HUMAN

EVOLUTION

Sadr K. School of Geography, Archaeology and Environmental Studies University of the Witwatersrand Private bag 3 Wits 2050 South Africa sadrk @geoarc, wits.ac.za

Keywords: South Africa, first millennium AD, Archaeological work.

Vol. 2 0 - n. 4 ( 2 1 7 - 2 3 0 ) - 2 0 0 5

From foraging to herding: the west coast of South Africa in the first millennium AD Current history textbooks state that the first livestock and ceramics were brought to the southern tip of Africa by Khoespeaking herders from northern Botswana. Recent archaeological work on the west coast of South Africa, however, contradicts this scenario. The evidence points instead to local adoption of livestock and ceramic technology. In the absence of any evidence for immigrants, it is assumed that livestock and the associated new ideas were passed along from one group of southern African foragers to another, eventually reaching the west coast around 2000 years ago. At first, the new items seem to have had little effect on the local economy. But by the mid-first millennium AD a new type of site had appeared in the landscape. These sites, which have an abundance of sheep and seal bones, as well as broken pots used for pouring liquids, were perhaps feasting sites. They suggest a more complex social and economic organization than had hitherto been suspected for this time and place.

Introduction Europeans first came into contact with some of the inhabitants of southern Africa when Bartholomew Diaz rounded the Cape of Good Hope in 1488. By 1652, the occasional willingness of the locals to barter their livestock had led to the establishment of the first permanent settlement of Europeans at what is now Cape Town. Where did the local herders, the so-called Hottentots, and their livestock come from? This question has interested historians for some time and many different answers have been put forward (for a short list see Sadr 1998: 102). Today, the majority view is that the Hottentots were Khoe-speakers who had originally come from what is now northern Botswana (Elphick 1985; Smith 1992; Boonzaier et al., 1996). It is thought that they acquired livestock and the knowledge of herding from immigrant, Bantu-speaking farmers. With their newly acquired livestock, the Khoekhoe branch of the Khoe-speakers are thought to have migrated to the west and south coasts of southern Africa, where they co-existed with the local Bushmen hunter-gatherers until Europeans and smallpox erased both cultures in the early 18th Century.

218

SADR

The majority view, however, is beginning to unravel. Recent studies have revealed a number of problems. For example, the earliest sheep bones on the south and west coasts have now been directly dated to the first centuries AD and BC (Sealy & Yates 1994; Henshilwood 1996), a time when the Bantu-speaking farmers and herders perhaps had not yet left the Lake Victoria area (Phillipson 1993:188). The question of how the indigenous southern Africans first acquired livestock has thus been re-opened. Another problem recently revealed is that the earliest ceramics across southern Africa show too much stylistic variation to fit the majority view (Sadr 1998). If immigrant Khoe-speakers had introduced the earliest ceramics along with the first sheep, one might have expected more similarities in decoration styles and vessel shapes between the earliest pots on the west coast of South Africa and those in northern Botswana. A third problem is evident in the excavated animal bones. The majority view implies a long stretch of pastoralism on the west coast of South Africa since 2000 years ago, but the animal bones from the sites excavated there only show a few isolated peaks in livestock counts (Sadr 2002: table 1). Pastoralism was perhaps not as constant as the majority view implies.

Table 1, Comparative statistics of the six Kasteelberg sites. 1

2

3

4

5

6

7

8

9

10

11

12

13

g

=o

|

E

~

O {:

|

~

,~

~

.-=

,~

:

~

~

.12

(/) KBA KBD KBE KBM KBN KBG 1-2 I KBG 1-2 It KBG 3-5 I KBG 3-5 II KBG 3-5 III KBG 3-5 IV KBG 3-5 V

tj AD 813-1007 AD 789-1020 AD 581-774

z

190 168 78

z

z

z

t,~

u

428 574 716

4789 14098 1688

11439 6954 3637

99 99 99

69 77 90

203 1123 66

113 76 66

1,31 2,12 0,76

3 2 0

1,6 1,2 0

321 798 623 634 1985 1053 1606 901 63 & 6.

81 96 69 78 59 68 68 65 40

14 22 4,3 5,3 4,6 4,1 0,4 0 0

12 84 2 6 8 3 6 3 2

2 20 0 0 3 7 4 4 0

0,47 1,36 0,26 0,22 0,39 0,31 0,38 0,19 0,07

3 0 4 9 45 30 33 20 8

1,8 0 0,9 3,4 2,2 4,2 2,6 2,9 4,5

53BC-AD80 168 28 383 AD 744-972 119 34 1753 465 21 385 AD 869 43-23 BC 267 15 285 2016 97 820 AD 642-882 AD 444-1007 719 31 435 1693 BC 1264 5 1085 2575 BC 679 0 344 177 0 56 > 2575 BC * percentage of columns 4, 5 & 6 in total of columns 3 , 4 , 5 ** percentage of column 4 in total of columns 3 & 4.

z

z

~

z

FROM FORAGING TO HERDING: THE WEST COAST OF SOUTH AFRICA

219

Problems like these weaken the majority view and allow the counter-proposition that the earliest livestock did not reach the south and west coasts of South Africa by a migration of Khoe-speaking herders from the north, but by a process of diffusion whereby indigenous hunter-gatherer groups passed on the knowledge of herding and ceramic manufacture, as well as live sheep, along existing exchange networks (Sadr 1998, 2003). This would better account for the distinct regional styles in ceramics, and for the variable emphasis on herding in different parts of southern Africa at different times. This alternative model in effect says that various groups of local foragers across southern Africa adopted herding around 2000 years ago. To test this alternative view on the west coast, six archaeological sites were excavated on a hill called Kasteelberg, located 150 km north of Cape Town on the Vredenburg peninsula (Figs. 1 & 2). The relevant results are briefly outlined below. Overall, the six sites tend to refute the majority view (Sadr et al., 2003). They support instead the alternative view that local foragers first adopted herding in this area. Below are presented several possible scenarios to account for the economy and social organization of the west coast herder-foragers during the first millennium AD.

Background to archaeological research on Kasteelberg That Kasteelberg contains rich, later Stone Age, ceramic-bearing sites had been known for some time (Rudner 1968). Starting in the 1980's, Andrew Smith from the University of Cape Town excavated four sites on Kasteelberg. These proved to be rich in ceramics as well as sheep bones, and the name Kasteelberg came to be associated with the earliest remains of Khoekhoe herders on the Atlantic coast of South Africa (Klein 1986; Smith 1992). Particularly rich were the two sites known as KBA and KBB, which together seemed to cover the full sequence of the last 2000 years. In an effort to prove that the Khoekhoe herders had remained economically and culturally separate from the San hunter-gatherers during the past two thousand years (contra Elphick 1977, 1985; Schrire & Deacon 1989), Andrew Smith and his colleagues found what they thought were distinct archaeological signatures for San foragers sites and those of Khoekhoe herders (Smith et al., 1991). Subsequently, Sadr (et al., 1992) conducted an archaeological survey on the Vredenburg peninsula to further test the validity of this "two cultures: two economies" model.

220

SADR

Pen

~okm Fig 1. The west coast of South A.fi'ica.

Fig 2. Sites on Kasteelberg.

f/,ow~

~

~1

FROM FORAGING TO HERDING: THE WEST COAST OF SOUTH AFRICA

221

The results were mixed. On the positive side, the survey confirmed that the first millennium AD sites in this area could be divided in two groups on the basis of artifact frequencies. In one group ceramics were common while retouched stone tools were rare; in the other, this pattern was reversed. On the negative side, however, there seemed to be little difference in artifact styles, hence perhaps "cultures," of these two groups of sites. On sites of both groups the same type of ceramics could be found and the stone tools were also similar in technology of manufacture and choice of raw material. The only evidence for a distinction between the occupants of the two groups of sites that could be interpreted as cultural rather than functional was a difference in ostrich eggshell bead diameters (Smith et al., 1991). The results of the survey were thus inconclusive and could be used both ways: to support the Smith model, or that of Elphick, Schrire and Deacon who thought that the Khoekhoe herders and San foragers had largely mixed their cultures and economies in the first millennium AD. To shed more light on the subject, a project was launched in 1999 to excavate six sites on Kasteelberg (supported financially by the Swan Fund, the Wenner-Gren Foundation, and the ACACIA project of the University of Cologne, the Universities of Botswana, Witwatersrand and Cape Town). Surface collected artifacts suggested that all six sites were of the first millennium AD, and that three of them (KBA, D & E) had the traits we expected from Khoekhoe herder sites, while the other three (KBM, N & G) showed the characteristics we expected from San hunter-gatherer sites (for details of our expectations see Smith et al., 1991; Sadr et al., 2003). Radiocarbon dates subsequently confirmed that all six sites dated to the first millennium AD. KBG has the longest sequence extending back to at least 4000 years ago (Fig. 3). The presumed Khoekhoe herder sites KBA, D and E (together abbreviated to ADE) have no dates older than the mid-first millennium AD, while the presumed San hunter-gatherer sites, KBM, N & G (abbreviated to MNG), cover a longer sequence and show much continuity in their material culture since the mid-Holocene. At first glance, this seems to support the majority view: indigenous San versus newly arrived Khoekhoe herders. Things, however, proved to be more complex. First, faunal analyses showed that the first millennium occupations at all six sites contained a significant proportion of sheep bones relative to hunted mammals (Sadr et al., 2003: 30). This was contrary to our expectation that the hunter-gatherer sites should have no sheep bones while the herder sites should have many. Thus, a simple distinction between herders and hunter-gatherers was not evident in the faunal remains at the six sites. Second, we found a curious pattern in that the proportions of protein obtained on the shoreline were higher in the ADE sites (Sadr et al., 2003: 30). This suggested that

222

SADR

1500 KBA 1000

500 AD 1 BC

-"

KBE)

KBN i --~

|W~mn

I

B

KBE

K~3~v~

500 KBG 1000 1500 2000 2500

|p

3000

Fig 3. Calibrated radiocarbon dates flora the six Kasteetberg sites.

instead of herders vs. hunter-gatherers, these six Kasteelberg sites might represent something like a shoreline- and an inland-oriented population. Third, analyses of ostrich eggshell beads from the first millennium AD occupations showed no significant size differences (Sadr et al., 2003:31). Our expectation, following Smith et al., 1991, was that the ADE sites would have beads on average larger than 6 mm in diameter while the MNG sites would have beads on average smaller than 5 mm. In fact, all six first millennium AD occupations had average bead sizes over 6mm. It was only in the pre-first millennium AD levels at KBG that small beads were found. In other words, the bead sizes showed an increase through time, but no indication for a cultural distinction between the shoreline- and inland-oriented first millennium AD populations on

FROM FORAGING TO HERDING: THE WEST COAST OF SOUTH AFRICA

223

Kasteelberg. Thus, the majority view of culturally and economically distinct Khoekhoe herders and San hunter-gatherers on the west coast during the first millennium AD seemed refuted. The question became: "how then did the occupants of these six sites relate to each other?" Several answers seemed possible. These are briefly described below.

Social and economic dynamics on first millennium Kasteelberg The first alternative explanation was that the shoreline and the inland oriented sites on Kasteelberg represented seasonal occupations by the same group of people. In support, it could be pointed out that the MNG sites generally have less bone and artifacts per cubic meter of deposits than the ADE sites (Sadr et al., 2003: 29-30). This suggested they were occupied repeatedly, but each time for a short duration. Two of these MNG sites, KBM & N, are located on the southern side of the hill, where they are exposed to the strong summer winds that batter Kasteelberg today. This suggested they might only have been occupied in winter, when they would have been well protected from the prevailing northerly winds. The idea of a winter occupation for the MNG sites was complemented by the position of KBA and D atop Kasteelberg. In the winter of 1993 it was observed that low clouds often obscured the crown of Kasteelberg. It seemed that the hilltop sites KBA and D might thus represent summer occupations. It helped the scenario that KBE was in a position exposed to the northerly winds of winter, and thus perhaps only occupied in the summer. But the faunal remains oppose this agreeable scenario: seal bones are particularly common on the supposedly summer occupied ADE sites. Yet the faunal analysts noted that seal pups, the bones of many of which are represented on KBA, wash up mainly in the winter (Klein & Cruz-Uribe 1989). This contradicts the idea of summer occupation on the hilltop, and as a result the seasonal scenario seems less tenable now. A second hypothesis to explain the relation between the Kasteelberg shoreline- and inland-oriented sites emphasizes a slow transition to more intensive use of marine resources. In this scenario, the inland-oriented MNG sites represent the conservative element among the population, with an occupation history that at KBG can be traced back to at least 4000 years ago. The shoreline-oriented ADE sites, none of which are older than the mid-first millennium AD, represent the new economy. In this scenario, the transition to intensive use of marine resources was completed in the first few centuries of the second millennium AD, with a major shift in settlement distributions to coastal sites (Sadr et al., 1992). Our survey in 1992 seemed to support this hypothesis.

224

SADR

Alas, preliminary results from a major radiocarbon dating project now underway are casting doubts on the validity of the early second millennium AD coastal settlement pattern. Most sites during that period seem to have been inland occupations. If that turns out to be the case, the argument that the late first millennium shoreline-oriented sites represent an economic transition becomes less convincing. A third hypothesis to explain the relations between the first millennium AD sites on Kasteelberg concerns social inequality. In this scenario, higher status members of the community occupied the ADE sites, while commoners occupied the lower lying MNG sites. This hypothesis is supported by two observations. First, a physically higher location is often a sign of status in many parts of the world, including in pre-colonial southern Africa. Secondly, the hilltop ADE sites have an abundance of ceramic pots and sheep bones compared with the MNG sites further down the hill (Table 1). Given that they were only introduced to this area a few centuries earlier, both sheep and ceramics may have been considered valuable exotics on Kasteelberg. If that were the case, their abundance on the hilltop sites could be interpreted as an indication of the occupants' wealth. The weak link in this scenario is the site of KBB, which lies near the base of Kasteelberg. Excavated by Andy Smith in the mid-1980s, KBB also has layers dated to late first millennium AD (Smith 1987). The material remains in the early layers of KBB resemble those of the ADE sites, with a wealth of sheep bones and potsherds (Smith 1987; Klein & Cruz-Uribe 1989; Sadr & Smith 1991). Here then is a site with a high status material signature set in a low status, hill-base position. This tends to weaken the "one culture: two economic classes" scenario, but the idea may still have some merit, especially in the modified version set out below. This final hypothesis to explain the relation between the two sets of first millennium AD sites on Kasteelberg proposes that the sheep bone and ceramic rich ADE sites (as well as KBB) were special purpose locations for feasts, while the stone-tool rich MNG sites represent the everyday habitation camps of the indigenous west coast herder-foragers (this does not exclude the possibility that at times the ADE sites may also have acted as everyday habitation camps). In effect, this scenario proposes a "one culture: one economy" model, but with a relatively complex herder-forager culture wherein occasional feasts were held to consume special foods. Much has been written on competitive feasting as a mechanism for accumulating power among "complex" or "trans-egalitarian" hunter-gatherer societies (see Hayden 2002 for an overview). Complex hunter-gatherers are difficult to identify archaeologically. Some of the material correlates of such a society might be food storage, reduced residential mobility, increased population density, socio-economic differentiation,

FROM FORAGING TO HERDING: THE WEST COAST OF SOUTH AFRICA

225

social division of labor, developed systems of exchange, warfare and intensive ceremonial and social activities such as feasting (Hayden 1990; Arnold 993). Some of these features can be seen among various hunting and gathering groups on the south and west coasts of South Africa during the last 3000 years. The massive shell middens of the west coast, for example, suggest harvesting, food storage and a delayed return economic system between two and three thousand years ago (Henshilwood et al., 1994; Jerardino 1996). Some of the burials at Oakhurst on the south coast include fairly rich grave goods, which might indicate nascent social complexity (Goodwin 1938; Mitchell 2002: 159). The dozens of grinding grooves found in the bedrock around the site KBB, which probably date to the early second millennium AD, seem to indicate intensive production for exchange or ritual (Boonzaier et al., 1996: 21). And, overall, a rise in population density is well documented for the late Holocene on the west coast (Mitchell 2002: 188). Thus, some evidence for complex hunter-gatherers can be found in the archaeology of the south and west coasts of South Africa before and after the first millennium AD. Complex herder-foragers on Kasteelberg during the late first millennium AD thus would not be as far-fetched an idea as it might sound. But were the herder-foragers feasting on Kasteelberg? The principal characteristic of such feasts is the planned consumption of large quantities of special food and beverages (Hayden 2002: 166). This activity may be archaeologically visible in a high "consumption index": the ratio between food and drink debris versus other discarded artifacts of daily use. Indeed, as the hypothesis would lead one to expect, the ADE sites do have a higher consumption index than the MNG sites (Table 1). Particularly notable is the abundance of sheep bones on the hilltop sites K B A and D. Sheep, exotic and rich in fat, were surely special on the west coast at this time, and the super-abundance of their bones at KBA and D fits well the idea of mass butcheries for feasting purposes. Seal bones are also well represented on the ADE sites and their fat would presumably have rendered them fit for a feast as well. The large quantities of potsherds fit the feasting scenario too. The potsherds all come from a single type of vessel: a small spouted jar evidently used for pouring liquids. Residue analyses currently under way will probably confirm that they were used mainly for sheep's milk, and perhaps also for rendering and pouring seal fat as has been shown to be case for somewhat later pottery from this area (Patrick et al., 1985). Fat and milk, as ethnographic information from many parts of the world confirm, would almost certainly have been considered luxuries in a herder-forager population. Thus, the abundance of broken pots on the ADE sites is also consistent with the feasting scenario.

226

SADR

Even the hilltop position of KBD and KBA can provide support for the feasting scenario. As the archaeologists working at Kasteelberg over the past twenty years can readily confirm, the summit of the hill, with its commanding views and protective boulders, is certainly a fine place for a barbecue. Feasts could presumably have taken place anytime during the year, and the faunal evidence that points to multi-seasonal occupations at KBA (Klein & Cruz-Uribe 1989: 92) is thus in keeping with this scenario. The other possible feasting site on Kasteelberg is KBB at the base of the hill, which may represent the location used in winter when the summit of Kasteelberg was covered in clouds, or when the effort of schlepping the food so far did not seem worthwhile for one reason or another. Just south of the Vredenburg peninsula, the sites of KTB and DSM also display a high consumption index (Smith et al., 1991). If these also represent feasting sites, it is interesting to note that both are located on hilltops with commanding views and protective boulders. Overall then, the evidence can be taken to suggest that during the late first millennium AD, the local population of the Vredenburg Peninsula developed the concept of feasting on sheep and seal, on the impressive and centrally located Kasteelberg. According to this alternative view, the dichotomy between sites in the first millennium AD archaeological record, which had previously been interpreted as the distinction between Khoekhoe herders and San hunter-gatherers (Smith et al., 1991), and which more recently had been interpreted as a distinction between inland- and shoreline-oriented factions within the local population (Sadr et al., 2003), in fact may represent a geographically and chronologically specific manifestation of social complexity among the indigenous herder-foragers, expressed through the medium of feasts.

Summary and Conclusion The majority view of the history of herding on the west coast (for example Smith et al., 1991; Boonzaier et al., 1996) calls for the existence of two culturally and economically distinct populations in the area during the last 2000 years: the Khoekhoe herders and the San foragers. But six recently excavated first millennium AD sites on the hill Kasteelberg fail to support this "two cultures: two economies" model (Sadr et al., 2003). All six sites show the same stylistic traits in pottery, ostrich eggshell beads, and stone raw material preference (if this last can be considered style). Although such stylistic similarity does not necessarily prove that the occupants belonged to the same culture (for example, Hodder 1982), it certainly weakens the proposition that two dis-

FROM FORAGING TO HERDING: THE WEST COAST OF SOUTH AFRICA

227

tinct cultures occupied Kasteelberg during the first millennium AD. As for two economies, the excavation of the six sites revealed a distinction between inland- and shoreline-oriented occupations, but both fall within the ambit of a mixed economy of herding, hunting and gathering. A clear distinction between herders and foragers is thus not found on Kasteelberg during the first millennium AD. If neither as Khoekhoe vs. Bushmen, nor as herders vs. foragers, how then did the first millennium AD occupants of the inland- and shoreline-oriented sites on Kasteelberg relate to each other? A number of possible answers to this question, such as seasonally specific occupations, status differences in the society, as well as the idea of a transition from one economy to another were considered. All were found wanting. A possible answer that now seems to fit the evidence better is feasting. In this scenario, the dichotomy of sites on Kasteelberg represents everyday living sites, and special purpose sites for occasional feasts. The "inland-oriented" MNG sites represent the everyday living sites. They seem to have been short-term, but repeated camp occupation. The feasting sites, on the other hand are those previously labeled as "shoreline-oriented." These, the ADE sites, only appeared in the mid-first millennium AD. Given that sheep were being herded in this landscape since 2000 years ago and seal were being exploited for far longer, why do the feasting sites not pre-date the mid-first millennium AD? The evidence suggests that the introduction of livestock to the west coast around 2000 years ago, far from representing the immigration of economically specialized foreigners as the majority view would have it, was an event of only minor economic significance (Sadr 2003). At some time, sheep and the idea of making ceramic pots arrived in this landscape, perhaps via long established networks of gift giving and exchange among indigenous southern African foragers and hunter-gatherers. Initially, as the continuity in material remains at KBG shows, the sheep and pottery were simply another item or two added to the existing technological and subsistence menu of the west coast foragers. Here, a useful conceptual analogy might be the goat herding foragers of the central Kalahari in the twentieth century (Tanaka 1980; Ikeya 1993; Kent 1993). They also retained their hunting and gathering way of life and their strict ethos of sharing in an egalitarian community, in spite of their privately owned domesticated animals (Kent 1993). In the Kalahari, the goats represented simply a resource which in addition to providing a regular supply of milk, provided emergency meat, and a was valuable item that could be bartered for other products such as tobacco.

228

SADR

In similar fashion, herding during the early first millennium AD on Kasteelberg may have represented merely a continuation of the indigenous forager culture to which were grafted a few sheep and ceramic vessels. But by the mid-first millennium AD, something had drastically changed to produce sites that can be interpreted as locales for sheep and seal feasts. What happened and why? These questions cannot be fully addressed yet, but a number of correlating variables are worth pointing out. They may in the future prove to be the key to understanding this significant cultural transformation towards more "complex" or "trans-egalitarian" behavior on Kasteelberg. One important variable is climate. Palaeo-climatic evidence suggests a shift from a warm and presumably drier to a cold and perhaps wetter climate around the mid-first millennium AD (for e.g., Talma & Vogel 1992). Another important and perhaps related variable is the evidence for a population increase in the Vredenburg Peninsula during the late first millennium AD. This population increase is visible in the proxy evidence provided by radiocarbon dates. The Vredenburg peninsula has so far produced only 10 dates in the first half, but 32 dates in the second half of the first millennium AD. It is possible that the increase represents an influx of populations from neighboring areas: there is some evidence that in the Elands Bay area, 100 or so km further up the coast, the number of C14 dates decline in the second half of the first millennium AD. If this complementary pattern in the radiocarbon dates indeed represents a population movement from Elands Bay to the Vredenburg peninsula around the mid-first millennium AD, the question becomes why? Did the success of shepherding on the peninsula (perhaps helped by the new climatic regime) produce a food surplus that could be manipulated for social prestige through feasting, which in turn attracted the people from Elands Bay? Or did the people from Elands Bay in fact bring the idea of feasting with them? They certainly had experience of manipulating food surplus in earlier centuries at the so-called "mega" shell middens, which were more common around Eland's Bay than on the coast of the Vredenburg Peninsula (Jerardino 1996). There are probably many other variables that still have to be taken into account as well. The investigation of these and similar questions should eventually shed more light on the question of feasting at Kasteelberg. For the moment, it seems clear enough that the weight of evidence is against the majority view of two cultures and two economies on the west coast of South Africa during the first millennium AD. The possibility of a local development of social complexity among the herder-foragers of the Vredenburg peninsula through the agency of com-

FROM FORAGING TO HERDING: THE WEST COAST OF SOUTH AFRICA

229

petitive feasts h i g h l i g h t s the n e e d for a g e n e r a l r e - e x a m i n a t i o n o f the p r e - c o l o n i a l history o f this part o f the world: the ancestral C a p e K h o i s a n m a y not h a v e b e e n as " p r i m itive" as the h i s t o r y b o o k s suggest.

References Arnold, Jeanne E. 1993. Labor and the rise of complex hunter-gatherers. Journal of Anthropological Archaeology 12: 75-119. Boonzaier, Emile, Candy Malherbe, Andrew Smith & Penny Berens 1996. The Cape herders: a history of the Khoikhoi of southern Africa. Cape Town and Johannesburg: David Philip Elphick, Richard 1977. Kraal and Castle. New Haven, CT: Yale University Press. Elphick, Richard 1985. Khoikhoi and the founding of white South Africa. Johannesburg: Raven. Goodwin, A. John H. 1938. Archaeology of the Oakhurst shelter, George. Transactions of the Royal Society of South Africa 25: 229-324. Hayden, Brian 1990. Nimrods, piscators, pluckers, and planters: the emergence of food production. Journal of Anthropological Archaeology 9:31-69. Hayden, Brian 2002. Hunting and feasting: health and demographic consequences. Before Farming 1:166-176. Henshilwood, Christopher 1996. A revised chronology for pastoralism in southernmost Africa: new evidence of sheep at c. 2000 b.p. from Blombos Cave, South Africa. Antiquity 70: 945-9. Henshilwood, Christopher, Peter Nilssen, and John E. Parkington 1994. Mussel drying and food storage in the late Holocene, south-west Cape, South Africa. Journal of ~Teld Archaeology 21: 103-9. Hodder, Ian 1982. Symbols in Action: ethnoarchaeological studies of material culture. Cambridge: Cambridge University Press. Ikeya, Kazunobu 1993. Goat raising among the San in the central Kalahari. African Studies Monograph 14(1): 39-52. Jerardino, Antonieta 1996. Changing social landscapes of the western Cape coast of southern Africa over the last 4500 years. PhD thesis, University of Cape Town. Kent, Susan 1993. Sharing in an egalitarian Kalahari community. Man 28:479-514. Klein, Richard G. 1986. The prehistory of Stone Age herders in the Cape Province of South Africa. South African Archaeological Society Goodwin Series 5: 5-12. Klein, Richard G. & Kathryn Cruz-Uribe 1989. Faunal evidence for prehistoric herder-forager activities at Kasteelberg, Western Cape Province, South Africa. South African Archaeological Bulletin 44: 82-97. Mitchell, Peter 2002. The archaeology of southern Africa. Cambridge: Cambridge University Press. Patrick, Mary, Andrew Smith & A.J. de Koning 1985. Gas-liquid chromatographic analysis of fatty acids in food residues from ceramics found in the south-western Cape, South Africa. Archaeometry 27:231-46. Phillipson, David W. 1993. African archaeology. Cambridge: Cambridge University Press. Rudner, Jalmar 1968. Strandloper pottery from South and South West Africa. Annals of the South African Museum 49:441-663.

230

SADR

Sadr, Karim 1998. The first herders at the Cape of Good Hope. African Archaeological Review 15: 101-132. Sadr, Karim 2002. Ancient pastoralists in the Sudan and in South Africa. In Jennerstrasse 8 (eds.) 1ides

of the Desert: contributions to the archaeology and environmental history of Africa in honour of Rudolph Kuper, pp. 471 -484, K61n: Heinrich-Barth-Institut. Sadr, Karim 2003. The Neolithic of southern Africa. Journal of African History 44: 195-209. Sadr, Karim & Andrew Smith 1991. On ceramic variation in the southwestern Cape, South Africa. South African Archaeological Bulletin 46:107-115 Sadr, Karim, John Gribble & Gail Euston-Brown 1992. The Vredenburg peninsula survey, 1991 / 1992 season. In Guide to the archaeological sites in the southwestern Cape, compiled by A.B. Smith & B. Miitti, pp. 41-43. Cape Town: University of Cape Town. Sadr, Karim, Andrew Smith, Ina Plug, Jayson Orton & Belinda Mtitti 2003. Herders and foragers on Kasteelberg: interim report of excavations 1999-2002. South African Archaeological Bulletin 58: 27-32. Schrire, Carmel & Janette Deacon 1989. The indigenous artefacts from Oudepost I, a colonial outpost of the VOC at Saldanha Bay, Cape. South African Archaeological Bulletin 44:105-113. Sealy, Judith & Royden Yates 1994. The chronology of the introduction of pastoralism to the Cape, South Africa. Antiquity 68: 58-67. Smith, Andrew B. 1992. Pastoralism in Africa. Johannesburg: Witwatersrand University Press. Smith, Andrew B. 1987. Seasonal exploitation of resources on the Vredenburg peninsula after 2000 BE In Papers in the prehistory of the Western Cape, South Africa, edited by J. Parkington and M. Hall, pp. 393-402. British Archaeological reports International Series 332(ii). Smith, Andrew B., Karim Sadr, John Gribble & Royden Yates 1991. Excavations in the south-western Cape, South Africa, and the archaeological identity of prehistoric hunter-gatherers within the last 2000 years. South Afi-ican Archaeological Bulletin 46:71-91. Talma, A.S. & J.C. Vogel 1992. Late Quaternary palaeotemperatures derived from a speleothem from Cango Caves, Cape Province, South Africa. Quaternary Research 37: 203-13. Tanaka, Jiro 1980. The San hunter gatherers of the Kalahari: a study in ecological anthropology. Tokyo: University of Tokyo Press.

Received: December 20, 2004

Accepted: March 20, 2005