‘‘An Ape’s View of the Oldowan’’ RevisitedTHOMAS WYNN, R. ADRIANA HERNANDEZ-AGUILAR, LINDA F. MARCHANT, AND WILLIAM C. MCGREW

In 1989, Wynn and McGrew published an explicit comparison betweenOldowan technology and what was then known of chimpanzee technology.1

They compared the range and variety of tools, adaptive role of tools, carryingdistances, spatial cognition, manufacturing procedures, and modes of learning.They concluded that everything archeologists had reconstructed about thebehavior of Oldowan hominins could be accommodated within the ape adaptivegrade; that is, a paraphyletic group united by overall similarities in anatomy and,in this case, behavior. The only Oldowan activities that were almost unknown formodern apes were the long-distance transport of objects and direct competitionwith carnivores, which was implied by meat acquisition activities. ‘‘In its generalfeatures Oldowan culture was ape, not human. Nowhere in this picture needwe posit elements such as language, extensive sharing, division of labor, orpair-bonded families, all of which are part of the baggage carried by the termhuman.’’1:394

Much has changed since the origi-nal article was conceived in the1980s. Primatological research hasexpanded the variety of tool-basedactivities documented for nonhumanprimates. The range of tool-usingspecies has also expanded to include

orangutans, capuchins, and maca-ques. Archeological research hasextended the known geographic andtemporal distribution of Oldowansites, and provided greater detail tothe reconstructions of Oldowanactivities and procedures. The theo-

retical climate in paleoanthropologyhas shifted as well, with greater

attention being paid to concepts

drawn from ecology, demography,

primatology, and even cognitive sci-

ence. But most importantly, few

paleoanthropologists would now

attribute language, extensive sharing,

division of labor, or pair-bonded

families to the hominins responsible

for the Oldowan; in this regard, the

1989 argument has become effec-

tively moot. Yet a fundamental per-

spective remains unchanged. Paleo-

anthropologists continue to celebrate

the advent of flaked stone technology

as the crossing of an evolutionary

Rubicon. ‘‘These materials herald a

new chapter in the human evolution-

ary record and mark a significant

departure from the rest of the primate

world: the onset of a technology-

based adaptation in which synthetic

tools supplemented the biological

repertoire of these creatures.’’2:3

The idea that the advent of flakedstone tool technology marked a qual-itative leap in hominin evolution iswidely held in paleoanthropology.Was the Oldowan a new chapter inhuman evolution and the startingpoint for greater things to come? Orwould it be more fruitful if the Oldo-wan were understood as the continu-ation of a way of life that had beenin place for millions of years? Takingan ape’s perspective on the Oldowanis not just a clever experimentdesigned to clarify and highlight thesimilarities and differences betweenapes and humans; it is a call to placeearly hominin tool users in theirappropriate evolutionary context.Lithic technology may eventuallyhave provided the selective contextfor more human-like developmentsamong some hominins, but these

ARTICLE

Thomas Wynn is Professor of Anthropology at the University of Colorado Colorado Springs.He has published extensively in Palaeolithic archaeology, with an emphasis on cognitive evo-lution. His books include The Evolution of Spatial Competence (Illinois 1989), The Rise ofHomo sapiens: The Evolution of Modern Thinking (with F. Coolidge, Wiley-Blackwell 2009),and How to Think Like a Neandertal (with F. Coolidge, Oxford, forthcoming). E-mail: twyn-n@uccs.eduR. Adriana Hernandez-Aguilar is a researcher at the Centre for Ecological and EvolutionarySynthesis, Department of Biology, University of Oslo, Norway. She studies the behaviouralecology of chimpanzees in a dry habitat, Ugalla, in Tanzania. Her articles include ‘‘Savannachimpanzees use tools to harvest the underground storage organs of plants’’ PNAS 2007(with J. Moore and T.R. Pickering). E-mail: r.a.hernandez-aguilar@bio.uio.noLinda F. Marchant is Professor and Chair of the Department of Anthropology at MiamiUniversity. She has published on manual laterality and culture in primates, and studied wildchimpanzees and bonobos in Democratic Republic of Congo, Senegal, Tanzania, andUganda. Her co-edited volumes include Great Ape Societies (Cambridge UP, 1996) andBehavioural Diversity in Chimpanzees and Bonobos (Cambridge UP, 2002). E-mail: marchalf@muohio.eduWilliam C. McGrew is Professor of Evolutionary Primatology at the University of Cambridge.He has studied the ethology and ecology of wild African primates, especially chimpanzees,from Tanzania to Senegal. His books include The Cultured Chimpanzee: Reflections onCultural Primatology (Cambridge University Press, 2004) and Chimpanzee Behavior in theWild (Springer, 2010). E-mail: wcm21@cam.ac.uk

Key words: technology; material culture; adaptive grade; chimpanzee; early hominins

VVC 2011 Wiley-Liss, Inc.DOI 10.1002/evan.20323Published online in Wiley Online Library (wileyonlinelibrary.com).

Evolutionary Anthropology 20:181–197 (2011)

developments were nowhere appa-rent in its earliest hominin manifes-tation. Indeed, as we hope to showhere, Schick and Toth’s2:3 felicitouscharacterization of the Oldowan as‘‘the onset of a technology-based ad-aptation in which synthetic toolssupplemented the biological reper-toire of these creatures’’ is accurateprecisely because it is a characteriza-tion of the ape adaptive grade.

METHODOLOGICALCONSIDERATIONS

How does one make a meaningfulcomparison of such different datasets? Ethological descriptions of non-human primate tool use and tool mak-ing, supplemented by primate archeol-ogy, appear to be a richer source ofdata than are the few thousand piecesof stone and bone offered up by theentire Oldowan archeological record.How, without appearing paltry, canthe static data of archeology bematched against the dynamic recordof living, active, tool-using primates?The prospects are not as dire as theymight seem. The task of evolutionaryscience is to take descriptive data,apply appropriate theoreticallygrounded methods, and draw infer-ences. Despite having largely differentdata sets, both primatology and paleo-anthropology draw inferences aboutthe adaptive significance of tool-basedactivities. It is these inferences thatcan be compared. Archeology has alarge battery of techniques by which itcan reconstruct many of the activitiesresponsible for archeological remains.Yes, much of what ethologists

observe is inaccessible from thearcheological record of the deep past(for example, the sex of tool users).Moreover, due to lack of preserva-tion, it is very possible, even likely,that there were hominin tool-basedactivities that we fail to detect. How-ever, if everything we do detect issimilar to what we know of livingapes, we have no grounds for assum-ing that undetected behavior wassomehow more human-like.

NEW PRIMATE EVIDENCE

Over the last 20 years, field studiesof wild chimpanzees have continued

to expand, as older sites (Bossou,3,4

Gombe,5–8 Mahale,7,8 Taı̈9,10) addedmore decades or were revitalized(Budongo11); middle-aged sitesemerged to join the ranks of thosewith habituated subjects (for exam-ple, Kalinzu,12 Kanyawara,13 andNgogo,14,15); and new sites addedluster to the range (among them Fon-goli,16,17 Gashaka,18 Goualougo,19–22

and Semliki23). The range of ethnog-raphy now available20,24 was unima-ginable 20 years ago. Comparativestudies25,26 have made the most ofthis rich tapestry of diversity.

On the other hand, established

bonobo field sites remain few, with

only Lomako27 and Wamba28 carry-

ing on. The bright spark comes from

the first fully habituated study popu-

lation at Lui Kotale, which has

yielded surprises, such as social

hunting of monkeys.29 At Mondika,30

habituation of a wild population of

western lowland gorillas finally has

been achieved, but the material cul-

ture of the species remains virtually

nonexistent. Orangutans are now

firm members of the elementary

technology and culture club,31

thanks to breakthrough research,especially at Suaq Balimbing,32–34

but their tool repertoires still looksparse in comparison with those ofchimpanzees.Similarly, in captive populations of

great apes, chimpanzees lead theway, especially via the elegantlydesigned experiments of Whiten andde Waal and their students.35 Usingmultiple groups at various laborato-ries, they have established the spe-cies’ ability to learn socially by imita-tion. More comparative work byTomasello and Call’s36 group at Leip-zig has tackled various topics incomplex cognition, from mental timetravel to cooperation. More recently,Hare and his colleagues37 haveshown that bonobos in an Africansanctuary (Lola ya Bonobo) yieldsimilarly complex capacities thatseem to be species-unique.

FIELD STUDIES

Field study of free-ranging apes innature is still the main source of newinformation about tool-using activ-ities. Typically, this comes fromlong-term, continuous research onhabituated subjects followed over thecourse of their lifetimes. For chim-panzees, there are now nine estab-lished field sites where subjects canbe followed daily from nest to nest,in ecotypes ranging from rain forestto savannah. In addition, severalnewer sites where the apes are stillunhabituated have also reported newtypes of tool use.

Chimpanzees

Tool-assisted hunting

Pruetz and Bertolani38 reported thefirst habitual use of tools by chim-panzees to secure vertebrate prey.The apes of Fongoli, Senegal, usedsharpened sticks as skewers to huntlesser bushbabies (Galago senegalensis)from their nests in tree cavities. Mak-ing these instruments involved a hier-archically organized sequence of up tofive steps, flexible enough to allowsome steps to be omitted and othersrepeated. Some of the observed epi-sodes involved more than one toolbeing used to probe a single cavity.

The task of evolutionaryscience is to takedescriptive data, applyappropriate theoreticallygrounded methods, anddraw inferences. Despitehaving different datasets, both primatologyand paleoanthropologydraw inferences aboutthe adaptivesignificance of tool-based activities. It isthese inferences thatcan be compared.

182 Wynn et al. ARTICLE

Two individuals manufactured theirtools before ascending the tree. Therate of success was low; in only one ofthe 22 observed cases was the preyseen to be captured. This is similar tothe solitary hunting of monkeys atother chimpanzee sites. Interestingly,adult females or immatures did mostsuch tool hunting, in contrast to chim-panzee hunting of bigger prey else-where, done mainly by adult maleswithout tools.

Digging tools

Hernandez-Aguilar, Moore, and Pick-ering39 reported for the first time thatchimpanzees use tools to excavate theunderground storage organs (USOs)of plants (Fig. 1). The chimpanzees ofIssa, in Ugalla, Tanzania, were unha-bituated, and indirect evidence wasused to infer this novel behavior:holes, spat-out fibrous wadges,knuckle prints, feces, and associatedtools. The apes reused two of the 11sites that were identified. Three typesof tools were recovered: sticks, frag-ments from a fallen log, and pieces oftree bark. Microwear on the tools con-firmed their use for digging. Else-where, other exploitation of USOs isdone without technology.40,41

Cleaving tools

Koops, McGrew, and Matsuzawa42

found indirect evidence that unhabi-tuated chimpanzees at Nimba,Republic of Guinea, West Africa, usetools to reduce Treculia africanafruits into easier-to-handle pieces.Fruits of this species are spherical,large, dense, and heavy (up to 8.5 kg)and thus difficult to bite into. Twotypes of tool were used: movablestone and wooden ‘‘cleavers,’’ andstone outcrop (nonmovable) anvils.Such cleavers had not been reportedbefore. The chimpanzees did not usethe two types of tools in combination,as in nut-cracking, but rather usedthem independently. This is the firstreport of the use of two differenttypes of percussion tools to executethe same task and achieve the samegoal. The Treculia fruits were carriedup to 55 m from the tree of origin.

Tool sets, composite tools,

and compound tools

A tool set is ‘‘two or more tools[used] in an obligate sequence toachieve a single goal.’’43,44 Brewerand McGrew45 first described a toolset. In that case, a female chimpan-

zee used four types of tools in seriesto extract honey. Since then, tool setshave been seen in wild communitiesfor obtaining honey, sap, and socialinsects (Fig. 2).19,21,22,46,47 The larg-est known tool set comprises fivetools used to get honey.46 A specificorder in the use of each tool of theset is needed to reach the goal.A composite tool is ‘‘two or more

objects used simultaneously and com-plementarily to achieve a goal.’’43,44

Each of these objects can be used sin-gly, but its usefulness is diminished.Examples of composites are the use ofa small branch to push in and pull outa leaf sponge in a tree-hole containingwater for drinking48 and the use ofwands to dip for driver ants while usingbent-over saplings as perches.49 At Bos-sou, Carvalho and coworkers50,51 sys-tematically studied the use of a custom-ary composite tool: portable anvils andhammer stones for nut-cracking.A compound tool is ‘‘two or more

components combined as a singleworking unit’’.43 The most wide-spread example is the leaf ‘‘sponge,’’in which several leaves are com-pressed together to sop up water.26,52

Another example, from Bossou, isthe wedge stone inserted under astone anvil to level and to stabilizethe surface for nut cracking.3 Theuse of up to two wedges used at thesame time to stabilize an anvil hasbeen reported in a natural nut-crack-ing site in Bossou.51 Thus, the com-pound anvil is used with a hammer,making, in total, a composite tool.

Excavation of chimpanzee

nut-cracking sites

Mercader, Panger, and Boesch53 werethe first to excavate a chimpanzee siteusing standard archeological techni-ques. The chimpanzees of Taı̈, IvoryCoast, had been observed using thissite intermittently for nut-crackingover a period of 26 years until thedeath of its source (nut) tree. Nut-shells were associated with stones,and hammers were intentionallybrought to the anvils; stone hammerswere fractured and shattered, appa-rently accidentally. This researchshowed that the chimpanzees’ reuseof activity areas for nut crackingresulted in accumulations of stone

Figure 1. Chimpanzee tool used at Ugalla to dig for USOs and log from which it wasextracted. Photo by R. Adriana Hernandez-Aguilar. [Color figure can be viewed in theonline issue, which is available at wileyonlinelibrary.com.]

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 183

and plant refuse at specific points onthe landscape, forming sites, and thatthe tools exhibited wear patternscharacteristic of these activities.Also at Taı̈, Mercader and col-

leagues54 conducted the first studyof an ancient chimpanzee site andnoted the type of raw material, mor-phology, dimensions, and patternsof use-wear on the stone assem-blage. Based on these characteris-tics and on the taxonomic identifi-cation of the nut species repre-sented in the starch assemblage, theresearchers concluded that the exca-vated site was the product of chim-panzee nut-cracking activities in theremote past. They also found prod-ucts of intentional systematic flak-ing of stone and inferred that spo-radic human activities had contrib-uted to the assemblage.Radiometric dating of material fromsmall lightning-caused fires revealedthat chimpanzees have been usingstone tools at the locale for at least

4,300 years, suggesting that nut-cracking was transmitted from gen-

eration to generation over at leastthat time span.

Figure 2. Adult male chimpanzee of theMoto community of the Goualougo Triangle Ape Project, Republic of Congo, uses a tool set. He uses apuncturing tool to create an access tunnel into a subterranean Macrotermes nest, while holding an herbaceous fishing probe in his mouth.Photo by Ian Nichols, National Geographic Society. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 3. Natural chimpanzee nut-cracking site at Bossou, with hammers, anvil, andoil-palm nuts. Note wear patterns on stone, resulting from repeated use. Photo copyrightSusana Carvalho. [Color figure can be viewed in the online issue, which is available atwileyonlinelibrary.com.]

184 Wynn et al. ARTICLE

Orangutans, Gorillas,and Bonobos

Although orangutans had sporadi-cally been seen to use unmodifiedobjects as tools,31 only in the mid-1990s was regular tool-making andflexible tool use reported for thisAsian ape.34 Fox, Sitompul, and vanSchaik32 further described task-de-pendent modification of tools. All ofthe manufactured tools were used intrees in two foraging contexts:extracting insects or honey fromtree-holes and seeds from hard-shelled fruits. In the second context,tools often were made before access-ing the fruit tree. Tools also were fre-quently reused to extract seeds fromdifferent Neesia fruits. The arborealapes transported a tool in theirmouths between feeding sites in thesame tree or to different trees. Fromthe list of behavioral patterns for sixwild populations of orangutans inBorneo and Sumatra, as presentedby van Schaik and coworkers,33 15types of tool use can be identified, 10of which are either customary (forexample, seed extraction tool, leafnapkin) or habitual (for example,leaf glove, branch as swatter) in atleast one study site. This list did notinclude all tool use cases everobserved in wild orangutans. Otherexamples include the use of a leafybranch as an umbrella during rain32

or as a branch hook.55 van Schaik,van Noordwijk, and Wich56 describedyet another type of technology, leafcarrying, a habitual behavior seen atonly one of seven wild orangutansites, Tuanan. The leaves were alwaysplucked from the same species of treeand carried for up to 50 meters andfor up to half an hour; they weresometimes put down while feeding,but were picked up again and takento the nest site, where they were usedto make nest covers or linings.Gorilla tool use is only anecdotal.

Breuer, Ndoundou-Hockemba, andFishlock57 reported for the first timethe use of tools by this species in na-ture: A female detached a branchand thrust it into a pool to test thewater’s depth or the stability of thesubstrate, then used it as a walkingstick for postural support. Anotherfemale detached the stem of a shrub,

inserted it into the ground andleaned on it for support while col-lecting aquatic herbs; after that sheplaced the tool on swampy groundand used it as a bridge to crossquadrupedally.

Thirteen types of tool use arereported for bonobos (at two studysites: Wamba and Lomako), but onlytwo are customary (branch drag,buttress beat) and four habitual26

(rain hat, rain cover, leaf clip bymouth or by fingers). A single type oftool has been seen (four times) to beused in subsistence: the mosssponge, equivalent to the leaf spongeof chimpanzees.27,28 No bonobo innature has been seen to make a tool.

Old World and New WorldMonkeys

The productivity of field researchersin documenting the tool use reper-toires of apes such as Pongo and Panhas been matched by research on NewWorld and Old World Monkeys (Boxes1 and 2). In particular, the generaMacaca and Cebus have yielded star-tling patterns of object manipulationof stones and tool use. Especially pro-vocative is the convergence of percus-sive technology in Macaca fascicularis,the long-tailed macaque, and Cebuslibidininosus, the bearded capuchin, toextract food that is encased and inac-cessible without tool use (Fig. 4).These new findings provide fertileground for primate archeology andexpand modeling exercises to morephylogenetically distant taxa that mayallow us to scrutinize the intersectionof habitat, socio-ecology, and tool use.They also raise the possibility thatwhat we here term an ape adaptivegrade is yet another example of taxo-nomic myopia.

FIELD EXPERIMENTATION

Reconciling the contrast betweentightly controlled but artificial settingsin captivity versus uncontrolled butecologically valid settings in nature isa perennial dilemma. One way out ofit is to transfer experimental methodsinto the wild, embedding managedsettings into the daily lives of apes,who may or may not volunteer to takepart. This approach means controlling

as many natural variables as possiblewithout biasing the subjects’ normalbehavior. Matsuzawa’s4 researchgroup at Kyoto University, working atBossou in Guinea, has pioneered this,especially in studies of stone tool usein cracking nuts. The researchersmake available arrays of lithic rawmaterials that control for key varia-bles, such as size, weight, shape, typeof stone.18,21

Chaines Operatoires

Carvalho and coworkers51 appliedchaines operatoires technique, whichinvolves the description of an actionsequence from acquisition of rawmaterials to discard of exhaustedtools, in a study of wild chimpanzees.They demonstrated the existence ofoperational sequences in nut-crackingtechnology, both by observation andexperimentation at Bossou (outdoorlaboratory and natural forest) and byindirect evidence at Diecke (naturalforest), Guinea. They found that forall steps from the selection of raw ma-terial to the final discard of a tool, se-quential actions were needed to reachthe goal; the apes exhibited flexibilityin making this sequence longer andmore complex. Chimpanzees discrimi-nated the functions of a tool based ontool features (size and weight). Theauthors found diversity in the stoneassemblages of the three sites studied.Spatial analysis of the percussion

revealed different activity areas andthe existence of three strategic choicesof resource exploitation: using toolsand nuts at the nut tree; carrying nutsto a different place with tools or tak-ing stones to join previously carriednuts; or transport of tools and nuts toa spot away from others.Systematic study of one type of

composite tool, the hammer-and-anvil, in Bossou’s outdoor labora-tory50 revealed that chimpanzees sys-tematically selected anvils and ham-mers not only independently, but alsothat individuals preferred particularhammer-and-anvil combinations.

LABORATORY EXPERIMENTATION

Apes have figured in laboratory-based behavior studies since theearly studies of Wolfgang Koehler on

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 185

chimpanzees, almost a century ago.Modern studies seek to recreatenatural phenomena in controlledconditions (for example, tool use inextractive foraging) to probe the lim-its of nonhuman abilities, the pointbeing to model counterpart capaci-ties important in human evolution(such as stone tool making) or to testcognitive capacities, both individualand social (for example, cooperativeproblem-solving).

Bonobo Knapping

A long-term experimental projectdesigned to study the ability of bono-bos to make and use Oldowan-typestone artifacts began in 1990 with ayoung male named Kanzi.58 Over thecourse of several weeks, Kanzilearned, by watching a humanmodel, how to knap stone flakes anduse them for cutting. At first he wasreluctant to use the model’s knap-ping gesture of striking a core heldin one hand with a hammer held inthe other, and instead invented a

technique in which he forcefullyhurled cores onto a hard substrate.He eventually acquired the standard,hand-held gesture, after he was pre-vented from breaking cores by pad-ding the hard surfaces, and exhibitedrapid improvement.58–61 Schick andcolleagues62 later reported the con-tinued refinement of Kanzi’s stonetool-making capabilities, althoughKanzi still preferred to make tools bythrowing rather than by the standardknapping gesture. After a period ofthrowing as the preferred techniquefor flake production, Kanzi resumedknapping in a much more efficientmanner than in the earlier trials andslowly shifted to a preference forknapping over throwing. He seemedto understand that successful flakingdepended on hitting the core in aprecise and systematic way and con-tinued to improve the bimanual per-cussion, eventually achieving effectiveand efficient flaking27,28,58,61,62

Conchoidally flaked stone artifactsmade by bonobos are similar in impor-tant features to Oldowan artifacts.63

Toth, Schick, and Semaw64 madedetailed comparisons of early Oldo-wan stone tools from Gona, Ethiopia,with those made by experiencedbonobos (Kanzi and Panbanisha) andhumans. The comparison, whichmainly focused on the products ofknapping, showed that bonobo arti-facts shared many similarities withOldowan artifacts. Although theywould be clearly recognizable as arti-facts in an archeological context, theyexhibited less precision, less reduction,fewer flake scars, and more batteringfrom unsuccessful blows, which wasinterpreted as a possible product ofbiomechanical and cognitive limita-tions of the apes.64,65 Whiten, Schick,and Toth63 concluded that, ‘‘basedon the proficiency in the acquisitionof stone tool making by bonobos,the acquisition of stone tool makingproficiency in these bonobos makesit probable that a basal ability forthe effective stone-to-stone percussionnecessary for stone tool manufactureexisted in our common Pan-homininancestor, along with a capacity to

Box 1: Old World Monkey Tool Use

The macaques (Macaca spp.) area varied and widespread evolution-ary radiation of Catarrhine mon-keys, living in a variety of habitatsfrom seashore to mountain foothills,mostly in Asia.106 Some membersof the genus, such as rhesus(M. mulatta) and long-tailed (M. fas-cicularis) macaques have been well-studied, especially in laboratories.However, the longest-running stud-ies are of Japanese macaques (M.fuscata), which date to the pioneer-ing work of Imanishi and his stu-dents from the 1950s to the pres-ent.107 Despite this attention, therewere, until recently, no reports ofhabitual tool use in any populationof macaques, in nature or captivity.Hence, they rarely have been incor-porated into human evolutionarymodeling. Now, this has changed.Astonishingly, long-tailed maca-

ques in Thailand have been reportedto pluck and use human hairs astools to floss their teeth.108 Laterresearch shows that macaque moth-

ers in the presence of their infantsappear to teach their offspring thebehavioral pattern.109 This customwill leave no archeological record,as the artifacts are organic, but itreminds us that novel behavior canalways turn up, given opportunityand inventiveness.Much better-known is stone han-

dling, a complex of behavioral pat-tern that involves pebbles beingmanipulated by sitting monkeys,usually at feeding areas.110 Themanner of manipulation variesimmensely: 45 components havebeen catalogued, such as rubbing,tapping, stroking, and mouthing.111

None of the stone handling is tooluse; that is, the stones are not usedto accomplish any visible goal, butsome of the behavioral patternsprobably leave recognizable modifi-cations that could be subject tolithic analyses. The complex variesgreatly across populations of wildmacaques, suggesting cultural ori-gins, but is most commonly

expressed in provisioned groupsversus unprovisioned ones.112

Most impressive among macaqueelementary technology is the use ofstones as hammers to extract food-stuffs from natural casings. Long-tailed macaques on islands in theAndaman Sea use percussion tocrack open nuts and molluscs onanvils and to chip open oystersattached to the intertidal sub-strate.113,114 These pounding andaxe hammers differ in size andshape and display diagnostic wearpatterns according to their differentmethods of use. There seem to bestrong functional convergences inpercussive technology between themacaques, capuchins, and chimpan-zees, but these remain to be studiedin detail.No other Old World monkeys

show habitual or customary elemen-tary technology, but these recentrevelations remind us never to closethe book on nonhuman primate ar-cheological possibilities.

186 Wynn et al. ARTICLE

socially transmit the essence of theskill.’’43:427 Replication of this experi-ment with other ape genera may allowextension of this conclusion to a com-mon ape ancestor.Since 1989, the picture of non-

human primate technology-basedadaptations has broadened dramati-cally. From termite fishing at Gombeand nut cracking at Taı̈, which formedthe core of the 1989 analysis, theknown tool-based activities of chim-panzees have expanded to encompassa much greater variety of tools, includ-ing tool sets, composite tools, andcompound tools, and a greater varietyof foods accessed and processed.Moreover, it is clear that much of thisactivity depends on sequentially andhierarchically organized actions thatallow flexibility in task completion.66

Chimpanzees represent the high end ofa range of reliance on tool-assistedactivities67 now documented for otherapes, and also monkeys (see boxes).This record of technology-based adap-tations provides a rich context fromwhich to view the earliest flaked stonetechnology, the Oldowan.

THE OLDOWAN

From the perspective provided by

nonhuman primate tool making and

tool use, particularly in chimpan-

zees, the earliest evidence of homi-

nin tool use appears very familiar in

regard to the multiplicity of differ-ent tools and the variety of usesto which they were put. The follow-ing account of the Oldowan updatesthe one used in 1989, with emphasisplaced on new discoveries andanalyses.

Box 2: Capuchin Monkey Technology

Capuchins (Cebus spp.) are awidely distributed evolutionary radi-ation of Platyrrhine (New World)monkeys, found from Honduras toArgentina, and from primary rainforest to arid scrub.115 One species,C. apella, is famous not only as alongstanding subject of laboratorystudies, but also for its ‘‘train-ability,’’ often appearing in featurefilms. This species is famous for itsintelligence, but its natural behaviorshows no signs of tool use in thestrict sense; only a few instances ofanecdotal tool use have beenobserved. The most studied speciesin nature is C. capucinus, the north-ernmost member of the genus andthe only one ranging into CentralAmerica.116 Based on long-termstudies, these white-faced capuchinsshow notable parallels with chim-panzees, such as social hunting,meat-sharing, and interpopulationalbehavioral variation. However, inthe last 10 years, it is C. libidinosus

(bearded capuchin) that has leapt toprominence through several studiesin northern Brazil in savannah-likeecotypes (caatinga, cerrado).117 Theirimpressive technological repertoire,including lithics, has captured theattention of primatologists and non-primatologists alike.Bearded capuchins at two study-

sites, Fazenda Boa Vista and Serrada Capivara, regularly use hammersand anvils to crack open several spe-cies of palm nuts. They use stonesas hammers, and rocky boulders oroutcrops and fallen logs as anvils.They prospect, collect, transport,and reuse stones to open the hard-shelled palm fruits, which are pref-erentially placed for cracking inworn depressions (‘‘pits’’) in theanvils.118 Various types of lithic rawmaterials, some rare, are exploited.The hammers used are heavy, asmuch as 25%-40% of averageadult body weight. Experimentalstudies done in nature show the

monkeys to be discerning users thattake into account the mass, size,and composition of stones, andthe species of nut, in makingchoices that optimize hammeringefficiency.119

The tool-kit of the bearded capu-chins goes beyond percussive ex-tractive technology.120 Some groupshabitually use pointed stones astrowels to dig up tubers or to openup wood to get insects. Sticks areused as probes to access embeddedfood or water or even to skewer ver-tebrate prey such as lizards. Suchtools may be made by reduction,such as stripping of twigs andleaves. One population uses stonebanging as an aggressive auditorydisplay121; in another, individualsthrow stone at one another to initi-ate social interaction. Overall, thetool kits of capuchin monkeys rivalor even exceed those of chimpanzeepopulations in scope and complex-ity, if not yet in size.

Figure 4. A long-tailed macaque in Thailand uses a stone ‘‘axe-hammer’’ to crack mol-lusk shells. Photo by Michael Gumert. [Color figure can be viewed in the online issue,which is available at wileyonlinelibrary.com.]

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 187

The Oldowan Technocomplex

Oldowan assemblages are made upof differing combinations of unmodi-fied stone, stone hammers, knappedcores, knapped flakes, and debitage.Some assemblages, such as FLK 22(Zinj) at Olduvai, occasionallyinclude flakes with minimally modi-fied edges. There are no tools thatsuggest attention to overall shape,and few convincing examples of bonetools. These artifact categories werethe result of repeated application ofhard hammer knapping of flakes, ei-ther by direct percussion in which ahammer held in one hand struck acore held in the other, or a bipolartechnique in which the core is posi-tioned on a stone anvil and struckwith a hammer stone. There were notool ‘‘types’’ in the modern culturalsense of the term.2,65,68 Instead, thehominins deployed a variety ofprocedures in specific circumstances,resulting in the range of artifactspreserved archeologically. There wasno need for ‘‘mental templates’’ inthe sense of preconceived tool cate-gories or shapes.

When, Where, and Whom?

The known age of sites labeled asOldowan now appears to havespanned two million years. The sitesat Gona, in Ethiopia, still boast

the earliest reliably dated stone toolsat 2.6 Ma (Figs. 8 and 9),69–71

although indirect evidence of cut-marked bone now suggests the useof stone flakes as early as 3.4 Ma.72

However, in the absence of stonetools, many archeologists are skepti-cal of this early Dikika evidence.73

The Oldowan sites at Peninj, whichdate to only 1.4 Ma, anchor the morerecent end.74 The late dates for Pen-inj make it especially difficult to at-tribute agency; besides Paranthropus

and Homo habilis, Homo erectus isalso a candidate. Some archeologists

now argue that late Oldowan

assemblages differed technologically

from early ones, though there is no

agreement on this point.2 The possi-bility of technological change within

the Oldowan tradition is counterbal-

anced by a well-established fact of

the archeological record: Flake and

core assemblages that meet the basic

parameters of assignment to the Old-

owan continued to be manufactured

in various parts of the world up to

historic times. Thus, in order to

sharpen the focus of our compari-

son, we limit our Oldowan sample to

sites dating between 2.6 Ma and 1.7

Ma.The geographic distribution of

sites attributed to the Oldowan nowextends from North Africa and theIberian peninsula to Dmanisi in theCaucasus, Swartkrans in SouthAfrica (that is, most of Africa),southern Europe, and southwestAsia.75 These temporal and geo-graphic boundaries encompass arange of habitat types and hominintypes. Identifying the hominins re-sponsible for the Oldowan remainsproblematic. The best direct associa-tion is still the one with Paranthro-pus boisei and Homo habilis in BedI at Olduvai Gorge. The closest

Figure 6. A chimpanzee cracks oil palm nuts with a stone hammer and anvil at Bossou. Indi-vidual on the left watches, holding a hammer stone. Photo by Susana Carvalho. [Color fig-ure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 5. A chimpanzee infant at Bossou watches its mother crack oil palm nuts withstone hammer and anvil. Photo by Susana Carvalho. [Color figure can be viewed in theonline issue, which is available at wileyonlinelibrary.com.]

188 Wynn et al. ARTICLE

association with the earliest Oldo-wan artifacts at Gona is withAustralopithecus garhi. It is no lon-ger tenable to assign the Oldowansolely to larger-brained forms. Hom-inins with relative brain sizes withinthe ape range may have manufac-tured any or all of these earlieststone tools.

Tool Use

Oldowan hominins used stonetools for butchery and, perhaps,other food processing. Evidence ofthis consists of fragmentary boneassociated with tools at several Oldo-wan sites and, more tellingly, the na-ture of damage on the bones. Manyof the bones exhibit cut marks pro-duced by stone flakes. The bonescome from a range of animals,mostly herbivores ranging in sizefrom 10 kg to over 2,500 kg,although ungulates that rangedbetween 100-350 kg were most com-mon.76 Bone damage indicates thatthe hominins used stone tools to per-form two butchery tasks: sharp flakesto cut meat from bone and roundedcobbles to split long bones for mar-row. Although associated bone hasbeen found at several Oldowan sites,one site in particular, FLK 22 (Zinj)at Olduvai, remains the bestdescribed.77–80 FLK was situated in a

densely wooded spot within 200 m ofa spring. The hominins carried partsof carcasses to the site, where theyprocessed them further. Archeolo-gists continue to disagree about howthe hominins accessed the carcasses.Some favor hunting or confronta-tional scavenging,79 while othersfavor more passive forms of scaveng-ing, such as pilfering from leopardkills cached in trees.76,77 Reliance oncarcasses of medium to large ungu-lates, however they were acquired,put these hominins in either physical(confrontational scavenging) or eco-logical competition with members ofthe carnivore guild, including notjust lions, leopards, and hyenas, butalso saber-toothed cats. What cannotbe determined from the archeologi-cal record, at least not yet, is howimportant meat-eating was to thediet of Oldowan hominins. From themore than one- million-year span ofthe Oldowan (as considered here)there is only one site, FLK 22 (Zinj)at Olduvai, where evidence of meatconsumption is extensive.

Some Oldowan sites have unmodi-fied cobbles and stones carried infrom elsewhere. Some of thesemanuports exhibit evidence of use inbashing. Mora and de la Torre81

state that the number and size ofthese tools argues for bashing notjust marrow bones, but also plantfoods, including USOs.

Transport

The recently excavated sites atGona,69–71 Koobi Fora,82,83 Lokala-lei,84,85 and Kanjera86,87 give a moredetailed picture of artifact transportthan the one presented in 1989 byOlduvai and Koobi Fora alone. Inparticular, it is now clear that by 2.6Ma hominin knappers carried flakedcores from a source to locationswhere they were further reduced.The telltale cortical flakes that indi-cate the first steps of core reductionare missing, or at least few, at manyof these sites. It also appears thathominins occasionally carried awaylarge flakes.2,65 This pattern of trans-port has suggested to many archeolo-gists that the Oldowan knappers hada sophisticated lithic procurementsystem.70,71,82 However, this need notbe the most parsimonious explana-tion. The hominins could simplyhave picked up knapped cores fromone spot and moved them to anotheror, more likely, knapped a core at itssource to test its potential andreduce its weight. Neither practicewould have required elaborate pro-curement systems or plans of action.Testing of cores may simply havebeen a component of foragingpatterns. The hominins did preferparticular raw materials, and theirchoices reflected not only ease offlaking but also durability.87 Hun-dreds of millennia of knapping pro-vided ample opportunity to learnwhich type of stone was good andwhich was not. Such discriminationis not cognitively difficult. The homi-nins did not rediscover it every gen-eration; they learned it from observ-ing adults, a long-established apepattern. Such social learning doesnot obviate the possibility that differ-ent populations and even species ofhominins independently inventedstone knapping. In fact, this is verylikely, based on the convergence inpercussive technology in differentprimate species.88

It is the distance that some of thisstone was carried that set homininsapart. Most raw material was carrieda few hundred meters at most, usuallyfrom a nearby gravel source. Butsome was carried over 10 km. Thereare a few such examples from Olduvai

Figure 7. A chimpanzee selects hammer and anvil stones from an experimental array atBossou. Note nuts in the mouth. Photo by Susana Carvalho. [Color figure can be viewedin the online issue, which is available at wileyonlinelibrary.com.]

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 189

(gneiss from ca. 13 km)80 and acorroborating (and earlier) set fromKanjera, where cobbles of variousraw material were transported from aconglomerate 10-13 km away.86 AtKanjera, more than 25% of the rawmaterial appears to have been carriedfrom this distant source. Braunand colleagues86 suggested that such

long-distance transport is evidence ofa procurement system of some com-plexity. Whether complexity is theappropriate term or not, it indicateseither a series of short distance trans-ports or a willingness to carry stonefor hours.

Oldowan hominins also carriedthe body parts of animals, but the

distance is almost impossible tocalculate. The Olduvai sites of FLKZinj and FLK N were within a fewhundred meters of a permanentwater source that appears tohave been a common death site.89

There is no evidence that thebody parts were carried further thanthis.

Figure 8. Flaked pieces from Gona EG10 and EG12. Battering on artifact ‘‘f’’ indicates use as hammer.122 Copyright � 2010, Elsevier, bypermission. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

190 Wynn et al. ARTICLE

Procedures

Since 1989, developments in lithicanalysis, including replication stud-ies,64 core refitting,84 knappingerrors,90 scar patterns on cores,74

and spatial concepts91 have expandedunderstanding of Oldowan knappingprocedures to include the following:use of direct hard-hammer technique;use of different procedures ondifferent raw materials (for example,bipolar for quartz and direct percus-sion for lavas); searching cores forknappable angles; systematic rotationof cores when platforms or produc-tion surfaces failed; removal of hingeor step flake-marred productionsurfaces (core rejuvenation); use offlake scars for platforms; contiguousplacement of knapping blows (topo-logical spatial concepts); and occa-sional unifacial trimming of flakes(but only in a few assemblages).It is at this level of procedure that

current Oldowan specialists havestaked claims for cognitive complex-ity and a qualitative differencebetween the technology of apes and

early hominin knappers. Based onrefitted cores, Roche84,85 has arguednot only that the knappers at Lokala-lei 2C were ‘‘cognitively complex,’’but that it is possible to distinguishdifferent levels of complexity amongOldowan assemblages. Semaw,Rogers, and Stout71 demurred on thelatter point, but accepted the generalattribution of complexity, as do mostarcheologists working in the field.However, Roche’s claims for hier-archical complexity in Oldowanprocedures can easily be matched,for example by the hierarchical com-plexity employed by gorillas whenmanipulating thistles,92 chimpanzeesusing tool sets to get honey,46 andthe Galago hunters at Fongoli.38

One of the few examples of chainesoperatoires that invokes theoreticallygrounded concepts is Haidle’s,93

which compared decision points andattention shifts evident in an episodeof chimpanzee termite extractionwith an episode of Oldowan meatcutting. Attention shifts are impor-tant components of executive prob-lem-solving ability and working

memory.94,95 Haidle found that thechimpanzee and Oldowan exampleswere almost indistinguishable in thisregard. With few exceptions, claimsfor Oldowan cognitive complexityrarely have been grounded in well-defined cognitive concepts. Whenthey have been, as in Haidle’s use ofworking memory and in the 1989article’s1 use of Piagetian spatial con-cepts, the cognition of Oldowanhominins has fallen squarely in theape range.

DISCUSSION

Since 1989, primatological researchhas dramatically expanded science’sunderstanding of the technology-based adaptations of apes. Amongthe great apes, chimpanzees continueto show the most complex and flexi-ble tool use. Some behavioral pat-terns found only in chimpanzeesinclude using one type of raw mate-rial to make several kinds of tools(for example, a leaf used as sponge,napkin, or fishing probe)96; makingone type of tool from various raw

Figure 9. Flaked pieces from Gona OGS 7.122 Copyright � 2010, Elsevier, by permission. [Color figure can be viewed in the online issue,which is available at wileyonlinelibrary.com.]

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 191

materials (fishing probe made ofgrass, bark, vine, or twig)3,96; usingtwo kinds of tool to achieve the samegoal (fishing probe and digging stickto obtain termites)19,21; using twotypes of percussive tools to solve thesame task (hammer or anvil to dis-mantle large fruit)42; and using onetool for several functions (dip, lever,and pound in honey extraction).20

Although some types of tool use areuniversal to all chimpanzee popula-tions, the tool kits of chimpanzee com-munities differ markedly, highlightingthe impressive variation in materialculture for the species.26,96,97,98

After five decades of studying wildchimpanzees, new types of tool usecontinue to be found, thus extendingthe impressive repertoire of this spe-cies.67 Some examples of novel tooluse come from dry, open, and sea-sonal habitats (Fongoli, Ugalla), andsome from rainforest (Nimba). Some

of the new tool use reports come

from unhabituated subjects (Ugalla,

Nimba) where indirect evidence

guided their discovery. Newly

reported types of tool use, such as

tool-assisted hunting, USO digging,

and cleaver percussion, shorten the

list of tool-assisted behavioral pat-

terns believed to be uniquely homi-

nin. Hunting vertebrates and harvest-ing USOs with tools are thought tobe key adaptations in human evolu-tion, but as components of chimpan-zee foraging their adaptive signifi-cance must now be extended to acombined Pan-hominin clade, atleast. Although chimpanzees havenot been seen to use tools to makeother tools, they may unintentionallyproduce a second tool during tooluse. At Bossou, during the pestlepounding of oil palms, the pounderproduces a sponge of drenched fiberthat is then used repeatedly to suckup the sap.99 In nut cracking, Car-valho and colleagues51 reported thata large flake detached from an anvilwas then reused as a hammer. Theystressed that although such fractur-ing may have been accidental, theresult was a tool produced byanother tool. Unintentional productsof chimpanzee percussion activitieswere also identified in the excavatedarcheological assemblages at Taı̈.53,54

Although accidental stone tool-mak-ing by apes is not the same as inten-tional knapping of stones by homi-nins, it is an important occurrenceand allows for the hypothesis thatthe unintentional fracture of anvilmargins was the first step in the evo-lution of stone knapping.100

Bossou’s outdoor laboratory resultson nut cracking show that chimpan-zees exhibit systematic patterns ofselection, transport, use, reuse, modi-fication, and discard.51 Application of

the chaine operatoire method to chim-panzee percussion reveals that chim-panzee sequences resemble thosereconstructed for hominins. Carvalhoand coworkers51 suggested that vari-able strategic choices of resource ex-ploitation shown in chimpanzee nutcracking are flexible, dynamic, oppor-tunistic, and energy-saving, as hasbeen said of the Oldowan.

Nut-cracking chimpanzees selectthe types of stones that are optimalfor specific tool-using activities8 basedon features such as size and weight.51

Chimpanzees transport stone tools(hammer, anvil, wedge) and nutsacross the landscape.9,42,51,54 How-ever, Mercader53 noted that the totaldistance from the source at which astone was collected to the placewhere it was used last and exhaustedremains unknown. Besides tools usedin nut cracking, object transport inchimpanzees or other primates hasnot been systematically studied. Dur-ing tool-making and use, chimpan-zees modify the raw materials bothintentionally and unintentionally, pro-ducing distinctive evidence of theactivities performed, which is distin-guishable from natural damage. Forexample, during nut cracking stonetools are pitted, abraded, flaked, andfractured, leaving diagnostic use-wearpatterns.51,53,88 Spatial concentrationsof refuse from chimpanzee activitiesat specific spots on the landscape andthe recurring reuse of these placesforms sites. Depending on the type ofraw materials used or deposition con-ditions, these assemblages may enterthe archeological record, challengingthe traditional view of hominins asthe only creators of archeologicalsites.51,88,100–103

Orangutan tool use resembles thatof chimpanzees in several ways: vari-ation of tool kits across populations,tools used in extractive foraging, andcompound tools (leaf sponge, leafwipe, leafy-branch umbrella). Wenow know that bonobos can learnsocially to make stone tools, mastercomplex stone-modifying techniques,improve skills by practice, inventnew ways to make flakes, and createtools that share some characteristicswith Oldowan tools.64 Social learningin wild chimpanzees seems to occurthrough observation and trial-and-error, mainly from mother to off-spring.5,6,10,96,104 Long-term studiesof bonobo stone tool-making in cap-tivity suggest that the main mode oflearning is through observation ofothers doing the activity.60,63,64

The ‘‘Apeness’’ of the Oldowan

Tables 1 and 2 compare ape andOldowan technology using the same

Newly reported typesof tool use, such as tool-assisted hunting, USOdigging, and cleaverpercussion, shorten thelist of tool-assistedbehavioral patternsbelieved to be uniquelyhominin. Huntingvertebrates andharvesting USOs withtools are thought to bekey adaptations inhuman evolution, but ascomponents ofchimpanzee foragingtheir adaptivesignificance must nowbe extended to acombined Pan-homininclade, at least.

192 Wynn et al. ARTICLE

format that was set out in 1989.Many of the tool-based activitiesdocumented for apes through obser-vation and experimentation have alsobeen documented archeologically forthe Oldowan hominins. However, ifanything, living apes demonstrate agreater variety of technical activitiesthan can be reconstructed for theOldowan. Given the likelihood thatthe archeological record under-represents the variety of Oldowantool-based activities, it is more in-formative to condense these lists evenfurther to clarify the overarching simi-larities. Living apes and Oldowanhominins share the following:Use of tools to access food: Apes

use tools to access otherwise hard-to-get foods, including insects, nutmeats, honey, water, brains, marrowand USOs. Oldowan hominins used

tools to access meat from carcasses,marrow from bones, and perhapsalso plant foods, including USOs.Paleoanthropologists often makethe implicit, erroneous assumptionthat technology is ‘‘optional’’ forchimpanzees (Pan troglodytes) butwas obligate for the hominins whomade Oldowan tools. The reality isdifferent: All populations of well-studied chimpanzees depend ontechnology to function adaptively inthe environments in which they live.Arguably, without the use of tools,the chimpanzees’ intake of animalprotein (for example, insects) orplant protein (such as nuts) and fatwould be seriously diminished,affecting individual survival.

Use of tools to process foods:Apes use tools to break fruit intomanageable sizes. Oldowan homi-

nins used tools to butcher carcassesinto manageable sizes and alsopounded meat and perhaps USOs.Discrimination and selection of

raw materials: Apes select optimalraw materials for use as hammersand probes. Oldowan homininsselected raw material for knappingbased on ease of flaking and durabil-ity of resulting flakes.Selection of tools in advance of

use: Chimpanzees select sticks toskewer bushbabies, vines to probe fortermites, and stones to crack nuts, allin advance of episodes of use. Oldo-wan hominins selected cores forknapping and flakes for later use.Carrying tools and food: Apes

carry tools and occasionally food toprocessing sites. Oldowan homininscarried cores, flakes, and parts ofcarcasses to processing sites.

TABLE 1. Tools and Procedures

Ape tools and procedures Oldowan tools and procedures2,65,68,71,74,75,81,83,84,122

Known tools Known tools� Stone hammers50,51 � Stone hammers� Stone anvils3,50,51,100 � Stone anvils� Flexible probe6,45,46 � Flaked cores� Stiff probe45,46,48,49 � Unmodified flakes� Wooden hammers9,10 � Modified flakes� Wooden anvils9,10 � Battered cobbles� Digging tool39 � Unmodified cobbles� Spear/sharpened stick17 � Modified bone (rare)� Stone ‘cleaver’42

� Wooden ‘cleaver’42

� Sponge5,52

Groupings/types Groupings/types� Ad hoc � Ad hoc� Standardization from selection of raw materials of aptsize and strength3,50,51

� Some modalities linked to raw material (e.g., size)

Manufacturing procedures Manufacturing procedures� Raw material selection and transport � Raw material selection and transport� Use of simple topological spatial concepts(proximity, boundary, order)

� Use of simple topological spatial concepts (proximity,boundary, order)

� Bite, strip, chew, bend � Direct hard hammer knapping and bipolar knapping� Different procedures linked to different raw materials � Searching for knappable angles

� Contiguous placing of knapping blows� Use of flake negatives as platforms� Core rotation� Core rejuvenation� Unifacial trimming of flakes (late)

Tool sets/combinations/composites Tool sets/combinations/composites� Tool kit19,28,31 � Hammer 1 core� Tool set – e.g. Pounder, perforator, enlarger,collector, swab22,45–47

� Hammer 1 anvil

� Tool composite – e.g., Hammer 1 anvil48,50,51� Hammer 1 core þ anvil (bipolar)

� Compound tool – e.g., anvil 1 wedge; leafsponge3,51

� Hammer 1 [food] þ anvil

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 193

Reuse of activity areas: Apesreuse activity areas, including nut-cracking stations, occasionally foryears, and termite mounds for deca-des. Oldowan hominins reused proc-essing sites perhaps intermittentlyfor years.105

Hierarchically organized proce-dures: Apes use hierarchically organ-ized procedures in preparing andusing tools and tool sets. Oldowanhominins used hierarchically organ-ized knapping procedures. The num-ber of steps and subroutines isdirectly comparable. When ape

chaines operatoires are contrasteddirectly with Oldowan chaines opera-toires using the same conventions, asin Haidle’s93 analysis, they appearremarkably similar.

Use of flexible procedures: Infield experiments at Bossou, chim-panzees adjusted their procedures todifferent local problems. Oldowanhominins applied different proce-dures to different raw materials andadjusted procedures to differentqualities of cores.

Cultural differences: Differentcommunities of the same species use

different socially learned technicalsolutions. Different Oldowan localitiesexhibit different socially learned knap-ping procedures.Experiments with captive bonobos

have established that they can learnto knap stone. These experimentshave added another perspective onpossible cognitive differencesbetween apes and Oldowan homi-nins.58,59,61,62,64 Although both Kanziand Panbanisha have learned toknap, neither has achieved the skilllevel of Oldowan knappers.64 It is farfrom clear why not. Toth argues that

TABLE 2. Uses and Chaines Operatoires

Ape tool uses and chaines operatoires Oldowan tool uses and chaines operatoires

What is processed? What was processed?75–79,81

� Termite mound47,49 � Marrow bones of small to large herbivores� Ant nest49 � Soft tissue (flesh, tendons) of small to large herbivores� Bee hive20 � Roots, stems, corms, etc.� Nut50,51

� Hard shelled fruit34,100

� Big fruit42,100

� Vertebrate prey17

� Water52

� Honey20,45,46

How is it processed? How was it processed?75–79

� Pound, prise, cleave object � Splitting bone diaphyses with cobble hammers� Insert, extract contents � Slicing flesh and tendons with flakes� Contain, concentrate prey � Pounding roots, stems, corms, and perhaps flesh59

� Sop liquids

Where is processing done? Where was processing done?75–79,89

� At mound, nest, hive � At death sites of herbivores� Stream bed � More often away from death sites at protected places� Kill site� In trees� Beneath trees

What is carried? What was carried?2,65,69,78,79

� Raw material – lithic and organic56,96 � Raw material� Hammer3,9,50,51 � Hammers� Anvil3,50,51 � Flaked cores� Probe5,96 � Large flakes� Nut50,51 � Cobbles� Fruit51 � Body parts of herbivores� Carcass5

How far are things carried? How far did they carry things?80,82,83,86,89

� Nut – 265 m � Mostly a few hundred meters from nearby streams or� Hammer – 500 m conglomerates� Probe – 1 km � Occasionally as far as 13 km� Carcass – 6 km

Chaines operatoires (example of termite extraction,distilled from Haidle93)

Chaines operatoires (example of meat cutting,distilled from Haidle93)

� 7 phases � 7 phases� 4 foci � 4 foci� 9 steps � 14 steps� 4 shifts in focus � 4 shifts in focus

194 Wynn et al. ARTICLE

Kanzi does not examine coresclosely. If true, this could be a differ-ence in spatial perception (ability toisolate patterns in complex back-grounds91) or directed attention.There are other equally plausibleexplanations: Perhaps Kanzi lacksanatomical and neuromotor featuresthat have evolved in hominins alongwith stone knapping or perhapsKanzi is less able to copy a complexmotor sequence. From refits, weknow that the hominin knappers atLokalalei 2C all used the same orvery similar procedures of platformidentification and core rotation,84

which suggests that these homininsmay have been more adept at copy-ing motor procedures than are Kanziand Panbanisha.In sum, the tool-based activities of

Oldowan hominins fit easily into thepattern of technology-based adapta-tions documented for modern apes.Indeed, tool-based activities appearto be an important component of theape adaptive grade, with Oldowanhominins being a derived variant.We are not arguing that there were

no differences between early homi-nins and other apes. The ape adapt-ive grade encompasses gorillas andbonobos, who demonstrate little tooluse; orangutans, who regularly makeand use tools; and chimpanzees, whoexhibit a great deal. One advantageof an ape’s perspective is that it facil-itates identification of characteristicsthat were different. As in 1989, twostand out: Oldowan hominins regu-larly carried tools farther than anyliving ape, occasionally much far-ther; early hominins were eitherdirect or indirect competitors withmembers of the carnivore guild,which living apes are not.14 Thesedifferences are not trivial. However,they do not indicate that these homi-nins were behaving in significantlyhuman-like ways. ‘‘At most one canargue that the Oldowan pushed thelimits of ape grade adaptations; itdid not exceed them.’’1:394

Implications for UnderstandingEarly Hominin Evolution

The Oldowan was not a proto-human technology. Indeed, there

was nothing uniquely or even espe-cially humanlike about it, unless onewishes to define human as the abilityto knap invasive flakes. The implica-tions of this conclusion for theunderstanding of human evolutionare significant:

The Oldowan was not a newadaptive grade, but a variation on anold one. As Wynn and McGrewobserved in the original article, thisassessment does not disparage Oldo-wan hominins. Apes had been verysuccessful and very varied for mil-lions of years before the advent offlaked stone technology, and it islikely that several were tool-users. Itis within this context that the Oldo-wan should be understood.

Human-like technical elementsmade their appearance after the Old-owan. A noncomprehensive listwould include: curated toolsintended for repeated future use;shape imposition and shared stand-ards for shape; semiotic load, or theuse of tools to communicate infor-mation about the tool user to othertool users; linguistic instruction fortool fabrication.

In a techno-behavioral sense, Homoerectus sensu lato was the intermediateform between ape and human. As anextension of the ape adaptive grade,the Oldowan can accommodate multi-ple hominin agents, including Australo-pithecus, Paranthropus, and Homo.This perspective predicts that theremay have been technological traditionsdeveloped by members of an ape cladethat produced an archeological record.Not all archeological remains must beassumed, a priori, to have been pro-duced by hominins. Indeed, distin-guishing ape from hominin archeologi-cal records may be a problem.88 Thetechnology of Pan troglodytes and othernonhuman primates offers a richsource of ideas for modeling not onlythe earliest hominin technology, butalso ape and hominin technology thatantedated 2.6 Ma.

REFERENCES1 Wynn T, McGrew WC. 1989. An ape’s view ofthe Oldowan. Man 24:283–298.

2 Schick K, Toth N. 2006. An overview of theOldowan industrial complex: the sites and thenature of their evidence. In: Toth N, Schick K,

editors. The Oldowan: case studies into the ear-liest Stone Age. Gosport IN: Stone Age InstitutePress. p 3–42.

3 Matsuzawa T. 1994. Field experiments on theuse of stone tools by chimpanzees in the wild.In: Wrangham RW, McGrew WC, De WaalFBM, Heltne PG, editors. Chimpanzee cultures.Cambridge: Harvard University Press. p 351–370.

4 Matsuzawa T, Humle T, Sugiyama Y, editors.2011. The chimpanzees of Bossou and Nimba.Tokyo: Springer.

5 Goodall J. 1986. The chimpanzees of Gombe:patterns of behavior. Cambridge MA: BelknapPress.

6 Lonsdorf EV. 2006. What is the role of moth-ers in the acquisition of termite-fishing behav-iors in wild chimpanzees (Pan troglodytesschweinfurthii)? Anim Cogn 9:36–46.

7 Nishida T, editor. 1990. The chimpanzees ofthe Mahale Mountains: sexual and life historystrategies. Tokyo: University of Tokyo Press.

8 Nishida T, Zamma K, Matsusaka T, Inaba A,McGrew WC. 2010. Chimpanzee behavior inthe wild. An audio-visual encyclopedia, Tokyo:Springer.

9 Boesch C, Boesch H. 1984. Mental map inwild chimpanzees: an analysis of hammer trans-ports for nut-cracking. Primates 25:160–170.

10 Boesch C, Boesch-Achermann H. 2000. Thechimpanzees of the Tai Forest: behavioral ecol-ogy and evolution. Oxford: Oxford UniversityPress.

11 Reynolds V. 2005. The chimpanzees of theBudongo forest: ecology, behaviour, and con-servation. Oxford: Oxford University Press.

12 Hashimoto C, Furuichi T, Tashiro Y. 2000.Ant dipping and meat eating by wild chimpan-zees in the Kalinzu forest, Uganda. Primates41:103–108.

13 Kahlenberg SM, Emery Thompson M,Wrangham RW. 2008. Female competition overcore areas in Pan troglodytes schweinfurthii,Kibale National Park, Uganda. Int J Primatol29:931–947.

14 Watts DP. 2008. Scavenging by chimpanzeesat Ngogo and the relevance of chimpanzee scav-enging to early hominin behavioral ecology.J Hum Evol 54:125–133.

15 Mitani J, Watts D, Amsler S. 2010. Lethalintergroup aggression leads to territorial expan-sion in wild chimpanzees. Curr Biol 20:507–508.

16 Pruetz J. 2007. Evidence of cave use by sa-vanna chimpanzees (Pan troglodytes verus) atFongoli, Senegal: implications for thermoregu-latory behavior. Primates 48:316–319.

17 Pruetz J, Bertolani P. 2009. Chimpanzee(Pan troglodytes verus) behavioral responses tostresses associated with living in a savanna-mosaic enviornment: implications for homininadaptations to open habitats. PaleoAnthropol-ogy 2009:252–262.

18 Fowler A, Sommer V. 2007. Subsistencetechnology of Nigerian chimpanzees. Int J Pri-matol 28:997–1023.

19 Sanz C, Morgan D. 2007. Chimpanzee tooltechnology in the Goualougo Triangle, Republicof Congo. J Hum Evol 52:420–433.

20 Sanz C, Morgan D. 2009. Flexible and per-sistent tool-using strategies in honey-gatheringby wild chimpanzees. Int J Primatol 30:411–427.

21 Sanz C, Morgan D, Gulick S. 2004.New insights into chimpanzees, tools, andtermites from the Congo Basin. Am Nat 164:567–581.

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 195

22 Sanz C, Schoening C, Morgan D. 2010.Chimpanzees prey on army ants with special-ized tool set. Am J Primatol 72:17–24.

23 Hunt K, McGrew WC. 2002. Chimpanzeesin the dry habitats at Assirik, Senegal, andat Semliki Wildlife Reserve, Uganda. In: BoeschC, Hohmann G, Marchant L, editors. Behaviou-ral diversity in chimpanzees and bonobos.Cambridge: Cambridge University Press. p 35–51.

24 Mitani J, Watts DP, Muller M. 2002. Recentdevelopments in the study of wild chimpanzeebehavior. Evol Anthropol 11:9–25.

25 Lycett SJ, Collard M, McGrew WC. 2010.Are behavioral differences among wild chim-panzee communities genetic or cultural? Anassessment using tool-use data and phyloge-netic methods. Am J Phys Anthropol 142:461–467.

26 Whiten A, Goodall J, McGrew WC. 1999.Cultures in chimpanzees. Nature 399:682–685.

27 Hohmann G, Fruth B. 2003. Culture inbonobos? Between-species and within-speciesvariation in behavior. Curr Anthropol 44:563–571.

28 Ingmanson EJ. 1996. Tool-using behavior inwild Pan paniscus: social and ecological consid-erations. In: Russon A, Bard K, Parker S,editors. Reaching into thought: the minds ofgreat apes. Cambridge: Cambridge UniversityPress. p 190–210.

29 Surbeck M, Hohmann G. 2008. Primatehunting by bonobos at LuiKotale, SalongaNational Park. Curr Biol 18:R906–907.

30 Doran-Sheehy D, Conklin-Brittain N, Lod-wick J, Mongo P. 2009. Male and female west-ern gorilla diet: preferred foods, use of fallbackresources, and implications for ape versus OldWorld monkey foraging strategies. Am J PhysAnthropol 140:727–738.

31 Galdikas BMF. 1982. Orang-utan tool-use atTanjung Puting Reserve, Central IndonesianBorneo (Kalimantan Tengah) J Hum Evol10:19–33.

32 Fox EA, Sitompul AF, Van Schaik C. 1999.Intelligent tool use in wild Sumatran orangu-tans. In: Parker ST, Mitchell RW, Miles HL,editors. The mentality of gorillas and orangu-tans: comparative perspectives. Cambridge:Cambridge University Press. p 99–116.

33 van Schaik C, Ancrenaz M, Borgen G, Galdi-kas BMF, Knott CD, Singleton I, Suzuki A,Utami SS, Merill M. 2003. Orangutan culturesand the evolution of material culture. Science299:102–105.

34 van Schaik C, Fox EA, Sitompul AF. 1996.Manufacture and use of tools in wild Sumatranorangutans: implications for human evolution.Naturwissenschaften 83:186–188.

35 Whiten A, Horner V, de Waal FBM. 2005.Conformity to cultural norms of tool use inchimpanzees. Nature 437:737–740.

36 Tomasello M, Call J. 1997. Primate cogni-tion. New York: Oxford University Press.

37 Hare B, Kwetuenda S. 2010. Bonobos volun-tarily share their own food with others. CurrBiol 20:R230–231.

38 Pruetz J, Bertolani P. 2007. Savanna chim-panzees, Pan troglodytes verus, hunt with tools.Curr Biol 17:412–417.

39 Hernandez-Aguilar RA, Moore J, PickeringTR. 2007. Savanna chimpanzees use tools toharvest the underground storage organs ofplants. Proc Natl Acad Sci USA 104:19210–19213.

40 Hockings K, Anderson J, Matsuzawa T.2009. Flexible feeding on cultivated under-ground storage organs by rainforest-dwelling

chimpanzees at Bossou, West Africa. J HumEvol 71:636–646.

41 Wrangham RW, Cheney D, Seyfarth R,Sarmiento E. 2009. Shallow-water habitats assources of fallback foods for hominins. Am JPhys Anthropol 140:630–642.

42 Koops K, McGrew WC, Matsuzawa T. 2010.Do chimpanzees (Pan troglodytes) use cleaversand anvils to fracture Treculia africana fruits?Preliminary data on a new form of percussivetechnology. Primates 51:175–178.

43 McGrew WC. 2010. Chimpanzee technology.Science 328:579–580.

44 McGrew WC. 2010. In search of the lastcommon ancestor: new findings on wild chim-panzees. Philos Trans R Soc London, B365:3267–3276.

45 Brewer SM, McGrew WC. 1990. Chimpan-zees use of a tool-set to get honey. Folia Prima-tol 54:100–104.

46 Boesch C, Head J, Robbins MM. 2009.Complex tool sets for honey extraction amongchimpanzees in Loango National Park, Gabon.J Hum Evol 56:560–569.

47 Deblauwe I, Guislain P, Dupain J, vanElsacker L. 2006. Use of a tool-set by Pan trog-lodytes troglodytes to obtain termites (Macro-termes) in the periphery of the Dja BiosphereReserve, southeast Cameroon. Am J Primatol68:1191–1196.

48 Sugiyama Y. 1997. Social tradition and theuse of tool-composites by wild chimpanzees.Evol Anthropol 6:23–27.

49 McGrew WC. 1974. Tool use by wild chim-panzees in feeding upon driver ants. J HumEvol 3:501–508.

50 Carvalho S, Biro D, McGrew WC, Matsu-zawa T. 2009. Tool-composite reuse in wildchimpanzees (Pan troglodytes): archaeologicallyinvisible steps in the technological evolutionof early hominins? Anim Cogn 12(suppl 1):103–114.

51 Carvalho S, Cunha E, Sousa C, MatsuzawaT. 2008. Chaines operatoires and resourceexploitation strategies in chimpanzee nut-crack-ing (Pan troglodytes). J Hum Evol 55:148–163.

52 Sousa C, Biro D, Matsuzawa T. 2009. Leaf-tool use for drinking water by wild chimpan-zees (Pan troglodytes): acquisition patterns andhandedness. Anim Cogn 12(suppl 1):115–125.

53 Mercader J, Panger M, Boesch C. 2002.Excavation of a chimpanzee stone tool site inthe African rainforest. Science 296:1452–1455.

54 Mercader J, Barton H, Gillespie J, Harris J,Kuhn S, Tyler R, Boesch C. 2007. 4,300-year-old chimpanzee sites and the origins of percus-sive stone tecnology. Proc Natl Acad Sci USA104:3043–3048.

55 Fox EA, bin’Muhammad I. 2002. Brief com-munication: new tool use by wild Sumatranorangutans (Pongo pygmaeus). Am J PhysAnthropol 119:186–188).

56 van Schaik C, van Noordwijk M, Wich S.2006. Innovation in wild Bornean orangutans(Pongo pygmaeus wurmbii). Behaviour 147:839–876.

57 Breuer T, Ndoundou-Hockemba M, FishlockV. 2005. First observation of tool use in wildgorillas. PLoS Biol 3:e380.

58 Toth N, Schick K, Savage-Rumbaugh S, Sev-cik R, Rumbaugh D. 1993. Pan the tool-maker:investigations into the stone tool-making andtool-using capabilties of a bonobo (Pan panis-cus). J Archaeol Sci 20:81–91.

59 Savage-Rumbaugh S, Fields W. 2006. Rulesand tools: beyond anthropomorphism. In: TothN, Schick K, editors. The Oldowan: case studies

into the earliest Stone Age. Gosport, IN: StoneAge Institute Press. p 223–242.

60 Savage-Rumbaugh S, Fields W, Segerdahl P,Rumbaugh D. 2005. Culture prefigures cognitionin Pan/Homo bonobos. Theoria 20:311–328.

61 Savage-Rumbaugh S, Toth N, Schick K.2007. Kanzi learns to knap stone tools. In:Washburn D, editor. Primate perspectives onbehavior and cognition. Washington, DC: Amer-ican Psychological Association. p 279–291.

62 Schick K, Toth N, Garufi G, Savage-Rumbaugh E, Rumbaugh D, Sevcik R. 1999.Continuing investigations into the stone tool-making capabilities of a bonobo (Pan paniscus).J Archaeol Sci 26:821–832.

63 Whiten A, Schick K, Toth N. 2009. The evolu-tion and cultural transmission of percussive tech-nology: integrating evidence from palaeoanthrool-ogy and primatology. J Hum Evol 57:420–435.

64 Toth N, Schick K, Semaw S. 2006. A com-parative study of the stone tool-making skills ofPan, Australopithecus, and Homo sapiens. In:Toth N, Schick K, editors. The Oldowan: casestudies into the earliest Stone Age. Gosport, IN:Stone Age Institute Press. p 155–222.

65 Toth N, Schick K. 2009. The Oldowan: Thetool making of early hominins and chimpan-zees compared. Ann Rev Anthropol 38:289–305.

66 Byrne R. 2007. Culture in great apes: usingintricate complexity in feeding skills to trace theevolutionary origin of human technical prowess.Philos Trans R Soc London, B 362: 577–585.

67 Sanz C, Call J, Morgan D. 2009. Design-com-plexity in termite fishing tools of chimpanzees(Pan troglodytes). Biol Lett 5:293–296.

68 Schick K, Toth N, editors. 2009. The cuttingedge: new approaches to the archaeology ofhuman origins. Gosport, IN: Stone Age InstitutePress.

69 Semaw S. 2000. The world’s oldest stoneartefacts from Gona, Ethiopia: their implica-tions for understanding stone technology andpatterns of human evolution between 2.6–1.5million years ago. J Archaeol Sci 27:1197–1214.

70 Semaw S, Rogers M, Quade J, Renne P, But-ler R, Dominguez-Rodrigo M, Stout D, Hart W,Pickering T, Simpson S. 2003. 2.6-million-year-old stone tools and associated bones from OGS-6 and OGS-7, Gona, Afar, Ethiopia. J Hum Evol45:169–177.

71 Semaw S, Rogers M, Stout D. 2009. Insightsinto late Pliocene lithic assemblage variability:the East Gona and Ounda Gona South Oldo-wan archaeology (2.6 million years ago), Afar,Ethiopia. In: Schick K, Toth N, editors. Thecutting edge: new approaches to the archaeol-ogy of human origins. Gosport, IN: Stone AgeInstitute Press. p 211–246.

72 McPherron A, Alemseged A, Marean C,Wynn J, Reed D, Geraads D, Bobe R, Bearat H.2010. Evidence for stone-tool-assisted consump-tion of animal tissues before 3.39 million yearsago at Dikika, Ethiopia. Nature 466:857–560.

73 Dominguez-Rodrigo M, Pickering T, BunnH. 2011. Configurational approach to identify-ing the earliest hominin butchers. Proc NatlAcad Sci USA 107:20929–20934.

74 de la Torre I, Mora R, Dominguez-RodrigoM, de Luque L, Alcala L. 2003. The Oldowanindustry of Peninj and its bearing on thereconstruction of the technological skill ofLower Pleistocene hominids. J Hum Evol44:203–224.

75 Toth N, Schick K, editors. 2006. The Oldo-wan: case studies into the earliest Stone Age.Gosport, IN: Stone Age Institute Press.

76 Blumenschine R, Pobiner B. 2007. Zooarch-aeology and the ecology of Oldowan hominin

196 Wynn et al. ARTICLE

carnivory. In: Ungar P, editor. Evolution of thehuman diet: the known, the unknown, and theunknowable. Oxford: Oxford University Press. p167–190.

77 Blumenschine R, Prassack K, Kreger CD,Pante M. 2007. Carnivore tooth-marks, micro-bial bierosion, and the invalidation of Domi-nguez-Rodrigo and Barba’s (2006) test of Oldo-wan hominin scavenging behavior. J Hum Evol53:420–426.

78 Bunn H, Kroll E. 1986. Systematic butcheryby Plio/Pleistocene hominids at Olduvai Gorge,Tanzania. Curr Anthropol 27:431–452.

79 Dominguez-Rodrigo M, Barba R. 2006. Newestimates of tooth mark and percussion markfrequencies at the FLK Zinj site: the carnivore-hominid-carnivore hypothesis falsified. J HumEvol 50:170–194.

80 Leakey M. 1971. Olduvai Gorge Vol. 3. Cam-bridge: Cambridge University Press.

81 Mora R, de la Torre I. 2005. Percussion toolsin Olduvai Beds I and II (Tanzania): implica-tions for early human activities. J AnthropolArchaeol 24:179–192.

82 Braun D, Harris J, Maina D. 2009. Oldowanraw material procurement and use: evidencefrom the Koobi Fora formation. Archaeometry51:26–42.

83 Braun D, Harris JWK. 2009. Plio-Pleistocenetechnological variation: a view from the KBSMbr., Koobi Fora Formation. In: Schick K,Toth N, editors. The cutting edge: newapproaches to the archaeology of human ori-gins. Gosport, IN: Stone Age Institute Press. p17–32.

84 Delagnes A, Roche H. 2005. Late Pliocenehominid knapping skills: the case of Lokalalei2C, West Turkana, Kenya. J Hum Evol 48:435–472.

85 Roche H, Blumenschine R, Shea J. 2009.Origins and adaptations of early Homo: whatarchaeology tells us. In: Grine F, Fleagle J, Lea-key R, editors. The first humans - origin andearly evolution of the genus Homo: Springer. p135–147.

86 Braun D, Plummer T, Ditchfield P, FerraroJ, Maina D, Bishop L, Potts R. 2008. Oldowanbehavior and raw material transport: perspec-tives from the Kanjera Formation. J ArchaeolSci 35:2329–2345.

87 Braun D, Plummer T, Ferraro J, DitchfieldP, Bishop L. 2008. Raw material quality andOldowan hominin toolstone preferences: evi-dence from Kanjera South, Kenya. J ArchaeolSci 36:1605–1614.

88 Haslam M, Hernandez-Aguilar RA, Ling V,Carvalho S, de la Torre I, McGrew WC, Mer-cader J, Mora R, Petraglia M, Roche H andothers. 2009. Primate archaeology. Nature 460:334–339.

89 Bunn H, Mabulla A, Dominguez-RodrigoM, Ashley G, Barba R, Diez-Martin F, RemerK, Yravedra J, Baquedano E. 2010. WasFLK North level 1–2 a classic ‘‘living floor’’of Oldowan hominins or a taphonmicallycomplex palimpsest dominated by large car-nivore feeding behavior? Quaternary Res74:355–362.

90 Hovers E. 2009. Learning from mistakes:flaking accidents and knapping skills in theassemblage of A.L. 894. In: Schick K, Toth N,editors. The cutting edge: new approaches tothe archaeology of human origins. Gosport, IN:Stone Age Institute Press. p 137–150.

91 Wynn T. 2010. The evolution of humanspatial cognition. In: Dolins F, editor. Spatialperception, spatial cognition. Cambridge: Cam-bridge University Press. p 213–236.

92 Byrne R, Corp N, Byrne J. 2001. Estimatingthe complexity of animal behaviour: howmountain gorillas eat thistles. Behaviour 138:525–557.

93 Haidle M. 2010. Working-memory capacityand the evolution of modern cogntive potential.Curr Anthropol 51(suppl 1):S149–S166.

94 Baddeley A. 2007. Working memory, thought,and action. Oxford: Oxford University Press.

95 Normand E, Boesch C. 2009. SophisticatedEuclidean maps in forest chimpanzees. AnimBehav 77:1195–1201.

96 McGrew WC. 1992. Chimpanzee materialculture: implications for human evolution.Cambridge: Cambridge University Press.

97 Whiten A, Goodall J, McGrew WC, Nishida T,Reynolds T, Sugiyama Y, Tutin CEG, WranghamRW, Boesch C. 2001. Charting culturalvariation in chimpanzees. Behaviour 138:1481–1516.

98 Schoening C, Humle T, Moebius Y, McGrewWC. 2008. The nature of culture: technologicalvariation in chimpanzee predation on armyants revisited. J Hum Evol 55:48–59.

99 Yamakoshi G, Sugiyama Y. 1995. Pestle-pounding behavior of wild chimpamzees atBossou, Guinea: a newly observed tool-usingbehavior. Primates 36:489–500.

100 Marchant L, McGrew WC. 2005. Percussivetechnology: chimpanzee baobab smashing andthe evolutionary modeling of hominid knapping.In: Rous V, Bril B, editors. Knapping stones: thenecessary conditions for a uniquely humanbehaviour. Cambridge: MacDonald Institute forArchaeological Research. p 341–350.

101 Stewart F, Piel A, McGrew WC. n.d. Livingarchaeology: artifacts of specific nest site fidel-ity in chimpanzees. J Hum Evol. In press.

102 Hernandez-Aguilar RA. 2009. Chimpanzeenest distribution and site reuse in a dry habitat:implications for early hominin ranging. J HumEvol 57:350–364.

103 Sept J. 1992. Was there no place likehome? A perspective on early hominid archaeo-logical sites from the mapping of chimpanzeenests. Curr Anthropol 33:187–207.

104 McGrew WC. 2004. The cultured chimpan-zee: reflections on cultural primatology. Cam-bridge: Cambridge University Press.

105 Potts R. 1988. Early hominid activities atOlduvai. New York: Aldine.

106 Thierry B. 2007. Unity in diversity: lessonsfrom macaque societies. Evol Anthropol16:224–238.

107 Watanabe K. 2001. A review of 50 years ofresearch on the Japanese monkeys of Koshima:status and dominance. In: Matsuzawa T, editor.

Primate origins of human cognition and behav-ior. Tokyo: Springer. p 405–417.

108 Watanabe K, Urasopon N, MalaivijitnondS. 2007. Long-tailed macaques use human hairas dental floss. Am J Primatol 69:940–944.

109 Masataka N, Koda H, Urasopon N, Wata-nabe K. 2009. Free-ranging macaque mothersexaggerate tool-using behavior when observedby offspring. PLoS ONE 4:e4768.

110 Huffman MA. 1984. Stone-play of Macacafuscata in Arahsiyama B troop: transmissionof non-adaptive behavior. J Hum Evol 13:725–735.

111 Leca JB, Gunst N, Huffman MA. 2007. Jap-anese macaque cultures: inter- and intra-troopvariability of stone handling patterns across 10troops. Behaviour 144:251–281.

112 Leca JB, Gunst N, Huffman MA. 2008.Food provisioning and stone handling in Japa-nese macaques: a comparative study of tentroops. Am J Primatol 70:803–813.

113 Gumert MD, Gluck M, Malaivijitnond S.2009. The physical characteristics and usagepatterns of stone axe and pounding hammersused by long-tailed macaques in the AndamanSea region of Thailand. Am J Primatol 71:594–608.

114 Malaivijitnond S, Lekprayoon C, Tandava-nittj N, Panha S, C C, Hamada Y. 2001. Stone-tool use by Thai long-tailed macaques (Macacafascicularis). Am J Primatol 69:227–233.

115 Fragaszy DM, Visalberghi E, Fedigan LM.2004. The complete capuchin: the biology ofthe genus Cebus. Cambridge: Cambridge Uni-versity Press.

116 Perry S, Manson JH. 2008. Manipulativemonkeys: the capuchins of Lomas Barbudal.Cambridge, MA: Harvard University Press.

117 Ottoni EB, Izar P. 2008. Capuchin monkeytool use: overview and implications. EvolAnthropol 17:171–178.

118 Visalberghi E, Fragaszy DM, Ottoni EB,Izar P, de Olivera MG, Andrade FRD. 2007.Characteristics of hammer stones and anvilsused by wild bearded capuchin monkeys (Cebuslibidinosus) to crack open palm nuts. Am J PhysAnthropol 132:426–444.

119 Fragaszy DM, Greenberg R, Visalberghi E,Ottoni EB, Izar P, Qing L. 2010. How wildbearded capuchin monkeys select stones andnuts to minimize the number of strikes per nutcracked. Anim Behav 80:205–214.

120 Mannu M, Ottoni EB. 2009. The enhancedtool-kit of two groups of wild bearded capuchinmonkeys in the Caatinga: tool making, associa-tive use, and secondary tools. Am J Primatol71:36–45.

121 Moura ACA. 2007. Stone banging by wildcapuchin monkeys: an unusual auditory dis-play. Folia Primatol 78:36–45.

122 Stout D, Semaw S, Rogers M, Cauche D.2010. Technological variation in the earliestOldowan from Gona, Afar, Ethiopia. J HumEvol 58:474–491.

VVC 2011 Wiley-Liss, Inc.

ARTICLE ‘‘An Ape’s View of the Oldowan’’ Revisited 197

The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.