I once overheard a children’s conversation in a New York playground. Two young girls were discussing the size of dogs. One asked: “Can a dog be as large as an elephant?” Her friend responded: “No if it were as big as an elephant, it would look like an elephant.” How truly she spoke.
22 | Sizing Up Human Intelligence
A | HUMAN BODIES
“Size,” Julian Huxley once remarked, “has a fascination of its own.” We stock our zoos with elephants, hippopotamuses, giraffes, and gorillas; who among you was not rooting for King Kong in his various battles atop tall buildings? This focus on the few creatures larger than ourselves has distorted our conception of our own size. Most people think that Homo sapiens is a creature of only modest dimensions. In fact, humans are among the largest animals on earth; more than 99 percent of animal species are smaller than we are. Of 190 species in our own order of primate mammals, only the gorilla regularly exceeds us in size.
In our self-appointed role as planetary ruler, we have taken great interest in cataloging the features that permitted us to attain this lofty estate. Our brain, upright posture, development of speech, and group hunting (to name just a few) are often cited, but I have been struck by how rarely our large size has been recognized as a controlling factor of our evolutionary progress.
Despite its low reputation in certain circles, self-conscious intelligence is surely the sine qua non of our current status. Could we have evolved it at much smaller body sizes? One day, at the New York World’s Fair in 1964, I entered the Hall of Free Enterprise to escape the rain. Inside, prominently displayed, was an ant colony bearing the sign: “Twenty million years of evolutionary stagnation. Why? Because the ant colony is a socialist, totalitarian system.” The statement scarcely requires serious attention; nonetheless, I should point out that ants are doing very well for themselves, and that it is their size rather than their social structure that precludes high mental capacity.
In this age of the transistor, we can put radios in watch-cases and bug telephones with minute electronic packages. Such miniaturization might lead us to the false belief that absolute size is irrelevant to the operation of complex machinery. But nature does not miniaturize neurons (or other cells for that matter). The range of cell size among organisms is incomparably smaller than the range in body size. Small animals simply have far fewer cells than large animals. The human brain contains several billion neurons; an ant is constrained by its small size to have many hundreds of times fewer neurons.
There is, to be sure, no established relationship between brain size and intelligence among humans (the tale of Anatole France with a brain of less than 1,000 cubic centimeters vs. Oliver Cromwell with well above 2,000 is often cited). But this observation cannot be extended to differences between species and certainly not to ranges of sizes separating ants and humans. An efficient computer needs billions of circuits and an ant simply cannot contain enough of them because the relative constancy of cell size requires that small brains contain few neurons. Thus, our large body size served as a prerequisite for self-conscious intelligence.
We can make a stronger argument and claim that humans have to be just about the size they are in order to function as they do. In an amusing and provocative article (American Scientist, 1968), F. W. Went explored the impossibility of human life, as we know it, at ant dimensions (assuming for the moment that we could circumvent—which we cannot—the problem of intelligence and small brain size). Since weight increases so much faster than surface area as an object gets larger, small animals have very high ratios of surface to volume: they live in a world dominated by surface forces that affect us scarcely at all (see previous essay).
An ant-sized man might don some clothing, but forces of surface adhesion would preclude its removal. The lower limit of drop size would make showering impossible; each drop would hit with the force of a large boulder. If our homunculus managed to get wet and tried to dry off with a towel, he would be stuck to it for life. He could pour no liquid, light no fire (since a stable flame must be several millimeters in length). He might pound gold leaf thin enough to construct a book for his size, but surface adhesion would prevent the turning of pages.
Our skills and behavior are finely attuned to our size. We could not be twice as tall as we are, for the kinetic energy of a fall would then be 16 to 32 times as great, and our sheer weight (increased eightfold) would be more than our legs could support. Human giants of eight to nine feet have either died young or been crippled early by failure of joints and bones. At half our size, we could not wield a club with sufficient force to hunt large animals (for kinetic energy would decrease 16 to 32-fold); we could not impart sufficient momentum to spears and arrows; we could not cut or split wood with primitive tools or mine minerals with picks and chisels. Since these all were essential activities in our historical development, we must conclude that the path of our evolution could only have been followed by a creature very close to our size. I do not argue that we inhabit the best of all possible worlds, only that our size has limited our activities and, to a great extent, shaped our evolution.
B | HUMAN BRAINS
An average human brain weighs about 1,300 grams (45.5 ounces); to accommodate such a large brain, we have bulbous, balloon-shaped heads unlike those of any other large mammal. Can we measure superiority by the size of our brains?
Elephants and whales have larger brains than ours. But this fact does not confer superior mental ability upon the largest mammals. Larger bodies need larger brains to coordinate their actions. We must find a way to remove the confusing influence of body size from our calculation. The computation of a simple ratio between brain weight and body weight will not work. Very small mammals generally have higher ratios than humans; that is, they have more brain per unit of body weight. Brain size does increase with body size, but it increases at a much slower rate.
If we plot brain weight against body weight for all species of adult mammals, we find that the brain increases at about two-thirds the rate of the body. Since surface areas also increase about two-thirds as fast as body weight, we conjecture that brain weight is not regulated by body weight, but primarily by the body surfaces that serve as end points for so many innervations. This means that large animals may have absolutely larger brains than humans (because their bodies are bigger), and that small animals often have relatively larger brains than humans (because body size decreases more rapidly than brain size).
The correct criterion for assessing the superiority in size of our brains. The solid line represents the average relationship between brain weight and body weight for all body weights among mammals in general. Superiority in size is measured by upward deviation from this curve (i.e., “more” brain than an average mammal of the same body weight). Open circles represent primates (all have larger brains than average mammals). C is the chimpanzee, G the gorilla, and A the fossil hominid Australopithecus: erectus covers the range of Homo erectus (Java and Peking Man); sapiens covers the field for modern humans. Our brains have the highest positive deviations of any mammal. (F. S. Szalay, Approaches to Primate Paleobiology, Contrib. Primal. Vol. 5, 1975, p. 267. Reproduced with the permission of S. Karger AG, Basel)
A plot of brain weight vs. body weight for adult mammals points the way out of our paradox. The correct criterion is neither absolute nor relative brain size—it is the difference between actual brain size and expected brain size at that body weight. To judge the size of our brain, we must compare it with the expected brain size for an average mammal of our body weight. On this criterion we are, as we had every right to expect, the brainiest mammal by far. No other species lies as far above the expected brain size for average mammals as we do.
This relationship between body weight and brain size provides important insights into the evolution of our brain. Our African ancestor (or at least close cousin), Australopithecus africanus, had an average adult cranial capacity of only 450 cubic centimeters. Gorillas often have larger brains, and many authorities have used this fa
ct to infer a distinctly prehuman mentality for Australopithecus. A recent textbook states: “The original bipedal ape-man of South Africa had a brain scarcely larger than that of other apes and presumably possessed behavioral capacities to match.” But A. africanus weighed only 50 to 90 pounds (female and male respectively—as estimated by Yale anthropologist David Pilbeam), while large male gorillas may weigh more than 600 pounds. We may safely state that Australopithecus had a much larger brain than other nonhuman primates, using the correct criterion of comparison with expected values for actual body weights.
The human brain is now about three times larger than that of Australopithecus. This increase has often been called the most rapid and most important event in the history of evolution. But our bodies have also increased greatly in size. Is this enlargement of the brain a simple consequence of bigger bodies or does it mark new levels of intelligence?
Evolutionary increase in human brain size (dotted line). The four triangles represent a rough evolutionary sequence: Australopithecus africanus, ER-1470 (Richard Leakey’s new find with a cranial capacity just slightly less than 800 cc), Homo erectus (Peking Man), and Homo sapiens. The slope is the highest ever calculated for an evolutionary sequence. The two solid lines represent more conventional scaling of brain size in australopithecines (above) and great apes (below). (“Size and Scaling in Human Evolution,” Pilbeam, David, and Gould, Stephen Jay, Science Vol. 186, pp. 892–901, Fig. 2, 6 December 1974. Copyright 1974 by the American Association for the Advancement of Science)
To answer this question, I have plotted cranial capacity against inferred body weight for the following fossil hominids (representing, perhaps, our lineage): Australopithecus africanus; Richard Leakey’s remarkable find with a cranial capacity of nearly 800 cubic centimeters and an antiquity of more than two million years (weight estimated by David Pilbeam from dimensions of the femur); Homo erectus from Choukoutien (Peking Man); and modern Homo sapiens. The graph indicates that our brain has increased much more rapidly than any prediction based on compensations for body size would allow.
My conclusion is not unconventional, and it does reinforce an ego that we would do well to deflate. Nonetheless, our brain has undergone a true increase in size not related to the demands of our larger body. We are, indeed, smarter than we were.
23 | History of the Vertebrate Brain
NATURE DISCLOSES the secrets of her past with the greatest reluctance. We paleontologists weave our tales from fossil fragments poorly preserved in incomplete sequences of sedimentary rocks. Most fossil mammals are known only from teeth—the hardiest substance in our bodies—and a few scattered bones. A famous paleontologist once remarked that mammalian history, as known from fossils, featured little more than the mating of teeth to produce slightly modified descendant teeth.
We rejoice at the rare preservation of soft parts—mammoths frozen in ice or insect wings preserved as carbonized films on beds of shale. Yet most of our information about the soft anatomy of fossils comes, not from these rare accidents, but from evidence commonly preserved in bone—the insertion scars of muscles or the holes through which nerves pass. Fortunately, the brain has also left its imprint upon the bones that enclose it. When a vertebrate dies, its brain quickly decays, but the resultant hole may be filled by sediment that hardens to produce a natural cast. This cast can preserve nothing of the brain’s internal structure, but its size and external surface may faithfully copy the original.
Unfortunately, we cannot simply use the volume of a fossil cast as a reliable measure of an animal’s intelligence; paleontology is never that easy. We must consider two problems.
First, what does brain size mean? Does it correlate at all with intelligence? There is no evidence for any relationship between intelligence and the normal range of variability for brain size within a species (fully functional human brains range from less than 1,000 to more than 2,000 cubic centimeters in volume). The variation among individuals within a species, however, is not the same phenomenon as variation in average values for different species. We must assume that, for example, average differences in brain size between humans and tuna fish bear some relationship to a meaningful concept of intelligence. Besides, what else can paleontologists do? We must work with what we have, and brain size is most of what we have.
Secondly, the primary determinant of brain size is not mental capacity, but body size. A large brain may reflect nothing more than the needs of the large body that housed it. Moreover, the relationship of brain size to body size is not a simple one (see previous essay). As animals get larger, brains increase in size at a slower rate. Small animals have relatively large brains; that is, the ratio of their brain weight to body weight is high. We must find some way to remove the influence of body size. This is done by plotting an equation for the “normal” relationship between brain weight and body weight.
Suppose we are studying mammals. We compile a list of average brain and body weights for adults of as many different species as we can. These species form the points of our graph; the equation fitting these points indicates that brain weight increases about two-thirds as fast as body weight. We can then compare the brain weight of any given species with the brain weight for an “average” mammal of that body weight. This comparison removes the influence of body size. A chimpanzee, for example, has an average brain weight of 395 grams. An average mammal of the same body weight should have a brain of 152 grams according to our equation. A chimp’s brain is, therefore, 2.6 times as heavy as it “should” be (395/152). We may refer to this ratio of actual to expected brain size as an “encephalization quotient”; values greater than 1 signify larger than average brains; values less than 1 mark brains that are smaller than average.
But this method imposes another difficulty on paleontologists. We must now estimate body weight as well as brain weight. Complete skeletons are very rare and estimates are often made from a few major bones alone. To pile difficulty upon difficulty, only birds and mammals have brains that completely fill their cranial cavities. In these groups, a cranial cast faithfully reproduces the size and form of the brain. But in fishes, amphibians, and reptiles, the brain occupies only part of the cavity, and the fossilized cast is larger than the actual brain. We must estimate what part of the cast the brain would have occupied in life. And yet, despite this plethora of difficulties, assumptions, and estimates, we have been able to establish, and even to verify, a coherent and intriguing story about the evolution of brain size in vertebrates.
California psychologist Harry J. Jerison has recently marshaled all the evidence—much of it collected during his own labors of more than a decade—in a book entitled The Evolution of the Brain and Intelligence (New York, Academic Press, 1973).
Jerison’s major theme is an attack upon the common idea that vertebrate classes can be arranged in a ladder of perfection leading from fish to mammal through the intermediary levels of amphibian, reptile, and bird. Jerison prefers a functional view that relates the amount of brain to specific requirements of modes of life, not to any preordained or intrinsic tendency for increase during the course of evolution. The potential “brain-body space” of modern vertebrates is filled in only two areas: one occupied by the warm-blooded vertebrates (birds and mammals), the other by their cold-blooded relatives (fish, amphibians, and modern reptiles). (Sharks provide the only exception to this general rule. Their brains are much too big—quite a surprise for these supposedly “primitive” fishes, but more on this later.) Warm-blooded vertebrates, to be sure, have larger brains than their cold-blooded relatives of the same body size, but there is no steady progress toward higher states, only a correlation between brain size and basic physiology. In fact, Jerison believes that mammals evolved their large brains to meet specific functional demands during their original existence as small creatures competing on the periphery of a world dominated by dinosaurs. He argues that the first mammals were nocturnal and that they needed larger brains to translate the perceptions of hearing and smell into spatial p
atterns that animals active in daylight could detect by vision alone.
Jerison presents a variety of intriguing tidbits within this framework. I hate to confute a comfortable item of received dogma, but I must report that dinosaurs did not have small brains—they had brains of just the right size for reptiles of their immense dimensions. We should never have expected more from Brontosaurus because large animals have relatively small brains, and reptiles, at any body weight, have smaller brains than mammals.
The gap between modern cold- and warm-blooded vertebrates is neatly filled by intermediate fossil forms. Archaeopteryx, the first bird, is known from fewer than half a dozen specimens, but one of them has a well-preserved brain cast. This intermediate form with feathers and reptilian teeth had a brain that plots right in the middle of the unfilled area between modern reptiles and birds. The primitive mammals that evolved so rapidly after dinosaurs became extinct had brains intermediate in size between reptiles and modern mammals of corresponding body weights.
We can even begin to understand the mechanism of this evolutionary increase in brain size by tracing one of the feedback loops that inspired it. Jerison computed the encephalization quotients for carnivores and their probable prey among ungulate herbivores for four separate groups: “archaic” mammals of the early Tertiary (the Tertiary is the conventional “age of mammals” and represents the last 70 million years of earth history); advanced mammals of the early Tertiary; middle to late Tertiary mammals; and modern mammals. Remember that an encephalization quotient of 1.0 denotes the expected brain size of an average modern mammal.