W. H. Calvin's THE ASCENT OF MIND (Chapter 8)
Home Page || Public Bookmarks || Science Surf column || The Calvin Bookshelf
A book by
William H. Calvin
The Ascent of Mind (Bantam 1990) is my book on the ice ages and how human intelligence evolved; the "throwing theory" is one aspect.
   My Scientific American article, "The emergence of intelligence," (October 1994) also discusses ice-age evolution of intelligence. Also see Wallace S. Broecker, "Massive iceberg discharges as triggers for global climate change," Nature 372:421-424 (1 December 1994) and his "Chaotic Climate" Scientific American article (November 1995 issue).
AVAILABILITY is challenging.
Many libraries have it (try the OCLC on-line listing), but otherwise it’s strictly used bookstores (and German and Dutch translations).
The Ascent of Mind
Ice Age Climates and
the Evolution of Intelligence

Copyright ©1990 by William H. Calvin.

You may download this for personal reading but may not redistribute or archive without permission (exception: teachers should feel free to print out a chapter and photocopy it for students).



The Ambitious Ape's Guide to a Bigger Brain

Progress in science is achieved in two ways: through new discoveries, such as x-rays, the structure of DNA, and gene splicing, and through the development of new concepts, such as the theories of relativity, of the expanding universe, of plate tectonics, and of common descent. Among all the new scientific concepts, perhaps none has been as revolutionary in its impact on our thinking as Darwin's theory of natural selection.

the evolutionary theorist Ernst Mayr, 1988

Fidalgo Island is the northern sister to Whidbey, but considerably more mountainous than glacial till ought to be: Mount Erie is a hundred rocky stories tall, and would have made an excellent easy-to-climb lookout for the Indians. Its view is of the Skagit River delta to the southeast, where great flocks of migrating birds winter, over to the mountainous Olympic Peninsula in the southwest, the San Juan Archipelago and Vancouver Island to the west -- and of course the volcanic Mount Baker to the north and Cascade Mountains stretching to the east. You can almost see out to the Pacific Ocean from atop Mount Erie, survey the entire domain of those Indians who lived around here. There were megafauna hereabouts, probably hunted during the meltback by newly arrived Indians. Looking southwest over to the Olympic Peninsula, I can see the area where a mastodon skeleton was found.
      Did the hunters use this viewpoint to spot pods of whales cruising around? Hunting whales might have been their variant of big-game hunting, once the mammoth and mastodon disappeared from the glacial grasslands about 10,000 years ago. The whale hunt would have demanded even more organization and cooperation and planning ahead than the mammoth hunts, what with keeping those hollow logs in seaworthy condition.
      Individually digging up clams and snagging salmon in streams would have sufficed much of the time. But inventing boats would have meant a boom time, just as surely as if the climate had dramatically improved.

BOOM-TIME PSYCHOLOGY is somewhat familiar to us, as I mentioned earlier (bullish stock market speculation, decline in savings, increases in borrowing, more risk taking everywhere -- even higher hemlines and more daring décolletage in womens' fashions). Boom-time reproductive physiology is more obscure.
      Until we know what the boom time's proximate mechanisms are -- what aspect of the environment that even children are sensing, what hormones they use to implement the change -- and what that mechanism's disadvantages might be, we won't be able to adequately evaluate the proposition that boom times alone could work the ratchet to yield ever-larger brains and ever-smaller teeth. The mere fact that the succession of boom times started 2.5 million years ago with the ice-age melt-off cycles, and that this nicely overlaps the period of hominid encephalization, speaks in its favor.
      In evolutionary arguments, it is no longer enough to demonstrate that something could have done the job, given enough time. By compound-interest reasoning, any slight advantage can eventually do the job. There are usually multiple ways to do the job, and the one that gets there first on the fast track tends to preempt the niche. The speed of the cycle is always important -- and especially with this proposed encephalization cycle, simply because the four-fold increase in hominid brain size in only 2.5 million years is "almost unbelievably fast" (in the words of Ernst Mayr) by the standards of natural selection. We need fast tracks, even if slow tracks might have sufficed.
      This rapidity provides an important constraint on proposed explanations for what happened since the australopithecines: most proposals for how hominid encephalization evolved are too leisurely to explain the Great Encephalization. Spanning the same period as the fourfold encephalization are the evidence for the ice ages, for prolific toolmaking, and for hominid hunting. Might toolmaking or hunting do the trick?
      Tool use is, of course, shared with quite a number of animals including the birds. The fancier types of toolmaking during the ice ages are mostly associated with hammering techniques -- but hammering isn't unique to hominids either. Female chimp nut-cracking involves surprisingly sophisticated skills in positioning and grading of delicate blows; one wonders how much early hominids needed to improve on our common heritage in order to produce the early toolkits that sufficed until the hand ax.
      It is also hard to see what the growth curve for toolmaking was like during the relevant period. Remember that the brain size of Homo erectus doubled during a period when the dominant toolkit persisted without major improvements, a period of a dozen ice ages starting 1.5 million years ago; this is hardly suggestive of man-the-toolmaker being the driving force behind hominid evolution during this period.
      So what about hunting?

MANY SKILLS ARE IMPORTANT for human hunting: detecting the prey, outsmarting and outmaneuvering it, and killing it while avoiding injury. Many other human attributes facilitate our hunting endeavors: our social organization, reproductive strategies, and communications skills.
      But carnivores are clearly experts at outsmarting and maneuvering. Snatching the defenseless young hiding in the underbrush, or outrunning small mammals, is also practiced by both baboons and chimpanzees. Both hunt cooperatively, baboons chasing gazelles into the arms of a fellow baboon, chimps moving to guard escape routes and attempting to draw off adult pigs so that other chimps can snatch their young. What used to be thought of as uniquely human hunting skills are often shared with other animals -- animals that haven't experienced rapid brain enlargement.
      What then are the uniquely human aspects of hunting, and what role did they play in hominid brain evolution? Clearly, hominids could have adapted chimp-like maneuvers to chase competing scavengers away from dead meat; it probably did not require a bigger brain to make this small modification to ape behavior. But, though I love to look at all those microscopic marks on teeth and bones, I tend to question scavenging as an important evolutionary path: the food chain would limit prehumans to populations similar to those of the existing top predator, if they made their living that way. That's not usually the way to a new niche or population boom.
      Projectile predation seems to be a form of hunting not practiced by other mammals in competition for the same resources: this "action at a distance" hunting is a very important invention. It reduces the chance of injury to the hunter, keeps one out of range of horn and hoof. From an evolutionary standpoint, throwing is not a one-step invention: it has aspects such as accuracy and length of throw that may be improved, time and again, for additional advantages, generating a long growth curve. The type of throw, the distance of the throw, the weight thrown, the accuracy of the throw, the suitability of the object thrown -- all can be improved again and again.
      Chimpanzees certainly throw, but the thrown object (usually a branch, sometimes a rock) is primarily used as a threat to chase off leopards or intimidate fellow chimps in dominance displays. The accuracy of the threat throw is largely irrelevant, so long as it threatens to generate a blink reflex by its high angular velocity. Chimps throw both underarm and overhand, using much the same postures and motions that human children utilize; only the occasional chimp attains a reputation among human observers as a thrower.
      While they obviously have the neural and musculoskeletal machinery for the basic throws, chimps may not have the precision timing neural circuits needed for accurate launch at distant small targets. No one has ever measured chimpanzee throwing accuracy, to the best of my knowledge. Were chimps even half as accurate as humans, however, I think we would have heard about it: they'd be the terror of Africa, and (given how they love meat) they'd be eating meat every day.
      Minimal accuracy (a "side of the barn throw" in baseball phraseology) might have limited hominids to throwing at large nearby targets -- but there weren't very many blind mammoths. So how did we get started? What bootstrapped hominid hunting? What might we recommend to an ambitious ape? One clue, in my opinion, is the earliest fancy tool.

ABOUT THE EARLIEST STONE TOOL of fancy design was the Acheulean hand ax. It's almost as fancy as the arrowhead (first seen during the last ice age but mostly in the 10,000 years since the melt-off). The Acheulean hand ax is far, far older: it was the most prominent feature of the Acheulean toolkit made by Homo erectus between 1.5 and 0.3 million years ago. It is found everywhere from the tip of Africa to Europe to South Asia, made of whatever local rocks were handy.
      There is only one problem: for more than a century, no one could seem to figure out what the Acheulean hand ax was especially good for. For archaeologists, it has been like one of those "What is it?" exhibits in the children's room at a museum, where the children attempt to guess what the covered pan on a pole was once used for. To preheat beds with coals from the fire is not a modern problem, what with other forms of heating; I'm not sure that our guesses about hand ax usefulness are much better than the children's guesses about the pan on a pole.
      Labeling the Acheulean creation a "hand ax" was certainly a major error, though the name has stuck anyway for various reasons. The sharpened edges of the typical hand ax continue all around its perimeter, and so would do a lot of damage to any hand that attempted to use a hand ax for chopping: it would, so to speak, bite the hand that held it.
      The archaeologists' fallback position is that perhaps it was used for separating meat from skin and bone. But a flesher is hardly an important item in a toolkit, since split cobbles work so well for the purpose already. A hand ax (especially one with a broken edge) could certainly do double-duty as a flesher, but some other function must account for its singular features:
      1) it is bilaterally symmetric,
      2) usually has a point,
      3) usually has a sharpened edge all the way around, and
      4) it is also usually flattened, something like a discus.

The exceptions are interesting. There are some with blunt back ends, just as there are some (called Acheulean cleavers) without a point. But they may simply be broken versions of the classic shape; that's the default position to take concerning such variants until they are shown otherwise.
      Surely we can do better than the position taken by some frustrated archaeologists: that it was a ceremonial item, functionless in the everyday sense of the word. "Form for form's sake" certainly exists, but it is subject to fads and fashions -- the Acheulean hand ax would have to be the all-time-record fad, extending over Africa and Eurasia for more than a million years! What use requires all of those four features, a use that would inhibit further variations in the usual manner, so that the design would remain stable for a very long time? It must be nearly perfect for some important task to achieve such an all-time-record for design stability.
      Because its shape is reminiscent of the spear point and arrowhead, there was an early suggestion (H. G. Wells mentions it in his 1899 book, Tales of Space and Time) that the hand ax was thrown at animals while hunting. This suggestion floundered because the back end of the hand ax is so unsuitable for attachment to a spear (hafting didn't appear until well after hand-ax days): the rear edge of a classic hand ax is carefully rounded and sharpened. Throwing it without a shaft seems a bit silly too: how would one keep the point oriented forward in flight? Any explanation for the function of the hand ax needs to explain that point, those all-around edges, that symmetry, that flattening.
This unsatisfactory state of affairs lasted until an intrepid undergraduate at the University of Massachusetts made a fiberglass replica of a big Acheulean hand ax and gave it to some varsity discus throwers to experiment with. Eileen O'Brien took her cue from a 1965 suggestion by a South African anthropologist, M.D.W. Jeffreys: that the smaller hand axes could be thrown with spin, perhaps into a flock of birds. The replica indeed spun well; that flattened shape and bilateral symmetry are very useful for setting a spin. O'Brien and her two athletic friends discovered a totally unsuspected aerodynamic property of their hand-ax replica: in mid-flight, it would turn on edge and land that way. Indeed, the hand ax would usually slice into the ground and bury its point. Now, as you probably recall from your own experience, having the Frisbee turn edge-on shortly after launch is something that happens to all inexperienced Frisbee throwers -- but those experienced discus-throwers couldn't keep it from happening. It seemed to come with the shape.
      And the tendency to land edge-on matches up with a previously puzzling aspect of the archaeology: hand axes are often found in dried-up ponds and lakes and creeks, sometimes standing on edge! This strongly suggests that hand axes were indeed thrown at animals visiting the waterhole to drink -- that hominids were practicing an old carnivore trick, lying in wait at the only waterhole.
      O'Brien's experiments were a major advance, but they left many questions unanswered: Waterhole predation ought to work with any old handy rock; the painstaking preparation of this rock seems excessive. Why the sharpened edges all around? If spin is nice, why not just use a flat slab of rock, broken to be symmetrical? The answer implicit in these experiments was that a "spinning ax" could do a lot more damage than a rock: by landing on edge (especially a sharpened edge), all of the force is concentrated on a thin edge. But why the point?

THERE THE MATTER RESTED for nearly a decade; I had to puzzle over it for four years before I stumbled upon an interesting clue. It seemed to me that the hand axes were not being thrown at individual animals but at whole herds. Teaching introductory biology for the first time while writing The River that Flows Uphill had reminded me of why animals cluster into herds or schools: to protect against predators.
      As herd size increases, there are more individuals on the periphery of the herd exposed to predators -- but the average animal is safer. The percentage of the herd on the periphery will drop as the herd size increases. That's why there is "safety in numbers." For a small herd, half are exposed on the periphery; tenfold larger, and most of the herd is protected inside that vulnerable outer ring. To a physiologist, this is just another surface-to-volume ratio problem of the kind familiar from thermoregulation, from why an animal needs a circulatory system to move oxygen around, if larger than the size where diffusion suffices.
      But lobbing a rock up over, and thus into, a herd gets around this restriction of only the peripheral ones being vulnerable; you circumvent a two-dimensional design with a lob into the third dimension! Furthermore, herds cluster ever more tightly together when feeling threatened -- which would only make matters better for the hunter lobbing rocks into their midst as fewer rocks would fall between animals. Even when you miss, it's easier next time!
      You aim at the herd, not any one individual animal: it is a "side of the barn" throw rather than a precision throw. And knowing what I did about how hard it was to throw with precision, I thought that lobbing into herds was likely to be a good entry-level technique for the beginning hunter. Invention in behavior tends not to be the "light bulb" flashing on, the bright idea after contemplation -- it tends to be an old way of doing things, converted to a somewhat similar task, one that turns out to hit upon something valuable. After this invention, adaptations streamline the behavior and eventually the body style itself. Chimps can probably throw well enough to hit a herd, though probably not with sufficient consistency to hit an isolated animal from any distance (and no second chances: the animal runs away after the first launch).
      There is just one problem with hitting a herd animal in this way: most lobbed rocks that strike it would hit its back and bounce off -- an unlikely way to kill an animal. On the rare occasions when a rock hit the animal on its head or spine, it might have conveniently collapsed -- but otherwise the hunters would likely be left with an angry animal running away, with a good head start on the pursuers. Even if knocked down, the animal could likely have gotten up and run away before pursuers arrived.
      Ah, but when I thought about it some more, I realized that if the animal should be knocked down, it might be further injured by its fellow herd animals -- they would stampede when the hunters launched. Even if the herd didn't trample the injured animal, they would delay it getting back on its feet. This might give the hunters time to run up and club the animal, or perhaps throw stones from up close at its head.
      I was especially impressed with this scenario when I realized that there was a perfect transition from known behaviors of chimpanzees: while chimps do throw rocks, my primate ethologist friends tell me, they particularly like to throw big tree branches after flailing them around furiously. Such a branch, lobbed into a herd lapping up the lake at sunset, would land just as the herd was wheeling around and starting to run away -- so it would often trip an animal or two, expose them to trampling by the rest of the herd, delay them enough so that the hunters could corner them and polish them off. If chimps lived among herds of grazing animals, the more patient chimps could easily practice such a technique. If they ran out of branches, they would probably throw their other favorite projectile, big rocks.
      If that's the way hominids got started hunting, how did they ever arrive at a fancy scheme such as making Acheulean hand axes? What is it about flattened bilateral symmetry, a point, and sharpened edges all around? So I decided to fiddle around with throwing hand axes.

I TOO FINALLY ENLISTED THE AID of an experienced discus thrower, Gareth Anderson, and we repeated the O'Brien experiments with five crude hand axes from southern Algeria and a fiberglass replica of a fancy flattened one. They all exhibited the same aerodynamic peculiarity as the giant replica that O'Brien tested: they tended to land on edge, even if thrown horizontally like a Frisbee.
      Gareth and I had picked a well-worn soccer field for this experiment; it had close-cropped grass and many worn spots, and the ground had been softened up by a Seattle drizzle the day before. So when a hand ax landed and then bounced away, we could see the gouge it left behind. Gareth would retrieve the hand ax and bring it back to fit into the hole in the ground, trying to figure out its orientation when it landed. And because of packed dirt adhering to the hand ax, we could usually see the place along the perimeter of the hand ax that hit the ground first -- and it was no preferred place. Since the hand ax was spinning, it rotated after impact and the point eventually poked into the ground. Sometimes the point would snag the ground and impale the hand ax, just as in the O'Brien experiments. Thus the point helps stop the hand ax -- meaning that, in the case of an animal target, it would cause the animal to stagger much more than when the rock merely bounced free.
      So if the soccer field were instead the back of a zebra or gazelle, the projectile would no longer bounce off their backs like a rock would -- but rather transfer most of its forward momentum to the animal. The animal might not be able to right itself in time, before collapsing, due to an interesting neurological peculiarity: injury to the back in a four-legged animal causes the legs to flex, as when an animal scrapes its back on an overhanging tree branch or rock and the hindquarters hunch down to free the skin from the sharp obstruction. To keep from collapsing sideways after a hand ax impact on its near side or its back, the animal needs to extend its legs on the far side -- but the back injury from the sharpened edge of the hand ax would tend to make it flex the hindlimbs instead. Thus the reflex protection against toppling would be countermanded.
      And that's when the pointed front end of the hand ax finally began to make some sense. It would spin around and tend to bury itself in the skin (or snag a roll of skin pushed up by the forward motion of the hand ax landing). This would not only transfer much of the hand ax's forward momentum to the animal -- but it would yank on the just-incised skin.
      A clean cut of the skin is not necessarily painful if you're busy with something else, as I discovered myself one night as a child playing hide-and-seek after dark: I got a big cut on an ankle (from the nearly buried stump of a newly sawed-off bush) that I didn't notice until my mother complained at me ten minutes later, for tracking something red into the house and across the carpet. One of the things that amazes medical students during their first duty in the hospital emergency room is how many patients with a bad cut or scrape (and even broken bones) will claim that it doesn't hurt (someone finally compiled some statistics: 37 percent claim no pain for several hours after injury, though almost everyone hurts a half day later).
      But what is guaranteed painful is to manipulate the cut skin edges (just ask a surgeon: they can often continue operating after local anesthesia wears off, so long as they don't touch the skin edges; when they start to place stitches is when the patient requests a booster dose). The spinning hand ax, incising the skin and then snagging its point to yank on the new incision, ought to produce a powerful withdrawal reflex that lowers the hindquarters. Even a small hand ax might cause enough sharp pain to make a big animal suddenly collapse. If the animal were standing alone, it might still get up in time to run away from the approaching hunters -- but with a herd stampeding past, just being knocked down might prove fatal.
      And so lobbing branches and then rocks into herds visiting waterholes looks like a good way to make the transition from chimpanzeelike behaviors to hominid hunting -- without improving the brain's timing abilities at all. That's the basic invention for hunting. Making a "spinning-snagging ax" (as we ought to rename the hand ax, though I suspect that "killer Frisbee" will win out!) probably doubled and tripled the yield, permitted hominids to graduate from small gazelles (for whom a thrown rock might have sufficed) to the larger herd animals such as zebra. To make further improvements beyond that, you have to improve throwing accuracy so that you can hit small herds or single animals.
      Note that the lobbing technique won't work against anything except targets that are tightly packed together, at least not until accuracy improves quite a lot. That's why I don't think that this invention was important for aggression within a hominid species. Yes, a tendency toward mayhem probably existed in our common ancestor (newly-installed silverback male gorillas practice infanticide, and chimps savagely beat up "enemy" chimps), and yes, accurate throwing would have allowed intermediate prehumans additional ways of committing mayhem. Attacking one another is definitely a potential way of shaping up prehumans to be bigger and better fighters -- which, judging from the history of warfare, may well have played some role at some point in the ape-to-human transition. But the shift from gathering-snatching-scavenging to successful waterhole hunting was not a major step along that path; it was instead a major step in food acquisition that would not work well against fellow hominids (unless as tightly packed as a herd!). And this invention was probably of "new niche" proportions, the sort of thing that can create a new species and spread them around the continents.
      What might any of these aspects of side-of-the-barn throwing have to do with juvenilization? Certainly they might produce boom-time conditions as they increased the hominid population size that could be supported. But I think that precision throwing came later, and that it has a much better tie with juvenilization.

PRECISION THROWING is what children work up to, as they develop their throwing skills, starting with the high-chair food fling, developing into the kindergartner's unaimed sidearm lob, and gradually progressing to the overhand direct trajectory that can reliably hit a small target, a technique mastered by elementary-school-aged children.
      At Laetoli, Tanzania, Mary Leakey found rocks that appear to have been carried in from outcrops a good hike away; these 2-million-year-old manuports (from "hand-carried") would seem suitable for threat throwing, as in warding off scavengers from a kill or butchery site. But it is also obvious that, at some point long before baseball, our ancestors began throwing apple-sized stones with accuracy. Barbara Isaac has surveyed museum collections, finding various examples of rocks that may qualify as ancient throwing artifacts, smoothed and with thumb grips, etc. While they could have been merely used for threat throws (they aren't heavy enough for side-of-the-barn throws into waterhole herds), they tend to suggest precision aimed throwing, something closer to the modern style where one "gets set," launches with care, and practices the technique with small variations.
      Precision aim has a much better growth curve than does the waterhole lob. Hitting the head of the prey is an obvious improvement. Maintaining accuracy while standing farther away is important, not only because of the "approach distance" of prey animals (the distance at which they decide you've come close enough and move away), but because throwing twice as far (using a relatively flat trajectory rather than a high-angle lob) means throwing about twice as fast. This creates a bonus: the projectile arrives with as much as four times the kinetic energy (or "stopping power"), enabling ever larger animals to be felled with precision throwing techniques.
      There are growth curves in materials as well as technique: graduating to the spear, boomerang, and other throwing sticks such as the knobkerrie. But both faster throws and more accurate throws are always better and better, provided that the brain can cope with the more precise timing requirements for letting loose of the projectile. Both of these timing-dependent throwing aspects make considerable demands on brain reorganization, as most brains are incapable of fancy timing.

THERE IS A BOTTLENECK that needs to be overcome in order to throw with more-than-an-ape's accuracy. A crucial skill is accurately timing the moment when the hunter lets loose of the projectile. Release too soon and the rock lobs too high, lands behind the target. Release too late, and it hits the ground in front of the target. The "launch window" is the range of useful release times; it shrinks to submillisecond values for reasonable throws to rabbit-sized targets.
      The only known way of achieving such one-millisecond-in-a-thousand timing precision with jittery neurons (individually no better than about ten-milliseconds-per-hundred) is to assign many timing neurons to the same task. The heart has the same problem: individual heart cells don't discharge anywhere as rhythmically as a heart; only when hundreds are massed together does the regular beat emerge. Applied to making timing more predictable for throwing, you have to wonder where the extra cells come from: this averaging technique is extremely "cell hungry." It is not something to be implemented merely by quadrupling the traditional brain area for muscle sequencing, the premotor cortex. Nor by tripling everyone's favorite candidate for a precision delay-line timing device, the cerebellum. We are talking of hundred- and thousandfold increases in the numbers of brain sequencing circuits that need to be temporarily synchronized.
      It reminds me of expanding the choir to include the whole audience, when singing the Hallelujah Chorus. I suggest that the only practical way for the brain to achieve such numbers is temporarily to synchronize large areas of cerebral cortex, utilizing the widespread intracortical connections between the various areas. Neurons, especially those outside the traditional sensory cortical "receiving areas," do not seem committed to single functions; they enjoy widespread inputs from multiple sensory modalities. This generalized wiring suggests that neurons can be "borrowed" from their primary task (if, indeed, they have one). In such a manner, "getting set" to throw may serve to assign many neurons to a choral-like parallel assembly; after they function briefly in tandem, to determine the moment of projectile release, most are presumably unhitched from this temporary duty and return to their regular assignments. The Darwin Machine outlined in Chapter 2 would provide a simple way of sorting through the different throwing options while ending up with many clones, handy for the choral performance needed for precise timing.

DOES PRECISION have anything to do with juvenilization? The answer is a conditional yes, based on such "Law of Large Numbers" arguments. Because primate neocortex exhibits a tendency to eliminate some of its widespread interconnections during postnatal development, there is a progressive reduction in synaptic connections with age, a carving process that Daniel Dennett and J. Z. Young suggested a quarter-century ago. In addition to detaching synapses, reduced connectivity also occurs by neuron death in some cortical regions; a monkey's motor cortex loses a third of its neurons during infancy and the juvenile period (though very few during adulthood).
      Some individuals have a tendency to mature early; they might incidentally slow down these two carving processes, conserve widespread connections into adulthood. They might be better throwers, everything else being equal, able to recruit more cerebral assistants on the occasions when a lot of helpers were temporarily needed. Our big heads may be only an epiphenomenon of a developmental solution to the temporary synchronization requirement for accurate throwing, as our ancestors juvenilized to retain the more widespread intracortical connections of juvenile animals into adulthood (or, more likely, juvenilized during a boom time for the usual reproductive race, but didn't drift back later because throwing success kept the more juvenilized versions well fed as the climate worsened).
      Happily, neocortical pruning has the requisite long "growth curve" that the proximate mechanism would need if the three-part cycle is to be repeatedly used. The synaptic reduction curve peaks at eight months after birth in the current model of Homo sapiens, so further juvenilization in the future might allow ever larger assemblies for precision purposes.
      Obviously, that's not all there is to throwing (or infants might be the best baseball pitchers! Everything else often isn't equal). One trade-off is in motor skills, which are a prominent part of childhood development; all of the precise timing in the world won't do any good if the muscles downstream of the controller aren't up to carrying out the commands. Or the cortical commands go to more muscles than they should for precise movements; each corticospinal neuron makes connections to many levels of the spinal cord and thus many muscles, and these too are edited during postnatal development (though probably on a different time schedule than within-the-cortex connections).

WHAT THIS WORLD NEEDS is a beach with discus-shaped rocks -- which these islands seem to lack, though I keep looking. Sunbaked and windblown, I've been musing that the different kinds of waterfront correspond to the ages of humankind.
      Homo habilis (and probably the australopithecines as well) would have loved a shingle beach such as the ones on Cape Cod, with all those nicely smoothed throwing rocks that fit the hand, so handy for the darwinian toolmaking technique as well.
      For Homo erectus we need the discus-covered waterfront, the sort of place where you find good "skipping stones" these days. There were some lakes in the Sahara (well, at least during some Pluvial period in Homo erectus days) whose beaches were likely paved with genuine Acheulean hand axes. Apparently hand axes were lost in the mud, and sank even deeper as the worms churned the bottom sediments. As the lake expanded, the shoreline moved back -- and so new regions accumulated lost hand axes as well. After a while most of the lake bottom was paved with lost hand axes! Let the lake dry up and some surface dirt erode away, and you have exposed a sea of hand axes.
      Some sand dunes in southern Algeria have recently shifted and exposed exactly such ancient lake beds as my scenario hypothesizes: my archaeologist friend said that the whole lake bed appeared to be covered with hand axes. Presumably our ancestors could have mined such lake beds for ready-made hand axes -- they would have considered it Hand ax Heaven! For more recent inhabitants of the Sahara, those newly exposed tools for the taking would have been the Pleistocene equivalent of our oil wells and coal mines. Buried wealth, and long before fossil fuels.
      The waterfront symbolizing Homo sapiens is riprapped, covered with broken concrete and imported boulders, symbolizing both our abilities to look ahead to trouble during next winter's storms (but not far enough to build well back from the shore, for that once-in-a-lifetime storm). And symbolizing our tendency to pave over nature with manhandled stones, rendering the beautiful into the ersatz.
      For the biocomputer age to come, Silico sapiens and such, what else but the golden sands where silicon and human skin already lie in close contact?

THROWING LOOKS LIKE A FAST TRACK to a bigger brain, given that more precision is always better and that each increment in precision timing always requires a doubling of the number of neurons synchronized together. Throwing has a nice relation to juvenilized brains, given the possibility of juvenilization conserving connections that would otherwise be broken. Finally, temperate zone hunting is under enormous selection pressure -- and as "pumping the periphery" suggests, the ice ages are likely to have spread the temperate zone genes around the low latitudes within several ice age cycles even if hunting wasn't important in the tropics.
      In temperate climates, winter selects for hunting skills. Lacking food storage techniques (which conflict with the need to move around, e.g., to follow the herds), meat becomes the major source of calories and salt for a few months. A hunter's offspring could starve if he or she missed the target: winter means that gathering cannot serve as a backup for a few months each year. Only grass remains nutritious in any quantities throughout the winter, thus accounting for the popularity of grazing; eating the muscles created (at less than ten percent efficiency) by that grass is a popular way of making a living in wintertime.
      Winters are important for a fast-track evolutionary reason: they happen once a year, producing annual waves of selection that shape up the species to better fit the environmental opportunities and hazards. If the three-phase postulate and the throwing-recruitment analysis should both prove correct, it gives us a fast-track ratchet for pumping up brain size.
      In comparison, the tropical savannah seems a most unlikely setting for rapid evolution, even if it does provide optimal conditions for fossil deposition (like the hand axes, the margin of an expanding lake provides an excellent setup for preservation of skeletons) and recovery (the Rift Valley has been splitting apart recently and exposing old layers). One must ask if the Rift is not analogous to the streetlight in that old joke about why the tipsy fellow was crawling around looking for a lost item under the streetlight (no, that wasn't where he lost it, but the light was better there). The Rift has been very useful for answering What and When questions, e.g., about a fourfold encephalization in a mere 2.5 million years, since it provides minimum dates for important features. But we are skating on thin ice when we assume the Rift will also answer those Where, Why, and How questions. There may have been faster tracks elsewhere (with spread back into Africa), and the temperate zone is a likely candidate at some point.

HOW DOES THIS ARGUMENT CIRCUMVENT the previous objection to big heads per se? Juvenilization plus slowing allows for a significant fraction of the population to escape the birth canal bottleneck. And there is a degree of decoupling between the features under positive and negative selection pressures:
      1) Selection for juvenilization via generation-time shortening or hunting success happens first, and primarily on the frontiers.
      2) Selection for slowed somatic development then occurs, and not just on the frontier but throughout the population (since the ice advance causes frontier-type genes to permeate the tropics).
      3) Frontier hunting selects for the fast half of the sexual maturity variants; the birth canal bottleneck selects for the slow half of the somatic development rate variants, and the frontier survivors are the fraction of that subpopulation having both traits. And they are the ones that get all the extra babies, the next time that the ice sheets melt, the ones who are exposed to yearly episodes of winter.

WITH SO MANY of the pieces of this jigsaw puzzle still missing, it is difficult to be confident of any proposed scenario. The evolution of humans has only happened once; almost everything that happened along the way can therefore be argued to be important for shaping our present capabilities. That's one of the reasons why fast-track arguments are so important in sorting through possibilities. The groupings of the pieces are becoming clearer (thanks to both hard-earned new data and reevaluations of traditional data), and the proposed links between hunting, big brains, and the juvenilization family of traits suggest plausible solutions to one part of the puzzle.
      Now if only the lessons of hand ax heaven were known in earlier centuries: One of the reasons that the cannon was so effective when first introduced was because opposing generals were fond of infantry formations that clustered soldiers together. They make rather easy targets, even for the inexpert gunner -- a lesson that I suspect was first learned several million years ago with herds visiting waterholes.

The need is not really for more brains, the need is for a gentler, a more tolerant people than those who won for us against the ice, the tiger, and the bear. The hand that hefted the ax, out of some blind allegiance to the past, fondles the machine gun as lovingly. It is a habit man will have to break to survive, but the roots go very deep.
Loren Eiseley, The Immense Journey, 1957

Email || Home Page || The Calvin Bookshelf || Table of Contents || to End Notes || Continue reading next chapter