|A book by|
William H. Calvin
UNIVERSITY OF WASHINGTON
SEATTLE, WASHINGTON 98195-1800 USA
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 its 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).
Ratcheting Up Brain SizeWe like to imagine that preindustrial peoples endured (and endure) less stress than we do-- that, although they may have lacked physical amenities, they spent peaceful days weaving interesting fabrics and singing folk songs. But the psychic stresses of the simple life are, in fact, far greater than those experienced by the most harried modern executive. It is one thing to fret over a tax return or a real estate deal, and quite another to bury one's children, to wonder if a fall's harvest will last the winter, or to watch one's home wash away in a flood.
To grow up surrounded by scarcity and ignorance and constant loss -- whether in an African village or a twentieth-century urban slum -- is to learn that misery is usually a consequence of forces beyond one's control and, by extension, that individual effort counts for naught. And there is ample evidence that such a sense of helplessness is often associated with apathy, depression, and death -- whether in laboratory animals or prisoners of war.... Modernization, through such mechanisms as fire departments, building codes, social insurance, and emergency medical care, has cushioned most of us against physical, psychic, and economic disaster. But, more importantly, it has created circumstances in which few of us feel utterly powerless to control our lives. We now take for granted that we are, in large part, the masters of our own destinies, and that in itself leaves us better equipped to fight off disease.
Leonard A. Sagan, 1988
Since Washington's capital is really in Olympia rather than atop Seattle's misnamed hill, state legislators will be held responsible for environmental policy in a way that few others can be: If they allow pollution of the oceans and atmosphere in such a way that excess clouds reflect sunlight back out into space (and so cool the Earth), then the glacier will plow its way down Puget Sound again and loom outside their windows. If instead they allow warming pollution that melts the polar ice caps via the greenhouse effect, then the capitol dome will be submerged underwater by the rising sea level (perhaps it will have to move to Capitol Hill after all -- we're a potential refugia for public servants!).
Juneau, capital of Alaska, is even more exposed, what with the large Mendenhall Glacier already on the outskirts of the port town. Washington, D.C., in comparison, is only exposed to rises in the sea level -- unless, of course, the body politic fouls up in a really big way and causes an all-time-record glaciation that makes it well past New York City.
THE NATURE OF EXPLANATION is, as Darwin noted, particularly troublesome in evolutionary and ecological matters. Our approximations to explanation are often inadequate: contemplating an ancestor's skull, a physical anthropologist may label a feature (for example, the "brow ridge"), suggest a function for it, and then pass on to contemplate another piece of the puzzle. For some purposes such an "explanation" suffices (living as I do in Seattle, I like Grover Krantz's joke that brow ridges function as a visor for keeping rain out of the eyes!).
But often the more serious version of this one-feature-at-a-time explanatory exercise isn't very enlightening. There are parts, process, and product -- and it is often difficult to understand function without appreciating the transforming processes involved. In the case of our own evolution during the ice ages, about all we know are a few of the parts (thestones and bones, and a stand-in for our ancestor: the "average ape"), and the end product (us). The process is what transforms the parts into the product.
Around a state capital such as Olympia, everyone knows about the "legislative process" (it transforms problems and public opinion into laws and budgets). And they're not likely to confuse the process with the law itself, nor with the various constituent pressures that interact to transform the first draft of a piece of legislation into the much-amended statute that finally takes effect as the law of the land.
Personally, I always associate the legislative process with cold winter mornings and fog-shrouded freeways in the predawn light (the Washington State Legislature convenes in January). To lobby the legislators or testify at committee hearings involves long drives in the dark, sipping coffee and eating what was available at the bakery en route, while you plot tactics. The legislative process involves lots of quick conversations with busy people, trading information. Different interests may be "balanced" when the committee marks up a proposal, amending it into something quite different than the way it started out. Even if it passes one house of the legislature, it has to go through the same committee process again in the other house of a bicameral legislature, where a different set of pressures come to bear, where it competes for attention with other matters. The proposal often has to go through the cycle twice, as one house will amend the bill in a way that requires the other house's consent. What emerges from the legislature still has to get past the governor. At least in Olympia, only about 17 percent of the bills that start ever manage to finish; the rest die along the way.
The evolutionary process is a lot like that: many pressures to balance, multiple times through some subsidiary loops, with overall progress dependent on opportunities afforded by the failures or stalemates of unassociated proposals.
Have we merely labeled evolution, or have we explored it as a process? We have certainly identified some of the parts: the primate brain and some of the primate behaviors which were likely to be modified or augmented. We have identified some of the product: our fourfold larger brain, our plan-ahead intelligence, our versatile language, and a variety of other features. And regarding process, we have attempted to identify what influences the rate of evolution: mutations vs. permutations, waves of selection, conversions of function, isolation in refugia or other islands, new niche expansions when the rules are off, etc. These give us some glimpses of the processes that were surely involved in the transformation of an apelike species into humans.
But, when you compare it all to the legislative process, you see that we haven't really discussed human evolutionary process in any specific way, in the sense of playing off one pressure against another, or the multiple loops before an advance is made. To be specific, you have to get behind the facade of a "big brain gene" and ask what developmental programs were altered, and when. What conversions of function took place? What was the new niche in each of the "model years"? Can the process be repeated for additional progress, or must you have a series of different selective pressures, one after the other, to keep making progress? Was there backsliding, or was it prevented by some feature of the pumping process?
ON THE SHORES OF PUGET SOUND, Mount Rainier appears in the south, the Olympic Mountains to the west, the Cascades to the east. The long beaches are swept clean by the tides, with their piles of driftwood at the high water line. The Sound's tidal excursion is twice as large as on the Pacific coast beaches outside the Strait of Juan de Fuca.
You'd think that it would be half, not twice. Tides become smaller and smaller as one passes Copenhagen and goes further into the Baltic. Similarly when going along the Spanish coast to the east of the Strait of Gibraltar. So why are tides bigger inside Puget Sound than they are outside?
This paradox confounds the student who has forgotten about resonance. Puget Sound itself is a nice size for water to slosh back and forth within, and at a rate that the coastal tidal rhythm can reinforce. Just as you want to push on a playground swing at a rate which corresponds to its natural pendulum "ticking" rate, so a basin with a natural slosh interval of 13 hours will be most effectively driven by the Pacific coast's 13-hour interval between high tides. The sloshing water is the swing, the next coastal tide is the periodic push.
And so within Puget Sound, the tidal range is enough to cover and uncover a typical one-story house -- six times as large as the tidal range in Hawaii, which wouldn't even come up to the bottom of the windows. A good low tide uncovers a lot of beach around here. Thanks to how water sloshes around the ocean basins, both the Atlantic and Pacific coasts of the U.S. have high tides twice every day (the shallow Gulf of Mexico has to get by on once-each-day highs).
Today ocean-going ships plow their way through the waters of Puget Sound, having come through the Strait of Juan de Fuca bound for the ports of Seattle or Tacoma. Amidst the reflections of the sun off the windswept waters, there are sailboats. Fishing boats loll offshore. Here on Whidbey Island, halfway between Seattle and the San Juan Islands, are many summer homes. But there are still long stretches of beach without adjacent habitations, Most of them are protected by high bluffs, a dozen stories of cascading sand and clay (and, in places where the ancient peat bog burned and baked the slightly-more-ancient clay layer, red brick!).
There are places where the high bluffs protect from the winds of winter, and there the original fishermen often lived, starting sometime after the glacier retreated to Canada. Small fishing villages grew up in such spots, close to the beaches with their plentiful supplies of shellfish and firewood. The tidal range is why the beaches of Puget Sound have been so attractive to humans, as the shellfish were regularly exposed, free for the taking.
Drinking water drips out of a nearby cliff for much of the year, groundwater left over from the rains of winter. Only after a dry summer would they have had to walk some distance down the beach to the nearest creek. These American natives may not have set up camp by the creek -- because the sheltering cliffs had been eroded down by the water runoff, the beach would have been exposed to the winter winds out of the north. But for drinking and cooking, the seeps in the bluff were quite adequate most of the time; for washing, there was always plenty of salt water.
In fact, the fresh water supply was a little too adequate on occasion. The Puget Sound region is sometimes soaked by the rains of winter; as the ground becomes sodden with water, the flow in such seeps becomes quite a stream. All the movement of water within the bluff lubricates its sand and silt so that it can move around and work its way downhill a little, closer to the seashore. Every few generations, the fishing village would get an unpleasant surprise: a whole section of the bluff would collapse during a heavy rain, the mudslide inundating the village. If the people were not alert, they could lose their lives as well as their possessions. They likely dug around in the cold mud trying to locate lost possessions, but many were missed. How long did it take for people to realize that dwellings should be located some distance from the water source? How long did this caution take to become part of the cultural heritage that was reliably passed on to future generations?
Fishing villages didn't last very long anyway, because a severe winter storm coinciding with high tide would, once a century or so, sweep into the village and wash it out to sea, leaving nothing for their owners to recover. And nothing for the archaeologists of the twentieth century -- who have, however, found some old mudslides and excavated them. Mudslides tend to do more damage than the layer of ash that buried Pompeii, but they do provide the archaeologist with a blurred snapshot in time of a Stone Age hunter-gatherer band of the kind that specialized in fishing. Most of the villages preserved by the mudslides are not much older than the European rediscovery of the Americas five centuries ago -- but that is, at least, before metal hooks and outboard motors modified the traditional Stone Age ways of the coastal Indians.
Under the mudslides, of course, one also discovers that not all survived: Human bones are also found.
MADONNA-AND-CHILD was a favorite subject of Renaissance painters. Here one sees a madonna and child that isn't a work of art.
Human bones had been eroding out of a headland, a few more found on the beach every winter after a high tide had lapped against the shoreline. The local Indian tribe asked the archaeologists to excavate the area so that the bones of their presumed ancestors could be properly reburied a safe distance away from the shoreline.
In the process of excavating a skeleton, my archaeologist friend discovered another skeleton nestled within its pelvis. A pregnant woman and her unborn child. A reminder that pregnancy is still hazardous in most of the world.
WE TEND TO TAKE OUR BIG BRAINS for granted. Even those with a working knowledge of evolution often make the mistake of assuming that big brains would naturally evolve by slow increments: we assume that a bigger brain is a smarter brain. And since a smarter brain is surely a better brain, then it is not surprising that, analogous to compound interest, we should have bootstrapped ourselves up to a much bigger brain. After all, some people naturally have somewhat bigger heads than others, so all it takes is some natural selection for the obviously useful variant.
There is something very wrong with this commonplace explanation: it ignores the enormous natural selection against bigger heads. Maybe bigger brains are indeed better for something, but it would have been bought at an enormous price, extorted over and over again at each little increment along the way to a brain four times larger than that of our presumed ancestors, the australopithecines.
Actually, it isn't clear that bigger brains are even necessary; an ape-sized brain reorganized to facilitate language and plan-ahead might work equally well. Yet the truly horrendous problem with bigger-heads-are-better should have been obvious long before anyone got around to noticing that someone's hat size didn't correlate with how smart he was: big heads cause a lot of trouble at childbirth. Big heads not only kill themselves but, moreover, others carrying similar gene combinations: their mothers. Thus all potential siblings (and occasionally some of the still-dependent prior children of that mother as well), many likely to carry those same gene combinations, will also be eliminated from the surviving gene pool.
It is hard to imagine any form of natural selection that is more powerfully negative; modern genetic diseases such as hemophilia pale by comparison. Big heads are a candidate for the worst genetic disease of all time. By all rights, any straightforward tendency toward bigger heads should have been promptly squelched.
Those who nonetheless argue bigger-is-smarter-is-better should realize that a small increment in intelligence would have had to be overwhelmingly better ever to establish a somewhat larger brain. The next increment would have had to be overwhelmingly better than the previous miracle, and so on. While perhaps anything is possible given a long enough time and compound interest, bigger-brain cleverness per se seems unlikely as a source for the fastest encephalization on record, fourfold in a mere 2.5 million years.
It makes you wonder how bigger-brains-are-better ever became established in the first place as the dominant explanation for human evolution. If women had been the scientists doing the theorizing, I suspect that we would have long ago abandoned the notion and gone in search of a better idea.
Big heads, however, nonetheless happened. And so there is presumably some way around this problem. Something else must have been under frequent selection pressure, with big heads as an unwanted side effect that was dragged along. This suggests that big heads were achieved by some decoupled backdoor route, rather than via straightforward selection for variants in brain size. And indeed big heads come as part of a package, a panoply of linked features called juvenilization (or paedomorphosis or, in even older literature, fetalization) that has been a repeated theme of vertebrate evolution.
BIG HEADS, RELATIVE TO BODY SIZE, are most readily achieved by exactly the same process used by those salamanders at Mount Rainier: early puberty. We know that brain size, as such, isn't the determinant of cleverness, since elephants and dolphins aren't the leaders in that department. Furthermore, among modern humans a large brain is no sign of genius; despite centuries of looking for a correlation, geniuses keep coming in a variety of head sizes.
Brain size considerations remind me of seeing Mount Rainier from sea level: it's the relative size that counts, given that brain size tends to scale up with adult body size across mammals. But despite our realization that we need to normalize brain size in some manner, there is really no rational reason for talking about the brain/body ratio: losing some weight around your waist, and thereby increasing your brain/body ratio, might be a wise move but it won't make you smarter more generally. And if women are smarter than men, it probably isn't because of their larger brain/body ratio (that they typically owe to earlier sexual maturity than men, and to their lower levels of testosterone).
So it is hard to imagine why brain size would be under natural selection for its advantages -- especially when the disadvantages of an increased brain/body ratio are so immediate and so horrendous. For it is the bigger head relative to the smaller body that gets us into so much trouble: If hip size had increased commensurately, no birth canal bottleneck would have developed.
Yet it is precisely brain/body ratio that increases with juvenilization. And so an adult woman has to give birth with (by the standards of earlier generations) the narrow-hipped body of an adolescent girl. True, hip size in women does increase with childbearing; true, short adult women cannot find something that fits in the children's section of a clothing store, thanks to the hip size disproportion. But whatever the hip size compensation has been, it has been insufficient: it cannot explain the fourfold larger brain of modern humans compared to apes and the australopithecines. So if the boom time physiology of the ice ages produced juvenilizations, selection against big heads would surely have followed.
WOULD THAT REVERSE the trend that produced juvenilization? There might have been some other way of compensating, of having your cake and eating it too. Back-and-forth need not imply maneuvering along a one-dimensional track; when there are many degrees of freedom in a developmental system, an advance on one track may be partly compensated by a retreat on another. It's similar to the way a cook can raise the oven temperature but shorten baking time; time and temperature are two of the major themes the chef varies (along with ingredients and the order of mixing) in looking for better versions of a dish.
What are the typical "variations on a theme" in human development that might have been involved? Variation in head size is not, as such, a major theme. The major themes are robust-to-gracile, short-to-tall, time of puberty, rate of somatic development, plus various behavioral traits such as bold/cautious, etc. And so it is useful to discuss such prominent themes, rather than postulate random changes in this or that.
** Body size varies and short-average-tall is also partially heritable.
** Early in this century, the pioneer anthropologist Franz Boas noted that there was a considerable variation in the rate (slow, average, fast) at which infants and children add on to height and weight; he called it the "tempo of growth."
** Another heritable variant, not merely part of somatic development, is the time of sexual maturity: mothers with early menarche tend to have daughters who reach sexual maturity early too.
These themes are somewhat interdependent (time of puberty affects height, for example) but let us discuss them as if heritable genes (note the plural) for stature existed, as if somatic development rate genes existed, and as if sexual development rate genes were separate too. And that nutrition influenced them all.
Now suppose early puberty had happened to an early hominid -- for whatever reason (boom-time physiology, or some advantage of juvenilized adults such as behavioral plasticity, or even throwing skill). What next?
A single phenotypic trait -- height, for example -- may be influenced by a number of different genes; conversely, a single gene [whose alternative versions are called alleles] may influence the development of various traits. Furthermore, a particular favorable value of a trait may be attained, by different members of a species, through different allele combinations. We cannot assume in a human population that all persons of a given height have the same combination of alleles for controlling height. There may be a substantial number of alternative genetic patterns that, holding environment constant, would produce people of the same height.
Herbert A. Simon, 1983
WE'RE NOT SHORTER on average than at least one adolescent specimen of Homo erectus dated to 1.6 million years ago: he was 168 centimeters tall (known in several idiosyncratic countries as 5'6") and, had he survived to adulthood, would probably have reached 180 centimeters (5'11"). You might expect that we would be considerably shorter, since early puberty tends to reduce body size.
So it seems likely that stature has re-enlarged via some other gene affecting stature, just as those juvenilized axolotls become larger than the land-loving salamanders. If repeated juvenilizations have occurred in the hominid lineage, we would all be miniature pygmies if some re-enlargement trend had not supervened. While many of the influences on stature -- such as the improved diet and fewer childhood diseases of industrialized countries in the last century -- only affect the phenotype (body style) and not the genotype (the genes carried by that body and passed on to offspring), stature is nonetheless relatively heritable.
Natural selection likely operated on these variants in the genotype, e.g., bigger bodies for better throwing distances, better nursing of babies during involuntary fasting, better abilities to undertake long migrations to distant patches of food, or better protection from predators. Competition between individuals can presumably enlarge the average species stature, just as harem mating systems and male competition have caused male gorillas to become twice as large as females. And in the context of the temperate zone where someone was occasionally trapped out in a blizzard, the reduced surface-to-volume ratio that goes along with bigger bodies would have increased the survival time in freezing conditions because of lengthening the time it takes to reach a life-threatening internal temperature (body size is indeed larger at high latitudes). Bigger females have bigger birth canals. For these and other reasons, bigger bodies are sometimes better, despite costing more to build and operate.
Rather as a baker might have tried raising the oven temperature but shortening the baking time, we now have a population whose body style (and genome, because natural selection has been operating) is somewhat juvenilized compared to their ancestral population, but whose body size has re-enlarged via another genetic route. What now?
Man, with his remarkable brain, developed the use of fire, but, even apart from considerations of brain power, as F. W. Went has pointed out, only a creature of man's size could effectively control that fire. It happens that a small campfire is the smallest fire that is reliable and controllable. A still smaller flame is too easily snuffed out and a larger one too easily gets out of control. Prometheus was just large enough to feed the flame and keep from getting burnt.
the architect Peter K. Stevens, 1974
THE BIRTH CANAL BOTTLENECK comes next because, without further changes, bigger-headed fetuses are going to start getting stuck during childbirth (if they hadn't already had trouble at the smaller stature). This in turn will start selection operating on another common variation-on-a-theme, somatic developmental rate -- just due to their genes, some children gain height and weight more slowly than others.
We knew that some more changes were going to be necessary because juvenilization by itself tends to suggest a shorter childhood -- indeed, its truncation by early sexual maturity. But the monkey-to-ape and ape-to-human transitions show exactly the opposite: a lengthening of childhood. This paradox is resolved if we assume that a slowing of general body development (selected from that variation-on-a-theme that Boas observed) has been superimposed on juvenilization, moving the earlier menarche back out to its original year and even beyond. It's the relative rates of somatic and sexual development that control childhood's tempo and the resulting adult shape, just as it is the relative rate of growth in the north and south sides of a flower stem that cause it to bend south toward more sunshine.
The main reason to believe that slowing has actually happened is that slowed development is more general than just childhood. Most life phase durations (conception-to-birth, birth-to-weaning, weaning-to-menarche, adult span) have been nearly doubled in going from monkey to ape. And nearly doubled again in going from ape to human. Though human gestation would at first appear to constitute an exception (it is only several weeks longer than in apes), this doubling rule seems to apply there too: human infants do not attain the same developmental landmarks as newborn apes until many months after birth, for a total internal-plus-external "gestation time" about twice that of chimpanzees.
This halving of the rate of the somatic developmental clock throughout pre- and postnatal life also needs explaining; I'm surely not the first to suggest that it was the solution to the childbirth problem presented by that big head that came along with juvenilization. If there had been a way of slowing only prenatal development without concomitant slowing of postnatal development, it might have done the job too -- but the more generalized slowing may have been the only variant available.
Because juvenilization makes the adult head relatively larger and the adult pelvis relatively smaller, repeated juvenilizations will eventually run into trouble when the baby's head can no longer get through the pelvic outlet. The gene combinations that result in early puberty and normal somatic developmental rates will then be edited out, unfortunately via maternal mortality rather than merely unsuccessful fetuses (but therefore at a much faster rate, because of the kin selection practiced by the unsuccessful fetus). The same would be true for faster-than-average somatic development genes. The gene combinations of precocity and slowed somatic developmental rates will get by, provided parturition is not equally delayed.
So long as the surviving mother can cope with raising a relatively fragile premature infant, the gene pool would soon come to be dominated by the genes for slower-than-average somatic development. This escape route for big baby heads would seem to require slowed somatic development superimposed upon the accelerated sexual maturity; our longer life spans after birth may be largely a side effect of the slowing of somatic development needed to work around the birth canal bottleneck.
Thus we get the sequence of 1) juvenilization via faster-than-average sexual development, 2) re-enlarged stature via other taller-than-average genes, and 3) slower-than-average somatic developmental rate. And because of the carryover of slowed development into postnatal life, the usual time scale is stretched; the number of years that it takes to get to puberty may have moved back out beyond what it was before the changes started to take place. Body size is also potentially back to the norm. Only head size is still increased, along with a few other uncorrected side effects such as reduced tooth size, flatter faces, and other such juvenile features.
Eureka? Only if the three-part cycle can be repeated quite a few times. And body style doesn't backslide.
Heredity is particulate, but development is unitary. Everything in the organism is the result of the interactions of all genes, subject to the environment to which they are exposed.
the evolutionary biologist Theodosius Dobzhansky, 1961
WALKING ALONG THE WHIDBEY BEACHES, one cannot help but be impressed by cycles. The twice-a-day tides. The twice-a-month extreme tides at new and full moons caused by the moon, earth, and sun approximately lining up and so exaggerating the pull on the oceans. There are even greater extremes every 170 days when the moon is threatening an eclipse because of being very near the earth-sun line and pulling in the same line as the sun. And if the sun is at its closest to earth (perihelion) while the moon is also at the minimum in its elliptical orbit at such a time (perigee), the low and high tides reach their extremes (that's fairly rare, but the moon at perigee coincident with new or full moon happens every 9.3 years).
I assume that the coastal Indians noticed the relation between moon phases and the tides; they might even have made the connection between eclipses and extreme low tides (and, instead of fearing eclipses, considered them a portent of good clamming prospects, what with all that uncovered food).
All this is not quite as obvious as the cycles of the seasons due to the tilt in the earth's axis. The severity of winter and summer also varies, due to all those orbital parameters that cause the ice age cycles -- but that is also on too long a time scale for anyone to notice without generation-spanning record-keeping (such as the weather records of the last century).
There are also cycles we don't understand (though chaos theory is helping), such as why the local shellfish undergo a population crash every so often -- or why "red tides" occur and the shellfish become poisonous (there is a potent neurotoxin from a dinoflagellate, Gonyaulax tamarensis, which produces paralysis in humans; Gonyaulax thrives in low salinity, such as near river mouths). Given how the natives of that fishing village probably relied on shellfish, those would have been hard times, making fishing skills extremely important until the clams become safe to eat again. The local salmon are easier to catch in rivers than here -- but they too have their cycles, the salmon returning to local rivers increase every few years and then drop back.
A cycle implies that the process can be repeated, that it doesn't run out of steam (pendulums swing back up after reaching the bottom of their arc because their potential energy has all been converted into kinetic energy; they stop and reverse when all the energy of motion has been reconverted into potential energy). The shellfish and the salmon are probably locked into some sort of back-and-forth, boom-and-bust relationship with their food sources or their parasites and predators, seen more easily in the ups and downs of the Arctic lynx and hare populations (and modeled by those same equations that gave us r and K terminology; just remember that hare populations can oscillate even without lynxes around!).
Can this evolutionary process of juvenilization-reenlargement-slowing become a repeating cycle? Is it like a college course that can be repeated for additional credit? Or has it run out of steam after the three phases, like most inventions in evolution -- run to the end of its growth curve, with no further progress possible? The classic example of a limited growth curve is hairlessness: when something such as swimming success starts selecting for variants with less body hair, there is a limit beyond which further selection cannot operate (as you can only become so naked).
Can our slightly juvenilized (but re-enlarged and slower-growing) hominid, with its slightly larger-than-ape brain and its slightly flatter face and slightly smaller teeth, be subjected to yet another round of selection exactly like the first one? Can boom-time physiology (or some specific advantage of a juvenilized body style) select for juvenilization again? Can the resulting population then re-enlarge? Will the birth canal bottleneck then again select for slower-than-average somatic development rate genes?
WELL, WHY NOT? If any one of the selection pressures is removed, the cycle won't repeat. If any one of the three processes runs out of growth curve, it'll stall. If anything is invented that can break the cycle -- such as cesarean sections or really big hips -- it should stop. Otherwise, it ought to cycle until the disadvantages balance out the advantages. There might be counterpressures (if you are a hyper-robust australopithecine and need big teeth for processing plant food, this counterpressure might have prevented further rounds of juvenilization). Something, for example, happened after a similar juvenilization transition from the Old World monkeys to the apes: they seem to have stabilized for 30 million years rather than repeating the cycle of juvenilization and slowing.
Might body size have counterpressures? There are situations (islands with dwarf elephants, for example) where small body size is common. But, at least in the temperate zones, our typical adult stature seems better than a pygmy-like stature; while bigger-is-better may not extend to 250 centimeters stature, it may apply in the 100-200 centimeter range if one judges from the latitude data on aboriginal populations -- and so another juvenilization episode would again reposition stature on the lower half of its growth curve, ready to re-enlarge again seemingly forever. The ever-more-helpless infant may well have required some prerequisites: the kangaroo's pouch may suffice for its helpless infant, but we probably adapted with the aid of bipedal locomotion for infant carrying and the two-parent family for provisioning mother and child.
So we are left with whether there were even earlier sexual maturity gene versions around (apparently so, judging from the heritability of early menarche). So what is the juvenilization advantage that comes under selection? Whatever that feature is, it must have had a very long growth curve -- where more and more was always better and better -- for it to have been used repeatedly during the last 2.5 million years. That's a big order. What's so good about juvenilization that has such characteristics?
Reproductive races in boom times might work, if repositioning by slowed somatic development suffices. But it really ought to backslide readily in hard times. I've got another candidate for why juvenilization was so useful, one with a spectacular growth curve. It is just what I'd recommend to an ambitious ape, wanting the brain capacity for language.The scientist [J.B.S.] Haldane,
brooding upon the future,
has speculated that we will even further
prolong our childhood and retard maturity
if brain advance continues....
[But ultramodern man has] happened already.
Back there in the past, ten thousand years ago.
The man of the future,
with the [even bigger] brain,
and the small teeth....
Those who contend that
because of present human cranial size,
and the limitations of the human pelvis,
man's brain is no longer capable
of further expansion, are mistaken.
Cranial capacities of almost a third more
than the modern average have been attained
among the Boskop people [of southern Africa]
and even in rare individuals among other,
less [juvenilized] races.
Loren Eiseley, The Immense Journey, 1957