W. H. Calvin's THE ASCENT OF MIND (Chapter 2)
Home Page || Public Bookmarks || Science Surf column || The Calvin Bookshelf
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 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).


2
INCREMENTING INTELLIGENCE:
A Principle of Nature?

Who taught the raven
in a drought
to throw pebbles
into a hollow tree,
where she espied water,
that the water might rise
so as she could come to it?
Francis Bacon (1561-1626)
Brain size. Cleverness. Intelligence. Versatility. Being "smart," thoughtful, able to plan ahead. What do they all have to do with one another? There is clearly much overlap in the connotations of such words. Since they are also self-congratulatory, we have to watch out for anthropocentric tunnel vision as we try to get a grip on the problem by comparing various animals, seeing what's so good about innate intelligence.
      Brain size seems especially crude as an index, as if the brain were only a container for what was really important, rather than the working machinery of consciousness. We know that our way of thinking isn't simply a matter of absolute brain size: individuals with a two-liter brain aren't twice as smart as those with a one-liter brain. If you correct for body size (the ratio of brain weight to body weight is the usual measure, though obviously inadequate), you get a somewhat improved correlation of size with some aspect of cleverness. But there are still all sorts of exceptions: the squirrel monkeys, fairly average among the New World monkeys, have a much higher brain/body ratio than all other monkeys -- but the capuchin monkeys seem by far the cleverest of the New World monkeys, almost ape-like in some respects.
      And cleverness? Intelligence? What do they mean? Except for the great boost that language gives us, are we humans all that much more clever than the apes? An airplane flight from London to Budapest gave me plenty of time to think about this (despite the nice view of London after takeoff, Europe was entirely covered by clouds, undoubtedly contributed by the North Atlantic Current). I already had plenty of incentive to reflect on it, given that I was shortly scheduled to explain the evolution of intelligence to a group of astronomers, who were gathering to discuss the prospects of detecting extraterrestrial life forms.

GENES NEED ONLY be approximately correct, as a little behavioral versatility can do the rest. While this versatility during life may not alter the genes passed on to offspring, it does serve to shape up those genes: behavior can drag along anatomy. This was recognized by three scientists in 1894; though often called the Baldwin Effect, it probably ought to be called the Morgan-Baldwin-Osborn Effect. Perhaps we would understand it more intuitively were it called the Old-Family-Recipe Effect.
      Anyone who has ever asked for a copy of "that wonderful recipe" knows that the recipe card is always faded, flour-encrusted, written in a style of handwriting favored by some first-grade teacher of long ago, and smeared by several ancient droplets of an unknown fluid. And so when you transcribe it onto a new card to carry home with you, some copying errors are likely.
      What's worse, the donor of this recipe has long since stopped consulting the recipe card: she just bakes from memory and, over the years, has improved the cake (or whatever) considerably beyond what would result from faithfully following her written recipe. Indeed, she has no idea how much her "handful of flour" departs from the half-cup that the recipe calls for, or how inaccurate the temperature setting on her oven has become. Still, she has found the winning combination (you did, after all, ask for the recipe) and so her point-of-departure version of the recipe comes to be copied with an unintentional mutation or two.
      This commonplace situation suggests a simplified scheme for how cake-baking contests at county fairs could "cause" better cakes to evolve. Pretend for a moment that success in baking cakes obeys the following rules:

1. Each participant inherits a randomly altered copy of her parent's recipe for a cake. Perhaps a teaspoon of baking soda is changed into a tablespoon's worth. Or the 385° baking temperature into 335°. Or some other such alteration in the mix of ingredients, amounts, times, and temperatures.

2. The cook can modify the recipe during her lifetime, but only by memory, not by amending the recipe card. Indeed, since the recipe card is merely the point of departure for experimentation, it need never be consulted again (until finally copied).

3. There are contests to select the better cakes, and the winners and runner-ups are the ones most likely to have offspring attracted by the cake-baking contests in some future decade. Note that winners don't train offspring at cooking (in this simplified scheme): they only pass on their point-of-departure version of the recipe. The only thing that experience, i.e., the recognition of good variations, does in the long run is to make the winners' offspring more likely to become contest-minded cake bakers.

4. The judging doesn't change criteria over the years ("good taste is eternal").

The recipe's mutations are usually worse than the original. In any generation, of course, an off-on-the-wrong-foot cook who is, nonetheless, skillful at fiddling the recipe may hit upon the combination that constitutes the optimal recipe; inheritance is not fate (but she cannot pass on this winning combination as such, just the degraded recipe card). Yet on the average, the copying errors that move away from the optimal make it less likely that unwritten variations in the recipe ("a lifetime of experience") will hit on the optimum.
      Because losers tend not to have offspring that participate in such contests (the losers don't get asked for a copy of their recipe), diverging recipes are more likely to die out. And so there will be a slow convergence in copying errors toward the optimal combination, just by carving away the other combinations. The optimal recipe may never be written down, but the population of written recipes in use gets closer and closer to the combination of ingredients, amounts, times, temperatures, and assembly procedures that will satisfy the expert tasters of cakes.
      Allowing a son or daughter to learn the parent's hard-earned variations on the recipe would represent Lamarckism: inheritance of acquired-during-life characteristics. This "Training Effect," of course, happens with real cooks and their offspring; we encourage this mode of transmission with schools and books. But we theorists may temporarily leave such influences out of explanations, just to demonstrate that the whole population of written recipes (or whatever) can nonetheless shift closer and closer to the unwritten optimal even without the additional Lamarckism (in the case of biological inheritance, we also leave instruction out because there is little evidence for it).
      Adding some version of Lamarckian shaping has two interesting effects: cakes converge on the optimal even more quickly, but the written recipes converge more slowly than they would otherwise. (In the terminology of evolutionary biology: With Lamarckism, the phenotypes evolve faster but, paradoxically, the genotypes evolve slower!) Should there be a "lost generation" that never learns to cook from their expert parents, the grandchildren will have to start over from instruction cards that haven't been shaped up anywhere as far as they might otherwise have been. While shaping up the "written version" may be safer in the long run, one has to first survive the short run -- and climates often shift so rapidly that survival depends on changing food-finding strategies just as quickly (in the cake analogy, suppose that next year's judges went sour on sugar, all trying to lose weight because of a new preventive medicine campaign against obesity). And so both the Old-Family-Recipe Effect and the Training Effect may prove essential in the short run because the judging criteria have changed.
      In the analogy, the individual ingredients-and-procedures are the genes, the recipe is the sperm-or-ovum, and the whole population of cake recipes is the genome. And, of course, the cake is only the recipe's way of getting a copy made of itself. The Selfish Recipe has struck again.

THE ABILITY TO DO SOMETHING COMPLEX isn't, by itself, a sign of intelligence. The earth's ocean-atmosphere-icecap system is quite complex, without even being alive. Spiders weave complex webs, but are hardly versatile designers. Ants build high-density dwellings that are air-conditioned with a sophistication that, until this century, was beyond the engineering abilities of mere humans. Learning and memory are not necessarily signs of sophisticated abilities either. All sorts of animals, such as earthworms, can learn and exhibit long-lasting memories. Pigeons have even learned quite fancy category discriminations, learning to pick out pictures of sad humans from happy humans. But when an animal does something both novel and complicated -- after only several unsuccessful attempts -- that's at least clever.
      Observation learning is the most obvious case of such speedups in acquiring abilities, where one animal imitates the novel actions of another. Insight is another, where an animal seems to contemplate the situation and then does the effective thing without any trial-and-error. A dog on a leash, who is prevented from getting to food because its leash has become snagged around an intervening tree, may never solve the problem except by rambunctious trial-and-error. A chimp, on the other hand, can take one look at the situation, immediately retrace its steps, and disentangle its leash from the obstruction.

Man is an imitative animal.
This quality is the germ
of all education in him.
From his cradle to his grave
he is learning to do
what he sees others do.
Thomas Jefferson, Writings

LEARNING BY OBSERVATION AND IMITATION is not uniquely human. As the cats demonstrate. Psychologists had a hard time training cats to press bars or run mazes; rats do such things easily. Since the number of trials that it takes to produce flawless performance is the basic measure of learning among comparative psychologists, the recalcitrant cats were coming in last, behind the slowest rats in the ratings.
      This contradicted common sense, so psychologists persevered and finally found a cooperative cat that would consent to learn their task. And the way that they trained the next cat was simply to allow it to be a sideline spectator, while they put the trainable cat through its paces. When the spectator was then placed in the apparatus, it naturally tried out the tricks for itself. And so got the idea very quickly, faster than a rat would have done. The bar-pressing problem posed by the psychologists just wasn't sufficiently interesting by itself; the way to engage a cat's attention is to let it observe another animal.
      Observation learning is probably how the neighborhood cats have discovered our pet cat's entrance to the basement. I had constructed this hole in the wall such that our cat had to jump the height of a countertop in order to reach the opening, thus breaking the scent trail that other animals could easily follow. And this indeed cuts down on the number of midnight marauders who require evicting. The raccoons still come to visit (we see their muddy paw prints on the glass of the front door, from where they peered inside during the night) but it has been a decade since one ransacked our kitchen, back in the days of a ground-level cat door. Yet once a year, a neighbor's cat will appear in our kitchen, with that tentative "just looking around" poise, shortly after our cat has arrived indoors. It probably saw the leap to the opening, and mimicked the behavior.

FORESIGHT LIES BEYOND INSIGHT (if one were to attempt to construct a rating scale for animal cleverness) but only if there is something unique about the plan. A squirrel hoarding nuts for wintertime at the behest of its hormones doesn't count. But as I discussed in my previous book, The Cerebral Symphony:

Chimpanzees come the closest to human-level novel planning when they engage in little deceptions (a behavior rarely observed in monkeys). A chimpanzee who comes upon a bountiful food resource -- say, a tree full of ripe fruit -- usually utters a joyful ``food cry'' that quickly attracts the other chimpanzees of the band, who similarly exclaim in delight upon seeing the bounty. But if the first chimp sees that there are only a few fruit to be had, it may keep quiet, attempting to silently eat all the fruit before any other chimp wanders along.
      Foresight-prompted deception occurs when the lone chimp, hearing the approach of other chimps and worried that it will be deprived of the rest of its feast, leaves the limited bounty, casually strolls over in a different direction, and issues a food cry in the midst of dense foliage -- where there is no food! This decoys the other chimps away from the limited supply of fruit. While the others are excitedly looking around the false site, the first chimp circuitously returns to the true site and finishes off the feast.
      So it seems as if the chimpanzee can foresee the scenario of losing its remaining feast to competitors, and that it can spin a decoy scenario that involves ``telling a lie.'' One might argue that these deceptions are only occasionally novel: losing food to a higher-ranking animal is an everyday occurrence, and most decoy deceptions are probably just repeats of an earlier success. But still, there is some element of novelty in the animal's ``first lie'' that begins to look like the scenario-spinning deceptions common in humans.
Now if all chimpanzees did such things, we might simply consider it an innate behavior, wired into their brains before birth. And if we knew that the chimp had learned to do this by mimicking the success of a frequent companion, we also might be less impressed. Only if we were convinced that a chimpanzee spun alternative scenarios, picked and chose between them, spun more scenarios when dissatisfied with the early choices, etc., would we worry that apes were closing in on the uniquely human scenario-spinning abilities that we associate with contemplative consciousness.
      And when we say intelligent rather than merely smart or clever, we are often implying a substantial amount of looking-ahead, judging the probable consequences of a novel course of action. Doing something nonstandard, rather than what your genes tell you is the appropriate thing to do, is usually risky. Just as most gene mutations are not an improvement (many lead to spontaneous abortions), most behavioral innovations are disastrous, absent foresight. The only way that humans get by with so many inventive behaviors, performed for the first time ever, is that we can do a lot of trial-and-error in our heads as we contemplate acting, as we "get set." We simulate a course of action before acting, provided that we have the time to spare. And we discard most of the plans before acting on them, rating them unsafe, inappropriate, or uninteresting. Another key aspect of intelligence is the ability to perceive order in a situation that appears disorderly, all those collections of objects where you're supposed to deduce the feature that characterizes all but one, so as to spot the odd man out. What's been surprising is how often a chimpanzee can solve the ones that two-year-old babies can solve. Pigeons do surprisingly well, too.

THE ANIMAL INTELLIGENCE PROBLEM has caused some investigators to emphasize that animals can do almost anything that humans can, except use language to express the results or pose the questions. They suggest that the appropriate "null hypothesis" is that language is the main difference between apes and humans, that most of the "intelligence" differences are merely secondary to the mental structures that come with language.
      The great neurologist of a century ago, John Hughlings-Jackson, said: "We speak not only to tell other people what we think, but to tell ourselves what we think." But don't animals think, and without our kind of language? Yes, all animals "think" to some extent -- all can decide what to do next, evaluating their environment and choosing between standard alternatives -- but without language we lose the richness of the choices available to the thoughtful person, and we miss out on much of our ability to invent novel alternatives.
      The tragic problems of "feral children" are sometimes used to illustrate this point, but they always have a multitude of social and medical problems caused by the neglect. Children born deaf, and never exposed to sign language, illustrate how an otherwise-normal human upbringing that omits language leaves the unfortunate child lacking in basic abilities. The neurologist Oliver Sacks described such an 11-year-old deaf boy who was never exposed to sign language:

Joseph saw, distinguished, categorized, used; he had no problems with perceptual categorization or generalization, but he could not, it seemed, go much beyond this, hold abstract ideas in mind, reflect, play, plan. He seemed completely literal -- unable to juggle images or hypotheses or possibilities, unable to enter an imaginative or figurative realm. And yet, one still felt, he was of normal intelligence, despite these manifest limitations of intellectual functioning.
Joseph's deafness escaped diagnosis and compensatory early education in sign language; he was considered "retarded" or "autistic" for most of the critical years of his childhood. Language allows far greater levels of abstraction, permits us to build up mental models for how the world works, allows us to pose questions, craft answers.
      How does the brain organize itself to do that? How do we weave together that linear tapestry that we call a text or a speech?

BRAINSTORMING techniques illustrate one explicit way of synthesizing a sentence or proposition. This way of thinking is one that we probably don't share at all with the apes (even if they should have the neural machinery, they're usually too impatient!). We attempt to generate dozens of ideas, the wilder the better -- but hold off evaluating them until quite a few have been generated. That way, we get a lot of variations on a theme out on the table. Then we shape up the best ones a little further, using our factual and aesthetic judgments.
      This creativity-promoting technique is a lot like the processes of darwinian evolution, where a boom time serves to suspend judgment until a lot of variations are out there, broadening the characteristics of a species (like those dozens of unevaluated ideas). And then judgment time arrives, usually in the form of a worsening climate, and only the versions survive that perform well in that particular climate. Might the brain be using darwinian techniques most of the time, not merely when formally brainstorming? Might the subconscious be the dozens of mostly-nonsense candidates, vying to be what we are conscious of? Deciding what to say next is a simple example of the brainstorming technique, though we usually do it so unconsciously as to be unaware of most intermediate steps. Imagine four planning tracks, each able to hold on to a string of words, keep them in order. Start with a series of words that are in the forefront of your short-term memory, probably because you've recently used or heard or read them; they'll each have some connections to other words in your vocabulary (cat might evoke dog, bite might evoke eat, etc.). Stringing some of them together in a random order will usually yield nonsense (pretend that the four tracks are merely the best out of a hundred such tracks). But some will approximate reasonable sentences, when you judge the string of words against your long-term memories of reasonable English-language sentences. Most of those will be inappropriate to the situation you're currently in, so that current-situation judgment will deflate your "good" ratings of otherwise reasonable sentences.
      Now try another round of brainstorming: erase those low-rated strings, take the top-rated string of Round One ("The dog bit the mailman"), and try variations on it -- which you store in the erased tracks. This "noisy copy" makes mistakes just like the ones in genetics, sometimes using a related word instead of the original (as in a thesaurus). And so you'll get mailperson as an occasional substitute for mailman, or perhaps letter-carrier or just person. There will be a hundred such strings in the hundred planning tracks: the original plus 99 variations on its themes, of which the top four might be worth talking about. These 99 new tracks are again judged against your memories of what might be grammatical and what might be suitable. If you prefer letter-carrier to other related words, you might wind up with "The dog bit the letter-carrier" as your most common string of words. If it seems good enough to cross your personal threshold for converting thought into action (perhaps because it has finally cloned itself into a majority of the planning tracks), you might even speak the sentence.
      Many rounds of this shaping-up process would likely yield more literate sentences, and occasionally novelty: concepts that had never been linked together before. It's very much the way in which natural selection shapes up a population of biological individuals, which is why I call it a Darwin Machine.
      The various Darwin Machines are each characterized by a somewhat different sequence of information units. A sequence of DNA nucleotides, in the case of genes. Amino acid sequences, in the case of an individual antibody of the immune system. And, in the case of mental plans for what to do next, we are creating new sequences of sensory schemas (e.g., nouns) and movement subprograms (e.g., verbs).
      Besides the obvious usefulness for our kind of beyond-the-apes language, this Darwin Machine shaping-up method is also handy for scenarios, devising plans of action that involve many linked steps. Most random scenarios won't work, and it is nice to be able to figure that out before acting. If you've done exactly the same series of actions before (as that deceptive chimp might have done), fine – but what if the scenario is unique? Novelty in biological evolution usually results in spontaneous abortions; in behavior, most novel actions will get you into trouble. Making a working model of what is likely to happen next, inside your head before acting, is the way to have your cake and eat it too.
      While remembered environments are less detailed than real ones, this off-line simulation and testing operates in milliseconds-to-seconds rather than the centuries-to-millennia of biological speciation. If you've time to contemplate the problem, you can do thousands of generations, shaping up alternatives. Unless you are as unfortunate as the deaf Joseph, you can create a metaphorical world in your head, within a matter of minutes – and using the same techniques as darwinian evolution took to evolve the physical world in eons.
      But where might humans get those hundred planning tracks, that ability to shape up better and better plans? Why don't apes do the same thing? Our best clue is whatever evolved our left-brain's special ability to order things serially.

IS THE ELABORATION OF FORESIGHT, good old think-before-you-act, the particularly human aspect of intelligence that evolution somehow augmented, not language per se?
      There are three main theories for where this foresight has come from, in evolutionary terms. The English psychologist Nicholas Keynes Humphrey, for example, would emphasize that social intelligence is all-important: that a up-and-coming chimp is always trying to predict what a dominant animal will do in response to an initiative, is often recruiting help by building coalitions, and otherwise solving social problems (that influence access to mates) rather than environmental ones (that affect survival). On this theory, social foresight bootstraps cleverness. This would make it analogous to the way that the gorilla's harem mating system tends to exaggerate male body size.
      A second theory is that augmented foresight (and, indeed, language) resulted from a conversion of function, that the natural selection was not for foresight itself but rather for the forerunner function, before conversion. However, let me start with a few words about the third, which is the conventional "natural selection" reasoning for becoming smarter and smarter -- and a few more words about why one cannot be satisfied with it, which I like to call the Fermi Principle.

THE LAKESHORE TOWN of Balatonfûred in Hungary, an hour or two southwest of Budapest, was the scene of the International Astronomical Union's bioastronomy symposium. "Bioastronomy" is sometimes considered the IAU's euphemism for what is commonly known as the search for extraterrestrial intelligence, or SETI (actually, it is quite appropriate: They searching for biology in general, not intelligence in particular). About 150 scientists met for a week. We were mostly radio and optical astronomers, plus some chemists, and a few odd brain-behavioral people like me.
      I suspect that many of us were curious to see the Hungarian culture that had produced so many mathematicians, scientists, artists, and composers. However, the visitor to Hungary is immediately disoriented by discovering that the language is completely impenetrable, totally unlike any familiar European language (though there is a distant relationship to Finnish and Estonian). Fortunately, many of the highway and railroad station signs are bilingual -- Hungarian and German.
      Hungary was a particularly appropriate place for a SETI discussion, given that famous quip by the Hungarian physicist Leo Szilard a half-century ago. Once at lunch, the Italian physicist Enrico Fermi tried to point out the absurdity of the favorable estimates of intelligent life elsewhere by asking, "If they are so probable, then where are they? They should be here already, we should have seen them by now." After all, there are stars far older than ours: life elsewhere could have had a ten-billion-year head start.
      After a pregnant pause, Szilard answered, "Perhaps they are already here. But we call them Hungarians."

EXACTLY THE SAME OBJECTION as Fermi's can be raised to our common assumption that becoming intelligent, or at least smart, is what evolution is all about. It seems so self-evident that being smart is better than not. But, if so, we should now be surrounded by smart animals, exploiting sheer intelligence rather than brute strength and low cunning. Where are they?
      Well, the primates, and indeed many of the mammals, are often clever. But really useful features tend to be reinvented by evolution. Photoreceptors have been independently invented over 40 times in various invertebrate lineages: partway out a branch on the tree of species, photoreception will appear and persist. Powered flight was invented at least three times after the insects did it: by the flying dinosaurs, by the bats, and by the birds (not to mention all the jumping spiders, gliding mammals, "flying" fish, even a snake that glides between tropical treetops). Is being smart a similar sterling feature of evolution, rediscovered many times?
      If ape levels of cleverness are your criterion, then the answer is no. Even lowering your standards to the abilities exhibited by monkeys and bears still yields only one major lineage: mammals. But if we take a somewhat lower standard of intelligence, say the cleverness of a rat or racoon, we can find several more lineages besides the mammals. The birds have gone on to develop clever crows, ravens, gulls, and vultures; the Egyptian Vultures bomb ostrich eggs with stones when the eggs themselves are too large to haul aloft and crack by dropping them. The big-brained crows and ravens are mischievous in ways that tend to suggest they get bored. And one invertebrate phylum, the mollusks, has also gone on to develop cleverness of a rat-racoon level: the octopus has impressed researchers with its versatility, especially when it comes to catching crabs.
      So count three examples of the independent evolution of rat-racoon levels of cleverness, all probably associated with omnivorous diets and the necessity for a dozen different techniques for detecting and outsmarting prey. Why might there be a "varied diet" requirement for evolving cleverness? There are, of course, some clever animals that presently have monotonous diets, such as the marine mammals that presently make their living in the same manner as the fish-eating fish. The big brains of the dolphins and whales don't seem to be currently needed for many of their characteristic food-finding behaviors, given that fish-sized brains suffice. The land-dwelling ancestors of the marine mammals probably specialized in eating shellfish in the intertidal. They gradually learned to swim well enough to exploit schools of fish offshore, and finally miniaturized their limbs and converted to streamlined body forms via a thick layer of fat, rounding out their shapes (sea otters, who rely on fur rather than fat for insulation, have probably returned to the sea rather recently, compared to the 100-million-year time scale of the seals, dolphins, and whales).
      But having a mammalian brain means that they can sometimes invent clever techniques, the way that orca ("killer whales") may herd small fish into a corral of bubbles. They swim around blowing bubbles, to create a circular curtain, that causes the fish to turn around and head back towards the middle. Then the orca soar up through the corral towards the surface, mouths wide open, scooping up fish. Laying down a bubble curtain is exactly what hatchery workers do, when wanting to net a lot of fish -- but I think that orca invented the technique, long before humans.
      The big brain of the gorilla isn't really needed for its 60-pounds-a-day diet of leaves and bamboo. And while they can be playful, wild gorillas exhibit little of the behavioral versatility of the chimps and haven't been observed to do anything as fancy as the orca's funneling of fish. Gorillas (and, for that matter, orangutans and the lesser apes) seem to have retreated into a vegetarian niche that severely limits where they can live. Given the low quality of the food, they need dense forest to provide the needed quantities (and an enormous gorilla-length gut to digest them). Humans who retreat from our ancestral diet that valued meat to being vegetarians can at least cook their food (which expands the choice enormously, via inactivating toxins and softening bonds).
      Obviously, cleverness isn't just useful in finding food and avoiding predators. It can also facilitate acquiring mates, surely one of the major advantages of social intelligence. In societies with a dominance hierarchy, the position in the hierarchy tends to influence reproductive success -- and so the ability to build alliances, pacify the angry, get around a watchful alpha male and consort with a female unobserved, will all aid reproductive fitness.
      The most obvious aspect of male competition for females is body size -- the bigger gorilla tends to win the fights with smaller males, and so a harem-type mating system leads to an arms race in body size. But since a gene augmenting testosterone production is located on the male-only Y chromosome, it's mostly bigger bodies for the males (they're now about twice the size of females). In contrast, male cleverness in winning females in other spheres of action should tend to improvements in both male and female cleverness in following generations (there is only room for several genes on the Y chromosome, so most are located on the 22 pairs and X chromosome common to both males and females), just as female cleverness in keeping sick infants alive has undoubtedly benefitted both sexes, not just females.
      Thus both environmental selection and sexual selection could operate on cleverness and so shape up the population to evolve into increasingly more clever animals. Even a minor improvement can eventually confer a major advantage. Consider the fourfold brain size increase of humans over the apes, most of which happened in the last 2.5 million years: it only required an average increase of one-millionth of a percent per generation. Compound interest has done the rest. Or so the story goes.

Why should not Nature have taken a leap from structure to structure? On the theory of natural selection, we can clearly understand why she should not; for natural selection can only act by taking advantage of slight successive variations; she can never take a leap, but must advance by the shortest and slowest steps.
Charles Darwin, On the Origin of Species, 1859
SUCH IS THE STANDARD REASONING for intelligence by adaptations, the argument why, given enough time, biology ought to evolve our kind of intelligence: We just used increments in cleverness for more efficient food-finding, predator evasion, or creating mating opportunities.
      The efficiency type of argument always seems to point to inevitable progress. "Since evolving intelligence is a general principle of nature, we don't need to bother with the details -- it'll happen, one way or another." Perhaps that is a little exaggerated, but it is what the physicists and astronomers rely upon, when they argue the probabilities of finding intelligence "out there."
      Why, then, are the evolutionary biologists so uniformly skeptical about the SETI story? It's not that they believe intelligence is surely rare elsewhere -- they just point out that progression in intelligence is a suspect proposition, that efficiency leads even more often to dead ends. They are better acquainted with all those branches of the tree of animals that don't seem to be going anywhere, those stabilities into which evolution settles.
      A familiar stability is embodied in the Peter Principle, the late Lawrence J. Peter's humorous suggestion that all experienced bureaucrats are incompetent. This is because, as a reward for past service, they've finally been promoted to a level for which their abilities prove insufficient. And so they receive no further promotions, limited by reaching their "level of incompetence." This stability means, in Peter's formulation, that the higher echelons of the bureaucracy are filled by people who are well-suited to one level below their final rank. Biological evolution isn't quite like that (nor are real organizations!), but there are many stabilities that similarly limit progress. Indeed, species often "paint themselves into a corner" by overspecialization.
      Evolution is also full of good-enough solutions that remove a feature from exposure to natural selection -- and so a Rube Goldberg scheme may persist without improvement. "Satisficing" is Herbert Simon's term (from satisfy, as opposed to optimize) that he uses to describe the analogous situation – the failure to optimize seen in psychology and economics. Satisficing is probably why only three lineages have developed rat-raven-octopus cleverness: most were clever enough for their way of making a living and the brain changes that did chance to come along had as many liabilities as they did advantages.
      But, assuming that dead-end stabilities aren't reached, how fast will evolution progress toward cleverness? It depends on exposure to natural and sexual selection. As the Younger Dryas story suggests, there are sometimes severe waves of natural selection (in the following chapter, I will elaborate on the various selection cycles that may have played a role in our evolution). But what else influences the speed with which new functionality develops?
      Evolution isn't just "shaping up" via adaptations, though that is the first explanation we always try out for size, when contemplating a feature that evolution has produced. There are at least two other major routes to new functionality: 2) Sometimes a feature is shaped by natural selection for another feature, one that is linked to the first feature because they share a common developmental mechanism (as when selection for precocious puberty also serves to produce shorter stature and smaller adult teeth). And 3) sometimes a new function is simply invented, a new use emerging for old anatomy. One hesitates to invoke these less common explanations until the simple adaptationist reasoning is tried out.
      Yet sometimes the simple route just doesn't work very well. I like to imagine what Nicholas Copernicus would have been subjected to, had there been scientific meetings of our modern kind in sixteenth century Europe: "But my dear Copernicus, surely it is simpler to assume just one rotation, that of the sun around the earth? Occam's Razor says we should pick the simplest explanation, does it not? Why this messy, unaesthetic business of assuming two rotations, the absurd postulate of the earth rotating in orbit around the sun, plus the earth rotating around some axis through the frozen northlands? Gentlemen, if the earth were spinning like some child's top, I'd fly off my feet and out that window! And I assure you that my feet remain planted firmly on the ground." No wonder Copernicus was reluctant to publish until he was dying -- he could imagine the pointed questions.
      Simplicity is relative: it depends on how many things you're trying to explain at once. Just imagine what you'd have asked Copernicus or Galileo at a scientific meeting if you didn't much care about those occasional retrograde motions of the planets across the night sky, didn't think them very important compared to the sun and moon. We see a closely analogous situation when attempting to figure out what happened in human evolution: with adaptations one can usually, given a sufficiently good imagination, figure out a plausible reason why a feature might have been useful. We explain things one feature at a time -- just like that fellow Ptolemy, the Roman astronomer of the second century A.D. who simply added on another "epicycle" for each problem that needed a solution, building up a descriptive model of the motions of the heavens with dozens of rotations around different centers (rather than Copernicus' two).
      One can, presumably, "explain" everything in human evolution in that mosaic manner -- but how enlightening will that be? Explicating many disparate features with one stroke of the theorist's pen, proposing an explanatory structure that not only explains with economy but is framed to be fallible (potentially falsifiable) -- that's considered the sign of a more promising theory. In contemplating our present task, we see that there are easily a hundred features by which we humans differ from the apes -- not just language, but also plan-ahead intelligence, accurate throwing, concealed ovulation, relative brain size, hand anatomy, body hair. Not to mention pseudo-monogamy and our predilections for wagering and playing all sorts of serial-sequential games. Were there a hundred different lines of improvement, as the prevalent Ptolemaic adaptationist reasoning seems to envision -- or just a few basic inventions, each of which had multiple effects via developmental linkages or conversions of function?

WORSE YET, efficient adaptations can actually slow down the evolution of complex behaviors. That is because a major stimulus toward more elaborate organisms has been the fluctuating climate: if evolution were fast enough to track it, we'd likely see body styles fluctuating back and forth along the same path that the weather takes -- getting bigger or smaller, more or less hairy, earlier or later maturing. But with little sustained, long-term change.
      Yet evolution is often too slow to track the Earth's climate, especially given those episodes of abrupt climatic change like the Younger Dryas where the climate shifts dramatically within one generation. That generation either has what it takes, or it dies.
      And so those variants that happen along, capable of surviving various extremes of climate, will have an advantage over those aforementioned one-climate-at-a-time efficient trackers. The very slowness of evolution relative to climate change serves as a drive toward more complex organisms, those with the machinery for handling both kinds of environment. And complexity is the overall trait that underlies intelligence, primarily because new capabilities emerge from combinations of mechanisms: rather than compound interest, we have compounded mechanisms, such as those dozen behavioral strategies that omnivores need for finding their various kinds of food. The SETI meeting in Hungary offered a perfect example of what compounded mechanisms might provide, which I incorporated into my talk.
      We humans track the seasons by varying the clothing we wear. When we travel to Hungary in the summer, I noted, we have the problem of guessing whether or not we will need warm clothing. (This brought an appreciative chuckle from the audience at Balatonfüred, as the first few days of the conference had been too chilly for swimming or windsurfing offshore at lunchtime.) Those who always carry both winter and summer outfits will be safest. Those who carry only enough for one climate at a time will be less burdened. Because carry on luggage may suffice, they may get the only available taxicabs while the cautious await their checked baggage. If the weather was completely unpredictable, then everyone would have to carry along both winter and summer clothing. As long as climate fluctuations occur slowly, the more efficient packers may outreproduce those clothing-for-all-seasons types burdened by the need to be so versatile.
      But sometimes new properties arise from having both sets of clothing available at the same time (perhaps a winter coat or umbrella could be pressed into service as a sail for summer windsurfing on Lake Balaton?). And sometimes compounded mechanisms confer new "emergent" properties, quite unlike anything existing. They are true innovations, not just predictable improvements.
      This means that capabilities occasionally arrive unheralded by gradual predecessors. In the familiar case of bird flight origins, natural selection for thermal insulation shaped forelimb feathers up to the threshold for flight. Natural selection for a better airfoil shaped feathers thereafter. But the switchover from one track to another was presumably a surprise, leaving the protobirds to explore their newfound abilities rather as we might try to figure out a holiday gift that arrived without an instruction manual. The protobird's experiments were very different from adaptations, where the animal already knows how to perform and the improvement is merely a matter of efficiency.
      Inventions are the novelties in evolution, though you'd think that shaping-up streamlining was what it was all about, when reading most of the literature (most of the people doing the arguing are primarily concerned with bone-based comparative anatomy, not the broader viewpoint of comparative physiology). But adaptations are only improvements on a basic design; what we're talking about is the invention itself before streamlining, which is often a matter of a conversion of function. Nature does take leaps, and the physiological conversions of function are even faster than those anatomical leaps envisaged by proponents of punctuated equilibria and hopeful monsters.

In considering transitions of organs, it is so important to bear in mind the probability of conversion from one function to another....
Charles Darwin, 1859
THIS SECOND MAJOR ASPECT of evolutionary change is something that people often forget, fixated on what Darwin said about "a leap from structure to structure" being unlikely. But that's structure: Darwin also emphasized the role of conversions of function without anatomical change. Adaptationists often conflate the two, probably because their focus is on bones where structure is indeed closely related to function. There's more to bodies than bones.
      Darwin's teaching example of a functional conversion was the fish's swim bladder: a fish extracts gases from the blood and inflates the swim bladder just enough so the fish neither sinks nor floats to the surface (pilots will recognize this as a biological version of the "trim tab" on a rudder). Darwin suggested that when fish crawled ashore, they started exchanging blood gases with the outside air by converting the swim bladder's then-obsolete function to the new one, breathing air. For efficiency, many lobes of the swim bladder were developed and, somewhere along this path, we rename it "the lung." But a conversion of function, as in the case of those reptilian feathers on forelimbs, need not initially require an anatomical change (though, of course, they tend to follow as the new function comes under natural selection for efficiency).
      Life coming ashore surely involved quite a lot of compounding of mechanisms, as intertidal animals have to survive both in the water and in drying conditions; they acquire compounded mechanisms in consequence, organs for both environments not unlike the way that some animals (such as humans and horses) have both hair for insulation and sweat glands for getting rid of excess heat. When finally ashore, early land animals had some obsolete organs available for conversion, such as the swim bladder and the salt glands.
      Might intelligence have been aided by some conversions of function, perhaps in brain machinery?

THE BRAIN IS PROBABLY BETTER at new uses for old things than any other organ of the body. Sometimes two digestive enzymes, which each evolved separately for a different food, can act in combination to digest a third foodstuff; occasionally, nature really does provide something for nothing. (Yes, I know that this is profoundly anti-Calvinist; there is a Puritanical streak in modern evolutionary thinking that seems to require us to look for a function's antecedents in their usefulness to that very function, not some other one.) But a brain can easily combine sensory schema and movement programs in new ways, since it tends to use a common currency.
      From whatever source, an excitatory or inhibitory input is first converted into positive or negative millivolts; nerve cells then add and subtract in this substitute value system. For one input to substitute for another, it only needs to produce similar voltage changes in the relevant nerve cells. One can add apples and oranges to get so many pieces of fruit.
      This means that omnivores, with their compounding of behavioral programs for detecting and outmaneuvering many kinds of prey, can make innovations more easily than an animal evolved for eating a monotonous diet. Indeed herbivores have smaller brains than omnivores of the same body weight. Horses and bears have similar body size, but the bear's brain is somewhat larger and it is forever outsmarting the human designers of garbage cans for national parks.
      It is hard to talk about "basic units" of brain function but, for the present purpose, sensory schemas and movement programs will suffice; even if you haven't heard of them separately, you've heard of their combination, the reflex. Schema is the general term for the template inside the mind that detects a sensory pattern in time and space; movement programs such as breathing can often be decomposed into subprograms, such as for inspiration and expiration. When a schema and movement program are firmly linked, we tend to call the combination a reflex, as when the silhouette of a hawk overhead causes a baby bird to crouch down in concealment. We once thought that all movements were guided by reflexes; now we know that some are innate, capable of being executed spontaneously and without any sensory guidance. Feedback tends to be important when first learning a new task, or when the task is quite varied (each time I pick up my tea cup, its weight is somewhat different and my posture has probably altered too) -- feedback helps shape up the movement program.
      What is surprising about schemas is what seems to suffice -- some are quite crude, not even the equivalent of a cartoon sketch. Some shore birds, for example, may recognize their own young by proximity to the nest: let a chick stray outside the parents' territory, and it will be attacked when it returns, as if a total stranger. Male flickers (a colorful woodpecker) have a black "moustache" stripe on the side of their heads; paint such a stripe on a female, and her mate will attack her as if she were a total stranger. The cuckoo practices parenthood piracy successfully because the "foster parents" fail to recognize their young except via a brightly colored throat, which cuckoo chicks mimic and so are fed, even when absurdly larger in size than the foster parents. Some birds will preferentially incubate the larger eggs in a nest -- and so one can see a small bird sitting atop a large chicken egg placed in its nest by experimenters, ignoring its own small eggs.
      Absurd? Evolution often is -- that's because good-enough solutions may suffice (Simon's "satisfice"), and evolution never gets around to finding solutions for the occasional problems. Our first-generation household robots are going to be characterized by similar stupidities, and we will tell each other stories of how our robot threw out the umbrellas with the trash, mistaking the umbrella stand for a wastebasket.

INBORN SCHEMAS certainly exist, but the hawk-overhead protective crouch shows how indirectly the detection may be accomplished. We eventually realized that chicks initially crouch down when any bird flies over, but soon habituate to the more familiar species that they see every day. Then only rarely-seen birds trigger the reflex. Some birds are rare because they are exotics, just passing through. But other species are few in numbers because they are at the top of the food web; birds that eat other birds cannot be as numerous as their prey species. So the simple habituation serves, at the cost of some false alarms, to tune up the chick to the local predator species, whatever it is. Thanks to the population statistics of various species in the food web, all it needs as "inborn" is the generalized bird-overhead template, an ability to learn new schemas (special cases of the more general inborn type), and the ability to use learned schemas to cancel the primitive reflex.
      Movement programs can be tuned up too, enhancing and suppressing features with both genetic variations and learning within individual lifetimes. The horse's "pacing" gait, where both left legs move forward together, then both right, is infrequently seen in nature, but selective breeding can bring it out. A few humans have the ability to wiggle their ears, and it seems likely that many others could learn to do so with sufficient shaping by a skillful coach. Some movement variations turn out to be useful in certain situations; a dog that tends to circle a few times before lying down would, in the context of grasslands, create a better nest for itself. Charles Darwin saw this potential for variation and selection of such behaviors in his 1872 book on the expression of emotions.
      Nature is always throwing up new variations, thanks to the shuffling of genes when making new sperm and ova. We tend to think of the unusual phenotypes as "defective" (the 15 percent of children who have difficulty learning to read) or as "gifted" (unusual musical and artistic abilities) -- but we are all just variations on a series of themes, thrown up for the present environment to evaluate. We are all nonstandard because there is no standard (that "escape from the Platonic essence" was the initial ingredient of Darwin's great insight that allowed him to conceive how evolution works). When the variations are easy to see (thin and fat, short and tall, light skins and dark), we give them names -- but when they just involve brain wiring, as many of them do, then they are less readily recognized.
      When a new way of making a living comes along, perhaps extracting insects from holes with a probe, these brain variations make it easier for some individuals to learn the new task. Perhaps they have a predisposition to chew on the ends of sticks (like some children I know) and so are likely to manufacture better "fishing" sticks. More of their offspring survive than others, and so variations on the new theme get tried out, some of which are even better at fishing for insects. Eventually some body features change, in addition to the more subtle brain features, as when the precision grip modifications are made to the fingers.
      Because learning within an individual lifetime is easier than brain-wiring variations in successive generations, which are in turn easier than gross body changes, behavior tends to lead the way in exploring new evolutionary pathways. Squirrels that seek food in the tops of trees may have to climb all the way up a tree, then all the way back down, and across the ground to another tree. But if its ability to leap across rocks on the ground can be extended to leaping between branches in the tree tops (perhaps because a variation arose that had less fear of heights), then it might become a more efficient food-finder, despite the fraction of the population fatally injured by falls. If the climate then cooled, so that the forest thinned out, squirrels that could glide between trees would become the most efficient at feeding their offspring. And so we might see the squirrel's skin become flabby, as those with that variation would now have a more suitable airfoil for gliding from the top of one tree to halfway down a neighboring tree. They'd get to the food faster, and out-fox the foxes watching from the ground for a squirrel to descend.

SO OUR KIND OF INTELLIGENCE may not be the inevitable outcome of some general principle of nature. And, while it might happen via gradual adaptations, surely there were a number of speeding-things-up surprises along the way as animals discovered previously untried combinations of sensory schemas and movement programs that proved handy for new ways of finding food, avoiding predators, or acquiring mates.
      The natural history of intelligence may turn out to be a prolonged version of the progress we've recently seen from special-purpose computers to the modern general-purpose computer. The basic techniques evolved in the nineteenth century with the programmable Jaccard loom, the punched card sorting machines, and the mechanical hand-operated calculators. By the time of World War II, special-purpose computers were constructed for pointing antiaircraft guns and breaking ciphers -- and these machines were internally so similar that we began to see general-purpose computers, able to switch from one task to another. In less than a half-century, schoolchildren possess computers far more versatile than those once nurtured in air-conditioned warehouses by legions of experts.
      Did we get our general-purpose brains, capable of tasks like reading that were never involved in their evolution, via a similar series of special-purpose adaptations for finding fruit and catching meat? Did our fruit-finding ancestors owe their versatility to something similar? Certainly developmental coupling, mechanism compounding, and functional conversions are -- each of them -- a theme as important as the usual adaptationist efficiency. But we must look at the interaction between a flexible diet and a changeable climate to see how hominid brains might have been "pumped up." It doesn't take climate change as abrupt or dramatic as the Younger Dryas to pump up behavioral versatility; the ice ages have had plenty of merely rapid changes as well.
      And even if we explain the origins of Homo sapiens, there is still the problem of accounting for how Hungarians happened.
     

[As] my conclusions have lately been much misrepresented, and it has been stated that I attribute the modification of species exclusively to natural selection, I may be permitted to remark that in the first edition of this work, and subsequently, I placed in a conspicuous position -- namely at the close of the Introduction -- the following words: "I am convinced that natural selection has been the main but not the exclusive means of modification." This has been of no avail. Great is the power of steady misrepresentation....
Charles Darwin,
in a late edition of On the Origin of Species
The great synthesizer who alters the outlook of a generation, who suddenly produces a kaleidoscopic change in our vision of the world, is apt to be the most envied, feared, and hated man among his contemporaries. Almost by instinct they feel in him the seed of a new order; they sense, even as they anathematize him, the passing away of the sane, substantial world they have long inhabited. Such a man is a kind of lens or gathering point through which thought gathers, is reorganized, and radiates outward again in new forms.
Loren Eiseley, 1973

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