Artifacts of Material History

African Paleolithic Artifacts: Questions with Quartzite Tools from Northcliff, South Africa

Figure 1 Northcliff – Quartzite Hand Tool Acheulian Tradition, most likely ca. 500-300 ky (Photo Garth Hall)

By Garth C. Hall –

This paper addresses the African stone age somewhat myopically, focusing on artifacts and not much on the hominins who made and used them. It also focuses geographically on South Africa, yet, at least in the Early and Middle Stone Ages, the literature indicates that the stone industries were fairly similar at least throughout Africa if not also in Europe and Asia. In the paper, I raise several questions, for which answers may or may not be available from published sources and may prompt further research.

The Span of Time

The span of time addressed in this paper extends perhaps as far back as 2.5 million years ago (Ma) but also extends through to the putative end of the stone age, roughly 2,000 years ago (2 ka) in South Africa. In palaeoarchaeology, our focus on stone artifacts is increasingly crucial the further back in time our exploration takes us. Clearly, what makes palaeolithic material so important is the fact that ancient hominin skeletal remains and other biological material associated with ancestral living readily decompose and are usually absorbed into soils and leached away. Only in rare cases, have stone-age hominin skeletal remains been found, for example, in dolomite and limestone caves or rock shelters with overhangs. In most cases, all that we find are the stone artifacts. Prior to the conventionally defined stone age, however, discoveries of manual cut marks in faunal bones going back as 3.4 Ma have been found in Ethiopia associated with an Australopithecine species (McPherron, et al, 2010) and stone tools associated with hominin fossils in Kenya have been dated to 3.3 Ma (Harmand, et al, 2015). 

Stone-Age Eras

Palaeoarchaeologists typically refer to long-past eras in terms of the dominant industry that produced tools and weapons. The three industries I have reviewed for this paper are:

The Oldowan Industry, spanning approximately 2.6 Ma through 1.5 Ma (Harris, et al, 2013), was characterized by stone artefacts typically with a naturally rounded side that can be held in one hand, with flakes hammered away from the other sides, thus leaving sharp edges useful for chopping. The majority of the African Oldowan sites are in the eastern Rift Valley of Ethiopia, Kenya, and Tanzania; however, a site in Algeria and two close-by sites in South Africa are geographical outliers. Two of the oldest sites in Ethiopia included proximate faunal bone fragments with cut marks, confirming a shift of diet towards meat eating. Other more recent sites yielded numerous bone cut marks. Even though the Oldowan Industry is typically bounded by the time frame stated above, I wonder whether tool-makers in later eras may have sporadically and locally used similar techniques. For example, Figure 1 shows an artifact found by my father, Eric Hall, in South Africa (no site record available and therefore no site-based dating). 

The Acheulean Industry, spanning from around 1.5 Ma to 300 ka (Harris, et al, 2013) was characterized by artifacts that were produced typically from core rocks, from which flakes were knapped, usually bifacially. Acheulean artefacts are generally referred to as cutting tools. Handaxes were common artifacts, which may have had a similar manufacturing process as that for cores simply used as a source of flakes. In Africa, documented Acheulean sites are more numerous than the documented earlier Oldowan sites (ibid) and are generally located in open, more elevated, drier parts of the continent, with inclusion of a number of coastal beach sites. Several of the Acheulean sites, such as Olduvai in Kenya, West Turkana in Ethiopia, Sterkfontein and Swartkrans in South Africa, are coincident with Oldowan sites (ibid)– with Oldowan artifacts found in lower stratification at Olduvai and, at least at West Turkana, representive of Oldowan-Acheulean progression or overlap in hominin species and tool making (Proffitt and Martín-Ramos, 2019). Together, the Oldowan and Acheulean industries are usually considered as comprising the Early Stone Age.

The Middle Stone Age, spanning from approximately 300 ka to 50 ka, comprised many lithic sub-industries in South Africa and many sites show overlap with Acheulean production in the first half of the period (Wurz, 2013). Middle Stone Age artifacts trend towards smaller handaxes, blades and points than Acheulean, with a greater emphasis on points, notably bifacial points, at some sites. Several of the most studied sites are cave sites in the Cape coastal area, with names including Klasies River, Blombos and Diepkloof. Several subindustries in South Africa, especially in the later years of the Middle Stone Age, showed distinct differences from one another in lithic technology, which may indicate that throughout Africa transfer of methods from one community to the next was weak or occurred slowly.

Questions that arise relative to the Early Stone Age and Middle Stone Age are endless. To raise a few: Why did the progression of tool making technology take so incredibly long? To think of the number of generations of hominins using very similar technology is mind boggling. Perhaps technology advancement was limited by absence of creativity or the acute stress of survival due to food periodic shortages and disease. (The same question extends to Neanderthal and Homo Erectus stone technology in Europe and Asia, respectively.) Fast-forwarding to the Late Stone Age (next section), perhaps it was acceleration of creativity in the years since 50 ka that enabled or propelled Homo Sapiens to leave Africa bound for Europe and Asia? Perhaps an acceleration of creativity was accompanied language development?

A further question relates to locations of Early Stone Age sites in Africa. As delineated in maps published by Harris, et al (2013), the well-studied sites are clustered mainly around the Rift Valley in Tanzania, Kenya, and Ethiopia, as well as a few in southern Africa and in North Africa (Morroco and Algeria). Do those clusters occur where palaeoarchaeologists and their kin just happened to come across remains, perhaps in caves or areas where artifacts were not buried unfathomably in deep layers of soil? I think, in part the answer to this question is yes, but I am also attentive to the disease question I pose below. Did those clusters imply superior food sources, especially meat? Unlikely, I think, since a map hypothetically showing only abundance of potential meat supply, would include at least the moister plains. Did those clusters coincide with adequate water sources? I think the answer is yes, although again a map hypothetically showing mainly secure water supply would highlight the larger rivers and lakes. Did those clusters coincide with areas that were relatively free of malaria and other insect-carrying diseases? Since I believe most of those sites are relatively malaria free based on today’s climate, I think this may be a question worthy of exploration – even though any answers may not be conclusive.

Figure 2: South Africa: Oldowan Style Artifact

The Late Stone Age in South Africa spanned from approximately 50 ka (Stuart, 2011) to perhaps as late as 2 ka, with the arrival of iron age technology; a set of iron-age sites were dated to almost 2 ka by Vogel and Fuls (1999). The Late Stone Age is characterized generally by artifacts such as hafted microlithic tools, bored stones used as weights on digging sticks, bows and arrows, bones shaped as fishing equipment (Villa, et al, 2012). It is also characterized by artwork such as beads of shell and ostrich shell, and engraved decoration on bone and wood. However, Stuart (2011) points to evidence of year-round coastal fishing and hunting (e.g., tortoises and birds) on the southern Cape coast after 50 ka as indicative of population growth. Migration of Homo sapiens (humans) from Africa into Europe and Asia, which occurred in the Late Stone Age may have been enabled in part by human newly-acquired skills as exhibited by these tools, and perhaps by DNA change (ibid).

Stone-Age Workshops

Authors documenting discoveries of Acheulean and Middle Stone Age sites refer, in many cases, to finding an extraordinarily large numbers of stone artifacts. For example, Walker, et al (2013) documented an archaeological investigation in Kathu Townlands, in the northern Cape, South Africa, in which excavation of ten small test trenches up to 6 feet deep (to bedrock) yielded approximately 17 lbs. of artifacts per cubic foot of excavation. The artifacts are described by the authors as predominantly flakes made from ironstone sourced from nearby outcrops, identified as consistent with the Fauresmith subindustry (transitional between Acheulean and Middle Stone Age), of which the earliest findings are dated approximately 500 ka (Herries, 2011). Walker, et al (2013) refer to anecdotal reports by a public works foreman in the 1980s that local gravels dominated by stone artifacts were excavated and used for roadworks.

In the early 1970s at the edge of a quartzite hillside west of Northcliff, Johannesburg, South Africa, I came across trenches that had recently been excavated in a natural terraced area by developers laying water pipes for a new housing development. The total length of the trenches was in the order of 1,000 feet (not measured). Along each portion that I observed closely, there was an extraordinary number of quartzite flakes, packed most densely at 12-18 inches depth and less densely below 18 inches to the bottom of the trench (approximately 4 feet depth). In portions of the trenches, depth was limited by bedrock, which presumably called for subsequent blasting. The 12-18 inch depth of dense artifact material was consistent with the so-called “pebble marker” attributed in most African soil profiles to the action of moles, termites and other insects/worms in the root zone. The flakes I found were apparently each made by using single strike (no obvious attempt at edge or biface shaping). Their typical size was 1½ to 3 inches and their typical thickness was ½ to 1 inch. They were consistent with products of the Acheulean Industry but I did not attempt to determine their age. I observed no wear due to use of the flakes. They all had the appearance of being discarded. Unfortunately, to my chagrin I did not keep any of the flakes, took no photographs, and did not document my observations. Based on the length of the trenches and the density of flakes, I estimate that the terraced area may have held a volume of artifacts that would fill at least several eight-cubic-yard dump trucks. 

A map of the Northcliff location is provided in Figure 3. 

Unknown to me until I began to write this paper, excavation in 1962-64 by Beaumont (1969), at a Northcliff site approximately one mile north-east of the trenched location described above, produced approximately 400 stone artifacts, mainly quartzite, which the author described as “Chelles-Archeul type” and characterized as “very early stone age.” The author did not perform any stratigraphic, faunal or radio-isotope dating to narrow down the age range of the artifacts. The author’s determination of artifact age relied upon comparison of form against those in published findings from several other South African sites. The author estimated, based on his excavation pits, that the full two-acre site represented by his excavation would yield at least 100,000 artifacts. The author also stated (ibid) that “Earlier Stone Age and First Intermediate artifacts have been noted at Corriemoor.” (The trenched site I described above is at the southern edge of a suburb previously known as Corriemoor, adjacent to Northcliff, but I have no knowledge of precisely where in Corriemoor Beaumont refers to.) Although I did not keep any of the flakes from the trenched site described above, I have other quartzite artifacts from the Northcliff area found by me or my father, including ones shown in Figure 1 above and Figure 4.

Interestingly, Mason (ibid) also documents finding a large number of stone artifacts at a site (which he classified as primarily Middle Stone Age) referred to as “Aasvoëlkop.” Aasvoëlkop is the predecessor name of Northcliff ridge. Questions that arise about these stone-age workshops include whether they were of long duration, such as hundreds of thousands of years, or relatively short-lived. Soil profile aging would assist in answering this question. In both the Kathu Township and Northcliff trench sites, the depth of the soil profile containing artifacts suggests use of these sites was by no means transitory. Did the stone-age artisans locate their work in those sites due to the proximity of good source rock? Presumably yes, but there must have been other factors such as strategic co-location with animal prey. Strangely, Kathu Townlands is in a semi-arid area, where small and large game densities were likely always low. Although the quartzite ridges like Northcliff, even where undeveloped today, do not themselves have large densities of small and large game, the lower flatter landscapes to the north of these ridges may well have had such. Presumably, the artisans carried their selected tools (the “good” ones) to encampments closer to good water sources and good hunting locations.

Figure 3: Map of Northcliff Area near Johannesburg with adjacent Quartzite Ridges where many stone tools derive (Image from Garth Hall)

Figure 4 : Handaxe / core from the Northcliff area, scale same as Figure 1 above

Interestingly, Mason (ibid) also documents finding a large number of stone artifacts at a site (which he classified as primarily Middle Stone Age) referred to as “Aasvoëlkop.” Aasvoëlkop is the predecessor name of Northcliff ridge.

Documentation by Mason (1969) of a number of stone-age sites in the northern provinces of South Africa (previously known collectively as the Transvaal) indicate that at least several of those sites, e.g., the Wonderboom site (near modern day Pretoria/Tshwane), approximately 30 miles north of Johannesburg, have deep deposits of rubble containing a large number of stone artifacts. My qualitative observation informed by photographs in this publication that this (and other similar sites) were occupied for long periods consistent with accumulation of deposits up to 10 feet deep.

More broadly, what do our studies and understandings of stone tools tell us about the lives and cultures of the hominins who made and used the tools, especially about food? In asking this question, I am treading a path of query well worn by numerous respected paleoarchaeologists. In the case of the Northcliff environment, described above, the sheer mass of stone tools might suggest that hominin lives revolved around hunting, although some of the tools would have been helpful in digging tubers from the soil, for example. The question must address efficiency in securing food. In gathering food (as opposed to hunting), perhaps ancient activities can, for example, be compared in that environment with gathering by a modern-day troop of baboons. The gatherers work perhaps subconsciously but intelligently nonetheless against an energy budget. Most wild non-forested African environments today cannot yield enough food for a group of gatherers unless they cover a wide area, perhaps several square miles, each day. Stone digging tools must have helped, of course, but the limiting factor would have been availability of edible tubers and roots (as well as spatial density of berries and small creatures such as lizards, termites and locusts).

Numerous findings and exploration of stone tools from the Early and Middle Stone Age enable our following of a single thread, so to speak. The stones are all that is left in most cases, so we draw all possible inferences from them while being relatively starved of other tangible remains to enhance our understanding. Only in the Late Stone Age, as illustrated by Stuart (2011), do we find evidence in coastal caves of progression over time in the efficiency of hunter-gathering diet — such as the declining size of tortoise bones and new evidence of fishing. From this evidence, Stuart and others infer population growth and emigration of Homo sapiens out of Africa.


Beaumont, P., 1969. A Very Early Stone Age Site at Northcliff Extension 2, Johannesburg. South African Journal of Science65 (3), 65-71.

Harmand, S., Lewis, J., Feibel, C. et al., 2105. 3.3-million-year-old stone tools from Lomekwi 3, West Turkana, Kenya. Nature 521, 310–315.

Harris, J.W.K., Braun, D.R., Pante, M., 2013. Archaeological Records / 2.7 Myr-300 000 Years Ago in Africa.Encyclopedia of Quaternary Science (Second Edition), 59-66.

Herries, A.I., 2011. A Chronological Perspective on the Acheulian and its Transition to the Middle Stone Age in Southern Africa: The Question of the Fauresmith. Int J Evol Biol.; doi:10.4061/2011/961401 

Mason, R., 1969. Prehistory of the Transvaal, Witwatersrand University Press, Johannesburg.

McPherron, S., Alemseged, Z., Marean, C. et al., 2010. Evidence for stone-tool-assisted consumption of animal tissues before 3.39 million years ago at Dikika, Ethiopia. Nature, 466, 857–860.

Proffitt, T. and Martín-Ramos, C. (2019). Oldowan/Acheulean Succession at Olduvai Gorge, Wiley Online Library.

Stuart, R., 2011. Out of Africa & The Evolution of Human Behavior, does human behavior have a cultural or genetic heritage. The Forum At Poly, Cal Poly Social Sciences Department

Villa, P., Soriano, S., Tsanova, T., et al., 2012. Border Cave and the Beginning of the Later Stone Age in South Africa. Proceedings of the National Academy of Sciences109 (33) 13208-13213; DOI: 10.1073/pnas.1202629109

Vogel, J.C. and Fuls, 1999. A. Spatial Distribution of Radiocarbon Dates for the Iron Age in South Africa. South African Archaeological Bulletin 54, 97-101.

Walker, J.H., Lukich, and V. Chazan, M., 2014. Kathu Townlands: A High Density Earlier Stone Age Locality in the Interior of South Africa, Plos One

Wurz, S, 2013. Technological Trends in the Middle Stone Age of South Africa between MIS 7 and MIS 3. Current Anthropology, 54, Supplement 8, S305-S314.