Technology and Tradition in the Eastern Arctic, 2500 BC-AD 1200: A Dynamic Technological Investigation of LithicAssemblages from the Palaeo-Eskimo Traditions of Greenland

by Mikkel S0rensen. Museum Tusculanum Press, University ofCopenhagen, Copenhagen, DK. 418 pp. ISBN 978-87-635-3167-2(hardcover) US$89.00. 2012.

FOR DECADES, THE SOPHISTICATEDStone tools found among Palaeo-

Eskimo sites spanning the North Ameri-can Arcdc and Greenland have inspiredthe imaginations of archaeologists.Through detailed typological analyses,these tools have been used to establishregional culture-histories to understandhow and when these original peopleslived in the far North. These pioneeringstudies laid the foundation upon whichmore recent analyses, like this one, arebased. Mikkel S0rensen cites two sourcesof inspiration for this work, which pre-sents the findings of his 2006 Ph.D. dis-sertation, completed at the Universityof Copenhagen's Department of Prehis-toric Archaeology. The first is S0rensen'sown fascination with the tiny lithicartifacts found among Palaeo-Eskimosites in Greenland. The second is anemergent "reordering" (p. 18) of Palaeo-Eskimo culture-history based tipon newinvestigations of previously studied sitesand extant artifact collections. The bookis published by the University of Copen-hagen's Museum Tusculanum Press,which prints dozens of works each yearthat are written by university affiliates,like S0rensen. The book is lengthy at 418pages, and includes seven chapters andone appendix.

Chapter 1 presents the muld-layeredgoals of the study. First, S0rensen aimsto focus on the acdons of individuals tofacilitate the description of a new and"dynamic" view of Arctic prehistory. Heproposes a "French formulated meth-odology," known as chaîne opératoire(CO) as the most effective way of achiev-ing this. Second, S0rensen endeavorsto construct precise definitions for thefunctional tool types that comprisePalaeo-Eskimo toolkits. This, he argues,will permit Arctic archaeologists to inter-pret what kinds of lithic producdon anduse activities occurred within settlementsites. The final goal is to achieve a newinterpretation among the five identifiedPalaeo-Eskimo cultures in Greenland,and their North American and Siberianrelatives.

Chapter 1 continues with a briefoverview of previous work in Greenland.S0rensen states that more recent studies,like his, are challenging existing inter-pretations, which are described as beingshaped more by the historical developnment of research than actual events inprehistory. Previous work in the Cana-dian Arctic, especially that of MoreauMaxwell, is more harshly criticized aswidely inaccurate since, according toS0rensen, it was based on mixed assem-

CAN ADÍAN JOURNAL OF ARCHAEOLOGY/JOURNAL CANADIEN D'ARCHÉOLOGIE 37: 349-353 (2013)

350 • MILNE, S.B.

blages from multi-component sites thatyielded questionable radiocarbon dates.

After outlining his two researchproblems (one methodological; oneculture-historical), S0rensen describesbriefly, and selectively, previous Arcticlithic studies he considers technological.While S0rensen does not define whatconstitutes a technological analysis, itbecomes apparent that he is largelyreferring to those that use, or at leastmention however cursorily, the Frenchmethodology or CO. This position arti-ficially sets up the study as being amongthe first to view Palaeo-Eskimo lithictechnology from a dynamic technologi-cal perspective. What S0rensen fails toacknowledge is that there is indeed abody of research that has examinedPalaeo-Eskimo lithic technology usingsuch a perspective. S0rensen's oversightto mention these studies is argvtably tiedto a larger debate in the field of lithicanalysis that focuses on the epistemologi-cal differences between CO and what areknown as reduction sequence models(RSM). At the heart of the debate iswhether CO is fundamentally differentanalytically than RSM.

Chapter 2 presents a thinly researchedoverview of this debate that negativelygeneralizes North American lithic stud-ies, which commonly use RSM, in such away that he can discount them, unjustly,as inferior and methodologically anti-quated. CO and RSM are essentiallyreduction sequence models, which is,ironically, well illustrated by S0rensen'sFigure 2.1 on p. 31. However, one of themain points of divergence between COand RSM is the belief that those who usea GO approach can access prehistoriccognition, which is one of S0rensen'sgoals in this book.

S0rensen goes on to describe the his-torical context for the CO approach and

CANADIAN JOURNAL OF ARCHAEOLOGY 37 (2013)

acknowledges some of the ambiguitiesassociated with its use. He clarifies hisuse of the CO term in that it "describesthe physical reduction sequences thatall together consist of many artifacts"(p. 31 ). He continues stressing that thesereduction sequences are carried out onthe basis of what he calls the productionconcept. It is the production concept, ormental template, that is handed downgenerationally through one's culture.Based on this information, S0rensenargues that a "technological tradition" asit is identified archaeologically equatesto a people or culture (p. 35-37). Thetoolkits used by a culture are describedas being standardized and specialized,which necessitate a particular produc-tion process to create and maintain.The repeated creation and maintenanceof this standardized toolkit requiresknowledge, know-how, and apprentic-ing. What S0rensen is trying to establishis that people learn how to make stonetools through apprenticeship with thosein one's social unit who have the knowl-edge of the craft. Because novices learnfrom experts who possess the mentaltemplates for how to make tools a certainway, CO analysis can determine to whatdegree artifact assemblages are relatedor not given observed similarities anddifferences in the attributes they dis-play. The tightness of the similarities, intheory, should be indicative of sharedmental templates or production con-cepts and hence cultural affinity suchthat discrete social groups or culturescan be identified and their degree ofrelatedness or interaction assessed.

S0rensen goes on to explain thatwhen a contemporary knapper, like him-self, learns how to make the stone tooltypes used by a people in the past, theycan gain a hermeneutic understandingof that culture through the establish-

TECHNOLOGY AND TRADITION IN THE EASTERN ARCTIC • 351

ment of "overlapping horizons" (p. 39).As such, studying the results of replica-tion experiments affords the same tech-nical insights as studying the remains ofarchaeologically derived materials. Thecontemporary knapper can, arguably,get inside the head of the prehistoricknapper by repeating the same reduc-tion sequences, thus gaining insight intoprehistoric cognition.

S0rensen's methodology comprisesfour approaches. The first is lithic refit-ting of formal tools. The second is a tech-nological dynamic classification, whichinvolves classifying lithic material asderiving from the production of blades,bifaces, cores, etc. However, one mustconceive of the reduction from start tofinish so that if a core is being made,it begins with the original raw materialand then the steps of the sequence areenvisioned until one reaches the finalfinished product. The third approach iscontemporary lithic replication (whichprovides the knowledge to apply thesecond approach). Last, mental refit-ting is carried out in conjunction withreplication experiments in the sense thatone studies the negative flake scars ontools to determine the how flakes weredetached and in what order.

Chapter 2 concludes with a descrip-tion of S0rensen's analytical procedure,which includes the analysis and compari-son of two sites for each of the identifiedPalaeo-Eskimo cultures that inhabitedGreenland.

Chapter 3 describes the formation oflithic raw materials found throughoutGreenland and their respective macro-scopic properties. The photographs foreach raw material are excellent and pro-vide a strong visual sense of the diversityof available toolstones.

Chapter 4 is a daunting presentationspanning 226 pages of incredibly dense

descriptions of sites and site histories,lithic raw materials, artifact assemblages,and CO sequences for specific tooltypes. S0rensen's artifact drawings areprolific and well laid out in terms of thedifferent phases of tool production hehas identified. However, representativephotographs of the artifact assemblagesand individual artifacts are not includedmeaning the reader cannot evaluate forthemselves the actual artifacts in ques-tion; instead one must rely on Sorensen'sperception of their idealized forms. Thedata presentation is wanting in severalareas. For example, axes for frequencydistributions are not always labeled andnone of them include absolute numbersfor the variable data summarized. Thisis especially frustrating when examiningthe distributions for lithic debitage (seefor example Figure 4.1.12). The tablesare also lacking artifact totals for eachartifact type.

While formal artifact types are dis-cussed in detail, the lithic debitageanalyses, when included, are cursory andclearly of secondary importance. Precisedefinitions for flake attributes examinedare not provided so it is challenging tounderstand the meaning of what appearin some instances to be idiosyncraticterms (e.g., the "butt" of a flake; p. 122).Moreover, on page 328 S0rensen essen-tially discounts the analytical value oflithic debitage since his version of atechnological analysis, which privilegesformal tool types, is "qualitative" and assuch, examining things like the relativeproportions of debitage to tools, which isquantitative, is insignificant.

Chapter 5 begins with a discussionof raw material procurement patternsthat S0rensen has identified amongthe sites included in the study for eachPalaeo-Eskimo culture. The reductionmethod for each Palaeo-Eskimo culture

JOURNAL CANADIEN D'ARCHÉOLOGIE 37 (2013)

352 • MILNE, S.B.

is then presented. S0rensen advocatesthe importance of burins as sensitivechronological markers in Palaeo-Eskimoculture-history, which sharply contrastswith other studies that identify thesetools as multi-functional, unstandard-ized, and thus unsuitable for typologicalclassification (e.g. Barton et al., J FieldArch 23(1)). The chapter concludes withan evaluation of the methodology used,which S0rensen claims is dynamic andnon-typological. He believes ardentlythat all artifact types created by Palaeo-Eskimo peoples were precisely worked,making them amenable to morphologi-cal and technological classification, nomatter the stage at which they are foundin the archaeological record.

Chapter 6 presents S0rensen's inter-pretations of the CO he's identifiedamong the Greenlandic Palaeo-Eskimocultures. Surprisingly, S0rensen heavilyweights the presence/absence of artifacttreatments like edge serration, grinding,and notching as culturally sensitive indi-cators. These attributes are at the coreof the existing typologies that S0rensenheavily criticizes as static and of limitedutility throughout the study.

The "reordering" of Palaeo-Eskimoculture-history achieved through thisstudy includes the revelation thatGreenland was circum-populated twiceby Saqqaq and Independence I (whoare viewed as contemporaneous enti-ties), and by S0rensen's newly identifiedGreenlandic Dorset. The Thule regionis described as more culturally complexthan previously thought in that it has beenoccupied, albeit sporadically, by as manyas six Palaeo-Eskimo cultures, which nowinclude "Canadian Pre-Dorset" and "EarlyCanadian Dorset." Lastly, the Saqqaq,Independence I, and Greenlandic Dorsetperiods are described as remarkablystable and unchanged given the intense

CANADIANJOURNAL OF ARCHAEOLOGY 37 (2013)

technological conservatism exhibited bytheir respective lithic toolkits.

Chapter 7 presents S0rensen's evalu-ation of his methods and his speculationof how CO can be applied in futureArctic studies. It becomes clear to thereader that the net goal of the study wasto establish a series of idealized artifacttypologies, which include preciselydefined phases in the overall reductionsequence for each tool type. Archae-ologists are meant to compare finishedformal tools, and even tool blanks andpreforms, to these typologies so as toidentify the Palaeo-Eskimo culture thatmade them. Production concepts andmethods of reduction equate to culturaltraditions; therefore, when one findsstone tools, even those that are unfin-ished, that fit a specific phase of the CO,one can attribute them to that cultureand assign them a relative date.

What is so difficult to reconcile afterreading this book is that S0rensen tearsdown existing typologies as inadequateonly to advocate replacing them withones that are more rigid and idiosyn-cratic. In fact, the CO that S0rensenpresents stifle the potential to identifymeaningful patterns of lithic artifactvariability in Palaeo-Eskimo assemblagesbecause they assert that each person ineach culture always followed the sameproduction concept unfailingly, almostrobotically. This central tenet of CO(i.e., production concept) has been criti-cized as "flawed" (Andrefsky, / ArcAaeo/Res 17:68) and "unpersuasive" (Shott,Lithic Technology 28 (2): 100) because itleaves no room to identify how peoplein the past adapted their lithic technolo-gies to deal with situational contingen-cies in their everyday lives. One wouldthink that living in the Arctic wouldpresent a myriad of challenges relatingto raw material availability, subsistence

TECHNOLOGYAND TRADITION IN THE EASTERN ARCTIC • 353

resources, climate, social organization,among others; therefore, it seems highlyunlikely that Palaeo-Eskimo peoplesdid not adapt their lithic technologyin order to deal with them. People arepeople and human behaviour is not thatproscriptive.

In a thorough review of CO, Bar-Yosef and Van Peer ( Current Anthropology50(1)) demonstrated in a replicativeexperiment that there is more than oneway to produce the same ardfact type; inother words, equifinality can yield simi-lar results in the archaeological recordno matter how well defined a CO maybe. While the replication experimentsthat S0rensen has carried out are impres-sive, a major caveat noted with such stud-ies and the CO defined from them is thatthe idealized prodtiction concept cre-ated by the contemporary knapper "risksbeing a construction in [their] mind"and in all likelihood was never appliedby toolmakers belonging to the culturaltradition it is meant to exemplify (Bar-Yosef and Van Peer, Current Anthropology50(l):108; see also Andrefsky,/ArcAa^o/Res 17:68). The a priori statements thatS0rensen makes based on his replica-tive studies as reflecdng realides among

Palaeo-Eskimo toolmakers are in dangerof being just that, particularly when herejects the use of quantitative analysesin favour of a subjective and untestablequalitative approach.

This book will no doubt be of valueto scholars who are interested in CO.It is attractively illustrated, includingthe appendix, and presents detailedsite maps. The book may be of inter-est to other specialists in lithic analysisbut it will arguably be of limited utilityto those studying Palaeo-Eskimo lithicsthat do not follow the core tenets of CO.While some archaeologists are toutingCO as the best methodological choicefor future Palaeo-Eskimo lithic analyses,it behooves Arctic researchers to criti-cally assess the purported contributionsit can make versus those that it reallycan. Replacing old artifact typologieswith new ones will not do much to moveour interpretadons of these populationsforward.

S. Brooke MilneDepartment of Anthropology andCenter for Earth Observation Science(CEOS), University of ManitobaWinnipeg, MB

JOURNAL CANADIEN D'ARCHÉOLOGIE 37 (2013)

Copyright of Canadian Journal of Archaeology is the property of Canadian ArchaeologicalAssociation and its content may not be copied or emailed to multiple sites or posted to alistserv without the copyright holder's express written permission. However, users may print,download, or email articles for individual use.