Speaker:         Prof Judith Sealy

The many caves and open-air archaeological sites on the Robberg Peninsula have long been known to contain a wealth of archaeological material, some of it many hundreds of thousands of years old. Plett and surrounding areas were desirable places to live in those days, just as they are today, and for many of the same reasons – including mild weather year-round and good fishing!  The best-known cave is Nelson Bay Cave, excavated by archaeological teams in the 1960s and 1970s. It contains perhaps the most complete record of any site in South Africa of occupation over the past 22000 or so years, as well as older remains dating to before 60000 years ago.

 Nelson Bay Cave provides an excellent picture of life in the past, but questions remain. One important question is “Are the patterns seen at Nelson Bay Cave typical of other sites in the area, or are some of them idiosyncratic features of this particular cave?”

This can only be answered by investigating more caves. We do, of course, have a good deal of evidence from Matjes River Rock Shelter, at Keurboomstrand, but the excavations there were not well documented and we lack detailed information about what was found, and where. We also know, from the bone chemistry of human skeletons found at Matjes River and on the Robberg Peninsula, that between 4500 and 2000 years ago, different groups of people with different ways of life lived on either side of the Keurbooms/Bitou estuary. They were all hunter-gatherers, but the folk who lived at Robberg and Plettenberg Bay were specialized marine-oriented hunter-gatherers for whom fish, seals and seabirds were major food items. People who lived at Matjes River, on the other hand, ate much more mixed diets in which plant foods and the meat of terrestrial animals were more important. This is a relatively recent finding, published in the academic journal Current Anthropology in 2006.

In order to explore these issues further, a team from UCT undertook excavations in June/July 2007 and 2008 in the very large cave on the western side of Robberg, just beyond The Gap. Many locals know this cave as East Guanogat. Archaeologists often call it Hoffman’s Cave, after Hoffman, a former Director of the National Museum in Bloemfontein, who dug a trench there in (we think) 1958. Hoffman himself called it “Robberg Cave”, but as there are many caves on Robberg, this isn’t a very helpful name. It is, as everyone who has walked around Robberg knows, an enormous cave with a very large shell midden spilling out of the mouth. The path around the peninsula bypasses the cave immediately below the mouth. The surface inside the cave is very soft, so our first problem was how to move about in the cave without causing damage. With a great deal of help from the staff of CapeNature, we filled sandbags with beach sand and laid these down, together with coir mats, to make pathways across the site. We decided to work back from the western side of the trench left by Hoffman, cleaning the surface, identifying the different depositional layers, then removing them one by one using small mason’s trowels. All excavated material was sieved and the sand used to fill more sandbags, while shells, bones, stone artefacts and anything else we found were placed in bags for further study at the university.

We found a lot! The midden is packed full of shells, fish bone, estuarine grass (Zostera sp.) brought into the site for use as mattresses and cushions, as well as rarer finds such as mammal bone, stone artefacts, bone and ostrich egg-shell beads and other items. Some, but not all of the finds have now been studied. Shellfish were a major food for the inhabitants of the cave, and Katharine Kyriacou has identified the discarded shells for her masters thesis. She found that brown mussels were the most common shell species, with pear-shaped limpets second. This is interesting, since both these species are found today on the steep rocky slopes immediately beneath the cave. Siffies and alikreukel were also collected and eaten regularly. Karen van Niekerk has studied the fish bones for her doctoral thesis. She found that the inhabitants of the cave caught 29 different kinds of fish, of which the most common were yellowtail, galjoen, blacktail, harders and strepies. Species such as yellowtail and galjoen were almost certainly line-caught, while harders were probably taken with nets.

We also found stone artefacts, and items fashioned from bone, shell and even ivory. Stone artefacts of this period consist mostly of crude flakes used as rough cutting tools. A great deal of work was, however, devoted to manufacturing finely crafted objects in other materials. We found perfectly symmetrical bone points that were probably used as arrow-tips, and bone beads made by ringing and snapping bird-bones, then grinding the snapped edges smooth. We also found parts of one or more bowls made from the shells of fresh-water turtles. The upper parts of the shells were kept intact, and the undersides removed and the edges smoothed to produce beautiful and useful bowls. As in other coastal sites of this period, we found surprisingly little evidence of fishing equipment. We did recover fragments of twine that may have been parts of nets, and small stones with a groove all the way round them that were probably used as line sinkers. These are known also from Nelson Bay Cave and other south coast sites. We do not, however, know what these ancient fishermen used as hooks: nothing resembling a hook has been found at this site or any other dating from this time. Presumably they were made of some perishable material that has not survived.

It is remarkable that the entire 1.7-metre depth of midden that we excavated accumulated in only 1000 years, between about 4500 and 3500 years ago. Such a thick deposit is unusual, and offers a wonderful opportunity to get a detailed picture of changes through time. This period is also well represented at other sites in the southern Cape, from which we can infer that populations were large and population densities high at this time. We do not yet know why this was, but I suspect that the way that people were dividing up the landscape, as shown by the separation between groups on either side of the Keurbooms/Bitou estuary, was a response to population pressure and presumably pressure on food and other resources. The changes that this brought about within hunter-gatherer society, and the ways that they coped with the challenges of rapid population growth are of considerable interest, and resonate with some of the problems we face in our own society today.

Why did we not find anything older than 4500 years? At the base of our excavations, we encountered a sand-dune composed of very fine, wind-blown sand. Zenobia Jacobs, of the University of Wollongong in Australia, has dated this for us to about 7000 years ago. At that time the sea level was a little higher than today, and the shape of the shoreline was therefore slightly different. I think that the dune we see in the cave is the same one that occurs today on the side of Robberg between the cave and the ‘island’, but that it was displaced slightly towards the mainland during the time of higher sea level. It is quite possible that there is older archaeological material underneath the sand-dune, but it would be very difficult to reach it. The sand is very fine, so as soon as it is exposed it dries out and then starts to ‘run’. We do not know how deep the dune is, but it is certainly more than a metre thick – possibly several metres – so any serious attempt to dig through it would require substantial boards or plates securely fixed in place to prevent the sides of the trench falling in.

It is also interesting that the site lacks deposits more recent than 3500 years ago. Perhaps there was once more recent material on top, but this has been eroded away. Alternatively, by 3500 BC the cave may already have been so full of other people’s rubbish that it was no longer an attractive place to live, and other camp-sites were chosen instead. Nelson Bay Cave has a number of layers dating between 3500 and 2000 years ago, so people were living on Robberg at that time. After 2000 years ago, there is much less evidence of occupation. In fact, all along the southern and western coastlines of South Africa, people harvested marine foods much less intensively after 2000 years ago. This was just part of the disruption of old ways of life as the first domesticated sheep and cattle were brought into South Africa, and people turned from hunting and gathering to sheep- and later cattle-herding. Robberg is clearly not an ideal environment for these activities, and early herders would have gravitated towards grassier environments. 

Evidence from this site is making an important contribution to understanding the long-term history of South Africa. The hunter-gatherer communities who have lived in South Africa for many thousands of years have a dynamic history that is of considerable interest, although it is not yet well known. It is especially important for archaeology to document the variation in hunter-gatherer lifestyles, because it is becoming increasingly clear that there are many ways to be a hunter-gatherer. What determines which path a particular community follows? Is it to do with climatic and environmental conditions? The demands of increasing or decreasing population size? Cultural choices? As we go further back in time, archaeological remains tend to be less well-preserved and therefore more difficult to interpret. Along the south coast, we have some sites such as Pinnacle Point, Blombos, Klasies River and others that contain evidence of early modern humans who were, of course, also hunter-gatherers. Reliable interpretations of sites like these depend crucially on understanding the range of options for how to make a living as a hunter-gatherer in this environment.