A radically different kind of faculty
Riding a pony in the mountainous country of Lesotho, I would
become nervous when the narrow, rock-strewn trail we were
negotiating skirted the edges of steep precipices. I found
myself paying close attention, at such places, to where my
horse was placing its hooves. On several occasions, I saw
it test footholds, before shifting its full weight onto them.
These were reassuring acts. My animal had the “horse
sense,” I realized, to choose “wise” or
“judicious” ways to negotiate that potentially
hazardous trail, without possessing anything like a human
level of intelligence. Nature had, it seemed, equipped it
with cheap but effective neural machinery dedicated to the
production of sophisticated all-terrain mobility, rather than
bothering with the more general and abstract forms of intelligence.
We use a word for behaviors of that kind that was, until recently,
beginning to sound quaint or obsolescent: instincts.
Natural selection is as capable of assembling complex instincts
as it is of producing complex body shapes. Consider, for example,
the inborn “promptings” and “guidance”
which will allow a young Arctic tern to make its first migratory
flight from northern Canada down to Antarctica. The bird’s
parents will leave before it does, so it has to rely entirely
on a genetic “navigation package” to complete
that ten-thousand mile journey. Ingenious experiments relating
to the directions in which caged birds orient themselves in
planetariums have shown that some migratory bird species have
“star maps” written into their DNA, and it’s
not unlikely that Arctic terns are also equipped with “instinctive”
sky charts of that kind. It’s difficult to imagine how
chance variations – each separately advantageous to its
possessor – could have accumulated to put together such
a star map, as well as the “clocks,” “compasses”
and other specialized structures which constitute the tern’s
“avionics.” Natural selection is, however, as
Richard Dawkins’ 1997 Climbing Mount Improbable
demonstrates so convincingly, equal to tasks of that kind.
While the mills of natural selection can, therefore, grind
exceedingly fine, they also grind very slowly. Bat
species took millions of years to develop the ability to send
out the sound-pulses whose echoes locate the insects they
feed on, while the insect species which execute spiraling
crash-dives when they hear those sound-pulses, took similarly
big chunks of time to evolve that defense. If insects could
have understood what was happening when they heard bats’
“sonar” pulses, then they could have adopted countermeasures
such as the crash-dive immediately. Because they don’t
have that kind of understanding, the impulse to make power-dives
in response to sonar pulses had to be constructed entirely
by the accumulation, over thousand of generations, of chance
variations in the reactions of individuals to such sonar pulses.
Animals of all kinds are equipped to react to particular
situations in such uncomprehending but appropriate ways. Humans
themselves are not required to understand that spiders can
give them dangerous bites, or that contact with excrement
can pass contagious diseases to them. Specialized neural circuits
tell us instead – in a peremptory way that bypasses logic – that
spiders look frightening and that excrement is disgusting.
Much of what we think and do is powerfully affected by instincts.
We don’t have to be taught to fear heights, enjoy sweet
tastes, find someone beautiful, or seek the respect of our
fellow-humans. A multitude of instincts shape our behavior
so effortlessly and powerfully that we tend to be blind to
their existence. Activities with large instinctual components,
like walking, eating, and talking, seem so much like the natural
order of things to us, that it takes a neurological disability
to bring home to us that they are, in reality, the extraordinary
and unlikely productions of “purpose-built” neural
systems.
Even the moral dimension of our species’ existence,
often thought of as a kind of “opposite pole”
to the instinct-dominated “animal” part of our
make-up, may have originated in, and still be influenced by,
the kind of “special-purpose” neural circuitry
which characterizes instinctive behavior. “[T]he first
foundation or origin of the moral sense,” Darwin reasoned
in The Descent of Man, “lies in the social
sense...”
Animals endowed with the social
instincts take pleasure in each other’s company, warn
each other of danger, defend and aid each other in many ways.
These instincts are not extended to all the individuals of
the species, but only to those of the same community. As they
are highly beneficial to the species, they have in all probability
been acquired through natural selection.
* * *
Natural selection can, as we’ve seen, construct instinctive
behaviors whose complexity and effectiveness strain our credulity.
“There is,” Daniel Dennett declares in his Darwin’s
Dangerous Idea, “simply no denying the breathtaking
brilliance of the designs to be found in Nature.”
Time and time again, biologists baffled
by some apparently futile or maladroit bit of bad design
in nature, have come to see that they have underestimated
the ingenuity, the sheer brilliance, the depth of insight
to be discovered in one of Mother Nature’s creations.
Francis Crick has mischievously baptized this trend in the
name of his colleague Leslie Orgel,
speaking of what he calls ‘Orgel’s Second Rule:
Evolution is cleverer than you are.’
“Orgel’s Second Rule” is unassailable.
The level of “ingenuity” which has gone into the
assembly of the instinctive behaviors we’ve been talking
about, exceeds that of our species by a very wide margin.
And yet the world of instinctive behavior has limitations.
Instinctive behaviors can only evolve in response to situations
that arise over and over in the life of a species, over thousands
of generations. The mechanism which assembles them isn’t
flexible enough to respond appropriately to the enormous world
of opportunities which don’t present themselves in this
repetitive, “stereotyped” way. If a species could,
therefore, retain the indispensable benefits which instinctual
behaviors confer on it, but develop, in addition, a method
of devising useful responses to opportunities lying beyond
the reach of the instinct-building mechanism, it would
gain access to a cornucopia of new resources.
The human family has developed such a method. It did so by
evolving “general-purpose” computational machinery
that can “think up” or invent useful
responses to this previously unexploitable class of opportunities.
That computational machinery hasn’t just enabled us
to respond to a much broader range of opportunities –
it has also enabled us to respond to such opportunities very
quickly. Invention and improvisation take place on
what biologists call the “ontogenetic” level.
They’re produced, in other words, “in real time,”
by individuals rather than being evolved “phylogenetically”
– by species – over thousands of generations.
This has given hominids an overwhelming advantage over all
other members of the biosphere. The psychologist Leda Cosmides
and the anthropologist John Tooby, whose work I’ve relied
on heavily in this chapter, describe that advantage, and some
of its effects, as follows:
Instead of being constrained to
innovate only in phylogenetic time, ...[humans]... engage
in ontogenetic ambushes against their antagonists – innovations
that are too rapid with respect to evolutionary time for their
antagonists to evolve defenses by natural selection. Armed
with this advantage, hominids have exploded into new habitats,
developed an astonishing diversity of subsistence and resource
extraction methods, caused the extinction of many prey species
in whatever environments they have penetrated, and generated
an array of social systems far more extensive than that found
in any other single species.
* * *
The world of “ontogenetic” solutions isn’t
entirely closed to non-hominid organisms. Some species have
been able to respond to novel opportunities by expanding existing
instinctive behaviors: robins and tits have learned, for instance,
to peck open the aluminum caps on milk bottles in Britain
(and, in the case of the tits, passed this “discovery”
on to other members of their species by example).
Other non-human animals use general-purpose computational
machinery analogous to that evolved by our species, to “think
up” responses to novel opportunities. Ravens raised
by Berndt Heinrich in Maine, were able, for instance, to get
hold of meat suspended from a perch by two-foot pieces of
string, by pulling a length of the string up with their beaks,
standing on it with a foot, and then repeating that operation.
This solution was not, in Heinrich’s view, discovered
by trial and error. For several hours after he had fastened
the meat-and-string device to the perch, none of the birds
came near it. Then one of them abruptly flew to the perch,
and used the “beak over foot” method to haul the
meat up without further ado. The method seems, therefore,
to have “occurred to” that bird as the result
of a mental examination of the problem. Corvids – the
group to which ravens, crows, magpies and blue jays belong – are
a very smart family. In Richmond, British Columbia, I witnessed
crows dropping hazel nuts which they had brought from a nearby
tree onto a busy street during rush-hour, so that the cars’
tires would break their shells. It was a startling sight to
see those birds swooping down into the traffic to retrieve
the kernels of those nuts, and to steal them from one another.
A Japanese macaque monkey became famous for learning to separate
the cooked rice given to her by humans from sand by floating
the former away from the latter in water, a method which some,
but not all, members of her band were able to imitate. The
world of invention and improvisation does not, moreover, seem
to be restricted to particularly clever groups like primates
and corvids: most if not all vertebrates may, in a rudimentary
and limited way, be able to “think up” beneficial
responses to novel opportunities. It would be surprising,
in fact, to learn that relatively intelligent invertebrates
like squids and octopuses don’t possess some degree
of “general-purpose” computational ability.
To say without qualification, however, that “both humans
and some non-human organisms can invent beneficial new ways
of doing things” would be as misleading as saying without
qualification that “both humans and earthworms are sensitive
to light.” Earthworms don’t have eyes, and they
can only detect changes in the intensity of light falling
on the front ends of their bodies with the aid of light-sensitive
cells situated in that region. Humans, on the other hand,
can use light to create exquisitely precise mental images
of their surroundings. The difference between the light-manipulating
powers of humans and those of earthworms is a roughly accurate
metaphor for the difference between the inventive power of
hominids and that of non-hominids. The former is so much greater
than the latter that it appears, for all practical purposes,
to be a radically different kind of faculty.
* * *
The ability to “think up” solutions to novel
problems seems to rest on “what if” questions.
“What if,” a member of our species might ask herself,
“I take the noose that I’ve tied into this cord,
and position it across a path that small antelopes like duikers
use, while tying the other end of the cord to a branch?”
To answer that question, her brain will produce a kind of
“movie” of the consequences likely to flow from
this action. The plot-line of that “movie” will
conform to her idea of how the causal structure of the real
world works. Her understanding of that structure could be
good enough to allow her to predict correctly that duikers
aren’t likely to spot a noose stretched across a path
which passes through tall grass, and that they could, therefore,
put their heads into such nooses and then pull them tight
in their efforts to escape. If too many duikers manage to
pull their heads out of those nooses, (in either her mental
movie or in reality) then sequels will appear, produced by
the original snare-designer or by others, in which young trees
whose crowns are bent down and secured to the ground with
hair-trigger connections, might spring upright to hoist the
hapless little beasts off their feet when those connections
are disengaged by a tug on the noose.
Hominids have become so proficient at using such “movies”
or “models” to evaluate possibly advantageous
new behaviors, that they’ve become the sole occupants
of what John Tooby, Irven de Vore and Leda Cosmides speak
of as “the cognitive niche.”
One might suppose that our duiker-trapper was able to “think
up” her snares because she had emancipated herself from
the “primitive” instinctual functions of her brain,
but that doesn’t appear to be the case. The “general-purpose”
computational machinery which our species has evolved hasn’t
turned its back on our special-purpose neural structures – our
instincts – as if they were poor relatives. It employs
them, instead, and seems, indeed, to be dependent
upon them. Our duiker-trapper’s general-purpose computational
machinery might have been overwhelmed, for instance, by the
volume and complexity of the calculations required to model
the physics of the snares she was considering, if it didn’t
have a “cheat sheet” of “internal”
or “intuitive” physics to refer to.
General-purpose or inventive intelligence is also thought
to employ instincts or “special-purpose modules”
to help avoid what cognitive theorists and specialists in
the field of artificial intelligence refer to as “combinatorial
explosions”: the fact that even a small increase in
the elements of a problem leads to exponential – i.e.
explosive – growth in the number of ways those elements
can combine with each other. The nine-step game of tic-tac-toe
can, for instance, unfold in 362,880 different ways. Computers
can review all these ways in real time in order to avoid the
ones that lead to loss or stalemate, but humans can’t
ordinarily manage a calculation of that size.
The world of all possible chess games is much too big to
be subjected to a review of this kind by either humans or
computers. (Computers beat us at chess only because they can
construct bigger “trees” of consequences for particular
moves than we can.) The process of human inventiveness can’t
and doesn’t, therefore, involve sorting through enormous
numbers of combinations by brute force. Trying to design a
duiker-snare by considering all the possible ways that the
raw materials in your environment can be processed and combined,
would be like trying to write a book on French cooking by
generating all the permutations which the characters printable
by your computer, including blank spaces, can assume in a
200-page sequence. Although those permutations include what
Daniel Dennett refers to in his Darwin’s Dangerous
Idea as a “Vast” number of books on French
cooking, an attempt to generate even one of by a process of
random iteration, would be Vastly unlikely to succeed on a
schedule relevant to you, your publisher, or, indeed, the
lifetime of the solar system.
Our duiker-snare designer could not, therefore, have succeeded
in her task if her thinking wasn’t “shepherded”
toward potentially productive areas with the help of well-stocked
caches of instinctual, personal and cultural information,
as well as input from her biological drives and emotions.
Cognitive theorists refer to this “shepherding”
process with words related to steering (cybernetic) or learning
(heuristic), but we’re still a long way from understanding
it. Its results are, however, familiar enough: while human
thought-processes are fallible, they can make breathtaking
leaps of discovery and invention. Some of those leaps are
made in at least partly conscious ways; others occur without
the intervention of conscious logic, as “revelations,”
intuitions and gut feelings.
* * *
The “general-purpose” or “abstract”
intelligence we’ve been talking about is obviously an
immensely beneficial tool. It may seem surprising, therefore,
that, in over four billion years of biological evolution only
one biological family – that of the hominids – has
managed to develop it. Presumably abstract intelligence took
more time to evolve than the modules which run specific behaviors,
because the complexity of the neural systems which run the
former, is of a higher order than that of those which manage
the latter. It’s conceivable, too, that an abstract
intelligence of our kind could not have come into existence
before the “operating system” of instinctual behaviors
on which it appears to depend, reached a critical level of
richness and flexibility. Speculating along similar lines
more than a century ago, William James suggested that the
power of the human intellect stems from the fact that our
species has more, rather than less, instincts than other animals
do.
It may be a truism to say this, (and my ancient Shorter OED
defines truisms as statements which are obviously true but
often of limited importance) but our ability to engage in
general-purpose computation must have began to appear, like
any other biological system, as soon as it became both beneficial
and feasible for natural selection to assemble it. Having
made that appearance, it has, however, given our species the
power to transform the biosphere so profoundly, that no other
organism on this planet may get the opportunity of evolving
it again.
* * *
If the unprecedented mental power of our species is derived
from the “general purpose computers” we’ve
been speaking about, then language has provided the “modems”
which allowed those computers to increase their power by exchanging
information. General-purpose intelligence, language, and technology
are so strongly interdependent in our species, that it’s
hard to imagine any one of those three things attaining the
power it presently possesses, in isolation from the other
two.
As we saw in Chapter 9, the earliest hominid tool-making
populations, going back to 2.5 million years ago, were dominated
by right-handers. That, we concluded, suggests that the brains
of at least some of the late australopithecine species may
already have displayed the “strong lateralization”
in which one hemisphere is notably larger than the other.
As we saw, too, in that chapter, the presence in one of the
hemispheres of the modern human brain – usually the left
one – of a large and extensive “language archipelago,”
appears to explain why that hemisphere is larger than the
other. We noted, also, in Chapter 9, that the 1.8-million-year-old
KNM-ER 1470 skull from Lake Turkana, Kenya, midway in size
between the skull-capacities of the australopithecines and
erectus, displays a noticeable degree of brain lateralization,
as well as a concavity on its inner surface which could indicate
that Broca’s area – an important “island”
in our species’ “language archipelago” – had
already become enlarged in this kind of hominid.
The anatomy of the bottom of several Homo erectus
skulls suggests, finally, that the descent of the human larynx – a
development which has given our tongues room to move up and
down, and side to side, to create resonance chambers capable
of producing a large number of vowel sounds – may already
have begun almost two million years ago. Language of some
kind may, therefore, already have developed by the beginning
of the Pleistocene or earlier. That’s not an outlandish
conjecture. It would be startling, on the contrary, if a biological
system as complex and powerful as the human ability to communicate
via language had taken only thousands, rather than millions
of years to evolve.
Technology, the other concomitant of our species’ “general-purpose
computing machinery,” unquestionably made an early appearance
in the evolutionary history the hominid family. As we’ve
seen in several contexts, hominids living in Ethiopia’s
Middle Awash region some 2.5 million years ago, identified
and carried with them the “isotropic” rocks that
are suitable for making tools, knocked cutting-flakes off
them thereafter, and then used those flakes to butcher the
carcasses of horse-sized animals.
* * *
For a long time hominids must have lived – like the “clever”
animals and birds we spoke about earlier – on the edge
of the terrain of invention and discovery. By 2.5 million
years ago, however, the tools and tool-marked bones of the
Middle Awash show clearly that at least one hominid species
had entered that terrain. Some time would go by before that
new faculty would give our family the ability to “explode
into new habitats,” and “cause the extinction
of many prey species,” but hominids may already have
caused the disappearance of one kind of prey animal before
2.5 million years ago.
Several species of giant tortoise were living in Africa at
the beginning of the Pliocene, 5.3 million years ago. By about
3 million years ago, giant tortoises had disappeared from
that continent. The late Wilhelm Schüle (1929-1997),
an archeologist who was affiliated with Freiburg University’s
Institut für Ur- und Frühgeschichte, argued that
hominids were responsible for that disappearance. Schüle’s
investigation of the disappearance, around eight thousand
years ago, of giant tortoise species from islands in the Mediterranean,
convinced him that those giants had become extinct soon after
humans had first reached the islands on which they’d
been living. The Mediterranean disappearances were, he realized,
part of a world-wide tendency for giant tortoise species to
become extinct soon after humans or hominids reached the continents,
regions or islands which those reptiles inhabited. He concluded
from this pattern, that the disappearances of the giant tortoise
species which had lived in Africa during the Miocene and the
earlier Pliocene, were likely connected to the rise, on that
continent, of the human family.
The extermination of those African tortoises would have taken
place, Schüle reasoned, when hominids “adopted
a more carnivorous diet during the Upper Miocene or Lower
Pliocene.” While I agree with Schüle’s idea
that hominids were responsible for the disappearance of the
African giants, I don’t believe that those disappearances
were caused by the rise of, or by an intensification of, meat-eating
in our family. As we saw in Chapter 10, our family probably
became carnivorous even before it split off from the chimpanzee
line around seven million years ago. The disappearances of
the giant tortoises of Africa resulted, in my view, from an
“ontogenetic ambush” constituted by the relatively
abrupt discovery, by a family which had already been carnivorous
for millions of years, that the carapaces of large tortoises
could be smashed open with rocks.
* * *
At least two genera of giant tortoises inhabited early-Pliocene
Africa: Stigmochelys and Centrochelys. Contemporary
estimates suggest that the biggest members of these genera
might have exceeded six foot in length, and reached a weight
of about 800 lbs. It’s possible, too, that Pelusios
and/or other genera of African terrapins (i.e. fresh-water
turtles) may also have grown much larger in at this time than
they presently do. (France de Lapparent de Broin of the Département
de la Terre du Muséum d’Histoire naturelle in
Paris, provided me with invaluable personal communication
and published materials in relation to the tortoises and terrapins
discussed in this chapter. Dr. de Lapparent de Broin is not,
of course, responsible for any mistaken conclusions I might
have drawn from the publications to which she referred me.)
In India, a member of the genus Megalochelys (named
in 1837 by Falconer and Cautley from Plio-Pleistocene deposits
in the Siwalik Hills) measured up to ten feet from the front
of its carapace to the back, and reached some 2,000 pounds
in weight. This monster survived, it seems, somewhat longer
than Africa’s giants – it was still around in the
early Pleistocene. Other giant tortoise species – possibly
also members of Megalochelys, although their taxonomy
has not yet been sorted out satisfactorily – inhabited
Sumatra, Java and other East Indian islands in the early Pleistocene.
(As we’ll see presently, hominids entered the Southern
parts of Asia shortly before the Pleistocene began some 1.8
million years ago.) Giant tortoises living on Flores Island,
which wasn’t joined to the Asian mainland by the falling
sea-levels of the Pleistocene glaciations like nearby Java
and Bali were, survived until about 800,000 years ago, when
Homo erectus managed to settle Flores by crossing
the water-barrier – ten to fifteen miles wide during the
height of the glaciation – which separated it from the
mainland.
Evolution enlarged the front and back shell openings of the
extinct “saddle-back” species from Rodrigues island
in the Mascarenes, and of a few varieties of the Galapagos
species. This increased the reach of their heads for browsing
purposes, and gave their legs more mobility. Species with
modifications of this kind could have been vulnerable to large
continental predators like lions, bears and hyenas. Giant
tortoises such as those living on the African, Eurasian and
American continents, which retained “normal-sized,”
shell-openings (i.e. relatively small ones), would, on the
other hand, have been secure from predation, despite the fact
that they would not have had the power to flee from, or harm,
a would-be predator.
They would, however, only have had one line of defense against
animals wanting to gain access to their meat – the fact
that they could withdraw into an impregnable carapace. “I
was always amused,” Darwin wrote about the Galapagos
giants,
...when overtaking one of these great monsters
as it was quietly pacing along, to see how suddenly, the
instant I passed, it would draw in its head and legs, and
uttering a deep hiss would fall to the ground with a heavy
sound, as if struck dead.
Exclusive reliance on this defense would, sooner or later,
have made these animals vulnerable to a predator who was starting
to explore the world of ontogenetic innovation. By exposing
a rich store of meat to view, and protecting it only by enclosing
it in a hard structure, nature was setting the Pliocene predators
of Africa the same kind of intelligence test that Berndt Heinrich
set his ravens in Maine, and primatologists set when they
hang bananas from ceilings to see if chimpanzees can reach
them by manipulating boxes and poles. As long as no member
of Africa’s Pliocene predator-guild could pass that
test, that continent was as viable an environment for giant
tortoises as any oceanic island.
Chimpanzees have taken the first steps (but only
the first steps) toward developing the skills needed to smash
the carapaces of tortoises. Some chimps have been seen to
break open the shells of hard nuts such as those of Panda
oleosa with rocks, after placing them on other rocks or on
tree-roots which serve as “anvils.” Females are
generally more adept than males at acquiring this behavior,
and not all members of the bands in which this behavior has
been observed can, apparently, do so. Cruder forms of “percussive
technology,” like beating hard-shelled fruits against
tree trunks, have been seen throughout the chimpanzee’s
range, but the “hammer and anvil” method of breaking
open nuts has been seen only in West African populations.
It seems to represent, therefore, a kind of “high-water
mark” in chimpanzee tool-use, practiced only by a few
relatively gifted animals in a localized area.
No chimp has, as far as I know, extended its nut-breaking
skill to smashing open the shells of tortoises or turtles.
Baboons in South Africa’s De Hoop nature reserve have
been seen to eat tortoises belonging to the small Chersina
angulata species, but they use their hands and teeth
to open the relatively fragile shells of that species’
juveniles. It seems safe to say that breaking open a giant
tortoise’s armor – a task which would have required
the purposeful, two-handed manipulation of relatively large
rocks for an extended period of time – is very far beyond
the capabilities of any primate outside the hominid family.
It’s highly likely, on the other hand, that this task
was already within the power of beings who had developed
the ability to use hammerstones to knock sharp cutting flakes
off other rocks, and use those flakes to butcher horse-sized
mammalian carcasses. Giant tortoises probably become vulnerable
to our family, in fact, before any of its members learned
to make stone cutting-tools. As we saw in the previous chapter,
hominid tool-use – including the use of unmodified rocks – must
have come into existence long before our family was manufacturing
stone implements.
We saw, too, in that chapter, that our ancestors have probably
been eating meat since some time before the hominid family
started its separate existence. We also know, from archeological
finds in Africa, Asia, Europe and the Americas, that tortoises
were a valued source of meat for hominids throughout the Pleistocene.
The relatively fatty flesh of those animals would, pound for
pound, have been richer in calories than that of most other
available meat. Judging by accounts written in our era, it
would probably have tasted good too. William Dampier, an English
pirate-naturalist based on the Galapagos during the seventeenth
century, wrote that the meat of the tortoises of those islands
was “...so sweet, that no pullet eats more pleasantly.”
* * *
Hominids probably started their tortoise-eating career by
opening the shells of very young tortoises with their teeth,
the way present-day baboons do. From there, they might have
started breaking open the carapaces of somewhat larger individuals
with “crude” percussive technology, i.e. by smashing
the tortoise itself against a rock, the way present-day chimps
smash hard-shelled Strychnos fruits against tree-trunks.
It probably took a relatively long time for them to move from
that method to the point where they began to use rock “hammers”
to break into the armor of the biggest tortoises.
Although the ontogenetic innovations can literally arise
overnight, the mental machinery which makes such discoveries
and inventions possible must, as we’ll see in Chapter
13, have been constructed, by natural selection, over thousands
of generations. The hominid ability to break open the carapaces
of giant tortoises and terrapins with tools would, therefore,
have manifested itself in a relatively gradual way.
One or more of Africa’s giants might, therefore, have
had time to evolve a smaller body-size in response to hominid
predation. Size-reduction can – if a species has time
to undergo it – be a very effective countermeasure against
the threat of human-caused extinction: small tortoise species
are, as we saw in Chapter 1, much less vulnerable to such
extermination because they can reproduce more rapidly, exist
in larger numbers, and hide more effectively.
I think it’s likely, for these reasons, that Africa’s
biggest present-day tortoise, Centrochelys sulcata,
the spur-thighed tortoise of the Sahel (which can weigh up
to 200 lb in exceptional cases) and/or Stigmochelys pardalis,
the leopard tortoise of South-Eastern Africa (which normally
reaches 40 lb, but has been known to get to 90) are dwarfed
descendants of the giant Centrochelys and Stigmochelys
species of the Pliocene. Size-reduction among Africa’s
tortoises probably continued during the Pleistocene, even
though the giants had already disappeared by the beginning
of that epoch. In their article “Middle and Later Stone
Age large mammal and tortoise remains from Die Kelders Cave,”
published in 38 (1) 2000 Journal of Human Evolution,
Richard Klein and Kathryn Cruz-Uribe report that “[t]he
tortoises tend to be much larger in the MSA [Middle Stone
Age] layers than in the LSA [Later Stone Age] ones...”
Although some or all the the African giant tortoises and
terrapins may have escaped extermination at the hands of our
family by dwarfing, there’s no doubt that the fate of
the overwhelming majority of giant tortoise species exposed
to our species has been outright extinction. The giants which
still lived on dozens of islands after their continental counterparts
had disappeared, are only represented today by eleven subspecies
of Geochelone nigra, found in the Galapagos, and
by two or three closely-related members of the genus Dipsochelys
which survive on islands forming part of the Republic of Seychelles.
At the time of the birth of Christ, a rich radiation of giant
tortoise species still extended across Madagascar, the Comores,
the main Seychelles islands, the Aldabra group, the Glorieuses,
the Amirantes and the Mascarenes. The surviving Indian Ocean
species are geographically and genetically widely separated
from their Galapagos counterparts, but they reach approximately
the same size: their carapaces are about 4 feet long measured
over the curve, and they commonly reach 500 lb. in weight.
Esmeralda, who lives on Bird Island near the main Seychelles,
(and happens, incidentally, to be male) is often claimed to
be the world’s largest tortoise. He’s thought
to be around 180 years old. In 1989 his carapace was five
feet ten inches long, and he weighed 657 lb.
The reason why tortoises inhabited so many islands in the
planet’s seas and oceans is that they can remain afloat
and alive for extraordinary periods of time without access
to food or fresh water. Their longevity means, moreover, that,
after making landfall on an island, they can wait around for
a century or more for a mate to be washed up by the same current
that brought them to their new habitat (although many colonizations
probably started by the arrival of a single female carrying
fertilized eggs). Some of the tortoise species which colonized
islands in this way, may not have been giant-sized when they
first arrived – enlargement of their species would have
taken place, instead, in a relatively rapid way, after their
arrival. On relatively small islands, factors such as reduced
predation and a limited food supply, lead to a relatively
rapid size-increase in small species, and a decrease in the
size of big ones like elephant or deer. The hobbit-sized descendants
of Homo erectus which hunted Jack Russell-sized rats
on Flores island show that the human family wasn’t exempt
from the workings of this dynamic.
* * *
Could climate-change have killed off the planet’s giant
tortoises? There’s no question that many tortoise species,
large and small, were exterminated by this agency. The giant
Meiolania tortoises and the side-necked turtle species
that inhabited Antarctica early in the Cenozoic or “age
of mammals,” and the soft-shelled turtles that lived
on Greenland and on other islands in the vicinity of the North
Pole at that time, were clearly victims of the present ice-age.
Many tortoise and terrapin species would not have had the
option of retreating away from the poles to escape the temperature-drop
that the present ice age was bringing to the higher latitudes.
The fact that giant Cheirogaster tortoises disappear
from Northern Europe by about five million years ago, but
survive in the southern regions of that continent until at
least 2 million years ago, must also be the result of high-latitude
cooling. It’s hard to understand, however, how climate-change
could have exterminated Cheirogaster in the southern
parts of Europe. Small tortoises, are, after all, still widespread
and relatively abundant there, and there’s no evidence
that large tortoise species would have been more vulnerable
to cold than these surviving species are. Even though giant
tortoises drop off the paleontological radar screen in Southern
Europe around two million years ago, it’s not inconceivable,
therefore, that they could have survived there, until hominids
made their first appearance in that region a million years
ago or more.
It’s very unlikely that climate change was responsible
for the disappearance of giant tortoises whose ranges included
the relatively warm regions of Africa, Eurasia and the Americas.
As we’ll see in the appendix to this book, those regions
became only marginally cooler during the cold phases of the
glacial cycles. As we’ll see, too, in the appendix,
the biomes inhabited by tortoises or terrapins – whether
jungle, savanna or desert – grew, shrank and/or shifted
in response to those cycles, but none of them ceased to exist.
The Pliocene disappearance of the African giants was not,
moreover, accompanied by the extinction of large tortoise
species on any of the other warm-climate land-masses: the
South Asian species only disappeared, as we’ve seen,
in the early Pleistocene, and the Americas did not lose their
giant tortoises until the end of that 1.8-million-year-long
Epoch. The two giant tortoise species that inhabited Madagascar,
whose climate is closely tied to that of Africa, survived
until after the birth of Christ. Nor was the disappearance
of the African giants part of any wider extinction-spasm in
Africa itself, whether of tortoises or any other organism.
* * *
The only direct evidence we have of humans exterminating
giant tortoises is found in the historical accounts of tens
of thousands of tortoises being killed for food by sailors
in the last few hundred years in the Seychelles, the Mascarenes
and the Galapagos. Taking advantage of these animals’
ability to live without food or water for extended periods
of time, sailors would bring large numbers of them on board
alive, often stacking them upside-down in the ships’
holds to kill them later as their meat was required.
The evidence implicating our family in the giant-tortoise
extinctions which took place before this time is circumstantial,
but that doesn’t necessarily mean that it’s insufficient.
In a classic 1894 case which helped to build the Anglo-American
law of evidence, Makin v. Attorney General of New South
Wales (an appeal from an Australian court to the British
Privy Council), a man and his wife were accused of murdering
a child whose mother had paid them to take him into their
care. The child’s corpse had been dug up in the yard
of a house they’d been renting. The Makins’ defense
was that the child had died naturally, and that “if
they were guilty of anything, it was merely of having improperly
buried the child.” That plea might well have succeeded
if the court hadn’t admitted evidence that no less than
eleven corpses of other children had been dug up in the yards
of various houses occupied by the accused.
Without being told about the other eleven corpses, a jury
might have decided that there was, say, a one in ten chance
that the child in question could have died naturally. That
would probably have been enough to constitute, in the minds
of the jurors, the “reasonable doubt” that the
criminal law requires for an acquittal. Add one more corpse,
however, and the one in ten chance that I’ve assigned
to the possibility of a single child dying naturally, becomes
one in a hundred. Dig up a further ten, and it reaches one
in 1,000,000,000,000. The Privy Council had no trouble affirming,
at any rate, that evidence of the other eleven corpses was
admissible. The law is normally cautious about admitting evidence
of previous wrongdoing with which the accused wasn’t
charged, (“similar fact evidence”) because of
its potential to prejudice the jury against the accused unfairly,
but such evidence can be admitted, the Council decided, if
its probative value outweighs that potential for prejudice.
In the Makin case, the probative value of twelve corpses
was, as we’ve seen, overwhelming. The frequent recurrence
of the “hominids arrive, giant tortoises disappear”
sequence which impressed Wilhelm Schüle, establishes
a pattern whose probative power is, in my view, as irresistible
as the “Makins take in children, children die”
sequence. The former sequence manifests itself for the first
time when hominids arrive in South Asia around two million
years ago. It is repeated when Homo erectus reaches
Flores island 800,000 years ago, and re-appears when Homo
sapiens reaches Australia some 50,000 years ago. It is
seen again when humans enter and settle the New World around
15,000 years ago; when they reach the previously inaccessible
islands of the Mediterranean sea between 7,000 and 8,000 years
ago; when they reach the Caribbean islands some 6,000 years
ago; when they make it to the Canaries off the north-west
coast of Africa, presumably at about the same time; when they
sail to New Caledonia island, some eight hundred miles east
of Australia about 3,500 years ago; when they settle Madagascar
around 1,500 years ago; and, finally, when Arab, Persian,
Portuguese, Dutch, French and English ships start making landings
on the last undiscovered Indian Ocean islands between the
time when Madagascar was settled, and the seventeenth century.
These disappearances establish beyond any reasonable doubt,
that the hominid family was responsible for the extermination
of the vast majority, if not all, of the giant tortoise species
which inhabited territories newly settled by its members.
They show clearly, therefore, that giant tortoises, with their
single line of defense against predation, are enormously vulnerable
to extermination at the hands of our family. That oft-demonstrated
vulnerability, coupled with direct archeological proof that
hominids have found tortoise meat a tasty, easy-to-utilize
food-package for nearly two million years, constitute a strong
indication that the giant tortoise species which disappeared
from Africa in the later Pliocene were exterminated and/or
dwarfed by hominids, as the result of an early push by the
latter into the terrain of ontogenetic innovation.
CHAPTER 12–
Still-marvelous
but significantly reduced
