unexpected and unknown (Magnusson in Sahlins 1988:15). In this process of crisis and change in relation to something unknown, memories come to represent “the good old days” returned to and idealized: “Every new day is more modern than yesterday. The modern is created continuously every day and gives many people the feeling of not being able to follow with their time. The modern and modernity ares thus put up against something other, namely DJDLQVWWKHWUDGLWLRQDOVRFLHW\ZLWKLWVWUDGLWLRQDOYDOXHV« LW VWDQGVIRUDWLPH when…all humans had their place and knew it…and where all people were willing WRXSKROGWKDWRUGHUµ /XW]HQ Lutzen continues and sees how this longing for a golden age is out of sci HQWLÀFUHDFKUHGXFLQJKLVWRU\DQGDUFKDHRORJ\WRWKHVSKHUHRILGHRORJLFDO ÀFWLRQ $QWKRQ\*LGGHQVVHHVWKLVSUDFWLFH DVDUHÁH[LYHSUDFWLFHDQGWKXVDQ REMHFWLÀHGSDVW FUHDWLQJKLVWRU\DVDPRGHUQSKHQRPHQRQDQGQRWDSULQ ciple that can be generalized and applied to all epochs. In his opinion, it is DYHUVLRQRIWKHVSHFLÀFTXDOLW\RIUHÁH[LYLW\ZLWKLQPRGHUQLW\,ZLOOWU\WR GLVFXVVWKLVQRWLRQKHUHWKURXJKDUHÁHFWLRQRQ´WKHSDVWLQWKHSDVWµDQG KRZ WKH DWWHPSW WR EULGJH WKH DE\VV RI WLPH SRVVLEO\ KDV LQ D ORQJWHUP SHUVSHFWLYHVRPHWKLQJWRVD\RIUHÁH[LYLW\DVDPRUHIXQGDPHQWDODVSHFWRI cultural reproduction with long threads reaching far back in time. 158 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” The institutionalization of the past-as-object Science shuns tradition and systematically tries to break its power. Thus, by an active dissociation to and critique of tradition, it is possible to break up conventions by the relativization of them, for example, when as an anthro SRORJLVW \RX FRPH KRPH DIWHU D ORQJ ÀHOGWULS DQG ´VHHµ \RXU RZQ HPER died habits in a new light (Marcus & Fischer 1986). According to my view, WKHFXOWXUDOFULWLTXHPD\PHWDSKRULFDOO\EHSDUDOOHOHGZLWK*DGDPHU·VHSLV temological concept of the fusion of horizons: “Only in meeting the unfamiliar…will I be provoked into realizing that my horizons are to narrow” (Gadamer in Lübcke 1987:163 ff), and as a result of that meeting you are forever changed. The point is that Gadamer, as a philosopher discussing the epistemological grounds for “modernity”, shows by this how he has made the past into his tory and thus to an object out there and therefore something “unfamiliar” that both can and must be investigated. ,QSUH(QOLJKWPHQW(XURSHWKHSDVWZDVQRWDQREMHFW´RXWWKHUHµLWZDV as a mythical past, where: “Oral narratives telescoped, expanded and rear UDQJHGVHJPHQWVRIWKHSDVWLQOLQHZLWKWKHVLJQLÀFDQFHDWWULEXWHGWRWKHPµ (Lowenthal 1985:220). According to the “mechanical” perspectives of this time, “Humankind, society and nature were predetermined by a static and therefore unchangeable order (Jensen 2002:42); i.e., in this case, the literary “canons” of the day. It is understandable that an interest in time, change and history in general, was not a central issue then. Despite this, the Swe dish archaeologist Ola Jensen continues, speaking of old artifacts: “..people had in earlier times…been interested in prehistoric objects. However, they where not to a greater extent considered to contribute to our knowledge of the past, but mainly obtained their value and explanations from the text” (Jensen 2002:43). In the late 18th and early 19th century, things changed. According to Ola Jensen, the past, history and antiquities became subjects of enormous inte rest, and soon, archaeology came to be developed (Jensen 2002:42). From WKLVWLPHWKHPRGHUQHSLVWHPHDQGKLVWRU\ )RXFDXOWLQ-HQVHQ 2002:42) became the main principle of how to form and structure know ledge and understanding. An eternal search was embarked upon that addres sed issues such as the very roots of humankind itself and the role of people LQDQHYHUFKDQJLQJZRUOG+XPDQLW\EHFDPHDVXUURJDWHIRU*RGLQWKDW SHRSOHZHUHQRZWKHVHOIPDGHHQJLQHHUVRIKLVWRU\6LPXOWDQHRXVO\WKHORVV of faith in a divinely ordained history (Arwidsson 2003), as the Enlightment proceeded, made the past not only remote, but also fearsomely different. This probably came to be by the comparative scrutiny of datable text and manifold contacts with exotic peoples, making it obvious how unlike the 159 Kjel Knutsson present our own previous thinking was. Consciousness that the past was un like the present, that people in other times and places did things differently thus came to be central to progressive Western thought. This disembedding process of modernity (Hornborg 1994), created a situation where the past as an object became visible as “a past before us” (Ingold 1996), something that had to be “explained” or understood in relation to a radical revision of how the world was understood at that time; a radical revision of the important narrative that on a less abstract, personal level was discussed referring to &KULVWLQD'RFWDUH·VOLIHKLVWRU\DQGWR*LGGHQV·V Halbersham (1999:17 in Dobres 2000:54) speaks of this period as the rise RIDPRGHUQVHOIGHÀQLQJVXEMHFWZKLFKDVSLUHVWRWDNHFRQWURORILWVRZQ social relationships, rather than leaving those structures that shape its eve ryday existence to the authority of tradition carried by texts and/or narrati ves: “The once sacred, mystical and poetic aspects of culture were replaced by the craft, method, dispassionate and rational study of tangible matter”. %\WKLVUHHYDOXDWLRQRIWKHSUHVHQWWKHZRUOGEHFRPHVGLVFXUVLYHDQGRE MHFWLÀHG,QWKHVDPHYHLQWKHSDVWEHFRPHVDQREMHFWWKDWFDQRUPXVWEH investigated because it is “from the outside”. But as the Swedish anthropolo gist Alf Hornborg has put it: “If contextualism is served by a monist epistemolog y, we may conversely conclude that the “disembedding” tendencies of modernity are part and parcel of Cartesian GXDOLVPµ +RUQERUJ ,QIDFW´'HFRQWH[WXDOLVDWLRQDQGREMHFWLÀFDWLRQFDQ be seen as two sides of the same coin here”. Thus, the decontextualized relics become objects detached from the world of the “cultivated human” (Bauman 1990:185) because the code that explai ned that world had crumbled; when the “body of collective wisdom, rules and proscriptions” (Dobres 2000:5) had shattered. ,W LV QRW VXUSULVLQJ WKDW DW WKLV IXOFUXP LQ UHFHQW KLVWRU\ ZH ÀQG WKH JURZWKRI´DUFKDHRORJ\µLQLWLDWLQJDSURFHVVZKHUHE\ZHLQÀQGH[ SHUWV WHFKQLFDOO\" VNLOOHG SHUVRQV SHHSLQJ LQWR D PDQPDGH KROH LQ WKH ground. Archaeology, then, is an institution that as its subject explains the built environment of relics and sacred places, an environment that due to a break in the chain between the past, the present and Ego, was made “unfa PLOLDUµDQG´IRUHLJQµ7KLVGHP\VWLÀFDWLRQRIFXOWXUHDQGWUDGLWLRQVDVZH saw, also relates to disengaged procedures of knowledge, the formation of KLVWRULFDOGLVWDQFHVZKRVHÀ[LW\RQDPRUHJHQHUDOOHYHOGLVWDQFHG8VIURP Them. Modernity in this scenario represents: “(a) decisive transformation in pure ontology” (Cassirer 1951:38). Through the modern episteme, objects as relics of the past thus came to be considered on the basis of their own internal lives and biographies 160 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” (Jensen 2002:45). This puts the question under the changed conditions of the present, meaning that the: “knowledge seeking subject has become self-relating” (Habermas 1989) rather than informed by tradition. Soon this loss would become its own tradition or ideology, an ideology that ironically shunned tradition. Domestication of the past – science as conventional thinking within tradition Archaeology then, as represented by the archaeologist under the parasol, is a representation of the way modernity has toDSSURDFKWKHSDVWDVVWUXFWXUH because the past is made into an object through disembedding. This need for GLJJLQJUHFRUGLQJDQGUHFRQVWUXFWLQJIRUPHUOLIHLVDFKLOGRIPRGHUQLW\·V feverish “search for ontological security and thus for origins”. Is this active search in the scrap heaps of past events the result of a “break LQWKHFKDLQµDSDVWDQGDSUHVHQWWKDWLVLQFRQÁLFW",VDUFKDHRORJ\UHDOO\D practice that is invoked in the wake of a lost or threatened identity or is it just “a job”? “The job” being a metaphor for an embedded, unquestioned view of archaeology, archaeology as a naturalized craft legitimized through the routi nes of everyday practice. I will argue that this has bearing on the crucial and complex relation between embodied knowledge and discursive knowledge in cultural reproduction (Barrett 2001). .DQWLVVDLGWREHWKHÀUVWWR H[SOLFLWO\ EUHDNZLWKWKHPHWDSK\VLFDOKHULWD ge (read tradition), divert philosophy from the True and Eternal and concen WUDWHRQZKDWIRUPRVWSKLORVRSKHUVFRQVLVWHGRIWKHFRQIXVHGQRQH[LVWLQJ the occasional and volatile. The Transient? This is where we see the crack that makes modernity possible, the loss of truth in the critical moments that Giddens speaks of (Giddens 1984). These critical moments no doubt reveal, through disembedding, your cultural code, and it has been the task of critical theory in general to unmask many of the false ideals, repressive practices, ex clusionary identities and other fallacies of science and modern culture. “This sort of wisdom and its applications become problematic, however, when the forms of criticism in which it was originally achieved are turned into academic genres of critique and commentary, and the insights are recast as established and legitimaWHGDFDGHPLFNQRZOHGJHµ 6WHQOXQGPDQXVFULSW Following Stenlund, the problem with much archaeology today, since it has a long history of pretensions of being “a science”, is that the “unfamiliar” in the past has become just too familiar again and thrown us into something much more like a “cultural” practice characterized by “routines”; a dealing 161 Kjel Knutsson with the past taking the form of a cultural “tradition” and is thus more akin to a domestic “recalled and chronicled past”. In his book “The Order of things: an archaeology of the human sciences (1970), bearing on the French KLVWRULFDOHSLVWHPRORJLVWV0LFKHO)RXFDXOWWKXVWDONVDERXW´WKHXQUHÁHFWHG preconditions for our knowledge”. As scientists, David Lowenthal says, we are, “too easily swayed by a spu rious likeness, as seeming continuity”. We should be wary of anything fa miliar. This, he continues, precludes the domestication of the past. Or as J. Harvey puts it: “separation is required conceptually in order that continuity and sameness can be made apparent. “the past is brought into the present, but the process requires that its otherness be addressed”. (Ingold 1996:221). Or with the worlds of Sherry Ortner speaking of anthropologists (but this could actually go for any scientists forever trapped in the ideology of a distant view), ´,WLVRXUFDSDFLW\ODUJHO\GHYHORSHGLQÀHOGZRUNWRWDNHWKHSHUVSHFWLYHRIWKHIRONV on the shore, that allows us to learn anything at all-even in our own culture-beyond what we already know”. ..”the importance of maintaining a capacity to see the otherQHVVHYHQQH[WGRRUEHFRPHVPRUHDQGPRUHDFXWHµ 2UWQHU Even scientists too easily “gloss over…past social realities” (the historical re cord) thus “weakening contradictions” giving the narrative a “powerful sim SOLFLW\µPDNLQJWKHVFLHQWLÀFWH[W KLVWRU\ LQWRDOHJHQG ,QJROG 7KHSDVWLVLQWKLVFRQWH[WQRWDIRUHLJQFRXQWU\EXWRXURZQKRZHYHUÀOWH red or sanitized. 7KHGDLO\URXWLQHVRIVFLHQWLÀFZRUNIXQFWLRQWROHJLWLPL]HZKDWHYHUWUD dition we work within. Thus, in a way, we live and work in accordance with DQHPERGLHGQRQGLVFXUVLYHXQGHUVWDQGLQJRIWKHFRQWHQWDQGDLPVRIRXU “science. The past is in such situations made to look familiar. “Even for academics, most of the time, the past is not a foreign country but our own”. This said because the past is seen as another present. Films are regarded as SRSXODUWDOHVVHWLQRWKHUWLPHVEULGJLQJSDVWDQGSUHVHQW´7KHVDPHPHQ talities are shown to animate mythical or medieval as modern folk” ( Ingold 1996:206). “Elemental passions are enacted on a timeless stage”. Since conventions are embodied habits and therefore “unseen”, they are the fundamental, axiomatic preconditions for thinking. The discussed idea RI WKH ´ÀHOG RI DUFKDHRORJ\µ WRGD\ PD\ YHU\ ZHOO EH XQGHUVWRRG DV VXFK an unseen “habit”. If it is so, then its role as a strategy to resolve an identity crisis, which was why it was originally formed as a practice, is lost; it is just something we do in our culture so to speak. Modernity has to start asking TXHVWLRQVDERXWLWVHOIDJDLQ-XVWOLVWHQWR3LHUUH%RXUGLHX·VYLVLRQRIWKHWUXH SUDFWLFHRIVRFLRORJ\DVDVFLHQWLÀFGLVFLSOLQH 162 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” “The sociologist whot studies his own world, the most intimate and familiar, should not like the anthropologist make the exotic familiar, but make the familiar exotic. He does this by breaking with the immediate intimacy he feels for the life-forms and way of thinking that makes them foreign, because they are to well known” (Bourdieu Although he is totally wrong about the anthropologist, at least as they want LWWRGD\ZKDWZHVHHKHUHLVDFXOWXUHWKDWKDVUHLÀHGGLVHPEHGGLQJDQGWKXV REMHFWLÀFDWLRQ7KHKLVWRULDQ($&DUU DVFLWHGLQ'REUHV LQWKH same vein sees distanciation as the (positive, I assume) sign of the modern scientist: “The historian who is most conscious of his own situation is also most capable of transcending it…Man´s capacity to rise above his social and historical situation seems to be conditioned by the sensitivity with which he recognizes the extent of LQYROYHPHQWLQLWµ &DUU Scientists are thus expected to challenge straight in the eye an ontology that makes the world known and safe. Since ontological security is one of the fun damentals of the human existence, this idea of continuous critique for good UHDVRQVVRRQEUHDNVGRZQDQGWKHVFLHQWLÀFZRUNLVWXUQHGLQWRDFXOWXUHRU tradition. The good reason being of course the need for “ontological safety”. Lowenthal (1985) sees a division between “the scientist” and the public opi QLRQ7KHLGHDRIDVSOLWNHSWRSHQLQZHVWHUQWKLQNLQJDQGUHLÀHGWKURXJK the idea of a “science”, thus seems to have been overtaken by cultural action; cultural action by real people doing real things, albeit getting its legitimate H[LVWHQFHZLWKLQD´ÀHOGµLGHRORJLFDOO\FXVKLRQHGLQDFXOWXUHRI´DGLVWDQW view”. If archaeology does not actively keep this split open, it has, in my view, ORVWWKHEDVLVIRULWVH[LVWHQFHDVDVHSDUDWHÀHOG+HQFH´KLVWRULFDOHPSDWK\µ is legitimate from a cultural point of view, a way to present informed and LQWHUHVWLQJQDUUDWLYHVEXWQRWZLWKLQWKHÀHOGFDOOHGVFLHQFHDQGWKXVQRWDU FKDHRORJ\$UFKDHRORJ\DVPDQ\RWKHUSURFHVVHVRUUHÁH[LYLW\LOOXVWUDWHVDQ important force in cultural reproduction: the urge to form cultural conven WLRQVDQGWUDGLWLRQV:KHQWUDGLWLRQVDUHWKUHDWHQHGDUHÁH[LYHSHUVSHFWLYHLV growing, but this is soon to be formalized into unquestioned habits again. 163 Kjel Knutsson Strain or interest in cultural change We have seen how cultural change may be understood as the result of a FRQÁLFWEHWZHHQWKHVWUXFWXUHRIDQLGHDWLRQDOV\VWHPDQGGDLO\SUDFWLFH,WLV QRWRQO\W\SLFDOIRUPXQGDQHOLIHEXWIRXQGDOVRLQWKHFRUHDUHDRIUHÁH[LYH practice, science. Sherry Ortner has discussed the motivation for action in situations of cri sis in terms of two theoretical concepts that have bearing on this discussion of an enforcing crisis: strain theory and interest theory: “although pragmatic rationality is certainly one aspect of motivation, it is never the only one” (…) “need, fear, suffering, desire etc. must surely be part of motivation as well. Further: “the idea that actors are always pressing claims, pursuing goals, ad vancing purposes (...) may simply be an overly energetic view of how and why people act…”. Actors within strain theory (Geertz 1973c) “are experiencing the complexities of their situations” (Ortner 1984:151). Here it is not the ag gressive aggrandizer character that is the engine in cultural process (Hayden EXWSHRSOHLQWKHUHYHUVHVLWXDWLRQKDYLQJORVWFRQÀGHQFHDQGRQWROR gical security, people under threat and strain… ”While strain theory does not rectify the psychological shortcomings of interest theory, it does at least make for a more systematic exploration of the social forces shaping motives more than interest theory does. Indeed, one may say that strain theory is a theory of the social, as opposed to psychological production of “interests”, the latter being seen less as direct expressions of utility and advantage for actors, and more as LPDJHVRIVROXWLRQVWRH[SHULHQFHGVWUHVVHVDQGSUREOHPVµ 2UWQHU I will argue here that the strain a crisis puts on people, can be seen as a human condition. This condition, where malfunctioning cultural codes are made discursive, is further, I assume, an important part of the dynamics of the “structuration” of society (Giddens 1987). As part of a general theory of a human propensity to react in situations of trouble in social interaction, these situations may always have been there. As we will see below, these periods of paradigmatic changes and discu rsiveness in relation to cultural codes, the loss of the certainty of how the world is created by tradition, might be the true seeds of cultural change. Archaeology, although still hypothetically, may very well aid in developing, expanding on and providing substance to such theory of “historical struc turation” because we are at times “unbound, freely acting and voluntary thinking individuals” (Arwidsson 2003:150), and the past is an object out there. A materialized discursive practice, such as lithic technology, is a key WRVXFKDGLVFXVVLRQRIWKHORQJWHUPVLQFHWKLVZD\RIFRPPXQLFDWLQJKDV 164 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” metaphorical relations to cosmology as mnemonic devices in reproduction and construction of culture bearing narratives. Knowledge and know-how To be able to reconstruct an ancient craft, present day archaeologists or lithic specialists of the kind gathered at The Skilled Production and Social Repro duction Workshop, read stones, stones found by accident, through surveys or on archaeological excavations (Fig 2). I thus see muscle memory as a vehicle for discursive knowledge, aiding in the implementation of analyses of, and correct reproduction of the material world. Two questions pertinent to our under standing of this activity have bearing on my topic: the will or urge to do it LQWKHÀUVWSODFHDQGWKHDFWXDOVNLOOQHFHVVDU\WREHDEOHWR7KHZLOOWRGR so has a different background of course (as symbolic capital), but in general it is typical for modernity and somehow contributes to the construction of the grand narrative. The stones are reinterpreted in the present and are the Figure 2. A lithic analyst reconstructing old technologies today. Skill manifests itself as the embodied ability – know-how – to reconstruct technical processes, but also as the ability to “read” the stones thus getting close to their discursive qualiWLHV7KLVUHÁH[LYHUHXVHRIROGWHFKQRORJLHVDVUHFLSHVIRU WHFKQLFDOSURGXFWLRQLVDUHSUHVHQWDWLRQRIUHÁH[LYLW\RIPLQG typical for the distanced “modern”. Mikkel Sørensen and Errett Callahan at Lejre Research Centre, 2005. 165 Kjel Knutsson sources for and stand as representations of past events in that narrative. The past as material culture is thus actively engaged in the present and made in telligible in relation to our view of the world. The skill necessary to be able to replicate and read stones is a matter of bodily and cognitive experience, something we have several examples of at the workshop. This is the technically skilled practioner “who knows”. The archaeologist or lithic analyst then, in my argument, is but one, the latest one, in a long sequence of “archaeologists”. The lithic craft person thousands of years ago, “returning” to old campsites of their past, may ac tively have used the relics found to recreate a craft using the same objecti fying strategy. Not only as “discursive objects or phenomena of the subjects (craftsmans) cognitive experience”, but based on the “real” qualities they possess, qualities that “shape both our perception of them and our cohabitation with them” (Olsen 2003:88 ,WLVWUXHWKDW´ « PRUHWKDQDQ\PDWHULDOOLPLWDWLRQVLQKHUHQWLQWKHPHFKDQLFDO DQGFKHPLFDOSURSHUWLHVRIQDWXUDOUHVRXUFHV OLNHOLWKLFV DQGRUHFRORJLFDOFRQGLtions, it is human agents and their webs of social relations and values that are central WRWKHGD\WRGD\ UH SURGXFWLRQRIWKHLUPDWHULDOFRQGLWLRQVµ 'REUHV %XW WKH VNLOOHG OLWKLF FUDIWVSHUVRQ DOWKRXJK ÀUPO\ VLWXDWHG LQ KLV RU KHU cultural context, the craft being reproduced from within so to speak could DFFRUGLQJWRP\YLHZUHDGWKHVWRQHVDOVRZLWKLQDPRUHDKLVWRULFXQLYHU salizing frame. Here the “intertwined social and material constitutions of material practice” has temporarily been made separate (the Present in Mau ULFH %ORFK·V WHUPV knowledge as seen by Jacques Pelegrin (1990), or re-collective remembering DFFRUGLQJWR%HUJVRQ 0XOODUNH\LQ2OVHQ 2003:97)), making “the mechanical and chemical properties” (Dobres 2000 and thus regularities of fractures in brittle solids (Knutsson 1998) more open IRUHYDOXDWLRQ,QWKLVZD\NQRZOHGJHRUUHFROOHFWLQJPHPRU\LVHPSKD sised in the process, mimicking the distanced analytical work of the natural scientist in the true Baconian tradition. This distant view is thought of as being typical of the modern mind. But this is only the surface; underneath this technical relationship to past materialities there is another relation, the relation to history that forces people to “return to the sources” (Bourdieu 6LQFHWKHVWUXFWXUHRIWKHFXOWXUDOÀHOGLQ%RXUGLHX·VZRUOGUHODWHV to its VSHFLÀFhistory, this return cannot stand apart from it. 166 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” 5HÁH[LYLW\DQGDUFKDHRORJ\ The archaeologist can only be understood as a socially informed agent crea ted within the frame of humanity´s all encompassing historicity; as the result of a “drive” to relate to the past as society derives from antecedent condi tions, highlighting time and history as causal factors (Gosden 1994). Relating to practice theory, the dialectics of structuration theory and its localization of causality of agency and structure in time and space, one could revitalize the actuality of “the question” once creating the subject of archaeology and formulate a real relationship between the past and the present: to situate our selves in a radical political present. Here one could, for example, try to add substance to the theory of structuration through an investigation of whether WKHUHDUHDQ\ORQJWHUPKLVWRULFDOFRQVHTXHQFHVRIVWUXFWXUDWLRQWKHRU\WKDW go beyond the anthropological/sociological time perspective (Axel 2002; Drugge & Johansson 1997), questions of whether there are any relationships between disembedding and an activated generative thinking in the reuse of the past (for a discussion of the past in the past, see e.g. Gillian 2001; van '\NH $OFRFN%UDGOH\ RUZKHWKHUWKHUHDUHDQ\ORQJWHUPKLV WRULFDOO\IRUPHGVWUXFWXUHVWKDW´LQÁLFWRQWKHPDNLQJRIWKHQHZSUHVHQWµ (Thörn 1997:13). The archaeologist, to return to the situation under the parasol, is from this point of view, “skilled”. He/she is the technically skilled practioner that by corporeal engagements with experiences of material culture can discuss WKH SDVW IURP D GHWDFKHG SRVLWLRQ SHUVRQLÀHG LQ WKLV ZRUNVKRS E\ OLWKLF DQDO\VWVUHFRQVWUXFWLQJROGWHFKQRORJLHVYLDGLUHFWUHODWLRQVWRFHUWDLQDKLV WRULFDVSHFWVRIWKHSUHVHQW ÀJ 7KHFRQFHSWXDOO\VNLOOHGSUDFWLRQHUZKR understands and uses the past in the meaningful context of the present. I have mentioned the nostalgia boom and see this as an expression of fun damentally the same type of ontological insecurity that formed the precondi WLRQVIRUWKHJURZWKRIPRGHUQLW\,WLPSOLHVDQREMHFWLÀHGSDVW,KDYHDOVR PHQWLRQHGWKHVSHFLÀFGLOHPPDRIPRGHUQLW\LQWKHSUHVHQWVLQFHWKHUHFHQW FULWLTXHRIWKHFXOWXUHRIUHÁH[LYLW\KDVVHULRXVO\WKUHDWHQHGWKHRQWRORJLFDO position of the scientist and therefore the very ground on which modernity rests. Both these present day phenomena relate to an explicitly formulated or implicit existential worry and may be discussed from the perspective of WKHORQJWHUPDQGDVDWKUHDWHQHGFXOWXUHRIWKHSUHVHQWWKXVKRZFXOWXUDO UHSURGXFWLRQPD\EHUHODWHGQRWRQO\WRWKHSDVWEXWWRDQREMHFWLÀHGSDVW 167 Kjel Knutsson Returns and the construction of Narratives The archaeologist setting the stage for this paper returned to a place of pre vious activities, albeit based on an institutionalized need to relate to the past, put up the parasol and started to dig. What could easily be discussed in relation to a theory of structuration, in terms of the past and its materiality, are signs of returns, archaeologists at work in the past. This has been discus sed already (World archaeology 1999, vol. XXX; Gillian 2001; van Dyke & $OFRFN DQGWKHDUFKDHRORJLFDOUHFRUGMXVWRYHUÁRZVZLWKH[DPSOHV In my talk at the conference I provided quotations from some excavation reports showing that these periods of active return do exist. ”Scattered older iron age pottery fragments are a common feature on most Stone Age Sites in the region”. “The large Mesolithic sites were also visited during the Neolithic” No doubt, people did return to places of their own past and it can therefore be argued that they had, at least at times, an active relationship to it. To be able to deal with this as part of a more serious theoretical problem, I think ZH VKRXOG VWDUW E\ ORRNLQJ D ELW FORVHU DW WKH VSHFLÀF SURFHVVHV UHODWHG WR these possible returns. We thus need to explicate material culture and its cul tural context in accordance with the theme of his workshop, technology and LWVVSHFLÀFUROHLQVHTXHQFHVRIHPEHGGLQJDQGGLVHPEHGGLQJWHQGHQFLHVLQ reproduction and try to understand the relation between technology and the skilled “practioner who can tell”. This relates not only to technical reproduc tion of a craft, but to the urge to do so, and therefore also to the mythical and structural levels of material culture. We have to look at the context of ancestral legitimation carried by material culture. The thing as a materialized discourse in the construction of Narrative We are all surrounded by objects in the built and natural environment: “the VHEULJDGHVRIQRQKXPDQDFWRUVWKDWconstrainGLUHFWDQGKHOSRXUGD\WRGD\ activities” (Olsen 2003:88). Some of these are relics that come from past events and in accordance with the theme of the Skilled production workshop, I will concentrate on re lics from technological events. Sites where the traditional archaeological ma terials, such as stone tools and debitage are found may have been, as they are 168 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” in modern eras, important social arenas for cultural reproduction. But the site is thus also a place of tension and therefore promotes cultural change. Material culture, functioning as vital parts of this communication, not least, as I will show below, as references to the ancestral past, will then be active both in conserving structure and in inducing change (Weiner 1992), or as Mark Edmonds says in talking about lithics; “... The creation of technolog y, the form that it takes, and the manner of its subsequent deployment, serve as powerful media through which people reproduce some of the basic categories of their social and material world. For the same reason, traditions of making and using may also serve as a point of departure in the negotiation of new relations and new meanings.´ (GPRQGV Thus, we have to situate and understand the activities (technological) within the larger, meaningful social practices of which they are part and, above all, link them to material representations. $VUHSUHVHQWDWLRQVRIDQLGHDOL]HGSDVWUHODWHGLQ0DUVKDOO6DKOLQV·VYHU sion to ancestors or cultural heroes, the dramatis personae in important nar UDWLYHV VRPH RI WKHVH QRQKXPDQ DFWRUV RU WHFKQRORJLFDO SURGXFWV LQ WKH GLVFXVVHGSURFHVVRIGLVHPEHGGLQJDQGGHP\VWLÀFDWLRQRIFXOWXUDOFRGHVLQ situations of crisis, must have become “unfamiliar“ and thus had to be rede ÀQHG7KHLUYDOXHLQWKHSDVWKDGWREHUHIRUPXODWHGLQWKHSUHVHQWFRQWH[WV reintroducing them into and helping to formulate the logic of a new cul tural code, a new origin and a logic balancing the structure of social reality DQGGDLO\SUDFWLFH$QGHYHQPRUHFRQFUHWHO\LQ*RVGHQ·VZRUGV´/RQJHU sweeps of recursiveness are solidly material, as it is the enduring nature of material culture that makes possible life on a scale greater than that of the individual” (Gosden 1994:137). The past is thus also radically material and thus an aspect of the structu res that “constrain, direct and help our day to day activities” (Olsen 2003:88) DQGDVZHZLOOVHHZHDUHDOVRQRW´IUHHÁRDWLQJLQWHOOHFWVGHWDFKHG«IURP the physical world”. But this is not all, as I have tried to argue; this physical world, these traces of past events, are not always embedded as a past with us EXWREMHFWLÀHGLQGLVHPEHGGLQJDQGWKXVDFWLYHO\HQJDJHGLQFXOWXUDOUHSUR duction as a past before us. As such, despite the fact that they have been de prived of most of their original cultural explanations causing a split between the structures (including material culture) of the social reality and cultur ally conditioned behaviours, they still as signs “constrain, direct and help”. They are involved in social interaction as agents. As lithic analysts we un derstand these constraints in the form of debitage characteristics informing us about concepts of method technique. Their physical appearances are thus not open to any interpretation. Their new lives as agents are thus formed 169 Kjel Knutsson E\WKHWHFKQLFDOO\VNLOOHGFUDIWVPDQ·VERGLO\H[SHULHQFHV,WLVLQWKLVGLDOHF tic web of material production and social reproduction that the “physicality of technical activities intersect with the lived meaningful world of human agents” (Dobres 2000:127). As such, the archaeological material record must be understood as a material discourse always in the process of being read. Narratives and material discourses An important aspect of the construction and reproduction of identities and thus the context of a reused past are collective rites and rituals. But what keeps together a series of such collective manifestations are “Narratives” (Thörn 1997:24). Narratives or myths contribute a perspective that in a given situation gives coherence and meaning to life (White 1987; Jameson 1989) by giving a name to a human collective and by localizing it in time and space. In this vein, the English historian Eric Hobsbawm in an often cited paper talks of “invented tradition”, by which he means “a set of practices, normally governed by overtly or tacitly accepted rules and of a ritual or symbolic nature, which seek to inculcate certain values and norms of behaviour by repetition, which automatically implies continuity with the past” (Hobsbawm 1996:1). In this process, old materials are reused to construct new traditions for new purposes (Regner 1999:17). It is in this latter function I will discuss skill and social reproduction, where the past, material culture and skilled production are brought together, skilled production becoming the vehicle for the construction of new tradi tions through a materialized Narrative. The sacred time of origin constitutes in these myths a distant yet vital past against which the present can be contrasted. “The associative conventions of representation in myth make the intermediate times disappear”, while the historical “narrative” tends to focus on middle periods or sequences of chan ge which makes the process from past to present seem visible and compre hensible” (Hastrup 1987:261). They represent two different ways of linking past and present typical for cultures with and without written history. That is, whatever narrative is produced, the disorderly past (Lowenthal) may look ordered by hindsight. Because origin points are necessary, false beginnings (Conkey 1993) that imply unidirectional causality (Dobres 2000:11) we have to, the Swedish sociologist Håkan Thörn says, speak of a beginning rather than origins. Beginnings do not exclude that what is the beginning is a con WLQXDWLRQRIVRPHWKLQJWKDWKDVDOUHDG\KDSSHQHG7KDWDV\PEROLFÀJKWKDV an historical beginning does not exclude that it has inherited patterns from HDUOLHUÀJKWVSDWWHUQVWKDWYDU\DQGDFKLHYHQHZPHDQLQJV 7K|UQ We thus stand in front of a continuously enfolding production of meaning, being an amalgamation of the past and the present. 170 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” Narratives, historical or mythical, are important in cultural reproduction. They insert the individual into a collective and in a space/time frame encom SDVVLQJWKHLQGLYLGXDO·VH[SHULHQFH,WSUHVHQWVDQVZHUVWRTXHVWLRQVVXFKDV who you are; from where you come; where we are heading. Narratives thus construct perspectives that in a given moment give cohesion and meaning to the world. These narratives are so fundamental to cultural reproduction and VRXQLYHUVDOO\SUHVHQWWKDWWKH\PD\EHXQGHUVWRRGLQDKLVWRULFWHUPV´7KH all encompassing process of narrative is actually thought to be the central function of the human consciousness” (Jameson 1989). As an example I point out the connection between material discourses RU PHWDSKRUV DQG P\WK HJ 3HWHU 5RH·V ZRUN RQ 6RXWK $PHULQGLDQ PD terials. Speaking of the mythic level of style, he shows the tight relation between aesthetics and its semiotic value. Myths, we are told, are “sacred stories that…explain the form (structure) and behaviour (process) of things (and sociality) and how they came to be differentiated. Myth therefore can be an important determinant of style. In these situations, the style of material culture represents the overt message in myth, shown by the iconography of material culture. They are surface manifestations of the unconscious patter ning of “structure”, a structure accessible at the mythical level. Further on in the paper I will discuss these types of relationships in a FDVHVWXG\RIWKH/DWH*ODFLDOKXQWHUJDWKHUHUVRIWKH1RUWK:HVW(XURSHDQ plain. The Norwegian archaeologist Ingrid Fuglestvedt has proposed that these groups of reindeer hunters had an animistic relation to their environ ment. In the arts and crafts in such societies, Roe tells us: “The artefact symbolically represents its typological aspect, and recapitulates in the contexts of its processual fabrication and use, the pervasive animism of the world view of the groups that create it. This view of “object as myth, myth as object” is based on WKHDQLPLVWLFSULQFLSOHRIWKHWRWDOVSLULWXDOLW\RIWKHZRUOG«µ 5RH Technology in this setting is more than utilitarian items; the artefacts are mythical transforms of creatures which act as the major “natural symbols” in cosmology. One example is the myth of how to explain different levels of skill in basket making in Waiwai in Shefarimo. ´/RRNLQJDWDEDVNHWDQGDSSUDLVLQJLWVZDUDWDSLSDWWHUQIURP8UXÀUL·VVNLQZLOO “trigger” a recitation of this myth, just telling of the myth will “call up” the artefact as example. This process of “mythic empiricism” shows the intimate relationship between myth and material culture. The whole stylistic process, from the selection of raw materials, through the production of the form and its appropriate decoration, to the sex that employs this technolog y and the type of task for which it is designed, are all systematically linked to the cosmos in animism technolog y. Microcosm recapituODWHVPDFURFRVPµ 5RH 171 Kjel Knutsson The site as a sacred place Artefacts, their production and use obviously have a strong symbolic and VHPLRWLFORDGLQUHSURGXFLQJZRUOGYLHZVWKXVH[SODLQLQJWKHZRUOGWKURXJK DPDWHULDOL]HGGLVFRXUVHZLWKP\WKLFDOFRQWHQW,QWKHÀUVWTXRWDWLRQIURP %DUEDUD%HQGHU·VERRNRQODQGVFDSHWKDWIRUPVDQLQWURGXFWLRQWRWKHSD per, not only the material “sacred” remains of past events are important in the construction of narratives explaining the world, but the landscape in which they are situated. “Landscapes contain traces of past activities, and people select the stories they tell, the memories and histories they evoke, the interpretative narratives that they weave, to further their activities in the SUHVHQWIXWXUHµ %HQGHU No doubt the landscape in different ways has to be incorporated into the life world and thus the narratives of groups in habiting that world. Perhaps DVWKH´ZHHSLQJELUGVRXQGZRUGµVRQJVRIWKH.DOXOLDQG*LVDORFHUHPR nies in New Guinea evoke powerful images of landscapes, paths and places through which, as they harden in the course of the singing; “living people UHFRQQHFWZLWKWKHLUDQFHVWRUVLQVHHQDQGXQVHHQZRUOGVµ )HHOH\+DUQLFN in Ingold 1996:215 f). If this animated world was inside the heads of these people, it must at the same time have served as a mnemonic device, bridging the present with past events through material culture. The built environment, the named and used landscape, is thus a break to “ceaseless change” and might explain the long sequences of continuity in material symbolism that we see in many prehistoric setting. But, as Lo ZHQWKDOVHHVWKLVFRQWLQXLW\WKURXJKWKHPHPRU\RIPDQPDGHHQYLURQPH nt: “we lack their spontaneous and unselfconscious use of their own cultural conventions” (Lowenthal In Ingold 1996:209). “…we can never fully enter their perceptual world”. That is as a past that is with us. “Nothing repli cates the past as it was for those who lived it as their present” (ibid:209f). None of us doubts that “people select the stories they tell”, the narrative is written from the historical position of the writer, but, as I will argue, there are constraints, constraints activated through the ”discriminant judgement of the (perceptually) skilled practioner” (Ingold 1996:48). The landscape is WKXVFRQWLQXRXVO\UHUHDGDOEHLWPRUHVRLQFHUWDLQSHULRGVWKDQRWKHUVEXW always in a dialogue between the historical present and historical/mythical narratives represented in and carried by sacred places and relics. I have already pointed out the drive for meaning that is triggered as re production for some reason fails. History as a past before us is activated in these circumstances. Not only archaeology but sociology was established as an answer to the same type of cognitive insecurity that characterizes these 172 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” types of situations. For most of the sociologists at the turn of the century, the historical perspective was naturally important (Thörn 1997:22). Thus, re ÁH[LYLW\DVDIXQGDPHQWDOVWUDWHJ\LQVRFLDOUHSURGXFWLRQPLJKWEHHOXFLGDWHG E\DKLVWRULFDOUHÁHFWLRQ2UUDWKHULQOLQHZLWKZKDW,WDONHGDERXWLQWKH beginning of the paper, it is about genealogy, a genealogy of a certain mode RIDWWHQWLRQLQUHODWLRQWRUHDOLW\5HÁH[LYLW\LQFXOWXUDOUHSURGXFWLRQ Genealogy is a method whereby one tries to construct an understanding for the soil from which, for example, modern humans come from and in which they still live. What the genealogist asks her/himself is not what ac tually happened in history, but rather what is still alive from that time and in which special way it is living and thus contributes to the society as “struc ture”. The point of departure is always a question that is put in and about the present (Beronius 1991:14). What I want to illustrate in the archaeological discussion below by putting this question from the standpoint of a prehis WRULFDFWRULVKRZWKHSDVWLVVWLOOOLYLQJLQWKHSUHVHQWHYHQLQDORQJWHUP perspective. Not as an origin, but as the result of an unfolding production of meaning with strands of historical events interfering with the production of a new present. Instead of origins we can thus speak of a beginning that will not exclude that the beginning itself is a continuation that has inherited patterns from earlier processes, patterns that are varied in a new way and WKDWFRQWLQXRXVO\FKDQJH,WLVWKHZD\WKLVSURFHVVUHODWHVWRUHÁH[LYLW\WKDW however, is of prime interest here. Genealogy in the Late Glacial/Early Holocene The Late Glacial on the north European plain was a period of dramatic cli matic and environmental turbulence. Almost within a lifetime the climate FKDQJHGIURPZKDWPXVWKDYHEHHQWKHZDUPWKRIDERUHDOIRUHVWWRLFHFROG tundra – altering the conditions for life (Fig. 3) (after Engelmark & Buck land 2005). If, like Descola has proposed: “(…) social relationships provide DFRQFHSWXDOPRGHOIRUKXPDQQDWXUHUHODWLRQVKLSVµ 'HVFRODLQ Hornborg, 1994:9) then a changing environment must have had important consequences for the social world of these people. Once useful cultural codes became obsolete, an existential worry must have spread among the groups inhabiting the area. 7KHSDVWLQ,QJROG·VZRUGVDVDFRQFHSWXDOPR del “became alien to present ( some 13000 years ago) experience, it was not any more generative of that experience”. To trace or rewrite the genealogy of that present situation became important, a new past had to be created. 173 Kjel Knutsson Figure 3. Temperature variations in the Late GlacialEarly Holocene in relation to changes in material culture symbolism or “archaeological cultures”. The new came to be what we know as the Ahrensburgian expressed through a peculiar public material symbolism. But, as the Swedish sociologist Martin Thörn sees it, to understand how the new was new, there has to be some knowledge of the identity of the old (Thörn 1997:20). That is, whether the QHZLQLWVVSHFLÀFKLVWRULFDOSRVLWLRQZDVDUHFRQVWLWXWLRQDQGUHLQWHUSUH tation of tradition. Thus, it is now that the past may enter the arena, a past represented by relics and sacred places as vehicles for the narratives. Thus, what can we say about the relationship between past and present in that cri tical time period? I will try to give a historical perspective on that radical now WKDWZHGHÀQHDVWKH$KUHQVEXUJLDQWUDGLWLRQ7KLVWUDGLWLRQZDVWKHODWHVW RIDVHTXHQFHRIELJJDPHKXQWHUJDWKHUHUVLQWKH/DWH*ODFLDORIQRUWKHUQ Europe. Some of the earlier groups may be the historical substrate on which the Ahrensburg public symbolism was moulded. We meet these groups through the remains of their campsites now lur king under the turf of a rural landscape in northern Denmark and Germany. 2QWKHVLWHVZHÀQGUHPQDQWVRIFXOWXUDOO\GHÀQHGWHFKQRORJLFDOSUDFWLFHWK rough lithic technologies, reduction sequences or rather “chaîne de opératoire” varying over time. The lithic technological traditions we call the Hamburg, Federmesser, Bromme and Ahrensburg groups are found throughout the area; most si tes are small and probably hunting stations left there by task groups during inland hunting. Although living in a changing environment, we know that people from these groups revisited old campsites, for example Sölbjerg in VRXWKZHVWHUQ'HQPDUN (Vang Petersen & Johansen 1995:22) Settling close to but many times respecting the outline of the old camp, similar activities organized similarly in space seem to have been conducted, 174 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” as for example at Jels (Holm & Rieck 1992:Fig. 12). If not only going there to exploit the concentration of biotic “energy” through migrating animals (reindeer) steered by a maximizing strategy, why did they return? What did the Federmesser and Ahrensburgians think of that old camp and the lithics lying around? The point I want to make based on my discussion earlier in the paper but without going into any deeper analysis, is that they knew that site, they had an active relationship with it, it was at that time already part of their built environment and known past and thus a “sacred place”. The narrative of its former inhabitants was hooked on to those places in the landscape and onto those relics lying around; the latter may have functioned as actors and representations of dramatis personae in the important collective narratives. Perhaps there were even ritual gatherings where the stories were enacted and important social relations of society reproduced through rites and rituals. As the climate deteriorated and the old narratives had to be renewed, a rereading of the “frozen” material discourse represented by the sacred remains may have been one strategy to renew the exemplary material of a myth in accordance with the ongoing historical present. Did this period RIWXUPRLOUHVXOWLQDKLVWRULFDOUHÁHFWLRQDQGLIVRhow did these people reach over the abyss of time? Bridging cultural difference, bridging the abyss of time 7KURXJKH[SHULPHQWDWLRQDQGWKXVH[SHULHQFLQJÁLQWNQDSSLQJ%R0DGVHQ WKHVNLOOHG'DQLVKÁLQWNQDSSHUKDVKLPVHOIWULHGWREULGJHWKHDE\VVRIWLPH by reconstructing lithic technologies of the Hamburg, Bromme and Ahrens EXUJ FXOWXUHV WKURXJK DFWLYHO\ UHUHDGLQJ WKH VWRQHV IURP WKH SUHVHQW GD\ “Sacred places of origins”. He thereby concludes that “The marked changes in the composition of artefact types and the surprisingly varied exploitation of raw materials during the last millennia of the Ice Age is radical” (Mad sen 1995:16) (my translation). Although explained in a different theoretical context, the observations made show that the material culture in the Late Glacial when seen as public symbolism, as a material discourse explicating a cultural ethos of the groups discussed, indicates paradigmatic changes in the sphere of cultural reproduction during this period. I will not go into detail of the differences in this short paper, but looking at the chaîne opératoire of the Hamburg, Federmesser, Bromme and Ahrens burg cultures, some aspects important to the present argument can be pre sented. Looking at changes in method and technique concepts of the lithic 175 Kjel Knutsson 2ETURN OF THE OBLIQUE POINT CULTURE CULTURE 2ETURN OF THE ONE SIDED BLADE CORE Figure 4. Blade cores from the Late Glacial Hamburg and Ahrensburg cultures. Core geometry, method concept and detachment technique is identical. This may be an example of the reuse of and actual reconstruction based on relics (Hamburg material) found on 2000 year old “sacred sites” by Bromme groups thus transforming the public symbolism to what archaeologists call the Ahrensburg culture. UHGXFWLRQVWUDWHJLHVWZRWKLQJVVWDQGRXW &KDQJHVLQÁDNLQJWHFKQLTXHV indicate that the Hamburg and Ahrensburg groups used similar or identical soft impactors. 2) The core reduction method, the culturally conditioned V\VWHPRIRUJDQL]LQJÁDNHGHWDFKPHQWVDQGWKXVRIIRUPLQJWKHJHRPHWU\RI the blade core is identical in the Hamburgian and Ahrensburgian “cultures” based as it is on the unifacial, bipolar core (Fig. 4). 7KH)HGHUPHVVHUDQG%URPPHFXOWXUDOVHTXHQFHÀOOLQJWKHJDSRIF years between Hamburg and Ahrensburg cultures shows a distinctly diffe rent lithic tradition. Anders Fischer, discussing the technology of the Federmesser, Egtved ÀQGLQFRPPHQWVRQGHWDLOVRIWKHWHFKQLTXHDQGWKHÁDNLQJLPSOH ments used. 176 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” ´:LWKUHVSHFWWRWKHKDUGQHVVRIWKHÁDNLQJLPSOHPHQWWKH(JWYHGÁLQWFOHDUO\GHviates from inventories which indisputably belong to the Hamburg and Ahrensburg FXOWXUHVµ )LVFKHU By looking at the fracture indications (see Madsen 1992), he can show that the Federmesser lithics have been reduced by using hard impactors, something he shows, with reference to Bo Madsen and Söncke Hartz, is also used in the %URPPHFXOWXUH)LVFKHUIXUWKHUFRPPHQWVRQWKHPHWKRGRIÁDNHUHPRYDO from blade cores: ´2QHPXVWQDPHO\LQWKLVFDVHDOVRDVVXPHWKDWWKHÁLQWNQDSSHUTXLWHH[FHSWLRQDOO\ forgot the systematic trimming of the core edges, and that he did not as usual work with bipolar cores with striking platforms laid obliquely to each other and with blade ÁDNLQJDORQJDOLPLWHGSDUWRIWKHFRUHVLGH$OWRJHWKHULWVKRXOGEHUXOHGRXWWKDW the Egtved inventory originated in the Hamburg or Ahrensburg cultures” (Fischer The distinction between the Ahrensburgian/Hamburgians on the one hand and the Bromme/ Federmesser on the other thus did not only relate to varia tion in techniques. If the bipolar and unifacial blade core was typical of the former groups, the unipolar blade core was typical for the latter ,W FDQ EH K\SRWKHVL]HG WKDW WKH $KUHQVEXUJLDQ /DWH *ODFLDO KXQWHU JDWKHUHUV· OLWKLF WHFKQRORJ\ EODGH PDNLQJ LV D FRS\ RI WKH +DPEXUJLDQ Modus or Gestures, based in a form of formalization of the sensuous expe rience. The fact that methods and techniques are similar indicates that the past as Hamburgian relics authored some 2000 years earlier were actively “read” by the Brommian and/or Federmesser groups, and thus indicates a UHÁH[LYHYLHZRIWKHROGÁLQWVZKHUHE\WKHPXVFOHPHPRU\DFWHGDVDYHKLFOH IRU D PXFK PRUH FRQVFLRXV VWUDWHJ\ EDVHG RQ DQ ´REMHFWLÀHGµ DQDO\VLV RI ÁDNLQJ VWUDWHJ\ DQG W\SH RI LPSDFWRU 7KH WHFKQLFDO UHDGLQJ RI WKH ÁLQWV and the reconstruction of the operative chain, was however culturally inter preted and contextualized just like any authored “text”. There is no room here to expand on this argument, which would be quite possible. It can be concluded, however, that the cultural reproduction in Late Glacial northern Europe included an activated past looked on by the “skilled practioner who could tell” at least technically, within an objectifying strategy. Historicity LQFRUSRUDWHGVRPHWKLQJRIDQREMHFWLÀHGSDVW%XWDV,ZLOODUJXHWKHFROOHF tive memory might have introduced itself as real history by culture bearing QDUUDWLYHVFDUULHGE\WKHUHXVHGUHOLFVLQWKHODQGVFDSHRYHUPLOOHQQLD7KH fact that the Hamburgian tradition was reused was no coincidence but had historical roots. The need to or urge to reconstruct this technology in all its minute de tails, including the tanged points, must be understood as a measure of its 177 Kjel Knutsson importance for the Late Glacial hunters, indicating its relation to “webs of VRFLDOUHODWLRQVDQGYDOXHVWKDWDUHFHQWUDOWRWKHGD\WRGD\ UH SURGXFWLRQ RI WKHLU PDWHULDO FRQGLWLRQVµ 'REUHV %DVHG RQ WKH ÀQGV RI ODUJH amounts of reindeer corpses in dead ice hollows in the Tunnel valley close to Ahrensburg in northern Germany, some of which seem to be part of sacri ÀFHVWKH1RUZHJLDQDUFKDHRORJLVW,QJULG)XJOHVWYHGWZDQWVWRXQGHUVWDQG the reindeer as central in the cosmology of the Late Glacial Ahrensburgian reindeer hunters. She speaks of them, based on a generalizing anthropology, LQWHUPVRI´QRQKXPDQSHUVRQVµWKDWVWDQGLQDVSHFLÀFUHODWLRQVKLSWRWKH human societies: “I have interpreted the reindeer as the good thing…This implies a kinship with religious and moral connotations between man and reindeer” (Fuglestvedt 2004:5).This interpretation is especially interesting LQWKLVFRQQHFWLRQEHFDXVHLWSUHVHQWVDSRVVLEOHH[SODQDWLRQIRUWKHVSHFLÀF change in material communication we see at this time. ,WPD\KDYHEHHQWKHHQIRUFLQJFOLPDWHGHWHULRUDWLRQWKDWUHLQWURGXFHG the reindeer in the landscape of the Brommean groups. Reindeers that may have been part of the mythological social memory of times past when rein deers where still a role model for human interaction in the Hamburg cul WXUHLQWKH2OGHU'U\DV7KHUHLQWURGXFHGUHLQGHHUVLQWKH<RXQJHU'U\DV PD\ZLWKUHIHUHQFHWR0DUVKDOO6DKOLQV·VVWXG\RIWKHGUDPDWLFFKDQJHVRQ Hawaii in 1788, be called the Captain Cook of the Bromme people; an ani mal of mythical dimensions possibly representing one of the dramatis personae (Sahlins 1988) of culture bearing narratives and thus the catalyst for the new history. But exactly as the case was with Captain Cook and the Hawaiians, the attempt to activate the content of the myths introduced cultural changes. The reuse of the tanged point and the blade technology and the activated ideology that they carried, formed the new present, the Ahrensburg culture. The Hamburgian relics carried an ideology of mobility moulded on the QRZSK\VLFDOO\SUHVHQWPRELOHUHLQGHHUDQGSHUKDSVVRFLDOÁX[JRLQJZLWKLW This may have been what triggered the expansionist character of the Ahrens burgians and thus explain why they entered the long journey of colonization to the north. We can see their technology/ideology as archaeological sites in the coastal areas to the north at the onset of the Holocene. Traces of pione ering settlements are found along the west Norwegian border (Waaraas 2001; Fuglestvedt 2001; 2004), expanding into the Finnmark coastal areas (Tom mesen 1996; Grydeland 2005) and as far east as the northern coast of the Kola 3HQLQVXOD :RRGPDQ 7KLVLVDOVRZKHUHWKHÀQDOSDUWRIWKLVSDSHU starts, the colonization of western and northern Scandinavia and the proces ses whereby the Late Glacial beginnings, once again mediated by the skilled FUDIWVPDQFRQWULEXWHGVWUDQGVRILQÁXHQFHRQWKHFRQVWUXFWLRQRIFROOHFWLYH identities through a materialized social memory, bridging millennia. 178 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” Aesthetics, nostalgia and cultural change in the north The past is omnipresent as sites with relics in the windblown hills in the north Norwegian coastal area today, and must have been so also in the Early +RORFHQH ZKHQ WKH ÀUVW FRORQLVW DSSHDUHG LQ WKH DUHD )LJ 8VLQJ WKH same tactics as when discussing the Late Glacial in southern Scandinavia, I will take a short look at the changes in material culture in the early part of the Holocene and relate this to the concept of historicity and genealogy in FXOWXUDOUHSURGXFWLRQ,Q%M|UQDU2OVHQ·V RYHUYLHZRIWKH)LQQPDUN prehistory, the early part of the Holocene is characterized by three distinct changes in material culture, named period I to III. Period I is the coloniza tion phase characterized by the material culture of the Ahrenburgian legacy LQFOXGLQJVLQJOHHGJHGDQGWDQJHGSRLQWV ÀJD 3KDVHWZRLVFRPSDUDEOH to the early microblade phase in southern Norway and is characterized by microblade production from conical cores and the use of local and exotic raw materials. Phase III is characterized by the appearance of single edged and tanged point together with transverse and oblique points in the assem blages. We know very little about the blade industry during this period and thus cannot discuss the details of the chaîne opératoire. It is from this last pe riod III, however, that I will look at the possible reuse of the past as relics. It is from this standpoint that the genealogy of the tanged point, its relation WRDVSHFLÀFKLVWRULFDOWUDGLWLRQFDUULHGE\P\WKVDQGPDWHULDOFXOWXUHIRXQG RQ´VDFUHGVLWHVµDQGWKHSRVVLEOHLPSRUWDQFHRIUHÁH[LYLW\LQFXOWXUDOUHSUR duction can be discussed. Figure 5. The omnipresent past in northern Norway today. Sites with debitage from knapping events like this one from the colonist period in the Preboreal, must have been as visible to the inhabitants of this area in later periods of the prehistory and may thus have functioned as material representations of the “sacred times of origins”. 179 Kjel Knutsson In 1993, I excavated a small hunting camp in the south Swedish mountain areas close to Tärnaby in southern Lapland. The site, the remnants of a small hut with a central hearth situated on a small rock in the lake Tärna, contai QHGTXDUW]LWHÁDNLQJGHEULVIURPWKHIDEULFDWLRQRISRLQWVDQGODUJHDPRXQWV of arrow points related to this event. The formal variability was large, going IURP WZRHGJHG WDQJHG SRLQWV RYHU VLQJOH HGJHG SRLQW WR REOLTXH SRLQWV and transverse arrowheads (Knutsson 1998). The material had, due to point IRUPVVWURQJDIÀQLWLHVWRERWKSHULRG,DQGSHULRG,,,LQWKH1RUZHJLDQ FKURQRORJ\DQGLWZDVQRWXQWLO,JRWWKH&GDWHVWKDWLWZDVFOHDUWKDW the site belonged to period III, as three different analyses gave a date of c. 6500 BP (c 5400 cal BC) (Knutsson 2005: Manninen 2005). Scanning the region for comparable material it soon turned out that this period was of an “expansionist” character, as it was at this stage in the settlement of northern Norway that the inland was occupied (Olsen 1993). Apart from the known coastal sites, a few localities had been found during survey in inland northern Norway and still fewer (Devdis and Auksojavrri) excavated. Recent work in northernmost Finland has revealed yet another group of sites ZLWKVLPLODULQYHQWRU\DQG&GDWHV .DQNDQSllPDQXVFULSW0DQQLQHQ 2005). Apart from Rastklippan, a handful of surveyed and excavated sites of this time period with its characteristic lithic industries are found throughout northernmost Scandinavia today. I will now turn to the past from the van tage point of the beginning of this tradition around 7000/7500 BP (6400 cal BC) (Olsen 1993). Departing from the notion of the visibility of “the sacred past” in the north Norwegian coastal area represented by the period I lithic assemblages on sites in the coastal area, it comes as no surprise that the change in ma terial culture as shown by period III lithics has strong bearing on period I (Fig. 6:a and b). As a matter of fact, points from the two periods are at times impossible to separate from each other. If culture is materialized as public symbolism (see Ortner 1984 above), the drastic change in material culture in this period indicates a paradigmatic cultural change. The fact that period III so heavily bears on period I aesthetics seem to indicate that period III lithics actually is a material representation of an active effort to copy forms and processes found on period I sites, here, similar to what was discussed concerning the Hamburg culture above, seen as the sacred time of origins providing mnemonic devises for a mythical “history”. Assuming that paradigmatic changes around or slightly before 6500 BC IRUVRPHUHDVRQKDGIRUFHGWKHKXQWHUJDWKHUHUVRXWRIWKHLUHPEHGGHGUH SURGXFWLRQDQGWKXVFDOOHGIRUDQRVWDOJLFEDFNZDUGJD]HDQGDQREMHFWLÀ cation of culture bearing narratives, the reappearance of the oblique point tradition as shown in Fig. 6a and b may be interpreted as the result of a pe 180 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” Figure 6 a, b. Oblique arrowheads from two different periods of the north Scandinavian prehistory. Although almost identical they belong to two different material culture symbol systems separated by 2000 years. This may be an example of reuse of the aesthetics from the past whereby the sign qualities of the ancient stones represented something “known and safe”. riod of crisis, whereby the sacred times of origin were activated in attempts to reconstitute part of a crumbling world view. Through an active effort to maintain its culture bearing narratives it materialized as a socially charged technology in the present, thus once again part of the ongoing material dis FRXUVH$V-HDQHWWH:HLQHUSXWVLWLQUHODWLRQWRHIIRUWVWRUHSURGXFHRQH·V culture, the past is “activated to bring a vision of permanence into a social world that is always in the process of change”. Writing of this change in a paper in 2004 (Knutsson 2005), I did not pre VHQWDQ\DWWHPSWDWLQWHUSUHWDWLRQRIZK\WKLVUHLQVFULEHGSDVWZDVDFWLYDWHG VSHFLÀFDOO\DWWKLVWLPHLQQRUWKHUQ6FDQGLQDYLD,WLVWHPSWLQJKRZHYHULQ WKHOLJKWRI,QJULG)XJOHVWYHGW·VGLVFXVVLRQRIUHLQGHHUVDVQRQKXPDQSHU sons being part of the hunter gatherer sociality at that time, and the idea of aesthetics and its relation to ontological security, to discuss the return of the oblique arrowheads, devises used to kill reindeer, as caused by the metapho rical relation to the reindeer tied to the cosmology of these groups. In a recent paper on the early history of the reindeer in Finland the aut hors (Rankama & Ukkonen 2001) can show that mountain reindeer must have been present in northernmost Scandinavia in 11500 BP. This animal entered in the Bromme world around this time, came to northernmost Scan dinavia soon thereafter, and must have been important to early colonists of 181 Kjel Knutsson Ahrensburg descent in the area. The two authors further conclude that the forest reindeer invaded Scandinavia from the east around 6000/6500 BC. ”The wild reindeer in Finland thus appear to derive from two distinct regions and to have used two separate routes in colonizing the area. The forest reindeer migrated into Finland directly from its eastern glacial refugia in Siberia. The mountain reindeer that invaded northernmost Finland from the northwest were descendants of the European Pleistocene reindeer that had migrated northwards along the Norwegian coast freed from ice during the Late Glacial period. At present, a north-eastern migUDWLRQURXWHWKURXJKWKH.RODSHQLQVXODFDQQRWEHGHÀQLWHO\HOLPLQDWHG'XULQJWKH postglacial Climatic optimum the two subspecies met in northern Lapland, where their distribution areas overlapped. This resulted in the sharing of some morpholoJLFDOIHDWXUHVµ 5DQNDPD 8NNRQHQ Björnar Olsen has in this context proposed that it was the pine forest expan ding from the south during this period that formed the conditions for the inland settlement that we see in period III (Olsen 1993:40). If this expansion also carried with it the forest reindeer we do not know, but according to 8NNRQHQDQG5DQNDPDWKHWLPLQJLVJRRG7KHQRUWKHUQKXQWHUJDWKHUHUV at this time therefore did not only meet a new environment, but also had to DGMXVWWRDQHZQRQKXPDQDFWRUEHFDXVH ”From the point of view of the human hunter, the two subspecies represented very different challenges. They also played different roles in the overall economic systems of the human populations exploiting them. The mountain reindeer was a dominant game species for populations that largely depended on it for their winter subsistence, while the forest reindeer was for the most part only part of a much more varied diet” 5DQNDPD 8NNRQHQ In the argumentation above, the reindeer was seen as an important actor with historical roots in the cosmology of the Late Glacial hunter gatherers RQWKHQRUWKZHVW(XURSHDQSODLQ7KHIRUFHIXOFOLPDWHGHWHULRUDWLRQWKDW UHLQWURGXFHGDQHZODQGVFDSHDQGZLWKLWUHLQGHHUVDOVRUHLQWURGXFHGDP\W KRORJLFDOO\FRQVHFUDWHGUHLQGHHUKXPDQ7KHVHUHLQGHHUVEHFDPHWKH&DS tain Cook of the Brommean hunter gatherers, a mythological messenger and therefore the catalyst of the new history. But as in Captain Cook, it was the attempt to activate the mythological material to save the ideological struc ture that simultaneously introduced the winds of change. ,Q WKH QRUWK UHLQGHHUV ZHUH DOUHDG\ SUHVHQW DQG SDUW RI WKH OLIHZRUOG RIWKHHDUO\KXQWHUJDWKHUHUV,IWKHUHLQGHHUZDVUHLQWURGXFHGLQVRXWKHUQ Scandinavia in the Late Glacial, then a new type of reindeer with a totally different behaviour was introduced in northern Scandinavia c. 6000/6500 BC. Assuming with Marshall Sahlins that “if the legend heroes and the pri 182 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” me movers in history all the way down to the dramatis personae of mundane life, represent the concretization of cultural concepts and classes” (Sahlins 1988:37), then the reindeer and its mythological context may have been a mirror of the human sociality in this area. The introduction of a new actor, the forest reindeer with a different ecology must have caused problems. Here UHDOLW\WDNHVWKHIRUPRIQRQKXPDQSHUVRQV´UHLQGHHUSHRSOHµLQFRQÁLFW with the mythologically anchored ideological structure that reproduced the cultural codes, were introduced. The interpretation discussed points towards a context for the change in material communication, taking the form of the oblique arrowhead tradition. I thus present the hypothesis that changes in the natural environment intro GXFHDQHZDFWRULQWKHOLIHZRUOGRIWKHQRUWKHUQKXQWHUJDWKHUHUVRFLHWLHV DQHZ´QRQKXPDQSHUVRQµWKDWGLGQRWIROORZWKHFXOWXUDOO\GHÀQHGDQG WUDQVPLWWHGUXOHVIRUEHKDYLRXUDQGWKHUHIRUHKDVWREHUHLQVFULEHGLQWKH that present. For the archaeologist who approaches oral cultures, it is of the greatest importance to creatively use the discursive qualities in the material FXOWXUHDVH[SUHVVLRQVIRUDP\WKRORJLFDOUHZULWLQJ,QVLWXDWLRQVOLNHWKHVH the relics from past events that carried the Narratives became “documents” that had to be reinterpreted. The sequences of an active relationship to the past in the Late Glacial and Early Holocene that has been discussed here illustrate the historicity within which we all are situated. The detailed analysis and the copying of Hamburg culture lithics show how this must be understood as conducted by a person looking at the world from a detached position, formalization of a sensuous experience and in reconstruction of technical processes mimicking the mo dern lithic analyst. The Early Holocene reuse of the past is different. Here the aesthetics of cultural materials, relics, seem to be important. Common for both is the active and discursive relation to the past as relics. In traditional societies, the past is cherished and symbols appreciated be cause they contain and immortalize the experiences of earlier generations (Giddens 1997:42). Although normally embedded in cultural reproduction as “a past with us” sensu Ingold (1996), it may be integrated into the Narratives WKDWFUHDWHRUGHULQOLIH7KURXJKDIRUPRIUHÁH[LYLW\LQGD\WRGD\DFWLRQ all people in a group routinely supervise activities and thus have contact with the reasons for behaving like they do. Tradition is in this scenario a way to LQWHJUDWHWKLVUHÁH[LYHVXSHUYLVLRQRIEHKDYLRXULQWLPHDQGVSDFH7KLVSOD ces each activity in continuity between the past, the present and the future. As I have tried to show in this example by placing myself in a radical now DWWZRLPSRUWDQWDQGKLVWRULFDOO\VSHFLÀFYDQWDJHSRLQWVLQWLPHDQGVSDFH located in the late Pleistocene and early Holocene, we can see how continui WLHVWKURXJKDVHTXHQFHRIUHUHDGLQJVRIFXOWXUDOPDWHULDOVUHDSSHDUDVSDUW 183 Kjel Knutsson of a material culture within a new context as the production of a meaningful new present. The chaîne du opératoire with its materialized sequence of tech nological events and gestures, no doubt had a strong textual quality making the rereading and its incorporation into a new material discourse historically UHOHYDQW3HUKDSVWKLVFDQEHFDOOHGDQ´LPDJLQDU\WUDGLWLRQµLQ*LGGHQV·V word (Giddens 1997:43), because its identity is founded on an argument ba VHGRQUHÁH[LYLW\LQHVVHQFHDQLQYHQWHGWUDGLWLRQ%XWDQLQYHQWLRQFDUULHG by reactivated relics constructed through copying by the skilled lithic prac tioner “who could tell”. Individual interest or collective strain 6NLOOLVQRUPDOO\UHODWHGWRGLVFXVVLRQVRIFUDIWVSHFLDOL]DWLRQLQWKHUHSURGXF WLRQRIKLHUDUFKLFDOVWUXFWXUHVFRQWUROOHGE\DJJUDQGL]HUV,Q%ULDQ+D\GHQ·V ZRUOG +D\GHQ WKHDJJUDQGL]HULVDELRORJLFDO$SHUVRQWDNLQJFRQWURO over the collective knowledge in times of relaxed social control, for example in times of abundance. In this connection we could also remember John %DUUHWW·V ´FULVLV RI VXVSLFLRQµ ZKHQ KH WDONV DERXW WKH UHODWLRQ between how to go on and discursive knowledge. ”But there were moments when practioners stood apart from the world of their acWLRQVDQGORRNHGXSRQWKDWZRUOGGLVFXUVLYHO\7KH\REMHFWLÀHGFHUWDLQFRQGLWLRQVDV a strateg y for acting upon them. Such moments of analysis may have arisen when things did not work, at moments of crisis, or at moments when political authorities sought to extend their authority, to objectify, and thus to act upon, the lives of others. In such situations the dominant social agents were akin to social theorists, formulating a ritual or a political theory of their own world in an attempt to control and analyse that world through their actions. They ascribed an identity for others, objectifying such communities in the legislative schemes of political control. Hierarchically structured forms of systemic integration may therefore be expected to contain agents who objectify some aspects of the social system upon which they may then act”. Here the interest of aggrandizer individuals is seen as the force that intro duces change through reformulating how the world should be understood. Although based on different ideas of what instigates change, both see the discursive situation, which arises when conditions change, as a slot where a “dominant agent” is given the possibility to and thus takes control over the LQWHUSUHWDWLRQRIWKHSDVWWRIXUWKHU´DFWLYLWLHVLQWKHSUHVHQWIXWXUHµ7KLV GHYRWLRQWRLQWHUHVWWKHRU\KDVDWOHDVWLQ+D\GHQ·VYHUVLRQHYROXWLRQDU\XQ derpinnings in terms of the “natural” presence of aggrandizing behaviour. Strain theory as discussed above seems in this perspective on the Brom 184 A genealogy of reﬂexivity: the skilled lithic craftsman as “scientist” me/Ahrensburg tradition to give a better explanation of culture change. The social context of lithic production seems, at least in the sources available to us, the former hunting camps on the tundra, to be one of equal access to skill DQGNQRZOHGJH,UHIHUWRÁDNLQJÁRRUVDVWKH\DSSHDULQIRUH[DPSOH+DP burg culture at Jels (Holm & Rieck 1992), the Federmesser culture at Olde KROWZROGHDQG$KUHQVEXUJFXOWXUHDW6|OEMHUJ 9DQJ3HGHUVHQ -RKDQVHQ 1995). The skill necessary to reread and reproduce the technical operations of the Hamburg culture which forms the Ahrensburg idiom seems to be a group knowledge continuing to be so as the technology and its associated value system changed. $JHQHDORJ\RIUHÁH[LYLW\RQDUFKDHRORJ\ skill and modernity 0RGHUQLW\DVFULVLV KDV UHLÀHG WKH GLVFXUVLYH VLWXDWLRQ RI SDUDGLJPDWLF change and the scientists are thus ideologically, through the institutions of science, made into eternal cultural critics. Due to this enforcing ideology WKH\PXVWDFWLYHO\UHODWHWRVWUXFWXUHWKHSDVW,IWKHSDVWLVVHHQDVPDWHULDO traces, then old sites, artefacts, through their sheer materiality, set the stage for the present. Of course we do not leave modernity and its disembedding tendencies DQGWKXVLWVREMHFWLÀFDWLRQRIWKHZRUOGVLPSO\EHFDXVHLWLVDQLGHRORJ\ZKR se pointlessness has been made visible during the “science wars”. This way of understanding has too much “contributed to constituting the social world we now live in” (Giddens 1984:xxxv), and thus made the practice of “science” into a craft steered by conventions. We repeat the diagnostic question for a PRGHUQLW\WKDWUHÁHFWVDERXWLWVHOI7KLVGLVWDQWYLHZRIWKHGLVWDQWYLHZDV $OI+RUQERUJFDOOVLW LVSDUWDQGSDUFHORIPRGHUQLW\·VIXQGDPHQWDO agenda and the very writing of this paper is a physical representation of it. ,QVWHDGRIOHDYLQJPRGHUQLW\·VFDOOIRUGLVVRFLDWLRQDQGUHÁH[LYLW\EHFDXVH of its inability to account for important aspects of culture as social relation VKLSVYDOXHV\VWHPVFXOWXUDOHSLVWHPRORJLHVDQGZRUOGYLHZV,KDYHWDNHQ WKHÀUVWSHUVRQDOVWHSGLVFXVVLQJPRUHJHQHUDOO\LWVSRVVLEOHLPSRUWDQFHLQ cultural reproduction over long sweeps of time. I thereby have to reclaim the need for the problem of actuality that Kant saw, this true “wish to know”. Foucault reminded us of the importance of this message sent 300 years ago. “The subversive thinker is the one who explores the actuality of the present. They are the legitimate heirs of the Kantian critique” (Foucault 1989). That is, the question is formed by and has to be put under the changed circum VWDQFHVRIWKHSUHVHQWDQGLVWKXVDUDGLFDOL]DWLRQRIUHÁH[LYLW\DVFULWLTXH 185 Kjel Knutsson 7KLVQRWLRQRIWKHSUHVHQWWDNHVXVFORVHWR0LFKHO)RXFDXOW·VJHQHDORJLFDO method that was introduced above as a way of understanding the process whereby the past is meaningful only as a means of getting a perspective on the unfolding of history in a radical now. Following the Swedish sociologist, Thörn, it can be said that “to understand how the new is new, there has to be some knowledge of the identity of the old”. 1RGRXEWUHÁH[LYLW\LVIXQGDPHQWDOWRFXOWXUDOUHSURGXFWLRQDQGFKDQJH and although the distant past as material culture is continuously reread in UHODWLRQWRWKHSUHVHQWWKHVSHFLÀFKLVWRU\RIWKDWSDVWLVDOZD\VSDUWRIWKH rereading. The lithic craftsman, past and present, has a special relation to WKLVVSHFLÀFSDVWDVDVNLOOHGUHDGHURIWHFKQLFDORSHUDWLRQV+RZHYHUHP EHGGHGLQDVSHFLÀFKLVWRULFDOO\IRUPHGFXOWXUHWKHHPERGLHGNQRZKRZ EHFRPHVDYHKLFOHIRUFRQVWUXFWLRQRIDQHZZRUOGWKURXJKUHÁH[LYLW\DQG REMHFWLÀFDWLRQ 186 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Anders Högberg Continuity of place: actions and narratives Abstract -lUDYDOOHQLVWKHQDPHRIDEHDFKULGJHDORQJWKHVRXWKDQGVRXWKZHVWFRDVW of Scania in the southern part of Sweden. Large amounts of tool preforms, SDUWLFXODUO\ IRU 6FDQGLQDYLDQ VTXDUHVHFWLRQHG D[HV KDYH EHHQ IRXQG RQ three sites along this beach ridge. The several thousand preforms represent tool types from the Early Neolithic to the Early Bronze Age. The three sites have not been much noticed in recent archaeological research. With a basis in a discussion of action and technology these three places are analysed and interpreted as representing traditions involving repeated actions over a long period of time. It is suggested that the production and deposition of the preforms can be seen as an investment for the future. Introduction In recent years, several archaeological studies in Sweden and elsewhere have focused on the continuity of place and on phenomena that extend over long periods of time (e.g., Karsten 1994; Burström 1999; Rudebeck & Ödman 2000; Högberg 2002; Bergren & Celin 2004). In these studies, the analysis of activities according to the traditional archaeological periods is replaced by an emphasis on traditions that have persisted over a much longer time. The focus has often been on the “conspatiality” rather than on the contempora neity (Burström 1999). This perspective creates dynamics between separate and individual events and the overall impression of all these events, i.e. between individual ac tion and general patterns. Since the archaeological understanding of general patterns is built on traces of individual actions, intentionality is important. In this text, intentionality is not seen only in relation to the creativity and choices of the individual, but also in relation to the accessibility, i.e. what is available to allow action to take place (Hodder 2000:22 pp). Historical and material conditions create possibilities and establish limits. It is within social life that individual creativity has the chance to prosper. This creativity, the intentions of individuals, is part of the interaction between the individual DQGWKHFRPPXQLW\ %DUUHWW ,QWKLVZHÀQGWKHG\QDPLFVLQWKH study of conspatiality, and ideas concerning the relationship between indi vidual actions and the general context of these actions. 187 Anders Högberg The purpose of this text is to look into how similar actions, repeated over time on a beach ridge along the Scanian coast, may have created places of VSHFLDOVLJQLÀFDQFHDQGFKDUDFWHU7KHIRFXVLVRQDSDUWLFXODUFDWHJRU\RI DUWHIDFWVSUHIRUPVIRU6FDQGLQDYLDQVTXDUHVHFWLRQHGD[HVWKDWZHUHGHSRV ited on the beach ridge (called Järavallen) during the period from the Early Neolithic until the Early Bronze Age, c 4000–1700 BC. The preforms repre sent thousands of individual actions, all of which were repeatedly carried out RQVSHFLÀFORFDWLRQVGXULQJDYHU\ORQJWLPH )LJ Fig. 1. A selection of preforms from the beach ridges. Photo by Malmö Museum. The beach ridge Järavallen and the production sites -lUDYDOOHQLVWKHQDPHRIDEHDFKULGJHDORQJWKHVRXWKDQGVRXWKZHVWHUQ coast of Scania in the south of Sweden. It is situated around 5 metres above VHDOHYHODQGZDVIRUPHGE\WKHSRVWJODFLDOWUDQVJUHVVLRQVDQGUHJUHVVLRQV of the Litorina sea. The ridge mainly consists of stone, gravel and sand and LQVRPHORFDWLRQVVPDOOÁLQWQRGXOHVDUHSURPLQHQW$WWKUHHGLIIHUHQWORFD tions along the ridge, there are natural deposits of large amounts of high TXDOLW\ÁLQW +|JEHUJet al. 2001). These places are named Sibbarp, Barse bäck and Östra Torp (Fig. 2). Fig. 2. Map of Sweden and Scania with the sites Sibbarp, Barsebäck and Östra Torp indicated. 188 Continuity of place: actions and narratives 7KHVHVLWHVVKRZHYLGHQFHRIDQDEXQGDQWSURGXFWLRQRI6FDQGLQDYLDQÁLQW D[HSUHIRUPV+LJKTXDOLW\'DQLDQÁLQW +|JEHUJ 2ODXVVRQLQSULQW KDV been available and utilised for an extended tool preform production (Fig. 3). Fig. 3. At the beach ridge Järavallen at Östra Torp, geological OD\HUVRI'DQLDQÁLQWDUHYLVLEOH7KHSKRWRVKRZVWKHDXWKRU WHVWLQJWKHTXDOLW\RIWKHÁLQW3KRWRE\'HERUDK2ODXVVRQ However, what is so extraordinary and remarkable about these places is not only the evidence of the production of thousands and thousands of preforms carried out at the locations, but also the fact that large amounts of preforms have been left at these sites. During intensive exploitation of the beach ridge by extensive sand and gravel extraction during the late 19th and early 20th century, huge amounts RI SUHIRUPV IRU VTXDUHVHFWLRQHG D[HV ZHUH IRXQG .MHOOPDUN Rydbeck 1918; Hansen 1929; Högberg 2002). Thousands of preforms were collected. For example, in 1912 at Barsebäck, more than 1 000 preforms were found at one single occasion in the ridge within an area of c. 400 square me tres (Althin 1954). The preforms represent various types of axes that can be dated to periods between the Early Neolithic and the Early Bronze Age, and which were produced in accordance with different technological approaches. From this it is clear that they represent many separate actions, performed on different occasions (Fig. 4). Excavations have showed that many of the preforms were transported to settlements or other production sites in the vicinity of the EHDFK ULGJHV IRU WKH ÀQDO NQDSSLQJ RI WKH D[HV 6DUQlV 1RUG 3DXOVVRQ 2001; Högberg 2002). 189 Anders Högberg Fig. 4. A straight axe preform and a preform with a broad edge. Note WKHGLIIHUHQFHLQVL]HDQGDSSOLFDWLRQRIWKHÁDNHVFDUVEHWZHHQWKH preforms, which indicates different technological approaches. The preform to the left is c. 22 cm long. Photo by Malmö Museum. 7RGD\WKHVHÀQGVDUHVFDWWHUHGDQGPDQ\SUHIRUPVDUHLQFOXGHGLQFROOHF tions of the local farms. Circa 400, most of them from Barsebäck and Sib barp, are included in various museum collections. On several occasions, these preforms left on the beach ridges have been interpreted as discarded DQGQRQIXQFWLRQDO HJ*ORE6DORPRQVVRQ 7KHEDVLVIRUWKLV interpretation is the idea that if the preforms had been suitable for tool pro duction, they would not have been left behind at the sites: … one has to remember that it is only the discarded preforms that have been left. 7KHDGHTXDWHSUHIRUPVZHUHWDNHQDZD\ 6DORPRQVVRQ A closer study of the axe preforms in museum collections reveals, however, that most of them are of such quality that regardless of any technological or UDZPDWHULDOFRQVLGHUDWLRQVWKH\FRXOGKDYHEHHQPDGHLQWRÀQLVKHGD[HV Hence, the preforms were not left behind because they were discarded due to technological considerations or because they were not suitable for further tool production (Fig. 5). 190 Continuity of place: actions and narratives Fig. 5. Preforms from the beach ridge Järavallen at Sibbarp. Note – the two preforms to the upper right are preforms for Late Neolithic/Early Bronze Age bifacial sickles, not for axes. The preform to the lower left is c. 15 cm long. Photo by Malmö Museer. $QRWKHUDVSHFWRIWKHVHÀQGVLVWKDWWKH\KDYHEHHQLQWHUSUHWHGDVWKHUH sult of actions performed over a short period of time. The places have been labelled as either axe factories or places with ritual depositions. However, looking at the typological aspect of the preforms, it is obvious that they represent preforms for axes which can be dated to different time periods throughout the whole Neolithic and into the Bronze Age (Fig. 6). 191 Anders Högberg )LJ6FDQGLQDYLDQÁLQWD[HVFDQEHGLYLGHGLQWRGLIIHUHQWW\SRORJLFDO JURXSV7KHVHJURXSVKDYHFKURQRORJLFDOVLJQLÀFDQFH7KHÀJXUHVKRZV three different types of axes from three different time periods. The axe preforms from the beach ridges represent different types of axes from different chronological periods. From Vang Petersen (1993). Thus, thousands and thousands of fully usable and workable preforms have been produced and then left behind at these sites. The preforms represent various types of axes which can be dated to periods between the Early Neo lithic and the Early Bronze Age. For the subsequent discussion it is impor tant as a point of departure to state that: x An extended production of axe preforms has been carried out at these beach ridges. x A huge number of preforms have been left behind at these places. x The preforms were not left behind because they were discarded due to technological considerations. x The preforms were not left behind on single occasions, but through out an extensive time period. 192 Continuity of place: actions and narratives 7KHVLJQLÀFDQFHRIWKHVLWHV It is clear that the beach ridges were production sites for large amounts of axe preforms. The production was extensive and preforms were transported to other places in the vicinity for further knapping. (It is not unreasonable to assume that the preforms were also transported further away, although this hypothesis will not be discussed here.) Repeated production over an exten sive time period created places which were literally scattered with preforms. The places were dynamic in the sense of transformations through different seasons. With autumn storms, preforms were washed out of eroded parts of the ridges; in winter the ridges were covered with ice and snow; come spring, the preforms were once again visible through the melting ice and snow, rain and waves; in summer the preforms became covered with sea weed and plants. Why, then, were so many preforms left at the beach ridges? Fully func tional preforms were produced, but left behind without being processed into ÀQLVKHG D[HV ,Q RUGHU WR GLVFXVV SRVVLEOH UHDVRQV IRU WKLV , KDYH FKRVHQ to take as a point of departure the actual process of production. Although the production may be only one of several possible reasons for the repeated use of the sites, it is an activity that connects the use of the sites over time. +HQFHWKHSURGXFWLRQPD\EHVHHQDVWKH´SHUSHWXDOYDULDEOHµ7KHVLJQLÀ cance of the production and of its organization, and the way that its various purposes and aims came to material expression, probably varied over time. However, the activity that was constant at the beach ridges in the long term was the production of axe preforms. Action, production and technology Because the point of departure in this study is production, and because pro GXFWLRQLPSOLHVWHFKQRORJ\LWLVLPSRUWDQWWREULHÁ\GLVFXVVWKHEDVLVIRUWKH VWXG\RIÁLQWWHFKQRORJ\ZKLFKLVDSSOLHGKHUH7HFKQRORJ\LVFUHDWHGDQG used by people and is therefore something which joins human thought with material action (Schlanger 1994:143). Technology, i.e. the physical creation of things, is a social phenomenon. Intellectual thought is, in its action, a formu lation. The performing action is a practical thought. Intellectual thought is manifested in practical action (Schlanger 1994:143). The socially constituted thoughts which have shaped technology are manifested, in action, through the technology. This line of argument is only used here to stress the social implications of technology. The established differences between theoretical 193 Anders Högberg thoughts and practical actions (Bourdieu 1977; Broady 1989), or between NQRZOHGJHDQGNQRZKRZ 3HOHJULQ$SHO LVLPSRUWDQWWRQRWH but will not be further discussed here. $ FKDUDFWHULVWLF RI ÁLQW NQDSSLQJ LV WKDW VHYHUDO SRVVLELOLWLHV H[LVW FRQ cerning the choices of available techniques and methods for the production RIVSHFLÀFREMHFWV$PRQJWKHGLIIHUHQWVWUDWHJLHVVXLWDEOHIRUÁLQWNQDSSLQJ there is the option of choosing the most suitable for the purpose: Whilst the production of stone tools takes place within broad physical and mechanical constraints imposed by the raw material, the artisan is nevertheless capable of implementing a number of different strategies to create a particular artefact ( Ed monds 1990:57 7KHDFWXDOFKRLFHRIUDZPDWHULDODQGÁLQWNQDSSLQJVWUDWHJLHVKRZHYHU is not solely based on the evaluation of technological and methodological VXLWDELOLW\ :LWKLQ WKH FXOWXUDO IUDPHZRUN ZKHUH WKH ÁLQW NQDSSHU ZRUNV IXQFWLRQDO DQG WUDGLWLRQDO UHTXLUHPHQWV FUHDWH QHHGV IRU VSHFLÀF REMHFWV These needs control the choice of strategies concerning with which raw ma terial, techniques and methods the object will be made (Pelegrin 1990). As WKHÁLQWNQDSSHUKDVWKHSRVVLELOLW\ZLWKLQWKHIUDPHZRUNRIWKHWUDGLWLRQLQ ZKLFKKHRUVKHZRUNVWRFKRRVHDVWUDWHJ\IRUWKHWDVNWKHÁLQWNQDSSHU·V DQG WKURXJK WKH ÁLQW NQDSSHU WKH FRPPXQLW\·V DWWLWXGH WR ÁLQW DV D UDZ PDWHULDOLVPDQLIHVWHGLQWKHUHVXOWVRIWKHÁLQWNQDSSHU·VZRUNWKDWLVWRVD\ in the product and the waste material from the production. Each product is thereby the result of a chain of cultural choices in relation to function, tech nology and requirements. Each product and the handling of each product is DQLQGLFDWLRQRIDFRQFHSWLRQRIWKHÁLQWDVUDZPDWHULDODQGWKHUHIRUHWKH product gives a clue to the understanding of the many decisions that created its shape and use (Högberg 2001a, 2001b). Action is essential in technology. Through the study of actions it is pos sible to approach the culturally conditioned choices which have created the prerequisites for the production of various objects. Therefore it is important WREULHÁ\GLVFXVVWKHSUHPLVHVWKDWDUHXVHGLQWKLVFRQWH[WLQRUGHUWRVWXG\ technology and action. Axe preforms and chaîne opératoire Technological studies of complex forms of production in which action has EHHQWKHREMHFWHJWKHSURGXFWLRQRIVTXDUHVHFWLRQHGÁLQWD[HV +|JEHUJ ÁLQWGDJJHUV $SHO RUSRWWHU\ YDQGHU/HHXZ KDYHVKRZQ that specialized production is based on strategies that include several stages 194 Continuity of place: actions and narratives of production. These stages are all performed within the framework of what is available to allow action to take place. By studying actions it is possible to JDLQNQRZOHGJHRIWKHFRQGLWLRQVIRUWKHVHDFWLRQV$ZHOOHVWDEOLVKHGZD\WR study action, technology, tool production and the handling of tools is by way of chaîne opératoire (Inizan et al. 1992; van der Leeuw 1994; Eriksen 2000). Technology and action are intimately connected to the result of actions, i.e., the material culture. It is through the objects that the possibility to in YHVWLJDWHDFWLRQVLVSUHVHQWHG7KHUHIRUHLWLVLPSRUWDQWWREULHÁ\UHWXUQWR the axe preforms from the beach ridges in order to investigate how a chaîne opératoire for these objects may be analysed. First, however, we must take a look at the premises for a general chaîne opératoire for production for Scandi QDYLDQVTXDUHVHFWLRQHGD[HV )LJ )LJ$QHZO\PDGHVTXDUHVHFWLRQHGD[HWRJHWKHUZLWKWKHÁDNHV from its production. Flint knapper Thorbjørn Petersen. Photo by Anders Högberg. The various steps in axe production have been thoroughly investigated in several studies (Arnold 1981a, 1981b; Hansen & Madsen 1983; Olausson 1983a, 1983b; Madsen 1984; Nordquist 1988, 1991; Knarrström 1997; Hög EHUJ ,WLVFRPPRQWRXQGHUVWDQGWKLVSURGXFWLRQDVGLYLGHGLQWRÀYH different stages, with the accomplishment of each stage being dependent upon the results of the preceding one (Fig. 8). 195 Anders Högberg I II III A IV B I – Collection of raw material II – Knapping of preform A – Knapping of preform III – Knapping of blank IV – Axe ready for polishing B – Shaping of the axe V – Polished axe C – Polishing V C Fig. 8. Chaîne opératoireIRUD[HSURGXFWLRQZLWKÀYHVWDJHVDQGWKUHHORFDWLRQV related to each stage. Square-sectioned axes are produced with what is known as the quadrifacial method. The basis of this method is a four-sided surface ÁDNLQJWKDWUHVXOWVLQDEODQNZLWKVTXDUHGRUUHFWDQJXODUFURVVVHFWLRQ7KH production of square-sectioned axes can be divided into various working stages, with the accomplishment of each stage being dependent upon the result of that SUHFHGLQJLWIURPVWDJHRQHZKHQVXLWDEOHUDZPDWHULDOLVREWDLQHGWRVWDJHÀYH DQGWKHÀQLVKLQJSROLVKLQJRIWKHD[H Steps one and two are generally regarded as having been carried out at the raw material source, place A in Fig. 8 (Hansen & Madsen 1983; Knarrström 1997; Högberg 1999), and steps three and four as having been carried out ei ther at large production sites or at settlement sites, place B in Fig. 8 (Hansen 0DGVHQ.QDUUVWU|P+|JEHUJ 6WHSÀYHWKHÀQDOSRO ishing of the axe, is generally considered to have taken place at yet another location, place C in Fig. 8 (Hansen & Madsen 1983). 7KLVPHDQVWKDWWKHD[HSURGXFWLRQFDQEHGLYLGHGLQWRÀYHSURGXFWLRQ stages and into three locality stages, depending on the place where the dif ferent production stages were carried out (Fig. 8). If we place the preforms from the beach ridges in this chaîne opératoire, they represent stage one and two in the production chain, and, expressed in spatial terms, as belonging to place A. It is important to note in this context that the preforms were 196 Continuity of place: actions and narratives part of a chain of action until they were removed from it. In other words, it ZRXOGKDYHEHHQSRVVLEOHWRPDNHWKHVHSUHIRUPVLQWRÀQLVKHGD[HV,QRWKHU FRQWH[WVÀQGVRIVRFDOOHGV\PEROLFD[HSUHIRUPVRUD[HLPDJHVKDYHEHHQ made (e.g. Björhem & Säfvestad 1989). These objects are naturally shaped or RQO\VOLJKWO\NQDSSHGÁLQWVWKDWKDYHWKHVDPHVKDSHDVDFWXDOD[HSUHIRUPV EXWWKH\FRXOGQHYHUKDYHEHHQVKDSHGLQWRÀQLVKHGD[HVGXHWREDGTXDO LW\RULUUHJXODULWLHVLQWKHFKRVHQÁLQWQRGXOHV7KHD[HSUHIRUPVIURPWKH beach ridges are not of this type, but instead fully functional preforms. A chaîne opératoire for the preforms from the beach ridges In accordance with the chaîne opératoire presented above, it is clear that the D[H SUHIRUPV IURP WKH EHDFK ULGJHV KDG D OLIHF\FOH WKDW ZHQW IURP VWDJH two in the production chain directly to deposition. The hypothetical normal OLIHF\FOHRIDQD[HIURPSURGXFWLRQWRYDULRXVIRUPVRIXVHDQGÀQDOO\WR discard and deposition (Fig. 9), has been passed over and the preforms were GHSRVLWHGEHIRUHHYHQEHFRPLQJÀQLVKHGWRROV Fig. 9. Hypothetical normal life-cycle for an axe. The preforms from the beach ridges were physically never part of this kind of life-cycle. They were deposited before even EHFRPLQJÀQLVKHG tools. 197 Anders Högberg +RZHYHUWKHÀQGVIURPSURGXFWLRQVLWHVZLWKLQWKHEHDFKULGJHVUHYHDOWKDW other preforms from the beach ridges were kept within the production chain DQGPDGHLQWRÀQLVKHGD[HV$VDFRQVHTXHQFHRIWKLVGLVFXVVLRQWKHTXHV tion arises how a chaîne opératoire for axe production concerning axe preforms from the beach ridges may have been organised. Perhaps the preforms left behind were part of the chaîne opératoire, in which the production of axes based on preforms from the beach ridge actually contained the production of more preforms than axes, and consequently resulted in a larger production of preforms than axes? A chaîne opératoire for the production should perhaps not be seen as a linear process – from one preform to one axe? Considering the amount of preforms produced and left behind, a chaîne opératoire for the production of axes from the beach ridge preforms should perhaps be seen as consisting of an initial production of one, two or more axe preforms, resulting in the production of one polished axe (Fig. 10). Raw material Preforms Depostion of preforms at the beach ridge + Transportation of preforms to other places, for further knapping Resulting in a polished axe, ready for use 198 Fig. 10. Suggested chaîne opératoire for the production and lifecycle for the beach ridge preforms. A number of ÁLQWQRGXOHVDUHVHOHFWHG for the production of two or more preforms. One of the preforms is deposited on the beach ridge, while others are brought along WREHPDGHLQWRÀQLVKHG axes at other locations. These actions, both the deposition of preforms on the beach ridge and the transportation of preforms to other locations, were part of the chain of production for axes. Continuity of place: actions and narratives In conclusion, it may be established that the axe preforms from the beach ULGJHVFRXOGZHOOKDYHEHFRPHÀQLVKHGD[HV7KHSURGXFWLRQRID[HVLQFOXGHG the production of preforms which were removed and knapped into axes and the production of preforms which were deposited on the ridges. Concerning WKH ODWWHU WKH K\SRWKHWLFDO QRUPDO OLIHF\FOH RI D[H SUHIRUPV ZDV VNLSSHG and the preforms were deposited without having been made into axes. The beach ridges in a new light Previous interpretations of the beach ridges include a dualism between the terms axe factories and ritual sites. Here I would like to discuss the beach ULGJHVDQGWKHÀQGVRID[HSUHIRUPVZLWKWKHDPELWLRQWRXQLWHWKHVHLQWHU pretations. Central concepts in this discussion are technology and action, where technology and action are seen as culturally constituted and as mani festing the thoughts that have shaped them. Archaeological studies of prehistoric societies and of human action in the SDVWLQHYLWDEO\UDLVHTXHVWLRQVRIUDWLRQDOLW\7KHUHKDVEHHQDORQJVWDQGLQJ tradition to assume either that humans in the past had the same conception of rationality as people today, or that they were mainly irrational (Damm 1998). This has been the case concerning studies of practical actions such as production and exchange – usually regarded as rational, and concern ing abstract phenomena such as ritual and religion – usually regarded as irrational. In this study, the ambition is to look beyond these dualistic terms and to view them as integrated parts of the prehistoric society, or rather as opposites which never existed during prehistory as we see them today (e.g. Edmonds 1999). However, the concepts remain in this text, but only with the ambition to be used as analytical tools. Production sites, stock of goods and axe factories 6HYHUDOVWXGLHVKDYHVKRZQWKDWSHRSOHYDOXHDFFHVVWRKLJKTXDOLW\UDZPD terials and that various social groups make efforts to acquire such materi DOV7KHUHODWLRQVKLSEHWZHHQSODFHVRIQDWXUDODVVHWVDQGWKHVLJQLÀFDQFH of these places as raw material sources has been stressed (Højlund 1979; Hansen & Madsen 1983; Edmonds 1995). A basic reason for the use of the beach ridges for the production of axe preforms is, of course, the natural FRQGLWLRQVRIWKHVHVLWHV7KHKXJHDPRXQWRIKLJKTXDOLW\UDZPDWHULDOLQ 199 Anders Högberg the ridge was a prerequisite for the use of the place for this purpose. The production of preforms, both those which were transported elsewhere to be knapped into axes, and those which were left on the site, was extensive. Because the preforms were produced over a long time period, it is, as previously mentioned, not reasonable to interpret them as axe factories in the modern sense of the word; i.e. as “industrial production with a centrally organised division of labour and mechanised production steps, connected E\DFRPPRQH[HUWLRQDQGDLPLQJDWPDVVSURGXFWLRQµ Nationalencyklopedin 1997–98). The preforms have obviously been accumulated over a long time. %XW ZKDW ZDV WKH UHDVRQ IRU WKLV" :LWKLQ YDULRXV WHFKQRFRPSOH[HV LGHDV of availability exist. Several anthropological examples have been described, where tools and raw materials have been stored for future use (Binford 1983). The tools and raw materials have not always been stored with the intention to be used by those who stored them. Binford describes this with the term “insurance gear”, and explains the term by using the words of a Nunamiut spokesman: Every time men go out for something they have space in the pack or on the sled on the way out. Good men always say what can I carry that may help someone in the IXWXUH0D\EHWKH\GHFLGHWKDWZKHUHWKH\DUHJRLQJWKHUHLVQRÀUHZRRGVRPD\EH they take out some extra. Maybe there is no good stone for using with Strike-aLight, so maybe they take out some extra to leave out there in case somebody needs it later. In the old days … fellows always carried out shiny stones for making tools and left them all over the place so if you needed them they would be around. ( Binford 1983:271 If this tradition of solidarity is transferred to the discussion of the pre forms from the beach ridges, it would mean that an organised habit and tradition stated that supplementary preforms should always be produced and left at the site for future use. If this was the case, the preforms were actually a stock of goods, although not in a modern sense of “stock for the keeping of VHPLPDQXIDFWXUHGSURGXFWVDQGÀQLVKHGSURGXFWVLQWHQGHGIRUVDOHµ Nationalencyklopedin 1997–98). However, the argument that these preforms were produced in order to be made into axes at a later occasion is problematic, considering the great number of preforms. This number reveals that the site, after many repeated visits, must have been virtually covered with preforms. Hence, it is unlikely that the notion of availability was the reason for the production of preforms for later use. There were already enough preforms to easily pick up directly from the beach. The tradition to leave all these preforms on the beach ridge probably had another reason than to secure the future availability of preforms. 200 Continuity of place: actions and narratives Ritual deposits Water and various natural formations are often closely connected with ritu als and have been considered important symbols in human conceptions of the relation between the human being and the surrounding world (Karsten 1994; Koch 1998; Edmonds 1999; Bradley 2000; Rudebeck & Ödman 2000). Water and natural formations may be seen as representing aspects of human cosmologies (Rudebeck & Ödman 2000). The beach ridge is a place which connects these attributes, a manifested natural formation located directly by the water. The tradition to deposit objects in or in the vicinity of wet land areas during the Neolithic has been thoroughly studied. The objects thus deposited are usually interpreted in terms of ritual offerings (Svensson 1993; Karsten 1994; Hallgren et al. 1997; Koch 1998). Axes were obviously one of the typical types of objects in these depositions (Karsten 1994). It is clear that also raw material extraction and the production of axes may be interpreted in ritual terms (Edmonds 1995; Rudebeck 1998). One example WKDWPLJKWEHPHQWLRQHGLV*DEULHO&RRQH\·VVWXG\RIVRFLDODQGULWXDODV pects of axe production and axe production sites in Ireland and Great Britain (Cooney 1998). Cooney describes axe production as an activity connected to ritual and the sites, where this was carried out as permeated with ritual and symbolic aspects (Cooney 1998:110). Should we regard the preforms left at the beach ridges as the material ex SUHVVLRQRIULWXDO"$FRPPRQEDVHIRUGHÀQLWLRQRIULWXDOLV´DVWDQGDUGLVHG LQVWLWXWLRQDOLVHGEHKDYLRXUZLWKV\PEROLFVLJQLÀFDQFHLQZKLFKWKHULWXDOLV V\PEROLFLQWKHVHQVHWKDWFRQYHQWLRQDOEHKDYLRXUH[SUHVVHVDGHHSHUO\LQJ meaning of a religious, magic or other kind” (Nationalencyklopedin 1997–98). +HQFHULWXDOVDUHUHJXODWHGDQGWKHUHLVDFXOWXUDODJUHHPHQWRQWKHVLJQLÀ cance of the conventions. However, studies have shown that there may be URRPIRUVLJQLÀFDQWYDULDWLRQZKHUHDJUHHPHQWVRILQGLYLGXDOSDUWVRIWKH ritual need not necessarily exist (Damm 1998). In a study of social and ritual aspects of raw material extraction, axe production and axe use in the high ODQGVRI1HZ*XLQHD+¡MOXQGKDVVKRZQKRZWKHVLJQLÀFDQFHRID[HVZLWK in a society varies in norm and practice (Højlund 1979). Axes are reserved for men. The male axes and the use of them are associated with a complex social and ritual set of rules concerning how, when and why the axes may be used. This is the norm. However, there are examples when women have sometimes started to use old axes to cut wood within the household. Hence, the norm is challenged by practice. This practice is not normatively accepted and it is not very common, although existent. Although the norm expresses XQLW\WKHVRFLDODQGULWXDOVLJQLÀFDQFHRIWKHD[HLQWKLVVRFLHW\LVDPELJX 201 Anders Högberg ous. Rituals may be regarded as open to interpretation and it is not necessary WRDVVXPHWKHSUHVHQFHRIDJHQHUDOXQDQLPLW\7KHULWXDODQGVRFLDOVLJQLÀ cance of things and performances may change, from person to person, from FRQWH[WWRFRQWH[WDQGIURPRQHWLPHWRDQRWKHU7KHXQLÀHGLPSUHVVLRQRI the ritual is nevertheless a perception of unanimity. Anthropological studies have shown that stones, animals, celestial bod ies and various natural phenomena are often part of rituals. They are the SDUDSKHUQDOLDRIKXPDQFRVPRORJLHV HJ/pYL6WUDXVV +RZHYHUWKH structures of meaning and the internal relationship within this paraphernalia KDYHSURYHGWREHDEVWUXVHDQGGLIÀFXOWWRGHÀQH 7KHDFFXUDWHLGHQWLÀFDWLRQRIHYHU\DQLPDOSODQWVWRQHKHDYHQO\ERG\RUQDWXUDO phenomenon mentioned in myths and rituals is a complex task for which the ethnographer is rarely equipped. Even this is not however enough. It is also necessary WRNQRZWKHUROHZKLFKHDFKFXOWXUHJLYHVWKHPZLWKLQLWVRQV\VWHPRIVLJQLÀFDQFHV Of all these minute details, patiently accumulated over the centuries and faithfully transmitted from generation to generation, only a few are however actually employed IRUJLYLQJDQLPDOVRUSODQWV RUVWRQHV DVLJQLÀFDQWIXQFWLRQLQWKHV\VWHP /pYL Strauss 1962:53 p This reveals the complexity in studies of the meaning of various details in rituals, and may be perceived as discouraging. However, what is of im portance here is that the meaning of objects and details in ritual is actually stated. They are part of the ritual and they are important in the ritual, irre spective of whether the meaning is elusive or not. This fact has important consequences for this study. The task is to study the objects, the axe preforms from the beach ridge sites, as possible paraphernalia of a cosmology and as PDQLIHVWDWLRQVRIULWXDO7KHIRFXVLVQRWRQWKHVSHFLÀFVLJQLÀFDQFHRIWKHVH preforms, but rather on the material expression of rituals. The shaping of the preforms, their deposition on the ridges and the transport of selected SUHIRUPVWRRWKHUVLWHV IRUÀQDONQDSSLQJ PD\DOOEHVHHQDVLQWHJUDOSDUWV RIDQRUPDWLYHEHKDYLRXUZLWKWKHV\PEROLFVLJQLÀFDWLRQRIFRQVWLWXWLQJWKH sites as essential in the collective memory of the community. The important thing in this study is to focus on everyday actions, but also to regard the commonplace as possibly integral to ritualised behaviour. In this context, the preforms were the material expression, the paraphernalia, of this behav LRXU7KHFRVPRORJLFDOVLJQLÀFDQFHRIWKHVLWHVZDVPDQLIHVWHGLQWKHDFWLRQ to leave behind a part of the production. 202 Continuity of place: actions and narratives The beach ridge and the preforms – an expression of production and ritual It is clear that some kind of habit, custom, practice or tradition existed which created these sites, where fully functional preforms were produced and left behind on countless occasions. However, there is no reason to see this as the product of either practical/functional or ritual reasons. The function al and the ritual are often different aspects of the same context, so closely intertwined so as not to allow a separation, except for analytical purposes /pYL6WUDXVV 7KHULWXDOVRIGDLO\OLIHDOZD\VH[LVW %DUUHWW 7KH SUHVHQFH RI ÁLQW QRGXOHV DQG DFWXDO SURGXFWLRQ RI SUHIRUPV IRU D[HV ZHUHREYLRXVUHDVRQVIRUWKHVLJQLÀFDQFHRIWKHVLWHVDVUDZPDWHULDOVRXUFHV and production sites. Axe preforms were produced because useful axes were UHTXLUHG7KHSUHVHQFHRIGHELWDJHIURPD[HSURGXFWLRQHJÁDNHVDWVHW tlement sites are evidence of an extensive axe production. That axes were DOVRXVHGIRUYDULRXVWDVNVLVUHYHDOHGE\WKHPDQ\ÀQGVRIZRUQEURNHQDQG discarded axes in settlement dumps, and by the often complete and unused axes deposited in burials and as offerings in wetland areas. However, the actual leaving behind of preforms on the beach ridges must also have in volved some additional tradition. Some kind of cultural notion, a mentality, PXVWKDYHH[LVWHGZKLFKXUJHGRUVWLSXODWHGWKHÁLQWNQDSSHUVQRWRQO\WR produce, but also to leave preforms behind. Continuity of place – actions and ideas The beach ridges along the Scanian coast reveal a continuity of place, a “con spatiality”, of long duration. The actions taking place there have been guided by expectations of the place, and of the activities that have been performed there, and the expectations have come to a material expression through a VSHFLÀFVHWRIDFWLRQV:KDWFRQQHFWVWKHXVHRIWKHSODFHWKURXJKWLPHLV therefore not only the place itself, but the place together with the activities that have taken place there. However, the use of the beach ridges through time does not imply that a “continuity of place” should be comprehended as a continuous and repeated knapping of preforms, from the Early Neolithic to the Early Bronze Age. In his dissertation, Per Karsten discusses the tradition of axe offering during the Neolithic (Karsten 1994). He considers this as a persistent tradi tion, a tradition and custom which existed throughout the entire Neolithic. The way the depositing of axes was carried out, i.e. the action in itself, varied 203 Anders Högberg through time, but the general idea, the tradition, was most probably the same. Hence, the actions at the beach ridges may have varied and changed through time. Perhaps it was customary during one time period to leave behind one preform for each preform that was taken away, while during another time period it was customary to leave behind one preform for ten preforms taken away. The result might be that preforms produced and left behind during the Early Neolithic were picked up and taken away for further shaping dur ing the early Middle Neolithic. Thus, there may have been great variation in how the activities that took place at these sites were performed, although the same type of objects was involved. However, the thoughts about the place and the meaning of the actions that took place there seem to have been last ing. On repeated occasions, during a very long time period, the places have been visited with the purpose of manufacturing preforms and during this long time period, preforms have been left at the site. Consequently, the factors linking the use of the beach ridges through time were the conception of the places and the ideas concerning how to act WKHUH,KDYHIRXQGLWWHPSWLQJWRLQYHVWLJDWHKRZWKLVIRUPHGSHRSOH·VSHU ception of the place. The people who sporadically or on a daily basis visited WKHEHDFKULGJHRUPRYHGDURXQGLQLWVDUHDVKRZGLGWKH\HQYLVDJHWKHVH SODFHV" +RZ ZDV D ÀQG RI D SUHIRUP IRU DQ (DUO\ 1HROLWKLF SRLQWEXWWHG D[HFRPSUHKHQGHGE\DÁLQWNQDSSHUZKRYLVLWHGWKHSODFHGXULQJWKH/DWH 1HROLWKLFLQRUGHUWRSURGXFHDSUHIRUPIRUDEURDGHGJHGWKLFNEXWWHGD[H" How did this person “read the older design” and how did he or she perceive WKHFUDIWWKDWLWPDQLIHVWHG":KDWWKRXJKWVZHUHHYRNHGDERXWSUHYLRXVÁLQW knappers, the craft, the place and the community of which those knappers were a part? Oral tradition – narratives beyond the beach ridge Narratives and oral traditions as social interactions and as tools for com munication are important to human beings and may be regarded as general DQGFURVVFXOWXUDOSKHQRPHQD 'DXQ 7KHUHDUHHQGOHVVH[DPSOHVRI the communicative and constitutional possibilities of narratives (Fiske 1993). Existential conditions and the fundamental meanings of human life are in vestigated by way of narratives. Narratives of origins, being and the future, and their association with human beings, events, objects and places are, and KDYH DOZD\V EHHQ D IXQGDPHQWDO SDUW RI P\WKV DQG ULWXDOV /pYL6WUDXVV 1962; Bourdieu 1977; Andersson et al. 1997). Narratives are essential, both to individuals and to communities. 204 Continuity of place: actions and narratives A central point of departure in this study is the notion of places and land VFDSHVDVVRFLDOO\VLJQLÀFDQWGXULQJSUHKLVWRU\+XPDQEHLQJVPDNHVSDWLDO DUUDQJHPHQWV%DVHGRQQRUPVWKHHQWLUHHQYLURQPHQWDQGVSHFLÀFSODFHV are conceived of in terms of intention and use. The cultural organisation of the landscape manifests a spatial organisation of established meanings, norms and values: The landscape is redolent with past actions, it plays a major role in constituting a sense of history and the past, it is peopled by ancestral and spiritual entities, forms SDUWDQGSDUFHORIP\WKRORJLFDOV\VWHPVLWLVXVHGLQGHÀQLQJVRFLDOJURXSVDQGWKHLU relationship to resources ( Tilley 1994:67 3ODFHVRIUDZPDWHULDOH[WUDFWLRQDUHVLJQLÀFDQWLQWKLVFRQWH[W HJ(G monds 1995, 1999; Cooney 1998). If certain places have been ascribed mean ings that are persistent through time and if one can accept that the meanings of places are manifested in material culture, then production sites like the beach ridges discussed here may offer insights into past human thoughts (Edmonds 1999). The sites were attractive due to natural conditions, and the availability of KLJKTXDOLW\ÁLQW )LJ Fig. 11. The beach ridge at Östra 7RUSVFDWWHUHGZLWKÁLQWQRGXOHV Photo by Anders Högberg. 205 Anders Högberg 7KLVFDXVHGSHRSOHWRVHHNRXWWKHVHSODFHVLQRUGHUWRH[WUDFWWKHÁLQWIRU tool production. Repeated visits through many generations turned these places into meeting places which, by way of the craft, assembled both liv LQJDQGGHDGLQWKHVHQVHRIPHPRULHVRIDQFHVWRUV$ÁLQWNQDSSHU PDQ RU ZRPDQ YLVLWLQJ WKH SODFH DFFRPSDQLHG E\ D SURVSHFWLYH ÁLQW NQDSSHU DFKLOG ZRXOGKHUHPHHWWKHÁLQWFUDIWRIHDUOLHUJHQHUDWLRQVLQWKHVKDSH of tools and styles, hundreds of years old. In this meeting, generations of FUDIWRIÁLQWNQDSSLQJZHUHSDVVHGRQIURPROGHUREMHFWVDQGIRUPVWKURXJK WKHSUHVHQWDQGRQWRIXWXUHÁLQWNQDSSLQJ7KHVLJQLÀFDQFHRIWKHVLWHLQ WKH PLQG RI LWV XVHUV DQG YLVLWRUV ZDV WKXV YHULÀHG DQG UHSURGXFHG 7KH preforms left behind by earlier generations may in this have functioned as DUHPLQGHURIWKHVLJQLÀFDQFHRIWKHVLWHDQGRIWKHWUDGLWLRQDQDIÀUPD tion and legitimisation of the present by way of the past and a guideline for future action. Referring to the past is a strong argument in the creation of legitimacy and also constitutes a future warrant of authorities, powers and rights. Access to and the use of the places may have been manifested in this NLQGRIWUDGLWLRQ+HQFHDÁLQWNQDSSHUZKRFRXOG´UHDGµDQGXQGHUVWDQG ÁLQWWHFKQRORJ\ZDVDOVRWKHSHUVRQZKRSRVVHVVHGWKHNQRZOHGJHRIKRZ WRLQWHUSUHWWKHSDVW&RQVHTXHQWO\WKHSULYLOHJHRIWKHÁLQWNQDSSHUZDVWR have a code to the past at his or her disposal. Acknowledgement This text is a reworked version from the article “Production Sites on the Beach Ridge Järavallen. Aspect on Tool Preforms, Action, Technology, Ritual and the Continuity of Place” published in the journal Current Swedish Archaeology (Hög berg 2002). 206 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Jan Apel Skill and experimental archaeology Abstract This paper examines a way of combining the unique experience and knowl HGJHWKDWÁLQWNQDSSLQJH[SHULPHQWVSURGXFHZLWKDPRUHGLVWDQWDQGUD WLRQDOZD\RIFRQGXFWLQJUHVHDUFKZKLFKLVWKHDUFKDHRORJLVW·VSRLQWRIGH parture. In line with a French epistemological tradition, it is suggested that a relational research process, in which these two perspectives are consciously merged, is needed in order to elucidate the social aspects of technology. In RUGHUWRLOOXVWUDWHWKLVDVVHUWLRQWKH/DWH1HROLWKLFÁLQWGDJJHUSURGXFWLRQ in southern Scandinavia is studied with concepts that allow for a relational perspective. Introduction Archaeologists tend to interpret technology from two different perspectives (Dobres 1995; Roux 1999; Torrence 2001). Those inspired by Contextual Archaeology, Culture History and Phenomenology stress the importance of understanding traditional craftsmanship from “the native point of view”, ZKLOH VFLHQFHRULHQWDWHG UHVHDUFKHUV PDLQWDLQ WKDW DUFKDHRORJLFDO UHPDLQV of craft production in themselves are mute and that interpretations must be based on comparative studies conducted from a distance. I suggest that our knowledge of the past as well as the present could gain by consciously merg ing these two perspectives. This ambition is in line with the epistemological tradition in the French philosophy of science, where these two sides of the research process – realism and rationalism – are regarded as complementary LQVFLHQWLÀFUHDVRQLQJDQGWKDWWKHSRVLWLRQLQJLQVXFKFOHDUFXWHSLVWHPR ORJLFDOFRXSOHVODFNIRXQGDWLRQLQVFLHQWLÀFSUDFWLFH$FFRUGLQJO\VFLHQFHLV not entirely a rational practice or simply a question of describing empirical phenomena – it is both at the same time (Bachelard 1984:1 ff; Bourdieu et al. 1991; Broady 1991:387 ff; Bourdieu 2004). As a consequence, concepts such as “technology” and “skill” should be studied in ways that, on the one hand, DFFHSWWKHSHUVRQDOH[SHULHQFHRIWKHH[SHULPHQWDOÁLQWNQDSSHULQIRUPDQW and on the other hand, formalise these experiences with different forms of objectifying techniques. )RU VFLHQFHRULHQWDWHG DUFKDHRORJLVWV LW KDV EHHQ HVVHQWLDO WR FUHDWH LQ strumental categories that primarily are connected to the technical aspects RI VNLOO VHH EHORZ VLQFH VFLHQWLÀF FODVVLÀFDWLRQ PLJKW UHYHDO SDWWHUQV RI interest to us today, of which the prehistoric agents themselves were una 207 Jan Apel ware. A phenomenological approach, on the other hand, aims at contextual understanding and this requires that the researcher share categories with the agents whose technology is investigated. Thus, the analysis has to be based on folk categories that are embedded in conceptual skill and that are needed WRXQGHUVWDQGVNLOOIURPDFXOWXUDOVSHFLÀFSRLQWRIYLHZ7KLVLPSOLHVWKH XVHRIVRPHIRUPRIHWKQRPHWKRGRORJLFDOUHVHDUFKVWUDWHJ\WKDWVWULYHVWR wards understanding the experience of the people that are actually involved LQWKHFUDIWWKDWLVVWXGLHG'XULQJÁLQWNQDSSLQJH[SHULPHQWVIRULQVWDQFH this tradition would take the emicMXGJHPHQWVRIWKHH[SHUWÁLQWNQDSSHUIRU granted. In archaeology, these two perspectives were originally discussed in the typology debate of the 1950s and 60s (see Malmer 1965). However, while this discussion revolved around the question of whether or not archaeological types were “real” (realism) or “constructed” (rationalism) the possibility of merging these perspectives was never considered, probably because the sub MHFWRI$UFKDHRORJ\GXULQJWKLVWLPHZDVXQGHUWKHLQÁXHQFHRID´FORVHGµ positivist epistemology that denied the importance of subjective experience. As concerns experimental archaeology, the severe critique of emic approaches delivered by processual archaeologists, such as David Hearst Thomas (1986) and Lewis Binford, must be seen in this light. This may also explain why the chaîne opératoire DSSURDFKRULJLQDOO\FUHDWHGE\$QGUp/HURL*RXUKDQZLWKLQ WKHIUDPHVRIDWUXO\UHODWLRQDOHSLVWHPRORJ\ /HYL6WUDXVV WHQGHGWR be reduced to a tool for empirical descriptions of reduction sequences in its YDULRXV$QJOR6D[RQYHUVLRQV $XGRX]H.QXWVVRQ $SHOin prep). In the 1980s and 90s, a relational perspective was advocated in a debate of the archaeological use of relational versus formal analogies. In this discus VLRQWKHLPSRUWDQFHRIWKHLQVLGHSHUVSHFWLYHZDVHPSKDVLVHGSHUKDSVDV DUHVXOWRIWKHIDFWWKDWLWFRLQFLGHGZLWKWKHSRVWSURFHVVXDOWKHRUHWLFDOGH bate (Wylie 1985; Ravn 1993). While the notion that most archaeologists use VHQVXDOO\EDVHGDVZHOODVGLIIHUHQWIRUPVRIREMHFWLI\LQJWHFKQLTXHVLQWKHLU research is self evident, there are great advances to be made if these different phases in the research process are used in a consistent manner. An emic (in side) perspective needs to be combined with a modernistic (outside) perspec tive if we want to go beyond that which is purely cultural and ideological. This is not least important when social aspects of prehistory are discussed and, as I see it, it is a prerequisite for archaeology in general (Apel 2001:9). 208 Skill and experimental archaeology From substance to relation Interpretations produced by proponents of the two traditional perspectives have one thing in common: they have a substantial character whether aiming at instrumental and technically orientated descriptions – based on experi ments or distanced observation – or at an understanding of the ideological esoteric rules and regulations that surround traditional technology. In this context, the term “substantial” refers to a phenomenon that has an intrinsic value; an inherent value that can be appreciated by all humans and not only those who have knowledge of the culture in which the phenomenon occurs. 'XULQJWKHSDVW\HDUVLWKDVEHHQÀUPO\HVWDEOLVKHGE\HGXFDWLRQDOVRFLRO ogy that a homology exists between the objective assets of individuals (both symbolic and real) and their cognitive structures, i.e. their personal thoughts and opinions. For two instructive examples from Paris and Stavanger re spectively, see Bourdieu (1984) and Rosenlund (2000). These studies dem onstrate empirically that substantivist interpretations are ill suited if social issues are to be investigated, since social facts can only be relationally de ÀQHG%RXUGLHX KDVLOOXVWUDWHGWKLVSUREOHPLQKLVGLVFXVVLRQRI´WKH opinion”. An opinion is put forward by an individual and is, in this respect, based on a subjective experience. However, since the opinion is brought for ZDUGE\DSHUVRQZLWKDÀ[HGVRFLDOSRVLWLRQEDVHGRQKLVRUKHUV\PEROLF and material assets, it cannot be regarded solely as springing from the indi vidual. The opinion is inevitably coloured by the objective social position of its owner. Any research process that aims toward social interpretations should keep this in mind. Thus, neither of the two traditional approaches, that is the technical/conceptual dichotomy, can be used to properly discuss social aspects, and since one of the aims of this paper is to argue for the so cial aspects of technology, we have to choose a different path, and aim for a merging of technical and conceptual skill. Archaeologists in general accept that few artefacts have an intrinsic value independent of the position in the social space of its owner. It is surprising that the awareness of the relational value of things has not resulted in a simi lar awareness of the relational aspects of social interpretations themselves. By practising a relational research strategy we are forced to clearly articulate DQGFRPELQHWKHLQDQGRXWVLGHSHUVSHFWLYHVLQDQDO\VLVDVZHOODVLQLQWHU SUHWDWLRQ DQG EH DEOH WR XQGHUVWDQG VSHFLÀF HYHQWV LQ UHODWLRQ WR JHQHUDO processes. This may allow us to move beyond the descriptions of individual disparate historical events, leading to an “illusion of transparity” (Bourdieu et al. 1991:109) that can be produced by the inside perspective, on the one hand, and by emphasis on the outside perspective on technology and function, on the other hand. 209 Jan Apel A relational perspective on skill might be formulated as in Figure 1. In order to study the conceptual side of skill, contextual and historical information is required, while the practical side can be studied and analysed from a distant perspective. Since skill, in this context, is regarded as a social fact, it can only EHGHÀQHGDVDUHODWLRQRIWKHVHHOHPHQWV Practical experience Relation SKILL Figure 1. A relational view of skill. Concepts I will explore this possibility by using results from controlled archaeological H[SHULPHQWVRQWKHSURGXFWLRQRI/DWH1HROLWKLFÁLQWGDJJHUV7KLVWHFK nology was conducted in a society that most probably did not adhere to the economic rules of present day western societies. However, as we will see, this does not mean that the discussion lacks implications for the view of technol RJ\LQRXURZQVRFLHW\$UFKDHRORJLFDODQGHWKQRDUFKDHRORJLFDOVWXGLHVRI material culture may, in fact, contribute to an deeper understanding of how material culture affect us today simply because the distant perspective allows us to register aspects that appear natural to us in our own social sphere and, thus, become unobservable. $GHÀQLWLRQRIWHFKQRORJ\ In this context, technology is regarded as a coherent system of artefacts, be haviours and knowledge that can be handed down from one generation to DQRWKHU 6FKLIIHU 6NLER 7KLV GHÀQLWLRQ HQKDQFHV WKH LPSRU tance of the reproduction of crafts through the generations, something that pertains to the Neolithic production of daggers. The empirical example deals with the organisation of traditional craftsmanship with respect to sequence of gestures and procedures into different stages of production as well as the LQWUDDQGLQWHUVLWHVSDWLDOGLVWULEXWLRQRIWKHVHVWDJHV7KHSRLQWRIGHSDU ture is that traditional technologies were vehicles for the reproduction of a social order and we assume that the reproduction of traditional technologies through the generations was based on kinship. Convincing arguments for this has been presented elsewhere (see for instance Guglielmino et al. 1995; Shen QDQ 6WHHOH6DQWLOOR)UL]HOO6KHQQDQDQG6WRXW 210 Skill and experimental archaeology The Production of Late Neolithic Flint Daggers in southern Scandinavia 'XULQJWKHSHULRGEHWZHHQFDO%&GLDJQRVWLFÁLQWGDJJHUVZHUH produced in southern Scandinavia. Over 13,000 of these daggers are known to be distributed over a fairly large area in northern Europe (Fig. 2). Since LWLVLPSRVVLEOHWRVWXG\WKHSUHKLVWRULFSURGXFWLRQRIGDJJHUVDWÀUVWKDQG we have reconstructed the production process by conducting experiments and comparing the results with waste products and preforms from prehis WRULFVLWHV$QLQLWLDODLPLVWRGHÀQHWKHQHFHVVDU\WHFKQLFDOVNLOOVQHHGHGWR EHDEOHWRFDUU\RXWGLIIHUHQWSURGXFWLRQVWDJHVLQWKHPDNLQJRIÁLQWGDJ gers and thus forming the basis of an interpretation for the structure of the apprenticeship system that guaranteed the reproduction of the technology through at least 24 generations. )LJXUH7KHVL[PDLQ6FDQGLQDYLDQÁLQWGDJJHUW\SHV $SHO 2001:234, Fig. 8:1). 211 Jan Apel 7UDGLWLRQDOO\WKHUHDUHWZRZD\VRIGHÀQLQJVWDJHVLQWKHH[SHULPHQWDOSUR GXFWLRQ RI D ÁLQW WRRO DQG WKHVH WZR ZD\V FRUUHVSRQG WR WKH GLVWLQFWLRQ between realism and rationalism (Apel 2001:130 f.). Proponents of an inside SHUVSHFWLYHVXJJHVWWKDWVWDJHVVKRXOGEHGHÀQHGIURPWKHH[SHULPHQWDOÁLQW NQDSSHU·VSRLQWRIYLHZ7KXVDWUDQVLWLRQEHWZHHQWZRVWDJHVLVGHÀQHGE\ DPDMRUFKDQJHLQWKHWHFKQLTXHDQGPDQQHURIWKHÁLQWNQDSSHU :KLWWDNHU 1994:201). In order to be a useful tool for archaeologists working with pre KLVWRULFVLWHVWKLVFKDQJHKDVWREHVLJQLÀFDQWHQRXJKWRDIIHFWWKHSURGXF tion debitage in a recognisable way. Some researchers argue that a natural SURGXFWLRQ VWDJH RFFXUV ZKHQ D VWDJH GHÀQHG LQ WKLV ZD\ FRLQFLGHV ZLWK the completion of the mental template that is needed to secure a continued reduction (Callahan 1979; 1986). Proponents of an outside perspective, on WKHRWKHUKDQGSUHIHUWRZRUNZLWKVWDJHGHÀQLWLRQVWKDWDUHLQGHSHQGHQW RIDQ\VXEMHFWLYHO\GHÀQHGWHFKQRORJLFDOFKDQJHVDQGLQVWHDGDUHEDVHGRQD GLYLVLRQRIWKHUHGXFWLRQVHTXHQFHLQWRDUWLÀFLDOO\QRPLQDOO\GHÀQHGVWDJHV ÁDNHV VWDJHÁDNHV VWDJHHWF ,WFDQEHVXVSHFWHGWKDW YHU\IHZÁDNHW\SHVRISDUWLFXODUVWDJHVZLOOEHGHÀQHGLIVXFKDGLYLVLRQRI the reduction sequence is used on an archaeological material. This is due to the fact that researchers who consciously avoid the personal experience of good craftsmen will miss out on the deep understanding that experience gives. However, this understanding has to be balanced by a more formal ap SURDFKLILWLVWREHXVHIXOLQDEURDGHUVFLHQWLÀFFRQWH[W ,Q WKH SURGXFWLRQ H[SHULPHQWV ZLWK W\SH ,9 'DQLVK ÁLQW GDJJHUV WKDW I conducted together with Errett Callahan, eight production stages were GHÀQHG DFFRUGLQJ WR WKH emic perspective: (1) Obtaining raw material, (2) 5RXJKRXW 3ULPDU\3UHIRUP 6HFRQGDU\3UHIRUP 7HUWLDU\3UH IRUP *ULQGLQJRIEODGH 3DUDOOHOÁDNLQJDQG 5HWRXFK 7KHUHDUHWZRUHDVRQVZK\WKHVHSURGXFWLRQVWDJHVPXVWEHGHÀQHGLQWKH production debitage from the experiments. First of all, from a general epis temological point of view, in line with the introduction of this paper, it was LPSRUWDQWWRVXSSOHPHQW&DOODKDQ·VemicGHÀQLWLRQRIWKHSURGXFWLRQVWDJHV ZLWKDQLQGHSHQGHQWGHÀQLWLRQEDVHGRQDQDQDO\VLVRIWKHSURGXFWLRQGHELW DJH6XFKDGHÀQLWLRQLVDOVRQHFHVVDU\ZKHQSUHKLVWRULFVLWHVZLWKGDJJHU UHGXFWLRQVHTXHQFHVDUHGLVFXVVHGLQUHODWLRQWR&DOODKDQ·Vemic perspective. 6LQFHWKLVZDVDPDMRUDLPRIWKHH[SHULPHQWVGLDJQRVWLFÁDNHVIURPVHYHUDO RIWKHSURGXFWLRQVWDJHVZHUHGHÀQHGLQWKHSURGXFWLRQGHELWDJH DWKRURXJK GHELWDJHVWXG\RIWKHH[SHULPHQWDOÁDNHVZLOODSSHDULQ&DOODKDQ$SHO 2ODXVVRQPV 'LDJQRVWLFÁDNHVIURPVHYHUDORI&DOODKDQ·VVWDJHVKDYHWKHQ been used to identify dagger production sites in archaeological contexts as well (Callahan, Apel & Olausson ms). 212 Skill and experimental archaeology $Q LQYHVWLJDWLRQ RI SUHKLVWRULF ÁLQW GHELWDJH IURP H[FDYDWHG VLWHV LQ -XW land conducted in 1999 reveals that different bifacial production stages were conducted on different locations in the landscape (Apel 2001:199 ff.). Ac FRUGLQJO\HDUO\VWDJHV VWDJHV ZHUHFRQGXFWHGRQVHFOXGHGVSRWVQHDU WKHUDZPDWHULDOVRXUFHVDWDFHUWDLQGLVWDQFHIURPWKHVHWWOHPHQWV$W\SL FDOH[DPSOHLV)RUQ VRQ'MXUVODQG *ORE +HUHWKHNQDSSLQJÁRRU was positioned on a beach below the steep, white chalk cliffs that contain ODUJHSLHFHVRIÁLQWLQFOXGLQJDKLJKTXDOLW\VHQRQLDQYDULDQWDSSURSULDWHIRU WKHSURGXFWLRQRIÁLQWGDJJHUV,QWHUHVWLQJO\WKHSURGXFWLRQGHELWDJHIURP these sites seems to contain a high proportion of knapping errors (Olausson 2000:129; Apel 2001:189 ff.). Because of the fact that a high proportion of knapping errors can be a sign of inexperience, we might suggest that appren tices were conducting a major part of the earlier stages at these sites. The present (generative)-----------------------------------------------------------The past in the present (repetitive) To the site small beach nodules blades square axe preforms dagger preforms On the site From the site Beach nodules were turned into scrapes, stickles and borers with direct and pressure techniques. Artefacts as well as debitage were used and deposited on the site. Blades were used as harvesting knives and as preforms for transverse arrowheads. Debitage and harvesting knives were deposited on the site. Axe preforms were knapped and ground into finished thickbutted axes by artisans. Some of the debitage were used as preforms for scrapers and other tools Dagger preforms were knapped into finished lancet-shaped daggers by artisans. Some of the debitage was turned into bifacial, heart-shaped, arrowheads. Transverse arrowheads were used during war and hunting parties. Thick-butted axes as well as selected flakes were deposited in graves, hoarded and were distributed by a large scale exchange network for prestigious goods Lancet-shaped daggers, bifacial arrowheads and selected bifacial thinning flakes were deposited in graves, hoarded and were distributed by a large scale exchange network for prestigious goods )LJXUH)ORZFKDUWRIWKHÁLQWLQGXVWU\DW0\UK|MUHODWHGWRWKHWKHRUHWLFDOFRQFHSWV of Maurice Bloch’s concepts “the present” and “the past in the present” (Bloch 1989, see also Apel 2001). In short, the generative aspects of material culture are, just as the generative grammar of linguist Noam Chomsky (1957), more open to individual variation and consequently also to change. In this technological sphere we can expect a certain degree of individual variation. Within the repetitive technological sphere, normative rules that to a high degree are culturally transmitted between generations will govern the technological outcome. In this sphere, formal variation will be low since the technology also carries an important social message. 213 Jan Apel The later stages of dagger production, however, were conducted at the set tlements and the production debitage is often found secondarily deposited in depressions belonging to dwelling structures. Typical examples of such sites are Myrhöj, a settlement with evidence of at least three longhouses containing dagger preforms and production debitage (Jensen 1973; Apel 2001) and Gug, where production debitage was located in a depression, probably belonging to a longhouse similar to the ones at Myrhöj (Brøndstedt 1957; Simonsen 1982; Olausson 2000; Apel 2001). On these settlements, evidence of high quality knapping as well as knapping of lower quality has been recognised. A closer examination of the production debitage from Myrhöj revealed that the production of everyday items, as well as of more elaborate artefacts such DVWKLFNEXWWHGD[HVDQGGDJJHUVZDVFRQGXFWHGRQWKHVLWH )LJ ,WKDV DOVREHHQVXJJHVWHGWKDWWKHUDQJHLQWKHTXDOLW\RIWKHÁLQWLQGXVWU\DW*XJ investigated by Olausson (2000), may be explained in the same way (Apel 2001:199). Thus the early dagger production stages were conducted at se FOXGHGDQGSULYDWHSODFHVSUHIHUDEO\QHDUWKHQDWXUDOÁLQWVRXUFHVZKLOHWKH ÀQDOVWDJHVZHUHFRQGXFWHGRQWKHVHWWOHPHQWVLWHV,QRUGHUWRLQYHVWLJDWH LIWKLVLQWUDVLWHYDULDWLRQRIGDJJHULVUHODWHGWRGLIIHUHQWOHYHOVRIVNLOOZH now need a generalised way of evaluating the degree of skill that is based in &DOODKDQ·VH[SHULHQFH $FFRUGLQJ WR %DFKHODUG D VFLHQWLÀF EUHDNWKURXJK LV DFKLHYHG through a rupture with the folk categories connected to the object of study. In the present case, when the question concerns the degree of skill needed WRFRQGXFWGLIIHUHQWVWDJHVLQWKHSURGXFWLRQRI6FDQGLQDYLDQÁLQWGDJJHUV we need to make a clear break with the conceptions of skill that everyday language supplies. This was achieved by classifying the stages according to D .QRZOHGJHDQG E .QRZKRZWZRFRQFHSWVWKDWZHUHLQWURGXFHGLQWR DUFKDHRORJ\E\-DFTXHV3HOHJULQ 7RXVHDQHXURSV\FKRORJLFDOWHUPL nology, assumed to have transhistorical qualities, we may say that knowledge LVDGHFODUDWLYHVHPDQWLFPHPRU\IXQFWLRQZKLOHNQRZKRZFRUUHVSRQGVWR DQRQGHFODUDWLYHSURFHGXUHSHUFHSWXDOPHPRU\IXQFWLRQ )LJ LQIRUPD tion that is acquired within the body (muscle memory). A subjective judge PHQWRIWKHUHODWLYHGHJUHHRIWKHRUHWLFDONQRZOHGJHDQGSUDFWLFDONQRZKRZ was made for each production stage (Figs 5 & 6). Accordingly, the stages that rely on theoretical knowledge and that demand a lower degree of practical NQRZKRZDUHWKRVHWKDWFDQEHFDUULHGRXWE\DSSUHQWLFHVZLWKOLWWOHH[SHUL ence. Stages that require years of practical training, on the other hand (for instance stages 5, 7 and 8), can only be executed by experienced knappers. 214 Skill and experimental archaeology Figure 4. Relationship between theoretical knowledge, practical know-how and skill. The neuro-psychological scheme in the upper SDUWRIWKHÀJXUHLVIURP Nilsson (2004), the key words in the lower part RIWKHÀJXUHLVIURP$SHO (2000:59). Long-time memory Declarative memory Semantic memory Non-declarative memory Episodic memory Peceptual memory Embodied History History Knowledge (Connaissance) Communicative Explicit Conscious Conceptual Figure 5. The stages of dagger production graded according to the degree of practical know-how (from Apel 2001:42, Fig. 2:5). Figure 6. The stages of dagger production graded according to the degree of theoretical knowledge (from Apel 2001:42, Fig. 2:6). 3 Apprentice/youth 1 6 Know-how (Savoir faire) Acting Intuitive Unconscious Practical SKILL 1 2 6 2 Procedural memory 3 Learning by observation, simple recipe 4 Journeyman/young 4 5 Moderate recipe 8 7 5 Master/Old 8 7 Complex recipe It is intriguing to see that the stages that demand a high level of practical NQRZKRZZHUHDFWXDOO\FRQGXFWHGLQWKHVHWWOHPHQWV&RQYHUVHO\WKHVWDJHV that are quite easy from a practical point of view have obviously been car ULHGRXWRQVHFOXGHGVSRWVQHDUWKHUDZPDWHULDOVRXUFHV$VZHKDYHVHHQ this pattern is further strengthened by the investigations carried out on the prehistoric production debitage (Olausson 2000). 215 Jan Apel Conclusion We might be content to accept the view that it is our role, as archaeologists, WRGHVFULEHKRZDFHUWDLQWHFKQRORJ\ZDVSHUFHLYHGLQDVSHFLÀFSUHKLVWRULF context. Concerning, for instance, traditional iron production in Kenya, it would make sense to point out that iron making is surrounded by esoteric UHJXODWLRQWKDWPD\VWHPIURPDIHDURIVXSHUQDWXUDOSRZHUV²ÀUHFHUWDLQ sounds etc. – involved when iron oxide is transformed into iron. We would also have to accept that the reason for the taboos – for example the prohibi tion for women to attend the area around the furnace during smelting – is due to the belief that the presence of women will poison the ore and make LWXVHOHVV+RZHYHULIZHDUHVDWLVÀHGZLWKGHVFULELQJWKHVHFXOWXUDODVSHFWV we will have little chance of understanding social aspects that might not be apparent to the agents themselves. While I agree that technologies can only be understood through the people who engage in them (Barndon 2004:35), I do not believe that our main goal should be to understand the technology according to its own logic. If we aim at an understanding of social facts we have to move beyond pure observation and put more effort into relating the technology and its participants to their place in a larger social room. The IROORZLQJTXRWHIURP%RXUGLHX·VVWXG\RIWKHV\PEROLVPRIWKH.DE\OLDQ agrarian calendar, illustrates this point: 8QGHUVWDQGLQJ ULWXDO SUDFWLFH LV QRW D TXHVWLRQ RI GHFRGLQJ WKH LQWHUQDO ORJLF RI symbolism but restoring its practical necessity by relating it to the real condition of its genesis, that is, to the conditions in which its functions, and the means it uses to REWDLQWKHPDUHGHÀQHG(Bourdieu 1977:114). :KDWVHHPVWRXQLWHWKHSURGXFWLRQRIÁLQWGDJJHUVGXULQJWKH/DWH1HR lithic and Early Bronze Age periods in southern Scandinavia with traditional iron work is the fact that the stages in the production that involve a lesser GHJUHHRINQRZKRZZHUHFRQGXFWHGLQVHFOXGHGVSRWVLQWKHODQGVFDSH3UH sumably these stages were also surrounded by esoteric regulations (Fig. 7). It would not have been possible to argue for this interpretation without mak ing a clear break with, on the one hand, the conception of skill that ordinary ODQJXDJHSURYLGHV LHWRGLYLGH´VNLOOµLQWR´NQRZOHGJHµDQG´NQRZKRZµ and on the other hand, with the observations of the way the in which sym bolic and esoteric rules and regulations were involved in different stages RI WKH SURGXFWLRQ E\ UHODWLQJ WKHP WR WKH GHJUHH RI SUDFWLFDO NQRZKRZ involved). 216 Skill and experimental archaeology Stages conducted in public display on the settlements + 5 7 8 Master Knowledge Adult Journeyman Young adult + 6 Apprentice Youth 2 1 Know-how Stages surrounded by taboos and esoteric regulation. Conducted on secluded places )LJXUH7KHVHYHQSURGXFWLRQVWDJHV DQGDJULQGLQJVWDJH * JUDGHGDFFRUGLQJWR the degree of theoretical knowledge and practical know-how. Note that there is a negative correspondence between production stages that demand a high degree of know-how and those that involve a high degree of secrecy and presumably also esoteric regulation and taboos. 217 Jan Apel It should be obvious to us that the difference between the inside and outside SHUVSHFWLYHVFDQDOVREHIRXQGLQRXURZQVRFLHW\,WLVIDLUO\HDV\WRÀQG examples of how certain technologies are protected from insight by restrict ing the theoretical knowledge involved. I have collected several examples IURPZHVWHUQVRFLHWLHVDQGLQWKLVFRQWH[WLWPLJKWVXIÀFHWRUHIHUWRRQHRI these. I have a friend who worked at Arlanda airport, transporting baggage IURPDLUFUDIWVWRWHUPLQDOV,IEDJJDJHZHUHORVWDUHSRUWKDGWREHÀOHGRQ a special computer. This procedure was effectively guarded by a few workers ZKRFDUHIXOO\FRQWUROOHGUHFUXLWPHQW7KHRIÀFLDOOLQHZDVRIFRXUVHWKDW all were supposed to make the registrations. The registration procedure in itself was simple and demanded no sophisticated training and thus, control had to be executed by exclusion. While it may be possible to make this kind of study in our own contexts, it is easier to make them in a cultural context of which you are not a part. This is the true advantage of working from a GLVWDQWSHUVSHFWLYHDQGWKLVLVZK\DUFKDHRORJ\PD\FRQWULEXWHWRWKHÀHOGRI material culture studies. Once we avoid the temptation to only try to understand how the prehis toric agents experienced their world, thus avoiding making “accounts of ac FRXQWVµ %RXUGLHX DQGLQVWHDGSXWRXUHIIRUWVLQWRFUHDWLQJVFLHQWLÀF categories that break with folk categories, the point may very well be reached when the distance between “them” and “us”, cultural as well as chronologi cal, disappears. The knowledge gained through studying material remains of people in other contexts may then be used to study aspects of our own soci ety because we are forced to see ourselves from the outside. We might even realise that the symbolic economy in evidence in traditional technologies is uncomfortably familiar. Perhaps it is now time for archaeologists, engaged in material culture studies, to focus less on writing cultural historical interpre tation of how it once was and instead develop a study of material culture that may help us to understand ourselves through our relation to things. 218 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Leslie Harlacker Knowledge and know-how in the Oldowan: an experimental approach Abstract The successful production of stone tools requires both the empirical knowl HGJHRIVWRQHSURSHUWLHVDQGIUDFWXUHPHFKDQLFVDQGWKHSUDFWLFDO´NQRZ how” acquired only through experience. It has been argued that for sim SOH0RGH,ÁDNHSURGXFWLRQDPRQJPRGHUQKXPDQVNQRZOHGJHRIVWRQH properties and effective striking angles is more important to success than LVWKHNQRZKRZJDLQHGWKURXJKH[WHQVLYHSUDFWLFHKRZHYHULWLVXQFHUWDLQ whether this argument holds true for the earliest hominid stone toolmak ers, who were in the process of inventing lithic technology and may have relied equally on both components of skill. This issue can be explored us ing biomechanical and technological information derived from both human DQGQRQKXPDQSULPDWH ERQRER ÁLQWNQDSSHUVRIDOOVNLOOOHYHOVDOORZLQJ kinematic and kinetic parameters to be compared among novice and expe rienced knappers. If knappers of varying skill levels demonstrate similar ÁDNLQJELRPHFKDQLFVWKLVZRXOGVXSSRUWWKHK\SRWKHVLVWKDW0RGH,ÁDN ing relies primarily on knowledge, as it would imply that the difference in DUWLIDFWVSURGXFHGDPRQJWKHJURXSVLVGXHPRVWO\WRWKHQRYLFHV·ODFNRI knowledge about proper striking angles and platforms. If, however, they GHPRQVWUDWHGLIIHUHQFHVLQWKHLUÁDNLQJPHFKDQLFVWKHVLWXDWLRQEHFRPHV PRUHFRPSOLFDWHGZLWKDJUHDWHUUROHVXJJHVWHGIRUNQRZKRZ Introduction The determination of skill from archaeological evidence is a question that has received increased attention from scientists over the past several years (Roux 1994; Roux et al. 1995; Minar 2001). For researchers interested in the Palaeolithic, the primary intent in this regard has been to determine the degree of cognitive and social sophistication implied by various stone tool technologies (Wynn 1979; Wynn 1981; Gibson and Ingold 1993; Toth et al. 1993; Schick et al. 1999) as part of an overall picture of the evolution of hominid intellectual capabilities, including the origins and development of language. Although some studies have focused on the physiological aspects of toolmaking (Marzke & Shackley 1986; Hamrick et al. 1998; Marzke et al. 1998), these studies have been intended to elucidate the identity of the earli est toolmakers and potential capabilities of fossil hominids more than to shed light on their skills per se. 219 Leslie Harlacker Previous investigations into hominid technological skills have been of two main types: analyses of archaeological lithic assemblages (e.g., Kibunjia 1994; Semaw et al. 1996) and experimental investigations into the techniques necessary to produce a given range of artifacts (e.g., Toth 1985; Schick and Toth 1994; see also extensive references in Johnson 1978). Taken together, these elements allow for an estimation of the skills (and by inference, the cognitive capabilities) of the occupants of a particular site. However, the in terpretive step from skill assessment to cognitive and/or social assessment is RIWHQQRWJLYHQVXIÀFLHQWH[SOLFDWLRQ6SHFLÀFDOO\EHFDXVHVNLOODVVXFKLVQRW DOZD\VGHÀQHG RUHYHQH[SOLFLWO\PHQWLRQHG EXWLVUDWKHUXVXDOO\DVVXPHGWR EHSUHVHQWWRWKHGHJUHHWKDWDJLYHQWHFKQRORJ\LVGLIÀFXOWIRUPRGHUQH[SH rimentalists to replicate, technologies that are more complicated to modern eyes are generally taken as indicative of greater cognitive capacity and/or so cial complexity (Wynn 1979; Toth and Schick 1993; Mithen 1996). Although this is probably often the case, it does not represent a full exploration of what assessment of skill can tell us. This is especially true in the case of the earliest Lower Palaeolithic (the Oldowan), which is marked by comparatively very simple technologies whose cognitive and/or behavioral implications may not be fully assessed by a cursory assessment of the skill involved in their pro duction. In other words, simply to say that the technology is quite simple and can be produced using a relatively small skill set may be correct, but has not exhausted the interpretive possibilities. Researchers such as Jacques Pelegrin (Roux & Pelegrin 1989; Pelegrin 1993) have provided a useful interpretive model in their studies of various WLPHSHULRGV%\SURYLGLQJDZRUNLQJGHÀQLWLRQRIVNLOOGUDZQIURPUHOHYDQW literature, they allow for more nuanced interpretations of hominid cognitive and social complexity to be drawn from the lithic evidence. This paper will IROORZWKHLUH[DPSOHLQFKRRVLQJDGHÀQLWLRQRIVNLOOXSRQZKLFKWREDVHWKH discussion. As mentioned above, this approach is especially useful in analyz ing Oldowan technology. The aim here will be to further clarify the aspects of skill involved in the successful production of Oldowan lithic technology using both technological and biomechanical information. 'HÀQLWLRQRIVNLOO 6NLOO FDQ EH GHÀQHG DV WKH FDSDFLW\ WR DFKLHYH D JLYHQ JRDO XVLQJ DYDLOD ble resources, which results from some sort of learning process (Roux et al. )RUWKHSXUSRVHVRIWKLVSDSHUVNLOOZLOOEHPRUHVSHFLÀFDOO\FRQ sidered to consist of two components, knowledge and know-how, following pre vious researchers (e.g., Roux and Pelegrin 1989, Apel 2001). Knowledge can be 220 Knowledge and know-how in the Oldowan: an experimental approach understood to include the explicit, declarative information necessary for the performance of a given task (Pelegrin 1990:118) – simply stated, the things that the individual must know in order to successfully perform the task, and ZKLFKEHFRPHSDUWRIWKHNQDSSHU·VH[SOLFLWPHPRU\ $SHO .QRZO edge can be gained through the learning process, either through observation or during direct instruction, or by trial and error. Know-how, on the other hand, LVSHUKDSVPRUHGLIÀFXOWWRGHÀQH$VLPSOLHGE\$SHO LWVDEVHQFH accounts for the frustrating experience of knowing exactly what needs to be done in order to achieve a particular result, but being unable to actually do LW.QRZKRZZRXOGLQFOXGHWKLQJVVXFKDVPXVFOHPHPRU\DQGDQLQWXLWLYH “feel” for an activity that is connected with body movements (Apel 2001:28); in essence, the elements of successful performance that cannot be taught or learned but must be obtained through extensive practice (Pelegrin 1990:118), and of which the knapper may not be explicitly conscious. Knowledge and know-how: the Oldowan In order to consider Oldowan toolmaking from this perspective, it is nec HVVDU\WRÀUVWHVWDEOLVKWKHIDFWRUVWKDWLQÁXHQFHWKHSURGXFWLRQDQGIRUP of Oldowan artifacts, and to then classify those factors as being indicative SULPDULO\RIHLWKHUNQRZOHGJHRUNQRZKRZ7KHOLVWRIVXFKIDFWRUVLVIDLUO\ short, including a variety of properties intrinsic to the stone raw material, such as lithology, fracture properties, and the physical characteristics of the cobbles themselves (size, shape, amount of cortex). Other factors are intrin sic to the toolmaker, and include physical strength as well as physiological characteristics of the arm and hand (which are less variable today than in our evolutionary past). It is argued here that, at a minimum, the factors intrinsic to the stone fall under the heading of knowledge. For instance, the aspiring toolmaker needs at least a basic understanding of appropriate stone types, of conchoidal fracture, and of appropriate platforms and striking angles, in order to suc FHVVIXOO\REWDLQXVHDEOHÁDNHV7KLVLQIRUPDWLRQFRXOGEHREWDLQHGWKURXJK WULDODQGHUURU DVZDVOLNHO\WKHFDVHIRUWKHYHU\ÀUVWWRROPDNHUV RUWKURXJK a learning process that might have involved observation of and/or teaching by more experienced knappers. By contrast, physiological characteristics as well as the intuitive “feel” mentioned above would fall under the heading of NQRZKRZDORQJZLWKWKHPXVFOHPHPRU\WKDWGHYHORSVZKHQDQDFWLYLW\LV practiced repeatedly. It might be assumed from the foregoing that knowledge is more important WKDQNQRZKRZLQ2OGRZDQWRROPDNLQJEHFDXVHDUJXDEO\PRUHFRQWULEX 221 Leslie Harlacker WLQJIDFWRUVDUHDWWULEXWDEOHWRNQRZOHGJHWKDQWRNQRZKRZ7KLVKDVEHHQ suggested indirectly by Apel (2001:21), who indicates that possession of the NQRZOHGJHUHIHUHQFHGDERYHHQDEOHVDQRYLFHWRVXFFHVVIXOO\UHPRYHÁDNHV from a core, without substantial experience in swinging the hammerstone. However, it should be recalled that both the lists of factors compiled above and the argument just outlined are conceptualized in terms of modern hu mans (without, it should be noted, the element of performance consistency being taken into account). All of the potential Oldowan toolmaker candida tes were, quite literally, a different animal, and at least the earliest of them ZHUH LQYHQWLQJ WKH ÀUVW VWRQH WHFKQRORJ\ IURP VFUDWFK )RU WKHVH HDUOLHVW toolmakers, with their very different cognitive and behavioral repertoire, NQRZKRZPD\KDYHEHHQHTXDOO\DVLPSRUWDQWDVNQRZOHGJHDVWKH\EHJDQ WRJDLQFRQWURORYHUÀQHUXSSHUOLPEDQGKDQGPDQLSXODWLRQVWKDQSHUIRU med in our arboreal past. They were performing an activity – stone knapping – that may have had few analogues in their previous activity repertoire, gain LQJWKHQHFHVVDU\NQRZOHGJHDQGNQRZKRZWKURXJKUHSHDWHGWULDODQGHU ror. Of course, these early toolmakers are no longer present for observation, EXWZHGRKDYHPRGHUQJUHDWDSHV VSHFLÀFDOO\FKLPSDQ]HHVDQGERQRERV and ourselves as modern analogues from which to triangulate. 7KHSUHVHQWFRQWULEXWLRQDLPVWRDSSO\WKHDXWKRU·VFXUUHQWH[SHULPHQWDO research into Mode I technology to the question of the relative contributions RINQRZOHGJHDQGNQRZKRZWRVXFFHVVIXO2OGRZDQÁDNLQJ7KLVUHVHDUFK involves examining both biomechanical and technological data to character L]HWKHEDVLFÁDNLQJDUPVZLQJPRWLRQDQGFRUUHODWHELRPHFKDQLFDOIDFWRUV such as forces and torques with the technological attributes of the artifacts produced, among both human and nonhuman knappers of varied experience levels. The ultimate goal is to apply the results to the palaeoanthropological record, shedding light on toolmaker capabilities among different hominid WD[D$OWKRXJKSHUKDSVWKHVWXG\·VUHOHYDQFHWRTXHVWLRQVRIVNLOOLVQRWLP mediately obvious, it should be possible to use some of the results to shed OLJKWRQWKHNQRZOHGJH²NQRZKRZEUHDNGRZQIRU2OGRZDQWHFKQRORJ\ Once the patterns of both technological attributes and biomechanical dif ferences among knappers of various experience levels are determined, pre dictions related to the above question can be tested. For instance, if knappers DFURVVWKHH[SHULHQFHGLVWULEXWLRQH[KLELWVLPLODUÁDNLQJPHFKDQLFV2OGRZ DQÁDNLQJPLJKWEHDVVXPHGWRUHO\PRUHRQNQRZOHGJHWKDQRQNQRZKRZ alternatively, if the biomechanical characteristics differ among experience OHYHOV D JUHDWHU UROH IRU NQRZKRZ ZRXOG EH VXJJHVWHG 7KH UDWLRQDOH IRU WKHVH SUHGLFWLRQV LV DV IROORZV LI ÁDNLQJ PHFKDQLFV DUH VLPLODU DFURVV WKH sample, the differences in the artifacts produced would be due mostly to the knowledge brought to bear on the task (presumably, more experienced knap 222 Knowledge and know-how in the Oldowan: an experimental approach pers would have a better understanding of these factors and would therefore DFKLHYHJUHDWHUÁDNLQJVXFFHVV ,IKRZHYHUÁDNLQJPHFKDQLFVDUHGLIIHUHQW it is possible that these differences are at least partly responsible for the arti factual differences among the knappers, with the differences in mechanics EHLQJDWWULEXWDEOHWRGLIIHULQJOHYHOVRINQRZKRZDVVRFLDWHGZLWKGLIIHUHQW DPRXQWVRIWLPHVSHQWÁLQWNQDSSLQJ7KHSDWWHUQRIWHFKQRORJLFDODWWULEXWH GLIIHUHQFHVPD\DOVRVKHGOLJKWRQWKHNQRZOHGJH²NQRZKRZTXHVWLRQLI novices stand out statistically in terms of technological attributes, a greater role for knowledge would be suggested. If, alternatively, differences are seen DFURVVWKHERDUGNQRZKRZOLNHO\SOD\VDJUHDWHUUROHWKDQSUHYLRXVO\VXJ gested: given the simplicity of the knowledge involved, it is likely that mas tery of it would occur relatively early in the learning process, with further LPSURYHPHQWVLQSURGXFWVEHLQJDWWULEXWDEOHWRNQRZKRZ The remainder of this paper describes the study methodology and re views the results that are currently available from the standpoint of the issues raised above. Experimental design and methodology This study was designed to obtain data from as wide a range as possible of knappers, both human and nonhuman primate, within certain practical con VWUDLQWV )RUW\QLQH KXPDQ VXEMHFWV ZHUH UHFUXLWHG WR SDUWLFLSDWH URXJKO\ divided into three experience classes (novice, intermediate, and advanced) for sample selection purposes. Thirteen of the subjects are novices, 19 in termediates, and 17 are advanced knappers; 39 are men and 10 are women; DQGÀYHDUHOHIWKDQGHG7KHPDMRULW\RIWKHQRYLFHVDQGDIHZH[SHULHQFHG participants were recruited on the Indiana University campus, with the re PDLQGHUSDUWLFLSDWLQJDWHLWKHUDÁLQWNQDSSLQJÀHOGVFKRRORUDQ2OGRZDQ WKHPHGNQDSLQ7KHNQDSSLQJVFKRROJURXSZLOOEHRISDUWLFXODULQWHUHVWDV they participated after several days of intensive instruction by a single master ÁLQWNQDSSHU 1RYLFHVDUHGHÀQHGDVKDYLQJKDGQRSUHYLRXVNQDSSLQJH[SHULHQFHDW DOO([SHULHQFHGNQDSSHUVLQFOXGHGLQWKHVWXG\UDQJHIURPWZRZHHNV·WR \HDUV·NQDSSLQJH[SHULHQFHXQVXUSULVLQJO\WKHQWKHGLVWLQFWLRQEHWZHHQ intermediate and advanced knappers is somewhat less rigid. For subject re FUXLWPHQWDQGSUHOLPLQDU\DQDO\WLFDOSXUSRVHVLWKDVEHHQEDVHGRQVXEMHFWV· questionnaire responses and an initial technological survey of the artifacts produced. In general, knappers placed into the advanced category have had at least ten years of experience regardless of knapping intensity; those ad YDQFHGNQDSSHUVZLWKOHVVWKDQWHQ\HDUV·H[SHULHQFHWHQGWRNQDSIDUPRUH 223 Leslie Harlacker frequently than is usual for experienced subjects as a whole. Intermediates DYHUDJH\HDUV·H[SHULHQFHDQGDGYDQFHGNQDSSHUVDYHUDJH\HDUV7KH analyses reported here maintain these preliminary distinctions; however, the results suggest that it will be interesting to investigate the differences among JURXSVGHÀQHGE\PRUHÀQHO\GUDZQGLVWLQFWLRQVLQH[SHULHQFHOHYHO 7KH VDPSOH ZDV DOVR WHVWHG IRU WKH LQÁXHQFH RI SRVVLEOH FRQIRXQGLQJ factors – physical characteristics and activity patterns that might theoreti FDOO\LQÁXHQFHRUDOWHUÁDNLQJELRPHFKDQLFV DQGWKXVSRVVLEO\WKHDUWLIDFWV produced). These factors included arm and hand injuries, however minor, DQGSDUWLFLSDWLRQ LQFOXGLQJWLPHVSHQW LQDQ\QRQNQDSSLQJDFWLYLW\UHTXLU ing extensive arm and hand use (reported activities included martial arts, woodworking, and the playing of various musical instruments); these data were collected on the subject questionnaire. The three groups are statistically identical (p < 0.05) for all these factors. In addition to the human subjects, two bonobo subjects with several \HDUV·NQDSSLQJH[SHULHQFHZHUHDOVRLQFOXGHG 7RWKet al. 1993; Schick et al. 1999). One, Kanzi, has been knapping for more than ten years; his sister, Panbanisha, has been knapping for about six years. Knapping task and subject participation The knapping task in this study was simply to produce sharp, useable stone ÁDNHV (DFK VXEMHFW ZDV ÀOPHG DFFRUGLQJ WR D PHWKRG GHVFULEHG EHORZ while reducing two stone cores, one greenstone and one chert. This method ZDV FKRVHQ IRU D QXPEHU RI UHDVRQV ,W LV QRQLQYDVLYH DQG SRUWDEOH DQG its expense is relatively moderate compared to other biomechanical analysis techniques. These practical concerns were especially important with respect to the bonobo subjects; they also allowed for a larger human sample to be included. 3ULRUWRWKHÀOPHGWULDOVQRYLFHSDUWLFLSDQWVZHUHJLYHQEULHIYHUEDOLQ VWUXFWLRQFRQVLVWLQJRIWKHNQRZOHGJHQHFHVVDU\WRUHPRYHÁDNHV LHFREEOH choice, platform selection, and striking angle). The arm swing was not mod eled for them, but they were able to observe other subjects (usually no more WKDQDERXWÀYHRWKHUVXEMHFWVZHUHDYDLODEOHIRUREVHUYDWLRQE\DQ\JLYHQ novice). Novices were then required to participate in a practice session to EHFRPHIDPLOLDUZLWKWKHWDVNDQGWRHQVXUHWKDWDVDPSOHRIÁDNHVZRXOG be available for each subject; this practice session was also offered to more experienced knappers, but most declined. 6XEMHFWVFKRVHWKHVWRQHVXVHGLQWKHÀOPHGWULDOVIURPDVDPSOHRISUH PHDVXUHGSUHZHLJKHGFREEOHV7KHDXWKRUFKRVHFREEOHVIRULQFOXVLRQLQ 224 Knowledge and know-how in the Oldowan: an experimental approach WKDWVDPSOHEDVHGRQSUREDEOHHDVHRIÁDNLQJWKXVDOWKRXJKFREEOHVKDSHV varied, most cobbles presented to the subjects were not highly spherical or irregularly shaped. Although this does make it impossible to draw valid in ferences concerning cobble choice differences among knappers of different skill levels, it was considered more important to increase the probability that ÁDNHV ZRXOG EH SURGXFHG IURP DV PDQ\ FRUHV DV SRVVLEOH GXULQJ ÀOPLQJ WKHUHE\ LQFUHDVLQJ WKH VDPSOH RI ÁDNHV WKDW PLJKW EH GLUHFWO\ DWWULEXWDEOH to a given hammer strike. Hammerstones were chosen from a small sample of granite, sandstone, and greenstone cobbles of varied shape weighing bet ween 480 and 550g; this restriction was imposed in order to make the bio PHFKDQLFDOUHVXOWVPRUHHDVLO\FRPSDUDEOHDPRQJVXEMHFWV+DPPHUVWRQHV were allowed to remain in the sample offered to subjects until their mass fell below 480g, at which time they were replaced by a similar, but heavier, VWRQH$OORIWKHGHELWDJHSURGXFHGLQSUDFWLFHDQGGXULQJÀOPLQJZDVFRO lected, and all artifacts over 20 mm in maximum dimension were retained for analysis. The task parameters were similar for the bonobo subjects, except that the stone raw materials were somewhat larger than those presented to the hu man subjects. This was necessary because, due to the anatomy of their hands, bonobos are not able to comfortably manipulate and knap the smaller cores presented to the human subjects; conversely, most human subjects would ÀQGLWGLIÀFXOWWRPDQDJHVWRQHVRIDVL]HWKDWERQRERVÀQGFRPIRUWDEOH %RQRERVZHUHDOVRÀOPHGUHGXFLQJPRUHWKDQRQHFRUHSHUUDZPDWHULDO SDUWO\EHFDXVHRIGLIÀFXOWLHVLQSHUVXDGLQJWKHDSHVWRPDLQWDLQWKHPRVW desirable body orientations with respect to the camera, and partly because observation of previous video recordings of their knapping styles seemed WRUHYHDOPRUHLQWUDVXEMHFWYDULDWLRQWKDQVHHQLQW\SLFDOKXPDQNQDSSHUV making additional trials desirable. Filming: method and analysis 6XEMHFWV ZHUH ÀOPHG VLPXOWDQHRXVO\ ZLWK WZR 5HGODNH /RFDP KLJKVSHHG ÀOPFDPHUDVSRVLWLRQHGDWDSSUR[LPDWHO\DGHJUHHDQJOHWRHDFKRWKHU ÀUVWZKLOHNQDSSLQJWKHJUHHQVWRQHFRUHWKHQZKLOHNQDSSLQJWKHFKHUWFRUH The goal was to obtain at least three analyzable strikes per core per subject; FDWDORJLQJRIWKHÀOPVUHYHDOHGWKDWIRUWKHYDVWPDMRULW\RIVXEMHFWVEH tween four and six analyzable strikes per core were obtained. Camera speeds were set at 200 frames per second in order to facilitate the collection of data as close to the instant of impact as possible (in this case, data can be collected XSWRVHFRQGEHIRUHLPSDFW 2QFHGHYHORSHGÀOPVZHUHSURMHFWHGRQWR 225 Leslie Harlacker a Houston Instruments Complot digitizer, and 21 body landmarks (Fig. 1), the core, and the hammerstone were digitized every fourth frame for each trial analyzed. (This digitizing rate allows meaningful biomechanical data to be obtained while minimizing the time spent digitizing; the digitization of all frames of a trial was found not to contribute additional meaningful information.) Figure 2. The DLT control object, also known as “Sputnik.” Figure 1. Body landmarks to be digitized for each trial (only visible landmarks are indicated). 7KUHHGLPHQVLRQDO ORFDWLRQDO GDWD IRU FRUH KDPPHUVWRQH DQG ERG\ landmarks were obtained using a process known as Direct Linear Transfor PDWLRQ '/7$EGHO$]L] .DUDUD %ULHÁ\'/7LQYROYHVÀUVWÀOPLQJ DWKUHHGLPHQVLRQDOFRQWUROREMHFWRINQRZQGLPHQVLRQV )LJ SODFHGDW the approximate spot where the activity of interest will occur. The digitized coordinates of the control object points, along with the known dimensions of the object, allow a series of camera parameters to be calculated giving the relative locations and angles of the two cameras, along with other techni 226 Knowledge and know-how in the Oldowan: an experimental approach FDOLQIRUPDWLRQ 'DSHQD $QDGGLWLRQDOVHWRIWKUHHWRÀYHUHIHUHQFH points is also digitized; these points allow the calculation of a moveable ref HUHQFHIUDPH VHWRI[\]D[HV UHOHYDQWWRWKHDFWLYLW\VRWKDW'FRRU dinates need not be reported with respect to a reference frame attached to the control object. Once the camera parameters and new reference frame are computed, they are used along with digitized landmark data to pro GXFHWKUHHGLPHQVLRQDOORFDWLRQGDWD7KHVHWKUHHGLPHQVLRQDOFRRUGLQDWHV are then used to obtain biomechanical information about the subject with a computer program customized for the activity in question. Information thus obtained includes body segment locations, velocities, and accelerations KHOSIXOLQFKDUDFWHUL]LQJWKHÁDNLQJPRWLRQ YHORFLW\DQGDFFHOHUDWLRQGDWD for the hammerstone; and kinetic information such as the forces and torques produced through the arm joints, which can be related to patterns of muscle use. These data are output at time intervals of .02 second. Careful viewing RIWKHÀOPVVKRXOGHYHQWXDOO\DOORZWKHVHGDWDWREHPDWFKHGZLWKWKHH[DFW ÁDNHVGHWDFKHGE\WKHEORZIRUZKLFKWKHELRPHFKDQLFDOGDWDZHUHREWDLQHG at minimum, the biomechanical characteristics of each subject will be able to be correlated with his or her artifacts in general. Results Both the lithic and the biomechanical results presented here are of a prelimi nary nature. The analyses reported below for the experimental lithics include all artifacts (cores and debitage) produced by the human subjects during the ÀOPHGWULDOVSUDFWLFHDUWLIDFWVKDYHQRW\HWEHHQDQDO\]HG7KHELRPHFKDQL cal data presented were obtained from one advanced subject during a pilot study conducted several years ago (Dapena et alLQSUHVV DQDO\VLVRIÀOPVLV ongoing, including both human and bonobo subjects. Inclusion of bonobo lithic data must also await further analysis. Lithic Analysis A wide range of technological attributes were collected for the lithic materi als (Fig. 3). Each artifact was measured using a Mitutoyo dial caliper to the nearest .02mm, and weighed to the nearest 1g using an electronic scale; an gles were measured using a standard goniometer. All statistical calculations were performed using SPSS version 11.0. 227 Leslie Harlacker Artifact Attributes Collected Hammerstones Raw material Cores Debitage Raw material Raw material Maximum dimension (mm) Original % cortex ,GHQWLÀFDWLRQ Length (mm) 2ULJLQDOÁDNHVFDUV Maximum dimension (mm) Breadth (mm) Original maximum dim. Length (mm) Thickness (mm) Maximum dimension (mm) Breadth (mm) Weight (g; pre- & post use) Length (mm) Thickness (mm) Breadth (mm) Relative thickness (Th:Br) Thickness (mm) Elongation (Br:L) Th:Br, Th:L, Br:L Platform type Flaking mode Platform breadth Percent cortex Platform thickness ÁDNHVFDUV VFDUVIURP subject # platform scars # edges # dorsal scars Edge length, edge angle # dorsal scars from subject Max. dim. largest scar (MDLS) Dorsal cortex % MDLS from subject Core angle (ext. pl. angle) MDLS:max dimension Bulb angle (int. pl. angle) Flakes removed Weight &ODVVLÀFDWLRQ Termination type :HLJKW SUH SRVWÁDNLQJ Toth type (1987) Figure 3. Metric and nonmetric lithic attributes collected. 228 Knowledge and know-how in the Oldowan: an experimental approach Cores $WRWDORIFRUHVZHUHSURGXFHGGXULQJÀOPHGWULDOVURXJKO\ZHUH RIHDFKUDZPDWHULDO6RPHVXEMHFWVZHUHÀOPHGPRUHWKDQRQFHRZLQJWR WHFKQLFDOGLIÀFXOWLHVZLWKWKHÀOPLQJHTXLSPHQWWKXVWKHWRWDOQXPEHURI cores exceeds two per subject. Approximately one quarter of the cores were produced by novices, with the remainder evenly split between the two expe rienced groups. The analyses presented here pool both raw materials; future analyses will consider the greenstone and chert artifacts separately. Relatively few nonmetric attributes were examined for the cores; those of LQWHUHVWLQFOXGHÁDNLQJPRGH FKLHÁ\XQLIDFLDORUELIDFLDO DQGW\SRORJLFDO FODVVLÀFDWLRQ W\SHVZHUHORRVHO\EDVHGRQ/HDNH\·VW\SRORJ\ 2YHUDOOWKH UHVXOWVUHÁHFWPRUHH[WHQVLYHFRUHUHGXFWLRQDPRQJH[SHULHQFHGNQDSSHUV 3HDUVRQ FKLVTXDUH WHVWV ZHUH FDUULHG RXW FRPSDULQJ WKH WKUHH JURXSV LQ WHUPV RI ÁDNLQJ PRGH DQG W\SRORJLFDO SDWWHUQ IRU ERWK WHVWV WKH JURXSV GLIIHUHGVLJQLÀFDQWO\ SDQGSUHVSHFWLYHO\ ,QWHUPVRIÁDNLQJ PRGHWKHQRYLFHVGLGQRWÁDNHDQ\FRUHVELIDFLDOO\PRVWH[DPSOHVRIELID FLDOÁDNLQJRFFXUDPRQJDGYDQFHGNQDSSHUVZLWKDIHZH[DPSOHVDPRQJLQ termediates. Examination of type frequencies among the groups reveals that QRYLFHVKDGDPXFKKLJKHUUDWHRIDUWLIDFWVFODVVLÀHGDVFREEOHV PHDQLQJ WKDWQRÁDNHVZHUHUHPRYHG WKDQGLGHLWKHURIWKHH[SHULHQFHGJURXSVDG ditionally, most of the “tool types” such as choppers and scrapers occurred among advanced knappers, while intermediates displayed the highest rela WLYHIUHTXHQF\RIFRUHV ÁDNHVUHPRYHGEXWQRUHVHPEODQFHWRWUDGLWLRQDO “tool” types). Considering that the knapping task was simply to remove use DEOH ÁDNHV WKH JUHDWHU IUHTXHQF\ RI ELIDFLDO ÁDNLQJ DQG ODUJHU QXPEHU RI ´WRROVµDPRQJDGYDQFHGNQDSSHUVVXJJHVWVWKDWWKH\ZHUHWKHPRVWHIÀFLHQW JURXSLQWHUPVRIFRUHUHGXFWLRQRIWHQPDQDJLQJWRÁDNHPRUHWKDQRQH face of their cobbles and even produce some “tools” despite the highly con VWUDLQHGHQYLURQPHQWRIWKHÀOPHGWULDOZKHUHHDFKFREEOHZDVNQDSSHGIRU less than one minute. Intermediates only occasionally displayed this level of VNLOODQGQRYLFHVUDUHO\GLG7KHVHUHVXOWVZHUHFRQÀUPHGE\WKHUHVXOWVQRW presented here, of paired group comparisons. A total of 23 metric attributes were recorded for each core (Fig. 3). As a ÀUVWVWHSLQWKHDQDO\VLVRQHZD\$129$WHVWVZHUHFRQGXFWHGWRGHWHU PLQHZKLFKYDULDEOHVGLIIHUHGVLJQLÀFDQWO\DPRQJWKHWKUHHVXEMHFWJURXSV 6WDWLVWLFDOO\ VLJQLÀFDQW GLIIHUHQFHV S ZHUH GHWHFWHG IRU RI WKH PHWULFYDULDEOHV )LJ DQDGGLWLRQDOWKUHHYDULDEOHVKDGSYDOXHVEHORZ 6XFKYDULDEOHVZLOOEHEULHÁ\FRQVLGHUHGKHUHEHFDXVHWKHUHODWLYHVLPSOLFLW\ RI0RGH,WHFKQRORJ\PD\PDNHLWZRUWKZKLOHWRH[DPLQHQHDUVLJQLÀFDQW UHVXOWVDVZHOODVWKRVHWUDGLWLRQDOO\FRQVLGHUHGVLJQLÀFDQW 229 Leslie Harlacker Group Means, Core Metric Attributes Original % cortex Original # scars Novice Intermediate Advanced 6LJQLÀFDQFH 89.8 88.6 83.6 .136 1.9 2.3 2.2 .720 Original MD 127.6 132.4 126.0 .341 Max. dimension 122.6 124.9 111.7 .003* Length 120.3 123.8 110.6 .006* Breadth 87.3 86.1 83.5 .541 Thickness 52.2 54.6 50.1 .277 Th:Br .62 .64 .61 .570 Th:L .44 .45 .46 .880 Br:L .74 .71 .76 .221 Percent cortex 85 77 61 .000* Flake scars 4.5 5.2 6.2 .031* Scars from subject 2.5 3.0 4.1 .010* Edge angle 80 80 75 .095 Edge length 79.8 69.8 73.6 .490 Number of edges .38 .50 .71 .051* MDLS 51.1 61.8 74.6 .000* MDLS from subj. 47.8 60.2 73.4 .000* MDLS:Length .43 .51 .69 .000* MDLSsubj:L .41 .50 .68 .000* 1XPEHURIÁDNHV 4.8 5.9 9.2 .016* Original weight 692 801 744 .096 3RVWÁDNLQJZW 646 697 540 .012* )LJXUH 6LJQLÀFDQWDWOHYHORQHZD\$129$ 7KHSDWWHUQRIVWDWLVWLFDOO\VLJQLÀFDQWUHVXOWVVXJJHVWVWKDWPXFKRIWKHYDUL DWLRQDPRQJWKHWKUHHJURXSV·DUWLIDFWVFDQEHDWWULEXWHGWRWKHLQWHQVLW\RU degree of reduction. Maximum dimension, length, percent cortex, number RIÁDNHVFDUVDQGÁDNHVSHUFRUHDOOUHÁHFWWKHGHJUHHWRZKLFKHDFKFRUHZDV reduced; more tangentially, the relative and absolute largest scar dimensions do as well. The means quoted in Fig. 4 indicate that in general, more experi enced knappers tend to produce lighter cores with smaller linear dimensions, OHVVFRUWH[PRUHÁDNHVSHUFRUHDQGODUJHUVFDUV8QVXUSULVLQJO\WKHQH[SH 230 Knowledge and know-how in the Oldowan: an experimental approach rienced knappers are able to carry out much more extensive core reduction ZLWKLQWKHFRQÀQHVRIWKHH[SHULPHQWWKDQDUHQRYLFHV,WLVDOVRRILQWHUHVW WRQRWHWKDWPRUHH[SHULHQFHGNQDSSHUVDOVRWHQGHGWRSURGXFHVLJQLÀFDQWO\ more “useful” edges on their cobbles, and these edges had substantially more acute angles. These qualities are intimately related to reduction intensity as well (more reduction naturally resulting in more edges) and are mentioned here in that regard; it is unlikely that subjects were intentionally trying to produce useable edges on their cores. Although relatively little can be deduced from these results concerning cobble choice among the subjects, a few points can be made. Raw material, cobble shape, and weight (to a certain extent) were already somewhat con VWUDLQHGLQWKHVDPSOHPDGHDYDLODEOHIRUÀOPHGWULDOVVRWKHGLIIHUHQFHVZH might expect to observe in cobble choice would primarily involve the degree RI FRUWH[ DOUHDG\ UHPRYHG EHIRUH WKH ÀOPHG WULDO ([SHULHQFHG NQDSSHUV tended to choose cobbles that were slightly heavier than those chosen by QRYLFHVDQGWKHLUFREEOHVDOVRKDGVOLJKWO\PRUHSUHH[LVWLQJVFDUVWKRXJK WKHVHGLIIHUHQFHVZHUHQRWVLJQLÀFDQW7KHUHLVKRZHYHUHYLGHQFH SUHVHQW HGEHORZ WKDWDGYDQFHGNQDSSHUVPD\KDYHPDGHEHWWHUXVHRIWKHLUFREEOHV· SUHH[LVWLQJVFDUVWKDQGLGRWKHUVXEMHFWV More detailed comparisons are necessary to determine the relative con WULEXWLRQVRINQRZOHGJHDQGNQRZKRZWRWKHUHVXOWVGHVFULEHGDERYHDO though these results do indicate some differences among the groups, it is QRWSRVVLEOHWRVD\IURPWKHPZKHWKHUWKHGLIIHUHQFHVDUHGXHWRQRYLFHV· lack of knowledge or lack of practice. To this end, paired group comparisons RIPHWULFDWWULEXWHPHDQVZHUHSHUIRUPHGXVLQJWWHVWV3DLUHGJURXSWWHVWV were also done using skill level cut points – in other words, novices were compared to all experienced knappers, and advanced knappers were com pared to all other subjects. %RWKDQDO\VHVXVLQJVNLOOOHYHOFXWSRLQWVUHYHDOHGDQXPEHURIVLJQLÀFDQW differences, with the advanced vs. all others comparison revealing more dif ferences than the novice vs. experienced comparison (Fig. 5). Paired com parisons revealed that advanced knappers differ from both of the other two groups on many more variables than novices and intermediates differ from one another. Details are provided below; however, it is of interest to note at this point that the pattern just mentioned provides initial support for an LQFUHDVHGUROHIRUNQRZKRZLQ0RGH,SHUIRUPDQFH²LINQRZOHGJHZHUH substantially more important, we might expect that intermediate knappers would be more similar to advanced knappers than to novices, given the rela WLYHO\VPDOODPRXQWRINQRZOHGJHLPSOLHGE\VXFFHVVIXO0RGH,ÁDNLQJ$W least as regards the core metrics, this does not seem to be the case; intermedi ates are more similar to novices than to advanced knappers. 231 Leslie Harlacker 6LJQLÀFDQW'LIIHUHQFHVIURP0L[HG*URXS&RPSDULVRQV Novices v. Experienced Percent cortex .001 Advanced v. All Others Original % cortex .048 ÁDNHVFDUV Maximum dimension .001 Scars from subject Length .002 .000 .049 # edges .054 Percent cortex MDLS .002 ÁDNHVFDUV MDLS from subj. .001 Scars from subject .005 MDLS:Length .001 Edge angle .029 MDLSsub:L .000 # edges .024 MDLS .000 MDLS from subj. .000 1XPEHURIÁDNHV MDLS:Length .000 MDLSsub:L .000 1XPEHURIÁDNHV 3RVWÁDNLQJZHLJKW )LJXUH9DOXHVDUHVLJQLÀFDQFHUHVXOWVIURP6WXGHQW·VWWHVWV Novices vs. experienced knappers $V)LJLQGLFDWHVHYHQIRUDWHFKQRORJ\DVVHHPLQJO\VLPSOHDV0RGH,ÁDN ing, there are substantial differences in performance between those who have any experience and those with no experience whatsoever. Novice knappers were able to reduce their cores much less extensively than were experienced NQDSSHUVWKH\OHIWDVLJQLÀFDQWO\KLJKHUSHUFHQWDJHRIFRUWH[RQWKHLUFRUHV DQGSURGXFHGVLJQLÀFDQWO\IHZHUVFDUVDQGOHVVGHELWDJHSHUFRUH7KHODUJHVW ÁDNHVFDUVRQWKHLUFRUHVDUHVLJQLÀFDQWO\VPDOOHUWKDQWKHODUJHVWÁDNHVFDUV RQH[SHULHQFHGNQDSSHUV·FRUHVERWKLQDEVROXWHWHUPVDQGDVDSURSRUWLRQ RIFRUHVL]HVXJJHVWLQJWKDWWKH\DUHOHVVHIÀFLHQWWKDQH[SHULHQFHGNQDSSHUV at utilizing the surface area of the core. It is interesting to note that novices produced cores that are statistically similar to those produced by experienced knappers for a number of linear dimensions and shape ratios; the cobbles they chose are also similar in terms RI PD[LPXP GLPHQVLRQ DQG SUHH[LVWLQJ VFDU FRXQW /LQHDU GLPHQVLRQ similarities can be partially explained by the pooling of intermediates with advanced knappers, while shape similarities are expected given the nature 232 Knowledge and know-how in the Oldowan: an experimental approach of the knapping task. However, these results also might suggest that, given the indications of more extensive reduction among experienced knappers, at least some individuals with experience are reducing their cores in such a way as to have minimal impact on linear dimensions. 2QH ÀQDO DWWULEXWH PHULWV PHQWLRQ KHUH $OWKRXJK SRVWÁDNLQJ ZHLJKW GRHVQRWGLIIHUVLJQLÀFDQWO\RULJLQDOFREEOHZHLJKWLVQHDUVLJQLÀFDQW S WKLVUHVXOWZRXOGOLNHO\ULVHWRVLJQLÀFDQFHZHUHLWQRWIRURQHQRYLFH·V highly atypical choice of a cobble weighing in excess of 1 kg. Novices would be expected to choose lighter cobbles, which tend to be easier for a learner WRPDQDJHWKHIDFWWKDWWKHLUÀQLVKHGFRUHVDUHQRWOLJKWHUWKDQH[SHULHQFHG NQDSSHUV·ÀQLVKHGFRUHVSURYLGHVDQRWKHUVXSSRUWIRUWKHDUJXPHQWWKDWH[ perienced knappers reduced their cores more intensively. Advanced knappers vs. all others Advanced knappers stand out even more than do novices when compared WRWKHUHVWRIWKHVDPSOHQRWRQO\LVWKHGLIIHUHQFHVLJQLÀFDQWIRUPRUHYDUL DEOHVEXWWKHUHVXOWVWKHPVHOYHVDUHRYHUDOOPRUHKLJKO\VLJQLÀFDQW )LJ The pattern noted for the previous comparison holds here, with the addi tion of original percent cortex, maximum dimension, length, edge angle, DQGSRVWÁDNLQJZHLJKW7KHÀUVWRIWKHVHGHVHUYHVFRPPHQWDVLWUHYHDOV WKDWDGYDQFHGNQDSSHUVFKRVHFREEOHVZLWKVLJQLÀFDQWO\OHVVFRUWH[WKDQGLG the other subjects. Combined with the results to be discussed concerning platform scars, this result suggests that advanced knappers were choosing FREEOHV ZLWK DQ H\H WRZDUGV XVLQJ DQ\ SUHH[LVWLQJ ÁDNH VFDUV DV DLGV LQ WKH ÁDNLQJ SURFHVV D NQRZOHGJHEDVHG DFWLRQ WKDW DSSHDUV WR KDYH EHHQ overlooked by the intermediate knappers (novices lacked this knowledge en tirely). (Given the size of the difference in absolute terms, it is unlikely that this difference in cobble choice had a measurable effect on other variables.) As noted for the previous comparison, much of the difference between these two groups is related to reduction intensity, this time even more strik LQJO\$GYDQFHGNQDSSHUVSURGXFHFRUHVZLWKVLJQLÀFDQWO\OHVVFRUWH[PRUH ÁDNHVFDUVPRUHHGJHVDQGODUJHUÁDNHVFDUV ERWKDEVROXWHO\DQGDVDSUR SRUWLRQRIFRUHOHQJWK 7KHLUFRUHVDUHOLJKWHUDQGWKH\SURGXFHPRUHÁDNHV per core. Additionally, in terms of linear dimensions, their cores have smaller maximum dimension and length measurements; this suggests, along with the results mentioned above, that it is the intermediate knappers who are reducing cores in such a way as to leave all linear dimensions similar to novices. All of the above points to advanced knappers achieving the greatest reduction intensity. 233 Leslie Harlacker The similarities between advanced knappers and all others have primarily to do with the original size of the cobbles chosen, which is expected given the highly constrained nature of the sample made available. Ratios indicating core shape are also similar across the sample, an expected result given that the subjects were not striving to impart any particular shape to their cores. Finally, similarities in thickness and breadth seem to simply indicate that core reduction was more likely to produce differences in length than in these dimensions. Paired group comparisons $V PHQWLRQHG DERYH HDFK SRVVLEOH SDLU RI JURXSV ZDV FRPSDUHG XVLQJ W WHVWV7KHVH UHVXOWV WR D ODUJH H[WHQW FRQÀUP WKHUHVXOWVMXVW GLVFXVVHG VR they will not be reported extensively here. Paired comparisons do, however, clarify previous arguments that the intermediate knappers are driving those results, revealing that although advanced knappers produce quite different debitage from either of the other two groups, novices and intermediates are VWDWLVWLFDOO\ TXLWH VLPLODU GLIIHULQJ VLJQLÀFDQWO\ RQO\ LQ WKDW LQWHUPHGLDWHV leave less cortex behind, produce larger scars, and choose heavier cobbles to ÁDNH$OWKRXJKWKHVHUHVXOWVGRVXJJHVWWKDWLQWHUPHGLDWHVUHGXFHWKHLUFRUHV more extensively than novices do, the difference is not as profound as that between the other pairs, nor is it supported by as many individual variables. Implications Overall, the technological analysis of the cores points to a rather complex pattern of differences among the subjects. Although novices, as expected, stand out statistically from experienced knappers, advanced knappers stand out equally dramatically when compared with a pooled group of novices and LQWHUPHGLDWHV)LQHUJUDLQHGFRPSDULVRQVEHWZHHQSDLUVRINQDSSHUJURXSV reveal a surprising similarity between intermediate and novice knappers. This raises interesting implications for assessing the relative contributions RINQRZOHGJHDQGNQRZKRZWRNQDSSHUSHUIRUPDQFHDFURVVWKHH[SHULHQFH groups. It is logical to interpret the difference between novices and all ex perienced knappers as being due primarily to knowledge factors; because although novices were verbally given the necessary knowledge, it is far from certain that they effectively assimilated this knowledge into their perform ance. However, it would be unwise to apply that reasoning to interpret the difference between advanced knappers and all others. The knowledge re quired to produce Mode I artifacts is limited enough that it can be acquired 234 Knowledge and know-how in the Oldowan: an experimental approach within a short time span (Apel 2001:29), an assertion supported by the au WKRU·VSHUVRQDOH[SHULHQFHSDUWLFLSDWLQJLQDÁLQWNQDSSLQJVFKRROWKHUHIRUH it would not be reasonable to assume that intermediate knappers possessed appreciably less of this knowledge than the advanced group, especially con sidering that most intermediates had been knapping for over a month. Thus, DOWKRXJKQRYLFHVPD\QRWKDYHKDGVXIÀFLHQWNQRZOHGJHWRSHUIRUPFRQVLVW HQWO\LQWHUPHGLDWHVFHUWDLQO\GLG\HWWKH\VWLOOGLGQRWÁDNHDVVXFFHVVIXOO\DV WKHDGYDQFHGNQDSSHUVZKLFKSRLQWVWRWKHSRVVLELOLW\WKDWNQRZKRZSOD\V a larger role than previously suspected in Mode I knapping performance. This argument is supported by the similarity between novice and intermedi ate knappers; the main difference between these two groups seems to be that LQWHUPHGLDWHVSURGXFHGVOLJKWO\PRUHÁDNHVWKDQQRYLFHVZKLOHÀQLVKLQJXS with largely similar cores. Additional support for this view can be found in the results of the debitage analysis. Debitage $WRWDORIÁDNHVDQGIUDJPHQWV!PPLQPD[LPXPGLPHQVLRQZHUH collected and analyzed. The results presented here will include analyses lim LWHGWRWKHZKROHÁDNHVDVWKH\UHWDLQWKHPRVWWHFKQRORJLFDOLQIRUPDWLRQ DGGLWLRQDODQDO\VHVDUHFXUUHQWO\LQSURJUHVVWKDWLQFOXGHDOOSODWIRUPEHDU LQJGHELWDJH ZKROHÁDNHVVSOLWÁDNHVDQGSUR[LPDOIUDJPHQWV 7KHZKROHÁDNHVUHSUHVHQWRIWKHGHELWDJHDERXWRIWKHP are chert and 30% greenstone. Advanced knappers produced approximately RIWKHZKROHÁDNHV$QXPEHURIQRQPHWULFDWWULEXWHVZHUHH[DPLQHG for the sample, including raw material, platform type, termination type, DQG7RWKW\SH 7RWK &KLVTXDUHWHVWVUHYHDOHGVLJQLÀFDQWGLIIHUHQFHV among experience levels for raw material and termination type (p<.002 for both variables). Frequency data indicate that advanced knappers produced PDQ\PRUHJUHHQVWRQHÁDNHVWKDQGLGWKHRWKHUWZRJURXSVFRPELQHGERWK DEVROXWHO\DQGDVDSURSRUWLRQRIWRWDOÁDNHVSURGXFHGLWUHPDLQVWREHVHHQ whether biomechanical differences can account for this apparent perform DQFHGLIIHUHQWLDODQGE\H[WHQVLRQZKHWKHUNQRZKRZEHFRPHVPRUHLPSRU WDQWZKHQDPRUHGLIÀFXOWUDZPDWHULDOLVNQDSSHG1RYLFHVDOVRSURGXFHG more step and fewer feather terminations than did experienced knappers, VXJJHVWLQJSHUKDSVOHVVVNLOOHGSHUIRUPDQFHRWKHUWHUPLQDWLRQW\SHVZHUH too infrequent for conclusions to be drawn. $OWKRXJKWKHJURXSVGLGQRWGLIIHUVLJQLÀFDQWO\LQWHUPVRIWKHSDWWHUQ of Toth types produced, it is worth noting that advanced knappers produced ÁDNHVRIW\SH,99, QRQFRUWLFDOSODWIRUP LQODUJHUQXPEHUVWKDQGLGRWK 235 Leslie Harlacker ers, which supports the evidence previously discussed for the cores indicat ing greater reduction intensity among advanced knappers. The similarity in SODWIRUPW\SHDFURVVJURXSVPD\EHUHÁHFWLYHRIWKHVLPSOLFLW\RIWKHWHFK QRORJ\RULWPD\UHÁHFWPHFKDQLFDOFRQVWUDLQWV K\SRWKHVL]HGWRRSHUDWHE\ 'LEEOHDQG3HOFLQ WKDWPRGHOÁDNHUHPRYDODVGHWHUPLQHGE\FHUWDLQ SODWIRUPFKDUDFWHULVWLFV ZKLFKZRXOGLPSO\WKDWDPRQJWKHÁDNHVDFWXDOO\ removed, a relatively narrow range of platform traits would be expected). Group Means, Whole Flake Metric Attributes Novices Intermediates Advanced 6LJQLÀFDQFH Maximum dim. 42.7 43.2 49.6 .030* Length 34.3 35.5 40.7 .048* Breadth 34.8 36.4 38.9 .298 .094 Thickness 6.7 7.7 8.8 Rel. th (Th:Br) .20 .21 .23 .153 Elongation (Br:L) 1.1 1.1 1.1 .594 Platform breadth 21.0 20.6 23.6 .260 Platform thickness 7.3 6.4 7.6 .363 Platform scars .43 .32 .58 .013* Dorsal scars 1.9 1.6 1.7 .439 Scars from subject 1.7 1.2 1.4 .036* % dorsal cortex 42 41 42 .938 Core angle 79 83 80.6 .047* Bulb angle 102 98 103.6 .050* Weight 15 20 25 .233 )LJXUH 6LJQLÀFDQWDWOHYHORQHZD\$129$ 7KHPHWULFYDULDEOHVZHUHVXEMHFWHGÀUVWWRDRQHZD\$129$WHVWWRGLVFHUQ SDWWHUQVRIVLJQLÀFDQWGLIIHUHQFHDPRQJWKHJURXSV )LJ QRQSDUDPHWULF FKLVTXDUHWHVWVZHUHXVHGWRFRQÀUPWKHVHUHVXOWV7KHUHVXOWVLQGLFDWHVXE VWDQWLDOVLPLODULW\LQÁDNHFKDUDFWHULVWLFVDFURVVH[SHULHQFHJURXSVZKLFKLV SHUKDSVXQVXUSULVLQJJLYHQWKDWRQO\ZKROHÁDNHVZHUHFRQVLGHUHGZLWKLQ the context of a very simple experimental task; indeed, it is perhaps more remarkable that any differences could be found. These differences include maximum dimension, length, number of platform scars, number of dorsal VFDUVPDGHGXULQJÁDNLQJFRUHDQJOHDQGEXOEDQJOH)ODNHZHLJKWZDVDOVR VLJQLÀFDQWXVLQJWKHFKLVTXDUHWHVW 236 Knowledge and know-how in the Oldowan: an experimental approach ,QJHQHUDOQRYLFHVSURGXFHGVPDOOHUÁDNHVERWKLQOLQHDUGLPHQVLRQVDQG weight, although they are close to the intermediates; advanced knappers are more separated from the rest in this regard. Advanced knappers also pro GXFHGPRUHSODWIRUPVFDUVWKDQGLGWKHRWKHUJURXSVZKLFKPD\EHDUHÁHF WLRQRIWKHLUPRUHHIIHFWLYHXVHRISUHH[LVWLQJÁDNHVFDUVLQLQLWLDWLQJWKHLU RZQÁDNLQJ LHEHWWHUSODWIRUPFKRLFH 7KHVHUHVXOWVVXSSRUWWKHFRQWHQ tion from the core analysis that advanced knappers reduced their cores more intensively. Fig. 6 also reveals some surprising results. Flakes produced by novice knappers have the highest number of dorsal scars, followed by advanced and LQWHUPHGLDWHVDGGLWLRQDOO\WKHVHÁDNHVKDYHFRUHDQGEXOEDQJOHVVLPLODUWR DGYDQFHGNQDSSHUV·ÁDNHVZKLOHLQWHUPHGLDWHVKDYHPRUHREWXVHFRUHDQG more acute bulb angles. In contrast to all the other results presented thus far, these variables place novices closer to advanced knappers than to in termediates. The results regarding platform angles may be explained as due to reduction intensity differences: although all subjects had the same (very brief) amount of time to reduce their cores, advanced knappers produced substantially more debitage than did other knappers. Their average core and bulb angles may approach novice values because the highly constrained cir cumstances of the experiment did not allow them to manipulate their cores and choose platforms as carefully as they would normally, forcing them to utilize less favorable platform angles as the trials progressed. Contrary to ZKDWZHPLJKWH[SHFWWKHJUHDWHUQXPEHURIGRUVDOVFDUVIRXQGRQQRYLFHV· ÁDNHVPD\UHÁHFWWKHLULQH[SHULHQFHQRYLFHVDUHPRUHOLNHO\WRKDYHUHSHDW edly struck their cores in approximately the same place, thus guaranteeing WKDWWKHLUÁDNHVZRXOGKDYHPRUHGRUVDOVFDUVWKDQZRXOGÁDNHVUHPRYHGE\ DQH[SHULHQFHGNQDSSHUXWLOL]LQJPRUHRIWKHFREEOHVXUIDFH,WLVGLIÀFXOWWR VD\ZKHWKHUWKLVLQGLFDWHVDODFNRINQRZOHGJHRUNQRZKRZERWKDUHOLNHO\ to be at work. 'HVSLWHWKHIRUHJRLQJWKHZKROHÁDNHVSURGXFHGE\HDFKJURXSGRQRW point toward differential reduction intensity as strongly as do the core re VXOWVPRVWGLIIHUHQFHVLQZKROHÁDNHVKDYHWRGRZLWKRYHUDOOVL]HZKLFKLV JUHDWHUIRUDGYDQFHGNQDSSHUV2WKHUGLIIHUHQFHVLQWKHZKROHÁDNHVLQGLFDWH a higher, more consistent level of skilled performance among advanced knap pers (termination type, platform scars, core and bulb angles, dorsal scars). Variables that would indicate reduction intensity, such as the dorsal surface cortex percentage and Toth type pattern, are similar among the groups. For the present study, the analysis of the cores provided the best evidence of differential reduction intensity, while the debitage analysis provided other information. 237 Leslie Harlacker Implications 7KH ZKROH ÁDNH DQDO\VLV SUHVHQWHG DERYH SURYLGHV DGGLWLRQDO VXSSRUW IRU the conclusion that advanced knappers show greater reduction intensity than do less experienced knappers, and additionally furnishes evidence for more highly skilled, consistent performance among the advanced knappers. :KHWKHUWKLVGLIIHUHQFHLVGXHWRJUHDWHUNQRZOHGJHRUNQRZKRZDPRQJ the most experienced knappers is not certain; however, a preliminary esti mation can be made. As previously mentioned, advanced knappers removed greater numbers RIODUJHUÁDNHVWKDQGLGWKHRWKHUNQDSSHUV6RPHIHDWXUHVRIWKHVHÁDNHVLQ GLFDWHGJUHDWHUNQRZOHGJH LHWKHXVHRISUHH[LVWLQJÁDNHVFDUVWRSURYLGH striking platforms), and some features were ambiguous (core angles might indicate choice or a response to the constraints of the experiment). Consid HUDWLRQRIDIHZRWKHUYDULDEOHVWKDWGLGQRWVKRZVLJQLÀFDQWLQWHUJURXSGLI ferences may help clarify the issue. In particular, platform dimensions were statistically similar across groups; interestingly, core angles were almost iden tical for novices and advanced knappers, with intermediates standing out. (YLGHQFHIURPFRQWUROOHGH[SHULPHQWV 'LEEOHDQG3HOFLQ LQ dicates that for a given core angle, there is a certain range of platform thick QHVVHVWKDWZLOODOORZÁDNHGHWDFKPHQWVDQGWKDWWKLVUDQJHLVVPDOOHUIRU more obtuse angles (like those seen here). The similarity for both these vari ables for novices and advanced knappers suggests that novices perhaps only UHPRYHGÁDNHVZKHQWKH\KDSSHQHGWRKLWXSRQWKHSURSHUYDOXHVIRUWKHVH YDULDEOHVZKLOHDGYDQFHGNQDSSHUVSURGXFHGPRUHÁDNHVEHFDXVHWKH\KDG better control over these variables – in other words, they were able to strike WKHLUFRUHVPRUHSUHFLVHO\DQLQGLFDWRURIJUHDWHUNQRZKRZ,QWHUPHGLDWHV have slightly higher core angles and slightly thinner platforms than do the RWKHUWZRJURXSVVXJJHVWLQJWKDWWKHLULQWHUPHGLDWHQXPEHURIÁDNHVSUR duced may be due to their exercising less control than the advanced knappers EXWVWLOOPDQDJLQJWRUHPRYHPRUHÁDNHVWKDQWKHQRYLFHVGXHWRVWULNLQJ the cores with more force (which is necessary with higher platform angles). 3DLUHGJURXSFRPSDULVRQVDQGDQDO\VHVLQFOXGLQJDOOSODWIRUPEHDULQJGHELW age are underway, which should provide more insight into the debitage pat terns presented preliminarily at present. 238 Knowledge and know-how in the Oldowan: an experimental approach Discussion Taken as a whole, the results for the lithic analysis conducted to date support WKHFRQFOXVLRQWKDWNQRZKRZLVDQLPSRUWDQWFRQWULEXWRUWR0RGH,ÁDN ing performance, which contrasts with previous expectations that knowl edge would hold much greater importance, and that once the knowledge was (quickly) mastered, successful and consistent performance could quickly be achieved. The present study suggests that although knowledge is indeed quite important, its chief contribution may be in the very earliest stages of skill acquisition, as indicated by the clear distinction between the cores produced by novices versus all experienced knappers and by the pattern of VLPLODULWLHVDQGGLIIHUHQFHVLQYROYLQJWKHZKROHÁDNHV1RYLFHVGRVWDQGRXW however, for both the cores and (preliminarily, for most variables) the whole ÁDNHV WKH\ UHVHPEOH LQWHUPHGLDWHV PRUH WKDQ LQWHUPHGLDWHV UHVHPEOH DG vanced knappers. The simplicity of the knowledge involved here makes it unlikely that this is due primarily to steadily increasing levels of knowledge among more experienced knappers; the surprisingly striking distinction be tween advanced knappers and all others reinforces this assertion. It seems OLNHO\WKDWDOWKRXJKDFHUWDLQOHYHORIHDVLO\DFTXLUHGNQRZOHGJHLVQHFHVVDU\ IRU ÁDNH UHPRYDO LQFUHDVHG NQRZKRZ DFTXLUHG RQO\ WKURXJK ORQJ SUDF WLFHLVQHHGHGWRSURGXFHWKHNLQGRIHIÀFLHQWFRQVLVWHQWSHUIRUPDQFHVHHQ among the advanced knappers. These results will be enhanced in the future by more extensive analyses of the data collected, including separate analyses of each raw material. The results discussed above clearly demonstrate that there are differences LQ 0RGH , ÁDNLQJ SHUIRUPDQFH DPRQJ LQGLYLGXDOV RI GLIIHUHQW H[SHULHQFH levels, at least insofar as this performance may be assessed through the ar tifacts produced. In simplest terms, more experienced knappers tend to re GXFHFRUHVPRUHH[WHQVLYHO\SURGXFLQJPRUHÁDNHVWKDWDUHJHQHUDOO\ODUJHU in size than those produced by less experienced knappers. It has been sug JHVWHGDERYHWKDWWKLVLVGXHWRDGYDQFHGNQDSSHUV·PRUHHIIHFWLYHXVHRIWKH knowledge presumed to be shared to some degree by all the other knappers LHJUHDWHUNQRZKRZ $GGLWLRQDOVXSSRUWIRUWKLVLQWHUSUHWDWLRQZRXOGEH REWDLQHGLIÁDNLQJELRPHFKDQLFVDUHIRXQGWRGLIIHUVXEVWDQWLDOO\DPRQJWKH three experience level groups, with less variation occurring among the most experienced individuals. It is to this evidence we now turn. Unfortunately, analysis of the biomechanical data is in a very preliminary state, but informa tion from the pilot study can provide an example of the sort of data obtained, and some preliminary inferences can be made. 239 Leslie Harlacker Biomechanical analysis The pilot study, conducted in 1992 (Dapena 1993; Dapena et al. in press), involved one advanced knapper. The methodology used was the same as that described for the present study, although subject body position was slight O\ GLIIHUHQW 7ZR WULDOV ZHUH DQDO\]HG RQH WULDO LV GHÀQHG DV ODVWLQJ IURP the instant the hammerstone loses contact with the core until the instant before the hammerstone strikes the core on the next arm swing. Kinetic FKDLQPRGHOLQJZDVXVHGVRWKDWWKHDUPZDVPRGHOHGDVDIRXUOLQNNLQHWLF chain consisting of upper arm, forearm, hand, and hammerstone. Using this approach, forces and torques at the elbow, for instance, are understood as resulting from a combination of elbow musculature action and the forces exerted on the forearm by upper arm motion through the elbow joint. In ad dition to joint torques, hammerstone velocity was also calculated throughout the trials. All results presented below pertain to this study. Results Hammerstone speeds, unsurprisingly, were greatest just before impact; speeds were 8.3 and 9.0 m/s for the two trials. Because the subject also moved the core upward to meet the hammerstone, a combined impact speed was calcu lated, giving a result of 8.8 and 10.1 m/s respectively. Wireframe sequences were drawn for the trials, and torque plots were generated; an example in volving the shoulder torques is given (Fig. 7). These sequences reveal that WKHÁDNLQJPRWLRQLVQRWDVLPSOHSODQDUÁH[LRQDQGH[WHQVLRQEXWUDWKHUD FOHDUO\WKUHHGLPHQVLRQDORYHUDUPPRWLRQDQGWRUTXHSORWVUHYHDOWKDWWKH VKRXOGHU PXVFXODWXUH LV PRVW LPSRUWDQW LQ SURGXFLQJ WKH ÁDNLQJ PRWLRQ with the elbow musculature also contributing; movement across the wrist MRLQWLVODUJHO\DÁDLODFWLRQFDXVHGE\PXVFOHDFWLYLW\IXUWKHUXSWKHNLQHWLF chain (Dapena et al. in press). 7KHVHGDWDDOORZDUHODWLYHO\GHWDLOHGSLFWXUHRIWKHÁDNLQJPRWLRQWREH developed. For example, consider the shoulder torque plot given in Fig. 7, which depicts the muscle activity surrounding the shoulder along the three axes shown to the left of the plot. The negative torques produced from times $WKURXJK(VKRZWKDWWKHVKRXOGHUZDVÁH[LQJDQGH[WHUQDOO\URWDWLQJDQG DEGXFWLQJGXULQJWKLVWLPHWKHHOERZZDVÁH[LQJ7KHVHPXVFOHDFWLRQVKDOW the downward motion of the core and accelerate the arm and hammerstone upward. About .25 second before impact, the torques reverse as the knapper swings the hammerstone toward the core and the elbow extends (Dapena et al. in press). A similar picture involving all three joints and velocity data 240 Knowledge and know-how in the Oldowan: an experimental approach IRUKDPPHUVWRQHDQGFRUHZLOOEHGHYHORSHGIRUDOOVXEMHFWVÀOPHGIRUWKH SUHVHQWVWXG\DOORZLQJIRUDIXOOHUSLFWXUHWREHREWDLQHGUHJDUGLQJWKHÁDN LQJDUPVZLQJ Figure 7. The sequence of wire-frame renderings at the top depicts the downswing. The torque plot for the shoulder is at bottom right, and an illustration of the three torques is at bottom left. It is interesting to note that, in comparison with athletic activities such as baseball pitching, Mode I knapping appears to require relatively little strength (Dapena et al. in press); the torques produced through the elbow and shoulder are substantially less in knapping than in such sports activities. 7KLVPLJKWLQLWVHOIEHWDNHQDVHYLGHQFHWKDWNQRZKRZLVOHVVLPSRUWDQW – strength develops over time, and if little is needed, then extensive practice ZLOOQRWEHQHFHVVDU\+RZHYHUWKHUHLVPXFKPRUHWRVXFFHVVIXOÁDNHSUR duction than generating the needed force, such as control and consistency. $QLQLWLDOYLHZLQJRIWKHÀOPVUHYHDOVDZLGHUDQJHRIYDULDWLRQLQWKHEDVLF ÁDNLQJPRWLRQZLWKVRPHVXEMHFWVGLVSOD\LQJPRUHFRQWURODQGFRQVLVWHQF\ WKDQRWKHUVWKHELRPHFKDQLFDOGDWDWREHREWDLQHGIURPWKHVHÀOPVZLOOHQ able the characteristics of the arm swing to be related to the quality of the artifacts produced, allowing us to determine both the mechanics of a suc cessful swing and how frequently those mechanics are to be found among each experience group. Initial data suggest that there will be noticeable dif IHUHQFHVLQÁDNLQJELRPHFKDQLFVDPRQJLQGLYLGXDOVDQGSRVVLEO\EHWZHHQ groups as well, supporting the argument made on the basis of the lithics that NQRZKRZPD\SOD\DVXEVWDQWLDOUROH 241 Leslie Harlacker Conclusions This study represents an attempt to apply an experimental approach to the question of skilled performance regarding the earliest known technology, 0RGH , RU 2OGRZDQ ÁDNLQJ 6NLOO KDV EHHQ GHÀQHG DV D FRPELQDWLRQ RI NQRZOHGJHDQGNQRZKRZZKRVHUHODWLYHFRQWULEXWLRQVPD\EHDVVHVVHGWR give a more detailed picture of the capabilities implied by the technology. Although results are still fairly preliminary, and many more analyses are planned or underway, it seems safe to conclude at this point that knowl edge is not solely or perhaps even primarily responsible for the variation REVHUYDEOHLQ0RGH,ÁDNLQJSHUIRUPDQFH1RYLFHVZKRFDQEHH[SHFWHG to possess rather less knowledge than experienced knappers, do produce artifacts that stand out statistically; however, advanced knappers also stand RXW IURP WKHLU OHVVH[SHULHQFHG FROOHDJXHV ZLWK LQWHUPHGLDWHV UHVHPEOLQJ novices more than advanced knappers in many respects. (Biomechanical results are preliminarily expected to follow a similar pattern.) Because the knowledge needed in the context of this experimental study is so limited, it seems reasonable to conclude that knowledge is quite important in shaping performance among novices, but that as more experience is gained, it is the DFTXLVLWLRQ RI NQRZKRZ UDWKHU WKDQ DGGLWLRQDO NQRZOHGJH WKDW UHVXOWV LQ IXUWKHUJDLQVLQSHUIRUPDQFHHIÀFLHQF\DQGFRQVLVWHQF\ This conclusion has implications for the study of hominid technologi cal evolution in general. Although it is recognized (Toth and Schick 1993) that stone tool technologies represent at least a minimum indication of the cognitive sophistication possessed by their makers, breaking down skilled SHUIRUPDQFHLQWRNQRZOHGJHDQGNQRZKRZHQDEOHVIXUWKHUGLUHFWLRQVWREH suggested. Cognitive attributes that are assumed to be present on the basis of Oldowan technology tend to be based upon the factors previously mentioned that would fall under the heading of knowledge, such as raw material choice DQGIUDFWXUHPHFKDQLFV 7RWKDQG6FKLFN +RZHYHULINQRZKRZ LV DOVR TXLWH LPSRUWDQW LQ UHÀQLQJ 2OGRZDQ ÁDNLQJ SHUIRUPDQFH DQG LQ FUHDVLQJHIÀFLHQF\DQGFRQVLVWHQF\DGGLWLRQDOEHKDYLRUDOLQIHUHQFHVFDQEH suggested on the basis of Oldowan technology. For instance, since fairly extensive experience seems necessary to consistently and quickly produce ODUJHQXPEHUVRIÁDNHVZHPLJKWLQIHUWKDW2OGRZDQKRPLQLGVZRXOGKDYH EHQHÀWHGIURPÀQGLQJWKHWLPHWRSUDFWLFHWKHLUWRROPDNLQJVNLOOV7KHODUJH QXPEHUVRIWRROVUHFRYHUHGIURPPDQ\2OGRZDQVLWHVKLQWVDWDQHIÀFLHQF\ that may indeed imply that such practice took place. Indeed, the acquisition RINQRZKRZWKURXJKSUDFWLFHPD\KDYHEHHQHYHQPRUHLPSRUWDQWIRUHDUO\ hominids, many of whom possess anatomy that differs from our own. The 242 Knowledge and know-how in the Oldowan: an experimental approach additional work to be done on the data sets discussed in this paper will soon UHVXOWLQPRUHGHÀQLWHFRQFOXVLRQVFRQFHUQLQJNQRZOHGJHDQGNQRZKRZLQ WKH2OGRZDQDQGHQDEOHVXFKLQIHUHQFHVWREHUHÀQHGIXUWKHUHQKDQFLQJ our understanding of hominid technological evolution. Acknowledgements This study follows from my dissertation research, which was made possible by a grant from the L. S. B. Leakey Foundation, as well as assistance from CRAFT Research Center and Friends of CRAFT, Inc. Support during the DQDO\VLVDQGGLVVHUWDWLRQZULWLQJSURFHVVLVEHLQJSURYLGHGYLDDQ$PHULFDQ Fellowship from the American Association of University Women Educa tional Fund. My advisors, Nick Toth, Kathy Schick, and Jesus Dapena have been sources of invaluable discussion, suggestions, encouragement, and as sistance during the data collection process and beyond, for which I am most grateful. I also thank them for their comments on a draft of this chapter, and their permission to present the pilot study results in this article. Dr. Da SHQDDOVRSURYLGHGPXFKDSSUHFLDWHGDVVLVWDQFHZLWK)LJ,ZRXOGOLNHWR H[WHQGP\KHDUWIHOWWKDQNVWR(UUHWW&DOODKDQZKRDOORZHGPHWRÀOPDWKLV ÁLQWNQDSSLQJÀHOGVFKRRO DQGSURYLGHGPXFKWKHRUHWLFDOIRRGIRUWKRXJKW GXULQJVHYHUDOFRQYHUVDWLRQV DQG6WHYH:DWWVZKRDOORZHGPHWRÀOPDWKLV 2OGRZDQWKHPHGNQDSLQWKHVHRSSRUWXQLWLHVZHUHLQYDOXDEOHWRREWDLQLQJ DZHOOEDODQFHGVXEMHFWVDPSOH6SHFLDOWKDQNVJRWR6XH6DYDJH5XPEDXJK and Bill Fields for allowing me to include Kanzi and Panbanisha in my study, DQG IRU SURYLGLQJ PXFKDSSUHFLDWHG DVVLVWDQFH GXULQJ RXU GDWD FROOHFWLRQ session. Scott Brish and Melanie Everett assisted with the human subject ÀOPLQJ VHVVLRQV -DFN /DJRQL SURYLGHG WKLQ VHFWLRQV RI WKH UDZ PDWHULDOV and Paul Jamison provided ongoing statistics advice. I am grateful for all of their contributions. Thanks are also due to Rogers Quarry in Martinsville, Ind., who allowed me to collect hammerstones, greenstone, and practice stone from their waste piles. Finally, I must thank all 51 of my subjects, both human and bonobo, for their time and efforts – they are much appreciated – as well as Jan Apel and Kjel Knutsson for inviting me to present my work at the symposium. 243 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Tuija Rankama, Mikael A. Manninen, Esa Hertell and Miikka Tallavaara Simple production and social strategies: do they meet? Social dimensions in Eastern Fennoscandian quartz technologies Abstract Lately, anthropologists and archaeologists have emphasized the social di mensions of all technologies. However, it is also well known that the quality and quantity of lithic resources affected the organisation of the prehistoric XVHRIOLWKLFV,QDUHDVZKHUHJRRGTXDOLW\NQDSSDEOHUDZPDWHULDOZDVHLWKHU scarce or absent, simple lithic technologies often prevailed. These are not easily converted into reconstructions of social strategies of the past. This paper discusses the problems and possibilities in deriving social information IURPOLWKLFDVVHPEODJHVLQ(DVWHUQ)HQQRVFDQGLDZKHUHJRRGTXDOLW\ÁLQW does not exist. Introduction Due to its geological history, Eastern Fennoscandia (Fig. 1) is largely devoid RIJRRGTXDOLW\ÁLQWOLNHUDZPDWHULDOV7KLVKDVEHHQWKHPDLQIDFWRUDIIHFW ing the archaeological lithic record. The lithic assemblages found within the SUHVHQWGD\ERUGHUVRI)LQODQGUHSUHVHQWDOPRVWH[FOXVLYHO\GLIIHUHQWNLQGV of ”ordinary production” (sensu Pelegrin 1990). Technologies were mainly EDVHGRQVLPSOHSODWIRUPDQGELSRODURQDQYLOFRUHV2QO\DIHZH[DPSOHV of more elaborate technologies of knapped stone have been reported (e.g., Rankama 1997). The most common lithic raw material throughout the pre historic era was vein quartz, available either as rounded cobbles or as quar ried pieces. Other local raw materials, such as quartzite, porphyry, sandstone DQGTXDUW]FU\VWDOVZHUHXVHGWRDOHVVHUGHJUHH7KHÁLQWWHFKQRORJ\ZKLFK according to traditional thinking, relied almost exclusively on imported raw materials (Manninen et al. 2003), is in line with the rest of the simple knap ping technologies. 245 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara Figure 1. Map of Eastern Fennoscandia. The geographical term Fennoscandia covers an area consisting of Norway, Sweden, Finland, the Kola Peninsula, and parts of Russian Karelia. The geoORJLFDOO\GHÀQHG)HQQRVFDQGLDQ6KLHOGLV GHYRLGRIQDWXUDOVRXUFHVRIÁLQW From the beginning of the Mesolithic to the end of the Stone Age the ar chaeological stone tool record in Finland consisted of two parts: knapped and ground stone tools. This division of stone tool technology is by no PHDQVDVFOHDUFXWDVLWVHHPVRULQGHHGWKHRQO\SRVVLEOHZD\WRFODVVLI\WKH lithic material. Technologically, as well as functionally, these two categories RYHUODSZLWKHDFKRWKHUHJVRPHD[HVZHUHPDQXIDFWXUHGÀUVWE\NQDSSLQJ and then grinding, and some projectiles were knapped whereas others were ground. In this paper the term ”lithics” refers to knapped stone alone. The table in Figure 2 gives an idea of the general outline of the lithic raw material situation in Finland from the beginning of the Mesolithic to, and LQFOXGLQJWKH(DUO\0HWDO$JH,WPXVWEHHPSKDVL]HGWKDWWKHWDEOHUHÁHFWV the traditional archaeological knowledge concerning the matter and does not go into detail, especially about all the variety of other than knapped raw ma WHULDOV,WLVFOHDUWKDWRQJRLQJUHVHDUFKPD\FKDQJHLWFRQVLGHUDEO\:HKDYH also omitted Lapland from the table, because the raw material situation there is different and including it would have made the table overly complicated. 246 Simple production and social strategies: do they meet? calBC Period Quartz Flint Other major raw materials Ground slate, quartzite, jasperoid 1900 Early Metal Age Dominant (DVWHUQVWUDLJKWEDVHGÁLQWELIDFHV imported (copied in Finland in quartzite and quartz) 2350 Late Neolithic Dominant ,PSRUWHGZHVWHUQÁLQWVLFNOHVDQGD few daggers Ground slate, porphyry, soapstone etc., indigenous and imported rocks 3200 Corded Ware Dominant 1RÁLQW Ground diabase and other indigenous rocks 3600 Late Comb Ware Dominant ,PSRUWHGHDVWHUQÁLQWGHFUHDVHV bifaces, but less raw material Ground slate, porphyry, other indigenous rocks, imported igneous rock 4000 Typical Comb Ware Dominant Fair amount of imported eastern ÁLQWVPDOOELIDFHVDQGUDZPDWHULDO Ground slate, porphyry, other indigenous rocks, imported igneous rock 5000 Early Comb Ware Dominant 6RPHLPSRUWHGÁLQW Ground slate, porphyry, other indigenous rocks, imported igneous rock 7000 Litorina Mesolithic Dominant 1RÁLQW Ground and pecked slate and diabase, indigenous and imported rocks 9000 Ancylus Mesolithic Dominant 6RPHÁLQWLPSRUWHGE\HDUOLHVW immigrants Ground and pecked slate and diabase, indigenous and imported rocks Figure 2. Generalised outline of lithic raw material use in Finland during the Stone Age and Early Metal Age. The lithic technology, and especially the quartz technology in Eastern Fen noscandia, seems very uniform throughout the Stone Age and so far it has not been possible to distinguish regional or chronological traditions within it. This is partly due to the fact that technological studies of quartz assem EODJHV KDYH XQWLO WKH PLGV EHHQ YHU\ UDUH EXW DOVR UHÁHFWV WKH IDFW that few regionally or chronologically distinct tool types seem to exist. If we accept that cultural and social traditions always affect the technology HPSOR\HGE\DVSHFLÀFJURXS /HPRQQLHU WKHIDFWWKDWWKHUHDUHFOHDU regional differences in other sections of the material culture, such as ground stone tools and pottery styles, gives reason to expect these to also be found within the quartz technology. How, then, can these differences be detected and social meaning be given to them? 7KLVSDSHUGLVFXVVHVGLIIHUHQFHVLQTXDUW]DQGÁLQWVWXGLHVDQGWKHSURE lems and possibilities in deriving social information from predominantly very simple lithic technologies. Its emphasis is on vein quartz technology. Our perspective on stone tool research will largely be Fennoscandian or North European, LQFRQWUDVWWRWKHPRUHFRQYHQWLRQDOÁLQWEDVHGUHVHDUFKWUDGLWLRQV to the south and west. The current research paradigm in Finland is based on the work carried out in Sweden, especially in Uppsala, during the 1980s and 1990s (e.g. Callahan 1987; Callahan et al. 1992; Knutsson 1988a). A similar research orientation can also be seen in other parts of Fennoscandia (e.g. Nærøy 2000; Olofsson 2003). 247 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara “The past in the present” in lithic research When working with assemblages representing simple lithic technologies, one must be aware of the way in which the history of research tends to guide our thinking. For example, it is not often that one comes across studies GLVFXVVLQJELSRODURQDQYLOWHFKQRORJLHVHVSHFLDOO\LQ(XURSHDQOLWHUDWXUH Ironically, it is almost more common to see drawings of bipolar cores mis interpreted as burins or wedges, a mistake that is a direct consequence of a lack of knowledge concerning this kind of technology. This is an example of what could be called the past in the present in lithic research, i.e., a conven WLRQWUDQVPLWWHGLQWKHVFLHQWLÀFFRPPXQLW\FRQFHUQLQJKRZOLWKLFVWXGLHV should be conducted and what to study. Conventionally, simple technologies are not interesting and are therefore easily overlooked. In quartz studies the ZHLJKWRIWKHVHVFLHQWLÀFFRQYHQWLRQVKDVFDXVHGVHYHUHSUREOHPVRYHUWKH years (Knutsson 1998). The fact that vein quartz fractures somewhat differ HQWO\WKDQÁLQWZDVQRWH[SODLQHGLQDVDWLVIDFWRU\ZD\XQWLOTXLWHUHFHQWO\ (Callahan et al. 1992; Knutsson 1998; Rankama 2002). :KHQVWXG\LQJORQJODVWLQJWUDGLWLRQVRIYHLQTXDUW]XVHRQHKDVWRWU\ WRDYRLGGLUHFWFRPSDULVRQVZLWKÁLQW$WWKHVDPHWLPHRQHLVIDFHGZLWK WKHIDFWWKDWQHDUO\DOOWKHUHVHDUFKOLWHUDWXUHLVEDVHGRQÁLQWRURWKHUÀQH grained raw materials that are for the researcher, as they were for the pre KLVWRULFÁLQWNQDSSHUPXFKHDVLHUWRZRUNZLWKWKDQYHLQTXDUW],QYHLQ quartz assemblages such things as ripple marks, bulbs of percussion, and VRPHWLPHV HYHQ FRPSOHWH ÁDNHV DUH XQFRPPRQ 7KLV FDXVHV GLIÀFXOWLHV when research avenues that have been found fruitful in other materials are DWWHPSWHGLQYHLQTXDUW]7KHTXDUW]DQDO\VWUXQVLQWRGLIÀFXOWLHVQRWRQO\ because many of the traditional research methods are not applicable, but also because of the pressure exerted by the current trends in lithic research that may not prove practicable in vein quartz, either. To make things even more complicated, some varieties of quartz, such as smoky quartz, rose quartz, and quartz crystal, often do fracture much like ÁLQWDQGIRUWKLVUHDVRQDPRQJRWKHUVPXVWEHFRQVLGHUHGQRWWREHWKH same raw material as vein quartz. This does not mean, however, that vein TXDUW]LVDEDGUDZPDWHULDO²LWLVMXVWYHU\GLIIHUHQWIURPÁLQW,QRUGHUWR JLYHDQDO\VWVXVHGWRVWXG\LQJÁLQWDSRVVLELOLW\WRXQGHUVWDQGWKHQDWXUHRI YHLQTXDUW]DVVHPEODJHVZHZLOOEULHÁ\GLVFXVVWKHGLIIHUHQFHVEHWZHHQYHLQ TXDUW]DQGÁLQW 248 Simple production and social strategies: do they meet? Problems in Eastern Fennoscandian quartz assemblages Lithic analysts aiming to understand the social aspects of vein quartz use will encounter serious problems compared with those one has to cope with LQÁLQWVWXGLHV0RVWRIWKHVHSUREOHPVGHULYHIURPYHLQTXDUW]PDWHULDOLW self and it does not matter what kind of a research strategy one is following. )LUVWWKHSURYHQLHQFHRIWKHUDZPDWHULDOFDQQRWEHGHÀQHGYLVXDOO\RUDW OHDVWDWWKHPRPHQWHYHQWKURXJKJHRFKHPLFDOPHWKRGV7KLVFDXVHVREYL ous problems for studies of mobility and settlement patterns. In addition, it means that individual knapping events cannot easily be distinguished at KDELWDWLRQVLWHV5DZPDWHULDOFODVVLÀFDWLRQVHJPLQLPXPQRGXOHDQDO\VLV FI/DUVRQ .RUQIHOG DQGWKHEHQHÀWRQHDFTXLUHVWKURXJKWKHP are therefore not available in vein quartz studies. Admittedly, some types of quartz, such as smoky quartz, are distinctive enough to be recognized in analysis, but they are much less common in prehistoric assemblages. Questions concerning raw material procurement strategies are not easy to address, either. Vein quartz is a practically ubiquitous raw material in the landscape and occurs both as rounded cobbles in moraine ridges, lake shores, and so on, and as veins in the bedrock, from which it can be quarried with relative ease (cf. Broadbent 1979; Manninen & Valtonen 2002). 'LVWLQJXLVKLQJÁDNHVFDUIHDWXUHVLQYHLQTXDUW]LVDOVRSUREOHPDWLF6LQFH YHLQTXDUW]LVDPDFURFU\VWDOOLQHDQGSDUWO\WUDQVSDUHQWDQGWUDQVOXFHQWPD WHULDO WKDW RIWHQ KDV LQWHUQDO ÁDZV LW GRHV QRW KDYH HDVLO\ REVHUYDEOH DQG VPRRWK IUDFWXUH VXUIDFHV RI WKH VDPH NLQG WKDW ZH PHHW LQ ÁLQWOLNH UDZ PDWHULDOV 'LIIXVH RU QRQH[LVWHQW EXOEV RI SHUFXVVLRQ DQG WKH DEVHQFH RI ULSSOHPDUNVPDNHÁDNHVFDUGLUHFWLRQVRIWHQGLIÀFXOWWRGHWHFWHVSHFLDOO\LQ ÁDNHIUDJPHQWV7RFRPSOLFDWHWKLQJVIXUWKHULUUHJXODUVXUIDFHIHDWXUHVLQ vein quartz can be mistakenly interpreted as retouching (Siiriäinen 1981:8p; Schäfer 1990:89; Lindgren 1998; Cornelissen 2003:13). )LJXUH5HÀWWHG TXDUW]ÁDNHVIURPWKH Leakšagoadejohka 3 site in Utsjoki, Finnish Lapland, showing radial and bending fractures (Manninen 2003:Fig.11). 249 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara $OWKRXJKWKHFKDUDFWHULVWLFVRITXDUW]DQGÁLQWRYHUODSWKHJHQHUDOFKDUDFWHU RIYHLQTXDUW]LVYHU\GLIIHUHQWIURPÁLQW/LNHÁLQWTXDUW]KDVDFRQFKRLGDO fracture. Usually the dominating fracture types in vein quartz, however, are UDGLDODQGEHQGLQJIUDFWXUHV )LJ WKDWEUHDNWKHÁDNHLQWRVPDOOHUIUDJ ments (see Callahan et al. 1992; Huang & Knutsson 1995; Rankama 2002). Because of this, a lithic tradition based on vein quartz is likely to make use of these fracture types and only secondarily of the conchoidal fracture (see 5DQNDPD5DQNDPD 7KHIUDJPHQWDWLRQRIÁDNHVLQDGGLWLRQ PDNHVWKHGHELWDJHORRNXQIDPLOLDUWRDQDQDO\VWXVHGWRVWXG\LQJÁLQWOLNH UDZ PDWHULDOV ,Q DGGLWLRQ WKH IUDJPHQWDWLRQ DOVR PDNHV UHÀWWLQJ VWXGLHV PRUHFRPSOLFDWHGVLQFHHYHQWKHIUDJPHQWVRIDVLQJOHÁDNHPD\KDYHWREH ÀWWHGEDFNWRJHWKHUEHIRUHDWWHPSWLQJWRUHÀWDÁDNHUHPRYDOVHTXHQFH VHH Manninen 2003:63pp). Simple technologies, especially the use of bipolar reduction, are often ex plained as the result of an attempt to maximise the use of lithic raw mate rial, and this is undoubtedly often the case (e.g., Andrefsky 1994:384). To a degree, the use of different reduction methods and techniques is also known WREHPDWHULDOVSHFLÀF4XDUW]XVHUVDOORYHUWKHZRUOGVHHPWRKDYHUHOLHG on bipolar reduction (Siiriäinen 1977; Flenniken 1980; Huang & Knutsson 1995; Kuhn et al. 1996). The frequent use of bipolar reduction on quartz is QRWVLPSO\EHFDXVHRWKHUPHWKRGVRUZD\VWRSURGXFHÁDNHVFRXOGQRWKDYH been employed. It is well known, for instance, that microblade and bifacial concepts have been applied to quartz. The main reason for the wide use of bipolar reduction is probably the fact that quartz seems to be especially well suited for it (cf. Flenniken 1980; Kuhn et al. 1996), and therefore the use of bipolar reduction has been an active choice made by the knappers. It is worth noting, however, that although quartz knapping methods are rather similar around the world, there are also clear differences in assem blages separated in time and space. No doubt these are the result of different traditions manifested as different kinds of behavioural acts. Although the TXDUW]DVVHPEODJHVLQ(DVWHUQ)HQQRVFDQGLDGRQRWPHHWDQ\RIWKHÀYHD[HV or points Gero (1989:93pp) has suggested as helpful when assessing social information in stone tools (rarity of raw material, artifact size, artifact lon gevity, number of production stages, restrictiveness of production), we will in the following paragraphs consider some current research trends in studies pertaining to social strategies in prehistory and their applicability to Eastern Fennoscandian quartz studies. 250 Simple production and social strategies: do they meet? Social dimensions in lithic technologies Anthropologists and archaeologists studying technology have lately stressed the importance of the social dimensions of all human technologies (e.g., Lemonnier 1992; 1993; Pfaffenberger 1992; Dobres 2000). No matter how simple a human technology is, it is nevertheless largely dependent on its social surroundings, since, for example, the acceptance of a technological innovation by a community may be dependent on the existing social system (Lemonnier 1992). Discussions of age groups and gender in archaeology have been impor tant and opened up new insights into the study of archaeological assemblag es (e.g., Gero & Conkey 1991; Moore & Scott 1997; Kamp 2001). They have reminded us that archaeological assemblages are the result of the different actions of a variety of individuals and groups. As a result, lithic assemblages, often treated as cold statistics, have become more human. In line with these discussions, learning processes and apprenticeship in general have become important research topics (e.g. Minar & Crown 2001). Age and learning Several scholars have recently published papers dealing with the learning SURFHVVHV LQYROYHG LQ PRUH HODERUDWH PHWKRGV RI ÁLQWNQDSSLQJ )LVFKHU 1990; Pigeot 1990; Högberg 1999; Apel 2001). The simple knapping meth ods employed in prehistoric Finland did not require a prolonged learning SURFHVV%HJLQQHUVOHDUQHGWKHPHWKRGVHPSOR\HGLQDVSHFLÀFWUDGLWLRQE\ observing how the older people used stone. Children playing tried to copy WKHLUSDUHQWV·DFWLRQV VHH.QXWVVRQ)LQOD\&URZQ DQGDW some point, when old enough, were shown the raw material sources and the correct ways to manufacture the necessary edges for tools and weapons. One could easily envision initiation ceremonies tied to certain age groups and segments of transmitted knowledge. Unfortunately, learning processes of this kind are only rarely observable in the archaeological record (but see Knutsson 1983; Finlay 1997), because of the simple and uniform appearance of the end products, regardless of whether they were made by a skilful knap per or a beginner. )URPWKHUHVHDUFKHU·VSRLQWRIYLHZWKHSUREOHPZLWKWKHUDWKHUVLPSOH technologies is the low level of cognition and motor skill required to per form them. It is easy to employ a method that does not require previous knowledge and special learning, in which the common gestures of everyday life are enough to perform the action successfully. This makes it possible 251 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara IRUDQLQGLYLGXDOWRVZLWFKEHWZHHQGLIIHUHQWVLPSOHÁDNLQJPHWKRGVZKHQ ever need arises, provided that there are no social or cultural restrictions to prevent it. From this it also follows that we usually cannot distinguish individual knappers from vein quartz assemblages and study, e.g., their level of craftsmanship. In an important early paper, Knutsson (1986) addressed the role of chil GUHQ DV SRVVLEOH PDQXIDFWXUHUV RI VPDOO ELSRODURQDQYLO FRUHV $OWKRXJK KH DEDQGRQHG WKH H[SODQDWLRQ WKH FRQVLGHUDWLRQ RI FKLOGUHQ·V SUHVHQFH DW DUFKDHRORJLFDOVLWHVLVRILPSRUWDQFHKHUH6LQFHDQRGXOHVPDVKLQJNLQGRI ELSRODURQDQYLO UHGXFWLRQ LV SUREDEO\ WKH VLPSOHVW ZD\ WR UHGXFH VWRQH LW seems likely that it is also the way children started to practice quartz knap SLQJ0RGHUQSHRSOHZLWKQRSUHYLRXVNQRZOHGJHRIÁLQWNQDSSLQJDOVRUH duce rocks into smaller pieces this way (Callahan 1987:63). At sites where children were present their role in producing part of the assemblage we are VWXG\LQJWRGD\PD\KDYHEHHQFRQVLGHUDEOH7KLVFDQKDYHDVLJQLÀFDQWHIIHFW on the platform vs. bipolar reduction ratio that several authors have studied LQ6ZHGHQDQG)LQODQGHVSHFLDOO\ZKHQDSSOLHGWRLQWHUVLWHFRPSDULVRQV $OWKRXJKFKLOGUHQDUHGLIÀFXOWWRGHWHFWLQ(DVWHUQ)HQQRVFDQGLDQOLWKLF assemblages, in some cases it is still possible. An illuminating example of a FKLOG·VNQDSSLQJSURGXFWZDVLGHQWLÀHGDPRQJWKHELSRODURQDQYLOFRUHVDW the Rävåsen site in Southern Ostrobothnia, dating to the Late Comb Ware Period. Among the analysed lithic material, consisting mainly of different SURGXFWVRIELSRODURQDQYLOTXDUW]UHGXFWLRQDELSRODUO\EDWWHUHGIHOGVSDU “core” was observed (Hertell & Manninen in press). The feldspar is of a light colour and resembles vein quartz very much. Feldspar, however, was not used to make tools at the site and does not fracture the same way as quartz or any other commonly used lithic raw material when bipolarly knapped. It is therefore very likely that this “core” is the product of a child imitating an older knapper but not yet able to recognize the correct raw material, i.e., not SRVVHVVLQJWKHWDVNVSHFLÀFNQRZOHGJH Gender Some researchers have emphasized that the making of stone tools must have EHHQHYHU\ERG\·VEXVLQHVV *HUR/LQGJUHQ +RZHYHULW is evident that everyone in the past did not always do everything. Women are known to have made and used lithic artefacts in many societies/communities and their impact on the formation of the lithic assemblages must in those FDVHVKDYHEHHQVLJQLÀFDQW *HUR +RZHYHULWLVHTXDOO\OLNHO\WKDWLQ other societies women did not participate in stone tool manufacture and use. 252 Simple production and social strategies: do they meet? It is even possible that men normally made stone tools, e.g. scrapers, even if women mostly used them (Osgood 1940:440p). As Gero has pointed out, the assumption that tool manufacture and use, which is often a male activity in contemporary Western world, was restricted RQO\WRPHQLQSUHKLVWRU\LVÁDZHG *HUR ,WLVRIFRXUVHHTXDOO\ZURQJ to argue that men or women must have done something in the past simply because we want it to be so, that is, because that is the way we think the ZRUOGVKRXOGZRUN7KHKXQWHUJDWKHUHU·VRUHDUO\KRUWLFXOWXUDOLVW·VZRUOG view does not need to agree with our present concept of equality. $UFKDHRORJLFDO ÀQGV )OHQQLNHQ DQG HWKQRJUDSKLF H[DPSOHV VHH Knutsson 1988b:14) indicate that in some parts of the world small unmodi ÀHGTXDUW]ÁDNHVZHUHKDIWHGLQDPDQQHUWKDWLVDQDORJRXVWRWKHXVHRI backed microliths. It has been suggested that during the Stone Age in Fin land (Hertell & Manninen in press) and Sweden (Lindgren 1994) small quartz ÁDNHVDQGÁDNHIUDJPHQWVZHUHDOVRXVHGWKLVZD\&RQVHTXHQWO\LWLVSRV VLEOHWRH[WHQG)LQOD\·V LGHDVDERXWPXOWLSOHDXWKRUVKLSLQWKHXVHRI microliths also to Fennoscandian quartz technologies. Eastern Fennoscandian quartz studies have not, however, so far proceed HGIDUHQRXJKWREHDEOHWRDVVLJQVSHFLÀFSDUWLDORUFRPSOHWHchaînes opératoires to gender or sex groups. The continuous lithic scatter at most sites sug gests that at many sites everyone must have been familiar with and possibly involved in quartz technology. As noted above, due to the short history of studies of quartz technology, the analysis of reduction strategies has not so IDUEHHQDEOHWRGLVWLQJXLVKVSHFLÀFUHJLRQDOO\RUFKURQRORJLFDOO\UHVWULFWHG quartz working traditions, let alone those of smaller groups or individuals. Lindgren (1996; 2003), however, has studied this question in Eastern Middle Sweden and suggested that the two quartz reduction strategies observed at Mesolithic sites in the area could be related to social groups, possibly divided along gender lines. Power and Prestige In various archaeological and ethnographic contexts elaborate stone objects KDYH EHHQ DVVRFLDWHG ZLWK SRZHU DQG SUHVWLJH HJ $SHO EXW ÀQG ing comparable examples among simple technologies is a different matter. Quartz is a common mineral in Eastern Fennoscandia. This makes it un likely that it would have been an exotic commodity during the Stone Age, although some special types of quartz, such as smoky quartz and quartz crystals, may again constitute a special case. In addition, as a result of the nature of the raw material, artefacts made of quartz are not large. This, to 253 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara gether with the informal character of the tools, may suggest that the visual appearance of quartz was not an important aspect to those for whom it was the everyday raw material. However, the white and shiny colour as well as the transparent and translucent nature of the raw material must have en hanced the visibility and the visual appreciation of quartz artefacts. These FKDUDFWHULVWLFVPD\HYHQKDYHKDGDVSHFLDOPHDQLQJIRUWKHXVHUVRIÁLQW ZKRZHUHQRWDFFXVWRPHGWRWKHPLQVWRQH FI%DQJ$QGHUVHQ&DUOLH 1999; Darvill 2002). If we accept that the ordinary, simple production of vein quartz imple ments was available to practically every member of a community, it is dif ÀFXOWWRGLVWLQJXLVKDQ\IHDWXUHVWRZKLFKRQHFRXOGDWWDFKSUHVWLJHYDOXH However, one part of the lithic record in Eastern Fennoscandia was almost certainly not available to everyone, and it could be suggested that its social meaning could be transmitted by extension, as it were, to quartz implements. :LWKWKLVZHPHDQWKHLPSRUWHGÁLQWREMHFWVWKDWDSSHDULQ)LQQLVK6WRQH Age material more or less simultaneously with Comb Ware pottery that have recurrently been found in burials together with other prestigious materials, such as imported amber (e.g., Edgren 1966:93, Table 3; Torvinen 1979; Ha linen 1997). 7KHPDLQÁLQWLPSOHPHQWW\SHVIRXQGLQWKHVHEXULDOVDUHSUHVVXUHÁDNHG bifacial points of various shapes (e.g., Manninen et al. 2003), which are not easy to manufacture from vein quartz. Nevertheless, vein quartz bifaces were PDGHVRPHWLPHVZLWKFRQVLGHUDEOHVXFFHVV,WLVQRWGLIÀFXOWWRLPDJLQHWKDW a bifacial quartz point brought prestige not only to its owner but also to its skilful manufacturer, who may or may not have been the same person. In a sense, then, it could perhaps be said that in a simple technology, the same way as in more elaborate technologies, complicated objects brought prestige to their owner and manufacturer. Social dimensions in quartz assemblages? Since quartz was an integral part of the lives of the Stone Age inhabitants of Eastern Fennoscandia, it is obvious to us that quartz processing must, indeed, have been a social act in the past communities. We have discussed above some of the social dimensions that become apparent in quartz as VHPEODJHVZKHQORRNLQJDWWKHPIURPDSHUVSHFWLYHEDVHGPDLQO\RQÁLQW studies. But what could the vein quartz assemblages in themselves offer for the study of prehistoric social strategies? One aspect of lithic technology that can be studied especially well using predominantly vein quartz assemblages is the way in which raw material selection and use was controlled socially. 254 Simple production and social strategies: do they meet? The general character of quartz technology in Eastern Fennoscandia is quite uniform. This raises questions since, due to the differences between vein TXDUW]DQGÁLQWOLNHPDWHULDOVWKHUHPXVWDOVRKDYHEHHQGLIIHUHQWZD\VRI looking at vein quartz as a raw material in prehistoric times. When forced to XVHTXDUW]DNQDSSHUFRPLQJIURPDWUDGLWLRQEDVHGVWULFWO\RQÁLQWSURE DEO\ORRNHGIRUWKHPRUHÁLQWOLNHSLHFHVRIUDZPDWHULDO HJTXDUW]FU\VWDOV that enable one to carry out a chaîne opératoireOHDUQHGZLWKÁLQWHYHQZKHQXV ing quartz. A knapper from a mainly vein quartz using tradition used quartz in a different manner because he or she was not looking for a substitute for ÁLQWEXWDSLHFHRITXDUW]WKDWZDVVXLWHGIRUNQDSSLQJLQWKHZD\OHDUQHG from other knappers used to vein quartz. As a consequence, the assemblages produced by these two hypothetical knappers turned out very different (cf. Manninen 2003:81p). These different kinds of perception concerning quartz can, in fact, be GHWHFWHGLQWKHDUFKDHRORJLFDOPDWHULDO7KHDERYHPHQWLRQHG&RPE:DUH SHULRGÁLQWDQGTXDUW]ELIDFHVDUHDJRRGH[DPSOH%LIDFLDOSUHVVXUHÁDNLQJ ZDVQRWXVHGLQZRUNLQJTXDUW]EHIRUHWKHÁLQWUDZPDWHULDOLPSRUWEHJDQ DQGDIWHULWHQGHGLWZDVQRWXQWLOWKHVRFDOOHGVWUDLJKWEDVHGDUURZKHDG PDGHPDLQO\RIÁLQWDQGÀQHJUDLQHGTXDUW]LWH ZDVLQWURGXFHGLQWKH(DUO\ Metal Age, about a thousand years later, that quartz bifaces reappeared in the Finnish archaeological record. The usually quite crude appearance of the quartz bifaces indicates clearly that vein quartz is not a raw material that lends itself well to making for PDOWRROV,WLVPXFKEHWWHUVXLWHGWRPDNLQJRQO\VOLJKWO\PRGLÀHGHGJHV for composite and expedient tools. This disparity between vein quartz and ÁLQWOLNHUDZPDWHULDOVLVWKHUHDVRQIRUWKHXQLIRUPLW\LQWKHYHLQTXDUW] assemblages: the differences are so pronounced that they generated a unique OLWKLFWHFKQRORJ\QRWGLUHFWO\FRPSDUDEOHWRÁLQWWHFKQRORJLHV FI5DQNDPD 2003). However, as simple as this kind of technology may seem, plenty of task VSHFLÀFNQRZOHGJHKDGWREHOHDUQHG,WLVNQRZQIRUH[DPSOHWKDWQRQ potters cannot estimate the dimensions of a pot that can be manufactured IURPDJLYHQEDOORIFOD\ZKHUHDVSRWWHUVFDQ :DOODHUW3rWUH 7KLV reminds us of how individuals relying on, for instance, vein quartz had to learn through practice how much quartz would be needed in certain tasks or over the winter, or how long a certain piece of raw material or a tool edge might last. A pragmatic approach to lithic reduction methods and the use of lithics, such as the one suggested here to have developed in Eastern Fennoscandia, ZRXOG KDYH PDGH D WHFKQRORJ\ ÁH[LEOH DQG ZHOO VXLWHG IRU D FRORQL]DWLRQ SURFHVV VHHcNHUOXQGet al. 2003:xxxvii). In Finland the use of simple knap 255 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara ping methods, however, was not restricted to highly mobile groups. Quartz UHGXFWLRQLQVHPLVHGHQWDU\SLWKRXVHYLOODJHVVHHPVWRKDYHEHHQPRVWO\DV simple as among the more mobile groups (see, e.g., Rankama 2002), perhaps HPSKDVL]LQJLWVÀUPDVVRFLDWLRQZLWKFXOWXUDOSUHIHUHQFHV,QWKLVVHQVHWKH concept of “the correct ways to manufacture the necessary edges of tools and weapons” (above) also has a social dimension: it can mean simply tech nologically correct, but also socially correct, i.e., the way we (as opposed to they) do it. Since culture is reproduced through learning (see Minar 2001), the fact that vein quartz maintained its importance as a raw material, even when RWKHUPRUHHDVLO\ÁDNHDEOHUDZPDWHULDOVZHUHDYDLODEOHFRXOGEHVHHQDVDQ indication of socially transmitted preferences. It is also a good example of the fact that good conchoidal fracturing per se was not necessarily the factor that made a raw material preferred in prehistoric times. It should be kept in mind that according to ethnoarchaeological examples, conchoidal fractur LQJKDVQRWDOZD\VEHHQLPSRUWDQWWRÁLQWXVLQJJURXSVHLWKHU 0LOOHU White & Thomas 1972:278p). 7KHÁLQWOLNHPDWHULDOVQHYHUWKHOHVVPXVWKDYHKDGDVSHFLDOVLJQLÀFDQFH to the traditionally vein quartz using groups (see Holm 2003). The factor WKDWPDGHÁLQWOLNHUDZPDWHULDOVVSHFLDOKRZHYHUZDVQRWWKDWWKH\ZHUH petrologically different. Petrological categorizations are a modern construc tion and do not necessarily correspond to the categories used by prehistoric knappers (see Stout 2002:704 for an ethnographic example). It is possible WKDW SUHKLVWRULF NQDSSHUV FODVVLÀHG UDZ PDWHULDOV DFFRUGLQJ WR IUDFWXULQJ qualities and considered petrologically different stones to be the same raw PDWHULDO&RQVHTXHQWO\VRPHTXDUW]HVDQGVRPHÁLQWVPD\KDYHEHHQFODVVL ÀHGDVWKHVDPHUDZPDWHULDOLQSUHKLVWRULFWLPHV Were the technologies so simple after all? We have grouped most of the Eastern Fennoscandian lithic technologies DERYHXQGHUWKHWHUPVLPSOHWHFKQRORJ\7KHVHVRFDOOHGVLPSOHWHFKQROR gies, like all technologies, should, however, always be seen as parts of larger technological systems, and attempts should not be made to explain them in isolation from other components of the systems. Ethnoarchaeological data allows us to estimate, for instance, that making a haft with stone tools prob ably took a much longer time than making the simple stone blade of the tool (see Hayden 1979). If the simple quartz technologies of Eastern Fennoscan dia were used mainly to produce edges for composite tools it would probably be the hafts that had most of the visible cultural or social meanings attached to them. 256 Simple production and social strategies: do they meet? It must be emphasized that even though the general picture of quartz tech nologies in Fennoscandia and adjacent areas seems rather simple, this does not automatically mean that the whole technological system was simple. Quartz tools that have residues of adhesive material still attached to them have been found both in Sweden (Broadbent 1979:Fig. 78) and in Finland (Meinander 1954:Abb. 22). This implies that at least these quartz ”tools” actually were only a small part of the complete tool. The complete tools may KDYHEHHQYHU\FRPSOH[DQGWLPHFRQVXPLQJWRPDNH If one considers, for instance, the Eskimo scrapers with their handles HUJRQRPLFDOO\VKDSHGWRÀWWKHRZQHU·VKDQG HJ1HOVRQ3O;/,; RU the Kwakiutl wood working adze in Figure 4, the term “simple technology” LVQRWWKHÀUVWWKLQJWKDWVSULQJVWRPLQG7KHIDFWWKDWWKHEODGHZDVHDV\WR manufacture does not mean that the tool as a whole was simple. For this rea VRQRQHVKRXOGQRWKDVWLO\LQWHUSUHWVLPSOHORRNLQJUHWRXFKHGTXDUW]ÁDNHV RUÁDNHIUDJPHQWVDVH[SHGLHQWWRROVDOWKRXJKPDQ\RIWKHPSUREDEO\ZHUH This, of course, is valid for other raw materials, as well. Since stone tools are often the only part of a technological system that has survived, one cannot estimate from the lithics how sophisticated or “skilled” a given technological system was, even if mainly expedient stone tools were used. The technology of the Wola (Sillitoe & Hardy 2003) is a good example of this. Figure 4. Kwakiutl adze for woodworking (from Buschan 1922:Abb. 28a). 257 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara New perspectives in Fennoscandian lithic studies At this point it is time to consider what the new approaches recently in troduced in Fennoscandian quartz studies could in the future offer to the subject at hand. The examples above have made it clear that many of the PHWKRGVERUURZHGIURPOLWKLFVWXGLHVGHDOLQJZLWKÁLQWOLNHUDZPDWHULDOV will not get us very far. It is also obvious that, most probably due to the spe cial character of vein quartz, formal tool typologies attempted in order to bring order to vein quartz assemblages in Fennoscandia have failed to work (Siiriäinen 1981; Knutsson 1998). Fracture analysis Fracture analysis, which was developed in Sweden in the early 1990s, has HQDEOHGUHVHDUFKHUVWRXQGHUVWDQGWKHIUDJPHQWDWLRQRITXDUW]ÁDNHVGXULQJ UHGXFWLRQDQGJLYHPHDQLQJWRWKHTXDUW]ÁDNHIUDJPHQWVWKDWZHUHSUHYL ously considered impossible to analyse (e.g., Callahan et al. 1992; Lindgren 1996; Räihälä 1999; Rankama 2002). Due to its novelty there naturally is still much to be tested in the reliability of some of the applications of the method (cf. Manninen 2003:55p) but understanding the fragmentation that occurs in vein quartz reduction and recognising the fragment types is essential for all types of quartz analyses. The way in which quartz fractures in distinct fragment types (Callahan et al. 1992; Rankama 2002) offers possibilities not available when studying as semblages produced by technologies in which these fracture types were not taken advantage of. For instance, if particular fragment types were selected as tool blanks, it is easier to see which piles of debitage should actually be regarded as collections of blanks put aside for future use, and which ones should be considered pure debris from tool production (see Rankama 2002; 2003). Chaîne opératoire analysis The chaîne opératoire or technological approach (sensu Chazan 1997) has reached Eastern Fennoscandia slowly, but the possibilities it offers to quartz studies are considerable. Although the methods used to reduce vein quartz were simple, at least in Finland there are, nevertheless, differences in them WKDWJREH\RQGWKHWHFKQRORJLFDOWULSDUWLWLRQ IUHHKDQGSODWIRUPSODWIRUP 258 Simple production and social strategies: do they meet? RQDQYLOELSRODURQDQYLO VXJJHVWHGE\&DOODKDQ LQKLVSLRQHHULQJ work on the lithic technology of Eastern Central Sweden. Simple operational schemes do not have as many readily observable char acteristics that can be considered cultural or social markers as the more elab orate methods (see Pelegrin 1984:84). Clear differences can, nevertheless, be seen even in simple knapping methods even when all the sequences of a chaîne opératoire cannot be detected. Figure 5 compares two instances: a partial operational sequence from the Leakšagoadejohka 3 site in Utsjoki, )LQQLVK/DSODQGUHFRQVWUXFWHGE\WHFKQRW\SRORJLFDODQDO\VLVDQGUHÀWWLQJ (Manninen 2003) and a core from the Rävåsen site in Southern Ostroboth nia (Hertell & Manninen in press). Both examples represent simple platform methods. Site /HDNäDJRDĄHMRKND Rävåsen Site type 6PDOO[PNQDSSLQJÁRRUQHDUWKHUDZPDWHULDO source (river bed) Large occupation site with pit houses Raw material Rounded cobble, c. 3-5 kilograms, shape slightly angular 5RXQGHGFREEOHFJUDPVVKDSHRYDODQGÁDW Source River bed near site Unknown Stage 1 Cobble opened by smashing against rocks 2SHQLQJÁDNHVWUXFNXVLQJQDWXUDOURXQGHG surface as platform Stage 2 Large piece (c. 2 kilograms) decorticated with large ÁDNHUHPRYDOV $OWHUQDWHELIDFLDOÁDNLQJXVLQJVFDUVIURPSUHYLRXV removals as platforms Stage 3 6HYHUDOÁDNHUHPRYDOVIURPDVLQJOHSODWIRUP Discoidal core (280 grams) rejected with part of the cortical surface remaining Stage 4 &RUHWDNHQDZD\IURPNQDSSLQJÁRRU Figure 5. Interpretation of partial chaînes opératoires from two Finnish Stone Age sites: the Leakšagoadejohka 3 site in Utsjoki, Finnish Lapland, and the Rävåsen site in Southern Ostrobothnia. The differences in these two examples cannot at present be directly connect HGZLWKVSHFLÀFFXOWXUDORUVRFLDOWUDGLWLRQV,WLVLPSRUWDQWWRQRWHKRZHYHU that the knapping methods represented by the examples are clearly distin guishable from each other. This gives hope that more examples of different operational schemes can be detected in future analyses and interpreted as cultural or social markers, making chaîne opératoireDQDO\VLVWKHORQJDZDLWHG substitute for formal tool typologies in vein quartz studies (cf. Manninen 2003). The chaîne opératoire concept understood as the study of the whole system of stone tool manufacture, maintenance, and use and expressed, e.g., as a ÁRZPRGHO VHH5DQNDPD5DQNDPD PD\DOVREULQJIRUWK differences in attitudes towards different stone tools and raw materials that PD\KDYHVRFLDORUFXOWXUDOVLJQLÀFDQFH 259 Tuija Rankama, Mikael A. Manninen, Esa Hertell & Miikka Tallavaara Conclusion ,QWKLVSDSHUZHKDYHEULHÁ\GLVFXVVHGWKHSUREOHPVRIGHULYLQJVRFLDOLQ formation from the most common artefact category in the Stone Age as semblages of Eastern Fennoscandia, namely, vein quartz tools and debitage. 7KHGLIÀFXOWLHVWKDWPXVWEHDGGUHVVHGLQYHLQTXDUW]VWXGLHVKDYHEHHQFRQ WUDVWHGZLWKWKHSRVVLELOLWLHVDYDLODEOHLQÁLQWVWXGLHV7KHVHGLIÀFXOWLHVDUH largely the result of the nature and appearance of vein quartz as compared ZLWKÁLQW,WLVDOVRZRUWKQRWLQJWKDWWKHGLIÀFXOWLHVEHFRPHSUREOHPVRQO\ when research methods and objectives found fruitful in studying other raw materials are uncritically applied to vein quartz studies. It should be remembered that although “quartz as a raw material has a bad reputation among archaeologists” (Cornelissen 2003:2) it is dangerous to assume that this was also the case among Stone Age peoples. Qualities that make a raw material easy to study do not necessarily make it better to XVH7KHVWDUWLQJSRLQWLQOLWKLFVWXGLHVVKRXOGQRWEHLQFRQYHQWLRQDOLGHDV of what is a good raw material or what are the most sophisticated or skilled ZD\VRIÁLQWNQDSSLQJ$VNLOOHGTXDUW]XVHUIRULQVWDQFHNQHZKRZWRPDNH WKHEHVWXVHRIWKHSURSHUWLHVRIWKDWSDUWLFXODUUDZPDWHULDOLQVSHFLÀFFXO tural, social, and environmental circumstances. Quartz studies require an orientation that differs from the one employed LQVWXGLHVRIÁLQWOLNHUDZPDWHULDOV$OWKRXJKTXDUW]ZDVVRPHWLPHVXVHG LQWKHVDPHPDQQHUDVÁLQWWKLVZDVQRWDOZD\VWKHFDVH&RQFKRLGDOIUDF ture was not always a factor that made a raw material desired. In addition, like wines, quartz assemblages are not all alike. Some quartz assemblages have clear counterparts in other raw materials, but at least in Eastern Fen QRVFDQGLDDGLIIHUHQWFRPSOHWHO\YHLQTXDUW]EDVHGWHFKQRORJLFDOWUDGLWLRQ also developed. ,WLVGLIÀFXOWWRVD\WRZKDWGHJUHHWUDGLWLRQDOFXVWRPVULWXDOVRUWDERRV PD\KDYHUHVWULFWHGWKHXVHRITXDUW]LQWKLVWUDGLWLRQWRVSHFLÀFLQGLYLGXDOV JURXSVRUJHQGHUV*UDYHÀQGVLQGLFDWHWKDWGXULQJFHUWDLQSHULRGVVRPH quartz items were considered valuable enough to be placed among funeral offerings. It should be remembered, however, that grave furnishings did not QHFHVVDULO\KDYHWKHVDPHPHDQLQJVDPRQJKXQWHUJDWKHUHUVDVLQRXUFRQ temporary or in agricultural contexts (cf. Knutsson 2000) and that raw mate rial was not necessarily the factor that regulated the value of an artefact. At Eastern Fennoscandian Stone Age sites quartz is normally scattered all over the habitation area, suggesting again that vein quartz was available to everyone. The knowledge concerning quartz procurement, reduction, use, maintenance, and discard was transmitted from generation to generation and 260 Simple production and social strategies: do they meet? learned in everyday situations. However, to what degree we will be able to distinguish the social dimensions of quartz use from other features, such as the effects of mobility, raw material availability, and so on, in our archaeo logical material, is a completely different matter. In Finland lithic analyses have so far concentrated on individual sites. As a consequence, it is not yet possible for us to try to understand the effects of different complex factors on the formation of the lithic record in a complete settlement system. Quartz reduction in Eastern Fennoscandia seems mainly to have been UDWKHUVLPSOHDQGWKHNQRZKRZQHHGHGWRZRUNTXDUW]HDVLO\DFTXLUHG,W also seems that the chaînes opératoires were not complicated, in other words, WKH\ GLG QRW LQYROYH VHYHUDO DQGRU GLIÀFXOW SURGXFWLRQ VWDJHV ,W LV OLNHO\ that tool production was not restricted or specialized. These things make some of the recent research trends in the anthropology of technology, such as the study of learning processes, the transmission of skills, and the like, XVXDOO\LOOVXLWHGIRUTXDUW]VWXGLHV7KHUHIRUHWKHVWXG\RIVLPSOHWHFKQROR gies should not be restricted to only those social strategies that have been investigated in connection with complex lithic technologies. Technologies that are based predominantly on vein quartz but have also used other raw materials offer a better opportunity to study, e.g., the social aspects of raw material use and the mechanisms of the transport of exotic UDZPDWHULDOVWKDQGRWHFKQRORJLHVUHO\LQJH[FOXVLYHO\RQÁLQW,WLVHVVHQWLDO however, to continue looking for cultural and social variation also within pure quartz assemblages. It is equally essential that the methods employed in these studies are not selected on the basis of their popularity in current lithic UHVHDUFKEXWSXUHO\RQWKHEDVLVRIWKHLUDSSOLFDELOLW\WRWKHVSHFLÀFFKDUDF WHULVWLFVRIYHLQTXDUW]DVVHPEODJHV5HÀWWLQJDQGUHSOLFDWLRQXVHGWRJHWKHU with chaîne opératoire analysis, for example, are ways to explore the potential of TXDUW]DVDVRXUFHRIVLJQLÀFDQWVRFLDOLQIRUPDWLRQ Acknowledgments The authors wish to thank the organisers of the symposium Skilled Production and Social Reproduction for the opportunity to participate and bring forth our ideas about Eastern Fennoscandian lithic assemblages in a social context. 7KH ZULWLQJ RI WKH FRQIHUHQFH FRQWULEXWLRQ DQG WKLV SDSHU ZDV SDUWO\ À nanced by a grant form the Jenny and Antti Wihuri fund to Tuija Rankama. 261 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Bradford Andrews Skill and the question of blade crafting intensity at Classic Period Teotihuacan Abstract 'HÀQLQJWKHVFDOHDQGRUJDQL]DWLRQRI&ODVVLF3HULRG7HRWLKXDFDQ·VREVLG ian tool industry is an important research objective. One question of interest LVZKHWKHULWVFUDIWVPHQZHUHIXOOWLPHRUSDUWWLPHVSHFLDOLVWV7KLVSDSHU DVVHVVHVWKHFUDIWVPDQVNLOOUHÁHFWHGE\VXUIDFHFROOHFWLRQVIURPZRUNVKRSV as a means for inferring specialist labor intensity. Skill is measured by tabu ODWLQJWKHIUHTXHQFLHVRIDWWULEXWHVWKDWUHSUHVHQWFRUHEODGHSURGXFWLRQHU URUV&RPSDULQJWKLVLQIRUPDWLRQWRFRUHEODGHZRUNVKRSVIURP(SLFODVVLF ;RFKLFDOFRLQGLFDWHVWKDW7HRWLKXDFDQPD\KDYHKDGRQO\SDUWWLPHEODGH producers. Introduction This paper addresses an issue related to the organization and scale of Classic $' 7HRWLKXDFDQ·V )LJ REVLGLDQWRROLQGXVWU\7KHVHWZRGL mensions of production encompass the social contexts in which crafts were made, how many craftsmen were involved, how often they worked, and how many items were produced. By the 1980s, claims were being made that Te RWLKXDFDQ FRQWDLQHG KXQGUHGV RI SDUW DQG IXOOWLPH ZRUNVKRSV VHUYLFLQJ the needs of consumers not only in the city, but also throughout much of the Mesoamerican region (Sanders and Santley 1983; Santley 1984; Santley et al. 1995; Spence 1981). These claims have been used to support inferences about WKHRYHUDOOFRPSOH[LW\RI7HRWLKXDFDQ·VHFRQRP\ 0LOORQ Sanders et al. 1979). In a landmark article in 1986, John Clark (1986) suggested that Teoti KXDFDQ·V REVLGLDQ WRRO LQGXVWU\ KDG EHHQ JUHDWO\ H[DJJHUDWHG $GGUHVVLQJ one issue he raised, I examine the same data used to support the proposed PRQXPHQWDOVFDOHRIWKHFRUHEODGHLQGXVWU\WRH[SORUHWKHTXHVWLRQRIODERU LQWHQVLW\GHÀQHGKHUHDVWKHDPRXQWRIWLPHDVSHFLDOLVWVSHQWPDNLQJFUDIWV 6SHFLÀFDOO\,H[DPLQHDUWLIDFWVIURPWKH6DQ0DUWLQFRPSOH[DUJXDEO\RQH of the most intensive obsidian tool workshops that is located in the north HDVWHUQ VHFWRU RI WKH FLW\ )LJ $QGUHZV 6SHQFH 1981, 1986). Unfortunately, the inferential validity of these data is suspect 263 Bradford Andrews because they were collected from the surface (Clark 1986). My purpose here, therefore, is to use them to formulate an alternative perspective about labor intensity at the workshop that will require future testing with subsurface information. Figure 1. The Locations of Teotihuacan and Xochicalco in Greater Mesoamerica. Figure 2. The Location of the San Martin Complex in Teotihuacan. 264 Skill and the question of blade crafting intensity at Classic Period Teotihuacan I suggest the San Martin data, when compared to valid subsurface infor PDWLRQ IURP D ZRUNVKRS DW (SLFODVVLF $' ;RFKLFDOFR )LJ LQGLFDWH WKDW 6DQ 0DUWLQ·V FUDIWVPHQ PD\ KDYH EHHQ SDUWWLPH VSHFLDOLVWV ,IVRWKHQZHPXVWZRQGHUZKHWKHURWKHUJHQHUDOO\OHVVLPSUHVVLYHEODGH SURGXFLQJFRQWH[WVLQWKHFLW\ZHUHDOVRRSHUDWHGE\SDUWWLPHVSHFLDOLVWV This perspective has important implications for understanding the nature of FUDIWSURGXFWLRQDW7HRWLKXDFDQDQGWKHFLW\·VRYHUDOOLQWHUGHSHQGHQWVRFLR economic structure. The following discussion begins by describing the San Martin and Opera WLRQ+ZRUNVKRSVDQGWKHQGHÀQHVVNLOODVLWSHUWDLQVWRWKLVVWXG\1H[W, address the artifact attributes used to measure craftsman skill and how they ZHUHTXDQWLÀHG7KHWKLUGVHFWLRQSUHVHQWVWKHDQDO\VLVRIFUDIWVPDQVNLOO 7KHÀQDOVHFWLRQGLVFXVVHVWKHUHVXOWVDQGLPSOLFDWLRQVRIP\VWXG\ The Workshops :KLOHVXEVXUIDFHGDWDDUHQHHGHGWRFRQÀUPZKHWKHUPDQ\RI7HRWLKXDFDQ·V “workshops” were actually production contexts, the inhabitants of the San Martin complex were involved in the specialized manufacture of blades and bifaces during the Classic period (Fig. 3). The workshop consists of three conjoined residential compounds covering about 2 ha, housing an estimated 140 to 300 people (Spence 1986). The array and density of obsidian tool byproducts at the workshop, and their direct association with residential ar chitecture,1 suggest that it was occupied by cooperatively organized extended families2LQYROYHGLQÁDNHGVWRQHWRROSURGXFWLRQ $QGUHZV Figure 3. Map of the San Martin Workshop Complex at Teotihuacan. 265 Bradford Andrews At Xochicalco, excavations of the Operation H workshop have revealed a FRQVLGHUDEOH DPRXQW RI FRUHEODGH E\SURGXFWV DQG NQDSSLQJ WRROV DVVRFL ated with multiple activity areas (see Hirth n.d. for a detailed discussion of these data). This context was a single large residential compound with three conjoined patios (Fig. 4) that housed an extended family of 10 to 20 people (Hirth 2000:tables 7.1 and 7.3). As such, both the San Martin and Operation H workshops represent comparable social contexts of production, although the family unit at San Martin was much larger. Figure 4. Map of the Operation H Workshop at Xochicalco. The Concept of Skill Since my question addresses crafting intensity, it revolves around the con FHSWRISDUWWLPHYHUVXVIXOOWLPHVSHFLDOL]DWLRQDGLFKRWRP\WKDWLVWRRFRQ straining because it forces workshops into two categories. Instead, I advocate placing workshops along a continuum that allows us to rank relative levels of labor intensity. This perspective is useful because it permits more realistic FRPSDULVRQVRILQWHQVLW\WKDQWKHWUDGLWLRQDOSDUWWLPHIXOOWLPHGLFKRWRP\ (Andrews 2003; Costin 1991, n.d.; Tosi 1984). The labor intensity at which specialists were involved in production can be inferred from measures of craftsman skill (Andrews 1999, 2003). This perspective assumes that, in most cases, more skilled craftsmen will be those who worked most often because they were the most practiced at the tech niques or gestures necessary for making their crafts.3 By techniques or ges WXUHV,DPUHIHUULQJWRWKHFRQFHSWRINQRZKRZGHÀQHGDVWKHLQWXLWLYHO\ GLUHFWHG ERG\ PRYHPHQWV DVVRFLDWHG ZLWK PDNLQJD VSHFLÀFFUDIW XQGHU D 266 Skill and the question of blade crafting intensity at Classic Period Teotihuacan VSHFLÀF VHW RI FRQGLWLRQV HJ UDZ PDWHULDO W\SH DQG IRUP DQG NQDSSLQJ WRROV NQRZKRZLVDIRUPRISK\VLRORJLFDOPHPRU\WKDWFDQRQO\EHKRQHG through practical experience (Pelegrin 1990:118). For this discussion, I distinguish two qualitatively different types of skill: DUWLVDQDODQGHIÀFLHQF\$UWLVDQDOVNLOOUHIHUVWRWKHSK\VLFDOPRWLRQVDELOL ties associated with the production of high quality ceremonial/wealth items (Apel 2001, Clark 1986:44, Clark and Parry 1990; Rathje 1975:414, Sheets 1978:66). This type of craft production usually favors a high level of labor invested per unit output (Hayden 1998:11); emphasis is placed on producing an aesthetic item that transmits social information (Helms 1993, Inomata 2001:324, Torrence 1986:45). Consequently, artisanal skill is measured by the RXWZDUGTXDOLWLHVRIDÀQHO\FUDIWHGSURGXFW7KLVW\SHRIVNLOOFDQEHGLI ÀFXOWWRTXDQWLI\EHFDXVHLWLVRIWHQEDVHGRQHPLFDHVWKHWLFTXDOLWLHV ,QFRQWUDVW,VXJJHVWWKDWHIÀFLHQF\VNLOOLVJHQHUDOO\DVVRFLDWHGZLWKWKH production of utilitarian items. Here, lower levels of labor are invested per product to make a practical item that may be more uniform and standardized at higher levels of specialization (Hayden 1998:2; Schiffer and Skibo 1997). ,QFRQWUDVWWRDUWLVDQDOVNLOOHIÀFLHQF\VNLOOPXVWEHPHDVXUHGE\ORRNLQJ at populations of items. One approach is to assess the proportion of errors LQDQDVVHPEODJHEHFDXVHHIÀFLHQF\VNLOOVKRXOGEH´«SRVLWLYHO\FRUUHODWHG with specialization, and therefore industries with fewer mistakes…will be more specialized than those [with more] mistakes or less command over the productive process” (Costin 1991:40). #ONTINUUM OF ,ABOR )NTENSITY 0ART TIME INTENSITY &ULL TIME INTENSITY Figure 5. Continuum of Labor Intensity. (IÀFLHQF\VNLOODVDPHDVXUHFDQEHFRQFHSWXDOL]HGDVDSDUWWRIXOOWLPH continuum (Fig. 5). Minimal labor investment in the absence of economies RI VFDOH SUHVXSSRVHV WKH YLHZ WKDW VSHFLDOLVWV DW WKH SDUWWLPH HQG RI WKH VSHFWUXPKDYHORZHUHIÀFLHQF\VNLOOWKDQWKHLUIXOOWLPHFRXQWHUSDUWV3DUW WLPHVSHFLDOLVWVDUHOHVVHIÀFLHQWEHFDXVHWKH\GRQRWHQJDJHLQWKHQXPEHURI repetitive activities often associated with a standardized production process 5LFH ,QFRQWUDVWIXOOWLPHSURGXFHUVDUHPRUHSURÀFLHQWDWSHU forming the gestures related to their efforts because constant repetition of these actions promotes greater competence. If economies of scale come into play, then the range of activities performed by each worker is reduced lead LQJWRDQLQFUHDVHLQWKHIUHTXHQF\RIVSHFLÀFDFWLYLW\SHUIRUPDQFH &RVWLQ 267 Bradford Andrews QG5DWKMH5LFH7RUUHQFH $VDUHVXOWIXOOWLPHVSHFLDOLVWVDUHH[SHFWHGWRPDNHIHZHUHUURUVWKDQSDUW time specialists, all other things being equal. ,HYDOXDWHHIÀFLHQF\VNLOOE\ORRNLQJDWSURSRUWLRQVRIHUURUVLQWKH6DQ 0DUWLQ DQG 2SHUDWLRQ + DVVHPEODJHV %\ GHÀQLWLRQ HUURUV UHSUHVHQW WKH failure to reach a desired outcome (Clark 2003:222). Accordingly, to identify HUURUVRQHPXVWEHUHODWLYHO\FHUWDLQRIDFUDIWVPDQ·VLQWHQWLRQ &ODUN For Mesoamerican pressure blades, the intention was straightforward: the production of long, regular blades with parallel sides and prismatic cross sections. Errors *LYHQ WKH EODGHPDNHU·V LQWHQWLRQ PRVW UHVHDUFKHUV DJUHH WKDW WZR FRP PRQ FRUHEODGH HUURUV DUH KLQJHG DQG RYHUVKRW EODGHV )LJ $QGUHZV 2003; Clark 1997, 2003; Clark and Bryant 1997; Crabtree 1972; Santley et al. 1995; Sheets 1975, 1983). A hinged blade does not travel the length of the core, but rather, terminates with a fracture plane that curves outward from LWV YHQWUDO WR GRUVDO VXUIDFH +LQJHV FDQ EH LGHQWLÀHG IURP WKH VFDUV WKH\ leave behind on cores (Fig. 6a), or on the distal ends of blade sections (Fig. E ,QFRQWUDVWRYHUVKRWEODGHVRFFXUZKHQDEODGH·VYHQWUDOVXUIDFHFXUYHV inward towards the bottom of the core, thereby removing considerably more core distal mass than intended (Fig. 6c; Clark and Bryant 1997:123; Crabtree 1972). The forces promoting these errors are described in detail elsewhere (Andrews 2000; Clark and Bryant 1997; Santley et al. 1986; Sheets 1978). Figure 6. Errors used in the analysis: a) hinge scars visible on a core artifact, b) proximal blade section with a hinge termination, c) distal section with an overshot termination. 268 Skill and the question of blade crafting intensity at Classic Period Teotihuacan Identifying errors is relatively easy. The more challenging proposition is de YLVLQJDPHWKRGZLWKZKLFKWKH\FDQEHPHDQLQJIXOO\TXDQWLÀHG(OVHZKHUH -RKQ&ODUN PDLQWDLQVWKDWHUURUVPXVWEHTXDQWLÀHGDJDLQVWDPHDVXUH RI EODGH RXWSXW %\ GHÀQLWLRQ KRZHYHU VXLWDEOH EODGHV ZHUH H[SRUWHG WR the consumers so they will not be available to count in workshop deposits. :LWKRXWDGRXEWHUURUVDVXQVXLWDEOHE\SURGXFWVZLOOEHRYHUUHSUHVHQWHGDW workshops, a phenomenon Clark (1997, 2003) refers to as the Brandon Bias. Discarded items such as core artifacts, however, can be used to estimate blade output. First, one must estimate the number of blades available from WKHDYHUDJHVL]HGFRUHUHGXFHGDWDVLWH7KLVQXPEHUFDQWKHQEHPXOWLSOLHG by the number of core artifacts to estimate the blade output represented by an assemblage. Consequently, the number of blades with hinge terminations can be divided by the estimated output to derive a hinge frequency. This method assumes that the proportion of core artifacts in an assem blage has not been altered. This assumption is unlikely to hold true for the San Martin data because it is a surface collection. Prospecting of this loca tion since the Classic, for usable tool stone or simple curiosity, has probably resulted in the differential removal of artifacts such as exhausted4 cores be cause of their unique shape and size.5 As a result, the core artifacts in the San 0DUWLQFROOHFWLRQSUREDEO\GRQRWDFFXUDWHO\UHÁHFWLWVEODGHSURGXFWLRQ $VDQDOWHUQDWLYHWR&ODUN·V PHWKRGIRUTXDQWLI\LQJHIÀFLHQF\VNLOO ,VXJJHVWDQDSSURDFKWKDWTXDQWLÀHVHUURUVDVUDWLRVRIVSHFLÀFFODVVHVRI related artifacts (Andrews 1999, 2003; Hirth and Andrews n.d.a). This op tion is based on the assumption that any blade has a direct potential rela tionship to the incidence of a single error. In other words, a blade is either VXFFHVVIXOO\RUHUUDQWO\GHWDFKHG6LQFHWKHNQDSSHU·VLQWHQWLRQZDVWRPDNH IXOOOHQJWKVWDQGDUGL]HGSULVPDWLFEODGHVHUUDQWKLQJHGSUR[LPDOVHFWLRQV FDQ EH TXDQWLÀHG SHU WRWDO QXPEHU RI SUR[LPDO VHFWLRQV /LNHZLVH HUUDQW RYHUVKRWEODGHVFDQEHTXDQWLÀHGSHUWRWDOQXPEHURIGLVWDOVHFWLRQV7KHVH UDWLRVSURYLGHDPHDVXUHRILQSXWWKDWUHÁHFWVWKHUHODWLYHSURSRUWLRQRIVXF cessfully produced blades. Although successful blades were exported, data from numerous Mesoa PHULFDQ EODGH ZRUNVKRSV LQGLFDWH WKDW VLJQLÀFDQW SHUFHQWDJHV RI VQDSSHG blade segments were discarded (Hirth et al. 2000, 2003; Clark and Bryant 1997; Healan 1986:table2, 1990:cuadro 1; Parry 2002:table 4.1 & 4.2). Discard of these artifacts relates to the segmentation of blades into usable sections DQGWKHPRGLÀFDWLRQRIEODGHVLQWRPRUHVSHFLDOL]HGWRROVVXFKDVSURMHF tile points and eccentrics. Snapped blade segments that were too small to use effectively were discarded. These artifacts, therefore, provide the basis for quantifying the relative rates of hinge fracture and overshot errors. In FRQWUDVWWRFRUHDUWLIDFWVSRVWDEDQGRQPHQWVFDYHQJLQJRIVPDOOGLVFDUGHG 269 Bradford Andrews EODGH VHFWLRQV LV XQOLNHO\ WR KDYH RFFXUUHG RQ D VLJQLÀFDQW VFDOH , PDLQ tain, therefore, that this method is more appropriate for quantifying errors present in the San Martin surface collection. Three variables are examined in this analysis (Tab. 1): 1) the percentage of hinged proximal sections per total number of proximal sections, 2) the percentage of distal sections with overshot terminations per total number of distal sections, and 3) the number of hinge scars per core artifact. Variable 1 Hinged Proximals/ Total Number of Proximals Variable 2 Overshot distals/Total Number of Distals Variable 3 Number of Scars/Core Artifact Table 1. Variables used in the study. The data While both blades and bifaces were made at San Martin, my analysis is con FHUQHGZLWKEODGHPDNLQJVNLOO)XUWKHUPRUH,KDYHVHOHFWHGRQO\WKH3D chuca obsidian artifacts to assess the skill of the San Martin craftsmen be FDXVHWKLVPDWHULDOFRQVWLWXWHVDERXWSHUFHQWRI7HRWLKXDFDQ·VFRUHEODGH assemblages (Andrews 2002; Spence 1981). In contrast, I use only the gray Ucareo and Zacualtipan obsidian artifacts to assess skill of the Operation H FUDIWVPHQEHFDXVHWKHVHPDWHULDOVFRQVWLWXWHSHUFHQWRI;RFKLFDOFR·VFRUH blade material (Hirth 2002:83). I have chosen to do this because evidence suggests that fracture mechanics vary according to raw material; hence, my DQDO\VLVIRFXVHVRQWKHWRROVWRQHPRVWIDPLOLDUWRWKHFRUHEODGHFUDIWVPHQ at each workshop. Theoretically, it should provide a measure of their highest levels of respective skill. Focusing on these data, the errors from the San Martin and Operation +ZRUNVKRSVFDQEHTXDQWLÀHG2QFHDJDLQYDULDEOHLVWKHSHUFHQWDJHRI hinged proximals per total number of proximal sections. There are hinge terminations on 1.9 percent of the San Martin proximal sections and 2.0 percent of the Operation H proximal sections (Tab. 2). Variable 2, the per centage of overshot terminations per total number of distal sections, there are overshot terminations on 9.2 percent of the San Martin distal sections, and 7.3 percent of the Operation H distal sections. 270 Skill and the question of blade crafting intensity at Classic Period Teotihuacan Variables 1) hinged prox./prox 2) overshots/distals 3) hinge scars/core San Martin Operation H (Teotihuacan) (Xochicalco) 1.9% 2.0% (7 hinged prox./365 prox.) (123 hinged prox./6,148 prox.) 9.2% 7.3% (13 overshot distals/142 distals) (169 overshot distals/2,317 distals) 1.1 scars 1.9 scars (144 hinges/133 cores) (170 hinges/91 cores) 7DEOH5HVXOWVRIWKH(IÀFLHQF\6NLOO$QDO\VLV Once again, variable 3 represents the number of hinge scars per core arti fact (Tab. 2). In this case, the hinge scar frequency is standardized against WKHEODGHVFDUVRQWKHFRUHDUWLIDFWVWKHPVHOYHVZKLFKGLUHFWO\UHÁHFWHLWKHU successful or unsuccessful (hinged) blade removals. Core artifacts include exhausted complete cores, core sections, and core tops. Since exhausted FRUHVUHSUHVHQWWKHODVWVHULHVRIEODGHUHPRYDOV,LQFOXGHODUJHUGLDPHWHU artifacts such as core tops and core sections, generally produced earlier in the sequence, to acquire data on hinge occurrence throughout the whole EODGHPDNLQJVHTXHQFH7KHUHDUHKLQJHVFDUVSHUFRUHDUWLIDFWLQWKH6DQ Martin data, and 1.9 hinge scars per core artifact in the Operation H data. Discussion The hinge data suggest that relative to the craftsmen at Operation H, those at San Martin may have been slightly more adept at controlling the incidence of hinge fractures. In contrast, the incidence of overshot terminations sug gests that the Operation H craftsmen had the upper hand. Taken together, however, these data may indicate that the craftsmen from both workshops had similar levels of skill. If so, then these patterns also may have implica tions for inferring the labor intensity of the San Martin craftsmen. The San 0DUWLQZRUNVKRSZDVRULJLQDOO\SODFHGLQ6SHQFH·V 5HJLRQDO:RUN VKRSFDWHJRU\7KLVFDWHJRU\VXSSRVHGO\UHSUHVHQWV7HRWLKXDFDQ·VPRVWLQ WHQVLYHODUJHVFDOHZRUNVKRSVZLWKIXOOWLPHFUDIWVPHQWKDWVXSSOLHGODUJH TXDQWLWLHVRIWRROVIRUH[WUDORFDOFRQVXPSWLRQ 6DQWOH\et al. 1995; Spence 1981, 1987). %DVHGRQWKHHVWLPDWHGODERULQWHQVLW\RI;RFKLFDOFR·V2SHUDWLRQ+FUDIWV PHQKRZHYHU,VXJJHVWWKDWWKHIXOOWLPHVWDWXVRI6DQ0DUWLQ·VFUDIWVPHQ may have to be reconsidered. 271 Bradford Andrews Labor intensity at Operation H was calculated according to the amount of time needed to produce blades at Xochicalco. This estimate was derived from several categories of artifacts, including core tops found on the work VKRS·VÁRRU +LUWKDQG$QGUHZVQGE DQGTXDQWLÀHGH[SHULPHQWDODVVHVV PHQWVRIEODGHPDNLQJDFWLYLWLHVXVLQJ;RFKLFDOFR·VFRUHEODGHWHFKQRORJ\ (Flenniken and Hirth 2003; Hirth et al. 2003). The estimate was based on a monthly production cycle, to account for SHULRGLFZDVWHGLVSRVDODQGWKHQFRQYHUWHGLQWRDÀJXUHRIDQQXDORXWSXW 7KH ODERU LQWHQVLW\ RI 2SHUDWLRQ +·V FUDIWVPHQ WKH KLJKHVW DPRQJ WKUHH GRPHVWLFFRUHEODGHZRUNVKRSVDW;RFKLFDOFRZDVDERXWFUDIWVPDQGD\V per month, or 420 craftsman workdays per year (Hirth and Andrews n.d.b). Accordingly, the number of production locales strongly suggests that Op eration H had two to three craftsmen operating simultaneously, although it FRXOGKDYHKDGDIXOOWLPHFUDIWVPDQ +LUWKQG ,IVRWKHQWKH6DQ0DUWLQ FUDIWVPHQPD\DOVRKDYHEHHQSDUWWLPHEHFDXVHWKH\DSSHDUWRKDYHKDGD OHYHORIHIÀFLHQF\VNLOOVLPLODUWRWKHFUDIWVPHQDW2SHUDWLRQ+$WWKHYHU\ least, this evidence casts doubt on the assertion that the San Martin crafts PHQZHUHKLJKO\LQGXVWULRXVIXOOWLPHSURGXFHUV This study has used the same data set to reach a conclusion that runs contrary to the previously proposed nature of production at the San Martin workshop. That aside, I realize that numerous objections can be raised to this exploratory analysis, including 1) its use of surface data, and 2) the tech nological comparability of the assemblages. Setting aside the obvious need for systematic subsurface samples of data, I believe the issue of technological compatibility is essential to consider. For LQWHUVLWHVWXGLHVRIVNLOORQHWKLQJIXWXUHUHVHDUFKPXVWDGGUHVVLVWKHUHDO LW\WKDWWKHFRUHEODGHWHFKQRORJLHVDW7HRWLKXDFDQDQG;RFKLFDOFRGLIIHUHG 7KLVIDFWRUPDNHVFRPSDULVRQVRIHIÀFLHQF\SUREOHPDWLF &RVWLQQG $O WKRXJK WKH FRUHEODGH WUDGLWLRQ LV SDQ0HVRDPHULFDQ WKH DUUD\ RI UHGXF tion techniques varied widely (Hirth and Andrews 2002). For example, core platforms at Teotihuacan were single facet, whereas they were pecked and ground at Xochicalco. Pecking and grinding produced tiny fracture cones that facilitated easier crack initiation during blade removal. Since Teotihua can platforms were single facet, requiring generally more force to press off a EODGHSHUKDSVJUHDWHUHIÀFLHQF\VNLOOZDVUHTXLUHGWRSUHYHQWWKHRFFXUUHQFH of hinge terminations. Experimental research has shown that blade remov als requiring more force can be more likely to hinge because, depending on RQH·VWHFKQLTXHWKHH[WUDHIIRUWFDQFRPSURPLVHDEODGHPDNHU·VSUHFLVLRQ (Andrews 2003).6 If so, then the hinge fracture data presented in this paper would tend to suggest that the San Martin craftsmen were more skilled. &OHDUO\VXFKWHFKQRORJLFDOGLIIHUHQFHVPD\VLJQLÀFDQWO\DIIHFWWKHIHDVLELO 272 Skill and the question of blade crafting intensity at Classic Period Teotihuacan LW\RIXVLQJVNLOOGDWDWRFRPSDUDWLYHO\LQIHUUHODWLYHOHYHOVRIEODGHPDNLQJ intensity at Teotihuacan and Xochicalco. These questions require further H[SHULPHQWDOUHVHDUFKEHIRUHZHUHOLDEO\PDNHLQWHUVLWHFRPSDULVRQVRIHI ÀFLHQF\VNLOO Conclusion I have counted the number of artifacts with hinges and overshot attributes and derived ratios of these artifacts to support inferences about the rela WLYHOHYHOVRIEODGHPDNLQJODERULQWHQVLW\DWWKH6DQ0DUWLQDQG2SHUDWLRQ +ZRUNVKRSV%DVHGRQWKHUHODWLRQVKLSEHWZHHQSUDFWLFHDQGSURÀFLHQF\ (Clark 2003), errors made during the production process were used to infer UHODWLYHOHYHOVRIEODGHFUDIWLQJLQWHQVLW\$OORWKHUWKLQJVEHLQJHTXDORQH·V skill should relate to the time spent making blades. Since the measures of skill for the San Martin and Operation H craftsmen appear similar, I suggest WKDWERWKZRUNVKRSVPLJKWKDYHKDGVLPLODUOHYHOVRISDUWWLPHLQWHQVLW\ )ROORZLQJ &ODUN·V DUWLFOH DGGUHVVLQJ WKH QDWXUH RI 7HRWLKXDFDQ·V obsidian tool industry, some studies have approached the issue of craft pro duction in the city more critically (Sheehy 1992; Widmer 1991). The earlier model positing a system of monumental scale, however, still seems to be prevalent in the literature (Millon 1988; Santley and Alexander 1993; Santley et al.6DQWOH\DQG3RRO:LGPHUVHH)HLQPDQ>@ for relevant commentary). Despite the problems with using the San Martin surface data, the intent of this study was to continue to question the earlier LGHDVDERXW7HRWLKXDFDQ·VFUDIWHFRQRP\,IWKH6DQ0DUWLQZRUNVKRSZDV DSDUWWLPHFRQWH[WRIEODGHSURGXFWLRQWKHQLWLVSUREDEOHWKDWWKHUHZDV enough aggregate demand for blades to support the amount its craftsmen could have been produced. Although local demand would have been high in a city of 125,000 people (Millon et al. 1973), the conclusion reached here is consistent with the sug gestion that the San Martin workshop and others like it probably did not export vast quantities of tools beyond the city (Clark 1986). This conclusion LVDOVRLQOLQHZLWK5DQGROSK:LGPHU·V VXJJHVWLRQWKDWWKHODSL GDU\ZRUNVKRSVDW7ODMLQJDGLGQRWSURGXFHIRUH[WUDORFDOH[SRUW,IWUXH WKHQLWLVOLNHO\WKDWIXWXUHUHVHDUFKZLOOIXUWKHUHVWDEOLVKWKDWWKHIDUÁXQJ LQÁXHQFHDQGLQWHUGHSHQGHQWVWUXFWXUHRI7HRWLKXDFDQ·VHFRQRP\KDVEHHQ overstated. 273 Bradford Andrews Acknowledgements I would like to extend my gratitude to Kjel Knutsson and Jan Apel for pro viding me with the opportunity to contribute to this volume. Data analysis of the Teotihuacan surface collections and the excavation and collection of the data from Xochicalco was provided by the National Science Foundation research grants 9420202, 9496188, and 9121949. I am indebted to Kenneth Hirth, Jeffrey Flenniken, and Gene Titmus for their help, assistance, and guidance throughout all steps in this research. Many thanks also to Pedro Baños and Seferino Ortega who kindly located the Teotihuacan collections for me and provided me space in which to analyze them at the lab in San -XDQ7HRWLKXDFDQ0RUHRYHU-RKQ&ODUN·VPHWLFXORXVIHHGEDFNRQWKHRULJL nal manuscript was invaluable. Anna Backer, Cathy Costin, Rand Gruebel, Akiko Nosaka, and Alan Reed also provided valuable comments. Finally, I greatly appreciate the efforts of Oralia Cabrera and Kristin Sullivan who or ganized the symposium “Craft Production at Terminal Formative and Clas sic Period Teotihuacan, Mexico” for the 69th annual meeting of the SAAs in which this paper was originally presented. Notes 1) The San Martin complex has never been excavated but holes for planting nopal cactus and a barrranca cut along its eastern margin reveal the direct association of ÁDNHGVWRQHWRROE\SURGXFWVZLWKGRPHVWLFKRXVHIHDWXUHV $QGUHZV 7KH location of the workshop on the periphery of the city, and the predominance of Middle to Late Classic ceramics suggest that it was primarily occupied during the Tlamimilolpa and Xolalpan phases (A.D. 300 to 650). Consequently, it is unlikely WKDW WKH ÁDNHG VWRQH WRRO E\SURGXFWV UHSUHVHQW GXPSV FRPSRVHG RI UHIXVH IURP workshops located elsewhere in the city. Given the considerably reduced population RIWKH3RVWFODVVLFFLW\FRQFHQWUDWHGZHVWRIWKH$YHQXHRIWKH'HDG GRZQ from a Classic high of 125,000 [Sanders et al. 1979]), it is unlikely that Epiclassic $' FUDIWVPHQZRXOGKDYHWUDYHOHGDOOWKHZD\WRWKHQRUWKHDVWHUQSH riphery of the former Classic city to dump their trash. 2) Widmer (1991:144) has proposed a similar model for the organization of lapidary production in the large extended family compound of Tlajinga 33. 274 Skill and the question of blade crafting intensity at Classic Period Teotihuacan ,UHFRJQL]HWKDWVNLOOLVDOVRUHODWHGWRRQH·VLQQDWHDELOLWLHV 2ODXVVRQ I assume here, however, that for workshops in extended family households, the more innately skilled individuals would have been those encouraged to excel in blade production. ([KDXVWHGFRUHVDUHWKRVHGLVFDUGHGDWWKHHQGRIWKHLUXVHOLYHV3UHVXPDEO\ FRUHEODGHNQDSSHUVUHJDUGHGWKHPDVWRRVPDOOIRUPDNLQJEODGHV 5) Another factor that can potentially affect the number of core artifacts in an as semblage is their differential consumption as a result of lateral cycling. For example, at Otumba and Xochicalco many exhausted cores were cycled into the lapidary LQGXVWU\WRPDNHEHDGV 2WLV&KDUOWRQ+LUWKet al. 2003, n.d.). In addition, a VLJQLÀFDQWQXPEHURIFRUHWRSVDQGFRUHVHFWLRQÁDNHVLQ;RFKLFDOFR·VZRUNVKRSV ZHUHIXUWKHUPRGLÀHGLQWRYDULRXVVFUDSLQJWRROVWKDWZHUHWKHQH[SRUWHGIRUFRQ sumption elsewhere in the city. These behaviors obviously affect the accuracy of any estimate of blade output based on core artifact frequency in a given assemblage. 6) Another technological factor that must be considered is the degree to which different blade removal techniques might have affected error rates. Experimental research has shown that many of the blades at Xochicalco were probably reduced with a handheld technique best suited to the small prismatic cores prevalent in the city (Flenniken and Hirth 2003). At Teotihuacan, in contrast, the pressure blade producing sequence started with much larger polyhedral cores. The size of these artifacts probably required a footheld technique permitting the generation of much more force than its handheld counterpart. More experiments are needed to see if, and under what conditions (i.e. core size, platform type), both techniques are associ ated with appreciably different rates of production errors. 275 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Mikkel Sørensen Rethinking the lithic blade GHÀQLWLRQWRZDUGVDG\QDPLF understanding Abstract 7KLV SDSHU GLVFXVVHV WKH FODVVLÀFDWLRQ RI WKH OLWKLF EODGH GHÀQLWLRQ LQ D historical perspective. Nowadays, there is a general agreement that lithic artefacts are produced intentionally through human action; this change of SHUFHSWLRQPXVWDOVREHUHÁHFWHGLQRXUGHÀQLWLRQV$VDQLOOXVWUDWLRQWRWKLV ZD\RIWKLQNLQJDQHZWHFKQRORJLFDOEODGHGHÀQLWLRQLVSUHVHQWHG 7KH EODGH GHÀQLWLRQ LV XVHG LQ D QHZ WHFKQRORJLFDO VWXG\ RI WKH (DUO\ Mesolithic Scandinavian (Maglemosian) lithic industry – a period in which VHYHQGLIIHUHQWEODGHSURGXFWLRQPHWKRGVKDYHEHHQLGHQWLÀHG7KHVHYHQ concepts, which form a process of technological change during the Maglem osian, are interpreted and explained by means of modern replication. 6RPH PDLQ EHQHÀWV RI WKH DSSOLFDWLRQ RI WHFKQRORJLFDO GHÀQLWLRQV DUH 1) we can focus on human behaviour in prehistory; 2) The traditional ar chaeological culture concept can be dissolved and human prehistory can be UHLQWHUSUHWHGDV´WUDGLWLRQVµGHÀQHGE\DPRQJRWKHUVWKHXVHRIVSHFLÀF technologies. Introduction to the problem 7UDGLWLRQDOOLWKLFDUWHIDFWGHÀQLWLRQVKDYHEHHQDGYRFDWHGE\DUFKDHRORJLVWV whose main goal was to construct a chronology of human prehistory. Today, we have established a general chronology of prehistory and the research fo cus has changed to address questions such as: how did people live, how did they organise their social life, how did they solve daily problems, and what did they generally think about life? Despite the fact that our research focus KDV FKDQJHG WUDGLWLRQDO SRVLWLYLVWLF PHWULFDO DUWHIDFW GHÀQLWLRQV DUH RIWHQ VWLOOLQXVHDQGUHVHDUFKUHVXOWVDUHWKHUHIRUHRIWHQLQÁXHQFHGE\UHGXQGDQW static perceptions of human prehistory. One solution to this problem is to FUHDWHSUHVHQWDQGGLVFXVVQHZGHÀQLWLRQVZKLFKLQFOXGHRXUSUHVHQWNQRZO edge of the chaîne opératoire and lithic technology. This paper focuses on lithic DUWHIDFW GHÀQLWLRQV XVLQJ DQ H[DPSOH RI RQH DUWHIDFW W\SH WKH EODGH DQG viewing it from a historical perspective. 277 Mikkel Sørensen 7KHOLWKLFEODGHFODVVLÀFDWLRQLQDKLVWRULFDO perspective A lithic artefact type that we generally perceive as known and agreed upon is the blade. Blades are often not secondarily worked or retouched and were tradi tionally believed not to be tools. Neither can they be described as waste VLQFHWKH\RIWHQDUHPXFKWRRUHJXODUDQGDVVXFKEODGHVGRQRWÀWLQWRRXU pyramided hierarchy of artefacts (Fig. 1). This problem is often dealt with by placing blades in their own artefact category. Yet, drawing a line between waste and blades just results in more problems when classifying: where do preparation blades and crested blades belong? What about hinged and bro ken blades? What happens with the atypical or unsuccessfully made blades or rejuvenation blades? )LJXUH$OLWKLFDVVHPEODJHFODVVLÀHGXVLQJWUDGLWLRQDOW\SRORJ\$UWHIDFWVDUHSHUceived as parts of a hierarchy, where the formal tool types are at the top and the remainder (majority of the assemblage - so-called waste) at the bottom. But what about blades? Blades do not seem to be real tools, but at the same time are perceived to be WRRUHJXODUWREHFODVVLÀHGDVZDVWH2ZLQJWRWKLVEODGHVDUHRIWHQFODVVLÀHGDVDGLVtinct group situated between the tools and waste in the traditional artefact hierarchy. $SUREOHPZLWKWKLVFODVVLÀFDWLRQLVKRZZHGHÀQHEODGHVLQRUGHUWRVHSDUDWHWKHP IURPZDVWH ÁDNHV :KHUHGREURNHQEODGHVEHORQJDQGZKDWDERXWFRUWLFDODW\SLFDO blades or blade rejuvenations? 278 Rethinking the lithic blade deﬁnition: towards a dynamic understanding 7KURXJKRXW WKH KLVWRU\ RI OLWKLF DQDO\VLV DQG FODVVLÀFDWLRQ PDQ\ DW tempts were made by archaeologists to solve these problems through the use RIGLIIHUHQWEODGHGHÀQLWLRQV7KLVDXWKRUZLOOVWDUWRXWE\RXWOLQLQJDVKRUW KLVWRU\RIWKHEODGHGHÀQLWLRQVDQGFRQFHSWLRQVLQUHODWLRQWRWKHRUHWLFDODS proaches in archaeological history. 'XULQJWKHVRFDOOHG(DUO\(PSLULFLVP WKFHQWXU\ RQO\DIHZDUFKDH RORJLVWVZHUHFRQFHUQHGZLWKEODGHGHÀQLWLRQVDIDFWWKDWZDVSUREDEO\UH ODWHGWRWKHODWHDFNQRZOHGJHPHQWRIWKHROGHVW6WRQH$JH2QHRIWKHÀUVW DUFKDHRORJLVWV ZKR DQDO\VHG D ÁLQW DVVHPEODJH ZLWK D EODGH LQGXVWU\ ZDV Sophus Müller in his treatment of the shell midden at Ertebølle (Müller 1900). In order to classify and analyse the late Mesolithic assemblage, Müller SUHVHQWHGZKDWZDVSUREDEO\WKHÀUVWXVHRIDEODGHDQGDÁDNHGHÀQLWLRQLQ a Mesolithic context. By studying the assemblage carefully he stated: During the excavation, all detached removals that were thought to derive from the blade production were collected. Not only complete pieces, but also broken and small pieces; not just beautiful and regular blades in the most limited sense of the word, together with crested- and edged blades, but also irregular and possible blades. In all, every detached removal that was thought to be produced with the intention that it should be a blade was collected 0OOHU 7KLVDXWKRU·VWUDQVODWLRQ 7KLVVSHFLÀFEODGHGHÀQLWLRQLQFOXGHG“... all pieces that, more or less, have the (following) described unique morphology: the oblong, narrow and thin, intended and regular removal with a sharp and straight lateral edge, which is appropriate as a cutting edge” (Müller 1900:42) (This is the au WKRU·VWUDQVODWLRQ 7KLVGHÀQLWLRQZDVSXWIRUZDUGLQRSSRVLWLRQWRWKHIRO ORZLQJÁDNHGHÀQLWLRQ´WKHLUUHJXODUPRUHFRLQFLGHQWDOSDUWO\HLWKHUZLGH RU WKLFN ÁLQW GHWDFKPHQW VKDUS HGJHG EXW ZLWKRXW D ORQJHU ZHOO GHÀQHG cutting edge” 0OOHU 7KLVLVWKHDXWKRU·VWUDQVODWLRQ It should EHVWUHVVHGWKDW0OOHU·VEODGHGHÀQLWLRQRQO\FRQFHUQHGPDWHULDOIURPWKH (UWHE¡OOHH[FDYDWLRQDQGWKDWWKHGHÀQLWLRQLVWKHUHIRUHVSHFLÀFWRD/DWH Mesolithic artefact assemblage. It must also be noticed that Müller considered prehistoric intentions in the industry, and that it is on the basis of both morphology and ideas about LQWHQWLRQVWKDWKHVHSDUDWHVEODGHVIURPÁDNHV,QPDQ\ZD\V0OOHU·VGHÀ nitions must, therefore, be considered as well in line with what came out of a similar research interests at the end of the 20th century. During the period of Cultural (normative) Archaeology from 1900 to 1960 2OVHQ EODGHVDUHGHÀQHGYDJXHO\RQWKHEDVLVRIRQO\DIHZPRUSKR ORJLFDODQGPHWULFDOFULWHULD7KLVLVVHHQH[HPSOLÀHGLQWKHZRUNRI0DWKL DVVHQ ZKRSUREDEO\RQWKHEDVLVRI0OOHU·VZRUNGLYLGHGEODGHVLQ 279 Mikkel Sørensen regular and irregular pieces. This less precise terminology was generally not DUJXHGIRUDQGQRRYHUDOOGHÀQLWLRQRIEODGHVZDVSUHVHQWHGEHVLGHVLW During the period of Positivism in Scandinavia in the 1960s, objective GHÀQLWLRQV XVLQJ PHWULFDO FODVVLÀFDWLRQV DQG VRPHWLPHV TXDQWLÀFDWLRQV RI VSHFLÀFDWWULEXWHVZHUHRIWHQDWWHPSWHG$OWKLQ SXWIRUZDUGWKHµ$ EODGHµDQGµ%EODGHµFDWHJRULHVZKHUH$EODGHVKDYHSDUDOOHOVLGHVDQGD OHQJWK²ZLGWKUDWLRRIDWOHDVWZKLOH%EODGHVGRQRWKDYHSDUDOOHOVLGHV and a length–width ratio of less than 2:1. 0DOPHU LVWDNLQJWKHGHÀQLWLRQVHYHQIXUWKHUDLPLQJIRUDQ´RE MHFWLYHPHWULFDOWUXWKµ'HÀQLQJWKHDPRXQWRIGRUVDOULGJHVSURSRUWLRQDO to the regularity and the quality of the blade, he formed an index for blade UHJXODULW\ ,Q 0DOPHU WKXV UHXVHV $OWKLQ·V $ DQG % FDWHJRULHV EXW adds the amount of dorsal ridges and the thickness of the blades as metrical SDUDPHWHUV WR WKH GHÀQLWLRQ 0DOPHU $W WKH VDPH WLPH KH GHÀQHV microblades as shorter than 5 cm in length, with a length–width ratio of 5:1. It has to be mentioned that Malmer was primarily interested in chronologi FDOTXHVWLRQVDQGWKDW0DOPHU·VPHWULFDOGHÀQLWLRQRIPLFUREODGHVLVRQO\ UHOHYDQWIRU6FDQLDDQGIRUÁLQWDVVHPEODJHV+RZHYHULQJHQHUDOKLVGHÀQL tions were used from then on. From a present day perspective, these attempts can be criticised for two major reasons. First, we now are aware that objectivity is an illusion. The REYLRXV SDUDGR[ OLHV ZLWK WKH SRVLWLYLVWLF GHÀQLWLRQV WKHPVHOYHV QRWKLQJ seems more subjective than to force a lithic assemblage into some subjective chosen metrical categories. Second, through the use of one metrical blade GHÀQLWLRQ DQG WKH VHOHFWLRQ RI RQO\ VRPH DVSHFWV RI WKH EODGH SURGXFWLRQ sequence, the lithic analyst misses the opportunity to recognise different technologies, to specify blade concepts and culture/groups or to interpret social interactions through technology. ,Q6$6HPHQRYVXJJHVWHGDIXQFWLRQDOGHÀQLWLRQRIWRROW\SHV But since blades were produced using different techniques, and since blades JHQHUDOO\DUHSUHIRUPVIRUIXQFWLRQDOWRROW\SHVDQGKDGPXOWLSXUSRVHIXQF tions throughout prehistory, morphology still seemed to be more reliable for FODVVLÀFDWLRQ %RUGHV In the French tradition of lithic analysis, J. Tixier (1963) presented metri FDOGHÀQLWLRQVIRUEODGHV ODPHV PLFUREODGHV ODPHOOHV DQGÁDNHV pFODWV LQ KLVZRUNRIWKH(SL3DODHROLWKLFLQ1RUWK$IULFD7KHVHGHÀQLWLRQVDUHEDVHG on measurements of width/length and correspond in general with the earlier 6ZHGLVKGHÀQLWLRQVEXWLQFRQWUDVWWRWKH6ZHGLVKDWWHPSWV7L[LHUFRPSLOHG KLVGHÀQLWLRQVRQWKHEDVLVRIDVSHFLÀF (SL3DODHROLWKLF FRQWH[WDQGQRWDV JHQHUDOGHÀQLWLRQV In the French tradition, technology was linked to sociology very early on 280 Rethinking the lithic blade deﬁnition: towards a dynamic understanding (e.g. Mauss 1927, 1947), and studies involving artefact replication opened the eye for a description of the artefacts based on an understanding of the lithic knapping process. A sociological interpretation of the knapping process was VXJJHVWHG /HURL*RXUKDQ LQZKLFKDUWHIDFWVZHUHVHHQLQUHODWLRQWR human acts in the lithic production sequence: the use, the resharpening and WKHÀQDOGLVFDUG7KHWHFKQRORJLFDODQGVRFLRORJLFDOSHUFHSWLRQRIOLWKLFDUWH IDFWVDOVRUHVXOWHGLQDQDYHUVLRQWRWKHXVHRIJHQHUDOIRUPDOFODVVLÀFDWLRQV This was for instance clearly stated by F. Bordes in 1961: English-speaking authors, among others, make a distinction between true blades DQGEODGHOLNHÁDNHV $WUXHEODGHVKRZLQJWUDFHVRISUHYLRXVSDUDOOHOUHPRYDOV RQ LWV XSSHU IDFH DQG DOVR KDYLQJ PRUH RU OHVV SDUDOOHO HGJHV $OWKRXJK WKH GLVWLQFWLRQLVSHUIHFWO\YDOLGLQWKHRU\LWLVRIWHQGLIÀFXOWWRPDNHLQSUDFWLFHDQGZLOO WKHUHIRUHEHGLVUHJDUGHG 7UDQVODWLRQE\,QL]DQHWDO During the “New Archaeology” 1970–1990, numeric values were needed for a statistical treatment of the artefacts and, here, lithic assemblages presented a perfect subject of study, since they were both numerous and could be meas ured in many ways. The “new archaeologists” were readily able to use the SRVLWLYLVWLFPHWULFDOGHÀQLWLRQVIURPWKHVDVDEDVLVIRUDTXDQWLWDWLYH analysis. Today, we often discover that quantitative investigations based on PHWULFFODVVLÀFDWLRQVFRQWDLQDUWHIDFWVSURGXFHGZLWKGLIIHUHQWSXUSRVHVDQG that each metrical artefact category includes different technological produc tions. The results of such quantitative investigations are sparse, and can be GHVFULEHGDVVRPHVWDWLVWLFWHQGHQFLHVZKLFKDSSHDUWREHVLJQLÀFDQWLQWKH DVVHPEODJHDQGWKHDFKLHYHPHQWRIVLPSOHJRDOVVXFKDVVSDWLDOTXDQWLÀFD tion of the lithic distribution. The general problem with the quantitative analysis is that the question why there are statistical differences cannot be discussed, due to the above mentioned reasons, and therefore, essential questions about prehistoric human behaviour are often ignored when a statistical analysis is carried out. However, “New Archaeology” also generated an experimental “faction” in which researchers tried to formulate Binfordian “Middle Range Theories” by using experimental archaeology in combination with statistic methods (e.g. Madsen 1992; Hansen and Madsen 1983; Knutsson 1988). Many of these studies are based on solid technological knowledge, and in most studies fo FXVFKDQJHGIURPPHDVXUHPHQWVRIDUWHIDFWVWRZDUGVVWXGLHVRIVLJQLÀFDQW attributes and micro morphologies, an approach which certainly still is valid HJ3HOHJULQ 1HYHUWKHOHVVWKH´ROGµEODGHGHÀQLWLRQDOVRVXUYLYHGWKH ÀUVWJHQHUDWLRQRIH[SHULPHQWDODUFKDHRORJLVWVHYHQWKRXJKDPXFKGHHSHU technological understanding, of e.g. blade production, was gained and ex pressed during this period. 281 Mikkel Sørensen The new dynamic technological blade GHÀQLWLRQDQG0DJOHPRVLDQEODGHV &OHDUO\WKHWLPHKDVFRPHWRUHHYDOXDWHWKHH[LVWLQJEODGHGHÀQLWLRQEXW ZKDWIRUPVKRXOGLWWDNH"$VDUHVXOWRILQWHQVLYHUHSOLFDWLRQDQGUHÀWWLQJ studies carried out during the last two decades, we now have an elaborated NQRZOHGJHDERXWÁLQWWHFKQRORJ\DQGEODGHSURGXFWLRQ HJ%RGXHWDO Coulson 1986; Fischer 1990; Inizan et al. 1999; Madsen 1992; Pelegrin 1984a, 1984b, 1988, 1995, 2000, 2002, Pigeot 1990; Skar 1987; Sørensen 2000; Sø UHQVHQLQSUHVV DE 'XHWRWKLVLWVHHPVOLNHO\WKDWDQHZGHÀQLWLRQFRXOG arise from a dynamic technological understanding. Figure 2. Blade production by means of direct hard percussion using a quartzite stone hammer. The core is held at the thigh and struck at an angle of approximately 70 degrees. The blades are irregular and characterized by e.g. large bulbs, impact cones on the butt. A similar method is carried out during the Bromme Culture (Madsen 1992). 1RWHWKDWWKLVYHU\VLPSOHEODGHSURGXFWLRQFDQQRWEHGHÀQHGXVLQJDWUDGLWLRQDOGHÀnition due to its irregularity. However, the process (strategy and concept) reveals a true blade concept. Photo J. Sørensen. 282 Rethinking the lithic blade deﬁnition: towards a dynamic understanding A study of the blade concepts in the Early Mesolithic (Maglemosian, ca. 9000–6000 BC) in Scandinavia and Northern Germany (Sørensen in press D LQSUHVV E OHDGVWRWKHGHÀQLWLRQSUHVHQWHGLQWKLVSDSHU+DYLQJH[ amined the lithic chaînes opératoires, and especially the step involving blade production, on six Maglemosian sites and 17 additional sites, it became ap SDUHQWWKDWWKHWUDGLWLRQDOOHQJWKZLGWKGHÀQLWLRQVZHUHLQVXIÀFLHQW/LWKLF UHPRYDOVZHUHVHULDOO\SURGXFHGIURPFRUHVDQGXVHGIRUPLFUROLWKDQGWRRO production through the entire period. However only during the late Maglem osian (phase 3,4,5) (Petersen 1973) the serially produced removals made for tools and microliths has a morphology that can “allow” the term “blades” ZLWKLQ WKH GHÀQLWLRQ IURP WKH V ,W PDGH QR VHQVH WR WKLV DXWKRU WR describe the Maglemosian as a period where “blades” were produced only Figure 3. Blade production by means of direct hard percussion using a quartzite stone hammer. The core is held on the ground and struck at an angle of approximately 70 degrees. The blades are large, irregular and characterized by a distinct set of attributes: large bulbs, impact cones on the butt. A study of the prehistoric blade production and a comparison of the blades and their attributes to original and recently produced blades concluded that the direct hard hammer technique was used during the earliest Early Maglemosian Period in Southern Scandinavia. The production of big irregular blades was, during the Early Maglemosian and the Ertebølle Period, used to make preIRUPVIRU´ÁDNHD[HVµ1RWHWKDWWKLVYHU\VLPSOHEODGHSURGXFWLRQFDQQRWEHGHÀQHG XVLQJDWUDGLWLRQDOGHÀQLWLRQGXHWRLWVLUUHJXODULW\+RZHYHUWKHSURFHVV VWUDWHJ\DQG concept) reveals a true blade concept. Photo J. Sørensen. 283 Mikkel Sørensen within the last phases of this period, when the concept for this production existed throughout the entire period (see also Knutsson 1981). A further study of the lithic technology within the Maglemosian investi gated how the lithic blade production methods developed. The investigation comprised analysis of Maglemosian blade attributes from six Maglemosian sites (one from each phase), series of thoroughly documented replicative studies, and the use of knowledge about the relation between attributes and VSHFLÀFNQDSSLQJWHFKQLTXHV7KLVLQYHVWLJDWLRQUHVXOWHGLQWKHIROORZLQJLQ terpretation of the technological development in blade production methods and techniques during the Maglemosian: In the Early Maglemosian (early SKDVH D VLPSOH EODGH SURGXFWLRQ PHWKRG WHFKQRORJLFDO GHÀQLWLRQ LQ which irregular removals were serially produced by direct percussion with Figure 4. Blade production by means of direct soft percussion, using a soft stone (chalk/sand/limestone) or antler hammer. The core is held on the thigh, the platform edge is heavily trimmed by the hammer, and struck in an acute angle (approximately 20 degrees). The blades are thin, irregular and characterized by a distinct set of attributes: small bulbs, lib formation, splintered butts (“fracture lanquette”), and pronounced waves/ripples. Note that many blades produced by this simple blade production cannot EHGHÀQHGXVLQJDWUDGLWLRQDOGHÀQLWLRQGXHWRLUUHJXODULW\3KRWR-6¡UHQVHQ 284 Rethinking the lithic blade deﬁnition: towards a dynamic understanding DKDUGKDPPHUVWRQHIURPXQLSRODUFRUHVZHUHSHUIRUPHG )LJ ,QWKLV period, as in the rest of the Maglemosian, blades were used for microliths and blade tools. However, within the Barmosen group (Johansson 1990) a special blade method was used for the production of heavy broad blades: it ZDVSHUIRUPHGE\DKDUGKDPPHUIURPODUJHXQLSRODUFRUHVZLWKÁDWIURQWV WKHEODGHVZHUHXVHGDVÁDNHD[HSUHIRUPV )LJ 'XULQJWKHQH[WSKDVHV (late phase 0, and phase 1,2) a method in which thinner irregular remov als were serially produced by direct percussion with a soft stone or antler KDPPHUZDVFRQGXFWHG&RUHVZHUHJHQHUDOO\VLQJOHIURQWHGRIWHQXQLSRODU EXWVRPHWLPHVDOVRGXDOSODWIRUPHGSULVPDWLF$W\SLFDOFRUHDWWULEXWHGXU ing this period is the approximate angle of 70 degrees between front and platform (Fig. 4). Within the Late Maglemosian (phase 3), the technological climax of blade production arrives. Technically complex processes, which IRUWKHÀUVWWLPHLQYROYHDVHSDUDWHFRQFHSWIRUPLFUREODGHSURGXFWLRQDUH Figure 5. Blade production by means of pressure technique (debitage), using a composite pressure tool with antler tip. The core is held in a loose “V” shaped clamp, VXSSRUWHGDWWKHERWWRP7KHSUHVVXUHVWLFNLVÁH[LEOH'XULQJWKHPRPHQWRIGHELWage the stick is pushed while, at the same time, it is moved towards the edge of the platform, resulting in a blade removal. The blades are thin, regular and prismatic, characterized by the following attributes: a combination of lip and bulb, lens shaped butts, and regularity, lack of waves/ripples. Photo J. Sørensen. 285 Mikkel Sørensen FRQGXFWHGE\SUHVVXUHÁDNLQJ )LJ $WWKHVDPHWLPHLQGLUHFWSHUFXV VLRQ DSSHDUV IRU WKH ÀUVW WLPH LQ 6FDQGLQDYLD 7KH LQGLUHFW SHUFXVVLRQ LV HPSOR\HGIRUWKHPDFUREODGHSURGXFWLRQ )LJ %RWKPLFURDQGPDFUR blades are, during this period, very regular, prismatic and straight. Cores are typically used up to a circular conical “bullet shape”. Towards the end of the Maglemosian (phase 4,5) techniques are in general maintained, however core morphology is changed towards oblong keeled micro cores (handle cores) (Fig. 7) and heavier macro cores with (more) curved fronts, resulting in heav ier and more curved macro blades (Fig. 8). The reason for the increased cur vature might be that cores are now unsupported at their bottom (e.g. Bordes and Crabtree 1969). These late Maglemosian blade production methods are maintained, with minor changes in core morphology and technique, during the Kongemosian period (middle Mesolithic of southern Scandinavia), and perceived as the classical blade and microblade productions of the Meso lithic Southern Scandinavia. Figure 6. Blade production by means of indirect percussion, using a curved red deer punch. The core is held between the feet, supported at the bottom, while the punch is gently hit by a billet of hard wood (elm). The blades are thin, regular and prismatic, characterized by the following attributes: a combination of lip and bulb, lens shaped butts, and regularity, straightness, lack of waves/ripples. Photo J. Sørensen. 286 Rethinking the lithic blade deﬁnition: towards a dynamic understanding Concerning the blade production, the conclusion was that serially produced removals made for microliths and tools were produced throughout the entire Maglemosian by the same overall idea (serial production of blades from plain platforms). However, the removals were produced by means of different methods and techniques, which resulted in distinct changes in morphology of the removals. Further, the study demonstrated that seven different con cepts of blade production were conducted during the Maglemosian and that a technological change took place such that four traditions (groups), each with its own concept of blade production, existed diachronically during the Maglemosian (Fig. 9). The fact that blades were recognized through a study RIWKHOLWKLFSURFHVVDQGQRWE\PHWULFDORUPRUSKRORJLFDOGHÀQLWLRQVPDGH it possible to put forward an alternative technological chronology, which facilitates relative dating of blades, cores, punches and pressure tools from the Maglemosian (Sørensen in press (a,b)), whereas the traditional typologi cal chronology, consisting of 6 phases, is based only on frequencies of one artefact type: the microlith (Petersen 1973). Figure 7. Blade production by means of indirect percussion, using a red deer antler punch. The core is held between the thighs. The blades are regular, curved and prismatic, characterized by a distinct set of attributes e.g. a combination of lib and bulb, lens shaped butts, regularity and curvature. Photo J. Sørensen. 287 Mikkel Sørensen $QRWKHUSUREOHPFRQFHUQLQJDPHWULFDOEDVHGEODGHGHÀQLWLRQZDVWKDWLW included removals from steps of other technological productions, e.g. from the trimming of macro blade cores and from core axe production, but at the same time clearly excluded intentionally made, but fractured, blades. The question that arose while analysing the Maglemosian blade industry was basically how to describe and classify the lithic material from the step RIWKHEODGHSURGXFWLRQZLWKRXWEHLQJIRUFHGWRUHÀWWKHHQWLUHOLWKLFPDWH rial (which would be an impossible, illusory task)? This question was solved E\ GHÀQLQJ EODGHV DV UHVXOWV RI D SURGXFWLRQ PDGH E\ FOHDUO\ LGHQWLÀDEOH intentions. The material intention for the blade production was seen as the stage in the châine opératoire in which blades were chosen for tools (namely PLFUROLWKV ,QRWKHUZRUGVWKHQHZEODGHGHÀQLWLRQZRXOGFRQFHUQ´UHPRY als made with the intention of being tools or preforms for tools”. Serially produced removals, intended to be preforms or tools, from each period of Figure 8. Blade production by means of pressure technique (debitage), using an antler SUHVVXUHWRRO7KHFRUHLVÀ[HGLQDZRRGHQGHYLFHE\WZR´MDZVµ7KHSUHVVXUHVWLFN LVRQO\VOLJKWO\ÁH[LEOH'XULQJWKHPRPHQWRIGHELWDJHWKHVWLFNLVSXVKHGZKLOHDWWKH same time it is moved towards the edge of the platform, resulting in a blade removal. The blades are thin regular and prismatic, characterized by a set of attributes such as a combination of lip and bulb, relatively large lens shaped butts, regularity and lack of waves/ripples. Photo J. Sørensen. 288 Rethinking the lithic blade deﬁnition: towards a dynamic understanding the Maglemosian could then be termed “blades” and be described differ ently on the basis of morphology and attributes diagnostic to their technical manufacturing. According to the concept of the châine opératoire and in accordance with WRGD\·VWHFKQRORJLFDONQRZOHGJHRIGLDJQRVWLFDWWULEXWHVWKHIROORZLQJQHZ G\QDPLFDOWHFKQRORJLFDOGHÀQLWLRQRIDEODGHFDQEHSUHVHQWHG ”A blade is a serially produced removal made with the intention of being a tool or a preform for a tool. Blades in the same industry are produced by the same technique, method and mental representations and are characterised by a similar morpholog y and the same set of diagnostic attributes”. 7KLVGHÀQLWLRQLQFOXGHVVHULDOO\SURGXFHGUHPRYDOVZKLFKFRXOGQRWEHLQ FOXGHGLQHDUOLHUGHÀQLWLRQVRQJURXQGVRIWKHLULUUHJXODUPRUSKRORJ\HYHQ WKRXJKWKH\ZHUHVHULDOO\SURGXFHGIRUGLVWLQFWXVH7KHQHZEODGHGHÀQLWLRQ can also include blade fragments as long as they have the same morphology and set of attributes as other blades within the industry. 6RPHH[DPSOHVRIVHULDOO\SURGXFHGUHPRYDOVWKDWFDQQRZEHUHGHÀQHG as blades are: 7KH SURGXFWLRQ RI ODUJH UHPRYDOV IRU ÁDNH D[H SURGXFWLRQ 7KLV blade concept was predominantly used in the Early Maglemosian Period in Southern Scandinavia, (the Barmosen Group), (Johansson 1990) and also in the Ertebølle Period (Müller 1900) (Fig. 3). The production of removals for the manufacture of transverse ar rowheads in the Neolithic Period. This production concept can be described as a serial production of irregular broad blades from dis coidal cores. With regard to the Levallois method, intentional serially produced UHPRYDOVIURP/HYDOORLVFRUHVVKRXOGEHFODVVLÀHGDVEODGHV7KLV means that when a recurrent Levallois method (i.e. recurrent unipolar, -bipolarRUcentripetal ) method is applied (Boëda 1994) the results will EHUHPRYDOVGHÀQHGDVEODGHV0HDQZKLOHZKHQRQO\DVLQJOHUH moval is made from a prepared Levallois core, (i.e. the preferential or classical PHWKRG WKHLQWHQGHGUHPRYDOPXVWEHGHÀQHGDVDÁDNH,Q praxis, this difference in naming the Levallois removals will be hard to maintain because it is hardly possible to judge from the Levallois removal itself whether it is produced serially or preferentially. How ever, the preferential method, when investigated, is interpreted to be rare compared to the recurrent method (Schlanger 1996). Owing 289 Mikkel Sørensen to this, this author will argue that removals detached from Leval lois cores in general, and in agreement with the common perception (Inizan et al. VKRXOGEHGHÀQHGDVEODGHV $VDFRQVHTXHQFHRILQFOXGLQJLQWHQWLRQVLQDUWHIDFWGHÀQLWLRQVÁDNHVDV RSSRVHGWREODGHVFDQEHGHÀQHGDV x Waste removals (i.e. serially produced removals which are knapped off to shape cores and blanks). x Removals which are not serially produced. Figure 9. The Maglemosian chronology based on a study of blade technology (Sørensen in press (a, b)). Seven different concepts of blade production were conducted during the Maglemosian in a technological development . Four traditions (groups), each with its own concept of blade production, existed diachronically during the Maglemosian. The fact that blades were recognized through a study of the lithic process, and not by PHWULFDORUPRUSKRORJLFDOGHÀQLWLRQVPDGHWKLVDOWHUQDWLYHWHFKQRORJLFDOFKURQRORJ\ possible. The chronology facilitates relative dating of blades, cores, punches and pressure tools from the Maglemosian, whereas the traditional typological chronology, consisting of 6 phases, is based on frequencies of one artefact type: the microlith. (Analysis DQGÀJXUHE\06¡UHQVHQ)URPWKHVLV´7KH0DJOHPRVLDQ%ODGH&RQFHSWµ 6¡rensen in press)). 290 Rethinking the lithic blade deﬁnition: towards a dynamic understanding 291 Mikkel Sørensen Explaining the technological dynamical blade GHÀQLWLRQ 7KHQHZEODGHGHÀQLWLRQFDQEHVXSSRUWHGWKURXJKUHÀWWLQJUHSOLFDWLRQDQG WHFKQRORJLFDOG\QDPLFDOFODVVLÀFDWLRQ 5HÀWWLQJWKHLGHDOEXWWLPHFRQVXPLQJDSSURDFK $VWURQJDUJXPHQWIRUWKHQHZGHÀQLWLRQLVWKHVWXG\RIEODGHSURGXFWLRQ WKURXJKUHÀWWLQJ6HYHUDOLQYHVWLJDWLRQVRIchaînes opératoiresWKURXJKUHÀWWLQJ especially on Palaeolithic assemblages, clearly demonstrate that what we so IDUKDYHXQGHUVWRRGDVEODGHVLQJHQHUDODUHVHULDOSURGXFHGUHPRYDOVZLWK the same set of attributes and a similar morphology (e.g. Coulson 1986; Skar & Coulson 1989; Skar 1987; Pigeot 1990; Fischer 1990; Fiedorzcuk 1995; Johansen 2000; Bodu et al. 1990). Lithic replication: the experimental approach 'XULQJWZRGHFDGHVRISHUVRQDOH[SHULPHQWDWLRQZLWK0HVROLWKLFÁLQWWHFK nology, blade replication and comparison with prehistoric material, the au thor, as well as several other modern knappers (e.g. Bordes & Crabtree 1969; Flenniken 1987; Madsen 1992; Pelegrin 1984(b); 1988; 1991), have come to the conclusion that blades have to be serially produced in sequences with a consistent technology, i.e. the working tools (hammers, punches, billets, etc.), the angle of percussion, the force applied, the preparation of each blade (trimming), and the morphology of the core have to be identical to achieve even blades in a production. In addition, the (modern) knapper soon realises that the same main intention is required throughout the entire process of blade production in order to fully succeed. This intention can be described as a visualised representation of an ideal blade. It is generally experienced that this consistency in technique, method, and concept of the blade production results in removals (blades) with the same VHWRIGLDJQRVWLFDWWULEXWHVDQGDVLPLODUPRUSKRORJ\ )LJV 7KHWHFKQRORJLFDOG\QDPLFDOFODVVLÀFDWLRQWKHUHDOLVWLF approach 7KHWKLUGPHWKRGLVWKHG\QDPLFWHFKQRORJLFDOFODVVLÀFDWLRQZKLFKLVRXW lined by R. Schild (1980). Two observations suggest that blades are serially 292 Rethinking the lithic blade deﬁnition: towards a dynamic understanding produced: (1) Regular negative removal scars on core pieces and rejuvenation ÁDNHVVKRZWKDWDVHULHVRIUHPRYDOV i.e. sequence) have been produced. Since the scars have the same appearance (size and morphology) it is probable that the removals has a similar morphology, have been produced using the same technique and have the same set of attributes. (2) The fact that cores (at least in the European Palaeolithic and Mesolithic) are generally found fully ex hausted and not preserved from the step of blade production (ready for the next blade) suggests that blade production was generally carried out in one coherent process. Blades must therefore commonly be regarded as having been serially produced. The conclusion of the dynamical technological anal ysis of blades and blade production is that blades in most prehistoric cases were not produced as single events. Instead, a blade production must gener ally be characterised as a serial mass production from one core at a time. Discussion :KHQDQDO\VLQJOLWKLFDVVHPEODJHVDQGWU\LQJWRGHÀQHFRQFHSWVRQWKHEDVLV of blade productions, a methodological problem is that not all blade produc tions on a site or within a culture/group are technically of the same quality. This problem can be explained through differential skill levels, as not all in dividuals will be able to perform the ideal concept of the production within the same group. Several technological analyses of lithic artefact assemblages have suggested that children will play with lithics and simulate the produc tion process without being conscious of the concept, while apprentices, due WR ODFN RI NQRZOHGJH DQG NQRZKRZ ZLOO QRW DOZD\V EH DEOH WR FDUU\ RXW the chaînes opératoires typical for the technological tradition (e.g. Knutsson 1986; Apel 2001; Bodu et al. 1990; Fischer 1990; Högberg 2001; Pigeot 1990; Karsten and Knarrström 2003). In this respect, it seems clear that not every lithic chaîne opératoire in an assemblage can be described as typical for the culture/group itself. 7KH ORJLFDO VROXWLRQ WR WKLV PHWKRGRORJLFDO SUREOHP LV WR ÀQG DQG GH scribe those chaînes opératoires which contain (and lead to) the curated tools, and to regard these as the norm in the investigated group/culture. This ar gument can be supported using the rationale that those chaînes opératoires that contain curated tools must be produced by trained knappers who are able to perform the concept of the group/culture. As a general rule, a lithic analyst WU\LQJWRGHÀQHWKHW\SLFDOZRUNLQJFRQFHSWVLQDJURXSFXOWXUHVKRXOGDYRLG describing training sequences, expediently made household productions and simulations as intentions carried out typical for the group/culture. In con trast, archaeologists analysing a synchronous situation, describing single pre 293 Mikkel Sørensen historic events and social interactions, should be aware of all the different chaînes opératoire, since they represent different human actions, individuals, learning situations and social relations on the site. This author is well aware of the fact that the new dynamic technologi FDOEODGHGHÀQLWLRQZLOOEHPRUHGLIÀFXOWWRXVHWKDQWKHIRUPHUPHWULFDO GHÀQLWLRQV$UFKDHRORJLVWVXVLQJWKHQHZGHÀQLWLRQIRUDVVHPEODJHDQDO\VHV require a good general knowledge of lithic technology to be able to recognize the chaîne opératoireIRUWKHEODGHSURGXFWLRQDQGVWDWHWKHVSHFLÀFEODGHPRU phology. Unfortunately, the dynamic technological methodology is still not FRPPRQO\XVHGDQGRQO\VRPHSUHKLVWRULFFXOWXUHVJURXSVDUHVXIÀFLHQWO\ described, a fact that often leaves archaeologists who want to understand and work with lithic dynamics in a “pioneer” situation. Due to this situation, it is necessary to describe what kind of studies and methodologies are required for a dynamical understanding of lithic production. Fortunately, the solution to this question is well demonstrated in a number of case studies (e.g. Bodu et al. 1990; Boëda 1988; Coulson 1986; Fischer 1990, Johansen 2000; Madsen 1992; Pelegrin 1995; Schild 1980; Skar 1987) which all successfully use one or a combination of the following three methods in explaining technological PHWKRGVDQGFRQFHSWVUHÀWWLQJOLWKLFUHSOLFDWLRQDQGRUG\QDPLFDOWHFKQR ORJLFDOFODVVLÀFDWLRQ A conclusion is therefore that a dynamic technological understanding of lithics can be acquired through the use of one, or better still, a combi nation of the above named methods. Another issue this author would like WRVWUHVVLVWKDWWKHEODGHGHÀQLWLRQSUHVHQWHGFDQEHXVHGVXFFHVVIXOO\LQ DQDO\VLV ZKHUH WKH UHVHDUFK TXHVWLRQV DUH UHÀQHG WRZDUGV GLDFKURQRXV RU social matters, while in cases where the goals are limited and descriptive, PHWULFDOEDVHGEODGHGHÀQLWLRQVFDQFHUWDLQO\EHXVHG7KHFUXFLDOREMHFWLYH WKDWVKRXOGEHDFFHSWHGRQZDUGVLVWKDWPRUHWKDQRQHGHÀQLWLRQZLWKLQD problem area can exist, and that archaeologists should always explicitly state WKHFKRVHQGHÀQLWLRQ:HDOVRKDYHWRDFFHSWWKDWFKDQJHVDQGGHYHORSPHQWV RIRXUGHÀQLWLRQVZKLFKXSGDWHDQGUHÁHFWFXUUHQWDUFKDHRORJLFDOUHVHDUFK are a must if archaeology will develop. An important problem area encountered with the discussion of dynamic WHFKQRORJLFDOGHÀQLWLRQVLVWKHFRQWLQXHGXVHRIWKHWUDGLWLRQDODUFKDHRORJL cal term “culture concept”. The archaeological culture concept was put for ZDUGLQFORVHUHODWLRQWRVWDWLFPRUSKRORJLFDOGHÀQLWLRQVRIDUWHIDFWW\SHV and the associated theory centered on the idea that artefact types found con FHQWUDWHGLQVSHFLÀFUHJLRQVRUDUHDVFRXOGEHVHHQDVH[SUHVVLRQVRIKXPDQ cultures. The focus on the archaeological “culture” has ever since been both politically and archaeologically problematic. Archaeological cultures, deter mined on the basis of simple stylistic differences, have often been implicitly 294 Rethinking the lithic blade deﬁnition: towards a dynamic understanding understood as cultures/societies in an anthropological meaning, which has led to severe misunderstandings (e.g. Kossinna 1936). In changing the focus WRZDUGVSURFHVVHVDQGHPSOR\LQJG\QDPLFWHFKQRORJLFDOGHÀQLWLRQVDUFKDH RORJLVWVZLOOEHDEOHWRGHÀQHDUWHIDFWVLQUHODWLRQWRWKHSURFHVVIURPSUR FXUHPHQWWRGLVFDUGLQJWKHÀQLVKHGDUWHIDFW$QGDVERWKWHFKQRORJLFDODQG VRFLDOLGHDVDUHH[SUHVVHGLQWKHSURFHVV HJ0DXVV/HURL*RXUKDQ 1964; Lemonnier 1990) this methodology reveals and traces technological and social traditions, as well as individual expressions, whereas traditional morphological analysis primarily reveals stylistic changes. By implement ing a dynamic understanding of lithic technology and combining this ap proach with analysis of other cultural aspects, such as subsistence strategies, architecture and art, archaeologists are enabled to overcome (problematic) perceptions of lithic artefacts as undisputed, categorical trademarks of “ar chaeological cultures”. The advantage of this change can introduce light and shade into our perception of prehistory. Conclusions and perspectives 7KH QHZ G\QDPLF WHFKQRORJLFDO GHÀQLWLRQ LV ZKHQ XVHG LQ PDQ\ VWXGLHV DGYDQWDJHRXVWRWKHHDUOLHUGHÀQLWLRQVEHFDXVHLWVIRXQGDWLRQLVEDVHGXSRQ prehistoric human intentions as manifested in lithic sequences. The use of WKHQHZGHÀQLWLRQZLOOWKHUHIRUHDOVRIDFLOLWDWHLQWHUSUHWDWLRQVRIKXPDQDFWV and social aspects of the prehistoric life. ,Q D ZLGHU SHUVSHFWLYH WKLV DXWKRU ÀQGV WKDW RXU DUWHIDFW GHÀQLWLRQV should relate to human behaviour and cognition rather than to chosen met rical categories. The crucial focus of future research in lithic assemblages must therefore be to understand and explain `reasons and purposes´ in hu PDQDFWVUDWKHUWKDQWRXVHVXSSRVHGO\´REMHFWLYHµGHÀQLWLRQVWRJHWKHUZLWK numbers and statistics to create more archaeological cultures. One step to wards this goal is to expand our knowledge about prehistoric technology and LQFRUSRUDWHLQWHQWLRQVLQWRRXUDUWHIDFWGHÀQLWLRQV $IXWXUHFRQVHTXHQFHRIWKHXVHRIG\QDPLFWHFKQRORJLFDOGHÀQLWLRQV DV the one put forward in this paper) is that they can enable us to dissolve the problematic focus on “culture concepts” and favour human behaviour and cognition. Perhaps the most important advantage of changing our artefact GHÀQLWLRQVWRZDUGVWKHG\QDPLFDOWHFKQRORJLFDODSSURDFKLVWKDWSUHKLVWRU\ FDQEHUHGHÀQHGDQG´UHHVWDEOLVKHGµLQWHUPVRIVRFLDOWUDGLWLRQVRUVRFLH ties, based on the concepts, methods and behaviour of the people who made them. 295 Mikkel Sørensen Acknowledgements Thanks for discussions and proofreading to: Farina Sternke, PhD student (Centre for the Archaeology of Human Origins, University of Southamp ton), Sheila Coulson, archaeologist and professor (“Institut for Arkeologi” Oslo University), Anders Högberg, archaeologist (Malmø Heritage) and Bjarne Grønnow, archaeologist and leader of Sila (The Greenland Research Center at the National Museum of Denmark). 296 Chapter 3 From Experience to Interpretation Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Nyree Finlay Manifesting microliths: insights and strategies from experimental replication Abstract Experimental replication has been an underdeveloped tool for exploring routines of microlith production. This paper presents the results of two rep lication experiences and explores the constraints of traditional approaches, UHÁHFWLQJRQWKHLQIHUHQFHVIRUXQGHUVWDQGLQJPLFUROLWKPDQXIDFWXUHERWK in replication and in relation to later mesolithic archaeological assemblages from sites excavated as part of the Southern Hebrides Mesolithic Project, western Scotland. Concealment and visibility emerge as key metaphors for engaging with the various routines of microlith production and the perfor mativity of manufacture is explored as a means to consider both group and individual actions and identities. Introduction The aim of this paper is to consider insights derived from experimental rep lication and the dialogue that exists between modern engagements with mi crolith manufacture and the information that can be teased from detailed analysis of archaeological assemblages. Replication as a contemporary prac WLFH KDV WHQGHG WR IRFXV RQ WKH SURGXFWLRQ RI ¶H[FHSWLRQDO· LWHPV OLNH WKH Danish daggers (as evidenced by Apel 2001; Apel and Nunn this volume) rather than the apparently more simplistic objects that are often the main stay of the archaeological record. This has been clearly detrimental to our appreciation of the range of techniques and other subtle differences that can RFFXULQPDQXIDFWXUH,QWKLVSDSHU,ZLVKWRUHÁHFWRQWKHFRQVWUDLQWVRI traditional representation and present observations derived from the results of two short programmes of lithic replication. In terms of structure, the paper is divided into four sections where my point of departure moves from conventional ways of representing these artefacts to the presentation of ex SHULPHQWDODQGDUFKDHRORJLFDOSHUVSHFWLYHVWRÀQDOO\UHÁHFWRQSURGXFWLRQ as an exercise in revealing and concealing self and group identities. 299 Nyree Finlay Refocusing the microlithic gaze Traditions of representation and presentation of microliths in both text and in illustration tend to promote a dominant view of the microlith that is at odds with the mode of production. This creates what I have termed the ¶PLFUROLWKLFJD]H·RIFRQYHQWLRQDODQDO\VLV )LQOD\a; 2003). While the dorsal surface orientation is promoted to convey technical details such as ar ris scars and directionality of removals, it is highly unusual to see the ventral face depicted unless it is to show invasive retouch or edge damage. Yet, the simple fact remains that the conventional surface orientation of microliths as presented in archaeological illustration and discussion is not that experi HQFHG GXULQJ PDQXIDFWXUH 7KH PDMRULW\ RI PLFUROLWKV DUH PRGLÀHG IURP the ventral surface and this is the plane view of the piece when retouched. Therefore, there is a reverse symmetry of the experience of microlith manu facture versus the idealised view of the microlith as presented in convention al archaeological discourse. Equally, the vertical arrangement of microliths in lithic illustration is a feature of page layout that is more concerned with economy than a vehicle to challenge representation. Yet, if we consider the KDIWLQJRIWKHÀQLVKHGSLHFHWDNLQJWKHPRVWGRPLQDQWVWHUHRW\SHGUHFRQ struction of these pieces as the points and barbs of an arrowhead, then the directionality of movement and alignment is horizontal not vertical. The modes of presentation and the microlithic gaze also promote a particular plane symmetry that conditions form recognition (Rock 1973; Washburn 7KLVKDVDQLPSDFWRQW\SRORJLFDOFODVVLÀFDWLRQDQGFRQGLWLRQVWKH analysis of what are essentially the same forms (Finlay 2003). In addition the WZRGLPHQVLRQDOZRUOGYLHZWKLVFUHDWHVFRQÁLFWVZLWKWKHWKUHHGLPHQVLRQ DOUHDOLW\RIWKHÀQLVKHGSLHFH6LJQLÀFDQWO\WKHHIIHFWVRIWKLVPLFUROLWKLF gaze are brought into sharp focus during microlith manufacture. Modern microlithic replication Microliths have not been subject to the same traditions of experimental rep lication as other classes of lithic objects. In contrast to the wealth of litera ture available on the manufacture of bifacially retouched artefacts, such as projectile points (e.g. Callahan 1979; Young and Bonnischen 1984), there are few guides and sources that consider microlith manufacture. There are a few notable exceptions (for example Tixier 1963; Inizan et al. 1999) but WKHFRYHUDJHLVJHQHUDOO\UDWKHUVXSHUÀFLDODQGWKHUHKDVQRWEHHQWKHVDPH GHSWKRIHQJDJHPHQWDVLVWKHFDVHZLWKRWKHUPRUH¶FKDOOHQJLQJ·LWHPVVXFK 300 Manifesting microliths: insights and strategies from experimantal replication as bifaces. Microliths are perceived as quick and easy to manufacture, not requiring a high degree of lithic skill. There are no kudos attached to their contemporary reproduction; advertisements within the popular replication literature, such as Chips for exquisite replicas of these artefacts are rare if not QRQH[LVWHQW$VVXFKPLFUROLWKVDUHRYHUORRNHGQRWRQO\ZLWKLQWKHUHSHU toire of modern knappers, but detailed academic discussion of technological features is generally sparse. One aspect of microlith manufacture that tends WREHGLVFXVVHGLVWKHVLJQLÀFDQFHRIWKHPLFUREXULQWHFKQLTXHSDUWLFXODUO\ LQUHODWLRQWRWKH/HYDQWLQHPDWHULDO VHH*RULQJ0RUULV+HQU\ Kaufman 1995). The microburin has acquired almost monumental status in discussions of microlithic technology at the expense of other aspects of mi crolith manufacture such as the execution of retouch and sequences of modi ÀFDWLRQQRWWRPHQWLRQGLVFXVVLRQRIUHODWHGLVVXHVVXFKDVEODQNVHOHFWLRQ and breakage rates during manufacture. Yet while the microburin has been UHDGDVDFXOWXUDOVLJQLÀHUDQGDIIRUGHGDGHJUHHRIVLJQLÀFDQFHLWKDVUDUHO\ been the subject of detailed experimental replication particularly in recent decades (early exceptions include Vignard 1934; Barnes 1947). Equally, while the theoretical possibilities of explaining microlith variability as the product of remodelling was the topic of extended debate (Neeley & Barton 1994; )HOOQHU*RULQJ0RUULVet al. 1996), it was not informed by replication. Experiencing replication The general neglect of microliths in the broader replication literature pre VHQWHG GLIÀFXOWLHV ZKHQ GHVLJQLQJ D VHULHV RI UHSOLFDWLYH H[SHULPHQWV WR explore the construction of variability and its relationship to the manufac turing process. The lack of precedents for this type of research resulted in trial and error being one of the dominant forces, and it is evident that methodological strategies need to be further developed to explore issues of microlith manufacture effectively. However, the action of making micro liths is in itself instructive and informative, raising issues that will not de rive from the examination of archaeological material alone. Several sessions RIH[SHULPHQWDOUHSOLFDWLRQZHUHXQGHUWDNHQWRFRQWULEXWHWRGHÀQLWLRQVRI microlith production and explore the various aspects of manufacture and choices available with regard to technique. The impetus for this was the DQDO\VLVRIDQXPEHURIDUFKDHRORJLFDODVVHPEODJHVDQGUHÁHFWLRQVRQWKH inferences from replication are considered below. One of the dominant fac tors in much modern and informal lithic replication is the lack of familiarity with techniques and the limiting constraints of skill and ability. Knapping is not, after all, a common practice in contemporary society and this raises a 301 Nyree Finlay VXLWHRILVVXHVUHJDUGLQJWKHHIÀFDF\RIUHSOLFDWLRQDVDPRGHRIXQGHUVWDQG ing and the paradox of familiarity and dislocation that it often brings to our engagement with stonecraft. Microlith manufacture can be considered as an act with three or more parts: the manufacture of suitable blanks; the transformation of blank to PLFUROLWKDQGÀQDOO\WKHPDQLSXODWLRQDQGVHOHFWLRQRISLHFHVIRUXVHOHDGLQJ WRWKHLULQWHJUDWLRQLQWRDKDIW,WLVWKHÀUVWWZRHOHPHQWVWKDWDUHRIGLUHFW concern here. The production of suitably sized blanks is evidently the more skilful aspect of microlith production and the extent of standardisation at this stage no doubt impacts on the amount of retouch required to alter the SLHFHLQWRWKHGHVLUHGIRUP([SORULQJWKLVÀUVWSDUWRIWKHPLFUROLWKchaîne opératoire is constrained by the paucity of experienced knappers. As well as providing analogies for the past, modern experimental replication is also a form of contemporary social praxis. It is the latter aspect that is often RYHUORRNHG DQG ELDVHG YLD WKH DWWHQWLRQ JLYHQ WR WKH SURÀFLHQW LQGLYLGXDO at the expense of the novice (Finlay forthcoming). Moreover, the collective and communal aspects of tool production as a shared experience are often negated. The results of an informal session of replication reveal the variety of modes of retouch and enable us to consider the more performative aspects of production. As part of a group replication programme to consider con sistency in blade production (Finlay forthcoming), three individuals spent a couple of hours making microliths from the blanks they manufactured in an HDUOLHUVWDJH$WRWDORIEODQNVZHUHPRGLÀHG3ULRUWRWKHVHVVLRQWZR of the participants had made microliths on an informal basis and both had considerable experience of analysing Scottish Mesolithic material over sev eral decades, the third had no previous experience of microlith manufacture but had previous sporadic knapping experience over two years. A random selection of microliths from an archaeological assemblage was shown to the knappers as rough templates for size and shape. Given that this was an initial exploration into microlith manufacture, no explicit in structions about desired forms was given. Each person made microliths from their own blanks and the tendency was to work through all the suitable blanks from one core rather than select out the most suitable blanks from the entire sample. No instruction was given as to how the microliths should be manufactured and the use of the microburin technique was left to the discretion of the individual. The debitage from each microlith was collected and separately bagged. All three were seated at the same table, yet each had individual preferences in the techniques of manufacture (Fig. 1). Knapper 1 used the combination of an antler point and hammerstone fragment with a pointed edge, with the former the preferred tool. The anvil stone was sup 302 Manifesting microliths: insights and strategies from experimantal replication ported in the palm of the hand under a piece of leather and the antler used as percussion tool to modify the blanks. This method enabled a very me thodical and precise method of backing to be used. In contrast, knapper 2 preferred to use a small quartz hammerstone; the blank was placed directly RQWKHWRSRIDÁDWEHDFKSHEEOHDQYLOUHVWLQJRQWKHWDEOHRQWRSRIDOHDWKHU support, to collect the debris. An antler point was also occasionally used. Finally, knapper 3 used a very angular anvil stone and utilised the sharp edge of the anvil, the blank was held against the edge using the overhang to modify the blanks by the use of a pebble percussor. Figure 1. Group replication study: preferred techniques (knappers 1-3, left-right). In practice, the nature of the blank emerged as an important factor as it placed constraints on what could be produced and the time taken to mod ify pieces. Admittedly the sample of blanks available to the knappers were clearly constrained by their (in)ability to consistently produce small blanks WKDWZRXOGUHTXLUHWKHOHDVWDPRXQWRIPRGLÀFDWLRQ7KHEODQNVVHOHFWHGDV suitable for microlith manufacture were characterised by inner regular pieces WKDWZHUHEODGHVRUEODGHOLNHÁDNHV7KHFULWHULDXVHGWRVHSDUDWHRXWWKHVH pieces from the spread of debitage were the regularity and thinness of the blank and a degree of parallelism with feathered edges. The size and mor phology of the blank emerged as an important factor. Much of the time was taken up by the trimming of the blank, the thicker the piece the more effort required to trim the piece and this was particularly the case for knapper 1 and is represented schematically in Fig. 2. While length and breadth can be PRGLÀHGE\UHWRXFKWKHWKLFNQHVVRIWKHEODQNFDQQRWEHDOWHUHGZLWKRXW recourse to invasive retouch, unless the thickness of the piece is variable. Blank thickness was the stated reason for discard for one piece by knapper 1, where two attempts were made at manufacture, and three of the microliths made by knapper 2. 303 Nyree Finlay Figure 2. Group replication study: selection of microliths produced by knappers 1 and 2 (numbers refer to order produced, greytone depicts size of original blank). Figure 3. Group replication study: Knapper 3 ‘imitation’ microliths. The form of the blanks as well as the technique adopted placed constraints on the ability of knapper 3 to fashion microliths. These pieces could easily be distinguished by pronounced enclume retouch and the bidirectionality and irregularity of the retouch types (Fig. 3). It is questionable whether the con straints of the blank form would have been overcome by more conventional methods of retouching as none of these attempts even resembled microliths. In this case, inexperience and lack of familiarity clearly limited perform ance as much as the blanks worked and the method used. In this respect, this technique has produced pieces that generally approximate the shape of PLFUROLWKVEXWLQQRZD\UHÁHFWWKHFRUUHFWXVHRIWHFKQLTXHV3DUDOOHOVFDQ EHGLUHFWO\GUDZQZLWKH[DPSOHVRIFKLOGUHQ·VSOD\DQGLPLWDWLRQDVLGHQWLÀHG in the archaeological record (Högberg 1999; Högberg forthcoming). Here the novice knapper has produced pieces that imitate the basic form but used inappropriate methods of retouch to create them. 304 Manifesting microliths: insights and strategies from experimantal replication Slight differences in the method of manufacture could be discerned on the end products. However, it was not possible to differentiate between knap SHUV DQG IRU HVVHQWLDOO\ ERWK ZHUH XVLQJ WKH VDPH PRGH GLUHFW SUHV sure supported by an anvil, even though there were slight differences in their WHFKQLTXHDQGWKHÀQLVKHGSLHFHV7KHUHWRXFKRQWKHPLFUROLWKVSURGXFHG by knapper 1 is very regular and even, whereas that for knapper 2 is slightly PRUHYDULDEOH7KLVPD\UHÁHFWWKHSUHGRPLQDWHXVHRIDSHEEOHLQFRQWUDVWWR WKHDQWOHUSRLQWIDYRXUHGE\NQDSSHU+RZHYHULWPD\DOVREHDUHÁHFWLRQRI the extra time spent by knapper 1 and her more methodical trimming of the blank. Despite the use of a supporting anvil, enclume retouch is only present on one piece by knapper 2, which was made using an antler point. &KRLFHVRIPRGLÀFDWLRQ In order to explore the implications of subtle differences in technique and develop methodologies for recording variation, a separate programme of experimental replication was conducted by the author. Discussion will be based on the experiences of producing around 100 microliths as a discrete replication programme. The informal group replication outlined above re vealed that subtle differences were employed by these knappers who were all engaged in the same task of producing microliths. A range of techniques was adopted: the use of both stone and antler retouchers with the blank resting on an anvil stone. There is another method of executing the retouch and WKDWLVE\VFUDSLQJWKHEODQNDORQJWKHHGJHRIDÁLQWSHEEOHRUHGJHRIFRUH or chunk to modify the blank. This produces retouch that is akin to that produced by using a stone on an anvil. By supporting the blank in the hand it results in less breakage due to the fact that contact with the stone was more even than when resting on an anvil. Three techniques were used to make mi croliths, 67% were made using a stone to execute the retouch, 22% with an DQWOHUWLQHDQGE\VFUDSLQJWKHEODQNDORQJDÁLQWSHEEOH8VLQJDVWRQH percussor is the quickest and easiest method, the natural bevelled end of the VWRQHWHQGHGWRSURGXFHDPRUHFRQFDYHSURÀOHWRWKHUHWRXFKHGJHLQSODQ 7KLVPHWKRGDOVRJHQHUDWHGDODUJHDPRXQWRIÀQHGHELWDJHWKDWKDGWREH removed from the surface of the anvil to prevent breakage. Enclume retouch was only present on one in ten blanks. The use of an antler tine produced retouch that could not be macroscopically distinguished from that produced by stone. The size of the point enables more precision but it was much slower to execute. Finally, simply scraping the blank along a stone produced quite ÀQH UHWRXFK ZLWK D VPRRWK SURÀOH 7KLV PHWKRG SODFHV PRVW SUHVVXUH RQ the hand, as the piece had to be kept level. Notching was not possible un 305 Nyree Finlay less the stone had an angled edge, such as the use of a core edge. Where this was present, it could be used to create a notch and was effective at realising a microburin. The debitage produced by scraping the piece against another edge tended to be smaller in size fraction and this was more easily dispersed than with the other methods. Quantifying differences resulting from the various techniques was dif ÀFXOWWRLGHQWLI\RQWKHFRPSOHWHGSLHFHVJLYHQWKHYDULDWLRQLQWKHVL]HDQG form of the original blanks. The blanks used were not uniform in terms of size and thickness as well as the location of arris scars. Moreover, many were incomplete having lost their platforms during manufacture. Given that many of the blanks were already missing platforms, it was not necessary to use the microburin technique in every instance. It was also often easier to trim the proximal end with retouch rather than remove the bulb. While the FRQVWUDLQWVRIEODQNWKLFNQHVVRQWKHDELOLW\WRSURGXFHDÀQLVKHGPLFUROLWK can be anticipated, breakage resulting from the manufacturing process itself cannot always be predicted. A selection of the microliths produced are il lustrated in Fig. 4, these have the original form of the blank outlined in grey WRQH$VFDQEHVHHQIURPWKLVÀJXUHWKHH[WHQWRIWKHEODQNORVWGXULQJWKH manufacturing process is variable. The average reduction in length was 9.7 mm (8.8 mm Stdev) and 3.6 mm (3.17 mm Stdev) in width. Manufacture WLPHYDULHGIURPVHFRQGVWRPLQXWHVDQGZDVLQÁXHQFHGE\WHFKQLTXH and blank form. In these various experiences of producing microliths, there were clear limitations in terms of the character and type of form. The blanks used were not standardised in terms of their overall dimensions and techno logical attributes. This was one of the main limiting features, given the large degree of variability present in the blanks used. Yet this also encapsulates many of the problems with modern experimental replication per se, namely WKDW H[DPLQLQJ SURÀFLHQF\ LQ RQH HOHPHQW RI D JLYHQ FKDvQH RSpUDWRLUH LV FRQGLWLRQHGE\SURÀFLHQF\RUDWOHDVWFRQVLVWHQF\LQRWKHUDVSHFWV7KHDF commodation of variation in blank form is one of the key features in micro lith manufacture and this would also have been a factor in the past. Even if it is not so extreme an issue as seen in the replication experiences described here. It is clear that we need to see more detailed accounts of replicating microliths. The methodology of recording and illustrating the blanks prior to reduction at least enables the constraints of blank form to be represented, HYHQLIWKHLQÁXHQFHVDUHPRUHVXEMHFWLYHLQLQWHUSUHWDWLRQ Approximately 20% of attempts at manufacture ended in failure, with the irretrievable breakage of the blank. Breakage did not appear to reduce through time, as it was often a question of the original blank form. Many of these fragments do not even retain evidence of retouch, as they broke with WKHÀUVWDSSOLFDWLRQRISUHVVXUH%ODQNVZHUHEDFNHGXQWLODUULVVFDUVOLPLWHG 306 Manifesting microliths: insights and strategies from experimantal replication Figure 4. Individual replication: selection of microliths showing original blank size and extent of reduction required. further removal or the shape approximated the required form. The location of the dorsal arris scar affected the ease by which a blank could be further trimmed. This can be seen in some of the illustrations of the pieces made, E\FRPSDULQJWKHEODQNVZLWKWKHÀQDOSURGXFW6HYHUDORIWKHPLFUROLWKV have arris scars close to the retouched edge. The fact that the blanks used to create microliths were quite variable limits the inferences that can be made regarding the consistency in realising form and potential expressions of in dividuality. The issue of technique is an important one and has not been subject to much discussion within the wider archaeological literature on the Mesolithic. There are a number of options when faced with the manufacture RIDPLFUROLWK7KHVHDUHLQÁXHQFHGE\WKHQDWXUHRIWKHEODQNVHOHFWHGDQG how the retouch is to be executed. However, it is evident that the techno logical choices open to the mesolithic knapper would also be conditioned by experience and the routinised modes of production. Archaeological perspectives The experimental studies outlined above were conducted in conjunction with the examination of a suite of archaeological assemblages dating from FDO%&H[FDYDWHGXQGHUWKHDHJLVRIWKH6RXWKHUQ+HEULGHV0HVR lithic Project (SHMP) from the islands of Islay and Colonsay, off the Scottish west coast (Fig. 5; Mithen 2000). Several of these sites are palimpsests where the accumulated actions of several generations of mesolithic knappers are represented and where continuity in these basic forms and routines in pro duction are seemingly maintained over several millennia. Here the repeti tion of tasks would have connected present with past and future action. The archaeological assemblages created a number of challenges when exploring 307 Nyree Finlay PLFUROLWKSURGXFWLRQDQGWKHWUDQVIRUPDWLRQRIEODQNVWRÀQLVKHGPLFUROLWKV Detailed attribute analysis was conducted on 2600 microliths and fragments and around 440 related pieces such as microburins and truncations (Finlay 2000b). This was undertaken in conduction with detailed core and debitage analysis with the goal of reconstructing the microlith chaîne opératoire. Exami nation of the archaeological material illustrates some of the problems in try LQJWRGHÀQHDQGGHWHUPLQHEODQNSUHIHUHQFHVDQGWHFKQRORJLFDOFKRLFHVLQ microlith production. At these sites, two basic microlith forms predominate: backed bladelets and scalene triangles supplemented by a suite of other types as can be seen for two sites Bolsay Farm and Staosnaig (Fig. 5). Many retain bulbs of percussion. Determining original blank dimensions was problem DWLFJLYHQWKHH[WHQWRIPRGLÀFDWLRQDQGWKHIDFWWKDWRULJLQDOEODQNOHQJWK could only be ascertained for eight microliths out of 1530, although it could be gauged by combining microlith and microburin lengths. No apparent patterning was noted with respect to microlith form and the frequency of arris scars and types of distal terminations. Comparison of the size dimensions between scalene triangles and backed blades is also not VWDWLVWLFDOO\VLJQLÀFDQWWRVXJJHVWWKHVHOHFWLRQRIGLIIHUHQWEODQNVIRUFHUWDLQ forms. However, microwear analysis has shown that the more angular the scalene form in plan on the ventral surface, the more probable the presence of wear traces (Finlayson & Mithen 1997). This suggests that the primacy of the ventral surface during backing may also be important in terms of LQÁXHQFLQJVHOHFWLRQFULWHULD,WDOVRGHPRQVWUDWHVWKHWHQDFLW\RIWKHPLFUR lithic gaze and the failure to acknowledge the three dimensional properties of these objects. Several factors remain unknown at present; one of these is the direction of orientation of the blank during backing. While the surface orientation in terms of the execution of retouch is overwhelmingly ventral, what cannot be ascertained is the position the piece was held on an anvil QRULQGHHGZKHWKHUDPRUHÁXLGPRGHZDVDGRSWHGZLWKWKHSLHFHUXEEHG against another surface to facilitate the retouch. It is commonly assumed within the literature that truncating the blank with the use of the microburin technique is facilitated by using an overhang (e.g. Inizan et al. 1999: Fig. 33). However, this was not my experience of the technique during manufacture; rather it was achieved by notching the blank RQWKHÁDWVXUIDFHRIWKHDQYLODQGDSSO\LQJSUHVVXUHQRWDWWKHHGJH7KH ease by which the microburin technique was successful appeared to relate to the thickness of the blank and the angle at which pressure was applied. Several of the most successful microburins were accidentally produced while UHWRXFKLQJRUQRWFKLQJWKHEODQNDVLWUHVWHGÁDWRQWKHDQYLO$WWHPSWVWR remove some of the thicker bulbs had to be abandoned as these could not be easily notched and were too thick to apply the requisite pressure. Another 308 Manifesting microliths: insights and strategies from experimantal replication Figure 5. Location map and types of microlith present at archaeological case study sites. 309 Nyree Finlay observation was that many of the successful attempts at removing the bulb of percussion by this method did not result in a characteristic microburin product. Rather, fragments with some or no retouch present were created. There are other ways of realising a microburin such as creating the notch on a prominent edge and applying slight pressure. One of the most noteworthy aspects was how the use of the microburin technique did not always result in a discernible waste product, and that it frequently produced debitage that would be unrecognisable as such. Therefore, we can consider that the vis ibility of this method should be related to factors such as blank form, thick ness and more than likely to the subtle variations in executing the technique. Within the SHMP assemblages, there are examples where the constraints of blank thickness are clearly a determining factor in the preservation of distal microburins. In light of this, it is perhaps no surprise that despite the prevalence of distally retouched scalene triangle forms in the archaeological assemblages, most microburins are proximal and the ratio is 1 microburin: 4 complete microliths. In the SHMP assemblages there does seem to be some support for the ba sic routines of production as produced via replication; however, the micro OLWKLVDUFKDHRORJLFDOO\YLVLEOHDQGLGHQWLÀDEOHGXHWRWKHSUHVHQFHRIUHWRXFK The question remains as to whether suitably sized debitage and fragments KDGHTXLYDOHQWELRJUDSKLHV<HWZLWKLQWKHSURFHVVHVRIPRGLÀFDWLRQZHFDQ see consistency in practice even if this resulted in variability in the end prod uct (over 53% of the microliths analysed are unique combinations of the at WULEXWHVUHFRUGHG 7KHEDVLFDFWLRQVDQGURXWLQHVRIVLGHDQGHQGPRGLÀFD tion are present. The dominant mode for scalene triangles is the longitudinal backing of the left lateral side and distal end, but there are single examples from three sites that produce a scalene triangle morphology that is fashioned RQWUDQVYHUVHÁDNHVHJPHQWV(TXDOO\WKHUHDUHRQO\WZRSLHFHVZKHUHWKH internal geometry of the piece contradicts the basic scalene triangle form. Moreover, the evidence from microburins and lamelles à crans indicates that WKHORQJLWXGLQDODOWHUDWLRQZDVXQGHUWDNHQEHIRUHPRGLÀFDWLRQRIWKHHQGRI the piece. In terms of retouch, direct retouch from the ventral surface pre dominates and the frequency of enclume retouch is variable. Overall, it could RQO\EHLGHQWLÀHGRQRISLHFHVDQGDWRQHVLWH6WDRVQDLJ&RORQVD\LWLV present on 17% of the backed blades. While this could be used to support GLIIHUHQWSURGXFWLRQURXWLQHVLWLVDOVROLNHO\WRUHÁHFWRWKHUIDFWRUVVXFKDV blank character. Overall, there is little evidence to support the remodelling of microliths from one form to another, both in terms of the size ranges DQGWKHSUHVHQFHRILQYHUVHUHWRXFK7KXVWKHSDWWHUQLGHQWLÀHGZLWKLQWKLV dataset reveals variability in microlith form, but consistency in practice as far it can be discerned in terms of patterns of surface orientation and routines 310 Manifesting microliths: insights and strategies from experimantal replication of execution. Identifying the mode of retouch was more problematic and was limited solely to macroscopic examination. Here the constraints of the EODQNFUHDWHGGLIÀFXOWLHVLQDVVLJQLQJWKHSLHFHWRDSDUWLFXODUPRGHDQGWKH lessons from replication suggest that equating retouch with mode (given the XQGHUUHSUHVHQWDWLRQ RI enclume retouch) is problematic, it cautions against DVVXPLQJPRGHRIPRGLÀFDWLRQDQGKHUHPLFURZHDUHYLGHQFHDQGPRUHUHS licative studies would be useful. 7KHLPSUHVVLRQJDLQHGIURPWKHDUFKDHRORJLFDOVLWHVLVRQHRIVXSHUÀFLDO conformity in the routines of production. Yet the actions may have been FRQVWDQW LQ UHODWLRQ WR WKH VHTXHQFH RI PRGLÀFDWLRQ ZKLOH WKH PRGH RI EDFNLQJZDVKLJKO\YDULDEOH,QGHHGWKLVÁH[LELOLW\LQUHDOLVLQJWKHVDPHHQG product may have been a salient feature: variation within acceptable social constraints and functional needs. Discussion: the performativity of microlith production In this section I want to consider the production of microliths described above as a form of performance, and the more performative aspects of mi crolith manufacture as an exercise in concealment and (trans)formation. Christina Lindgren (2003) has explored how sources of raw materials are controlled and the location of production secluded and used to create social tensions within a group. Here I wish to consider how we can apply this meta SKRURIFRQFHDOHGDQGKLGGHQDFWLRQDWWKHOHYHORILQGLYLGXDO,QÁXHQFHG E\-XGLWK%XWOHU·VWDNHRQSHUIRUPDWLYLW\LQSDUWLFXODUWKHQRWLRQWKDWLWLV¶D UHLWHUDWLRQRIDQRUPRUVHWRIQRUPV·WRWKHH[WHQWWKDWLWDFTXLUHVDQDFWOLNH status and that ‘it conceals and dissimulates the conventions of which it is a UHSHWLWLRQ· %XWOHU ZHFDQFRQVLGHUWKHUROHRIPLFUROLWKSURGXF tion as the visible expression of social values and thereby consider its role in social reproduction. The idea of technology as performance has been dis cussed in the past (eg. Lechtman 1977; Lemmonier 1993), but it can be recast along with more recent perspectives on the social dimensions of technology WRUHFRQÀJXUHDJLYHQchaîne opératoire (Dobres 2000). The entire process of microlith production and use can be read as a play between visible and con cealed action. At times, the various events that contribute to the creation of a piece are either clearly visible or obscured. The production of blanks marks the most visible stage of production. Given the small size of the blank, the WUDQVLWLRQWRPRGLÀHGSLHFHLVDFFRPSOLVKHGE\VOHLJKWRIKDQG7KHDFWLRQ is often concealed from the maker, the hand obscuring the visibility of the 311 Nyree Finlay DFWLRQ+HUHLWLVZKDWLVNQRZQIHOWEXWSHUKDSVQRWVHHQWKDWLVVLJQLÀFDQW as the piece is altered via retouch. Thus, the simple action of backing can be seen as an extension of the self as the microlith is created with some of the agency of the maker transferred to the piece. To the onlooker, this PRGLÀFDWLRQWKLV WUDQV IRUPDWLRQLVDFFRPSOLVKHGRIWHQZLWKOLWWOHHIIRUW and frequently with some speed, yet it is not entirely visible. Clearly, the vis LELOLW\RIWKLVDFWLRQLVFRQGLWLRQHGE\WKHPRGHRIPRGLÀFDWLRQZKHWKHUDQ anvil stone is used and the manner the piece is held. It is more often than not felt rather than witnessed by the maker. Yet if the replication modes are analogous then it is questionable whether even the orientation of the piece in the hand could always be discerned by others without close scrutiny. In this sense, subtle variations serve to challenge and subvert the conventions of production. The metaphor of concealment whereby the piece is hidden in the hand continues into the hafted form. The use of mastic and the form of the KDIWPHDQVWKDWWKHÀQDOIRUPRIWKHSLHFHZRXOGDOVRQRWEHLGHQWLÀDEOHLQ WKHÀQLVKHGSURGXFW2IFRXUVHQRWDOOPLFUROLWKVZRXOGKDYHQHHGHGWREH hafted and the form of the complete tool form might not always conceal the form of the piece. However, the process is one where there is a clear interplay between concealment and visibility where what is known and not seen is im portant. This can be read in a number of ways for it is also a means by which WKHRZQHUVKLSDQGLGHQWLÀFDWLRQRIWKHPDNHUWKHLUDFWLRQDQGWKHDUWHIDFW are made present or absent. Individual action is revealed and equally hidden or suppressed at different stages. The wider visibility of the skills and tech QLTXHVXVHGLQEODGHSURGXFWLRQWKURXJKWRWKHPRUHFRQFHDOHGPRGLÀFDWLRQ of backing is one such shift. Another is the relative visibility of the microlith DWWKHSRLQWRIVHOHFWLRQDQGWKHFRQFHDOPHQWLQWKHKDIWLQJRIWKHPRGLÀHG piece. In this regard, microlith manufacture can be regarded as embodying HOHPHQWVRID¶WHFKQRORJ\RIHQFKDQWPHQW·LQWKHVHQVHGHYHORSHGE\*HOO (1998; 1999) in relation to the anthropology of art. The process of produc tion is captivating not only in terms of the performance of production but also in transforming some of the agency of the maker to the piece. Exploring how some of these metaphors of visibility and concealment are expressed in microlith production allows us to consider how these stages PD\ KDYH EHHQ LPEXHG ZLWK SDUWLFXODU VLJQLÀFDQFH QRW ZLWKVWDQGLQJ WKH subsequent impact of biography to create other meanings (Finlay 2000c). 'LIIHUHQW WHFKQLTXHV RI PRGLÀFDWLRQ LQWURGXFH VXEWOH GLIIHUHQFHV LQ WKH form of engagement with materiality. The challenge lies in developing a body of replication and modern experiences and engagements that can enable us to explore and identify such processes. Here our engagements like those GLVFXVVHGDERYHVWDQGDORQHDVZHPXGGOHWKURXJKWKHPRWRUDFWLRQVDQG constellation of knowledge required to complete such as task. In the past, 312 Manifesting microliths: insights and strategies from experimantal replication WKHFKRLFHRIWRROXVHGWREDFNWKHSLHFHPD\KDYHKHOGVLJQLÀFDQFHLQWHUPV of individual preferences as well as symbolic meanings. The position of the hands – whether hand held, supported on knees, resting on the ground or another surface – would have conditioned the visibility of various stages and OHDGWRSRWHQWLDOHODERUDWLRQRUÁRXULVKDVZRXOGWKHVSDWLDOORFDWLRQDQGUHO DWLYHERG\SRVLWLRQRIWKHPDNHULQUHODWLRQWRRWKHUVRUVLJQLÀFDQWIHDWXUHV We envisage this process of production to occur as a sequence of repetitive actions as the maker(s) multiplies the number of microliths and then selects RXWWKRVHIRUXVHRUKDIWLQJ7KHVHOHFWLRQDQGOD\LQJRXWRIWKHPRGLÀHGRU XQPRGLÀHGSLHFHVWKDWIRUPHGWKHÀQDOFRPSRVLWHDUWHIDFWZRXOGKDYHFUH ated further opportunities to make explicit the implicit ownership of pieces, to include or exclude others. Elsewhere I have suggested that microlith man ufacture afforded an ideal opportunity for multiple authorship (Finlay 2003). It is the potential communality of production that is a dominant motif. The design of composite implements mitigates against individualism in as much as the multiplicity of component pieces are quintessentially a collective prod uct. Moreover, there is considerable redundancy inherent in the multiplicity of suitable components. The meaning and the stages of transformation of stone into tool can be seen as one expression of the composite character of mesolithic technology. While the techniques used to create these artefacts are relatively simple, skill rests in the consistency of practice that produces VXLWDEOHEODQNV7KHHDVHE\ZKLFKPLFUROLWKVDUHPRGLÀHGDUJXHVIRUWKH VXSSUHVVLRQRIVNLOODQGLQGLYLGXDOLVPWKDWLVH[SUHVVHGLQDÁXLGZD\KHUH techniques can be realised and vary in subtle ways but which often end up conforming to tradition and ultimately concealed in the hafted element. 7RFRQFOXGHUHSOLFDWLRQLVDQXQGHUGHYHORSHGDQGQHJOHFWHGIDFHWRIRXU engagement with microlithic technologies. This can be related to the bi ases within lithic replication as contemporary praxis that privilege ‘elaborate NQDSSLQJ·DWWKHH[SHQVHRIPRUHEDVLFREMHFWV7KHUHLVDOVRWKHSHUFHSWLRQ that experimental replication has little to offer given the apparent simplicity of these artefacts. Yet as Lindgren (this volume) and Rankama et al. (this vol ume) argue in relation to bipolar quartz industries, there is much to learn and FKDOOHQJHLQRXURZQSUHFRQFHSWLRQVDERXWVRFDOOHG¶VLPSOH·WHFKQRORJLHV Microliths may well be easy to make but as objects, their apparent simplicity belies a potentially more complex set of meanings and readings. The ‘micro OLWKLFJD]H·KDVKDGDSURIRXQGLPSDFWRQZD\VRIERWKVHHLQJDQGWKLQNLQJ about these artefacts. Experimental replication has a valuable contribution to make at a number of levels from forcing a different type of engagement with the issues of tool production, through to providing comparative break age rates and the different attributes created by subtle variations in the ex ecution of retouch. The manifestation of microliths was and is an exercise 313 Nyree Finlay in (trans)formation. Thinking more metaphorically about this process allows us to consider the relationship between visible and invisible action and its implications for the reproduction of social knowledge. Acknowledgements I am extremely grateful to my three knappers Caroline, Bill and Alejandro. The replication experiences were conducted as part of my doctoral research at the University of Reading in 1994 and I thank Professor Steven Mithen for his support and the opportunity to work on the SHMP material. Attendance at Uppsala was partially funded by the British Academy Networks Scheme and support from the Department of Archaeology, University of Glasgow is also gratefully acknowledged. Lorraine McEwan produced Figure 5 and I am grateful for her advice and assistance. Finally, thanks are due to Jan Apel, Kjel Knutsson and all the other organisers and participants at the sympo sium for fruitful discussion and insights into skilled production, and lastly to Paul Duffy for drawing attention to the more obvious faults herein. 314 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. 0DUFLQ:üV Some remarks on contacts between Late Mesolithic hunter-gatherer VRFLHWLHVDVUHÁHFWHGLQWKHLUÁLQW technology: a case study from Central Poland Abstract &KRFRODWHÁLQWZDVRQHRIWKHSRSXODUUDZPDWHULDOVXVHGGXULQJWKH6WRQH Age in the Polish Lowlands. Both the range and the quality of its supply changed together with the cultural sequence from the Palaeolithic to the Early Iron Age. This paper deals with the problem of the use of choco ODWHÁLQWDPRQJWKH/DWH0HVROLWKLFKXQWHUJDWKHUHUVRFLHWLHVLQKDELWLQJWKH 9LVWXODEDVLQ7KHHFRQRPLFVWUDWHJLHVDQGÁLQWH[FKDQJHEHWZHHQÁLQWDF TXLULQJ PLQLQJ VLWHV ORFDWHG LQ WKH +RO\&URVV 0RXQWDLQV DQG KXQWLQJ camps found about 250 km to the north, will be discussed. The aspect of WUDQVPLWWLQJ EODGH FRUH SURFHVVLQJ FRQFHSWV LQ WKH VRFDOOHG -DQLVODYLFLDQ culture will also be discussed. A reconstruction of a contact network based on Mesolithic sites containing artifacts of the same Mesolithic culture will EHSUHVHQWHGWKURXJKWKHWUDFLQJRINQRZKRZDQGNQRZOHGJHLQWKHÁLQW ZRUNLQJRQWKHVHVLWHV7KLVZLOODOORZIRUDVSHFLÀFDWLRQRIWKHWHUULWRULDO UDQJHRIWKHVHKXQWHUJDWKHUHUVRFLHWLHVEHORQJLQJWRWKHVDPHWUDGLWLRQLQ the 1st half of the 5th millennium BC. Introduction In this paper I would like to discuss the interpretative possibilities of using ÁLQWWHFKQRORJLFDODQDO\VHVWRVKHGOLJKWRQFRQQHFWLRQVEHWZHHQ0HVROLWKLF societies. My investigation is based on assemblages of the Late Mesolithic -DQLVODYLFH&XOWXUH .R]âRZVNL 7KH´-DQLVODYLFLDQµ VLWHVDUHNQRZQ from the territory of Central and Eastern Poland, Western Belarus and West ern Ukraine. The discussion focuses on assemblages from the Vistula ba sin. One important aspects of this culture is the use of mined “chocolate” ÁLQW 6FKLOG &\UHN 6XOJRVWRZVND 7KLV ZDV RQH RI WKH most popular raw materials used during Polish Lowland prehistory from the Lower Palaeolithic until the Early Iron Age (Domanski & Webb 2000). Its popularity was probably the result of both its high quality and the rich outcrops. From an archaeological point of view it is important that the out 315 Marcin Wąs FURSVKDYHEHHQORFDOL]HGLQWKHQRUWKERUGHURIWKH+RO\&URVV0RXQWDLQV 6FKLOG ,WJLYHVXVWKHSRVVLELOLW\WRVWXG\SUHKLVWRULFFKRFRODWHÁLQW distribution. In the Mesolithic its use was differentiated; in the Early Mesolithic it was used only sporadically, but in the Late Mesolithic, an explosion of popularity took place (Cyrek 1981). In the Late Mesolithic there was also an expansion of the Janislavice Cul ture settlements in the area of the outcrops and their neighbourhood. Find LQYHQWRULHVIURPWKHVHVLWHVFRQWDLQDERXWFKRFRODWHÁLQWEXWLQGLYLG XDODUWLIDFWVRIWKLVNLQGRIÁLQWDUHNQRZQIURPVLWHVDVIDUDVNLORPHWUHV from the area of extraction (Schild et al.1975). Baltic Sea DĉBY 29 100 km 200 km TOMASZÓW II Vistula R. )LJ7KHIUHTXHQF\RIFKRFRODWHÁLQWRQWKHVLWHVRIWKH-DQLVODYLFH&XOWXUHLQWKH Vistula basin (after Cyrek 1995). The discussed sites are indicated. 7KHRFFXUUHQFHRIFKRFRODWHÁLQWLQWKHLQYHQWRULHVRIWKH-DQLVODYLFH&XO ture has been interpreted as the result of contacts between societies belong ing to this cultural unit. The connection was created only on the basis of FKRFRODWHÁLQWSUHVHQFHLQWKHDVVHPEODJHV &\UHN ,WZDVVHHQWKDWRQ the sites near the outcrops all chaîne opératoire stages are represented, whereas on the sites far from outcrops, mainly blades or microliths occur. Therefore, LQWKHHDUO\HLJKWLHVDPRGHORIFKRFRODWHÁLQWGLVWULEXWLRQZDVFUHDWHGWKDW VWDWHGWKDWWKHIUHTXHQF\RIFKRFRODWHÁLQWLQDVVHPEODJHVGHFUHDVHGHYHQO\ ZLWKWKHGLVWDQFHEHWZHHQWKHVLWHDQGWKHRXWFURSVDQGWKDWFKRFRODWHÁLQW is not in evidence on sites more than 200 km from the outcrops (Fig.1). This situation changed in the second half of the eighties after the discovery of 316 Some remarks on contacts between Late Mesolithic hunter-gatherers societies WKH'ęE\VLWH 'RPDľVND 7KLVVLWHLVVLWXDWHGDERXWNPQRUWK RIWKHRXWFURSVDQGFRQVLVWVRIDERXWDUWLIDFWVPDGHRIFKRFRODWHÁLQW (100% of the inventory). Therefore, a reconsideration of our earlier interpre tations is necessary. ,Q WKH IROORZLQJ WH[W WKH WHFKQRORJ\ RI ÁLQW SURFHVVLQJ IURP WKH IRO ORZLQJWZRVLWHVZLOOEHFRPSDUHGWKH'ęE\VLWHZKLFKLVLQWHUSUHWHGDV a typical hunting camp with numerous microliths, microburins and blades 'RPDľVND DQGWKH7RPDV]yZ,,VLWHZKLFKLVFRQQHFWHGZLWKH[ WUDFWLRQDQGSURFXUHPHQWRIFKRFRODWHÁLQWQRGXOHV 6FKLOGHWDO 7KH sites will be used for testing a hypothesis concerning connections between societies that in the Late Mesolithic of Vistula basin used the same raw mate rial for production of the same type of microliths. One site – Tomaszów II ²LVVLWXDWHGLQWKHRXWFURSDUHDDQGWKHRWKHU²'ęE\²NPIURPWKH outcrops (Fig. 1). Materials The general chaîne opératoire sequences in the two assemblages can, on the EDVLVRIUHÀWWLQJVWXGLHVEHVXPPDUL]HGLQWKHIROORZLQJZD\ x 7RPDV]yZ ,,²FRQWDLQV DOO VWDJHV RI H[SORLWDWLRQ IURP SUHFRUH procurement to microlith production. It is important to note that SURFXUHPHQW DQG SULPDU\ UHGXFWLRQ RI SUHFRUHV DQG EODGH FRUHV is well represented in the material. This situation is probably con QHFWHGZLWKWKHSUHVHQFHQHDUE\RIVKDIWVIRUÁLQWH[WUDFWLRQ x 'ęE\²UHFRQVWUXFWLRQRIWKHWHFKQRORJ\ZDVSRVVLEOHRQO\WKDQNV WRWKHUHÀWWLQJPHWKRG :ĊV $OPRVWDOOVWDJHVRISURGXFWLRQ KDGWDNHQSODFHKHUHIURPUHDG\SUHFRUHH[SORLWDWLRQWRPLFUROLWK resharpening. ,QWKHFRQWH[WRIWKHGLVFXVVHGSUREOHPRIFKRFRODWHÁLQWGLVWULEXWLRQLQWKH DUHDRIWKH-DQLVODYLFH&XOWXUHDFRPSDULVRQRISUHFRUHIRUPVIURPERWK sites is interesting. In my opinion, the sites were strategic points of initiation of exploitation and determined further blade production. ,QWKH'ęE\DVVHPEODJHDIRUPRISUHFRUHKDVEHHQUHFRQVWUXFWHGWKURXJK UHÀWWLQJ DVRFDOOHG´SKDQWRPFRUHµGXHWRODFNRIWKHDFWXDOFRUH )LJ It is very valuable because it displays the primary form that was brought to WKHVLWH6LQFHQRÁDNHVIURPWKHSUHFRUHIRUPDWLRQZHUHIRXQGLWVHHPV WKDWWKLVSURFHVVWRRNSODFHHOVHZKHUH7KHUHÀWWLQJVKRZVWKHFRQFHSWLRQ 317 Marcin Wąs RISUHFRUHSUHSDUDWLRQDQGLWVWUDQVIRUPDWLRQLQWRWKHFRUHIRUEODGHV7KH PLVVLQJFRUHKDGDQDUURZÁDNLQJVXUIDFHDORQJRIWHQUHMXYHQDWHGVWULNLQJ SODWIRUPDQDUURZWLSDQGQDWXUDOÁDWVLGHV)RUWKHSUREOHPGLVFXVVHGKHUH WKHDQDO\VLVRIWKHSUHFRUHLVRIWKHXWPRVWLPSRUWDQFH 0 3cm „PHANTOM CORE” „PHANTOM CORE” )LJ$UHÀWWHGSKDQWRPEODGHFRUHIURP'ĆE\VLWH 7KHSUHFRUHLVDÁDWQRGXOHZLWKFRUWH[VLGHV7KHGLPHQVLRQV DUHDERXW [ [ FP 7KH SUHÁDNLQJ VXUIDFH KDG D QDWXUDO FUHVW 7KXV QDWXUDO W\SHV RI ÁDW QRGXOHV ZHUH WUDQVIRUPHG LQWR SUHFRUHV ,Q WKLV FDVH LW ZDV done only by reduction of two shorter opposite narrow surfaces. Each of these was formed by strokes from the same direction, from one of the sides of the nodule, changing the orientation (180°). In this way a geometrical, UKRPERLGDOIRUPZDVREWDLQHG )LJ ,WLVVSHFLÀFEHFDXVHLWLVJHRPHWULFDO viewed from two directions: en face and from the side. The investigation of UHGXFWLRQVHTXHQFHVWKURXJKUHÀWWLQJGHPRQVWUDWHVWKHZD\LWZDVGRQH,W seems that geometrical relations between particular surfaces of this kind of SUHFRUHDOORZH[SORLWDWLRQWREHJLQDWRQHRIWZRSRVVLEOHSRLQWV,WGLGQRW LPSRVHRQO\RQHRULHQWDWLRQEXWWKHFKRLFHRILQLWLDOÁDNLQJZDVPXOWLSOH VWLOOUHWDLQLQJWKHJHRPHWU\RIWKHSUHFRUH0RUHRYHUHDV\UHMXYHQDWLRQRI 318 Some remarks on contacts between Late Mesolithic hunter-gatherers societies the striking platform or the preparation of the tip of the core was possible at every moment. I think this is a very universal and practical form. P1 Fig. 3. Scheme of the rhomboidal pre-core preparation based on WKH´SKDQWRPFRUHµUHÀWJURXS from the Deby 29 site (P1, P2 SRVVLEOHSRLQWVRILQLWLDWLRQ of blade core exploitation; arrows indicate direction of preparation). P2 P1 P1 P2 P2 0 3cm Fig. 4. Unexploited core form from the Tomaszów II site (after: Schild et al. 1983, Table XXIII: 6). 319 Marcin Wąs $ VLPLODU XQH[SORLWHG SUHFRUH IRUP ZDV IRXQG RQ WKH VLWH 7RPDV]yZ ,, )LJ ,WVGLPHQVLRQVDUHQHDUO\WKHVDPHDVWKHSUHFRUHIURP'ęE\ The preparation of the nodule is connected with the same type of surfaces as LQ'ęE\2QO\WKHGLUHFWLRQRIÁDNLQJLVGLIIHUHQWQRWIURPWKHEURDGHU VLGHVEXWIURPWKHQDUURZVXUIDFHV SUHÁDNLQJVXUIDFHV 7KHUHVXOWZDVD SUHFRUHZLWKQDWXUDOFRUWH[FRYHUHGVLGHVDPRUHRUOHVVFUHVWHGRSSRVLWH VXUIDFHDQGDÁDWSUHVWULNLQJVXUIDFHRUWLSRIFRUH9LHZHGIURPWKHVLGHLW VKRZVDUKRPERLGDOVKDSHVLPLODUWRWKHSUHFRUHIURP'ęE\ Discussion 7KHGHVFULEHGVLPLODULWLHVEHWZHHQWKHSUHFRUHVIURPWKH'ęE\DQG7R maszów II sites are, in my opinion, not accidental. They indicate the exist HQFHRIDUHDOVWDQGDUGL]HGFRQFHSWLRQRIÁDWFKRFRODWHÁLQWQRGXOHSUHSD ration for further exploitation. These similarities imply some form of con nection between the sites. This observation has important implications for prehistoric research, pointing towards two paths of explanation: 'HÀQLQJWKHGLVWDQFHRIPLJUDWLRQIRUUDZPDWHULDO 'HÀQLQJWKHVL]HRIDWHUULWRU\XQLWHGE\WKHVDPHWHFKQRFRPSOH[ 7KHÀUVWLQWHUSUHWDWLRQLVJHQHUDODFFHVVWRFKRFRODWHÁLQWVRXUFHV0LJUD tions of at least 250 km were possible for members of the Janislavice Culture (Cyrek 1978). The second explanation involves the existence of a contact and exchange QHWZRUN7KXVWKHSURGXFHURIDSUHFRUH ´VHQGHUµ ZDVDZDUHRIWKHOHYHO RI ´NQRZOHGJHµ DQG ´NQRZKRZµ RI WKH ´FRQVXPHUµ ² LH D SHUVRQ ZKR UHGXFHGWKHSUHFRUHRQDQRWKHUVLWH7KHLPSOLFDWLRQRIWKLVLGHDLVWKHSUHV ence of direct or indirect connections between different societies inhabit ing the Vistula basin in the Late Mesolithic. In this case, the form of the SUHFRUHIURP'ęE\LQGLFDWHVWKDWWKH0HVROLWKLFSHRSOHVHOHFWHGJRRG quality nodules for exploitation. From an emic perspective, this was a suc FHVVIXOSUHFRUHLWKDGDOOWKHFULWHULDQHFHVVDU\WRFRQWLQXHWKHSURGXFWLRQ RIEODGHV7KXVLWZDVIRXQGIDUIURPWKHUDZPDWHULDOH[WUDFWLRQDUHDDV DFRPSOHWHO\UHGXFHGSUHFRUHUHFRQVWUXFWHGE\UHÀWWLQJ,WVKRXOGDOVREH SRLQWHGRXWWKDWWKHEODGHFRUHZDVQRWH[KDXVWHGDWWKH'ęE\VLWHIXUWKHU exploitation probably took place at another site later on. 320 Some remarks on contacts between Late Mesolithic hunter-gatherers societies 7KHLGHQWLÀFDWLRQRIWKHVDPHLGHDRUDVLPLODUFRQFHSWRIPHWKRGRI SUHFRUHSUHSDUDWLRQDQGEODGHFRUHH[SORLWDWLRQRQWKHVLWHVIURPGLIIHUHQW regions of the Vistula basin is interesting from a sociological point of view DQGIRULQYHVWLJDWLRQVRIWKHSV\FKRWHFKQRORJLFDODVSHFWVRIÁLQWNQDSSLQJ (Roux 1990; Schlanger 1994). Especially interesting is the problem of “know OHGJHµDQG´NQRZKRZµZKLFKDUHQHHGHGIRUDVNLOIXOXVHRIWKHSUHFRUH (Pelegrin 1990; Karlin and Julien 1994). Inasmuch as this problem is easy to explain in assemblages from the region where sources of raw material are IRXQGWKHH[DPSOHIURPWKH'ęE\VLWHLVPRUHFRPSOLFDWHG,WKLQNWKDW WKHDVSHFWRIOHDUQLQJÁLQWNQDSSLQJLQWKH-DQLVODYLFLDQDVVHPEODJHVIURP WKHSHUVSHFWLYHRIWKHJHRJUDSKLFDOGLVWULEXWLRQRISUHFRUHVVWUHQJWKHQVWKH hypothesis about connections between societies from different parts of the Vistula basin. 7KHFRPSDULVRQSUHVHQWHGDERYHKDVVKRZQWKDWWHFKQRORJLFDOXQLÀFD tion is seen not only in standardized forms of microliths, blades or cores. In WKHFDVHRIFKRFRODWHÁLQWWKLVXQLÀFDWLRQLVDOVRFOHDULQWKHIRUPRISUH cores, and of course in the idea of its exploitation. I am convinced that this DUJXPHQWFRQÀUPVWKHJHQHWLFDVVRFLDWLRQEHWZHHQ/DWH0HVROLWKLFKXQWHU gatherers from the Polish Lowland. :KDWVKRXOGDOVREHSRLQWHGRXWLVWKHFURVVFXOWXUDOVWXGLHVWKDWGHP RQVWUDWHWKDWQRQORFDORULQDFFHVVLEOHPDWHULDOVDUHRIWHQXVHGDVV\PEROVRI SUHVWLJH 7DIÀQGHU 0D\EHWKLVDOVRDSSOLHVWRFRQQHFWLRQVRIH[RWLF raw materials and skilful technologies. This aspect should be investigated in IXWXUHVWXGLHVRIFKRFRODWHÁLQWGLVWULEXWLRQLQWKH-DQLVODYLFH&XOWXUH 321 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Kim Akerman High tech–low tech: lithic technology in the Kimberley Region of Western Australia Abstract Australia is one of the few countries in which it was possible to observe and document indigenous lithic technologies as recently, in some areas, as the latter half of the 20th century. Even today in the Kimberley Region of Western Australia and in the Western Desert, there exist Aboriginal men and women, who have made and used tools of stone, and of glass or ceramic in their younger years. ,QWHUPVRIOLWKLFVWKH.LPEHUOH\UHJLRQLVSULPDULO\NQRZQIRUWKHÀQHO\ SUHVVXUHÁDNHGVWRQHDQGPRUHUHFHQWO\JODVVVHUUDWHGSRLQWVXVHGWRDUP long, light, composite spears. They were also prized as items of exchange. 0RVWRWKHUVWRQHWRROVDSDUWIURPKDIWHGHGJHJURXQGVWRQHKDWFKHWVDQG FKLVHOVDUHEDVHGRQVLPSOHÁDNHVDQGÁDNHEODGHVZLWKYDU\LQJGHJUHHVRI secondary retouch. There is a marked contrast between the technological skills required to produce the points and those used to create the other stone tools used in the area. This paper will examine the lithic technologies of three distinct cultural areas of Kimberley – The Northern, the Fitzroy Basin and the Dampierland Peninsula – focusing to a degree on the distribution of types and also on the social contexts in which they were made and used. Keywords: Australia, Kimberley Region, lithics, points, adzes, hatchets, shell tools, bone tools. Introduction 7KH.LPEHUOH\5HJLRQRI$XVWUDOLDLQFRUSRUDWHVWKHQRUWKZHVWFRUQHURI the Continent. Its borders include the Indian Ocean on the west and the Timor Sea to the north. The Great Sandy Desert lies to the south of the Fitzroy River Basin, while the eastern border of the Region is the State bor der with the adjacent Northern Territory. It is a rugged region consisting of a massive central plateau bordered and dissected by several large river V\VWHPV$ORQJWKHFRDVWWLGHZDWHUVPD\ÁXFWXDWHQLQHRUPRUHPHWUHV7KH vegetation is of open sclerophyll woodland type with patches of vine thicket DQGMXQJOHHVSHFLDOO\LQWKHQRUWK7RQRQLQGLJHQRXVSHRSOHWKHUHDUHWZR VHDVRQV7KHÀUVWLVFDOOHGWKH¶:HW·DQGLVWKHWLPHRIVXPPHU 1RYHPEHU March) monsoons, during which most of the annual rainfall, varying from 400 mm to 1400 mm annually across the region, occurs. Winter is the cooler 323 Kim Akerman ¶'U\·SHULRGLQWKHPLGGOHRIWKH\HDU,QGLJHQRXVSHRSOHVDFURVVWKHUHJLRQ recognise six or seven different seasons, marked by very subtle changes in FOLPDWHDVZHOODVWKHKDELWVRIÁRUDDQGIDXQD The region is a complex one linguistically. At the time of contact, there ZHUHVRPHWKLUW\RQHGLIIHUHQWODQJXDJHVEHORQJLQJWRVL[GLVWLQFWODQJXDJH groups spoken in an area of some 345 350 square kilometres. Figure 1. Map of the Kimberley with proposed lithic zones overlaying contemporary indigenous exchange routes. 324 High tech–low tech: lithic technology in the Kimberley Region of Western Australia ,Q,GHVFULEHGWKHUHJLRQDVFRQWDLQLQJZLWKLQLWÀYHGLVWLQFWLQGLJH nous cultural blocs (Akerman 1979a 7KHVHFXOWXUDOEORFVZHUHGLI ferentiated primarily on the basis of social and ceremonial organisation. Two of these areas are very distinct and are in fact bounded by the other three EORFVZKLFKDOVRÁRZLQWRWKHDGMDFHQWJHRJUDSKLFUHJLRQVWRWKHVRXWKDQG east. In terms of lithics and lithic technologies, however, these blocs can be reduced (at a general level) to three broad areas or zones (Fig. 1). Zone 1 covers the central and northern Kimberley and the Ord River Ba sin. In the central and northern area, social organisation is based on a patri lineal moiety system, and cosmologically the focus was on Rainbow Serpent and/or Wanjina religious beings. In the Ord River Basin however social or JDQLVDWLRQZDVEDVHGRQWKHVXEVHFWLRQV\VWHPDQGWKH5DLQERZ6HUSHQWLQ a variety of guises, appears to be the dominant cosmological entity among a body of other ancestral creator beings that travelled across the country. On the coastal margin of this area, there was some reliance on maritime resources, but generally, the economy was geared to exploit the wide variety of environments, coastal, riverine, black soil plains and the dissected rocky Kimberley Plateau. =RQH consists of the desert margin where the Great Sandy and Tana mi Deserts abut the southern margin of the Kimberley Plateau. Here an arid zone economy, focussing on the seeds of various species of acacias and JUDVVHVH[SDQGVWKHUDQJHRIHQYLURQPHQWVH[SORLWHGEH\RQGWKHFRQÀQHVRI WKHULYHUVWKDWGHÀQHLWVQRUWKHUQERUGHUV5DLQERZ6HUSHQWVDUHWKHPDMRU cosmological beings, perceived as occupying major waterholes and springs LQDODQGVFDSHWKDWLVGHQVHO\FURVVHGE\IDUÁXQJGUHDPLQJWUDFNVRIERWK individual and groups of Ancestral Creator Beings. =RQH is that area of the Dampierland Peninsula and adjacent islands QRUWKRQ5RHEXFN%D\VRXWKDORQJWKHQRUWKHUQUHDFKRIWKH(LJKW\0LOH Beach. In this zone, the people possess neither moieties nor sections and had DPDULWLPHKXQWLQJDQGÀVKLQJHFRQRP\UHO\LQJRQO\XSRQWKHWHUUHVWULDO environment for plant foods and resources and reptiles and small marsupi als. These people maintain a cosmology and ceremonial life, primarily based on the activities of a pair of cultural heroes that is quite distinct from that found elsewhere in the Region. To the south, I have indicated on the map a fourth Zone, which is in fact an extension (in terms of lithics) of the coastal Pilbara area. Here social RUJDQLVDWLRQZDVEDVHGRQWKHVHFWLRQV\VWHPDQGWKHUHZDVFRQVLGHUDEOH ritual and economic interaction with people from the wider Pilbara region as well as the desert to the east and the Kimberley proper. 325 Kim Akerman Antiquity of human occupation of the Kimberley In the early 1970s, Charles Dortch excavated a number of sites in the Ord River Valley. Subsequently these sites were submerged beneath the dammed XSZDWHUVRI/DNH$UJ\OHDKXJHDUWLÀFLDOODNHRIQHDUO\FXELFPHWUHV of water. A basal date of about 18,000 yrs BP was obtained at Miriwun rock VKHOWHU 'RUWFKDQG5REHUWV 0RUHUHFHQWO\GDWHVLQH[FHVVRI \UV%3KDYHEHHQREWDLQHGE\6XH2·&RQQRUDW:LGJLQJDUUL %RZ GOHUDQG2·&RQQRU2·&RQQRU DQGRI\UVDW&DUSHQ WHU·V *DS 2·&RQQRU DQG DW 0LPEL &DYHV %DOPH :LGJLQJDUULOLHVRQWKHQRUWKZHVW.LPEHUOH\FRDVWZKLOH&DUSHQWHU·V*DS and the Mimbi Caves are in limestone ranges in the central south Kimberley. Organic materials including plant remains and marine shells (indicative of long distance trade) are found at the lowest levels at the latter sites. Some rock art in the region has been dated by luminescence dating of TXDUW]JUDLQVLQPXGZDVSQHVWVWKDWRYHUOD\WKHDQFLHQWSDLQWLQJVWREHDW least 17 500 years old (Roberts, R. et al. Stone tool utilisation in the ethnographic recent past The following observations are based, unless otherwise indicated, upon my own work in the region, which commenced in 1966. Technologies that are the same across the three Lithic Zones will only be described in any detail once and only differences in approach or other pertinent data further devel oped. Lithic Zone 1 Heat treatment of lithic materials does not seem to have been practised in the west, northern and central areas of this area, although heat was used to FDXVHÀQHJUDLQHGFRUWLFDODUHDWRH[IROLDWHIURPVKHHWVRIPDVVLYHTXDUW] ite (Akerman 1979b:144). Among the Wunambal speakers of the area, this process was called jaran. Fires were lit under boulders of suitable material or on the surfaces of quartzite sheets. While the burning destroyed stone in LPPHGLDWHFRQWDFWZLWKWKHÀUHVWRQHWKDWKDGOLIWHGIXUWKHUDZD\ZDVQRW 326 High tech–low tech: lithic technology in the Kimberley Region of Western Australia unduly affected. In the east, however, heat treatment was utilised to improve lithic materials. The lithic assemblage of this area consists of: 1. Ground-edged tools 6WRQHD[HKHDGVRUPRUHFRUUHFWO\KDWFKHWKHDGVZHUHELIDFLDOO\ÁDNHGE\ PHQIURPHLWKHUVXLWDEOHFREEOHPDWHULDORUIURPODUJHUTXDUULHGÁDNHEODQNV of suitable metamorphic or igneous rocks. While they do occur, axe quarries DUHUDUHLQWKHUHJLRQ7KHÁDNHGSUHIRUPVDUHWKHQHGJHJURXQGDQGKDIWHG by women. Men and women who were familiar with stone axe technology were alive well into the latter half of the last century. One of my close in formants had a scar on her right hand where she had been cut with a stone axe, when as a child as she impetuously reached into a hive full of honey being chopped out by her mother. Stone axes are tools that were primarily used by women in the Kimberley, for the extraction of native beehives from hollows in trees (Akerman 1979c Grinding was generally restricted to forming the edge and rarely covered more than a third of each face of the head. In the eastern area however, about WKH2UG5LYHU%DVLQD[HVRIWHQDSSHDUWREHERWKPRUHFDUHIXOO\ÁDNHGDQG more extensively ground than in the northern and central areas (Fig. 2). Figure 2. Hafted edge-ground axe and unhafted axe-heads from the Kimberley. 327 Kim Akerman +DIWLQJLVDFFRPSOLVKHGE\EHQGLQJDVWUDSRIÁH[LEOHZRRGDERXWWKHD[H head and lashing the two ends together, compressing the head between them. Softened beeswax, prepared by mixing it with pounded charcoal or one of several plant exudates, may be placed around the head prior to tying off the helve. In this region hafting adhesives were derived from either the mountain bloodwood tree (Eucalyptus dichromophloia); the white cypress pine, Callitris columellaris; and the porcupine grass or spinifex (Triodia pungens). Pecking as a means of shaping axes is rare and appears to be restricted to archaic examples, which may also be grooved to facilitate hafting (Dortch, C.E. 1977a (GJHJURXQGD[HWHFKQRORJ\KDVDQDQWLTXLW\RI \HDUVLQWKLVDUHD 2·&RQQRU 'LFNVRQ H[DPLQHV $XVWUDOLDQ VWRQH KDWFKHWV DQG RWKHU JURXQG HGJHWRROVIURPDSK\VLFLVW·VYLHZSRLQWFRQVLGHULQJPDQXIDFWXUHKDIWLQJ resins and other adhesives and operational dynamics in detail. 2. Grinding and pounding stones Seed grinding was not a major activity in this area and, unlike other regions, deliberately fashioned grindstones were not made. Suitable slabs or pieces of DEUDVLYHVWRQHZHUHXVHGWRJULQGDQGVKDUSHQVWRQHD[HKHDGVDQGODWHUZLWK the introduction of metals, iron spearheads and tomahawks. Ochre grinding was a also major function and small grooved grindstones were used to shape ERQHDQGZRRGHQSRLQWVDQGPRUHUHFHQWO\ZLUHWRROV%HGURFND[HJULQGLQJ platforms are common only in the eastern area of this zone. Pounding stones (mortars and pestles – for general terminology relating WRJULQGVWRQHVDQGSRXQGHUVVHH6PLWK ZHUHXVHGWRSURFHVV food and resource materials. Some fruits were pounded prior to being eaten; others such as the toxic fruit of Cycas media were hulled prior to leaching and subsequent cooking. In some instances, cooked meat and bones were pounded to a pulp prior to being eaten, especially by the very young or the elderly. Some native tobaccos were prepared for chewing by pounding and both avian and vegetal down was prepared for ceremonial use by being pounded with ochre of the required colour. Pounding stones consisted of DVWDEOHERXOGHURITXDUW]LWHZLWKDVXLWDEOHÁDWWLVKVXUIDFHRQZKLFKWRUHVW the object or substance being worked and a rounded river cobble of the same material that was used unhafted as a hammer. There is little evidence to sug gest that the butt end of the stone axe was used as a hammer or pounder in this area and, unlike the situation in some other areas of Australia, axeheads themselves do not seem to have been used as anvils or hammerstones. 328 High tech–low tech: lithic technology in the Kimberley Region of Western Australia Stone pounders were used to shape metal tools, and stone anvils used as a platform upon which spearheads were made show abrasion where they have been used to grind the edges of stone pieces being worked and also to reform the working areas of wood, bone or wire indenters. Resins were pounded in the course of manufacture or prior to being used. Hammerstones were usually fortuitously selected from available cobbles when required, that is, when stones were being checked for quality etc. in the course of hunting or travelling. At home bases however, cobble hammer stones were more carefully curated. Smaller hammerstones, used to preform points etc, were kept within a bark wallet along with other tools and points LQYDULRXVVWDJHVRIPDQXIDFWXUH %DOIRXU 8QKDIWHGEODGHDQGÁDNHDQGFRUHWRROV 8QVSHFLDOLVHGNQLYHVDQGVFUDSHUVZHUHPDGHIURPÁDNHEODGHV²LHÁDNHV ZKRVHOHQJWKZDVDWOHDVWWZLFHWKHZLGWKQRWWKHVSHFLDOLVHGSDUDOOHOVLGHG EODGHV DVVRFLDWHG ZLWK 1HROLWKLF (XURSH RU 0HVR$PHULFD 7KHVH EODGHV have been described as being produced in a similar manner to the Levallois SRLQWVRI(XURSH 'RUWFKDQG%RUGHV'RUWFKE Blades ( jarung:RURUD1JDULQ\LQODQJXDJH ZHUHXVXDOO\PDGHRITXDUW] LWHWKHFRPPRQHVWDYDLODEOHOLWKLFW\SHLQWKHDUHD0HQÁDNHGWKHEODGHV but they were used by both men and women for general cutting/scraping purposes. Men used them to cut the culms of Phragmites karka, used as spear shafts and also the long thin hardwood foreshafts. From branches of the VWURQJEXWVRIWZRRGRIWKHEDWZLQJFRUDOWUHH Erythrina verspertilio), men carved their spear throwers. Jarung were used to notch the side of a suitable branch of acacia or other wood prior to bending and causing it to split when making axe handles, or to trim sheets of bark that were used to form cradles and trays or water vessels. They were also used to remove the inner bark from some species of trees when making cordage or string. Jarung PD\ EH XVHG ZLWKRXW IXUWKHU UHWRXFK RU LI LW UHTXLUHG UHVKDUS ening, was unifacially trimmed by percussion. If, of a quality suitable for ODWHUUHGXFWLRQLQWRDELIDFHSUHVVXUHÁDNHGVSHDUSRLQWDjarung could also be termed yilera¶VSHDUSRLQWSUHIRUP· 6PDOOÁDNHVRIWHQVXLWDEOHWKLQRYDWHELIDFHWKLQQLQJÁDNHVRUiriyela, ZHUH VRPHWLPHV NHSW DQG FDUULHG WR EH XVHG DV VXUJLFDO NQLYHV SDUWLFX larly for cutting the distinctive patterns of cicatrices with which most adults, men and women were adorned. They were also used to cut small therapeutic ZRXQGVRQEDFNVDQGVKRXOGHUVRUDWWKHWHPSOHV6PDOOÁDNHVRIJODVVZHUH used for the same purpose as late as 1980 (Fig. 3). 329 Kim Akerman )LJXUH6PDOOJODVVÁDNHV used for cutting therapeutic and other cicatrices. Broken section of glass bottle used as a circumcision knife. Jarung and ilera were carried wrapped in bark wallets by women. Men carried jarung, tucked into a cord band on the upper arm, while iriyelaÁDNHVZHUHFDU ULHGWLHGLQWRWKHFKLJQRQRUKDLUEXQZRUQE\PRVWDGXOWPDOHV7KHKDLU often used to carry small objects, speartips, sinew, bone awls etc. Tradition ally oriented hunters would sometimes carry a pocketknife on the head by closing the blade on a lock of hair. 2WKHUZLVHDVLQPRVWDUHDVRI$XVWUDOLDDQ\VKDUSÁDNHVRIVWRQHRUSLHFH of glass that is available may be used for a single task, such as butchering and then discarded. It must be remembered that initial butchering of most terres trial game consists of making a small incision through which the alimentary tract and other internal organs are removed. Most animals are not skinned, and, apart from large sea animals – dugong, cetaceans and turtles, further division of meat only occurs after cooking. Large fragments and blocks of quartzite that presented cleanly broken HGJHVRIÜRUOHVVZHUHXVHGLQOLHXRIVWRQHD[HVWRRSHQEHHKLYHVRUSHU form other heavier chopping functions as required. These stones were usu ally discarded on completion of the task at hand. 4. Hafted knives/scrapers Hafted stone knives and adzes are not recorded for most of Zone.1. The exception occurs in the Ord River Valley in the eastern section of the area ZKHUHDQXQPRGLÀHGSRLQWHGFRUWLFDOÁDNH O PPZ PPW mm) struck from a quartzite river cobble was hafted with resin to a stout wooden handle. Known in Miriuwang as binbalang, the short stout tool (l = 200 mm) was said to have been used as a knife, chopper or chisel (Akerman DQG %LQGRQ +DIWHG NQLYHV PDGH IURP ERWK JURXQGHGJHG and naturally fractured fragments of pearlshell (Pinctada maxima) and other oysters with a resin handle were used on the coastal areas (McCarthy 1976:91; Akerman 1995:178). Occasionally, jarung was said to be provided with a resin grip on the proximal end of the blade. 330 High tech–low tech: lithic technology in the Kimberley Region of Western Australia 3UHVVXUHÁDNHGVSHDUKHDGV There have been a number of descriptions of the lithic technology of Kim berley spearpoints. One early and incorrect description of Kimberley pres VXUHÁDNLQJE\&OHPHQW GHVFULEHGKRZDERQHLQGHQWRUZDVLPPR bilised on an anvil and the piece of stone being worked was pressed against it. Unfortunately, both Balfour (1903:65) and more recently Jelínek (1975:176) EDVHG WKHLU XQGHUVWDQGLQJ RI .LPEHUOH\ SUHVVXUHÁDNLQJ RQ &OHPHQW·V ZRUN ([FHOOHQW GHVFULSWLRQV RI SUHVVXUH ÁDNLQJ KDYH EHHQ JLYHQ E\ /RYH (1917:25–6; 1936:74–5), Basedow (1925:367–70), Idriess (1937:59–62), Elkin ² DQG7LQGDOH ² 0RRUH SURYLGHVDUHFHQW detailed description of the technology involved. 8QOLNHSUHVVXUHÁDNLQJWHFKQLTXHVFRPPRQO\SUDFWLVHGLQERWKWKH2OG World and the Americas, the Kimberley craftsman sits on the ground with one leg tucked under the other, which is extended before him. A stone (about 15 x 15 x 12 cm) is placed in front of the knapper, to serve both as an abrader and as a working platform. This is covered with a cushion of Melaleuca bark or today cloth. The knapper usually holds the piece between thumb and IRUHÀQJHUZLWKRQHHGJHUHVWLQJRQWKHEDUN+HJUDVSVWKHLQGHQWRUDFURVV WKHSDOPRIKLVRWKHUKDQGZLWKWKHWLSHPHUJLQJEHORZWKHOLWWOHÀQJHU+H places the tip of the indentor on the upper margin of the object piece and braces his wrist and arm before applying pressure by leaning forward with his body, and pushing down and outward with his hand (Fig. 4). Figure 4. PressureÁDNLQJJODVVVSHDUhead. Kalumburu. 331 Kim Akerman :RRGHQLQGHQWRUVZHUHXVHGIRUWKHÀUVWVWDJHSUHVVXUHÁDNLQJDQGVWRXW DQGÀQHWLSSHGERQHLQGHQWRUVIRUWKHVHFRQGDQGWKLUGVWDJHVRISUHVVXUH ÁDNLQJUHVSHFWLYHO\,QWKHWKLUGDQGODVWVWDJHWKHWLSLVGHYHORSHGDQGODW HUDO PDUJLQV DUH VHUUDWHG RU GHQWDWHG $NHUPDQ DQG %LQGRQ describe margin treatment on the contemporary and archaeological biface points of northern Australia. &UDEWUHH LQYHVWLJDWHGWKH.LPEHUOH\PHWKRGRISUHVVXUH ÁDNLQJZLWKZRRGHQWRROVEXWDV,SRLQWHGRXW $NHUPDQd WKH points that Don was replicating were made of glass and that it was probably likely that they were made with a metal indentor rather than a wooden one. In any case, unless the point was exceptionally large it is likely that all nega WLYHÁDNHVFDUVFUHDWHGE\ÁDNHUHPRYDOZLWKDZRRGHQLQGHQWRUZRXOGKDYH been removed by use of the bone indentor. The technique, not the tool, may KDYHUHVXOWHGLQ&UDEWUHH·VRXWFRPHVvis a vis the Palli Aike points he was at tempting to replicate. Figure 5. Potential reduction trajectories of Kimberley point reduction VHTXHQFHV 0RGLÀHGDIWHU$NHUPDQ et al. 2002). Sound descriptions of the manufacture of glass points are provided by Por WHXV ² ,GULHVV LELG DQG0DKRQ\ ,GULHVVGH WDLOVWKHXVHRIUHGKRWZLUHWRGLYLGHJODVVERWWOHVLQWRVXLWDEO\VL]HGSLHFHV Long lengths of wire were heated in the mid section, and then twisted about the shoulder of the bottle and above the base. The resulting cylinder of glass was divided longitudinally by using hot wire to create a crack and control the 332 High tech–low tech: lithic technology in the Kimberley Region of Western Australia direction in which it travelled through the glass. I have also seen the removal RIWKHEDVHVRIMDUVDQGERWWOHVE\ÀOOLQJWKHYHVVHOZLWKVDQGDQGXVLQJWKH ZLUHLQGHQWRURUZLUHSRLQWRIDÀVKVSHDUWRSXQFKRXWFRQHVRIJODVVDURXQG the inner circumference of the base until it dropped free. The bottle was then divided longitudinally with hot wire as described above. The contemporary reduction sequence of stone Kimberley points follows one of several possible paths (Fig. 5). More recently Akerman et al. LQDQDWWHPSWWRLGHQWLI\DQG interpret residues and use wear on points in the archaeological record, pro vide a detailed description of Kimberley points; their production, uses to which they are put, and their role in ceremonial exchange. Kimberley composite spears, with a reed shaft and a thin hardwood fore shaft, were among the longest of Australian spears used in conjunction with a spear thrower, with lengths ranging from 2500 mm to 3500 mm. They were however extremely light, the mean weight of 6 spears was 170 g (Aker man 1978:486). These spears were thrown with long, light and slender spear WKURZHUV ÀWWHG ZLWK D KDUG ZRRG VSXU ODVKHG ZLWK VLQHZ DQG UHLQIRUFHG with Triodia or Callitris resin. Northern Kimberley spear throwers, ranging from 90 cm to 150 cm in length, are among the longest made in Australia. The low mass of the spears, coupled with the length of the Kimberley spear throwers, meant that they could be hurled extremely long distances. There are records of Kimberley spears being thrown 140 yards (about 128 m) with accuracy at 80 yards (about 73 m) being maintained (Stuart 1923:75, 106). In earlier times, point production was a major occupation for most men, when they were not involved in hunting or ceremonial activities (Love 1936:74; Porteus 1931:113). From both observation and experimental work, LWLVNQRZQWKDWWKHDYHUDJH O PP SRLQWFDQEHPDGHLQDERXW minutes. Larger stone points generally take longer. With bottle glass points, the main effort in creating points from pieces of curved glass is directed at reducing the concave section. The inner face of the glass is subjected to VHULDOÁDNHUHPRYDOVIURPHLWKHUHGJHXQWLOWKHIDFHLVUHODWLYHO\ÁDW2QO\ then does the knapper tackle the convex face, which is, in effect, set up for RSWLPXPÁDNHUHPRYDO$VLQJOHVHULHVRIÁDNHVLVUHPRYHGIURPHDFKHGJH on this face, before serrating the edge and accentuating the tip. A glass point about 20 cm long can be made from a rectangular piece of bottle glass in about 45 minutes. Glass points made at Kalumburu dur LQJWKHVZHUHOHDIVKDSHGDYHUDJLQJPPORQJDQGPPZLGH 6HUUDWLRQVRQWKHVHSRLQWVZHUHPLQXWHUHÁHFWLQJWKHUHJXODUDQGGHOLFDWH trimming of the margin as the knapper adjusted plan symmetry about longi WXGLQDOD[LV*ODVVLQWKHIRUPRIZLUHEURNHQWDEVÁDNHGSUHIRUPVRUHYHQ entire pieces of glass ovenware were often given as gifts to master craftsmen 333 Kim Akerman DVUHFHQWO\DVWKHPLGV,QHDUOLHUGURYLQJGD\VSDVWRUDOVWDWLRQV UDQFK es) would bring herds of cattle from the hinterland, to ports and abattoirs for shipping or processing. Aboriginal stockmen (cowboys) would often go to the bottle dumps associated with these settlements and collect glass to take as gifts and trade goods for their relatives back home. Worked glass in the form of reject preforms can be found around the old dumps and also at sites where the stockmen would camp while they tended the herds at stockyards, wells and tanks on the fringes of the town. Most points for everyday are usually less than 6 cm long, and six hours of knapping could produce about eighteen points suitable for hunting and ÀJKWLQJSXUSRVHV$VNLOOHGKXQWHUKXQWLQJWZRRUWKUHHGD\VDZHHNZRXOG require no more than six points and might replace or renew four points in a week. The constant production of points was as much to make goods for gifts or exchange, as it was maintain a constant supply of spear armatures (Akerman et al, ibid %HWZHHQÀYHDQGWZHQW\SRLQWVPD\EHKHOGE\D person at any one time. Because of the nature of the material generally available – either tough TXDUW]LWHVDQGFKHUW\KRUQIHOVRIVPDOOHUQRGXOHVRIÀQHUFU\SWRFU\VWDOOLQH PDWHULDOVODUJHSRLQWV O!PP DUHUDUHLQWKLVDUHD Points were used as spear armatures, as butchering knives (while hafted as spears and unhafted), as circumcision knives, for use in ceremonial decora tion and as important trade items. With regard to the latter function, smaller points were exchanged between close kin and friends and used for mundane purposes while larger points (including glass points) usually remained cir culating within the exchange system, until they exited the area. Large white chert points that entered this area from the south were also worn as paired head ornaments in some ceremonies. In this case, they were hafted with resin to a short stick handle and placed in a headband to project forward over the temples like a pair of horns. They could also be displayed project ing vertically from the ground. If stone points were not available facsimiles carved from wood and painted white were worn in a similar manner (Petri 1954: plates 4c, 13a and b). Lithic Zone 2 This region is of great interest, as within it there are a number of stone tech QRORJLHVWKDWLQFRUSRUDWHWKHLQGXVWULHVDVVRFLDWHGZLWK=RQHDQGÁDNHDQG FRUHWRROEDVHGLQGXVWULHVJHQHUDOO\DVVRFLDWHGZLWKDULG$XVWUDOLD*URXQG VWRQH DQG ÁDNHG VWRQH DUWHIDFWV DVVRFLDWHG ZLWK ZRRGZRUNLQJ UHÁHFW WKH importance in this area of hardwood and softwood spear throwers, bowls 334 High tech–low tech: lithic technology in the Kimberley Region of Western Australia and shields; as well as hardwood spears and boomerangs in this zone. These artefacts, apart from the soft wood spear throwers, were not made or used until very recently (as a response to tourism) in Zone 1, and then are only associated with metal tools. Heat treatment of stone was widespread and associated in particular with D ÀQHJUDLQHG ZKLWH FKHUW\ FDSURFN $NHUPDQ b). This was quarried from outcrops and breakaways and also occurred in places as relict features within the soil. In the latter case, the stone was mined, pits being dug in the VRLOWRH[SRVHERXOGHUVDQGIUDJPHQWVRIWKHPDWHULDO7KHSURFHVVRIKHDW treatment is said to have been instituted by Malu, the Red Kangaroo culture hero. In 1978, I recorded a Jaru language song cycle that described the Kan JDURRÁDNLQJERXOGHUVRIWKHVWRQHFUHDWLQJODUJHELIDFHEODQNV WKHVHUH VHPEOH VRPH IRUPV RI $FKHXOLDQ KDQGD[HV DQG KHDWWUHDWLQJ WKH EODQNV As the stone cooks, the hero constructs a spear, straightening the shaft and foreshaft sections of reed and cane over the hot coals on the stone oven. The two sections are united and the stone is then removed from the oven and a VSHDUKHDGLVFUHDWHG7KHKHDGLVDIÀ[HGWRWKHIRUHVKDIWZLWKUHVLQ)LQDOO\ the Kangaroo mounts the completed spear on his spear thrower and settles it into position with a shake of the wrist, which sets the spear vibrating and EDODQFHVLWUHDG\WRWKURZ,WZDVWKLVVRQJF\FOHWKDW&UDEWUHHUHIHUUHGWR in an interview with Errett Callahan in 1979 (Callahan 1979:31). Don was XQGHUWKHPLVXQGHUVWDQGLQJWKDWWKHGUDPDZDVUHHQDFWHGDWHDFKÀULQJ Firing could be successfully undertaken in about 24 hours with some grades of this material. 1. Ground-edged tools Hatchets made from dolerites and tuff were commonly used throughout the QRUWKHUQVHFWLRQRIWKLVDUHD0DQ\RIWKHPH[KLELWH[FHSWLRQDOO\ÀQHELID FLDOÁDNLQJRIWKHSUHIRUPDQGPRUHH[WHQVLYHDQGUHJXODUJULQGLQJRIWKH cutting edge than seen on hatchets from Zone 1. $OVRSUHVHQWDUHVPDOOJURXQGHGJHGVWRQHWRROVRIGROHULWHFKHUWRUVLO LFLÀHGWXIIWKDWUDQJHIURPPPLQOHQJWKWRDERXWPPDQGIURP PPPPZLGH7KLFNQHVVYDULHVEHWZHHQPP7KHVHVPDOOWRROVDUH generally ground normal to the long axis, although smaller examples may be wider than they are long. They are usually end hafted to short stout handles, but longer examples are hand held. They are termed adzes, rather than chis els, as they are not driven with a mallet, but rather chopped into the surface of wood being worked or dragged in a scraping or shaving motion across it 335 Kim Akerman $NHUPDQ DQG %LQGRQ +DIWLQJ ZDV DFFRPSOLVKHG E\ HLWKHU lashing the stone element into a split in the handle or further securing it with resin, or by embedding it in a mass of resin attached to the end of the handle. $NHUPDQ H KDVUHFRUGHGDQH[DPSOHRIDÀQHO\PDGHELIDFLDOO\ ÁDNHGSRLQW²D.LPEHUOH\SRLQWSUHIRUP²WKDWKDGEHHQWUDQVIRUPHGLQWRD QDUURZELWWHGHQJUDYLQJWRROE\JULQGLQJGRZQWKHWLSDQGDGMDFHQWDUHDV (OERZKDIWHG DG]HV RI WKH W\SH XVXDOO\ DVVRFLDWHG ZLWK 0HODQHVLD RU Polynesia for example do not occur in Australia. Australian adzes are tools ZLWKHLWKHUVWUDLJKWRUFXUYHGKDQGOHVRIZRRGWRZKLFKVPDOOJURXQGHGJHG RUÁDNHFXWWLQJHGJHVDUHÀ[HGXVXDOO\ZLWKDQDGKHVLYHRUE\ODVKLQJUHLQ forced by an adhesive. With the introduction of metal adzes were made of a similar type by forcing a sharpened section of metal – often a ground down shearing blade – into a split made in a short stout section of a tree branch. No lashings or DGKHVLYHVDUHXVHGWRÀ[WKHEODGHWRWKHKDIW 2. Grinding and pounding stones Unlike Zone 1, milled seeds of both acacias and grasses were important ele ments in the diet of the inhabitants of Zone 2. Both the broad river valleys in the northern section of the area and the more arid southern desert environ ment to the south support a wide range of plants that provide edible seeds. Grinding plates or millstones were of major importance (Fig. 6). These were VODEVRIVDQGVWRQHWKDWZHUHÁDNHGRUÁDNHGDQGKDPPHUGUHVVHGLQWRVKDSH and with one or more grinding grooves that extend across one or both faces and over one end (the distal end). They were used in conjunction with a smaller topstone or muller. Most grindstones show grooves indicative that ZHWJULQGLQJWHFKQLTXHVRIVHHGSUHSDUDWLRQDUHEHLQJXQGHUWDNHQ VHH/RZH DQG3LNH 0LOOVWRQHVWKDWKDGEHHQEURNHQRUZHUHVRUH duced to be useless, were recycled as mullers and also used as heat retention stones in some forms of cooking. Quarries of sandstone of milling quality ZHUHDVVRFLDWHGZLWKFRPSOH[P\WKRORJ\WKDWUHÁHFWHGWKHLUHFRQRPLFLP portance. Millstones were an important item exchange in this area. Apart from the processing of cycad fruit, mortars and pestles were put to similar uses as in Zone 1. &REEOHVRUKDQG\FKXQNVRIVWRQHEHWZHHQNLORVLQZHLJKWZHUHXVHG to outline and bruise off plates of thick bark from the red river gum (Eucalyptus camaldulensis) to make trays and shallow bowls. The outline of the desired artefact was pounded heavily until the sapwood was exposed. Con tinued pounding directed around the perimeter of the exposed sapwood and 336 High tech–low tech: lithic technology in the Kimberley Region of Western Australia against the rim of thick bark caused it to lift away. Wedges, crudely fashioned from broken sticks, were used to lift the bark if it adhered to the sapwood too tenaciously. The back of a steel tomahawk if available is often used today instead of a stone pounder. Figure 6. A Walmajarri woman wet grinding the seeds of the sedge (Fimbrystlis oxystachya), using a set of grindstones marangu/ jungari, topstone- and ngamanyan bottom stone. A small wooden tray set under the lip of the lower stone catches the milled seed slurry. Unprocessed seed is in the larger wooden dish beside her. 8QKDIWHGEODGHDQGÁDNHDQGFRUHWRROV 8QKDIWHGÁDNHDQGEODGHWRROVZHUHXVHGIRUDYDULHW\RISXUSRVHVLQFOXG ing cutting and scraping. As mentioned above, the widespread presence of the white chert meant that there was ample material to provide cutting tools ZLWKRXW PXFK FRQFHUQ IRU FXUDWLRQ RU FRQVHUYDWLRQ /DUJH ÁDNHV ZLWK D VXLWDEOHHYHQXQUHWRXFKHGPDUJLQZHUHXVHGDVKDQGDG]HVIRUÀQHÀQLVK ing work on hardwood artefacts as well as being used more vigorously as picks or gouges to remove large quantities of material while roughing out VRIWZRRGDUWHIDFWVSDUWLFXODUO\FDUU\LQJERZOV6PDOOVWHHSHGJHGVFUDSHUV PDGHWKHRQODWHUDOPDUJLQVRIWKLFNÁDNHVZHUHDOVRXVHGXQKDIWHGDVZRRG working tools 2QHVSHFLDOLVHGÁDNHWRROZDVDPPORQJÁDNHPPZLGH ZLWKWKHGLVWDOHQGLQYDVLYHO\ÁDNHGWRIRUPDFRQYH[HGJH7KHVHZHUHVDLG to be spoons (= jururri-DUXODQJXDJH XVHGWRHDWWKHFRRNHGSXOSRIWKH yam (Ipomea costata). I collected examples of this implement in 1977 and sub sequently similar, but larger implements, have been recorded by Graham Heavy Flakes and discarded cores without further trimming are used to remove bark and enlarge entrances to moth larvae tunnels excavated in eu 337 Kim Akerman calyptus trees. Once the tunnel is exposed at its maximum circumference, a barbed twig made by stripping a forked branch is introduced into the tunnel DQGWKHODUYDHDUH¶VSHDUHG·DQGGUDJJHGRXW7KHODUYDHDUHHDWHQUDZRU cooked. 6SHFLDOLVHGFRUHWRROVRIWKHW\SHNQRZQLQ$XVWUDOLDDV¶KRUVHKRRIFRUHV· DUHDOVRIRXQGLQWKLVDUHDWKHVHDUHXVXDOO\VLQJOHSODWIRUPFRUHVRIDERXW JZHLJKWXVHGDVFKRSSLQJRUDG]LQJWRROV0XOWLSODWIRUPHGH[ DPSOHVGRRFFXUKRZHYHU &RRSHU)LJV 7KH\DUHFKDUDFWHULVHG E\VHULHVRIVWDFNHGVWHSIUDFWXUHVDURXQGWKHSHULPHWHURIWKHSODWIRUPWKDW often deeply undercut the distal sections of the core. In use the platform VHUYHVDVWKHEDVHRIWKHWRRODQGWKHVWHSÁDNHGPDUJLQIRUPVDVWHHSHGJHG chopping and planing tool. Flenniken and White (1985:135) suggest that horsehoof cores are in fact normal producer cores showing platform prepa ration. I have, I think demonstrated that they are tools in their own right DQGWKDWWKHSURFHVVRIGHOLEHUDWHO\FUHDWLQJVWHSIUDFWXUHVZDVDQHGJHUH sharpening method that reduced gross weight loss and conserved the mass RIWKHPDWHULDO $NHUPDQ +RUVHKRRIFRUHWRROVGRQRWKDYH DFRQWLQHQWZLGHGLVWULEXWLRQDQGFRQIXVLRQUHLJQVDVPDQ\DUFKDHRORJLVWV cannot distinguish between horsehoof and normal pyramidal cores. +DIWHGÁDNHDQGEODGHWRROV 7KHPRVWGLVWLQFWLYHIRUPDOWRROW\SHIRXQGKHUHDUHWKHDG]HVWRQHV7ZR forms are recognised. 7KHÀUVWWHUPHGLQ$XVWUDOLDWKHEXUUHQDG]HLVIRUPHGZKHQRQHODWHUDO PDUJLQRIDÁDNHLVHPEHGGHGLQUHVLQRQDZRRGHQKDQGOHRUWKHSUR[LPDO end of a hardwood spear thrower. The cutting edge is formed by the other lateral margin which is unifacially trimmed from the ventral surface to form a steep, strong edge. Subsequent retouching with wood or stone percussors RUHYHQE\WRRWKSUHVVXUHOHDYHVDFKDUDFWHULVWLFDOO\VWHSIUDFWXUHGDSSHDU DQFH7KLVLVWKHUHVXOWRIXVLQJPLQLPXPIRUFHWRHIIHFWÁDNHUHPRYDOLQ RUGHUQRWWRGLVORGJHWKHVWRQHHOHPHQWIURPWKHUHVLQKDIWLQJPHGLXP 7KHRWKHUIRUPRIDG]HLVWKHWXOD7XODDG]HÁDNHVDUHPDGHRQÁDNHV with wide and narrow striking platforms and broad and prominent bulbs of SHUFXVVLRQ7KHGRUVDOVXUIDFHRIWKHÁDNHVRIWHQKDYHDGHHSQHJDWLYHVFDU left by the previous removal of a similar, but smaller prominently bulbed, ÁDNHLPPHGLDWHO\LQIURQWRIWKHSRLQWRISHUFXVVLRQRIWKHXOWLPDWHÁDNH 7KHVHÁDNHVKDYHEHHQWHUPHGE\0RRUH JXOOZLQJÁDNHV,Q VRPHDUHDVRI$XVWUDOLD7XODDG]HVFDQEHTXLWHODUJH VWULNLQJSODWIRUPV! mm for example are not uncommon on the Barkly Tableland in the North ern Territory or in northeast South Australia), but in the south Kimberley 338 High tech–low tech: lithic technology in the Kimberley Region of Western Australia are generally no more than 25 mm wide. Platforms may be cortical, plain or IDFHWHGEXWDUHDOZD\VDWDQH[WUHPHO\REWXVHÁDNLQJDQJOH ,QL]DQ5RFKH DQG7L[LHU RIWHQ!Ü7XODFRUHVPD\EHWDEXODUSLHFHVRIVWRQH LQFOXGLQJFREEOHVRUODUJHÁDNHV7KHLQWHQWLRQRIWKHNQDSSHULVWRSURGXFH DUHODWLYHO\VKRUWDQGZLGHÁDNHLQZKLFKWKHJUHDWHUSDUWRIWKHÁDNHFRQVLVWV of the prominent bulb. 7KHÁDNHLVUHGXFHGE\XQLIDFLDOUHWRXFKWKDWUHPRYHVDOOEXWWKHEXOE DQG WKH VWULNLQJ SODWIRUP LQWR D VHPLGLVFRLGDO VKDSH +DIWLQJ LV GRQH E\ HPEHGGLQJWKHSUR[LPDOHQGRIWKHÁDNHLQWRUHVLQDWWDFKHGHLWKHUWRDVSH cial wooden handle or onto the grip of a hardwood spear thrower. The tool is used either in one or both hands in either an adzing motion or as a scraper or gouge, the working edge being that opposite the striking platform and LQWKHVDPHSODQH$VZLWKEXUUHQDG]HVVXEVHTXHQWUHVKDUSHQLQJUHGXFHV WKHZRUNLQJHGJHOHDYLQJDVWHSIUDFWXUHGPDUJLQRQWKHGRUVDOVXUIDFHRI WKHÁDNH'LVFDUGRFFXUVZKHQLWKDVEHHQUHGXFHGWRWKHSRLQWZKHUHLWLV no longer serviceable. Either the step fractures have stacked up until it is impossible to remove them, or the remnant stone is too narrow to remain HPEHGGHGLQWKHUHVLQDQGSURYLGHDVXLWDEOHZRUNLQJHGJH7KHVH¶VOXJV·DUH characterised by the presence of the striking platform, a narrow transverse UHPQDQWRIWKHYHQWUDOVXUIDFHDQGWKHVWHSIUDFWXUHGÁDNLQJSUHVHQWRQWKH dorsal surface, they are usually rhomboidal in longitudinal section. Tula slugs may be recycled by rotating the stone in the resin hafting and using the pro MHFWLQJFRUQHURIWKHSODWIRUPYHQWUDOVXUIDFHMXQFWLRQDVDQHQJUDYLQJWRRO Tula adzes are usually associated with the working of hardwoods common to arid Australia. However, they are also used extensively to work the soft ZRRGRIWKHEDWZLQJFRUDOWUHH Erythrina vespertilio ,QWKLVDUHDVRIWZRRG coolamons or carrying vessels were used for carrying water and also for stor ing grain. There appears to be a correlation of those areas, where tulas have been observed ethnographically, with collection and storage of seeds and the SUHVHQFHRIODUJHVRIWZRRGERZOV6KLHOGVDQGWUD\VDUHDOVRPDGHIURPWKH same softwood in these areas. It is interesting to note that in some arid areas, such as the Western Deserts, where tulas occur in archaeological contexts but were not used ethnographically, that there is not the same emphasis on softwood artefacts. Another implement that occurs in this zone and which has been associ ated with tula adzes in other parts of Australia, is the pirri graver (Kam PLQJD 7KLV VSHFLDOLVHG WRRO LV PDGH E\ UHGXFLQJ D ÁDNH ZLWK any pronounced area of linear curvature, those that terminated in a hinge RUUHWURÁH[HGKLQJHZHUHLGHDOWROHDYHDQDUURZVWRQHLPSOHPHQWZLWKD smooth convex (in transverse section) base. The hinge is isolated by careful SHUFXVVLRQ ÁDNLQJ DQG KDIWHG WR D ZRRGHQ KDQGOH ZLWK UHVLQ VR WKDW RQH 339 Kim Akerman end protrudes to form a gouge edge. Fluting of both hard and softwood artefacts is a common practice in arid regions, and these tools were but one of a number of engraving tools made for the purpose. Far commoner was the use of lower incisors of large and small macropods and other marsupials such as possums. These teeth with their distal ends snapped off to create a KDUGFRQYH[VHFWLRQHGHQDPHOHGJHZHUHIDUPRUHGXUDEOHWKDQWKHLUVWRQH FRXQWHUSDUWV /DUJHU SLUULJUDYHUV ZLWK VKDOORZ DUFV RI ORZHU RI FXUYDWXUH KRZHYHU SURGXFHG ZLGHU DQG VKDOORZHU ÁXWHV WKDQ GLG DQ\ RI WKH WRRWK JUDYHUV3RVWFRQWDFWHQJUDYHUVZHUHPDGHIURPZLUHVHFWLRQVRIXPEUHOOD rib, and other foraged scraps of metal. Today Aboriginal artisans use com mercial axes, chisels and rasps for most woodworking. In the early historic period, late 19th–early 20th century, two specialised types of hafted knives were used in this zone. In the east, pointed macrob lades of chert were resin hafted to form knives of the type generally known as leilira. In Central Australia, these knives are usually hafted quartzite or VLOFUHWH PDFUREODGHV 7KH .LPEHUOH\ FKHUW EODGHV DUH JHQHUDOO\ VPDOOHU than those made of silcrete or quartzite but are used for the same purposes ²ÀJKWLQJDQGFHUHPRQLDORSHUDWLRQV ,Q WKH VRXWKHUQ DQG ZHVWHUQ DUHDV RI =RQH D UHVLQKDIWHG ÁDNH ZDV XVHGIRUFHUHPRQLDOSXUSRVHV7KLVWRRNWKHIRUPRIDVPDOOÁDNH²XVXDOO\ less than 40mm in length with convex lateral margins. One margin was hap KD]DUGO\UHGXFHGE\ÁDNLQJDQGWKHQFRYHUHGZLWKUHVLQOHDYLQJWKHRWKHU PDUJLQH[SRVHG7LQGDOH GHVFULEHVWKLVW\SHRINQLIHDVXVHG elsewhere in Western Australia, in some detail. The edges were initially used ZLWKRXWIXUWKHUWULPPLQJRUÀQHO\VHUUDWHGE\SUHVVLQJDJDLQVWWKHHGJHRI DQRWKHUÁDNH:KHQEOXQWHGJHVZHUHUHVKDUSHQHGE\XQLIDFLDOÁDNLQJE\ either percussion or pressure, including tooth pressure. I have seen instances where refurbishing by percussion has created a steep stepped edge rather WKDQWKHDFXWHLQYDVLYHO\ÁDNHGDQGWKHHGJHKDVEHHQWKHQJURXQGGRZQRQ DQHOHFWULFJULQGLQJZKHHOEHIRUHEHLQJUHVHUUDWHGZLWKWKHHGJHRIDVWHHO pocket knife. 3UHVVXUHÁDNHGVSHDUKHDGV As noted above, it is in this area with a relative abundance of massive white chert caprock and outcrops that heat treatment of stone was a regular fea ture of lithic technology. The production of spearheads was a major indus try – particularly focussed on the great ceremonial exchange cycles of the region. Most points produced appear to have been sent south and southeast into the desert, where the points were particularly sought after as ceremo nial knives (Fig. 7). Tindale (1965:156) records the collection of such a point 340 High tech–low tech: lithic technology in the Kimberley Region of Western Australia 960 km (600 miles) south of its likely place of origin. As noted earlier, large white stone points from this area were also sought after as ceremonial items in Zone 1. Figure 7. Large Kimberley points, important ceremonial exchange items and a point hafted as knife. One reason why these points, rather than large points made from other VWRQHZHUHHOHYDWHGLQYDOXHPD\EHWKHIDFWWKDWWKHKHDWWUHDWHGJOHDPLQJ white chert with an almost nacreous lustre, resembled points made of pearl shell. Pearl shell ornaments, by the end of the 19th century were increasingly important items of ritual exchange that originated in the Zone 3 and adja cent northwestern coastal areas. Pearl shell was seen as a manifestation of both water and lightning, and consequently had close associations with the DOOSRZHUIXO5DLQERZ6HUSHQWEHLQJVWKDWZHUHLQWHJUDOWRPDQ\LQGLJHQRXV FRVPRORJLHV $NHUPDQZLWK6WDQWRQ ,WKDVEHHQVXJJHVW HGWKDWFODVVLF.LPEHUOH\SRLQWSURGXFWLRQLQ=RQHÁRXULVKHGDIWHU +DUULVRQIRUWKFRPLQJ 7KLVDFFRUGVZHOOZLWKWKHH[SDQVLRQRIWKH importance and use of pearl shell in the indigenous Australia (Akerman with 6WDQWRQ In Zone 2 both the Kimberley composite spear, with stone and later glass or metal tip, and simple wooden spears with plain unbarbed tips or with lashed on wooden barbs were used for hunting. Shafts for either form of spear used in this area were acquired primarily by trade. The phragmites reed, integral to the construction of the composite spear, grew in Zone 1 and the best shafts for the simple wooden spears grew primarily in the south ern areas of Zone 2. Each form of spear required a different style of spear thrower. The solid wood spears were thrown with a shorter hardwood spear WKURZHURIWKHGHVHUWVW\OHRIWHQHTXLSSHGZLWKDUHVLQKDIWHGÁDNHGVWRQH adze or scraper at the proximal end. 7LQGDOH SURYLGHVGHWDLOHGGHVFULSWLRQVRISUHVVXUHÁDNLQJRI .LPEHUOH\SRLQWVLQWKLV]RQHDQG+DUULVRQ GLVFXVVHVFXO 341 Kim Akerman tural factors that ensured the continuation of spearhead production in the area. Harrison sees the continued production of points of stone or glass in WKHSDVWRUDOHUDDVERWKDJHQGHUDQGDFXOWXUDOLGHQWLÀHU&LWLQJWKHFRQ tinued production of points by Kimberley men incarcerated on the island prison of Rottnest, nearly 2000 km to the southwest of the Kimberley, Har ULVRQVHHVSUHVVXUHÁDNLQJDVDVNLOOWKDWVHWVWKH.LPEHUOH\PHQDSDUWIURP men of other regions. Prison production of points was also seen as a way in which artefacts, made by a few artisans, could become valuable artefacts of exchange between other Aboriginal men or their white gaolers (ibid:360 . As recently as 1986 a glass Kimberley point of 100 mm or more could com mand a price in excess of A$100 among Aboriginal men in Central Australia. Large, clear glass points were seen as valuable tools used for both sorcery DQGUDLQPDNLQJLQWKH&HQWUH Lithic Zone 3 $OWKRXJKWKHUHZDVDUHODWLYHDEXQGDQFHRIÁDNLQJTXDOLW\VLOFUHWHLQWKH form of tabular boulders or slabs of material; that occurred within the more commonly outcropping micaceous schists and sandstones of the area, the range of formal tool types was not as extensive as in the previous zones. 7KHUHZDVOLWWOHLQWKHZD\RIXOWUDPDÀFRUPHWDPRUSKLFURFNIRUWKHSUR GXFWLRQ RI JURXQGHGJH WRROV 8QEDUEHG VLPSOH VSHDUV WKURZQ E\ KDQG DORQH DQG D UDQJH RI ERRPHUDQJ W\SHV LQFOXGLQJ D ÀVKLQJ ERRPHUDQJ clubs were made of hardwood. Women used hardwood digging sticks/spears IRUKDUYHVWLQJVKHOOÀVKÀVKDQGVPDOOWXUWOHVIRXQGLQUHHISRROVH[SRVHG by the fall of the 10m tides of the area. All other artefacts were made from softwood, bark or shell. 1. Ground-edged tools As mentioned above, stone, such as basalts, dolerites etc, normally associ ated with the production of ground edged tools did not seem to occur in this ]RQH $UWHIDFWV PDGH RI WKHVH PDWHULDOV ZHUH LPSRUWHG /RFDOO\ JURXQG edged tools produced locally were made of crystalline silcrete. These ranged IURPKHDY\KDWFKHWKHDGVWR PXFKVPDOOHUÁDNHWRROVZLWKDJULQGLQJUH stricted to the immediate area of the cutting edge. *URXQGHGJHGD[HVDQGFKLVHOVVLPLODUWRWKRVHSURGXFHGLQ=RQHZHUH important trade items to this area. Most wooden artefacts apart from spears, clubs and boomerangs, were made of softwood – usually that of the helicop 342 High tech–low tech: lithic technology in the Kimberley Region of Western Australia ter tree (Gyrocarpus americanus). Pieces of baler shell (Melo amphora), fashioned LQWRJURXQGHGJHGNQLYHVDQGDG]HVWKDWZHUHXVHGERWKKDIWHGDQGXQKDIW HG $NHUPDQ ZHUHXVHGIRUPXFKRIWKHZRRGZRUNDQGDOVRVDLG to have been used for lighter butchering purposes. Other tools associated with woodworking were also made from shells. Drills, gouges and chisels, collectively known as juror, made by grinding a strong sharp edge on the anterior fasciole of Syrinx aruensis columellas were also made and used by the 1\XO1\XODQG%DUGLRIWKH'DPSLHUODQG3HQLQVXODRI:HVWHUQ$XVWUDOLD$ hatchet of baler shell, from Sunday Island, has been described by Davidson DQG,KDYHDOVRVHHQWZRJURXQGHGJHKDWFKHWKHDGVPDGHIURP pieces of clam shell at One Arm Point on the northern end of the peninsula. +DQGDG]HVZHUHDOVRPDGHIURPWKHVKRUWKHDY\LYRU\XSSHUWXVNVRIWKH PDOHGXJRQJ $NHUPDQQG 2. Grinding and pounding stones As with Zone 1, milling of grasses and other seeds was not of major impor tance, and grindstones made of shelly conglomerate and micaceous sand stones of varying grades appear to be associated with the manufacture and PDLQWHQDQFH RI VWRQH D[HV DQG VKHOO WRROV 6PDOO ! PP [ PP plaques of sandstone showing narrow grinding grooves were used to hone the shell tools. Larger grindstones were also used to clean, shape and polish the large valves of the pearlshell Pinctada maxima, into oval ornaments which are ubiquitous to the area. Pounding stones were used to process fruits of a wide variety of plants. Small cobbles and slabs of country rock were also used as anvil stones to open a range of gastropods that were important dietary elements and which ZHUHDOVRXVHGDIWHUSRXQGLQJWRDWWUDFWUHHIÀVKZLWKLQVWULNLQJGLVWDQFH with both spear and boomerang. Anvil stones are characterised by pocking and dimpling of one or more surfaces caused by the repeated impact endured LQWKHFUXVKLQJSURFHVVWKLVLVQRWXQOLNHWKHDQYLOZHDUIRXQGDIWHUSUR longed bipolar percussion technique of stone reduction. 8QKDIWHGEODGHDQGÁDNHWRROV In this area, there are numerous exposures of silcrete of varying grades. The ÀQHVWZHUHXVHGWRSURGXFHEODGHVDQGÁDNHVWKDWZHUHXVHGXQKDIWHGIRU both woodworking and butchering purposes. Apart from the initial produc WLRQRIWKHEODGHRUÁDNHWKHUHDSSHDUVWRKDYHEHHQOLWWOHPRGLÀFDWLRQRI edges. Secondary retouch when it does occur seems to have been limited to 343 Kim Akerman FUHDWLQJDFDVXDOEDFNLQJRIDEODGHRUÁDNHWRSURYLGHDPRUHFRPIRUWDEOH RUVDIHUWRROWRZRUNZLWK$QDQDO\VLVRIDFDFKHRIEODGHVDQGÁDNHVIRXQG DZD\IURPWKHTXDUU\VRXUFHLQGLFDWHVWKDWÁDNHVDQGEODGHVZLWKRQHHGJH more obtuse than the other (i.e. naturally backed), were the desired form. Unilateral, unifacial retouch was used to create this condition however, if it ZDVQRWSUHVHQWRQDEODGHRUÁDNHZLWKDVXLWDEOHHGJH2QHÁDNHLQWKLV sample that was much wider than it was long, was apparently selected be cause the broad striking platform area formed a suitable backing (Akerman 7KHVHEODGHVDQGÁDNHVZHUHSULPDULO\XVHGIRUWKHEXWFKHU ing of dugong (Dugong dugon) and green turtles (Chelonia mydas). Both these animals have thick hides, and in the case of the latter, a tough carapace. 6PDOOHUÁDNHVZHUHXVHGDVFDVXDONQLYHVZKHQWKHRFFDVLRQGHPDQGHGLW but broken fragments of shell were also used for these tasks. 3UHVVXUHÁDNHGVSHDUKHDGV 8QOLNHRWKHUDUHDVRIWKH.LPEHUOH\VWRQHWLSSHGVSHDUVZHUHQRWPDGHRU used on the Dampierland Peninsula. They did enter the area as exotic trade LWHPVWKDWZHUHSDVVHGRQEXWQRWXVHGLQIDFWWKH\ZHUHSUHVWLJHJRRGV Stone points similarly came into this zone, and were used occasionally as ritual knives. In the early historic period, the Catholic Church created a number of mis sion settlements on the Dampierland Peninsula. While some of these catered for the local indigenous population, others served also as centres to which Aboriginal people from other areas of the Kimberley (that is Zones 1 and 2) were resettled. On some sites in the sand dunes near these latter centres SUHVVXUH ÁDNHG JODVV .LPEHUOH\ SRLQWV DUH RFFDVLRQDOO\ IRXQG VXJJHVWLQJ that at least some of the immigrant men continued to practise the art of point making in their new home. Around the town of Broome, on the southwestern corner of the Dam pierland Peninsula, glass points in various stages of manufacture are often found. Like the glass points found on Rottnest Island mentioned earlier, these too are ascribed to the activities of Aboriginal prisoners brought into the regional prison from other areas. $NHUPDQDQG%LQGRQ UHSRUWRQRQHVLWHLQWKHVRXWKHUQDUHD of this zone where, within a general quarry and blade workshop context, a 3m x 3m area contained 44 bifacial and unifacial points, made of local mate rial. The points were in good condition and any damage appeared to have occurred subsequent to deposition. This feature of the general site area is interpreted as the result of a single episode of experimentation or demonstra tion of Kimberley point technology, within historic times. 344 High tech–low tech: lithic technology in the Kimberley Region of Western Australia Conclusions In the three zones described it is clear the lithic and associated glass technol ogy was not the same across the Kimberley (Fig. 8). Zone 1 Has ground edged axes and associated grindstones; mortars and pes WOHVVLPSOHÁDNHDQGEODGHNQLYHVDQGVFUDSHUVDQGSUHVVXUHÁDNHG.LPEHU OH\SRLQWV$SDUWIURPWKHKDIWHGFRUWLFDOÁDNHWRROVIRXQGLQWKHHDVWNQLYHV and scrapers were used unhafted. =RQH+DVJURXQGHGJHGD[HVDQGDG]HVDQGDVVRFLDWHGJULQGVWRQHVPRU tars and pestles and grindstones associated with the milling of seeds. As well DV WKH FDVXDO XVH RI ÁDNH WRROV WKHUH DUH UHVLQKDIWHG ÁDNHDG]HV KDIWHG HQJUDYHUVDQGWZRIRUPVRIUHVLQKDIWHGNQLIH+HDWWUHDWPHQWRIVWRQHZDV ZLGHO\SUDFWLVHGDQGODUJH.LPEHUOH\SRLQWVSURGXFHGVSHFLÀFDOO\IRUFHU emonial use and exchange. =RQH+DVJURXQGHGJHGD[HVRIORFDOVLOFUHWHLPSRUWVD[HVRIGROHULWHRU other stone, associated grindstones. Mortars and pestles occur and anvil VWRQHVDVVRFLDWHGZLWKVKHOOÀVKSURFHVVLQJ([WHQVLYHXVHRIJURXQGHGJHG VKHOOWRROVDQGDVVRFLDWHGJULQGVWRQHV8QKDIWHGEODGHVDQGODUJHÁDNHVRI VLOFUHWHZLWKQDWXUDOO\RUDUWLÀFLDOO\EDFNHGPDUJLQVXVHGIRUKHDY\EXWFK ering. Kimberley points imported but not used, evidence for manufacture of these points in the historic period by visitors from other regions of the Kimberley. We can see how in the Kimberley, three different suites of technology abut each other. Even though there is much social interaction between the three areas, including ceremonial and economic exchange systems and intermar riage, the technologies remain quite distinct. This is as much, or more, a fac WRURIFXOWXUDOLGHQWLÀFDWLRQWKDQDUHÁHFWLRQRIHQYLURQPHQWDORUHFRQRPLF constraints. Members of each proposed lithic or technological zone appear to be bound by certain cultural constraints from adopting technologies unique to their neighbours. Interestingly these cultural constraints are not applied HYHQKDQGHGO\,QWKHSDVWFHQWXU\FKDQJHVLQVRPHFHUHPRQLDODFWLYLW\DQG social organisation has occurred, as have some technological changes – the introduction of the didjeridoo from much further east, for example – with out destabilising the cultural identity of any one area. 345 Kim Akerman )LJXUH6RPHRIWKH.LPEHUOH\ÁDNHGVWRQHLPSOHPHQWVUHIHUUHGWR in the paper. L-R: silcrete blade, naturally backed at the tip, Zone 3; Top Row: quartzite blade jarung, Zone 1; above – two ground-edged adzes and pirri-graver, Zone 2; glass and quartzite spearheads showing resin hafting. Bottom Row: ground-edged axe head, Zone 1; three tula adze stones in various stages of reduction, Zone 2; large glass Kimberley point, Zone 1; prestige exchange Kimberley point of heat treated chert, Zone 2. The opportunity to observe traditional technological minutiae, among Aus WUDOLD·VLQGLJHQRXVSHRSOHVSHUPLWVDÀQHUXQGHUVWDQGLQJRIWKHFRPSOH[LWLHV RIKXQWHUJDWKHUHUUHVSRQVHVWRPDQ\RIWKHGD\WRGD\WDVNV7KLVLQFOXGHV the diverse ways in which stone tools are made and used within one small corner of the Australian continent. Acknowledgements My heartfelt thanks must go to Jacqui Ward of the Tasmanian Museum and $UWV*DOOHU\ZKRODERXUHGWRSURGXFHWKHPDSDQGÀJXUHVWKDDFFRPSDQ\ this paper. Val Hawkes, my wife, read and corrected the many drafts that emerged as the paper was written. Her criticisms were always concise and to WKHSRLQWHQVXULQJWKDWWKHÀQDOGUDIWZDVDEHWWHUSDSHU 346 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Per Falkenström A matter of choice: social implications of raw material variability Abstract Many studies on lithic technology have focused on raw materials, but vari ability has not always been a central theme. While technological strategies EHDUZLWQHVVWRVRFLRHFRQRPLFSDWWHUQVPRUHLQIRUPDWLRQFDQEHREWDLQHG by looking at one decisive initial step before reduction: the choice of raw ma terial. It is suggested that this choice has social connotations based on indi vidual experience as well as on collective ideas shared by group members. In addition, this may be subordinated to an ideological community, whereupon the choice of raw materials can be valued and ritualised in different ways. Such a choice includes further social considerations that are worth examin ing. In order to obtain additional information about the choice of raw ma terials and how they are communicated in society, examples are drawn from other social disciplines with illustrations from ethnographical literature and archaeological investigations. Introduction Raw materials offer important information about prehistoric societies. In this article it will be stressed that certain materials are chosen and acquired within established technological traditions. This implies a dialectal relation ship between individuals, groups and the materials they choose to use. It is WKHÀUVWVWHSLQDQRSHUDWLRQDOFKDLQDQGLWPLJKWZHOOEHDGHFLVLYHFKRLFH with regard to the technological tradition and future relations with other groups. Will the chosen material be approved within a given technological tradition? Can it be used for daily use or as prestige items in exchange? These questions are crucial in a discussion about the importance of raw materials, how they are used and valued in a social context where stone technology is a part of everyday life. In order to shed some light on this issue, the following themes will be addressed: 347 Per Falkenström • Implications with regard to distance between settlements and raw material sources • The raw material in social life on the local level • Raw materials in exchange networks on the regional level Concerning the choice of raw materials, an attempt is made to discuss these themes within a theoretical framework based on technological and social reproduction. The discussion will be considered partly in relation to ethno graphic studies and partly to material culture as it appears on a few archaeo logical locations in central Scandinavia (Fig. 1). In northern Scandinavia, diagnostic artefacts and absolute datings ac count for a high degree of variability, especially during the Mesolithic peri RGDERXW%3 2ORIVVRQ /HVVYDULDELOLW\FDQEHDVVLJQHG WRKXQWHUJDWKHUHUFRQWH[WVLQWKHVXEVHTXHQWSHULRGVDVODWHDVDERXW BP. However, the acquisition of different local materials should not be un GHUHVWLPDWHG /DQQHUEUR1RUHOOS $IHZGHFDGHVDJRWKLVSDWWHUQ was explained from a south Scandinavian perspective, i.e. that comparatively FRDUVHUDZPDWHULDOVZHUHXVHGDVVXEVWLWXWHVIRUÁLQW7RGD\WKLVYLHZKDV changed and is more focused on local and regional conditions (Knutsson et al. 1999:106). It must be stressed that lithic sources only encompass a limited part of the entire material culture. With this in mind, it is not possible to deal with the whole variety of societies practising stone technology. The main pur pose is rather to understand how raw materials can be used and valued in the societies discussed here. Hence, by using different sources I will discuss raw materials from a social perspective. They are seen as integral parts of an operational chain, which here is used in the same way as the French concept chaîne opératoire. Social perspectives on raw materials In societies where stone technology is a part of everyday life, people are in HYLWDEO\FRQIURQWHGZLWKOLWKLFPDWHULDOV7KHVLJQLÀFDQFHRIWKHVWRQHVPD\ vary a lot, but can barely be seen as isolated in relation to the whole society and its environment. There might be traditions about the landscape, where WRÀQGYDOXDEOHUHVRXUFHVDQGKRZWRXVHWKHP Technology is thus not necessarily limited to material culture. Attempts to widen and contextualise this concept will fortunately explain technologi 348 A matter of choice: social implications of raw material availability FDODVZHOODVVRFLRHFRQRPLFVWUDWHJLHV7KXVDFTXLVLWLRQRIUDZPDWHULDOV cannot be excluded from social organisation, division of labour or exchange networks. I see each acquisition event as a choice within a technological tra dition. This would require knowledge about the technological process, about the subsequent steps of lithic production and even how to cope with possible social tensions when essential information is learned and communicated. Certain production stages demand a high degree of individual skill, which can be attained only through practice. Although raw material quality can be estimated by experience, it can never be fully anticipated. The possibility of practising is thus governed by natural availability. Apel has shown that spe cialised manufacturers live in the vicinity of raw material sources. Far from the source there are reasons to expect only a few persons working with rare materials. Thus, we can expect more people taking part in lithic production where the sources were available to everyone (Apel 2001:28, 100). With the processual archaeology many methods were launched in order to deal with ancient societies. For instance, lithic analysis has made consid erable progress since the introduction of the chaîne opératoire concept in the ·V,QWKHIROORZLQJGHFDGHVIXQFWLRQDOLVWLFDSSURDFKHVZHUHFKDOOHQJHG when both tools and debris were related to social and economic conditions. Technology was divided into collective ideas in opposition to individual practice and experience. At the end of the century, more problem orient ed studies were triggered by microwear analysis and practical experiments. Later on, cognitive aspects were gradually brought into the concept of chaîne opératoire )qEORW$XJXVWLQV (ULNVHQ 1HYHUWKHOHVV WKH new perspectives are by no means unproblematic. Without an oral tradition, ÀQGFRQWH[WVZLWKIRUPDOWRROVDVZHOODVSURGXFWLRQGHEULVVWLOOSURGXFH limited information. Besides, our interpretations are more or less biased by current paradigms and personal experience. Additional information chan nels are needed. The initial stage of chaîne opératoire is characterised by either direct or in direct procurement strategies (Eriksen 2000:80). The former can be seen as direct contact with the natural source, including quarrying and eventual quality tests. The latter strategy has more to do with exchange or secondary use of the material. To address the question why certain raw materials have been chosen, a number of factors can be mentioned. These are partly functional or deter ministic explanations, involving the practical and economic aspects of stone procurement strategies. According to this view, certain raw materials are seen as suitable for performing certain tasks. Functional interpretations have DOVREHHQJLYHQWRTXDOLW\LQWHUPVRIÁDNLQJSURSHUWLHVRUH[SHFWHGXVHRI E\SURGXFWV 7DIÀQGHU 6RPHPDWHULDOVPLJKWDOVRKDYHEHHQFKRVHQ 349 Per Falkenström GXHWRVHWWOHPHQWSDWWHUQVRUVLPSO\E\ORFDODYDLODELOLW\ /DQQHUEUR1RUHOO 1987:68). Other factors have been emphasised in postprocessual research and stress the importance of rituals and social networks in connection with TXDUU\LQJ1HLWKHUSHUVSHFWLYHFDQEHLJQRUHGEXWVKRXOGEHQHÀWIURPHWK nographic material and anthropological analyses. Some contributions from other social disciplines will be discussed below. The use and meaning of rocks – two anthropological studies The papers presented in this volume make it clear that stone technology should be studied within a framework including both practical experiments DQGHWKQRDUFKDHRORJLFDOGDWD7KHVLJQLÀFDQFHRIUDZPDWHULDOVLVHVSHFLDOO\ pointed out in anthropological studies. Arnhem Land and Western Desert, Australia Tacon has studied a technological tradition in Arnhem Land, Australia. He particularly mentions the symbolic meanings of quartz and quartzite. These materials and associated rocks were regarded differently in myths and sacred rituals. Light and bright quartz was seen as a powerful symbol of life. Colour and workmanship further reinforced the link with the Ancestors. It also facilitated control of the stone source. Owning tools made from certain raw materials could even lead to increased prestige (Tacon 1991:198 p). Quartzite, on the other hand, was used for trade and ceremonial exchange of items such as spear shafts. While stone points were rare, they were popular exchange objects. The myths of quartzite was similar to that of quartz, but its power as a ready made object made it more valued (Tacon 1991:202). The myths also explain how the rocks were created and thereby reinforcing the contact with the power of Ancestral Beings that were attached to the rocks. “The Rainbow Serpent, one of the most potent of all Beings, swallowed other Ancestors and then was forced to vomit their bones, which then formed the rocky sandstone and quartzite escarpment and other geological features of the landscape.” (Tacon 1991:195) Material of distant origins is a repeated theme in the ethnographic literature. Materials used in social networks are those functionally superior to local materials. Although it is possible to increase the value of a tool through re touching, grinding etc., it is the origin of the material that is most valued or 350 A matter of choice: social implications of raw material availability even sacred. Access to certain quarries in the Western Desert of Australia is controlled and maintained through kinship alliances. The acquisition and use can even be controlled within the group, as women are not allowed to XVHFU\SWRFU\VWDOOLQHURFNV 7DIÀQGHUS The Highlands of Papua New Guinea Among axe making groups in Papua New Guinea, Burton emphasises the ritual behaviour at quarries. In an area at the Tuman River, several hornfels quarries are scattered along a valley. Each quarry is situated within a clan ter ritory and is owned by its members, although many quarries are overgrown DQGQRORQJHULQXVH7KHPDWHULDOLVÀQHJUDLQHGKRUQIHOVWKDWLVVXLWDEOHIRU making sharp edges. All clans at the Tuman River participated in quarrying H[SHGLWLRQVDQGWKH\GLGVRDWWKHVDPHWLPHZLWKLQWHUYDOVRIXSWRÀYH\HDUV However, successful making of axes was not only due to cooperation. Since quarrying was thought to be dangerous, an expedition must not be under taken without ritual precautions. Animals that appear in myths were killed for the spirit sisters, who were thought to control the quarries. In connection with these and similar hazardous activities, men were not allowed to meet women directly. During the expedition, the men lived in enclosed camps for PRQWKV(YHU\GD\WKHZRPHQZHQWWRWKHFDPSDQGGHOLYHUHGIRRGRQ wooden platforms. Quarrying activities were collective tasks and therefore all men involved had reasons to claim an axe. After sharing the raw material they announced that women could enter the camp, destroy it and beat their men in order to drive away the female spirits (Burton 1984:240 pp). The quarrying organisation employed hundreds of individuals. All lived in a tribal society where large men were the most respected persons, although without real power. Their decision seems to be a consequence of collec tive negotiations regulated in the belief system. Personal wealth was further restricted when a certain person of each clan was appointed to distribute the axe stone and thereby expressing equality principles. After that, the raw material was regarded as private property, although it engendered individu al prestige that was supposed to be transferred to the clan. Most preforms could be manufactured into comparatively small working axes, while only a few were large enough to be valued as prestige items and therefore used in competitive exchange networks between clans (Burton 1984:244). The fact whether the quarrying was launched by economic demands, so cial or ritual factors can be discussed. The natural prerequisites in the valley and the location of villages next to them cannot be ignored. One might sus pect a centralised society when looking at the number of participants and the 351 Per Falkenström amounts of stone, but on the contrary, one purpose of quarrying activities was to maintain exchange networks and prevent inequality. Rules and rituals ZHUHZD\VWRSURQRXQFHWKHVLJQLÀFDQFHRIWKHPDWHULDOLQFUHDVHWKHZRUWK and the exclusivity of the raw material source. Natural sources – cultural considerations With an archaeological perspective, what can be learned from the anthro pological studies mentioned above? They can barely be compared with pre historic societies concerning different sorts of geological deposits. Neither is the anthropological situation convenient to enlighten general prehistoric conditions. However, they show how raw materials can be used in mythol ogy, social structure and ritual behaviour. The studies underscore that it is important to consider raw materials in a wider cultural context. $ WHFKQRHFRQRPLF SHUVSHFWLYH LV ZHOO UHSUHVHQWHG LQ VWXGLHV GHDOLQJ with material remains, for instance raw materials, and how they are distrib uted in the landscape. The distribution of raw materials has been used to discern mobility patterns among Lower and Middle Palaeolithic populations LQ(XURSH:KLOHPRVWPDWHULDOVZHUHDFTXLUHGZLWKLQÀYHNLORPHWUHVIURP the occupation sites, very small quantities were acquired outside that zone. 7KH\GRQRWRQO\UHÁHFWPRELOLW\SDWWHUQV7KHSURFXUHPHQWVWUDWHJ\LVDOVR seen as a cultural norm with assumed social interaction between different UHJLRQV )qEORW$XJXVWLQV 6XUYH\VLQ%ULWWDQ\LQ)UDQFHVKRZWKDWRQO\ODUJHÁDNHVRIORFDOPDWHULDOV are left at the quarries and the cores are apparently taken away. This pattern has been interpreted as specialised quarry sites in contrast to settlements ZKHUH ÁDNHV RI DOO VL]HV KDYH EHHQ IRXQG 7KH TXDUULHV ZHUH VLWXDWHG RQ suitable locations with regard to migration routes of Mesolithic populations. In this case, the raw material quality was of secondary importance (Yven 2004:729 pp). Quarrying activities can be undertaken either at large outcrops in the ground or in tiny veins embedded in boulders. It would thus be helpful if prehistoric quarries can be found and be related to the distribution of each material respectively. Unfortunately, such comparisons are quite rare due to isostatic uplift, natural erosion and recent vegetation that prevent visibility of already extracted dykes and outcrops. In other cases, hornfels quarries are no longer detectable because of chemical erosion (Berg 2003:285). Ap parently, some quarries do not show any traces of use or only of temporary use, while other quarries have been used intensively. General trends can be observed over time but the lack of secured, contemporary events at quarries 352 A matter of choice: social implications of raw material availability PDNHVLWGLIÀFXOWWRHVWLPDWHWKHWLPHRUVRFLDORUJDQLVDWLRQQHHGHGWRIRUP the archaeological assemblages. Hence, more social factors should be consid ered to understand how raw materials are used in different societies. Although quarries often imply good quality, the locations are outnum bered by occasional, usually low quality, moraine deposits. The latter source has been devoted less research in archaeology, especially as similar collecting DFWLYLWLHVDUHGLIÀFXOWWRDVVHVVRQO\IURPPDWHULDOFXOWXUH+RZHYHUVRPH results on this problematic issue have been presented. In northern Finland it has been suggested that glacial quartz deposits have been collected from VWRQH ÀHOGV $ORQJ ZLWK RWKHU DFWLYLWLHV DOWKRXJK QRW QHFHVVDULO\ FRQWHP porary, the quartz was stored and even prepared on the spot (Alakärppä et al ,W LV GLIÀFXOW HQRXJK WR UHODWH FHUWDLQ UDZ PDWHULDOV WR VSH FLÀFTXDUULHV7KHSRVVLELOLWLHVRIORFDOLVLQJPRUDLQHGHSRVLWVZRXOGEHHYHQ more limited. It has already been discussed how raw materials can be valued LQUHODWLRQWRTXDUULHVDQGVRFLDOHYHQWV:LWK7DFRQ·VVWXG\LQPLQGLVLW possible to think of symbolic meanings concerning acquired materials from moraine deposits? Fig. 1. Area under study with locations mentioned in the text. 353 Per Falkenström Raw materials and their social context have been highlighted in a few stud ies dealing with axe production in Neolithic societies of central Sweden, just south of the study area in this paper (Fig. 1). In the mid 1990s, several locations were excavated, a few of which proved to be contemporary Early Neolithic settlements. A petrographic analysis of porphyrite materials re sulted in three raw material groups. Although no natural source could be found, each group corresponded to separate settlements. The results have been interpreted as that each settlement was a household unit in control of UDZPDWHULDODFTXLVLWLRQDVZHOODVSURGXFWLRQDQGXVHRIWKLQEXWWHGSRUSK\ rite axes (Sundström & Apel 2001:334). Unless the porphyrite was distributed outside the settlements it might be a valuable local resource. When looking at the technological strategy as a whole, it seems as though different reduction methods have been applied to GLIIHUHQWVRUWVRIUDZPDWHULDOV&KDQJLQJUHGXFWLRQPHWKRGVIXUWKHUUHÁHFW choices that are dependent on availability, hardness of the material and skill of the knapper (Eriksen 2000:86). This pattern is also what could be expect ed in an egalitarian society (Sundström & Apel 2001:332). If by egalitarian structures we mean absence of personal wealth, there will be ways to acquire personal prestige through exchange networks. Nevertheless, the egalitarian system is maintained as long as the prestige is transferred to the whole group. By comparison, how are raw materials used and valued in the Late Neolithic, a period during which more inequality structures can be expected? Lekberg has shown that long and angular Late Neolithic axes were ex tracted from quarries, while shorter axes were derived from rounded morai ne deposits. Both types have been used as working axes, but the appearance and length of the object, as well as the raw material source, might be signals of social prestige acquired by the owner (Lekberg 2002:198 p). In general, LWVHHPVWKDWH[WUDFWLRQIURPTXDUULHVLVPRUHWLPHFRQVXPLQJWKDQÀQGLQJ raw materials in moraine deposits. The latter is often used for functional SXUSRVHVDQGKDVOLWWOHVRFLDOVLJQLÀFDQFH5DZPDWHULDOVIURPTXDUULHVRQ the other hand, are more associated with personal status and prestige ob jects. Somehow, we can expect that the artefact value changes in an exchange system. In line with this argument, even moraine deposits might have sym bolic meanings. For similar reasons, the raw material value may change in different stages of the operational chain. In egalitarian societies it is reasonable to think of talented members who collect appropriate materials for intentional production and use, possibly even to establish or maintain social networks. By providing other groups with raw materials, objects and new ideas, it is more likely that social networks are maintained. The situation is somewhat different in a society with more ac centuated social structures, where specialists have to be provided by other 354 A matter of choice: social implications of raw material availability group members. Within such a redistribution system it is likely that techno logical knowledge is maintained within the family unit (Apel 2001:100). $V LQ %XUWRQ·V VWXG\ WKH D[HV PLJKW EH YDOXHG GLIIHUHQWO\ LQ WKH /DWH Neolithic society. These values are communicated between group members DVWKH\PLJKWUHÁHFWFRPSHWLWLRQDQGFRQWURORYHUUDZPDWHULDOVRXUFHVDQG those who produced the axes. It is possible that certain individuals or social JURXSVEHQHÀWHGIURPWKHXVHRIUDZPDWHULDOVDQGWKHSURGXFWV,IUDZPD WHULDOVFRQWULEXWHGWRLQÁXHQWLDOSRVLWLRQVLQH[FKDQJHQHWZRUNVSUHIRUPV UHDG\ PDGH REMHFWV RU VHFRQGDU\ PRGLÀFDWLRQV VKRXOG DOVR EH FRQQHFWHG with certain values. The function and value could be estimated already at the source. Ideas and activities that were carried out at the sources could thus be seen as integral parts of the natural disposition of resources, social structures and the belief system in society. Local investigations – regional implications Just to the north of central Sweden, the landscape relief is more accentuated in contrast to the plain terrain in the south. With smooth hills, lakes and river valleys the climate and vegetation of the area is more reminiscent of northern Sweden. This is also characteristic for the province of Dalarna in the interior. Even the cultural history has seen different research traditions. While certain areas are well known through excavations, the prehistory of Dalarna is known almost exclusively by surveys of vast areas. Figure 2. A Mesolithic cooking pit at Tjärna, Borlänge. Digitalisation by Christoffer Samuelsson. 355 Per Falkenström Similar to in southern Norway, there is a decreasing raw material variety at the end of the Mesolithic. Berg has interpreted this pattern as that there was much experimentation with local materials during most of the Mesolithic (Berg 2003:286). How can we explain this strategy? Conical microblade cores are among the oldest known artefacts from this region, dating back to at least 8000 BP (Olofsson 2003:77). They were made of a wide variety of materials. It is possible that the choice of material had more to do with a mobile settle ment pattern than with experimentation. Since few investigations have taken SODFHLWLVGLIÀFXOWWRDVVHVVPLJUDWLRQURXWHVDQGH[FKDQJHQHWZRUNVDVZHOO as changing patterns of raw material procurement. However, one exception deserves special attention. In 1998 a Late Iron Age settlement was excavated in the town of Borlänge in the province of 'DODUQD7KHVLWHFRQWDLQHGDFRRNLQJSLWZLWKÀUHFUDFNHGVWRQHVOLWKLFGHE itage and bones of moose (Fig. 2). Apparently the pit did not belong to the ,URQ$JHVHWWOHPHQWEXWZDVRIDQHDUOLHUGDWH,WFRQWDLQHGÁDNHVRITXDUW] SRUSK\U\DQGVLOLFLÀHGWXII7KHVHPDWHULDOVZHUHFKRVHQIRUSURGXFWLRQRI microblades. The 14&DQDO\VLVFRQÀUPHGWKDWWKHFRRNLQJSLWZDVIURPWKH Mesolithic, dated to about 7960±35 BP, and that it probably was part of a tem SRUDU\KXQWLQJVWDQG 6DQGEHUJLQSUHVV $OOWKHÁDNHVDQGFRUHVIRXQGLQWKH cooking pit were apparently deposited at the same time. They also indicated that at least three different materials were brought to the site and that both bipolar and microblade production were used simultaneously (Fig. 3). Core Microblade Flake Quartz Material 5 - 249 Porphyry 2 2 17 6LOLFLÀHGWXII - - 8 Figure 3. Distribution of raw materials and number of artefacts in the cooking pit at Tjärna, Borlänge. While awaiting more data from this district, distribution patterns of other materials must be relied on. Flint was imported into the area, partly from southern Scandinavia and partly from the west coast of Sweden and Norway. /RFDOPDWHULDOVVXFKDVMDVSHUDQGVLOLFLÀHGWXIIVKDUHHTXLYDOHQWFKDUDFWHULV tics and were probably highly appreciated due to their quality and colour. It is possible that these materials were regarded as prestige symbols. For groups living far from the sources, in particular, these and similar materials might be valuable resources reserved for manufacturing by persons with enough skill and knowledge. 356 A matter of choice: social implications of raw material availability In Los parish, surveys have yielded large amounts of lithic artefacts, main O\LQKXQWHUJDWKHUHUFRQWH[WV,QDGGLWLRQWRMDVSHUÁLQWDQGVLOLFLÀHGWXII locally available quartz and porphyries dominate the record. A research pro ject at Fågelsjö in Los parish was carried out on two sites: Kvarnbacken, site I, and Sandnäsudden, site II. Surveys in the area indicate several occupations since the Mesolithic period. The sites comprise about 6000 m2 each and dea ling with diagnostic artefacts both sites have been reoccupied. Though not totally investigated they could be delimited and the artefacts were thought to be representative for the investigated areas based on a system of squares that were excavated on regular intervals. The idea is that collected artefacts UHÁHFWWKHVDPHSURSRUWLRQVWKDWZHUHGHSRVLWHGLQSUHKLVWRULFWLPHV$VD result, distribution patterns on these sites show that locally available quartz and porphyry rocks were brought to the sites to be reworked and used there. 2Q WKHRWKHU KDQGUDUHPDWHULDOVOLNHÁLQWMDVSHUDQGVLOLFLÀHG WXIIZHUH found only as fragments and in small numbers (Fig. 4). Raw materials on Kvarnbacken (site I) and Sandnäsudden (site II) 700 600 500 400 Number, site I Number, site II Weight (g), site I Weight (g), site II 300 200 100 Weight (g), site I Number, site I Slate Sandstone Quartzite Quartz Jasper, fine Jasper, coarse Silicified tuff Layered porphyry tuff Porphyry Basic porphyrite Basic volcanic rock 0 Figure 4. Distribution of raw materials on two sites at Fågelsjön, Los parish. The calculations are based on artefact number and weight. Splinters are not included. 357 Per Falkenström 7KHIUDJPHQWDWLRQDQGQXPEHUVRIÀQHJUDLQHGPDWHULDOVSUREDEO\UHÁHFW UHF\FOLQJRIDWOHDVWMDVSHUDQGVLOLFLÀHGWXIIPD\EHXVHGIRUVRFLDOSXUSRVHV such as kinship and alliance negotiations. This would explain the raw mate rial proportions on site II. Perhaps they were regarded as valuable items in exchange networks. Site I differs in this respect, especially if we consider the domination of porphyry and quartz. These were exclusively local material DQGTXLWHHDVLO\DFTXLUHG7KLVVLWHUHÁHFWVDUDWKHUGLIIHUHQWOLWKLFVWUDWHJ\ ZLWK UDZ PDWHULDOVWKDWZHUHFKRVHQ WR ÀWORFDO DQG PRUHIXQFWLRQDO SXU SRVHV3HUKDSVWKHFKRLFHDQGXVHRIWKHVHPDWHULDOVKDGVRPHVLJQLÀFDQFH for those who visited the site, for instance as a way to express group identity (Falkenström & Lindberg in press). Experimentation with different materials might be a regular strategy in the Mesolithic, but this behaviour did not come to an end in the Neolithic. Foraging groups showed similar subsistence patterns throughout the Neo lithic and the Bronze Age. At the same time, technological and social chang es can be observed in northern Sweden by studying bifacial points or daggers made of certain local raw materials, apparently inspired by south Scandina YLDQÁLQWWHFKQRORJ\7KHVHKDYHEHHQFKRVHQIURPTXDUW]LWLFDQGSRUSK\U\ rocks. Other rocks were preferably used for production of knives, scrapers, line sinkers etc. The distribution can thus be followed in close connection to URFNVRXUFHV /DQQHUEUR1RUHOO Knowledge of each rock was required due to different composition and FKDUDFWHULVWLFDODWWULEXWHVRIWKHPDWHULDOV2QHVRUWRIUHGDQGÀQHJUDLQHG quartzite, in particular, available in the Venjan parish, was used to make bifacial points or daggers. Many preforms of this material have been found DQGLQGLFDWHWKDWWKHÀQDOSURGXFWVZHUHGLVWULEXWHGKXQGUHGVRINLORPHWUHV IURPWKHVRXUFHSHUKDSVDVDSDUWRIZHOOHVWDEOLVKHGH[FKDQJHQHWZRUNV Corresponding points of volcanic materials are very rare. Only small points have been found, probably because volcanic rocks show cracks and are al PRVW H[FOXVLYHO\ DYDLODEOH DV PRUDLQH GHSRVLWV /DQQHUEUR1RUHOO SS $OWKRXJKJHRORJLFDOFODVVLÀFDWLRQVPD\EHKHOSIXOWRWKHDQDO\VLVSHWUR JUDSKLFGHWHUPLQDWLRQVPD\LQÁXHQFHDUFKDHRORJLFDOLQWHUSUHWDWLRQV6RPH rocks are very rare and not always recognised in the landscape. These are described according to mineral composition and current geological proc esses. As a consequence, it is a challenge to cope with both archaeological DQGSHWURJUDSKLFDOFODVVLÀFDWLRQV While bifacial technique has been applied to quartzite, slate points have usually been sawed. By using this method, slate points have probably been HDVLHUWRPDNHDOWKRXJKPRUHWLPHFRQVXPLQJ7KLVGRHVQRWQHFHVVDULO\ mean that slate was made by everyone. On the contrary, slate from certain 358 A matter of choice: social implications of raw material availability sources and of certain colours and texture might have been chosen for other than functional reasons. The increased use of quartzitic silt stone and slate in the Neolithic and Bronze Age contexts of northern Sweden indicate changing social relations as well as an accentuated degree of territorial consciousness. The raw mate rial sources could thus be seen as places with restricted access. It could have special meaning to those who used it; perhaps they controlled the quarry. Nevertheless, I see the choice of raw materials mainly as collective expres sions of group identity, sharing similar ideas of how raw materials have been FKRVHQH[WUDFWHGXVHGUHXVHGDQGGLVFDUGHG'LIIHUHQWPDWHULDOVKDYHDOVR been produced, consumed and exchanged within and between groups. Conclusion To conclude, if raw material sources can be found they offer many perspec tives on technology, social organisation and landscape perceptions. Doing so, we have to apply local perspectives in order to understand social repro duction on certain sites. Use contexts of raw materials involve knowledge, skill, rituals, value, exchange etc. The factors are interconnected and should partly explain why certain raw materials have been chosen. Furthermore, common and locally available raw materials are used in an interactive dia logue between people and their environment. The ethnoarchaeological dimension is inevitably connected with a number of methodological problems. Although ethnographic sources are biased in many respects, the information can be used to raise new questions and lead to alternative interpretations. The social perspective in association with ex change and ritual performance are seen as important issues in this respect. In general, during the Mesolithic and Early Neolithic, the use of raw PDWHULDOV ÀW ZHOO LQWR HJDOLWDULDQ VWUXFWXUHV %\ WKH HQG RI WKH 1HROLWKLF distribution patterns show restricted use with attempts to control resources. This would also have consequences on social networks with renegotiations of kinship, alliances, exchange networks, etc. It is possible that some indivi GXDOVZHUHVSHFLDOLVHGPDQXIDFWXUHUVZKRGLVWULEXWHGUHDG\PDGHSURGXFWV 7KLV PLJKW OHDG WR LQFUHDVHG SUHVWLJH DQG VRFLDO VWUDWLÀFDWLRQ ZLWKLQ ORFDO communities. The raw material is thus an important aspect of an operational chain and technological traditions. It plays an active role among people and being so, the life of raw materials is simply a matter of choice. 359 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Per Lekberg Ground stone hammer axes in Sweden: production, life cycles and value perspectives, c. 2350–1700 cal. BC. Abstract 0\VWXGLHVRIVWRQHKDPPHUD[HVIURPWKH6ZHGLVK/DWH1HROLWKLF F 1700 cal BC) have shown two categories; shorter axes with rounded sec tions produced through the pounding of natural cobbles from beaches or ULYHUEHGVDQGORQJHUD[HVZLWKVTXDUHVHFWLRQVSURGXFHGE\WKHÁDNLQJRI quarried stone. The difference in accessibility and controllability of both raw material and technology between these categories makes it possible to XQGHUVWDQGWKHVLPSOHVKDIWKROHD[HGLVWULEXWLRQ²DQGWKXVWKHFRQWH[WVRI the previous studies and clusters of such contexts – from a value or wealth perspective. The accumulation of such value or wealth could then be con trasted against consumption and deposition habits, and studied with the aim to understand the contexts, their topographical setting, and thus create a social topography of the landscape. Introduction The social organization of society of eastern central Sweden in this period of time can be studied in the contextual formation of the landscape and the DFFXPXODWLRQRIZHDOWKDVLVGLVSOD\HGLQWKHGLVSHUVDORIÀQGVDQGPRQX ments that in Swedish research tradition are called Late Neolithic. As I will WRXFKXSRQWRZDUGVWKHHQGRIWKHSDSHUUHVHDUFKZRXOGEHQHÀWIURPKDU monising with European Bronze Age studies, thus considering the period as the Earliest Bronze age of Scandinavia. In fact, that is what I will call it throughout this paper. The archaeological understanding of the period has ORQJ EHHQ GLIÀFXOW IRU DUFKDHRORJLVWV RZLQJ WR D UHODWLYH ODFN RI UHVHDUFK LQDQGEH\RQGWKHÀHOGDQGVHOILQÁLFWHGDFDGHPLFLVRODWLRQ7KLVKDVOLP ited the interpretative potential of the few scattered excavations that have been conducted by leaving them with little or no conceptual framework. Consequently, archaeological studies on this period have suffered from a relative lack of source material and a fragmentary, sporadically mapped cul tural landscape, making it hard to get a grip on the period and to extract some knowledge from its remains. In this unclear situation, however, a few 361 Per Lekberg archaeological materials are potential sources for fruitful research. One of them, and maybe the largest in numbers, consists of ground stone hammer D[HVRUVRFDOOHGVLPSOHVKDIWKROHD[HVGHÀQHGZLWK6FDQGLQDYLDQDUFKDHR logical terminology. $QDEXQGDQFHRIWKHVHVWRQHD[HVKDVDSSHDUHGDVVWUD\ÀQGVLQSORXJKHG soil all over Scandinavia. The idea of this study is that they possibly repre sent a former cultural landscape of graves, votive sites and settlement sites, GHFD\HGWKURXJKPLOOHQQLDDQGÀQDOO\UHQGHUHGLQYLVLEOHE\WKHSORXJK,I contextualized once again, the axes might unveil this hidden cultural land scape and thus open it to research. In Norway and Sweden alone, the ham mer axes number in the tens of thousands (Östmo 1977). The provinces of VWIROGLQVRXWKHDVWHUQ1RUZD\'DOVODQG9lVWHUJ|WODQGLQZHVWHUQFHQWUDO Sweden and the Mälar valley area in eastern central Sweden are the node of the Scandinavian distribution. Following this, the predominant view of re searchers during the second half of the last century has been that the axes are Scandinavian artefacts. Only a very few archaeologists (for example Hagen KDYHFRQVLGHUHGWKHIDFWWKDWWKH\KDYHDSDQ(XURSHDQFRQWHPSRUDU\ Bronze Age distribution from the Aegean via the Balkans, the Black Sea steppes, Bulgaria and Belorussia up to Scandinavia. Many researchers of lat ter decades have come to the conclusion that however abundant in numbers, the hammer axes are too poor in typological elements to make a study worth while. Despite this, the hammer axes have been considered and studied dur ing the last century as one of the key artefacts of the Earliest Bronze Age of Scandinavia. The studies have mainly concentrated on typology for the sake of chronology, an approach that has opened up few paths towards a conceptual framework of society and cultural landscape that could be ar chaeologically fruitful in discussing the social and political structures of the Earliest Bronze Age. In 1957, when Mårten Stenberger stated that research on the hammer axes was an important task for further understanding of the Earliest Bronze Age, researchers had already been trying to get some sense out of these perforated stones for almost a century. My interest in Earliest Bronze Age society, its social structure and its political organisation led me to reconsider this mass material. As mentioned above, I assumed that the axes represented a structured use of the landscape and thus that they might prove vital in understanding this cultural land scape and the society behind it. Consequently, I found it necessary to adopt a different perspective on the study of the axes. This perspective included production, consumption and deposition studies, together with the contex WXDOL]DWLRQRIVWUD\ÀQGVDLPLQJDWXQGHUVWDQGLQJWKHIXUQLVKLQJDQGVRFLDO topography of the landscape. This social topography and the society that it mirrors is understood and described mainly from a Marxist theoretical 362 Ground stone hammer axes in Sweden standpoint as one of institutionalised inequality and accumulation of wealth. An insertion of this society in its chronological, cultural and geographical surroundings in Europe and western Asia indicates interesting results. The diffusion of ideas, knowledge, styles and material culture has been conduct ed within and between elite groups with vast contact networks. Studies of production, consumption and deposition Background and ideas In trying to bring about a contextualization of the axes, and thus the unveil ing of the hidden cultural landscape of Earliest Bronze Age central Sweden, the main idea is the following: If different contexts of deposition – graves, hoards, settlement sites – can be connected with regular, describable and cognitively understandable mor phological differences in the hammer axes, it should be possible to create DQDQDO\WLFDOOLQNEHWZHHQPRUSKRORJ\DQGFRQWH[WWKDWUHÁHFWVSUHKLVWRULF choices in the acts of deposition. Such a link makes it possible to discuss the reason for these choices, i.e. the norms for dealing with material objects, and thus it creates an opening for studying the society. Furthermore, the analyti cal link between morphology and context makes it possible to see and study the sum of these contexts, the cultural landscape, by mapping the different FRQWH[WXDO W\SHV 7KLV UHTXLUHV UHFRQWH[WXDOLVDWLRQ RI VWUD\ ÀQGV E\ PRU SKRORJ\7KLVLQWXUQGHPDQGVWKDWZHHVWDEOLVKVWUD\ÀQGVDVREMHFWVWKDW have once been deposited in contexts according to norms of deposition, but which later on have been robbed of their context by the plough. As I mentioned previously, more than one researcher over the years has considered the morphology of these axes as poor in typological elements. One might say that their morphology could be viewed as homogeneous in its heterogeneity. By this, I mean that the material as a whole has been homoge QHRXVHQRXJKWRPRWLYDWHWKHFODVVLÀFDWLRQRIWKHD[HVLQWRDVLQJOHFDWHJRU\ based upon morphological likeness. However, the material has been consid HUHG VXIÀFLHQWO\ KHWHURJHQHRXV DQG SRRU LQ W\SRORJLFDO HOHPHQWV WR UHMHFW most attempts at meaningful typological differentiation. In my view, this is largely a problem of perspective. If one regards all apparently intact axes as newly manufactured, it is understandable that a possibly inherent differen tiation of the material may remain invisible. Instead, I have chosen to view 363 Per Lekberg the relatively irregular dissimilarities in shape mainly as the results of the deposition of axes in different stages of consumption, and thus taken the bi ography of the objects into account in the archaeological analyses (Kopytoff 1986:64ff). As mentioned previously, I thought it possible that an analytical link between morphology and context could enable a discussion about the norms for deposition and other dealing with material objects, thus facilitat LQJDVWXG\RIWKHVRFLHW\.RS\WRII·VDQWKURSRORJLFDOO\ZHOOIRXQGHGWKHR retical discussion about the change in meaning and value in the biography of an object makes it evident that such an approach can deepen the analyses of WKHVRFLRHFRQRPLFVWUXFWXUHLPEHGGHGLQWKHFXOWXUDOODQGVFDSH Lives and lengths of axes Schematically, I argue that most axes were originally manufactured in rela WLYHO\ODUJHVL]HV²DERXWFP7KHUDZPDWHULDOZDVPLQHGLQVXLWDEOH dikes, probably situated not far from the place of production, or picked up as cobble blanks as suggested by Fenton (1984) for the Scottish material of similar axes. The technology applied in the production of the axes seems to have varied a great deal, depending on the nature of the raw material, and I will return to this more extensively below. Figure 1. Principal drawing of revision phases (”lives”) of an originally 25 cm long hammer axe. The straight dashed lines represent fractures by the shaft holes, the bowshaped dashed lines represent revisions to create new axeheads. Drawing by: Alicja Grenberger. 364 Ground stone hammer axes in Sweden If put to practical use, the originally long axe was then consumed in sequenc es (Fig. 1), any of which can be described as follows. After breaking in two DFURVVWKHVKDIWKROHWKHHGJHSDUWRIWKHD[HZDVUHZRUNHGLQWRDQHZD[HE\ GULOOLQJDQHZVKDIWKROHDQGFUHDWLQJDQHZEXWWE\NQDSSLQJDQGRUSHFN ing (Fig. 2). In most cases, the creation of a new butt completely annihilated WKHUHPQDQWVRIWKHROGVKDIWKROH6RPHWLPHVWKRXJKLWFDQEHVHHQDVD VPDOOSROLVKHGYHUWLFDOJURRYHLQWKHEXWWHQGDQGRUDVDKHDUWVKDSHGEXWW end on an axe that otherwise, to all intents and purposes, looks quite new. The sides of the axe were sometimes also reworked, in order to retain the proportions of the axe. This means that in many cases, all traces of rework ing have been extinguished from the reworked axe, so that it in fact looks like a newly manufactured axe. Repeated polishing of the edge is also a kind of consumption that slowly shortens the axe, thereby affecting the number RISRVVLEOHUHZRUNLQJV Figure 2. (a) Principal drawing of a visibly revised edge half of a hammer axe. The dashed line represents a revision to create a new axehead. Drawing by: Alicja Grenberger. (b) Revised edge half of a hammer axe from the settlement site Lugnet, Österuda parish, Uppland, central Sweden. Drawing by Alicja Grenberger At some point in the series of lives of an axe, it was taken out of practical use and deposited, almost always intentionally and as a result of norms of D[HGHSRVLWLRQLQRQHFRQWH[WRUDQRWKHU:LWKUHIHUHQFHWRWKHSULQFLSOHV presented above, axes could be deposited at any stage of consumption, con sequently being of varying fragmentation statuses and lengths when placed in the depositional context, and thus also when found in our time. This is, I think, the most plausible explanation of the morphological variation seen in the archaeological record of these axes. Indeed it is also a prerequisite for this study. 365 Per Lekberg As mentioned above, the hypothesis of consumption sequences or “lives” of hammer axes renders possible a discussion of the use value contra function in the depositional context and thus about the standards and norms of the depositing society. Also, a contextually based differentiation in the distribu WLRQRIÀQGVFDQEHYLHZHGDVWKHUHVXOWRISUHKLVWRULFFKRLFHVGHSHQGLQJRQ cultural standards in the depositing society. The varied picture of the stray ÀQGVLQRXUPXVHXPDVVHPEODJHVLVWKHUHVXOWRIWKHEULQJLQJWRJHWKHURI objects with different contextual origins, removed from their original depo sitional contexts, most often by ploughing, and thus this contextual differ entiation is hidden. For example, let us for a moment consider the lengths of hammer axes. The length variation in the material is, in my view, a result of the fact that axes were deposited in different stages of consumption. The question is, then, ZKHWKHUWKHVHGHSRVLWLRQVWRRNSODFH D UDQGRPO\ZLWKRXWEHLQJLQÁXHQFHG E\MXGJHPHQWVDERXWWKHOHQJWKRIWKHD[HRU E XQGHUWKHLQÁXHQFHRIDVHW of cultural rules or standards, stipulating what could be deposited where? If alternative (a) is the right one, it is required that the length distribu WLRQ ZLWKLQ D ODUJHHQRXJK JURXS RI KDPPHU D[HV IURP DQ\ JLYHQ (DUOL est Bronze Age, archaeological context category – graves, settlement sites, hoards or whatever – must cohere with the length distribution of the stray ÀQGVLQWKHPXVHXPFROOHFWLRQ)XUWKHUPRUHWKHUHVKRXOGQRWEHDQ\REYL RXV GLIIHUHQFHV LQ OHQJWK GLVWULEXWLRQ EHWZHHQ D[HJURXSV IURP GLIIHUHQW FRQWH[WVLIODUJHHQRXJKPDWHULDOVKDYHEHHQVWXGLHG The prerequisite, on the other hand, for establishing alternative (b) as the right one is that the length distributions of axes from different contexts should not be coherent, either in comparison with each other, or with the OHQJWKGLVWULEXWLRQRIWKHVWUD\ÀQGVLQWKHPXVHXPDVVHPEODJH ,ILWLVSRVVLEOHWRVKRZDFRQWH[WXDOGLIIHUHQWLDWLRQLQWKHVLPSOHVKDIW KROHD[HPDWHULDOLPSO\LQJWKDWWKHGHSRVLWLRQZDVJXLGHGE\FXOWXUDOVWDQG ards, it may be stated that: 1. The axe length (stage of consumption) was an important factor in the prehistoric choice of depositional context for the axe, and there fore 2. The axe length (stage of consumption) might be used as a differenti ating tool in attempting to form contextual hypotheses for distribu WLRQVRIVWUD\ÀQGV 0\ÀUVWDLPPD\WKHQEHH[SUHVVHGDVIROORZV%\LQYHVWLJDWLQJTXDOLWDWLYH similarities and differences between axe groups from different contexts, I 366 Ground stone hammer axes in Sweden ZLOOWU\WRHVWDEOLVKWKHFRQWH[WXDOO\EDVHGGLIIHUHQWLDWLRQRIVWUD\ÀQGVLH ÀQGVWRUQRXWRIWKHLUGHSRVLWLRQDOFRQWH[WV7KHDLPRIWKLVLVWRXQGHUVWDQG the patterns of production, consumption and deposition of the hammer axes and thus to create a tool for forming hypotheses about the organisation of the Earliest Bronze Age cultural landscape. Context studies I have conducted three studies of context populations, treating axes found in graves, axes found in hoard contexts and axes found on settlement sites. First, 55 hammer axes found in Earliest Bronze Age graves in Sweden (Seg HUEHUJ ZHUHFRPSDUHGZLWKD[HV VWUD\ÀQGV IURPWKHFHQ tral Swedish province of Uppland. The purpose of the study was to decide whether the distributions of sets of qualitative variables in the two materials deviated from or cohered with one another and to draw preliminary conclu sions about the fruitfulness of pursuing this path of investigation. The result was positive, since distinct differences were noted between the two distri EXWLRQV$IWHUWKLVWKHJUDYHFRQWH[WD[HVZHUHFRPSDUHGZLWKD[HV from hoards or votive contexts in the southern Swedish province of Scania (Karsten 1994). )LQDOO\WKHJUDYHDQGKRDUGÀQGVZHUHFRPSDUHGZLWKD[HVDQGGULOO cores found in excavated settlement sites from all over Sweden. The variables under discussion when comparing axes were the axe length and the damage to the axes. A prerequisite for the study was the presumption that all damage, DWOHDVWRQWKHFRQWH[WDIÀOLDWHGD[HVZDVLQÁLFWHGEHIRUHGHSRVLWLRQ,QRUGHU to compare the lengths of the axes, it has been necessary to include only axes that are not obviously damaged in a way that has affected their lengths. Axe length is to be understood as the distance, measured in centimetres, between WKHEXWWDQGWKHFXWWLQJHGJHRIWKHD[H)RUW\QLQHRIWKHD[HVIURPKRDUG contexts that meet this requirement have a length span of 9.7 cm to 30 cm, with a medium span of 17.9 cm. For 33 of the 55 axes from grave contexts, WKHVHÀJXUHVDUHFPWRFPZLWKDPHGLXPOHQJWKRIFP ,WVKRXOGEHQRWHGWKDWWKHÀQGVIURPERWKWKHKRDUGDQGWKHJUDYHFRQ texts tend to differ in length distribution, not only from each other but also LQFRPSDULVRQZLWKWKHVWUD\ÀQGV)XUWKHUPRUHWKHJUDYHFRQWH[WD[HVVHHP to be concentrated around the shorter intervals, while the hoard axes, on the other hand, have their pivot of distribution in the longer intervals. The really VKRUWD[HV VKRUWHUWKDQFP DUHODFNLQJLQWKHKRDUGFRQWH[WVZKLOVWWKHJUDYHFRQWH[WV ODFNUHDOO\ORQJD[HV ORQJHUWKDQFP). Moreover, ten of the undamaged axes from hoard contexts (20%) are axe preforms, while these objects are totally 367 Per Lekberg lacking in the grave contexts. There are 27 axe preforms, or about 8% of the total of 358 undamaged axes in the stray axes from Uppland. The term ´SUHIRUPµLVXVHGKHUHWRGHÀQHDQLQFRPSOHWHD[HWKDWFDQEHLGHQWLÀHGDV an Earliest Bronze Age hammer axe from its general shape but lacks, for LQVWDQFHDGULOOHGWKURXJKVKDIWKROH2QO\QLQH RIWKHD[HVIURP hoard contexts are damaged, while 22 (40%) of the 55 axes from graves are damaged. On the studied settlement sites, there are no undamaged axes, only fragments and preforms (Figs. 3, 4 and 5). There are altogether six preforms in the set WOHPHQWVLWHPDWHULDODQGÀYHRIWKHVHFRPHIURPRQHDQGWKHVDPHH[FDYD tion. Most of these preforms are comparatively short. )LJXUH7\SLFDOÀQGVIURPVHWWOHPHQWVLWHV3KRWRRIIUDJPHQWDU\KDPPHUD[HVIURP the settlement site Gröndal, Lunda and Husby-Ärlinghundra parishes, Uppland, central Sweden. With the courtesy of RAÄ/Uv Mitt. Figure 4. Photo of drill cores – results of the process of shaft-hole drilling with a pipe drill - from the settlement site Gröndal, Lunda and Husby-Ärlinghundra parishes, Uppland, central Sweden. With the courtesy of RAÄ/Uv Mitt. 368 Ground stone hammer axes in Sweden Figure 5. Photo of cobble axe preform, left before completion of the shaft-hole drilling, from the settlement site Gröndal, Lunda and Husby-Ärlinghundra parishes, Uppland, central Sweden. With the courtesy of RAÄ/Uv Mitt. In my view, the results presented above clearly show that there is a contextu ally explainable differentiation in length and quality, inherent in the simple VKDIWKROHD[HPDWHULDO&RPSDULVRQVRIWKHGLVWULEXWLRQRIOHQJWKDQGIUDJ mentation categories between the two analytical categories of stray found D[HVDQGFRQWH[WIRXQGD[HVVKRZWKDWWKH\PRVWOLNHO\UHSUHVHQWWKHVDPH structures of action, i.e. the formation processes for the two population are VRVLPLODULWLVKDUGWRGLVWLQJXLVKEHWZHHQWKHP )LJDE +RZHYHUZKHQ GLVVLPLODWLQJ WKH FRQWH[WIRXQG D[HV LQWR FRQWH[W FDWHJRULHV WKLV OLNHQHVV GLVDSSHDUV,QVWHDGWKHGLVWULEXWLRQVKRZVDFOHDUFRQWH[WGHSHQGDQWPRU phological differentiation between the axes from graves, hoards and settle ment sites (tab. 1; Fig. 7 and 8). Intact Table 1a. Distribution within different consumption-phase categories of all context-found axes. Table 1b. Distribution in percentage within consumption-phase categories of all context-found axes. Fragments Preforms Settlements 0 54 6 Sum 60 Graves 45 9 0 54 Hoards 35 2 10 47 Sum 80 65 16 161 Fragments Preforms 90% 10% Intact Settlements Sum 100% Graves 83% 17% Hoards 74% 5% 21% 100% Sum 50% 40% 10% 100% 369 100% Per Lekberg Figure 6a. Distribution into fragmentation categories of 1359 stray-found hammer axes from central Sweden (to the left; “lösfynd”) and 167 context-found hammer axes from southern and central Sweden (to the right:“kontextfynd”). 100 75 50 25 F H Ä 0 Lösfynd Kontextfynd 100 75 50 25 0 >17 cm 13-17 cm <13 cm Figure 6b. Distribution into length categories of 645 strayfound undamaged hammer axes from central Sweden (to the left:“lösfynd Up/Sö/Nä”) and the context-found undamaged hammer axes from southern and central Sweden (to the right:“kontextfynd”). Figure 7. Distribution in percentage of fragments (“fragment”), undamaged (“hela”) and blanks (“ämnen”) within context categories ´ERSODWVHUµ VHWWOHPHQW VLWHV´JUDYDUµ JUDYHV ´GHSnHUµ KRDUGV RIDOOFRQtext-found axes. Figure 8. Distribution in percentage of fragments (“fragment”), short undamaged axes/ blanks (“H/Ä<13cm”) and long undamaged axes/blanks (“H/ Ä>13cm”) within context cateJRULHV ´ERSODWVHUµ VHWWOHPHQW VLWHV´JUDYDUµ JUDYHV ´GHSnHUµ KRDUGV RIDOOFRQtext found axes. 370 Ground stone hammer axes in Sweden Thus, it is likely that Earliest Bronze Age Scandinavians were guided by norms delimiting the size and shape of axes suitable for deposition in dif IHUHQWFRQWH[WV5HFDOOLQJZKDW,SUHYLRXVO\VWDWHGDERXWWKHXVHOLYHVFRQ sumption and functional value of axes (see above), it may be reasonable to UHJDUGJUDYHD[HVDVKDYLQJEHHQGHSRVLWHGLQWKHLUODVWOLIHPRVWKRDUGD[HV LQRUEHIRUHWKHLUÀUVWOLIHDQGVHWWOHPHQWD[HVPRVWO\DIWHUWKHLUXVHOLYHV,Q DSHUVSHFWLYHRISRWHQWLDOSUDFWLFDOXVHYDOXHKRDUGD[HVPD\EHYLHZHGDV PRUHYDOXDEOHREMHFWVWKDQJUDYHD[HV)XUWKHUPRUHLWLVUHDVRQDEOHWRFRQ clude that the results show a tendency that can be used in forming contextual K\SRWKHVHVIRUVWUD\ÀQGVQDPHO\WKDWRYHUFPORQJD[HVDQGD[HSUH forms represent hoard contexts, whilst axes shorter than 13 cm commonly represent graves, and axe fragments and short preforms represent settlement VLWHV 6LPSOH VWXGLHV RI PRUSKRORJLFDO YDULDWLRQ LQ FRQWH[WDWWULEXWHG D[HV have hence been the basis for creating something that can be called a contex tual typology. This tool for contextualization has enabled the transformation RIORFDWLRQVRIVWUD\ÀQGVLQWRK\SRWKHWLFDOSODFHVIRUGHSRVLWLRQDOFRQWH[WV thus making a contextual furnishing of the landscape possible. Production, value and wealth The studies related above have dealt with the morphology as a product of choices and norms related to depositions of different kinds. However, the re sults of these studies have limited potential for analyses of the accumulation of wealth, and thus for aiding in the discussion of economic and political power (Earle 1997), if no acceptable line of argumentation can link different shapes of axes or even different kinds of contexts with higher or lower value in any evident way. Aiming at such a line of argumentation, and as a result of observing indications of different modes of production, I have focused on the production of hammer axes. An observation of production differences in the material, where some D[HVKDGREYLRXVO\EHHQSURGXFHGE\ÁDNLQJWHFKQLTXHZKLOHRWKHUVVKRZHG no signs of this, gave rise to the question if different kinds of raw material are more or less suited for the one or the other mode of production. An analysis of the raw material of a large number of hammer axes from four provinces in central Sweden showed, not surprisingly, that most of them were made of diabase (Kresten 1998). This kind of rock has been sought for and used in different cultures around the world for axe making (Olaus son 1983; Bruen Olsen & Alsaker 1984; McBryde 1984; Pitts 1996; Le Roux 1998). Consequently, no direct analytical ties could be established between signs of the mode of production, nor any difference as to what kind of rock to choose as raw material for the axe. 371 Per Lekberg Instead I looked more closely at the relationship between the production WHFKQRORJ\WKHGLIIHUHQWVRXUFHVRIUDZPDWHULDODQGWKHFURVVVHFWLRQRIWKH ÀQLVKHGREMHFW)HQWRQ DQGSDUWO\DOVR2ODXVVRQ EHOLH ves in secondary provenience for the raw material of most ground stone axes, considering that cobbles and loose rocks are relatively easy to obtain. This view has also dominated French research, but Pierre Petrequin and his col leagues sees this explanation as being too simple (Petrequin et al. 1998:282). 7KHLUÀHOGZRUNLQ,ULDQ-D\DKDVVKRZQWKDWRQO\D[HVQRWPHDQWWREHRE jects of value were made of cobbles. The reasons are said to be that cobbles from riverbeds etc. often were too dry for being suitable for knapping, and also that not enough raw material for an economically acceptable quality production could quickly be obtained. This division of the axe production – into household production and the making of valuables – is thus in Irian -D\DFRQQHFWHGZLWKDGLYHUVLÀHGSURFXUHPHQWDQGXVHRIUDZPDWHULDOV,Q the household production, preforms have been made out of small or me GLXPVL]H FREEOH EODQNV ZLWK VLPSOHWROHDUQ WHFKQRORJ\ 7KH SURGXFWLRQ of the large, valuable axes requires, on the other hand, often mining of out FURSVRUUHJXODUGLNHVDQGSDUWWLPHVSHFLDOLVWVZLWKUHODWLYHO\KLJKWHFKQR ORJLFDOVNLOOVWUDQVIRUPLQJWKHDQJXODUEODQNVLQWRVTXDUHVHFWLRQSUHIRUPV and axes (Fenton 1984:237; Petrequin et al. 1998, Fig.3). In some societies in Irian Jaya and New Guinea, the axe length and the rock it is made of are FRQQHFWHGWRWKHGLVSOD\RIVWDWXVRIWKHFDUULHURQO\KLJKTXDOLW\SUHIRUPV over 20 cm in length are considered valuable enough for status and prestige transactions (Petrequin et al. 1998:294ff; Burton 1984:244). These studies give example of societies where technological differences in the production of axes can be connected to a differentiated raw material procurement, on the one hand, and to value differences on the other. Our view of the social complexity of a society is highly dependant upon WKHZD\ZHVHHLWVWHFKQRHFRQRPLFVLWXDWLRQ7KHVWRQHD[HUDZPDWHULDO that Fenton (1984) describes can be found just about everywhere in the near neighbourhood of Scottish Neolithic and Bronze Age man, the procure ment of preforms is fast and easy, the raw material is hard to control and the production requires no advanced technological skills. On the other hand, McBryde (1984:268ff) shows how two neighbouring Australian groups co operate in working and controlling a certain dike known for its good raw material, a situation somewhat like the Tungei in New Guinea who display DQ LQWULEH FRRSHUDWLYH ZD\ RI ZRUNLQJ WKH GLNH DQG GLVWULEXWLQJ WKH SUH forms (Burton 1984:235ff). Also, at Plussulien in Bretagne, France, it has been estimated that several million tools have been made, equalling twelve people making about ten tools a day for about 2000 years of use (Le Roux DQGDW3ODQFKHUOHV0LQHVLQWKHHDVWRI)UDQFHWKHHVWLPDWHLV 372 Ground stone hammer axes in Sweden PDQGD\V·ZRUNSHU\HDUIRUWKH\HDUVRIXVHDFFHSWLQJFRQWLQXRXVXVH (Petrequin et al. 1998:304ff). The foremost purpose for mining, as opposed to collecting the raw mate rial, is wherever you choose to direct your ethnographic eyes, getting socially potent prestige objects. The value in such an object is not primarily func tional. Brian Hayden maintains that The purpose of creating prestige artifacts is not to perform a practical task, but to display wealth, success, and power. The purpose is to solve a social problem or accomplish a social task such as attracting productive mates, labor, and allies or bonding members of social groups together via displays of success”. Thus, he sees the logic and the strategies, i.e. the technology, connected to the production of prestige objects as being completely different than in the production of practical objects. Easily accessible raw material, low de mands on work intensity, organisation, technological knowledge and practi FDONQRZKRZDUHGLUHFWO\QHJDWLYHDQGGHVWUXFWLYHWRWKHYDOXHRIDSUHVWLJH WHFKQRORJ\%HFDXVHRIWKHGULYHWRFRS\KLJKVWDWXVEHKDYLRXUDQGSUHVWLJH objects, the raw material procurement for, and production of these objects PXVWWKXVEHFRQQHFWHGZLWKVXFKGLIÀFXOWLHVWKDWUHDVRQDEO\IHZFDQKDYH access to them (Hayden 1998:11f, 41). Table 2. Min-, mean- and maxlengths for 122 hammer axes with squared-off vs. rounded cross sections. Cross section MIN (cm) MEAN (cm) MAX (cm) Squared-oﬀ 4,7 13,8 25,5 Rounded 7,1 11,6 17,5 My results show that there is a difference in the length distribution of axes with rounded vs. square cross sections (table 2). In a studied material from Uppland, consisting of 122 hammer axes, the mean length for rounded sec tioned axes is 11.6 cm; no axe is longer than 17.5 cm, while square section axes means 13.8 cm and maxes 25.5 cm. Thus, length and cross section can not be said to be independent of each other. Instead, there is a clear tendency that long axes (longer than 18 cm) always have a square cross section and small axes (smaller than 12 cm) always have a rounded cross section. What could then be the reason for mining some of the raw material? Especially ZKHQLWLVFOHDUWKDWLWWDNHVORQJHUSHUÀQLVKHGD[HLWLVPRUHULVN\DQGUH quires a higher extent of organisation of labour mining raw material than collecting it. My view is that the long hammer axes with square sections ÀUVWDQGIRUHPRVWDUHREMHFWVRIYDOXHHYHQLIWKH\DOVRDUHSUDFWLFDOO\XVH able. In this case they are results of a prestige technology with the purpose 373 Per Lekberg of producing an abundance of objects for socially important transactions of different kinds. The existence of such axes in hoards or votive deposits strengthens the probability that they are objects of value. The ones that have not ended up in hoards or votive deposits have been used practically (and RI FRXUVH DV SUHVWLJH REMHFWV DW WKH VDPH WLPH DQG FRQVXPHG VKDIWKROH E\VKDIWKROH7KHVKRUWWRPLGGOHOHQJWKKDPPHUD[HVZLWKURXQGHGVHF tions, on the other hand, should be viewed as copies; produced locally out of cobbles that have been collected and pounded/pecked into shape and used mainly as household tools. But is the technology required for producing a long hammer axe really complicated enough for it to be described as a prestige technology? I have not made experiments of my own yet, but from the examples of such work I have read, and indeed referred to in this paper (for example Petrequin HWDO.DUVHWDO ,ZRXOGVD\WKDWLWLVGHÀQLWHO\KDUGHQRXJK Altogether, the working time, moment of risk and organisation of labour UHTXLUHGIRUÀQGLQJWKHULJKWUDZPDWHULDOPLQLQJLWDQGSURGXFLQJÁDZOHVV 30 cm long hammer axes or preforms for them should be at least as dif ÀFXOWDVWKHSURGXFWLRQRIWKHSUHVWLJHD[HVPHQWLRQHGLQWKH(XURSHDQRU 2FHDQLFH[DPSOHV´.QDSSHGEODGHVFPORQJRUPRUHQHHGDKLJKOHYHO NQRZKRZDQGFRXOGKDYHEHHQWKHUHVSRQVLELOLW\RIVSHFLDOLVWV«µ 3HWUH quin et al.1998:306). That is probably the reason why the Tungei tribe only FRQVLGHUHGD[HEODGHVRYHUFPORQJDVEHLQJYDOXDEOHHQRXJKIRUEULGH SULFHV DQG RQH UHDVRQ ZK\ FRPPRQ ZRUND[HV ZHUH VKRUWHU WKDQ FP (Burton 1984:244). 0RUHRYHU WKH GLVFXVVHG D[HV IURP )UDQFH DQG 1HZ *XLQHD DUH D[H EODGHVDQGWKXVODFNWKHVKDIWKROHRIWKHKDPPHUD[HV7KHVSHFLDOWLPH SHUVSHFWLYHV GLIÀFXOWLHV DQG KDUGVKLSV RI VKDIWKROH GULOOLQJ VKRXOG WKHQ also be taken into account. As far as one can see, the drilling was a time consuming phase of the production. Experiments conducted by the Dan ish amateur archaeologist N. F. B. Sehested at the end of the 19th centu ry showed what a delicate process it can be. Too vigorous drilling and not HQRXJKZDWHULQWKHJULQGLQJPHGLXPVHWWKHGULOOLQJPDFKLQHU\RQÀUHDQG created tiny heat cracks in the stone, rendering it useless. Too much water, on the other hand, meant that much more force had to be used for keeping the drilling speed reasonably high, and the excess force weakened the stone. $W 6HKHVWHG·V H[SHULPHQWV WKH GULOOLQJ WLPH SHU FP ZDV EHWZHHQ DQG 150 minutes, depending on type of stone, drilling apparatus, drill bit ma terial and grinding medium (Sehested 1884:26ff). Other experiments have shown that it takes about two hours per millimetre drilling in diorite, even if the speed is slightly higher with a pipe drill than with a massive drill (Rieth 1958:108; Fenton 1984:227). Also, the axe preform is weakened during and 374 Ground stone hammer axes in Sweden by the drilling, which probably constituted yet another threat to the produc WLRQRIORQJD[HVVWUHWFKLQJWKHWLPHVSDQDQGZRUNDPRXQWSHUÀQLVKHGD[H even more. This might be the reason for so many hoarded preforms exhibit LQJDSDUWO\ÀQLVKHGVKDIWKROH6KRZLQJLQWKLVZD\WKDWWKHSUHIRUPGLG QRWKDYHPLFURFUDFNVWKDWZRXOGUXLQLWLQGULOOLQJZRXOGFHUWDLQO\KDYH enhanced the value of the preform even more. In conclusion, a reasonable interpretation of these results is in my view that the raw material for axes with square sections derive from primary sources, while the raw material for axes with rounded sections is mainly cobbles picked from ridges, beaches, streams and riverbeds. The study of the relationship between raw material and technology, and the social and societal consequences of this relationship, on which this paper is founded is theoretically based upon and inspired by research by N. F. B. Se hested (1884); A. Rieth (1958); A. Laitikari (1928); M. B. Fenton (1984); J. Apel (2001); B. Hayden (1998); D. Olausson (1983, 1998 and 2000); E. Weiler (1994 DQG &7/H5RX[ -%XUWRQ 33HWUHTXLQHWDO / Sundström & J. Apel (1998); E. Callahan (1987); L. Kaelas (1959); G. Sarauw -$OLQ ,0F%U\GH /((QJOXQG -7DIÀQGHU c+\HQVWUDQG DQG($.DUVHWDO 7KHVWXG\KDVVKRZQWKDW there is a great divide resulting in two axe categories: those probably pro duced through the pounding of natural cobbles from beaches or riverbeds into shorter axes with rounded sections on the one hand, and on the other hand the longer axes with square sections, produced from preforms created E\WKHÁDNLQJRITXDUULHGVWRQH )LJDQG 7KHFOHDUGLIIHUHQFHLQDFFHV sibility and controllability of both raw material and technology between these categories has made it possible to understand the hammer axe distribution – and thus the contexts of the previous studies (as discussed above) and clusters of such contexts (as discussed below) – from a value or wealth perspective. The accumulation of such value, or indeed wealth, could then be contrasted against consumption and deposition habits (as displayed above) and studied with the aim set at understanding the contexts, their topographical setting and thus at creating a social topography of the landscape. Figure 9. Distribution of length within the groups of rounded (to the left: “rundade”) and square cross sections. 30 25 20 15 10 5 0 Rundade Kantiga 375 Per Lekberg Figure 10. A comparison between the share of axes with rounded (left) or square crosssections which are longer or shorter than the mean length for the entire analysed material. 100 75 50 25 4,7-12,7 cm >12,7 cm 0 Rundade Kantiga Value and wealth are crucial terms in discussing the political economy of a society (Earle 1997). The analytical link for connecting different raw materi als and different modes of production with differentiation in value has been established above. Connecting square sections with contexts of deposition in simple statistics shows that value has been deposited unevenly (Fig. 11) – with the most value in hoards, to a lesser extent in graves and to a yet lower GHJUHHRQVHWWOHPHQWVLWHV,QP\RSLQLRQWKLVUHÁHFWVDKDELWUHSUHVHQWLQJ a cultural standard regulating the handling and deposition of wealth in the society. Figure 11. Percentage of axes with rounded (“rundat”) vs. squared-off cross section on the different context groups ´ERSODWVHUµ VHWWOHPHQW VLWHV´JUDYDUµ JUDYHV ´GHSnHUµ KRDUGV 5HWXUQLQJRQFHDJDLQWRWKHLGHDRIXVHOLYHVRIWKHKDPPHUD[HVSUHVHQWHG above, this too can be studied from the value perspective within the theo UHWLFDOIUDPHZRUNRI,JRU.RS\WRII·VUHVHDUFKRQWKHELRJUDSKLHVRIPDWHULDO things (Kopytoff 1986). Here it is demonstrated that material culture, when used/consumed, can pass through stages of deterioration equalizing a change in value. This means something socially for the people using or even looking at the object. If they are familiar with the practical and cultural meaning of the object – how it can and cannot be used in the different stages or “lives” it passes through – they have biographical expectations on the object. As Helle Vandkilde has pointed out, there is a strong connection between the value of 376 Ground stone hammer axes in Sweden an object and its “exchange rate”. Objects that are interchangeable are usu ally of lower value than objects that are acceptable as gifts (Vandkilde 2000). Thus, what people do with objects can always be compared with the cultural norm for such dealings – it says something about a person if she deviates IURPWKLVQRUP7KLV²WKHOLQNEHWZHHQREMHFWFRQWH[WYDOXHQRUPLVDQ important point to reach when using archaeology in trying to understand something about the social or political economy of a society. If the norm can be established, so can the deviation, and the rate of deviation from the norm RIDVLQWKLVFDVHZHDOWKGHSRVLWLRQPLJKWEHWDNHQWRUHÁHFWDVLPLODUGLI ferentiation in ability to act according to this norm, and in turn connected with an inequality in the distribution of wealth as such in the society. Returning to the axes and their contexts discussed above, the distribution RIFRQVXPSWLRQSKDVHFDWHJRULHV SUHIRUPVFRPSOHWHD[HVDQGIUDJPHQWV has been shown to be uneven when comparing settlement sites, graves and hoards (Fig. 7 and tab. 1). The settlement sites include no complete axes, mostly fragments and a few preforms. The graves have mostly complete axes, and a few fragmented ones. The hoards display mostly complete axes, extremely few fragments and the highest percentage of preforms. Looking at lengths of complete axes and preforms (Fig. 8) it is obvious that the preforms of the settlement sites are short, the complete axes of graves are mostly short, and the complete axes and preforms of hoards are mostly long. Transformed LQWR XVHOLYHV DQG YDOXH LQ OHQJWK WKH QHZHVW DQG PRVW YDOXDEOH D[HV DUH deposited in hoards, while the oldest and least valuable objects are deposited on settlement sites – mainly as rubbish. The short, rounded preforms found on settlement sites are reworked cobbles, left for one reason or another. The axes that are deposited in graves are mostly short and display a rounded cross VHFWLRQZKLFKVKRZVWKDWWKH\DUHGHSRVLWHGVRPHZKHUHEHWZHHQPLGOLIH DQGWKHHQGRIWKHLUDUUD\RISRWHQWLDOUHZRUNLQJV To conclude, it is obvious that deposition of value is uneven between hoards, graves and settlement sites as context categories, and within these context categories. 7KHÀUVWVWDWHPHQWUHÁHFWVDGLIIHUHQFHLQQRUPVIRUGHSRVLWLRQRIYDOX DEOHVZKLOHWKHVHFRQGVWDWHPHQWUHÁHFWVWKHDELOLW\ZLWKLQVRFLHW\WRDGKHUH to this norm, i.e. the access to valuables. Like Helle Vandkilde (1996:316), I YLHZWKHKRDUGVRUYRWLYHRIIHULQJVDVWKHVRFLDOJURXSV·FROOHFWHGJLIWVWR higher entities or gods, with the function of marker and producer of pres tige and social position in relation to other groups. This means that if the norms and standards for deposition of valuables stipulated that the high est value be hoarded in bogs, for one reason or another, the members of society should do their best in that respect, depositing their most valued or expensive axe for the sake of the collective. The result of this would be that 377 Per Lekberg DQXQHYHQYDOXHGHSRVLWLRQZLWKLQKRDUGVZRXOGLQGHHGUHÁHFWDQXQHTXDO distribution of wealth within the social group. From a Marxist perspective, the whole collective hoarding or offering practice can be seen as a strategy for masking inequality, typical of a power structure of an institutionalised hi erarchy (Nordqvist 2001:60). Agreeing with this perspective, and accepting the results of the study of deposition and production, I would state that the distribution of value within votive hoards reveals a cross section of wealth distribution pattern of the society under study. Moving over to depositions in graves, I continue to agree with Vandkilde (1996:316) when she holds graves to be more individually oriented – grave JLIWVH[SUHVVLQJPRUHRIWKHPRXUQHUVKRQRXULQJWKHGHFHDVHGSHUVRQ·VVR cial position in life. However, if we accept the discussion about votive hoards above, the grave goods should be taken as being the next best – the objects that people can afford to lose after offering their most valuable pieces in the bog. In this perspective, the shortness of grave axes makes sense, since REMHFWVZLWKDIHZPRUHXVHOLYHVZRXOGKDYHEHHQQHHGHGIRUSUDFWLFDOSXU SRVHV 7KH YDOXH GLVWULEXWLRQ VHHQ LQ WKH FURVVVHFWLRQ RI WKH JUDYH D[HV VKRXOG UHÁHFW WKH SHUFHQWDJH RI GHFHDVHG ZLWK UHODWLYHV ZKR FRXOG DIIRUG grave gifts of higher or lower value. Votive offerings and graves can thus be viewed as manifested ideologies of the collective and the individual, but as ideological manifestations they ODFNUHÁHFWLRQVRIWKHVRFLDOUHDOLWLHVRISHRSOH VHH1RUGTYLVW 6HW WOHPHQW VLWHV FDQ LQFOXGH ERWK LGHRORJ\ DQG VRFLDO UHDOLW\ EXW WKH VSHFLÀF meaning of studying settlement sites in comparison with votive offerings and graves is the possibility of contrasting socially real with the culturally ideal. The distribution of value within the settlement sphere (that is: within and/or between farmsteads and/or villages) shows what could be afforded WREHXVHGSUDFWLFDOO\FRQVXPHGWRWKHHQGRIXVHOLYHVDQGWKHQGHSRV ited as rubbish. This means that when it comes to the distribution of value, the axes from settlement sites reveal to a higher degree the factual wealth distribution – that is (in generalised terms): the percentage of rich and poor farmsteads/villages. To sum up, the production and consumption traces within the context categories result in the following interpretation: 9RWLYHRIIHULQJV²UHÁHFWGLIIHUHQFHVLQZHDOWKDQGVRFLDOVWDWXVEHWZHHQ JURXSVJLYHDFURVVVHFWLRQRIWKHFDSLWDOGLVWULEXWLRQSURÀOHLQVRFLHW\ *UDYHV²UHÁHFWZKDW\RXFDQDIIRUGDIWHURIIHULQJWKHPRVWYDOXDEOHJLYH an estimate of the percentage of individuals with families that can afford richer or poorer grave gifts. 6HWWOHPHQWVLWHV²UHÁHFWWKHIDFWXDOGLVWULEXWLRQRIYDOXHEHWZHHQVHWWOH ment units; give an estimate of the percentage of rich and poor farmsteads/ villages. 378 Ground stone hammer axes in Sweden Looking at the axe materials attributed to context categories in this per spective (Fig. 11 and 12) one could thus express the economic power relations in the Earliest Bronze Age in southern and central Sweden: One fourth (1/4) of the settlement units are inhabited by the third (1/3) of the population that controls two thirds (2/3) of the wealth. This means that the settlement units of the economical elite should be relatively large – with large and/or many houses – settlement clusters and the forming of villages. Figure 12. The distribution of wealth expressed in axes with squared-off cross secWLRQV ´(QKHWHUµ VHWWOHPHQW XQLWV´,QGLYLGHUµ LQGLYLGXDOV ´.DSLWDOµ FDSLWDO Landscape furnishing 7KHDQDO\WLFDOWRROVSUHYLRXVO\IRUJHGDVEULHÁ\GHVFULEHGDERYHDUHLQLWLDOO\ put into operation in the mapping of a contextual and social landscape in east ern central Sweden. Subsequently, in a discussion of the cultural landscape WKXV GHÀQHG DQG E\ UHIHUULQJ WR VRFLDO WKHRU\ ODQGVFDSH WKHRU\ 0DU[LVW theory) relating the landscape to the social and political organisation of the inhabitants of that landscape, I try to explore some of the Earliest Bronze Age forms of social integration (Lindblom 1991:135; Gansum, Jerpåsen & Keller 1997:18; Appadurai 1986:38; Welinder 1992:45; Bender 1993:3; Hod der 1982; Küchler 1993:85ff; Johnston 1998:54; Knapp & Ashmore 1999:1ff; Schama 1995:35ff, 82f; Tilley 1993:81, 1994:19ff; Donham 1995:53ff, 193f; Bradley 2000:26ff, 152ff). The aim is to understand the landscape, furnish ing enough to enable a fruitful discussion about the societal context of such landscapes. The aim is set on understanding more about the social organi zation in eastern central Sweden during its Earliest Bronze Age. Two areas have been studied, one around lake Hjälmaren in central Sweden and one IXUWKHUWRWKHQRUWKHDVWLQ8SSODQG 379 Per Lekberg Figure 13. The study area around lake Hjälmaren, with an image of a landscape of the Earliest Central Swedish Bronze Age landscape, created through an interpolation of the distribution of hammer axes of different sizes. The interpolation is universal with linear drift.The legend shows, from top to bottom: Settlement areas (axe fragments); Short axes; Long axes; Many long axes. Map by Karl-Johan Lindholm. Figure 14. The study area around lake Hjälmaren, with an image of the distribution of wealth within a landscape of the Earliest Central Swedish Bronze Age, created through an interpolation of the distribution of hammer axes with rounded vs. squared-off cross sections. The interpolation is universal with linear drift. The legend shows, from top to bottom: Settlement areas (axe fragments); Axes with rounded cross sections; Axes with squared-off cross sections; Many axes with squared-off cross sections. Map by Karl-Johan Lindholm. 380 Ground stone hammer axes in Sweden The result of the mapping is a landscape structure, exhibiting a contextu DOO\GLYHUVLÀHGXVHRIODQGVFDSH]RQHVDQGODQGVFDSHHOHPHQWV7KHVKRUW JUDYHLQGLFDWLQJD[HVJHQHUDOO\H[KLELWOLQHDUGLVWULEXWLRQSDWWHUQVPDUNLQJ paths and roads along ridges, eskers or waterways, while the long, hoard/vo WLYHLQGLFDWLQJD[HVDUHPRVWRIWHQSODFHGLQWKHFRDVWDO]RQHRUDWVSHFLDO places along the inland paths, marking nodal points in the landscape, such as fords, harbours and other places of liminal transgression (Fig. 13 and 14). According to the distribution of axe fragments, the settlement sites are lo cated in generally quite evenly distributed clusters that increase in density in certain areas (Fig. 15). They seem to be equivalent to settlement spaces, i.e. ¶ODQGVFDSHURRPV·WKDWSRWHQWLDOO\IRUPSROLWLFDOO\LQWHUSUHWDEOHVHWWOHPHQW XQLWVRIWKH(DUOLHVW%URQ]H$JHODQGVFDSH )LJ *UDYHPDUNHGSDWKV DQGURDGVZLWKWKHLUKRO\ULYHUFURVVLQJVZHOOVDQGZDWHUKROHVOHGWRDQG from these settlement spaces. Furthermore, notable spatial differences in the distribution of valuables between the settlement areas are detected, indicat ing economic and political relations between centre and periphery and an unequal accumulation of wealth (Fig. 17; see also Lekberg 2002:273). Figure 15. The study area around lake Hjälmaren, with an image of settlement areas within a landscape of the Earliest Central Swedish Bronze Age, created through an interpolation of the distribution of fragmented hammer axes (dots). The interpolation is universal with linear drift. Map by Karl-Johan Lindholm. 381 Per Lekberg Figure 16. The study area around lake Hjälmaren, with site catchment areas of 10 km in diameter, around the hypothetical settlement areas. Map by Karl-Johan Lindholm. 4ECKENFÚRKLARING Figure 17. The study area around lake Hjälmaren, with the relative distribution of wealth, in the form of hammer axes with squared-off cross sections, within each site catchment area presented as a pie chart. More white colour in the chart equals relatively more axes with squared-off cross sections. Note that the absolute number in each pie chart may vary between 1 to c. 25. Map by Karl-Johan Lindholm. The core areas of this accumulation of wealth coincide geographically with areas of higher density in the distribution of the axes that indicate settle ment sites and thus theoretically a denser settlement structure and a larger population per square unit of area. Earlier mappings of the distribution of Middle Neolithic to Bronze Age artefacts in eastern central Sweden (Jensen 382 Ground stone hammer axes in Sweden 1989:115ff), amongst them Earliest Bronze Age hammer axes, show concen trations every 25 km or so along the coastal region. These concentrations ex hibit a considerable spatial continuity, since they are detectable at least from the Middle Neolithic and well into the Bronze Age. It is thus reasonable to believe that they indeed represent settlement areas in the form of political entities of some kind during a long period of their existence. The northern most of these concentrations coincides with my test area in Uppland. Here, the accumulation of wealth during the Earliest Bronze Age took SODFH LQ WKH IRUP RI ERWK ÁLQW GDJJHUV DQG YDOXDEOH KDPPHU D[HV LQ WKH same area as the concentration of early Bronze Age valuables. Given that the results are the same both around Hjälmaren and in the Uppland area, it LVUHDVRQDEOHWREHOLHYHWKDWPRVWRI-HQVHQ·VFRQFHQWUDWLRQVZRUNWKHVDPH way, and that there is in fact a strong continuity in inequality when it comes to the accumulation of wealth that has been made visible in the cultural landscape. I believe that this indicates a society based on principles of he reditary inequality – the mere landscape strategy of utilities cannot solely account for these structures. The results of excavations of Swedish Earliest Bronze Age settlement sites indicate that larger and larger houses were built during the period. The smaller houses of Middle Neolithic character con tinue to be built alongside these, however, but they are also built separately IURPWKHODUJHUKRXVHV7KLVDOLJQVZLWK3HU1RUGTYLVW·VFRQFOXVLRQV´7KH observations of collective burials in megaliths, a tension between collective and single graves, buried children in the collective graves and the differentia tion of the size of houses at settlements all signal the existence of hereditary VRFLDOUDQNLQJ6LQFHKRXVHVL]HGLIIHUHQWLDWLRQFRPPXQLFDWHVDVRFLRHFR nomic difference, the social rank seems to have been related to an economic dimension. This also points towards a chiefdom level of social integration during the LN because differences in lifestyles can be understood in terms of a social index referring to the degree of economic control” (Nordqvist 2001:216ff). Furthermore, the results of southern Scandinavian excavations make it reasonable to believe that village organization was in place, at least in some settlements (see for example Artursson 2000). My results show that the rich er areas were also the more densely populated and that more people lived in the richer units than in the poorer ones. In the perspective of economic and political power, this might amount to a hereditary elite, dominating clan territories with a core of densely populated, coastal areas, measuring c. 20 x 20 km and extensively settled “uplands”, maybe in constant negotiation or feud with neighbouring groups, along dominating means of communication, such as ridges, rivers, fjords and lakes. 383 Per Lekberg Social and cultural context of the elite The accumulations of valuables and exotica in the cores of at least the richer areas show that these elites were internationally connected, at least in the Bronze Age. It is, however, very likely that there was a continuity in elite control in these core areas and that these contacts developed as early as the later parts of the Middle Neolithic of Scandinavia. The continuity in the use and furnishing of the landscape can be observed back to these times, and the foundation of this landscape continuity might be sought in the reproduction within society of cognitive or mental maps, used to regulate the furnishing of the landscape into an arena of meaning, tradition and ideological repro duction (Cherry & Renfrew 1986:154f; Gansum, Jerpåsen & Keller 1997). Also, almost simultaneous shifts in material culture (for example stone ham mer axes) and ideology over vast areas seem to give evidence of direct con tacts between Scandinavia and southern Russia as well as the Middle East and the eastern Mediterranean area (Warren & Hankey 1989:11; Kristiansen 1998a:32; Gimbutas 1956:121; Shennan 1986:138ff; Apel 2001a:336; Podbor ský et al. 1993:229; Machnik 1997:152; Ebbesen 1997:75ff; Loze 1997:135ff; Rimantiené 1997:181ff; Buchvaldek 1997:43ff; Marsalek 1999:123ff; Knuts son 1995:190f; Frödin & Persson 1938:218ff; Bouzek 1985; 1997; Wace & Thompson 1912; Blegen et al. 1950; Dörpfeld 1902; Blinkenberg 1904:22; Banks 1967a:99ff). One way for the elites of eastern central Sweden to maintain their heredi tary elite status may have been by negotiation with commoners, using foreign but meaningful objects and other signs as symbolic capital, manifesting their kinship with gods and foreign rulers and their subsequent right to dominate and accumulate (Helms 1988:66ff, 1993, 1998:120ff; Kristiansen 1991:27, 1998a:540ff, 1998b:180f; Larsson 1999a:78, 1999b:57f; Willroth 1989:93f; Beck 1996:91f; Piggott 1950:273ff, 1967:134ff, 1983:91ff, 1992:52f; Treherne 1995:108ff; Malmer 1989:96; Renfrew 1993:187). This would, I argue, result in commoners striving for status, copying the material and immaterial signs of the elites, integrating them in regional or local modes of commoner cul ture, and thus devaluating them as symbolic capital. Such an ongoing devalu ation would force the elites to keep on travelling and maintaining contact ZLWKWKHLUIDUUHDFKLQJQHWZRUNVWRVHFXUHWKHLQWURGXFWLRQRIQHZV\PEROLF capital into the system (Kristiansen 1998b:333ff; Larsson 1999:49ff; Hay den 1998:33; Nordqvist 2001:258; Elster 1988, 1993; Olsen 1997; Nordqvist 1999). This process can be compared with a siphon; once the suction is es tablished, the system runs itself, fuelled by the urge for prestige. 384 Ground stone hammer axes in Sweden 6XFKDWKUHHZD\Vocial siphon-model²ZLWKLQWKHVSKHUHRIWKHHOLWHSHHU polity interaction networks, between this and the local/regional commoners, and within the regional commoner networks – provides a reasonable expla nation of diffusion and typological change in time and space, for material DVZHOODVQRQPDWHULDOFXOWXUH /HNEHUJ 7KHGHVFULEHGSURFHVVFRXOG have played an important role in preserving the social system, in the conti nuity and change of material and immaterial culture within the vast, trans continental network of cultural dialects that we call the European Bronze Age. In my view, this cultural network through time and space is not ma QLIHVWHGÀUVWDQGIRUHPRVWE\WKHPDQXIDFWXUHDQGXVHRIEURQ]HVDVVXFK but rather by a new way of life and outlook on the conditions for existence. Seen this way, the Bronze Age in Scandinavia starts with the introduction of institutionalised inequality sometime around the MN – LN transition, subsequently gaining acceptance as a way of life. 385 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Witold Migal 7KHPDFUROLWKLFÁLQWEODGHVRIWKH neolithic times in Poland Abstract 'XULQJWKHODVWWZRGHFDGHVWKHQXPEHURIVFLHQWLÀFZRUNVGHDOLQJZLWKWKH problems of Neolithic blade procurements using pressure technique has in FUHDVHGVLJQLÀFDQWO\5HVHDUFKRQWKLVVXEMHFWZDVFRQGXFWHGLQ)UDQFH 3H legrin, see this volume), Russia (Girâ) and Poland (this author). In all these FDVHVWKHUHVHDUFKHUVIRFXVHGRQODUJHÁLQWEODGHVLQ3ROLVKDUFKDHRORJLFDO terms known as macrolithic blades. The experiments revealed that in order WRGHWDFKVXFKODUJHUHFXUUHQWEODGHVYDULRXVIRUFHLQFUHDVLQJGHYLFHVZHUH needed. Unfortunately, no archaeological remains of such devices are pre served and there are no ethnographical analogies that can be used. The most DQFLHQWIRUFHLQFUHDVLQJGHYLFHWKDWFRXOGKDYHEHHQNQRZQLQWKHHDVW(X ropean Neolithic is the wine/oil press. The presence of grape seeds together with the oldest macrolithic blades is known from the territory occupied by people from the Tripole culture. The paper is an attempt to answer the ques tion about real economic conditions determining the existence of developed craft and potential markets for macrolithic blade half products. Introduction In the Later Neolithic, the area of central Poland and Ukraine was a centre for DQLQWHQVLYHSURFHVVLQJSURGXFWLRQDQGGLVWULEXWLRQRIÁLQWDUWLIDFWV %DOFHU 1975). One of the products typical for this period is the large blades made RI WZR W\SHV RI ÁLQW WKH JUH\ZKLWH VSRWWHG 7KXURQLDQDJH ÁLQW QDPHG ŒZLHFLHFKyZÁLQWIURPWKHQDPHRIWKHRXWFURS WKDWRFFXUVLQFHQWUDO3R ODQGDURXQGWKH9LVWXOD5LYHU )LJ DQGWKHFUHWDFHRXV6HQRQLHQDJHÁLQW that occurs in Wolhyn, western Ukraine. The Polish literature on the sub ject stresses the metric change that occurred when Neolithic peoples began producing up to 34 cm long blades (Fig. 2) with an average length of about FP 0LJDO 7KHVHODUJHVL]HGEODGHVDSSHDUHGGXULQJWKHPLGGOH period of the Linear Pottery Cultures connected with the Tiszapolgar stage, ZKLFK FRUUHVSRQGV WR VWDJHV %,%,, RI WKH 7ULSROH &XOWXUH 0DVVRQ et al. ,Q3RODQGWKHDUFKDHRORJLFDOFXOWXUHVZKLFKDUHFODVVLÀHGLQWKDW complex of cultures are “Wolhyn – Lublin White Painted Pottery culture” )LJV DQGWKHJURXS´:\FLęůH²=âRWQLNLµ )LJ =DNRœFLHOQD However, the largest development and distribution of blades processed in such a way were connected with the Funnel Beaker culture (see Balcer 1980, 387 Witold Migal 1983) (Figs. 6, 7) and the Tripole culture (Budziszewski 1995), adjacent to the Funnel Beaker culture in the south. Such blades and cores were also found among Corded Ware culture grave inventories in the southern part of Poland (Machnik 1966:50 p.). As far as we know today, this was the last ap pearance of macrolithic blade technology. )LJXUH$UHDRISURGXFWLRQRIORQJEODGHVDPRQJ1HROLWKLFVRFLHWLHV D ėZLHFLHFKyZ ÁLQWPLQHRQWKH9LVWXOD5LYHU E 7HUULWRU\RI:KLWH3DLQWHG3RWWHU\FXOWXUHRFFXSDWLRQ F 2XWFURSVRI6HQRQLHQÁLQWLQ9ROK\Q 388 The macrolithic ﬂint blades of the neolithic times in Poland Figure 2. Samples of regular blades from the White 3DLQWHG3RWWHU\FXOWXUH6WU]\ĤyZ/XEOLQGLVWULFW $IWHU=DNRĘFLHOQD6FDOH Figure 3. Long blade from the White Painted pottery FXOWXUH6WU]\ĤyZ/XEOLQ GLVWULFW$IWHU=DNRĘFLHOQD 1996. Scale 1:3. The morphology of the large cores that produced macro blades can vary DFFRUGLQJ WR SURGXFWLRQ FRQWH[W $PRQJ WKH :/:33 URXQGHG VWULNLQJ platforms were used and the cores had a conical shape similar to those of the Tripole culture (Fig. 4). Cores prepared within the Funnel Beaker Culture, RQWKHRWKHUKDQGDUHRIWHQÁDWKDYHWULDQJXODUVKDSHVDQGZHUHSUHSDUHG with two or three side crests (Fig. 5). 389 Witold Migal Figure 4. Core from the White Painted Pottery culture. Staszów, Kielce district. After =DNRĘFLHOQD6FDOH )LJXUH/DUJH6HQRQLHQÁLQW core from the White Painted 3RWWHU\FXOWXUHáüĤHN=Dklikowski. Lublin district. $IWHU=DNRĘFLHOQD Scale 1:3. 390 The macrolithic ﬂint blades of the neolithic times in Poland Blade production There are some factors that may indicate that these large blades were de tached with applied pressure. Above all it is the extraordinary regularity of sides and thickness (Figs. 6, 7). Another premise supporting this suggestion is the very strong reduction of striking platform before every detachment ²XSWRFPEHWZHHQWZRFRQVHFXWLYHEODGHV²WKDWLVYLVLEOHLQUHÀWWLQJV,Q the case of FBC, there is one additional factor which relates to the economy RIÁLQWUDZPDWHULDODQGWKHSUREOHPRILPPRELOL]DWLRQRIWKHLQYHVWLJDWHG ÁLQWFRUHVGXULQJSUHVVXUH7RJHWKHUZLWKRWKHUDUFKDHRORJLVWVRIWKH6WDWH Archaeological Museum in Warsaw we investigated one of the inventories of WKH)%&VHWWOHPHQW=ORWDQRWIDUIURPDGHSRVLWRIVSRWWHGÁLQW 0DWUDV]HN et al. ,WZDVSRVVLEOHWRUHÀWÁLQWIURPUHVLGXHSLWV2XUÀUVWLQWHQWLRQ ZDVWRXQGHUVWDQGGLIIHUHQWZD\VRIH[SORLWDWLRQRIDODUJHEODGH$UHÀWWLQJ of the blocks revealed a process of reshaping blade cores to quadrangular axes, typical for this cultural group. It was interesting to note that the analy VLVRIGLIIHUHQWPDWHULDOV WKURXJKREVHUYDWLRQRIÀQLVKHGD[HV VKRZHGWKDW this process was more or less typical for this tradition. Our analysis was also VXFFHVVIXOLQWHUPVRIGHÀQLQJWKHFKDUDFWHULVWLFGHELWDJHIRUWKLVW\SHRI production. The process can be described as follows: 1. Shaping the core together with all necessary crests was carried out FORVHWRWKHPLQHLQŒZLHFLHFKyZ 2. Exploitation of cores by pressure was probably carried out in only RQHRUWZRKRXVHKROGV HYLGHQFHIRXQGDWüPLHOyZ %DOFHU Bodiaki (Skakun 1996)) 'LVTXDOLÀFDWLRQRIFRUHV UHDVRQVL]HRIEODGHV 4. Accumulation of cores in huts (for example 22 specimens at Bodiaki, see Skakun 1996 or the Sapanów collection at State Archaeologi FDO0XVHXPLQ:DUVDZ RUUHZRUNLQJWKHPLQWRD[HV )%&=âRWD üPLHOyZVHH0DWUDV]HNet al. 2002). The shaping and reworking of cores into axes was no doubt executed with a punch technique and the production of blades by using a pressure technique. ,WLVHVVHQWLDOWKDWWKHWKLUGSRLQW²RQZKDWJURXQGVFRUHVZHUHGLVTXDOLÀHG – is explained. The loss of metric values of potential blades could be one UHDVRQIRUGLVTXDOLÀFDWLRQ1HYHUWKHOHVVPDQ\FRUHVDUHPXFKORQJHUWKDQ LQGLFDWHGE\WKHQHJDWLYHVRIEODGHV7KHPXVHXPDW=EDUDů :ROK\Q KDV 391 Witold Migal is a 42 cm long core with negatives no longer than 30 cm. At the same time, the discarded cores show that it would have been possible to extract more blades from them, smaller than 15 cm. Precision in the shaping of sides and backs of the cores leads us to the conclusion that a system of immobilization PXVWKDYHH[LVWHGZKLFKH[FOXGHGWKHFRUH·VIXUWKHUXVHDIWHUDUHGXFWLRQLQ dimension. Such a construction must have been large and stable enough to hold out against the great force acting on it in the detaching of blades. In my opinion, this explains both the rigorous shaping of cores and the later stage of discarding. Figure 6. Two of the longest blades from the Funnel Beaker culture in Poland. Both PDGHRIėZLHFLHFKyZÁLQW.DPLHĎáXNDZVNL.LHOFHGLVWULFW$IWHU%DOFHU Scale 1:3. 392 The macrolithic ﬂint blades of the neolithic times in Poland Figure 7a. One core for long blades from the large settlement RI)XQQHO%HDNHUFXOWXUHDWýPLHOyZ.LHOFHGLVWULFW6XFKFRUHV ZHUHXVXDOO\UHZRUNHGE\1HROLWKLFÁLQWNQDSSHUVLQWRÁLQW D[HVėZLHFLHFKyZÁLQW$IWHU%DOFHU6FDOH 393 Witold Migal Figure 7b. One core for long blades from the large settlePHQWRI)XQQHO%HDNHUFXOWXUHDWýPLHOyZ.LHOFHGLVWULFW 6XFKFRUHVZHUHXVXDOO\UHZRUNHGE\1HROLWKLFÁLQWNQDSSHUVLQWRÁLQWD[HVėZLHFLHFKyZÁLQW$IWHU%DOFHU Scale 1:3. 394 The macrolithic ﬂint blades of the neolithic times in Poland Devices for pressuring blades During the past 15 years, experimental research geared towards getting long blades with the help of devices that enhance pressure have been conducted in Europe. The lever was one mechanism constructed as a way of enhancing pressure. In a construction made by Giria, the core was fastened in a vice. The blades were detached from the core by increasing the pressure on the front of the core using the lever (Giria 1997). By applying this device, the researchers were able to make blades up to 27 cm long (pers. com.). My own construction was placed on a log on the ground as a support in which the core was made immobile in a carved notch. Blades were detached with an DQWOHUWLSSHGZRRGHQSROHSXWWKURXJKDKROHLQWKHXSSHUSDUWRIWKHORJ The point of support for the lever had been constructed as a hole in the wall opposite the core. This way of working was very compact and connected with the size and shape of cores. The longest blades were 21 cm, and I could exploit the core to about 15 cm of height. Further study will include an in vestigation of the bullet shaped high pressure blade cores of the Globular $PSKRUDH&XOWXUH %DOFHU ZKLFKW\SLFDOO\JDYHQDUURZDQGÁDW blades (Migal and Barska 2003). Both constructions for blade detachments mentioned above are results of an intellectual and practical game with clear visible inspirations (in my case WKHLPPRELOL]DWLRQV\VWHPLVDYDULDQWRI3HOHJULQ·VIRUN 7KXVWKHH[SHULPHQWDOZRUNUHVXOWHGLQPRUHTXHVWLRQV'LGWKHVHIRUFH enhancing devices really exist in central Eastern Europe? If similar devices were known for other purposes it is possible that comparable technical solu WLRQVZRXOGEHXVHGWRPDNHÁLQWEODGHV The oldest devices to multiply mechanical force used the principle of the lever Beyond doubt, the oldest known devices for enhancing human muscular force are so called simple machines, the most well known examples of which are the wedge, the lever, the screw and the pulley block or the windlass. The ZHGJHZDVZLGHO\XVHGLQÁLQWPLQLQJ(YLGHQFHIRUWKUXVWLQJZHGJHVKDV EHHQIRXQGLQPLQHVRIEDQGHGÁLQWDW.U]HPLRQNL FHQWUDO3RODQG ZKHUH ÁLQWDQGVWRQHZHGJHVZHUHXVHGIRUKRUL]RQWDOVSOLWWLQJRIOD\HUVRIOLPH stone oxfordian rock (Migal 2000). Wedges were not used as axes, but two dozen tools were simultaneously thrust into the ceiling, splitting off plates about 1.5 m in diameter. From the same mines we have evidence of work 395 Witold Migal with a wooden pole lever with which similar work was probably executed. The pole was about 12 cm in diameter and about 2 m long and was used for breaking off large blocks of cast rock. These mines were exploited by the Globular Amphorae population in the early third millennium cal BC. One problem is that long pressure blades appeared in Polish territories 500 years earlier. Furthermore, all such production seems to have existed within a spe FLÀFGHÀQLWHVRFLRHFRQRPLFVLWXDWLRQ Socio-economical transformation among the peoples of the later Linear Pottery cultures and the Tripole culture According to current theories, social transformations of the population of the later Linear Pottery cultures are connected with an agricultural econo my that led to changes in social relations. By reconstructing the size of the population of the Tripole culture, Russian researchers have shown that this cultural unit possessed many features of a “civilization”, which is marked by using the term “Tripole Civilization”. They underline its settled character as indicated by, e.g. the municipal character of the housing estate of the Near (DVW 7KHVH ´PXQLFLSDOLWLHVµ FRYHU KHFWDUHV H[DPSOHV DUH IRXQG DW Tsciszovka or Onoprievka in stage B I, and in the later stage (B II) they cover KHFWDUHV 9LGHLNR 7KLVLQGLFDWHVWKDW WKH DUHD RIWKH SURWRWRZQ0D\GDQHWVNR\H KHFWDUHV ZDVVHWWOHGE\c. 10,000 inhabit ants occupying 2800 buildings. On the other hand, we see advances in the ceramic technology and the development of exchange of copper items (see Videiko 1994:9). At the same time, similar changes occurred within the later Linear Pottery cultures (also called the Lengyel – Polar Complex). In both cases, the transformations were accompanied by a larger cultivated area (in comparison with the later Linear Pottery Culture), as well as an increased H[FKDQJH RI GLIIHUHQW JRRGV E\ ORQJ GLVWDQFH WUDGH FRSSHU ÁLQW DPEHU shells). Other differences seen in the archaeological material are e.g. the gi gantic storage amphorae that are accompanied by small cups or mugs with ears (see Sherratt 1987, 1995). Such sets are linked to the production and con VXPSWLRQRIZLQH *XPLľVNL ,QFRQQHFWLRQZLWKWKH7ULSROH&XOWXUH of stage B II, specialized pottery workshops are also known. This is evidence of a civilization in progress among the Middle Neolithic populations and of simultaneous specialization in the production of different types of goods. 396 The macrolithic ﬂint blades of the neolithic times in Poland Unfortunately, the factors mentioned above do not explain the existence RIH[FHSWLRQDOSURGXFWVVXFKDVPDFUROLWKLFÁLQWEODGHV7KLVLVQRWEHFDXVH their beautiful appearance and size were indispensable for production and distribution over large distances. One can successfully execute everyday tasks with far smaller and uglier blades. Thus we have to accept two essential IDFWRUVDVLQÁXHQFLQJWKHGHYHORSPHQWRIWKLVSURGXFW 7KHPDFUROLWKLFÁLQWEODGHVKDGWREHV\PEROVRIVRFLDOVWDWXVRU 2. They were the most important identifying symbols of tribe member ship. &RQFHUQLQJWKHVHFRQGIDFWRULWLVLQWHUHVWLQJWRQRWHWKDWWKHÀQLVKHGEODGHV were often distributed outside the range of the archaeological cultures that occupied the areas where extraction and procurement of raw materials took SODFH LQ RXU FDVH EODGHV IURP JUH\ VSRWWHG ÁLQW PDGH E\ )XQQHO %HDNHU producers). Thus, Funnel Beaker blades are found at the territory of White 3DLQWHG3RWWHU\FXOWXUHDW8NUDLQHDQGEODGHVPDGHRIÁLQWIURP:ROK\Q are found in the Funnel Beaker area. In the case of FBC, one essential factor is the megalithic activities visible in the large constructions of stone, wood and earth in Poland. The construc tions clearly show the high degree of organization of these societies through WKHH[HFXWLRQRIVXFKODUJHVFDOHXQGHUWDNLQJV,QWKHJUDYHLQYHQWRULHVZH RIWHQÀQGODUJHEODGHV2QWKHRWKHUKDQGVXFKEURNHQEODGHVDUHDOVRIRXQG DVWRROVLQODUJHVHWWOHPHQWVLQWKHYLFLQLW\RIWKH6ZLHFLHFKyZÁLQWPLQHV However, although this was not investigated, it seems possible that ready blanks of blades were more appreciated further away from the deposits and places of production. Such blades have larger dimensions and are in a better state of maintenance compared to blades on settlements near the deposit WKHORQJHVWNQRZQÁLQWEODGHPDGHRI6ZLHFLHFKyZÁLQWLVc. 34 cm long and comes from Kuiavia in central Poland, c. 300 km from the extraction area). 7KHUHLVQRHYLGHQFHWKDWWKHSHRSOHRIWKH)%&ZKRUHFHLYHGWKHÀQLVKHG blades far from the deposits, also were able to use this technique to produce EODGHVIURPORFDOÁLQW7KLVFOHDUO\LQGLFDWHVDKLJKVSHFLDOL]DWLRQZLWKLQDQ archaeological culture. Another transformation in lifestyle in comparison with the Linear Pot tery cultures took place when the people of the FBC occupied poor quality lands simultaneously with an increase in population. In the archaeological material, this is indicated by the appearance of new plants and changes in the system of cattle farming (see Kruk 1973). 397 Witold Migal Among others, Russian researchers underline that it is precisely during this SHULRGWKDWWKHROGHVWVHHGVRIGRPHVWLFDWHGJUDSHYLQHIRXQGRQDVHWWOH PHQWDW1RY\MH5XVHVWLDSSHDU7KHÀQGLVGDWHGWR %OQ and thus belongs to the phase BI – BII of the Tripole culture (Masson et al. 1982:191, 235). In the production of must, the largest quantity of juice should be extract ed from ripe grapes. The primary method to increase the amount of must is different ways of trampling and crushing the berries and straining the juice )UDQNHO ,WLVDZHOOLOOXVWUDWHGZD\RISURGXFWLRQDQGZDVSUREDEO\ used by producers of wine from early on. On the other hand, the refuse of VXFKDSURGXFWLRQFRQWDLQVODUJHTXDQWLWLHVRIMXLFHWKDWDUHGLIÀFXOWWRKDQ dle immediately. From the beginning, different machines were constructed for the purpose of increasing pressure to maximise the amount of must. A wide range of constructions existed and attest to the ingeniousness of people living in different wine producing regions. Examples of devices that have been recovered provided the basis for the experimental investigations con QHFWHGZLWKGHWDFKLQJODUJHEODGHVIURPÁLQW Figure 8. Some ideas how a machine for making long blades would be done. Based on historical- ethnological evidence. After Frankel 1999. Summary 0DFUROLWKLFÁLQWEODGHVZLWKOHQJWKVIURPWRFPDSSHDUGXULQJWKH Middle Neolithic of east central Europe together with groups of the Tripole FXOWXUHDQG:ROK\Q/XEOLQ:KLWH3DLQWHG3RWWHU\FXOWXUH$WWKHWLPHWKH\ ZHUH SURGXFHG LQ 8NUDLQH HJ :ROK\Q /DUJHEODGH SURGXFWLRQ ZDV DOVR developed in the later stage of the Funnel Beaker culture. Beyond question, WKLVSURGXFWLRQLVFRQQHFWHGZLWKWKHDSSOLFDWLRQRIIRUFHHQKDQFLQJGHYLF es. With the appearance of such blades we see changes in economy and ways of farming, including grape cultivation and wine production. My hypothesis unites these two facts and emphasizes that a similar construction based on the idea of a simple machine (Fig. 8) was used both for grape must extrac tion and blade detachment. However, the last blades of this type appeared in WKHHDUO\VWDJHVRIWKH&RUGHG:DUHFXOWXUHLQVRXWKHDVWHUQ3RODQG PLGrd millennium); after that, blades were produced differently. 398 Apel, J. & Knutsson, K., 2006. Skilled Production and Social Reproduction. SAU Stone Studies 2. Uppsala. Kim Darmark Flaked rhyolite from Jettböle: attempts at an experimental explanation Abstract 3LWWHGZDUH VLWHV VLWXDWHG LQ WKH cODQG ,VODQGV )LQODQG RIWHQ \LHOG KXJH quantities of lithic debris from a local rhyolite. This source of information has attracted very little scholarly attention. The aim of the present article is WRGHVFULEHDVDPSOHRIÁDNHGPDWHULDOIURPRQHRIWKHPRVWZHOONQRZQPLG GOH1HROLWKLFVLWHV-HWWE|OH7KHEDVLFUHGXFWLRQVWUDWHJ\VHHPVWREHKLJKO\ dependent on direct technique and to be of a quantitative nature. Tools are VHHPLQJO\LQIRUPDODQGODFNVHFRQGDU\PRGLÀFDWLRQ6WLOOWKHOLWKLFVIURP this site display some unexpected characteristics, which will be highlighted in contrast with material from a roughly contemporary site. Therefore, a second goal of this paper is to present the results of a series of experiments, ZKLFKDUHFRQGXFWHGVSHFLÀFDOO\LQDQDWWHPSWWRJHWDEHWWHUXQGHUVWDQGLQJ of the formation processes behind the archaeological material. Introduction -HWWE|OH LV WKH ELUWKSODFH RI 6WRQH $JH DUFKDHRORJ\ LQ WKH cODQG ,VODQGV When it was discovered in 1905, only a few years after the discovery of the ÀUVW3LWWHG:DUHVLWHLQHDVWHUQFHQWUDO6ZHGHQLWFKDQJHGWKHHDUOLHUQRWLRQ of the islands being uninhabited during the Stone Age. Excavations con GXFWHGEHWZHHQDQGUHVXOWHGLQODUJHDPRXQWVRIÀQGVHVSHFLDOO\ of the richly decorated pottery characteristic of the Pitted Ware tradition, and of lithic debris. The site became notorious when it became clear that several human skeletons were deposited in the cultural layer, a fact that led to an ongoing discussion about the possibility of cannibalistic practice (Ce GHUKYDUI1XxH] 7KHDQWKURSRPRUSKLFFOD\ÀJXUHVZLWKSDUDO OHOVLQWKHHDVWHUQ&RPEFHUDPLFWUDGLWLRQDUHZLGHO\NQRZQ 1XxH] Wyszomirska 1984). In 1999 and 2000, new excavations were conducted at Jettböle. During this time, a nearby settlement by the name of Bergmanstorp was also investigated. Bergmanstorp is situated 200 metres east of Jettböle and 4 to 5 metres higher above sea level. The 14&GDWLQJVIURP-,UDQJH between 3300 and 2600 BC (Storå & Stenbäck 2001: 65), and the datings from Bergmanstorp fall within the same interval (Storå & Stenbäck 2001: 399 Kim Darmark 63f.). However, the presence of atypical pottery might indicate an older pha se at Bergmanstorp (Stenbäck 2003). In eastern central Sweden, the Neolithic Pitted Ware Culture emerges around 3500 BC, and is characterized by a return to an economy highly dependent on hunting and gathering. The coastal regions became particu larly important, and osteological assemblages are often dominated by marine wildlife (Storå 2001b: 3f). 7KHPDLQÀQGFDWHJRU\DW3LWWHG:DUHVLWHVLQHDVWHUQFHQWUDO6ZHGHQLV the decorated pottery, while the amount of lithic debris varies to a consider DEOHGHJUHH (GHQPRHWDO ,QWKHcODQG,VODQGVVHYHUDOVLWHV GLVSOD\ODUJHTXDQWLWLHVRIÁDNHGPDWHULDOSUHGRPLQDQWO\UK\ROLWH7KHUK\R OLWHIURPcODQGLVDSRUSK\ULWLFURFNFRQVLVWLQJRIDJUH\PDWUL[ZLWKSKHQ RFU\VWVRITXDUW]DQGIHOGVSDU,WFDQEHIRXQGDORQJWKHEHDFKHVRIcODQG and to a certain degree in southwest Finland (Nuñez 1990). It is similar to porphyry occasionally found at Swedish sites on the mainland. Even though WKHUK\ROLWHRIWHQPDNHVXSWKHEXONRIWKHÀQGVIURPcODQGLF6WRQH$JH sites, very few studies have taken this material into account. The works that touch upon the subject often approach the material from a formalistic point of view, with the aim of identifying formal tools (Dreijer 1940; Meinander 1957; Nuñez 1990; Storå 1990). It has been demonstrated that this is not a YHU\IUXLWIXOZD\RILQYHVWLJDWLRQRXWVLGHWKHÁLQWEHDULQJDUHDVRI6ZHGHQ (Callahan 1987). 7KHDPRXQWRIOLWKLFVIRXQGQRWRQO\DW-HWWE|OHEXWDOVRDWcODQGLF3LW ted Ware sites in general, seems to be without rival in eastern central Swedish contexts, where lithic material is often relatively sparse. In order to illustrate WKLVIDFW,VFDQQHGWKURXJKVRPHH[FDYDWLRQUHSRUWVIURP6ZHGHQDQGcODQG (Larsson 1995; Olsson 1996; Olsson et al.1994; Storå 1995a; Storå 1995b; Storå & Stenbäck 2001; Welinder 1971), and earlier summaries (Martinsson 1985; Nuñez 1990). The comparison is presented in Fig. 1, which gives the reader an idea of the vast amounts of lithic debris that are unearthed at the VLWHVLQWKHcODQG,VODQGVDVFRPSDUHGWRWKDWHQFRXQWHUHGRQWKH6ZHGLVK mainland. 400 Flaked rhyolite from Jettböle: attempts at an experimental explanation Site Häggsta V (Swe) Häggsta IV (Swe) Överåda (Swe) Fagervik (Swe) Korsnäs (Swe) .ROVYLGMD cO -HWWE|OH, cO 6PLNlUU cO cVJnUGD cO Amount of debris 66 291 205 64 3159 11725 26208 20040 27232 m excavated 350,0 220,0 125,5 18,0 136,0 220,0 187,5 125,0 82,0 'HEULVP 0.18 1.32 1.63 3.55 46.45 53.30 139.78 160.32 332.09 Figure 1. Table showing the amount of lithic debris at various Pitted Ware sites from the Swedish mainland (Swe) and from Åland (Ål). The characteristics of the lithic assemblages $ VWXG\ KDV EHHQ FRQGXFWHG E\ WKH DXWKRU LQYROYLQJ ÁDNHG UK\ROLWH IURP 1999 years excavation at Jettböle. In order to establish the exact location of the 1905 excavation and to get a reference point as to the stratigraphy of the site, a small trench (4.5 x 2.5 meters) was opened up (Storå & Stenbäck 2001). The old excavation trench was encountered and the total area of undisturbed cultural layer excavated during 1999 was 7.5 m2 ,QDOOÁDNHVRIUK\ROLWH were collected during the excavation; 33 cores and two hammerstones of the same material were also recovered. In this paper, the material from Jettböle will be compared with material from trench 2 at Bergmanstorp, which was excavated in a similar manner (Storå 1999). The following characteristics have been noted in the studied material from Jettböle and Bergmanstorp (Darmark 2001): x The dominating reduction method employed at the site is a direct platform method, probably using a hard hammerstone, judging from the 15 hammerstones found within the trench. Very little evidence RIDELSRODUPHWKRGLVIRXQGZLWKLQWKHPDWHULDO2QHSODWIRUPÁDNH in rhyolite has been subjected to bipolar reduction but only to a li mited degree. One bipolar core in quartz was found at the site. The use of an anvil method has not been investigated, but the presence of anvils within the trench possibly indicates that this technique has been used. The lithics from Bergmanstorp also display a total lack of bipolar technique. 401 Kim Darmark x The amount of splinters within the Jettböle assemblage is large, 60%. This is valid even though the excavated material was sieved through 0.5 cm mesh, and even though no water sieving was conducted. The original amount of splinters could therefore be somewhat higher than shown by the excavated material. The 60% from JI99 can be compared to the mere 28% from Bergmanstorp, trench 2. x 7KHFRPSOHWHÁDNHVIURP-,DUHJHQHUDOO\VPDOOLQVL]H7KHÁDNHV have been categorized according to weight, in classes of 10 grams. $GRPLQDWLQJSDUW RIWKHFRPSOHWHÁDNHVIURP-,LVIRXQG WRZHLJKJUDPVRUOHVV:LWKLQWKLVJURXSPRVWRIWKHÁDNHVDUH EHORZ JUDPV $W %HUJPDQVWRUS RI DOO ÁDNHV DUH EHORZ grams. x )RUPDOWRROVDUHQRQH[LVWHQW$IHZÁDNHVKDYHEHHQREVHUYHGWR GLVSOD\VLJQVRIVHFRQGDU\PRGLÀFDWLRQPRVWO\LQWKHIRUPRIXVH wear (polish or micro retouches), but deliberate shaping of tools has not been found. x 7KHÁDNHVWKDW VKRZ WUDFHVRI XVH KDYH LQ FRPPRQ WKDWWKH\ DUH ODUJHLQVL]HRUUDWKHUWKH\ÀWQLFHO\LQRQH·VKDQG7KLVFRXOGRI FRXUVH EH D UHVXOW RI QRW XVLQJ PLFURZHDU DQDO\VLV WKH XVHZHDU has been observed with the naked eye only. Therefore, there is a possibility that only the larger tools used for rough work have been LGHQWLÀHGZKLOHVPDOOHUWRROVKDYHUHPDLQHGXQQRWLFHG x The cores are mostly polygonal in shape and are the result of an opportunistic strategy, where the toolmaker has used appropriate platforms and has not tried to uphold or create a platform suiting his purpose. Some cores have a constant platform. The relative frequen cy of cores is higher at Bergmanstorp, where the 74 cores constitute 2.4% of the total rhyolite material. The corresponding value in JI99 is 0.4% (33 cores). The overall impression after the survey of the material is that there are tech nological similarities. Judging from the cores, which are often exhausted, ERWKDVVHPEODJHVDUHWKHUHVXOWRIDUHGXFWLRQSURFHVVLQLWVÀQDOSKDVH7KH goal of the platform reduction at both sites seems to have been the produc WLRQRIODUJHUÁDNHVXVLQJDIUHHKDQGPHWKRG7KHPRVWVWULNLQJGLIIHUHQFH between the sites is in the amount of splinters at Jettböle that contrasts with 402 Flaked rhyolite from Jettböle: attempts at an experimental explanation WKHPDWHULDOIURP%HUJPDQVWRUS7KHODUJHSRUWLRQRIYHU\VPDOOÁDNHVDW Jettböle was seen as problematic, since the technology was described as be ing macrolithic. Further interpretations were hindered by the lack of an ex perimental frame of reference. The experiments Three series of experiments were conducted during the summer of 2003. 7KH SXUSRVH RI WKHVH ZDV WR JHW D ZHOOQHHGHG UHIHUHQFH WR WKH DUFKDHR logical material. The archaeological lithic material has been used in cultural interpretations of the site (Darmark 2001), but with hesitation, since the natural fragmentation was not known. Therefore, the experiments tried to imitate the anticipated reduction strategy described above. The lithic debris produced was sieved using a 0,4 mesh in order for it to be comparable to the excavated material. The following questions were asked: x What is the natural fragmentation of rhyolite? In what proportions DUHFRPSOHWHÁDNHVIUDJPHQWVDQGVSOLQWHUVSUHVHQWDWDNQDSSLQJ ÁRRU" x Are the fragmentation and the amount of splinters produced af fected by the weight of the hammerstone? This question arose since the hammerstones at Jettböle and Bergmanstorp are different. At Bergmanstorp, only smaller hammerstones (average 400 grams) were found, while the hammerstones at Jettböle ranged from 200 to 2300 grams. The character of the hammerstones was believed to be connected with the characteristics of the lithic debris. x How common is the occurrence of multiple fractures? Previous ex periments showed that the lack of platform preparation or prepara tion of hammerstones often resulted in the production of several ÁDNHVLQWKHVDPHEORZ7KHVHFRQGDU\ÁDNHVDUHRIWHQHDVLO\GLVWLQ JXLVKHGLQWKDWWKH\DUHVPDOOWKLQDQGKDYHDFUXVKHG´8VKDSHGµ platform. These had also been noticed in the archaeological material from Jettböle and their presence led to the idea that they were re VSRQVLEOHIRUWKHKLJKGHJUHHRIVPDOOÁDNHVDWWKLVVLWH 403 Kim Darmark 100% 80% 49 53 55 Splinters 60% Unclassified High Speed Fract. 10 40% Lateral Fract. 14 16 13 9 11 20% 27 Complete Flakes 22 18 Exp 2 (n=78) Exp 3 (n=119) 0% Exp 1 (n=263) Figure 2. The results of the experimental reduction (%). Figure 2 illustrates the results of the experimental reduction. Exp. 1 was car ried out with a heavy hammerstone of granite, weighing 2600 grams, Exp. 2 with a natural boulder of rhyolite with a weight of 1900 grams, while Exp. ZDVSHUIRUPHGZLWKDJUDPKDPPHUVWRQHRITXDUW]LWH7KHPDWHULDO REWDLQHGGXULQJWKHUHGXFWLRQKDVEHHQFODVVLÀHGRQDIRUPDOEDVLVLQWKH following categories: &RPSOHWHÁDNHV7KLVLVDÁDNHZLWKWKHSODWIRUPDQGGLVWDOSDUWLQ tact. No attention has been paid to the fact that many of the smaller ÁDNHVDUHWKHUHVXOWRIPXOWLSOHIUDFWXUHVDQGDUHWKHUHIRUHDE\SUR duct. • Lateral fracture fragments. This category includes proximal, medial and distal fragments. • High speed fractures. Includes middle and lateral fragments. 8QFODVVLÀHG. This category is a result of the fact that many rhyolite boulders have natural cracks within them. When the boulder is struck the force moves along these natural planes and the resulting ÁDNHRUIUDJPHQWLVKDUGWRFODVVLI\ • Splinters. Small fragments. 404 Flaked rhyolite from Jettböle: attempts at an experimental explanation The general trend is that the amount of splinters grows higher as the size of the hammerstone diminishes. The fragmentation in general follows the same trendline. The high speed fracture is marginally more common than WKHODWHUDOIUDFWXUH7KHDPRXQWRIFRPSOHWHÁDNHVREWDLQHGLVKLJKHVWZLWK the heaviest hammerstone. This is interesting, since the presupposition (judging from the archaeological material) was that the fragmentation would be higher using a larger hammerstone. It seems likely that the nature of the rhyolite, being a very hard, elastic kind of rock, is best reduced with a heavy hammerstone that can produce enough energy to allow the fracture to travel DOOWKHZD\DQGSURGXFHDFRPSOHWHÁDNH &RQFHUQLQJWKHTXHVWLRQRIPXOWLSOHIUDFWXUHVDVLPSOLÀHGZD\RIFDOFX ODWLQJWKLVKDVEHHQXVHG7KHLQGLYLGXDOÁDNHVKDYHQRWEHHQFODVVLÀHGDV WKHUHVXOWRIPXOWLSOHIUDFWXUHV,QVWHDGWKHQXPEHURIFRPSOHWHÁDNHVEH low 10 grams within the different series has been used as a relative measure ment to express this. Since the experiments were carried out with the goal of SURGXFLQJODUJHUÁDNHVWKHVPDOOHUÁDNHVDUHPRUHRIWHQWKDQQRWWKHUHVXOW of multiple fractures. 80 60 Number 40 % 20 0 Exp 1 Exp 2 Exp 3 Total Experimental )LJXUH7KHDPRXQWRIVPDOOÁDNHVZLWKLQWKHH[SHULPHQWDOVHULHVUHSUHVHQWLQJ the degree of multiple fractures. $VLVVHHQE\)LJXUHWKHDPRXQWRIVPDOOÁDNHVKHUHLQWHUSUHWHGDVWKH result of multiple fractures, is large with hammerstone 1 and diminishes with the size of the hammerstone. This is not necessarily a result of the weight of the hammerstone. The hammerstones used are made of different material. The granite of Hs 1 has the largest crystalline structure, and is therefore PRUHOLNHO\WRSURGXFHPRUHÁDNHVZKLOHERWKWKHUK\ROLWHDQGWKHTXDUW]LWH 405 Kim Darmark hammerstones retain a relatively smooth surface even after usage. The ex SHULPHQWVFOHDUO\VKRZWKDWVPDOOÁDNHVDUHFRPPRQO\SURGXFHGGXULQJWKH UHGXFWLRQRIODUJHÁDNHVDQGFRQVWLWXWHWKHEXON RIWKHFRPSOHWH ÁDNHFDWHJRU\ Comparisons between the experiments and the archaeological material The experiments now enable us to look upon the archaeological material through a frame of reference. Figure 4 gives us the following information regarding Jettböle: x 7KHDPRXQWRIFRPSOHWHÁDNHVLQWKHDVVHPEODJHLVDOPRVWLGHQWLFDO to the sum of the experiments. Since the hammerstones from Jett böle range from 200 to 2300 grams, this result is very encouraging. x The amount of fragments from Jettböle is lower than in the expe rimental series. Above all, this applies to the lateral fracture frag ments. x The percentage of splinters is somewhat higher than all the experi mental series show. Bergmanstorp (n=302) 30 Jettböle (n=2109) 24 Total experimental (n=460) 24 12 6 20 28 9 59 Complete Flakes 12 12 51 Lateral Fract. High Speed Fract. 18 Exp 3 (n=119) 11 16 Unclassified 55 Splinters 22 Exp 2 (n=78) 9 27 Exp 1 (n=263) 0% 14 13 20% 53 10 40% 49 60% 80% 100% Figure 4. The experimental series compared with the lithics from the discussed sites. 406 Flaked rhyolite from Jettböle: attempts at an experimental explanation &RQFHUQLQJ%HUJPDQVWRUSÀJXUHWHOOVXVWKDW x The amount of splinters is very low, 28 % compared to the 50 %, which is the average in the experiments. x There are a high proportion of lateral fracture fragments and com SOHWHÁDNHV x 7KHSURSRUWLRQRIXQFODVVLÀHGREMHFWVLVPXFKKLJKHUWKDQQRUPDO 80 60 40 20 0 Exp 1 Exp 2 Exp 3 Total Experimental Jettböle Bergmanstorp )LJ7KHSUHVHQFHRIÁDNHVEHORZJUDPVZLWKLQWKHH[SHULPHQWDOVHULHVFRPpared with the archaeological material. )LJXUHLOOXVWUDWHVWKHIDFWWKDWWKHSHUFHQWDJHRIVPDOOÁDNHVDW-HWWE|OHLV highly similar to what is obtained during experimental reduction as a result of multiple fracturing. Bergmanstorp, on the other hand, has relatively few FRPSOHWHÁDNHVEHORZJUDPV Conclusions and discussion Originally, Bergmanstorp was viewed as a primary site of reduction, where raw material had been tested in order to establish its quality. This explained WKHSUHVHQFHRIORZTXDOLW\UK\ROLWH7KHUHIRUH%HUJPDQVWRUSZDVVHHQWR UHSUHVHQW D NQDSSLQJ ÁRRU ZKLOH -HWWE|OH ZDV VHHQ DV UHSUHVHQWLQJ D VLWH where the lithic material had been reduced beyond what was functionally ex plainable (Darmark 2001). This interpretation is strongly challenged by the experiments presented in this article. It is clear that Jettböle is much closer to 407 Kim Darmark DQRULJLQDONQDSSLQJÁRRUWKDQ%HUJPDQVWRUSLV7KHRQO\DVSHFWLQZKLFK Jettböle differs from the experiments is in a relative lack of fragments. The experiments have also helped in solving the question of the presence of large DPRXQWVRIVPDOOÁDNHVZLWKLQDPDFUROLWKLFLQGXVWU\ZKLFKDUHQRZVHHQWR be the result of unintentional multiple fracturing. Bergmanstorp, on the other hand, contains a small amount of splinters, ZKLFKLVLQDFFRUGDQFHZLWKDUHODWLYHODFNRIVPDOOÁDNHVLQ WKLVFRQWH[W seen as the result of multiple fracturing. Combined with the high amount of ODUJHUFRPSOHWHÁDNHVWKHHPHUJLQJSLFWXUHLVWKDWRIDVLWHZKHUHUHGXFWLRQ has not taken place. Instead, the already reduced lithic material was brought to the place, possibly to be used as tools or as raw material for tools. The high UDWHRIFRUHVZLWKLQWKHWUHQFKGRHVQRWÀWZHOOLQWRWKLVSLFWXUHKRZHYHU A total of 33 cores were found in the small area of 7 m2 . Therefore, it seems more reasonable to view trench 2 as a refuse area. This might be further VWUHQJWKHQHGE\WKHKLJKDPRXQWRIORZTXDOLW\UK\ROLWH Whether the lithic assemblage at Jettböle is to be seen as primary or sec RQGDU\UHIXVH 6FKLIIHU LVGLIÀFXOWWRDVFHUWDLQ6SOLQWHUVDUHXVXDOO\ seen to indicate primary refuse owing to their small size that easily evades cleaning attempts (Healan 1995). If the knapping is done on a blanket or a fur, even small debris can be collected and disposed of elsewhere (Fladmark 1982). However, the close resemblance between the Jettböle assemblage and the experimental series does show that lithic reduction has been carried out in the vicinity. This is interesting, especially when compared to similar sites on the Swedish mainland. As stated in the beginning, very few Pitted Ware sites in eastern central Sweden show signs of a considerable production of stone tools (cf Gustafsson et al. 2003: 51f, 56). Judging from the depositional VWUDWHJLHVDQGWKHREYLRXVO\LQWHQVLYHVWRQHWRROSURGXFWLRQDWWKHcODQGLF sites, it seems there would be a difference in function between the Swedish DQGWKHcODQGLFVLWHV 7KHRSSRUWXQLVWLFDQGVWUDLJKWIRUZDUGQDWXUHRIWKHOLWKLFWHFKQRORJ\DW both Jettböle and Bergmanstorp would seem to be the result of a lack of skill in stone production. This, however, is implausible. The choice of a simple WHFKQRORJ\PXVWEHVHHQDVDVRFLDOVWUDWHJ\7KHORZGHJUHHRINQRZKRZ within the technology would ensure that the technology was available to DQ\RQH FI$SHO 7KH3LWWHG:DUHKXQWLQJJDWKHULQJFRPPXQLW\ living in an age of steadily increasing social complexity, consciously used the easily available technology as a means of reproducing one of the social values important to them: egality. 408 Reference List $EGHO$]L]<DQG.DUDUD+0'LUHFWOLQHDUWUDQVIRUPDWLRQIURP FRPSXWHUFRRUGLQDWHVLQWRREMHFWVSDFHFRRUGLQDWHVLQFORVHUDQJH photogrammetry. In: 3URFHHGLQJVRIWKH$638,6\PSRVLXPRQ&ORVH5DQJH Photogrammetry. Falls Church, Va.: American Society of Photogrammetry. Adams W. I. and Adams E. W. 1991 Archaeological typolog y and practical reality : DGLDOHFWLFDODSSURDFKWRDUWLIDFWFODVVLÀFDWLRQDQGVRUWLQJ Cambridge: Cambridge University Press. cNHUOXQG$*XVWDIVVRQ3+DPPDU'/LQGJUHQ&2OVVRQ(DQG:LNHOO R. 2003. Peopling a forgotten landscape. In: Larsson, L., Kindgren, H., .QXWVVRQ./RHIÁHU'DQGcNHUOXQG$ HGV Mesolithic on the Move. Papers presented at the Sixth International Conference on the Mesolithic in Europe, Stockholm Oxford: Oxbow Books. $NHUPDQ.%DOHUVKHOOLPSOHPHQWVIURPQRUWKZHVW$XVWUDOLDMankind. — 1976. An analysis of stone implements from Quondong, W.A. Occasional Papers in Anthropolog y$QWKURSRORJ\0XVHXP8QLYHUVLW\RI4XHHQVODQG ³1RWHVRQWKH.LPEHUOH\VWRQHWLSSHGVSHDUIRFXVVLQJRQWKHKDIWLQJ mechanism. Mankind — 1979a. The renascence of Aboriginal law in the Kimberley. In: R.M and C.H. Berndt (eds). Aborigines of the West. Their Past and Present. University of Western Australia Press, Western Australia. — 1979b. Heat and lithic technology in the Kimberleys, Western Australia. Archaeolog y and Physical Anthropolog y in Oceania — 1979c. Honey: an overview of its role in the life of the Aboriginals of the Kimberleys. Oceania — 1979d. Flaking stone with wooden tools, The Artefact 4/3and4: 79–80. — 1979e. An unusual stone chisel from Cherabun Station, West Kimberleys, Western Australia. The Artefact. DQG — 1993. The status of the horsehoof core. Records of the Australian Museum. 6XSSOHPHQW — 1995. The use of bone, shell, and teeth by Aboriginal Australians. In: Johnson, E. (ed.). Ancient peoples and landscapes. Lubbock, Texas: Museum of Texas Tech University. — and Bindon, P. 1983. Evidence of Aboriginal lithic experimentation on the Dampierland Peninsula. In: Smith, M. (ed.). $UFKDHRORJ \DW$1=$$6. Proceedings of Section 25A, Western Australian Museum, Perth. ³DQG%LQGRQ37KHHGJHJURXQGVWRQHDG]HDQGPRGHUQFRXQWHUSDUWV in the Kimberley Region, Western Australia. Records of the Western Australian Museum 409 Skilled Production and Social Reproduction — and Bindon, P. 1995. Dentate and related stone biface points from northern Australia, The Beagle, Records of the Museum and Art Gallery of the Northern Territory. Darwin, Northern Territory. — Fullagar, R. and van Gijn, A. 2002. Weapons and Wunan: production, function and exchange of Kimberley Points. Australian Aboriginal Studies. Canberra. — and Stanton, J. 1994. Riji and Jakuli: Kimberley pearl shell in Aboriginal Australia. Northern Territory Museum of Arts and Sciences, Monograph Series No 4. Darwin, Northern Territory. — (n.d). The use of bone, shell and teeth by Aboriginal Australians. Unpublished ms. Alakärppä, J., Ojanlatva, E. and Ylimaunu, T. 1997. Raw material sources and use RITXDUW]LQWKH.HPL7RUQLRDUHDQRUWKRIWKH*XOIRI%RWKQLD,Q+ROP L. and Knutsson, K. (eds). Proceedings from the Third Flint Alternatives Conference DW8SSVDOD6ZHGHQ2FWREHU Occasional Papers in Archaeology 16, Uppsala. Aldred, C. 1965. Eg ypt to the end of the Old Kingdom1HZ<RUN0F*UDZ+LOO Alix, P,. Pelegrin, J., and Deloge, H. 1995. Un débitage original de lamelles par SUHVVLRQDX0DJGDOpQLHQGX5RFKHUGHOD&DLOOH /RLUH)UDQFH Paléo 199. Althin, C. A. 1954. The chronology of the Stone Age settlement of Scania, Sweden. I The Mesolithic settlement. Acta Archaeologica Lundensia. Series in 4°, N° 1. Lund, Bonn. Anderson, D. D. 1970. Akmak. an early archaeological assemblage from Onion Portage, Northwest Alaska. Acta Artica SODWHV Andersson, R., Boman, K. and Borbás, I. 1997. Lyckad nedfrysning av herr Moro. Stockholm: Ordfront. Andersson, S. 1985. Tingens psykoanalys. In: Andersson, S., Johansen, T., Nilsson, G. and Österberg, D. (eds.). Mellan människor och ting. Göteborg: Korpen. Andrefsky, W. Jr. 1994: The Geological occurrence of lithic material and stone tool production strategies. Geoarchaeolog y: An International Journal 9/5: 375–391. Andrews, B. W. 1999. Craftsman skill and specialization:Iinvestigating the craft production of Prehispanic obsidian blades at Xochicalco and Teotihuacan, Mexico. Ph.D. dissertation, Department of Anthropology, Pennsylvania State University, University Park. — 2002. Stone tool production at Teotihuacan: What more can we learn from surface collections? In: Hirth, K. and Andrews, B. (eds). Pathways to Prismatic Blades: A Study in Mesoamerican Core-Blade Technolog y. Cotsen Institute of Archaeology, University of California, Los Angeles. — 2003. Measuring prehistoric craftsman skill: Contemplating its application WR0HVRDPHULFDQFRUHEODGHUHVHDUFK,Q+LUWK. HG Experimentation and Interpretation in Mesoamerican Lithic Technolog y. Salt Lake City: University of Utah Press. Apel, J. 2000. Kunskap, handlag och råmaterial: en discussion kring produktionen RFKNRQVXPWLRQHQDYVHQQHROLWLVNDÁLQWGRONDUTor 410 Reference List — 2001a. 'DJJHUVNQRZOHGJHDQGSRZHU7KHVRFLDODVSHFWVRIÁLQWGDJJHUWHFKQRORJ \LQ 6FDQGLQDYLDFDO%&. Coast to Coast Books 3. Uppsala: Department of Archaeology, Uppsala University. — 2001b. Några formella konventioner och ett halvdant teknologiskt handlag. In: Bolin, H., Kaliff, A. and Zachrisson, T. (eds). 0HOODQVWHQRFKEURQV8SSGUDJVDUNHRORJLRFK forskning kring senneolitikum och bronsålder. OPIA 27 and SAR 39 2001. Stockholm and Uppsala: University of Uppsala, Dept of Archaeology. Appadurai, A. 1986. Introduction: Commodities and the politics of value. In: Appadurai, A. (ed.) 1986. The social life of things: Commodities in cultural perspective. Cambridge: Cambridge University Press. Arnold, V. 1981a7HJHOEHUJHLQHVSlWQHROLWKLVFKH2VWVHHNVWHQVWDWLRQXQGLKUH Flintteknik. 8QLYHUVLWlW]X.|OQ, Köln. — 1981b. Ein aus Schlagabfällen rekonstruierbarer Flintdolch vom Tegelberg, *HPHLQGH4XHUQ.UHLV6FKOHVZLJ)OHQVEXUJOffa 38: ³5HÀWWLQJRIZDVWHPDWHULDOIURPGDJJHUSURGXFWLRQRI6LWH7HJHOEDUJ In: Czieslo, E., Eickhoff, N., Arts, N. and Winter, D. (eds). The Big Puzzle. ,QWHUQDWLRQDO6\PSRVLXPRI5HÀWWLQJ6WRQH$UWHIDFWV0RQ5HSRV. Holos: Bonn. $UWXUVVRQ0 HG 6WnQJE\VWDWLRQVVDPKlOOH%RSODWVRFKEHE\J gelselämningar från senneolitikum till yngre järnålder. Skåne, Vallkärra sn, väg 930. Arkeologisk förundersökning och undersökning. 5 $b8Y6\G5DSSRUW 2000:79. Lund. Arwidsson, H. 2003. Stiger vi op mod lyset? Om upplysning og romantik. I: Kruse, T. E. (ed). Historiske kulturstudier. Tradition-modernitet-antimodernitet. Roskilde: Roskilde universitetsforlag Routledge,. Atzeni, E. (Dir.) 2000. /HFROOH]LRQLOLWLFKHSUHLVWRULFKHGHOO·8QLYHUVLWjGL&DJOLDUL. Cagliari: Edizioni AV. $XGRX]H)/HURL*RXUKDQDSKLORVRSK\RIWHFKQLTXHDQGHYROXWLRQ Journal of Archaeological Research Axel, B. K. 2002. From the margins: historical anthropolog y and its futures. Durham: Duke University Press. Bachelard, G. 1984. 7KHQHZVFLHQWLÀFVSLULW Boston: Beacon Press. Balcer, B. 1975. Bemerkungen zur Feuersteinbearbeitung in der Kugelamforenkultur in Polen, in: Archaeologia Polona;9,, ³$6WXG\RIVRFLRHFRQRPLFDVSHFWVRI1HROLWKLFÁLQWZRUNLQJRQWKH example of the Funnel Beaker Culture (FBC) in: 8QFRQYHQWLRQDO$UFKDHRORJ \, 6FKLOG5 HG :URFâDZ:DUV]DZD.UDNyZ*GDľVN — 1983. :\WZyUF]RœýQDU]ęG]LNU]HPLHQQ\FKZQHROLFLH]LHP3ROVNL:URFâDZ:DUV]DZD .UDNyZ*GDľVNáyGŭ — 2002. Cmielow, Krzemionki, swiecichow. Zwiazki osady neolitycznej z kopalniami krzemienia. Zwiazki osady neolitycznej z kopalniami krzemienia. Wa,, Wyd. Inst. Archeologii PAN, Br., 190 S., 4°, Tab. Balfour, H. R. 1903. On the method employed by the natives of N.W. Australia in WKHPDQXIDFWXUHRIJODVVVSHDUKHDGV. Man 35: 65. — 1951. A native tool kit from the Kimberley District, Western Australia. Mankind 411 Skilled Production and Social Reproduction Balme, J. 2000. Excavations revealing 40,000 years of occupation at Mimbi Caves, south central Kimberley, Western Australia. Australian Archaeolog y %DQJ$QGHUVHQ6:K\DOOWKHVHZKLWHDQGVKLQ\VWRQHV"2QWKHRFFXUUHQFH RIQRQÁLQWOLWKLFPDWHULDORQ0HVROLWKLFLQODQGVLWHVLQVRXWKZHVWHUQ Norway. In: Holm, L. and Knutsson, K. (eds). Proceedings from the Third Flint $OWHUQDWLYHV&RQIHUHQFHDW8SSVDOD6ZHGHQ2FWREHU Occasional Papers in Archaeology 16. Uppsala. Banks, E. 1967. The Early and Middle Helladic small objects from Lerna. Part 1-II. Unpublished dissertation, University of Michigan, Ann Arbor. Barndon, R. 2004. A Discussion of magic and medicines in East African iron working: Actors and artefacts in technology. Norwegian Archaeolog y Review 37: Barnes, A. S. 1947. The technique of blade production in Mesolithic and Neolithic Times. Proceedings of the Prehistoric Society %DUUHWW-7LPHDQGWUDGLWLRQWKHULWXDOVRIHYHU\GD\OLIH,Q1RUGVWU|P+ cDQG.QDSH$ HGV Bronze Age Studies. Transactions of the British-Scandinavian &ROORTXLXPLQ6WRFNKROP0D\² Studies 6, The Museum of National Antiquities, Stockholm. — 1994. )UDJPHQWVIURPDQWLTXLW\$QDUFKDHRORJ \RIVRFLDOOLIHLQ%ULWDLQ%& Oxford and Cambridge. — 2001. Agency, the duality of structure, and the problem of the archaeological record. In: Hodder (ed). Archaeological theory today. Cambridge: Polity Press.. Basedow, H. 1925 The Australian Aboriginal, Adelaide: Preece and Sons. Baudou, E. 1995. Norrlands forntid. Ett historiskt perspektiv. Umeå: Cewe förlaget. Bauman, Z. 1990. Thinking sociologically. Oxford: Blackwell %HFN&:6SHFWURVFRSLFLGHQWLÀFDWLRQRI´DPEHUµDQG´EODFNUHVLQµIURP Asine. In: Hägg, R., Nordquist, G. and Wells, B. (eds.) Asine III. Supplementary VWXGLHVRQWKH6ZHGLVKH[FDYDWLRQV (Acta Instituti Atheniensis Regni 6XHFLDH;/9 6WRFNKROP Bender, B. 1993 (ed.). Landscape: politics and perspectives. Providence/Oxford, Berg Publishers. — 2001. Introduction. In: Bender, B. and Weiner, M. (eds). Contested landscapes. movement, exile, place. Oxford: Berg Publishers. Bennett, A. 1984. Karleby och Gärtuna. Bebyggelse och gravar från bronsålder och järnålder i Östertälje socken Södermanland. Riksantikvarieämbetet Rapport 89 1984:29. Stockholm. Berg, E. 2003. The spatial and chronological development of the Late Mesolithic Nøstvet period in coastal southeastern Norway from a lithic raw material SHUVSHFWLYH,Q/DUVVRQ/.LQGJUHQ+.QXWVVRQ./RHIÁHU'DQG cNHUOXQG$ HGV Mesolithic on the Move. Papers presented at the Sixth International &RQIHUHQFHRQWKH0HVROLWKLFLQ(XURSH6WRFNKROPOxford: Oxbow Books. %HUJUHQcDQG&HOLQ8%XUO|Y&gUHVXQGVI|UELQGHOVHQ5DSSRUW|YHU arkeologisk slutundersökning. Rapport 36. Malmö Kulturmiljö. 412 Reference List Beronius, M. 1991. Geneaologi och sociologi: Nietsche, Foucault och den sociala analysen. Stockholm/Stehag: Brutus Östlings bokförlag. Binford. L. R. 1978. Nunamiut ethnoarcaheology. Studies in Archeolog y. New York: Academic Press. — 1983. Working at archaeology. Studies in Archaeolog y. New York: Academic Press. %LUG-$&RPSDULVRQRI6RXWK&KLOHDQDQG(FXDWRULDO´ÀVKWDLOµSURMHFWLOH points. The Kroeber Anthropological Society Papers ³3DOHR,QGLDQGLVFRLGDOVWRQHVIURPVRXWKHUQ6RXWK$PHULFDAmerican Antiquity — 1988. Travels and archaeolog y in South Chile, edited by John Hyslop. Iowa City: University of Iowa Press. — and Bird, M. 1937. Human artifacts in association with horse and sloth bones in southern South America (Magallanes). Science %LVFKRII(DQG&DQpWR)0RQXPHQWVGHO·$JHGH3LHUUHHWGHODSpULRGH *DOORURPDLQHGDQVODYDOOpHGX*HUVRevue de Gascogne9, Björck, N. 2003. The Neolithic coastal settlements: cosmology and ideology in a Baltic Sea perspective. In: Samuelsson, C. and Ytterberg, N. (eds). 8QLWLQJ 6HD6WRQH$JH6RFLHWLHVLQWKH%DOWLF6HD5HJLRQ3URFHHGLQJVIURPWKH)LUVW8QLWLQJ6HD :RUNVKRSDW8SSVDOD8QLYHUVLW\6ZHGHQ-DQXDU\ Occasional Papers in Archaeology 33. Uppsala. Björhem, N. and Säfvestad, U. 1989. )RVLH,9%\JJQDGVWUDGLWLRQRFKERVlWWQLQJVP|QVWHU under senneolitikum. Malmöfynd 5. Malmö. Blegen, C. W., Caskey, J. L., Rawson, M. and Sperling, J. 1950: 7UR\WKHÀUVWDQG second settlements. Princeton. Blinkenberg, C. 1904. Archaeologische Studien. Copenhagen: Carlsbergfondet; Gyldendal: Nordisk forlag; Leipzig: O. Harrassowitz. Bloch, M. 1989. Ritual, history and power. Selected papers in anthropolog y. Monographs on Social Anthropology 58. London and Atlantic Highlands, N. J.:Athlone Press. — 1992. Prey into hunter:The politics of religious experience. Cambridge: Cambridge University Press. — and Parry, J. (eds.) 1982. Death and the regeneration of life. Cambridge: Cambridge University Press. Bodu, P., Karlin, C. and Ploux, S. 1990. Who´s who? The Magdalenian ÁLQWNQDSSHUVRI3LQFHYHQW )UDQFH ,Q&]LHVOD((LFNKRII6$UWV1 and Winther, D. (eds). 7KH%LJ3X]]OH,QWHUQDWLRQDO6\PSRVLXPRQ5HÀWWLQJ6WRQH $UWHIDFWV0RQ5HSRV Bonn: Holos. Boëda, E. 1988. Le concept Levallois et évaluation de son champ d´application. In: Otte, M. (ed). 1988. L´homme de neandertal. vol 4. Liège: Université de Liège. — 1993. Le debitage discoïde et le debitage Levallois récurrent centripète. Bulletin de la Société Préhistorique Française — 1994. Le concept Levallois: variabilité des méthodes. Paris: CNRS. Bonnichsen, R. and Morlan, R. 1974. Experiments in lithic technolog y: A cognitive approach to bifaces. Paper presented at the Symposium on Primitive Technology DQG$UW8QLYRI&DOJDU\$OEHUWD1RY 413 Skilled Production and Social Reproduction Bordaz, J. 1959a. First tools of mankind. Natural History /;9,,, — 1959b. The New Stone Age. Natural History/;9,,, — 1970. Tools of the Old and New Stone Age. Garden City, N.Y.: Natural History Press. Bordes, F. 1961.Typologie du Paleolitique Ancien et Moyen.3XEOLFDWLRQVGHO·,QVWLWXWGH 3UHKLVWRLUHGHO·8QLYHUVLWHGH%RUGHDX[0HPRLUH1R%RUGHDX['HOPDV — 1968. La question périgordienne. In: La Préhistoire: problèmes et tendances. Paris: &156 — 1968. The Old Stone Age. London: World University Library. ³5HÁHFWLRQVRQW\SRORJ\DQGWHFKQLTXHVLQWKH3DOHROLWKLFArctic Anthropolog y9, — and Crabtree, D. 1969. The Corbiac blade technique and other experiments. Tebiwa Borrero, L. A. 1986. Cazadores de Mylodon en la Patagonia Austral. In: Bryan, A. (ed.) New Evidence for the Pleistocene Peopling in the AmericasS2URQR Center for the Study of the Early Man. — Lanata, J. L. and Borella, F. 1988. Reestudiando huesos: Nuevas consideraciones sobre sitios de Ultima Esperanza. Anales del Instituto de la Patagonia, Serie Ciencias Sociales Boschian, G., Brilli, P., Falchi, P., Fenu, P., Martini, F., Pitzalis, G., Sarti, L. and 7R]]L&3ULPHULFHUFKHQHOO·DELWDWRQHROLWLFRGL&RQWUDJXGD 3HUIXJDV Sassari). Rivista di Scienze Preistoriche Bourdieu, P. 1977. Outline of a theory of practice. Cambridge: Cambridge University Press. — 1984. Distinction: A social critique of the judgement of taste. Cambridge, Mass.: Harvard University Press. — 1987. What makes a social class? On the theoretical and practical existence of groups. Berkeley Journal of Sociolog y — 1996. Praktiskt förnuft. Göteborg: Daidalos. — 2001. 6FLHQFHGHODVFLHQFHHWUpÁH[LYLWp3DULV5DLVRQVG·DJLU — 2004. 6FLHQFHRIVFLHQFHDQGUHÁH[LYLW\ Chicago: University of Chicago Press. ³&KDPERUHGRQ-&DQG3DVVHURQ-&The craft of sociolog y: epistemological preliminaries. Berlin: de Gruyter. Bouzek, J. 1985. The Agean, Anatolia and Europe: Cultural intereactins in the second millennium B.C. Studies in Mediterranean Archaeolog y Vol. 29, Göteborg: 3DXOcVWU|PV)|UODJ — 1997. Greece, Anatolia and Europe: Cultural interrelations during the Early Iron Age. Studies in Mediterranean Archaeolog y 9RO*|WHERUJ3DXOcVWURPV Forlag. %RZGOHU6DQG2·&RQQRU67KHGDWLQJRIWKH$XVWUDOLDQVPDOOWRRO tradition, with new evidence from the Kimberley, WA. Australian Aboriginal Studies %UDGOH\%3UHG\QDVWLF(J\SWLDQÁLQWLPSOHPHQWV$QLQGXFWLYH technological sequence. Newsletter of Lithic Technolog y, 414 Reference List ³$WHFKQLTXHRIGUDZLQJÁDNHGVWRQHNewsletter of Lithic Technolog y III/3: 53 – 55. — 1982. Flaked stone technology and typology. In: Frison, G. and Stanford, D. (eds). The Agate Basin Site. A record of the Paleoindian occupation of the Northwestern High Plains. New York: Academic Press. ³DQG6WDQIRUG'7KH1RUWK$WODQWLFLFHHGJHFRUULGRUDSRVVLEOH Paleolithic route to the New World. World Archeolog y Bradley, R. 2000. An archaeolog y of natural places. London: Routledge. — and Williams, H. 1998 (eds..) The past in the past. World Archaeolog y 30: 1. Broadbent, N. 1979: Coastal resources and settlement stability. A critical study of a Mesolithic site complex in northern Sweden. Aun 3. Societas Archaeologica Upsaliensis. — and Knutsson, K. 1975. An experimental analysis of quartz scrapers. Results and applications. FornvännenSS ³DQG.QXWVVRQ.1nJUDUHÁHNWLRQHUNULQJH[SHULPHQWHOODUNHRORJLRFK dess tillämpning i Norden. Tor;9,,,SS Broady, D. 1989. Kapital, habitus, fält. Några nyckelbegrepp i Pierre Bourdieus sociologi8+bIR8$UEHWVUDSSRUW 1989:2. — 1991. Sociologi och epistemolog y. Om Pierre Bourdieus författarskap och den historiska epistemologin. Stockholm: HLS Förlag. Brøndstedt, J. 1957. Danmarks oldtid I. Stenalderen. Copenhagen. Brown, J. 1995. Traditional Metalworking in Kenya. Cambridge Monograph in African Archaeology, 38; Oxbow Monograph 44. Oxford, Oxbow Books. %XFKYDOGHN0%HPHUNXQJHQ]XP$KRULVRQWLQ0LWWHOHXURSD. In: Siemens, P. (ed). (DUO\&RUGHG:DUH&XOWXUH7KH$KRUL]RQ²)DFWRU)LFWLRQ Arkeologiske Rapporter fra Esbjergs Museum. Esbjerg. %XG]LV]HZVNL-)OLQWPDWHULDOVIURPFHPHWHULHVRIWKH6RÀHYNDW\SH,Q &HPHWHULHVRIWKH6RÀHYNDW\SH%&, Baltic – Pontic Studies 3. Warszawa. Burström, M. 1999. Cultural diversity in the home ground: How archaeology can make the world a better place. Current Swedish Archaeolog y Burton, J. 1984. Quarrying in a tribal society. World Archaeolog y Buschan, G. 1922. (ed.) ,OOXVWULHUWH9|ONHUNXQGH. Stuttgart. Butler, J. 1993. Bodies that matter. London: Routledg Callahan, E. 1973. The Old Rag Report: a practical guide to living archeolog y. Dept. of Sociology and Anthropology, Virginia Commonwealth University, Richmond. (Revised in 2005 for republication.) — 1974. (ed.). Experimental Archeolog y Papers, 3. Dept. of Sociology and Anthropology, Virginia Commonwealth University, Richmond, Va. — 1974. 9DULDWLRQDQGLQWHUSUHWDWLRQRI9LUJLQLDÁXWHGSRLQWV Unpublished paper. — 1975. The non-returning boomerang: Evolution and experiment. Unpublished paper. — 1976a. A Lithic Workshop Symposium. Newsletter of Lithic Technolog y9 — 1976b. (ed.). Experimental Archeolog y Papers 4. Dept. of Sociology and Anthropology, Virginia Commonwealth University, Richmond, Va. — 1979. Craftsman Don Crabtree. Flintknappers Exchange 415 Skilled Production and Social Reproduction ³7KHEDVLFVRIELIDFHNQDSSLQJLQWKHHDVWHUQÁXWHGSRLQWWUDGLWLRQ$ PDQXDOIRUÁLQWNQDSSHUVDQGOLWKLFDQDO\VWVArchaeolog y of Eastern North America ³$QLQWHUYLHZZLWKÁLQWNQDSSHU-DFTXHV3HOHJULQContract Abstracts and CRM Archaeolog y ³'DQLVK'DJJHU$)OLQWNQDSSHUV([FKDQJH . Albuquerque, New Mexico: Atechiston. — 1985 A successful test model of the Type 1 Dagger. Unpublished manuscript available through Piltdown Productions, Lynchburg, Va. — 1985. Experiments with Danish Mesolithic microblade technology. Journal of Danish Archaeolog y ³7KHEDVLFVRIELIDFHNQDSSLQJLQWKHHDVWHUQÁXWHGSRLQWWUDGLWLRQ nd ed.). Archaeolog y of Eastern North America — 1987. An evaluation of the lithic technolog y in Middle Sweden during the Mesolithic and Neolithic. Aun 8. Uppsala: Societas Archaeologica Upsaliensis. — 1993 (cont). Piltdown Productions Catalog. Lynchburg, Va. Piltdown Productions. — 1995. Blades from Middle Neolithic Battle Axe Culture graves in Sweden: a preliminary technological study. In: Kuntsson, H. Slutvandrat? Aspecter på gYHUJnQJHQIUnQ5|UOLJWLOO%RIDVW7LOOYDUR Appendix 2. Aun 20. Uppsala: Societas Archaeologica Upsaliensis. — 1999; 1991. Cliffside Workshop sponsored by Piltdown Productions, Richmond, Virginia. ³7KHEDVLFVRIELIDFHNQDSSLQJLQWKHHDVWHUQÁXWHGSRLQWWUDGLWLRQ$ Manual for Flintknappers and Lithic Analysts. 4th ed. [1st ed. 1979, 2nd ed. 1990, 3rd ed. 1996]. Lynchburg, Va: Piltdown Productions. — 2001. Archaeological evidence of rotator cuff injury. Bulletin of Primitive Technolog y — 2002. The Tsirk challenge. Bulletin of Primitive Technolog y 24: 84. — 2002. Dagger Research Workshop. Dagger Research Team, Lynchberg, Virginia. — 2003. $SSUHQWLFHVKLSVWDJLQJDQGVRFLDOLQÁXHQFHLQ'DQLVK'DJJHUSURGXFWLRQ. Manuscript. — Apel J. and Olausson, D. in prep. Neolithic Danish Daggers: an experimental and analytical study. Manuscript. — Forsberg, L., Knutsson, K. and Lindgren, L. 1992. Frakturbilder. Kulturhistoriska kommentarer till det säregna sönderfallet vid bearbetning av kvarts. Tor 24: 27–63. &DQWHW-3/H&KDOFROLWKLTXHHWO·$JHGX%URQ]HHQ*DVFRJQHJHUVRLVH. Périgueux éd. 9pVXQD FROOHFWLRQ©$UFKpRORJLHVªGLU&&KHYLOORWYRO &DUGLFK$00DQVXU)UDQFKRPPH(*LHVVR0DQG'XUiQ9$ Arqueología de las Cuevas de “El Ceibo”. Provincia de Santa Cruz, Argentina. Relaciones de la Sociedad Argentina de Antropología;,9%XHQRV$LUHV Carlie, A. 1999. “Sacred white stones.” On traditions of building white stones into graves. Lund Archaeological Review 5: 41–58. 416 Reference List Carr, E. H. 1961. What is History? New York: Vintage. Cassirer, E. 1951. The Philosophy of the Enligthenment. Princeton: Princeton University Press. Cattaneo, G. R. 2002. Conjuntos instrumentals líticos durante la transición Pleistoceno/Holoceno en el maciso del Deseado. Paper presented at Quintas Jornadas de Arqueología de la Patagonia, Buenos Aires. &HGHUKYDUI%1HROLWLVNDOHUÀJXUHUIUnQcODQGFinska Fornminnesföreningens årsskrift &HOV$DQG'H3DXZ/&RQVLGpUDWLRQVVXUODWDLOOHGXVLOH[WHOOHTX·HOOH pWDLWSUDWLTXpHj6SLHQQHVjO·kJHGHODSLHUUHSROLHBulletin de la Société Anthropologique de Bruxelles Chabas, F. 1874. /HV6LOH[GH9ROJX5DSSRUWjOD6RFLHWHG·+LVWRLUHHWG·$UFKHRORJLH GH&KDORQVXU6D{QH&KDORQVXU6D{QH Chabot, J. 2002. 7HOO·$WLM7HOO*XGHGD,QGXVWULHOLWKLTXH$QDO\VHWHFKQRORJLTXHHW IRQFWLRQQHOOH$YHFXQHDQQH[HGH-DFTXHV3HOHJULQ &156)UDQFH VXUODUHFRQQDLVVDQFH du débitage par pression au levier.8QLYHUVLWp/DYDO&(/$7 &DKLHUVG·DUFKpRORJLH du CELAT 13, Série archéométrie 3). &KD]DQ05HGHÀQLQJ/HYDOORLVJournal of Human Evolution 33: 719–735 Cherry, J. F. and Renfrew, C. 1986. Epilogue and prospect. In: Renfrew, C. and Cherry, J. F. (eds). Peer Polity Interaction and Sociopolitical Change. Cambridge: Cambridge University Press. Chomsky, N. 1957. Syntactic Structures. The Hague: Mouton. Clark, G. 1970. The Stone Age Hunters. New York: McGraw Hill. &ODUN-()URPPRXQWDLQVWRPROHKLOOV$FULWLFDOUHYLHZRI7HRWLKXDFDQ·V obsidian industry. In: Isaac, B. (ed). Economic Aspects of Prehispanic Highland MexicoSS5HVHDUFKLQ(FRQRPLF$QWKURSRORJ\6XSSOHPHQW Greenwich, Conn: JAI Press. ³3ULVPDWLFEODGHPDNLQJFUDIWVPDQVKLSDQGSURGXFWLRQ$QDQDO\VLVRI obsidian refuse from Ojo de Agua, Chiapas, Mexico. Ancient Mesoamerica 8: — 2003. Craftsmanship and Craft Specialization. In: Hirth, K. (ed). Experimentation and Interpretation in Mesoamerican Lithic Technolog y. Salt Lake City: University of Utah Press. — and Bryant, D. D. 1997. A technological typology of prismatic blades and debitage from Ojo de Agua, Chiapas, Mexico. Ancient Mesoamerica — and Parry, W. J. 1990. Craft specialization and cultural complexity. Research in Economic Anthropolog y Clement, E. 1903. Ethnographical notes on the Western Australian Aborigines, with a descriptive catalogue of a collection of ethnographical objects from Western Australia by Schmeltz, J.D.E. Internationales Archiv fur Ethnographie. /HLGHQ Coe, W. R. 1959. Piedras Negras archeolog y: Artifacts, caches, and burials. Museum monograph. University Museum, University of Pennsylvania. 417 Skilled Production and Social Reproduction Collins, M. B. 1999. Clovis and Folsom lithic technology on and near the Southern Plains: Similar ends, different means. In: Amick, D. S. (ed.) Exploring 3DWWHUQDQG9DULDWLRQLQ)ROVRP/LWKLF7HFKQRORJ \/DWH3OHLVWRFHQH+XQWHU*DWKHUHUV of the North American High Plains, International Monographs in Prehistory, Archaeological Series 12. Ann Arbor: University of Michigan. Conkey, M. 1991. Humans as materialists and symbolists: Image making in the Upper Palaeolithic of Europe. In: Gero, J. M. and Conkey, M. W. (eds). Engendering Archaeolog y: Women and Prehistory. Basil Blackwell, Oxford. Cooney, G. 1998. Breaking stones, making places: The social landscape of axe production sites. In: Gibson, A. and Simpson, D. (eds.) Prehistoric ritual and religio:. Essays in honour of Aubrey Burl. Thrupp, Stroud, Gloucestershire: Sutton. Cooper, H. M. 1943. Large stone implements from South Australia. Records of the South Australian Museum$GHODLGH Cornelissen, E. 2003: On microlithic quartz industries at the end of the Pleistocene in Central Africa: The evidence from Shum Laka (NW Cameroon). African Archaeological Review9RO &RVWD/-DQG3HOHJULQ-8QHSURGXFWLRQGHJUDQGHVODPHVSDUSUHVVLRQj ODÀQGX1pROLWKLTXHGDQVOHQRUGGHOD6DUGDLJQH &RQWUDJXGD3HUIXJDV Bull. de la Société Préhistorique FrançaiseWRPHQS &RVWLQ&/&UDIWVSHFLDOL]DWLRQLVVXHVLQGHÀQLQJGRFXPHQWLQJDQG explaining the organization of production. Archaeological Method and Theory 56. Costin, C. L. n.d. The study of craft production. In: Mauschner, H. and Chippindale, C. (eds). Handbook of Archaeological Methods. Lanham, Md: Alta Mira Press. &RXOVRQ65HÀWWHGÁLQWQRGXOHVIURP6RQJD7HOHPDUN8QLYHUVLWHWV Oldsakssamling Årbok 1984/1985. Oslo. Courtin, J. 1974. Le Néolithique de la Provence. Paris : Klincksieck, 359 p. Mémoires de la Soc. Préhist. Française, n°11. &UDEWUHH'($VWRQHZRUNHU·VDSSURDFKWRDQDO\]LQJDQGUHSOLFDWLQJWKH Lindenmeier Folsom. Tebiwa — 1970. Flaking stone with wooden implements. Science ³$Q,QWURGXFWLRQWRÁLQWZRUNLQJOccasional Papers of the Idaho State 8QLYHUVLW\Pocatello: Idaho State University — 1973. Experiments in replicating Hohokam points. Tebiwa 16:24 Crown, P. L. 2001: Learning to Make Pottery in the Prehispanic American Southwest. Journal of Anthropological Research 57/4: 451–469. &\UHN.1LH]QDQH]DE\WNL]JUREXZ-DQLVâDZLFDFKZRM6NLHUQLHZLFNLHL QRZHREVHUZDFMHQDGW\P]HVSRâHP:LDGRPRœFL$UFKHRORJLF]QH ³8]\VNLZDQLHLXů\WNRZDQLHVXURZFyZNU]HPLHQQ\FKZPH]ROLFLH GRU]HF]\:LVâ\LJyUQHM:DUW\3UDFHL0DWHULDâ\0$L(6HULD$UFKHRORJLF]QD 28: ³2QWKHGLVWULEXWLRQRIFKRFRODWHÁLQWLQWKH/DWH0HVROLWKLFRIWKH Vistula basin, Archaeologia Polona 418 Reference List Damm, C. 1998. Forhistoriske ritualer: En diskussion omkring mening og handling. In: Bredholt Christensen, L. and Benedicte Sveen, S. (eds.) Religion og materiel kultur. Aarhus. Dapena, J. 1985. Computers in biomechanics. On line — 1993. 0HFKDQLFVRIWKHDUPVZLQJGXULQJÁDNLQJ. Unpublished report. — Anderst, W. J. and Toth, N. (in press). Mechanics of the arm Swing in stone ÁDNLQJ,Q7RWK1DQG6FKLFN.The Origins of Human Technolog y: Studies into WKH(DUO\6WRQH$JH 2OGRZDQ . Bloomington, Ind: CRAFT Press. Darmark, K. 2001. Att stycka en sten. Kvartskeratofyr som källmaterial. MA Thesis. Institute of Archaeology, Uppsala University, Uppsala. Darvill, T. 2002: White on blonde: quartz pebbles and the use of quartz at Neolithic monuments in the Isle of Man and beyond. In: Jones, A. and MacGregor, G. (eds.) &RORXULQJWKHSDVWWKHVLJQLÀFDQFHRIFRORXULQFUFKDHRORJLFDO research. (Conference of the European Arcaheological Association, Bournemouth, England), 1999. Oxford: Berg. 'DXQcDet allmänmänskliga och det kulturbundna. Stockholm: Prisma. Davidson, D. S. 1938. Stone axes of Western Australia. American Anthropologist 40: 'HODQW\*$OOWVRPlUIDVWI|UÁ\NWLJDV0RGHUQLVPRFKPRGHUQLWHW Cambridge: Polity Press. 'HOFRXUW9ODHPLQFN0/HVLOH[GX*UDQG3UHVVLJQ\GDQVOHQRUGRXHVWGH O·(XURSHBulletin des amis du Musée du Grand-Pressigny Descola, P. 1994. In the society of nature: a native ecolog y in Amazonia. Cambridge: Cambridge University Press. 'LEEOH+/DQG%DU<RVHI2 HGV 7KHGHÀQLWLRQDQGLQWHUSUHWDWLRQRI/HYDOORLV technolog y, Monographs in World Archaeology 23, Madison: Prehistory Press. ³DQG3HOFLQ-7KHHIIHFWRIKDPPHUPDVVDQGYHORFLW\RQÁDNHPDVV Journal of Archaeological Science Dickson, F. P. 1981. Australian stone hatchets: A study in design and dynamics. Sydney: Academic Press.. 'REUHV0$*HQGHUDQGSUHKLVWRULFWHFKQRORJ\2QWKHVRFLDODJHQF\RI technical strategies. World Archaeolog y — 2000. Technolog y and social agency: Outlining a practice framework for archaeolog y. Oxford: Blackwell. 'RPDľVND/2ER]RZLVNRNXOWXU\MDQLVâDZLFNLHMZ'ęEDFKZRM :âRFâDZVNLHVWDQRZLVNR3R]QDľ,QRZURFâDZ 'RPDľVNL0:HEE-$)ODNLQJSURSHULWLHVSHWURORJ\DQGXVHRI3ROLVK ÁLQWAntiquity Donham, D. L. 1995. History, power, ideology: Central issues in Marxism and anthropology. Studies in Marxism and Social Theory. Cambridge: Cambridge University Press and Paris: Editions de la Maison des Sciences de l´Homme. Dörpfeld, W. 1902. 7URMDXQG,OLRQ,(UJHEQLVVHGHU$XVJUDEXQJHQLQGHU9RUKLVWRULVFKHQ XQG+LVWRULVFKHQ6FKLFKWHQYRQ,OLRQ². Athens. 419 Skilled Production and Social Reproduction Dortch, C. E.1977a. Ancient grooved stone axes from an alluvial terrace on Stonewall Creek, Kimberley, Western Australia. Journal of the Royal Society of Western Australia —1977b. Early and late stone industrial phases in Western Australia. In: Wright, R.V.S. (ed). Stone Tools as Cultural Markers: Change, Evolution and Complexity, Camberra: Australian Institute of Aboriginal Studies. — and Bordes, F. 1977. Blade and Levallois technology in Western Australian prehistory. Quartar — and Roberts, R. G. 1996. An evaluation of radiocarbon chronologies at Miriwun rock shelter and the Monsmont site, Ord Valley, east Kimberley, Western Australia. Australian Archaeolog y 'UHLMHU0cODQGVlOGVWDEHE\JJHOVHFinskt Museum Drugge, U. and Johansson, M. 1997. Historisk sociologi. Studentlitteratur, Lund. Earle, T. 1997. The evolution of chiefdoms. In: Earle, T. (ed.) Chiefdoms: Power, Economy and Ideolog y. School of American Research Advanced Seminar Series. Cambridge: Cambridge University Press. . Ebbesen, K. 1981. Offerfundet fra Suldrup i Himmerland. Fra Himmerland og Kjær Herred. Aalborg. — 1997. Der Beginn der Steitaxtzeit. Arkæologiske Rapporter Edenmo, R., Larsson, M., Nordqvist, B., and Olsson, E. 1997. Gropkeramikerna – fanns de? In: Regionalt och interregionalt. Stenåldersundersökningar i Syd- och Mellansverige. Riksantikvarieämbetet UV. Stockholm. (GJUHQ7-lNlUOlJUXSSHQ(QYlVWÀQVNNXOWXUJUXSSXQGHU\QJUHVWHQnOGHU Finska fornminnesföreningens tidskrift 64. ³'HQI|UKLVWRULVNDWLGHQ,Q(GHJUHQ76DOR8DQG/HKWRVDOR +LODQGHU3/ HGV Kivikausi, pronssikausi, rautakausi. Suomen historia 1. Espoo. Edmonds, M. 1990. Description, understanding and the “chaîne opératoire”. Archaeological Review from Cambridge — 1995. Stone tools and society. Working stone in Neolithic and Bronze Age Britain. London: Batsford. — 1999. Ancestral geographies of the Neolithic: landscape, monuments and memory. London: Routledge. (ONLQ$33UHVVXUHÁDNLQJLQWKH1RUWKHUQ.LPEHUOH\$XVWUDOLD. Man 130: Elster, J. 1988. 9HWHQVNDSOLJDI|UNODULQJDUGöteborg: Korpen. — 1993. Explaining technical change. A case study in the philosophy of science. Cambridge: Cambridge University Press; Oslo: Universitetsforlaget. Engelmark, R. 1979, The Paleoenvironment. In: Broadbent, N. (ed.). Coastal Resources and Settlement Stability, Aun, vol. 3. Institute of North European Archaeology, Uppsala University. Uppsala. (QJOXQG/(6PHGHURFKPDNW,Q+MlUWKQHU+ROGDU(DQG5LVEHUJ& (eds). Hantverkets roll i samhället: produktion och reproduktion. Riksantikvarieämbetet UV Gal. Uppsala. 420 Reference List Eriksen, B. V. 2000. ”Chaîne opératoire”: den operative proces og kunsten at W QNHVRPHQÁLQWKXJJHU,Q(ULNVHQ%9 HG Flintstudier. En håndbog i V\VWHPDWLVNHDQDO\VHUDIÁLQWLQYHQWDUHUcUKXV (VWjFLRGD9HLJD630$QWLJXLGDGHVPRQXPHQWDOHVGR$OJDUYH7HPSRV prehistoricos. Imprensa Nacional,,, Fa1kenström, P. 1996. Spån och spånande. Mesolitiska storspånsindustrier i Dalarna och Härjedalen. MA Thesis. Institute of Archaeology, Uppsala University, Uppsala. Fagan, B. M. 1974. Men of the Earth. Boston: Little, Brown and Co. Falkenström, P. and Lindberg, K. in prep. Boplatslämningar vid Fågelsjön. Arkeologisk XQGHUV|NQLQJDY5DlRFK/RVVQ'DODUQD Uppsala University, Uppsala. )qEORW$XJXVWLQV-5DZPDWHULDOWUDQVSRUWSDWWHUQVDQGVHWWOHPHQWV\VWHPV in the European Lower and Middle Palaeolithic: Continuity, change and variability. In: Roebroeks, W. and Gamble, C. (eds.) The Middle Palaeolithic occupation of Europe. Leiden: University of Leiden. Feinman, G. M. 1999. Rethinking our assumptions: Economic specialization at the household scale in ancient Ejutla, Oaxaca, Mexico. In: Skibo, M. and Feinman, G. M. (eds). Pottery and People: A Dynamic Interaction. Salt Lake City: University of Utah Press. Fellner, R. 1995. Technology or typology?: A response to Neeley and Barton. Antiquity Fenton, M. B. 1984. The nature of the source and the manufacture of Scottish EDWWOHD[HVDQGD[HKDPPHUV3URFHHGLQJVRIWKH3UHKLVWRULF6RFLHW\ Fiedorczuk, J. 1995. Processing workshops and habitation sites of the Final 3DODHROLWKLF0D]RYLDQ&RPSOH[$YLHZIURPWKHSHUVSHFWLYHRIÁLQWDUWHIDFWV UHÀWWLQJ,Q6FKLOGW5DQG6XOJRVWRZVND= HGV Man and Flint. Warszawa. Finlay, N. 1997: Kid knapping: the missing children in lithic analysis. In: Moore, J. and Scott, E. (eds.) Invisible people and processes. Writing gender and childhood into European archaeolog y. London; New York: Leicester University Press. — 2000a. Microliths in the making. In: R. Young (ed.). Mesolithic lifeways: Current research in Britain and Ireland, /HLFHVWHU$UFKDHRORJ\0RQRJUDSK — 2000b'HÀQLQJWKHPLFUROLWKchaîne opératoire. In: Mithen, S. (ed.). HunterGatherer Landscape Archaeolog y: The Southern Hebrides Mesolithic. Cambridge: McDonald Institute. — 2000c. Deer Prudence. Archaeological Review Cambridge — 2003. Microliths and multiple authorship. In: Larsson, L., Kindgren, H., .QXWVVRQ./RHIÁHU'DQGcNHUOXQG$ HGV Mesolithic on the Move. Papers presented at the Sixth International Conference on the Mesolithic in Europe, Stockholm 2[IRUG2[ERZ%RRNV — forthcoming. Blank Concerns: issues of skill and consistency in the replication of Scottish Mesolithic Blade Production. In: Finlay, N. and Bamforth, D. (eds.) Skilful Stones: approaches to skill in lithic technolog y. New York: Plenum Press. Finlayson, B. and Mithen, S. 1997. The microwear and morphology of microliths from Gleann Mor. In: Knecht, H. (ed.) Projectile Technolog y, New York: Plenum Press. 421 Skilled Production and Social Reproduction Fischer, A. 1988. A late Palaeolithic Flint Workshop at Egtved, East Jutland. Journal of Danish Archaeolog y ³$/DWH3DODHROLWKLF´VFKRROµRIÁLQWNQDSSLQJDW7UROOHVJDYH'HQPDUN 5HVXOWVIURPUHÀWWLQJActa Archaeologica 60: 33–49. ³2QEHLQJDSXSLORIDÁLQWNQDSSHU\HDUVDJR$SUHOLPLQDU\ DQDO\VLVRIVHWWOHPHQWRUJDQL]DWLRQDQGÁLQWWHFKQRORJ\EDVHGRQFRQMRLQHG ÁLQWDUWHIDFWVIURPWKH7UROOHVJDYHVLWH,Q&]LHVOD((LFNKRII6$UWV1 and Winther, D. (eds). 7KH%LJ3X]]OH,QWHUQDWLRQDO6\PSRVLXPRQ5HÀWWLQJ6WRQH $UWHIDFWV0RQ5HSRV Bonn: Holos. Fiske, J. 1993. Kommunikationsteorier. En introduktion. Stockholm: Wahlström and Widstrand. Fladmark, K. R. 1982. Microdebitage analysis: initial considerations. Journal of $UFKDHRORJLFDO6FLHQFH. Flegenheimer, N., Amick D. and Bayón, C. (in press). Early strategies of raw material acquisition and use in the southern Cone. In: Morrow, J. and Gnecco, C. (eds.) Late Pleistocene Lithic technolog y: An Hemisphere Perspective. Flenniken, J. J. 1987. The Paleolithic Dyukai pressure blade technique of Siberia. Arctic Anthropolog y — 1980: Replicative systems analysis: A model applied to the vein quartz artifacts from the Hoko River site$GLVVHUWDWLRQVXEPLWWHGLQSDUWLDOIXOÀOOPHQWRIWKH requirements for the degree of Doctor of Philosophy, UMI Dissertation Services. Department of Anthropology, Washington State University. ³DQG:KLWH-3$XVWUDOLDQÁDNHGVWRQHWRROVDWHFKQRORJLFDOSHUVSHFWLYH Records of the Australian Museum — and Hirth. K. G. 2003. Handheld prismatic blade manufacture in Mesoamerica. In: Hirth, K. (ed). Experimentation and Interpretation in Mesoamerican Lithic Technolog y. Salt Lake City: University of Utah Press. )RUVVDQGHU-('HU2VWVNDQGLQDYLVFKH1RUGHQZlKUHQGGHWlOWHVWHQ Metallzeit Europas. 6NULIWHUXWJLYQDDY.XQJO+XPDQLVWLVND9HWHQVNDSVVDPIXQGHWL Lund XXII. Lund. Foucault, M. 1970. The order of things. An arcaheolog y of the human sciences New York: Pantheon Books. — 1972. The archaeolog y of knowledge and the discourse on language. New York: Pantheon Books. — 1978. The history of sexuality, vol 1: An introduction. New York: Pantheon Books. — 1989. Upplysningens, revolutionen och framstegets möjlighet. In: Östling, B. (ed.). 9DGlUXSSO\VQLQJ"6\PSRVLRQ. Stockholm/Stehag. — 1990. Politics, philosophy, culture: interviews and other writings. Lawrence Kritzman, Ed. Translated by Alan Sheridan. London and New York: Routledge. Frankel, R. 1999. Wine and oil production in antiquity in Israel and other Mediterranean countries,6KHIÀHOG(QJODQG6KHIÀHOG$FDGHPLF3UHVV Frazier, F. 1973. Production of Artifact Casts Using Epoxy Resins. Newsletter of Lithic Technolog y ,, Frison, G. and Bradley, B. 1999. The Fenn Cache. Clovis weapons and tools. Santa Fe, One Horse Land and Cattle Company. 422 Reference List Frödin, O. and Persson, A. W. 1938. $VLQH5HVXOWVRIWKH6ZHGLVKH[FDYDWLRQV ² Stockholm. Fuglestvedt, I. 2001. Pioneerbosetningens fenomenologi. Sørvest-Norge og Nord-Europa ²%3 Unpublished doctorate dissertation. Department of Archaeology, University of Bergen. –––– 2005. 3LRQHHUERVHWQLQJHQVIHQRPHQRORJL6¡UYHVW1RUJHRJ1RUG(XURSD ²%3 Stavanger: Arkeologisk museum i Stavanger. ³&RQWDFWDQGFRPPXQLFDWLRQLQ1RUWKHUQ(XURSH%3 A phenomenological approach to the connection between technology, skill and landscape. In: Knutsson, H. (ed). Pioneer settlements and colonization processes in the Barents region.9XROOHULP3DSHUVRQ+XQWHU*DWKHUHU$UFKDHRORJ\9XROOHULP Sweden. *DGDPHU+*Truth and method. New York: Continuum. Gansum, T., Jerpåsen, G. B. and Keller, Ch. 1997. Arkeologisk landskapsanalyse med visuelle metoder. $P69DULD 28. Stavanger.´ Geertz, C. 1973. The Interpretation of Cultures. New York: Basic Books. Gell, A. 1998. Art and agency: an anthropological theory. Oxford: Oxford University Press. — 1999 [1992]. The technology of enchantment and the enchantment of technology. In: Gell. A. and Hirsch, E. (eds.). The art of anthropolog y: Essays and diagrams. Monographs on social anthropology, 67. London; New Brunswick, N.J.: Athlone Press. Gero, J. M. 1989. Assessing social information in material objects: How well do lithics measure up? In: Torrence, R. (ed.). Time, Energ y and Stone Tools. New Directions in Archaeology. Cambridge University Press. Cambridge. ³*HQGHUOLWKLFVZRPHQ·VUROHLQVWRQHWRROSURGXFWLRQ,Q*HUR-0DQG Conkey, M. (eds). Engendering archaeolog y: women and prehistory. Oxford: Basil Blackwell. — and Conkey, M. 1991: (eds.) Engendering archaeolog y: Women and prehistory. Oxford: Basil Blackwell. *HVOLQ0%DVWLHQ*DQG)RXTXHW*(WXGHGXGpS{WGHODPHVGH©/D &UHXVHWWHªj%DUURXBulletin des Amis du Grand-PressignyQ ³%DVWLHQ*DQG0DOOHW1/HGpS{WGHJUDQGHVODPHVGH©/D&UHXVHWWHª %DUURX ,QGUHHW/RLUH Gallia Préhistoire — Bastien, G., Mallet, N. and Freslier, B. 1982. Le site de La Creusette, habitat et atelier. Bulletin des Amis du Grand-Pressigny Gibson, K. R. and Ingold, T. (eds.) 1993. Tools, language and cognition in human evolution. Cambridge: Cambridge University Press. Giddens, A. 1984. The constitution of society. An outline of the theory of structuration. Berkeley: University of California Press. — Introduction to sociolog y. New York: Norton. — 1996. Modernitetens följder. Studentlitteratur. Lund. — 1997. Självet och samhället i den moderna epoken. Göteborg: Daidalos. 423 Skilled Production and Social Reproduction Gimbutas, M. 1956. The prehistory of Eastern Europe. Part I: Mesolithic, Neolithic and Copper Age cultures in Russia and the Baltic area. American School of Prehistoric Research, Peabody Museum, Harvard University, Bulletin No. 20. Cambridge. Giria, E. U. 1997. Tehnologiceskij analiz kamennych industrij. Metodika mikromakroanaliza drevnih orudij truda. Saint Petersburg. Glennan, W. S. 1972. The Manix lithic industry: Early lithic tradition or workshop refuse? Unpublished ms. *ORE39(QÁLQWVPHGLHSn)RUQ VKuml1951l Gonçalves, V. S. 1989. Megalitismo e metalurgia no alto Algarve oriental; uma DSUR[LPDojRLQWHJUDGD&ROOHFWLRQ81,$54HVWXGRVHPHPRULDV 2. 2 vol. 566 p., 252 Fig. Lisbon: Imprensa Nacional, Unidade de Arqueologia do Centro de Arqueologia e Historia da Universidade de Lisboa. — 1992. Revendo as antas de Reguengos de Monsaraz. &DGHUQRVGD81,$54 2. 264 p. 48 photos h.t. Lisbon: Unidade de Arqueologia do Centro de Arqueologia e Historia da Universidade de Lisboa.. *RULQJ0RUULV1$WWKHHGJH7HUPLQDO3OHLVWRFHQHKXQWHUJDWKHUHUVLQWKH Negev and Sinai. British Archaeological Reports, International Series S361. Oxford. — Henry, D. O., Phillips, J. L., Clark, G.A., Barton, C.M. and Neeley, M.P. 1996. Patterns in the Epipaleolithic of the Levant: Debate after Neeley and Barton. Antiquity Gosden, C. 1994. Social beeing and time. Oxford: Blackwell. — and Lock, G. 1998. Prehistoric histories. World archaeolog y *RXOG5$0DQ·VROGHVWFUDIWUHFUHDWHGCurator;,,, *UDKDP5.LPEHUOH\HWKQRDUFKDHRORJ\. Records of the Anthropological Society of South Australia.$GHODLGH Grydeland, E. 2005. The pioneers of Finnmark from the earliest coastal settlements to the encounter with the inland people of Northern Finland. In: Knutsson, H. (ed). Pioneer settlements and colonization processes in the Barents region. 9XROOHULP3DSHUVRQ+XQWHU*DWKHUHU$UFKDHRORJ\9XROOHULP6ZHGHQ *XJOLHOPLQR&59LJDQRWWL&+HZOHWW%DQG&DYDOOL6IRU]D// Cultural variation in Africa: Role of mechanism of transmission and adaptation. 3URFHHGLQJVRIWKH1DWLRQDO$FDGHP\RI6FLHQFH86$ *XPLľVNL:*UyGHN1DGEXůQ\RVDGDNXOWXU\SXFKDUyZOHMNRZDW\FK:URFâDZ :DUV]DZD.UDNyZ*GDľVNáyGŭ Gustafsson, P., Lindholm, P. and Runeson, H. 2003. Gropar och keramik. In: Anund, J. (ed). Landningsplats-forntiden. Riksantikvarieämbetet arkeologiska undersökningar skrifter nr 49. Stockholm. Habermas, J. 1975. Communication and the evolution of society. London: Heinemann. — 1984. 7KHRU\RIFRPPXQLFDWLYHDFWLRQ9ROXPH,5HDVRQDQGWKHUDWLRQDOL]DWLRQRIVRFLHW\ Boston: Beacon Press. Hacking, J. 1999. Social konstruktion av vad? Stockholm: Thales. Hagen, A. 1954. Europeiske impulser i østnorsk bronsalder. In: 9LNLQJ Halbersham, M. 1999. Totalitarianism and the modern conception of politics. New Haven: Yale University Press. 424 Reference List Halinen, P. 1997 Kaustisen Kankaan asuinpaikka ja punamultahaudat. Muinaistutkija 2/1997:18–27. Hallgren, F., Djerw, U., af Geijerstam, M. and Steineke, M. 1997. Skogsmossen, an (DUO\1HROLWKLFVHWWOHPHQWVLWHDQGVDFULÀFLDOIHQLQWKHQRUWKHUQERUGHUODQGRI WKH)XQQDOEHDNHU&XOWXUHTor Hamrick, M. W., Churchill, S. E., Schmitt, D. and Hylander, W. L. 1998. EMG RIWKHKXPDQÁH[RUSROOLFLVORQJXVPXVFOHLPSOLFDWLRQVIRUWKHHYROXWLRQRI hominid tool use. Journal of Human Evolution Hansen, F. 1929. Smärre meddelande. Neolitiska “paleolitica”. Fornvännen. 0HGGHODQGHIUnQ.9LWWHUKHWV+LVWRULHRFK$QWLNYLWHWV$NDGHPLQnUJnQJ. Hansen, P. V. and Madsen, B. 1983. Flint axe manufacture in the Neolithic. An H[SHULPHQWDOLQYHVWLJDWLRQRIDÁLQWD[HPDQXIDFWXUHVLWHDW+DVWUXS9 QJHW East Zealand. Journal of Danish Archaeolog y Harrison, R. 2002. Kimberley spearpoints, cultural identity and masculinity in the north of Australia. Australia. Journal of Social Archaeolog y ³IRUWKFRPLQJ'DWLQJWKHDSSHDUDQFHRISUHVVXUHÁDNHG¶.LPEHUOH\SRLQWV· in northern Australia: New insights from the southeast Kimberley, Western Australia. Archaeolog y in Oceania. +DVWUXS.3UHVHQWLQJWKHSDVW5HÁHFWLRQVRQP\WKDQGKLVWRU\Folk 29: Hayden, B. 1979: 3DODHROLWKLFUHÁHFWLRQV/LWKLFWHFKQRORJ \DQGHWKQRJUDSKLFH[FDYDWLRQVDPRQJ Australian Aborigines. Canberra: Australian Institute of Aboriginal Studies; Atlanatic Highlands, N.J.: Humanities Press. — 1995. Pathways to power: Principles for creating socioeconomic inequalities. In: Price, T. D. and Feinman, G. (eds). Foundations of Social Inequality. New York: Plenum Press. — 1998. Practical and prestige technologies: the evolution of material systems. Journal of Archaeological Method and Theory Healan, D. M. 1986. Technological and nontechnological aspects of an obsidian workshop excavated at Tula, Hidalgo. In: Isaac, B. (ed). Economic Aspects of Prehispanic Highland Mexico. Research in Economic Anthropology, Supplement 2. Greenwish, Conn: JAI Press. — 1990. Informe preliminar de la investigaciones en Tula, Hidalgo, por la 8QLYHUVLGDGGH7XODQH,Q6RWRGH$UHFKDYDOHW' HG Nuevos Enfoques en el Estudio de la Litica. Instituto de Investigaciones Antropologicas, UNAM, Mexico. ³,GHQWLI\LQJOLWKLFUHGXFWLRQORFLZLWKVL]HJUDGHGPDFURGHELWDJHD multivariate approach. American Antiquity Helms, M. W. 1988. 8O\VVHVVDLO$QHWKQRJUDSKLFRG\VVH\RISRZHUNQRZOHGJHDQG geographical distance. Princeton: Princeton University Press. — 1993. Craft and the kingly ideal: Art, trade, and power. Austin: University of Texas Press. — 1998. $FFHVVWRRULJLQV$IÀQHVDQFHVWRUVDQGDULVWRFUDWV Austin: University of Texas Press. 425 Skilled Production and Social Reproduction Hempel, C. G. 1989. El dilema del teórico: Un estudio sobre la lógica de la construcción de teorías. In: Olivé, L. and Pérez Ransanz, A. R. (eds.) Filosofía de la ciencia: Teoría y observación. Mexico: Siglo XXI Editores.. Henry, D. O. 1989. From foraging to agriculture: The Levant at the end of the Ice Age. Philadelphia: University of Pennsylvania Press. Hertell, E. and Manninen, M. A. in press: Rävåsenin kvartsiaineisto. To be published in Swedish in Finskt Museum. Hill, J. N. 1978. Individuals and their artifacts: an experimental study in arcaheology. American Antiquity &RQWULEXWLRQVWR$UFDKHRORJLFDO Method and Theory; April 1978) Hirth, K. G. 2000. (ed.). Ancient urbanism at Xochicalco: The evolution and organization of a pre-Hispanic society: Archaeological Research at Xochicalco, 1. Salt Lake City: University of Utah Press. — 2000. (ed). Archaeological Investigations at Xochicalco. Results of the Xochicalco Mapping Project 2. Salt Lake City: University of Utah Press — 2002. Provisioning constraints and the production of obsidian prismatic blades at Xochicalco, Mexico. In: Hirth, K. G. and Andrews, B. W. (eds). Pathways to Prismatic Blades. Sudies in Mesoamerican Lihic Technolog y. Cotsen Institute of Archaeology, University of California, Los Angeles. — n.d. [2006] Flaked stone craft production in domestic contexts. In: Hirth, K. G. (ed). Obsidian Craft Production in Ancient Central Mexico. Salt Lake City: University of Utah Press. — and Andrews, B. W. 2002. Pathways to prismaatic blades: Sources of variation in Mesoamerican lithic technology, In: Hirth, K. G. and Andrews, B. W. (eds). Pathways to Prismatic Blades. Sutdies in Mesoamerican Lithic Technolog y. Cotsen Institute of Archaeology, University of California, Los Angeles. — and Andrews, B. W. n.d.a [2006] Craft specialization and craftsmen skill. In: Hirth, K. G. (ed). Obsidian Craft Production in Ancient Central Mexico, Salt Lake City: University of Utah Press. — and Andrews, B. W. n.d.b [2006] Estimating production output in domestic craft workshops. In: Hirth, K. G. (ed). Obsidian Craft Production in Ancient Central Mexico. University of Utah Press, Salt Lake City, Utah. — Andrews, B. W. and Flenniken, J. J. 2003. The Xochicalco production sequence for obsidian prismatic blades: Technological analysis and experimental inferences. In: Hirth, K. G. (ed). Mesoamerican Lithic Technolog y: Experimentation and Interpretation. University of Utah Press, Salt Lake City, Utah. — Andrews, B. W. and Flenniken, J. J. n.d. [2006] A technological analysis of Xochicalco obsidian prismatic blade production. In: Hirth, K. G. (ed). Obsidian Craft Production in Ancient Central Mexico. Salt Lake City: University of Utah Press. — Flenniken, J. J. and Andrews, B. W. 2000. Flaked obsidian tools and their behavioral implications. In: Hirth, K. G. (ed). The Xochicalco Mapping Project. Archaeological Research at Xochicalco, 2. Salt Lake City: University of Utah Press. 426 Reference List Hobsbawm, E. 1983. Introduction: Inventing traditions. In: Hobsbawm, E. and Ranger, T. (eds.). The Invention of Tradition. Cambridge: Cambridge University Press. Hodder, I. 1982. Symbols in action. Ethnoarchaeological studies of material culture. Cambridge: Cambridge University Press. ³$JHQF\DQGLQGLYLGXDOVLQORQJWHUPSURFHVVHV,Q'REUHV0$DQG Robb, J. (eds.) Agency in Archaeolog y. London and New York. +|JEHUJ$&KLOGDQGDGXOWDWDNQDSSLQJDUHD$WHFKQRORJLFDOÁDNH DQDO\VLVRIDPDQXIDFWXUHRID1HROLWKLFVTXDUHVHFWLRQHGD[HDQGDFKLOG·V ÁLQWNQDSSLQJDFWLYLWLHVRQDQDVVHPEODJHH[FDYDWHGDVSDUWRIWKHgUHVXQG À[HGOLQNSURMHFWActa Archaeologica — 2001a. Flint in the Malmö area. In: Knarrström, B. Flint: a Scanian Hardware. National Heritage Board, Sweden. Skånska spår – arkeologi längs Västkustbanan. Lund. — 2001b. Flinta under yngre bronsålder och äldre järnålder. Rapport inom projektet Öresundsförbindelsen. Rapport nr 37. Malmö Kulturmiljö. — 2002. Production sites on the beach ridge Järavallen. Aspect on tool preforms, action, technology, ritual and the continuity of place. Current Swedish Archaeolog y — Apel, J., Knutsson, K., Olausson, D., and Rudebeck, E. 2001. The spread of ÁLQWD[HVDQGGDJJHUVLQ1HROLWKLF6FDQGLQDYLD3DPiWN\$UFKHRORJLFNp;&,,. ³IRUWKFRPLQJ3OD\LQJZLWKÁLQWDQGOHDUQLQJWRNQDSWUDFLQJDFKLOG·VLPLWDWLRQ of adult work in a lithic assemblage. In: Finlay, N. and Bamforth, D. (eds.) Skilful Stones: Approaches to skill in lithic technolog y. Plenum Press. New York. ³DQG2ODXVVRQ'$WSUHVV6FDQGLQDYLDQÁLQW²DJXLGHIRUDUFKDHRORJLVWV0V Höiriis, O. 1997. Kampen om stenalderen. Antropologiska bud på vor oprindelse I forntid och nutid. Kuml Højlund, F. 1979. Stenøkser i Ny Guineas Højland. Betydning af prestigesymboler for reproduktionen af et stammesamfund. Hikuin Holm, J. 2003. Quartz, microblades and the meaning of life. In: Samuelsson, C. and Ytterberg, N. (eds.) 8QLWLQJ6HD6WRQH$JH6RFLHWLHVLQWKH%DOWLF6HD5HJLRQ 3URFHHGLQJVIURPWKH)LUVW8QLWLQJ6HD:RUNVKRSDW8SSVDOD8QLYHUVLW\6ZHGHQ-DQXDU\ ²Occasional Papers in Archaeology 33: 79–88. — and Rieck, F. 1992. Istidsjaegere ved Jelssöerne. Hamburgkulturen I Danmark. Skrifter fra Museumsrådet for Sönderjyllands Amt 5. Haderslev. Holmes, W. H. 1919. The lithic industries. Part 1, Bulletin 60. Handbook of Aboriginal American Antiquities. Bureau of American Ethnology. Honegger, M. 2001. /·LQGXVWULHOLWKLTXHWDLOOpHGX1pROLWKLTXHPR\HQHWÀQDOGH6XLVVH (Monographies du CRA 24) Paris: CNRS. Hornborg, A. 1994. Ecolog y as semiotics. Outlines of a contextualist Paradigm for human ecolog y. Working papers in Human Ecology 1. Lund University, Human Ecology Division. Lund. Howell, F. C. 1964. Early-Man7LPH/LIH Huang, Y. and Knutsson, K. 1995. Functional analysis of Middle and Upper Palaeolithic quartz tools from China. Tor9RO 427 Skilled Production and Social Reproduction +\HQVWUDQGc'HQHQNODVNDIWKnOV\[DQVRPDUEHWVUHGVNDSNordsvensk forntid. Skytteanska samfundets handlingar 6. Studies in North Swedish $UFKDHRORJ\8PHn Idriess, I. 1937 Over the range: sunshine and shadow in the Kimberley, Sydney: Angus and Robertson. Ihuel, E. 2004. La diffusion du silex du Grand-Pressigny dans le Massif armoricain au Néolithique. %XOOHWLQGHO·$VVRFLDWLRQGHV$PLVGX0XVpHGX*UDQG3UHVVLJQ\, Supplément n°2; Documents Préhistoriques du CTHSQ-RXp/qV7RXUV Editions La Simarre. Ingold, T. 1996. Key Debates in Anthropolog y. New York. ,QL]DQ0//HFKHYDOOLHU0DQG3OXPHW3$WHFKQRORJLFDOPDUNHU of the penetration into North America : pressure microblade debitage. Its origin in the Paleolithic of North Asia and its diffusion. In: Vandiver P. M., Druzik J. R., Wheeler G. S. and Freestone I. C. (eds.): Materials Issues in Art and Archaeolog y ,,, ³5HGXURQ%DOOLQJHU05RFKH+DQG7L[LHU-7HFKQRORJ\DQG terminology of knapped stone, followed by a multilingual vocabulary Arabic, English, French, German, Greek, Italian, Portuguese, Spanish. Préhistoire de la Pierre Taillée 5,1DQWHUUH&HUFOHGH5HFKHUFKHVHWG·eWXGHV3UpKLVWRULTXHV — Roche, H. and Tixier, J. 1992. Technology of knapped stone. Prehistoire de la Pierre Taillee0HXGRQ&HUFOHGH5HFKHUFKHVHWG·eWXGHV3UpKLVWRULTXHV ³5HGXURQ%DOOLQJHU05RFKH+DQG7L[LHU- Technolog y and terminolog y of knapped stone.1DQWHUUH&HUFOHGH5HFKHUFKHVHWG·eWXGHV3UpKLVWRULTXHV Inomata, T. 2001. The power and ideology of artistic creation: Elite craft specialists in Classic Maya society. Current Anthropolog y Jameson, F. 1989. Narrative as a socially symbolic act. London: Routledge. Jelinek, A. 1965. Lithic Technology Conference, Les Eyzies, France. American Antiquity Jelínek, J. 1975. The pictorial encyclopedia of the evolution of man. London: Hamlyn. Jensen, J. A. 1973. Bopladsen Myrhöj: 3 hustomter med klokkebægerkeramik. Kuml -HQVHQ2:%HWZHHQERG\DQGDUWHIDFW0HUOHDX3RQW\DQGDUFKDHRORJ\ In: Karlsson, H. and Jensen, O. (eds). Archaeological Conditions. Examples of Epistemlog y and Ontolog y. Göteborg: Bricoleur Press. Jensen, R. 1989. Bosättning och ekonomi – Inomregionala differenser i Mälardalen. In: Poulsen, J. (ed). Regionale forhold I nordisk Bronzealder 5. Jysk $UNHRORJLVN6HOVNDEV6NULIWHU;;,9cUKXV Johansson, A. D.. 1990. Barmosegruppen. Præboreale bopladsfund i SydsjællandcUKXV 8QLYHUVLWHWcUKXV Johansen, L. 2000. Fitting Facts. Institute of Archaeology and Ethnology. University of Copenhagen. Copenhagen -RKQVRQ//$KLVWRU\RIÁLQWNQDSSLQJH[SHULPHQWDWLRQ Current Anthropolog y Johnston, R. 1998. Approaches to the perception of landscape: philosophy, theory, methodology. Archaeological Dialogues 1998, 1. 428 Reference List Judd, N. 1954. Pueblo Bonito. Smithsonian Misc. Collections, Vol. 124. Washington. Judge, W. J. 1973. 3DOHRLQGLDQHFFXSDWLRQVRIWKHFHQWUDO5LR*UDQGH9DOOH\. Albuquerque: New Mexico University Press. Kaelas, L. 1959. Skafthålsyxor. FYND, Göteborgs och Bohusläns Fornminnesförenings Tidskrift .DPPLQJD-7KHQDWXUHRIXVHSROLVKDQGDEUDVLYHVPRRWKLQJRQVWRQH tools. In: Hayden, B. (ed.) Lithic use-wear analysis. New York: Academic Press. — 1985. The Pirri Graver. Australian Aboriginal Studies. Canberra. Kamp, K. A. 2001: Where have all the children gone? The archaeology of childhood. Journal of Archaeological Method and Theory 8/1: 1–34. Karlin, C. and Julien, M. 1994. Prehistoric technology: a cognitive science? In: Renfrew, C. and Zubrow, E. B. W. (eds.) The ancient mind. Elements of cognitive archaeolog y, Cambridge: Cambridge University Press. Kars, E. A. K., Kars, H. and McDonnell, R. D. 1991. Greenstone axes from HDVWHUQFHQWUDO6ZHGHQ$WHFKQRORJLFDOSHWURORJLFDODSSURDFK. Archaeometry 34/2. Karsten, P. 1994. Att kasta yxan i sjön. En studie över rituell tradition och förändring utifrån skånska neolitiska offerfynd. Acta Archaeologica Lundensia Series in 8°, No. 23. Lund. — and B. Knarström (2003). The Tågerup excavation.5LNVDQWLNYDULHlPEHWHW89 Syd. Lund. Kaufman, D. 1995. Microburins and microliths of the Levantine Epipalaeolithic: a comment on the paper by Neeley and Barton. Antiquity Kelterborn, P. 1980. Zur Frage des Livre de Beurre. JahrBuch des Schweizerischen *HVHOOVFKDIWIU8UXQG)UK*HVFKLFKWH — 1981. The livre de beurre method. )OLQWNQDSSHUV·([FKDQJH 4/3: 12–20. ³7RZDUGVUHSOLFDWLQJ(J\SWLDQ3UHG\QDVWLFÁLQWNQLYHVJournal of Archaeological Science Kibunjia, M. 1994. Pliocene archaeological occurrences in the Lake Turkana basin. Journal of Human Evolution Kidder, A.V. 1947. 7KHDUWLIDFWVRI8D[DFWXQ*XDWDPDOD Carnegie Institute of Washington. Publication 576. Washington. Kjellmark, K. 1903. En stenåldersboplats i Järavallen vid Limhamn. Antikvarisk WLGVVNULIWI|U6YHULJH. — 1905. Öfversikt af Sveriges stenåldersboplatser. <0(5 Knapp, A.B. and Ashmore, W. 1999. Archaeological landscapes: constructed, conceptualized, ideational. In: Ashmore, W. and Knapp, A.B. (eds). Archaeologies of Landscape: Contemporary Perspectives. Oxford: Blackwell. .QDUUVWU|P%1HROLWLVNÁLQWWHNQRORJLLHWWVNnQVNWUDQGRPUnGH,Q Karsten, P. (ed.) &DUSH6FDQLDP$[SORFNXU6NnQHVI|UÁXWQD. R AÄ Skrifter 22. Lund. Knudson, M. A. 1983. 2UJDQL]DWLRQDO9DULDELOLW\LQ/DWH3DOHR,QGLDQ$VVHPEODJHV. 3XOOPDQ:DVKLQJWRQ6WDWH8QLYHUVLW\5HSRUWVRI,QYHVWLJDWLRQV Knudson, R. and Muto, G. 1973. Available lithic reproductions. Newsletter of Lithic Technolog y,, 429 Skilled Production and Social Reproduction Knutsson, H. 1995. Slutvandrat? Aspekter på övergången från rörlig till bofast tillvaro. Aun 20. Uppsala: Societas Archaeologica Upsaliensis. — 2000. Two technologies – two mentalities. In: Knutsson, H. (ed.) Halvvägs kust till kust. Stenålderssamhällen i förändring. Coast to Coast Books 2. Uppsala: Department of Archaeology, Uppsala University: –––– 2003. Technology, mythology and the travels of the agricultural package in Europe. In: Budja, M. (ed.). Documenta Praehistorica XXVIII (8th Neolithic 6WXGLHV /MXEOMDQD –––– 2005 (ed). Pioneer settlements and colonization processes in the Barents region. 9XROOHULP3DSHUVRQ+XQWHU*DWKHUHU$UFKDHRORJ\9XROOHULP6ZHGHQ Knutsson, K. 1981. Innovation och produktutveckling. Aspekter på spåntekniken i Boreal och Atlantisk tid i Skåne. Kontaktstencil 20. Umeå. ³%DUQÀQQVGRP"Fjölnir 2/1: 8–11. — 1986. Några ord om barn, stötkantkärnor och Pièces Esquillées. Fjölnir 5(1): 29–39. — 1988a. Making and using stone tools. The analysis of the lithic assemblages from Middle 1HROLWKLFVLWHVZLWKÁLQWLQ9lVWHUERWWHQQRUWKHUQ6ZHGHQ Aun 11, Uppsala: Societas Archaeologica Upsaliensis. — 1988b. Patterns of tool use. Scanning electron microscopy of experimental quartz tools. Aun 10. Uppsala: Societas Archaeologica Upsaliensis. — 1998. Convention and lithic analysis. In: Holm, J. and Knutsson, K. (eds). 3URFHHGLQJVIURPWKH7KLUG)OLQW$OWHUQDWLYHV&RQIHUHQFHDW8SSVDOD6ZHGHQ2FWREHU . Occasional Papers in Archaeology, 16. Uppsala. — 2005. The Historical Construction of Norrland. In: Knutsson, H. (ed.). Arrival. &RDVWWR&RDVW%RRNV8SSVDOD ³DQG$SHO-PV(QERNRPVWHQDURFKPlQQLVNRU2PNODVVLÀNDWLRQHQDY slagna stenar. Uppsala. Koch, E. 1998. Neolithic bog pots from Zealand, Møn, Lolland and Falster. Nordiske Fortidsminder Serie B Volume 16. Köpenhamn. Kopytoff, I. 1986. The cultural biography of things: Commodization as process. In: Appadurai, A. (ed). The social life of things. Commodities in cultural perspective. Cambridge: Cambridge University Press. Kossinna, G. 1936 (1912 'LHGHXWVFKH9RUJHVFKLFKWHHLQHKHUYRUUDJHQGQDWLRQDOH Wissenschaft. 7th ed. Leipzig: Carl Kabitzsch Verlag. .R]âRZVNL6.Mesolithic in Poland: a new approach, Warszawa: Wydawnictwa Uniwersytetu Warszawskiego. Kresten, P. 1998. 6NDIWKnOV\[RUIUnQ8SSODQG1lUNH6|GHUPDQODQGRFK9lVWHUJ|WODQG Bestämning av bergartsmaterialen. Geoarkeologiskt Laboratorium, Analysrapport QXPPHU5$b89*$/ Kristiansen, K. 1991. Chiefdoms, states, and systems of social evolution. In: Earle, T. (ed). Chiefdoms: Power, Economy, and Ideolog y. Cambridge: Cambridge University Press. — 1998a. Europe Before History. Cambridge: Cambridge University Press. 430 Reference List — 1998b. A theoretical strategy for the interpretation of exchange and interaction in a Bronze Age Context. In: Mordant, C. Pernot, M. and Rychner, V. (eds). /·$WHOLHUGXEURQ]LHUHQ(XURSHGX;;DX9,,,VLqFOHDYDQWQRWUHqUH,,,3URGXFWLRQ circulation et consummation du bronze. Paris: CTHS Kruk J. 1973. 6WXGLDRVDGQLF]HQDGQHROLWHPZ\ů \QOHVVRZ\FK:URFâDZ:DUV]DZD .UDNyZ*GDľVN Küchler, S. 1993. Landscape as memory: The mapping of process and its representation in a Melanesian society. In: Bender, B. (ed). Landscape: Politics and Perspectives. Providence/Oxford: Berg. Kuhn, S. L. 1995. Mousterian Lithic Technology. An Ecological Perspective, Princeton: Princeton University Press. — Arsebük, G. and Howell, F.C. 1996. The Middle Pleistocene lithic assemblage from Yarimburgaz Cave, Turkey. Paléorient 22/1: 31–49. /DLWDNDUL$'LH6FKDIWORFKl[WHGHU6WHLQ]HLWYRQJHRORJLVFKSHWUR JUDÀVFKHP6WDQGSXQNWFinska Fornminnesföreningens Tidskrift+HOVLQJIRUV Landau, M. 1993, Narratives of Human Evolution. New Haven: Yale University. /DQQHUEUR1RUHOO03LORFKVSMXWVSHWVDUIUnQgYUH'DODUQD BA Thesis. Uppsala: Uppsala University. Larson, M. L. and Kornfeld, M. 1997. Chipped stone nodules: Theory, method and examples. Lithic Technolog y 22/1: 4–19 Larsson, L. and Olausson, D. 1982. Testing for the presence of thermal SUHWUHDWPHQWRIÁLQWLQWKH0HVROLWKLFDQG1HROLWKLFRI6ZHGHQJournal of Archaeological Science Larsson, M. 1995. Fagervik – en klassisk gropkeramisk boplats, Raä 23, Krokeks socken, Norrköpings kommun, Östergötland. 5DSSRUW89/LQN|SLQJ Linköping. Larsson, T. B. 1999a. Symbols, divinities and social inequality. In: Goldhahn, J, and Nordquist, P. (eds). Marxistiska perspektiv inom skandinavisk arkeologi. Arkeologiska studier vid Umeå universitet 5. Umeå. — 1999b. The transmission of an élite ideology: Europe and the Near East in the second millennium BC. In: Goldhahn, J. (ed). Rock Art as Social Representation. BAR International Series 794. Oxford. Latour, B. 1993. :HKDYHQHYHUEHHQPRGHUQ>1RXVQ·DYRQVMDPDLVpWpPRGHUQHV@ New York: Harvester Wheatshearf, Cambridge, Mass.: Harvard University Press. Lechtman, H. 1977. Style in technology – some early thoughts. In: Lechtman, H. and Merrill, R. (eds.) Material culture: styles, organisation and dynamics of technolog y. ( Proceedings of the American Ethnological Society, 1975). St. Paul, Minn.: West Publishing Co. Leeuw, van der S. E. 1994. Cognitive aspects of “technique”. In: Renfrew, C. and Zubrow, E. (eds.) The ancient mind. elements of cognitive archaeolog y. New directions in archaeology. Cambridge, Cambridge University Press. Lekberg, P. 2002. Yxors liv - människors landskap. En studie av kulturlandskap och samhälle i Mellansveriges senneolitikum. Coast to Coast Books 5. Uppsala: Department of Archaeology, Uppsala University. 431 Skilled Production and Social Reproduction Lemonnier, P. 1990. Topsy turvy techniques. Remarks on the social representation of techniques. Archaeological review from Cambridge — 1992: Elements for an anthropolog y of technolog y. Anthropological Papers No. 88, Museum of Anthropology. Ann Arbor: University of Michigan. —1993 (ed.). Technological choices. transformation in material cultures since the Neolithic. London: Routledge. /HURL*RXUKDQ$Le geste et la parole I: Technique et langage. Paris: Albin Michel. &XDGURVGHPRUIRORJtDGHVFULSWLYD,Q/HURL*RXUKDP$ HG La prehistoria. Barcelona: Editorial Labor. /H5RX[&76SHFLDOLVHGSURGXFWLRQGLIIXVLRQDQGH[FKDQJHGXULQJ the Neolithic in Western France: the example of polished stone axes. In: Edmonds, M. and Richards, C. (eds.) 1998: 8QGHUVWDQGLQJWKH1HROLWKLFRI1RUWK Western Europe, *ODVJRZ&UXLWKQH3UHVV LeTurneau P. 2001. Evidence of the role of bifacial cores in Folsom lithic technology. Current Research in the Plesitocene /pYL6WUDXVV& 7KHVDYDJHPLQG /DSHQVpHVDXYDJH Chicago: University of Chicago Press.. ³1RXVDYRQVOXLHWPRLHVVD\pGHIDLUHjSHXSUqVODPrPHFKRVH,QAndré /HURL*RXUKDQRXOHV9RLHVGHO·+RPPH Paris: Albin Michel. Linares, J.A., Nocete, F. and Saez, R. 1998. Aprovisionamiento compartido versus aprovisionamiento restringido: los casos de las canteras del III milenio A.N.E. del Andévalo (Huelva). RubricatumQ Lindblom, I. 1991. Movements in a landscape: on the spatial behavior of neolithic man in a moraine area, Vestfold, Norway. In: Grøn, O., Engelstad, E. and Lindblom, I. (eds). Social Space. Human Spatial Behaviour in Dwellings and Settlements. Odense University Studies in History and Social Sciences 147. Odense. Lindgren, C. 1994. Ett bipolärt problem – om kvartsteknologi under mesolitikum. Aktuell Arkeologi IV. Stockholm Archaeological Reports 29: 77–86. — 1996. Kvarts som källmaterial – exempel från den mesolitiska boplatsen Hagtorp. Tor 28: 29–52. — 1998. Shapes of quartz and shapes of minds. In: Holm, L. and Knutsson, K. (eds).3URFHHGLQJVIURPWKH7KLUG)OLQW$OWHUQDWLYHV&RQIHUHQFHDW8SSVDOD6ZHGHQ 2FWREHU. Occasional Papers in Archaeology 16. Uppsala. —2003. My way or your way. On the social dimension of technology as seen in the lithic strategies in eastern middle Sweden during the Mesolithic. In: Larsson, /.LQGJUHQ..QXWVVRQ./RHIÁHU'DQGcNHUOXQG$ HGV Mesolithic on the Move. Papers presented at the Sixth International Conference on the Mesolithic in (XURSH6WRFNKROPOxford: Oxbow Books — Hallgren, F. and Björck, N. 1999. The Mesolithic in eastern central Sweden. In: Boaz, J. (ed.) The Mesolithic in Central Scandinavia. Universitetets Oldsaksamlings 6NULIWHU1U2VOR Locht, J. L. and C. Swinnen. 1994. Le debitage discoïde du gisement de Beauvais (Oise). Aspects de la chaîne operatoire au travers de quelques remontages. Paleo 432 Reference List Löfgren, O. 1983. Kulturbygge och kulturkonfrontation. In: Hannertz, I. Liljeström, R. and Löfgren, O. (eds.) Kultur och medvetande. En tvärvetenskaplig analys. Lund, Akademilitteratur. Lomborg, E. 1973. 'LHÁLQWGROFKH'lQHPDUNV Nordiske Fortidsminder, series B in quatro, Bind 1. Köpenhamn. /RPERUJ(7KHÁLQWGDJJHUVRI'HQPDUN6WXGLHVLQFKURQRORJ\DQG cultural relations of the south Scandinavian Late Neolithic and reply to comments. Norwegian Archaeological ReviewYRODQG Lopez Castaño, C. E. 1999. Ocupaciones tempranas en las tierras bajas tropicales del valle PHGLRGHOUtR0DJGDOHQDVLWLR<21<RQGy$QWLRTXtDFundación de Investigaciones Arqueológicas Nacionales, Banco de la República, Santafé de Bogotá, D.C. /RYH-5%1RWHVRQWKH:RURUUD7ULEHRI1RUWK:HVWHUQ$XVWUDOLD Transactions, Royal Society of North-Western Australia, Perth, Vol. 41. cc — 1936. Stone Age Bushmen of Today Glasgow: Blackie and Sons. Lowe, P. and Pike, J. 1990. Jilji -life in the Great Sandy Desert. Broome, Western Australia: Magabala Books. Lowenthal, D. 1985. The Past is a Foreign Country. Cambridge: Cambridge University Press. Loze, I. 1997. The early Corded Ware culture in the territory of Latvia. In: Siemen, P. (ed). (DUO\&RUGHG:DUH&XOWXUH7KH$KRUL]RQ²)DFWRU)LFWLRQ Arkeologiske rapporter fra Esbjergs museum. Esbjerg. /EFNH39nUWLGVÀORVRÀ²ÀORVRIHUQD²GHÀORVRÀVNDVWU|PQLQJDUQD Stockholm: Forum. Lutzen, K. 2003. Nostalgi og genskabte faelleskaber – om velgörehed I .|EHQKDYQ,VOXWQLQJHQDIWDOHW,Q.UXVH7( HGV Historiske kulturstudier. Tradition-modernitet-antimodernitet. Roskilde: Roskilde universitetsforlag. Machnik, J. 1966. 6WXGLDQDG.XOWXUĊ&HUDPLNL6]QXURZHMZ0DâRSROVFH:URFâDZ Warszawa, Kraków. — 1997. Zwei Entwicklungswege der Schnurkeramikkultur in den Flussgebieten der oberen Weichsel, Bug und Dnestr. In: Siemen, P. (ed). Early Corded Ware &XOWXUH7KH$KRUL]RQ²)DFWRU)LFWLRQ Arkeologiske rapporter fra Esbjergs museum. Esbjerg. Madsen, B. 1984. Flint axe manufacture in the Neolithic: experiments with JULQGLQJDQGSROLVKLQJRIWKLQEXWWHGÁLQWD[HVJournal of Danish Archaeolog y 3: ³1RJOHQRPHQNODWRULVNHEHP UNQLQJHUWLOVWXGLHWDIÁLQWWHNQRORJL eksperimentelt og arkæologisk. Fjölnir ³+DPEXUJNXOWXUHQÁLQWWHNQRORJLL-HOV,Q+ROP-DQG5LHFN) HGV Istidsjaegere ved Jelssöerne. Hamburgkulturen I Danmark. Skrifter fra Museumsrådet for Sönderjyllands Amt 5. Haderslev. — 1995. Late Palaeolithic cultures of south Scandinavia: tools, traditions, technology. In: Larsson, L. (ed). The Earliest Settlement of Scandinavia. Acta Archaeologica Lundensia, Series in 80, No 24. Lund. 433 Skilled Production and Social Reproduction Mahony, D. J. 1924 Note on Making Spearheads in the Kimberley District, WA, Australian Association for the Advancement of Science 117, 474–5. 0DOHQIDQW0&DXYLQ0&DQG&KDIIHQHW*'pFRXYHUWHG·XQHLQGXVWULH PDFUROLWKLTXHUpFHQWHGHIDFLqVSUHVVLJQLHQj9DVVLHX[HQ9HUFRUV 'U{PH ,Q Comptes rendus de l‘Acad. des Sciences3DULV 0DOOHW1/H*UDQG3UHVVLJQ\VHVUHODWLRQVDYHFODFLYLOLVDWLRQ6D{QH Rhône. Supplément au Bulletin de la Société des Amis du Musée du Grand-Pressigny 2. Malmer, M. P. 1962. Jungneolitische Studien. Acta Archaeologica Lundensia. Series in 8o, No.2. Lund. — 1965. Synpunkter på Jungneolitischen Studien. Svar. Finskt Museum LXXI 1964: — 1969. Grobkeramikboplatsen Jonstorp Rä. Antikvarisk Arkiv 36. Stockholm. ³3ULQFLSOHVRIDQRQP\WKRORJLFDOH[SODQDWLRQRI1RUWK(XURSHDQ%URQ]H Age rock art. ,Q1RUGVWU|P+cDQG.QDSH$ HGV Bronze Age Studies. The Museum of National Antiquities Studies 6. Stockholm. Manninen, M. 2003. Chaîne opératoire - analyysi ja kvartsi. Esimerkkinä NYDUWVLQLVNHQWlSDLNND8WVMRNL/HDNäDJRDGHMRKND8QSXEOLVKHG0$WKHVLV University of Helsinki, Institute of Cultural Research, Department of Archaeology. —2005. Problems in Dating Inland Sites. Lithics and the Mesolithic in Paistunturi, Northern Finnish Lapland. In: Knutsson, H. (ed). Pioneer settlements and colonization processes in the Barents region.9XROOHULP3DSHUVRQ+XQWHU Gatherer Archaeology 1. Vuollerim, Sweden. ³7DOODYDDUD0+HUWHOO(6XEQHROLWKLFELIDFHVDQGÁLQWDVVHPEODJHV in Finland. Outlining the history of research and future questions. In: Samuelsson, C. and Ytterberg, N. (eds.) 8QLWLQJ6HD6WRQH$JH6RFLHWLHVLQWKH%DOWLF 6HD5HJLRQ3URFHHGLQJVIURPWKH)LUVW8QLWLQJ6HD:RUNVKRSDW8SSVDOD8QLYHUVLW\ 6ZHGHQ-DQXDU\Occasional Papers in Archaeology 33. Uppsala. — and Valtonen, T. 2002. Havaintoja esihistoriallisesta kvartsin käytöstä Utsjoen Paistunturissa. Muinaistutkija 1/2002: 35–44. Manolakakis, L. 1994. La Production des outils de silex dans les sociétés hiérarchisées de O·(QpROLWKLTXHHQ%XOJDULH(YROXWLRQWUDGLWLRQVFXOWXUHOOHVHWRUJDQLVDWLRQVRFLDOH. Thèse GH'RFWRUDWGHO·8QLYHUVLWp3DULV,3DQWKpRQ6RUERQQH0XOWLJUDSKLpHHW PLFURÀFKpH ³3URGXFWLRQOLWKLTXHHWpPHUJHQFHGHODKLpUDUFKLHVRFLDOHO·LQGXVWULH OLWKLTXHGHO·(QpROLWKLTXHHQ%XOJDULH SUHPLqUHPRLWLpGX,9qPHPLOOpQDLUH non cal.). Bull. de la Société Préhistorique Française ³/HVWUqVJUDQGHVODPHVGHODQpFURSROHGH9DUQDHVVDLG·LQWHUSUpWDWLRQ GHODYDOHXUG·XQPRELOLHUIXQpUDLUH,Q/%DUD\ 'LU Archéologie des pratiques IXQpUDLUHV$SSURFKHFULWLTXH$FWHVGHODWDEOHURQGHGXMXLQ *OX[HQ*OHQQH ) *OX[HQ*OHQQH%LEUDFWH&HQWUH$UFKpRORJLTXH(XURSpHQGX0RQW %HXYUD\ — 2005. Les industries lithiques énéolithiques de Bulgarie. Rahden/Westf.: /HLGRUIS ÀJ ,QWHUQDWLRQDOH$UFKlRORJLH%G Månsson, P. (ed.) 1998. Moderna samhällsteorier. Stockholm: Prisma. 434 Reference List Marcus, G. E. and Fischer, M. M. J. 1986. Anthropolog y as cultural critique: an experimental moment in the human sciences. Chicago: University of Chicago Press. Marsalek, J. 1999. .DWDNRPEQtNXOWXUQtNRPSOH[QD8NUDMLQHDYSULOHKO\FKREODVWHFK 3UDHKLVWRULFD;;,9 0DUWLQH])HUQDQGH]*/DWHSUHKLVWRU\EODGHSURGXFWLRQLQ$QGDORXVLD 6SDLQ ,Q5DPRV0LOODQ$DQG%XVWLOOR0$ HGV Siliceous Rocks and Culture. Granada. Martinsson, H. 1985. Ålands stenålder. Kronologi, komparativa studier samt försök till bosättningsmodell. MA thesis in archaeology. Stockholm. Marzke, M. W. and Shackley, M. S. 1986. Hominid hand use in the Pliocene and Pleistocene: Evidence from experimental archaeology and comparative morphology. Journal of Human Evolution Marzke, M. W., Toth, N., Schick, K., Reece, S., Steinberg, B., Hunt, K., Linscheid, 5/DQG$Q.1(0*VWXG\RIKDQGPXVFOHUHFUXLWPHQWGXULQJKDUG hammer percussion manufacture of Oldowan tools. American Journal of Physical Anthropolog y Masson, W. M. and Merpert, N. J. (eds.). 1982. Eneolit SSSR. Moskwa. Massone, M. 1987. Los cazadores paleoindios de Tres Arroyos (Tierra del Fuego Anales del Instituto de la Patagonia, Serie Ciencias Sociales Mathiassen, T. 1948. Danske Oldsager I, Ældre Stenalder. Danmarks biblioteksskole. København. 0DWUDV]HN%0LJDO:6DâDFLľVNL66NâDGDQNLIRUPUG]HQLRZ\FK]MDP\ 424 z osady kultury pucharów lejkowatych ze stanowiska „Nad Wawrem” w =âRWHMZRMŒZLęWRNU]\VNLH,Q0DWUDV]HN%DQG6DâDFLľVNL6 HGV .U]HPLHľ œZLHFLHFKRZVNLZSUDG]LHMDFKWarszawa. Mauss, M. 1927. Division et proportions des divisions de la sociologie. Année sciologique 2. Paris. — 1947. Manuel d´etnographie. Paris: Payot. — 1979. The Notion of Body Techniques. In: Mauss, M.: Sociolog y and Psycholog y. Essays. London: Routledge and Kegan Paul. 0D\HU2DNHV: El Inga. A Paleoindian site in the Sierra of Northern Ecuador. Transactions of the American Philosophical Society 76/4. Philadelphia. Mazzanti, D. 1997. Excavaciones arqueológicas en el sitio Cueva Tixi, Buenos Aires, Argentina. Latin American Antiquity McBryde, I. 1984. Kulin greenstone quarries: the social contexts of production and distribution for the Mt William site. World Archaeolog y McCarthy, F.D. 1976. Australian Aboriginal stone implements, including bone, shell and teeth implements. Sydney: Australian Museum. McGowan, K. 1962. Early Man in the New World. Albany, N.Y.: Doubleday and Co.; Garden City, N.Y.: Anchor Books. Meinander, C. F. 1954: Die Kiukaiskultur. Suomen muinaismuistoyhdistyksen DLNDNDXVNLUMD)LQVNDIRUQPLQQHVI|UHQLQJHQVWLVNULIW 53. Helsinki. — 1957. Kolsvidja. 6XRPHQPXLQDLVPXLVWR\KGLVW\NVHQDLNDNDXVNLUMD)LQVND fornminnesföreningens tidskrift 435 Skilled Production and Social Reproduction 0HQJKLQ2)$(O$OWRSDUDQDHQVHAmpurias;9,,;9,,, 0LJDO:3URZDG]HQLHSUDFDUFKHRORJLF]Q\FKZGXů\FKNRSDOQLDFK NU]HPLHQLD]SHUVSHNW\Z\EDGDľZ.U]HPLRQNDFK2SDWRZVNLFK,Q Borkowski, W. (ed.)0HWRG\EDGDľDUFKHRORJLF]Q\FKVWDQRZLVNSURGXNF\MQ\FK²JyUQLFWZR krzemienia, Warszawa. ³=DP\VâWHFKQRORJLF]Q\ZLyUDNU]HPLHQQHJR]:LQLDUJP'ZLNR]\ In: 0DWUDV]HN%DQG6DâDFLľVNL6 HGV .U]HPLHľœZLHFLHFKRZVNLZSUDG]LHMDFK Warszawa. ³DQG%DUVND.7KHUROHRIH[SHULPHQWDOÁLQWNQDSSLQJIRUWKH UHFRQVWUXFWLRQRI1HROLWKLFÁLQWSURFHVVLQJ 5 ($ 0LOOHU72-U6WRQHZRUNRIWKH;rWi,QGLDQVRI%UD]LO,Q+D\GHQ% HG /LWKLF8VH:HDU$QDO\VLV. New York. 0LOOHW5LFKDUG/$+DELWDWVHWDWHOLHUVGHWDLOOHDX1pROLWKLTXHÀQDOGDQVOD UpJLRQGX*UDQG3UHVVLJQ\ ,QGUHHW/RLUH WHFKQRORJLHOLWKLTXH. Thèse de Doctorat de O·8QLYHUVLWpGH3DULV, 3UpKLVWRLUH(WKQRORJLH$QWKURSRORJLH YROS 180 Fig. 21 tab. Paris. Millon, R. F. 1973. 8UEDQL]DWLRQDW7HRWLKXDFDQ0H[LFR7KH7HRWLKXDFDQPDSYROSDUW 1. Austin: University of Texas Press. — 1981. Teotihuacan: City, state, and civilization. In: Sabloff, J. (ed). Handbook of Middle American Indians, Supplement 1: Archaeolog y. Austin: University of Texas Press. — 1988. The last years of Teotihuacan dominance. In: Yoffee, N. and Cowgill, G. (eds). The Collapse of Ancient States and Civilizations. Tucson: University of Arizona Press. –––– Drewitt, B. and Cowgill, G. W. 1973. 8UEDQL]DWLRQDW7HRWLKXDFDQ0H[LFRYRO 7KH7HRWLKXDFDQPDSSDUW0DSV. Austin: University of Texas Press. Minar, C. J. 2001. Motor skills and the learning process: the conservation of Cordage Final Twist Direction in communities of practice. In: Minar, C. J. and Crown, P. (eds.) Learning and craft production. Journal of Anthropological Research ² — and Crown, P. 2001 (eds.). Learning and craft rroduction. Journal of Anthropological Research 57(4). Miotti, L. 1992. Paleoindian Occupation at Piedra Museo Locality, Santa Cruz Province, Argentina. Current Research in the Pleistocene Mithen, S. 1996. The prehistory of the mind: the cognitive origins of art, religion and science. London: Thames and Hudson. — 2000 (ed.). Hunter-gatherer landscape archaeolog y: the Southern Hebrides Mesolithic. Cambridge: McDonald Institute. Montelius, O. 1917. 0LQQHQIUnQYnUIRUQWLG,6WHQnOGHURFKEURQVnOGHU²EHVNULYQLQJDU Stockholm: Norstedt and Söners Förlag Moore, J. and Scott, E. (eds.) 1997: Invisible people and processes. Writing gender and childhood into European archaeolog y. London. Moore, M. W. 2000. Kimberley Spearpoints of Northwestern Australia, Chips 12/3: 5–17. 436 Reference List — 2003. Flexibility of stone tool manufacturing methods on the Georgina River, Camooweal, Queensland. Archaeolog y in Oceania Morgado Rodriguez, A. 2002. Transformacion social y produccion de hojas de silex durante la Prehistoria Reciente de Andalucia Oriental. La estrategia de la complejitad. Tesis Doctoral, Universidad de Granada. Muto, G. 1971. A technological analysis of the early stages in the manifacture of lithic artifacts. M.A. Thesis, Idaho State University. Müller, S. 1900. Kapitel V, Oldsager. In: Madsen, A. P., Müller, S., Neergaard, C. et al. Affaldsdynger fra Stenalderen i Danmark. København. — 1902. Flintdolkene i den nordiske stenalder. Nordiske Fortidsminder I. Køpenhamn. Nærøy, A. J. 2000: Stone Age living spaces in western Norway. BAR International Series 857. Nami, H. G. 1986. Experimentos para el estudio de la tecnología bifacial de las ocupaciones tardías en el extremo sur de la Patagonia Continental. PREP. Informes de Investigación — 1987. Informe sobre la segunda y tercera expedición en la Cueva del Medio. Perspectivas Arqueológicas para la Patagonia Austral. Anales del Instituto de la Patagonia — 1988. Arqueología experimental, tecnología, artefactos bifaciales y modelos. Estado actual del conocimiento en Patagonia y Tierra del Fuego. Anales del ,QVWLWXWRGHOD3DWDJRQLD 6HULH&LHQFLDV6RFLDOHV — 1991. Desechos de Talla y Teoría de Alcance Medio: Un caso de Península Mitre. Shincal — 1992. Noticia sobre la existencia de Técnica ”Levallois” en Península Mitre, Extremo Sudoriental de Tierra del Fuego. Anales del Instituto de la Patagonia (Serie &LHQFLDV+XPDQDV ³a. Observaciones sobre desechos de talla procedentes de las ocupaciones tempranas de Tres Arroyos (Tierra del Fuego, Chile). Anales del ,QVWLWXWRGHOD3DWDJRQLD 6HULH&LHQFLDV6RFLDOHV ³b. Aportes para el conocimiento de técnicas líticas del Pleistoceno Final. Análisis de artefactos bifaciales del Norte de Venezuela (Colección Edmonton, Canada). Relaciones de la Sociedad Argentina de Antropología XIX: ÀJXUDV ³3DOHRLQGLRFD]DGRUHVUHFROHFWRUHV\WHFQRORJtDOtWLFDHQHOH[WUHPRVXUGH Sudamérica Continental. In: Lanata, J. L. and Borrero, L. A. (eds.) Arqueología de Cazadores-Recolectores. Límites, Casos y Aperturas. Arqueología Contemporánea 5: — 1995. Nota sobre la presencia de núcleos preparados y lascas predeterminadas en Puerto Esperanza (Misiones, Argentina). Cuadernos del Instituto Nacional de Antropología y Pensamiento Latinoamericano — 1996. New assessments of early human occupations in the southern cone. In: Akazawa, T. and Szathmáry, E. J. E. (eds.) Prehistoric Mongoloid Dispersals. Oxford: Oxford University Press. 437 Skilled Production and Social Reproduction — 1997. Investigaciones actualísticas para discutir aspectos técnicos de los FD]DGRUHVUHFROHFWRUHVGHOWDUGLJODFLDO(OSUREOHPD&ORYLV&XHYD)HOOAnales GHO,QVWLWXWRGHOD3DWDJRQLD 6HULH&LHQFLDV6RFLDOHV ³7HFKQRORJLFDOREVHUYDWLRQVRQWKH3DOHRLQGLDQDUWLIDFWVIURP)HOO·VFDYH Magallanes, Chile. Current Research in the Pleistocene — 1999a. Variaciones sobre rocas cuarcíticas. Relaciones de la Sociedad Argentina de Antropología;;,9 ³E3UHSDUHGÁDNHFRUHWHFKQLTXHLQWKH1RUWKHDVWHUQ3DWDJRQLDQ&RDVW Argentina. Bulletin of Primitive Technolog y — 1999c. The Folsom biface reduction sequence: evidence from the Lindenmeier collection. In: Amick, D. S. (ed.) ([SORULQJ3DWWHUQDQG9DULDWLRQLQ)ROVRP/LWKLF Technolog y: Late Pleistocene Hunter-Gatherers of the North American High Plains, International Monographs in Prehistory, Archaeological Series 12. Ann Arbor: University of Michigan. — 2001. Consideraciones tecnológicas preliminares sobre los artefactos líticos de Cerro de Los Burros (Maldonado, Uruguay). Comunicaciones de los Museos de Historia Natural y Antropología de Montevideo,,, — 2003a. Experimentos para explorar la secuencia de reducción Fell de la Patagonia Austral. Magallania — 2003b%UHYHVREVHUYDFLRQHVDFWXDOtVWLFRH[SHULPHQWDOHVVREUHODWpFQLFDGH núcleos preparados de la costa Norpatagónica. In: Sanguinetti de Bórmida A. C. (ed.). Arqueología de la Costa Norpatagónica (in press). — n.d. Experiments to understand North and South American Late Pleistocene lithic reduction sequence: an actualistic and comparative study. Ms. ³QG7KHRUHWLFDOUHÁHFWLRQVRQH[SHULPHQWDOOLWKLFWHFKQRORJ\DQGDUFKDHRORJ\ VRPHLVVXHVRQDFWXDOLVWLFVWRQHWRROV·DQDO\VLVDQGLQWHUSUHWDWLRQ,Q1DPL+ G. (ed.) Experiments and Interpretation of Traditional Technologies: Essays in Honor of Errett Callahan (in prep.). — and Elkin, D. 1994. Aportes para la categorización de instrumentos de talla en base a su densidad. CEIDER. Revista de Estudios Regionales — Norton, M. R., Stanford, D. J. and Broster, J. B. 1996. Comments on Eastern &ORYLVOLWKLFWHFKQRORJ\DWWKH&DUVRQ&RQQ6KRUWVLWH %1 7HQQHVVHH River Valley. Current Research in the Pleistocene Nationalencyklopedin.´3URGXNWLRQµ´ODJHUµDQG´ULWXDOµ Neeley, M. P. and Barton, C.M. 1994. A new approach to interpreting late Pleistocene microlith industries in southwest Asia. Antiquity Nelson, E. W. 1899: The Eskimo About Bering Strait. Eighteenth Annual report of the Bureau of American Ethnology, 1896–97. Washington, D. C. Newcomer, M. H. 1971. Some quantitative experiments on handaxe manufacture. World Archeologist — 1975. “Punch technique ” and Upper Paleolithic blades. In: Swanson, E. (ed.) Lithic technolog y: making and using stone tools. La Haye: Mouton. 1LOVVRQ/*9nUDOnQJWLGVPLQQHQTvärsnitt Nordin, S. 1981. Historia som vetenskap. En essä om marxismen, historicismen och humaniora. Lund: Zenit.. 438 Reference List 1RUGTYLVW%%RSODWVIXQNWLRQXUÁLQWEHDUEHWQLQJVSHUVSHNWLY²VHWWLWLGRFK rum. Förhistoria i mellersta Halland. Riksantikvarieämbetet och Statens Historiska Museer. Stockholm. ³5HGXNWLRQVSURFHVVHUDYERSODWVÁLQWDIUnQ+DOODQG(QVSDWLDORFK NURQRORJLVNVWXGLH,Q%URZDOO+3HUVVRQ3DQG6M|JUHQ.* HGV 9lVWVYHQVNDVWHQnOGHUVVWXGLHU Gotarc Serie C. Arkeologiska skrifter No. 8 Göteborgs Universitet Institutionen för arkeologi. Göteborg. Nordquist, P. 1999. (JDOLWHWVEHJUHSSHW²NDQMlPOLNDVDPKlOOHQH[LVWHUD" In: Goldhahn, J, and Nordquist, P. (eds). Marxistiska perspektiv inom skandinavisk arkeologi. Arkeologiska studier vid Umeå universitet 5. Umeå. — 2001. Hierarkiseringsprocesser. Om konstruktionen av social ojämlikhet i 6NnQHI.U6WXGLD$UFKDHRORJLFD8QLYHUVLWDWLV8PHQVLV8PHn Nuñez, L. J., Varela, R., Casamiquela, V., Schiappaccasse, H., Niemeyer y C. Villagrán. 1994. Cuenca de Taguatagua en Chile: El ambiente del Pleistoceno Superior y Ocupaciones humanas. Revista Chilena de Historia Natural 1XxH]0&OD\ÀJXULQHVIURPWKHcODQG,VODQGVDQGPDLQODQG)LQODQG Fennoscandia archaeologica ,,, ³$ÁLQWTXDUW]VXEVWLWXWHLQWKHcODQG$UFKLSHODJR8QLYHUVLWHWHWV Oldsaksamling Årsbok 1989/1990. Oslo. ³&DQQLEDOLVPRQ3LWWHG:DUHcODQG"Karhunhammas Nunn, G. R. this volume. Using the Jutland Type IC Neolithic Danish dagger as a PRGHOWRUHSOLFDWHSDUDOOHOHGJHWRHGJHSUHVVXUHÁDNLQJ Oakley, K. P. 1968. Man the tool-maker. Chicago: University of Chicago Press. 2·&RQQRU6&DUSHQWHU·V*DSURFNVKHOWHU\HDUVRI$ERULJLQDO occupation in the Napier ranges, Kimberley, WA. Australian Archaeolog y 40: — 1999. \HDUVRI$ERULJLQDORFFXSDWLRQ.LPEHUOH\1RUWK:HVW$XVWUDOLD Canberra: Department of Archaeology and Natural History and Centre for Archaeological Research, Australian National University. Terra Australis 14. Olausson, D. 1983a. Flint and groundstone axes in the Scanian Neolithic. An evaluation of raw materials based on experiments.6FULSWD0LQRUD/XQG ³E/LWKLFWHFKQRORJLFDODQDO\VLVRIWKHWKLQEXWWHGÁLQWD[HActa Archaeologica — 1997. Craft specialization as an agent of social power in the south Scandinavian Neolithic, In: Schild, R. and Sulgostowska, Z. (eds.). Man and Flint. Proceedings of WKH9,,WK,QWHUQDWLRQDO)OLQW6\PSRVLXP:DUVV]DZD2VWURZLHF6ZLHWRNU] \VNL6HSWHPEHU . Warzawa. ³%DWWOHD[HVKRPHPDGHPDGHWRRUGHURUIDFWRU\SURGXFWV? In: Holm, L. and Knutsson, K. (eds.) 1998. Proceedings from the Third Flint Alternatives Conference DW8SSVDOD6ZHGHQ2FWREHU. Occasional Papers in Archaeology 16. Uppsala. — 2000. Talking axes, social daggers. In: Olausson, D. and Vandkilde, H. (eds.) Form, Function and Context. Material Culture Studies in Scandinavian Archaeolog y. Lund: Almqvist and Wiksell International. 439 Skilled Production and Social Reproduction Olofsson, A. 2003. Pioneer settlement in the Mesolithic of Northern Sweden. Archaeology and Environment 16. Umeå University. Olsen, B. 1993. Finmarks forhistorie. Tromsö. — 1995. Bosetning og samfunn i Finnmarks forhistorie (Settlement and Society LQ)LQQPDUN·V3UHKLVWRU\ Norwegian archaeological review, Oslo, Scandinavian University Press. —1997. Fra ting til tekst. Teoretiske perspektiv i arkeologisk forskning. Universitetsforlaget. Oslo. ³0DWHULDOFXOWXUHDIWHUWH[WUHPHPEHULQJWKLQJVNorwegian Archaeological Review — and Alsaker, S. 1984. Greenstone and diabase Stone Age unilisation in the 6WRQH$JHRI:HVWHUQ1RUZD\WHFKQRORJLFDODQGVRFLRFXOWXUDODVSHFWVRID[H and adze production and distribution. Norwegian Archaeological Review 103. Olsson, E. 1996. Stenåldersboplats vid Häggsta, Botkyrka socken, Botkyrka kommun, Södermanland5DSSRUW5LNVDQWLNYDULHlPEHWHW896WRFNKROP 1996. Stockholm. ³*UDQDWK=LOOpQ*DQG0RKU$.RUVQlVHQJURSNHUDPLVNJUDYRFK ERSODWVSn6|GHUW|UQ6|GHPDQODQG*U|GLQJHVRFNHQ5$bRapport 5LNVDQWLNYDULHlPEHWHW896WRFNKROP 1994:63. Stockholm. Ortner, S. 1984. Theory in anthropology since the sixties. Comparative Studies in Society and History Osborn, H. F. 1916. Men of the Old Stone Age, their environment, life and art. London: G. %HOO1HZ<RUN6FULEQHU·V6RQV Osgood, C. 1940. Ingalik material culture. <DOH8QLYHUVLW\3XEOLFDWLRQVLQ$QWKURSRORJ \ . New Haven: Yale University Press. Østmo, E. 1977. Schaftlochäxte und landwirtschaftliche Siedlung. Eine Fallstudie über Kulturverhältnisse im südöstlichsten Norwegen im Spätneolithikum und in der älteren Bronzezeit. Acta Archaeologica 48. Copenhagen. 2WLV&KDUOWRQ&2EVLGLDQDVMHZHOU\/DSLGDU\SURGXFWLRQLQ$]WHF Otumba, Mexico. Ancient Mesoamerica Parry, W. 2002. Aztec blade production strategies in the eastern Basin of Mexico. In: Hirth, K. and Andrews, B. (eds.). Pathways to Prismatic Blades: A Study in Mesoamerican Obsidian Core-Blade Technolog y. Cotsen Institute of Archaeology, University of California, Los Angeles. Parsons, K. M. and Pearson, W. L. 1991. Three Anzick style bifaces. Current Research in the Pleistocene Paton, R. 1994. Speaking through stones: a study from northern Australia. World Archaeolog y 3DXQHUR56(OVLWLR&XHYDGHOD/RFDOLGDGDUTXHROyJLFD&HUUR Tres Tetas (Estancia San Rafael, Provincia de Santa Cruz, Argentina). Anales de Arqueología y Etnología 440 Reference List — 2000. Colonización humana en la Meseta Central de Santa Cruz: Aportes al problema desde la actual evidencia arqueológica en estancias San Rafael y La María. Paper presented at Taller La Colonización del sur de América durante la WUDQVLFLyQ3OHLVWRFHQR+RORFHQR, La Plata. Pelegrin, J. 1981. Experiments in bifacial work. )OLQWNQDSSHUV·([FKDQJH 4/1: 4–7. ³5pÁHFWLRQVXUOHFRPSRUWHPHQWWHFKQLTXH/DVLJQLÀFDWLRQFXOWXUHOOHGHV industries lithiques. Otte, M. (ed.) Actes du Colloque de Liège du 3 au 7 octobre 1984. Studia Praehistorica Belgica 4. BAR International Series 239. Oxford. — 1984a. Approche technologique experimentale de la mise en forme de nucleus pour le debitage systematique par pression. In: Tixier, J. (ed). Prehistoire de la Pierre Taillée II. Paris: CRES — 1984b. Debitage par pression sur silex: nouvelle experimentations. Prehistoire de la Pierre Taillée II. Paris: CRES. — 1988. Debitage experimental par pression “du plus petit au plus grand”. Technologie prehistorique: — 1990. Prehistoric lithic technology: some aspects of research. Archaeological Review from Cambridge — 1991. Sur une recherche technique expérimentale des techniques de débitage laminaire. In: $UFKpRORJLHH[SpULPHQWDOHWRPH$FWHVGXFROORTXHLQWHUQDWLRQDOGH Beaune, 1988, Expérimentations en Archéologie: bilan et perspectives. Paris: Errance, t. — 1994. Lithic technology in Harappan times. In: Parpolla, A. and Koskikallio, P. (eds). 6RXWK$VLDQ$UFKDHRORJ \SURFHHGLQJVRIWKHth International Conference RIWKH(XURSHDQ$VVRFLDWLRQRI6RXWK$VLDQ$UFKDHRORJLVWV+HOVLQNL-XO\ Annales Academiae Scientiarum Fennicae B 271. — 1995. Technologie lithique OH&KDWWHOSHUURQLHQGH5RF'H&RPEH /RW HWGH/D&{WH 'RUGRJQH Cahiers du Quarternaire 20. — 1997. Nouvelles observations sur le dépôt de lames de La Creusette (Barrou, ,QGUHHW/RLUH Bulletin de la Société des Amis du Musée du Grand-Pressigny 34. — 2000. Les techniques de débitage laminaire au Tardiglaciaire: critères de GLDJQRVHHWTXHOTXHVUpÁH[LRQV,Q9DOHQWLQ%%RGX3DQG&KULVWHQVHQ M. (eds). L´Europe Centrale et Septentrionale au Tardiglaciaire. Actes de Table-ronde LQWHUQDWLRQDOHGH1HPRXUVPDL Nemours, A.P.R.A.I.F. Mémoires du Musée de Préhistoire d´Ile de France. Paris. ³/DSURGXFWLRQGHVJUDQGHVODPHVGHVLOH[GX*UDQG3UHVVLJQ\,Q Guilaine, J. (ed.) 0DWpULDX[SURGXFWLRQVFLUFXODWLRQVGX1pROLWKLTXHjO·$JHGXEURQ]H. Paris: Errance. — 2002. Principes de la reconnaisance des méthodes et techniques de taille. In: Chabot, J. (ed). Tell`Atij Tell Gueda. Analyse technologique et fonctionelle. Quebec, 8QLYHUVLWH/DYDO&DKLHUVG·DUKpRORJLHGX&(/$74XHEHF — 2003. Blade making techniques from the Old World: Insights and applications to Mesoamerican obsidian lithic technology. In: Hirth, K. (ed.) Experimentation and Interpretation in Mesoamerican Lithic Technolog y. Salt Lake City: University of Utah Press. 441 Skilled Production and Social Reproduction ³LQSUHS1RWHVWHFKQRORJLTXHVVXUOHVSLqFHVHQVLOH[GX*UDQG3UHVVLJQ\GH &KDODLQ&ODLUYDX[UDSSRUWpHVjODÀQGXème millénaire et au tout début du 3èmePLOOpQDLUH,Q3pWUHTXLQ3DQG3pWUHTXLQ$0 'LUV Les sites littoraux QpROLWKLTXHVGH&ODLUYDX[HWGH&KDODLQ -XUD WRPH,9'X)HUULqUHVDXJURXSHGH &ODLUYDX[ eHWeVLqFOHVDY-& Paris: Ed. de la MSH (Coll. Archéologie et Culture matérielle). — this volume. Long blade technology in the Old World: An experimental approach and some archaeological results. Perlès, C. 1990. /HVLQGXVWULHVOLWKLTXHVWDLOOpHVGH)UDQFKWKL $UJROLGH*UqFH WRPH,,/HV industries du Mésolithique et du Néolithique initial. Bloomington and Indianapolis: Indiana University Press, 288p. (Coll. Excavations at Franchthi Cave, Greece, vol.5). — 2001. 7KHHDUO\1HROLWKLFLQ*UHHFH7KHÀUVWIDUPLQJFRPPXQLWLHVLQ(XURSH. Cambridge (UK): Cambridge: Cambridge University Press. — 2004. /HVLQGXVWULHVOLWKLTXHVWDLOOpHVGH)UDQFKWKL $UJROLGH*UqFH WRPH,,,'X 1pROLWKLTXHDQFLHQDX1pROLWKLTXHÀQDO. Bloomington, Ind.: Indiana University Press. Petersen, E. B. 1973. A Survey of the Late Palaeolithic and the Mesolithic of Denmark. Kozlowski, S. K (ed). The Mesolithic in Europe. Warszawa. 3HWUHTXLQ33HWUHTXLQ$0-HXG\)-HXQHVVH&0RQQLHU-/3HOHJULQ J.and Praud, I. 1998. From the raw material to the Neolithic stone axe: production processes and social context. In: Edmonds, M. and Richards, C. (eds). 8QGHUVWDQGLQJWKH1HROLWKLFRI1RUWK:HVWHUQ(XURSH Glasgow: Cruithne Press. 3HWUL+6WHUEHQGH:HOWLQ1RUGZHVW$XVWUDOLHQ%UDXQVFKZHLJ Pfaffenberger, B. 1992: Social anthropology of technology. Annual Review of Anthropolog y 21: 491–516. Pfeiffer, J. E. 1972. The Emergence of Man. New York: Harper and Row. 3LJHRW17HFKQLFDODQGVRFLDODFWRUVÁLQWNQDSSLQJVSHFLDOLVWVDQG apprentices at Magdalenian Etoilles. Archaeological Review from Cambridge 9/1: Piggott, S. 1950. Prehistoric India. London: Penguin Books. — 1967. Ancient Europe from the beginnings of Agriculture to Classical Antiquity. Edinburgh: Edinburgh University Press. — 1983. The earliest wheeled transport, from the Atlantic Coast to the Caspian Sea. London: Thames and Hudson. — 1992. Wagon, chariot and carriage: symbol and status in the history of transport. New York: Thames and Hudson. Pitts, M., 1996. The stone axe in Neolithic Britain, Proceedings of the Prehistoric Society, Podborský, V. 1993. (ed.). Praveké dejiny Moravy. Vlastiveda Moravská Zeme a Lid. Nová rada. Svajzek 3, Muzejní a vlastivedná spolecnost v Brné. Brno. Politis, G. 1991. Fishtail projectile points in the southern cone of South America: an overview. In: Bonnichsen, R. and Turnmire, K. L. (eds.) Clovis: Origins and Adaptations. Corvallis: Oregon: Center for the Study of the First Americans. 442 Reference List 3RSOLQ)(WXGHFRPSDUDWLYHGHGHX[VpULHVGHFKDVVHODPHHQERLVGHFHUI QpROLWKLTXHVGHO·<RQQH )UDQFH HWLQGLHQQHGX0LVVRXUL 86$ ,QCongrès 3UpKLVWRULTXHGH)UDQFHèmeVHVVLRQ3URYHQFH ³/HVFKDVVHODPHQpROLWKLTXHVHQERLVGHFHUI,Q&DPSV)DEUHU+ HG /·,QGXVWULHHQRVHWERLVGHFHUYLGpGXUDQWOH1pROLWKLTXHHWO·$JHGHV0pWDX[3UHPLqUH UpXQLRQGXJURXSHGHWUDYDLOQVXUO·LQGXVWULHGHO·RVSUpKLVWRULTXH. Paris: CNRS. ³'HVFKDVVHODPHQpROLWKLTXHVHQERLVGHFHUIGHO·<RQQHGH6SLHQQHVHW SRXUTXRLSDVGX*UDQG3UHVVLJQ\,QEtudes sur le Néolithique de la région Centre. Actes du colloque interrégional sur le Néolithique6DLQW$PDQG0RQWURQGRFW Porteus, S.D. 193. The psycholog y of a primitive people: a study of the Australian Aborigine. London: Edward Arnold. Räihälä, O. 1999. Tutkimuksia Suomussalmen kivikautisesta asutuksesta kvartsien fraktuurianalyysin avulla. Studia septentrionalia 35. Rajamailla V, 1998: 117– 136. 5DQNDPD74XDUW]LWHDW8WVMRNL$OD-DOYH7KHIUDPHRIDFDVHVWXG\ 8QLYHUVLWHWHWV2OGVDNVDPOLQJcUERN 1989/1990:103–117. — 1997. Ala-Jalve. spatial, technological, and behavioral analyses of the lithic assemblage from D6WRQH$JH(DUO\0HWDO$JHVLWHLQ8WVMRNL)LQQLVK/DSODQG. BAR International Series 681. Oxford. — 2002. Analyses of the quartz assemblages of houses 34 and 35 at Kauvonkangas in Tervola. In: Ranta, H. (ed.). Huts and Houses. Stone Age and Early Metal Age Buildings in Finland. Jyväskylä. Helsinki. — 2003. Quartz analyses of Stone Age house sites in Tervola, southern Finnish Lapland. In: Samuelsson, C. and Ytterberg, N. (eds.). 8QLWLQJVHD6WRQH$JH VRFLHWLHVLQWKH%DOWLF6HDUHJLRQ3URFHHGLQJVIURPWKH)LUVW8QLWLQJ6HD:RUNVKRS DW8SSVDOD8QLYHUVLW\6ZHGHQ-DQXDU\² .Occasional Papers in Archaeology 33. Uppsala. —and Ukkonen, 2001. On the early history of the wild reindeer (Rangifer tarandus) in Finland. Boreas Rasmussen, L. W. 1990. Dolkproduktion og distribution in senneolitikum. Hikuin 5DWKMH://DVWWDQJRLQ0D\DSDQ$WHQWDWLYHWUDMHFWRU\RISURGXFWLRQ GLVWULEXWLRQV\VWHPV,Q6DEORII-DQG/DPEHUJ.DUORYVN\&& HGV Ancient Civilization and Trade. Albuquerque: University of New Mexico Press. Ravn, M. 1993. Analogy in Danish prehistoric studies. Norwegian Archaeological Review 5HJQHU(0\WKLVWRULDHOOHUDQQDWI|UÁXWHW"cWHUDQYlQGQLQJDYlOGUH anläggningar under vikingatid. In: Nordström, P. and Svedin, M. (eds.) Aktuell Arkeologi 9,, 5HQDXOW6(FRQRPLHGHODPDWLqUHSUHPLqUH/·H[HPSOHGHODSURGXFWLRQDX 1pROLWKLTXHÀQDOHQ3URYHQFHGHVJUDQGHVODPHVHQVLOH[]RQpROLJRFqQHGX EDVVLQGH)RUFDOTXLHU $OSHVGH+DXWH3URYHQFH ,Q'·$QQD$HW%LQGHU D. (Dir.) Production et identité culturelle. Actualité de la recherche. Actes de la deuxième VHVVLRQGHV5HQFRQWUHVPpULGLRQDOHVGH3UpKLVWRLUHUpFHQWH $UOHV . Antibes : (GLWLRQV$3'&$ 443 Skilled Production and Social Reproduction — Poupeau, G. and Dubernet, S. in prep. 5HFRQQDLVVDQFHGHO·HPSORLG·XQFRPSUHVVHXU GHFXLYUHSRXUODSURGXFWLRQGHODPHVGHODÀQGX1pROLWKLTXHOHGpS{WGHODPHVGH6DLQWH &pFLOHOHV9LJQHV 9DXFOXVH Renfrew, C. 1993. Arkeologi och språk. Det indoeuropeiska ursprungets gåta. Symposion. Stockholm; Stehag: Kulturhistoriskt bibliotek. — and Zubrow, E. B. W. (eds.) 1994. The ancient mind: elements of cognitive archaeology. Cambridge: Cambridge University Press. Rice, P. M. 1981 Evolution of specialized production: a trial model. Current Anthropolog y Rieth, A. 1958. Zur Technik des Steinbohrens im Neolithikum. Zeitschrift für Schweizerische Archäologie und Kunstgesichte WDIHOQ 5LPDQWLHQp5'HU$+RUL]RQW²(OHPHQWHLQGHU+DIINVWHQNXOWXULQ /LWDXHQ,Q6LHPHQ3 HG (DUO\&RUGHG:DUH&XOWXUH7KH$KRUL]RQ²)DFW or Fiction. Arkeologiske rapporter fra Esbjergs museum. Esbjerg. Roberts, R., Walsh, G. Murray, A. et al. 1997. Luminescence dating of rock art DQGSDVWHQYLURQPHQWVXVLQJPXGZDVSQHVWVLQQRUWKHUQ$XVWUDOLDNature Rock, I. 1973. Orientation and Form. London: Academic Press. Rodriguez, J. A. 1992. Arqueología del Sudeste de Sudamérica. In: Meggers, J. B. (ed.). Prehistoria Sudamericana: Nuevas Perspectivas. Washington D.C.: Taraxatum 5RH3*6W\OHVRFLHW\P\WKDQGVWUXFWXUH,Q&DUU&DQG1HLW]HO-( (eds). Style, Society and Person: Archaeological and Ethnological Perspectives. New York: Plenum Press. Rohner, J.R. 1970. Techniques of making plastic casts of artifacts from permanent molds. American Antiquity 5RRW0-$QDO\VLVRIVWRQHWRROVDQGÁDNLQJGHEULV6LWH'8$ )ROVRPRFFXSDWLRQRIWKH.QLIH5LYHUÁLQWSULPDU\VRXUFHDUHD3KDVH,,, (Part 1 In: Root, M. J. (ed). Archaeological Data Recovery at Lake Ilo National :LGOLIH5HIXJH'XQQ&RXQW\1RUWK'DNRWD,QWHUQ5HSRUWIRU,QYHVWLJDWLRQV DW'8$)LQDO5HSRUW3URMHFW5HSRUW1XPEHU&HQWHUIRU Northwest Anthropology, Department of Anthropology, Washington State University, Pullman. Rosen S.A. 1997. Lithics after the Stone Age: A handbook of stone tools from the Levant. Walnut Creek, Calif.: Alta Mira Press Rosenlund, L. 2000. 6RFLDOVWUXFWXUHVDQGFXOWXUDOFKDQJHV$SSO\LQJ3LHUUH%RXUGLHX·V approach and analytical framework. Doctoral Dissertation. Stavanger. Roux, V. 1990. The psychological analysis of technical activities: A contribution to the study of craft specialization, Archaeological Review of Cambridge ³7KHZKHHOWKURZLQJWHFKQLTXHGHÀQLWLRQDQGLGHQWLÀFDWLRQRQWKHEDVLV of ceramic surface features. Man and Environment — 1999. Ethnoarchaeology and the generation of referential models: The case of Harappan Carnelian beads. 8UJHVFKLFKWH0DWHULDOKHIWH — and Pelegrin, J. 1989. Knapping technique and craft specialization: an ethnoarchaeological investigation in Gujarat. Puratattva 444 Reference List ³%ULO%DQG'LHWULFK*6NLOOVDQGOHDUQLQJGLIÀFXOWLHVLQYROYHGLQ VWRQHNQDSSLQJ7KHFDVHRIVWRQHEHDGNQDSSLQJLQ.KDPEKDW,QGLDWorld Archaeolog y 5R]R\-*7\SRORJLHGHO·(SLSDOpROLWKLTXHIUDQFREHOJH/·pWXGHGX matériel brut et des microburins. Bulletin de la Société Préhistorique Française, Etudes et Travaux 5XELQRV3HUH]$5HFRSLODFLyQ\DQiOLVLVGHODVIHFKDVFDUERQRGHOQRUWH de la provincia de Santa Cruz (Argentina). In: Aguerre, A. M. (ed). Arqueología y Paleoambiente en la Patagonia Santacruceña Argentina. Buenos Aires. Rudebeck, E. 1998. Flint extraction, axe offering, and the value of cortex. In: Edmonds, M. and Richards, C. (eds). 8QGHUVWDQGLQJWKH1HROLWKLFRI1RUWK:HVWHUQ Europe.*ODVJRZ&UXLWKQH3UHVV — 2000. Tilling nature, harvesting culture. Acta Archaeologica Lundensia. Series in 80, No 32. Lund. — and Ödman, C. 2000. .ULVWLQHEHUJ(QJUDYSODWVXQGHUnU. Malmöfynd 7. Stadsantikvariska avd. Kultur Malmö. 5\GEHFN26OXWQDPDUNRFKPRVVI\QGIUnQVWHQnOGHUQL/XQGV universitets historiska museum, deras tidsställning och samband med religiösa föreställningar. )UnQ/XQGV8QLYHUVLWHWV+LVWRULVND0XVHXP6NULIWXWJLYHQPHG DQOHGQLQJDIPXVHHWVLQÁ\WWQLQJLGHVVQ\DKHP. Lund. Sahlins, M. 1988. Kapten Cooks död. Författarförlaget. Stockholm. Salomonsson, B. 1971. Malmö Stads historia. Första delen. Malmötraktens förhistoria. Utgiven på uppdrag av Stadsfullmäktige i Malmö. Malmö. Sandberg, F. in press. Unpublished report, Dalarnas museums arkiv, dnr 75/98. Sanders, W. T. and Santley, R. S. 1983. A tale of three cities: Energetics and urbanism in Prehispanic Central Mexico. In: Vogt, E. and Leventhal, R. (eds). Prehistoric Settlement Patterns: Essays in Honor of Gordon R. Willey. Albuquerque: University of New Mexico Press. — Parsons, J. R. and Santley, R. S. 1979. The Basin of Mexico: Ecological process in the evolution of a civilization. New York: Academic Press. Sanjek, R. 1991. The Ethnographic Present. MAN 6DQWLOOR)UL]HOO%+lQGHUQDVW\VWDNXQVNDS2PDSXOLVNDNXSROE\JJDUHV \UNHVNXQQDQGHRFKGHVVUHSURGXFWLRQ,Q+MlUWKQHU+ROGDU(DQG& Risberg (eds). +DQWYHUNHWVUROOLVDPKlOOHW²SURGXNWLRQRFKUHSURGXNWLRQ. Uppsala, Riksantikvarieämbetet UV GAL 6DQWOH\562EVLGLDQH[FKDQJHHFRQRPLFVWUDWLÀFDWLRQDQGWKHHYROXWLRQ of complex society in the Basin of Mexico. In: Hirth, K. (ed). Exchange in Early Mesoamerica. Albuquerque: University of New Mexico. — and Alexander, R. T. 1996. Teotihuacan and Middle Classic Mesoamerica: A Precolumbian World System. In: Mastache, A., Parsons, J., Santley, R. and Serra Puche, M. C. (eds). Arqueología Mesoamericana: Homenaje a William Sanders, vol 1. Mexico, Instituto Nacional de Antropología e Historia. 445 Skilled Production and Social Reproduction — Kerley, J. M. and Barrett, T. P. 1995 Teotihuacan period lithic assemblages from the Teotihuacan Valley, Mexico. In: Sanders, W. (ed). 7KH7HRWLKXDFDQ9DOOH\ 3URMHFW)LQDO5HSRUW9RO7KH7HRWLKXDFDQ3HULRG2FFXSDWLRQRIWKH9DOOH\3DUW Artifact Analyses. Occasional Papers in Anthropology, No. 20. University Park: Matson Museum of Anthropology, Pennsylvania State University. ³.HUOH\-0DQG.QHHERQH552EVLGLDQZRUNLQJORQJGLVWDQFH H[FKDQJHDQGWKHSROLWLFRHFRQRPLFRUJDQL]DWLRQRIHDUO\VWDWHVLQ&HQWUDO Mexico. In: Isaac, B. (ed). Economic Aspects of Prehispanic Highland Mexico. Research in Economic Anthropology, Supplement 2. Greenwich, Conn.: JAI Press. — and Pool, C. A. 1993. Prehispanic exchange relationships among Central Mexico, the Valley of Oaxaca, and the Gulf Coast of Mexico. In: Ericson, J. and Baugh, T. (eds). The American Southwest and Mesoamerica: Systems of Prehistoric Exchange. New York: Plenum Press Sarauw, G. and Alin, J. 1923. Götaälvsområdets fornminnen. Göteborg. Skrifter utgivna till Göteborgs stads trehundraårsjubileum. Sarnäs, P. and Nord Paulsson, J. 2001. gUHVXQGVI|UELQGHOVHQ6NMXWEDQRUQD%DQG (OLQHOXQG$%5DSSRUW|YHUDUNHRORJLVNVOXWXQGHUV|NQLQJ Rapport nr 9. Malmö Kulturmiljö. Sauzade, G. 1983. /HVVpSXOWXUHVGX9DXFOXVHGX1pROLWKLTXHjO·$JHGX%URQ]H. Editions du Laboratoire de Paléontologie humaine et de Préhistoire. Etudes quaternaires, mémoire 6. Paris. Schama, S. 1995. Skog: Landskap och minne. En civilisationshistoria. Värnamo: Wahlström and Widstrand. Schick, K.D. and Toth, N. 1994. Early Stone Age technology in Africa: A review and case study into the nature and function of spheroids and subspheroids. In: Corruccini, R. S. and Ciochon, R. L. (eds.) Integrative Paths to the Past: Paleoanthropological advances in honor of F. Clark Howell. Englewood Cliffs, N. J: 3UHQWLFH+DOO ³7RWK1*DUXÀ*6DYDJH5XPEDXJK(65XPEDXJK'DQG6HYFLN5 &RQWLQXLQJLQYHVWLJDWLRQVLQWRWKHVWRQHWRROPDNLQJDQGWRROXVLQJ capabilities of a Bonobo (Pan paniscus). Journal of Archaeological Science 832. Schiffer, M.B. 1987. Formation processes of the archaeological record. Albuquerque: University of New Mexico Press — and Skibo, J. M. 1987. Theory and experiment in the study of technological change. Current Anthropolog y Schild, R. 1971. Lokalizacja prahistorycznych punktów eksploatacji krzemienia F]HNRODGRZHJRQDSyâQRFQRZVFKRGQLPREU]HůX*yUŒZLęWRNU]\VNLFKFolia Quaternaria — 1976. Flint mining and trade in Polish prehistory as seen from perspective of WKHFKRFRODWHÁLQWRI&HQWUDO3RODQG$VHFRQGDSSURDFKActa Archaeologica Carpathica — 1980. Introduction to dynamic technological analysis of chipped stone assemblages. In: Plater, K. (ed). 8QFRQYHQWLRQDO$UFKDHRORJ \Warsaw. 446 Reference List — Królik, H. and Marczak, M. 1985. Kopalnia krzemienia czekoladowego w Tomaszowie, :URFâDZ:DUV]DZD.UDNyZáyGŭ — Marczak, M. and Królik, H. 1975. 3yŭQ\PH]ROLW3UyEDZLHORDVSHNWRZHMDQDOL] \ otwartych stanowisk piaskowych:URFâDZ:DUV]DZD.UDNyZ*GDľVN Schlanger, N. 1994. Mindful technology: unleashing the chaîne opératoire for an archaeology of mind. In: Renfrew, C and Zubrow, E. (eds). The ancient mind. Elements of cognitive archaeolog y. New directions in archaeology. Cambridge. — 1996. Understanding the Levallois: Lithic Technology and Cognitive Archaeology. Cambridge Archaeological Journal Schmitz, P. I. 1987. Prehistoric hunters and gatherers of Brazil. Journal of World Prehistory 6FKlIHU-&RQMRLQLQJDUWHIDFWVDQGFRQVLGHUDWLRQRIUDZPDWHULDO WKHLUDSSOLFDWLRQDWWKH0LGGOH3DODHROLWKLFVLWHRIWKH6FKZHLQVNRSI Karmelenberg. In: Cziesla, E., Eickhoff, S., Arts, N., Winter, D. (eds.) The %LJ3X]]OH,QWHUQDWLRQDO6\PSRVLXPRQ5HÀWWLQJ6WRQH$UWHIDFWV. Studies in Modern Archaeology, Vol. 1. Bonn. Segerberg, A. 1974. Den enkla skafthålsyxan av sten. Fyndomständigheter och dateringsproblem. BA Thesis. Uppsala University. — 1978. Den enkla skafthålsyxan av sten. Fyndförhållanden och dateringar. Tor ;9,, — 1999. %lOLQJH0RVVDU.XVWERUL8SSODQGXQGHU\QJUHVWHQnOGHUQ. Aun 26. Uppsala: Societas Archaeologica Upsaliensis. Sehested, N. F. B. 1884. $UFK RORJLVNH8QGHUV¡JHOVHU. København. Semaw, S., Renne, P., Harris, J. W. K., Feibel, C.S., Bernor, R. L., Fesseha, N. and 0RZEUD\.PLOOLRQ\HDUROGVWRQHWRROVIURP*RQD(WKLRSLD Nature Semenov, S. A. 1964. Prehistoric Technolog y, London: Adams and Dart. Sheehy, J. 1992 &HUDPLFSURGXFWLRQLQDQFLHQW7HRWLKXDFDQDFDVHVWXG\IURP7ODMLQJD Unpublished Ph.D. dissertation. Pennsylvania State University, University Park, Pa. Sheets, P. D. 1975. Behavioral analysis and the structure of a prehistoric industry. Current Anthropolog y Sheets, P. D. 1978. From craftsman to cog: Quantitative views of Mesoamerican lithic technology. In: Sidrys, R. (ed). Papers on the Economy and Architecture of the Ancient Maya. Monograph 8, Institute of Archaeology, University of California, Los Angeles. — 1983. Chipped stone from the Zapotitán Valley. In: Sheets, P. (ed). Archaeolog y DQG9ROFDQLVPLQ&HQWUDO$PHULFD7KH=DSRWLWDQ9DOOH\RI(O6DOYDGRU. Austin: University of Texas Press. Shennan, S. 1986. Interaction and change in third millennium BC western and central Europe. In: Renfrew, C. and Cherry, J. F. (eds). Peer Polity Interaction and Sociopolitical Change. Cambridge: Cambridge University Press. — 2000. Population, culture history, and the dynamics of culture change. Current Anthropolog y 447 Skilled Production and Social Reproduction — and Steele, J. 1999. Cultural learning in hominids: a behavioural ecological approach. In: Box, H. and Gibson, H. (eds). Mammalian Social Learning. Cambridge: Cambridge University Press. Sherratt, A. 1987, Cups that cheered. In: Waldren, W. H. and Kennard R. C. (eds). Bell Beakers of the West Mediterranean, British Archaeological Report, International Series 331. Oxford. ³$OFRKRODQGLWVDOWHUQDWLYHVV\PERODQGVXEVWDQFHLQSUHLQGXVWULDO cultures, In: Goodman, J., Lovejoy, P. E. and Sherratt, A. (eds.) Consuming habits: drugs in history and anthropolog y. London; New York: Routledge. 6KQLHUO·PDQ9$Who gets the past? Competition for ancestors among nonRussian intellectuals in Russia. Washington DC: Woodrow Wilson Center Press; Baltimore: Johns Hopkins University Press — 2001. The value of the past: Myths, identity and politics in Transcaucasia. Senri Ethnological Studies, no. 57. Osaka, Japan: National Museum of Ethnology. Siiriäinen, A. 1977. Quartz, chert and obsidian: a comparison of raw materials in a Late Stone Age aggregate in Kenya. Finskt museum 1974: 15–29. — 1981. Problems of the East Fennoscandian Mesolithic. Finskt Museum 1977: 5–31. Sillitoe, P. and Hardy, K. 2003. Living lithics: Ethnoarchaeology in Highland Papua New Guinea. Antiquity 77/297: 555–566. Simonsen, J. 1982. Bolig eller dødehus?0XVHHU,9LERUJDPW 11. Viborg. 6NDNXQ11/HUROHHWO·LPSRUWDQFHGXVLOH[GDQVOH&KDOFROLWKLTXHGXVXG HVWGHO·(XURSH VXUODEDVHGXPDWHULHOSURYHQDQWGHVIRXLOOHVGXFDPSHPHQW de Bodaki). In: /DSUpKLVWRLUHDXTXRWLGLHQ 0pODQJHVRIIHUWVj3LHUUH%RQHQIDQW Grenoble 6NDU%7KH6FDQLDQ0DJOHPRVH6LWH%DUUH0RVVH,,$UHH[DPLQDWLRQE\ UHÀWWLQJActa Archaeologica — and S. Coulson 1989. A Case Study of Rørmur II: a Norweigan Early Mesolithic Site. In: Bonsall, C. (ed). Mesolithic Europe. Edinburg: University of Edinburg. Slanger, N 1996 Understanding the Levallois: lithic technology and cognitive archaeology. Cambridge Archaeological Journal9RO Smith, M.A. 1985. A morphological comparison of central Australian seedgrinding LPSOHPHQWVDQG$XVWUDOLDQ3OHLVWRFHQHDJHJULQGVWRQHV. The Beagle. Occasional Papers of the Northern Territory Museum of Arts and Sciences Smith, P.E.L. 1966. Le Solutrean en France3XEOLFDWLRQGHO·,QVWLWXWGH3UHKLVWRLUHGH O·8QLYHUVLWHGH%RUGHDX[0HPRLUH1R%RUGHDX[ Sollas, W.V. 1915 (1924). Ancient hunters and their modern representatives. New York: Macmillan. 6¡UHQVHQ0)OLQWKXJQLQJRJÁ NNHIUHPVWLOOLQJLGHQWLGOLJH Maglemosekultur. Kulturhistoriske Studier. Sydsjællands Museum: — in press a. Teknologiske traditioner i Maglemosekulturen. En diakron analyse DIPDJOHPRVHNXOWXUHQVÁ NNHLQGXVWUL,Q(ULNVHQ%9 HG Sidste nyt fra maglemose- kongemosefronten.cUKXV$DUKXV8QLYHUVLWHWVIRUODJ 448 Reference List — in press b. The Maglemosian Blade Concept. E. B. (ed.) Copenhagen: Petersen. Spence, M. W. 1981. Obsidian production and the state in Teotihuacan. American Antiquity — 1986. Locational analysis of craft specialization areas in Teotihuacan. In: Isaac, B. (ed). Economic Aspects of Prehispanic Highland Mexico. Research in Economic Anthropology, Supplement 2. Greenwich, Conn: JAI Press. — 1987. The scale and structure of obsidian production in Teotihuacan. In: McClung de Tapia E. and Rattray, E. C. (eds). Teotihuacan: Nuevos Datos, Nuevos Síntesis, Nuevos Problemas. UNAM, Mexico. Stanford, D. and F. Broilo. 1981. Frank´s Folsom campsite. The Artifact Stenbäck, N. 2003. Människorna vid havet. Platser och keramik på ålandsöarna perioden I.UStockholm Studies in Archaeology 28. Stockholm. Stenlund, S. 2003. Philosophy and critique of culture. Talk at the Conference Making a Difference: Re-considering Humanism and the Humanities September 2003. Ms. Storå, J. 1990. Stenredskap och bosättningsmönster. En morfologisk och funktionell analys av stenfynd från Geta, Åland0$WKHVLVLQ$UFKDHRORJ\cER ³D6DOWYLNcVJnUGD'RNXPHQWHULQJDYVNDGHJ|UHOVHRFK undersökning av kulturlager från stenåldern 1992. 8QSXEOLVKHGUHSRUWDWWKH Museum of Åland. Mariehamn. ³E6DOWYLNcVJnUGD)RVIDWNDUWHULQJRFKSURYXQGHUV|NQLQJ 1994. 8QSXEOLVKHGUHSRUWDWWKH0XVHXPRIcODQG. Mariehamn. — 1999. Provundersökning av lägenheten Bergmanstorp Rnr 1:15 I Överby. Fornminnesområde Jomala 14.1 Jettböle. 8QSXEOLVKHGUHSRUWDWWKH0XVHXPRI Åland. Mariehamn. — 2001b. Reading bones. Stone Age hunters and seals in the Baltic. Stockholm Studies in Archaeology 21. Stockholm. — and Stenbäck, N. 2001. Provundersökning av stenåldersboplats Jomala 14.1 Jettböle 1999. Jettböle I. 8QSXEOLVKHGUHSRUWDWWKH0XVHXPRIcODQGMariehamn. Stout, D. 2002. Skill and cognition in stone tool production. An ethnographic case study from Irian Jaya. Current Anthropolog y 43/5: 693–722. Stuart, E. J. 1923. A Land of Opportunities. London: Bodley Head. Sulgostowska, Z. 2002. Flint raw material economy during the Late Glacial and (DUO\3RVWJODFLDOLQWKH2GHU'DXJDYD3U\SHW%DVLQ,Q)LVKHU/((ULNVHQ B. V. (eds). Lithic Raw Material Economies in Late Glacial and Early Postglacial Europe. Oxford: BAR International Series. Sundström, L. 2003. Det hotade kollektivet. Coast to Coast Books 6. Uppsala: Department of Archaeology, Uppsala Universit — and Apel, J. 1998. An early Neolithic axe production and distribution system ZLWKLQDVHPLVHGHQWDU\IDUPLQJVRFLHW\LQ(DVWHUQ&HQWUDO6ZHGHQF 3500 BC. In: Holm, L. and Knutsson, K. (eds). Proceedings from the Third Flint $OWHUQDWLYHV&RQIHUHQFHDW8SSVDOD6ZHGHQ2FWREHU. Occasional Papers in Archaeology 16. Uppsala. 449 Skilled Production and Social Reproduction — and Apel, J. 2001. An early Neolithic axe production and distribution system ZLWKLQDVHPLVHGHQWDU\IDUPLQJFRPPXQLW\LQHDVWHUQFHQWUDO6ZHGHQG B.C. In: Knutsson, H. (ed.) Halvvägs. Stenålderssamhällen i förändring. Coast to &RDVW%RRNV8SSVDODDQG*|WHERUJ'HSDUWPHQWRI$UFKDHRORJ\ Uppsala University.. Svensson, M. 1993. Hindby offerkärr – en ovanlig och komplicerad fyndplats. Fynd 1993: 1. Tacon. P.S.C. 1991. The power of stone: Symbolic aspects of stone use and tool development in western Arnhem Land, Australia. $QWLTXLW\ 7DIÀQGHU-7KHVHOHFWLRQRIOLWKLFUDZPDWHULDOVTor — 1998. The allure of the exotic. the social use of non-local raw materials during the Stone Age in Sweden. Aun 25. Uppsala: Societas Archaeologica Upsaliensis. Temme, M. 1982. Excavaciones en el sitio precerámico de Cubilán (Ecuador). Miscelánea Antropológica Ecuatoriana 7KRPDV'++XQWHUJDWKHUHUDUFKDHRORJ\,Q0HOW]HU'-)RZOHU'' and Sabloff, J. A. (eds). American Archaeolog y, Past and Future. Washington, D.C.: Smithsonian Institution Press. Thomas, R. A. and Warren, N.H. 1970. A middle woodland cemetery in central Delaware: excavations at the Island Field site. Bulletin of the Archeological Society of Delaware 8. Thörn, H. 1997. Rörelser I det moderna: politik, modernitet och kollegialitet. Stockholm: Rabén Prisma. Tilley, C. 1993. $UWDUFKLWHFWXUHODQGVFDSH 1HROLWKLF6ZHGHQ In: Bender, B. (ed). Landscape: Politics and Perspectives. Oxford: Berg — 1994. A phenomenolog y of landscape. Oxford: Berg. Tindale, N.B. 1965. Stone implement making among the Nakako, Ngadadjara and Pitjandjara of the great Western Desert. Records of the South Australia Museum. $GHODLGH ³$XVWUDOLDQ$ERULJLQDOWHFKQLTXHVRISUHVVXUHÁDNLQJVWRQHLPSOHPHQWV some personal observations. In: Plew, M. G., Woods, J. C. and Pavesic, M. G. (eds). Stone tool analysis: essays in honor of Don E.Crabtree. Albuquerque: University of New Mexico Press. Tixier, J. 1963. Typologie de l´Epipaleolithique du Maghreb. A.M.G. mémoire 2. Centre de Recherches Anthropologiques, Préhistoriques et Ethnographiques, Alger. ³3URFpGpVG·DQDO\VHHWTXHVWLRQVGHWHUPLQRORJLHGDQVO·pWXGHGHV HQVHPEOHVLQGXVWULHOVGX3DOpROLWKLTXHUpFHQWHWGHO·(SLSDOpROLWKLTXHHQ $IULTXHGX1RUG2XHVW,Q%LVKRS::DQG&ODUN-' HGV Background to evolution in Africa. Chicago and London: The University of Chicago Press. — 1972. Personal communication. — 1984. Le débitage par pression. In: Tixier, J. (ed). 3UpKLVWRLUHGHOD3LHUUH7DLOOpH Economie du débitage laminaire: technologie et expérimentation. In: Tixier, J. (ed). IIIe WDEOHURQGHGHWHFKQRORJLHOLWKLTXH0HXGRQ%HOOHYXHRFW3DULV &HUFOHGH5HFKHUFKHVHWG·eWXGHV3UpKLVWRULTXHV ³,QL]DQ0/DQG5RFKH+Préhistoire de la Pierre Taillée I: terminologie et technologie.3DULV&HUFOHGH5HFKHUFKHVHWG·eWXGHV3UpKLVWRULTXHV 450 Reference List Tommesen, T., 1996. The Early Settlment of Northern Norway. In Larsson, L. (ed.) The Earliest Settlement of Scandinavia and its relationship with QHLJKERXULQJDUHDV$FWDDUFKDHRORJLFD/XQGHQVLD6HULHVLQR1R Torrence, R. 1986 Production and exchange of stone tools. Cambridge: Cambridge University Press. ³+XQWHUJDWKHUHUWHFKQRORJ\PDFURDQGPLFURVFDOHDSSURDFKHV,Q 3DQWHU%ULFN&5/D\WRQ+DQG5RZOH\&RQZ\3 HGV Hunter-Gatherers. An Interdisciplinary Perspective. Cambridge: Cambridge University Press. Torvinen, M. 1979: Liedon Kukkarkosken kivikautinen kalmisto. Suomen Museo 1978: 37–80. Tosi, M. 1984. The notion of craft specialization and its representations in the archaeological record of early states in the Turanian Basin. In: Spriggs, M. (ed). Marxist perspectives in archaeolog y.. Cambridge: Cambridge University Press. Toth, N. 1985. The Oldowan reassessed: A close look at early stone artifacts. Journal of Archaeological Science — 1987. Behavioral inferences from Early Stone Age artifact assemblages: An experimental model. Journal of Human Evolution — and Schick, K. D. 1993. Early stone industries and inferences regarding language and cognition. In: Gibson, K. R. and Ingold, T. Tools, Language and Cognition in Human Evolution. Cambridge: Cambridge University Press.. ³6FKLFN.'6DYDJH5XPEDXJK(66HYFLN5$DQG5XPEDXJK'0 3DQWKHWRROPDNHU,QYHVWLJDWLRQVLQWRWKHVWRQHWRROPDNLQJDQGWRRO using capabilities of a bonobo (Pan paniscus). Journal of Archaeological Science 20: 7UHKHUQH37KHZDUULRU·VEHDXW\7KHPDVFXOLQHERG\DQGVHOILGHQWLW\LQ %URQ]H$JH(XURSHJournal of European Archaeolog y Truffreau, A. 1995. The variability of Levallois technology in northern France DQGQHLJKERULQJDUHDV,Q'LEEOH+/DQG%DU<RVHI2 HGV 7KH'HÀQLWLRQ and Interpretation of Levallois Technolog y, Monographs in World Archaeology 23. Madison: Prehistory Press. Turnbull, C. M. 1972. 1987. The Mountain People. New York: Simon and Schuster Van Dyke, R. M. and Alcock, S. E. (eds.) 2000. Archaeologies of memory. Oxford: Blackwell. Vandkilde, H., 1996. From Stone to Bronze. The Metalwork of the Late Neolithic and Earliest Bronze Age in Denmark. Jutland Archaeological Society Publications XXXII. Aarhus. — 2000. Material culture and Scandinavian archaeology: a review of the concepts of form, function and context. In: Olausson, D. and Vandkilde, H. 2000 (eds). Form, function and context. Material culture studies in Scandinavian archaeology. Acta Archaeologica Lundensia. Series in 8o 31. Lund. Vang Petersen, P. 1993, 1999. Flint fra Danmarks oldtid. Köpenhamn: Høst and Søn; Nationalmuseet. — and Johansen, L. 1995. Tracking Late Glacial reindeer hunters in Eastern Denmark. In: Larsson, L. (ed). The Earliest Settlement of Scandinavia. Acta Archaeologica Lundesia, Series in 8o 24. Lund. 451 Skilled Production and Social Reproduction Vaquer, J. 1990. Le Néolithique en Languedoc occidental. Paris: CNRS. 9HUQDQW-3'HWSRVLWLYDWlQNDQGHWVXSSNRPVWLGHWDUNDLVND*UHNODQGRes Publica Videiko, M. Y. 1994. Tripolye “pastoral” contacts. Facts and character of the LQWHUDFWLRQV%&,QBalt-Pontic Studies Vignard, E. 1934. Triangles et trapèzes du Capsien en connexion avec leurs microburins. Bulletin de la Société de Préhistoire Française Waaraas, T. A. 2001. 9HVWODQGHWLWLOHJ%RUHDOWLG MA Thesis in Archaeology. Universitetet i Bergen. Bergen. Wace, A. J. B. and Thompson, M. S. 1912. Prehistoric Thessaly; being some DFFRXQWRIUHFHQWH[FDYDWLRQVDQGH[SORUDWLRQVLQ1RUWK(DVWHUQ*UHHFHIURP Lake Kopias to the borders of Macedonia. Cambridge, University Press. Waldorf, D. C. 1993. 7KHDUWRIÁLQWNQDSSLQJ (Revised Edition). Branson: Mound Builder Books. :DOODHUW3rWUH+/HDUQLQJKRZWRPDNHWKHULJKWSRWVDSSUHQWLFHVKLS strategies and material culture, a case study in handmade pottery from Cameroon. Journal of Anthropological Research 57/4: 471–493. Warren, P. and Hankey, V. 1989. Aegean Bronze Age Chronolog y. Bristol: Bristol Classical Press. :ĊV07HFKQRORJLDNU]HPLHQLDUVWZDNXOWXU\MDQLVâDZLFNLHM. Doctoral Dissertation LQ$UFKDHRORJ\áyGŭ Washburn, D. K. 1989. The property of symmetry and the concept of ethnic style. In: Shennan, S. J. (ed.) Archaeological Approaches to Cultural Identity. London: Routledge. Weiler, E. 1994. Innovationsmiljöer i bronsålderns samhälle och idévärld. Kring ny teknologi RFKEHJUDYQLQJVULWXDOL9lVWHUJ|WODQG. Studia Archaeologica Universitatis Umensis 5. Umeå. — 1997. Kontinuitet och förändring i senneolitikum. Västsverige. In: Larsson, M. and Olsson, E. (eds). Regionalt och interregionalt. Stenåldersundersökningar I Syd- och Mellansverige. Riksantikvarieämbete UV. Stockholm. Weiner, A. B. 1992. Inalienable possessions. The paradox of keeping-while-giving. Berkeley: University of California Press. Welinder, S. 1971. Överåda. A Pitted Ware culture site in eastern Sweden. 0HGHODQGHQIUnQ/XQGV8QLYHUVLWHWV+LVWRULVND0XVHXP/XQG Welinder, S. 1992. Människor och landskap. Aun 15. Uppsala: Societas Archaeologica Upsaliensis. Wengler 1995. Levallois Technology in the Middle Paleolithic of Eastern Morocco. ,Q'LEEOH+/DQG2%DU<RVHI HGV 7KH'HÀQLWLRQDQG,QWHUSUHWDWLRQRI Levallois Technolog y, Monographs in World Archaeology 23. Madison: Prehistory Press. White, H. 1987. The content of the form: narrative discourse and historical representation. Baltimore: John Hopkins University Press. :KLWH-3DQG7KRPDV'+:KDWPHDQWKHVHVWRQHV"(WKQRWD[RQRPLF models and archaeological interpretations in the New Guinea Highlands. Clarke, D. L. (ed.) Models in Archaeolog y. London. 452 Reference List Whittaker, J. C. 1994. Flintknapping: Making and understanding stone tools. Austin: Texas University Press. Widmer, R. J. 1991. Lapidary craft specialization at Teotihuacan: implications for community structure at 33:S3W1 and economic organization in the city. Ancient Mesoamerica Widmer, R. J. 1996. Procurement, exchange, and production of foreign commodities at Teotihuacan: State monopoly or local control? In: Mastache, A. G., Parsons, R. S., Santley, R. S. and Serra Puche, M. C. (eds). Arqueología Mesoamericana: Homenaje a William Sanders, vol 1. Mexico: Instituto Nacional de Antropología e Historia. :LOOH\*7KH$UWLIDFWVRI$OWDUGH6DFULÀFLRV. Papers of the Peabody Museum, Harvard 64/1, Cambridge, Mass. :LOOURWK.+1RJOHEHWUDJWQLQJHURYHUGHUHJLRQDOHIRUKROGL6OHVYLJRJ Holsten i bronzealderens periode II. In: Poulsen, J. (Ed. ), Regionale forhold i QRUGLVN%URQ]HDOGHU Jysk Arkeologisk Selskabs skrifter XXIV. Höjberg. :LOPVHQ(1DQG5REHUWV)+-U/LQGHQPHLHU&RQFOXGLQJ Report on Investigations. Smithsonian Contributions to Anthropolog y 24, Washington D.C. :RRGPDQ37KH.RPVD&XOWXUH$UHH[DPLQDWLRQRILWVSRVLWLRQLQWKH Stone Age of Finnmark. Acta Archaeologica Wylie, A. 1985. The reaction against analogy. In: Shiffer, M. (ed). Advances in Archaeological Method and Theory 8. New York: Academic Press. Wynn, T. 1979. The intelligence of later Acheulian hominids. Man — 1981. The intelligence of Oldowan hominids. Journal of Human Evolution 10: :\V]RPLUVND%)LJXUSODVWLNRFKJUDYVNLFNKRVQRUGRFK|VWHXURSDV neolitiska fångstkulturer. Acta Archaeologica Lundensia. Series in 40 18. Lund. Yalçinkaya, I. 1995. Thoughts on Levallois Technique in Anatolia. In: Dibble, H. /DQG%DU<RVHI2 HGV 7KH'HÀQLWLRQDQG,QWHUSUHWDWLRQRI/HYDOORLV7HFKQRORJ \ Monographs in World Archaeology 23. Madison: Prehistory Press.: Young, D. E. and Bonnischen, R. 1984. 8QGHUVWDQGLQJ6WRQH7RROVDFRJQLWLYHDSSURDFK. Maine: Centre for the study of Early Man. Yven, E. 2004. Approche spatiale et territoriale des industries lithiques. &RQVWDQWHVHWYDULDQWHVGDQVO·RFFXSDWLRQGXVXEVWUDWJHRJUDSKLTXHHWOD gestion des matieres premieres lithiques au Mesolithique en Bretagne. 729 p. These de soutenance, Université de Brest (France), Brest. =DNRœFLHOQD$.U]HPLHQLDUVWZR.XOWXU\:Râ\ľVNR²/XEHOVNLHM&HUDPLNL Malowanej. Lublin. =LOKDR-$RÀFLQDGHWDOKHQHRFDOFROLWLFDGH&DVDVGH%DL[R &D[DULDV9LOD Nova de Ourém). Trabalhos de Arqueologia da EAM, 2. Lisboa: Colibri. 453 Skilled production and Social Reproduction Aspects of Traditional Stone-Tool Technologies During a five-day symposium in late August 2003, a group of archaeologists, ethno-archaeologists and flint knappers met in Uppsala to discuss skill in relation to traditional stone-tool technologies and social reproduction. This volume contains 20 of the papers presented at the symposium and the topics ranges from Oldowan stone tool technologies of the Lower Palaeolithic to the production of flint tools during the Bronze Age. The symposium was arranged by Societas Archaeologica Upsaliensis and the department of Archaeology at Uppsala University. ISBN 91-973740-6-7 ISSN 1404-8493 Published and distributed by: Societas Archaeologica Upsaliensis Villavägen 6G SE-752 36 Uppsala Sweden www.sau se

Log In

Preliminary experimental observations on a particular class of bifacial lithic artifact from Misiones province, northeastern Argentina.

Preliminary experimental observations on a particular class of bifacial lithic artifact from Misiones province, northeastern Argentina.

Related Papers

RELATED PAPERS