Why Life Is So Complex (A Brief History of Organic Life)

While we readily admit that the first organisms were bacteria-like and that the most complex organism of all is our own kind, it is considered bad form to take this as any kind of progression. . . . [One] is flirting with sin if one says a worm is a lower animal and a vertebrate is a higher animal, even though their fossil origins will be found in lower and higher strata.

—John Tyler Bonner

Among the technological feats of the twentieth century is the invention of "binary" chemical weapons. Two chemicals, harmless when separate, are toxic if combined. Safe to transport, deadly when deployed.

This was not an original idea. It had appeared millions of years earlier, in the form of the bombardier beetle. In one tank the beetle carries a harmless chemical mix. In another tank resides a catalyst. The beetle adds the catalyst to the mix upon deployment, creating a scalding substance that, via pliable rear-end nozzle, is showered precisely on nearby tormentors.

The bombardier beetle is an example of the evolution of complexity. Clearly, a beetle equipped with two separate munitions compartments and a spray nozzle is more complex than the same beetle lacking these accoutrements.

And this isn’t just any old kind of biological complexity. The beetle’s arsenal involves behavioral complexity. Toxic nozzles aren’t much good unless you can aim them and squirt them.

Aiming and squirting—like any impressive behavior—involves information processing, a command-and-control system. In some small measure, then, evolution’s elevation of the beetle to bombardier rank involved a growth in intelligence. In other lineages, of course, the evolution of intelligence—of behavioral complexity—has proceeded further. And we have binary chemical weapons, among other things, to show for it.

Was all of this in the cards? I don’t mean binary chemical weapons, or bombardier beetles, or human beings or any other particular thing or species on this planet. I mean the evolution of complexity and intelligence. Did basic properties of natural selection make it very likely that someday some animal would be smart enough to invent neat gadgets? And figure out that the earth revolves around the sun? And ponder the mind-body problem? Does biological evolution intrinsically favor the growth of biological complexity—including behavioral flexibility, and its underpinning, intelligence? Is this biological "progress" somehow natural?

It has long been unfashionable to answer yes. One big reason is the same big reason that made belief in directional cultural evolution so unfashionable: past political misuse.

Early this century, biological progressivism was dear to the hearts of "social Darwinists," who used it to justify things like racism, imperialism, and laissez-faire indifference to poverty. The logic behind social Darwinism—to the extent that it had a coherent logic— was something like the following: The suffering, even death, of the weak at the hands of the strong is an example of "survival of the fittest." And the "survival of the fittest" has God’s blessing. And how do we know that the "survival of the fittest" has God’s blessing? Because He built the dynamic into His great creative process, natural selection. And how do we know that natural selection is God’s handiwork? Because of its inexorable tendency to create organisms as majestic as ourselves, organisms worthy of admission to heaven! In short, biological progressivism was used to deify nature in all its aspects, and nature, thus deified, was invoked in support of oppression.

The philosophical confusions underlying social Darwinism have been much analyzed, and we needn’t repeat the exercise. Nor need we mull the question of whether nature is "good" or "bad"—at least, not until we enter the realm of spiritual speculation several topics hence. For now we can let matter rest with a simple and obvious point: the fact that an academic proposition can be misused doesn’t mean it’s not true.

So is it true? Is organic evolution directional? Do basic properties of natural selection pretty much ensure the evolution of complexity, including behavioral complexity, and thus, given long enough, the evolution of great intelligence? Some people say yes (often quietly), and some say no. The most prominent, persistent, and passionate sayer of no is the noted paleontologist Stephen Jay Gould. He has devoted two topics to denying "that progress defines the history of life or even exists as a general trend at all." To see progress in evolution, he says, is to indulge a "delusion" grounded in "human arrogance" and desperate "hope"—the hope that we are anointed by nature to sit on its throne, that we are "a predictable result of an inherently progressive process." Gould recommends that we wake up and smell the coffee, confront the harsh prospect that "we are, whatever our glories and accomplishments, a momentary cosmic accident that would never arise again if the tree of life could be replanted from seed and regrown under similar conditions." And he doesn’t mean "we" narrowly—not just Homo sapiens. If you replayed evolution on this planet, the chances of getting any species smart enough to reflect on itself are "extremely small."


Before seeing what is wrong with Gould’s claim, we need to first see that it isn’t as sweeping as it sounds. You might think that when he says progress is not a general evolutionary trend, he is saying that evolution doesn’t tend to produce more and more complex forms of life over time. But he isn’t. He concedes that the outer envelope of organic complexity may tend to rise—that "the most complex creature may increase in elaboration through time." Nor is he saying that the average complexity of all species hows no trend. "Life’s mean complexity may have increased," he allows.

Okay. So in what sense doesn’t complexity tend to grow via natural selection?

For starters, a few species have gotten less complex through evolution. And many species have gone long periods with little if any growth in complexity. Bacteria showed up billions of years ago, and there’s a lot of them still around, evincing no aspiration to climb higher on the tree of life. This point is widely accepted by biologists, as is its upshot: that "orthogenesis"—some sort of mystical inner impetus toward higher complexity, pervading all of life—doesn’t exist. Surely Gould is saying more than this?

Yes. He’s saying not only that bacteria are pretty simple creatures; he’s saying that they outnumber us. Or, as he puts it: "modal" complexity hows no tendency to grow; the level of complexity at which the greatest number of living things resides—the mode—has not changed noticeably since at least 2 billion years ago. Back then, most living things were about as complex as a bacterium. One billion years ago, ditto. Now, ditto.

Indeed, not only do bacteria outnumber us; they outweigh us. In fact, they outweigh just about anything, if you add up all the underground bacteria. Also, they can survive under lots of weird conditions. "On any possible, reasonable, or fair criterion, bacteria are—and always have been—the dominant form of life on earth."

Actually, some people who consider themselves reasonable and fair might opt for an alternative definition of dominance. For example: "ability to blow up the planet," or "ability to figure out how all life forms were created" or "ability to create whole new life forms," or just "ability to put bacteria under a microscope." Gould will have none of this talk. To make the most complex form of life on earth our bellwether for progressive trends would be to exhibit a "myopic focus on extreme values only."

Well, maybe. On the other hand, "extreme values" are the only reason most people care about the question of biological progress to begin with. Was something as complex as us, a behaviorally flexible as us, as smart as us—something with our "extreme values"— likely to evolve? Indeed, take away this question, and Gould himself probably wouldn’t be terribly interested in the subject. A his writing makes clear, his pet peeve with biological progressivism is its past political overtones, notably social Darwinism.

In short, the issue of "extreme values"—the issue of what happens at the outer envelope of biological complexity, where our species resides—is the only reason we’re having this discussion to begin with. Gould seems to realize this. For he doesn’t, in the end, rest his case on the stagnation of "modal" complexity. He returns at length to the supposedly extraneous issue of extreme values. Having conceded that the outer envelope of complexity may tend to grow, he proceeds to argue that this growth is not truly "directional," but rather "random."

What does he mean by random? Consider a drunken man walking down a sidewalk that runs east-west. Skirting the sidewalk’s south side is a brick wall, and on the sidewalk’s north side is a curb and a street. Will the drunk eventually veer off the curb, into the street? Probably. Does this mean he has a "northerly directional tendency"? No. He’s just as likely to veer south as north. But when he veers south the wall bounces him back to the north. He is taking a "random walk" that just seems to have a directional tendency.

If you get enough drunks and give them enough time, one will eventually get all the way to the other side of the street (notwithstanding traffic fatalities involving other, less lucky drunks). That’s us: the lucky species that, through millions of year of random motion, happened to get to the far north. But we didn’t get there because north is an inherently valuable place to be; indeed, if it weren’t for the brick wall, there would be just as many drunks south of the sidewalk as north of it, and the randomness of all our paths would be obvious. That is: if it weren’t for the fact that no species can have less than zero complexity, the history of life wouldn’t look like a natural progression. Gould writes: "The vaunted progress of life is really random motion away from simple beginnings, not directed impetus toward inherently advantageous complexity."

Again, as with Gould’s emphasis on stagnant "modal" complexity, one might ask how much this argument really matters for philosophical purposes. The question behind this whole exercise, remember, is whether the evolution of something as smart and complex as us was very likely. If the combination of a "random walk" and a "wall of zero complexity" leads people to conclude that the answer is yes, then, well, their answer is yes. If, as Gould fears, people are inclined to take a "yes" answer as evidence of higher purpose, they probably aren’t going to be too picky about the exact type of "yes." God, they will say, works in strange and wondrous ways.

Still, the more factors favoring the evolution of complexity, the more irresistible the "yes" answer is—the more likely evolution was to eventually reach a human level of intelligence. So it’s worth seeing if Gould is somehow overlooking "non-random" factors that are conducive to complexity. He is. They fall under the rubric of "positive feedback"—the evolution of complexity strengthens the logic behind the evolution of complexity, which strengthens the logic behind the evolution of complexity . . . and so on.


Consider, again, the bombardier beetle. Since there was a time when beetles didn’t exist, there must have been a time when no animals were specially adapted to kill and eat them. Then beetles came along. Then various animals did acquire, by natural selection, the means to kill and eat them. This expansion of behavioral repertoire, itself a growth in complexity, spurred a response: the bombardier beetle’s bombardieresque qualities—a counter-growth in complexity. Thus does complexity breed complexity.

One might expect that, given long enough, beetle predators will undergo a counter-counter-growth in complexity, some way to neutralize the beetle’s noxious spray. In fact, they already have. Skunks and one species of mice, the biologists James Gould (no relation) and William Keeton have written, "have evolved specialized innate behavior patterns that cause the spray to be discharged harmlessly, and they can then eat the beetles." Until the next round of innovation, at least.

The technical term for this dynamic is the same as the nontechnical term: an "arms race." Over the past two decades, various prominent biologists—Richard Dawkins, John Tyler Bonner—have noted how arms races favor the evolution of complexity. Gould, in the course of a two-volume meditation on the evolution of complexity, doesn’t mention the phenomenon.

Finding evidence of arms races in the fossil record is tricky. But one venturesome scientist methodically measured the remnants of various mammalian lineages spanning tens of millions of year and found a suggestive pattern. In North America, the "relative brain size" of carnivorous mammals—brain size corrected for body size—showed a strong tendency to grow over time. And so did the relative brain size of the herbivorous mammals that were their prey. Meanwhile, comparable South American herbivores, which faced no predators, showed almost no growth in relative brain size. Apparently ongoing species-against-species duels are conducive to progress.

Arms races can happen within species, not just between them. Ever spend months observing a chimpanzee colony in painstaking detail? A few primatologists have. The male chimps, it turns out, spend lots of time scheming to top each other. They form coalitions that, on attaining political dominance, get special sexual access to ovulating females—at the great Darwinian expense of less successful coalitions. So males with genes conducive to political savviness should on average get the most genes into the next generation, raising the average level of savviness. And the savvier the average chimp, the savvier chimps have to be to excel in the next round. And so on: an arms race in savviness—that is, an arms race in behavioral flexibility. There’s little doubt that this dynamic has helped make chimps as smart as they are, and there’s no clear reason why the process should stop where it is now.

Meanwhile, female chimps also exhibit political skills, of a somewhat different sort, that raise the survival prospects for their young. Here, too, genes for savviness should in theory not only prevail and fill the gene pool, but, having prevailed, create selective pressure for yet more savviness. It would always pay female chimps to be smarter than average, and, for that very reason, the average would keep rising. Positive feedback.

Natural selection, as described by Gould, has no room for this sort of directional dynamic. "Natural selection talks only about ‘adaptation to changing local environments,’ " he writes. And "the sequence of local environments in any one place should be effectively random through geological time—the seas come in and the seas go out, the weather gets colder, then hotter, etc. If organisms are tracking local environments by natural selection, then their evolutionary history should be effectively random as well." This would be good logic if environments consisted entirely of seas and air. But in the real world, a living thing’s environment consists largely of—mostly of—other living things: things it eats, things that eat it, not to mention members of its own species that compete with it and consort with it. And no one—not even Gould—denies that the average complexity of all species constituting this organic environment tends to grow. So the sequence of environments isn’t "effectively random" over time; there is a trend toward environmental complexity.

And it wouldn’t matter if we assumed, along with Gould, that back at the dawn of life the growth in average complexity was wholly random, like the stumbling drunk’s path. The fact would remain that, for whatever reason, environmental complexity started to grow. Species, in "tracking" this growth of complexity, can’t be described as stumbling around randomly. Their evolutionary change is, by Gould’s own definition, directional. And, since they are themselves part of the environment for other species, the process is self-reinforcing. More positive feedback.

A number of evolutionists have argued that if you look at a colony of chimps (our nearest living relatives), you can see some of the social dynamics that pushed our own ape ancestor in our direction, toward greater intelligence. Probably so. In any event, some non-random forces seem to have done the pushing. To the extent that we can judge from an imperfect fossil record, the growth in brain size—from Australopithecus africanus to Homo habilis to Homo erectus to early Homo sapiens to modern Homo sapiens—is brisk, with no signs of backtracking and little in the way of pauses. It looks for all the world like 3 million year of pretty persistent brain expansion.

How does Gould explain this trend? Not readily. The only explanation his worldview would seem to allow is a long series of lucky coin flips—the most serendipitous drunken walk in the history of drinking. Indeed, luck of this caliber might be enough to make you suspect that the coin flips were divinely guided! It’s no surprise that the creationist literature contains many approving citations of Gould’s work. If his view of natural selection were correct, I would be a creationist, too; natural selection would not be a plausible means of human creation—at least, not of such rapid human creation.

And it isn’t just our ancestor that, in Gould’s scheme, were so lucky. Mammalian lineages broadly exhibit a movement toward braininess. True, individual species can spend a long time without getting noticeably smarter. But examples of mammals—or for that matter multicellular creatures in general—evolving toward less braininess are vanishingly rare. What a lucky bunch of drunks animals are!


There is another sense in which luck might figure in evolution, and this sense was the centerpiece of Gould’s first topic-length assault on biological progressivism, Wonderful Life. The topic was about the fossils of the Burgess Shale, products of an apparently sudden (as these things go) expansion of biological diversity around 570 million years ago, at the beginning of the Cambrian period. This "Cambrian explosion" is the first well-documented flourishing of multicellular life. Drawing on the work of British paleontologists who had studied the Burgess Shale, Gould used the fossils as a case study in the decisive role of chance, of "contingency."

Gould’s argument was simple. Some of the fossils are very weird-looking, and seem to fit into none of the basic categories of animals that now populate the earth. These weird species, or their descendants, apparently went extinct—and through no evident fault of their own. They must have fallen prey to bad luck, some sudden, unpredictable shift in ecology for which their past evolution hadn’t prepared them. Had this random shift not happened, had these oddballs proved enduringly prolific, today’s tree of life would presumably look very different. Thus can a roll of the cosmic dice fundamentally alter the future of evolution.

Since Gould’s topic was published, his interpretation of these fossils has been challenged by a number of paleontologists. It now looks as if the Burgess Shale animals aren’t nearly as weird as Gould and some other researchers first thought; for the most part the animals fit readily into a standard taxonomic tree, and their descendants are with us still. In the case of a fossil so bizarre-looking that it was named Hallucigenia, Gould—following the prevailing interpretation—was looking at it upside down. Those baffling squiggly things on its "back" were legs. And those strangely spiky "legs" were spikes—armor, presumably the product of an arms race.

Notwithstanding this revisionism, part of Gould’s argument is surely valid. Whether or not the Burgess Shale animals are a case in point, species do go extinct because of cosmic rolls of the dice. A big meteor happens to head toward Earth and then—poof!—no dinosaur . This sort of sudden and unpredictable emptying of niches undoubtedly shapes future evolution.

To take an example of particular interest: if our ancestor had been wiped out through bad luck, then indeed, as Gould has repeatedly proclaimed, human beings would never have evolved. This point—in some ways the central point of Wonderful Life—is so unarguable that, so far as I know, it has never been argued against; no sober biologist would claim that there was some kind of inexorability to the evolution of Homo sapiens per se—you know, a species five or six feet tall with armpits, bad jokes, and all the rest. The only serious question is whether the evolution of some form of highly intelligent life was likely all along—some animal smart enough, for example, to be aware of itself.

Gould skirted this question in Wonderful Life, but he later said, in Full House, that the answer is no. It might be tempting to agree, if Gould’s argument about the drunken stumbling of life were valid. But given the manifest existence of arms races, and the manifest premium those races place on behavioral flexibility, and the ongoing growth in behavioral flexibility since animals showed up on the scene a bit before the Cambrian explosion, the temptation to agree with Gould is quite resistible.

Resistance is rendered even easier when we consider what, at the risk of anthropomorphizing nature, I can only call natural selection’s genius. Though a blind process that works by trial and error—and random trial, at that—it has a remarkable knack for invention, for finding and filling empty niches. It has adapted animals to life on land, underground, under water, in trees, in the air.

In the course of this niche-filling, natural selection doesn’t just invent remarkable technologies; it keeps reinventing them. Flight and eyesight are two technologies so amazing that they are commonly cited by creationists for their implausibility. Yet flight has arisen through evolution on at least three separate occasions, and eyes have been independently invented dozens of times.

Why are eyes such a favorite of natural selection’s? Because light is a terrific medium of perception. It moves in straight lines, bounces off solid things, and travels faster than anything in the known universe.

This isn’t to say that natural selection is single-minded in its devotion to light. The other familiar senses—smell, sound, touch, taste—are all amply represented in the animal kingdom, and are just the beginning of a long list of organic data-gathering technologies. Indeed, humankind’s vaunted twentieth-century advances in sensory technology almost seem like a long exercise in reinventing the wheel. We now have infrared sensors for night vision; rattlesnakes beat us to that one. We use sonar—old hat to bats, and standard equipment on dolphins. Some burglar alarms work by creating electric fields and sensing disturbances in them; so do some fish—the elephant-snout fish of Africa and the banded knifefish of South America (not to be confused with fish that use electricity to stun, such as the 450-volt electric catfish of Africa, or the 650-volt electric eel of South America). Of course, people fathomed the informational value of the earth’s magnetic field long before the twentieth century—with compasses—but not as long ago as natural selection did; some bacteria, and various more complex creatures, use the same trick for orientation.

Why is natural selection so attentive to sensory technologies? Because they facilitate adaptively flexible behavior. And what else facilitates adaptively flexible behavior? The ability to process all of this sensory data and adjust behavior accordingly. In other words: brains. Not our brains, necessarily, or even "brains" in the technical sense of the term, but rather intelligence as an abstract property. It is natural selection’s demonstrable affinity for certain properties—its tendency to invent them and nurture them independently in myriad species—that renders trivial Gould’s truism about how bad luck can wipe out any one species or group of species. At least, it is trivial so long as we’re discussing the likelihood of the evolution of the property of great intelligence, and not the evolution of a particular intelligent species.

Simon Conway Morris, one of the paleontologists whose Burgess Shale research Gould relied on most heavily in writing Wonderful Life, made this general point in a broadside critique of Gould’s interpretation of the Shale fossils. "[T]he role of contingency in individual history has little bearing on the likelihood of the emergence of a particular biological property," Morris wrote. The fates of individual species may depend on the luck of the draw. But the properties they embody were in the cards—at least, in the sense that the deck was stacked heavily in their favor.

Consider again the property of eyesight. Obviously, given that natural selection has invented it dozens of times, the chance extinction of even large groups of species possessing it wouldn’t much affect the big picture; the property of eyesight was destined for invention and reinvention. Can we make the same argument about intelligence? Not exactly, because intelligence is a different kind of property. If we define it broadly enough—as the processing of information to orchestrate adaptively flexible behavior— then it appeared too early, at the very base of the tree of life, to have kept getting reinvented later, in the tree’s branches. (Even E. coli bacteria "know" enough to find new environs if their surroundings are low in carbon.) Still, greater intelligence is something that has been invented billions of times. In all kinds of animal lineages—in mammals, fish, reptiles, insects, birds—there has been extensive growth in behavioral flexibility, and the growth has often come in mall increments. Add up all those increments, and what do you have? A pattern.

Gould writes: "Humans are here by the luck of the draw." True. But a human level of intelligence isn’t. Given long enough, it was very, very likely to evolve. At least, that’s my reading of the evidence. It is also the reading of some eminent evolutionary biologists, such as William D. Hamilton and Edward O. Wilson, though other eminences, such as Ernst Mayr, disagree.


Sometimes people (including Gould) skeptically ask why, if the evolution of intelligence was so likely, it took so long. It’s a fair question. Remember the great evolutionary thresholds noted in the previous topic—from prokaryote to eukaryote, from eukaryote to full-fledged multicelled animals? Sounds like a pretty brisk progression when you say it in one sentence like that, but in fact getting from stage one to stage two may have taken as long as 2 billion years. And it seems to have taken at least another 700 million years to get from eukaryotes to the squishy, blobbish multicelled animals that show up in the fossil record before the Cambrian explosion. Then things really get rolling: from squishy blobs to human beings in around 600 million years. Why did organic evolution spend so much time in pre-explosion mode, twiddling its thumbs?

You could ask the same question about cultural evolution. If it was so powerful, why did it take so long to get off the ground? What were all those people in the Middle Paleolithic waiting for? As it happens, the explanation for these two forms of early sluggishness— cultural evolution’s and biological evolution’s—are, broadly speaking, the same.

Think back in time to the primordial ooze, when the first cells had come into being. Each of them was a potential source of innovation; by genetic mutation, it could come up with a new "idea" for organic design, and that idea, if good, could spread. In a sense, these cells collectively constituted natural selection’s "brain."

But mutations—new "ideas"—don’t happen all that often, and anyway most mutant ideas aren’t good. So it’s going to take a whole lot of bacteria to constitute a "brain" big enough to think up good ideas very often. No doubt a good part of life’s early life was spent slowly but relentlessly raising the size of the brain. The process is comparable to the tens of thousands of years humans spent slowly increasing their numbers, until finally, around 15,000 years ago, the invisible brain was big enough to generate innovations at a more-than-glacial rate.

Of course, human innovation wasn’t totally lacking amid the sparse population of the Paleolithic; in fact, it had been slowly accelerating as population slowly grew. And much the same can be said about the early history of organic life. During those 2 billion years before the invention of eukaryotic cells, important work got done. For example, a series of obscure-sounding but important energy technologies appeared: first "autotrophy" then "photophosphorylation" (a way of getting energy from light) and then full-fledged photosynthesis, which incorporated photophosphorylation. So natural selection wasn’t just twiddling its thumbs during stage one. In fact, even symbiotic division of labor among cells made an appearance. Gooey mats of wall-to-wall bacteria, it now seems, consisted of two kinds of cells: the ones on top photosynthesized, and the ones on the bottom made a living breaking down the photosynthesizers’ waste products by fermentation.

These gooey mats illustrate a key point. The growth of natural selection’s "brain" consists not just of rising numbers of organisms, but, at least as crucially, of rising numbers of species. A fermenting bacterium can, by mutation, generate a whole different set of "ideas" than a photosynthesizing bacterium can—new approaches to fermentation, for one. Each new species opens up new "design space," expanding not just the chances of a good idea, but the spectrum of possible ideas.

There’s a second sense in which each new species expands design space. Each species is—like the photosynthesizing cell—a potential energy source, just begging natural selection to create a species tailored to exploiting it. In the case of the photosynthesizing cell—exploited by the fermenting cell—the form of exploitation was harmless: the new species simply recycled the waste of the old species. Often the form of exploitation is more predatory. But whatever the form—harmless, predatory, parasitic, mutualistic, whatever—it always seems to arrive sooner or later. Each new species opens up a potential niche, and natural selection excels at filling niches.

Consider the technological opportunities opened by the nectar in flowers. It has led to the hummingbird’s long bill and its soft, stationary flight, to the beehive’s amazing collective intelligence, and to many other marvels of flower exploitation. And because, from the flower’s point of view, these animals usefully transport pollen, flowers have evolved various ways of encouraging the transport. Some flowers briefly entrap beetles, coating them thoroughly in pollen before letting them head for another plant. Scarlet gilia plants in Arizona put out red flowers, which attract hummingbirds, and then shift to an all-white floral display, which attracts moths, after the hummingbirds leave in late summer. Some orchids have come to resemble bees or wasps or flies, thus attracting a male bee or wasp or fly that, in confusedly trying to mate with the flower, picks up its pollen (and sometimes, poignantly, leaves sperm behind).

The diversification of flower in turn diversifies their symbionts. Different species of hummingbirds have different bill lengths, depending on their favorite flowers, just as different bee species have different tongue lengths.

And so it goes. Growth in the number of species is assured, in the first place, by the expanse and heterogeneity of the earth; a vibrant, spreading species gets split up by mountains or rivers or deserts or meadows or oceans—or heer distance—and then its fragments adapt to the peculiar contours of the local ecosystem. And, because each new species itself defines a new potential niche for another species, the more species there are, the more there will be. Once again, complexity begets complexity by positive feedback, but in this case it is the complexity of the whole ecosystem that expands. Thus does the size and fertility of natural selection’s brain so assuredly grow—slowly at first, but inexorably.

Apparently the brain was getting pretty big and fertile by the time eukaryotes showed up. For the age of eukaryotes seems to have lasted only a third as long as the age of prokaryotes before the epic threshold to multicelled animals was crossed. And, then, around 600 million years ago, came the "explosion" in the diversity of animal life.

Why the "explosion"? Personally, I suspect that the abruptness of Cambrian creativity has been exaggerated by an imperfect fossil record; and that the main story is continued creative acceleration of the sort you’d expect from a process whose inventiveness grows with the number of past inventions. Still, there probably was some abruptness, and it’s instructive to ask why. Perhaps the most commonly cited reason is the coming of multicellular predation. Much as evidence of human war is abundant by the time cultural evolution takes off, evidence of predation shows up near the beginning of the Cambrian. Fossilized tracks of the now-extinct trilobite can be seen homing in on the tracks of a Cambrian worm—and only the trilobite’s tracks emerge from the intersection.

In addition to this anecdotal evidence of predation, there is the sudden trendiness, during the Cambrian, of body armor. Exoskeletons and shells presumably were round two in an arms race, predation having been round one. And, of course, some armaments are more high tech than body armor and sharp teeth. It is fitting that the trilobite, the first predator ever to leave a record of its crime, is also the first animal known to have had eyes. Once you’re hunting mobile prey, acute perception helps. Once you’re hunted, ditto. And acute perception isn’t worth much without rapid data processing.

Biological evolution’s gathering momentum, then, had come from two forces that also figured in cultural evolution. First, there was the gradual elevation of the likelihood of innovation, via the growth of a giant "brain." Then there was intensifying competition that raised selective pressure; in both biological and cultural evolution, zero-sumness stimulated non-zero-sumness. Why did the trilobite’s genes for legs come to play a nonzero-sum game with subsequently added genes for eyes—and with genes for processing and applying visual data? Because it’s a jungle out there.


We could say more about why evolution was so slow in the beginning. In particular, we could elaborate on how daunting some of the simple-sounding thresholds to higher organic organization are. The invention of eukaryotic cells, and of multicellular life, entail enough knotty mechanical problems to raise one’s respect for natural selection’s ingenuity.

At the same time, these problems hould caution biological progressivists, such as me, against too easily asserting the inexorability of complexity’s growth. Though in ome ways organic evolution is a tidier process than human history, in other ways it is more imponderable. And I must admit to feeling slightly less confident in estimating the probability of the biological evolution of a human-caliber brain than in estimating the likelihood—given that brain—of the cultural evolution of a global brain like the one now emerging.

Still, just as the number of independent evolutions of chiefdoms and civilizations valuably corroborates other arguments for cultural evolution’s directionality, the number of times that life has passed through particular organizational thresholds carries great weight. Multicellularity was invented over and over again—more than ten times, by some estimates. A proliferation of multicellular life, then, would seem to have been in the cards, with or without the Cambrian "explosion."

The likelihood of the earlier crossing of the eukaryotic threshold is a murkier issue. Of the eukaryotic cell’s various organelles—including energy-processing units such as mitochondria and chloroplasts—its defining element is the nucleus. And it’s not clear whether the nucleus was invented more than once. But organelles of one kind or another have been "invented"—sometimes in the symbiotic fashion described in the previous topic, sometimes not—on a number of different occasions. In this sense, at least, the logic behind the complexification of simple cells seems to have been powerful.

In 1951, the British zoologist J. W. S. Pringle wrote a technical paper called "On the parallel between learning and evolution." It is not a bad comparison. Natural selection is not just a process that "invents" new technologies, such as eyes; it implicitly "discovers" properties of the physical world, such as light’s reflection. It is this ongoing invention, and implicit discovery, that is an essential, predictable part of evolution by natural selection. The particular species embodying the "learning" are incidental—transient repositories of knowledge.

No technology has received more attention from natural selection, more refinement via natural selection, than intelligence. Everywhere around us is evidence of the tendency of intelligence to grow through evolution. The most spectacular example, with all due humility, is us. The human brain is the greatest product yet of the larger, throbbing, endlessly inventive "brain" that is the biosphere.

Of course, intelligence is not a one-dimensional thing. We are not just ten times as smart a dinosaur or a thousand times as smart as bombardier beetles or a million times as smart a bacteria. Our intelligence is qualitatively different from theirs. It has a number of attributes that, together, have created culture—a culture, moreover, that is rich enough to become an evolutionary force in its own right.

In that sense, some of the terminology in this topic has been a bit loose. I’ve spoken about the likelihood of evolution’s reaching a human "level" of intelligence—as if there were some simple ladder of animal IQ that life climbs, with "self-awareness" at its acme. My excuse, I guess, is that biologists, including Gould, often talk this way—and that, moreover, talking this way is fine for some purposes. But for the purposes of this topic’s argument, we’ll have to now get a little subtler in our conception of intelligence. If you want to know how likely the coming of cultural evolution was, it isn’t enough to argue that natural selection tends to create smarter and smarter things. These things have to be smart in several particular ways if full-fledged cultural evolution is to get rolling. Was this kind of smartness in the cards?

Next post:

Previous post: