d’Errico, F. & Henshilwood, C.S. 2007. Additional evidence for bone technology in the southern...

Journal of Human Evolution 52 (2007) 142e163

Additional evidence for bone technology in the southern AfricanMiddle Stone Age

Francesco d’Errico a,b,*, Christopher S. Henshilwood c,d

a Institut de Prehistoire et de Geologie du Quaternaire, CNRS UMR 5199 PACEA, Universite Bordeaux 1, avenue des Facultes, F-33405 Talence, Franceb Department of Anthropology, The George Washington University, 2110 G Street, NW, Washington, DC 20052, USA

c Centre for Development Studies, 5, Nygardsgaten, University of Bergen, N-5015, Bergen, Norwayd Institute of Human Evolution, University of the Witwatersrand, Johannesburg, South Africa

Received 28 February 2006; accepted 4 August 2006

Abstract

Few Middle Stone Age sites have yielded convincing evidence for a complex bone technology, a behavior often associated with the emergenceof modern cultures. Here, we review the published evidence for Middle Stone Age bone tools from southern Africa, analyze an additional ninebone artifacts recently recovered from Middle Stone Age levels at Blombos Cave, describe an unpublished bone tool from probable Middle StoneAge levels at Peers Cave, examine a single bone awl found at Blombosch Sands (an open site near Blombos Cave), and reappraise marked boneartifacts and a bone point recovered from Klasies River. To determine the chronological and cultural attribution of these artifacts, document bone-manufacturing techniques associated with the southern African MSA, and discuss the symbolic significance of the markings present on some ofthese objects we use (1) available contextual information; (2) morphometric comparison of Later Stone Age, Modern San, and purported MiddleStone Age projectile points; (3) analysis of the carbon/nitrogen content of bone tools and faunal remains from Peers and Blombos caves; and (4)microscopic analysis of traces of manufacture and use. Previously undescribed bone artifacts from Blombos Cave include a massive point man-ufactured on weathered bone, two complete awls and two awl tips manufactured on small-sized mammal and bird bone, a probable projectile pointwith a tang manufactured by knapping and scraping, a shaft fragment modified by percussion, used as retoucher and bearing a set of incised lineson the middle of the periosteal surface, and two fragments with possible engravings. The point from Peers Cave can be assigned to the MiddleStone Age and bears tiny markings reminiscent of those recorded on projectile points from Blombos and used as marks of ownership on San arrowpoints. The awl from Blombosch Sands and the bone point from Klasies River can be attributed to the Later Stone Age. Two notched objects fromKlasies are attributed to the Middle Stone Age and interpreted as tools used on soft material; a third object bears possible deliberate symbolicengravings. Although low in number, the instances of bone artifacts attributable to the Middle Stone Age is increasing and demonstrates thatthe bone tools from Blombos Cave are not isolated instances. New discoveries of bone tools dating to this time period can be expected.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Blombos Cave; Bone tools; Middle Stone Age; Symbolism; South Africa

Introduction

In this paper, we review the published evidence for bonetools in the southern African Middle Stone Age, reappraise

* Corresponding author. Institut de Prehistoire et de Geologie du Quatern-

aire, CNRS UMR 5199 PACEA, Universite Bordeaux 1, avenue des Facultes,

F-33405 Talence, France.

E-mail addresses: [email protected] (F. d’Errico), chris@

blomboscave.co.za (C.S. Henshilwood).

0047-2484/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jhevol.2006.08.003

some bone artifacts described previously, and analyze a num-ber of unpublished worked-bone pieces. Complex bone tech-nologies are commonly associated with behaviorally modernHomo sapiens in the European Upper Paleolithic (UP) andAfrican Later Stone Age (LSA). They are virtually absent atMiddle Paleolithic (MP) sites from Europe and the NearEast and are rarely found in the African Middle Stone Age(MSA). Exceptions in the latter two periods are bone toolsapparently made by Chatelperronian-associated Neandertalsat a few French sites (d’Errico et al., 2003; Bailey and Hublin,

mailto:[email protected]



143F. d’Errico, C.S. Henshilwood / Journal of Human Evolution 52 (2007) 142e163

2006), and the fully shaped bone tools from two African sites,Katanda dated to ca. 90 ka (Brooks et al., 1995; Yellen, 1998;McBrearty and Brooks, 2000) and Blombos Cave (BBC) datedto ca. 75e82 ka (Henshilwood et al., 2002; Jacobs, 2004;Jacobs et al., 2006; Tribolo et al., 2006). It is arguable thatthe start of the MSA at ca. 250 ka or earlier coincides withthe emergence of modern human behaviors, in particular thesystematic use of pigments (Barham, 1998, 2002; McBreartyand Brooks, 2000). Additionally, recent finds suggest thatmaterial culture indicative of behavioral modernity is presentin the African MSA, at least at some sites (McBrearty andBrooks, 2000; Henshilwood et al., 2001a,b, 2002, 2004;Henshilwood and Marean, 2003; d’Errico et al., 2005;Parkington et al., 2005).

The expectation, then, is that complex bone technologies,like other behaviors generally considered to be hallmarks ofmodernity, should be widely present in MSA sites, particularlythose postdating 100 ka. However, this seems not to be thecase. One reason may be that a relatively low number ofpost-100-kyr-old MSA sites have been excavated or dated bymodern standards compared to the many well-excavated anddated MP and UP sites from Europe. It is possible that, evenif bone tools were present in some MSA sites, they were notrecognized by the excavators and were discarded with thebulk of the recovered deposits, including unmodified bone,a practice not uncommon in early-to-middle twentieth-centuryexcavations (e.g., Wurz, 2000: 9). Another factor is that bonepreservation varies depending on the cave matrix and the ageof the deposits; in Western Cape archaeological sites, this isparticularly evident. Interiors of caves formed in quartzite(Table Mountain Group sandstones) have a high pH that rap-idly leaches bone and shell and results in a lag deposit ofonly lithics. Examples of such sites are Hollow Rock Shelter(Evans, 1994) and Dale Rose Parlour (Schirmer, 1975), bothof which contain Still Bay bifacial lithics similar to thoserecovered from the 75-kyr-old levels at BBC, but no bone orother organics. The presence of mobile dissolved carbonatesassists with bone preservation even in a quartzitic environment[e.g., Klasies River (Wurz, 2000) and the Pinnacle Point cavesat Mossel Bay (Marean and Nilssen, 2001)]. Here, calcitedunes overlie the caves and the gradual percolation ofCaCO3 into the sites contributes to the preservation of boneand shell. However, even within a calcrete cave, variation indiagenetic alteration due to the presence of acids (e.g., fromhearth charcoal) may result in the loss of some bone, particu-larly in older levels.

However, these factors may still not fully explain why bonetools are absent from many MSA sites. Bone tools may be oneof the possible, but not obligatory, outcomes of the acquisitionof modern traits. The presence or absence of bone tools atMSA sites may not signify whether the inhabitants were neces-sarily modern or nonmodern in their behavior. We know, for ex-ample, that long-bone shaft fragments and horn cores were usedas digging tools at the early hominid sites of Swartkrans, Sterk-fontein, and Drimolen (Brain and Shipman, 1993; Backwelland d’Errico, 2001). In a few cases, the tips of horn coresfrom these sites were shaped by grinding to increase their

penetrating power (d’Errico and Backwell, 2003). The recentreappraisal of the purported bone-tool collection from variousOlduvai localities (Backwell and d’Errico, 2005a) seems toconfirm that a number of objects interpreted as bone tools byLeakey (1971) and Shipman (1989) were deliberately knapped,in one case to manufacture an Acheulean hand axe, or used ashammers. Similar examples of bone-tool manufacture anduse, as well as a few examples of sequentially marked bone,come from Lower and Middle Paleolithic sites in Europe (Villaand d’Errico, 2001; d’Errico et al., 2003).

The systematic working of bone to produce artifacts, as atKatanda and BBC, does not seem repeated at other MSA sites,but bone artifacts are not entirely absent at all other MSAsites. There is a point from Mumbwa Cave (Pinto Llonaet al., 2000), considered as doubtful by d’Errico (2003), onepossible point and two gougelike tools from the later middlePleistocene layers of Broken Hill, Zambia (Barham et al.,2002), a few notched and marked objects from Klasies Riverand Apollo 11 (Wurz, 1999, 2000; McBrearty and Brooks,2000), and a bone point from a disturbed MSA context atKlasies River (Singer and Wymer, 1982). A polished bonepin and a notched rib fragment with ten evenly spaced notches,attributed to the late MSA, have been recovered from SibuduCave in KwaZulu-Natal (Cain, 2004). The bone pin comesfrom a level dated to between 26,000� 420 BP and35,200� 1800 BP. Microscopic analysis of the notches indi-cates they were made with the same tool and in a single ses-sion, thus precluding the possibility that they are a system ofnotation based on the accumulation of marks over time. Thenotched rib comes from a layer that is stratigraphically directlyabove a layer dated to 56.7� 2.3 ka. However, the AMS 14Cage estimate for this object is 28,880� 170 BP (GrA19670),suggesting either that the age is correct but it was out ofcontext or that it is older and the 14C date should be read asa minimum age.

An analysis of 28 bone tools from BBC (Henshilwoodet al., 2002) showed that 25 are awls made on long-bone shaftfragments and further manufactured by scraping, which werethen used to pierce soft material such as leather or pierce shellsto make beads (d’Errico et al., 2004). Three points, completelyfinished by careful polishing after being shaped by scraping,are probably projectile points made for hafting. Different treat-ment of these points, made for a different function, is notewor-thy. A high polish on these objects has no apparent functionalcause; rather, it seems to have been produced to give a distinc-tive appearancedan ‘‘added value’’dto these artifacts. Thecollection as a whole provides comprehensive evidence forsystematic bone-tool manufacture and use in the MSA. Al-though the BBC bone-tool assemblage presents features thatconvincingly demonstrate the technical ability and compe-tence of these people in shaping bone, we argue that this evi-dence cannot be used to support the hypothesis that the MSAas a whole, or in Western Cape, South Africa, is characterizedby the systematic production of formal bone tools. Neither canone conclude that the production of worked bone gives a mod-ern character to all of MSA material culture, as little is knownabout the relationship between bone working and behavior

144 F. d’Errico, C.S. Henshilwood / Journal of Human Evolution 52 (2007) 142e163

considered ‘‘modern’’ (see Henshilwood and Marean, 2003;Backwell and d’Errico, 2005b). Furthermore, not all MSAbone objects have been scrutinized for features that could sup-port a symbolic interpretation, but there are some examples.Microscopic analysis of a bone fragment from BBC markedwith eight parallel lines suggests that they are the result of de-liberate engraving, possibly made for symbolic purposes(d’Errico et al., 2001). Arguably, these are comparable tothe abstract patterns engraved on ochre fragments found inthe same 75-kyr-old layers (Henshilwood et al., 2002).

In this paper, we present our analyses of an unpublishedbone tool from probable MSA levels at Peers Cave, unpub-lished bone tools recently recovered from MSA levels atBBC, and a single bone tool found at Blombosch Sands, anopen site near BBC. We also reappraise marked-bone artifactsand a bone point recovered from Klasies River (Singer andWymer, 1982). Our goals are to determine the chronologicaland cultural attribution of these artifacts, discuss their sym-bolic value, and document bone-manufacturing techniquesassociated with the southern African MSA.

Methods

Each specimen was examined with a reflecting light micro-scope in order to check its state of preservation and identifyanthropogenic and natural traces of modification. Selectedareas on one BBC awl and the two Klasies River markedbones were replicated with Provil L impression material(Bayer, Germany). Positive casts, made in RBS resin (T2LChimie, France), were observed with a scanning electronmicroscope (SEM; Jeol 840A). Transparent replicas obtainedusing the same replication technique were also observed andphotographed digitally with a Nikon Coolpix 990 camera intransmitted light through a Wild M3C stereomicroscope. Thedistinction between natural and anthropogenic modificationsis based on criteria in the literature and analysis of modernand Pleistocene reference collections (Binford, 1981;Bonnichsen and Sorg, 1989; Hannus et al., 1997; Villa andd’Errico, 2001; Backwell and d’Errico, 2001, 2005a). Identifi-cation of shaping techniques and use-wear on archaeologicalspecimens is based on (1) data from experimental bone-toolmanufacture and use (Newcomer, 1974; Campana, 1980;d’Errico et al., 1984; Bergman, 1987; Shipman and Rose,1988; Choyke and Bartosiewicz, 2001; d’Errico and Backwell,2005; Backwell and d’Errico, 2005a,b); (2) experimentalreproduction and microscopic analysis of marks producedwith different tools and motions (d’Errico, 1995, 1998); and(3) comparative analysis of European Upper Paleolithic spearpoints, awls, and notched artifacts (Knecht, 1997; Villa andd’Errico, 2001), the MSA tools from BBC (Henshilwoodet al., 2001b), and LSA bone tools from Nelson Bay Cave(Henshilwood et al., 2001b).

The location and extent of worked areas, the manufacturingtechniques used and the order of the modifications, based onmicroscopic examination, were systematically recorded foreach bone artifact. Morphometric data collected included thewidth and the thickness of each bone point at 30 mm from

the tip of the tool. For comparative purposes, the same vari-ables were also recorded on nine ethnographic San Masarwaarrowhead bone points from the Kalahari Desert, six projectilepoints from the LSA levels of Nelson Bay Cave (Inskeep,1987), two bone points from the MSA levels at BBC, andone LSA bone point from BBC. All items are curated at theIziko-South African Museum, Cape Town.

Sites and descriptions of bone artifacts

Blombos Cave

Blombos Cave, situated near Still Bay in the southern Cape(34�250S, 21�130E), is some 100 m from the coast and 35 mabove sea level. The site was first excavated in 1992e1993(Henshilwood, 1995), with subsequent excavations bienniallyor annually between 1997 and 2004 (Henshilwood et al.,2001a, 2004). The interior of the cave contains at least55 m2 of visible deposit with an estimated depth of w4e5 m at the front and w2 m toward the rear. Recent discoveryof a further chamber within the cave suggests that the depositsare more extensive than was initially thought. The upperw80 cm of LSA deposits are radiocarbon dated to <2 ka(Henshilwood, 1995). Undisturbed, sterile aeolian dune sand(w10e50 cm) separates the LSA from the MSA units (Hen-shilwood et al., 2001a,b; Jacobs et al., 2003a,b; Jacobs, 2004).

The matrix in the MSA is composed mainly of aeolian,marine-derived dune sand blown in through the cave entrance,intercalated with marine shell, decomposed humic materials,and limestone. Because the cave formation is calcrete, thematrix is strongly alkaline and the preservation of organic ma-terials in the MSA levels is mostly good (Henshilwood et al.,2001a), but with some degradation of bone and shell, particu-larly in deposits close to the seasonally damp rear wall of thecave.

Three phases have been identified and named in the MSAlevels: M1, M2, and M3 (Henshilwood et al., 2001a). Datingby optically stimulated luminescence (OSL) and thermolumi-nescence (TL) methods has provided occupation dates for eachphase: ca. 70 ka for the aeolian hiatus lying on top of the MSAlayers, ca. 75e77 ka for the M1 phase (OIS 5a), ca. 82 ka forthe M2 phase (OIS 5a/b), and a suggested Eemian date, ca.125 ka or earlier, for the M3 phase (OIS 5e/6) (Jacobs et al.,2003a,b, 2006; Jacobs, 2004; Tribolo et al., 2006).

Principal markers of the M1 phase are bifacial foliatepoints, typical of the Still Bay, both complete and in variousstages of manufacture. More than 400 have been recovered.Silcrete is the dominant raw material, the nearest source forwhich is ca. 30 km away. Large numbers of small flakeswere recovered, often discretely grouped, indicating on-siteproduction of these artifacts (Henshilwood et al., 2001a).The site may have served as a lithic workshop. More than41 beads manufactured from Nassarius kraussianus gastropodshells have been discovered (Henshilwood et al., 2004;d’Errico et al., 2005). Two slabs of ochre engraved with geo-metric patterns and more than 15 bone tools come from theM1 phase (Henshilwood et al., 2001b, 2002).


The M2-phase markers include fewer Still Bay points andan increased frequency of bone-tool use. Quartzite and quartzreplace silcrete and are the preferred raw materials (Henshil-wood et al., 2001a). In the M3 phase, bifacial flaking disap-pears, silcrete is again the dominant raw material, but thereare fewer retouched tools than in M1. An unpublished studyby Soressi (pers. comm.) indicates that the M3-phase lithicsare unlike those described for the ca. 100e110-kyr-old MSAI Industry at Klasies River (Volman, 1984). Striated ochre, par-ticularly in large chunk form, is common in these M3 levels(Henshilwood et al., 2001a), and ochre-processing toolsinclude lower and upper grindstones and hammerstones. Ex-tensive hearths characterize the upper M3 layers. Dense shell-fish middens indicate a nearby shoreline. Faunal remains fromthe three MSA phases show that a wide range of terrestrial re-sources was exploited. More than a thousand fish bones (manyfrom large fish), marine shells, seals, and dolphins attest to ex-tensive exploitation of aquatic resources and suggest

exploitation patterns not dissimilar to those of LSA peoplein this region (Henshilwood et al., 2001a). Nine human teethare known from the MSA levels, but no other human skeletalmaterial has been recovered (Grine et al., 2000; Grine andHenshilwood, 2002).

1. BBC bone point: Hiatus phase, square E4, layer DUN,SAM-AA 8975. This is the distal fragment of a massive point,carved on a metaphyseal shaft fragment from a bovid of ClassII size (23e84 kg) (Fig. 1). The fibrous appearance of the un-modified periosteal surface, the presence of cracking parallelto the fiber structure on the periosteal surface, and its absenceon the modified areas indicates that the blank was alreadyweathered when shaped into a point (Fig. 1c). The weatheringmay have provided the craftsman with a blank requiring littlemodification to carve an elongated tool. The tip was shaped bygently grinding the splintered end on very fine sandstone withmotions parallel to the main axis of the object and withoutchanging the angle of grinding during the work. One full facet

Fig. 1. (a) Specimen SAM-AA 8975, bone point from square E4, layer DUN (hiatus) at BBC compared to a point (1b) found in the LSA layers of the site; (c, d)

close-up view of 1a showing the weathered appearance of the periosteal surface and the facets left on both aspects by the manufacture by grinding; (e) close-up of

1b showing oblique spindlelike striations left by grinding on a coarse matrix; scale in 1a,b¼ 1 cm; 1c,d¼ 1 mm, 1e¼ 500 mm.


Fig. 2. (a) Bone pin from the M1 phase, square H5d, layer CC at BBC; (b) area on the middle of the diaphysis with a wide composite striation left by shaping the

tool through scraping. The edges of the striations are smoothed and the remaining surface is covered by thin, longitudinal striations; (c) area close to the tip with

residues and scraping marks partially worn out by randomly oriented, thin striations, probably due to use wear; scale in 2a¼ 1 cm; 2b,c¼ 100 mm.

was ground, then the piece was turned and the next facet wasground. Repetition of this process produced four converging,elongated facets (Fig. 1c,d) covered by thin, parallel striationsthat are hardly visible using optical microscopy. This is theonly bone tool found so far at BBC that was shaped usingthis technique. Those from the MSA layers were shaped byscraping; the few tools from the LSA were shaped both byscraping and grinding. In the LSA, grinding at BBC wasapplied to fresh bone by vigorously rubbing the blank ona coarse, hard matrix, oblique or perpendicular to its mainaxis (Fig. 1e). The technique recorded on SAM-AA 8975 isalso absent in the bone-tool collections we examined fromother LSA sites (Henshilwood et al., 2001b), where grindingis very similar to that on the BBC LSA bone tools. Compari-son of the tip morphology of SAM-AA 8975, with its bluntoutline and polygonal section, with experimental and archaeo-logical bone awls (Henshilwood et al., 2001b; d’Errico et al.,2003) indicates that it would have been ineffective as a pierc-ing tool, even on soft materials. Practically, it is more likely tohave been a robust spear point. However, its morphology anddegree of finish are very different from those on objects inter-preted as spear points from the BBC MSA layers (see Henshil-wood et al., 2001b), including a new one described in thispaper (see below), and from the thin arrow points or possiblelinkshafts recovered from BBC LSA layers (Henshilwoodet al., 2001b).

2. BBC bone awl: M1 phase, square H5d, layer CC (noaccession number). This awl was made from the tibia ofa small-sized mammal, possibly a primate (Fig. 2). After re-moval of the distal epiphysis, the adjacent metaphyseal surfacewas pointed by longitudinally scraping the bone with an irreg-ular cutting edge. This left individual subparallel, often
Fig. 3. Tanged bone point from M1 phasedvarious aspects. Scale¼ 1 cm.


composite striations that decrease in density toward the middleof the diaphysis. Microscopically, two different use-wears aredetectable on the worked surface. In the area close to the tip(Fig. 2c), the wear consists of elongated striations oblique orperpendicular to the tip’s axis, probably due to use of thetool in a piercing activity (d’Errico et al., 2003). The shiny ap-pearance of the bone surface in the center of the diaphysis(Fig. 2b) is due to thinner longitudinal striations that havesmoothed the edges of the scraping marks in this area. Thepresence of this wear, very similar to that produced by exper-imentally handling bone (d’Errico, 1993), suggests that themidshaft was the zone most commonly in contact with theuser’s fingers during tool use.

3. BBC bone point: M1 phase, square I6a, layer CC (noaccession number). This point (Fig. 3) was made from thelimb-bone shaft fragment of a bovid of Class II size. The tipof the tool was formed from the compact bone present at themiddle of the diaphysis, with remnants of spongy bone stillpresent at the proximal end. The naturally pointed shape ofthe bone flake was straightened by longitudinally scrapingits converging edges and adjacent periosteal and medullaryaspects with an irregular cutting edge (Fig. 4b,c). The longtang that constitutes the base of the point was produced by re-moving, through repeated percussion, small flakes on the med-ullary aspect of the blank. Rough edges and high points left bythis technique were then carefully smoothed through scraping

Fig. 4. (a) Point (square I6a, layer CC at BBC) illustrated in Figure 3 with locations (brackets) of close-up views illustrated in 4d (top bracket) and 4e (bottom

bracket); (b, c) tip of the tool on the periosteal surface showing scraping marks and grooves left by root etching; (d, e) scraping marks on the tang regularizing its

edge (4d) and the arrises (edge) of removals (4e). Scale in 4a¼ 1 cm; 4b,c¼ 5 mm; 4d,e¼ 1 mm.


(Fig. 4d,e). The base of the tool was probably heated beforeshaping the tang, with the aim of hardening the bone to facil-itate its knapping. The use-wear polish on the tool tip showsthat the initial striations left by scraping were reduced throughutilization, suggesting that the tool was hafted and used.

4. BBC shaft fragment retoucher, M1 phase, square I5d,layer CA (no accession number). This long-bone shaft frag-ment of a Class III bovid (Fig. 5) presents a well-preservedsurface and evidence of spiral fractures, suggesting that thebone was fresh when it was broken. One end is modified bycontinuous scaled removals that created a morphology similarto that of an end-scraper. Microscopic analysis of these re-movals (Fig. 6a) reveals a light smoothing of the protrusionsthat can be attributed to use-wear, particularly as the rest ofthe object is so well-preserved. The periosteal surface closeto the opposite end bears a set of six thin grooves oblique tothe splinter axis, produced by the same stone tool (Fig. 6b).The curved outline of segments of these lines and the absenceof features characteristic of marks, such as splitting, multipleside-striations, and barbs generated by a cutting edge, indicatethat they were produced by a sharp lithic point that had incisedthe bone surface (d’Errico, 1998). The presence of a few linearimpact scars close together, one of which is crossed by anincised line, indicates that the fragment was used as a retoucherbefore being incised.

5. BBC bird bone point, M3 phase, square E3, layer AH,SAM-AA 7297. This bird humerus was recorded as comingfrom the M3 phase (layer AH) during the 1993e1997 excava-tions at BBC. It has been broken close to the distal epiphysis

and the fracture has been shaped into an awl by scraping(Fig. 7b). The entire shaft has also been longitudinally scraped(Fig. 7c). The wear that smoothes the traces of manufactureclose to the tip suggests that the tool was used before beinglost or disposed. Dark spots on the shaft indicate that thebone was heated after being shaped.

6. BBC bone tool tips, M2 phase, square G5b,layer CFB/CFC (no accession number). Both of these broken bone-pointtips were found in the same layer. The first (Fig. 8a) is prob-ably bird bone, and its thin outline is similar to those of com-plete specimens found in the BBC MSA (Henshilwood et al.,2001b: fig. 9h,i). Traces of manufacture by scraping (Fig. 8b)are almost completely erased by use-wear. The second speci-men (Fig. 8c) is a fragment of shaft from a small mammal,scraped on the fracture edges to shape a point (Fig. 8d,e).Traces of scraping close to the tip are fresher than those foundon the edge and could represent resharpening of the tool; theseare covered by spots of ochre.

7. BBC engraved bone, M1 phase, square E5a, layer CB (noaccession number). This long-bone shaft fragment from a smallmammal (Fig. 9) shows a sequence of oblique incisions on theperiosteal surface. These were made by a cutting edge, as in-dicated by their thin ends and the presence of side striations.The bone was heated after being incised, which resulted inchipping of primary-bone microflakes and incision damageon the left side. This and the presence of compacted sedimentdemonstrate the antiquity of the lines. The unusual conver-gence of two groups of incisions suggests that they weredeliberate.

Fig. 5. Midshaft fragments from M1 phase bearing engravings on the periosteal surface and possible retouch at one end. Scale¼ 1 cm.


8. BBC engraved bone, M3 phase, square E5b, layer CJ (noaccession number). Made from a shaft fragment of a large-mammal long bone, this incised piece from the M3 phasedated at ca. 140 ka (Jacobs et al., 2006) shows traces of intenseheating and two parallel incisions on its periosteal surface(Fig. 10b). The sinuous outline of the incisions and their innermorphology suggest that they were incised by a very sharp V-shaped point. The irregular morphology of the edges indicates

Fig. 6. Close-up views of the shaft fragments from BBC M1 phase, depicted in

Figure 5, showing the contiguous, possibly anthropogenic, smoothed removals

present at one end (a) and the engravings close to the opposite end (b). Arrow

in 6b indicates linear impacts interpreted as the consequence of an occasional

use of the bone as a retoucher. Scale¼ 1 cm.

that the two lines were traced after the heating of the bone and,apparently, prior to the final breakage that reduced its shape tothe present state. This supports the interpretation that the inci-sions are ancient. However, the incisions look fresh, and thereis no accumulation of sediment. Most of the bone excavatedduring the 1993e1997 seasons at Blombos Cave was washedand scrubbed with a brush at the Iziko-South African Museumshortly after recovery. It is clear that this piece was washed(Fig. 10), and it is likely that any material that accumulatedin these incisions was removed during washing. The factthat the incisions are limited to the periosteal surface and pre-cede the breakage of the bone strongly suggests that they areancient, but we cannot entirely rule out excavation damage(i.e., by the tip of a metal tool) as their source.

Peers Cave

Peers Cave, also known as Skildergatkop Cave, is situatedin a quartzite outcrop near Fish Hoek on the Cape Peninsula.The first excavation was a test pit in 1925. Bertie and VictorPeers later excavated the site in the late 1920s (Peers, 1929;Jager, 1949; Deacon, 1992), and most of the deposit wasremoved at that time. The excavators found mostly MSA ma-terial, but their stratigraphic control was poor. Only selectedartifacts were kept and the balance was dumped on the talus.Some of the Peers Cave excavated material is now curatedat the Iziko-South African Museum. Bifacial points, typicalof the Still Bay, were recovered, as were Howiesons Poort(HP) lithics. Although their results were not fully published,Peers (1929) suggested in an abbreviated report that the StillBay was sandwiched between HP layers. Goodwin and vanRiet Lowe (1929) disputed this sequence and suggested in-stead that the HP was the precursor to the LSA. However,Peers did state that the ‘‘true’’ Still Bay lay beneath the HP(Peers, 1929: 9). The problem was later exacerbated in themid-1900s when some archaeologists regarded lanceolatepoints, the markers of the Still Bay, as an integral componentof the HP rather than as a temporally separate MSA entity(Goodwin, 1952; Malan, 1955). At Goodwin’s suggestion,Jolly (1948) started fresh excavations at the cave. He reportedthat Still Bay bifacials lay at the bottom of the sequence; HPwas above it, followed stratigraphically by the LSA. Unfortu-nately, this excavation was also neither well-controlled norwell-reported, and confusion continued about the position ofthe Still Bay. Later, a small excavation by Anthony (1963,1967) did not help resolve the problem. However, she believedthat the skull of the ‘‘Fish Hoek’’ man, reported by Peers ascoming from the MSA levels, actually derived from theLSA. Radiocarbon dating of the skull later confirmed thiswith a suggested date of ca. 4.8 ka (Minichillo, 2005).

Peers Cave bone point, Squares H3,4,5, Talus. The bonepoint (Fig. 11a) described below was found by Jolly in SquaresH3, 4, 5 and came from the sloping surface (talus?) probablyleft by the Peers excavation. Whether it derives from MSA orLSA levels was not recorded. It is the distal end of a pointmade on a shaft fragment of a Class II bovid. The tip is miss-ing because of postdepositional damage, and a thick layer of


Fig. 7. (a) Bone awl from the M2 phase (square E3, layer CG) at BBC with locations (brackets) of close-up views 7b (top bracket) and 7c (bottom bracket); (b, c)

scraping marks and grooves left by root etching. Scale in 7a¼ 1 cm; 7b,c¼ 1 mm.

manganese homogenously covers most of the surface, makingit difficult to identify traces of manufacture and use. The prox-imal break is also postdepositional in origin. A small area nearthis break, where manganese is absent, reveals traces of man-ufacture in the form of longitudinal grooves produced by anirregular cutting edge that has scraped the surface along thetool’s main axis (Fig. 11b). These grooves are smoothed andthe exposed areas are slightly polished, as if the tool surfacewas deliberately polished and the tool used before breakage.Thirteen close, elongated marks, covered by manganese stain-ing, are recorded on one aspect of the object in the half nearthe proximal break. They were produced by pressure with

the same stone tool, probably the tip of a broken flake/blade.Dimensions of the Peers Cave point are close to those of theBBC bone tools interpreted by Henshilwood et al. (2001b)as spear points (Fig. 11c,d), and they are significantly differentfrom those of LSA points from Nelson Bay Cave, as well asKalahari San ethnographic bone arrowheads (Fig. 12).

Peers Cave bone point: Carbon/nitrogen analysis. Boneprotein is known to degrade over time, resulting in the lossof carbon and nitrogen (C/N). Hence, bones from LSA andMSA levels will have considerably different concentrationsof these two elements. There may be intrasite variations inthese relative concentrations, and thus samples should be


Fig. 8. (a, c) Bone point tips from BBC M2 phase (square G5b, layer CFB/CFC); (b) close-up view of scraping marks smoothed by use-wear; (d, e) intense scraping

of the tip. Scale in 8a,c¼ 1 mm, 8b,d,e¼ 1 mm.

compared from the same site (Henshilwood and Sealy, 1997).In a previous test, two shaped and polished bone tools anda random selection of eleven animal bones recovered fromMSA and LSA levels at BBC were tested for relative per-centages of C/N. The results show that both bone tools testedhave the same C/N ratios as MSA bone from the site and arethus unambiguously from the MSA levels (Henshilwood andSealy, 1997; Henshilwood, 2005). To test for relative percent-ages of C/N, bone samples were drilled from (1) the Peersbone point, (2) five animal bones from the Peers HP levels,(3) the ‘‘Fish Hoek’’ skull (recovered from uncertain prove-nance at Peers Cave), and (4) five animal bones from thePeers LSA.

The C/N ratio for the Peers bone point (Fig. 13; Table 1) plotsat the lower limit of the range for the animal bones from HPlevels and is clearly distinct from the C/N ratios of the LSA an-imal bones from the same site and the skull. The ‘‘Fish Hoek’’skull has a value that lies just above that of the LSA bones and isnow known to date to ca. 4.8 ka (Minichillo, 2005). The resultindicates that the Peers point likely derives from the HP or olderMSA levels at this site, possibly the Still Bay.

Blombosch Sands

Blombosch Sands is a dune cordon that covers approxi-mately 3 km2 and lies to the northwest of BBC, with the


Fig. 9. (a) Incised bone fragment from M1 phase (square E5a, layer CB); (b) close-up view of a group of converging lines. Scale¼ 1 mm.

coastal edge of the dunes being just above the cave mouth. Un-til the 1960s, this area was a shifting dune field fed by aeoliansand derived from Blombos beach, situated to the west. Plant-ing of indigenous and exotic vegetation has stabilized thesedunes over the past 40 years. The sands were first describedby Heese (n.d.) in the 1920s during his searches for archaeo-logical material along the Blombos coastline and interior:‘‘In between these bones of those animals (elephant, rhino,buffalo) to the right of the track about a dozen small Wiltonmiddens were exposed . there are several graves here; lowerdown two skulls were seen protruding out of the black sand’’(Heese, n.d.: 54e55). The Still Bay type-site, first describedby Heese after his discovery of bifacial lanceolate pointswithin the Kleinjongensfontein dune cordon, lies about15 km east of Blombosch Sands. Heese’s correspondence indi-cates that he and the Abbe Breuil, the French prehistorian, vis-ited the Blombosch Sands area to collect lithics in the late1920s (Heese, n.d.). These were later shipped back to the Mu-see de l’Homme in Paris. The bone tool described below de-rives from an open site located in the Blombosch Sandscordon, but the exact location of its discovery is not recorded.Wind and water erosion in the dunes frequently caused defla-tion of archaeological materials onto a common hard surface,

usually calcrete or calcarenite, and it is possible it came froma mixed horizon. ‘‘Open sites may be the result of wind- orwater action . along the dunes where the sand is shiftedfrom season to season by the prevailing winds . the sitesare thereby considerably disturbed and difficult to locate’’(Heese, n.d.: 3). The materials recovered at Blombosch Sands,now curated at the Iziko-South African Museum, were re-cently examined by Minichillo (2005), who confirmed thepresence of diagnostic Still Bay bifacial points.

Blombosch Sands bone point, SAM-AA 603. This bone toolwas produced on a long-bone shaft fragment of a Class II/IIIherbivore (Fig. 14a). Root marks, particularly intense on themiddle and proximal portions of the object, partially erasetraces of manufacture, consisting of spindlelike grooves (visi-ble to the naked eye) that were produced by grinding the ob-ject on a coarse, hard stone. The orientation of these striationsindicates that the body of the object was shaped by moving theobject obliquely to its main axis while its tip was shaped, orresharpened, by moving it perpendicularly. The same tech-nique was used to create a flat butt (Fig. 14b), suggestingthat the object functioned as an awl rather than as a spearpoint. Technologically and morphologically, this tool is verydifferent from the awls discovered in BBC MSA layers. The


Fig. 10. (a) Incised bone fragment from M3 phase (square E5b, layer CJ); (b) close-up view of the two incised lines. Scale¼ 1 mm.

coarse grinding used to manufacture this tool, common at LSAsites, is absent on BBC MSA bone tools, and none of the nu-merous bone awls found in those layers is completely shapedor show a butt similar to this point. AMS dating may help re-solve the age of this piece, but in our opinion, it is likely to beLSA.

Klasies River

The Klasies River (KR) sites are situated on the southernCape coast, about 40 km from Cape St. Francis. The excavatedsites are ‘‘main site’’ (caves 1, 1b, 1c, and shelter 1a), a sepa-rate small shelter, 1d, and caves 2 and 5 (Singer and Wymer,1982; Wurz, 2000). Main-site deposits have produced lithic ar-tifacts, marine shell, bone, and human remains attributed to H.sapiens. In cave 1 and shelter 1a, the accumulation starts atca.110 ka. The entrance to cave 1 was eventually sealed offby deposit, and later the deposit in 1a became high enoughto provide access to cave 2. The stratigraphy is complex,and due to erosion and geological movement, some material

from 1a slumped into cave 1. Cave 1 and shelter 1a were ex-cavated by Wymer in 1967e1968 (Singer and Wymer, 1982).The deposits in the cave are well stratified, and Wymer exca-vated by following the layering. There is still some confusion,however, in the interpretation of some of the excavated layers(Wurz, 2000: 9), and much of the material from these excava-tions is not well-detailed. Human remains were recoveredfrom the MSA I and MSA II levels.

In 1984, H.J. Deacon (1995) began new excavations in cave1 and shelter 1a that concentrated on rehabilitation of majorcollapses within the site to (1) obtain furtherdbut unselecteddsamples of fauna, (2) initiate a new dating program for theMSA levels, and (3) resolve stratigraphic problems (e.g.,>Deacon and Geleijnse, 1988; Thackeray and Kelly, 1988;Wurz, 2000; Wurz et al., 2003). The Klasies River bone arti-facts described below come from the MSA layers excavatedby Wymer in cave 1 and shelter 1a (Singer and Wymer,1982: 115). The finds consist of: (1) a bone point (SAM-42160) found in layer 19 in shelter 1a at the base of the HP;(2) a shaft fragment with possible engravings (SAM-AA


26733) from layer 20 in shelter 1a; (3) a notched bone (SAM-AA 27069) from the disturbed top of Layer 15 in cave 1; and(4) a notched bone and possible fragment of the same bone(SAM-AA 31820 and 31819) from layer 36 in the lowerMSA II in shelter 1a. It seems likely that the disturbed topof layer 15 is probably layer 16 of cave 1, which equateswith layer 36 of shelter 1a, suggesting that the notched piecesare contemporary (Wurz, 2000: 117).

Fig. 11. (a) Bone point from Squares H3,4,5, Talus at Peers Cave (Jolly, 1948

excavation); (b) area close to the proximal break with traces of manufacture;

(c) two bone points and a fragment from BBC MSA interpreted by Henshil-

wood et al. (2001b) as spear points; (d) close-up view of a BBC MSA point

(top right) with traces of manufacture by scraping on the remaining medullar

surface and polish on the exposed areas, similar to that documented on the

Peers point. Scale in 11a,c¼ 1 cm; 11b,c¼ 1 mm.

Layers 19 and 20 fall within Singer and Wymer’s (1982)HP layers. A date of ca. 80e60 ka for the HP has been sug-gested (e.g., Deacon, 1989; Thackeray, 1989), but more recentdates derived from thorium-uranium analyses of limestones in-dicates that the Klasies HP falls within the 70 to 60 ka range(Vogel, 2001). Based on thermoluminescence results, Feathers(2002) proposed that the HP falls within a window of 55 to60 ka. Further HP dates for Klasies of ca. 58e48 ka comefrom Thackeray (2002) and a weighted mean date of ca.56 ka for the Klasies HP from Tribolo et al. (2005). A suiteof dates in press and published for Diepkloof, Boomplaas,Border Cave, Rose Cottage Cave, and Klasies indicatea mean date of 59� 6 ka for the HP (even with two standarddeviations, the HP dates do not exceed 65 ka) (e.g., Milleret al., 1999; Grun and Beaumont, 2001; Jacobs, 2004, pers.comm.). An important implication of the new dates for theStill Bay and the HP presented above is that temporal continu-ity is not a given and that 10 kyr may separate the two indus-tries. The notched fragments (3 and 4 above) come from layersequivalent to Deacon’s SAS member (Deacon and Geleijnse,1988) and fall within the MSA II, with deposition during ox-ygen isotope stage 5c/b. Layer 16 was dated at ca. 89 ka byBada and Deems (1975). Using the uranium-series method,Vogel (2001) dated stalagmites from Layer 14 in the MSA IIlevels to between 85 and 101 ka.

1. KR bone point: SAM-AA 42160, Howiesons Poort, layer10c (e?), cave 1a. This elongated point, probably made froma shaft fragment of a Class II bovid, has a postdepositionalbreak at one end (Fig. 15a). It has been shaped by scrapingthe blank longitudinally with a retouched lithic cutting-edge,as demonstrated by the presence of the typical thin, deepgrooves left by this technique (Fig. 15b). The reduction inthe size near the proximal end suggests that this was originallya double-pointed tool, morphologically similar to LSA(Fig. 15c) and ethnographic San bone arrow points (Fig. 15d).The shape of these San arrow points depends on the waythey were hafted: a thin point, similar to that described

Fig. 12. Size of San bone arrow heads and bone points from LSA (Nelson Bay

Cave and BBC) and MSA (BBC) compared with points from Peers Cave and

Klasies River.


Fig. 13. Percentage of carbon and nitrogen measured in various animal bones from the LSA and MSA layers at Peers Cave, the Peers bone point, and the Fish Hoek

skull from the same site.

here, and a thicker spindlelike linkshaft are connected to-gether and the latter to the wooden arrow shaft using a vege-tal thread. Both points and linkshafts similar to those knownethnographically have been found at LSA sites (Fagan andVan Noten, 1971; Deacon, 1976), suggesting that the samehafting system was used by southern African LSA people(Deacon, 1984). Dimensions of the KR point are close tothe mean of dimensions recorded on LSA points fromBBC and Nelson Bay Cave and close to the upper limit ofthe range of dimensions recorded on Kalahari San bone ar-row points (Fig. 12).

2. KR engraved bone: SAM-AA 26733, Howiesons Poort,layer 20, cave 1a. This piece is a midshaft fragment ofa limb bone from a very large bovid (Fig. 16a), as indicatedby the thickness of the compact bone (8.5 mm). The periostealsurface shows traces of ancient weathering in the form of lon-gitudinal breaks and, at its center, the bone surface is exfoli-ated. Subsequently, the bone underwent a gentle mechanicalabrasion that has left a highly polished appearance on someareas. Four parallel lines, engraved by a sharp lithic point,cross the periosteal surface perpendicular to the fragment’smain axis. They are heavily eroded and filled with compactedsediment in places, demonstrating their antiquity. Changes in

the direction of the lines when crossing incipient breaks pro-duced by weathering indicate that the bone was already weath-ered when the engravings were made (Fig. 16b,c). In addition,the three longer lines start on an old surface of fracture, indi-cating that the fragment was already detached from a largerpiece when it was engraved. The subrectangular shape of theobject, with one long and two short sides, is a result of threeancient breaks. The remaining long side, located close to thespecimen identification label, is the result of a postdepositionalbreak.

3. KR notched bone: SAM-AA 27069, MSA II, Layer 15þ,cave 1. Singer and Wymer (1982: 115, Fig. 8.1) describedthis object as a burnt fragment of scapula or rib with dentic-ulations cut along one edge (Fig. 17a). Examination of thebreaks shows a superficial brown layer that covers a blackinner one, demonstrating that the object was heated. Absenceof spongy bone and the very compact appearance of the6-mm-thick bone suggest that the fragment cannot comefrom a scapula; rather, it is probably from the limb boneof a large bovid, perhaps the posterior face of a tibia, consid-ering its flat morphology. In the first step in the production ofthis tool, the edge of the bone was scraped in order to obtaina round section. The flat surfaces were also scraped with an

Table 1

Results for relative percentages of carbon and nitrogen for bone items recovered from various stratigraphic levels at Peers Cave by Peers (1929) and Jolly (1948)

(see also Fig. 13)

Sample Excavator Depth Sq/Unit Industry Run 1 Run 2 Mean

C N C N C N

P1 Jolly 12e2400 CþD2þD3þD4 HP 5.69 0.87 6.05 0.88 6.01 0.88





P6 Peers 7 ft Fish Hoek Man Albany 9.28 2.14 9.39 2.07 9.34 2.11

P7 Jolly 12e1400 E2,3,4,5 Surface LSA 11.54 3.63 11.63 3.66 11.59 3.65

P8 Jolly Upper 100 C,D 3 and 4 Surface LSA 15.20 4.71 15.16 4.70 15.18 4.71

P9 Jolly Upper 100 C,D 3 and 4 Surface LSA 13.83 4.33 13.84 4.35 13.84 4.34

P10 Jolly 2e1200 CþD3 LSA 10.79 3.24 10.93 3.27 10.86 3.26

P11 Jolly 2e1200 Cþ 3D LSA 12.10 3.66 12.02 3.60 12.06 3.63

P12 Peers Bone point H3,4,5 Sloping surface 1.69 0.19 1.67 0.20 1.68 0.20


irregular/retouched lithic cutting-edge (Fig. 17b,c). Scrapingmotions were, during the latter work, parallel or slightly ob-lique to the tool edge. After the initial preparation, numeroussuperficial notches were carved on the edge. Microscopic

Fig. 14. (a) Bone awl from Blombosch Sands located near BBC; (b) detail of

the flat butt of the tool showing traces of grinding. Scale¼ 1 cm.

analysis of these marks (Fig. 17d,f) indicates that theywere made by a single stroke of different robust blades,probably retouched, crossing the bone’s edge perpendicularly.The edge was cut either across the apex or on alternate sidesof it. This resulted in an accumulation of shallow, sometimesoverlapping, incisions that are difficult to discriminate visu-ally. The notches differentially preserve their inner morphol-ogy. Some have fresh edges and preserve well the internalstriations that are visible on experimental incisions madewith the same technique, while others are smoothed grooveswith no or barely visible internal striations. Both very freshand heavily smoothed notches are found close to each otheron the edge. The reason for such a pattern is use-wear thatextensively affected the whole edge, which led to intensepolishing of the most salient areas (Fig. 17f). This wear isresponsible for the gradual fading of the notches. The freshappearance of some of the notches is the result of their hav-ing been carved when the bone surface and previously madenotches were already worn. In sum, the accumulation ofnotches on this bone edge can be interpreted as a continuousrejuvenation of the area of the tool during its active use. Thenotches probably served to increase the tool’s abrasiveness.The type of wear, the directions of the striations recordedon them (Fig. 17f), and the shallowness of the notches allsuggest that the tool was used on a soft material and movedparallel to its main axis. Microscopic residues of redpigments present in the depths of some of the notchesindicate that ochre may have been involved in this activity.The high polish that smoothes the prominent areas betweenthe grooves left by scraping on both flat aspects(Fig. 17b,c) likely resulted from the tool being held by thehand during use.

4. KR notched bone: SAM-AA 31820 (Fragment of SAM-AA 31819): MSA II, Layer 16-17, cave 1a. This object(Figs. 18, 19) was made from a rib of a large bovid, possiblyan eland. We were unable to refit the two fragments asconjoined by Singer and Wymer (1982: 116), probably asa consequence of recent damage to the fragment. With theexception of extensive damage by roots, both fragmentsbear postdepositional and human-induced modifications thatare similar to those described for KR 27069: striations onthe flat aspects parallel or slightly oblique to the bone’sedge, produced by scraping the bone surface with a retouchedstone tool, and unevenly spaced notches of different sizescarved, in this case, on the natural edges of the rib. Notcheson the large fragment are present only on a portion of theedge and have been almost entirely removed by postdeposi-tional damage (Fig. 18). Those made on the small fragment(Fig. 19b) are, in some cases, deeper than the notches on KR27069 and were produced by the to-and-fro movement ofa retouched cutting edge; in other cases, they correspond tosingle strokes, perhaps added to rejuvenate the abrasivenessof the tool’s edge (Fig. 18c). As on 27069, the notched edgesof this piece are worn. This wear shows an intense polishingthat is present only on prominent areas between the notches(Fig. 18c). Spots of red pigments are visible on the flat as-pects of both fragments.


Fig. 15. (a) SAM-AA 42160: Bone point from Klasies River, Howiesons Poort, layer 10c (e?) cave 1a, found by Singer and Wymer (1982); (b) detail of the surface

with longitudinal striations produced by scraping; (c) probable arrow points from the LSA layers at Nelson Bay Cave; (d) San arrow heads composed of a thin

arrow point and a thicker linkshaft connected with vegetal thread. Scale 15a,c,d¼ 1 cm; 15b¼ 1 mm.

Discussion

Technological, chemical, morphometric, and contextualanalyses of bone objects from four southern African MSAsites provide new data on bone-tool manufacture during theMSA. The artifactual nature of the objects examined here isbeyond doubt for almost all of the specimens, the only excep-tions being the incised piece (Fig. 10) from the M3 phase andthe shaft fragment from Blombos Cave that is modified at one

end through retouch (Figs. 5, 6). Caution is required when oneassesses the artifactual nature of bone modified by percussion,especially where modification is limited and there is no com-pelling evidence that the bone was used as a tool. Continuousretouch along the edge of a bone can result from naturalcauses, such as carnivore activity (Villa and Bartram, 1996)or deliberate bone breakage during marrow extraction(Binford, 1981; Bunn, 1981, 1982; Gifford-Gonzalez, 1989;Bonnichsen and Sorg, 1989; Peretto et al., 1996). For this


Fig. 16. (a) Bone fragment from Klasies River (SAM-AA 26733: Howiesons Poort, layer 20, cave 1a) bearing four parallel engraved lines; (b) close-up view dem-

onstrating the weathered appearance of the periosteal surface; (c) change of direction of the engraved line when crossing a crack produced by weathering. Scale in

16a¼ 1 cm; 16b¼ 5 mm; 16c¼ 1 mm.

reason, taphonomic analysis of the whole faunal assemblage(Backwell and d’Errico, 2005a,b) should be incorporated inany assessment of flaked bones from a site. Preliminary anal-ysis of the faunal remains from the MSA layers at BBC hasshown that humans were the main agents of accumulationand that there are very few modifications to bone resultingfrom animal activity (Henshilwood et al., 2001a,b). The typi-cal modifications we recorded on the BBC lithic retoucher arenot observed on other well-preserved large shaft fragments.These observations, including the presence of localized wearon the removals, five engraved lines, possibly representinga decoration reminiscent of that made on ochre fragments,and evidence of this fragment’s use as a lithic retoucher, sug-gest the occurrence of a bone-tool type previously unrecordedin the MSA.

The use of modified large limb-bone midshaft fragments astools is not new at BBC. A large metatarsal fragment from theStill Bay layers is thought to have been used as a soft hammerfor knapping (Henshilwood et al., 2001b: Fig. 10), and modi-fications through retouch are observed on other bone toolsfrom BBC (Henshilwood et al., 2001b: 250; see also below).A detailed analysis of bone modification in the MSA levelsat BBC now in progress (Thompson, pers. comm.) will assistin deciding whether modifications to some of the bones aredue to natural causes or are anthropogenic in origin.

The reliability of available data on the stratigraphic prove-nience, cultural affiliation, and dating of the specimens exam-ined here varies significantly from site to site and in somecases is challenged by our first-hand analysis of the objects.The four newly discovered specimens from BBC and thefour from Klasies River both come from ‘‘stratigraphic’’ exca-vations. Specimen SAM-AA 8975, the bone point from theDUN layers sandwiched between the LSA and MSA, is the

only object from BBC whose cultural affiliation appears un-certain. The morphology of this piece and the technologyused to produce it are different from those of all other MSAand LSA bone tools so far recovered at the site. It is thus dif-ficult to establish from the morphology alone the derivation ofthe tool. It seems possible but unlikely that it was depositedwhen the ‘‘BBC Hiatus’’ sand was accumulating above theStill Bay levels as, first, there is no other archaeological mate-rial within this level and, second, optically stimulated datingof single sand grains from the DUN layer shows this hiatuswas not disturbed after deposition (Jacobs et al., 2003a,b).The C/N analysis and AMS dating of this point in the futuremay determine whether it should be attributed to the LSA orMSA.

The awl made of bird bone (E3 AH; Fig. 7) is the only bonetool recovered thus far from the M3 phase at BBC. It is pos-sible that this tool migrated into the M3 phase from the M2phase above, which is rich in tools of this type (Henshilwoodet al., 2001b). Future excavations of the M3 phase at BBC mayshow whether bone was shaped for use as tools during the Ee-mian or earlier in the southern Cape.

Unfortunately, the contextual information concerning thepoints from Peers Cave and Blombosch Sands is poor. How-ever, the attribution of the Peers point to the MSA is stronglyindicated by the C/N ratio. An MSA age is also supported bythe size of the tool and the manufacturing technique, making itremarkably similar to the three projectile points from the MSAin BBC.

This is not the case for the Blombosch Sands point. Al-though the size of this object is comparable to the MSA pointsfrom BBC, the flat morphology of the base, shaped duringmanufacture, suggests that it was an awl. Fully manufacturedawls are absent among the 26 tools of this type recovered from


Fig. 17. Notched bone from Klasies River (Singer and Wymer excavations, SAM-AA 27069: MSA II, Layer 15þ, cave 1). Scale in 17a¼ 1 cm; 17b,c¼ 1 mm;

17e¼ 1 cm; 17f¼ 100 mm.

the MSA layers in BBC, and none exhibits a similarly flat-tened base. Additionally, this tool bears clear traces of manu-facture by grinding on a coarse matrix, a technique not presentin the MSA in BBC but which is virtually identical to themethod of bone-tool manufacture in the LSA levels at BBC

and other LSA sites. The evidence suggests the BlomboschSands awl dates to the LSA.

Marked or worked bones, together with unidentifiable bonefragments, may not have all been collected during the Wymerexcavations at Klasies River, as was apparently the case for


Fig. 18. Notched rib from Klasies River (Singer and Wymer excavations, SAM-AA 31819: MSA II, Layer 16-17, cave 1a). The specimen originally refitted with

the specimen shown in Figure 19. Scale¼ 1 cm.

much of the bone and shell (see Wurz, 2000). Consequently, itis difficult to establish whether bone tools were routinely man-ufactured during the MSA occupations at KR. No bone toolswere recovered during the Deacon excavations of the HPlevels, but this also seems to be the case at other HP sites.Our analysis of the bone point from the base of the HP levelsreported by Singer and Wymer (1982) suggests caution iswarranted when one makes chronological and/or culturalattributions to the bone tools from the KR excavations.Technologically, morphologically, and dimensionally, theSAM-AA 4216 bone point is very similar to arrow pointsfound at LSA sites. Technological convergence might be oneexplanation for this similarity, but if so, then why is thereonly one specimen from Klasies and why are similar bonepoints not found at other HP sites? It is possible that bone toolswere confined only to some phases of the MSA and that theHP is not one of these phases. On the other hand, in thecase of this particular point, it may be that it derives fromoverlying LSA layers and has been erroneously attributed tothe HP.

The engraved bone fragment from the HP and the threeheavily notched bonesdall bearing the same characteristicmodifications and use-weardfrom the MSA II levels are dis-similar to bone artifacts from LSA assemblages and so theirMSA provenience is very likely to be correct.

The location, technique of manufacture, and outline of theengraved lines on the retouched shaft fragment (Figs. 5, 6) andthe burnt fragment (Fig. 9) from Blombos Cave may have beenengraved with symbolic meaning. At least two other objectsdescribed here were not used as tools, or are tools that bearnonfunctional modifications that also could have had a sym-bolic connotation. No straightforward functional explanation

can be argued for the thin equidistant lines engraved on theheavily weathered bone fragment from Klasies River(Fig. 16). The same applies to the tiny marks incised on thePeers Cave point (Fig. 11) that are reminiscent of marks foundon one of the BBC spear points (Henshilwood et al., 2001b:Fig. 18). Perhaps these are marks of ownership, similar tothose made by San on their arrow heads.

Conclusion

Our aim in this study was to refute the uniqueness of theMSA bone-tool industry at Blombos Cave and to enlargeour knowledge of the technological and typological variabilityof MSA bone industries. Our research confirms that bone arti-facts shaped by scraping are present in selected MSA levels atsome sites, that bone tools are so far absent from the HP, andthat bone was used during the MSA for the retouching of stonetools and was occasionally shaped by percussion flaking. Moreinformation has emerged regarding the use of bone projectiles;in particular, the Peers Cave point confirms the use of bone-tipped projectiles by MSA people. The tagged point fromBBC demonstrates that there was variation in the shape ofthese points and in hafting techniques. Although discreet,the presence of deliberate, arguably symbolic, markings onMSA bone objects is also demonstrated by our analysis.This indicates that bone, in addition to ochre, was a suitablemedium to embody symbolic markings. Thus, bone toolsmay have formed part of a symbolically mediated society.The symbolic significance of the marine-shell beads and theengraved ochre pieces, taken with the regular manufactureand use of bone tools, finely made bifacial points, and theprobable ability to fish, suggests a cognitive-behavioral


package not previously associated with Middle Stone Agepeople. Newly recovered information from BBC and otherMSA sites in Africa (Vanhaeren et al., 2006) suggests the ca-pacity for these behaviors is likely to have evolved over a longperiod of time.

Fig. 19. (a) Notched rib from Klasies River [Singer and Wymer excavations,

SAM-AA 31820 (Fragment of SAM-AA 31819): MSA II, Layer 16-17, cave

1a]. The specimen originally refitted with the specimen illustrated in

Figure 18; (b) close-up view illustrating the deep notches produced by the

to-and-fro movement of a retouched cutting edge; (c) salient area between

two notches showing a polish that was partially removed by a groove

produced by a single passage of the cutting edge; scale in 19a¼ 1 cm;

19b,c¼ 1 mm.

The number of bone artifacts from confirmed MSA sites inAfrica remains small. Nevertheless, in the light of our results,we believe that future well-controlled excavations of MSAsites and the re-examination of purported bone tools followinga multifaceted approach will likely increase the quantity ofthese artifacts. Inferring the significance of bone-tool manu-facture among MSA people and determining whether andhow bone-tool manufacture changed during the MSA in differ-ent geographic and environmental settings remain a challengefor Africanist archaeologists.

Acknowledgements

We thank Dominique Armand, Patricia Davis, LindsayHooper, Royden Yates, Zenobia Jacobs, Karen van Niekerk,and Marian Vanhaeren. This work was supported by grantsto F.D. from the European Science Foundation (Origin ofMan, Language and Languages-EC Sixth FrameworkProgram, contract no. ERAS-CT-2003-980409), the ServiceCulturel of the French Embassy in South Africa, the FrenchMinistry of Research (ACI Espaces et Territoires); and toC.H. from the European Science Foundation (Origin of Man,Language and Languages).

References

Anthony, B.W., 1963. Report on excavations at Peers’ Cave, Fish Hoek, C.P.

MarcheJune 1963. Unpublished manuscript, South African Museum,

Cape Town.

Anthony, B.W., 1967. Excavations at Peers Cave, Fish Hoek, South Africa.

Palaeoecol. Afr. 2, 58e59.

Backwell, L.R., d’Errico, F., 2001. Evidence of termite foraging by Swartkrans

early hominids. Proc. Natl. Acad. Sci. U.S.A. 98, 1358e1363.

Backwell, L.R., d’Errico, F., 2005a. The first use of bone tools: a reappraisal of

the evidence from Olduvai Gorge, Tanzania. Palaeontol. Afr. 40, 95e158.

Backwell, L.R., d’Errico, F., 2005b. The origin of bone tool technology and

the identification of early hominid cultural traditions. In: d’Errico, F.,

Backwell, L. (Eds.), From Tools to Symbols: From Early Hominids to

Modern Humans. Wits University Press, Johannesburg, pp. 238e275.

Bada, J.L., Deems, L., 1975. Accuracy of dates beyond the 14-C dating limit

using the aspartic acid racemization reaction. Nature 255, 218e219.

Bailey, S.E., Hublin, J.-J., 2006. Dental remains from the Grotte du Renne at

Arcy-sur-Cure (Yonne). J. Hum. Evol. 50, 485e508.

Barham, L.S., 1998. Possible early pigment use in south-central Africa. Curr.

Anthropol. 39, 703e710.

Barham, L.S., 2002. Systematic pigment use in the middle Pleistocene of

south-central Africa. Curr. Anthropol. 31, 181e190.

Barham, L.S., Pinto Llona, A.C., Stringer, C., 2002. Bone tools from Broken

Hill (Kabwe) Cave, Zambia, and their evolutionary significance. Before

Farming 2, 1e12.

Bergman, C.A., 1987. Hafting and use of bone and antler points from Ksar

Akil, Lebanon. In: Stordeur, D. (Ed.), La main et l’outil e manches et

emmanchements prehistoriques. Travaux de la maison de l’Orient, Lyon,

pp. 117e126.

Binford, L.R., 1981. Bones: Ancient Men and Modern Myths. Academic

Press, New York.

Bonnichsen, R., Sorg, M.H., 1989. Bone Modification. University of Maine

Centre for the Study of the First Americans, Orono.

Brain, C.K., Shipman, P., 1993. The Swartkrans bone tools. In: Brain, C.K.

(Ed.), Swartkrans: A Cave’s Chronicle of Early Man. Transvaal Museum

Monograph 8, Pretoria, pp. 195e215.

Brooks, A.S., Helgren, D.M., Cramer, J.S., Franklin, A., Hornyak, W.,

Keating, J.M., Klein, R.G., Rink, W.J., Schwarcz, H., Leith Smith, J.N.,


Stewart, K., Todd, N.E., Verniers, J., Yellen, J.E., 1995. Dating and context

of three Middle Stone Age sites with bone points in the Upper Semliki

Valley, Zaire. Science 268, 548e553.

Bunn, H.T., 1981. Archaeological evidence for meat-eating by Plio-Pleisto-

cene hominids from Koobi Fora and Olduvai Gorge. Nature 291, 574e577.

Bunn, H.T., 1982. Meat-eating and human evolution: Studies on the diet and

subsistence patterns of Plio-Pleistocene hominids in East Africa. Ph.D.

Dissertation, Princeton University.

Cain, C.R., 2004. Notched, flaked and ground bone artefacts from Middle

Stone Age and Iron Age layers of Sibudu Cave, KwaZulu-Natal, South

Africa. S. Afr. J. Sci. 100, 195e197.

Campana, D., 1980. An analysis of the use-wear patterns on Natufian and Pro-

toNeolithic bone implements. Ph.D. Dissertation, Columbia University,

New York.

Choyke, A.M., Bartosiewicz, L., 2001. Crafting Bone: Skeletal Technologies

Through Time and Space. BAR International Series 937. Oxford.

Deacon, H., 1976. Where Hunters Gathered: A Study of Holocene Stone Age

People on the Eastern Cape. South African Archaeological Society Mono-

graph 1, Claremont.

Deacon, H.J., 1989. Late Pleistocene palaeoecology and archaeology in the

southern Cape, South Africa. In: Mellars, P., Stringer, C.B. (Eds.), The Hu-

man Revolution: Behavioural and Biological Perspectives of the Origins of

Modern Humans. Edinburgh University Press, Edinburgh, pp. 547e564.

Deacon, H.J., 1995. Two late Pleistocene-Holocene archaeological deposito-

ries from the southern Cape, South Africa. S. Afr. Archaeol. Bull. 50,

121e131.

Deacon, H.J., Geleijnse, V.B., 1988. The stratigraphy and sedimentology of the

main site sequence, Klasies River, South Africa. S. Afr. Archaeol. Bull. 43,

5e14.

Deacon, J., 1984. Later Stone Age people and their descendants in southern

Africa. In: Klein, R.G. (Ed.), Southern African Prehistory and Paleoenvir-

onments. Balkema, Rotterdam, pp. 221e328.

Deacon, J., 1992. Peers Cave, ‘the cave the world forgot’. The Digging Stick 9, 2.

d’Errico, F., 1993. La vie sociale de l’art mobilier paleolithique. Manipulation,

transport, suspension des objets en os, bois de cervides, ivoire. Oxf. J.

Archaeol. 12, 145e174.

d’Errico, F., 1995. A new model and its implications for the origin of writing:

La Marche antler revisited. Cambridge Archaeol. J. 5, 3e46.

d’Errico, F., 1998. Palaeolithic origins of artificial memory systems: an evolu-

tionary perspective. In: Renfrew, C., Scarre, C. (Eds.), Cognition and

Material Culture: The Archaeology of Symbolic Storage. McDonald Insti-

tute Monographs, Cambridge, pp. 19e50.

d’Errico, F., 2003. The invisible frontier: A multiple species model for the

origin of behavioral modernity. Evol. Anthropol. 12, 188e202.

d’Errico, F., Backwell, L.R., 2003. Possible evidence of bone tool shaping by

Swartkrans early hominids. J. Archaeol. Sci. 30, 1559e1576.

d’Errico, F., Backwell, L.R., 2005. Searching for common ground in palaeoan-

thropology, archaeology and genetics. In: d’Errico, F., Backwell, L.R.

(Eds.), From Tools to Symbols: From Early Hominids to Modern Humans.

Wits University Press, Johannesburg, pp. 1e8.

d’Errico, F., Giacobini, G., Puech, P.-F., 1984. Varnish replicas. A new method

for studying worked bone surfaces. OSSA, Int. J. Skeletal Res. 9/10, 29e51.

d’Errico, F., Henshilwood, C., Lawson, G., Vanhaeren, M., Tillier, A.-M.,

Soressi, M., Bresson, F., Maureille, B., Nowell, A., Lakarra, J.,

Backwell, L., Julien, M., 2003. Archaeological evidence for the origins

of language, symbolism and music. An alternative multidisciplinary per-

spective. J. World Prehist. 17, 1e70.

d’Errico, F., Henshilwood, C.S., Nilssen, P., 2001. An engraved bone fragment

from ca. 75 kyr Middle Stone Age levels at Blombos Cave, South Africa:

implications for the origin of symbolism and language. Antiquity 75,

309e318.

d’Errico, F., Henshilwood, C., Vanhaeren, M., van Niekerk, K., 2005. Nassar-

ius kraussianus shell beads from Blombos Cave: Evidence for symbolic

behaviour in the Middle Stone Age. J. Hum. Evol. 48, 3e24.

d’Errico, F., Julien, M., Liolios, M., Vanhaeren, M., Baffier, D., 2004. Many

awls in our argument. Bone tool manufacture and use from the Chatelper-

ronian and Aurignacian layers of the Grotte du Renne at Arcy-sur-Cur. In:

Zilh~ao, J., d’Errico, F. (Eds.), The Chronology of the Aurignacian and of

the Transitional Technocomplexes: Dating, Stratigraphies, Cultural Impli-

cations. Instituto Portugues de Arqueologia, Lisbon, pp. 247e270.

Evans, U., 1994. Hollow Rock Shelter, a Middle Stone Age site in the Ceder-

berg. S. Afr. Field Archaeol. 3, 63e72.

Fagan, B.M., Van Noten, F., 1971. The HuntereGatherers of Gwisho. Musee

Royal de l’Afrique Centrale, Tervuren.

Feathers, J.K., 2002. Luminescence dating in less than ideal conditions: case

studies from Klasies River Mouth and Duinefontein, South Africa.

J. Archaeol. Sci. 29, 177e194.

Gifford-Gonzalez, D., 1989. Ethnographic analogues for interpreting modified

bones: some cases from east Africa. In: Bonnichsen, R., Sorg, M.H. (Eds.),

Bone Modification. University of Maine Centre for the Study of the First

Americans, Orono, pp. 179e246.

Goodwin, A.J.H., 1952. Early Man on the Cape Peninsula. In: Mabbut, J.A.

(Ed.), The Cape Peninsula. Maskew Miller, Cape Town, pp. 124e143.

Goodwin, A.J.H., van Riet Lowe, C., 1929. The Stone Age cultures of South

Africa. Ann. S. Afr. Mus. 27, 1e289.

Grine, F.E., Henshilwood, C.S., 2002. Additional human remains from

Blombos Cave, South Africa: (1999e2000 excavations). J. Hum. Evol.

42, 293e302.

Grine, F.E., Henshilwood, C.S., Sealy, J.C., 2000. Human remains from Blombos

Cave, South Africa: (1997e1998 excavations). J. Hum. Evol. 37, 755e765.

Grun, R., Beaumont, P.B., 2001. Border Cave revisited: a revised ESR chro-

nology. J. Hum. Evol. 40, 467e482.

Hannus, L.A., Rossum, L., Winham, R.P., 1997. Proceedings of the 1993 Bone

Modification Conference, Hot Springs, South Dakota. Occassional Paper

no. 1. Archeology Laboratory, Augustana College, Sioux Falls.

Heese, C.H.T.D. Archaeological Studies in the Cape Province. Unpublished

manuscript. Altstadt, Riversdale, Cape Province on file at Iziko: South

African Museum, Cape Town, n.d.

Henshilwood, C.S., 1995. Holocene archaeology of the coastal Garcia State

Forest, Southern Cape, South Africa. Ph.D. Dissertation. University of

Cambridge, Cambridge.

Henshilwood, C.S., 2005. Stratigraphic integrity of the Middle Stone Age

levels at Blombos Cave. In: d’Errico, F., Backwell, L. (Eds.), From Tools

to Symbols: From Early Hominids to Modern Humans. Wits University

Press, Johannesburg, pp. 441e458.

Henshilwood, C.S., Sealy, J.C., Yates, R., Cruz-Uribe, K., Goldberg, P.,

Grine, F.E., Klein, R.G., Poggenpoel, C., Van Niekerk, K., Watts, I.,

2001a. Blombos Cave, Southern Cape, South Africa: Preliminary report

on the 1992e1999 excavations of the Middle Stone Age levels.

J. Archaeol. Sci. 28, 421e448.

Henshilwood, C.S., d’Errico, F., Marean, C.W., Milo, R.G., Yates, R., 2001b.

An early bone tool industry from the Middle Stone Age at Blombos Cave,

South Africa: implications for the origins of modern behaviour, symbolism

and language. J. Hum. Evol. 41, 631e678.

Henshilwood, C.S., d’Errico, F., Vanhaeren, M., Van Niekerk, K., Jacobs, Z.,

2004. Middle Stone Age shell beads from South Africa. Science 304, 404.

Henshilwood, C.S., d’Errico, F., Yates, R., Jacobs, Z., Tribolo, C.,

Duller, G.A.T., Mercier, N., Sealy, J.C., Valladas, H., Watts, I.,

Wintle, A.G., 2002. Emergence of modern human behaviour: Middle

Stone Age engravings from South Africa. Science 295, 1278e1280.

Henshilwood, C.S., Marean, C.W., 2003. The origin of modern human behav-

iourdcritique of the models and their test implications. Curr. Anthropol.

44, 627e651.

Henshilwood, C.S., Sealy, J., 1997. Bone artefacts from the Middle Stone Age at

Blombos Cave, southern Cape, South Africa. Curr. Anthropol. 38, 890e895.

Inskeep, R.R., 1987. Nelson Bay Cave, Cape Province, South Africa: The

Holocene Levels. BAR International Series 357 (I) and (II), Oxford.

Jacobs, Z., 2004. Development of luminescence techniques for dating Middle

Stone Age sites in South Africa. Ph.D. Dissertation, University of Wales,

Aberystwyth.

Jacobs, Z., Duller, G.A.T., Wintle, A.G., 2003a. Optical dating of dune sand

from Blombos Cave, South Africa: IIdSingle grain data. J. Hum. Evol.

44, 613e625.

Jacobs, Z., Wintle, A.G., Duller, G.A.T., 2003b. Optical dating of dune sand

from Blombos Cave, South Africa: IdMultiple grain data. J. Hum.

Evol. 44, 599e612.


Jacobs, Z., Duller, G.A.T., Wintle, A.G., Henshilwood, C.S., 2006. Extending

the chronology of deposits at Blombos Cave, South Africa, back to 140 ka

using optical dating of single and multiple grains of quartz. J. Hum. Evol.

51, 255e273.

Jager, H.S., 1949. The Peers’ Cave, Tunnel Cave and Rock Shelters at Skilder-

gat, Fish Hoek: The Home of Pre-Historic Man. Fish Hoek Municipality,

Fish Hoek.

Jolly, K., 1948. The development of the Cape Middle Stone Age in the Skil-

dergat Cave, Fish Hoek. S. Afr. Archaeol. Bull. 3, 106e107.

Knecht, H., 1997. Projectile points of bone, antler and stone: experimental

explorations of manufacture and use. In: Knecht, H. (Ed.), Projectile Tech-

nology. Plenum Press, New York, pp. 191e212.

Leakey, M.D., 1971. Olduvai Gorge, Volume 3: Excavations in Beds I and II,

1960e1963. Cambridge University Press, Cambridge.

Malan, B.D., 1955. The archaeology of Tunnel Cave and Skildergat Kop, Fish

Hoek, Cape of Good Hope. S. Afr. Archaeol. Bull. 10, 3e9.

Marean, C.W., Nilssen, P.J., 2001. The Mossel Bay Archaeological Project

(MAP): Background and Preliminary Results from Test Excavations at

Pinnacle Point, Mossel Bay. Progress report prepared for the South African

Heritage Resource Agency, Stony Brook University, New York.

McBrearty, S., Brooks, A.S., 2000. The revolution that wasn’t: a new interpre-

tation of the origin of modern human behavior. J. Hum. Evol. 59,

453e563.

Miller, G.H., Beaumont, P.B., Deacon, H.J., Brooks, A.S., Hare, P.E.,

Jull, A.J.T., 1999. Earliest modern humans in South Africa dated by isoleu-

cine epimerization in ostrich eggshell. Quatern. Sci. Rev. 18, 1537e1548.

Minichillo, T., 2005. Middle Stone Age lithic study, South Africa: An exam-

ination of modern human origins. Ph.D. Dissertation, University of

Washington.

Newcomer, M.H., 1974. Study and replication of bone tools from Ksar Akil

(Lebanon). World Archaeol. 6, 138e153.

Parkington, J., Poggenpoel, C., Rigaud, J.P., Texier, J.P., 2005. From tool to

symbol: the behavioural context of intentionally marked ostrich eggshell

from Diepkloof, Western Cape. In: d’Errico, F., Backwell, L. (Eds.),

From Tools to Symbols: From Early Hominids to Modern Humans. Wits

University Press, Johannesburg, pp. 475e492.

Peers, B., 1929. Preliminary report on the archaeology of the Fish Hoeke

Noord Hoek valley. Unpublished manuscript, South African Museum,

Cape Town.

Peretto, C., Anconetani, P., Crovetto, C., Evangelista, L., Ferrari, M.,

Giusberti, G., Thunhohesten, U., Vianello, F., 1996. Approccio sperimen-

tale alla comprensione delle attivita di sussistenza condotte nel sito di Iser-

nia La Pineta (Molise-Italia). La Frattura intenzionale. In: Peretto, C. (Ed.),

I reperti paleontologici del giacimento do Isernia La Pineta. Cosmo Ian-

none Editore, Isernia, pp. 187e451.

Pinto Llona, A.C., Andrews, P., Barham, L.S., 2000. Bone tools. In:

Barham, L. (Ed.), The Middle Stone Age of Zambia, South Central Africa.

Western Academic & Specialist Press, Bristol, pp. 122e128.

Schirmer, G.R., 1975. An analysis of lithic material from Dale Rose Parlour,

Trappies Kop, Kalk Bay, Cape Peninsula. Unpublished archaeology addi-

tional report, University of Cape Town.

Shipman, P., 1989. Altered bones from Olduvai Gorge, Tanzania: Techniques,

problems, and implications of their recognition. In: Bonnichsen, R.,

Sorg, M.H. (Eds.), Bone Modification. University of Maine Centre for

the Study of the First Americans, Orono, pp. 317e334.

Shipman, P., Rose, J., 1988. Bone tools: An experimental approach. In:

Olsen, S.L. (Ed.), Scanning Electron Microscopy in Archaeology. BAR

International Series 452, Oxford, pp. 303e335.

Singer, R., Wymer, J., 1982. The Middle Stone Age at Klasies River Mouth in

South Africa. Chicago University Press, Chicago.

Thackeray, A.I., 1989. Changing fashions in the Middle Stone Age: the stone

artefact sequence from Klasies River Mouth main site, South Africa. Afr.

Archaeol. Rev. 7, 33e57.

Thackeray, J.F., 2002. Comment on the age of late Pleistocene cave deposits

associated with the ‘‘Howieson’s Poort’’ stone tool industry in South

Africa. Transvaal Museum website. http://www.nfi.org.za/palaeo/Howie

sons_Poort.htm.

Thackeray, A.I., Kelly, A.J., 1988. A technological and typological analysis of

Middle Stone Age assemblages antecedent to the Howiesons Poort (HP) at

Klasies River main site. S. Afr. Archaeol. Bull. 43, 5e26.

Tribolo, C., Mercier, N., Valladas, H., 2005. Chronology of the Howiesons

Poort and Still Bay techno-complexes: assessment and new data from

luminescence. In: d’Errico, F., Backwell, L. (Eds.), From Tools to Sym-

bols: From Early Hominids to Modern Humans. Wits University Press,

Johannesburg, pp. 493e511.

Tribolo, C., Mercier, N., Selo, M., Joron, J.-L., Reyss, J.-L., Henshilwood, C.,

Sealy, J., Yates, R., 2006. TL dating of burnt lithics from Blombos Cave

(South Africa): further evidence for the antiquity of modern human behav-

iour. Archaeometry 48, 341e357.

Vanhaeren, M., d’Errico, F., Stringer, C., James, S.L., Todd, J.A.,

Mienis, H.K., 2006. Middle Paleolithic shell beads in Israel and Algeria.

Science 312, 1785e1788.

Villa, P., Bartram, L., 1996. Flaked bone from a hyaena den. Paleo 8, 143e159.

Villa, P., d’Errico, F., 2001. Bone and ivory points in the Lower and Middle

Paleolithic of Europe. J. Hum. Evol. 41, 69e112.

Vogel, J.C., 2001. Radiometric dates for the Middle Stone Age in South

Africa. In: Tobias, P.V., Raath, M.A., Moggi-Cecchi, J., Doyle, G.A.

(Eds.), Humanity from African Naissance to Coming MillenniadCollo-

quia in Human Biology and Palaeoanthropology. Florence University

Press, Florence, pp. 261e268.

Volman, T.P., 1984. Early prehistory of southern Africa. In: Klein, R.G. (Ed.),

Southern African Prehistory and Paleoenvironments. Balkema, Rotterdam,

pp. 169e220.

Wurz, S., 1999. The Howieson’s Poort backed artefacts from Klasies River: an

argument for symbolic behavior. S. Afr. Archaeol. Bull. 169, 38e50.

Wurz, S., 2000. The Middle Stone Age at Klasies River. Ph.D. Dissertation,

University of Stellenbosch.

Wurz, S., le Roux, N.J., Gardner, S., Deacon, H.J., 2003. Discriminating

between the end products of the earlier Middle Stone Age sub-stages at

Klasies River using biplot methodology. J. Archaeol. Sci. 30, 1107e1126.

Yellen, J.E., 1998. Barbed bone points: tradition and continuity in Saharan and

sub-Saharan Africa. Afr. Archaeol. Rev. 15, 173e198.

http://www.nfi.org.za/palaeo/Howiesons_Poort

http://www.nfi.org.za/palaeo/Howiesons_Poort

d’Errico, F. & Henshilwood, C.S. 2007. Additional evidence for bone technology in the southern...

Documents

Transcript of d’Errico, F. & Henshilwood, C.S. 2007. Additional evidence for bone technology in the southern...