Fig. 1 shows two bifacially retouched MSA foliates from the central Sahara made from Quartzite, the larger one is 12,5 cm long. Foliates are part of four important African technocomplexes: the early Nubian complex, the Aterian, the Lupemban and the Stillbay complex in South Africa. The question, if the African bifacial foliate point production emerged independently in these technocomplexes, or if different regions and traditions were interconnected, is not resolved and remains an open gateway for cultural historical assumptions. Anyhow a transfer of people and/or ideas with very special toolkits does not contradict the MIS 5 Biome Model of North Africa.
Stillbay assemblages are rare and, with the exception of Sibudu Cave ( KwaZulu-Natal), and Apollo 11 (Namibia), all concentrated in the Cape Province of South Africa. Foliate shaped bifacially worked stone points (Fig. 2: Wikimedia Commons, the free media repository (Vincent Mourre / Inrap) are the hallmarks of the Stillbay techno-tradition. Bifacial roughouts are more common than the finished product and correspond to various stages of the reduction process during which flakes are produced by thinning and by shaping the bifacials as shown at Blombos. Bifacial points were most probably used as spear points, as indicated by use wear analysis, but also served as multifunctional tools and were used as knives.
At Blombos Cave, the majority of the bifacial points recovered were made on silcrete that was heat-treated before flaking. After applying hard- and soft-hammer techniques to shape the blank, the points were finely finished using an sophisticated pressure-flaking technique, which is also known from the Lupemban complex (see below).
At present the known Still Bay assemblages show temporal and spatial discontinuity and much variability. At this point it seems not to be possible to reconstruct technological trends or directional change, but this may be a consequence of a sampling bias, with only a handful sites with undisturbed stratigraphy. Data already available suggest for the Stillbay techno-complexes in several South African sites an age from end of MIS 5 to the beginning of MIS 4. Thermoluminescence dating undertaken at Diepkloof Rock Shelter, Sibudu and Apollo 11 indicate a duration for the Still Bay period of around 7,700 years, from 75,5 to 67,8 k.a. ago. Anyhow, newly detected bifacial points from the “pre-Stillbay” strata at Sibudu and Thermoluminescence data from Diepkloof with a mean age of 109 k.a. could indicate that the Stillbay phase started considerably earlier.
Another regional trend in the development of the Middle Paleolithic can be traced in North Africa. Here, two complexes, the Aterian and the Nubian Complex, were recognized. Reviewing the foliates that are shown in this post, they resemble a finely made foliate from the Aterian of the Kharga Oasis, published by Gertrude Caton-Thompson in her seminal work about the Aterian. After M. Kleindienst, the Aterian Unit here is dated between 100-50 k.a. BP.
The Aterian industry is characterized by the use of the Levallois primary reduction technology. The industry was intended for manufacturing points, flakes, and blades. Its diagnostic elements are stemmed pieces, primarily points with a retouched tip and stem. Stems are observed on side-scrapers, end-scrapers, borers, and burins, which indicate that the people widely utilized multifunctional composite tools and reliable hafting-techniques.
Lithic assemblages associated with the Aterian sites are dominated by side-scrapers of various modifications, and also include notched pieces. It is said, that at a later stage in the development of this industry bifacial foliate points became more common, but there is a lack of detailed information about this topic (Fig. 1&3).
OSL analysis yielded a date of 110 k.a. BP for the site of Dar-es-Soltan located near Rabat. The time when sites with similar industry existed in the Temara region is close to this value. The sample derived from the lower Aterian layers at the cave of Mugharet el’ Alyia is dated to the range between 81 ± 9 k.a. BP and 62 ± 5 k.a. BP. At Ifri n’Ammar (Mediterranean side of Morocco) the lowest stratum with tanged objects was dated to 145 ± 9 k.a. (late OIS6).
It is likely that the Aterian industry evolved during late OIS6/OIS 5e and existed for a long time (latest dates around 32 k.a. BP). Sites containing Aterian assemblages located in northwestern Africa seem not to be be older than similar MSA techno-complexes in Egypt. Here the Aterian it is at least as old if not older than in the Maghreb, even accounting for the OSL-estimate of 145 ± 9 for the proto-Aterian at Ifri n’Ammar . A few tanged elements occur in a number of Nubian Complex assemblages from the Nile Valley, such as E-78-11 and Arkin 5. There are some tangs present as well in MIS 5 assemblages at Bir Sahara and Bir Tarfawi area. One of the first researchers of the Aterian, G. Caton-Thompson (1946), considered this industry a flexible technological system tracing its roots to Sub-Saharan Africa. Some scholars link the origin of the Aterian to the Lupemban industry (see below). Given that Aterian assemblages include Nazlet Khater points, and also Nubian Levallois cores, Ph. Van Peer concluded that the Aterian culture belonged to lithic industries from the Nile Valley, and should be integrated it into the Nubian complex.
Many Early Nubian Complex surface scatters in upper Egypt/Sudan were detected by the Combined Prehistoric Expedition in the Sahara Desert led by F. Wendorf from 1962-1999 (Fig.4). As early as 1964/ 1965 the Guichards reported about non stratified assemblages with Nubian cores, Nubian Points, thick scrapers and bifacial foliates in the area that would later be flooded by the Aswan dam. The early Nubian Complex since then was suggested to be characterized exactly by this artifact spectrum. 50 years after, there are some stratified sites, that substantiate this view, but the evidence is still scare.
A Sangoan presence in Northeast Africa, at Site 8-B-11 on Sai Island in the northern Middle Nile Valley has recently demonstrated. Here, late Acheulian and Sangoan occupation levels are inter-stratified suggesting the contemporaneous presence of two technocomplexes during MIS 7. The Sangoan levels at 8-B-11 contain evidence of novel behaviors including the exploitation and processing of iron-oxide pigments and vegetable materials and, in the lithic domain, specialized re-tooling of composite tools with depleted core-axes.
In the 8-B-11 sequence, lanceolate foliates are stratified in an MIS 6 level overlying the Sangoan/Acheulian. This assemblage evidences Lupemban-like features, such as the use of a complex blade reduction system very similar to the one documented in the Lupemban at Kalambo Falls . In addition, the Nubian Levallois technique for the production of points is present, although in small numbers and may provide a link between the Lupemban with the early Nubian Complex. A contemporaneous small lithic assemblage recovered from the exploitation pit at Taramsa-8 (Upper Egypt) also evidence the presence of bifacial foliates during tthis time.
The Paleolithic sequence from Sodmein Cave in the Egyptian Eastern Desert, near Quseir, contains seven stratified archaeological levels from the Middle and Upper Palaeolithic. A huge multilayered hearth occurs in the lowest archaeological level (MP5), and was associated with only a few artifacts. Among the lithics associated with this feature is a Nubian1 subtype Levallois core. Also present was a fragment of a thin, bifacially flaked tool. MP5 was TL dated to 118+/-8 k.a. (MIS 5e). The Nubian Complex sensu stricto is mainly a MIS 5a phenomenon. Bifacial foliates disappear from the typological inventory, but Nubian points still occur.
The Lupemban is an early MSA industry in Central Africa and on the fringes of the Congo basin first described by Breuil more than 60 years ago. This technocomplex is characterized by the presence of bifacial lanceolate points, core-axes and backed blades. Unfortunately sites with contextual information are still rare. JD Clark suggested the heavy duty tool component points to wood-working, based on the association of the Kalambo Falls site in Zambia with deciduous woodland, and preserved wood at site. However, a number of other sites, such as those excavated in Kenya (Lake Victoria Basin) were clearly occupied open grassland or savanna areas.
At Kalambo Falls the MSA Lupemban assemblages are stratified between a late Acheulean, followed by a Sangoan and later strata with LSA material. The Acheulean to Middle Stone Age transition at this key site occurred within a broad time interval of 500–300 k.a. according to recent published OSL dates. In addition the radiometric dates for the Lupemban at Twin Rivers (Zambia) indicate OIS 7 or 6 ages. At site 8-B-11 at Sai Island, Sudan, the Lupemban also follows an Acheulean and Sangoan and is dated to the OIS6. Similar stratified but undated findings from Nubia are known from Arkin 5 and Khor Abu Anga . These sites are at the extreme fringes of the Lupemban interaction sphere, the data base for the Lupemban “heartland” in contrast, is extremely small: adequate paleoenvironmental proxy evidence exists for only 5% of Lupemban sites, only 17% have chronometric dates, and only 3% are dated using radiometric techniques capable of reaching beyond C-14. Stone Age sequences in Central Africa,suffer disproportionately from profound post-depositional disturbance , confounding attempts to isolate reliably and define precisely the industry, and to correlate and compare the cultural content of excavated assemblages.
It has been suggested, that the Lupemban coincides with the first sustained settlement of the Central Africa lowlands. In this scenario, the wooded parts of central Africa probably acted as human refugia at times, though the locations of optimal areas would have changed as humid rainforests and dry grasslands waxed and waned with glacial cycles. In this case the Lupemban’s seemingly long duration and continuity with the subsequent Tshitolian could indicate at high overall population stability in central Africa from MIS 6-2. High population stability also makes it plausible that the region contributed source populations to human dispersal both within and beyond Africa.
But even the contrary could be supposed: A more conservative interpretation of the record of lowland Central Africa might consider that composite tool using mobile human foragers only dispersed into rainforests 40 k.a, or possibly very much later: an interpretation that implicitly casts the Lupemban as a late variant of the MSA of peripheral relevance to the evolution of Homo sapiens.
Sacha C. Jones, Brian A. Stewart (Ed.): Africa from MIS 6-2: Population Dynamics and Paleoenvironments (Vertebrate Paleobiology and Paleoanthropology). The most interesting book about the African Record published during the last year!
Foliates from: AJ Arkell. The Old Stone Age in the Anglo-Egyptian Sudan / by A.J. Arkell. ( Sudan Antiquities Service occasional papers ; no. 1) 1949.