¹ See History in Africa 33(2006), 287–320CrossRefGoogle Scholar.

² This contrary interpretation of the hollows, with his dismissal of the Sirikwa traditions as a modern “exaggeration,” was set out by Huntingford in 1926/27 in a typed memorandum on “Ancient and Historical Monuments in Nandi” (Nandi District office, Kapsabet; copy in SOAS Library, Huntingford collection), and in his article, “Local Archaeology in Kenya Colony,” Journal of the East Africa and Uganda Natural History Society 6(1926), 3-25, esp. 13-15, 24–25Google Scholar; and, more emphatically, idem, “Remarks upon the History of Nandi till 1850” in ibid., ns 5(1927), 3, 6. It was reiterated, with minor variations, for a wider academic readership in “The Azanian Civilization of Kenya,” Antiquity 7(1933), 153-65, esp. 155-56, 162Google Scholar. Although that was the first occasion on which he proposed the name “Azanian” for his imagined, in effect alien, civilization, the notion was consistent with his earlier thinking. For later summary restatements, see his Nandi Work and Culture (London, 1950), 6–11Google Scholar, where this essentially irrelevant archeological digression is included merely to insist that the Nandi (and the rest of the Kalenjin) settled in an empty land already abandoned by the Azanians-and also his contribution to the Oxford History of East Africa I (ed. Oliver, R. and Mathew, G., London, 1963), 71–74Google Scholar.

³ See my Archaeology of the Western Highlands of Kenya (Nairobi, 1973)Google Scholar; also idem, “Hyrax Hill, and the Later Archaeology of the Central Rift Valley of Kenya,” Azania 33(1998), 73-112.

⁴ Since detailed large-scale maps were unavailable in the 1920s, Huntingford was rarely able to record the precise locations of sites, a drawback faced by any subsequent attempt to follow up on the ground Moreover, by the 1960s some features had suffered damage or destruction through farming, and this trend has continued apace since, following radical changes in land tenure and land use across the region.

⁵ Huntingford, , “Local Archaeology,” 20Google Scholar; idem in Man (1931), article 45; idem, “Azanian civilization,” 158-61 However, he avoided the gross fantasizing displayed by a contemporary, Wilson, G.E.H., “The Ancient Civilization of the Rift Valley,” Man (1932)Google Scholar, article 98, presenting a hodgepodge of largely imaginary archeological remains (including extensive agricultural terracing) in dispersed parts of Tanganyika and equally fantastic historical explanations. For my own comments see “‘Ancient Civilizations’ and Modern Agricultural Systems in the Southern Highlands of Tanzania,” Azania 4(1969), 1–13CrossRefGoogle Scholar; and for cattle-tracks, “The ‘Ancient Dams’ of Tanganyika Masailand,” Azania 8(1973), 105–14CrossRefGoogle Scholar. Quite plausibly, the appearance of Wilson's article of 1932 spurred Huntingford to set out his own archeological observations in western Kenya and to articulate his Azanian theory in Antiquity the following year. For another example of settler imagination running wild on the basis of fanciful archeological features, see the statement by Raymond Hook about Mount Kenya and surrounding country quoted by F.E. Zeuner (a highly respected geoarcheologist) in Man (1948), article 14.

⁶ Huntingford, , “Local Archaeology,” 16–20Google Scholar. He avoided mention of this feature-and of monoliths/menhirs generally-in “Azanian Civilization,” but revived the subject two years later: “Megaliths in Kenya,” Antiquity 9 (1935), 219–20Google Scholar. See also Huntingford, , “The Hagiolithic Cultures of East Africa: a contribution to the megalithic problem,” Eastern Anthropologist 3 (1951), esp. 123Google Scholar.

The way in which the Antiquity note of 1935 is introduced suggests that Huntingford was responding to a question forwarded by the journal's editor. It also appears that he was having second thoughts on whether these broken pieces of rock at Tobolwa, a granite tor on the West Nandi escarpment, really represented a former standing stone, deliberately shaped and erected there, rather than natural exfoliation and breakage. I visited Tobol-wa in 1962, and again in 1964 with a copy of Huntingford's description and illustrations in hand, but could discern no convincing sign of human work, although it seemed that some stones had been moved since Huntingford had sketched them in 1922 and 1924.

A photograph accompanying Huntingford's 1935 note does not resolve the issue (and its authorship is not acknowledged). But it is noteworthy as probably the only instance in Huntingford's numerous publications of a photograph being employed to illustrate an archeological object or feature. The same applies to his ethnographic recording. The one case of photographs being attached to a published article of his occurred through his being accompanied in the field by a physical anthropologist, O'Brien, Hon. P.L., for studying Okiek (Dorobo) forest-dwellers east of Nandi: “Modern Hunters,” Journal of the Royal Anthropological Institute 59(1929), 333–78Google Scholar, of which the final section (376-78) and photographs (plates XXVI-XXVIII) are all by O'Brien. Two of those plate-pages illustrate “the physical features of the Dorobo,” while the third has valuable illustrations of their dwellings. Otherwise, the photograph of a “young Nandi warrior,” which serves as frontispiece to Huntingford's book, The Nandi of Kenya (London, 1953)Google Scholar, looks like a stock picture, perhaps supplied by the government's information services. It had already been used in a set of twenty “tribal types” which were inserted into the end-pages of the military handbook, East African Background (Nairobi, 1945, prepared by Huntingford and C.R.V. Bell). Huntingford preferred to sketch by hand (with pencil and notebook) and to illustrate sites and objects with ink drawings, such as appear in most of his articles on archeology in Kenya (and one on northwestern Somalia, , “The Town of Amud,” Azania 13[1978], 181–86Google Scholar—a posthumous publication based on notes made at this “ruined town” in 1943 during wartime duty). If he ever used a camera in fieldwork, he appears to have done so sparingly.

⁷ Huntingford, , “Local Archaeology,” 9–10Google Scholar, mentioning pottery thus retrieved and illustrating a hearth “excavated” in a second example. This site, in the Kipkaren valley, was on the family's farm.

⁸ Ibid, 13, speculating rather vaguely on rafters and thatch; and in “The Distribution of Certain Culture Elements in East Africa,” Journal of the Royal Anthropological Institute 91(1961), 255-56, 271Google Scholar, more specifically illustrating a flat roof, this influenced by the supposed ethnographic parallel (then available in the published literature) of Iraqw semi-sunken rectangular houses in northern Tanzania. Leakey, Similarly Mary, “Report on the Excavations at Hyrax Hill, Nakuru,” Transactions of the Royal Society of South Africa 30/4(1945), 271-409, esp. 355–72Google Scholar, assumed (partly in deference to Louis) that each of the Sirikwa hollows at Hyrax Hill (Site II) contained a house with a flat earthen roof, again citing the supposed Iraqw parallel. She persisted with that interpretation despite her demonstrating so clearly in her meticulous excavations that the hollows were not rectangular but circular in their original construction, and equally the absence-which she candidly acknowledged-of any sign of holes for stout internal posts, which would have been essential for supporting a heavy flat roof. The Leakeys' interpretation of these Sirik-wa hollows was reached independently of Huntingford, and they had the advantage of having actually examined Iraqw houses in 1935. Mary Leakey's error here may be partly explained by her earlier exposure to British archeology, that of the Bronze and Iron Ages, and its terminology for the visible features of those “barbarous” times-hut-circles, pit dwellings etc.-which she thereupon applied in her Hyrax Hill report. In that respect her background was comparable to Huntingford's.

⁹ White African, esp. 162. This was mostly in the summer vacation of 1925, when he bought a motorcycle and traveled across England and Wales.

¹⁰ There is a useful account in Maier, Gerhard, African Dinosaurs Unearthed: the Tendaguru Expeditions (Bloomington, 2003)CrossRefGoogle Scholar. For cursory descriptions of the sites and explanation of the fossil bone scatters, see Parkinson, John, The Dinosaur in East Africa (London, 1931)Google Scholar.

¹¹ A partial exception would have been E.J. Wayland, who from 1919 was Commissioner for Geology in Uganda, with interests extending to archeology, former lake levels, and climatic sequences (see previous article). He visited Leakey at Elmenteita in 1927 and 1928. Another geologist, Erik Nilsson from Sweden, then investigating Rift Valley sediments for evidence of climate change, shared camp with Leakey for several weeks (White African, 201; Stone Age Cultures of Kenya Colony, 10, 12). Also, for part of the second expedition (1929) a young geologist, John D. Solomon, was specially recruited from Britain as assistant. Interestingly, a letter from Wayland to the East African Standard in late 1926, in response to an article of Leakey's in the same newspaper to publicise his mission, shows him being ahead of Leakey in a willingness to concede that stages of the Stone Age in Africa need not be derivatives of those of Europe, but might have had priority; see Lowe, C. van Riet, The Pleistocene Geology and Prehistory of Uganda, Part II: Prehistory (Kampala, 1952), 7–9Google Scholar. Wayland's career is surveyed in a “Memorial Number” of the Uganda Journal, 31(1967)Google Scholar; see especially, Merrick Posnansky, “Wayland as an Archaeologist,” ibid., 9-12.

¹² This was not an entirely fanciful aspiration. As a boy, Leakey had collected obsidian tools and flakes, doubtless mostly of the Late Stone Age, near the family home at Kabete (White African, 68-69, 80-81). Others had been collected in the Rift Valley, and a rich Early Stone Age site with Acheulian hand-axes had been reported by the eminent geologist, J.W. Gregory, whom Leakey had met in Nairobi (The Rift Valleys and Geology of East Africa, 1921, 221; that site, Olorgesailie, was relocated by the Leakeys in the 1940s). Moreover, in 1925/26 Leakey visited Prof. Hans Reck in Berlin, who had recorded mammalian fossils, as well as a human skeleton, at Olduvai in 1913 (White African, 175-78; and see the preceding article). Also broadly encouraging would have been the accumulating reports of Stone Age finds and sites in other parts of the continent, notably the lower Nile valley and South Africa.

¹³ White African, 183, 189-96; Stone Age Cultures, 1-5, 278-79, 283.

¹⁴ A few sherds of pottery were discovered in these supposedly Aurignacian layers at Gamble's Cave II, a finding in absolute disagreement with conventional archeological thinking at the time-and indeed until the present. However, encouraged by locating one sherd in situ and excavating it on the occasion of a visiting scientific party (so that a distinguished professor, J.H. Fleure, FRS, could be cited as witness), Leakey felt bold enough to report it in Stone Age Cultures (103-04, 120-21; see also idem, White African, 232-33, 257-58). This apparently fantastic claim caused concern among his supervisors and in the scientific establishment more generally, with questions about his field methods, as well as his bold and hasty conclusions, such as culminated in the reaction to the Kanam fossil jaw fragment discovered during the 1932 season.

The Gamble's Cave anomaly was largely resolved by new excavation (by Glynn Isaac) in the 1960s, followed by radiocarbon tests, which showed the lowest layers to be only some eight thousand years old, and the attribution of the lithic finds to the Aurignacian as known in Europe to be based on superficial resemblances (or, as some authorities would put it, parallel evolution). The revised dating is still very early for pottery by general world comparison, but fits well into the Middle African “aqualithic” context of the early Holocene, an attribution supported by the lakeshore situation of the level in question, as well as the subsequent identification, among the boxes of bone fragments from Leakey's excavations, of distinctive bone harpoons; Sutton, J.E.G., “The Aquatic Civilization of Middle Africa,” JAH 15(1974), 527–46CrossRefGoogle Scholar.

¹⁵ See Burkitt, M.C., South Africa's Past in Stone and Paint (Cambridge, 1928), esp. 16fGoogle Scholar; Goodwin, A.J.H. and Lowe, C. van Riet, The Stone Age Cultures of South Africa (Cape Town, 1929)Google Scholar. For later comments see Deacon, Janette, “Stone Age Research in Southern Africa” in Robertshaw, Peter, A History of African Archaeology (London, 1990), 43–45Google Scholar, and Robertshaw, “The Development of Archaeology in East Africa” in ibid., 80-81; Schrire, C.et al., “Burkitt's Milestone,” Antiquity 60(1986), 123–31CrossRefGoogle Scholar; and Shaw, Thurstan, “Goodwin's Graft; Burkitt's Craft,” Antiquity 65(1991), 579CrossRefGoogle Scholar. And for “the South Africanization of prehistory” in the 1920s/early 1930s-an endeavor in which J.C. Smuts, sometime prime minister and for a while president of the South African Association for the Advancement of Science, provided pointed encouragement as part of his “sub-imperialist” agenda-see Schlanger, N., “The Prehistory of Field-Marshal Smuts,” Antiquity 76(2002), 200–09CrossRefGoogle Scholar. Leakey must have realized the need for a compromise between his initial Eurocentric principles and the South African scientific initiatives during his visit to that country in 1929 for the joint convention of the British and South African Associations, where he presented a summary of his Kenya findings, published in South African Journal of Science 26(1929), 749–57Google Scholar. Unfortunately Leakey's memoirs tell us nothing about the content of the scientific meeting in Johannesburg, or about reactions to his work or contacts made, except its being “well worth the long journey:” Leakey, , White African, 253Google Scholar. He did not have time to visit any South African archeological sites.

¹⁶ This seems to be a way of accommodating Burkitt's idea (South Africa's Past, 86) of the continuity of particular cultural traits or modes of manufacturing stone tools (implying divergent or parallel lines of technical evolution). Later in Leakey's Kenya sequence another South African term, Wilton, is borrowed for microlithic materials (see below), a similarity being noted in particular with sites in Southern Rhodesia so labeled. This hybrid Kenya sequence, and Leakey's defense of his chosen terminology, are set out in Stone Age Cultures, 27-28. A concluding section of the book considers, rather cursorily, comparisons between Kenya and both South Africa and Europe, this having the look of a hastily concocted exercise to demonstrate conversance with the latest research and thinking. Leakey enlarged on such comparisons in his Stone Age Africa of 1936—a book deriving from a lecture series, combining acute observations with not always reflective thinking.

¹⁷ In both Adam's Ancestors (1934) and Stone Age Africa (1936) with their charts of cultural sequences, Leakey remained firmly committed to the European system conceptually, merely adapting it to his own findings and ideas and, where necessary, accommodating South African categories, but without embracing Goodwin's theoretical underpinning of the latter scheme. That procedure persisted, albeit under strain, in the fourth edition of Adam's Ancestors (1953), and in the synopsis of Leakey's work in Sonia Cole's Prehistory of East Africa (1954).

¹⁸ In particular, Clark, J. Desmond, The Prehistory of Southern Africa (Harmondsworth, 1959)Google Scholar; and his increasingly influential voice in the periodic meetings of the Panafrican Congresses of Prehistory, notably at Livingstone in 1955, with its Resolutions and the establishment of a “standing committee on Terminology:” see Clark, J.D., ed., Third Panafrican Congress on Prehistory (London, 1957) esp. xxxiii–ivGoogle Scholar. This concern for systematic nomenclature and conformity in documentation along scientific principles reached its high point at the Sixth Congress (Dakar) in 1967, which was preceded by a special Burg-Wartenstein symposium in 1965 that resulted in the volume Background to Evolution in Africa, ed. Bishop, W.W. and Clark, J.D. (Chicago, 1967)Google Scholar containing Recommendations on Stone Age terminology, and the simultaneous appearance of the Atlas of African Prehistory, also under Clark's coordination.

¹⁹ The bold concentration on finished tools of recognisable types and disregard for the bulk of the “waste” would, together with the crude statistics, be considered flawed—or at least grossly skewed—by a later generation of Stone Age archeologists. Leading that more meticulous and systematic approach were Mary Leakey and, from the 1960s, Louis' graduate assistant, Glynn Isaac, who—encouraged by Desmond Clark—further pioneered new environmentalist interpretations of Early Stone Age sites and assemblages.

²⁰ There are cursory “preliminary reports” (e.g., on the late Iron Age remains at Engaruka, discussed below) and allusions in his books to site and excavation reports as “forthcoming.” Leakey claimed he completed in 1934 a manuscript on Gamble's Cave for the journal Préhistoire—cf Stone Age Races, 47, 56—but it “got lost somewhere:” information from Glynn Isaac, who delicately enquired thirty years later. For this site there are “diagrammatic sections” in Stone Age Cultures, figures 15-16.

The model reports on Hyrax Hill by Nakuru and Njoro River Cave, excavated in the late 1930s, were virtually entirely written by Mary Leakey, who conducted the work on both sites. Louis' main contribution to these was in examining the skeletons. Similarly at a later stage of their careers, the detailed reports of excavations at Olduvai were almost entirely Mary's work. Louis' Olduvai Gorge of 1951—whose sub-title, A Report on the Evolution of the Hand-Axe Culture in Beds I-IV, betrays its conception two decades previously—like Stone Age Cultures, skips over essential site and excavation details.

The one striking exception is the report of a site in southern England which the Leakeys investigated while on leave in 1950: L.S.B. Leakey, Preliminary Excavations of a Mesolithic Site at Abinger Common (Research Papers of the Surrey Archaeological Society 3, 1951). This includes fine drawings of flint artefacts by Mary Leakey's distinctive hand, as well as a description of a “pit-dwelling” (35-38) with an allusion, for ethnographic comparison, to Dorobo (Okiek) dwellings on the Mau escarpment in Kenya.

²¹ Leakey, , Stone Age Cultures, 200Google Scholar; idem, Stone Age Races, 114. Moreover, his anecdotal account (in Leakey, , White African, esp. 218fGoogle Scholar) of the progress of excavation of Gamble's Cave—and of the dangerous rock collapse while members of the team were at work— suggests that enthusiasm for recovering finds, and especially human skeletal remains, overrode systematic control.

²² Leakey, , Stone Age Cultures, 201Google Scholar; idem, Stone Age Races, 113-14.

²³ M.D. Leakey, “Hyrax Hill.” Later again, Merrick Posnansky, working under the Leakeys' aegis in the same district, assumed at first that the Sirikwa remains at Hyrax Hill and Lanet were “Neolithic: The Neolithic Cultures of East Africa” in Actes du IVe Congrès Panafricain de Préhistoire (Léopoldville) (2 vols.: Tervuren, 1962), 2:273–79Google Scholar. Only afterwards did Posnansky appreciate that they belonged in the later Iron Age a few centuries back (“Excavations at Lanet,” Azania 2[1967], 89–114)Google Scholar. I have discussed these sites in Sutton, , “Hyrax Hill and the Sirikwa: New Excavations at Site II,” Azania 22(1987), 1–36CrossRefGoogle Scholar; and “Hyrax Hill and the Later Archaeology of the Central Rift Valley of Kenya,” Azania 33(1998), 73–112CrossRefGoogle Scholar.

²⁴ This turning event in Leakey's career is told by Cole, Sonia, Leakey's Luck, 94fGoogle Scholar, and Morell, Virginia, Ancestral Passions (New York, 1997), 84fGoogle Scholar, as well as by Louis, himself in White African, 310–11Google Scholar, for the discovery, and By the Evidence, 34f, for the sequel.

²⁵ See Leakey's account in his Stone Age Cultures and extra information inserted into Stone Age Races, and the conspectus by Sonia Cole, The Prehistory of East Africa, with its attempt to incorporate Mary Leakey's work of the late 1930s at Hyrax Hill and Njoro River Cave. For a more recent overview and interpretation, see my “Hyrax Hill” (1998). Leakey, (Stone Age Cultures, 243; Stone Age Africa, 194)Google Scholar, did hint that “Neolithic” cultures might be associated with agriculture and pastoralism, but it is not clear that he regarded such an economic criterion essential for the definition, or reflected far on that aspect when applying the label in the 1920s. He did not seem very concerned to search for evidence of cultivated plants (recovery and identification would have been technically difficult there and then) or to identify bones of cattle or goats/sheep in those early excavations. However, he did assume that the stone bowls and associated pestles were for grinding grain or other possibly cultivated food, although their normal occurrence with burials indicates a special funerary purpose for the bulk of those found. While that in no way rules out a connection with a food-producing economy, particularly pastoralism in these grasslands, it underlines the problematic issue of definitions and legitimate deductions from the material evidence.

He also had a Mesolithic category (Stone Age Cultures, chapter 8), another European term, to cater for materials apparently too recent for true Palaeolithic status, but not sufficiently advanced for the Neolithic accolade. In that category he placed certain sites with blade and microlithic tools, calling the former Elmenteitan, the latter Kenya Wilton, in the latter case after a South African type-site. The occurrence of pottery with both did not seem to him inconsistent with his definitions, even less so since he also found potsherds with what he took to be Aurignacian stone tools (see above) dating, so he thought, thousands of years back in the (Upper) Palaeolithic.

²⁶ The term “pastoral neolithic” became fashionable in the 1970s/1980s to subsume both the burials and activity sites in the Central Rift Valley and adjacent high grassland districts with evidence, either direct or inferred, of cattle and goats or sheep between roughly 1500 BCE and 500 CE. This dating is based largely on radiocarbon tests, a method unavailable when the Leakeys were working in the region. For commentary on the interpretation of radiocarbon results—and their misuse by naive archaeologists who have claimed an antiquity for cattle-herding in this region greater than anywhere else in the world—see Collett, D.P. and Robertshaw, P.T., “Problems in the Interpretation of Radiocarbon Dates: the Pastoral Neolithic of East Africa,” African Archaeological Review 1(1983), 57–74CrossRefGoogle Scholar.

²⁷ See Leakey, , Stone Age Cultures, 243Google Scholar, for an explanation of sorts for choosing the name “Gumban.” Part of the blame here may be attributable to pressure from Leakey's supervisors, who were cautioning against adopting European cultural labels which might imply hyperdiffusion. Apparently Leakey sensed, as his fieldwork progressed, that what he chose to call Gumban A was older than Gumban B, entailing that the first of these burials to be investigated, the Nakuru Burial Site, became Gumban B. The problem with that excavation, as already explained, was that two distinct types of pottery were combined among the finds, one late Iron Age with twisted-roulette decoration (Sirikwa), the other preiron (i.e., more genuinely “neolithic”) and arguably associated with the burials and stone bowls. But since the intrusive (Sirikwa) pottery was the more distinctive in decoration and shapes, it became the unfortunate hallmark of this Gumban B category.

Gumban A was best known for its even more exceptional pottery (nowadays usually called Nderit ware)-basket-shaped with all-over external decoration as well as internal grooving. But Leakey was never perfectly explicit about the distinguishing criteria of Gumban A and B-whether the pottery, the different methods of burial (essentially the styles of cairn: Stone Age Races, 99), or the shapes of stone bowls found within. (A further consideration comprised customs of tooth extraction: ibid, 97.) He seemed to regard all as relevant cultural indicators, even if they might not have stood a rigorous test of correlation. But it was the pottery, forming what he took to be two diagnostic sets, which seems to have weighed most decisively in his categorization of Gumban A and B-and caused the long-running confusion in the latter, one which Leakey never acknowledged. There was also a Gumban C, a pot in the Cambridge museum being so labeled. That designation was dropped by the time Leakey published his outline findings.

²⁸ Leakey, Stone Age Races, chapter 8, and idem, Stone Age Africa.

²⁹ M.D. Leakey, “Hyrax Hill”; and idem, Excavations at Njoro River Cave (Oxford, 1950); for commentary, my “Hyrax Hill.”

³⁰ Prehistory of East Africa, 1954 edition. In Cole's second edition of 1963 the confusion was compounded by splitting the “stone bowl cultures” between Neolithic and Iron Age, both Gumban A and B being treated in the latter chapter. This was an over-reaction to the creeping realization that Leakey's inclusion of roulette-decorated pottery (designated Gumban B) under Neolithic must be wrong. Despite these muddles and uncritical following of the Leakeys, Cole's effort at a synthesis was widely noted. Thus Clark, Grahame in his World Prehistory (Cambridge, 1961), 114Google Scholar; (2d ed.: Cambridge, 1969), 201, looking for economic signals in attempting a comprehensive outline, interpreted the findings from these sites near Nakuru as indications of the early spread of pastoralism this far south in Africaan insight which has been broadly accepted since, though not necessarily through identical reasoning.

³¹ Subsequent research has revealed varied manifestations of this complex—variously called “Pastoral Neolithic” or “terminal Late Stone Age” (blade or microlithic industries associated with pottery, ground stone etc)—both along the Rift and in the highlands to either side, in Kenya itself and extending into northern Tanzania. The term “Gumban” has been generally discarded, however, both for cairn types and for broader cultural assignations-although, being embedded in the literature, students still find themselves grappling with it! One also notices an insistence in some quarters on retaining Neolithic—and recently on recognizing its equivalents in other parts of East Africa—as if the term bestows some sort of academic kudos on the regional archeology.

³² See Ogot, B.A. and Kieran, J.A., eds., Zamani: a Survey of East African History (Nairobi, 1968, 1974)Google Scholar, relevant chapters; Ehret, C. and Posnansky, M., eds., The Archaeological and Linguistic Reconstruction of African History (Berkeley, 1982)Google Scholar, notably chapter 7 (“East Africa”) by Stanley Ambrose (104-57); and my commentaries in “Hyrax Hill and the Sirikwa,” Azania 22(1987), 22–28Google Scholar; and “Hyrax Hill,” 77-82. One obvious change has been the purging of Hamites (including “half-” and other hyphenated Hamites) from the vocabulary, although vestiges of this mentality may remain detectable behind the more “correct” classifications and attitudes.

³³ In fact, one type of feature which Huntingford recorded in a few places in Nandi and Uasin Gishu could perhaps have qualified for the Neolithic category as usually defined. These are the groups of large cairns, being burial monuments of a pastorally-inclined population which preceded the Sirikwa in the pre- or transitional Iron Age of the first millennium CE or possibly earlier. This dating is attested not only by the (non-rouletted) pottery and obsidian blades recovered from these cairns, but also stratigraphically, occasional examples being encroached or overlain by Sirikwa enclosures see: my Archaeology of the Western Highlands of Kenya, 47, 111–14Google Scholar; Huntingford. “Remarks upon the History of Nandi,” 7; idem, “Miscellaneous Records Relating to the Nandi and Kony,” Journal of the Royal Anthropological Institute 57(1927), 439; idem, “The Azanian Civilization of Kenya,” Antiquity 7(1933), 158, felt uncertain about lumping these cairns with the rest of the local archeology in his Azanian package. In this instance he was, atypically, more inclined to follow local rationalizations suggesting that they were burials of warriors killed in nineteenth-century inter-Maasai conflicts.

³⁴ Galloway, A., “Stone Structures on the Uashin Gishu Plateau,” South African Journal of Science 32(1935), 656–68Google Scholar. Galloway was a South African anatomist with an interest in archeology and a developing academic involvement in East Africa (subsequently spending part of his career at Makerere's medical school). The article contains useful descriptions, but is confused on local ethnology, regarding the Elgeyo (or Keyu, a Kalenjin section) as some sort of Maasai.

It was apparently Galloway's article which alerted James Walton—in his ethno-archeological study of southern African settlements and structures, African Village (Pretoria, 1956), esp. 105–11Google Scholar—to note Huntingford's Azanian hypothesis and to suggest wide-ranging parallels if not actual connections, notably with the (I)Nyanga remains of eastern Zimbabwe. Walton's book is beautifully illustrated, but the text, being based on a combination of personal observations and existing literature (some of it poorly handled), has a distinctly antiquarian ring.

³⁵ Cole, , Prehistory (London, 1954), 275–77Google Scholar; see also the revised and expanded edition of this book (New York, 1963), 316-19.

³⁶ Murdock, G.P., Africa: Its Peoples and Their Culture History (New York, 1959)Google Scholar—a work interesting for its independent and systematic attempt at a continental survey following North American principles. Although Murdock's chapter 25 subsumes the archeo-logical features which Huntingford identified as “Azanian,” so that the Megalithic Cushites may appear (to superficial readers) as the Azanians under another name, it appears that Murdock had not actually read Huntingford's article of 1933. His imagined dating of these Cushites, being based on broad culture-historical and comparative-linguistic pointers, is considerably older, in fact, while his overall historical reconstruction of the Kenya highlands, convoluted and misguided though it has been shown, is considerably more sophisticated. Murdock's reliance on Cole only compounded the unsoundness of the data cited, let alone any conclusions thence derived. The following chapter of Mur-dock's book, “Ancient Azanians,” deals with a different subject and region, the East African coast, although the author did assume a connection by arguing that “Megalithic Cushites … descended the few miles from the Kenya highlands.”

³⁷ Davidson, , Old Africa Rediscovered (London, 1959), chapter 8Google Scholar. For the American edition of this book, an alternative title, The Lost Cities of Africa, was preferred by the publisher. This alternative, arguably less close to Davidson's original message, has been perpetuated by the most recent editions/reprints on both sides of the Atlantic.

³⁸ Leakey, , “Preliminary Report on Examination of the Engaruka Ruins,” Tanganyika Notes and Records 1(1936), 57–60Google Scholar. Leakey was specially asked by the Tanganyika government to make an expert assessment of Engaruka, in the wake of rumors and press reports of this “lost metropolis.” He and Mary accordingly detoured from their route to Olduvai in 1935.

³⁹ See my site study, “Engaruka: an Irrigation Agricultural Community in Northern Tanzania before the Maasai,” Azania 33(1998), 1–37CrossRefGoogle Scholar. Leakey's indication of a quite recent date for Engaruka is broadly corroborated by later work. The current estimate for this specialized agricultural community is from the fifteenth century to the seventeenth century approximately.

⁴⁰ See above.

⁴¹ The inherent untenability of Davidson's picture of East African highland history was soon pointed out by critics, and in later editions of Old Africa Rediscovered/Lost Cities of Africa this section was modified.

⁴² Davidson, , Old Africa Rediscovered, 30–31Google Scholar. Here Davidson took Sonia Cole to task for thoughtlessly regurgitating, in her Prehistory of East Africa published only five years before, such a racist notion in attributing some dubious archeological features on Mount Kenya and elsewhere in the highlands to “intelligent” Hamites, and by implication denying any useful achievement to the ancestors of the existing people (1954, 277). But typically he framed his criticism—or ridicule rather—in a way to avoid anything that might be construed as a personal attack. Instead of mentioning Cole by name or citing the book and offending page, he politely alluded to “an otherwise serious anthropological student of East Africa.” That fastidiously civil manner of commenting notwithstanding, Davidson must doubtless have found Cole's attempt to synthesize the most recent part of East African “prehistory” (in effect to summarize the Leakeys' work of previous decades and to tack on various loose ends) frustrating in the extreme, in that it seemed to lead nowhere and offered no sense of a history for the existing people of East Africa. The role of “intelligent” but unspecified Hamites, as Cole seemed to imagine them, only increased the mystification and widened the gulf between ethnographic and archeological approaches to the region's past.

⁴³ Davidson, Old Africa Rediscovered, chapter 8. Despite Davidson's intention to completely reverse the prejudice, the title of this chapter, “After Axum,” resonates with the diffusionist assumptions of the Hamitic school, which imagined vague north-south connections down the “highland spine” of eastern Africa in ancient or medieval times. Here, as also with his treatment of Kush on the Middle Nile, he was anxious not to lose the ingredients of a story which could hold together broad regions of the continent. I intend to discuss this in a forthcoming article on African towns and deserted sites.

⁴⁴ Ibid, 194-95. Davidson's characterization of Lwoo, Hima, and Maasai as “immigrants” and “relatively barbarian hamites” may nowadays be deemed inaccurate as well as unfair. His anti-Hamitic stand became a somewhat undiscriminating anti-pastoralist bias, with pastoralism being as loosely defined as in the colonial tribal stereotypes.

Although he did not elaborate here, Davidson hinted that the same imagined pastoral influx from the north upset some higher civilization established in the interlacustrine region: see ibid., 249, alluding vaguely to the historical roots of the earthworks of Bigo in western Uganda, illustrated by plate 28, as if a westerly counterpart to Azanian stone-building. In general tone he was invoking the common idea of civilizations rising and falling, with the Roman Empire in particular in mind, a vein of thinking reflected in the Gibbonesque title to a later chapter, “Decline and Fall,” to explain what had happened to “Old Africa.” Compare Huntingford's veiled analogy, noted in the previous article (307), with Roman Britain, as understood in British educated circles of his generation, for his image of the Azanians. Davidson (with his intellectual left-wing publisher, Victor Gollancz), in appealing to enlightened thinking at the close of the colonial era, assumed a similar cultural background of his readership.

⁴⁵ Huntingford's late restatement of the Azanian hypothesis, this time with Engaruka included—in the Oxford History of East Africa I—was published in 1963, four years after Davidson's Old Africa Rediscovered. But it was very likely written before that, delays occurring between the planning and eventual appearance of the collaborative Oxford History.

⁴⁶ Having in the early 1960s-with Merrick Posnansky's encouragement-followed up Huntingford's archeological recording in Nandi, Uasin Gishu, and adjacent districts, I would be the first to acknowledge this-and also Basil Davidson's role as a medium in provoking a new look at old evidence and a search for positive historical interpretations-as attempted in my Archaeology of the Western Highlands of Kenya (Nairobi, 1973)Google Scholar. All the same, the unlikelihood of all the random materials on the East African landscape, the soundly documented as well as the fanciful titbits which made up the Azanian package, belonging to a single era and “civilization” ought to have been obvious sooner, since it was a question of simple logic, not one requiring archeological expertise. An early hint to this effect, though not based on close acquaintance with the region or the features in question, was by Mathew, Gervase, “The East African Coast,” in History of East Africa 1:100Google Scholar.