Wenner-Gren Foundation for Anthropological Research

Paradigmatic History of San-Speaking Peoples and Current Attempts at Revision [andComments and Replies]Author(s): Edwin N. Wilmsen, James R. Denbow, M. G. Bicchieri, Lewis R. Binford, RobertGordon, Mathias Guenther, Richard B. Lee, Robert Ross, Jacqueline S. Solway, Jiro Tanaka,Jan Vansina, John E. YellenSource: Current Anthropology, Vol. 31, No. 5 (Dec., 1990), pp. 489-524Published by: The University of Chicago Press on behalf of Wenner-Gren Foundation forAnthropological ResearchStable URL: http://www.jstor.org/stable/2743950Accessed: 17/07/2010 06:43

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=ucpress.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press and Wenner-Gren Foundation for Anthropological Research are collaboratingwith JSTOR to digitize, preserve and extend access to Current Anthropology.

http://www.jstor.org

CURRENT ANTHROPOLOGY Volume 3I, Number 5, December I990

? I990 by The Wenner-Gren Foundation for Anthropological Research. All rights reserved OOII-3204/90-3I05-0002$3.75

Paradigmatic History of San-speaking Peoples and Current Attempts at Revision'

by Edwin N. Wilmsen and James R. Denbow

The time when there were no archaeological data for the Kalahari and the prevailing paradigm persuaded us that all the archives were empty, when presumed foraging peoples were consigned to simplicity in social relations, when we believed we could explain ourselves through evolution alone is long gone. Neither an- thropological knowledge claims nor the peoples upon which those claims are based nor the policy decisions influenced by them are well served by denying this. Nevertheless, a "Kalahari San de- bate" has arisen around the question whether foragers are genuine or spurious. The identical question was raised by Fritsch against Passarge's "revisionism" in the first "Bushman debate" of I906. We consider the question itself spurious, arguing that "Bushman" and "San" are invented categories and "Kalahari foragers" an eth- nographic reification drawn from one of several subsistence strate- gies engaged in by all of Botswana's rural poor.

EDWIN N. WILMSEN is Research Fellow at the University of Texas at Austin (his mailing address: African Studies Center, Boston University, 270 Bay State Rd., Boston, Mass. o22IS, U.S.A.). Born in I932, he was educated at Texas A & M College (B. Architec- ture, i958), Massachusetts Institute of Technology (M. Architec- ture, i959), and the University of Arizona (M.A., I966; Ph.D., i967). He has taught at the University of Michigan and in the African Studies Center of Boston University and has been visiting professor at Rutgers, New York, and Bayreuth Universities and the City University of New York. He has done fieldwork on the Iron Age of Botswana and ethnographic and archival research on !Kung ecology, nutrition, and family formation and on the devel- opment of pastoralist economies in southern Africa. His recent publications include Land Filled with Flies: A Political Economy of the Kalahari (Chicago: University of Chicago Press, i989), the collection We Are Here: Politics of Aboriginal Land Tenure (Berkeley and Los Angeles: University of California Press, I989), and "The Antecedents of Contemporary Pastoralism in Western Ngamiland" (Botswana Notes and Records 2o). JAMES R. DENBOW is Associate Professor in the Department of Anthropology of the University of Texas at Austin. He was born in I946 and received his B.A. from the University of Illinois in I968 and his Ph.D. from Indiana University in I983. He has been

Curator of Archaeology and Head of the Department of Archaeol- ogy of the National Museum of Botswana and has conducted exca- vations of Late Stone Age, Iron Age, and Neolithic sites in Bo- tswana and the Congo. Among his publications are "A New Look at the Later Prehistory of the Kalahari" (Journal of African History 27:3-29), (with J. S. Denbow) Uncovering Botswana's Past (Gaborone: Government Printer, i99o), and "Congo to Kalahari: Data and Hypotheses about the Political Economy of the Western Stream of the Early Iron Age" (African Archaeological Review, in press).

The present paper was submitted in final form 2o iv 90.

"Der Buschmann ist das ungluckselige Kind des Au- genblicks." With this sentence, Siegfried Passarge, a German geographer, began the first account of a San- speaking people that may be called ethnographic. Die Buschmanner der Kalahari, first published in I905 and reprinted with minor changes in I907 (Passarge I9o5a, I907), was based on some two years of fieldwork during I896-98 in what is now Botswana and Namibia. Pas- sarge spent about nine months in the Dobe-NyaeNyae area described later by members of the Harvard Kalahari Project (Lee and DeVore I976),2 much of that time in the vicinity of CaeCae,3 where one of us (Wilmsen) has worked continuously since I973.

The publication of Die Buschmanner precipitated the first "great Bushman debate." Gustav Fritsch (i868),4 from whom Passarge took his opening sentence (which may be translated "The Bushman is the unluckiest child of our time"), began his criticism of Passarge's work by saying, "Surely, Mr. Passarge was himself the unluckiest child of our time, for he states . . . that I personally have seen very little if anything of independent Bushmen" (Fritsch I906:7I). Fritsch was stung because "Mr. Pas- sarge thinks that descriptions of the Bushmen given by earlier authors (myself included) should be considered a 'caricature'" (p. 72); he asked rhetorically, "Should not the designation 'caricature' better fit modern conditions, as they are the result of 40 years of political upheaval?"

i. Funding was provided by the National Endowment for the Humanities, the National Science Foundation, the National Geo- graphic Society, the Max-Planck-Institut, the Social Science Research Council, the Sonderforschungsbereich-Universitat Bay- reuth, and the Wenner-Gren Foundation for Anthropological Re- search.

2. Lee (I979:xvii-xviii, 9-28) gives a clear account of the genesis of the Harvard Kalahari Project in the formulation of his Ph.D. thesis under Sherwood Washburn and Irven DeVore at Berkeley in i962. The other participants in the project, most of them students of DeVore, who had moved to Harvard, joined five to eight years later. DeVore has published almost nothing on this project, while the other members have confined their reporting to the areas of their own specialties and some have repudiated or sharply questioned its major premises (Howell i986, Shostak i98i, Konner and Shostak I986). Thus, the Harvard Kalahari Project should be seen as largely shaped by and identified with Lee. 3. We follow the Botswana Place Names Commission in adopting Setswana spelling, whereby the click consonants become c = I (dental) and # (alveolar), q = ! (palatal), and x = 11 (lateral). Thus CaeCae is Lee's /Xai/xai, Gcwi is Silberbauer's G/wi, and so forth. Passarge used superscript numerals to designate clicks, but he was inconsistent in his orthography; his SSu2gnassi is Zhucoasi and 'Kai'Kai is CaeCae. 4. Fritsch spent three years, i863-66, in southern Africa and wrote two books on his observations there (Fritsch i863, i868).

489

490 CURRENT ANTHROPOLOGY Volume 3I, Number 5, December I990

And he urged his protagonist (pp. 76-77, our transla- tion)5-and all of us, too-to

take to heart. . . the Bushman natural disposition, those well-known ancient characteristics [that] ren- dered them almost outcasts. . . [and to recognize those] other living conditions and arrangements into which Mr. Passarge entered, and, I am convinced, un- fortunately brought about the astonishing depletion of the wilderness he describes as shutting out the hunting way of life, making it simply impossible.

The issue was, of course, who saw the real Bushmen. Lee rephrases that question in what he chooses to see

as a revived "Kalahari San debate" (I989): Are foragers genuine or spurious? (Solway and Lee I990). We consider the question itself to be spurious; "Bushmen" and "San" are invented categories (Wilmsen I983; I989a:24-32), while "Kalahari foragers" is an ethnographic reification drawn from one of several integrated subsistence strate- gies engaged in by all the poor-regardless of their ethnic or ethnographic categorization-of rural Bo- tswana, where the poorest tenth derives 25 % of its subsistence from foraging and the poorest half I2% from that source (Rural Income Distribution Survey I976:IO2). Furthermore, Taylor and Moss (I98 i) suggest that on average all Batswana derive up to 40% of their vegetable diet from gathered rather than cultivated sources, and Grivetti (I978, I979) documents that some Batswana and Bakgalagadi communities use a greater va- riety of wild plants than do Zhu.6 Lee (I979:I80-8o) thinks that data such as these, along with those of Scud- der (I 97 I) on the Gwembe Tonga, are an index of relative pressure on the environment and a "further indication of the security of the foraging way of life." The conceptual impoverishment of forager reification is most glaringly expressed by Yellen (I 990:I04), who calls on "anecdotal information" (his term) to conclude that

from the !Kung perspective, goats were essentially the same as any other medium-size animal (in that they provided a reasonable yield of meat and were rel- atively easy to carry), and cows were the same as other large creatures. If an animal was easy to obtain the !Kung ate it.... Similarly, they seemed to con- ceive of agriculture and wage labor undertaken for the

Bantu and anthropologists ... much as they perceived herding: as foraging resources just like any other.

That foraging and food production in the Kalahari are not ideologically separate pursuits but activities mutu- ally constructed within a forager/farmer symbolic reser- voir is well illustrated by Campbell's (i986:88-89) ob- servation that "many people [rural Batswana regardless of ethnicity] now deny they know the names of quite common edible plants, fearing that it may be thought that they eat them."

The Ancestry of Illusion

It is fair to say that, if Die Buschmanner had been more widely read, modern ethnographies of San-speaking peo- ples would resemble those written about the region's Bantu-speaking inhabitants, which do not purport to be describing a way of life "protected by the vast distances of the Kalahari" (Lee I979:33). Speth and Hitchcock (n.d.) will soon make available an English translation of all works on the Kalahari published by Passarge; the fol- lowing excerpts will reveal their relevance to the current "debate."

Passarge lamented what he thought to have been lost from "the life of Bushmen in earlier times" under the conditions of the present. In company with most Euro- pean ethnographers (including Fritsch) who had spent time among San peoples in other parts of southern Af- rica before him and the majority of those who came af- ter, he thought that an essential element of what it meant to be fully, truly human had been erased when the pure life of the hunter was replaced by domesticated pursuits. But he had no illusions; his observational pres- ent, I896-98, featured Zhu not only in their "tradi- tional" role as foragers of wild resources but also as hus- bandmen of cattle and goats, associates of Batawana and Ovaherero, and, in the immediately preceding decades but no longer then, participants in trade. For him their "earlier times," by which he meant later precolonial centuries as representative of prehistoric millennia, were the same except that husbandry, external contact, and trade were subtracted and foraging expanded to fill the void. Passarge (i9o5a:iI4) thought that the change had begun only a few decades before his arrival in Ngami-Namibia ("6o or more years earlier") and had been accelerated by the entry of European merchants into the region in the I 8 5 os. That is to say, he considered pure aboriginal "bushman" existence to have ended in the early igth century.7

5. "die Buschmanner Naturgemass ... man zu beherzigen hat, dass diese notorisch von Alters her gewohnheitsgemass als nahezu Vogelfrei [translated "i. without or deprived of rights; 2. jurispru- dentially outlaw, outcast; 3. figuratively outcast"; we use the term in all three senses] betrachtet wurden.... andere Gewohnheiten und Anlagen, wahrend die von Hrn. Passarge eingehend und, wie ich iuberzeugt bin, leider zutreffend beschreibene erstaundliche Ausrottung des Wildes dass ausschliessliche Jagerleben einfach un- moglich machte." 6. Lee and other members of the Harvard Kalahari Project continue to use "!Kung" for the people who refer to themselves as Zhuc6asi (Iu 'person,' 1'a 'finished or complete,' si 'plural suffix', hence "completed people"). !Xfi, which is the source of the term !Kung, is the self-referent name of a people who live in Angola and speak a dialect very similar to Zhuc6asi. We use Zhu, a common abbrevia- tion used by the people themselves. Lorna and John Marshall have adopted Westphal's (I963) spelling, Ju/wasi.

7. Seiner (I977[I9Io]:3I-32), who travelled in Ngamiland during I905-6, wrote that "in the northern Kalahari the circumstances are changing so rapidly that reconstruction of the former lifeways of the people is very difficult. . . . it appears that the Northem Bushmen are losing their racial character through a long-standing process of intermixture with surrounding Negro peoples." This was especially true, he said, of the Makwengo-the Khoe-speaking Kxoe (this is actually the same name: kwe 'Khoe', ma 'noun-class prefix') studied by Kohler (I975, I987), who resemble their Bantu- speaking neighbors in physiognomy and physique. This resem-

WILM SEN AND DENBOW Paradigmatic History of San-speaking Peoples I 49I

Passarge offered invaluable insights into the nature of San political organization and leadership, matters that have been treated rather loosely by modern ethnog- raphers;8 the following passage (I907:II4, our transla- tion)9 merits very close attention:

The Bushmen of the middle Kalahari are divided, as is well known, into two great peoples [linguistic divi- sions], these again into individual branches/clans and the clans into kindreds. Previously, these clans

formed complete political organizations. At the head of each of them stood a "Grootkapitain," as the Dutch-speaking Bushman expressed it, who was sup- ported by the "4aichadji" [xhaihasi], the family heads.... In the Xaudum Valley, I came upon a band of Zhu whose leader's name was Gumtsa [Kumtsaa]. But the headman of the clan was named "2Auka" [Xauka], who formerly kept a valley called Tsanni for himself; this must lie westward from Sodanna [Tshocanna].

If we read "local descent group" for "clan," we find that this is a reasonable description of Zhu political divisions with their leaders; it uses the (Zhu) term xhaiha 'leader, chief' (see n. 8 for cognate terms in Khoe languages) ap- propriately and suggests the existence of nqoresi.'0 Fur- thermore, it specifies two men as leaders whose names match those of the grandfathers of the heads of one CaeCae Zhu homestead-Gxau, whose father's father was Xauka, and Tcruka, whose father's father was Kumtsaa. Both are known to have had nqoresi to the northwest of Tshumqwe-Kumtsaa at Ceni (#eni, which could have been written Tsanni by a German-speaker) and Xauka at Qama (the Zhu name for Karakuwisa). These men would have been in their mid-to-late-5os in I896-98 when Passarge met them. All of this matches what is known of the ancestors of this homestead (Wilmsen i989a:2II, 267), and it would seem that Pas- sarge has given us a very valuable firsthand glimpse of Zhu leadership in place a century ago. In this and in his thinking about San political organization, Passarge (I906:4I3, our translation)" was considerably ahead of his time: "In Australia it is the same as in South Africa. It has long been the case, as far back as we have knowl- edge, that the Australian is not a lawless, wandering savage but among them that which appears to proceed without regulation is well ordered just as among truly literate peoples."

Fritsch (i960:72, our translation)'2 retorted,

I have published statements concerning the foregoing questions.... "The entire way of life prevents a tight political organization.". . . Mr. Passarge wants to

blance is not, however, necessarily attributable to relatively recent intermating between these peoples; the presence of other dark- skinned Khoi- and Khoe-speakers (the Damara, who speak Nama, a language closely related to Kxoe, the Heixum of northern Namibia, and the Tshukhwe Khoe-speakers on the Nata River) suggests that it may have a long and complex history. Seiner said that Makwengo lived in the Oschimpolofeld, which on most maps drawn before about i 900 includes northem Ngamiland and adjacent Namibia above NqauNqau, although most Kxoe now live in Caprivi. Those maps plus eyewitness accounts (e.g., Prinsloo and Gauche I933, who recount the story of the Trek Boers who stayed at NyaeNyae during i875-76) place Damara in the Omaheke sandveld that in- cludes the NyaeNyae area adjacent to Dobe. That some Makwengo were under Tawana jurisdiction in I9I4 is documented by the fact that Mathiba, kgosi (chief) of Batawana, paid ten "Masarwa [Zhu] and Makwengo [Kxoe] Bushmen" spies to scout "their old hunting grounds on the sandveld to the west" in order to keep track of German movements in Southwest Africa in the early years of World War I (Botswana National Archives S28/3, Anglo-German & European War [I 9I4-1). 8. Silberbauer (i98i:I34), for example, says, "There are some social concepts included in the vocabulary for which the G/wi have no cultural equivalent, for example, //xeixama (xheixa = chief)." It is difficult to imagine how a language could come to include a term for "chief" unless speakers of that language had some conception of "chiefliness." Lee (I979:344) also recognizes that Zhu has a "word for 'chief,' //kaiha, derived from the word //kai (wealth)," but in- sists that speakers of this language have no leaders. Snyman (I975:27, i29) derives 11'aixa (xhaiha 'master, leader, wealthy per- son') from 11'ai (xhai 'to be wealthy'), contrasted with gllaTkh6e (gxaakwe, 'poor person'), from glla) (gxaa 'to be poor'). Gcwi xheixa- (ma) appears to be cognate with Zhu xhaiha, the suffix ma being a gender marker in Gcwi. Thus, leadership among Gcwi and Zhu- as among Batswana and Ovaherero-is associated with wealth. Hitchcock (i980:25) gives Ilkaiha (xkaiha) for Kwa, spoken in east- em Botswana; this, too, appears to be cognate with xhaiha. This term does not appear to be derivable from any Bantu term (Tswana kgosi, Nguni kosi, Herero omuhona) and thus cannot be presumed to represent borrowing. Although Marshall (I976:i9i) retracts her earlier attribution of headmen to Zhu, she does so on the basis that it was inappropriate to extend kx'ao (she writes k"xau) 'owner' (or 'possessor') to encompass "headman." But in her (i960:348) earlier work she gives nliha and gaoxa as the principal words for "head- man" and "chief," with kxau (as she then wrote, kx'ao) having this meaning only secondarily. Gao in gaoxa derives from the name for God in his earthly manifestation, while nliha appears to be an attempt to write (l'hixa. Thus it appears that Marshall's retraction is unwarranted. 9. "Die Buschmannrasse zerfdllt in der Mittel-Kalahari, wie be- kannt, in zwei grosse V6lker, diese wiederum in einzelne Stamme und die Stamme in Sippen. Friuher nun bildeten die Stiimme ge- schlossene politische Organisationen. An der Spitze eines jeden Stammes stand ein Grootkapitain, wie der hollandisch redende Buschmann sich ausdriickte, der von den 4aichadji, den Familienoberhauptern, unterstuitzt wurde.... Im Schadumtal traf ich auf eine Horde Ssu2gnassi deren Hauptling Gumtsa hiesst. Aber der Oberhaiuptling der Stammes hiesst 2Auka und hielt sich seinerseit an einer Vley Tsanni auf, die westlich von Sodanna liegen muss."

io. Nqore (n!6r6) may be glossed "a place in land" or "country" and has the primary connotation of associating persons with tracts of land. It has both an inclusive-in the sense of the geographical region within which all Zhu live-and a restrictive meaning. In the restrictive sense, a Zhu refers to some locality-some demarcated land-as n!6r6 mima (n!bre 'country,' mi 'self,' ma 'possessive': "my country," the place where I belong/was born). I i. "In Australien gings aber ebenso wie in Sudafrika. Es hat sehr lange gedauert, bis man erkannte, dass die Australier nicht gesetz- los herumziehende Wilde seinen, sondern bei ihnen auch das nach bestimmten Vorschriften geordnet ist, was scheinbar ganz regellos verlauft." 12. "Ich habe zu der vorliegenden Frage ausdriucklich Stellung genommen. . . . "Die ganze Lebensweise verbietet ein straffere staatliche Organisation. " . . . Hr. Passarge mochte wissen, wie es im Angang der sechziger Jahre unter den Buschmannern ausgesehen hat und anstatt sich bei mir, dem einzig iuberlebenden Reisenden aus jener Zeit und jener Gegend Rat zu erholen, lasst er sich lieber von seinem alten Buschmanndiener etwas vorlagen ... durch seine Verlogenheit beriuchtigten Stamme angehorte."

492 1 CURRENT ANTHROPOLOGY Volume 31, Number s, December I990

know what it was like among the Bushmen in the [i8]6os, and instead of recovering this from me, the only surviving traveller of that time and of that undertaking, he preferred to hear from his old Bush- man servant something close to a pack of lies ... the mendacity of his infamous tribe.

Passarge (i906:4I3) responded that "certainly, among the now entirely subordinated Bushmen not only polit- ical but also social organization has been lost." He at- tributed this loss (igo5a:2o6) to the consequences of mercantile trade initiated by Europeans and the strug- gles among local leaders to capture the benefits of this trade for themselves; the most serious of these conse- quences he thought to have been the extermination of wild animals and destruction of the veld's natural pro- ductive capacity. Passarge (I907:IO, I88, our transla- tion)'3 records that in I897 at the very places where "the merchant Franz Muller, a fine observer and unques- tioned expert on the Kalahari," whose oxen rode out the rinderpest at Qubi in the Dobe area,

the number of Bushmen has declined since his trip at the beginning of the eighties. Before, they gathered in hundreds around his wagon and brought very many pelts to trade; now one sees in these same places ever fewer, often not a single person.... The extermina- tion of the wild animals and the failure of melons have brought about, in my opinion, this unfavourable prospect.

But, while it is clear that Passarge was at least partly inspired by the mystical romanticism of Tonnies (I957 fi8871) so prevalent in Central Europe at the end of the igth century, an essay on the foundation for ethno- graphic depiction of the Kalahari region (Passarge igo5 b) strongly suggests that he was also influenced by Marx. This may account for his sense of history in the Kalahari and his corresponding somewhat ambivalent rejection of Fritsch's (I8o) concept of "Bushmen" as Urrasse (original race).'4 His text is pregnant with tension be- tween his received romantic proclivities and a rather realistic if uncomplicated recognition of what unfolded before his eyes.

To begin with, unlike later ethnographers of San peo- ples, Passarge did not mention Bantu-speakers in the re- gion only to dismiss them as inconsequential to the eco-

nomic, political, and social lives of these peoples. Indeed, his narrative account of contemporary social and political organization was a vivid word picture of San and Bantu peoples embedded in a single social forma- tion. A few examples drawn from his text will make this clear. The first refers to Tawana masters and the tribute they extracted (Passarge I 907: I 2 I, our translation): 15

Kchossani [Dikgosana, members of the Tswana ruling class] demand a yearly tribute-maketo [makgetho 'taxes' (Brown I979 [i8751:177)1-in hides, ostrich feathers, strings of ostrich eggshell beads; in the Os- chimpolosandfeld [the NqauNqau area up to the Okavango River] also redwood [morotomadi (Pterocarpus angolensis), used for dressing leather and cosmetics]; and in the northern Kalahari, Wurzel- kautschuk [root rubber, exact species unknown]. The tribute is usually brought in by the Bushmen on their return home during the dry season, or sometimes dur- ing the rainy season when the Kchossani hunts in his region.... Harry ruled over the Aukwe [Cexai, Makaukau] in the southern Kaukauveld and Rabutuba ruled over the Ssu2gnassi [Zhuc6asi].

San peoples also provided abundant labor for their Bantu masters, much of it involuntary (Passarge I907:I22, our translation): 16

The Bushmen serve as trackers and drivers. They do that gladly because there will be lots of meat in the end. In May I898, Harry was hunting in the southern Kaukaufeld between 'Kai'Kai [CaeCael and 'Gam [Gcam] and all our Bushmen ran away to join him. But the Bushmen not only help with the hunt; they are also commandeered for hard work, to carry heavy loads, to build houses, and such. We met, for in- stance, a travelling Batauana [Motawanal in the Chansefeld [Ghanzi veldj in June I897 and all his be- longings were carried by a group of Bushmen ... bent under the weight, while their masters with Scham- bocks [sjamboks, whips made of thick hide such as that of hippo or giraffe] in their hands walked along- side.

I 3. "dass die Zahl der Buschmanner seit seinen Reisen im Anfang der achtziger Jahre ganz abfallend abgenommen habe. Fruiher hat- ten sie sich zu Hunderten um die Wagen gesammelt und mancher- lei Felle zum Handeln gebracht, jetz aber sehe mann an denselben Platzen immer nur wenige, oft keinen einzigen.... Die Ausrottung des Wildes und die Abnahme der Melonen hiitten seiner Ansicht nach so ungiunstig eingewirkt." Fritsch (i906:77) heartily endorsed these statements "which clarify how the 'hunter' became a 'gatherer."' 14. Passarge (I907:3) was initially scathing-"What an irony of fate! This one stable trait in his [the Bushman's] character [his love of freedom] is the cause of his inability to accept a higher culture" ("Welche Ironie des Schicksals! Gerade dieser eine einzige stabile Zug in seinem Charakter [Freiheitsliebel begrundet seine Unfahig- keit die Kultur anzunehmen")-but in the end (p. I 3 I) could also call Bushmen "those lost relics of primitive humans."

I 5. "Dem neuen Hern-Kchossani bet.-wird ein jahrlicher Tribut-maketo bet.-an Fellen, Straussenfedern, Molesta-Ketten, im Oschimpolosandfeld an die Kuangari auch an Rotholz und in der Nord-Kalahari an Wurtzelkautshuk auferleft. Der Tribut wird teils wahrend der Trockenzeit von dem heimkehrenden Busch- mannern eingezogen, teils wahrend der Regenzeit, wenn der Kchossani Jagdausfluige in sein Gebiet unternimmt. . So be- herrschte der Batauana . . . Harry die 2Aukwe des siudlichen Kaukaufeldes, Rabutubu die Ssu2gnassi." I6. "Die Buschmanner dienen als Spurer und Treiber. Das tun sie gern, denn hinterther gibts Fleisch in Huille und Fulle. Im Mai I898 jagd Harry im sudlichen Kaukaufeld zwischen 'Kai'Kai und 'Gam, und aller unsere Buschmanner liefen uns fort, um sich ihm anzu- schliessen. Aber nicht zur Jagd alein, auch zu schwerer Fronarbeit, zum Lastentragen, Hauserbauern u. a. werden die Buschmanner kommandiert. So passierte uns im Juni I897 ein reisender Batauana, der aus dem Chansefeld kam und seine Sachen wurden von Buschmannern getragen . . . keuchten sie unter der Last, wahrend die Herren mit dem Schambok in der Hand daneben gingen."

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples | 493

Passarge (I907:I2I, our translation)'7 also documented the pervasive involvement of Dobe-NyaeNyae-area San peoples in pastoralism:

Aside from the yearly tribute, the Kchossi [kgosi 'chief'] reserves the right of keeping his cattle in the region. He sends some of his animals with his own slaves to pans which hold water during the rainy sea- son or he hands some of his herds over to Bushman families. The Bushmen have care of the cattle, and at the beginning of the dry season must bring them back properly. They have for their part right to the milk, as much as is not necessary for the calves. Thus, for ex- ample, there was in January I897 a goat kraal of Kil- litibwes solely under the supervision of Bushmen in the Makabana Hills; in February it was farther south in the sandveld. The chief of the Aukwe at Garu had, at Djarutsa, a cattle kraal of the Batauana chief Sekumi [Sekgoma Letcholethebe] during the rainy season I897-98 and went with the cattle at the be- ginning of the dry season to the Debrafeld [both Djarutsa and Debra are in the NyaeNyae area].

He did not romanticize the conditions under which all this took place (I907:I22-23, our translation):'8

Unfortunately, they [Bushmen] are not accorded the rights which the overlord is supposed to administer according to African custom. There are all too often the worst transgressions, forced pillages of property, unjust requisitions for work, and most of all daily rapes of the women and girls. Often they take pretty

little girls away from their parents to use them later on as concubines. The Bushmen get very angry in such cases and their revenge is bloody. Quite a few Batauana have been victims of such revenge. In such cases, the Bushmen feel bloody retaliation in their turn and a chain of murder and killing follows the calamity of the Kaffirs. Such cases happen constantly. For instance, in May i897, a Batauana stole a Bush- man woman from Kamelpan in the Chansefeld. She escaped. He returned and found her with her hus- band, shot him, and took the woman with him again. The most violent actions done in my presence were by the Bakalahari [Bakgalagadi] at Kuschi [Khutsi'9].

Passarge (igo5 b: 34) came to the conclusion that Bush- men were among the desert proletarians of southern Af- rica, a status which he also assigned to most Bakgalagadi and to which he thought some Tswana and Herero groups themselves were being rapidly consigned by ac- celerated European intervention. His program did not require him to turn a blind eye to tribute payment, forced labor, kidnapping, murder, rape, and other instru- ments of oppression that Solway and Lee (i990) think of as barter at trade festivities, voluntary privilege, fierce resistance, independence, and other attributes of self- sufficiency. Passarge (igo5b:82) was of the opinion that if Bushmen had indeed voluntarily chosen to live a life of forager poverty in the midst of pastoralist plenty, then it "would be of great interest to gain perspective on this psychological problem." He understood that it was not choice but the result of particular historical circum- stances. His was a position to which we should return.

The Persistence of Illusion

The Kalahari, it seems, is plagued by caricatures (Den- bow i986). Solway and Lee (i990:io9-Io) name theirs the Coke Bottle syndrome, with its "imagery [of] forag- ing societies as so delicately balanced and fragile that they cannot accommodate innovation and change . .. [along with] the rueful recognition of the unlimited ca- pacity of 'advanced societies' to consume everything in their path." They attribute this caricature to the percep- tions of "revisionists"20 (in addition to the two of us, Gordon, Howell, Parkington, and Schrire are named) who, "in their zeal to discover links and dispel myths of

I7. "Ausser der Einziehung des jahrlichen Tributs hat der Kchos- sani das Recht, in dem betreffenden Gebiet Viehposten zu halten. Daher schickt er teils seine eigenen Sklaven mit seinem Vieh wahrend der Regenzeit zu den wasserhaltigen Vleys oder er uber- gibt den Buschmannfamilien Teile seiner Herden. Die Buschman- ner haben auf das Vieh aufzupassen und es beim Beginn der Trock- enzeit wieder richtig abzuliefern. Dafur haben sie ihrerseits das Anrecht auf die Milch, soweit die Tiere ohne Schadigung der Kal- ber sie abgeben konnen. So lag z. B. im Januar I897 ein Ziegenkraal Killetibwes lediglich unter Aufsicht von Buschmannern an den Makabana-Bergen, im Februar weiter suidlich im Sandfeld. Der Oberhiiuptling der 2Aukwe von 2Garu hatte in Djarutsa einen Viehkraal des Batauanahauptlings Ssekumi wahrend der Regenzeit I897/98 und zog mit dem Vieh mit dem Beginn der Trockenzeit ins Debrafeld." i8. "Leider begnugt man sich nicht mit dem Rechten, die der Lehnsherr nach afrikanischer Anschauung besitzt, vielmehr kommt es fortwahrend zu den grosssten Uberschreitungen, gewalt- samen Plunderungen des Eigentums, unberechtigten Re- quisitionen zu Arbeitsleistungen, vor allem aber sind Vergewalti- gungen der Frauen und Madchen an der Tagesordnung. Mit Vorliebe nimmt man hubsche Madchen als Kinder ihren Eltern fort, um sie spater als Konkubinen zu benutzen. Die Buschmanner verstehen aber gerade in solchen Dingen keinen Spass und rachen sich blutig. So mancher Batauana fiel ihrer Rachsucht zum Opfer. Die Buschmainner werden in solchen Fallen ihrerseits blutig ver- folgt, und eine Kette von Mord und Todschlag folgt der Ubeltat der Kaffern. Derartige Faille ereigneten sich fortwaihrend. So hatt z. B. im Mai I897 ein Batauana an der Kamelpan im Chansefeld eine Buschmannfrau geraubt. Diese entlief ihm. Er kehrte zurick, fand sie bei ihrem Mann, erschoss diesen und nahm die Frau wieder mit. Die frechsten Gewalttaten gestatteten sich zur Zeit meiner An- wesenheit im Chansefeld die Bakalahari von Kuschi."

I9. Khutsi, where Kent (i990) claims to have found full-time hunter-gatherers who are not locked into any patron-client rela- tionship with Bantu people in a persisting "hinterland" (Solway and Lee I990: I I I, I I4), is the most frequently visited game reserve in Botswana, largely because it is only three hours by car from the capital, Gaborone; it has been a locus of Kgalagadi and Kwena residence for centuries (Okihira I976, Ramsay 9o90). 2o. Lee (I979:433) thought of his own work as a " 'revisionist' account of a foraging people"; Passarge (I907:5) says that his were "in many ways totally different from everyone's previous views." We are certain that our work is producing a "radical revision of the received view of the Kalahari" (Wilmsen ig8ga:xiv), but to erect on this a category of "revisionists" is to reduce scholarly activity to tactics and scholarly discussion to polemics.

494 CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

pristinity, . . . are in danger of . . . doing violence of a different kind to the data, . . . [projecting] a spurious uniformity on a vast and diverse region" and robbing foragers of their history.21 But it is they themselves who revive Passarge's pessimistic notion that an essential element of humanity was erased when the pure life of the hunter was replaced by domestic pursuits. They set the stage (p. II8) for

the final act in the transformation of the Dobe-area !Kung from a relatively autonomous people.... The bush had always been the backdrop to economic change, giving the !Kung security and a degree of free- dom.... by I 970, however, four decades of intensive and expanding pastoralism had begun to take their toll on the capacity of the environment to support hunting and gathering.

Passarge was almost a century ahead of Solway and Lee in stage design and in writing final acts, yet their scripts are nearly identical: bushmanness can survive only in the bush; nothing significant intruded into the bush and thus into aboriginal bushmanness until three or four de- cades before the script writer arrived;22 prior to that there were only a few "long-standing but non-decisive linkages" (Solway and Lee i990:ii8); intruders, pas- toralists and traders, destroyed the bush world and brought an end to bushmanness. Much of this is true, if overdramatized: we wince with every "bush or tree cleared to make way for cultivation" (p. I 4); we cringe as goats browse away berry bushes, and the pounding of hooves destroying square kilometers rings in our ears (p. II 8). But in what years? Can the bush have been laid waste and bushmanness have vanished at Dobe before I890 and yet have persisted there until I970? Indeed they can, for the play is about allegory, not

actuality. Once established in scholarly and scientific- and, of parallel importance, popular-lexicons, the events, peoples, and categories that become the objects of allegorical discourse are transmuted into "indexical signs which perpetually point to their status as realities constituted independently of the process of re-pre- sentation itself" (Alonso i988:36). The categories "Bushman," "San," "hunter-gatherer," "forager," and so forth, are products of just such transmutation; they be- come objects and function to illuminate and legitimize a crucial area in Euroamerica's symbolic reconstruction of its own ontology. Neither Passarge nor the architects

of the Harvard Kalahari Project derived their construc- tions of primal categories from a systematic evaluation of evidence; rather, they appropriated them pretty much unchanged from this ontological discourse (see Lee I979:Xvii, I). It was just this discourse, Pratt (I985) tells us, that

formed "what Mr. Barrow saw in the land of the Bush- men" a full century before Passarge, in I797. The form of this discourse continues to shape modern ethnographies of "Bushmen," ethnographies that serve to authenticate Euroamerica's subjective representation of its own past by fitting an iconic "Bushman" into a prefigured cate- gory (Fritsch's Urrasse), often called "primitive" (not only in the igth century) and ethnographically labelled "hunter-gatherer" or "forager" today. By displaying objectified peoples as exemplars of this category existing "in a timeless present tense . . . not as a particular histor- ical event but as an instance of a pregiven custom or trait" (Pratt i985:i2o), ethnography validates the epis- temological program required by the ontological quest. Consequently, the intrinsic realities of these objects are in themselves of little or no interest. What is important is that the objects conform to a discursive narrative; while any of the parts may be questioned at any time, the ontological reconstruction itself becomes unchal- lengeable as a whole.

Hitchcock's remarks (i982:252-53, emphasis added) concerning contemporary Khoe-speakers on the Nata River, based upon his fieldwork there between I975 and I978, aptly illustrate the epistemological juxtaposition of icon and actuality that has been so characteristic of Kalahari ethnography:

Along the Nata River a specialized hunt leader, known as a Dzimba, emerged. This individual gradu- ally became more and more important in decision- making. Also, traditional medical practitioners, Cho K'aos, have become increasingly important in group affairs among Nata River Basarwa. Dances and cur- ings, over which doctors preside, have become large communal affairs, and there is a sense of increasing ritualization of various other aspects of life among Nata Basarwa. Leadership roles have also begun to emerge among settled Basarwa. In the mid-I970's I was able to observe a situation in which the Nata people elected their own leader.... The newly ap- pointed leader immediately began to adjudicate dis- putes, hearing cases in a Kgotla or tribal court.

Assumed here a priori are that innovation is always a function of "advanced" influence on atavistic peoples and that Khoe- and San-speakers are such peoples. Be- yond this, ethnographic data gathered over a period of three years or even half a decade are hardly sufficient to support wide-ranging, ad hoc "evolutionary" scenarios, and published archaeological data from this region refute paradigms that purport to link an iconic and static past with an only recently changing present. A stone-walled ruin of Zimbabwe type, Tora Nju ("house of God" in the local Khoe dialect), with a large associated assemblage of

2i. On the contrary, we state clearly that our research "has pro- found implications for understanding relations among contempo- rary southern African peoples. In particular, those relegated to the ethnographic categories 'Bushmen' and 'hunter-gatherers' are seen to have a history radically different from that hitherto assumed. It is clear that, rather than being static, uniform relics of an ancient way of life, San societies and cultures have undergone transforma- tions in the past 2ooo years that have varied in place and time in association with local economic and political alterations involving a variety of peoples" (Denbow and Wilmsen i986:i5I4). 22. A similarly late date was also postulated for Bantu-speaking peoples' entry into southern Africa (Laidler I938, Schofield I940, Schapera I952).

WILM SEN AND DENBOW Paradigmatic History of San-speaking Peoples | 495

stone tools lies only 40 km south of the Nata River mouth. The history of this place, with Zimbabwe sherds suggesting a beginning at about A.D. I500, is attributed to local Khoe, not Bantu, by all the people living in the region today. A second stone-walled ruin, Kubu, also with many associated stone tools, stands atop a small granite hillock 40 km to the southwest of Tora Nju across the barren Sua salt flats. Kubu is also of Zim- babwe type. At the northern end of Kubu hill, ceramics related to i ith-I 3th-century Leopard's Kopje wares found in present-day Zimbabwe extend the chronology of Iron Age penetration of the region back even farther. Faunal remains collected from this locati-on include fish vertebrae, clearly an import from the Nata or Botletli Rivers 8o-ioo km distant across the Makgadikgadi salt pans; blue cane-glass beads from the Indian Ocean trade further attest to transcontinental trade stretching to the East Coast. The presence of two stone ruins with well- known Zimbabwean symbols of social and political status suggests that the Makgadikgadi Khoe peoples were deeply engaged with the hegemonic state. This re- gion of central Botswana must have contained resources of economic and political importance to the Zimbab- wean state; the most important of these would have been salt.

In the Nata area an indigenous Khoe industry func- tions to this day, manufacturing and sending salt north- ward from the Kalahari into adjacent Zimbabwe. The fact that a market for locally produced salt exists, espe- cially in the face of competition from a widely distrib- uted mass-produced product, argues for at least as great a value for this good in precolonial times. The manufac- ture and transport of salt from the Makgadikgadi extend back to the period of Zimbabwe florescence in the I4th and i5th centuries and perhaps even ear-lier. The fact that today Khoe-speakers are recognized as both the "owners'' of the ruins and the suppliers of salt indicates that they have controlled this product for a very long time. There is thus no reason to assume that in earlier times their social and political forms were "less com- plex" or "developed" than those of their Bantu-speaking neighbors. In fact, the archaeological data from the wider Makgadikgadi-Botletli region suggest just the opposite. Throughout this region over a hundred sites have been found containing Late Stone Age tools, distinctive Khoe ceramics, and occasional cattle teeth. Stone Age settle- ments containing ceramics of known Early Iron Age type dating between A.D. 6oo and 8oo have also been discovered, indicating that exchange relations with neighboring communities to the north and east were es- tablished at an early date (Denbow i984a, b). The dis- tinctive Khoe ceramics which occur on many sites in this region indicate that an indigenous ceramic industry developed here perhaps as early as the beginning of the 2d millennium A.D. Although not yet dated, these pre- historic settlements reinforce historic igth-century ob- servations about the settled life-style of food-producing Khoe in this region and suggest that more diverse and complex social and economic forms had been developed (like the ceramic industry) by Khoe many centuries ear-

lier. Livingstone (1858:72-73) provides an eyewitness record for this, reporting that "the Batletle, a tribe hav- ing a click in their language, and who were found by Sebituane [ca. i829] to possess large herds of the great horned cattle.... seem allied to the Hottentot family." On the other side of the Kalahari, Carl Hugo Hahn (Neueste deutsche Forschungen i867), one of the first Europeans to enter northern Namibia, also told of the central position of "Bushmen" in "more complex" activ- ities such as the salt and copper trade in the I85os-'6os:

At the lowest estimate that I can make, 50 to 6o tons of copper ore must go yearly to Ondonga. The Bush- men [Heixum, Khoe-speaking western neighbors of Zhu] are so jealous of this trade that to this day they have not allowed strangers, not even people of On- donga, to see the places where they dig.... Also, other Bushmen [these may have been Zhu, as no other "Bushmen" were in the area] prepare salt from saltpans in the form of sugarloaves and bring them to Ondonga to sell, from where they go on to other tribes, so that the salt trade is fully as important as or even more important than that of copper.

None of this is-or was-petty barter. The importance of control over trade in salt and in some places its contri- bution to state formation are well documented in the prehistory of Africa. Far from being small-scale relics of the Paleolithic, isolated since early times from hierar- chical social systems, these peoples were an integral part of the social calculus of these diversified social, eco- nomic, and political formations.

In the northwestern Kalahari, evidence for salt manu- facture and for trade in wild game meat extends back to the middle of the Ist millennium A.D. At Divuyu in the Tsodilo Hills, dated between A.D. 5oo and 750, numer- ous remains of clay salt strainers were found, indicating low-grade salt manufacture from sediments or ash at the site. This and many other aspects of the assemblage sug- gest that the Divuyu people were newcomers to the re- gion, unfamiliar with its resources and not yet inte- grated into local economies. However, despite extensive excavations at the nearby site of Nqoma (A.D. 85o- io5o), which follows immediately and is in part a devel- opment from Divuyu, no salt strainers have been found. This implies that a shift in salt supply had occurred, transforming it from a locally manufactured good to an imported item. This and, again, many other aspects of the assemblage suggest that an integration of indigenous and entering economies had taken place. The presence of numerous remains of fish along with lechwe and water- bok, all restricted to permanent water environments, provides additional evidence for exchange networks which linked different environmental and production zones in the Kalahari, the Delta, and beyond. Glass beads, cowrie shells, and copper attest to even wider networks ultimately reaching as far as the Indian Ocean.

Further evidence for indigenous "Bushman" trade in animal products and for large-scale game drives has been available for many years. Two kilometers from Tora Nju are the remains of a very large V-shaped game trap

496 1 CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

known to have been used for generations before the i88os but not since (Denbow I984b). This place, too, is acknowledged to have been the exclusive property of local Khoe people. Passarge (I907:78-79) describes kilometers-long fences of thorn bush leading toward deep pit traps and quotes an account by Theophilus Hahn of a game drive in which over 2oo Bushmen partic- ipated in the use of one such fence-pit structure. These structures are mentioned specifically in the vicinity of Rietfontein and in the Epukiro Valley, where Thomas Baines recorded large numbers of them (Charles John Andersson, mounted on his horse, fell into one); both of these places are inhabited by, among others, people who speak Zhu or Cexai. In a refreshing departure from ata- vism, Passarge (I 907:8o, our translation)23 offers the opinion that Batswana, who also used game drives of this kind, must have learned the technique from Bush- men; he says, furthermore, that "this method of hunting awakens in me a different kind of emotion from what Hahn mentions. I should think and am convinced that a people employing such primitive means to attain huge successes by all working in unison and with persever- ance ... could not possibly attain such powerful achieve- ments without a political organization." For Passarge this included leadership,24 and, as we have seen, he went on to describe it as he understood it to have been, observ- ing that it had been largely superseded by Batawana headmen and chiefs. Lee (I979:84) recognizes that many Zhu leadership functions were vested in Tawana head- men but envisions this as having happened a century later, in I948.

Again, however, both archaeological and ethnographic data suggest that the "emergence" of hunt leaders and other political and ritual coordinating if not controlling roles among Khoe and San peoples was already well under way, and on linguistic evidence complete, many centuries ago. Hitchcock's observations indicate that newly elected Khoe leaders wasted no time learning how to adjudicate disputes in a kgotla (court) setting. Clearly, no learning process was involved, nor were there insur- mountable structural or cultural obstacles to the per- formance of such judgmental tasks. There is no "gradualism" here, and short-term studies carried out in the 2oth century should probably not pretend to recap- ture processes of transformation which occurred cen- turies earlier. The concordance of Nata Khoe terms such as xkaihas, meaning "chief" (Hitchcock i980:25) or "territory owner" (Hitchcock i982:243), with the virtu-

ally identical terms noted above in the central and northwestern Kalahari languages (Zhu xhaiha, Gcwi xheixama), for instance, supports the proposition that both Khoe- and San-speakers have been entirely familiar with the structure and praxis of nonegalitarian social forms for a very long time. Oral, historical, linguistic, and archaeological data are all consistent with the prop- osition that Khoe and San peoples, far from being icons of aboriginality, developed and controlled the means of production and trade over large parts of the Kalahari in- terior in earlier centuries, only being subordinated as "Bushmen" in the I gth century when relations of power in the region were unbalanced by European-inspired commodity production followed by colonial rule.

While Passarge's ontological program was in essence the same as that of modern ethnographers of San peo- ples, his premises, predicates, and postulates were en- tirely different. He recognized that aboriginal bushman- ness existed only in earlier times (friiheren Zeiten) and situated his iconic bushman firmly in the past, while in the present (der Jetztzeit) people called Bushmen lived in their own time engaged with other peoples in the entire range of activities and institutions that constituted an integrated social formation. Economic activities sur- rounding both pastoralism and foraging were among these, with the needs of pastoralism being paramount and pastoralists in control. In the epistemology of mod- ern ethnographers of San peoples, in contrast, the actual activity of foraging is equated with a constructed cate- gory "forager"; the activity itself and selected peoples observed to engage in it are transmuted into indexical signs, and San-speaking peoples become iconic bush- men-caricatures not only of themselves but also of a wider class of Pleistocene hunters and gatherers (repre- sented by such statements as "The primary reason the Basarwa have drawn so much anthropological attention is that many of them still practice a way of life that characterized most of human history" [Hitchcock and Ebert i984:3231).

Lee's ontological program leads him to strange twists of logic and odd interpretations of his own empirical data. He describes what he thinks of as the "first known visits to the interior by non-!Kung" in the I870s (I979:77-78), quoting first a Motswana:

When we Tswanas first came into this area.... [the Dobe Zhu] were very afraid of us and would hide whenever we came around.... So we just ruled them. There was no killing or fighting. We would say, "Come here, give me water" or "Come here, and bring me my horse" and they would bring it.... It was good that they were so afraid of us, because if they had tried to fight, we would have slaughtered them.

and then a Zhu:

They put us under the carrying yoke. We had to carry the meat that they shot from the kill sites back to the camps; and then at the end of the summer hunting season, a line of porters would carry bales of biltong

23. "aber diese Jagdmethode erregt in mir andere Empfindungen, als Hahn sie aiussert. Sie erweckt vor allem, sollte ich meinen, die Uberzeugung, dass ein Volk, das mit so primitiven Mitteln so gewaltiger Leistungen fahig ist, die nur ein geordnetes, vielkop- figes, lange Zusammenarbeiten ausgefuihrt werden k6nnen, not- wendigerweise auch geordnete soziale und politische Verhaltnisse besitzen muss." 24. Whether large-scale cooperative activities require formal lead- ership is much-debated, but they do require coordination and some sort of supervision; thus, while game drives in themselves do not necessarily mean that leaders were present, they do add weight to the rest of the evidence for leadership.

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples | 497

[dried meat strips] back to Tsau.... The !Kung also did their own hunting and brought the hides and bil- tong in for tobacco.

Unlike Passarge, who saw this as a form of subjugation and dispossession, Solway and Lee (i990:II6-I7) dis- miss it, along with the Zhu-Tawana economic relations described in the account of !Xamn!a that they quote, as barter, a "form of linkage [that] does not lead to incorpo- ration and loss of autonomy." One wonders what consti- tutes autonomy. People who were "just ruled" and would have been "slaughtered" if they had resisted, who were "put under the yoke" and "had to carry" the pos- sessions of others, were not autonomous in any sense.25 MacKenzie (i883:62-63) used the same terms for the eastern Kalahari when he observed that

the contest for the possession of certain villages of Bakalahari or Bushmen is a fruitful source of strife in Bechwana towns.... the vassals with all their pos- sessions are the subject of litigation and endless jealousies; and it needs all the skill of a chief to settle these matters between greedy and plausible rivals.... When one Bechwana tribe attacks another, the Bush- men and Bakalahari belonging to both are placed in the same category with cattle and sheep-they are to be "lifted" or killed as opportunity offers. In such cases, therefore, all Bakalahari and Bushmen flee into wastes and inaccessible forests, and hide themselves until the commotion is past.

The accounts of MacKenzie and Passarge make it clear that the outlines of the process of hegemonic subjuga- tion were similar throughout the Kalahari even if the outcomes displayed local variations.

One also wonders what constitutes dependency, for we are told that not only was "the setting up of !Kung- operated cattle posts in the interior the main business of the period i890-i925" (Lee I979:79) but also

the Tswana would gather all the hunting parties at a large water hole like /Xai/xai, and intense trading would take place at the site of the ox wagons (koloi). The !Kung would bring in their hides, furs, bales of biltong, and pots of wild honey; in exchange they would receive such items as balls of tobacco, clay pots, ironware and beads, and clothing of European manufacture. Large groups of !Kung would congre- gate, and there were dancing, trading, and marriage brokering among the !Kung themselves as well as trading with the Tswana. At the end of the festivities, the oxen were inspanned and driven back to Tsau.

But it was only pack and trek oxen that are said to have been taken back; breeding and milking stock remained in the area: "When Hxore water was dry, [CaeCae Zhu] . . .drove the cattle back to /Xai/xai" (Lee I979:79; Solway and Lee I990:I I6); Zhu "went with the cattle at the beginning of the dry season to the Debrafeld" (Pas-

sarge I907:i2i); and at the height of the dry season in September I903 Zhu are recorded to have stolen i 6o head (plus "calves just born") from two Tawana cattle posts belonging to Sekgoma Letcholethebe, kgosi of Batawana-mentioned specifically by Passarge (above)- in Dobe-NyaeNyae (Wilmsen Ig8ga: I34-39). Solway and Lee (I990:II7) think that "only a minority [I0-20% (Lee I979:80)] of Dobe-area people became involved [in these herding activities].... they were islands of pas- toralism in a sea of hunting and gathering." This is em- phatically not so; the family histories of every CaeCae household contain accounts of participation in herding and mercantile trade from the middle of the igth cen- tury (Wilmsen ig8ga:I97-225) to the present ("most of the men had had experience herding cattle" [Lee I979:409]). From Passarge's account it is hard to imagine that many people were outside the events he describes.

In contrast to Passarge, who recognized that Zhu in his time were already incorporated into a social formation dominated by Batawana and that this had profound eco- nomic, political, and social consequences for their cur- rent lives as well as for their history, Lee (I979:79) thinks-despite his own excellent evidence to the con- trary-that the "Tswana impact on the Dobe area at the turn of the century was modest." For him, "on the eve of the European colonial incursions, the !Kung were evi- dently occupying the interior on their own as hunter- gatherers [merely] producing a small surplus of furs and other desert products for barter with agriculturalists on the western margins of the Okavango" (Solway and Lee I990:II5). Contradictorily, while now wishing to situate San peoples in history, he still finds them to be "exemplars of a hunting and gathering way of life" (i989); to Solway and Lee (I990:I22) "foragers are, for whatever reason, people who have resisted the tempta- tion (or threat) to become like us." This is just Levi- Strauss (i967:47; i963:233-34) all over again:

foragers are societies, which we might define as "cold" in that their internal environment neighbours on the zero of historical temperature, [and] are, by their limited total manpower and their mechanical mode of functioning, distinguished from the "hot" societies which appeared in different parts of the world following the Neolithic revolution....

[that is to say, they are] "peoples without history". . . seeking, by the institutions they give themselves, to annul the possible effects of historical factors on their equilibrium and continuity.

Lee wants to have it both ways. Anything that hap- pened before I925 remains unimportant in his eyes (we do not tar Solway with the Dobe brush): "The Dobe area, 700 km north of Dutlwe, was far from the turmoil of igth-century colonial southern Africa.... the wave of black settlement did not reach them until I 925 " (Solway and Lee I990:II5). Even the Early Iron Age, now con- ceded to have impinged on Dobe (though only in the form of "a few fragments of pottery and a few iron imple- ments, best interpreted as evidence of intermittent trade

25. Not even in the "philosophical" sense that Silberbauer (i989:207) now wishes us to imagine.

498 CURRENT ANTHROPOLOGY Volume 3I, Number 5, December I990

with Iron Age settlements to the east and north" [p. II51), was a gentle breeze leaving no lasting impression, and more recent times remain similarly unruffled by events. One could hardly neighbor more nearly on the zero of historical temperature.26

The word "history" recurs in Lee's narrative, but there is little historical substance. Dobe history, long as ours from the Stone Age to the present, could be written in nine pages (I979:76-86), all of it "event history" under the heading A history of contact, useful mainly for de- termining calendar-year ages of persons who could be associated with particular events. And this is the nub: history for Solway and Lee-and for Lee (I979), at least, before their joint paper-is simple historiography, Braudel's 1'histoire elemental (Matthews I980), the writ- ing down of events that can be remembered. They do not understand history as historicity, the process through which social formations realize their transformations, nor do they appreciate that historiography is the process through which those transformations come to be under- stood after the events. Lee and others have long written, at times movingly, of the experience of persons as indi- viduals, but their "history" as social beings remains an iconic abstraction.

We have here not a "Coke Bottle" but a Kuhnian syn- drome, pure paradigm myopia, and one recognizes Fritsch's (i 906:74, our translation)27 discomfort:

The rich observations Mr. Passarge collected about the Bushmen could offer estimable opportunity to en- lighten present conditions in terms of the past. It is impossible for this to be the favourable outcome when the author, instead of beginning with the dis- coveries of the older investigator, discredits those very observations and plants his own modern opin- ions backwards into olden times.

The same intellectual forces that led Lee to overlook Passarge and archives ("Written records were virtually unknown for this area until the Marshalls arrived in I95I" [Lee 1979:76]) have required him to ignore the uncomfortable elements in the corpus of evidence he himself has played a key role in assembling. These forces are responsible both for the incompatibility between the "Harvard researchers' sincere desire to be sensitive to the !Kung's situation in present historical circumstances . . .[and] . . . their project of viewing the !Kung as a complex adaptation to the Kalahari desert, and an ex- ample of how our ancestors lived" (Pratt I98 6:49) and for "the widespread willingness among anthropologists (and, needless to say, of the public as well) to overlook the serious methodological flaws of the Harvard Kala- hari Project" (see Wilmsen I989a:I-63).28 The flaws in question are anecdotal scientism, uniformitarianism, and (added later by Lee) an undigested Marxism char- acterized mainly by naive historicism.

It is naive historicism, perhaps more than anything else, that allows Solway and Lee (i990:ii9) to dismiss serious, complex issues such as the adjudication of land tenure rights in modern Botswana with the remark that "the local blacks have a healthy respect for the San's determination to protect their water claims, in the last century by force of arms and today through the courts." Not only does this contradict Lee's own evidence quoted above ("[they] were very afraid of us and would hide whenever we came around") and MacKenzie's parallel evidence, but it makes a mockery of centuries of struggle among peoples over just those rights and the disparities in power that continue to deprive San peoples of land rights. But then, Lee (I 976:7 5) from the beginning rejects "the conceptions or folk view of [San] peoples about themselves and their land" in favor of their behavior toward it-adopting a narrow ecologism focused on in- strumental means for extracting subsistence from the environment in an unbounded universe. Although this was a well-regarded, indeed apparently liberating para- digm in the I96os, its distortions of both theory and practice had become apparent by the mid-I970s. The obstructions placed by this paradigm in the way of San attempts to gain recognition of their land tenure rights in the I98os are thoroughly aired by Gordon (i989), Hitchcock (i980), Wily (I979), Worby (i984, n.d.), and Wilmsen (i989b; see also contributors to Wilmsen I989C).

On one point we thoroughly agree with Solway and Lee (i990:i1i); the claims we make "go well beyond the reinterpretation of Kalahari archaeology." "We are deal- ing, after all, with . . . the way anthropology founds its knowledge claims" (Wilmsen I989a:xiv). Our work, along with that of Gordon, Howell, Schrire, and others, provides a potent corrective to simplistic distortions of

26. Lee thinks that the Dobe area was "accessible to Iron Age peoples with livestock for only a few months in years of high rain- fall,and even then only after an arduous journey" (Solway and Lee I 990: I I 5). But rigors are subjective, and walking or riding an ox (as was done until horses and donkeys were introduced) with one's herd is not the same as driving a car over difficult tracks that can hardly be called roads; the journey obviously was not arduous to those who went out to CaeCae for annual trade "festivals" or drove pack oxen to and from the Delta for tobacco or to the herders and hunters mentioned by both Passarge and Lee. Again, foragers are here endowed with abilities that others lack and simply assumed to be capable of overcoming natural obstacles that discourage others, especially pastoralists. As to accessibility, stable numbers (in thousands) of livestock have been recorded in the Dobe area for decades through severe drought, foot-and-mouth disease, and other difficulties; cattle are routinely moved in herd management strate- gies between CaeCae, Qubi, Dobe, and the Delta every year. The Dobe area was equally accessible to herders in past centuries, sometimes probably more so during the cycles of higher rainfall identified by geomorphological studies (Cooke I975, I983; Shaw I985). 27. "Die reichen Beobachtungen, welche Hr. Passarge auch iuber die Buschmanner gesammelt, konnten eine sehr schatzenswerte Moglichkeit darbieten, die jetzigen Verhiiltnisse aus den friuheren abzuleiten; dies kann aber unmoglich von gunstigen Erfolg sein, wenn ein Autor anstatt von Erfahrungen den alten Forscher aus- zugehen, damit anfangt, dieselben zu diskreditieren und eine modernen Anschauungen riuckwarts in die alten Zeiten zu verpflanzen. "

28. Judging by some of the comments on Solway and Lee's article, this willingness still persists to a degree. "Revisionists" are ac- cused of "stridency," and this, rather than examination of sub- stance, is given as grounds for rejecting their cause. This is simply a rationalization of complacency in a paradigm.

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples | 499

the ontogeny of past and present Kalahari social forma- tions. The theoretical issue is transparent. Do we look at peoples encased in "cultures," in closed societies that are reifications of that which is considered "unique" in their lives? Or do we look at them engaged in social formations continuously shaped in an arena of economic and political action?

The End of Illusion

We will now briefly sketch those social formations, set in the political economy of the region, and their transfor- mations through the past I,500 years as they appear to be deducible from data. A large number of sites in the eastern hardveld Kalahari of Botswana were established by pastoralists during the 7th-Iith centuries A. D.; the earlier of these contain ceramics that are similar to others that are distributed widely in Zimbabwe and northern Transvaal by A.D. 6oo. Iron and copper tools and ornaments, most if not all of them manufactured locally, are abundant. Rondavel houses of the kind still commonly made of a cowdung-clay plaster applied to wattle frames in the region are preserved at some places. The agropastoral economic form of these settlements is evident in the remains of cattle, sheep, and goats, which at some of the larger sites make up 8o% of the faunal assemblage, the remaining 2o% being of hunted wild animals. Dung vitrified by burning marks the presence of kraals on the majority of the 320 sites known from this period; at the larger sites, these reach diameters of i00 m and depths of I 50 cm, evidence that large herds were kept. Sorghum and cowpeas appear to have been the principal crops. Finally, cane-glass beads manufac- tured in India or Arabic Asia and cowrie and Conus shells, some of species that live only in estuaries of the Indian Ocean, were found at four of the largest sites. This is certain evidence that before i,ooo years ago ag- ropastoral peoples in eastern Botswana participated in exchange networks that reached the east coast of the continent.

A tripartite hierarchy of settlements may be discerned in terms of site size, location, length of occupation, pro- portion of exotic trade items, and relative numbers of domestic stock; social stratification of the inhabitants seems to be clearly indicated (Denbow I982, I983). Toutswemogala, one of the largest of these sites, covers about ioo,ooo m2; it was occupied (perhaps with inter- ruptions) for about 500 years and contains a very large kraal deposit as well as many trade items. Second-level sites, such as Taukome and Thatswane, cover an area of about io,000 m2 and appear to have been occupied for only 200-300 years. Kraal deposits, while large, and trade goods, though numerous, do not match those of Toutswemogala. At the tertiary sites, kraals are small, averaging about 30 m in diameter, ceramics are numer- ous, and stone artifacts typical of the Late Stone Age of southern Africa occur. Significantly, stone artifacts are absent from the higher-order sites. Given the concentra- tion of lithics at these third-level sites, it is possible that

a fourth level in which hunting and gathering was more important than herding and farming remains to be found.

A further indication of social stratification is provided by analysis of the age at slaughter of cattle at these sites. At the later chiefly site of K2, and at Mapungubwe a century later, prime young-to-middle-adult animals- most desirable as food but important for reproducing the herd-were killed in far higher proportions than juvenile and aged ones. At Taukome, in contrast, juveniles (prob- ably yearling bull culls and runts) and old, postreproduc- tive animals were most often slaughtered, this is the strategy practiced in rural Botswana today by subsis- tence farmers who emphasize herd maintenance at the expense of meat production. It appears that, at this early date, an elite was already able to extract prime food re- sources-along with a surplus product distilled in value as exotic trade goods-for its own use from subordinate classes. Many if not most of the prime animals slaugh- tered at K2 must have been obtained from lower-ranked locations, for a sustained offtake of breeding stock would soon have reduced the resident herd to unsustainable numbers. It may even be appropriate to think of tertiary- site herders as cattle managers for an elite rather than as cattle owners in their own right.

The history of pastoralism in the western sandveld Kalahari has a similar chronology although it differs in a number of significant social and economic details. Di- vuyu is a fully developed Early Iron Age site very rich in ceramics, iron and copper tools and ornaments, and ivory. Its ceramics have design affinities to sites roughly contemporary with it in central Angola (Clark I968, Martins I976, Ervedosa i980) and to the Early Iron Age sequence now emerging from north of the Congo/Zaire River, which dates between the 2d century B.C. and the 5th century A.D. (Denbow, Manima-Moubouha, and San- viti I988, Denbow I990). Goats (and possibly sheep) were mainstays of the Divuyu economy, but cattle ap- pear to have been rare and to have been kept elsewhere. Two marine shells of Atlantic Coast origin (Cerithiidae) and two iron pendants were found at Divuyu; the pen- dants are virtually identical to specimens of the same age found in Shaba Province, Zaire (van Noten I982: fig. 3 I). These items indicate that, during the first centuries A.D., this northern margin of the Kalahari was part of a wider sphere of production and exchange extending throughout a large portion of the Angolan and Kongo river systems. Fishbones and freshwater mussel shells at Divuyu, probably from the Okavango Delta some 70 km away, are further evidence for such exchange. Relatively small communities of Bantu- and Khoisan-speakers ap- pear to have intermingled throughout the region on rela- tively equal terms.

At present, only two other sites with Divuyu ceramic affinities are known: Nqoma, on a low plateau of the female Tsodilo Hill, and Xaro, on the verge of the Okavango Delta. For the period between A.D. 700 and i0oo, however, a number of agropastoral sites are now known in Ngamiland. Nqoma, the main components of which are dated to the gth and ioth centuries, is the

500| CURRENT ANTHROPOLOGY Volume 3I, Number s, December I990

largest of these. Cattle, some of which were of a hump backed variety, were paramount in the economy of this site, but sheep/goats were also common. Sorghum, mil- let, and possibly melons were grown, but mongongc nuts and Grewia berries-along with wild faunal re- mains-indicate that foraging continued to be impor- tant. Rondavel houses were constructed, and an elabo- rate variety of ivory, iron, and copper ornaments and many iron tools were made on the site. Cane-glass beads and marine mollusc shells, including money or ring cowrie (Cypraea sp.), provide firm evidence that Nqoma was an important local center in intracontinental trade networks that extended from the Indian Ocean coast by the gth century. Freshwater mussels and fish continue to have been imported from Okavango communities. Stone tools are abundant at Nqoma, indisputably a thor- oughly Iron Age site, especially in a section where tuy- eres, cultivated grains (sorghum and millet), and East Coast trade items (cane-glass beads and cowrie shells) are associated with a burned-clay hut floor; this is clear evidence that no one-to-one association between the presence or proportions or frequencies of stone tools and "foragers" can be made.

Matlapaneng, northeast of Maun on the eastern side of the Okavango Delta, is contemporary with Nqoma. Cat- tle, sheep, and goats were kept here, and sorghum, mil- let, and cowpeas were grown. Although Matlapaneng was as large as Nqoma and had all the same material characteristics, there are two quite significant differ- ences between the sites. Matlapaneng is not nearly so rich in metal ornaments or East Coast trade goods; in this respect, it resembles the secondary sites of the east- ern hardveld, such as Taukome, rather than Nqoma. In addition, Matlapaneng ceramics are typologically allied with those of the northeast and the hierarchical east, and this must mean that its social, economic, and political ties were in that direction rather than northwestward as those of the Tsodilo sites had been until this time. It appears that the dominant centers of the hardveld, Toutswemogala and others, were extending their eco- nomic interests into the western sandveld; that some Nqoma ceramic motifs display eastern influence is evi- dence that these interests were penetrating deeply into the west. It would seem that this was the beginning of hegemonic domination from the east that was con- solidated by about A.D. I000.

A series of smaller sites is assignable to the Ist millen- nium; three have now been dated. In the Delta area, small undated components containing ceramics similar to those of Matlapaneng are present at Lotshitshi and near Tsau. At CaeCae ceramics, iron, and cattle (Wilm- sen I978) are contemporary with Nqoma and Mat- lapaneng (Denbow and Wilmsen I983). At the nearby sites of Qubi, Magopa, and Qangwa (Yellen I970, I973; Yellen and Brooks I988; Wilmsen I988), similar ce- ramics as well as iron occur in small quantities. All the sherds from these sites are thin and appear to have come from sm4ll bowls or dishes. In the adjacent NyaeNyae area of Namibia, Kinahan and Kinahan (i984:2i-22) recovered ceramics from I3 of 2o inves-

tigated sites; they state that some of these ceramics may be related to charcoal-tempered Early Iron Age wares in Botswana.

Discernible in the foregoing is a hierarchical site struc- ture not unlike that described for the eastern hardveld during these same centuries but less elaborate. Nqoma appears to be at the apex of this western settlement hierarchy; it was about 2o,ooo m2 and was occupied (per- haps intermittently) for about 400 years. Nqoma yielded a predominantly pastoral fauna, with an even higher pro- portion of cattle to sheep/goats than found at the larger centers in the east; it contained moderate numbers of exotic trade items. In addition, it produced the richest (both in quantity and variety) and most elaborate metal ornament inventory known for any site of its time in southern Africa (Miller et al. n.d.a, b). At CaeCae, in contrast, although occupation debris is found through- out an area of more than a square kilometer, settlement seems to have occurred in small clusters similar to pres- ent-day homesteads in the area, which also are spread over more than a square kilometer. There is no way to determine how many of these clusters may have been occupied simultaneously, but it is unlikely that aggre- gate occupation area at any given time would have ex- ceeded 2,000 m2. Information provided for Qubi (Yellen I973) indicates that the same may be said for this loca- tion. The other sandveld locations, though known only from isolated, small test excavations, appear to be simi- lar in most respects. Contemporary levels at Lotshitshi are also in this range. As already noted, ceramics and metal occur in very small numbers in these sites, cattle (also in very small numbers) are known only from CaeCae and Lotshitshi, and exotic trade items are absent from all current inventories. Excavations at Qogana, on an island deep in the Okavango Delta (Denbow and Wilm- sen I986), on the other hand, contain large numbers of ceramics, not a single stone tool, and a faunal assem- blage made up entirely of wild game. The archaeological data clearly indicate that Nqoma articulated the sand- veld and the delta within a wider, integrated regional economy-an economy in which diversity rather than uniformity would have been favored.

It appears that an elite was established at Nqoma that was able to exercise sufficient hegemony over the inhab- itants of secondary settlements to appropriate to itself the overwhelming preponderance of imported goods (glass beads and marine shells) that entered the western sandveld, as well as the bulk of the locally manufactured surplus product (metal and ivory ornaments) that was not exported. Some local products must have been ex- ported, for we may assume that imports were desired, thus were expensive and had to be obtained for value; otherwise, they would be more widely distributed among many sites. Judging from the presence of contem- porary ceramics at CaeCae, Qubi, Magopa, Qangwa, and the NyaeNyae sites, it seems likely that the Nqoma elite was also able to extract the required exchange value from its subordinates.

Solway and Lee (I990:II5) dismiss these ceramics as betokening nothing more than a bit of barter, of no con-

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples I 50I

sequence to the barterers. Yellen, whose excavations in I968 uncovered the majority of these ceramics, did not report them because "the assumption was that this stuff was so churned up that you couldn't say anything at all about it.... we couldn't even figure out what questions to ask" (Yellen I989). In response to our work, Yellen and Brooks (i 988:ii) are now reevaluating these earlier excavations; their current proposal is that the sherds were picked up as curiosities (this is not a new notion; Walker [I983:88] notes that in similar circumstances it was for a time the favored refuge for those who did not like to see Bambata ceramics surrounded by all those Late Stone Age lithics). These same authors have long rejected the evidence for cattle at CaeCae, arguing first that the I, i 5o-year-old specimen of Bos taurus ex- cavated by Wilmsen (I978, i982; Denbow and Wilmsen I983) was misidentified and now that it is intrusive. But the data are simply too numerous and too consistent to be so easily dismissed. There is no reason to suppose that any of the sandveld sites is seriously mixed and therefore the evidence from it marred by intrusions. The 23 radiocarbon dates available for CaeCae and Magopa (2i obtained by Yellen, 2 by Wilmsen) are extremely well correlated with depth (r = o.86); Yellen and Brooks (i988:i0; Brooks I989) now concede that the stratig- raphy is not mixed to any significant degree. We must, therefore, consider the significance of these few sherds scattered about those places in the western sandveld.29 While it appears to be true that Early Iron Age and later ceramics are associated with pastoralists and agro- pastoralists in southern Africa, it does not necessarily follow that these ceramics functioned in pastoral pro- duction; indeed, some of the most common forms are patently unsuitable for such functions. Hendrickson (i986:20-25, 37-40), who has reconstructed vessel shape and size profiles recovered from Nqoma and Mat- lapaneng, reports that small bowls, many of them shal- low dishlike forms with capacities of a liter or less, pre- dominate. These dishes had relatively thin walls and were often elaborately decorated; many had a red slip applied to exterior surfaces. The majority of these ves- sels thus appear to have been designed for food prepara- tion and serving, while wood and leather containers were used for high-risk purposes such as milking as they are in the Kalahari today.

If Early Iron Age ceramic vessels were in fact designed primarily for personal or household use as cooking/ serving/eating dishes, they would have been important as material manifestations of the relations of production in the pastoral/pastoro-foraging social formation of the time. As such they would have been invested with asso- ciational value as attributes of the dominant centers, with possession of them conferring prestige. Accord-

ingly, ceramics (especially personal serving dishes) would have been desirable prestations in social relations between those dominant elite centers and their rural sources of supply. So as not to debase them, the centers can be expected to have parceled out these prestations carefully, limiting them perhaps to senior members of local descent groups (or some such persons) who would have been responsible for channelling a locally produced surplus-a form of precapitalist commodity-back to the centers. Local products that could have been com- moditized in this way include ivory, rhino horn, ostrich feathers and eggshells, aromatic woods (Cape sandal- wood [Spirostachys africana]) and corms (Mariscus spp.), red dye woods (Pterocarpus angolensis), limonitic red pigments, and furs of animals such as jackal (Canis mesomelas) and fox (Otocyon megalotis). The pigments (known from several sources in the CaeCae-Qubi- Qangwa area but absent from the Delta and Tsodilo) could have been the source of media for red slips on Nqoma and Matlapaneng bowls/dishes as well as for paintings on rock walls at Tsodilo. Dye woods are known to have been highly prized trade items at least as early as the i6th century (Martin I972), when they were used in dressing leather goods, as they still are today in southern Africa; Passarge found them to be an important tribute item at the end of the igth century, and it is not difficult to suppose that they were equally desired in the past. The occurrence of a similar pattern of site in- ventory at several locations in the western sandveld suggests that these locations participated on the same level in a social formation dominated economically by Nqoma (about ioo km away on average) and possibly other similar centers that are as yet undiscovered. The fish remains recovered from the Tsodilo sites provide indisputable indications of the Delta side of this economy.

Thus, in the earliest well-documented period (A.D.

6oo-iooo) of agropastoral penetration into the Kalahari, a regional differentiation of settlement organization and an associated difference in social formations can already be discerned in the archaeological record. In the east there was clearly an appropriation of indigenous Khoisan forager and pastoro-forager systems by Bantu-speaking peoples who colonized the area in numbers and quickly established a hierarchy of settlements around their cen- tral towns. Smaller sites with distinctive Khoe ceramics and stone tools such as those known to occur along the Botletli River (which links eastern Botswana with the northwestern Kalahari) may reflect in their size and con- tent the domains of pastoro-foragers whose position in the imposed social hierarchy into which they were incorporated was economically subordinate to that of pastoralists. It appears that by the end of the Ist mil- lennium A.D. eastern Kalahari communities were differ- entiated socially and economically in a manner similar to that of historically known and contemporary social formations found in that same region. The presence of sites with non-Iron Age ceramics in the Botletli River region adds further complexity to prehistoric social for- mations in the Kalahari.

29. Solway and Lee (i90o:ii5) mention only the ceramics ex- cavated at CaeCae by Wilmsen. In addition, Yellen recovered I6 sherds from several levels (including Early Iron Age) at that site as well as io from Magopa and 6 from Qubi (also including Early Iron Age sherds) in I968-69; these went unreported until I988 (Yellen and Brooks I988:8-9).

502 CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

In the western half of the subcontinent, agropastoral- ist economies with ironworking and transcontinental exchange networks were introduced at the same time as in the east. Bantu-speaking peoples were surely involved in the process in the northern peripheral zones where transmission must have taken place, and there is some historical evidence that they penetrated much farther south. But Bantu hegemony as it now exists was not established in the western Kalahari until the mid-igth century, and at first only to a limited extent. It appears, rather, that early agropastoral economies were trans- ferred among indigenous cattle-keeping pastoro-foragers, who could only have been Khoisan-speakers, and en- tering Bantu-speaking small-stock herding and hor- ticultural ironworkers. The mechanisms for these trans- ferrals are not yet entirely clear but probably followed long-established lines of interaction and then became internally differentiated according to local conditions. Forager and herder polities were less hegemonically structured in this area than in the east, as they are known not to have been in historic time until disrupted in the igth century first by Tswana and then by European capitalism.

A series of sites dating from the i2th to the igth cen- tury documents the presence of ceramic- and metal- using agropastoralists in the sandveld into the time of recorded history. As trade with the Indian Ocean in- creased in scale in the I 3th century, however, and gold became a major export from the interior while fine India cloth was imported (e.g., at Ingombe Ilede on the Zam- bezi River [Phillipson I977:I93; Phillipson and Fagen I969]), the nearer peoples on the eastern highlands of present-day Zimbabwe organized trading networks to their own benefit. The routes to the ocean from the west were truncated beginning in the I.2th century by Mapun- gubwe and then by the developing Zimbabwean states, Great Zimbabwe and Khami (Birmingham i983:25; Gar- lake I973; Huffman i982). The influx of exotic foreign exchange items into the western Kalahari was reduced, perhaps eliminated, just after the beginning of the 2d millennium. From then on, the western Kalahari was relegated to the status of producer rather than receiver in the Asian trade, for reasons that were not environmental but political. Tsodilo today-as in the time of historical records-has less available water than and supports only 2o% as many people and herds as CaeCae, yet it is Tsodilo that was the site of the Nqoma center, with all its evidence for extensive trade. That these hills were also ritually important to Khoisan, as is evidenced by thousands of rock paintings (many of which depict cat- tle), provides a symbolic as well as a material link be- tween Bantu and Khoisan in the region. The collapse of Nqoma very strongly points to intervention by domi- nant powers from the east. Salt, ivory, and cattle from Namibia-Ngamiland probably continued to enter the Angola-Kongo trade to the north as these products are known to have done in the igth century (Miller I976; I983:I2I, I27). In this trade, shells, metal, and dried fish were among

the items distributed in return, but we cannot document

their entry into Namibia-Ngamiland until the Portu- guese initiated an Atlantic-oriented trade in the i6th century. Beads from this trade are found at the beginning of the I7th century at VunguVungu (Sandelowsky I979) and Xaro on the Okavango (Denbow and Wilmsen i986), at Homasi in the NyaeNyae pans (Kinahan I986), and in the upper levels of Nqoma. It was, incidentally, this i6th/ I 7th-century Portuguese trade that brought to- bacco to southwestern Africa; Zhu could not have traded with "Goba" or anyone else for this New World domesticate before this time as Solway and Lee (99go:ii6) claim. Sherds of a type called Bushman ware by Kinahan and Kinahan (i 984:2a2) are most numerous in the NyaeNyae sites; these have been radiocarbon-dated to the late i6th century (Kinahan i986:ii5). Similar ceramics with radiocarbon dates in the same century are found at Xomqoisi and Depression Shelter at Tsodilo, as well as at Magopa and CaeCae (Yellen and Brooks i988, Wilmsen i988); cattle remains are also identified at CaeCae in this period. Ceramics, iron, copper, and cow- ries reported from Xgi by Brooks and Yellen (I979:28) with unspecified radiocarbon dates of I IO, 495, and 8io years B.P. (Helgren and Brooks i983) may fall into this time period. Ceramics, iron beads, pink porcelain beads, and a maize cob recently recovered from the White Painted Shelter at Tsodilo-although as yet undated (L. Robbins and A. Campbell, personal communication)- fall into this period of Portuguese-inspired trade. It is clear that ancient trading routes continued to function despite the arrogation of luxury items by eastern hege- monic powers. Wikar, Brink, and van Reenan (Moosop I935, I947) recorded that this trade had reached the Orange River by the mid-i8th century and mentioned specific Bantu- and Khoisan-speaking peoples who were involved in it; the archaeological sites just listed, which are in-and surround on all sides-the Dobe area, give evidence that the peoples living there were participants in this trade. Given the magnitude and scope of this trade, involving the introduction of an important food crop, maize, and a highly prized narcotic, tobacco, it is very difficult to imagine that it passed through any place without effect.

It was into this political-economic environment that the igth-century European traders entered. Far from wandering off into the wilds hoping to find a fountain of fortune, these profit-seeking businessmen were follow- ing leads given to them by indigenous people long politi- cally and economically engaged with each other and seeking gain for themselves. Even Solway and Lee (i990:ii6) concede that Zhu look back upon this time as one of "intense social activity and economic pros- perity." This influx of Europeans, trading in ivory, os- trich feathers, and cattle, brought goods of many kinds and the force of arms into the region and transformed its economy and society once again. Van Zyl (Lee I979:78; Solway and Lee I990:II6) was not the only European in the Dobe area in this period: Andersson found oxen and calabashes in a "Bushman werft" at Mokoro in I859; Robert Lewis (Karobbie to Herero) left his name as Lewisfontein at what is now Qangwa in I863; Axel

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples | 503

Eriksson (Karuapa Katiti) shot elephants with Lewis at CaeCae at this time; the Boer Trekkers spent about i8 months at Debra in I875-76; William Wilkerson was killed in the area in I878 for his arrogance; James Todd, who hunted and traded with Eriksson and Lewis, is re- membered as Dowtli at CaeCae; some of Muller's cattle survived rinderpest at Qubi in I896 (see Wilmsen I989a for details on these and several other men who can be documented to have traded in the area). The traders turned ancient footpaths into oxwagon spoors some of which became well-worn roads-it took Schinz (I89I:358) just 36 hours to travel from Karakobis to Lewisfontein in I885. The prosperity of the time was paid for in the long run by dismemberment of the indige- nous social formation that prepared the way for the en- clavement of colonially tribalized peoples in the area (cf. Vail I989). A sherd from an English white-body cup with an embossed rose and a fragment of an iron spoon found at the Kupi site near Qubi are residues of this trader activity in the Dobe-NyaeNyae area.30

As suddenly as it was set in the I 85os, the commercial scene dissolved in the i89os. During the previous de- cades, the entire region had pulsed with activity; everybody had had a piece of the everyday action. Now the region seemed as empty and remote as it was later conceived to be. The Kalahari had been sucked dry of commoditizable wild animals, but it was only selec- tively destroyed, and those San and Khoe-as well as Herero and Tswana poor-who found it necessary could manage reasonably well on the diet of nuts and berries left to them, with odd additions of flesh from the re- maining animals, as they are recorded by ethnographers to have done. The remoteness imagined by outside ob- servers was not indigenous but created by the collapse of mercantile capital, which in its genesis and growth had dismembered earlier, well-worn native links of commu- nication. Neither the economy nor the social rela- tions nor the requirements of the people were as they had been a half-century before. The reins of external con- trol passed from the hands of merchant hunter-traders to those of policy-conscious colonial administrators. Where merchants (and missionaries) had dismembered networks, administration proceeded to enclave the seg- ments.

In contrast to the insistent intensity of the history we have condensed above, Lee's anecdotal presentation is no doubt seductive to some; it is a compelling story, compellingly told-"islands of pastoralism in a sea of hunting and gathering" sounds so good; "intense trading around ox wagons (koloi) where dancing and marriage brokering took place" sounds so fraternal, so gay; "a dozen boreholes in Bushmanland" sound so many, and we all have the fragile environment in mind and are thirsting to know why our exemplars can no longer be as we want them to be, for us. But can an endeavor with pretensions to scientific or humanistic status make any contribution on the basis of ad hoc assertions? We are

30. Mary Beaudry identified these igth-century materials.

certain the answer is no. Tanner (i984:626) alludes to this in a review of Leacock and Lee's (i982) Politics and History in Band Societies:

papers on the !Kung and G/wi San .. . demonstrate the unique position of the San in recent forager stud- ies, since their ecological and economic adaptations are presented as indigenous and largely beyond exter- nal influences.... [readers] would have benefited from a general overview of all Kalahari foragers in their relations with the sedentary groups and the states of the region, so that the conditions for !Kung and G/wi autonomy would be clear to the non- specialist.

We have presented a condensed overview of the mass of evidence available to provide that clarity.

Beyond this, we suggest that the evolutionary ecolog- ical paradigm carried to the Kalahari by Lee (onto which he much later grafted a pedestrian Marxism with "some- thing similar to Engels' primitive communism" as its central feature [Tanner i984:626]) has been discredited by his account. The conflict between the goals and the strategies of the Harvard Kalahari Project identified by Pratt is never resolved. Discussing the genesis of his work in the preface of The !Kung San, Lee (I979:xvii) records that he and his thesis committee (led by Sher- wood Washburn and Irven DeVore) recognized that

the undertaking faced some formidable difficulties of theory and methods. First, how were we to avoid'the implicit racism and biological reductionism of ear- lier anthropological work on the subject? Many nine- teenth-century writers had treated contemporary "savages" as "living fossils" or "missing links," an approach that had become thoroughly discredited.

But almost immediately, on the first page of the in- troduction to the book (I979:I), we read that

it is necessary first to consider why the hunting and gathering societies are important for social science. Peoples who live by hunting and gathering-Inuit (Eskimo), Australian aborigines, the Kalahari San- are among the few remaining representatives of a way of life that was, until io,ooo years ago, a human uni- versal.

Among these listed "hunting and gathering" peoples, only the Kalahari San have achieved the unique position in modern forager studies spoken of by Tanner, because only for them has a unique exemplary status been claimed. As Yellen (I984:54) puts it, "these San peoples were selected as a kind of narrow and opaque window to the Pleistocene." After living three years among them, Lee (I972:342) found the claim to be valid and placed Zhu "on the threshold of the Neolithic"; later (I979:6) they were "a foraging society on the threshold of a triple transformation to agriculture, to feudalism, and to capi- talism." And right up until this day, Yellen (i99o:io2b) can say that Zhu have "retained their foraging 'mental- ity.' " Surely to remain among the few representatives of a way of life that everyone else gave up io,ooo years ago

504 CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

is to be a living fossil.3' If one has a history one is not on the threshold of an earlier time; one may forage and do nothing else without retaining an atavistic forager men- tality and without being any more representative of for- agers io,ooo years ago than are modern Bantu agropas- toralists of early farmers in the Near East or the New World.

In the hunting model narrative, individual San- speakers are real persons who may have distinctive expe- riences, but San society is conceived to be a living fos- sil-human to be sure, indeed normatively so: "The San !Kung were preeminently UNESCO's universal man, hunter-gatherers on the open savannah, living en famille the human way of life" (Haraway i988:248-49). For Lee (I979:6), despite lip service to history, it was fieldwork among Zhu that could "begin to place this 'ahistorical' society into history to get a sense of its trajectory in historical materialist terms as a foraging society" on the thresholds just named. That these are not merely metaphorical thresholds is unambiguous in his concep- tion of the "nature of contact" between Zhu and Batawana described by Passarge at the close of the igth century (p. 40I):

The Tswana were a chiefdom with the beginnings of internal ranking, and the !Kung were immediately ac- corded a position at the bottom of the Tswana social scale. But it should be emphasized that in the Dobe area the San were not enslaved or enserfed; nor were they propelled into the cash economy. The nature of the contact, in socioevolutionary terms, occurred be- tween more or less adjacent stages within the se- quence of precapitalist social formations.

This must be read to mean that, as no social agency had placed them at the bottom of the social scale, their natural slot was predetermined by their evolutionary status-a status so radically different that it could not articulate in any meaningful way with the next higher one. Lee's most recent statements must be read to mean that he still believes this to be so, despite rather hopeful appeals to dependency theory: having "survived" a long list of political and economic interventions by Bantu and European intruders, "the !Kung became dependent only as a consequence of the inability of their land to support a foraging mode of production.... if [that mode] was so fragile, why did it persist?" (Solway and Lee I990: iI8, i20). Yellen (i990: ioo) still thinks that "the people [Zhu but not Herero or Tswana living in the very same places] we met were behaving much as their ancient ancestors had." How does this differ in its underlying premise from Service's (i962:8) view of forager society as con- stituted by ancient cultural forms preserved into the present?

The reasons for the Harvard Project's dilemma lie in Lee's adoption of a narrow ecologism (Wilmsen

19 89a: 6I); again, the dimensions of the dilemma are set out in the first pages of The IKung San (p. xviii):

the ecological approach was essential, because through it we could explore comparatively the con- tinuities and discontinuities in subsistence, ener- getics, spatial organization, group structure, and demography.... the more conventional social an- thropological categories-kinship, marriage, ritual, descent, and ideology-would be much more difficult to deal with from a long-term evolutionary perspec- tive.

Lee approaches the hunting model that springs from this manifesto narrowly by identifying a series of elements and their combination in a process of production. When this model, unchanged, is rebaptized the "forager mode of production" it falls into the "weak usage" of the con- cept (Clammer I978:i2), "the first 'pitfall" to be avoided in analysing pre-capitalist models" (Taylor I979:I5I). Taylor describes this pitfall as follows (p. i62): "the con- straints imposed by the conditions of production, them- selves determined by the development of the productive forces, express the conditions for the reproduction of the social formation. Phenomena such as kinship, religion, etc., are analysed as 'functional necessities' in relation to the level of development of the productive forces." Despite his Durkheimian group-structural function- alism, Lee eschews "the network of social relations an- thropologists are so fond of studying" in favor of "a net- work of energy relations" the study of which "anchors the ephemera of social life on the foundation of the natu- ral sciences" (Lee I979:250-5 I). It would be difficult to frame a clearer declaration: social relations are epiphe- nomena, secondary consequences of energy flow.

This is excessive scientism, for the wedding of Durk- heim and Steward is not at all radical. Steward's formu- lation of the band level of sociopolitical organization- with its overtones of Tonnies in recitations reminiscent of Gemeinschaft of blood, of locality, and of mind-was Durkheimian mechanical solidarity taking as its point of departure "human ecology or the modes of behavior by which human beings adapt themselves to their envi- ronment" (Steward I938:I). This was precisely Wash- burn's program, of which Lee's work was a part: to study "evolution primarily as a method of understanding hu- man behavior" (Haraway I 988:2 3 5).

Lee's anecdotal scientism is shored up by a spurious precision; nowhere does he present precise, system- atically collected data dictated by hypotheses whereby relations among variables may be examined in their very variation rather than in the fixed modes he envisions (Wilmsen I983). His extrapolation of his often cited three-week, single-season (i969) caloric input/output analysis (itself accurate and replicable) not only to all seasons but essentially to all years runs counter to the vast literature on drought cycles and production differ- entials in the Kalahari and, indeed, to his own discus- sions concerning variation in pastoral production. His continued insistence that forager production is more than adequate even in conditions of severe drought (Sol-

3 I . It must be said, however, that Lee recognizes that San individ- uals-as persons-are not fossils (I979: I): "They are humans like ourselves with a history as long as the history of any other human group."

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples I 505

way and Lee 99go:i20)32 is completely at variance with his report (I979:454) that "when we are brought into closer contact with their [Zhu] daily concerns, we are alternately moved to pity by their tales of hardship and repelled by their nagging demands for gifts." (One must bear in mind that this pitiable level of daily life was the condition found in the I9 6os, when according to Solway and Lee [I990: iI8] Zhu were still living in bush security and freedom.) More serious, he uses his input/output analysis as the basis for entirely erroneous assertions regarding Zhu dietary adequacy and nutritional well- being (for contrasting views, see Marshall I968:94; Trus- well and Hansen I968, I976; Tobias I975; Harpending and Wandsnider I980; Konner and Shostak I986; Wilm- sen i989a:225-57, 303-I5, 344, 35I-52; Wilmsen and Durham I988). His (I979:254) response to criticism, however, aims mainly to rationalize his methodology. As Johnson (I975:30I-3) says, this is beside the point; a researcher's belief that a single sample is intuitively rep- resentative does not substitute for adequate sampling.33 Solway and Lee (i990:ii8) continue to speak in quasi- quantitative terms: "mongongo nut harvests noticeably diminished in the i980S," "drilling of a dozen boreholes . . . in the early I980S aggravated these trends by lower- ing the water table," "food relief between I980 and I987 ... further deepened dependency." There are no data to support these speculations; they are simply journalistic appeals to legitimate concerns.34

Lee remains faithful to Steward's vision of band soci- eties "which evince a less complicated history, whose structure is simpler in content and form, and whose in- stitutions are extensively patterned by subsistence ac- tivities. " His own formulation of the hunting model (Lee and DeVore i968:ii-12) is vintage Durkheim filtered through Steward and buttressed by the evolutionist con-

sensus of Lubbock, Morgan, and Tylor. Marx does not enter into this model and does not appear in the refer- ence list or the index of Man the Hunter or, for that matter, in Kalahari Hunter-Gatherers (Lee and DeVore I976), and the "forager mode of production" (Lee I979: I I7) is simply a rephrasing of the original Durkhei- mian formulation. Indeed, since this characterization of foragers stems from a notion of natural constraints that probably should not be considered Marxian (Kahn and Llobera i98i:320), it is doubtful that it is a "mode of production" at all. Surely it is an undigested Marxism that imposes intricately technical homeostatic feedback mechanisms from systems theory (themselves under in- tense critical scrutiny in biological sciences) upon a highly problematic base/superstructure metaphor with only some four scattered pages of discussion (Lee I979:4, 436-37). And it is naive historicism to argue today that "each historical epoch exhibits a distinct cluster of fea- tures centered on a distinctive mode of production" and then in one sentence to associate "the culture core" with the mode of production (Lee I979:4). Such static analytical concepts do not lend themselves to the identification and explication of diversity and its rela- tion to transformation and change.

The smooth transition from unacknowledged Durk- heimian sociology linked with Stewardian cultural ecol- ogy to Marx-Engelsian historical materialism is never theorized; changes in terms are never addressed. Just as the hunting model of Man the Hunter becomes the for- ager mode of production of The !Kung San without any mention of its ancestry or discussion of how this can occur, so the "forager" becomes the "communal" mode of production (Lee i988)-practiced (behold!) not just by foragers who forage but by those who hoe and herd and even some who scavenge on an urban fringe- without pause to reflect, let alone determine what if anything this transition means.35 All it means is para- digm repair-normal science practiced in a normal, or- thodox way; the premises, predicates, and postulates, the very terms of all three formulations are the same. The major premises remain those identified by Haraway (i988:234): "potential, plasticity, human universals, and shared threat." Uniformitarianism is essential (Lee I979 :434-37):36

Foraging was the way of life that prevailed during an important part of human history. The modern for- agers do offer clues to this way of life.... the uni- formitarian approach is a method for treating the quantitative ecological data in a more rigourous way in order to sort out which aspects of foragers' be- haviour are central and closely linked to structure

32. Lee's anecdotal approach is apparent here; he tells us (Solway and Lee i99o:i12) that "in the drought of I964 Herero crops failed and cows were dry yet the San persevered without evident difficulty. In fact, the Herero women were observed gathering wild foods alongside their San neighbours." But Herero also persevered, as did their herds; and Herero women gather wild foods and Herero men hunt assiduously and successfully in every year, indeed with as much success as the most productive Zhu hunters (Wilmsen and Durham i988). Lee glosses over the dietary stress felt differen- tially-both seasonally and yearly-by Zhu, whose weights fluc- tuate seasonally in response to food availability, and Herero, whose weights remain stable (for data see Wilmsen i989a:3o3-i2); he also ignores Howell (I976). 33. When Lee's original input/output study was carried out in I964, ethnography was in what in the present context might be called its Stone Age of statistics. Not only was his procedure con- sidered adequate but his results were judged sufficient ground for grand generalities such as Sahlins's (i968) "original affluent soci- ety." Lee should therefore not be singled out for censure on this basis. 34. The assertion that mongongo nut crops are diminishing may be derived from a misinterpretation of Wilmsen's data collected dur- ing the I970S. In I979 almost no nuts were produced (Wiessner [personal communication] said that I976 was also an extremely poor year); in ig80, however, a bumper crop was produced almost none of which was harvested because drought relief food was abun- dant (see Wilmsen and Durham i988 for data and protocols). We have seen no post-ig80 data.

3S. That it means nothing substantive is suggested by Lee's (i988) attribution to the communal form of exactly the same predicate terms that he had previously assigned to the forager. 36. Lee thus brushed aside Levi-Strauss's (i968:35o) admonition, during the I966 "Man the Hunter" conference in the course of which the hunting model was legitimized, that "we should not try to use these recent hunter-gatherers to reconstruct events and con- ditions in the prehistory of mankind."

506 CURRENT ANTHROPOLOGY Volume 3i, Number 5, December I990

and which aspects of their behaviour are situa- tional.... therefore, when we discuss sharing and other central features of !Kung foraging, we are not simply looking at a central practice unique to them, but rather an expression of a universal theme in the foraging mode of production. Repairs are deemed necessary because the paradigm

has come under severe attack, not least because of its imprecision and inconsistency. When Zhu can have traded for tobacco before Columbus discovered America (Solway and Lee I990: I I 5), when cash can have been the medium of economic transactions in i gth-century Ngamiland (Lee I979:40I), we must be in another world-a world, perhaps, in which autonomy is just be- ing able to say no and run off into the bush, in which being put under a yoke and made to carry someone else's goods is independence, in which the implications of ap- pearing to be entirely subservient in the villages of neighbors by day but quite autonomous alone in the for- est at night can go unexamined-even slaves are auton- omous when asleep. In the actual world, the day of uni- versal man has passed; "Man the Hunter, once authorized by the human rights and antiracist docu- ments of the first decades of the United Nations, seem[s] a poor bearer of the new liberatory discourses and poli- tics regarding sexual and racial difference" (Haraway i988:252). The time when there were no archaeological data for the Kalahari and the prevailing paradigm per- suaded us all that archives were empty, when thought- to-be-foraging peoples were consigned in just about everyone's mind to simplicity in social relations, when we believed we could explain ourselves to ourselves through evolution alone is long gone. Neither anthro- pological knowledge claims nor the peoples upon which those claims are ultimately based nor the policy deci- sions influenced by them are well served by denying this historical fact.

As Karp ( I986: I 3 5) has written, "emancipatory social science can only be achieved through analyses that con- tain an element of auto-critique, which attempt to ex- amine how the conditions of research defined in the widest sense determine the research conclusion." To achieve this elusive end, it is necessary to reexamine the premises, predicates, and postulates of our paradigms against both evolving theory and emergent empirical evidence-not turn their flaws against "revisionists" set up to save them.37 The real problem with the hunting

model/forager mode of production is that, by postulating universal man on the basis of constraints operating on an obsolescent hunter (gatherer)38-to paraphrase Marks and Trapido's (I987:8) insight into the southern African context-holds out the prospect of evolution for some suppressed individuals while avoiding degeneration of the categories in which hegemonic power-bearers placed themselves. The anthropological concept of the closed culture serves the program admirably as it does all divi- sive ideologies-in a southern African context, not only imported authoritarian apartheid but also that of demo- cratic indigenous governments39 (cf. Dubow I987). Fa- bian (I986:45) reiterates the point: anthropological theo- ries that explain variation among "native" peoples in natural terms, such as heredity or stage attainment, rather than in moral or political terms and that con- struct scales of nearness/remoteness from civilization allow the placement of "cultures"-along with, un- avoidably, regardless of intention, the individuals in them-in adjacent stages of evolution (or "develop- ment") without confrontation of the segregating conno- tations-evolutionary, ethnic, racial-that this implies. The problems raised are not peculiar to the Kalahari and are among the most closely examined in anthropology today (see Comaroff I987). "In this context, remodelling a human way of life in the I98os will perhaps mean more the dismantling than the managing and rebuild- ing" of the model (Haraway i988:252).

That providing historical detail to their histories robs "foragers" of those histories (Solway and Lee i990:i22) is obscurantist rubbish, rooted in the model certainty that "foragers" are different.40 Vansina (1985:43I) con- denses the argument eloquently: "the time should be long past that hunter-gatherers are seen as 'roots of heaven' . . . [with] little or no real history as their lives have been similar generation after generation." The time has come to stop setting stages with imaginary final acts and to come to terms with issues. We see no reason to

37. Lee has changed, as he often insists and Wilmsen (I983:I8) recognizes, but only with respect to the time-I97oS-I98os-that postdates Harvard Project fieldwork with "foragers" (Haraway i988:249). Indeed, he sets his own prior ethnographic authority for both bushman and postbushman phases of his work: "the end of my main period of fieldwork in I969 found the !Kung still primar- ily hunter-gatherers ... the dramatic changes of the I970S occurred after my main field period" (Lee I979:86), when those "foragers" are depicted as suddenly embracing with enthusiasm or, some- times, resisting to the bitter end a wider world. Yellen (i990:96, Io2d) thinks of this as a "swift disappearance" of foraging brought about by "major changes in social norms" but finds it "puzzling" and says that "it is fortunate that a rather detailed portrait of the

Kung's traditional culture was compiled before the onset of dra- matic change." 38. The female "gatherer" was added in an attempt to coopt the first, feminist challenge to the hunting hypothesis, with its "funda- mental organizing axis of male dynamism in hominization" (Hara- way i988:252). 39. For example, the Attorney General's Chambers"'Opinion in Re Common-law Leases of Tribal Land," 23 January I978: "As far as I have been able to ascertain, the Masarwa [Bushmen] have always been true nomads. . . . it appears to me that . . . true nomad Masarwa can have no rights of any kind except rights to hunting." That opinion was handed down in the context of parliamentary debate over legislation that would privatize large blocks of com- munal land; in that debate, the then commissioner of lands cited the anthropological dogma that "Bushmen have no territories" as justification for the involuntary removal of San-speakers from lands they had occupied for many generations (Hitchcock i980:24; for a broader discussion see Hiatt I989; Maddock I989; Wilmsen i989a:i58-27I; i989C). 40. No one would argue that providing historic detail to centuries of economic, ideological, political, and social participation in a common European arena robs Europeans of their individual or col- lective history; indeed, it is just that detail which is avidly sought by individuals and nations alike.

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples 1 507

sustain a sterile "debate" about paradigms and will con- tinue to engage evidence in a discourse that takes theory and practice rather than personalities and positions as its dialectical poles.

Comments

M. G. BICCHIERI

Department of Anthropology and Museum, Central Washington University, Ellensburg, Wash. 98926, U.S.A. I4 VI 90

Reviewing recent publications related to hunter- gatherers is often a sad undertaking because, while con- taining substantive data and challenging ideas, they fre- quently fall into the genre of internecine struggles for dominance. I feel that Wilmsen and Denbow are more concerned with discrediting than with presenting and arguing their viewpoint. While questioning the credibil- ity of Solway and Lee's sources, their own case is ex- pressed in such terms as "strongly suggest," "appear to be deducible," "can be assumed," and "not difficult to suppose." The evidence they use to clarify their stand is condensed in a way more appropriate to a "comment" than to a full-scale article. A number of other perplexing problems remain: Does one become "indigenous" by be- ing born in the area or by arriving ahead of others? And ahead by a few hours, or by a thousand years? Is "Bantu" a linguistic or an ethnic label? Has the identity of the rock painters been unequivocally established? Maps would have been helpful, particularly for readers whose specialty is not the writers'.

Prehistoric reconstruction, too, remains problematic. The extrapolation of artifactual and behavioral "facts" across time and space requires strict data control. Prehis- toric data are objective but silent; they are made to speak through subjective interpretations of contemporary data. Wilmsen and Denbow's analytical construct is particu- larly vulnerable in this respect because their evidence is drawn from a very wide spatial and temporal spectrum. My personal experience on the Iramba plateau of Tan- zania, for example, suggests that the kraal may shift around the residential hut as many as five times per season in response to climatic and other factors. Given the great seasonal variation claimed for the region Wilm- sen and Denbow discuss, the correlation of archaeolog- ical kraals with herd sizes is far from certain.

I think that students of hunting-gathering societies do acknowledge the antiquity of contacts between food- gathering and food-producing peoples. Modern small- scale societies are found in areas and under conditions that are not only ill-suited to agropastoral enterprise but also far from ideal for hunting-gathering. It seems rea- sonable to postulate that, in this context, today's food collectors live in areas environmentally and demograph- ically marginal to the environments in which collecting strategies evolved. Given this refuge-area premise, one

can assume long-term and pervasive contact with inva- sive middle-range and state-system societies. That change-causing contact is a correlate of evolutionary ad- aptation is undeniable, but acknowledging change and assessing its effects are quite different things. For the "simpler" dwellers of the Kalahari, as for all societies, contact is not synonymous with loss of cultural identity.

Concern should instead be focused on the effects of such situationally variable contacts on hunting- gathering strategies. Among the Hadza of Tanzania, for example, the centuries-old agropastoral presence is un- disputed. The effect of such contact, however, varies yearly, seasonally, microenvironmentally, and, most rel- evant to this discussion, in degree of intrusion. This variation is well understood and verbalized by at least one band of Hadza, who until recently, when they were forcibly evicted, treated the variations as options whose selection caused limited alteration of their cultural cog- nition (see my work and Woodburn's).

Wilmsen and Denbow are right-we need to incorpo- rate new information and modify existing premises. Nevertheless, maintaining a critical stance should favor rather than impede our search for generalizations. It is legitimate to look for a hunting-gathering cultural pro- totype, to appreciate the time depth of its adaptation, and to relate it to the evolution of agropastoral econo- mies. As well-adapted food collectors proliferated over time, competition increased and food collecting became less efficient until it gave way, at different rates and in varying modes, to food production. This major shift in need-satisfying strategy was accompanied by a shift from an ethos of accommodation and integration to one of dominance over human and nonhuman elements. These are reasonable and significant although not ulti- mate generalizations; they should not be dismissed without proper and careful consideration.

LEWIS R. BINFORD

Department of Anthropology, University of New Mexico, Albuquerque, N.M. 87 I3 I, U.S.A. I 4 VI 90

This article makes several very important points and provides readers with a clear picture of the controversy currently being played out with regard to the widely published, much used and abused materials reported from the Kalahari area since Lee's work in the early I960S. I think it is clear that (a) during the active period of Lee's fieldwork he was not employing a Marxist per- spective and (b) his subsequent attempts to make his work appear Marxist have not been very successful. The term "revisionist" clearly annoys the contemporary crit- ics of Lee's characterizations of his Kalahari experience. Perhaps correctly, they see themselves as "setting things straight" and therefore as not revisionists but construc- tive critics. Perhaps justifiably, they see Lee's "conver- sion experience" as more aptly considered revisionism, and they point with skill to the inconsistencies in his work and to his linguistic accommodations, which are of

508 CURRENT ANTHROPOLOGY Volume 3I, Number 5, December I990

little substance. In presenting this point of view, Wilm- sen and Denbow make a strong and convincing case.

What is not being debated here or in other strongly stated arguments for "history as historicity, the process through which social formations realize their transfor- mations," is the degree to which this position yields adequate explanations as opposed to other approaches. For instance, according to Wilmsen and Denbow "the evolutionary ecological paradigm carried to the Kalahari by Lee . .. has been discredited by his account." Ironi- cally, some of Lee's views have been shown to be both inadequate and incorrect by staunch advocates of that paradigm (see Hawkes, Hill, and O'Connell i982:394; Hill and Hawkes I983:I40; Hurtado et al. i985:25). Falsification of a proposition does not prove the "truth" of one's intellectual biases.

The substantive issues separating the so-called re- visionists from Lee are not theoretical but existential: Were the Dobe !Kung isolated remnants of a pattern of life that was more common in the past, as Lee argued? If they were, then the descriptions he offered could be pro- jected into the past under a simple assumption of stabil- ity. (In fact, many researchers fell into this trap, as is evidenced by the "little Bushman model" of the Pleis- tocene offered, for instance, by Isaac [I978a, b] in his interpretation of the archaeological materials from Old- uvai Gorge and widely circulated by Leakey [I98I; Leakey and Lewin I977, I978].) Existential characteriza- tions are generally rather easy to evaluate, and falsi- fication impacts the theoretical biases of the persons set- ting forth the falsifying argument only if the intellectual implications are clearly linked to theories about the way the world works.

Much of the discussion of "paradigms" and "critical theory" seems to be aimed at establishing the "truth" of ontological propositions as a prerequisite for investigat- ing the world of experience. In science there is no such requirement for productive learning to occur. Determi- nation of who is "right" does not ensure that alterna- tives are necessarily better or more productive in the long run, only that some learning may occur when alter- natively reasoned research is conducted. The fact that some learning occurs does not ensure that all further learning will automatically flow from the new "conven- tionalism" that substitutes one set of ontological propo- sitions (e.g., that people are engaged in social formations continuously shaped in an arena of economic and polit- ical action") for another.

"Paradigms" used as they have been in this debate are the tools of missionaries and "word warriors" rather than scientists dedicated to the evaluation of the utility of our ideas about the world of experience. Neither Lee nor the "revisionists" are addressing the issue of how we can productively learn about variables from the contem- porary world of experience and the ways in which they interact to condition the organization of social life. Is it not reasonable to ask how hunting and gathering activi- ties and their variations differentially condition labor organization, cooperation, and scales of "corporateness"?

What about regional organizations and social articula- tions and their consequences for things that archaeolo- gists like to talk about, such as trade and exchange? Is it not possible to find cases in the modern world in particu- lar in which variables can be measured, monitored, and documented?

According to the Boasian view, the very fact of history ensures that comparisons are impossible because of the historical uniqueness of each case, era, or sample of ex- perience. This nihilistic posture is no more productive for learning than the one that assumes that a modern case is a static survival from a past era that can simply be projected back in time categorically. The former objects to uniformitarian assumptions in general; the latter makes uniformitarian assumptions about descriptions of experience (cases), not variables or relationships among variables. Perhaps both approaches are intellectual dead ends.

ROBERT GORDON

Anthropology Department, University of Vermont, Burlington, Vt. 05405-OI68, U.S.A. 29 v go

Wilmsen and Denbow are to be applauded for irrevo- cably destroying the myth of the "wild" Bushman/San/ !Kung/Zhu (or whatever one wants to call them). It is a dangerous myth because, as the ethnographic film maker John Marshall argued after viewing Paul My- burgh's People of the Great Sandface at the recent Northeastern Anthropological Association meeting, it has consequences for those labelled Bushman/San/Zhu. Wilmsen and Denhow make a persuasive case for a re- gional interactional perspective; it will no longer be pos- sible to argue, out of a conservative ethnographic partic- ularism, that they and others have ignored "diversity."

In his recent book on the political economy of the Kalahari Wilmsen (i989a) uses historical maps exten- sively to demonstrate the extent of interaction and trade in the Kalahari. Limitation of space may have prevented him from dealing with this map making as "the servant of colonial plunder," recognizing that "the vision and knowledge constituted by the map both preceded and legitimized the appropriation of territory" (McClintock I988: I 5 I). Now he has started to extend his discourse on the Kalahari to include the critical role of violence, a topic that scientists in the region have preferred to ignore.

While the Passarge-vs.-Fritsch debate was bitter, it was hardly "great." Smith (i987) has shown how struc- tural features inherent in the social organization of Ger- man anthropology-the dominance of the Berlin intel- lectuals, generational rivalry, regional differences, and ideological disparities-made such a "debate" inevita- ble. It was not the first debate on those labelled Bush- men; whether Bushmen were sui generis or simply im- poverished Khoi is a question that dates back at least to the pages of the South African Commercial Advertiser

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples I 509

of I829. That it is also a hardy perennial is evident from the debate in German cultural-historical anthropology as to whether Bushmen constitute an example of pri- mary or secondary "primitivity." Why does this debate continually resurface? Is it because of academic igno- rance, or is it because if "wild" Bushmen did not exist we would have to invent them?

Given this context, I am uneasy about having Passarge carry the banner of the "revisionists." He was not the first person to write an ethnography on Bushmen. That distinction must surely go to Hahn (i870). Hahn at least spoke Nama (a Khoi language) fluently; the same cannot be said for Passarge. While Passarge may have spent two years travelling in the Kalahari, his actual research amounted to a few months at the most. Moreover, his "fieldwork" was conducted in the wake of a rinderpest epidemic that had created exceptional conditions. His major interpreter was a Dutch-speaking Bushman, which increased the possibility of misinterpretation significantly. Most of his information was hearsay derived from white traders or Bechuanas (who had their own political agendas), since he found it difficult to get information directly from Bushmen. Far from being in- fluenced by Marx, Passarge was a radical conservative. His I897 Kalahari research was financed by Rhodes's British South Africa Company and led directly to the annexation of the Ghanzi Block, an area in which Cape Europeans were settled while the original (largely Bush- man) residents of the area were dispossessed. Lord Lugard, who was on the same expedition as Passarge, complained strongly about Passarge's ill-treatment of in- digenes (Perham i95 6:6oo). Passarge believed that Bush- men were on a closed development path, being incapable of adapting to agriculture or pastoralism (I907:I32). He concluded that the only policy in a settlement situation was to exterminate them: "What can the civilized hu- man being manage to do with people who stand at the level of that sheep stealer? Jail and the correctional house would be a reward, and besides do not even exist in that country. Does any possibility exist other than shooting them?" (p. I24, my translation).' Passarge, in short, was simply regurgitating the settler ideology au- thorizing expansionism. In contrast, Fritsch and espe- cially his associates, such as von Luschan, were estab- lished liberal, indeed anticolonial, Berlin professors with a well-known antiracist record (see, e.g., Hirschfeld I973). "Revisionists" are not necessarily or even usually reactionaries, and indeed, as Sontag (i98i) reminds us in her discussion of Riefenstahl's Last of the Nuba (the ideological cousin of much San ethnography), it is ro- mantics engaged in recording diverse "traditional" cul- tures before the onslaught of the world system who may be inclined to be so.

MATHIAS GUENTHER Department of Sociology and Anthropology, Wilfrid Laurier University, Waterloo, Ont., Canada N2L 3C5. IO VI 90

"No one of us ought to issue vetoes to the other, nor should we bandy words of abuse. We ought, on the con- trary, delicately and profoundly to respect one another's mental freedom: then only shall we bring about the in- tellectual republic....." Hassan (i986:520) commends these wise words of William James to the proponents of postmodernism in a key article that, among other things, laments the tendency of postmodernism to erode in its advocates "civil commitment, tolerant beliefs, critical sympathies." It seems appropriate to recall James's admonition in the context of the article before us, which is written very much along postmodernist lines. I wonder whether the negativism, indeed nihilism, of that discourse-its indeterminacy, fragmentation, de- canonization, self-lessness and depth-lessness, un(re)pre- sentability and irony (Hassan i986:504-6)-have not affected Wilmsen and Denbow's style. The debate between the "Kalahari revisionists" and their critics has become ever more heated, the Great Debate of hunter- gatherer studies. In some ways this has contributed to the theoretical advance of the field, challenging some of its fundamental assumptions and theories and forcing on it a process of self-correction. However, the tone of the debate has become much too strident for my taste; it is full of polemic sound and fury, posing such threats to scholarship as overstatement of the case, selective pre- sentation or misreading of the data, mistranslation of non-English sources, a straw-man approach to the oppos- ing view, and bathwater-and-baby-tossing. I will not, for lack of space, develop all of these points here; instead I will focus on a single issue of methodology that reveals some of these problems of scholarship-the treatment of the work of Siegfried Passarge.

Wilmsen and Denbow praise Passarge both for the quality and quantity of the empirical data he presents and for the theoretical analysis he undertakes of these data, declaring that had this work "been more widely read, modern ethnographies of San-speaking peoples would not . . . purport to be describing a way of life 'protected by the vast distances of the Kalahari.' "They hail Passarge as both the first scholar to have written an ethnographic account of the Bushmen and the first "re- visionist," a voice in the wilderness of romantic evolu- tionists depicting the Bushmen as part of a wider ag- ropastoral Tswana society and polity and as a people with herds and external trade. This appraisal of Passarge is based on misapprehension of both the empirical and the conceptual component of his work.

Passarge's ethnographic data are sketchy, often anec- dotal and ad hoc, conjectural, and secondhand. His two years of fieldwork in the Kalahari were devoted primar- ily to its geomorphology and geology (Passarge I904). The ethnographic studies of this geographer were on the whole incidental and preliminary, especially in the

i. Was soll der Kulturmensch mit Leuten anfangen, die auf dem Standpunkt jenes Schafdiebes stehen! Gefangnis und Zuchthaus waren Belohnung, existieren ausserdem in jenem Lande gar nicht. Bleibt da etwas anderes ubrig als Erschiessen?

5IO | CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

Dobe-NyaeNyae region (see Passarge I904:400-4i6). To flesh out his sometimes meagre information he drew on published works on the Bushmen by writers such as Werner, Schinz, Gentz, and Hahn. While in the field he availed himself as much of non-Bushman informants such as Batawana and white traders as of Bushman ones; a Dutch-speaking Nharo man named Koschep was his key informant and the source of virtually all of his San data.

Further, the vast bulk of his ethnography on both con- temporary and pre-contact Bushmen is not about the Zhu of the Kaukauveld-the Dobe-NyaeNyae area-but about the Nharo (or //Aikwe, as Passarge calls them) of the Ghanziveld and, to a lesser degree, the !Kung- speaking #Xau//ei (or #Aukwe). The illustrations in his book refer primarily to the Nharo, and the vernacular terms provided are from Nharo. While he does make the occasional reference to the "Ssu2gnassi" (Zhu) of the Kaukauveld (for example, on pp. II4 and ii 6), these ref- erences tend to be based not so much on observation as on extrapolation. That the more accessible Ghanziveld had been more thoroughly penetrated by Bantu-speaking and Orlam (Hottentot) pastoralists and European travel- lers, traders, and settlers from both the east and the west is well known and undisputed. (In this connection, there is in fact no indication that the German trader Franz Muiller was ever in the Dobe area; Passarge [I907:7-8, I0, I23; I904:map supplement, pl. i] clearly places both Muller and Kubi in the Chansefeld.) What Passarge's work shows, then, is that the Ghanzi Bushmen were part of a wider non-Bushman trade and political sys- tem-something that is neither new nor controversial, having been recognized in many of the works written by anthropologists about this area (Silberbauer I965, I98I; Silberbauer and Kuper I966; Russell and Russell I975; Guenther i986). Whether or not such was also the case with respect to the Kaukauveld is an open question.

Apart from the empirical unsuitability of Passarge's work for the task of presenting "San [i.e., Zhu] and Bantu people embedded in a single social formation," Pas- sarge's conceptualization of Bushman social organiza- tion and "external relations" can hardly be said to fit any revisionist programme. For Passarge the Bushmen, now as in the past, are "a hunting people par excellence (I907:ii9); hunting is bred in the bone, a phylogenetic, racial trait ("since time immemorial the Bushmen have been a race bred exclusively for the hunt") (p. i28).1 Hunting is the basis upon which are "built all of their social and economic relations, all of their laws and rights, their entire political organization" (p. j g).2 The fanciful and gratuitous way in which Passarge recon- structs the "Buschmann Reich" reveals some of the seri- ous anthropological deficiencies of the work of this

(Prussian) non-anthropologist (noted already by Schott [I955:I34]). This Reich consisted of large tribes with paramount chiefs (Oberhauptling) and chiefs (Haupt- ling) ruling over warrior/hunters and maintaining the integrity of their land and possessions by means of well- defined concepts of land tenure and property rights, as well as concepts of territoriality and defence against trespass. Passarge also deals with the psychology of hunters: an irrepressible yearning for freedom is its key element, such that any attempt at sedentarization and agropastoralism, let alone enserfment, would be doomed to failure; Bushmen would simply run off into the veld and the "Freiheit des Lebens der Steppe" or die (pp. 3, I24-31). Such a romantic-pessimistic view of the non- foraging and non-free-roaming Bushman and his society as doomed is of course squarely at odds with the re- visionist view of Bushmen as having for centuries or millennia been the economically marginalized, enserfed, and underclassed proletariat of their hegemonic food- producing neighbours.

Given the flaws of Passarge's data and their inapplica- bility to Wilmsen and Denbow's target area, as well as the lack of conceptual consonance between their ac- count and his, one wonders why they invest so much effort in reviving and reinstating the work of this distin- guished Kalahari geographer and not so distinguished Bushman anthropologist.

A final point of criticism has to do with mistransla- tion: in a conflation of two passages separated in the text by two pages, Wilmsen and Denbow offer as a transla- tion of "der Oberhauptling ... hielt sich seinerzeit an einer Vley Tsannie auf" (Passarge I907:II6) "the head- man . . . kept a valley called Tsanni for himself." The correct translation should be something like "the paramount chief . . . stayed in a valley called Tsanni." Translating the phrase the way they do allows them to fit this ethnographic fragment into their theoretical agenda as yet another bit of ethnohistorical evidence for the institutionalization of Zhu leadership and n!ore ownership.

RICHARD B. LEE

Department of Anthropology, University of Toronto, Toronto, Ont., Canada M5S iAi. 28 vi 9

By the end of this paper there is hardly a crime in the annals of science or politics to which the Harvard Kalahari Group is not a party; even the dispossession of the San of their land by their neighbors is laid at our doorstep (n. 39). But surely if we were as incompetent and mendacious as Wilmsen and Denbow claim, we would have been thrown out on our ears long ago. The fact that our ethnographic work and the models of hunter-gatherers derived from it continue to be critically engaged suggests that the paradigm we espouse is not quite ready for the scrap heap of history. At the recent Conference on Hunting and Gathering Societies, held in Fairbanks, Alaska, May-June I990, about 8o% of the

I. "Die Buschmanner sind eine seit Urbeginn ausschliesslich fur die Jagd geziichtete Rasse." 2. "Auf der Jagd bauten sich alle sozialen und wirtschaftlichen Verhaltnisse auf, alle Gesetze und Rechte, die ganze politische Or- ganisation."

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples | 5II

I33 papers presented situated themselves explicitly or tacitly within the broad ecological, social, and historical agendas of Man the Hunter, Woman the Gatherer, and Politics and History in Band Societies. Of perhaps greater significance is the fact that a number of native organizations endorsed the conference and supported it financially, unaware apparently that "hunter-gatherer" is a non-category, that they have been unwitting dupes of naive ecologism, and that their forebears have been proletarians (or entrepreneurs) all along.

There are many theoretical issues that could be dis- cussed at this point, but a prior and much more compel- ling issue is the quality of the evidence. By the vehe- mence of their attack Wilmsen and Denbow convey the impression that their critique has a solid empirical foun- dation; the words "irrefutable facts" are used more than once. But the empirical foundation is extremely weak, based on outright error and compounded by faulty infer- ence and misinterpretation of the historical record. It is hard if not impossible to take the theoretical issues seri- ously when even the most elementary points of fact are garbled, distorted, or just plain wrong-deficiencies that no amount of post-modern rhetoric can conceal. A few examples drawn from a paper now in preparation (Guen- ther and Lee n.d.) will illustrate this point.

In a summary paragraph intended to drive home the intimate and early involvement of the NyaeNyae-Dobe area in the world of merchant capital, Wilmsen and Denbow write:

This influx of Europeans, trading in ivory, ostrich feathers, and cattle, brought goods of many kinds and the force of arms into the region and trans- formed its economy and society once again. Van Zyl (Lee I979:78; Solway and Lee i990:ii6) was not the only European in the Dobe area in this period; An- dersson found oxen and calabashes in a "Bushman werft" at Mokoro in I859; Robert Lewis ... left his name as Lewisfontein at what is now Qangwa in I863; Axel Eriksson shot elephants with Lewis at CaeCae at this time; the Boer Trekkers spent about i8 months at Debra in I975-76; ... The traders turned ancient footpaths into oxwagon spoors some of which became well-worn roads-it took Schinz (i89i:35) just 36 hours to travel from Karakobis to Lewisfontein in I885.

Every statement in this paragraph is incorrect. i. Despite Wilmsen's efforts to place him in the Dobe

area (i989a:I12-I4), there is no evidence that Charles John Andersson ever reached it and a large body of evi- dence placing him elsewhere in February I859. In his own account of his trek to the Okavango, Andersson in mid-January is moving north, away from the Dobe area, on the dates Wilmsen claims he was moving towards it (Andersson i86i:I87). By February I2, Andersson is in the Omuramba Ovambo, 300 km west of the NyaeNyae- Dobe area (pp. 2oo-2o5). He continues to work his way north until he reaches the river March i 8-2o (i 86 I:2 2 9-

2i). These routes of travel are confirmed by both igth-

century and modern sources on Andersson (e.g., Neueste deutsche Forschungen I867; Tabler I973:5; Wallis I936). Therefore the "Makgoro" referred to by Wilmsen cannot be the "Mokoro" of the Dobe area, and the asso- ciation of Andersson with the Dobe area is spurious.

2. Neither here nor in his original mention (Ig8ga: i20) of Robert Lewis's trading for cattle in the Dobe area in I863 and having given his name to Lewisfontein (Qangwa) does Wilmsen cite a reference. Tabler (I973: 68), however, records that Lewis "traded and hunted elephants in the Kaokaoveld and Ovamboland from mid- I863 till perhaps I867." Since these areas are 500-800 km west of Qangwa, it would have been physically im- possible for Lewis to have been in Qangwa in i863. Fur- thermore, there is no evidence to link him with Lewis- fontein, which, according to Landell-Mills (Peters I972), may have been named after one of the Dorsland trekkers who arrived in the Dobe area around i876 at the earliest (see also De Kock I948).

3. Axel Eriksson did not arrive in the territory until i866 and spent the next five years hunting and trading in Ovamboland and the Kunene (Tabler I973:37), 500- 8oo km north-west of CaeCae (see also Winquist I978: 69-7 ).

4. The Boer Trekkers, also known as the Dorsland or Thirstland Trekkers, could not have spent I8 months at Debra in I875-76, since they only reached Ghanzi from the Transvaal in January I876 (Tabler I973:32). (They did spend the two years I876-77 at Reitfontein, 250 km south of the Dobe area, and did explore NyaeNyae dur- ing that period. In I878 they were found at Leeupan, 125 km north of the NyaeNyae-Dobe area.) Far from being powerful traders, the Dorsland Trekkers were refugees from the Transvaal looking for a place to settle and in I879-80 were the object of an international relief effort (Tabler I973:3I-34). After being driven out of the Grootfontein area in I887, they finally settled in Hum- pata, Angola.

5. Both here and in Wilmsen's book (Ig8ga:I2o) much is made of Schinz's statement that it took him only 36 hours to go from Karakobis to Lewisfontein, evidently to illustrate how well-travelled were the routes to and through the NyaeNyae-Dobe. In validating this claim, however, we encounter some major problems. The dis- tance between the two places measured on modern maps is I80-200 km depending on the route. According to the historical geographer Serton (McKiernan I954:i8-21), the average daily distance covered by igth-century ox- wagon was i9-26 km per i2-hour day, with a maximum of 32 km. Even if Schinz had gone non-stop he could not have covered more than 96 km, leaving him well over 80 km short of his destination. An examination of Schinz's highly inaccurate map (I8gI:appendix) shows that he has placed both "Karakobis" and "Lewisfontein" at least ioo km away from their actual positions and much closer to one another than they are. The text offers no clue to resolve the confusion, but without further clarification Schinz's statement about speed of travel cannot be invoked as evidence of anything except his

5I21 CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

unreliability as a reporter. There are good grounds for arguing that he never even reached Lewisfontein or Karakobis. Passarge's I904 map of the NyaeNyae-Dobe area (pl. i) shows Schinz's route passing through Gam on the southern margin of the area, 65 km south of Qangwa-Lewisfontein.

One could of course argue that I am nitpicking: What if a few Europeans weren't where they were supposed to be if the overall picture of mercantile penetration and devastation is accurate? This picture is the cornerstone of Wilmsen's (i989a) account of foragers as an impover- ished rural underclass the roots of whose dispossession go back over i,ooo years. Close examination of his sources reveals that the evidence for this position is any- thing but irrefutable. By conflating accounts of trade from all over the Kalahari he makes these accounts ap- pear by implication to apply to the Dobe area when the historical record reveals nothing of the sort. As Solway and I have noted, no one is denying for a moment that some parts of the Kalahari were the sites of intense and sustained mercantile activity; but this is not news, hav- ing been noted 25 years ago (Lee i965). The question is: was the Dobe-NyaeNyae area such a site? The historical record, including the data assembled by Wilmsen him- self, answers emphatically in the negative. Wilmsen's ( I989a: figs. 3.4, 3.7, 3.9, 3. II) overlays of place names or putative trade routes are applied to maps on which the original historical documents show a virtual blank. Readers are encouraged to consult the maps and accom- panying information in the sources themselves to evalu- ate the level of geographical knowledge and commercial activity in this part of the interior during the period in question (e.g., Andersson I856, I86I; Baines I864; Chapman I97I; Galton I853; McKiernan I954; Reise I859; Neueste deutsche Forschungen I867; Herero-Land I878; Vedder I938). Whereas Wilmsen (I989a: fig. 3.5) depicts the NyaeNyae-Dobe area as the centre of a net- work of trade routes in the mid-Igth century, he pres- ents not a single document for any of these routes, nor could we find any. Passarge himself does not help the case when he writes (I 907: i I8, my translation),' "There do not seem to be any trade routes through Bushman land, rather the trade went on between two close tribes." And Galton, writing in I853, at the beginning of the colonial trade era in northern Namibia, notes of the Her- ero peoples immediately to the west of NyaeNyae-Dobe that apart from the north-south axis "the Damaras have no communication whatever with any other country, a broad land dividing them from the natives to the east, and the sandy tract by the seashore bounding them to the west" (I97I [i8531: I99). The NyaeNyae-Dobe area in the period I850-I900

was far from the entrepot that Wilmsen and Denbow portray. In fact, with the exception of a few years in the I870s, the sources indicate that it was rather a backwa-

ter, bypassed by most of the major trade routes. This is the picture strongly conveyed by Schinz (e.g., I 89 I:3 5 7- 62) and Passarge (e.g., I 904:407- I 5; I 907: I I 9, I 3 I -3 3). The !Kung San traded when they could, defended them- selves when they had to, and during the long intervals between trading expeditions went about their business. These early fragmentary accounts offer glimpses of a rel- atively resilient and opportunistic hunting and gathering adaptation not in isolation from the wider world but in contact with it. In broad outline this is the picture that a dozen pre-Wilmsen ethnographers documented in the I950S, '6os, and '70s.

Even in the absence of evidence to support the case that the NyaeNyae-Dobe area was the centre of mercan- tile activity one could fall back on the weaker but still potent argument that the mercantile activity elsewhere in the region was sufficient to transform it. Here we return to the "Coke Bottle in the Kalahari" syndrome (Solway and Lee I990:IO9-Io): any level of mercantile activity, however small, has power and reach sufficient to penetrate all the social formations to the point where they are "sucked dry of commoditizable wild animals" and "selectively destroyed," but surely it is of interest whether a trading expedition arrives once a week, once a month, once a year, or once a decade? And whether the value of goods taken annually is $ioo,ooo, $IO,OOO, or $ioo? Or is the mere presence of any mercantile activity sufficient to transform hunter-gatherers into serfs, wage slaves, or members of a rural underclass? Is the question of historical and cultural specificity no longer of any relevance in the age of post-modern political economy? These are the broader questions raised by the paper under review.

Finally, I wonder what the connection is between the basic weakness of the historical case and this paper's extraordinary rhetoric.

ROBERT ROSS

Faculteit der Letteren, Rijksuniversiteit te Leiden, Postbus 95I5, 2300 RA Leiden, The Netherlands. I4 vI 90

Wilmsen and Denbow describe the debate between Fritsch and Passarge as the first great Bushman debate. This, through no fault of theirs, is not accurate. Some 70 years earlier, Dr. John Philip, the superintendent of the London Missionary Society in southern Africa, and Donald Moodie were the main protagonists in a consid- erably more vehement argument on the South African past which covered a wide range of issues-and indeed led to a clutch of libel suits-but which had as one of its main foci the origin and status of the Bushmen within what was then the Cape Colony. It was, I would con- tend, this controversy that set the agenda for most sub- sequent discussions of pre-industrial South African his- tory (Ross n.d.).

Philip (I828) argued in essence that the Bushmen against whom the Dutch colonists waged a genocidal

i. "Handelstrassen scheinen durch das Gebiet der Buschmanner nicht bestanden zu haben, vielmehr erfolgte der Austauch von Stamm zu Stamm."

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples I 5I3

war in the i8th and early igth centuries were people forced to take up a predatory way of life because their previous pastoralist existence had been destroyed by co- lonial conquest. Moodie (I84I), in contrast, adduced a variety of quotations from the earliest records of the Cape Colony to show that "'considerable numbers of these people [the Bushmen] existed in their present state previous to the occupation of the country by Europeans' and that they were, as they still are, the scourge of every people possessing cattle." It was an argument with di- rect political consequences. Philip was attempting to show that the actions of Europeans towards the autoch- thonous peoples of southern Africa had been and still were essentially unjust. If this was the case, then a new, more humanitarian course of action was required. In- deed, he managed to persuade the influential British Par- liamentary Committee on Aborigines to recommend such a change of policy. Moodie, on the other hand, was employed by the Cape government to find evidence which would rebut such views and justify the continua- tion of the status quo ante. Perhaps for the first time (at least among the colonists) but certainly not for the last, South Africa's past was contentious precisely because it could be mined for arguments about the country's pres- ent and future.

With this in mind, it is instructive to inquire into the hidden agenda behind the Fritsch-Passarge exchange. It would be naive to believe that Germans writing in I 907 on the Omaheke (as the Herero called the desert into which they had been driven by German colonialists a year earlier) would have been motivated exclusively by disinterested scientific concern. I must admit that I can- not see whether there are deeper ideological disagree- ments to be found in the exchange or whether it was merely impelled by Fritsch's pique at having been ig- nored by a man whom he considered his client. It may well be that I have failed to see through the code in which they wrote, or it may be that on fundamentals they agreed. What is certain, however, is that Passarge was writing within the discourse of German colonial- ism. To me, at least, he seems to be foreshadowing the "Thank God the Germans came" line of argument used notably by Heinrich Vedder (Lau I98I), arguing that colonization saved the Bushmen from destruction- though he does admit that it had been the introduction of firearms which had first caused the problem and won- der whether Afrikaner settlement near Ghanzi would not exacerbate the problem (I 907: I I 9, I 23).

Passarge, then, was certainly not a naive observer. Ex- actly how his political and personal concerns may have influenced his ethnography is, however, not something that Wilmsen and Denbow make clear. I take it that the promised translation of Passarge's works concerning the Kalahari will address this issue and provide more infor- mation about a fascinating, if somewhat repellent, man (see his comments on colonial policy in South Africa, with their great dislike for any form of humanitarianism [I908:34I], or his justification for the maintenance of slavery in Cameroon after the German colonial conquest [i895:526-27]). It may also explain why he chose to

adopt a tactic which is most unusual for an ethnog- rapher, namely, describing the area of his fieldwork as atypical. To Passarge the Kalahari was "a region created for hunting peoples" except (at least according to his map) for the small area to the west of the Okavango swamps where he, and later so many others, did fieldwork. In his map of the economic potential of southern Africa he designated precisely that area, along with the environs of Lake Ngami, as suitable for pas- toralists (Passarge I908:I70). Wilmsen and Denbow have not provided such basic source criticism on Pas- sarge, merely commenting on the length of time that he spent in the Kalahari. This would no longer be seen as a sufficient justification for the work of any ethnographer. The fact that Passarge supports their position does not absolve Wilmsen and Denbow from a responsibility to- wards his work which they assume with some relish with regard to that of their opponents.

In the absence of such cross-examination of their key witness, I am not convinced by Wilmsen and Denbow, except that I have no reason to doubt that the inferences they have drawn from the archeological data are rela- tively solid. Nevertheless, I think that they are proba- bly right in their basic contentions. These contentions derive, however, in the first instance from their exami- nation of written texts produced in the early years of colonial penetration, as indeed is the case for virtually all discussions of foragers in southern Africa from John Philip on (e.g., Elphick I977, Schrire I980, Parkington I984). The logics of analogy and back projection, jus- tified or otherwise, which run from these into the more distant past are complex and are made even more so by the need to take into account the activities of the Dutch in the Cape and the arrival of firearms and European traders in the Kalahari, to take only the case at hand and that which I know best. Building the arguments which are necessary requires more than the piling up of favour- able quotations.

In principle, the same procedures can be used to build analogies from the ethnographic present (of Richard Lee, for instance) to the distant past as from early colonial times to the centuries which preceded them, though of course the results would necessarily be somewhat more uncertain. It is not necessary to claim that the Zhu were isolated in order to make statements about them which have some potential relevance to the study of southern Africa before the introduction of domestic stock more than 2,ooo years ago and of the rest of the world even longer ago. Indeed, as this exchange has shown, it is counterproductive to do so. What we surely need now is to move beyond the sharply polarized positions that have been taken and ask whether it is possible, given the Zhu (and other Bushmen) were not isolated groups of foragers when they were investigated but also given that they acquired a considerable proportion of their suste- nance from hunting and gathering, to draw any general inferences about this way of life from their study, and, if so, which? Such an investigation must necessarily take into account the modern, historical, and archeological evidence of interaction with pastoralists, agricultural-

5I4 CURRENT ANTHROPOLOGY Volume 31, Number 5, December I990

ists, and colonialists, if only to discount it. It is surely illegitimate to extrapolate directly from the present to the past, except in the ways in which historians and archeologists always do, and would be so even without the problems caused by the interaction with pastoralists and others. Rather, descriptions of the behaviour of men and women in some ethnographic present, whether Wilmsen's, Lee's, Passarge's, or Philip's, though valuable in themselves as evidence for the time at which they were created, can only be used heuristically for other periods, as potential generators of ideas which need to be tested in some way or other when applied outside the very immediate temporal and spatial context in which they were made.

JACQUELINE S. SOLWAY

New College, University of Toronto, Toronto, Ont., Canada M5S IAI. 2o vi 90

It is difficult to know how to respond to an article that is so negative in tone, destructive in intent, and sin- gleminded in its attack on a group of scholars, particu- larly Richard Lee. Wilmsen and Denbow claim to take the intellectual and moral high road and close by saying that they will continue to engage evidence in a dialectic of theory and practice while others, presumably Lee and myself, are concerned with personalities and positions. Yet their article reads like an inquisition. Does one re- spond in like tones? Does one "wince" at every laboured attempt to harness i 9th-century sources to their po- lemic, "cringe" at every insult hurled in Lee's direction, "behold" as they penetrate layers of ideology and inade- quate theory to see truth in the Kalahari where all before them have been tainted by "overtones of Tonnies, " mired in "postwar biological humanism," etc.? Are we to accept that they have cornered the market on "actual- ity" while others have simply written allegories about invented people?

It is disconcerting to see one's words come back as parody, quoted out of context and with all qualifiers re- moved. There are numerous examples, but one will suffice. In their closing paragraph Wilmsen and Denbow paraphrase Lee and me as saying that "providing histor- ical detail to their histories robs 'foragers' of those his- tories." This is indeed "obscurantist rubbish," but we never said it. The original passage reads: "imposing agra- rian discourse on hunter-gatherers . . . robs the latter of their history" (Solway and Lee I 990: I2.2). Their reading misses both its point and its spirit. We were talking about the theorizing of history and not about the addi- tion of more historical detail, which is something we welcome.

Wilmsen and Denbow chide us for not understanding "history as historicity, the process through which social formations realize their transformations." Our case- study periodization is not, however, simply a chronicle of events but a model in which the latter phases repre- sent systems of regional articulation. They also accuse

us of subscribing to a simplistic ecological determinism in which San dispossession is attributed to environmen- tal degradation. Looking to the Southern Kalahari case, San dependence on Bantu-owned water sources has been a critical factor, but, as we note, this has more to do with changing property and power relations than with the changing water table (Solway and Lee I990:II4, II9).

The use of igth-century sources by Wilmsen and Den- bow demands mention. In an age when "ethnographic authority" is anything but taken for granted (Clifford I983) and ethnographic style and method are themselves subjects for study, it is surprising that Wilmsen and Denbow grant so much ethnographic authority to Pas- sarge. Our firsthand observations about contemporary drought-relief feeding and borehole drilling in the Kalahari are dismissed as "quasi-quantitative" while Passarge's descriptions are presented as "considerably ahead of his time." His use of terms such as "clan, "band," and "headman" goes virtually unquestioned. Contemporary ethnographers would acknowledge these terms' histories of competing definitions and would situate their usage within an anthropological discourse. I also question the manner in which MacKenzie's words are generalized to the whole Kalahari. MacKenzie's ob- servations of the contestation between Tswana leaders over their peripheries were written from the powerful and wealthy trading capital of the Ngwato chiefdom. This was the mercantile crossroads of central-southern Africa (Parsons I977:II3), and the political and eco- nomic fortunes at stake were significantly greater here than in Ngamiland, by comparison a political back- water.

Wilmsen and Denbow's paper is ambitious in its at- tempt to draw broad conclusions from vast regions and periods of prehistory from archeological evidence. The arguments offered are intriguing but highly speculative and at times contradictory. For example, we learn that the whole Kalahari has been enmeshed in hegemonic relations for over i,ooo years, with European incursion representing continuity with a long line of power and profit seekers: "the dominant centers of the hardveld ... were extending their economic interests into the west- ern sandveld; . . . this was the beginning of hegemonic domination from the east that was consolidated by about A.D. IOOO." Yet at another point in the text we find that "Khoe and San peoples, far from being icons of aboriginality, developed and controlled the means of production and trade over large parts of the Kalahari in- terior in earlier centuries, only being subordinated as 'Bushmen' in the igth century." Aside from the obvious discrepancy between these statements one has to ques- tion whether the archeological evidence is sufficient to suggest that any group of people was able to consolidate hegemony over the whole Kalahari i,ooo years ago.

A more general concern is the underlying assumption about cultural difference. I completely agree with Wilm- sen and Denbow in their plea for the abandonment of a reified view of closed cultures in favour of an analysis of continuously shaped social formations. But in rejecting the idea of closed cultures how does one account for

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples 5I 15

cultural difference? I get the impression from Wilmsen and Denbow that cultural difference in the Kalahari emerges only in a win-or-lose situation. It appears that all Kalahari residents have been striving for the same ends and all would have lived in the same manner and used the same material objects if only they could. Diver- sity exists, according to their argument, because it is favoured by the economy. To argue for the existence of cultural difference in the Kalahari that is not simply the product of unequal access to resources is not to be an apologist for the obvious inequalities that exist in the Kalahari today and that have existed throughout parts of the Kalahari in former times. Nor is it to grant any in- evitability or "naturalness" to these differences. Wilm- sen and Denbow's analysis seems to deny peoples the capacity to coexist in a culturally differentiated regional setting where one people's hegemony is not the sa- lient social condition. In a related formulation Sacks (I976:565) has coined the term "state bias" to refer to "a hierarchical way of conceptualizing [which is] character- istic of state societies." Regarding women, this bias posits that sexual equality can only come in the form of androgyny. Whatever truth this has in state societies, there is no reason to assume a priori that the same condi- tions of inequality characterize pre-state societies. Surely people can develop and reproduce cultural differ- ence without its simply resulting from domination or subordination.

It is a shame that Kalahari studies have reached such polemical depths. For the sake of Southern African scholarship perhaps all involved would do well to heed the Kgalagadi proverb: "Bulls of the same kraal should fight with their horns down."

JIRO TANAKA

Center for African Area Studies, Kyoto University, Kyoto 6o6, Japan. io vi 90

It cannot be denied that ecological studies of the San, including Lee's and my own, have paid relatively little attention to the interpretation of the present-day life of the San in a historical context. This lack of historical concern is not, however, a defect unique to the ecolog- ical studies of the San. It is a problem inherent in mod- ern ethnography, which chooses small communities as study subjects and attempts to extract from them the principles of their social structures. Archeological and historical methods, which view human societies dia- chronically, can complement the ethnographic method. Indeed, over the past decade various revisionists have contributed much to the recognition of the importance of historical considerations in ethnography. Wilmsen and Denbow, however, place too much emphasis on their own ethnohistorical approach, while paying too lit- tle attention to the wealth of detailed data collected ethnographically. Their seemingly endless display of archeological interpretations, intended to ensure the "end of illusion," fails to provide readers outside their own field with a reliable basis for fair judgment. They

are against a "narrow ecologism"; we should be equally opposed to a "narrow archeologism."

Since behavioral patterns (including thoughts and so- cial structure) leave no fossils, it is worthwhile studying the ecology and society of extant nonhuman primates and man. Anthropologists have come to realize, through their fieldwork, that a great variety of cultural/social sets can be formed in more or less the same environ- ment. In this sense, Solway and Lee's inference that the socioeconomic relations between the Bantu and the San must have been diverse is reasonable. How can we con- clude from only fragmentary archeological evidence that a system in which the San were the slaves of the Bantu covered the whole Kalahari area? Is this not exactly the kind of uniformitarian conclusion that Wilmsen and Denbow object to?

One of the reasons that Wilmsen and Denbow place the San at the bottom of the Bantu's social system is the difference between sites in the age at which cattle were slaughtered, as determined from excavated remains. From the observation that at some sites the remains are mostly those of prime animals they conclude that an elite monopolized prime food resources. Many studies on pastoral peoples, however, including my own in East Africa, have revealed that most male calves are castrated and only a few (about i out of 30) saved for reproduction. It is common practice to slaughter these castrated males for meat when they reach their prime. Here the ethno- graphic data do not support Wilmsen and Denbow's in- terpretation based on archeological evidence alone.

I do not think that Lee's attempt to apply Marxism to the !Kung San society is successful. Nevertheless, I be- lieve that his ecological analyses and detailed descrip- tions of the !Kung's social behavior patterns are of great value. His work is still first-class ethnography.

The longer I have spent with the San, the more I have become convinced that their unique "egalitarianism" is not an illusion. Even if historical evidence confirms that close links have long existed with the outside world, it does not lessen the importance of describing this cul- tural uniqueness. The Pygmies have lost their languages, and their "clientship" with the Bantu is more evident than the San's. Nevertheless, many anthropologists re- gard the Pygmies' fine recognition of nature, unique cul- tural value system, behavioral codes, and so forth, as invaluable sources for the understanding of man. The same applies to the San.

The importance of studies of hunter-gatherers, as Sol- way and Lee point out, is the opportunity they offer to understand another way of living in our age of global destruction of nature. Moreover, we should not restrict ourselves to hunter-gatherers. Our team of researchers in African studies at Kyoto University has been studying peoples that live very close to nature, whether through hunting-gathering, pastoralism, or slash-and-burn ag- riculture. The aim of our studies is not to deprive them of their history, economy, or politics but to draw lessons from the close relationship between their livelihood and nature. Good fieldwork is possible only if the field- worker feels, in his heart, respect for the people he stud-

5I6 CURRENT ANTHROPOLOGY Volume 31, Number 5, December I990

ies. Which respects the San more deeply, the group that is content with calling them "the poor" or the one that has expressed appreciation of their cultural uniqueness? Even if there are some defects in his work, as Wilmsen and Denbow point out, as a fieldworker and an anthro- pologist I feel more sympathy with Lee.

JAN VANSINA

Department of Anthropology, University of Wisconsin, Madison, Wis. 53706, U.S.A. 28 v go

As is argued by Wilmsen and Denbow and demonstrated by Wilmsen (ig8ga), foragers of the Kalahari have been part of a wider society that included pastoralists and farmers for well over a millennium. The evidence is overwhelming, and when a truth so evident provokes further debate it is no wonder that some of us betray some irritation. This irritation leads to a few errors and overstatements here. The first sentence surely must be translated as "The Bushman is the unhappy (unlucky) child of the moment," and it does not present Wilmsen and Denbow's case very well. Statements about ceramic similarities at Divuyu, about Angolan sites, and about sites in the Congo are premature, and the iron pendants from Divuyu are not of the same age as those from Sanga. But one cannot capitalize on such blemishes to deny the obvious.

Why is this evident truth not accepted? Clearly be- cause to do so requires a major shift in paradigms and in the study of forager societies everywhere. In the end, data such as these undermine the basis of any compara- tive anthropology based on sociocultural evolutionary theory and force us to reconsider the premises on which a sound comparative anthropology must be built. To deny the evidence merely postpones the inevitable.

JOHN E. YELLEN

Anthropology Program, National Science Foundation, Washington, D.C. 20550, U.S.A. I4 VI 90

Although the Wilmsen-Denbow argument is weakened by two serious conceptual flaws-the failure to consider the essential aspects which characterize a foraging soci- ety and the assumption that such entities can exist only in isolation-I direct my comments to a more limited issue. From its inception in the early I96os the "Harvard Kalahari Group" focussed its attention on a series of northern Kalahari waterholes which are located west of the Okavango Swamps and which straddle the Bo- tswana-Namibia border. For convenience I term this area "the Dobe region." It includes only a small part of the entire range occupied by San-speaking peoples and is en- vironmentally distinct from both the eastern Botswana hardvelt and the more heavily vegetated mopane wood- land which rings the Okavango itself. (For the reader unfamiliar with the details of southern African geogra- phy a good map is essential.) Because the Wilmsen and Denbow article consists primarily of an attack on the

"Harvard" work, this area is key. They argue that in the ist millennium A.D. a "dominant" Early Iron Age soci- ety appeared along the western margin of the Okavango and incorporated the Dobe-region San in a subservient position as suppliers of local products into a hierarchical system. Since they postulate a prehistoric trade in items such as ivory, pigment, red dye, and aromatic woods and postulate that in return Early Iron Age peoples provided items such as ceramic vessels, it is valuable to analyze the archaeological information which alone can provide the foundation for such a conjecture.

In a i 989 American Anthropological Association sym- posium, James Ebert presented a critique of the Wilm- sen-Denbow Western Okavango archaeological research. He noted the soft sandy matrix in which these sites are located, stated that neither archaeologist had dealt ade- quately with the problem of admixture, and argued that conclusions drawn from their data must be treated with skepticism. To evaluate this assertion, I reviewed the published record on these Western Okavango sites and was surprised to learn that it is limited to assertions and two brief articles which deal only with fauna. None of the sites have been properly published, and therefore it is impossible to evaluate on an objective basis the conclu- sions that Wilmsen and Denbow draw. For none of these excavations are there site plans, stratigraphic sections or any discussion of stratigraphy, or numerical data for ceramic types, lithics, and other remains (with the ex- ception of fauna). The kinds of information that ar- chaeologists use to judge the technical competence of their colleagues are also lacking. Discussions of sam- pling and excavation strategies are not provided. One can conclude that Early Iron Age peoples were present along the western margin of the Okavango by the early Ist millennium A.D., but it is dangerous to say more than that. Wilmsen and Denbow provides no reliable data on site sizes or the amount of exotic or imported goods. The information presented does not support a conclusion that any of the Okavango sites served as nodes in a hierarchical system.

At the other end of the "Okavango-Dobe connection," Wilmsen as well as Brooks and myself excavated and published on the site of /Xai /Xai (CaeCae), which docu- ments human occupation in the Dobe region during the last ca. 3,500 years. Because of the inconsistencies in Wilmsen's statements, his data are difficult to use. Pot- sherds which as he first describes them appear to contain limestone or bone temper and conform to Hambukushu or Historic Lungwebongu ware later become charcoal- tempered wares of the Early Iron Age. Discussion with Richard Redding, the archaeozoologist who studied the /Xai /Xai faunal remains, reveals that the "cow" from this site cannot be confirmed. Redding states that teeth of domestic cattle and the African Cape buffalo, which also occurs in the region, are distinguishable solely on a size basis, and discriminations can only be confidently made with large samples. (The "cattle" remains at /Xai /Xai consist of a single mandible fragment.)

Surveys conducted in the Dobe region by a number of archaeologists as well as excavations conducted by

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples 1 5I7

Brooks and Yellen at /Xai /Xai and a similar site at Mahopa (Magopa) (Brooks and Yellen I988) present a consistent picture. First, despite intensive survey, no Early Iron Age sites have been located in the Dobe re- gion. Secondly, as is attested by small amounts of ceram- ics and iron, by the mid-ist millennium A.D. contact in some form between the Okavango and Dobe regions had been established. The crucial issue is the nature of this interaction. While Wilmsen and Denbow assume that it must have been hierarchical, many other possibilities exist. Because the lowest of the three stratigraphic units at both /Xai /Xai and Mahopa predate the Early Iron Age in southern Africa, one can analyze stone assemblages to establish a baseline and use this to determine the effect of Early Iron Age contact. This effect, as the Yellen and Brooks article demonstrates, is nonexistent, or at least archaeologically invisible: the stone tool kit and basic range of activities it reflects remain unchanged after the appearance of Early Iron Age sites to the east. (Wilmsen and Denbow seriously misstate my position, and the reader is referred back to this unfortunately obscure orig- inal source.) If the provision of dye woods and the like to an Okavango-based Early Iron Age system became a cen- tral focus of the Dobe-area San, one would expect to see some local impact in the form of either changed tool kit or major inputs of exotic materials. Neither, however, occurs. The traditional !Kung hxaro or exchange system described by Wiessner (I990) provides, I believe, an ex- planation more consistent with the archaeological evi- dence.

Reply

EDWIN N. WILMSEN AND JAMES M. DENBOW Boston, Mass., U.S.A. 2 VIII 90

We are pleased that most commentators, even the ma- jority of those opposed to our argument (e.g., Guenther), agree that the issues we raise are important and require a great deal of further discussion. These issues are among those in the center of current anthropological discourse and, as is becoming ever more clear, have critical impli- cations for the peoples that are the objects of this dis- course. We are gratified that our work has stimulated ethnographers of San-speaking peoples to look seriously at earlier writings; this development has been one of the main goals of our presentation.

Several commentators give us high marks for the skill and effectiveness of our presentation, and Binford cor- rectly characterizes our view of ourselves as construc- tive critics. It is therefore astonishing to be accused by more partisan colleagues of nihilism, stridency, threats to scholarship, and insult. We share Vansina's irritation with this delaying action, which diverts attention from the critique itself. Response of this kind is a measure of the centrality of the issues we raise. It should not be

necessary in scholarly discourse, however, to point out that understandings change, or ought to change, in the light of new data and evolved theory, and when this happens no one's integrity, intelligence, ability, or sin- cerity is called into question. We agree completely with Binford and Ross that understandings, theories, and data must be applied to larger domains than those from which they were derived and constantly tested in these contexts. Our work is directed to just that purpose and provides perspective and historical context to the pro- cesses of social, political, and economic change that have previously been lacking for the region. This refor- mulation is necessary in order to ask questions about foraging as an economic strategy rather than as a mental template and about ethnicity as an ongoing construct of social relations in a polyvalent political context rather than as the simple product of norms of behavior. A new foundation is essential to the further development of current anthropological issues; we approach its con- struction neither as "word warriors" nor as fence mend- ers but as architects.

Some commentators are disturbed that we have been unable to reproduce here all the arguments and data we have previously published; this criticism is perhaps an inevitable consequence of challenging long-familiar work. In response, we can only ask that interested read- ers examine a large enough sample of our work for the full force of our presentation to become clear. They will find in it far more data than Yellen acknowledges. In particular, we recognize the truth of Bicchieri's com- ment that many students of foraging societies (Bicchieri among them) do acknowledge the antiquity of a wide range of relations between foraging and other peoples; indeed, we are often inspired by the work of some of our Australian and Arctic colleagues. But such acknowledg- ment has come late to the Kalahari, and in this essay we address only that small part of the earth. Likewise, we appreciate the supporting contributions of the staunch ecologists cited by Binford; in the broader work from which our essay is drawn, we are acutely aware of the impact of ecological variables on historical trajectories and try to take them into account. The larger questions regarding labor relations, social articulation, and others listed by Binford are addressed in that broader work, par- ticularly in Wilmsen (ig8ga) and Wilmsen and Vossen (I990).

We do not argue that "for the 'simpler' dwellers of the Kalahari" contact is "synonymous with loss of cultural identity" (Bicchieri), nor do we "deny peoples the capac- ity to coexist in a culturally differentiated regional set- ting" (Solway). Rather, we argue that the ethnographic landscape observed by 2oth-century ethnographers was fashioned through the very process of differentiation, not through the preservation in some desert refuge of a hypothetical Pleistocene ethos. Far from denying coexis- tence, we emphasize that each group has defined itself to itself in relation to others, not in terms of some univer- sal hunter-gatherer value system or Heideggerian will to autonomy (Solway and Lee I990). The image of a mono- lithic and self-conscious group with innate character

5I8 CURRENT ANTHROPOLOGY Volume 31, Number 5, December I990

traits, such as a deep will to freedom (Fritsch, Passarge, Solway and Lee), whose existence can be fundamentally explained by ideational factors is a teleological historiog- raphy that suppresses the real, historical complexity of San-speaking peoples, whether taken individually or as a whole. This is really the crux of the matter. While we would agree with Tanaka that "studying the ecology and society of extant nonhuman primates and man" can be an important avenue to learning, we would not attempt to create a link between studies of San "cultural uniqueness" as hunters "close to nature" and past "be- havioral patterns (including thoughts and social struc- ture)."

Our engagement with the Passarge-Fritsch exchange extends well beyond this essay, as Ross anticipates, and is predicated on Santayana's observation that those who are unaware of the past are condemned to repeat it. We set their "great Bushmen debate" in quotation marks to emphasize the irony of the fact that our challenge to the man-the-hunter model as applied in the Kalahari is being met by Lee, Yellen, and others who share similar vested interests at the end of the 2oth century in almost exactly the same way as Fritsch responded to Passarge's chal- lenge at its beginning, and apparently for similar rea- sons. We welcome Ross's addition of another three- quarters of a century to this dreary "debate." By the same textual convention, we extended that irony to Lee's revived "Kalahari San debate," thereby setting both in their common ontogeny. Although we have challenged a scholastic orthodoxy, we did not initiate this current manifestation of a "debate" the basis of which we explicitly reject as spurious. We are con- vinced, along with Gordon, that this "debate" resur- faces-indeed, is always implicit-in Euro-American thinking about "primitive" peoples for reasons having little to do with those peoples themselves. We sketched our position on this matter in the essay, and the reasons are examined at length in Land Filled with Flies.

We neither praise nor hail Passarge, as Guenther con- tends; he certainly does not carry our banner (Gordon). Passarge may have been a fascinating and somewhat reprehensible man (Ross), and he did make statements concerning colonial race relations that are unsavory to many people today. But, as Ross points out, these things are not unambiguously apparent in his Bushman publi- cations; why this should be so is a matter of continuing investigation. It may be that his romantic view of the Bushmen led him to conceptualize them in different terms, a condition upon which Passarge holds no mo- nopoly. It is, at least, very clear that the sentence Gor- don quotes was made in a context of concern that Af- rikaners would finish off what he thought Bantu peoples had begun, the genocide of Bushmen; Ross notes this. In fact, we quote his feeling that the decline in Bushman numbers is an "unfavorable prospect." This is hardly regurgitating settler ideology authorizing expansion (Gordon). Whatever Passarge may have felt obliged to report to Rhodes would have been irrelevant in any case; the British administration had annulled the BSAC con- cession to Ghanzi two years previously, had made its own reconnaisance of the area, and had assembled its

own contingent of Boers, who were already trekking to- ward Ghanzi.

Perhaps we should not cast stones so selectively: Fritsch may have been a liberal anticolonialist, but he also was convinced that Bushmen were living represen- tatives of an Urrasse; Galton, whose reputation in bio- logical and social sciences is secure, was a founder of the Eugenics Society, which advocated selective breeding of "superior" humans; Livingstone was one of the most ardent, active, and some (e.g., Listowel I974) say un- pleasant colonialists of his time. How the personal con- victions and behavioral traits of these persons color their observations and reporting is indeed relevant, but the passages we quote from Passarge are not disputed by Guenther and Gordon; they are simply disparaged for having been written by a putatively unpleasant person.

Guenther tries hard to discredit Passarge's ethnog- raphy, but on peculiar grounds: Passarge has "some- times meagre information"; "he drew on published work"; and "he availed himself . . . of non-Bushman informants." We all suffer from the former condition, and any good ethnographer seeks information from as many sources as possible. The fact that Passarge's pri- mary informant spoke Dutch appears to disturb Guen- ther (and Gordon), as if the man did not also speak Nharo and probably at least Setswana as well-judging by the number of Tswana vernacular terms given, correctly, by Passarge (however, some of Passarge's !Kung terms are wrong). Guenther notes that the bulk of Passarge's eth- nographic description is on the Ghanzi Nharo; that is true, and when we quote from him we make sure that the place being described is identified. Passarge's recon- struction of the Bushman past is, to be sure, overblown, but we do not make use of it. At the same time, his description of the conditions he observed is confirmed by Lee's informants (both Tswana and Zhu). Beyond this, Lee (I979:396) tells us that "early in the century the Motswana Murubela, leader of the Kubu clan, prac- ticed a rough form of frontier justice in the Dobe area that is still remembered with awe. In a fight over adul- tery at !Kangwa, one !Kung man had killed another. Murubela was summoned, heard the evidence, and shot the offender on the spot." It would be difficult to imag- ine a firmer substantiation of Passarge's general depic- tion of "rough frontier justice" than this. The names of persons and places Passarge gives for the Dobe- NyaeNyae are confirmable, and he uses terms for leader- ship and land tenure correctly. None of this is either challenged or contradicted by Passarge's detractors. His romantic image of Bushmen was at odds with his obser- vations of their life; he solved his dilemma in a typically romantic, and no doubt colonialist, way: by predicting their ultimate demise. We, of course, reject his image and his solution (as well as Yellen's [i9861 similar pre- diction of inevitable, though not yet accomplished, de- mise through assimilation), but this does not, as Guen- ther supposes, bar us from using his observations. After all, our image of San peoples is very different from Lee's and Yellen's, yet we incorporate Yellen's data and in Land make extensive positive use of Lee's observations and a number of his interpretations.

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples 1 5 I 9

Translation has, as Guenther intimates, a subjective element. Wilmsen takes this into account by consulting dictionaries in use at the time Passarge and Fritsch wrote. For example, today the most common English gloss of aufhalten (infinitive of "hielt sich ... auf") when used with respect to locality is "to stay." But Cassell's New German-English/English-German Dictionary (i888, revised i906) gives inter alia "to sustain, keep open," and "sich den Feind aufhalten" ('to keep off the enemy'). It seems essential to try to capture these con- temporary nuances, especially as the passage to which Guenther objects (which occurs in a single sentence, not over two pages (see n. 9]) was written in the context of Passarge's notions about Bushman group land tenure. In this context, "kept for himself" seems a reasonable gloss and reflects the Zhu notion that the senior member of a local descent group was the titular "owner/possessor" of the group's land-a notion generally accepted by mod- ern writers on the matter.

We will not address Lee's special pleading or the false issues of identity and cultural values he raises. To the extent that San-speakers may be thought of as an ethnic "group," it is because of their historically created posi- tion in the intersection of processes of ethnogenesis and class formation taking place in Botswana and Namibia (see Wilmsen I989a, Denbow and Wilmsen I986, Vos- sen and Wilmsen I990, Wilmsen n.d.). Neither Zhu nor any other San peoples' identity or cultural values depend on their being foragers, nor do they lose either when they engage in economic activities other than foraging. As for the errors Lee attributes to Wilmsen's data, Andersson was indeed heading north (actually, northeast) searching for the Okavango River; he was lost and trying to find local guides when he made the side trip to 'Makgoro. The sketch map of his route (State Archives Service, Windhoek n.d. A83[I]:I24) is reproduced in Wilmsen (i989a:I13); Lee relies only on Andersson's published narrative extractions from his unpublished diaries. What Wilmsen says about Robert Lewis is that he was trading in the Kaukauveld (a much larger area of which Dobe is a small part) beginning in i863. Tabler (I973:68) records that Lewis, along with four other Europeans, was "stand- ing at Leeupan to trade with the Dorsland Trekkers" in July i878; this is squarely in Zhu country, even if Lee wants to draw his Dobe line tighter. Neither Lewis nor Eriksson is remembered in the area by his European name, but both men's Herero names, Karobbie (Tabler I973:68) and Karuapa Katiti (Tabler I973:39), are re- corded, and Wilmsen elicited the oral histories to which Lee refers by using them (see Wilmsen i989a). Lewis must have been the source of "Lewisfontein," although there is no definite proof. Landell-Mills, in an editorial parenthesis in Peters (I972), did attribute the name to the Dorslanders, who he said (confusing the two sets of Boers) had arrived at the end of the I gth century, but this is of no consequence; no one named Lewis is mentioned among the Dorsland Trekkers (i870s), and the Ghanzi Trekkers (i897-98), among whom there were several Lewis families, came too late. It is not clear why Lee disputes our brief mention of the Boer Dorslanders; he admits that they were in the area during the entire time

mentioned. His objection to the citation of Schinz is equally unclear. Schinz's map is sketchy, but not so much so as to place Lewisfontein at Gcam; Guenther is at pains to discredit the evidence that Lee now relies on. Even if Schinz mistook Gcam for Qangwa, that is still in Lee's Dobe area and would add another traveller on the road used by the Dorslanders, the Lewis party, McKier- nan's group (of which Wilkerson was one), and others. Interested readers will find these and many other trad- ers' activities extensively discussed in Land.

Yellen's comments have been made by him before and need not be answered in detail here. It would be hard to find a correlation between depth and date closer than o.86; indeed, Yellen and Brooks (i988) accept the associ- ation of specimens with depth and dating when it fits their assumptions. Wilmsen (i988) discusses at length why the disputed animal should be cow rather than buf- falo, pointing out that this specimen is comparatively unimportant relative to the greater amount of ceramic and metal material available, particularly for the last 400 years (most of this obtained by Yellen in the late I96os but not reported until i988).

We fully agree that sites must be published in site- report format. We have done this for the eastern Bo- tswana sites (Denbow I983, n.d.) and CaeCae (Wilmsen I978), and preparations for detailed publication of the Ngamiland sites are under way. A detailed description of ceramics from those sites would neither add to nor sub- tract from our present argument and in any case is inap- propriate in an article of this sort (readers are, however, referred to Hendrickson [i985] on Tsodilo and Delta ceramics). In this paper, we stress the coordinate differ- ential faunal, ceramic, and lithic frequencies to access the dynamics of social relations in the region. It is the evidence for intensification of wild meat production at- tested to by faunal remains paralleled by changes in pot- tery and stone-tool proportions that forms evidence for the development of linkages between Tsodilo and other sites during the last half of the Ist millennium A.D.

Yellen's reference to Ebert's remarks is grasping at straws. No one has demonstrated substantial movement in Botswana soils, certainly not Yellen or Ebert, who merely assume it. Denbow (I98I, i983) tested the as- sumption by fitting potsherds as a routine part of ceramic anaysis; in eastern Botswana, about 90% of all mated sherds came from the same or immediately adja- cent io-cm levels. At Tsodilo and Delta sites, features containing reconstructible vessels, burned house floors and structures with concentrations of carbonized grains, and human burials with associated grave goods have all been found intact despite differences in density and size of these items. Radiocarbon dates obtained from several laboratories confirm our assessments of internally con- sistent stratigraphic sequences and compare well with dates for related assemblages in other parts of southern Africa.

Yellen continues to chide Wilmsen for not having rec- ognized, in I978, the significance of charcoal temper in Early Iron Age pottery, a failing he has readily conceded (Wilmsen i982; i988:38); it was just in that year that Denbow, working alone in eastern Botswana, was begin-

520 | CURRENT ANTHROPOLOGY Volume 31, Number S, December I990

ning to unravel that significance. Wilmsen (I978) esti- mated an age of about 8oo years for the earliest of the 373 sherds he recovered from CaeCae; radiocarbon dates place them only 300-400 years earlier. This plus the fact that a significant proportion of Early Iron Age sherds contain other tempers (limestone and bone among them) along with charcoal is omitted by Yellen here.

Tanaka seems to have misunderstood our discussion of differential slaughter strategies; his East African data exactly match ours from Early Iron Age sites. Regarding Bicchieri's comment on East African kraals, shifting a kraal several times a season may result in a larger area covered in dung, but it will not lead to vitrified deposits I50 cm deep unless substantial numbers of animals are kept.

The only substantive quarrel our critics have with our argument pertains to the locus of Lee's and Yellen's fieldwork (where one of us, Wilmsen, has also worked continuously since I973). They concede the rest of the Kalahari to our demonstration of widespread social net- works throughout recent millennia; only that small corner artificially delimited as the Dobe area is to be spared. We do not portray this Dobe area as an entre- pot; we simply demonstrate that it, too, was part of these networks. This, in itself, is not startling. Wiessner (i986:iog) demonstrates that any account of Zhu in- teraction must consider an area within a 200-300-km radius of the locality of concern. This radius, when cen- tered on Dobe, encompasses the Okavango Delta, Ghanzi, and all the other Ngamiland-Namibia places mentioned in this discussion. Lee (I979:365) himself re- cords that "initiates and their families would often travel 70 to ioo kilometers to join chomas" (male initia- tion ceremonies where extensive trading took place). This would bring Dobe Zhu to their relatives at Gcam, Leeupan, and Nokaneng on the Delta.

Anyway, the whole exercise is pointless; all members of the three central landholding Zhu homesteads at CaeCae (one of the largest settled locations in Lee's Dobe area) and their close affines in a fourth (there are nine Zhu, six Herero, and one Tswana homesteads in all) trace some ancestors to late i gth-century Ghanzi or to the north as far as the Okavango River (Wilmsen i 989a: I99-2 i) . Yellen suggests that hxaro networks can account for all ceramics, metal, and exotics found in Dobe archaeological sites. So far as Zhu in the last cen- tury or so are concerned, Wilmsen (i989a:26i-63) has already made this point and explained why it should be so. To argue thus, Yellen must accept our premise of widespread social formations, thereby negating the idea of isolates. But hxaro cannot account for all exchange at all times. Aside from the fact that we are not entitled to project uncritically a 2oth-century description of ex- change relations over more than i,ooo years of history, Wiessner (I977:246) is emphatic that hxaro relations ex- ist only among primary and close collateral kin. Thus, either Dobe Zhu kinspeople were potters, metallurgists, and stock-keepers or they were engaged with peoples who were and exchanged with them on a basis other than hxaro. In either case, our position is vindicated.

The apparent hostility of some Kalahari ethnographers

to our work does not seem to be widely shared. We ap- preciate Tanaka's graceful accommodation of our argu- ments to his views. Neutral anthropologists and histo- rians understandably ask for more data but do so out of interest in what they have seen of our arguments; many of the data they seek are to be found in our cited publica- tions. We are confident that when the dust settles Sol- way's mistaken impression that we see the world in win-or-lose terms and the Kalahari as a generalized whole (cf. Gordon's opinion that we can no longer be accused of this position, which in fact we never held) will settle with it, and many of us will join in answering Vansina's call to reconsider the premises on which a sound comparative anthropology must be based.

References Cited ALONSO, A. 1988. The effects of truth: Re-presentations of the

past and the "imaging of community." Journal of Historical Sociology 1:33-57.

ANDERSSON, C. J. I856. Lake Ngami. London: Hurst and Black- ett. [RBL]

. i86i. The Okavango River: A narrative of travel, explora- tion, and adventure. New York: Harper [RBL]

BAINES, T. I864. Exploration in South- West Africa. London: Longmans. [RBL]

BIRMINGHAM, D. I983. "Society and economy before A.D. 1400," in History of Central Africa. Edited by D. Birmingham and P. Martin, pp. I-29. London: Longman.

BRO 0 KS, A. I989. Past subsistence and settlement patterns in the Dobe area: An archaeological perspective. Paper presented at the 88th annual meeting of the American Anthropological As- sociation, November.

BROOKS, A., AND J. YELLEN. 1979. Archaeological excavations at #gi: A preliminary report on the first two field seasons. Bo- tswana Notes and Records 9:21-30.

BROWN, T. 1979 (?I875). 3d edition. Setswana-English dictionary. Braamfontein: Pula Press.

C A M P B E L L, A. I 986. The use of wild plant foods and drought in Botswana. Journal of Arid Environments ii:8I-9I.

C HAP MAN, J. 197 I. Travels in the interior of South Africa: I 849- I863. Cape Town: Balkema. [RBL]

CLAMMER, j. I968. "Concepts and objects in economic an- thropology," in The new economic anthropology. Edited by J. Clammer, pp. i-2o. New York: St. Martin's.

CLARK, J. D. I968. Furtherpaleo-anthropological studies in northern Lunda. Publicac6es Culturas do Museu do Dundo 78.

CLIFFORD, J. I983. On ethnographic authority. Representations I:II8-46. [IJSS]

COMAROFF, J. I987. Of totemism and ethnicity: Consciousness, practice, and the signs of inequality. Ethnos 52:301-23.

COOKE, J. 1975. The paleoclimatic significance of caves and adja- cent landforms in western Ngamiland, Botswana. Geographical Journal 141:430-44.

. I983. "The evidence from northern Botswana of Late Quaternary climatic change," in Late Cainozoic paleoclimates of the Southern Hemisphere. Edited by J. Vogel, pp. 265-78. Rotterdam: Balkema.

DE KO CK, W. J. 1948. Robert Lewis of Damaraland. Africana Notes and News 6(I): I4-15. [RBL]

DEN BOW, J. I982. "The Toustwe tradition: A study in socio-eco- nomic change," in Settlement in Botswana. Edited by R. Hitch- cock and M. Smith, pp. 73-86. Johannesburg: Heinemann.

. I984a. "Prehistoric herders and foragers of the Kalahari: The evidence for i 500 years of interaction," in Past and present in hunter-gatherer studies. Edited by C. Schrire, pp. 175-94. Or- lando: Academic Press.

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples I 52I

. I984b. Report on an archaeological reconnaissance of the BP Soda Ash lease, Makgadikagi Pans, Botswana. MS, Govem- ment of Botswana, Gaborone, Botswana.

- I986. A new look at the later prehistory of the Kalahari. Journal of African History 27:3-28.

. 1990. Congo to Kalahari: Data and hypotheses about the political economy of the westem stream of the Early Iron Age. African Archaeological Review. In press.

DENBOW. J., A. MANIMA-MOUBOUHA, AND N. SANVITI. I988. Archaeological investigations along the Loango Coast, Congo. NSI: Bulletin de Liaison des Arch6ologues du Monde Bantu 3:37-42.

DENB OW, J., AND E. WILMSEN. I983. Iron Age pastoralist settle- ments in Botswana. South African Journal of Science 79:405-8.

1 I986. The advent and course of pastoralism in the Kalahari. Science 234:1509-15.

DUBOW, S. I987. "Race, civilization, and culture: The elaboration of segregationist discourse in the inter-war years," in The poli- tics of race, class, and nationalism in twentieth-century South Africa. Edited by S. Marks and S. Trapido, pp. 71-94. London: Longman.

ELPHICK, R. 1977. Kraal and castle: Khoikhoi and the founding of White South Africa. New Haven: Yale University Press. [RR]

ERVEDOSA, C. I980. Arqueologia angolana. Lisboa: Ediq&oes 70. F A B I A N, j. I986. Hindsight: Thoughts on anthropology upon

reading Francis Galton's Narrative of an explorer in tropical South Africa (i853). Critique of Anthropology 7:37-49.

FRITS CH, G. I863. Die Eingeborenen Siid-Afrikas. Breslau: Fer- dinand Hirt.

i868. Drei Jahre in Sud-Afrika. Breslau: Ferdinand Hirt. i88o. Die afrikanischen Buschmanner als Urrasse.

Zeitschrift fur Ethnologie 12:289-300. .I906. Die Buschmanner der Kalahari,by S. Passarge.

Zeitschrift ftir Ethnologie 38:7I-79. GALTON, F. I97I (i853). Narrative of an explorer in tropical

South Africa. New York: Johnson Reprint. [RBLJ GARLAKE, P. I973. Great Zimbabwe. London: Thames and Hud-

son. GORDON, R. I989. "Can Namibian San stop dispossession of their

land?" in We are here: Politics of aboriginal land tenure. Edited by E. Wilmsen, pp. I38-5 4. Berkeley, Los Angeles, and London: University of Califomia Press.

. n.d. The Bushman myth and the making of a Namibian underclass. Boulder: Westview Press. [RG]

GRIVETTI, L. I978. Nutritional stress in a semi-arid land: Exami- nation of Tswana agropastoralists of the eastem Kalahari, Bo- tswana. American Journal of Clinical Nutrition 3I:I204-20.

I979. Kalahari agro-pastoral hunter-gatherers: The Tswana example. Ecology of Food and Nutrition 7:235-56.

GUENTHER, M. I986. The Nharo Bushmen of Botswana: Tradi- tion and change. Hamburg: Helmut Buske Verlag. [MG]

GUENTHER, M., AND R. LEE. n.d. Some problems of historical interpretation in Kalahari ethnography. MS. [RBLJ

HAHN, THEOPHILUS. I870. Die Buschmanner. Globus i6:65-68, 8i-85, I02-5, I20-23, I40-43, I53-55. [RG]

HARPENDING, H., AND L. WANDSNIDER. I980. Population structures of Ghanzi and Ngamiland !Kung. MS.

HARAWAY, D. I988. "Remodelling the human way of life: Sher- wood Washbum and the new physical anthropology, I950- I980," in Bones, bodies, behavior: Essays on biological an- thropology. Edited by G. Stocking, pp. 2o6-59. Madison: University of Wisconsin Press.

HASSAN, I. I986. Pluralism in postmodem perspective. Critical Inquiry I2:503-20. [MG]

HAWKES, KRISTEN, KIM HILL, AND JAMES F. O'CONNELL. I982. Why hunters gather: Optimal foraging and the Ache of eastem Paraguay. American Ethnologist 9:379-98. [LRBI

HELGREN, D., AND A. BROOKS. I983. Geoarchaeology at #gi, a Middle and Later Stone Age site in the northwest Kalahari. Journal of Archaeological Science IO:I8 I-97.

HENDRICKSON, A. I986. Early Iron Age ceramics from north- westem Botswana: The evidence from Matlapaneng, N!oma, and Divuyu. M.A. thesis, New York University, New York, N.Y.

Herero-Land: Land und Leute. I878. Petermanns geographische Mitteilungen 8:306-II. [RBLI

HIATT, L. I989. "Aboriginal land tenure and contemporary claims in Australia," in We are here: Politics of aboriginal land tenure. Edited by E. Wilmsen, pp. 99-117. Berkeley, Los Angeles, and London: University of California Press.

HILL, KIM, AND KRISTEN HAWKES. I983. "Neotropical hunting and the Ache of eastern Paraguay," in Adaptive responses of native Amazonians. Edited by R. Hames and W. Vickers, pp. 139-88. New York: Academic Press. [LRB]

HIRSCHFELD, MAGNUS. 1973. Racism. Port Washington: Ken- nikat. [RG]

HITCHCO CK, R. I980. Tradition, social justice, and land reform in central Botswana. Journal of African Law 24:1-34.

. i982. "Patterns of sedentism among the Basarwa of east- ern Botswana," in Politics and history in band societies. Edited by E. Leacock and R. Lee, pp. 223-68. Cambridge: Cambridge University Press.

HITCHCOCK, R., AND J. EBERT. I984. "Foragingandfoodproduc- tion among Kalahari hunter/gatherers," in From hunters to farmers. Edited by J. D. Clark and S. Brandt, pp. 328-48. Berke- ley, Los Angeles, and London: University of California Press.

HOWELL, N. 1976. Demography of the Dobe !Kung. Orlando: Ac- ademic Press.

. I986. Images of the Tasaday and the !Kung: Reassessing isolated hunter-gatherers. MS

HUF F MAN, T. I982. Archaeology and ethnohistory of the African Iron Age. Annual Review of Anthropology 11: 133-50.

HURTADO, ANA MAGDALENA, KRISTEN HAWKES, KIM' HILL, AN D H I L L A R D KA P L AN. I 985. Female subsistence strategies among Ache hunter-gatherers of eastem Paraguay. Human Ecology I3:I-28. [LRBI

ISAAC, GLYNN L. I978a. The food-sharing behavior of protohu- man hominids. Scientific American 238(4):90-io6. [LRB]

. I978b. Food sharing and human evolution: Archaeological evidence from the Plio-Pleistocene of East Africa. Journal of Anthropological Research 34:3 I I-25. [LRB]

JOHNSON, A. I975. Time allocation in a Machiguenga commu- nity. Ethnology I4:30I-IO.

KARP, I. I986. Agency and social theory: A review of Anthony Giddens. American Ethnologist I3:I3I-37.

KENT, S. I990. Comment on: Foragers, genuine or spurious? Situating the Kalahari San in history, by J. Solway and R. Lee. CURRENT ANTHROPOLOGY 3 I: I 3 I-32.

KINAHAN, j. I986. Settlement patterns and regional exchange: Evidence from recent Iron Age sites on the Kavango River, northeastern Namibia. Cimbebasia 3:IO9-I6.

KINAHAN, J., AND J. KINAHAN. I984. An archaeological recon- naissance of Bushmanland and southem Kavango. Report to the South West Africa Department of Agriculture and Nature Con- servation, Windhoek.

KOHLER, 0. I975. "Der Khoisan Sprachbereich," in Die Volker Afrikas und ihre traditionellen Kulturen. Edited by H. Baumann, pp. 305-37. Wiesbaden: Fritz Steiner.

. "Allgemeine und sprachliche Bermerkungen zum Feldbau nach oral Texten der Khoe-Buschleute," in Afrikanische Wild- beuter: Internationales Symposion, Tagungberichte. Edited by F. Rottland and R. Vossen, pp. 205-73. Sankt Augustin.

KONNER, M., AND M. SHOSTAK. I986. "Ethnographic roman- ticism and the idea of human nature: Parallels between Samoa and the !Kung San," in The past and future of !Kung ethnog- raphy. Edited by M. Biesele, pp. 69-76. Hamburg: Helmut Buske.

LAI D L E R, P. I 938. South African native ceramics: Their charac- teristics and classification. Transactions of the Royal Society of South Africa 27:29-46.

LAU, B. I98I. "Thank God the Germans came: Vedder and Nami- bian historiography," in Collected seminar papers, vol. 2. Edited by C. Saunders. Cape Town: University of Cape Town. [RR]

LEAKEY, RICHARD. I98I. The making of mankind. New York: Dutton. [LRBJ

LEAKEY, RICHARD, AND ROGER LEWIN. I977. Origins. New York: Dutton. [LRB]

522 CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

. I 978. People of the lake: Mankind and its beginnings. Garden City: Anchor Press/Doubleday. [LRBJ

LEE, R. i965. Subsistence ecology of !Kung Bushmen. Ph.D. diss., University of Califomia, Berkeley, Calif. [RBL]

. i969. "!Kung Bushman subsistence: An input-output anal- ysis," in Environment and cultural behavior. Edited by A. Vayda, pp. 47-79. New York: Natural History Press.

. I972. !Kung spatial organization: An ecological and histor- ical perspective. Human Ecology I:I25-47.

9 1976. The !Kung's new culture. Science Year, pp. I80-95. 9 1979. The !Kung San: Men, women, and work in a forag-

ing society. Cambridge: Cambridge University Press. . i988. "Reflections on primitive communism," in Hunters

and gatherers, vol. I. Edited by T. Ingold, J. Riches, and J. Wood- burn, pp. 252-68. Oxford: Berg.

. I989. The !Kung in question: Evidence and context in the Kalahari San debate. Paper presented at the 88th annual meet- ing of the American Anthropological Association, Washington, D.C., November.

LEE, R., AND I. DE VORE. i968. "Preface," in Man the hunter. Edited by R. Lee and I. DeVore, pp. vii-ix. Chicago: Aldine.

LtVI-STRAUSS, C. i963. Totemism. Translated by R. Needham. Boston: Beacon Press.

. i967. The scope of anthropology. Translated by S. Ortner Paul and R. Paul. London: Jonathan Cape.

. i968. "The concept of primitiveness," in Man the hunter. Edited by R. Lee and I. DeVore, pp. 349-5 2. Chicago: Aldine.

LISTOWEL, J. I974. The other Livingstone. Cape Town: David Phillip.

LIVINGSTONE, D. i858. Missionary researches and travels in South Africa. London: John Murray.

MC CLINTOCK, ANNE. I988. Maidens, maps, and mines: The reinvention of patriarchy in colonial South Africa. South At- lantic Quarterly 87:I47-92. [RG]

MAC KENZIE, j. I883. Day dawn in dark places. London: Cassel. MC KIERNAN, G. I954. The narrative and journal of Gerald

McKiernan in South West Africa, I874-I879. Edited by P. Ser- ton. Cape Town: Van Riebeck Society. [RBL]

MADDO CK, K. I989. "Involved anthropologists," in We are here: Politics of aboriginal land tenure. Edited by E. Wilmsen, pp. I 5 5-76. Berkeley and Los Angeles: University of California Press.

MARKS, S., AND S. TRAPIDO. Editors. I987. The politics of race, class, and nationalism in twentieth-century South Africa. Lon- don: Longman.

MARSHALL, L. I976. The IKung of Nyae Nyae. Cambridge: Har- vard University Press.

MARTIN, P. I972. The external trade of the Loango coast, I576- I870. Oxford: Clarendon Press.

MARTINS, R. DE SOUSA. I976. A estaa&o arqueol6gica da antiga Banza Quibaxe. Contribui9des para o Estudo da Antropologia Portuguesa 9:244-306.

MATTHEWS, S. I980. On history. Chicago: University of Chicago Press.

MILLER, D., N. VAN DER MERWE, E. WILMSEN, AND J. DEN- BOW. n.d.a. Copper and iron artefacts from the Early Iron Age of Divuyu, Botswana. South African Archaeological Bulletin. In press.

. n.d.b. Copper and iron artefacts from the Early Iron Age of N!oma, Botswana. South African Archaeological Bulletin. In press.

MILLER, J. I976. Kings and kinsmen: Early Mbundu states in An- gola. Oxford: Clarendon Press.

. I983. "The paradoxes of impoverishment in the Atlantic zone," in History of Central Africa. Edited by D. Birmingham and P. Martin, pp. I I8-5 9. Cambridge: Cambridge University Press.

MOODIE, D. I84I. An inquiry into the justice and expediency of completing the publication of the authentic records of the col- ony of the Cape of Good Hope relative to the aboriginal tribes. Cape Town: A. S. Robertson. [RR]

MOSSOP, E. I935. The journal of Henrik Jacob Wikar (I779) and the journals of facobus Coetse Janz (1760) and Willem van Reenan (I79I). Cape Town: Van Riebeck Society.

. I947. The journals of Brink and Rhenius. Cape Town: Van Riebeck Society.

Neueste deutsche Forschungen in Sud-Afrika: Von Karl Mauch, Hugo Hahn und Richard Brenner, I 866 und I 867. I 867. Peter- manns geographische Mitteilungen 8:28i-98. [RBL]

0 KI HI R 0, G. I976. Hunters, herders, cultivators, and traders: In- teraction and change in the Kalagadi, nineteenth century. Ph.D. diss., University of California, Los Angeles, Calif.

PARKINGTON, j. I984. "Soaqua and Bushmen: Hunters and rob- bers," in Past and present in hunter-gatherer studies. Edited by C. Schrire, pp. I5 I-74. Orlando: Academic Press. [RR]

PARSONS, N. I977. "The economic history of Khama's country in Botswana, I844-I930," in The roots of rural poverty in central and southern Africa. Edited by R. Palmer and N. Parsons, pp. II 3-42. Berkeley: University of Califomia Press. [jss]

PASSARGE, S. I895. Adamaua: Bericht iiber die Expedition des Deutschen Kamerun-Komitees in den Jahren I893-94. Berlin: Dietrich Reimer. [RR]

I 904. Die Kalahari. Berlin: Dietrich Reimer. [MG] 9go5 a. Die Buschmanner der Kalahari. Mitteilungen von

Forschungsreisenden und Gelehrten aus dem Deutschen Schutzgebieten, pp. I94-287. Berlin: Mittler.

. Igosb. Die Grundlinien im ethnographischen Bilde der Kalahari-Region. Zeitschrift der Gesellschaft fur Erdkunde zu Berlin. Berlin: Mittler.

. I906. Response to Fritsch. Zeitschrift fur Ethnologie 38:4II-I4.

. I907. Die Buschmanner der Kalahari. Berlin: Dietrich Reimer.

. I908. Siidafrika: Eine Landes-, Volks- und Wirt- schaftskunde. Leipzig: Quelle and Meyer. [RR]

PERHAM, MARGERY. I956. Lugard: The years ot adventure i858-I898. London: Collins. [RG]

PETERS, M. I972. Notes on the place names of Ngamiland. Bo- tswana Notes and Records 4:2I9-33 [RBL]

PHILIP, j. i828. Researches in South Africa. London: James Dun- can. [RR]

PHILLIPSON, D., AND B. FAGEN. I969. The date of the Ingome Ilede burials. Journal of African History IO: I99-204.

PRATT, M. I985. "Scratches on the face of the country, or, What Mr. Barrow saw in the land of the Bushmen," in Race, writing, and difference. Edited by H. Gates, Jr., pp. II 9-43. Critical Inquiry 12.

. I986. "Fieldwork in common places," in Writing culture: The poetics and politics of ethnography. Edited by J. Clifford and G. Marcus, pp. 27-5o. Berkeley, Los Angeles, and London: University of California Press.

PRINSLOO, J., AND J. GAUCHE. I933. In die woeste Weste: Die Lydensgeskiedenis van die Dorslandtrekkers. Pretoria: J. H. de Bussy.

RAMSAY, j. I990. A history of the Bakwena dynasty of south-cen- tral Botswana, c. i820-i940. Ph.D. diss., Boston University, Boston, Mass.

Reise der Herm Hugo Hahn und Rath im sudwestlichen Afrika. I859. Petermanns geographische Mitteilungen 7:299 [RBL]

ROSS, R. n.d. "Donald Moodie and the origin of South African historiography," in Beyond the pale: Essays on the history of colonial South Africa. Middletown: Wesleyan University Press. [RR]

Rural Income Distribution Survey. I976. Gaborone: Cen- tral Statistics Office.

RUSSELL, M., AND M. RUSSELL. I975. Afrikaners of the Kalahari. Cambridge: Cambridge University Press. [MG]

RUST, F. I960. Deutsch-Nama Worterbuch. Windhoek: Rheinische Mission in Sudwestafrika. [MJc]

SACKS, K. I976. State bias and women's status. American An- thropologist 78:565-69. [JSS]

SAHLINS, M. "Notes on the original affluent society," in Man the hunter. Edited by R. Lee and I. DeVore, pp. 85-89. Chicago: Aldine.

SANDELOWSKY, B. I979. Kapako and Vungu Vungu: Iron Age sites on the Kavango River. South African Archaeological Soci- ety Good win Seri es 3: 52z-61I.

WILMSEN AND DENBOW Paradigmatic History of San-speaking Peoples | 523

SCHAPERA, I. i952. The ethnic composition of Tswana tribes. London School of Economics Monographs on Social Anthropol- ogy I I.

SC H IN Z, H. I89I. Deutsch-Siidwest-Afrika: Forschungsreisen durch die deutschen Schutzgebiete Gross-Nama- und Herero- land, nach dem Kunene, dem Ngami-See, und Kalahari, I884- I887. Oldenburg and Leipzig.

S C H O F I E L D, J. I 948. Primitive pottery. South African Ar- chaeological Society Handbook 3.

S CH OTT, R. I955. Die Buschmanner in Siidafrika: Eine Studie iuber Schwierigkeiten der Akkulturation. Sociologus 5: I 32-49. [MG]

SCHRIRE, C. I980. An inquiry into the evolutionary status and apparent identity of San hunter-gatherers. Human Ecology

8:9-32. [RR] SCUDDER, T. I97I. Gathering among African woodland savan-

-nah cultivators: A case study, the Gwemba Tonga. University of Zambia Institute of African Studies Zambian Papers 5.

SEINER, F. I977 (i9io). Die Buschmanner des Okavango und Sam- besigebietes der Nord-Kalahari. Translated by H. Vierich Esche. Botswana Notes and Records 9:3I-36. (First published in Globus 97.)

SERVICE, E. I96 2. Primitive social organization. New York: Ran- dom House.

S HAW, P. I985. Late Quaternary landforms and environmental change in northwest Botswana: The evidence of Lake Ngami and the Mababe Depression. Transactions of the Institute of British Geographers IO:333-46.

SHOSTAK, M. I981. Nisa: The life and words of a IKung woman. Cambridge: Harvard University Press.

SILBERBAUER, G. I965. Bushman survey report. Gaborone: Gov- ernment Printer. [MG]

. I98I. Hunter and habitat in the central Kalahari Desert. Cambridge: Cambridge University Press.

. I989. On the myth of the savage Other. CURRENT AN- THROPOLOGY 2o:2o6-7.

SILBERBAUER, G., AND A. KUPER. I966. Kgalagari masters and Bushman serfs. African Studies 25:I7I-79. [MG]

S M I T H, W OO D R U F F. I 9 8 7. "Anthropology and German colonial- ism," in Germans in the tropics. Edited by A. Knoll and L. Gann. Westport: Greenwood Press. [RG]

SOLWAY, J., AND R. LEE. I990. Foragers, genuine or spurious? Situating the Kalahari San in history. CURRENT ANTHROPOLOGY

3I:IO9-46. SONTAG, SUSAN. I98I. Under the sign of Saturn. New York:

Vintage Books. [RG] SPETH, J., AND R. HITCHCOCK. Editors. n.d. The Kalahari eth-

nography of Siegfried Passarge. MS. STEWARD, J. I938. Basin-plateau aboriginal socio-political

groups. Bureau of American Ethnology Bulletin i2o. TABLER, E. C. I973. Pioneers of South West Africa and Ngami-

land: I738-I880. Cape Town: Balkema. [RBL] TANNER, A. I984. Review of: Politics and history in band soci-

eties, edited by E. Leacock and R. Lee (Cambridge: Cambridge University Press, i982). American Ethnologist II:626-27.

TAYLOR, F., AND H. MOSS. I98I. Final report on thepotential for commercial utilization of veld products. Vol. 2. Gaborone: Veld Products Research.

TOBIAS, P. I975. "Anthropometry among disadvantaged peoples: Studies in southern Africa," in Biosocial interrelations in popu- lation adaptations. Edited by E. Watts, F. Johnston, and G. Las- ker, pp. 287-305. Paris: Mouton.

TONNIES, F. I957 (i887). Gemeinschaft und Gesellschaft. Trans- lated by C. Loomis and J. McKinney. New York: Harper and Row.

TRUSWELL, A., AND J. HANSEN. I968. Medical and nutritional studies of !Kung Bushmen in northwest Botswana: A prelimi- nary report. South African Medical Journal 42:I3 38.

. I976. "Medical research among the !Kung," in Kalahari hunter-gatherers. Edited by R. Lee and I. DeVore, pp. I66-94. Cambridge: Harvard University Press.

VAIL, L. I989. The creation of tribalism in southern Africa. Berkeley, Los Angeles, and London: University of California Press.

VAN N O T EN, F. I 98.2. The archaeology of Central Africa. Graz: Akademische.

VAN SINA, J. I985. "Do Pygmys have a history?" Proceedings of the International Symposium on African Hunters and Gather- ers, Sankt Augustin. Edited by F. Rottland and R. Vossen, pp. 43I-46. Sprache und Geschichte in Afrika 7(I).

VEDDER, H. I938. South-West Africa in early times. Oxford: Ox- ford University Press. [RBL]

WALKER, N. I983. The significance of an early date for pottery and sheep in Zimbabwe. South African Archaeological Bulletin 38:88-92.

WALLIS, J. P. R. I936. Fortune myfoe: The story of Charles John Andersson. London: Jonathan Cape. [RBL]

WEST P HAL, E. I963. The linguistic prehistory of southern Africa: Bush, Kwadi, Hottentot, and Bantu linguistic relationships. Africa 33:237-65.

WIESSNER, P. I977. Hxaro: A regional system for reducing risk in the !Kung Bushmen. Ph.D. diss., University of Michigan, Ann Arbor, Mich.

. I986. "!Kung San networks in a generational perspective," in The past and future of !Kung ethnography. Hamburg: Hel- mut Biske.

. I990. Comment on: Foragers, genuine or spurious? Situat- ing the Kalahari San in history, by Jacqueline S. Solway and Richard B. Lee. CURRENT ANTHROPOLOGY 3 I: I 35-36. [JEYJ

WILMSEN, E. I978. Prehistoric and historic antecedents of an Ngamiland community. Botswana Notes and Records IO:5-I8.

. I 982. "Exchange, interaction, and settlement in north- western Botswana: Past and present perspectives," in Settle- ment in Botswana. Edited by R. Hitchcock and M. Smith, pp. 98-IO9. Johannesburg: Heinemann.

. I983. The ecology of illusion: Anthropological foraging in the Kalahari. Reviews in Anthropology io:9-20.

. I988. The antecedents of contemporary pastoralism in western Ngamiland. Botswana Notes and Records 20:29-39.

. I989a. Land filled with flies: A political economy of the Kalahari. Chicago: University of Chicago Press.

. I989b. "Those who have each other: Land tenure of San- speaking peoples," in We are here: Politics of aboriginal land tenure. Edited by E. Wilmsen, pp. 43-67. Berkeley, Los Angeles, and London: University of California Press.

. I989c. Editor. We are here: Politics of aboriginal land tenure. Berkeley, Los Angeles, and London: University of Cal- ifornia Press.

. n.d. The real Bushman is the male one. Botswana Notes and Records 2i. In press.

WILMSEN, E., AND D. DURHAM. I988. "Food as a function of seasonal environment and social history," in Coping with un- certainty. Edited by G. Harrison and I. de Garine, pp. 5.2-87. Cambridge: Cambridge University Press.

WILMSEN, E., AND R. VOSSEN. I990. Labour, language, and power in the construction of ethnicity in Botswana. Critique of Anthropology. In press.

WILY, L. I 979. Official policy toward San (Bushmen) hunter- gatherers in modern Botswana: I966-I978. Gaborone: Na- tional Institute of Development and Cultural Research.

WINQUIST, A. H. I978. Scandinavians in South Africa. Cape Town: Balkema. [RBL]

WORBY, E. i982. Anthropology comes to "class": Concepts of so- cial form and human history in contemporary social theory. B.A. honours thesis, McGill University, Montreal, Que., Canada.

I984. The politics of dispossession: Livestock develop- ment policy and the transformation of property relations in Botswana. M.A. thesis, McGill University, Montreal, Que., Canada.

. n.d. "Not to plow my master's field": Discourses of eth- nicity and the production of inequality in Botswana. MS.

YELLEN, J. I970. A report to the Botswana government on ar- chaeological investigations in western Ngamiland. MS, Na- tional Museum, Gaborone, Botswana.

. I973. Notes to accompany archaeological materials shipped to the National Museum, Botswana. MS, National Museum, Gaborone, Botswana.

524 CURRENT ANTHROPOLOGY Volume 3I, Number S, December I990

. I984. "The integration of herding into prehistoric hunting and gathering economies," in Frontiers: Southern African ar- chaeology today. Edited by M. Hall and G. Avery, pp. 5 3-64. Cambridge: Cambridge University Press.

. I986. The process of Basarwa assimilation in Botswana. Botswana Notes and Records I7: I 5-23.

. I 989. The ethnoarchaeology of !Kung foragers. Paper pre-

sented at the 88th annual meeting of the American An- thropological Association, November.

. I990. The transformation of the Kalahari !Kung. Scientific American 262:96-io5.

YELLEN, J., AND A. BROOKS. i988. The Late Stone Age archaeol- ogy of the !Kangwa and /Xai/Xai Valleys, Ngamiland. Botswana Notes and Records 2o:5-27.

Institutions

The problems of human evolution are diverse; they in- clude questions not only about the when and where of the story but also about the how and the why. The area may therefore be conceived as inherently an interdisci- plinary one, requiring specialised contributions from a wide range of researchers. This is already well recog- nised in the United States (for example, in the Human Behavior and Evolution Society and in multidisciplinary programs at Ann Arbor and Berkeley). In the United Kingdom, however, human-evolution researchers in such disciplines as psychology, social anthropology, ethology, genetics, and linguistics are frequently aca- demically isolated, while many of those in palaeontol- ogy and archaeology have become conscious of the need to incorporate findings and ideas from beyond their tra- ditional domains of spade- and detective work. It was to meet this need that the Human Evolutionary Interdis- ciplinary Research Group (HEIR) was established in Sep- tember I989, following the circulation of a draft manifesto among active researchers and the setting up of a steering committee comprising Clive Gamble and Paul Graves (Southampton), Tim Ingold (Manchester), Graham Richards (Polytechnic of East London), Chris Stringer (British Museum), and Andy Whiten (St. An- drews). Initial financial support has been provided by the Royal Anthropological Institute, and an inaugural meet- ing was held on May i 9, I 990, at Sheffield University on the suitably broad theme "Causal Models in Human Evolution."

Bernard Wood (Liverpool) opened the programme with "Causes of Hominid Evolution," stressing that the ques- tions of how and why cannot be properly answered be- fore we understand about who and when. Henry Plotkin (University College London), following with "Evolution in the Brain and Evolution of the Brain," argued the necessity for an evolutionary framework to explain not only the historical processes by which brains have arisen but also the neural processes that operate within them.

Tim Ingold's paper "On Dissolving the Distinction be- tween Proximate and Ultimate Causation" focussed on the traditional distinction between adapting organisms and external environments, suggesting that this is a mis- leading dichotomy that diverts attention from the active role of organisms in transforming the relational field as a whole-that there is more to evolution than sim- ply changing frequencies of elements. Robert Barton (Sheffield) then offered a critique of current primatolog- ical approaches to the evolution of intelligence in his "Evolution, Cognition, and the Comparative Method: The Mismeasure of Monkeys?"-emphasizing the dan- gers of anthropocentrism and challenging the idea that social and non-social intelligence can be empirically compared. Jim Hurford (Edinburgh), in "Evolution of the Language Acquisition Device," considered the evolu- tionary ramifications of Chomskyan linguistics, and finally Graham Richards addressed the reflexive charac- ter of human-evolution studies and their cultural mean- ing as origin myths. All the papers stimulated lively and often heated discussion, serving to highlight both the value and the difficulties of cross-disciplinary communi- cation. Predictably, perhaps, this seemed particularly ev- ident in the relationship between social anthropological and ecological perspectives, and Robin Dunbar suc- ceeded in raising one or two eyebrows with his comment that "genes have nothing to do with sociobiology."

Considerable time was also given to general discus- sion about the structure and aims of the organization. It was agreed that there should be a six-monthly newslet- ter and at least one meeting per year, but it was recog- nized that perhaps the main value of the group is likely to lie equally in the lines of communication that it opens up amongst individual members and the opportu- nity for generating smaller informal subgroups of re- searchers. For further information, write: Clive Gamble, Department of Archaeology, University of Southamp- ton, Highfield, Southampton SOg 5NH, England.