Evolutionary 235

ARTICLES

Older Than the ? Rethinking the Emergence of Hominin Use

MELISSA A. PANGER, ALISON S. BROOKS, BRIAN G. RICHMOND, AND BERNARD WOOD

Intentionally modified stone first appear in the hominin archeological record ized occurrences often containing about 2.5 mya. Their appearance has been variously interpreted as marking the onset hundreds or even thousands of stone of tool use by hominins or, less restrictively, the origin of hominin .1,2 artifacts and associated fossil speci- Although the 2.5 mya date has persisted for two decades, several related but distinct mens.5 Low-density sites with fewer questions about the origin and evolution of hominin tool use remain to be answered: than 100 stone artifacts, referred to as Did hominins use tools before 2.5 mya? Did hominins use unmodified stones as tools scatters, have also been described.6 before 2.5 mya? Does the earliest appearance of stone tools in the archeological The artifacts range from -made record represent the earliest use of intentionally modified stones as tools? flakes with clear striking platforms, bulbs of percussion, and dorsal scars indicative of multiple prior flake re- Using information from primatol- stone tools (Fig. 1). We consider sev- movals from the core7 to amorphous ogy, functional morphology, phylog- eral scenarios to explain why stone lumps and shattered fragments of ex- eny, archeology, and paleoanthropol- tool manufacture and use might not otic raw materials.8 ogy, we argue that before 2.5 mya have left archeological traces prior to Hard-hammer percussion, the most hominins may have used tools, includ- 2.5 mya and conclude by suggesting frequent technique used to manufac- ing unmodified and possibly modified means to test our hypotheses. ture Oldowan-type tools, apparently involved striking a hand-held hammer THE EARLIEST KNOWN DIRECT stone against a hand-held core.9 How- ever, it is likely that some Oldowan EVIDENCE OF HOMININ Melissa Panger is a Postdoctoral Research tools were made by placing a core Fellow in the Department of Anthropology TOOL USE and the Center for the Advanced Study of against a substrate and then striking it Paleobiology at The George Wash- Research on the evolution of tool- with another stone or by striking or ington University. E-mail: [email protected] throwing a stone against a hard sur- Alison Brooks is Professor and Chair of based behavior in hominins focuses 9 the Department of Anthropology at The on modified stone tools and fossilized face. Microwear analyses suggest George Washington University and Pro- faunal remains showing wear consis- that some Oldowan tools were used to fessor, Center for the Advanced Study of Human Paleobiology. Research Associ- tent with tool use. Here, tool use is process large vertebrates by cutting ate in Anthropology, National Museum of defined according to Beck3: “The ex- meat from bone or, using percussion Natural History, the Smithsonian Institu- ternal employment of an unattached techniques, to gain access to bone tion. E-mail: [email protected] 10,11 Brian Richmond is an Assistant Professor environmental object to alter...an- marrow. Other tools may have in the Department of Anthropology and the other object...when the user holds been used for digging9 or for cutting Center for the Advanced Study of Human wood and siliceous plant tissues.11,12 Paleobiology at The George Washington or carries the tool during or just prior University and Research Associate of the touse...”(p. 10). “Modified” stone Oldowan tools have been described in Human Origins Program at the National tools are those that have been inten- detail elsewhere.13,14 Museum of Natural History, the Smithso- nian Institution. E-mail: [email protected] tionally modified. “Hominin” refers to When stone tools are used to pro- Bernard Wood is the Henry R. Luce Pro- members of the human clade or, in cess large vertebrates, they often leave fessor of Human Origins in the Depart- other words, components of the lin- distinctive patterns of damage on the ment of Anthropology at The George 15,16 Washington University and Adjunct Se- eage that separated from the last com- fossilized remains. Therefore, fau- nior Scientist at the National Museum of mon ancestor of and spp. 8 nal remains showing modifications Natural History, the Smithsonian Institu- tion. E-mail: [email protected] to 5 mya, and that persist as Homo inconsistent with natural breakage sapiens.4 but consistent with use pro- The earliest known tools are modi- vide indirect evidence of stone tool Key words: stone tools, tool use, homi- fied stone tools belonging to the Old- use.17 Currently, the earliest such evi- nid evolution owan and dating to about 2.5 dence also dates to about 2.5 mya.17 to 2.0 mya (Table 1). Most Oldowan Bone tools dating to 2 to 1.5 mya Evolutionary Anthropology 11:235–245 (2002) tools, however, have been recovered have also been found in both East and DOI 10.1002/evan.10094 14,18 Published online in Wiley InterScience from sites dating from 2 to 1.5 mya, Southern . Although the (www.interscience.wiley.com). and are known primarily from local- bones are not extensively modified, 236 Evolutionary Anthropology ARTICLES

idence. Potentially important indirect evidence comes from primatology, functional anatomy, and molecular phylogenetics.

DID HOMININS USE TOOLS BEFORE 2.5 MYA? There is now compelling evidence that the hominins and Pan are sister clades24,25 (but see Satla, Klein, and Takahata,26 for a more cautious but still supportive interpretation of the data). Because the great apes (espe- cially ) and use tools, parsimony suggests that the last common ancestor of Pan and Homo (8 to 5 mya) used tools27–30 (see Box 1 and Fig. 2). Because most early hominins are associated with nonarid environ- ments such as riparian forests, densely wooded habitats, and mosaic habitats with some grasslands,31,32 we Figure 1. Estimations of the earliest hominin tool use using various lines of evidence. This assume that such high-energy, diffi- figure represents the estimated time of the earliest hominin tool use using three different sources of evidence. The Direct Evidence of tool use is represented by accumulated cult-to-access foods such as nuts, so- modified stone tools, bone tools, and percussion or cut marks on bones that survive to the cial insects, and honey were available. present in the fossil or archeological record, the earliest known dating to about 2.5 mya We further assume that raw materials ( A). Arrow B represents anatomical traits (such as a relatively large brain and for making tools, among them sticks, manipulative hands) from the hominin fossil record that are consistent with, correlated with, grasses, leaves, and appropriate stones, or necessary for primate tool use, including stone-tool use (modified and unmodified). The were also available to all early homi- earliest known hominins (about 3.2 mya) for which we have appropriate anatomical data nins. There is little doubt that homi- have an anatomy consistent with tool use, possibly including modified stone-tool use. Because anatomical evidence necessary for determining relative brain size and hand nins, given a body mass comparable morphology are currently not available for earlier hominins, this date may change as more to or greater than that of chimpan- information becomes available. Arrow C represents the estimated time of the earliest zees, had the strength to perform a hominin tool use (comparable to that observed among great apes, including unmodified variety of tool-using activities, includ- stone-tool use) derived using phylogenetic inference (see Fig. 2). The related event is the ing the manufacture and use of stone split between the panins and hominins; that is, the origin of the hominin clade at about 8 to tools. (Capuchins, which, at 2 to 3 kg, 5 mya. Thus, 5 mya is a conservative estimate. are about the size of a house cat, are macroscopic use-wear patterns and ies,22 and archeological and ethno- able to make stone tools). Thus, we microscopic analyses suggest that the graphic studies of symbolic behav- assume that the ecological impetus, bones were most likely used for dig- ior.23 We suggest that, in much the raw materials, and necessary strength for tool use were available to early ging, cutting, or termite fishing.18,19 same way, researchers interested in Additionally, some early modified the evolution of tool use benefitby hominins from the time of the split looking beyond traditional lines of ev- stone tools exhibit wear patterns typ- between hominins and panins. ical of , suggesting that TABLE 1. Early Oldowan Tool Sites (Older Than 2 My) wooden tools may also have been part of the early hominin toolkit.12 Site or Locality Name Region, Country Age References Although it is practical to use early Gona Hadar, 2.6–2.5 My 50 Oldowan tools and fossilized faunal Bouria , Ethiopia c. 2.5 My 17 remains having wear consistent with Valley, 2.4–2.34 8 tool use as the earliest indication of (Members E and F) Ethiopia My Lokalalei West , 2.34 My 7 hominin tool use we might be better served by turning to more indirect Senga 5Ab Semliki Valley, c. 2.3 My 105 methods. Alternative sources of evi- Democratic Republic dence about the of Congo are routinely explored, for example Kada Hadar Member Hadar, Ethiopia c. 2.33 My 36 through information regarding vocal KS-1 and KS-2 Kanjera South, Kenya c. 2.2 My 32 and neural anatomy that is consistent a Faunal remains with cut marks, but no stone tools, were recovered from this site. with language,21 ape-language stud- b The date and associations for this site are not secure.13 ARTICLES Evolutionary Anthropology 237

Box 1. Nonhuman Primate Tool Use

Tool use has been observed in a wide variety of animal taxa (see Beck3 for a review), but here we focus on tool use by nonhuman here for two specific reasons. First, their close relationship to modern humans makes their tool use particularly relevant to the evolution of tool use by hominins. Sec- ond, the type of tool use found among most animals is context-specific and is represented by stereotyped behaviors that show little variation among individ- uals and populations. “Intelligent” tool use, exhibited by some primates, refers to a more flexible tool-use repertoire29 and is the type of tool use argued to have important implications for homi- nin evolution. Tool manufacture or use by a vari- ety of primates has been reported, but habitual tool use is largely limited to free-ranging and captive Pongo Figure 1. Raw material of tools commonly used by free-ranging chimpanzees. These schematic drawings of materials that chimpanzees commonly use as tools are not to pygmaeus, Pan troglodytes, and cap- scale and do not necessarily represent the exact species or type actually used: a. tive Pan paniscus, Gorilla gorilla, and Grass, used as probing tools, may be intentionally modified before use; b. Leaves, when capuchin monkeys (Cebus spp.). chewed or folded, are used as leaf sponges or “napkins”; c. Sticks, used as tools for “Habitual” refers to tool-use events probing for social insects, honey, and such, may be intentionally modified before use; that are repeated by several individu- d. Stone, used as hammer or anvil to open hard-shelled nuts, may be unintentionally als in a population over time.67 Chim- modified during use. panzees (Pan troglodytes)inall known populations spontaneously and use hammers, probes, and albifrons.85 As with chimpanzees28,63 and habitually manufacture and use scrapers made of sticks.72,73 Gorillas and ,72,73 complex forag- tools. Some of the most common (Gorilla gorilla) and bonobos (Pan pa- ing behaviors, including tool use, are types of tool use observed in free- niscus) have yet to be seen habitually known to vary across capuchin pop- ranging chimpanzees are the manu- manufacturing or using tools in the ulations.62 facture and use of probing tools wild (but see Ingmanson74). However, Several captive primates have made of sticks, bark, grass, or other all of the great apes, including oran- been taught to make modified stone vegetation68 and the manufacture gutans,75 gorillas,76 chimpanzees,77 tools, among them Cebus apella,86 and use of leaf sponges69 (Fig. 1). (For and bonobos,78 readily use tools in Pan paniscus,41,87 and Pongo pyg- review, see McGrew28 and Whiten captivity. maeus.88 In the Pan and Cebus stud- and co-workers63). Several West Afri- Capuchins (Cebus spp.), medium- ies, although the animals were only can populations also use sized New World monkeys, are the shown hard-hammer percussion stone and wooden hammers and an- only nonhominoid primates known to techniques, the primates not only ac- vils to access a variety of nuts,27,69–71 use tools habitually. Studies have quired these techniques, but also de- but chimpanzees have never been shown that captive capuchins manu- veloped innovative techniques involv- observed intentionally modifying facture and use “ape-equivalent” ing pounding and throwing cores stone tools in the wild. Recent re- types of tools.67,79 Currently, wild ca- against hard substrates to produce search has demonstrated that chim- puchins are known to use tools only flakes. To date, no stone-tool-making panzee stone-tool use leaves a sig- rarely. The few such observations nonhuman primate has attained the nal, in the form of stone and plant that have been made include, for ex- skill to produce the acute-edged bi- remains, that is recoverable in the ar- ample, use of a wooden club by C. facial and polyhedral cores associ- cheological record.38 capucinus,80,81 wooden probing tools ated with many Oldowan sites87 or Some free-ranging and “leaf-wrapping” by C. capuci- the multiple sequential flake removals (Pongo pygmaeus) populations habit- nus,62,82 hammers and anvils by C. seen even in some of the earliest ually and spontaneously manufacture apella,83,84 and leaf containers by C. known Oldowan assemblages.7 238 Evolutionary Anthropology ARTICLES

hominins are almost certainly sister clades, it is probable that the common ancestor of chimpanzees and humans used stones as tools, though they may not have intentionally modified them (Fig. 2). Furthermore, given that the use of stone tools as hammers and anvils is within the realm of free-rang- ing chimpanzees and capuchins, the arguments regarding the morphologi- cal requirements necessary for un- modified stone tool use are similar to those in the preceding section. Most notably, by at least 3.2 mya, A. afaren- sis appears to have been capable of manipulative skills that exceeded those of extant apes. In addition, the hominin brain was comparable to that of chimpanzees in morphology and relative size, suggesting that early hominins were sufficiently intelligent to use stones as hammers and anvils. Unlike tools made of more ephem- Figure 2. Abbreviated primate phylogeny. Living primate genera that habitually use tools eral material like sticks and leaves, ϭ ϭ are in bold (see Box 1). NWM New World monkeys, OWM Old World monkeys. The stone tools are likely to be preserved logos are placed where the principle of parsimony suggests that tool use most likely evolved in the archeological record. The re- in primates. The question mark next to the axe logo implies that the data supporting placement of the axe are suggestive but not as compelling as the data supporting cent excavation of a chimpanzee placement of the other axe logos. stone-tool site in West Africa provides one model for how an unintentionally modified stone-tool assemblage might appear in the paleo-archeological Data from fossilized cranial re- (from the AL 333 site) attributed to A. record.38 The chimpanzee assemblage mains and endocasts indicate that afarensis date to about 3.2 mya and shows some similarities to a few early hominins living before the evolution show morphological features consis- Oldowan sites, for example in terms of of sensu stricto, perhaps tent with greater manipulative ability stone piece density. However, the lack as early as 3.2 mya, had relative brain than modern apes have (Box 2). Based of cores and the high proportion of sizes as large as or larger than those of on this and the fact that all hominoids stone shatter smaller than 20 mm in modern great apes.33–35 Australopithe- have hands capable of fine manipula- the chimpanzee assemblage suggests cus africanus, dating to about 2.75 mya, tion, including tool use, it is logical to that such sites may not be easily rec- exhibited encephalization greater than assume that the earliest hominins also ognizable in the archeological record. that in modern apes (specimens Sts 17 possessed hands capable of tool use. However, the chimpanzee assemblage and Sts 5),34 while A. afarensis, dating At this time, the fossil evidence does include a few flakes with sharp to about 3.2 mya,36 had brains compa- alone does not lead to a definitive con- edges. Because stone hammer use can rable in size to those of Pan (specimen clusion on the absence or presence of unintentionally create sharp-edged AL 162-28).33 The limited information tool use among our early ancestors, stones, it is possible that this type of available about brain morphology indi- but it does imply that hominin -tool use preceded the inten- cates that the brains of early hominins use was possible by at least 3.2 mya. tional production of stone flakes. As resembled those of extant hominoids in The combination of primatological, Sigaut39 noted, Pitt-Rivers proposed a many ways and had, as well, some sim- phylogenetic, and fossil evidence sug- similar connection 125 ago. In ilarities to the brains of modern hu- gests that hominins living before 2.5 1875, Pitt-Rivers, as quoted by Si- mans.37 At present, data on brain size mya almost certainly had the capacity gaut39 extrapolated from early reports and morphology are not available from to use tools (Fig. 1). of chimpanzees using unmodified the hominin fossil record before 3.2 stone implements to crack nuts to mya. However, the fact that by this time DID HOMININS USE suggest that: the hominin brain was comparable to UNMODIFIED STONES AS that of great apes in discernible mor- “In using stones as hammers, TOOLS BEFORE 2.5 MYA? phology and as large in relative size sug- they would occasionally split- gests that hominins were sufficiently in- Chimpanzees in West Africa living . . . The creature would be led to telligent to use tools. under natural conditions habitually perform the motions which had The relatively comprehensive sam- use stone tools as hammers and anvils been found effectual in splitting ple of early hominin hand bones (Box 1). Therefore, because Pan and the stone before applying it to ARTICLES Evolutionary Anthropology 239

the purposes for which it was to cis longus on the pollical distal pha- tools, or engaged in other manipula- beused...Andwemaythere- lanx,43 a third metacarpal designed tive tasks, such as food preparation, fore consider it probable that for stability,43,44 and a flat distal radi- that are more complex than those should any evidences of us.4,45 These human-like characteris- practiced by extant great apes before [hu]man[s] be hereafter discov- tics suggest that the hand of A. africa- the earliest-known stone tool record. ered in [M]iocene beds, they will nus dating to 2.4 to 3 mya was It remains to be seen whether or not be associated with...large adapted for manipulation to a degree these features appear in hominins crude flakes” (that is, single- not evident in earlier hominins (Box earlier than A. afarensis. flaked tools) (p. 32–33). 2). Furthermore, faunal remains with The proposal that hominins manu- factured and used modified stone Therefore, Pitt-Rivers argued that stone tool cut marks have been found tools before 2.5 mya is indirectly sup- long before Oldowan tools were coeval with another , ported by some of the earliest-known known, early modified stone tools the recently discovered taxon Australo- Oldowan tools. A subset of early Old- were most likely preceded by stone pithecus garhi, which dates to about 2.5 owan tools, although crude compared tools similar to those used by modern 17 mya. to later tool industries, includes well- chimpanzees. Phylogenetic and avail- The hand of A. afarensis is more made flakes with clear striking plat- able fossil evidence are consistent primitive than those of the later forms, bulbs of percussion, and dorsal with this notion. 46 hominins, but it is still derived rel- scars indicative of multiple prior flake ative to that of apes in features that removals from the core, implying that DOES THE EARLIEST are important for manipulation.47 their makers understood the fracture APPEARANCE OF STONE Most importantly, the ratio of the mechanics of stone.7,36,49,50 At some of thumb length to finger length resem- the earliest known Oldowan sites such TOOLS IN THE as Gona50 and Lokalalei,7 the stone ARCHEOLOGICAL RECORD assemblages show relatively elaborate REPRESENT THE EARLIEST USE The recent excavation schemes in which various OF INTENTIONALLY MODIFIED of a chimpanzee stone- raw materials were gathered from STONES AS TOOLS? stream beds, transported throughout tool site in West Africa the landscape, and reduced as part of Much research has been devoted to provides one model for a complete stone-reduction sequence characterizing morphological traits in which as many as thirty flakes were that are consistent with improved how an unintentionally removed from a single core. In addi- hard-hammer percussion abilities modified stone-tool tion, a few of the flakes were re- (see Box 2). Hard-hammer percussion touched. Therefore, it is unlikely that is thought to be particularly impor- assemblage might these tools represent the first attempts tant for the creation of flake and core appear in the paleo- at stone , the earliest lithic tools because it allows for much technology, or the first hominin tools. greater control than do other flaking archeological record. techniques such as throwing a stone FUTURE DIRECTIONS against a hard surface. However, the requisite morphological refinements bles that of the modern human hand If hominins used tools, possibly in- in the hand would enhance the perfor- more closely than it does the hands cluding modified stone tools, before mance of a variety of manipulative be- of extant nonhuman hominoids.48 2.5 mya, why is there no evidence of haviors involving a variety of materi- The hand of A. afarensis also fea- that tool use? Tools made of ephem- als. Therefore, morphological factors tures a second metacarpal with the eral material such as wood and grass that improve the ability to use the distinctly human ability to rotate for are unlikely to have been preserved. hard-hammer percussion technique grasping and cupping the hand.48 It Furthermore, unmodified stone tool (for example, those associated with has been argued that the second use, if similar to the chimpanzee case, grip capabilities and stress resistance) may be difficult to recognize in the metacarpal of A. afarensis has char- observed in fossil hominins would be archeological record. But why has no acteristics that would facilitate the consistent with stone-tool modifica- direct evidence of modified stone-tool firm pad-to-side pinch and three-jaw tion and use,40–42 but might also be use before 2.5 mya been found? At chuck grips that modern humans of- related to other manipulative tasks. many of the sites that contain the ear- Early hominins predating the stone- ten use when replicating and using liest Oldowan assemblages, earlier de- 47 tool record have some of the human- Oldowan tools. These features posits are devoid of flaked stone. For like adaptations for manual dexterity were present by about 3.2 mya. example, in places like Hadar, in East that characterize the earliest known Thus, adaptations for manual dex- Africa, abundant fossil hominin re- stone tool-makers, Homo habilis s. s. terity appear well before the earliest mains have been found in deposits and robustus. Hand fos- known evidence of modified stone dated between 3.4 and 2.8 mya, but sils attributed to A. africanus boast a tools. These adaptations suggest that stone artifacts do not appear until 2.3 broad apical tuft, apical spines, a early hominins used or made tools, mya.13 Tools dating to about 2.5 mya marked insertion site for flexor polli- possibly including modified stone have been found at nearby Gona50 and 240 Evolutionary Anthropology ARTICLES

Box 2. Morphology Associated With Primate Tool Use

The behavioral convergence be- tween the great apes and capuchins (see Box 1) helps identify the mini- mum morphological requirements for primate tool use. The great apes have relatively opposable thumbs and thus are able to make pad-to-pad contact between the index finger and thumb, although not as completely as hu- mans can. They are also capable of a wide range of manipulative behav- iors.40,87 Capuchins have a pseudo- opposable thumb89 and are capable of both power and precision grips.90 In addition, the great apes and capu- chins have large brains relative to body size when compared with other nonhuman primates.91 A bonobo, an orangutan, and sev- eral capuchin monkeys have made and used modified stone tools in cap- tivity. Thus, these species possess the minimum morphological require- ments for stone-tool use and manu- facture. However, these and other primates with comparable hand mor- phology do not habitually make stone tools. Thus, the fact that all hominins Figure 1. Hand anatomy related to manipulative skills and stress resistance. Many of known to date have these “minimum” these features have been related to tool use and manufacture, but might also be requirements shows that they have related to other uses of the hand (for example, manipulation involved in food process- the ability to modify stone but does ing). See text and Box 2 for the likely functional significance of these features and their presence in some early hominin taxa. MC ϭ metacarpal. not provide evidence of such behav- ior. Hard-hammer percussion, which ion and pronation of the metacarpal; evidence of stone tools, fossil hominin provides greater control than do and, most importantly, a high thumb- hands had derived features related to cruder flaking techniques such as to-finger-length ratio. Hand fossils enhanced manipulative skills. throwing a stone against a hard sur- belonging to A. africanus have skele- Hand bones attributed to Homo face, almost certainly has different tal features such as broad distal habilis s.s. (for example, OH 7), morphological requirements than the phalangeal tuberosities and ungual which are coeval with Oldowan stone modification by nonhuman pri- spines, which are associated with tools,92,93 also have broad distal mates seen to date. broad, fleshy fingertips that improve phalangeal tuberosities94 and show Some of the oldest known fossil object handling. They also show an evidence of a flexor pollicis longus hand bones, those of A. afarensis and insertion area on the pollical distal tendon92 (but see Susman and A. africanus, exhibit derived features phalanx, which is evidence of the Creel94). The hand of H. habilis s.s. that are related to enhanced manipu- presence of a flexor pollicis longus also resembles that of humans in lative function as compared to that of muscle; that muscle, which is rare in having a broader, less contoured apes. The hand of A. afarensis (for great apes but present in humans, is pollical carpo-metacarpal joint than example, those from the AL 333 site) important for strength in thumb flex- does the ape hand.94,95 This has a second metacarpal with an ion. Also, the styloid process of the is thought to improve stress dissi- asymmetric head and volar-proximal third metacarpal in A. africanus might pation across the joint. Evidence projection of the joint surface, fea- have helped to stabilize its articula- tentatively suggests that the hand of tures that are thought to allow slight tion with the carpal bones and trans- H. habilis s.s. had relatively human- pronation of the proximal phalanx to fer stresses associated with tool-re- like thumb-to-finger proportions.96 accommodate object shape during lated activities. The flat distal radius These features suggest that H. ha- grasp; relatively transverse orienta- of A. africanus probably increased the bilis s.s. had grasping abilities con- tions of the second metacarpal joints range of wrist extension, which is im- sistent with complex manipulation with the capitate and trapezium, portant for hand and finger move- including stone-tool manufacture which are thought to allow more flex- ments. Thus, before the oldest current using hard-hammer percussion. ARTICLES Evolutionary Anthropology 241

Box 2. Morphology Associated With Primate Tool Use (continued)

Although the connection is not uni- flexor pollicis longus,98 a broad pollical the hand bones are attrib- versally accepted (see Trinkaus and metacarpal head and shaft with a hu- utable to P. robustus, then its hand was Long97) the robust australopith Paran- man-like base,99 and a relatively mo- as well, if not better, adapted for ma- thropus (or ) robustus bile, human-like wrist joint.45,100 All of nipulation than was that of H. habilis has been linked with the Oldowan tools these features are consistent with ad- s.s.98 However, the morphological fac- at Swartkrans.98,99 The Swartkrans fos- aptations for manual dexterity and tors that facilitate hard-hammer per- sils exhibit broad distal phalangeal tu- power that would be useful for the cussion techniques might also facilitate berosities, a site for insertion of the manufacture and use of stone tools. If other manipulative tasks.40,41

faunal remains with tool-cut marks been manufacturing and using modi- for pre-2.5 mya sites emphasize iden- dating to approximately 2.5 mya have fied stone tools long before the earliest tifiable faunal remains, but a shift to- been found at Bouri, in the Middle currently known Oldowan tools. How- ward recovery of samples of unidenti- Awash, coeval with Australopithecus ever, if these tools were not concen- fiable long bone fragments would garhi.17 Similarly, in Southern Africa, trated in time and space (especially if increase chances of finding cut marks sites such as and Maka- they involved lone artifacts with min- or percussion before 2.5 mya. These pansgat have yielded hominin fossils imal flaking episodes), they would es- types of data, unless directly associ- dated to between 3 and 2.5 mya,51,52 sentially be archeologically “invisible” ated with primate remains, would not but flaked stone tools do not appear in (see also Klein13). indicate which species made and used these areas until about 2 mya.53 Several lines of potential evidence the tools, but they would support the Direct evidence older than 2.5 mya may, in the future, help to test the hypothesis that stone tool use existed may be absent because hominins were hypothesis that hominins used tools, before 2.5 mya. not using any tools, or at least modi- including stone tools, before 2.5 mya. Methods for obtaining data on tool fied stone tools, before 2.5 mya. For Obviously, the discovery of modified use independent of the recovery of example, the earliest known appear- and unmodified stone and nonlithic modified tools and bones processed ance of modified stone tools could tools dating to before 2.5 mya in the by tools would also be helpful. One have coincided with a dietary shift, fossil record would support the hy- promising avenue of research involves such as a new reliance on large verte- pothesis. However, there are two min- examining aspects of hand skeletal brates. The processing of large verte- imum requirements for finding such morphology that are sensitive to bio- brates is a behavior not seen in any tools: they have to exist, and they mechanical stresses.57 Some aspects living nonhuman primate, although a must be recognizable as artifacts. of bone, such as cortical robusticity,58 variety of nonhuman primates do eat Specific use-wear patterns on trabecular architecture,59 and shaft vertebrate prey smaller than 10 kg. stones not associated with deliberate curvature60 model and remodel in re- Therefore, it is possible that modified flaking would provide evidence of sponse to biomechanical loads and stone-tool use was first developed to hominin tool use before 2.5 mya.9 For are thus sensitive to daily use. Ways in aid in the processing of large prey, example, the repeated use of stone which the biomechanics of tool-mak- although microwear analyses indicate hammers and anvils to open hard nuts ing and tool use are distinct from that early Oldowan tools were also produces distinctive use-wear pat- other uses of the hand42,61 suggest used in a variety of ways not directly terns on both the hammers and anvils that significant tool-related activities related to the processing of verte- that can be recognized in the archeo- may leave stress signatures in the brates. logical record.27,55,56 In addition, the hand skeleton. Therefore, important It is also possible, however, that stone assemblage recently excavated areas of future research include im- around 2.5 mya changes occurred that at a chimpanzee nut-cracking site in proving our understanding of how caused behaviors that previously had West Africa demonstrates that low- tool-related activities are biomechani- been archeologically “invisible” to be- density, unintentionally modified cally distinct from other (for example, come archeologically “visible” (see stone tool assemblages are recover- locomotor) hand uses and investigat- Box 3). The chance of finding a lone able through excavation.38 It also pro- ing what osteological evidence might in situ, especially one with vides one model for what a pre-Old- provide a durable record of an indi- only one or two flakes struck, is like owan stone tool technology might vidual’s biomechanical history. finding the proverbial needle in a hay- look like in the archeological record. An important point that researchers stack. Moreover, a single-surface Faunal remains exhibiting cut or interested in hominin tool use should stone artifact is difficult to date and percussion marks due to modified keep in mind is that because tool use could easily be dismissed due to the stone-tool use dating to before 2.5 varies intraspecifically among primate possibility of contamination from mya would also provide support for populations,62,63 uniform tool use later sites (for example, see Howell, the hypothesis that modified stone across hominin populations should Haesaerts, and de Heinzelin54). In tools were being used before 2.5 not be expected. Therefore, absence of other words, hominins could have mya.16,17 Current collection strategies evidence of a specific type of tool use 242 Evolutionary Anthropology ARTICLES

Box 3. Site Formation Models to Explain the “Sudden” Appearance of Modified Stone Tools 2.5 Mya Oldowan tools have almost always been found in large concentrations, leading several researchers to suggest that early lithic sites represent areas where stone tools were concentrated by early hominins in both space and time (for example, home bases, butch- ering sites, and “workshops” for mak- ing tools).13,101,102 In most of these cases, the hypothesis is that when hominins began to make modified stone tools they accumulated the tools in the same geographic area over a relatively short period (see Fig. 1, A 3 C), resulting in the conclusion that the early “accumulated” sites represent the origin of modified stone tool use. But this interpretation is only one of several possible explanations of the evidence (Fig. 1). For example, Potts102 suggested that hominins may have manufac- tured and used stone tools before 2.5 mya, but that the tools do not show up in the fossil record because they were not cached before 2.5 mya (see Fig. 1, B 3 C). Another scenario for the appearance of stone tools in the archeological record at 2.5 mya has been proposed by Brooks and Lad- en,103 who attribute it to climate change that led to an alteration in habitat and available foods that changed the way hominins used the landscape (Fig. 1, D). Figure 1. Site formation models that could explain the “sudden” appearance of mod- Brooks and Laden103 argue that the ified stone tools at 2.5 mya. This figure represents a range of site-formation scenarios differences in the stability of certain consistent with accumulated stone-tool sites in the archeological record dating to 2.5 landscape features can make the dif- mya. The figure is separated into three main time frames: before 2.5 mya (an unspec- ified period between 5 and 2.5 mya), about 2.5 mya, and the present. Time is repre- ference between artifact-producing sented vertically in the entire figure, and space is represented three-dimensionally in behaviors being “visible” and “invisi- the subfigures A–D. The rounded objects represent modified stone tools. The tools to the ble” in the archeological record. For right of the Present time frame represent an Oldowan tool “site.” A: Modified stone tools example, trees have a much shorter were not made or used by hominins at the specified time frame; B: Modified stone tools life span than do geologic formations were distributed at low density in both time and space; C: Modified stone tools were like outcrops. Let us assume concentrated in both time and space; D: Modified stone tools were deposited at low density per use event, but at the same location over a long period of time. These are that two populations engage in a schematic figures demonstrating general distribution patterns; the represented tools shade-dependent, artifact-producing are not necessarily to scale in time or space. Before 2.5 mya, A and B are possible, and behavior at similar frequencies. One either situation could lead to C or D at about 2.5 mya. Any one of the four scenarios (A population lives in an environment to C, A to D, B to C, or B to D) could produce the accumulation of modified stone tools with a limited number of shade trees represented by Oldowan sites in the contemporary archeological record. and no other options for shade (for example, a savanna), while the other resent the focal points for the artifact- riod than they would be around the population lives in an area with no producing behavior. Because rock trees. Therefore, landscape use trees but a similar number of areas outcrops have a much longer poten- around the rocks would be much where shade is provided by rock out- tial life span than trees do, behaviors more stable over a geologic time crops (for example, a desert). Thus, would be concentrated around the scale, and would produce sites with shade trees and rocky outcrops rep- rock outcrops over a much longer pe- higher artifact density than would the ARTICLES Evolutionary Anthropology 243

Box 3. Site Formation Models to Explain the “Sudden” Appearance of Modified Stone Tools 2.5 Mya (continued)

more dispersed sites associated with regions in Africa, the Kalahari Desert in the early archeological record. They trees. Thus, the sites around the out- Western Botswana and the Ituri Forest do show, however, that 2.5 mya may crops would be more “visible” in the in the Democratic Republic of Congo. not mark the emergence of modified archeological record, and would mimic These site-formation scenarios do stone tool use, but instead may be cached sites. This pattern is supported not exhaust the possible scenarios the first time that the behavior be- by ethnoarcheological data from two for concentrations of stone tools in came archeologically visible.

at one hominin site does not necessar- capacity to use unmodified stones as of hominin tool use. It also encour- ily imply that other populations tools. ages integrative studies of early homi- within the same species did not use 3. Does the earliest appearance of nin life ways, and has drawn attention that specific tool type. In addition, as stone tools in the archeological record to nontraditional lines of research recent research has demonstrated, represent the earliest use of intention- that may help refute or support the crucial information about human or- ally modified stones as tools? The rel- hypothesis that hominins used tools, igins may come from outside East and ative complexity of a subset of early possibly including modified stone Southern Africa, such as the Sahel Oldowan tools implies that hominins tools, before 2.5 mya. Evidence from a and rainforests of Central and West most likely had experience modifying range of disciplines will provide a 64,65 Africa. An archeological record stone tools before 2.6 mya. Morpho- more comprehensive understanding dominated by evidence from East and logical evidence older than 2.6 mya of the evolution of tool use by homi- Southern Africa may not fully reflect does not preclude stone tool manufac- nins than will an approach that is the technological chronology of homi- ture. limited to lithic evidence and the nin populations elsewhere on that products of stone tool use. With char- continent. Therefore, the available direct evi- acteristic prescience several decades dence of tool use in the archeological ago, Isaac66 implied that researchers record potentially underestimates the CONCLUSIONS need to look beyond the conventional origin of hominin tool use by millions archeological record for evidence Regarding the questions posed in of years (Fig. 1). What we may be de- about the evolution of human behav- this paper: tecting at 2.5 mya is an intensification ior. He stated, in regard to stone tools, 1. Did hominins use tools before or spatial reorganization of previous that “we need to assess the limits to 2.5 mya? Because panins and homi- tool-related behaviors resulting from the amount of blood that can realisti- nins are sister clades, parsimony sug- ecological exigency, related or unre- cally be squeezed from these stones” gests the last common ancestor of Pan lated to new dietary opportunities (p. 5). and Homo (8 to 5 mya) used tools. such as a drier climate or the con- Furthermore, the earliest available sumption of large mammals by homi- data on hominin brain size and mor- nins. ACKNOWLEDGMENTS phology and hand morphology imply Early modified stone tools provide We thank Norman Birchfield, Phyl- that hominin tool use was possible by insights into the behavior and ecology lis Dolhinow, John Fleagle, Richard at least 3.2 mya. The combination of of early hominin taxa. However, plac- Klein, Daniel Lieberman, Elsworth primatological, phylogenetic, and fos- ing all our emphasis on the stone tools Ray, Karen Schmidt, David Strait, sil evidence suggests that hominins and the products of tool use that have John Yellen, anonymous reviewers, living as early as 8 to 5 mya, and al- been recovered from the fossil or ar- and especially Julio Mercader for most certainly by 3.2 mya, had the cheological record to the exclusion of their valuable comments on various capacity to use tools. alternative sources of evidence may drafts of this paper. While working on 2. Did hominins use unmodified strongly skew reconstructions of early this manuscript, Melissa Panger was stones as tools before 2.5 mya? Be- hominin life ways. Moreover, by fo- supported by National Science Foun- cause the use of unintentionally mod- cusing on lithic evidence we inevitably dation IGERT Post-Doctoral Fellow- ified stone tools is within the realm of constrain investigations of the origin ship; Bernard Wood was supported by free-ranging chimpanzees and capu- and evolution of tool use by hominins. the Henry Luce Foundation and a GW chins, the evidence of the possible use Our questions are restricted to the grant to the Center for the Advanced of unmodified stone tools by early evolution of the earliest known, accu- Study of Human Paleobiology. hominins is similar to that for ques- mulated modified stone tools. This tion 1. Therefore, the combination of omits important aspects of the evolu- primatological, phylogenetic, and fos- tion of tool use and the emergence of REFERENCES sil evidence suggests that hominins modified stone tool use. 1 Jurmain R, Nelson H, Kilgore L, Trevathan W. living as early as 8 to 5 mya, and al- This review has re-evaluated the 1997. Introduction to physical anthropology. most certainly by 3.2 mya, had the way we study the origin and evolution Belmont, CA: Wadsworth Publishing. 244 Evolutionary Anthropology ARTICLES

2 Bobe R, Behrensmeyer AK, Chapman RE. structing pharyngeal dimensions. J Hum Evol 41 Schick KD, Toth N, Garufi G, Savage-Rum- 2002. Faunal change, environmental variability 36:487–517. baugh ES, Rumbaugh D, Sevcik R. 1999. Con- and late hominin evolution. J Hum Evol 22 Fouts RS. 1973. Acquisition and testing of tinuing investigations into the stone tool-making 42:475–497. gestural signs in four young chimpanzees. Sci- and tool-using capabilities of a bonobo (Pan pa- 3 Beck B. 1980. Animal tool behavior: the use ence 180:978–980. niscus). J Archaeol Sci 26:821–832. and manufacture of tools by animals. New York: 23 Dibble HL. 1989. The implications of stone 42 Marzke MW, Marzke RF. 2000. Evolution of Garland Press. tool types for the presence of language during the the human hand: approaches to acquiring, anal- 4 Wood B, Richmond BG. 2000. Human evolu- middle . In: Mellars P, Stringer C, ed- ysing and interpreting the anatomical evidence. J tion: taxonomy and paleobiology. J Anat 197:19– itors. The human revolution: behavioural and bi- Anat 197:121–140. 60. ological perspectives on the origins of modern 43 Ricklan DE. 1987. Functional anatomy of the 5 Harris JWK, Capaldo SD. 1993. The earliest humans. Edinburgh: Edinburgh University hand of Australopithecus africanus. J Hum Evol stone tools: their implications for an understand- Press. p 415–432. 16:643–664. ing of the activities and behaviour of late plio- 24 Ruvolo M. 1997. Molecular phylogeny of the 44 Marzke MW, Marzke RF. 1987. The third cene hominids. In: Berthelet A, Chavaillon J, ed- hominoids: inferences from multiple indepen- metacarpal styloid process in humans: origin and itors. The use of tools by human and non-human dent DNA sequence data sets. Mol Biol Evol 14: functions. Am J Phys Anthropol 73:415–432. primates. Oxford: Claredon Press. p 196–220. 248–265. 45 Richmond BG, Strait DS. 2000. Evidence that 6 Isaac GL, Harris JWK. 1980. A method for de- 25 Gagneux P, Varki A. 2001. Genetic difference humans evolved from a knuckle-walking ances- termining the characteristics of artefacts be- between humans and great apes. Mol Phylogenet tor. Nature 404:382–385. tween sites in the upper member of the Koobi Evol 18:2–13. 46 Stern JT, Susman RL. 1983. The locomotor Fora Formation, East Lake Turkana. In: Leakey 26 Satta Y, Klein J, Takahata N. 2000. DNA ar- anatomy of Australopithecus afarensis. Am J Phys REF, Ogot BA, editors. Proceedings of the 8th chives and our nearest relative: the trichotomy Anthropol 60:279–317. PanAfrican Congress of and Quater- problem revisited. Mol Phylogenet Evol 14:259– 47 Marzke MW. 1997. Precision grips, hand mor- nary Studies, Nairobi. Nairobi: The International 275. phology, and tools. Am J Phys Anthropol 102:91– Leakey Memorial Institute for African Prehis- 27 Boesch C, Boesch-Achermann H. 2000. The 110. tory. p 19–22. chimpanzees of the Taı¨ Forest. Oxford: Oxford 48 Marzke MW. 1983. Joint function and grips of 7 Roche H, Delagnes A, Brugal JP, Feibel C, Ki- University Press. the Australopithecus afarensis hand, with special bunjia M, Mourrel V, Texier PJ. 1999. Early hom- 28 McGrew WC. 1992. Chimpanzee material cul- reference to the region of the capitate. J Hum inid stone tool production and technical skill ture: implications for . Cam- Evol 12:197–211. 2.34 myr ago in West Turkana, Kenya. Nature bridge: Cambridge University Press. 49 Ambrose SH. 2001. Paleolithic technology 399:57–60. 29 Parker ST, Gibson K. 1977. Object manipula- and human evolution. Science 291:1748–1753. 8 Merrick HV, Merrick JPS. 1976. Archeological tion, tool use and sensorimotor intelligence as 50 Semaw S, Renne P, Harris JWK, Feibel CS, occurrences of earlier age, from feeding adaptations in cebus monkeys and great Bernor RL, Fesseha N, Mowbray K. 1997. 2.5- Shungura Formation. In: Coppens Y, Howell FC, apes. J Hum Evol 6:623–641. million--old stone tools from Gona, Ethio- Isaac GL, Leakey REF, editors. Earliest man and 30 Tanner N, Zihlman A. 1976. Women in evolu- pia. Nature 385:333–336. environments in the Lake Rudolf Basin: stratig- tion, part I: innovation and selection in human 51 Reed KE, Kitching JW, Grine FE, Jungers raphy, paleoecology and evolution. Chicago: Chi- origins. Signs 1:585–608. WL, Sokoloff L. 1993. Proximal femur of Austra- cago University Press. p 574–584. 31 Reed KE. 1997. Early hominid evolution and lopithecus africanus from Member 4, Makapans- 9 Schick KD, Toth N. 1993. Making silent stones ecological change through the African Plio-Pleis- gat, . Am J Phys Anthropol 92:1–15. speak. New York: Simon and Schuster. tocene. J Hum Evol 32:289–322. 52 Clarke RJ, Tobias PV. 1995. Sterkfontein 10 Bunn HT, Kroll EM. 1986. Systematic butch- 32 Plummer T, Bishop LC, Ditchfield P, Hicks J. Member 2 foot bones of the oldest South African ery by Plio-Pleistocene hominids at Olduvai 1999. Research on late Pliocene Oldowan sites at hominid. Science 269:521–524. Gorge, . Curr Anthropol 27:431–452. Kanjera South, Kenya. J Hum Evol 36:151–170. 53 Kuman K. 1996. The Oldowan Industry from 11 Toth N. 1997. The artefact assemblages in the 33 Falk D. 1985. Hadar AL 162-28 endocast as Sterkfontein: raw materials and core forms. In: light of experimental studies. In: Isaac GL, edi- evidence that brain enlargement preceded corti- Pwiti G, Soper R, editors. Aspects of African ar- tor. research project, Vol. 5, Plio- cal reorganization in hominid evolution. Nature chaeology: papers from the 10th Congress of the Pleistocene archaeology. Oxford: Claredon Press. 313:45–47. PanAfrican Association for Prehistory and Re- p 361–388. 34 Kappelman J. 1996. The evolution of body lated Studies. Harare: University of Zimbabwe 12 Keeley LH, Toth N. 1981. Microwear polishes mass and relative brain size in fossil hominids. J Publications. p 139–146. on early stone tools from Koobi Fora, Kenya. Hum Evol 30:243–276. 54 Howell FC, Haesaerts P, de Heinzelin J. 1987. Nature 293:464–465. 35 McHenry HM. 1988. New estimates of body Depositional environments, archeological occur- 13 Klein RG. 1999. The human career: human weight in early hominids and their significance to rences and hominids from Members E and F of biological and cultural origins. Chicago: Univer- encephalization and megadontia in “robust” aus- the Shungura Formation (Omo Basin, Ethiopia). sity of Chicago Press. tralopithecines. In: Grine FE, editor. Evolution- J Hum Evol 16:665–700. 14 Leakey MD. 1971. . London: ary history of the “robust” . 55 Joulian F. 1996. Comparing chimpanzee and Cambridge University Press. New York: Aldine de Gruyter. p 133–148. early hominid techniques: some contributions to cultural and cognitive questions. In: Mellars P, 15 Potts R, Shipman P. 1981. Cutmarks made by 36 Kimbel WH, Walter RC, Johanson DC, Reed Gibson K, editors. Modeling the early human stone tools on bones from Olduvai Gorge, Tan- KE, Aronson JL, Assefa Z, Marean CW, Eck GG, mind. Cambridge: McDonald Institute for Ar- zania. Nature 291:577–580. Bobe R, Hovers E, Rak Y, Vondra C, Yemane T, York D, Chen Y, Evensen NM, Smith PE. 1996. chaeological Research. p. 173–189. 16 Blumenschine RJ, Selvaggio MM. 1988. Per- Late Pliocene Homo and Oldowan tools from the 56 Goren-Inbar N, Sharon G, Melamed Y, Kislev cussion marks on bone surfaces as a diagnostic Hadar Formation (Kada Hadar Member), Ethio- M. 2002. Nuts, nut cracking, and pitted stones at of hominid behaviour. Nature 333:763–765. pia. J Hum Evol 31:549–561. Gesher Benot Ya’aqov, Israel. Proc Natl Acad Sci 17 De Heinzelin J, Clark JD, White T, Hart W, 37 Falk D, Guyer J, Redmond JC, Conroy GC, USA 99:2455–2460. Renne P, Woldegabriel G, Beyene Y, Vrba E. Recheis W, Weber GW, Seidler H. 2000. Early 57 Lieberman DE. 1997. Making behavioral and 1999. Environment and behavior of 2.5-million- hominid brain evolution: a new look at old endo- phylogenetic inferences from hominid fossils: year-old Bouri hominids. Science 284:625–629. casts. J Hum Evol 38:695–717. considering the developmental influence of me- 18 Brain CK. 1985. Cultural and taphonomic 38 Mercader J, Panger MA, Boesch C. 2002. Ex- chanical forces. Ann Rev Anthropol 26:185–210. comparisons of hominids from Swartkrans and cavation of a chimpanzee stone tool site in the 58 Jones HN, Priest JD, Hayes WC, Tichenor CC, Sterkfontein. In: Delson E, editor. Ancestors: the African rainforest. Science 296:1452–1455. Nagel DA. 1977. Humeral hypertrophy in re- hard evidence. New York: Alan R. Liss. p 72–75. 39 Sigaut F. 1993. How can we analyse and de- sponse to exercise. J Bone Joint Surg 59:204– 19 Backwell LR, D’Errico F. 2001. Evidence of scribe technical actions? In: Berthelet A, Chavail- 208. termite foraging by Swartskrans early hominids. lon J, editors. The use of tools by human and 59 Huiskes R, Ruimerman R, van Lenthe GH, Proc Natl Acad Sci USA 98:1358–1363. non-human primates. Oxford: Claredon Press. p Janssen JD. 2000. Effects of mechanical forces 20 Schmandt-Besserat D. 1978. The earliest pre- 381–397. on maintenance and adaptation of form in tra- cursor of writing. Sci Am 238:50–59. 40 Marzke MW, Wullstein KL. 1996. Chimpan- becular bone. Nature 405:704–706. 21 Lieberman DE, McCarthy RC. 1999. The on- zee and human grips: a new classification with a 60 Richmond BG. 1999. Reconstructing locomo- togeny of cranial base angulation in humans and focus on evolutionary morphology. Int J Primatol tor behavior in early hominids: evidence from chimpanzees and its implications for recon- 17:117–139. primate development. J Hum Evol 36:A20. ARTICLES Evolutionary Anthropology 245

61 Marzke MW, Toth N, Schick KD, Reece S, Manufacture and use of tools in wild Sumatran 89 Napier JR. 1960. Studies of the hands of living Steinberg B, Hunt K, Linscheid RL, An K-N. orangutans: implications for human evolution. primates. Proc Zool Soc, London 134:647–657. 1998. EMG study of hand muscle recruitment Naturwissenschaften 83:186–188. 90 Costello MB, Fragaszy D. 1988. Prehension in during hard hammer persussion manufacture of 74 Ingmanson EJ. 1996. Tool-using behavior in Cebus and Saimiri: I. Grip type and hand prefer- Oldowan tools. Am J Phys Anthropol 105:315– wild Pan paniscus: social and ecological consid- ence. Am J Primatol 15:235–245. 332. erations. In: Russon AE, Bard KA, Parker ST, 91 Stephan H, Baron G, Fraham HD. 1988. Com- 62 Panger MA, Perry S, Rose L, Gros-Louis J, editors. Reaching into thought: the minds of the parative size of brains and brain components. In: Vogel E, MacKinnon KL, Baker M. 2002. Cross- great apes. Cambridge: Cambridge University Steklis HD, Erwin J, editors. Comparative pri- site differences in foraging behavior of white- Press. p 190–210. mate biology. New York: Alan R. Liss. p 1–39. faced capuchins (Cebus capucinus). Am J Phys 75 Russon AE, Gladikas BMF. 1992. Imitation in 92 Napier JR. 1962. Fossil hand bones from Anthropol 119:52–66. ex-captive orangutans (Pongo pygmaeus). J Comp Olduvai Gorge. Nature 196:409–411. 63 Whiten A, Goodall J, McGrew WC, Nishida T, Psychol 107:147–161. 93 Leakey LSB, Tobias PV, Napier JR. 1964. A Reynolds V, Sugiyama Y, Tutin CEG, Wrangham 76 Parker ST, Kerr M, Markowitz H, Gould J. new species of the genus Homo from Olduvai RW, Boesch C. 1999. Cultures in chimpanzees. 1999. A of tool use in zoo gorillas. In: Gorge. Nature 202:7–9. Nature 399:682–685. Parker ST, Mitchell RW, Miles HL, editors. The 94 Susman RL, Creel N. 1979. Functional and 64 Mercader J. 2002. Forest people: the role of mentalities of gorillas and orangutans. Cam- morphological affinities of the subadult hand African rainforests in human evolution and dis- bridge: Cambridge University Press. p 188–193. (OH 7) from Olduvai Gorge. Am J Phys An- persal. Evol Anthropol 11:117–124. 77 Kohler W. 1927. The mentality of apes. Lon- thropol 51:311–331. 65 Brunet M, Guy F, Pilbeam D, Mackaye HT, don: Kegan Paul, Trench, Trubner. 95 Trinkaus E. 1989. Oldovai Hominid 7 trape- Likius A, Ahounta D, Beauvilain A, Blondel C, 78 Jordan C. 1982. Object manipulation and zial metacarpal 1 articular morphology: con- Bocherens H, Boisserie J-R, Bonis Ld, Coppens tool-use in captive pygmy chimpanzees (Pan pa- trasts with recent humans. Am J Phys Anthropol Y, Dejax J, Denys C, Duringer P, Eisenmann V, niscus). J Hum Evol 11:35–39. 80:411–416. Fanone G, Fronty P, Geraads D, Lehmann T, Lihoreau F, Louchart A, Mahamat A, Merceron 79 Visalberghi E. 1990. Tool use in Cebus. Folia 96 Aiello L, Dean C. 1990. An introduction to G, Mouchelin G, Otero O, Campomanes PP, Leon Primatol 54:146–154. human evolutionary anatomy. London: Aca- MPd, Rage J-C, Sapanet M, Schuster M, Sudre J, 80 Chapman CA. 1986. Boa constrictor preda- demic Press. Tassy P, Valentin X, Vignaud P, Viriot L, Zazzo A, tion and group response in white-faced cebus 97 Trinkaus E, Long JC. 1990. Species attribu- Zollikofer C. 2002. A new hominid from the Up- monkeys. Biotropica 18:171–172. tion of the Swartkrans Member 1 first metacar- per of Chad, Central Africa. Nature 418: 81 Boinski S. 1988. Use of a club by a wild white- pals: SK 84 and SKX 5020. Am J Phys Anthropol 145–151. faced capuchin (Cebus capucinus) to attack a 83:419–424. 66 Isaac GL. 1977. Squeezing blood from stones. venemous snake (Bothrops asper). Am J Primatol 98 Susman RL. 1988. Hand of Paranthropus ro- In: Wright RVS, editor. Stone tools as cultural 14:177–179. bustus from Member I, Swartkrans: fossil markers: change, evolution, and complexity. 82 Chevalier-Skolnikoff S. 1990. Tool use by wild evidence for tool behavior. Science 240:781–784. New Jersey: Humanities Press. p 5–12. cebus monkeys at Santa Rosa National Park, 99 Susman RL. 1994. Fossil evidence for early 67 McGrew WC, Marchant LF. 1997. Using the Costa Rica. Primates 31:375–383. hominid tool use. Science 265:1570–1573. tools at hand: manual laterality and elementary 83 Boinski S, Quatrone R, Swats H. 2000. Sub- 100 Marzke MW. 1971. Origin of the human technology in Cebus spp. and Pan spp. Int J Pri- strate and tool use by brown capuchins in Suri- hand. Am J Phys Anthropol 34:61–84. matol 18:787–810. name: ecological contexts and cognitive bases. 101 Lovejoy CO. 1981. The origin of man. Sci- 68 McGrew WC, Marchant LF. 1992. Chimpan- Am Anthropol 102:741–761. ence 211:341–350. zees, tools, and termites: hand preference or 84 Fernandes MEB. 1991. Tool use and preda- 102 Potts R. 1991. Why the Oldowan? Plio-Pleis- handedness? Curr Anthropol 33:114–119. tion of oysters (Crassostrea rhizophorea)bythe tocene toolmaking and the transport of re- 69 Boesch C, Boesch H. 1990. Tool use and tool tufted capuchin, Cebus apella apella, in brackish sources. J Anthropol Res 47:53–176. making in wild chimpanzees. Folia Primatol 54: water mangrove swamp. Primates 32:529–531. 103 Brooks A, Laden G. n.d. Environmental de- 86–99. 85 Phillips K. 1998. Tool use in wild capuchin terminants of site visibility: comparative ethno- 70 Sakura O, Matsuzawa T. 1991. Flexibility of monkeys (Cebus albifrons trinitatis). Am J Prima- archaeology in the Kalahari Desert and Ituri For- wild chimpanzee nut-cracking behavior using tol 46:259–261. est with implications for the “emergence” of stone hammers and anvils: an experimental anal- 86 Westergaard GC, Suomi SJ. 1996. Hand pref- human culture. ysis. Ethology 87:237–248. erence for stone artefact production and tool-use 104 Isaac GL. 1976. Plio-Pleistocene artifact as- 71 Yamakoshi G. 1998. Dietary responses to fruit by monkeys: possible implications for the evolu- semblages from East Rudolf, Kenya. In: Coppens scarcity of wild chimpanzees at Bossou, Guinea: tion of right-handedness in hominids. J Hum Y, Howell FC, Isaac GL, Leakey REF, editors. possible implications for ecological importance Evol 30:291–298. Earliest man and environments in the Lake Ru- of tool use. Am J Phys Anthropol 106:283–295. 87 Toth N, Schick KD, Savage-Rumbaugh ES, dolf Basin. Chicago: University of Chicago Press. 72 Fox EA, Sitompul AF, van Schaik CP. 1999. Sevcik R, Rumbaugh D. 1993. Pan the tool-mak- p 552–564. Intelligent tool use in wild Sumatran orangutans. er: investigations into the stone tool-making and 105 Harris JWK, Williamson PG, Verniers J, In: Parker ST, Mitchell RW, Miles HL, editors. tool-using capabilities of a bonobo (Pan panis- Tappen MJ, Stewart K, Helgren D, De Heinzelin The mentalities of gorillas and orangutans: com- cus). J Archaeol Sci 20:81–91. J, Boaz NT, Bellomo RV. 1987. Late Pliocene parative perspectives. Cambridge: Cambridge 88 Wright RVS. 1972. Imitative learning of a hominid occupation of the Senga 5A site, Zaire. J University Press. p 99–116. flaked stone technology: the case of an orangu- Hum Evol 16:701–728. 73 van Schaik CP, Fox EA, Sitompul AF. 1996. tan. Mankind 8:296–306. © 2002 Wiley-Liss, Inc.