The Price of Altruism (28 page)

captures this relationship by explaining how the number of copies made (
w
) of the different heights (
z
) determine the average height of the new group (
).

This was a general-selection equation: It would hold true for everything from a child choosing radio channels to the earth preserving fossils to the culling of chemical crystals in far-off galaxies. But it could also be applied to biological traits like baldness and strength and crooked teeth, and, most profitably, to the evolution of social behaviors like altruism. All one needed to do was have
z
stand for the trait and
w
for its fitness (“copies” of traits simply mean their fitness), and, like a rabbit pulled out of a conjurer’s hat, the covariance equation told you how it would evolve from one generation to the next. George hadn’t done all the work yet, but already saw that his was a more abstract approach than using coefficients of kinship and could be made to work just the same: The spread of altruism could be tracked via statistical covariance of the character with fitness rather than calculations of the pathways of relatedness. Hamilton’s
r
B > C notwithstanding, altruism depended on association, not family.
²⁸

The reason was immediately clear to George: Natural selection is indifferent to why individuals end up together in groups; whether it’s due to common descent, or similarity in traits, or any other pretext doesn’t matter. On the other hand, since covariance could be made to treat relatedness as a statistical association rather than a measure of common ancestry, relatedness could actually be negative. What this meant mathematically was that while under conditions of a particular association altruism could evolve, under the conditions of another association
spite
could evolve: Everything depended on the environment. Spitefulness wasn’t just the selfish harming of others to help oneself; it was doing harm to oneself in order to harm one’s enemy even more. Explaining its evolution was therefore a similar problem to explaining the evolution of altruism: Both behaviors reduced fitness but existed nonetheless. It was a possibility Hamilton had entirely overlooked.

George pondered the larger meanings. Was there really nothing special about altruism? In evolutionary terms only a thin blue line seemed to separate it from spite. What determined whether a living being should act kindly or with malice had nothing to do with an “essence” or “inner core”—both, after all, resided within us. Instead, if the surrounding creatures were similar altruism could evolve; if they were different, spite was the solution. Pure unadulterated goodness was a fiction.

It was a shocking thought. And yet strangely, on further reflection, it was also laden with hope. For the adaptive success of altruism depended on the social environment, on society. Short of some unsullied divine morality, goodness could flourish if it was recognized as important. Institutionalize cooperation and you kill competition; valorize self-interest and you penalize altruism. Virtue was already within us but needed to be helped along. Perhaps Skinner held a piece of the truth after all: Create the right conditions and goodness would see the light.

There was still no word from
Nature
about the antlers paper, but George was optimistic. Most important now was this new selection math. Finally he’d been sticking to one problem and not jumping around as he’d always done. “I think this work I’m doing,” he wrote to Alice, emphasizing the adjective, “is really going to lead to something
important
.”
²⁹

Then, on September 24, he relayed the news:

“He liked it so much that he took me to meet the department chair,” he carried on in an excited letter to Annamarie. “90 minutes later,” he continued to a friend, “I walked out with a room assigned to me, with keys, plus request for curriculum vitae so that they could make it official about giving me an honorary appointment.”
³¹

A complete unknown walking off the street into the chair’s office and being given keys to a room of his own in arguably the world’s greatest department of human genetics in a matter of minutes? It was the miracle George Price had been waiting for.

Tucked away on little Stephenson Way east of Euston Station, the Galton Laboratory at UCL was a storied home. Great names had been attached to it—Karl Pearson, R. A. Fisher, J. B. S. Haldane, Lionel Penrose—and now Harry Harris at the helm. “Like the chambers and corridors of some vast battleship,” an observer remarked, “its rooms often seem to be below surface even if they are not. Its parquet and paneling are easily overlooked: its underlying grandeur is subordinated to the practical demands of intellectual inquiry.”
³²

Cedric Austin Bardell Smith, Hamilton’s old boss, was the Weldon Professor of Biometry into whose office George had walked. A Leicester-born Quaker five years George’s senior, CABS was known for his gentle heart and mathematician’s quirky sense of humor. What did Jesus mean when he said, “Heaven equals ax
²
+ bx + c”? was an example (answer: It’s a parabola); another was the invention of a new system of arithmetic based on counting from one to five and replacing all numbers greater than that with the same number subtracted by ten and printed upside down. As a student at Cambridge he and three equally off beat friends formed the Trinity College Mathematical Society, publishing solutions to arcane problems under the pseudonym Blanche Descartes, a mythical Frenchwoman still referred to in the mathematical literature.
³³

One example was the problem of whether it is possible to cut a square into smaller squares each of which is different. The solution gave birth to the “square squared,” a notion that proved highly useful in the design of electrical networks. Another example was the counterfeit coin problem: If you have twelve pennies, one of which is counterfeit and differs in that it is slightly heavier, and are given a pair of balance scales—what is the smallest number of weighings that will pick out the false coin? Cedric’s solution ended up pioneering the field of search theory, a branch of mathematics commonly used in computing, economics, and in locating airplane crashes and lost mountaineers. (The answer to the problem is 3.) Most of all, though, Cedric Smith had been a protégé of J. B. S. Haldane, inventing powerful methods to map genes on chromosomes. In fact he had succeeded him. He was one of the world’s leading biostatisticians.
³⁴

Meanwhile the skeletons of George’s past continued to haunt him. He had left his wife, abandoned his daughters, been a lousy son to his aging mother. His behavior was self-destructive, people said, and deep inside him he knew it was true. He was still “daydreaming about torturing Ferguson.” Back in America, Julia had come into her inheritance and could take care of the kids now. He was essential to no one; if he died not a soul in the world would be the worse for it. The UCL appointment was flattering but wouldn’t pay the bills. Unless something extraordinary happened he planned to kill himself, he wrote to Annamarie, “since it isn’t worth the bother of working just to stay alive.”
³⁵

And yet…

Family, strangers, altruism, spite—the ideas swam in his head like drunken piranhas. To tame them he’d need to turn to science. In a way, he knew, they were his lifelines. Under the positively impressed, somewhat flabbergasted gaze of CABS he hunkered down once again and went to work.

His goal was clear: to fathom the mystery of family. Mate choice, fatherhood, individual interest versus common good—these were the issues he would tackle. It was to be a clean affair, and perfectly rational, nothing like the mess he had made of his own life. Developing mathematical tools for making evolutionary inferences would be the only way “to protect against biasing effects of emotional prejudice.” Besides, quoting Haldane, with whose work he now began to become familiar, “an ounce of algebra is worth a ton of verbal argument.”
³⁶

In a direct translation of the optimization work he had done for IBM on the “register problem,” he set out to model human behavior. An “optimal” behavioral strategy was one that would maximize the frequency of an individual’s genes in the next generations. Just as Ardrey and Morris had done, the method would be to consider a problem facing tribes of twenty to fifty hominids in the Middle and Upper Pleistocene, imagine a number of alternative behavioral strategies that might serve as solutions to the problem, and then to compare them with present behavior. Under the assumption that our ancestors were very likely to have developed genetically optimal behavior and to have maintained it for a long time—after all,
Homo sapiens
was a highly successful species—the strategy most similar to observed behavior today would likely have been the one that had evolved.

He started with basics. How, for example, did our ancestors allocate food? One optimal solution could have been complete sharing and cooperation: promiscuous, noncompetitive mating, cooperative rearing of the young with little or no recognition of individual motherhood and fatherhood, and retaliation against anyone out for himself. Just as with the antler model, an individual would increase his fitness by cooperating with others and thereby avoiding punishment, and by helping to punish others deficient in cooperation and thereby causing them to cooperate. It would have been a veritable Stone Age Plato’s Republic.

But there was an alternative. What if the tribe chose cooperation in hunting by adult males, but individual and family action in all other areas? Hunting spoils would first be divided among the males, who would then distribute their share to women and children as a matter of personal choice. Such a system, George quickly saw, would favor monogamy, or at least something close to it. The reason was that if a man tended to keep the meat to himself when food was scarce so that his “wife” and children suffered severely while he ate comfortably, he would, on average, leave fewer descendants. Genes correlated with such behavior, therefore, would soon diminish in the tribe, and the behavior become less common. On the other hand, if a man was a wonderful provider and yet tended to swap wives every few years, most of the time he’d be providing food for the children of other men while neglecting his own older off spring; therefore, since gene frequency is a ratio rather than an absolute amount, the better he was at providing, the more effective he’d be in decreasing the frequency of his own genes in the group.
³⁷

True, there existed sexually promiscuous societies that deviated substantially from monogamy. But considering the enormous changes in living conditions that had occurred over the last twenty thousand years, it was remarkable how much the vast majority of humanity seemed to behave in rough accordance with the family model. Hypothesis 2 was a better bet than hypothesis 1.

Family, then, had developed under selective pressures related to food distribution at a time in human evolution when hunting by all-male bands became important—perhaps when hominids came down from the trees and began walking the savannas. Amazingly, exalted “fatherhood” might have been an optimal solution to the mundane challenge of securing daily grub. Even the heights of love were just an invention to oil optimality. After all, George wrote to an instrument-maker friend in the States from the days when he was thinking of building Skinner his Teaching Machine, love couldn’t be an automatic consequence of “reinforcement” since it doesn’t necessarily bring happiness. Something more powerful, like genetic evolution, had to be responsible.
³⁸