Fire
Swartkrans cave, situated in the Sterkfontein valley about
thirty miles north-west of Johannesburg, is the single richest
hominid site in Africa. As we’ll see presently, its
oldest deposits – Members 1 to 3 – have yielded, between
them, the skeletal remains of two hominid species that were
living in the early Pleistocene, as well as tools made of
both bone and stone, and animal bones butchered by hominid
meat-eaters.
Excavating grid square W3/S2 in Member 3 of this treasure-house
of early-Pleistocene hominid life on March 21, 1984, George
Moenda, working with Bob Brain, found an apparently charred
piece of bone. Had this specimen come from a later archeological
excavation, Brian tells us,
…I would have had no hesitation
in assuming that it had been partly charred in a fire, but
at an estimated age in excess of a million years, the presence
of burnt bone in a hominid living area required further verification.
He accordingly took a sample of it to the National Physical
Laboratory of South Africa’s Council for Scientific
and Industrial Research, where analysis confirmed that the
blackening of the bone was, in fact, caused by fire rather
than chemical staining.
By the time the excavation of Member 3 was completed, 270
pieces of burned bone had come to light. Although a few of
them had been found grouped in clusters, it’s clear
that those burned bones were, for the most part, the products
of a great many separate fires:
In one grid square, W3/S3, burnt
bones occur in 23 excavation spits, each 10 cm deep, indicating
that the bones were heated in frequently recurring fires during
the deposition period of this stratigraphic unit, which may
have spanned several thousand years.
The distribution of these burned bones throughout such a
large portion of the six-meter-deep gully of calcified debris
which constitutes Member 3, doesn’t only suggest a long
history of fire-use in or near the entrance to the Member
3 gully; it also excludes the possibility that the burned
bones could have “contaminated” that Member, i.e.
found their way into it at some relatively recent time.
In order to find out whether the bones were burned in natural
fires, or in fires made and/or tended by the hominids who
lived in that area, it was necessary to establish the temperatures
to which they had been subjected. Grass or brush fires typically
burn cooler than 250 degrees Celsius, while hearths or campfires
can reach or exceed the 600 degree mark. In a test "hearth"
fire made with wood from Celtis africana, an African
relative of the elm which commonly grows in the entrances
of karst caves like Swartkrans, Brain measured temperatures
of up to 860 degrees Celsius. A test fire made with the wood
of Acacia karoo, another tree which is presently
common in the vicinity of Swartkrans, peaked at 688 degrees.
There is, of course, no absolute rule that brush or veld
fires will not exceed 250 degrees Celsius. Grass fires can
ignite dead trees, or parts of trees, to produce, in localized
patches, the high temperatures characteristic of hearths.
It’s unlikely, however, that anything beyond a small
percentage of a collection of naturally-burned bones (which
would be more-or-less evenly distributed on the veld and not
concentrated in the vicinity of logs) will be heated to those
temperatures.
To begin the task of determining the temperatures at which
the Swartkrans bones were burned, Brain experimentally heated
a large selection of fresh and weathered bones in a kiln at
selected temperatures. He found that the color of these bones
changed to dark brown or black in the vicinity of 300 degrees
Celsius. At higher temperatures this discoloration faded and
the bones started becoming “calcined.” At about
600 degrees they would become grayish-white in color, and
tend to be lighter in weight.
In order to create a histological or microscopic record of
these changes, Brain cut the fresh, defleshed radius of a
hartebeest into seven transverse segments, and heated those
segments, respectively, to 200, 300, 400, 500, 600, 700 and
800 degrees Celsius for thirty minutes, cooling them slowly
thereafter. Thin sections prepared from each of these segments
displayed more-or-less unique microscopic features, characteristic
of the particular temperature to which it had been subjected.
Another portion of each of those bone-segments was sent to
Andrew Sillen, then of the Archeometry Laboratory of the Department
of Archeology at the University of Cape Town, for chemical
analysis.
With the knowledge they had gained from this controlled heating,
Brain and Sillen then subjected the burned bones found at
Swartkrans’ Member 3 to histological and chemical analyses,
establishing that well over half of them had been subjected
to temperatures of 300 degrees or more. Forty-seven percent
had become calcined and grayish-white in color, and were found
to have been heated to temperatures approaching or exceeding
the 600-degree mark. Dr. Anne Skinner and her colleagues at
Williams College in Williamstown, Massachusetts, have recently
confirmed these findings by using ESR to test samples of the
bones in question for the presence of the different kinds
of free radicals which are created by burning at different
temperatures.
The picture that has emerged from these analyses makes it
highly unlikely that anything but a small minority of those
bones could have been burned in natural fires. We know, in
addition, that hominids ate meat off at least some of the
burned bones found in Member 3, because four of them display
cut- and/or chop marks made by stone tools. (No cutmarked
bones have been found outside Member 3.) Of the 59,218 unburned
bone fragments found that Member, a total of twelve (0.02%)
are cut- or chop-marked; of the 270 burned bones found there,
four (1.48%) are marked in that way. The fact that only one
in 4,934.8 of the unburned bones are cutmarked, while one
in 67.5 of the burned bones are marked in that way, would
be a puzzling one if we assumed a “natural” or
non-hominid cause for the bone-burning, but it makes complete
sense if we accept that the burning of those bones was, like
their cut-marking, carried out by hominids.
With the exception of the lower bank of Member 1, in which
three specimens were found, no other burned bones were found
in any of the other Swartkrans members although each of those
other Members yielded, as we’ve seen, plenty of other
evidence of hominid remains and tools. Could this mean that
the entrance to Member 3 had, at the relevant time, some feature
like, say, a dolomite overhang which made it convenient to
build fires in its immediate vicinity, a feature which the
other Members lacked while they were accumulating? Or could
it indicate that the hominids only learned to use fire around
the time that Member 3 started accumulating? Or that the hominids
in question had only been capable of harvesting naturally-ignited
fire before the accumulation of Member 3, (which could explain
why three burned-bone specimens were found in the Member 1)
and that they had learned, by the time Member 3 was accumulating,
to make fire rather than relying on the relatively
infrequent occurrence of naturally-ignited fires in that area?
The latter two possibilities are both reconcilable with East
African evidence which suggests, as we’ll see presently,
that fire-use by hominids started between 1.7 and 1.6 million
years ago, while faunal and ESR dating suggest that Swartkrans’
three older Members accumulated in the vicinity of 1.6 million
years ago.
Eight hundred seventy-seven stone specimens adjudged to be
hominid artifacts were found in Members 1, 2 and 3. These
hominid-fractured stones, which consist mostly of cores and
unmodified waste, aren’t easy to classify. Clark Howell,
of U.C. Berkeley’s Laboratory for Human Evolutionary
Studies, thinks they “approximate the Developed Oldowan
of eastern Africa…” Howell points out, however,
that a number of bifaces, a pick-like piece, and a cleaver-ended
specimen, obtained from dump breccia residues presumably originating
from Member 3, raise the possibility of this Member contained
an Acheulian element.
Member 3 also yielded forty-one bone specimens – mainly
fragments of long bones and horn-cores – whose points
are scratched and rounded in ways that show that they were
used to for digging. A total of 84 such implements were found
in Members 1, 2 and 3, and a few more have been unearthed
at nearby Sterkfontein and Driemolen. Brain concluded from
experimental work done with modern replicas of those tools,
that the originals were used to unearth plant storage organs
like the bulbs of Ledebouria, and corms of the genus
Hypoxis, both of which are still found in the vicinity
of Swartkrans. Experimental work by Lucinda Backwell and Francesco
d’Errico, has persuaded those researchers that the width
and orientation of the striations on those tools, may approximate
more closely the wear-pattern created by digging in termite
mounds.
A delicate, awl-like artifact from Member 3, SKX 37052, has
a point which is marked by both circumferential and longitudinal
scratches, together with polish. “This tool may well
have been used,” Bob Brain and Pat Shipman wrote,
…for piercing holes in skins and other
soft materials, as similar microscopic wear has been documented
on experimentally made and used awls.
The evidence discussed here suggests that
the Swartkrans hominids of Member 1-3 times may well have
made simple carrying bags from animal skins, in which they
transported their tools, as well as possibly their gathered
food. This could explain the evidence for the apparent use
of the same tools over successive days or weeks.
* * *
Hominid remains – mostly fragmented like the majority
of the bone specimens found in that cave – are extremely
abundant at Swartkrans. Skeletal parts representing an estimated
116 individuals of the species Paranthropus robustus
have been found in the cave’s three “old”
Members. Homo – presumably Homo erectus – is
present in Swartkrans, too, but in much smaller numbers: fragments
(including one lower jaw) represent perhaps three individuals
in Member 1, and two in Member 2. The fact that no remains
clearly identified with Homo were found in Member
3 does not mean that this genus was absent from Swartkrans
area during the time that Member was accumulating: lions and
sabertooth cats appear even less frequently in the Swartkrans
deposits than Homo does, but it would be unrealistic
to conclude from that fact, that big cats must necessarily
have been absent, or even rare, in that area during the times
in which their remains don’t show up.
Which of the two human-like beings whose remains have been
found at Swartkrans made the tools, and tended the fires in
which the Member 3 bones were burned?
When Louis Leakey’s family unearthed a specimen of
both Paranthropus and the gracile hominid OH7 in
1.8 million-year-old deposits at Olduvai in which stone tools
had been found, they were confronted – at least in regard
to the tools – with a similar question. Leakey conceded
that it was possible that both hominid species had made and
used those tools, but inclined to the view that the species
which he called Homo habilis “was the more
advanced tool-maker, and that the Zinjanthropus [i.e.
Paranthropus] skull represents an intruder (or a
victim) on the Homo habilis living site.”
It’s by no means impossible that Paranthropus
could, on occasion, have become a victim of Homo.
Homo erectus had, as we’ve seen, become an
expert hunter of large animals in the early Pleistocene, and
there’s no reason to think that large primates would
have been excluded from their list of prey species. Erectus
is thought to have butchered giant gelada baboons later in
the Pleistocene, on the basis of 700,000-old-year-old evidence
found at the DE/89 B site at Olorgesailie in Kenya, and we
know that contemporary humans kill and eat chimpanzees. It’s
interesting, too, to note in this regard, that the burned
bones found in Swartkrans’ Member 3 include a Paranthropus
finger-bone. (Those bones also include one each of a baboon,
a zebra, a guinea fowl, a dassie and a meerkat, while the
rest are those of a group of antelopes and other split-hoof
animals which zoologists refer to as bovids.)
The relationship between Homo erectus and Paranthropus
would have been a strange one from the point of view
of present-day humans. Erectus – less intelligent,
we can assume, than sapiens was to become –
would have been sharing a habitat with a fellow-primate which
was probably quite a bit more intelligent than present-day
chimps or gorillas. Randall Susman’s anatomical investigations
have led him to the conclusion that Paranthropus’
hands were well-adapted to precision grasping and the use
of tools. It’s clear that Paranthropus and
Homo were genetically isolated from each other, but
there’s no reason why simple technological memes could
not have flowed between them. It would not be surprising,
at any rate, if digging-tools like those found at Swartkrans,
as well as wooden implements of various kinds, were used by
both hominids. One could speculate that Paranthropus
might have used crude stone tools – and perhaps even
the discarded Acheulian tools they would occasionally have
come across. Acheulian tools are, however, so frequently and
reliably associated with Homo erectus, and associated
with it, moreover, for so long after Paranthropus
becomes extinct around .9 million years ago, that there can
be little doubt that erectus was the developer and
practitioner of the of the Acheulian industry (and, indeed,
its immediate predecessor, the Developed Oldowan).
In the absence of evidence suggesting otherwise, I’ll
continue to presume, therefore, that advanced technologies
like the making of Acheulian tools and the use of fire were
the exclusive preserves of the big new hominid whose brain-capacity
was about 50% larger than that of both its Pliocene ancestors
and its robust contemporaries.
* * *
When did the fire-use evidenced in Swartkrans take place?
Unlike many of the bones and hominid-associated objects from
East Africa, which are often conveniently sandwiched between
easy-to-date layers of volcanic ash, those found in South
Africa’s limestone caves are notoriously difficult to
date. The fossil- and artifact-bearing deposits of Swartkrans
entered that cave at different times, falling into it or sliding
down “talus” slopes through different entrances
or holes in its roof. At times when the cave was accessible,
some objects are thought to have been carried into it by four-footed
carnivores or hominids. The body of debris presently found
in that cave constitutes, therefore, what Brain understatedly
calls “a depositional mosaic of considerable complexity.”
In a professional involvement with the cave that has lasted
more than half a century, Brain has resolved that mosaic into
five more-or-less distinct “Members.” Those Members
represent, in Brain’s view, depositional episodes which
alternated with erosional phases, in cycles which were probably
driven by precipitation changes triggered, in their turn,
by the 40,000-year glacial cycles which the planet was experiencing
up to a million years ago. The deposits are numbered in the
order in which they appear to have accumulated in the cave,
with Member 1 being the oldest, Member 2 being the second-oldest
and so on.
In the 1960s and 70s Elizabeth Vrba – then of the Transvaal
Museum, now of Yale – made a detailed and comprehensive
study of the fossil bovids from Swartkrans and other hominid-bearing
cave deposits. On the basis of when those bovid species had
made first and last appearances in deposits elsewhere in Africa – deposits
which could be securely dated by stratigraphic and
other means – she concluded in 1982 that a portion of
Member 1, the so-called “Hanging Remnant,” was
between 1.8 and 1.5 million years old. With the exception
of the burned and/or cutmarked bones we’ve been discussing,
the hominid-associated objects found in Members 2 and 3 don’t
differ significantly from those of Member 1. Apart from the
fact that Homo doesn’t appear in Member 3,
the hominid remains recovered from the three “old”
Members don’t differ from each other either. The animal
ensembles found in those three Members are also substantially
similar. Members 1, 2 and 3 are presumed, for these reasons,
to have accumulated in relatively rapid succession to each
other.
A 2004 review of the fauna of Swartkrans’ Members 1-3
by Darryl de Ruiter, now of Texas A&M, yielded an estimate
for the age of those Members which was very close to the result
produced by Elizabeth Vrba’s study of Member 1’s
“Hanging Remnant.” De Ruiter concluded that, although
Members 2 and 3 were contaminated by some recent material,
all three of the older Members were deposited in the neighborhood
of 1.6 million years ago.
Direct ESR testing of two hominid teeth and two bovid teeth
by Rainer Grün, Darryl Curnoe and others have produced
findings which are in agreement with those of Vrba and de
Ruiter, yielding an age of between 1,470,000 and 1,790,000
with a median of 1.63 million years for the “Hanging
Remnant” of Member 1 in which those teeth were found.
As one might expect from deposits which accumulated before
the 1.4 million-year-ago “hemorrhage” of megafaunal
diversity we talked about in Chapter 12, Swartkrans’
Members 1, 2 and 3 contain several species which disappeared
in that hemorrhage – the cheetah-like Chasmaporthetes
hyena makes two appearances, along with the machairodonts
Dinofelis (one appearance in Member 1) and Megantereon
(one appearance in Member 3).
In the sense that God is (as Ludwig Mies van der Rohe famously
remarked) “in the details,” the truth about hominid
fire-use in the early Pleistocene is, I believe, discernable
in the details of Bob Brain’s meticulously-documented
excavation and analysis of the burned bones in Swartkrans’
Member 3. Those details do not, in my view, merely suggest
that hominids may have been using fire by 1.5 million
years ago. They raise, instead, an overwhelming likelihood – a
practical certainty – that they were doing so.
* * *
Because signs of fire-use are not, until the later Pleistocene,
found as frequently as, say, stone tools are, Ian Tattersall
suspects that fire-use may have been a “rather intermittent”
part of human life in the earlier parts of that Epoch. I’m
prepared to accept that small and isolated hominid populations,
like the approximately 10,000 members of sapiens confined
to Tasmania by a rise in the sea-level at the end of the last
glaciation, may have lost the art of fire-making, but it doesn’t
seem likely to me that a such a powerful and useful technology
could, at any time after it was developed, have been abandoned
or lost throughout the entire African and South Asian range
which our family occupied in the Early Pleistocene. The fact
that fires feed on organic material – mainly wood – which
is preserved far less frequently than stone artifacts, is,
in my view, a more likely explanation for the fact that we
don’t find the signs of fire-use as consistently as,
say, stone tools.
But stone artifacts are themselves able to provide information
about early hominid fire-use. When “siliceous”
stone such as basalt or quartz (as distinct, say, from “calcereous”
material such as limestone) undergoes prolonged exposure to
temperatures exceeding 250 degrees Celsius, (which marks,
as we’ve seen, the borderline between the “campfire”
and “brush fire” temperature-ranges) its surface
undergoes characteristic color and texture changes, developing,
in particular, dimple-like scars called “potlid fractures.”
A survey by Brian Ludwig of Rutgers University of stone artifacts – and
of the waste pieces or “debitage” produced in
the making of such artifacts – revealed a telling pattern
of thermal alteration. Ludwig personally examined almost 40,000
pieces of hominid-modified stone from nearly fifty sites in
the Olduvai and Turkana basins, covering a period of 2.5 million
to less than one million years ago. In doing so, he found
no potlid fractures on any specimens dating from before about
1.6 million years ago. After that time, however, a small but
consistent proportion of the stone does display potlid fractures,
over a wide variety of sites including Olduvai where, despite
the preservation of a rich store of hominid-modified material,
no signs of fire-use had previously been found.
The other area surveyed by Ludwig, the Turkana basin, had,
however, yielded signs of hominid fire-use prior to his survey.
These were first identified in the 1970s , when Jack Harris,
who had been Ludwig’s teacher at Rutgers, examined several
patches of apparently baked reddish earth at a cluster of
sites known as FxJi-20, situated in the 1.6 million-year-old
Okote tuff at Koobi Fora. Harris and some of his colleagues
became interested in those patches when they realized that
they resembled the present-day spots on which local tribespeople
had made overnight campfires. These apparently-baked patches
were situated in an area which had yielded a number of hominid
artifacts, some of which were apparently discolored and/or
fractured by heat. It also contained a scattering of faunal
remains which included the lower jaw of a Paranthropus – KNM-ER
3230.
A 1978 suggestion by Harris that the reddish patches were
remnants of fire-places or hearths, provoked a skeptical response.
They could, his critics asserted, have been created by bush
fires. Alternately, their discoloration could have been caused
by either fungal invasion or the precipitation of iron particles.
Ralph Rowlett, an anthropologist from the University of Missouri,
then established, by thermoluminescence tests of four of the
discolored patches conducted in association with Charles Peters,
that those patches had been heated more recently than the
volcanic soil in which they were situated. That considerably
diminished the possibility that they could have resulted from
non-thermal processes. Rowlett also eliminated the possibility
that they had been directly caused by lightning strikes by
examining known strike sites in Africa and the United States.
He concluded that the discolored Koobi Fora patches – which
were between 12 to 20 inches in diameter – would have
been much smaller if they had represented lightning strikes.
Those patches were not, moreover, associated with fulgurites,
the glassy bits of fused sand that are produced by such strikes.
Rowlett, Randy Bellomo (then with the University of South
Florida, now of the University of Wisconsin-Milwaukee) and
others found melted crystals in the patches which indicated
that the fires causing them had produced temperatures close
to 400 degrees Celsius. (Grass- and brush fires do not, as
we’ve seen, usually exceed 300 degrees Celsius.) We’ve
seen, though, that grass- or brush fires on Africa’s
savannas can and do ignite logs and/or dead tree-stumps, which
can burn hotter than the fire which ignited them. The likelihood
that discolored patches had been produced in this way was
tested by two different methods.
Firstly, an archeomagnetic analysis of several possible fire-places
at Koobi Fora’s FxJi-20 Main site by Randy Bellomo yielded
evidence that they had been re-heated several times over a
period of several years. It’s not unlikely that camp-fires
would be re-lighted repeatedly on particular spots which tradition
and convenience might have established as hearths, but highly
unlikely that “natural” ignition of logs and stumps
would have occurred repeatedly on the same places.
Secondly, Ralph Rowlett (together with Robert Graber and
Michael Davis) examined the phytoliths left by various kind
of fires, including the those which had discolored the suspected
Koobi Fora fireplaces. Phytoliths are microscopic bodies of
opaline silica formed in epidermal and other cells of the
growing plant, from silica dissolved in ground water originally
absorbed by the plant’s roots. They typically survive
both the decay and incineration of the plant which produced
them, and can often be used to identify the species or, at
least, the general botanical category of that plant. Rowlett
and his colleagues reasoned that a fire consisting of a naturally-ignited
log or stump would only leave one kind, or, at any rate, relatively
few kinds, of phytolith. Fires which were, on the
other hand, started by hominids, would have left many kinds,
because they would have been kindled and fed with grass, leaves,
twigs, sticks and logs from a variety of plant species.
A comparison of phytolith residues of test camp-fires made
by students (who were not told what the nature of the study
was) in savanna-like surroundings, and of stumps that had
been experimentally set alight, confirmed that the former
produced a heterogeneous collection of phytoliths, while the
latter left a residue which tended to be homogenous.
Rowlett and his associates noted another difference between
the test fires in which logs and stumps were ignited, and
those in which camp fires had been made: the former left residues
with irregular shapes, while the latter left basin- or lens-shaped
residues.
Three of the four discolored patches at Koobi Fora’s
FxJi 20 E site, were basin-shaped, while a fourth, situated
some distance from the others, was irregular. The basin-shaped
patches were found to contains even more kinds of phytoliths
than the test camp-fires had, while the remaining, irregular
one contained only one kind, and was presumed, accordingly,
to have been made by naturally-ignited tree.
Swartkrans and Koobi Fora are not the only sites at which
researchers have found possible evidence of early-Pleistocene
fire-use by hominids. Hominid-used fire may also have discolored
the clumps of earth discovered in Chesowanja in Kenya, which
were also dated to about 1.6 million years b.p.
Israel has provided a link between these early African sites,
and later European evidence of hominid fire-use. Burned seeds,
wood, and flint found at Gesher Benot Ya’aqov in the
Northern part of that country, suggest that humans were, nearly
790,000 years ago, controlling fire at this site. The diverse
collection of plant material burned there, includes three
species which had edible fruits or seeds: olive, wild barley,
and wild grape. The distribution of burned fragments of flint
at the site, was, moreover, suggestive to investigators from
the Hebrew University in Jerusalem and Bar-Illan University
in Ramat-Gan, that burning had occurred on specific spots,
possibly suggesting the locations of hearths.
* * *
Bob Brain was careful to point out that the hominids who
burned the Swartkrans bones had not necessarily learned how
to make fire. He thought it was, instead, likely
that they had harvested their fire from naturally-ignited
flames.
How might that “harvesting” have begun? Birds
and mammals often converge on African veld fires to prey on
small mammals and insects flushed out by the flames. Hominids
could initially have been drawn to such fires for the same
reason. Eventually, the inventive, curious natures of those
beings might have prompted them to literally “play with
fire” at upwind or non-threatening edges of those burns.
That could have taught them enough about the dynamics of fire
to enable them to “harvest” flames or embers,
and use them to build and maintain fires at places of their
own choosing.
How frequently would natural fires have occurred? In Southern
Africa, the overwhelming majority of such fires burn in September
and October when dry electric storms ignite grass or brush
before the summer rains begin. In an article entitled “Fire
behaviour in the Kruger National Park,” published in
the Journal of the Grassland Society of Southern Africa
in 1985, W. Trollope and A. Potgieter report that lightning-caused
fires can be expected to burn any given area in the Kruger
National Park at intervals ranging between five and ten years.
Several subsequent studies have confirmed that a large proportion
of Kruger’s savanna remains unaffected by natural fire
for years at a time. Daniel Koen, who has studied the issue
of naturally-caused highveld fires intensively to formulate
a fire-policy for Suikerbosrand nature reserve just south
of Johannesburg, tells me that, even though Suikerbosrand
is one of the most lightning-prone areas in South Africa,
many parts of the reserve can go for years without being burned
by lightning-caused fires. The incidence of lighting-strikes
varies widely from year to year, and periods in which lightning-caused
fires don’t occur in the Reserve, or affect only a small
part of it, can be interrupted by a summer in which lightning
strikes are widespread and frequent. Under such circumstances,
successive lightning-ignited fires can start in close proximity
to each other in a single season, if each is extinguished
by rain before it burns all the available fuel in that area.
As we’ve just seen, though, lightning-caused fires
normally take place much less frequently than this. By August
of 2005 I had, for instance, been waiting more than two years
for the veld around my favorite walking trail in Suikerbosrand
reserve, Bokmakierie, to burn. (My impatience had to do with
the fact that fire, followed by rain, transforms Bokmakierie
into a verdant Alpine meadow, in which a well-beyond-Alpine
diversity of plants begins to flower.) Most of the Bokmakierie
area was finally burned in late 2005, whose rainy season was
characterized by an exceptional number of electric storms,
by one or more lightning-ignited fires, and by another started
by a visitor or staff member next to a public road.
I’m impressed – especially in view of the fact
that natural, harvestable fires would probably have occurred
relatively infrequently – by the abundance and the
ubiquity of burned bone found in Swartkranz’s Member
3. I suspect, moreover, that a great many “archeologically
invisible” fires – i.e. fires which didn’t
contain bone – must, in addition, have been made in that
Member’s catchment area: bones would not always have
been available to the fire-users, and bone produces, beside
that, an acrid, unpleasant-smelling smoke when it’s
thrown into a fire. Taken together with the fact that fire-use
seems, from the investigations of Rowlett, Bellomo, Ludwig
and others, to have become a widespread phenomenon in Africa
around 1.6 million years b.p., the high frequency of hominid-tended
fires evidenced at Swartkrans prompts me to speculate that
the beings who tended those fires could, by the time Member
3 was accumulating, have overcome their dependence on harvested
fire, and learned to kindle flames at will.
Hominids who were still dependant on harvesting
fire, would probably have learned, after a while, to keep
their fires going for relatively long periods of time, and
to transport embers over relatively long distances. Days or
weeks after such harvestings, however, inattention, sleep,
local fuel exhaustion, or rain would inevitably have deprived
them of fire for months or even years at a time. Each of those
losses would have been a bitter disappointment, because the
harvesters would, by that time, have experienced the comfort – and
the enormous increase in personal safety – that fire could
bring to their nights. Finding a way of kindling a flame would,
therefore, have become a very pressing issue indeed.
Would the average erectus have been able to find
such a way? No more, I imagine, than the average sapiens
would have been able to invent the polymerase chain reaction
in the late twentieth century. (The PCR, a cornerstone of
the biotech industry, makes it possible to multiply strands
of DNA to large quantities for a great many scientific, medical,
forensic and industrial purposes. It earned Kary Mullis, who
invented it in 1981, a Nobel Prize, the prestigious Japan
Prize, and a place in the National Inventors Hall of Fame.
In 1991 Hoffman-La Roche bought the patent for this process
from Mullis’ former employer for $300 million – possibly
the highest price ever paid for a piece of intellectual property.)
Acheulian tools, which were used for purposes
as disparate as digging in the earth, butchery, leatherwork,
woodwork and scything soft plant material, attest to relatively
high levels of ingenuity. So does the harvesting, transportation,
and maintenance of fire. It doesn’t seem unlikely to
me, therefore, that one or more erectus counterpart
of Kary Mullis could have noticed, somewhere between 1.7 and
1.5 million years ago, that sparks struck while knapping certain
kinds of stone, could fall, still glowing, onto dry organic
material and set it smoldering, and that he (or she or they)
could have exploited that state of affairs to make the discovery
that Charles Darwin regarded as “the greatest ever made
by man, excepting language.”
* * *
One of the characters in Ernest Hemingway’s The
Sun Also Rises asks another how he went bankrupt. “At
first slowly,” the other replies, “and then quickly.”
That’s the schedule on which exponential growth takes
place – invest money at compound interest, and the return
mounts up slowly at first, and then quickly. One dollar invested
at an annually compounding rate of 14.8153622%, will yield
a million dollars at the end of a century, but it will take
just under 90 years to earn a quarter of a million dollars,
almost 95 years to reach the half-million dollar mark, and
nearly 98 to produce three-quarters of a million.
Technological evolution is as complex as hominid life itself.
The growth of its power could not, therefore, be expected
to trace out the smooth upward curve which a simple mathematical
exponent produces. We would expect, instead, to see a generally
exponential increase, punctuated by relatively abrupt
rises, plateaus and even local reversals. Like our one-dollar
investment, that “generally exponential” increase
would have seemed to dawdle along for the first eighty or
ninety percent of its history, before accelerating dramatically
toward the end of the Pleistocene. This initial dawdle has
given rise to the conventional view that the first two million
years of hominid tool-making were characterized by what Carl
Swisher and his co-authors refer to in their Java Man
as a “mind-numbing lack of innovation.” It has
also led many paleontologists to conclude that what they regard
as “advanced” abilities of our species, such as
the use of language, the creation of art, and the hunting
of big game, were all acquired very recently in a “Great
Leap Forward” or “Human Revolution” which
took place less than 50,000 years ago.
If you imagine, though, that our one-dollar investment produced
nothing of value in the first seventy or eighty years of its
existence – that it was characterized by a “mind-numbing”
lack of growth during that time – then you haven’t
grasped the nature of exponential increase. After fifty years,
that one-dollar investment will have earned one thousand dollars.
That’s only a one-thousandth of the million dollars
it would earn after a century, but it’s also a thousand
times more than the starting amount. The inventive capacity
of Homo erectus had reached a comparable point around
1.5 million years ago: its products were still very crude
compared to the devices and processes which would make
H. sapiens a world-conqueror toward the end of the Pleistocene,
but they had already conferred on their inventors a degree
of power which was overwhelming in the naïve world of
the early Pleistocene. Another million and a half years would
have to go by before members of our family would become able
to make the warm clothes and footwear that would allow them
to penetrate Northern Eurasia and go from there to the New
World, but cultural evolution had, by the time the Acheulian
revolution was taking place, already begun to outpace its
genetic progenitor, “surprising” an ecosystem
whose other members were still exclusively reliant on genetic
evolution, and overwelming many of them in the “ontogenetic
ambushes” that we talked about in Chapter 11.
Fire-use alone must have added enormously to the ecological
power of its discoverers. Fire could repel predators, and/or
reveal their presence at night. It must also have been used
to drive prey animals, and, in some circumstances, to kill
them. Through its ability to neutralize bacteria and toxins,
and to release nutrients, hominid-controlled fire would, in
addition, have caused what Jonathan Kingdon calls “a
vast extension in the food base.” This would probably
have caused an increase in the population of erectus,
which would, in turn, have increased the impact of its other
ontogenetic innovations.
Despite its immense importance, fire-use was, however, only
one manifestation of the mental illumination which was allowing
erectus to increase its ecological power so dramatically
in the early Pleistocene. It seems probable, for instance,
that this species was, by that time, also exchanging and refining
ideas through the use of language.
Evidence for the latter proposition – which I reviewed
briefly in Chapters 9 and 11 – includes the possibility
that a brain structure comparable in achitecture and function
to Broca’s area was present in Leakey’s KNMR 1470
find; the anatomical evidence that hominid brains might have
become more-or-less strongly lateralized by the early Pleistocene;
the archeological evidence of right-handedness which lends
support to the latter proposition; and the apparent descent
of the larynx evidenced in several Homo erectus skulls.
In Chapter 7 of his 1994 Origin of Humankind, Richard
Leakey concludes from these indications that the earliest
members of the genus Homo were already language-users.
But language-use didn’t arise abruptly – either
30,000 years ago, or with the advent, in the early Pleistocene,
of what we’ve arbitrarily defined as the genus Homo.
Both Steven Pinker’s 1994 Language Instinct and
Terrence Deacon’s 1998 The Symbolic Species – The
co-evolution of language and the brain, provide support
for Raymond Dart’s 1925 thesis that Australopithecus
had, back in the Pliocene, already “laid down the foundations
of that discriminate knowledge of the appearance, feeling
and sound of things that was a necessary milestone in the
acquisition of articulate speech.”
* * *
We can imagine, therefore, that instructions, warnings, and
encouragement would have been shouted back and forth as a
group of early-Pleistocene erectus closed in on a
Deinotherium (or, indeed, a Loxodonta) with
their tool-sharpened wooden spears, driving it, perhaps, toward
a concealed ditch which they had dug with their big Acheulian
picks. If we could somehow transport a herd of present-day
Loxodonta or Elephas back into the early
Pleistocene, its members would probably turn on those erectus
attackers, kill one to two of them, and scatter the survivors
like chaff. As we’ve seen, however, the high levels
of comprehension, social cohesion, and instinctive aggression
toward hominids which would enable them to do so, would not
have existed in the early Pleistocene.
Early-Pleistocene elephant species would obviously have had
a more varied and effective set of behavioral defenses against
predators than the giant tortoises we talked about in Chapter
11, but they must, in comparison with their modern-day counterparts,
have relied more heavily on their size, and on the thickness
of their outer “walls,” than on behavioral flexibility,
to repel predators. The machairodont or sabertooth cats had
evolved “siege engines” – powerful forequarters
and necks, and elongated upper canines – to breach the
castle-like fortifications of these and other megaherbivores.
Regardless of the danger that this equipment would have posed
to individual megaherbivores – especially immature and
older ones – it did not threaten any of the species against
which it was directed with extermination. The relatively sudden
acceleration in human ingenuity which took place in the early
Pleistocene did, however, pose a threat of that kind.
Just as gunpowder rendered castles obsolete in the early middle
ages, our family’s rapidly growing ability to invent
and improvise had a fatal impact on species whose defense
strategies were focused too exclusively on large size and
a fortified exterior. Most of Africa’s megaherbivores
disappeared, as we’ve seen, around 1.4 million years
ago, together with the machairodont cats and giant hyenas
which were dependant on them.
* * *
We discussed, in Chapter 1, the counterintuitive truth that
vulnerability to human-caused extinction increases with size.
The fact that the great majority of the twenty-odd proboscidean
or elephant-like species that were in existence at the beginning
of the Pleistocene have become extinct should not, therefore,
be a surprising one. It’s surprising, on the contrary,
that two of those species, Loxodonta africana and
Elephas maximus, have survived into the present. Their
survival tells us that the Acheulian “avalanche”
which overwhelmed Africa’s other proboscideans was moving
slowly enough that phylogenetic evolution managed, in the
cases of those two species, to stay ahead of it.
By the time Loxodonta and Elephas were
subjected to the relatively fast surge in our species’
power which took place in the late Pleistocene – the surge
which killed off all the proboscidean species that inhabited
Northern Eurasia and the New World – those two survivors
were behaviorally equipped by their early-Pleistocene trials
with our family, and by more than a million years of subsequent
conflict and co-evolution with that family, to withstand it.
Table of Contents
