The transition to farming in eastern Africa: new faunal and dating evidence from Wadh Lang'o and Usenge, Kenya.

The transition to farming in eastern Africa: new faunal and dating evidence from Wadh Lang'o and Usenge, Kenya. Introduction One of the most important developments in African archaeology The continent of Africa has the longest record of human activity of any part of the world and along with its geographical extent, it contains an enormous archaeological resource. Scholars have studied Egyptology for centuries but archaeologists have only paid serious attention to the rest andcognate disciplines over the last few decades has been the increasingsophistication so��phis��ti��cate?v. so��phis��ti��cat��ed, so��phis��ti��cat��ing, so��phis��ti��catesv.tr.1. To cause to become less natural, especially to make less naive and more worldly.2. of the types of model used to document and explain theso-called 'Bantu Expansion'. In the 1960s and 1970s, thespread of Bantu languages Bantu languages,group of African languages forming a subdivision of the Benue-Niger division of the Niger-Congo branch of the Niger-Kordofanian language family (see African languages). , early farming communities and the knowledgeof iron working across vast areas of sub-Saharan Africa were generallyregarded as having been coterminous co��ter��mi��nous?adj.Variant of conterminous.Adj. 1. coterminous - being of equal extent or scope or durationcoextensive, conterminous , and to have occurred relativelyrapidly as a consequence of population growth and migration (e.g. Oliver1966; Hiemaux 1968; Posnansky 1968). In Eastern Africa, social andeconomic changes have been traditionally chronicled by means of a rangeof pottery types: Kansyore pottery marks the Later Stone Age (to3000-2400 BP) while Nderit and Elmenteitan pottery labels pastoralcommunities of the fifth to second millennia BP (Table 1). Urewepottery, named after the type of site where it was first documented(Leakey et al. 1948), has long been regarded as the main fossiledirecteur of the transition from Later Stone Age (LSA) mobilehunter-gatherer societies to Early Iron Age (EIA (Electronic Industries Alliance, Arlington, VA, www.eia.org) A membership organization founded in 1924 as the Radio Manufacturing Association. It sets standards for consumer products and electronic components. ) sedentary farmers(Posnansky 1968; Phillipson 1977). As more data (and [sup.14]C dates) have become available and awider range of possible sources has been taken into consideration, ithas become increasingly apparent that the processes of demographic,linguistic, economic and technological change were vastly more complexthan initially presumed (e.g. Vansina 1995; Ehret 2001; Salas et al.2002), and may well have operated independently of one another. This inturn has encouraged scholars to explore alternative models (e.g.Robertson & Bradley 2000; Karega-Munene 2002; Lane 2004) and todevelop a more regional focus to their enquiries (e.g. MacLean 1994/5;Eggert 2005; Kusimba & Kusimba 2005). As the number of such regionalstudies has accumulated, knowledge of the distribution and dating ofearly farming and iron-working sites has certainly improved. In Eastern Africa, however, with the notable exception of studiesof iron-smelting technology and its associated symbolism (e.g. Schmidt1997), the archaeological manifestations of other aspects of thesesocieties remain poorly understood (Reid 1994/5). Current modelsconcerning the emergence and spread of domestication domesticationProcess of hereditary reorganization of wild animals and plants into forms more accommodating to the interests of people. In its strictest sense, it refers to the initial stage of human mastery of wild animals and plants. across the regionare also further constrained by the relative lack of plant remains andlarge faunal assemblages from securely dated contexts (Young &Thompson 1999; Marshall 2000), and a dearth of well-stratified sitesspanning the transition from hunting-gathering-fishing to herding andfarming (Sutton 1994/5: 267-8). Here, we report on preliminary results from two recent excavationprojects in Nyanza Province Nyanza Province of Kenya, on Lake Victoria, is one of Kenya's seven administrative provinces outside Nairobi; it is in the southwest corner of Kenya. Nyanza includes part of the eastern edge of Lake Victoria and is inhabited predominantly by the Luo. , Kenya which shed new light on some of theseissues, focusing in particular on the ceramic, lithic lith��ic?1?adj.Consisting of or relating to stone or rock.Adj. 1. lithic - of or containing lithium2. lithic - relating to or composed of stone; "lithic sandstone" and faunalevidence from the multi-component site at Wadh Lang'o, excavated bya team from the National Museums of Kenya (NMK NMK National Museums of KenyaNMK Natural Multimedia KeyboardNMK Network Management Knowledge ) in collaboration withthe British Institute in Eastern Africa (BIEA BIEA British Institute in Eastern Africa ), and subsequently byAshley, and the lithic, fauna and ceramic data sets from the site ofUsenge 3, excavated by Lane and Ashley as part of a broader BIEAinvestigation of the later Holocene landscape archaeology Landscape archaeology is a body of method and theory for the study of the material traces of past peoples within the context of their interactions in the wider (typically regional) social and natural environment they inhabited. of northernNyanza Province (see Figure 1). [FIGURE 1 OMITTED] Wadh Lang'o The site of Wadh Lang'o (GrJd9) was first encountered in 1999,during an impact assessment of areas under threat from the constructionof a dam and hydroelectricity plant on the Sondu River, South Nyanza.The discovery of extensive scatters of ceramic types including'Later Stone Age' Kansyore, 'Pastoral Neolithic'Elmenteitan, and 'Early Iron Age' Urewe, as well as latermaterials, over an area of at least 0.6ha, and evidence of in situ In place. When something is "in situ," it is in its original location. preservation of deposits prompted further investigation by a joint NMKand BIEA team a few months later. During this phase eight test-pits,mostly 1 x lm, were excavated at various localities across the site soas to determine the depth of deposits and quality of preservation. Thisdemonstrated the presence of a well-stratified sequence of depositsextending back at least to the pre-ceramic LSA, with successive layersof Kansyore, Elmenteitan, Urewe, post-Urewe 'Middle Iron Age'(MIA) wares, and historic Luo material, in each case associated withwell-preserved faunal assemblages (Onjala et al. 1999). In view of the obvious significance of these deposits forunderstanding cultural and economic developments in the region duringthe later Holocene, and the possible threat to the site posed by plannedconstruction of a secondary hydroelectric plant in the vicinity,supplementary funds were provided by NMK and the BIEA for a rescueexcavation at the site in June and July 2001. This phase ofarchaeological work was directed by Frederick Odede and Isaya Onjalafrom NMK, with the BIEA acting in an advisory capacity. As part of themitigation work, a 5 x 5m trench was excavated a few metres upslope fromTrench 5 of the 2000 season. The natural subsoil subsoilLayer (stratum) of earth immediately below the surface soil, consisting predominantly of minerals and leached materials such as iron and aluminum compounds. Humus remains and clay accumulate in subsoil, but the teeming macroscopic and microscopic organisms that make was reached at c. 3.2mbelow ground surface and in all, seven broad horizons were identifiedspanning the pre-ceramic LSA to historic/modern Luo horizons. Because ofthe dry, ashy ash��y?adj. ash��i��er, ash��i��est1. Of, relating to, or covered with ashes.2. Having the color of ashes; pale.ash nature of the soil some divisions within these may havebeen missed during excavation; there was, however, very little evidencefor bioturbation bi��o��tur��ba��tion?n.The stirring or mixing of sediment or soil by organisms, especially by burrowing or boring.[German : bio-, bio- + Latin turb or other forms of post-depositional disturbance,especially below 0.4m. A third and final phase of excavation was carried out at the sitein January 2004 by Ceri Ashley, aimed at collecting additional samplesfor dating and analysis. This entailed the excavation of two 2 x 2m testpits to the south-west and south-east of the main 2001 excavation (UnitsA and B respectively), which reached to a maximum depth of 1.88m.Broadly corresponding with the occupational sequence identified inprevious investigations, five major phases of site activity wererecognised; the lowermost deposit ('K') of silty loam wasassociated with Kansyore ceramics (Figure 2), and contained a piledstone cairn cairn,pile of stones, usually conical in shape, raised as a landmark or a memorial. In prehistoric times it was usually erected over a burial. A barrow is sometimes called a cairn. . Overlying overlyingsuffocation of piglets by the sow. The piglets may be weak from illness or malnutrition, the sow may be clumsy or ill, the pen may be inadequate in size or poorly designed so that piglets cannot escape. this, three major ashy deposits containing aseries of isolated lenses of burning suggestive of suggestive ofDecision making adjective Referring to a pattern by LM or imaging, that the interpreter associates with a particular–usually malignant lesion. See Aunt Millie approach, Defensive medicine. temporary hearthswere defined as PN 1-3, and were principally associated with Elmenteitanceramics of the Pastoral Neolithic tradition. Above PN3, Unit A revealeda series of mixed ashy/loam deposits associated with Urewe and MIAceramics ('U'), whilst Unit B showed limited Urewe-relatedactivity, with only a few sherds recovered from the lowermost 5-10cm of'H', a thick occupational hiatus measuring c. 0.Sm depth. InUnit A this 'H' deposit was also recognised, albeit thinner,occurring as a sterile layer of rounded gravelly grav��el��ly?adj.1. Of, full of, or covered with rock fragments or pebbles: a gravelly beach.2. Having a harsh rasping sound: a gravelly voice. sand. Above'H', the uppermost layers ('L') are associated withmodern Luo activity at the site. [FIGURE 2 OMITTED] Unlike the only other comparable multi-component site in theregion, Gogo Falls, where issues of stratigraphic stra��tig��ra��phy?n.The study of rock strata, especially the distribution, deposition, and age of sedimentary rocks.strat integrity andbioturbation have impeded discussion (Robertshaw 1991; Karega-Munene2002), the stratigraphic sequence at Wadh Lang'o is intact, and thelower deposits of interest here are also sealed from modern activity bythe undisturbed occupational hiatus ('H'). As a result, thissite represents a key interpretive resource, and an initial programme ofdating was undertaken by the Oxford Radiocarbon Dating Service (Table2). Perhaps surprisingly, considering the potential temporal span ofsuch a multi-component site (cf. the dated sequence at Gogo Falls, whichminimally spanned 1600 years and conceivably may have extended, withhiatuses, over six millennia), occupation of Wadh Lang'o appears tobe restricted to c. 700 years (excluding the modern/historic Luophases). This circumstance is significant as it suggests a settlementsequence that is temporally inter-connected, despite the discretechanges in strata and ceramics observed. Indeed as Table 2 shows thereis significant overlap at the 2-sigma range between the dates associatedwith Elmenteitan- and Urewe-bearing deposits. Furthermore, while thereis no direct overlap between the Kansyore-related date (OxA-14506) andthe earliest Elmenteitan related date (OxA-14505), the interval betweenthem is insubstantial, and the single available date from the'K' deposits was sourced deep within the context rather thanat its terminal boundary. This notion of inter-connection andrelationship between the different stratigraphic/occupational phases isalso substantiated by associated evidence from ceramics, lithics andfaunal remains, as discussed below. Faunal and artefactual adj. 1. of or pertaining to an artefact.2. made by human actions.Adj. 1. artefactual - of or relating to artifactsartifactual evidence from Wadh Lang'o The assemblages from Wadh Lang'o, as at Usenge 3, were subjectto a high level of recovery: all deposits were dry sieved through 5mmmesh sieves, and 5-251t bulk samples taken for flotation and wet sievingfor each context and/or excavation spit. Ceramic evidence, althoughdemonstrating the presence of typologically discrete traditions, whencollated with stratigraphic evidence demonstrates that discrete ceramicstyles were present/in use contemporaneously at certain key periods ofsite occupation (Ashley 2005). This is particularly notable in theinterface of Elmenteitan and Urewe ceramics in Unit B during theterminal phase of PN3 (AID 280-400). Here the quite distinct ceramictraditions (Figure 3) are found in co-existence for a short period, andare indeed found in some instances to be using similar raw materialsources in their clay. [FIGURE 3 OMITTED] Similarly, the Urewe and the so-called MIA ceramics were found indirect association in Unit A in 'U' deposits. Previously,Robertshaw (1991) had tentatively suggested that MIA ceramics mightrepresent a devolution of Urewe (the transition from the EIA to the MIA,hence the name), thereby reflecting later occupation of the broaderregion. Evidence from Wadh Lang'o, only the second site containingMIA ceramics to be reported to be spoken of; to be mentioned, whether favorably or unfavorably.See also: Report , suggests that this was not the case, andthat instead there was coeval co��e��val?adj.Originating or existing during the same period; lasting through the same era.n.One of the same era or period; a contemporary. use of both ceramic traditions, wellwithin the accepted time-frame for Urewe-using societies (OxA-14499,OxA-14500) and that MIA ceramics cannot therefore be a laterchronological degeneration. Analysis by Oula Seitsonen of lithics recovered in the 2001 and2004 seasons (Seitsonen 2004; in prep.), supports these arguments forlevels of continuity and inter-connection between different phases ofsite use, set against contrasting evidence for change. For instance,lithics from contexts analogous to the 'K' deposits show areliance on quartz with some limited use of chert chert:see flint. , quartzite quartzite,usually metamorphic rock composed of firmly cemented quartz grains. Most often it is white, light gray, yellowish, or light brown, but is sometimes colored blue, green, purple, or black by included minerals. , obsidian,basalt basalt(bəsôlt`, băs`ôlt), fine-grained rock of volcanic origin, dark gray, dark green, brown, reddish, or black in color. Basalt is an igneous rock, i.e., one that has congealed from a molten state. and other igneous ig��ne��ous?adj.1. Of, relating to, or characteristic of fire.2. Geologya. Formed by solidification from a molten state. Used of rocks.b. Of or relating to rock so formed; pyrogenic. rocks (Table 3). These raw materials were usedto fashion a microlithic mi��cro��lith?n. ArchaeologyA very small blade made of flaked stone and used as a tool, especially in the European Mesolithic Period.mi technology reminiscent of the'nondescript' microlithic collections recorded in associationwith Kansyore ceramics elsewhere around Victoria Nyanza (Table 4).However, these Kansyore lithic assemblages are not as undefined as hasbeen suggested previously, and the technology actually ex ofstandardisation (Seitsonen in prep.). At Wadh Lang'o, a combinationof bipolar and platform reduction, that was partly based on formalmicroblade production (blade to flake ratio of 1:6), was used. Attentionwas also paid to the differing flaking characteristics of the rawmaterials, the rock-crystal-like quartz with few or no internal cracks,chert and obsidian being selected preferentially for tool and microbladeproduction. As shown in Table 3, large utilised blades and outils ecailles (inthis case bipolar cores; no tool-like outils ecailles were observed inthe assemblage) are present in the 'PN' deposits, althoughmicroliths still dominate. Also, based on assessments of obsidian colour(cf. Robertshaw 1991: 94), obsidian appears to have been mainly derivedduring the 'PN' from the same Central Kenyan Rift Valley rift valley,elongated depression, trough, or graben in the earth's crust, bounded on both sides by normal faults and occurring on the continents or under the oceans. sources as were utilised during the Kansyore phases. At the very least,this points to recurring connections between the two areas, perhaps inthe form of social exchange networks, which could have also been thesource of other cultural influences. Nevertheless, a significant shiftdoes occur with the transition to the 'PN' layers, in thatnon-local obsidian comes to dominate the raw material used (Table 3).Due to their different flaking characteristics, quartz and obsidianrequire a completely different set of motor skills (and by implicationmental template) from the stoneworker and, as might be expected, bladetechnology was more extensively used with obsidian than in thequartz-based Kansyore assemblage (blade to flake ratio of nearly 1:3).As in the 'K' levels, both platform and bipolar reduction wereused, although there was higher reliance on the bipolar technology inthe later phases of the reduction sequence than in the preceding'K' layers. This most probably mirrors attempts to preservethe exotic obsidian raw material. The changes in formal tool type frequencies, artefact See artifact. productionand tool repair may also suggest a shift towards more curated technologyin 'PN' layers. This might mirror a shift to a more sedentarysettlement pattern, perhaps closer to the collector end of theforager-collector continuum (sensu Binford 1979). However, even the'K' deposits exhibit signs of collector-like behaviour, sothis development might simply be a difference in emphases (Seitsonen inprep.; see also Dale et al. 2004 for discussion of Kansyore-usingcommunities as delayed-return hunter-gatherers). These patterns of rawmaterial exploitation and lithic technology In archeology, Lithic Technology refers to a broad array of techniques and styles to produce usable tools from various types of stone. The earliest stone tools were recovered from modern Ethiopia and were dated to between two-million and three-million years old. are largely replicated inthe 'U' layers (Table 4). This again suggests high levels oftechnological continuity from 'PN' to 'U' layers, aswell as an intriguing reliance on lithic tools into the putative'Iron Age', albeit with a possible shift towards less curatedtechnology. Faunal analysis of the collections from Ashley's excavations(and test-pits 2, 4 and 5 excavated in 2000) conducted by Paul Harvey <noinclude></noinclude>For the Stuckist artist, see Paul Harvey (artist).Paul Harvey Aurandt (born September 4, 1918), better known as Paul Harvey, is an American radio broadcaster for the ABC Radio Networks. (2005), also revealed interesting signs of economic and subsistencecontinuity between different stratigraphic/occupational phases (althoughit needs to be noted that the total number of identifiable specimensfrom each horizon is low, as is the overall faunal sample from theKansyore levels). Kansyore-associated fauna show a wide range of wildtaxa being exploited, including larger bovids such as eland eland(ē`lənd), large, spiral-horned African antelope, genus Taurotragus, found in brush country or open forest at the edge of grasslands. Elands live in small herds and are primarily browsers rather than grazers. and buffalo(Table 5). However, 46 per cent of the sample (based on NISP NISP National Industrial Symbiosis Programme (UK)NISP National Industrial Security ProgramNISP Neutron Instrument Simulation PackageNISP National Individual Security ProgramNISP Nutrition Services Incentive Program ) was foundto represent domesticated do��mes��ti��cate?tr.v. do��mes��ti��cat��ed, do��mes��ti��cat��ing, do��mes��ti��cates1. To cause to feel comfortable at home; make domestic.2. To adopt or make fit for domestic use or life.3. a. ovicaprines, with virtually all body partsexcept the radius being represented. The age profile (although ratherspeculative as it is based on assessment of the fusion ages of bonesowing to owing toprep.Because of; on account of: I couldn't attend, owing to illness.owing toprep → debido a, por causa dethe small sample of teeth recovered for sheep and goat),suggests a 15 per cent drop in individuals over two years of age and aprogressive drop thereafter, broadly consistent with a managementstrategy aimed at increasing herd size while still utilising the flockfor meat (Harvey 2005: 43). This mixed wild/domesticate profile isparticularly notable as previous research has tended to emphasise therole of wild or aquatic resources in this nominal hunter-gatherercommunity (see Sutton 1994/5). The exception is Karega-Munene (2002),who has argued for a significant domesticate (ovicaprines) presenceamong Kansyore-using communities at Gogo Falls. However, issues ofstratigraphic integrity prompted the original excavator ex��ca��va��torn.An instrument, such as a sharp spoon or curette, used in scraping out pathological tissue.excavator (eks´k to urge cautionover this association (Robertshaw 1991). The present evidence from WadhLang'o of securely provenanced domesticates associated withKansyore ceramics therefore seems to substantiate Karega-Munene'searlier assertions. Fauna associated with the Elmenteitan ceramic levels show asignificant decrease in the quantity of wild animal exploitation (22 percent) and also the appearance of Bos sp. for the first time (althoughovicaprines continued to be exploited). Nevertheless, the pattern ofbody part representation (BPR (Business Process Reengineering) See reengineering. BPR - Business Process Re-engineering ) and age profile for sheep and goat remainsimilar to those for the Kansyore levels, suggesting some level of herdmanagement continuity. Similarly, 'U' deposits show an almostanalogous distribution and range of wild/domesticated fauna to those of'PN' (21:79 per cent), reiterating the earlier notion of closecompatibility between the Elmenteitan- and Urewe-using communities. There are, however, some faunal differences between strata, whichindicate shifting patterns of animal exploitation and consumption; forexample there is a change between Kansyore and Elmenteitan levels in theways in which wild and domestic fauna were butchered and possibly cooked(Table 6). The fish remains from different levels also hint at changesover time. The Kansyore levels are dominated by cichlids (>70 percent), with Cyprinidae comprising 22 per cent of the total; this patternis reversed in the PN levels, with Cyprinidae rising to 59 per cent andcichlids dropping to 33 per cent. While this may represent evidence fora change in fishing strategies or perhaps species selection, moreprecise species identification is required than was achievable in thetime available and the reference collections to hand during this stageof the analysis. Perhaps of more significance is that while the ratiooffish off��ish?adj.Inclined to be distant and reserved; aloof.offish��ly adv.off to mammal is roughly the same for the Kansyore and PN levels,there is a much higher proportion of mammal bones to fish in the Urewelevels, which may be indicative of dietary change. Usenge 3 The Usenge 3 (GqJa 3) site was first located during archaeologicalsurvey in North Nyanza District in November 2000 below a low bluff,which probably represents a former shoreline (cf. Temple 1964), on thenorth-western side of Lake Saru (Figure 4). During this survey at leasteight shell-midden sites with Kansyore and later deposits were locatedalong this bluff, the nearest of which (Usenge 1, GqJa 1) is about 120maway to the west (Lane et al. 2006). Usenge 3 is currently subject tohoe hoe,usually a flat blade, variously shaped, set in a long wooden handle and used primarily for weeding and for loosening the soil. It was the first distinctly agricultural implement. The earliest hoes were forked sticks. cultivation, and two of the test-pits excavated (B & C) showedsigns of modern disturbance (including a child burial). Nevertheless, athird 5 x 3m unit (Trench A) revealed intact deposits, apparentlyprotected from modern interference by a nearby hedge. Like Usenge 1,Usenge 3 showed an early use of Kansyore ceramics in association with adense build up of shellfish debris (Figure 5). Above this horizon, Ureweceramics were encountered in a rare state of horizontal preservationincluding a moulded clay stand for holding a large pot and a series ofceramic scatters, the most impressive of which probably represents adiscrete rubbish midden middendungheap. . The uppermost occupational horizon revealed aseries of pits and hollows and was clearly linked to recent Luo activityat the site, with roulette-decorated ceramics dominating the materialculture assemblage from these contexts. Based on the available AMS AMS - Andrew Message System radiocarbon dates (Table 2), the dated sequence recovered concurs withthe extant chronology for the wider region, but also shows significanthiatuses of occupation between the different horizons. [FIGURES 4-5 OMITTED] Faunal and artefactual evidence from Usenge 3 Kansyore ceramics from Usenge 3 fall within the typical typologicaldefinition, being dominated by simple wide-mouthed bowls with taperedlips and extensive stabdrag/impressed covering patterns. Associatedlithics also conform to Verb 1. conform to - satisfy a condition or restriction; "Does this paper meet the requirements for the degree?"fit, meetcoordinate - be co-ordinated; "These activities coordinate well" the known definition, being primarily of quartzwith low quantities of formal microlithic tools. Analysis of the faunalevidence by Sada Mire mire(mer) [Fr.] one of the figures on the arm of an ophthalmometer whose images are reflected on the cornea; measurement of their variations determines the amount of corneal astigmatism. miren. indicates that, in contrast to Wadh Lang'o,subsistence strategies at Usenge 3 during the Kansyore phase were almostexclusively based on hunting, gathering and fishing. Exploitation ofaquatic resources is abundantly displayed in the concentrated shellmatrix and high quantities offish bones, as well as wild mammals such ashippo and sitatunga sitatunga:see bushbuck. , and different reptiles (Table 7). Of the sixidentifiable fish taxa, cichlids and Protopteorus aethiopicus (Africanlung-fish) are the most common, with each making up roughly 27 per centof the identifiable elements. A few domesticates (< 1 per cent of thetotal) are also present, and as with the Kansyore levels at WadhLang'o these are limited to ovicaprines (Mire n.d.). However, asthe [sup.14]C dates show, this is an earlier instance of domesticateexploitation within a nominal hunter-gatherer society, being c.1000-1500 years older than those examples dated at Wadh Lang'o. The second occupational phase in the mid-first millennium ADclosely correlates with known dates for Urewe-related activity inwestern Kenya, with the nearby type site of Urewe/Yala Alego also datedto the fifth century AD (Soper 1969). Fauna from this horizon shows amodest shift in the mixture of wild/domesticated (ratio of 81:19 percent) exploitation (Table 8), and the first presence of Bos sp.suggesting an extension of the domesticate portfolio. Nevertheless,alongside this apparent expansion into larger animal husbandry animal husbandry,aspect of agriculture concerned with the care and breeding of domestic animals such as cattle, goats, sheep, hogs, and horses. Domestication of wild animal species was a crucial achievement in the prehistoric transition of human civilization from it isstill notable that large wild ungulates ungulates, ungulataanimals with hooves; cattle, sheep, goat, pig, horse and many wild and other domesticated species. continue to be exploited intandem Adv. 1. in tandem - one behind the other; "ride tandem on a bicycle built for two"; "riding horses down the path in tandem"tandem , with remains of water buffalo water buffalo:see buffalo. water buffaloor Indian buffaloAny of three subspecies of oxlike bovid (species Bubalus bubalis). Two have been domesticated in Asia since the earliest recorded history. recovered within the potteryscatters found at this level. Similarly, a large range of small wildmammals and fish were also found and, overall, hunting and fishing maywell have remained the basis of the subsistence economy A subsistence economy is an economy in which a group generally obtains the necessities of life, but do not attempt to accumulate wealth. In such a system, a concept of wealth does not exist, and only minimal surpluses generally are created, therefore there is a reliance on renewal . Moreover,following arguments in Marshall & Stewart (1994), the faunalassemblage Faunal Assemblage is the archaeological or paleontological term for a group of associated animal fossils found together in a given stratum.The principle of faunal succession is used in biostratigraphy to determine each biostratigraphic unit, or biozone. suggests that these were specialist and accomplished huntersrather than opportunistic kills made by individuals primarily engaged ina domestic economy. For example, Marshall & Stewart (1994:14-15)argue that the danger involved in hunting buffalo is such that onlyexperienced hunters would attempt it. Equally, the range of smallmammals and fish argues for an exploitation and understanding of thewhole spectrum of wild resources. Hence, while there appears to be anexpanding domesticate sector (which suggests economic change), theexploitation of wild resources remained an integral element of thesubsistence calendar, albeit with possible changing strategies from thepreceding horizon. For instance, among the identifiable fish remainsfrom the Urewe levels, Claridae make up 40 per cent of the identifiableelements, while both cichlids and Protopterus aethiopicus declineslightly to 25 per cent and 24 per cent, respectively. The lithic evidence shows even greater levels of continuity betweenthe technologies found in association with both Kansyore and Ureweceramics, once again reinforcing the notion of long-term continuity.There are no significant differences in the assemblages connected todifferent stratigraphic contexts (Seitsonen in prep.), a point which isreminiscent of observations made at Nsongezi rockshelter (Nelson &Posnansky 1970) and Ugunja (Mosley & Davison 1992: 132-4). Thelithic technology used is also largely analogous with that recorded forthe Wadh Lang'o 'K' layers and at other Kansyore sites inthe area. Throughout the sequence lithics are characterised by arelatively expedient, quartz-based technology. Also, while chert,quartzite and igneous rocks were used, no obsidian was found (Table 3).This might be connected to the greater age of Usenge 3 compared with thedated Kansyore levels at Wadh Lang'o, especially since theas-yet-undated lowest excavation spits of the 2001 trench at WadhLang'o were also totally devoid of obsidian. As with WadhLang'o, the lithic technology at Usenge 3 seems more standardisedthan has been described previously for other quartz-based Kansyoreassemblages (Robertshaw et al. 1983: 34). Both platform and bipolarreduction were used and a clear microblade element is present, althoughin a smaller proportion than at Wadh Lang'o (blade to flake ratiois 1:8). Only a few formal retouched tools were encountered, microlithsbeing most common (Table 4). The high reliance on local raw materials, as well as somecharacteristics of tool production and curation suggest that theinhabitants of Usenge 3 were possibly logistically more mobile on alocal level, stayed at the site for shorter periods than was the case atWadh Lang'o, and possibly lacked some of the contact networksavailable to the inhabitants of Wadh Lang'o (Seitsonen in prep.).The observed differences in the lithic data from the two sites might beconnected to varying topographical, chronological, cultural and/orfunctional differences between them. However, these inferences remaintentative, and more studies of the Kansyore and later lithic assemblagesare needed to support them. Whilst economic and technological evidence points to significantlevels of continuity across the horizons, ceramic evidence seemssuperficially to show a more obvious disjunction disjunction/dis��junc��tion/ (-junk��shun)1. the act or state of being disjoined.2. in genetics, the moving apart of bivalent chromosomes at the first anaphase of meiosis. , with the firstappearance of Urewe ceramics. This ceramic, which is the hallmark of thesupposed EIA package right across the Great Lakes region The Great Lakes region can refer to: Great Lakes region (North America) African Great Lakes region , ischaracterised by well-made and crafted vessels exhibiting a range ofdistinguishing features including dimpled bases, bevelled rims and anextensive and complex decorative palette of incised patterns. Within theUsenge 3 assemblage eight of the possible 262 reconstructable vesselsperfectly conform to this model (see Figure 6a-c); however, the majorityare distinct in many respects and require separate attention. [FIGURE 6 OMITTED] This differentiation is particularly notable because it does notrepresent an entirely new ceramic tradition to Urewe; rather theceramics superficially resemble Urewe in many of the typologicalfeatures (e.g. forms, faceted rims), but yet are sufficiently dissimilarin the details of technological production and morphology to beconsidered distinct. Most striking are the differences in the fabric ofthe Usenge 3 vessels and their relative lack of decoration. In contrastto the typical fabric matrix of Urewe vessels, which is generally fairlywell sorted with only isolated inclusions, the ceramics from Usenge 3are coarse and heavily tempered by large and irregular quartziteinclusions. This high incidence of tempering has weakened the claymatrix and results in a fractured and friable friable/fri��a��ble/ (fri��ah-b'l) easily pulverized or crumbled. fri��a��bleadj.1. Readily crumbled; brittle.2. Relating to a dry, brittle growth of bacteria. vessel body, liable tofissure fissure/fis��sure/ (fish��er)1. any cleft or groove, normal or otherwise, especially a deep fold in the cerebral cortex involving its entire thickness.2. a fault in the enamel surface of a tooth. along the lines of protruding inclusions (see Figure 6d-e).These irregularities create an uneven surface that is not suitable forthe detailed decoration typically associated with Urewe ceramics, and itis notable that in the Usenge 3 collection less than 3 per cent of thesherds are decorated. This is in contrast to figure ranges of 66-85 percent found in more typical Urewe assemblages from right across the GreatLakes region (Van Grunderbeek 1988; Ashley 2005). These distinctions are intriguing and cannot be linked to any knownintra-Urewe variations such as Posnansky's Devolved Urewe(Posnansky 1973; Posnansky et al. 2005) or later variants recentlyrecognised from research in Buganda (Reid 2002; Ashley 2005). Theseexamples all retain a higher level of morphological and technologicalsimilarity to the typical Urewe ware, and are also believed to representchronological devolutions, which cannot be the case at Usenge 3 wherethe ceramics are dated well within the accepted Urewe-using time-frame.Instead it is suggested that the ceramics encountered at Usenge 3represent a unique and highly localised phenomenon, which is restrictedto the immediate lake basin environment, and is defined by thepre-existing socio-cultural and economic system. Reconstructing thechaine operatoire of Usenge ceramic production, it is clear that thehigh levels of investment and habitual skills (pyrotechnical py��ro��tech��nic? also py��ro��tech��ni��caladj.1. Of or relating to fireworks.2. pyrotechnic Resembling fireworks; brilliant: a pyrotechnic wit; pyrotechnic keyboard virtuosity. , aesthetic,motor) evident from the typical finely made Urewe ceramic are missing atUsenge, and that a simplified version of the ware is selected instead.This suggests that the community behind the Usenge ceramics did not havea long-term familiarity with the process of Urewe ceramic production,and/or did not have the social structure or desire to invest in suchsemi-specialist craft production. Combined with the lithic and faunalevidence that attest to deeply rooted continuity over the period of siteuse, it is suggested that these hybrid ceramics are the product of theexisting hunter-gathering community who were slowly and intermittentlyappropriating and adapting the trappings of the farming lifestyle,including its material cultures, resulting in a creolised ceramic uniqueto the region. Discussion The evidence outlined above represents a significant empiricalcontribution to the understanding of the late Holocene archaeology of arelatively poorly known area. In contrast to many other sites in thewider region which have suffered from deflation and bioturbation, Usenge3 and Wadh Lang'o provide excellent stratigraphic integrity,ensuring chronological, artefactual and faunal detail that affordinterpretive depth and security. The work presented here is, of course,partial, and further research will no doubt add clarity to some of thedebates; for instance, archaeobotanical investigation could potentiallyrespond to the recent suggestion, based on phytolith phy��to��lith?n.A minute particle formed of mineral matter by a living plant and fossilized in rock. data, that Kansyoreceramic users may have been progenitor pro��gen��i��torn.1. A direct ancestor.2. An originator of a line of descent.progenitorancestor, including parent.progenitor cellstem cells. banana cultivators as early asthe fourth millennium BC (Lejju et al. 2006). Nevertheless, despite the incomplete picture, even at thispreliminary stage there is ample evidence that the transition to farmingwas a complex, fluctuating and perhaps incomplete, process. Forinstance, and in support of Karega-Munene's earlier assertions withreference to Gogo Falls, the quality of stratigraphic preservation atWadh Lang'o conclusively proves the presence of domesticates (infairly significant proportions) within the economy of the ostensible Apparent; visible; exhibited.Ostensible authority is power that a principal, either by design or through the absence of ordinary care, permits others to believe his or her agent possesses. hunting-gathering-fishing community using Kansyore pottery. With theavailable dates it may be that this adaptation at Wadh Lang'orepresents a very late transition in the history of Kansyore-usingcommunities. Nevertheless, even if this is a terminal transformation, itstill warrants a re-adjustment in existing explanations of thetransitions to animal husbandry and domestication, demonstrating thatthe process may have been gradual and localised rather than the dramaticresult of a 'Pastoral Neolithic' revolution. Equally pertinent to this kind of debate is the evidence fromUsenge 3, which points unequivocally to domesticated ovicaprines inassociation with Kansyore ceramics c. 10001500 years before the firstdated occurrence of Urewe ceramics in northern Nyanza, and only slightlylater than the reported first occurrence of ovicaprines at Gogo Falls(see Karega-Munene 2002: 101). Given the evident continuity in lithictechnology and raw material selection at Usenge 3, and the associatedceramic evidence for the development of a distinctive ceramicrepertoire, termed here 'Contact Urewe', the sequence atUsenge 3 may well represent an instance of 'neolithization' byan entirely autochthonous autochthonous/au��toch��tho��nous/ (aw-tok��thah-nus)1. originating in the same area in which it is found.2. denoting a tissue graft to a new site on the same individual. community of hunter-gatherer-fishers, as alsotentatively postulated for the nearby site of Ugunja (Mosley &Davison 1992: 134). Certainly, the analysed data from both sitesundermine the still prevalent notion that the process of domesticationaround Victoria Nyanza formed part of a broader 'package' thatsimultaneously entailed the adoption of iron metallurgy and the spreadof Bantu languages. Similarly, preconceptions of the shift from Elmenteitan-usingcommunities of the Pastoral Neolithic to the assumed impact of migratingUrewe-using agriculturalists need to be re-assessed, as the evidencefrom [sup.14]C dates, ceramics and lithics at Wadh Lang'o allargues for significant and substantial overlap and continuity betweenthe otherwise discrete stratigraphic phases. Thus the transition from'Stone Age' to 'Iron Age' was a rather fluidprocess. In this regard, it should also be noted that similarobservations can be made with reference to the southerly expansion ofearly herding communities along the Eastern Rift between c. 4500-3000BP, as evidenced by the variable associations between Nderit pottery,Eburran Phase 5 LSA lithics and wild and domestic fauna found ondifferent Savanna savannaor savannah(both: səvăn`ə), tropical or subtropical grassland lying on the margin of the trade wind belts. Pastoral Neolithic sites (e.g. Gifford-Gonzalez 1998:128-9; see Table 1). The ceramic evidence is also illustrative, showing that duringKansyore- and Elmenteitanusing periods at Wadh Lang'o and theKansyore levels at Usenge 3 a limited range of vessel forms was in use,generally restricted to moderately sized bowls with large open mouths,suitable for a range of functions, but not specifically orientatedtowards a single role. The exception to this is the rare addition of aspout to the Elmenteitan vessels, an addition which is clearly designedfor liquids, and was possibly a response to the growing pastoral elementand the importance of milk. Urewe/Contact Urewe ceramics on the otherhand have a much wider range of forms and sizes, with similarmulti-purpose hemispherical bowls, but also open plates/dishes,narrow-mouthed bowls and globular globularresembling a globe.globular hearta spherical cardiac silhouette, usually greatly enlarged and lacking the detailed outline of the right and left atria and apex. Characteristic of pericardial effusion and cardiomyopathy. jars, each of which is restricted infunction by its shape (e.g. plates are not suitable for storage). Thecombined evidence therefore seems to indicate a declining role forroasted foodstuffs foodstuffsnpl → comestibles mplfoodstuffsnpl → denr��es fpl alimentairesfoodstuffsfood npl → in the Urewe-using period, and a growingsophistication of pot-based food production and consumption. As MacLean(1998) has argued, this broadening of ceramic repertoires is acharacteristic of Urewe sites throughout the Great Lakes region, and maywell signal wider sociological changes associated with thereconfiguration of gender relations. On the basis of the evidence from Usenge and Wadh Lang'o, wewould also argue that unlike traditional explanations of the transitionto farming in this region which identified change as profound andover-arching, change and transformation are instead to be recognised inmore nuanced differences, as deeply embedded cultural processes such ascooking and ceramic production adapt and transform according to according toprep.1. As stated or indicated by; on the authority of: according to historians.2. In keeping with: according to instructions.3. circumstance and against the context of perhaps several 'movingfrontiers' of domestication (cf. Gifford-Gonzalez 1998; Lane 2004).After all, 'domestication' is surely as much something in themind as it is a practical processes, and the evidence archaeologistsfind may need to be interpreted in symbolic and structural terms as wellas with reference to more conventional chronological and functionalcriteria. Conclusions In the light of these points, we suggest that just as widerdiscourse is recognising that the Bantu expansion model is notuniversally homogenous homogenous - homogeneous and replicating, the question of change andsocial transition must be re-focused to look at local conditions,collapsing the scale of analysis from the generic to the specific. Thus,it is clear that typological change in ceramics does not automaticallyherald a socio-cultural shift, as is implicit in Adj. 1. implicit in - in the nature of something though not readily apparent; "shortcomings inherent in our approach"; "an underlying meaning"underlying, inherent its widespread use as afossile directeur, and that ceramic boundaries may not be binding.Instead, the process of change needs to be examined more carefully andsubtly, using the available resources to develop a more nuancedperspective that does not make a priori a prioriIn epistemology, knowledge that is independent of all particular experiences, as opposed to a posteriori (or empirical) knowledge, which derives from experience. behavioural assumptions aboutthe material culture patterning. Acknowledgements The excavations at Wadh Lang'o were funded principally byKenGen Ltd. as part of a broader package of financial support formitigation work in connection with the Sondu Miriu dam andhydroelectricity generating plants, with supplementary funds andlogistical support provided by the National Museums of Kenya, the KenyaMuseums Society and the British Institute in Eastern Africa. CeriAshley's excavations at Wadh Lang'o and her ceramic analyseswere conducted as part of her PhD studies, with generous support fromthe Arts & Humanities Research Board (now AHRC AHRC Asian Human Rights CommissionAHRC Arts and Humanities Research Council (UK)AHRC American Homeowners Resource CenterAHRC Army Human Resources CommandAHRC Association for the Help of Retarded Children ); ORADS/AHRC (ORADS2004/2/13); the William Lambarde Memorial Fund (Society of Antiquaries Society of Antiquaries can refer to: Society of Antiquaries of London Society of Antiquaries of Scotland Society of Antiquaries of Newcastle-upon-Tyne Royal Society of Antiquaries of Ireland );Institute of Archaeology The Institute of Archaeology is an academic department of University College London (UCL), in the United Kingdom. The Institute is located in a separate building at the north end of Gordon Square, Bloomsbury. Awards; UCL UCL University College LondonUCL Universit�� Catholique de LouvainUCL UEFA Champions LeagueUCL Upper Confidence LimitUCL University of Central LancashireUCL Upper Control LimitUCL Unfair Competition LawUCL Ulnar Collateral Ligament Graduate School; and the CentralResearch Fund of the University of London For most practical purposes, ranging from admission of students to negotiating funding from the government, the 19 constituent colleges are treated as individual universities. Within the university federation they are known as Recognised Bodies . Lithic analyses of the WadhLang'o material were funded by the University of Helsinld and theBritish Institute in Eastern Africa. Excavation and analyses of theUsenge material was funded by the British Institute in Eastern Africa.We thank the Government of Kenya for permission to conduct fieldresearch, Dr George Abungu (former Director General), Dr Idle Farah(Director General) and Dr Mzalendo Kibunjia (Director, Sites &Monuments) of the National Museums of Kenya for their support and twoanonynous referees for their helpful comments on an earlier draft. Received: 5 June 2006; Accepted: 21 August 2006; Revised: 5September 2006 Note: With respect to the Wadh Lang'o site note that, to date, onlythe ceramic and faunal assemblages from Ashley's excavations, someof the 2000 test-pit fauna, and the entire lithic assemblage have beenthe subject of detailed study (Ashley 2005; Harvey 2005; Seitsonen 2004,in prep.). Work is still ongoing on the ceramics and fauna from the1999-2001 excavations by, respectively, Frederick Odede (MasenoUniversity) and Mary Prendergast (Harvard University) as part of theirPhD studies. The AMS dating of additional charcoal samples and analysisof the plant remains recovered by flotation are also planned. Theinterpretation of the remains from the Wadh Lang'o site musttherefore be regarded as provisional. References ASHLEY, C.Z. 2005. Ceramic Variability and Change: A perspectivefrom Great Lakes Africa. Unpublished PhD thesis, University College,London. BINFORD, L. 1979. Organization and formation processes: looking atcurated technologies. Journal of Anthropological Research 35:172-97. CLIST CLIST Command ListCLIST Center for Labor Information and TrainingCLIST Comma Separated List , B. 1987. A critical reappraisal of the chronologicalframework of Urewe Early Iron Age industry. Muntu 6: 35-62. DALE, D., F. MARSHALL & T. PILGRIM. 2004. Delayed-returnhunter-gatherers in Africa? Historic perspectives from the Okiek andarchaeological perspectives from the Kansyore, in G.M. Crothers (ed.)Hunters and Gatherers in Theory and Archaeology: 340-75. Carbondale:Southern Illinois University, Center for Archaeological Investigations,Occasional Paper No. 31. EGGERT, M.K.H. 2005. The Bantu problem and African archaeology, inA. Stahl (ed.) African Archaeology: A Critical Introduction: 301-26.Oxford: Blackwell. EHRET, C. 2001. Bantu expansions: re-envisioning a central problemof early African history. International Journal of African HistoricalStudies 34: 5-41. GIFFORD-GONZALEZ, D. 1998. Gender and early pastoralists in EastAfrica., in S. Kent (ed.) Gender in African Prehistory prehistory,period of human evolution before writing was invented and records kept. The term was coined by Daniel Wilson in 1851. It is followed by protohistory, the period for which we have some records but must still rely largely on archaeological evidence to : 115-37. London:Altamira. HARVEY, P. 2005. Transition on the shore of Lake Victoria: A faunalanalysis of Wadh Lang'o. Unpublished MSc dissertation, UniversityCollege London “UCL” redirects here. For other uses, see UCL (disambiguation).University College London, commonly known as UCL, is the oldest multi-faculty constituent college of the University of London, one of the two original founding colleges, and the first British . HIERNAUX, J. 1968. Bantu expansion: the evidence from physicalanthropology confronted with linguistic and archaeological evidence.Journal of African History 9:505-15. KAREGA-MUNENE. 2002. Holocene Foragers, Fishers and Herders ofWestern Kenya (BAR International Series 1037) (Cambridge Monographs inAfrican Archaeology 54). Oxford: Archaeopress. KUSIMBA, C.M. & S.B. KUSIMBA. 2005. Mosaics and interactions:East Africa, 2,000 b.p. to the present, in A. Stahl (ed.) AfricanArchaeology: A Critical Introduction: 392-419. Oxford: Blackwell. LANE, P.J. 2004. The 'moving frontier' and the transitionto food production in Kenya. Azania 39: 243-64. LANE, P.J., C.Z. ASHLEY & G. OTEYO. 2006. New dates forKansyore and Urewe wares from northern Nyanza, Kenya. Azania 41: 123-38. LEAKEY, M.D., W.E. OWEN & I.S.B. LEAKEY. 1948. Dimple-basedPottery from Central Kavirondo, Kenya. Nairobi: Coryndon MemorialMuseum, Occasional Paper No. 2. LEJJU, B.J., P. ROBERTSHAW & D. TAYLOR. 2006. Africa'searliest bananas? Journal of Archaeological Science 33: 102-13. MACLEAN, M.R. 1994/5. Late Stone Age and Early Iron Age settlementin the Interlacustrine region: a district case study. Azania 29-30:296-302. --1998. Gendered technologies and Rendered activities in theInterlacustrine Early Iron Age, in S. Kent (ed.) Gender in AfricanPrehistory: 163-77. London: Altamira. MARSHALL, F.B. 2000. The origins and spread of domestic animals inEast Africa, in R.M. Blench blench?1?intr.v. blenched, blench��ing, blench��esTo draw back or shy away, as from fear; flinch.[Middle English blenchen, from Old English blencan, & K.C. MacDonald (ed.) The Origins andDevelopment of African Livestock: archaeology, genetics, linguistics andethnography: 191-221. London: UCL Press. MARSHALL, F.B. & K. STEWART. 1994. Hunting, fishing and herdingpastoralists of western Kenya: the fauna from Gogo Falls.Archaeozoologia 7: 7-27. MIRE, S. n.d. Preliminary report on the fauna from Usenge 1 and 3.Ms. on file at the BIEA library, Nairobi. MOSLEY, P.N. & S. DAWSON. 1992. Ugunja: a new Kansyore (Oltome)site. Azania 27: 129-34. NELSON, C.M. & M. POSNANSKY. 1970. The stone tools from there-excavation of Nsongezi rock shelter. Azania 5:119-206. OLIVER, R. 1966. The problem of Bantu expansion. Journal of AfricanHistory 7:361-76. ONJALA, I., M. KIBUNJIA, F. ODEDE & G. OTEYO. 1999. Recentarchaeological investigation along the Sondu Miriu River, Kenya. Azania34:116-22. PHILLIPSON, D.W. 1977. The Later Prehistory of Eastern and SouthernAfrica. London: Heinemann. POSNANSKY, M. 1968. Bantu genesis--archaeological reflexions.Journal of African History 9:1-11. --1973. Terminology in the Early Iron Age Eastern Africa withparticular reference to Dimple Base wares of Lodui Island, Uganda.Panafrican Congress for Prehistory and Related Studies, V: 577-79. POSNANSKY, M., D.A.M. REID & C.Z. ASHLEY. 2005. Archaeology onLolui Island, Uganda 1964-5. Azania 40: 71-100. REID, D.A.M. 1994/5. Early settlement and social organization inthe Interlacustrine region. Azania 29-30: 303-13. --2002. Recent archaeological discoveries in Buganda and theirimplications for archaeological heritage management. Uganda Journal 48:87-103. ROBERTSHAW, P.T. 1991. Gogo Falls: Excavations at a complexarchaeological site east of Lake Victoria. Azania 26: 63-195. ROBERTSHAW, P.T., D. COLLETT, D. GIFFORD & B.N. MBAE. 1983.Shell middens on the shores of Lake Victoria. Azania 18: 1-43. ROBERTSON, J.H. & R. BRADLEY. 2000. A new paradigm New ParadigmIn the investing world, a totally new way of doing things that has a huge effect on business.Notes:The word "paradigm" is defined as a pattern or model, and it has been used in science to refer to a theoretical framework. : the AfricanEarly Iron Age without Bantu migrations. History in Africa 27: 287-323. SALAS, A., M. RICHARDS, T. DE LA FE, M.-V. LAREU, B. SOBRINO, P.SANCHEZ-DIZ, V. MACAULAY & A. CARRACEDO. 2002. The making of theAfrican mtDNA landscape. American Journal of Human Genetics Human geneticsA discipline concerned with genetically determined resemblances and differences among human beings. Technological advances in the visualization of human chromosomes have shown that abnormalities of chromosome number or structure are surprisingly .71:1082-1111. SEITSONEN, O. 2004 Lithics after the Stone Age in East Africa: WadhLang'o case study. Unpublished MA thesis, University of Helsinki The University of Helsinki is not to be confused with the Helsinki University of Technology.The University of Helsinki (Finnish: Helsingin yliopisto, Swedish: Helsingfors universitet. SEITSONEN, O. In prep. Lithics use at Kansyore sites in EastAfrica. SCHMIDT, P.R. 1997. Iron Technology in East Africa: Symbolism,Science and Archaeology. Bloomington: Indiana University Press Indiana University Press, also known as IU Press, is a publishing house at Indiana University that engages in academic publishing, specializing in the humanities and social sciences. It was founded in 1950. Its headquarters are located in Bloomington, Indiana. . SOPER, R. 1969. Radiocarbon dating of "Dimple-Based" warein Western Kenya. Azania 4: 148-53. SUTTON, J.E.G. 1994/5. The Interlacustrine region. Azania 29-30:263-9. TEMPLE, P.H. 1964. Evidence of lake-level changes from the northernshoreline of Lake Victoria, Uganda, in R.W. Steel & R. MansellProthero (ed.) Geographers and the Tropics: Liverpool Essays: 31-56.London: Longmans. VAN GRUNDERBEEK, M.-C. 1988. Essai d'etude typologique deceramique urewe de la region des collines au Burundi et Rwanda. Azania23:11-55. VANSINA, J. 1995. New linguistic evidence and 'The BantuExpansion'. Journal of African History 36: 173-95. YOUNG, R. & G.B. THOMPSON. 1999. Missing plant foods: Where isthe archaeobotanical evidence for sorghum sorghum,tall, coarse annual (Sorghum vulgare) of the family Gramineae (grass family), somewhat similar in appearance to corn (but having the grain in a panicle rather than an ear) and used for much the same purposes. and finger millet in EastAfrica?, in M. van der Veen (ed.) The Exploitation of Plant Resources inAncient Africa: 63-72. London: Plenum. Paul Lane (1), Ceri Ashley (2), Oula Seitsonen (3), Paul Harvey(4), Sada Mire (4) & Frederick Odede (5) (1) Department of Archaeology, University of York This article is about the British university. For the Canadian university, see York University. The University of York is a campus university in York, England. , UK(Email:paul.lane@orange.fr) (2) Cotsen Institue of Archaeology, University of California, LosAngeles UCLA comprises the College of Letters and Science (the primary undergraduate college), seven professional schools, and five professional Health Science schools. Since 2001, UCLA has enrolled over 33,000 total students, and that number is steadily rising. , USA (3) Department of Archaeology, University of Helsinki, Finland (4) Institute of Archaeology, University College London, London, UK (5) Department of History, Maseno University, KenyaTable 1. Chronology of major ceramic traditionsmentioned in the text.Period Ceramic ChronologyLater Stone Age (LSA) Kansyore 8200-2400 BP (Kusimba & Kusimba 2005) 8000-3000 BP (Dale et al. 2004)Pastoral Neolithic/ Nderit 4500-3000 BP (Gifford- Eburran 5 Gonzalez 1998)Pastoral Neolithic Elmenteitan 3300-1300 BP (Karega- (PN) Munene 2002)Early Iron Age (EIA) Urewe 2550-1000 BP (Clist 1987)Table 2. AMS Radiocarbon dates from Wadh Lang'o and Usenge 3. Sample Calibrated ageSite Number Date BP range (2 sigma)Wadh Lang'o OxA-14499 1449 [+ or -] 28 AD 610-700Wadh Lang'o OxA-14500 1484 [+ or -] 26 AD 590-690Wadh Lang'o OxA-14501 1741 [+ or -] 28 AD 280-440Wadh Lang'o OxA-14502 1698 [+ or -] 28 AD 300-470Wadh Lang'o OxA-14503 1746 [+ or -] 28 AD 280-400Wadh Lang'o OxA-14504 1742 [+ or -] 27 AD 280-400Wadh Lang'o OxA-14505 1819 [+ or -] 28 AD 170-310Wadh Lang'o OxA-14506 1989 [+ or -] 28 AD 1-120Usenge 3 Beta-186498 170 [+ or -] 40 AD 1950-1890 AD 1910-1950Usenge 3 Beta-190746 1560 [+ or -] 40 AD 410-600Usenge 3 Beta-190747 3310 [+ or -] 40 1690-1500 BCUsenge 3 Beta-186499 3240 [+ or -] 70 1680-1390 BC Provenance (all woodSite charcoal samples)Wadh Lang'o Unit A, Urewe/MIAWadh Lang'o Unit A, Urewe/MIA hearthWadh Lang'o Unit A, UreweWadh Lang'o Unit B, UreweWadh Lang'o Unit B, mixed Urewe/ElmenteitanWadh Lang'o Unit B, mixed Urewe/ElmenteitanWadh Lang'o Unit B, earliest ElmenteitanWadh Lang'o Unit A, KansyoreUsenge 3 Luo depositsUsenge 3 Urewe ceramic middenUsenge 3 Kansyore shell middenUsenge 3 Kansyore shell middenTable 3. Lithic raw material use at Wadh Lang'o and Usenge 3.Site & Phase Obsidian Chert QuartzWadh Lang'o--U -857 19 16 94% 2% 2%Wadh Lang'o--PN -4884 102 55 96% 2% 1%Wadh Lang'o--K -192 652 1904 5% 19% 54%Usenge 3 -- 33 661 -- 4% 82%[SIGMA] 5933 806 2636Site & Phase Quartzite Other [SIGMA]Wadh Lang'o--U 3 16 911 < 1% 2%Wadh Lang'o--PN 14 50 5105 < 1% 1%Wadh Lang'o--K 204 568 3520 6% 16%Usenge 3 11 97 802 1% 12%[SIGMA] 232 731 10338Table 4. Composition of the Wadh Lang'o and Usenge 3lithic assemblages. Large Other utilised retouchedSite & Phase Microliths Scrapers blades piecesWadh Lang'o--U 82 9 6 50 9% 1% 1% 5%Wadh Lang'o--PN 347 61 29 78 7% 1% 1% 2%Wadh Lang'o--K 93 67 4 54 3% 2% < 1% 2%Usenge 3 3 2 -- 5 < 1% < 1% -- 1%[SIGMA] 525 139 39 187 Cores (outils Utilised ecailles pieces &Site & Phase in parentheses) debitage [SIGMA]Wadh Lang'o--U 24 (23) 740 911 3% 81%Wadh Lang'o--PN 203 (409) 4387 5105 4% 86%Wadh Lang'o--K 194 (120) 3108 3520 6% 88%Usenge 3 15 (7) 777 802 2% 97[SIGMA] 436 9012 10338Table 5. Taxonomic quantification of the Wadh Lang'o mammal, bird andfish and fauna by NISP & MNI. Kansyore Elmenteitan UrewePhase:Taxon NISP MNI NISP MNI NISP MNIBaboon Papio sp. 3 1 19 2Bohors Reedbuck Redunca redunca 3 1 11 2 3 1Buffalo Syncerus caffer 3 1 7 1 8 1Canine Canis sp. 1 1 1 1Cattle Bos sp. 14 2 1 1Dik Dik Madoqua sp. 2 1Duiker Cephalophinae 2 2 9 2 1 1Eagle Owl Bubo sp. 1 1Eland Taurotragus oryx 6 1Hartebeest Alcelaphus buselaphus 1 1 2 1Hyena Crocuta crocuta 1 1Impala Aepyceros melampus 3 3 58 3 7 2Klipspringer Oreotragus oreotragus 1 1Oribi Ourebia ourebi 4 2Ovicaprine 41 3 494 18 124 5Steenbok Raphicerus campestris 2 1Thompson's Gazelle Gazella thomsoni 12 3 35 3 10 2Warthog Phacochoerus africanus 1 1 1 1Total 76 18 659 41 157 15Cyprinidae 32 3 1377 28 7 2Bagridae 4 1 15 1Cichlidae 4 3 765 8Claridae 8 1 169 5Citharinidae 4 1Protopterus aeathiopicus 20 3Total 48 8 2350 46 7 2 No. unidentified, all levels = 1550Table 6. Notable characteristics of the Wadh Lang'o faunalassemblage.Variable Kansyore ElmenteitanWild: Domestic % 54:46 22:78Mammals: Fish % 23:77 24:76Cut marks: 4% of total, light 6% of total; General shallow cut marks predominantly shallow slice marksCut marks: on Common, on limb Present, mostly wild fauna bones & restricted to consistent with phalanges, disarticulation consistent with skinningCut marks: on None >5% of all sheep/ domestic goat; mostly fauna slices on long bones consistent with disarticulationBurning 14% of total, only 7% of total, all limb bones elements (irrespective of (irrespective of whether wild or whether wild or domestic) domestic)Variable UreweWild: Domestic % 21:79Mammals: Fish % 62:38Cut marks: 5% of total, General typically deep & numerous -- possibly indicative of use of iron toolsCut marks: on Present on 60% of wild fauna buffalo; only on vertebrae, possibly associated with removal of skullCut marks: on 9% of all sheep/ domestic goat; mostly fauna slices on long bones consistent with disarticulationBurning 2% of total, all elements, but only domestic animalsTable 7. Taxonomic quantification of fauna from Kansyore and Urewelevels, Usenge 3 (NB many bones, and especially fish bones from theKansyore levels were heavily concreted, hence the overall weights ofdifferent parts of the assemblage should be treated with caution). Kansyore UrewePhaseTaxon NISP MNI NISPWild Mammals Antelope 1 1 Bovid small 11 5 7 Bovid large 2 2 Canine Canis sp. 2 1 Buffalo Syncerus caffer 2 Bushbuck Tragelaphus scriptus 1 1 1 Common Duiker Sylvicapra grimmia 1 Duiker Cephalophinae 1 1 Gazelle Gazella sp. 5 3 Giraffe Giraffa camelopardalis 1 1 1 Hippo Hippopotamus amphibius 1 1 3 Herpestine Mongoose Herpestidae family 1 1 Mongoose -- generic Herpestidae family 1 Pig Sus scrofa 1 1 Sitatunga Tragelaphus spekeii 13 1 2 Thomson's Gazelle Gazella thomsoni 2 2 Warthog Phacochoerus africanus 1 1 1Domestic Mammals Cattle Bos sp. 2 Ovicaprines 2 1 5Primates Generic, non-human primate 3 2 Baboon Papio sp. 3 1 Black & white colobus Colobus sp. 8 1 Vervet Monkey Chlorocebus pygerythrus 9 1 Human Homo sapiens sapiens 1 1Birds Bird -- generic 4 3 3 African darter Anhinga rufa 1 1 Eagle/Hawks 1 1 Kori bustard Ardeotis kori 6 1 Laughing dove Streptopelia senegalensis 1 1Rodents Rodent -- generic 3 2 5 Giant Elephant Shrew Rhynchocyon chrysopygus 1 1Reptiles/Amphibians Snake -- generic 1 Mamba (snake) Dendroaspis sp. 2 2 Nile Monitor Lizard Varanus niloticus 2 2 1 Marsh Terrapin Pelomedusa subrufa 4 2 Turtle 2Total identified mammal/ bird/reptile 95 46 37Unidentified mammal No. 379 Wt. 1400g No. 681Identified fish species No. 880 Wt. 75g No. 373Unidentified fish No. 3992 No. 1796 UrewePhaseTaxon MNIWild Mammals Antelope Bovid small 3 Bovid large Canine Canis sp. Buffalo Syncerus caffer 2 Bushbuck Tragelaphus scriptus 1 Common Duiker Sylvicapra grimmia 1 Duiker Cephalophinae Gazelle Gazella sp. Giraffe Giraffa camelopardalis 1 Hippo Hippopotamus amphibius 2 Herpestine Mongoose Herpestidae family Mongoose -- generic Herpestidae family 1 Pig Sus scrofa Sitatunga Tragelaphus spekeii 1 Thomson's Gazelle Gazella thomsoni Warthog Phacochoerus africanus 1Domestic Mammals Cattle Bos sp. 2 Ovicaprines 2Primates Generic, non-human primate Baboon Papio sp. Black & white colobus Colobus sp. Vervet Monkey Chlorocebus pygerythrus Human Homo sapiens sapiensBirds Bird -- generic 1 African darter Anhinga rufa Eagle/Hawks Kori bustard Ardeotis kori Laughing dove Streptopelia senegalensisRodents Rodent -- generic 2 Giant Elephant Shrew Rhynchocyon chrysopygusReptiles/Amphibians Snake -- generic 1 Mamba (snake) Dendroaspis sp. Nile Monitor Lizard Varanus niloticus 1 Marsh Terrapin Pelomedusa subrufa Turtle 1Total identified mammal/ bird/reptile 22Unidentified mammal Wt. 1031gIdentified fish species Wt. 20gUnidentified fishTable 8. Notable Characteristics of the Usenge 3 Faunal Assemblage. Kansyore UreweWild: Domestic % * 99: > 1 81:19Mammals: Fish % 17:83 16:84Cut marks: general Undiagnostic cut on > 3%, on unidentifiable one long-bone and 4 cranial elementsCut marks on wild No observation > 1%, particularly fauna on BuffaloCut marks on None as only domestic One sheep/goat bone domestic fauna fauna recovered were sheep/ goat teethBurning Difficult to say due Similar observations to colour as Kansyore modification and heavy concretion* Excluding birds, humans and rodents.