The symbolic significance of these finds suggests Middle Stone Age people were behaving in a cognitively modern way and had the advantages of syntactical language at least 80,000 years ago.
Excavations carried out since 1991 at Blombos Cave provide snapshots of life in the Middle Stone Age (MSA) in the southern Cape, South Africa (Blombos Cave web site) Three phases of MSA occupation have been identified named M1, M2 and M3. Dating by the optically stimulated luminescence (OSL) and thermoluminescence (TL) methods have provided occupation dates for each phase: these are c. 73 ka for the M1 phase (OIS 5a/4), c. 80 ka for the M2 phase (OIS 5a) and c. 100 - 140 ka for the M3 phase (OIS 5d - 5e/6).
Occupations in each phase are relatively brief and the depth of deposit per layer is mostly less than 10 cm. Hiatus periods separate the Late Stone Age (LSA) deposits from those of the MSA, and almost sterile deposits between the M2 and M3 phases indicate a long period of non-occupation. This depositional history suggests that Blombos Cave was occupied sporadically and for relatively short periods of time. Artefacts considered unusual for the MSA were recovered from the M1 phase including bone tools, marine shell beads, engraved ochre, and bone tools from the M2 phase. In the M3 phase ochre is a dominant raw material and ochre processing tools were recovered.
Blombos Cave is some 100 m from the coast and 35 m above sea level. The cave is a wave-cut bench in Mio/Pliocene Wankoe Formation aeolian deposits. Interior cave deposits, including those in recesses, cover more than 80 m². About 20 m² of the MSA has been excavated to a depth of c. 2 m below the original surface. The depositional history of the MSA levels is complex. Probably just prior to the accumulation of the M3 phase large calcrete roof blocks up to 3 m thick were dislodged creating a variable and uneven floor surface. Further rockfall onto the MSA deposits occurred after the M3 phase occupation at c. 130 -140 ka.
Subsequent human occupation left debris scattered over and around these blocks up to a height of more than 2 m. Compaction has caused deposits to drape over and around large spalls with some examples of near vertical layering. Near the rear and side cave wall MSA deposits have in some instances shrunk leaving a gap that later filled with LSA deposits. In places large rocks have shifted or broken causing shearing of deposits and infills. Despite these anomalies most MSA deposits are in situ and undisturbed. By identifying and excluding material from contaminated areas archaeologists are confident of the integrity of more than 95% of recovered MSA material.
Principal markers of the M1 phase are bifacial foliate points, typical of the Still Bay, both complete and in various stages of manufacture. More than 400 have been recovered. Silcrete is the dominant raw material and the nearest source is c. 30 km. Large numbers of small flakes occur indicating on-site production of these artefacts. The site may even have served as a lithic workshop. More than sixty beads manufactured from Nassarius kraussianus gastropod shells have been recovered. Twenty seven of these beads may derive from a single personal ornament. Two chunks of ochre engraved with geometric patterns and more than fifteen bone tools come from the M1 phase. M2 phase markers are the more than twenty bone tools and a marked reduction in bifacial technology. In the M3 phase bifacial flaking and bone tools disappear. Silcrete is still dominant but there are fewer retouched tools. Striated ochre, particularly in large chunk form, is common in these levels. Ochre processing tools include lower and upper grindstones and hammerstones. Dense shellfish middens characterise the lower layers with very large hearths.
Faunal remains from the three MSA phases show that a wide range of terrestrial resources were exploited. More than a thousand fish bones, many from large fish, marine shells, seals and dolphins attest to extensive exploitation of aquatic resources and suggest exploitation patterns not dissimilar to that of LSA people in this region. Nine human teeth, mostly deciduous, have been recovered from the MSA levels but no other human skeletal material. The teeth probably derive from fairly gracile individuals and are similar to samples recovered at Klasies River Caves and De Kelders.
The M1 phase (oxygen isotope stage 5a/4) occupation occurs during a period of falling sea levels (c. 60 – 70 m below present sea levels and 10-25 km from present coastline) that is arguably colder than during M2. Donax serra, a sand burrowing white mussel occurs in the M1 phase suggesting beach conditions in front of the cave. Densities of shell are lowest in this phase (17•5 kg per m3) probably because of the distance of the coast from the cave. M2 phase occupations fall within OI 5a with sea levels c. 25 m lower than present and a coastline less than 5 km from the present shore. Intermediate densities of shell occur in M2 (31•8 kg per m3). Climatic conditions may have been temperate and warmer than during M1 occupations. The upper part of the M3 phase (CH/CI layers) is a high density shell midden (68•4 kg per m3) suggesting sea levels similar to the present. These upper levels arguably fall just after the Eemian (OI 5d; c. 100 ka. Temperatures were probably 1 to 2 °C warmer than present with higher sea levels (+3 m?). An OSL date of c. 143 ka (OI 6) for a low density occupation level (layer CJ) in M3 suggests the M3 phase may need further subdivision when more dates are available.
The symbolic significance of the marine shell beads, engraved ochre pieces, taken with the regular manufacture and use of bone tools, finely made bifacial points and the probable ability to fish suggests a cognitive behaviour package not previously associated with Middle Stone Age people. The capacity for these behaviours is likely to have evolved over a long period of time. A key question is why the particular innovations at Blombos Cave are seemingly not apparent at other MSA sites dating to the same time period or just later in South Africa or the rest of Africa?
One answer is that culture complexity and innovation is a heuristic strategy. The evolution of complex traditions does not necessarily drive the evolution of still more sophisticated imitations or traditions. The fitness of the innovations at Blombos Cave may not have been selected for during, say, the Howiesons Poort. The HP groups by c. 65 kyr had converged on novel ideas that, for them, had maximal fitness. Another is that sites with Still Bay elements in the Western Cape are rare, and at most organics are not preserved, and/or the sites were poorly excavated. That these innovations simply did not spread to the rest of Africa, or even southern Africa, is one possibility, alternately the evidence has yet to be recovered. We know little of demographic dynamics at c. 75 ka. A population crash at c. 72 ka, after the Mt. Toba eruptions, may have precipitated the loss of these traditions.
These and many other questions will need to be addressed before we understand the origins and evolution of modern behaviour. For example, what drove the Blombos innovations? Cultural traditions are capable of sensitive adaptations to ecological change and local environment. Ecological circumstances were probably conducive to population growth during the M2 phase at c. 80 ka but deteriorated in the southern hemisphere after c. 75 ka. Could demographic pressure on coastal resources, coupled with a retreating coastline, have been responsible for stimulating behavioural and cultural innovation? We believe that Blombos Cave may be one of the key sites for understanding the development of fully modern human behaviour during the Late Pleistocene in Africa.