The period between 300 and 30 ka was critical for the evolution of modern humans. Our mitochondrial DNA indicates that fully modern humans emerged around 200 ka. Projectile weapons that help define the epoch first appeared. Clear signs of self-adornment and symbolism also turn up during the Middle Palaeolithic. All of these developments took place in Africa, and the last two are reflections of the increased efforts by archaeologists in the continent from which we all originated. There is a long way to go to match the density of sites from which later periods in human history have been outlined in Europe, but progress is accelerating. One great hindrance has been dating sites, for the Middle Palaeolithic lies in a time zone where the Ar-Ar and ¹⁴C methods are ineffective. A developing chronological ‘workhorse’ for this difficult period depends on the way in which exposure of sand grains to sunlight ‘heals’ the defects in their molecular structure formed when radioactive isotopes in soils emit ionising radiation. Artificial illumination of sand grains containing these defects causes them to luminesce. The degree of luminescence is related to the time over which the defects have built up. Optical dating relies on grains having been exposed at the surface for a time to ‘reset’ the luminescence clock, and then being buried so that new defects can accumulate. Having lots of sunlight and a superabundance of bare sand, Australia has become a hotbed of research into optical dating of events associated with its peopling during the last ice age. Expertise developed there has been applied to many Middle Palaeolithic sites in Southern Africa (Jacobs, Z. et al. 2008. Ages for the Middle Stone Age of Southern Africa: Implications for human behaviour and dispersal. Science, v. 322, p. 733-735).

Archaeological work in South Africa and Namibia has revealed two distinct stone industries in the Middle Palaeolithic, both of which made hafted weapons that would have made hunting more efficient than the whatever weapons were used in earlier times – the most distinctive of the preceding Lower Palaeolithic tools was the bifacial hand axe, whose use is obscure. Both cultures involved the earliest recognisable ornamentation, such as shell beads and materials engraved with symbols, together with indirect evidence for the use of hematite and goethite pigments for body painting (see When and where ‘culture’ began in EPN of November 2007). Genetic evidence famously places modern human origins and their global migration out of Africa within this time frame. So, dating the archaeological sites as accurately as possible is a crucial importance, and a tremendous start has been made by the multinational team lead by Zenobia Jacobs of the University of Woolangong in Australia. Optical ages span 90 to 30 ka, with clusters between 71.9 to 71 ka and 64.8 to 59.5 ka, with a statistically significant gap of about 6.7 thousand years between them. When compared with climatic-change indicators from the Antarctic ice record the developmental episodes do not seem to correlate clearly with any specific warm of cool periods, though the earlier spans the time of the Toba super-eruption in Indonesia and the later one was a period of warming. So any environmental cause for the technological and cultural changes is unclear. However, both fall within the estimated time span of the genetic ‘bottleneck’ between 80 and 60 ka, and the most likely times for the initial ‘Out of Africa’ migrations, probably across the Straits of Bab el Mandab linking Eritrea and Arabia across the Red Sea shallowed by ice-cap linked falls in global sea level.

Childhood and families

Human females are unlikely to break 10 seconds for the 100 metres because of their sashaying gait. It can’t be helped, being due to the evolution of the pelvic girdle of bipedal females to deal with birthing of infants with increasingly large heads. Supposedly, the human female pelvis is now close to the limit that will permit walking on two legs. Such problems do not plague other living primates partly because their young have small heads relative to their bulk, and pelvic anatomy is not constrained by an habitually upright gait. It seems not to have been an ‘issue’ for australopithecines either: they did not possess ‘child-bearing hips’. The intermediate species, Homo erectus, despite having a 1 Ma fossil record (maybe as long as 1.8 Ma for the Asian form) only recently provided substantial pelvic remains (Simpson, S.W. et al. 2008. A female Homo erectus pelvis from Gona, Ethiopia. Science, v. 322, p. 1088-11092). In the words of the authors, this pelvis is ‘obstetrically capacious’ and demonstrates that female skeletal evolution responded to increasing foetal brain size: it would have permitted infants with heads 30 to 50% of the adult size to have been born. Homo erectus has been widely supposed to have had a tall willowy frame analogous to that of fully modern human inhabitants of tropical savannahs, yet the Gona woman was stocky. So, environmental influences seem to have had less of an evolutionary role than the advantages of greater brain development before birth. That places H. erectus even more firmly on the human line; indeed greater in utero brain development seems to have taken place than in modern humans.

The Gona pelvis demands re-evaluation of how foetal and childhood development has progressed over the last two million years (Gibbons, A. 2008. The birth of childhood. Science, v. 322, p. 1040-1043), the unique attributes having appeared during the evolution of our own genus. Among chimpanzees, infants can fend for themselves, with a little help from elders, after 3 years old. Street children from Asia and South America need to be 6 before they can survive without parental care. Growth lines on teeth that appear week by week reveal that previous age estimates for a number of immature australopithecines whose first adult molars had erupted were large overestimates: instead of 6 they point to 4 years old. Another signal feature of human development is the lengthy period to full development (marked by the eruption of the 3^rd molar as well as the end of significant growth in stature). The average age when human child bearing begins is around 19, while chimpanzees start at about 11. A fresh examination of the famous Turkana Boy’s skeleton, an H. erectus, that uses tooth microstructure reduces his age at death from 13 to 8, suggesting an earlier onset of independence than in modern children. He grew much more quickly too, and would have reached adulthood somewhat earlier: around 14.5 years old. The picture with Neanderthals is not completely clear, some tooth studies suggest that their children grew significantly more quickly than modern ones, other studies point to the same rates or even longer development if adult brain sizes of Neanderthals are taken into account (larger on average than those of modern humans). Using average life expectancy of gatherer-hunter humans and chimps who survive dependent childhood – 45 and 70 years respectively – along with evidence for child development, suggests that australopithecines could have reached 45 while H. erectus adults could have expected to reach 60 years old.

There are other differences that begin to slot into space with the new data. Both human and chimpanzee females have a similar child-bearing period of around 20-25 years. The difference is that, on average, the natural interval between births is about half as long for human mothers as for chimpanzees. The greater number of human offspring gives a greater chance of the survival of some to reproduce themselves. On the other hand, slower child development places a greater burden on mothers, even after weaning. So there is quite a contradiction between the evolutionary effects, if only child-mother relationships are taken into account. This contradiction was resolved, to some extent, by a seminal paper in the late 20^th century by a group of anthropologists from the Universities of Utah and California (see O’Connell, J.F., Hawkes, K. & Blurton Jones, N.G. 1999. Grandmothering and the evolution of Homo erectus. Journal of Human Evolution, v. 36, p. 461-485). They focussed on the potentialities of the early onset of infertility or the menopause among women relative to its appearance among female chimpanzees, which gives, on average, a 30 year non-child-bearing period to older women. This approximately coincides not only with child-rearing periods for their daughters, but for their granddaughters as well. The ‘grandmothering’ hypothesis for human development centres on the great evolutionary advantages of post menopausal women assisting with child rearing. O’Connell et al. suggested that this arose among H. erectus, as far back as 1.8 Ma, and the Gona pelvis together with other new views of H. erectus development add considerable weight to that concept. As well as freeing younger women for food gathering, the cultural significance of older women caring for children adds another dimension that may link to the advantages of delayed post-weaning development that we see today, albeit in many annoying contexts!