Universiteit Leiden

nl en

Research project

Finding resolution for the Middle to Later Stone Age transition in South Africa

This project investigates the causes of the major archaeological change in the period of 40.000-20.000 BC in South Africa.

2017 - 2024
Gerrit Dusseldorp
NWO Vidi grant NWO Vidi grant

The development of modern human behaviour is best documented in the South African archaeological record. Early examples of complex behaviours appear during the Middle Stone Age (~300.000 - 20.000 years ago). However, they are present only intermittently. From 40.000 a major change in the stone artefacts occurs. This heralds the beginning of the Later Stone Age in which modern behaviour is assumed to be consistently present. The “Finding resolution for the Middle to Later Stone Age transition” research team investigates the causes of the major archaeological change.

The site of Shongweni, South Africa. At Shongweni, excavations focused on the Iron Age occupations of two rockshelters yielded indications for the presence of earlier Later Stone Age deposits. The team aims to re-open excavations at this site to clarify the character of these deposits.

Unfortunately, the transitional period (~40 000 – 20 000 years ago) represents a gap in our knowledge of human development in South Africa. This lacuna obscures pivotal developments from view. First, it falls in a period of globally unstable climates in the run-up to the most severe part of the last Ice Age. Second, the development of the gracile anatomy characteristic of historically known South African populations likely coincides with this period. Third a major technological reorganization takes place, encompassing a shift from prepared core technologies (also used by archaic hominins like Neanderthals) to miniaturized toolkits. This technological change was likely accompanied by changes in subsistence economy and social organization.

To remedy the hiatus in our understanding of the final stage of the development of modern behaviour in the region, we are undertaking a major new research effort. The effort focuses on renewed fieldwork at key sites that document the transitional period. The project team will contribute new chronological, technological and environmental data. This data will be used to test the hypothesis that the development of the microlithic industries characteristic of the Later Stone Age developed as a response to the increasingly cold climates between 40.000 and 20.000 years ago. The issue of the interrelationship between technological change, environmental change and subsistence has been intensively studied by the PI for the Middle Stone Age. The current project greatly expands these analyses.

The site of Umhlatuzana, South Africa.


The team will conduct re-newed excavations at key locations in KwaZulu-Natal: Umhlatuzana and Shongweni. Both were excavated earlier, but these excavations did not focus on the period between 40 000 and 20 000 years ago. The renewed excavations will specifically concentrate on the deposits and employ a range of state-of-the-art analytical techniques to the materials.

Umhlatuzana was subject of rescue excavations in the 1980s related to the development of a motorway next to the site. As the site has seen vandalism in recent years and there are plans to expand the motorway, new research here is imperative. Earlier research as focused mainly on two important phases from the Middle Stone Age represented at the site. However, a transitional phase between the Middle and Later Stone Age was reported. This phase will be the focus of our fieldwork.

At Shongweni, excavations focused on the Iron Age occupations of two rockshelters yielded indications for the presence of earlier Later Stone Age deposits. The team aims to re-open excavations at this site to clarify the character of these deposits.

Micromorphological analysis

The stratigraphy will be studied in microscopic detail to better understand how the deposits were formed and under which environmental conditions. The analysis can clarify how quickly the deposits were formed and if there was any mixing of remains from different phases. The analysis will also be employed to study the use of structures like hearths and bedding by the site’s occupants.


Two different dating techniques will be applied to the sites. First, AMS radiocarbon dating. This high-precision method was unavailable at the time of the original excavations. We will use it to refine the conventional radiocarbon chronologies that were originally obtained.

Second, we use Optically Stimulated Luminescence (OSL). As radiocarbon dating can only be applied to deposits of younger than 50 000 years, most archaeological sites from the Middle Stone Age have been dated using OSL. This technique relies on the amount of the earth’s background radioactivity trapped in the archaeological sediments. By applying this method to the deposits we ensure chronological comparability within the existing Middle Stone Age framework.

Environmental analysis

We use geochemical methods to determine the character of the vegetation around the sites. These methods rely on the presence of different isotopes of the same chemical elements. Different types of plants use these atoms in different proportions and by determining how the proportions of different carbon, oxygen and hydrogen atoms change through the sequence, we can track changes in the local climate.

We will also study microscopic plant remains to determine the exact species that were brought into the cave. Not only can this give us more detailed information on the environment, it can also give insight into things like the diet that people ate!

Archaeological analysis

Stone tools of the earliest Later Stone Age are only known from a small number of sites and few of those have been comprehensively studied. We do not know if the development of the Later Stone Age was similar across the region, or if different developments took place in different areas. The team will conduct a technological analysis of the stone tools of Umhlatuza and of Shongweni. This will clarify the technological character of the reorganization in stone tool manufacture, which is the first step into unravelling the causes of this major change.

Animal bones, plant remains and molluscs have been reported from earlier excavations, but not been systematically studied. By determining which animal species were brought into the sites and by studying charcoal and other remains, crucial details on the way of life of people living in South Africa during the coldest part of the last Ice Age will be provided.

Reidsma F.H., Sifogeorgaki I., Dinckal A., Huisman H., Sier M.J., Os B. van & Dusseldorp G.L. (2021), Making the invisible stratigraphy visible: a grid-based, multi-proxy geoarchaeological study of Umhlatuzana Rockshelter, South Africa, Frontiers in Earth Science 9.

Sifogeorgakis E., Klinkenberg M.V., Esteban I., Murungi M., Carr A.S., Brink V.B. van den & Dusseldorp G.L. (2020), New excavations at Umhlatuzana Rockshelter, KwaZulu-Natal, South Africa: a stratigraphic and taphonomic evaluation, African Archaeological Review 37(4): 551-578.

Reidsma F.H., Sifogeorgaki I, Dinckal A., Huisman D.J, Sier M.J., Os B. van & Dusseldorp G.L. (17 mei 2021), Integrating geoarchaeological techniques to reveal the invisible stratigraphy at Umhlatuzana rockshelter, South Africa: a grid-based approach. Developing International Geoarchaeology 17 5 2021-19 5 2021 [conference poster].

Dusseldorp G.L. Sifogeorgaki I. (2020), Excavations at Umhlatuzana rockshelter, Southern African Society for Quaternary Research Newsletter.

Sifogeorgakis I. & Dusseldorp G.L. (19 september 2019), Revisiting Umhlatuzana Rock Shelter, South Africa: First geoarchaeological results. ESHE 19 9 2019-21 9 2019 [conference poster]. 

Sifogeorgakis I. & Dusseldorp G.L. (2019), Revisiting Umhlatuzana rockshelter, KwaZulu-Natal: First results and perspectives of a geoarchaeological study. ASAPA conference 3 7 2019-5 7 2019 [conference poster].

Sifogeorgakis I. & Dusseldorp G.L. (2019), Stratigraphic assessment of Umhlatuzana Rock Shelter, South Africa. Developing International Geoarchaeology 2019 17 6 2019-21 6 2019 [conference poster].

The project’s results will be published open access in the scientific literature, so anyone can read the results of our research.

We also try to put out regular updates in the form of weblog and social media posts. Blog posts in Dutch will be published on the plaform Wetenschap.nu and announced on the @Keilmesser twitter account.

Published so far (in Dutch):

This website uses cookies.  More information.