Lone Survivors

by Chris Stringer

  This approach gave rise to a large collaborative project called RESET (Response of Humans to Abrupt Environmental Transitions), in which I am involved. Over a five-year research period, RESET is correlating tephras from their volcanic sources to where they fell in deep ocean and lake sediments, and even farther into important archaeological sites in Europe, western Asia, and North Africa. The aim of RESET is to investigate the effects of climate and environmental changes on the human populations of the region, including the last Neanderthals and the first moderns. The tephras themselves are markers of volcanic eruptions, of course, most of which were only local and short-lived in their effects, but a few did have major—even global—impacts, as we will see later in this chapter.

  Using volcanic deposits to date human fossils has a long history, as I explained in relation to Olduvai Gorge earlier, and the mapping of outputs from successive volcanic eruptions has played an important role in refining the age of many important fossil sites in East Africa, including Omo Kibish in Ethiopia. The two most complete human fossils from there, the Omo 1 skeleton and the Omo 2 braincase, were found in 1967 by a team led by Richard Leakey and were important in early proposals for a recent African origin. But although there were initial estimates that the material was over 100,000 years old, some of these were based on the application of uranium-series dating to shells in the deposits—not the most reliable material for such determinations—and so doubts remained. Over thirty years after the original discoveries, an international team led by the anthropologist John Fleagle returned to the Kibish region, relocated the 1967 find-spots, and found further fossils and stone tools. Both Omo 1 and Omo 2 were originally recovered from the lowermost portion of the massive Kibish Formation, a series of annual but episodic sediments laid down by the ancient Omo River when it periodically flooded, before it entered Lake Turkana. These deposits lie about one hundred kilometers farther north than its present delta, close to the Ethiopian border with Kenya. Occasionally, volcanic eruptions deposited volcanic ash and pumice over the river and lake sediments, and these can be dated through their contained argon. A layer of ash about three meters below the location of Omo 1 was placed at about 196,000 years old, while a second ash about fifty meters above the location was dated to about 104,000 years. Because there were also clear signs of geologic erosion (the removal of sediments when the river and lake level fell) between the level of Omo 1 and the higher ash, it seemed likely that the age of Omo 1 was much closer to the age of the 196,000-year ash than to the 104,000-year one.

  Additional indirect support for this came from much farther afield, on the seabed of the Mediterranean. During ancient monsoon periods, rain and snowmelt in the Ethiopian highlands sent annual floods pouring into the sources of the River Nile, causing sapropels (dark layers of sediment) to be deposited when these waters eventually flowed out into the Mediterranean. A particularly strongly marked sapropel can be dated from its position in Mediterranean seabed cores to about 195,000 years, suggesting that it correlates perfectly with the major monsoon event that sent floods in the opposite direction down the Omo River, producing the vast deposits of the lower part of the Kibish Formation, in which the Omo 1 skeleton and the underlying volcanic ash were found. The Omo 2 braincase was a surface find rather than a fossil excavated from sediments (which was the case for the Omo 1 partial skeleton), but the surrounding location consisted of the lowermost part of the Kibish Formation. Thus the team that revisited the region and published the new dating work remains confident that Omo 1 and Omo 2 are very close in age, at about 195,000 years, despite some strong contrasts in their level of modernity—something to which I will return in chapter 9.

  Another instance in which Mediterranean sapropels provided clues about events deep within the African continent concerns the “greening” of the Sahara about 120,000 years ago. Today, the Sahara is the largest hyperarid region on Earth, with an annual recorded rainfall as low as one millimeter across much of its vast extent. But as is well known from archaeological finds and rock paintings of animals and people deep within the desert, only 6,000 years ago the Sahara was a wetter place of grasslands, lakes, and gallery forests, fringing extensive river systems. What is less well known is that 120,000 years ago the Sahara was even wetter than that and was able to support a widespread population of Middle Paleolithic hunters and gatherers. There are many clues to this, including chemical signs of freshwater deposition and dark sapropels in deep-sea cores off the coast of Libya, both of which indicate powerful river sediments during the last interglacial. A similar signal was detected from plant-derived chemicals in dust deposits off the West African coast; analysis shows that the level of water-dependent plants (such as trees rather than grasses) peaked around 115,000 years ago, with a briefer second peak from about 50,000 years.

  In addition, as revealed in radar images from satellites, huge river channels now lie buried beneath desert sands, some of which are five kilometers wide and run for eight hundred kilometers. As the earth scientist Nick Drake and his colleagues showed, around 120,000 years ago the desert was covered by an interconnecting network of rivers and lakes, forming humid corridors that stretched from massive southern lakes such as Fazzan and Chad all the way to the Mediterranean. These corridors allowed typical African fauna and flora to flourish for at least 20,000 years, and as plant resources and game proliferated, so did the humans who lived off them. For the last hundred years, travelers and archaeologists have collected Middle Paleolithic tools from the surface of the Sahara, often well away from any modern oases, and we now know that many of those accumulations date from the green Sahara over 100,000 years ago.

  The tools include triangular stone points with a tang or shoulder, which was presumably used to mount them as projectiles on a wooden handle. These iconic artifacts characterize the Aterian industry, first recognized at the Algerian site of Bir el-Ater; this industry was made by a very robust and large-toothed variety of early Homo sapiens, comparable to the people we know from Herto in Ethiopia. It seems very likely that the increased humidity of many parts of Africa at this time led to population growth and an important sharing of ideas across Africa, as formerly isolated regions became connected by habitable corridors. Cultures that used shell beads and red ocher pigment to signal to each other seem to have spread across the whole known range of early modern humans at that time, from South Africa to Morocco and even into adjoining western Asia, at Skhul and Qafzeh. In Israel, elevated rainfalls produced the huge Lake Samra, which extended far beyond the now shrunken Dead Sea basin, about 75,000 to 135,000 years ago.

  But as the beginnings of the last Ice Age began to grip, these balmy interglacial conditions were not to last in Africa. We can trace the effect of this climatic downturn on the peoples of southern Africa from about 75,000 years ago through two important and innovative stone tool industries: those at Still Bay and Howiesons Poort. As well as sophisticated stone artifacts (heat-treated to improve flaking qualities in the case of the Still Bay), both possessed beads made of seashells or ostrich eggshells, and both used red ocher symbolically. The Still Bay industry is known from only a handful of spots in southern Africa, while the Howiesons Poort was much more widespread, with at least thirty sites ranging from the best-known locations on the southern coasts at places like Klasies River Mouth Caves, to the edges of the Namib Desert and the mountains of Lesotho. While it was thought that the Still Bay preceded the Howiesons Poort, they both lay well beyond the limit of effective radiocarbon dating, so methods like uranium series and ESR had been employed to place them in relation to each other, but with a poor fix on their respective durations.

  The breakthrough came in 2008 when a team of dating specialists including Zenobia Jacobs and Bert Roberts combined with archaeologists such as Hilary Deacon and Lyn Wadley to apply the latest luminescence dating techniques to single grains of quartz from the sites, using the same laboratory procedures throughout. Fifty-four samples were obtained from widely dispersed sites that contained either both industries or
one or the other. The results were striking: rather than spanning periods of 50,000 years or more, as some other methods had suggested, both the Still Bay and the Howiesons Poort industries were rather brief cultural episodes that seemingly appeared and vanished quite suddenly over large areas of southern Africa. Remarkably, the Still Bay lasted only a few millennia, around 72,000 years ago, while the Howiesons Poort appeared around 65,000 years ago, ending abruptly at about 60,000 years. Moreover, the people who succeeded the Howiesons Poort only returned after a gap of a few thousand years and were making more conservative Middle Stone Age tools (comparable to the Middle Paleolithic of western Eurasia), apparently without the innovations of their predecessors.

  It is, of course, possible that the manufacturers of the previous industries did not disappear but simply moved to locations that have not so far provided any archaeological records. (For example, they might have relocated farther out on the coastal shelves, which are now submerged.) But they do not seem to have reappeared even at a later date, suggesting that these really were brief episodes, like a light turning on and then being extinguished, perhaps forever. Environmental deterioration in the face of rapid climate change has been invoked to explain these episodic patterns, and I will return to their significance in chapter 8. Meanwhile, I would like to examine a global event that has controversially been claimed to lie behind even wider changes in human populations and behaviors, including the innovations of the Still Bay industry: the eruption of the Toba volcano in Sumatra.

  About 73,000 years ago, the large island of Sumatra in Indonesia was the source of the most powerful volcanic eruption of the last 100,000 (some calculations suggest 2 million) years. The eruption was about a thousand times larger than the famous Mount St. Helens in Washington State in 1980, and it expelled the equivalent of about 1,000 cubic kilometers of rock in the form of ejecta of many different sizes, as well as huge volumes of water vapor and gases. Thick ash deposits from the eruption were found in cores stretching from the Arabian to the South China seas, and some archaeological sequences in India are interrupted by ash falls several meters thick. The undoubted scale of the eruption led to some sensational claims about its effects on the Earth through a resultant “volcanic winter,” where the whole planet would have lacked summers for many years, as a result of clouds of dust and droplets of sulphuric acid residing in the upper atmosphere. The resultant drop in temperature and the lack of summer sun would have devastated plant growth and everything that relied on it, including the early human populations of the time. Some suggested it destabilized the Earth’s climate for a thousand years, or that it even triggered a global ice age, shrinking human numbers to only a few thousand people. On the other hand, studies of faunas in southeast Asia, closest to the eruption, suggested that any effect was minor and short-lived, since they were not devastated. Moreover in India, archaeological sequences studied by Mike Petraglia and his colleagues similarly indicate that the impact on human populations was not severe. I have been very cautious about Toba’s effects on humans globally. (After all, the Neanderthals living in temperate Europe and the Hobbit living in Flores in Indonesia, as well as our ancestors in Africa, certainly survived the effects of Toba somehow.)

  However, two recent studies by Alan Robock and his colleagues and Claudia Timmreck and her colleagues, using different models of its effects around the globe, do point to a severe, if shorter-lived, impact. Their work did not back up the idea that it could have triggered a glacial advance, but did conclude that it could have produced up to a decade of cold, dry, and dark conditions, serious enough to affect plant and animal life on land and sea, but perhaps not totally devastating. On the other hand, new analyses of land sediments and pollen in a core from the Bay of Bengal, involving the leading proponent of the Toba effect on early humans, Stanley Ambrose, did find signs of a long period of desiccation in India at the time of the Toba eruption. Unfortunately for simple scenarios, the apparent extinction of the South African Still Bay people after a short florescence came about 2,000 years after Toba, if the present dating evidence is accurate, although some might argue that their innovations were an outcome forced by the environmental degradation that the eruption brought about.

  Now let’s move on about 35,000 years to the environments and time scales around the time of the extinction of the last Neanderthals. Of course, if the Neanderthals passed on their genes to modern humans, as we will discuss in chapter 7, they did not become entirely extinct, since some of their DNA lives on within us. Nevertheless, as a population with their own distinctive bodily characteristics they vanished, and there are many scenarios constructed around what may have happened to them. Explanations have ranged widely from suggestions of imported diseases to which they had little natural immunity, through to economic competition from, or even conflict with, early modern humans. Until recently the view of climate change in Europe at this time was rather simplistic, leading to climate being ignored as a factor in Neanderthal extinction: they became extinct before the peak of the last Ice Age, they had survived cold conditions before, and they were adapted physically, and probably culturally, to cope with climatic downturns. Many explanations (including my own) had instead focused on the direct impact of modern humans on the Neanderthals, and the inherent superiority of people like the Cro-Magnons. But rich paleoclimatic records from cores in ice caps, the sea floor, and lake beds now reveal a startling complexity in climatic change at this time, with many rapid oscillations. This led to new ideas about their extinction, including those of two of my friends, Clive Finlayson and John Stewart, who consider that the Neanderthals were probably on the way to extinction anyway, and that moderns had little or nothing to do with their demise. For example, Clive thinks that early moderns had honed their adaptations on the plains of Africa, a very different environment from that to which the Neanderthals were adapted in Europe; hence the two species had different ecological preferences and never really overlapped, competed with each other, or interbred. According to this view, the Neanderthals faded away about 30,000 years ago as their preferred mixed habitats finally vanished from their last outposts, in places like Gibraltar.

  In my own case, around the year 2000, I took part in a collaboration called the Stage 3 Project (Marine Isotope Stage 3 lasted from about 30,000 to 60,000 years ago), led by Tjeerd van Andel and based in Cambridge. We used fluctuations in temperature recorded in a Greenland ice core and a lake core in central Italy to reconstruct hypothetical “stress curves” for Europe, based on two factors of equal weight, both of which were assumed to be bad for humans, whether Neanderthal or Cro-Magnon: low temperatures and rapid destabilizing fluctuations in temperatures in either direction—higher or lower. The approach was simplistic in that it did not attempt to model other factors such as changes in rainfall, snow, and wind chill, which would also have had important impacts on the human population of Europe and their survival prospects. The stress curves we generated showed a mild climatic phase around 45,000 years ago, which perhaps correlated with the migration of moderns into Europe, but the “climatic stress” peaked at around 30,000 years, rather than the subsequent glacial maximum, coinciding with the last known records of the Neanderthals or their stone tool industries in places like Gibraltar and the Crimea. Such stressful conditions would no doubt have affected both Neanderthal and Cro-Magnon populations, sharpening the competition for diminishing resources in environments where their ranges overlapped. But only Homo sapiens came through those crises.

  A rather more sophisticated modeling of conditions in Europe when the last Neanderthals overlapped with the Cro-Magnons was published in 2008. In this work William Banks and his colleagues used the location of stone tool assemblages dated between about 37,000 and 42,000 years, which were thought to identify the presence of Neanderthals or Cro-Magnons in particular regions. Then, treating the distributions as though they represented a species of mammal rather than stone tools, they used ecological modeling to reconstruct the environmental preferences and tolerances of the
two populations, and the ranges they each should have been able to occupy at the time, according to those preferences. The time span covered two mild phases interrupted by a short but severe cold snap at about 39,000 years. This was not when the Campanian Ignimbrite was deposited farther to the east, but when the Atlantic was chilled for several hundred years by the southward flow of an armada of icebergs (a Heinrich event, discussed further in chapter 4).

  The results showed that, before the cold snap, the Neanderthals should have been widely distributed, and indeed they were. During the Heinrich event, both populations shrank in their modeled and actual distributions, in the face of the environmental deterioration. But when conditions ameliorated at about 38,000 years, although the warmer and wetter conditions should have encouraged both populations, the moderns bounced back while the Neanderthals did not. The modeling also showed that the Neanderthal and Cro-Magnon populations were attempting to exploit similar ecological niches; in practice the moderns increased the breadth of theirs through time, at the expense of the Neanderthals. While in the earlier phases the modern niche did not include central and southern Iberia, as time went on the Cro-Magnons increasingly expanded southward toward outposts of Neanderthal survival such as Gibraltar.

  This interesting work shows how such modeling can be done, and it should be possible to further refine the analyses, as ultrafiltered radiocarbon dates and the use of correlation tools like microtephra become increasingly available. A more direct attempt to estimate the relative population sizes of the last Neanderthals and the first moderns in western Europe came from the Cambridge archaeologists Paul Mellars and Jennifer French. They mined a large data set recording the extent in area of each of the last Neanderthal sites in southwestern France and those of the succeeding Aurignacians in the same region. Similarly they compared data on the number of stone tools each human population left behind in their sites, and the amount of food debris they generated. Multiplying all these together, they concluded that the early modern population was about ten times the size of the preceding Neanderthal one. This might imply that the moderns swamped the Neanderthals, but at the moment we can’t reliably place them together as direct competitors in the European landscape over any precise length of time, only infer that they probably did coexist.