The Price of Altruism (24 page)

Read The Price of Altruism Online

Authors: Oren Harman

BOOK: The Price of Altruism
4.89Mb size Format: txt, pdf, ePub

It was a mortal blow to the doctrine of the “greater good.” Practically overnight,
Adaptation and Natural Selection
became a classic and group selection a theory to stay away from like the plague. Finally “Wynne” had been felled, though, like Sisyphus, he would keep on fighting his battle. Williams ended with a tip of the hat to Fisher: Beyond a doubt, fulmars and red grouse notwithstanding, natural selection produces adaptations optimizing individual fitness. “I am convinced,” he wrote triumphantly, “that it is the light and the way.”
50

Back from Brazil, Hamilton had news of his own: In countless species of ants, thrips, wasps, beetles, and mites there are
extraordinary
sex ratios, major departures from Fisher’s pristine one-to-one. Hamilton, unlike Fisher (and even Williams), would know: He had found them peeling back bark from capirona and kapok trees, stripping weeds of larvae beneath shallow jungle streams, and munching on figs with internal surprises.
Mellitobia acasta
was an example: A tiny parasitic wasp, its female lays her eggs inside the living pupae of bumblebees. When the eggs hatch they eat their way out of the pupa, but not before each female engages in sex with the sole male, their only brother. After all, it makes sense for their mother to use the confined body of the pupa to lay as many female eggs as possible and only one male to inseminate them: Once they’ve been fertilized, they can fly away to lay their own eggs in another poor bumblebee pupa, while their exhausted brother and lover, the wingless male, wastes away in the abandoned cocoon.
51

By now Hamilton had a research position at Imperial College Field Station near Ascot, an institution eminent in the study of insect ecology. Around its rotten-wood grounds in Silwood Park, in the asphalt cracks of his cramped Berkshire home, and on walks in the royal forests around Windsor he discovered the Western European cousins of his Amazonian friends. Parasitism, he was learning, could get complicated—and fast: Sometimes two females laid their eggs in the same caterpillar or pupa, sometimes progeny dispersed and sometimes they didn’t; sometimes ruthless males, like “maniacal homicides armed with knives,” fought over amorous sisters to the death. Each situation changed the rules of fitness entirely. How would a miniature
acasta
know what precise proportion of male-to-female eggs to lay?

To figure out himself what the answer should be, Hamilton journeyed into London many days to use the UCL computer center at Gordon Square. It wasn’t a pretty affair: Programming in the dinosaur FORTRAN language, which still preserved its nineteenth-century Hollerith punched-card six-space markings—“those once-useful claws that are now diminished to useless toe nails,”—Hamilton would queue along the old kitchen staff’s route to the basement, roll of five-punched tape ready in hand, watch as the technician set the toggles to guide the “magic monster” Mercury computer, walk across to collect the spaghetti-like tape spewing chatteringly out the far side, and rush to an adjacent room to view the results via a teletype. It was a far cry from sitting in front of a laptop.
52

But despite the hassle and beneath the natural complications, the teletypes disclosed a pristine river of logic. Whether you were an ant in a weevil egg, a wasp in a wild fig, or a mite inside your own mother, there always existed an optimal solution to the problem of the precise proportion of female to male eggs to lay. Hamilton called it the “unbeatable strategy,” and discovered to his great astonishment that, with no Mercury and toggles and teletype and FORTRAN, the little critters always found it on their own.

It was mind-blowing. Mites “calculating” fractions as precise as 3/14ths? Hamilton was godless, but this was as close to faith as anything. Now, he knew, he’d discovered a powerful tool that would help to fathom nature’s wonders. Back when he was a student at Cambridge he had idly read von Neumann’s great book, never imagining that it might have anything to do with biology. But there was no mistaking that here was an evolutionary analog to the prisoner’s dilemma: Fitness was a “payoff,” the opposing sex-determining genes the “players.” In parasitic wasps and sex ratios he was dealing in the theory of games.
53

No one seemed to notice at the time that extraordinary sex ratios provided the very test that Williams had sought to apply to group selection.
54
Hamilton himself contributed to the confusion: Extraordinary sex ratios couldn’t be an adaptation for the good of the population, he wrote, since even a tiny bit of outbreeding would destroy any “altruistic” genes.
55
No, once again this was a family affair; small spaces like pupae, after all, were like isolated cottages. In line with the times, parasitic wasps were interpreted as just another nail in the coffin of the “greater good”: It was gene frequencies that were being maximized, after all, and they could only do it by manipulating the behavior of related individuals.

Laying down his pencil, Hamilton might have smiled. First
r
B > C and now sex ratios: The mathematics of the “gene’s-eye-view” was turning biology into a “hard” predictive science. Fisher had been right: Selection
was
a mechanism for generating an exceedingly high degree of improbability; even altruistic behavior was fashioned by its unsentimental, ruthless cull. He ended his equation-filled article, in the grand natural-history tradition of humble cooperation, by thanking a Dr. Bevan “for information about bark beetles” and a Dr. Lewis “for information about thrips.”

 

 

Group or Individual, Optimization or Chance: Where had true goodness come from? Wright and Haldane and Fisher had begun the work, uniting Darwin and Mendel, selection and genes, in a bold evolutionary synthesis. They had argued among themselves over the relative role of mutation, drift and selection, but their project had been one of foundation. Now, on the back of Wynne-Edwards and the “greater good,” younger men like John Maynard Smith, George Williams, and Bill Hamilton were penetrating further into Nature’s puzzles. Theirs had been a thought experiment as audacious as Einstein’s, only from the point of view of the gene rather than a man traveling on a beam of light. Finally, they thought, they were solving the mysteries of behavior.

In the Amazonian jungle in 1964, Hamilton received a letter from a girlfriend in London. She was breaking up with him. That week in the forest he had met two Brazilian kids, Romilda and Godofredo. Alone and morose, he was having serious doubts whether he’d ever marry or have children of his own. Turning to their parents, he offered to educate and take care of them as a foster parent back in England.

Now, four years later, his glum misgivings had proved premature. He was married to a wonderful woman, Christine, and thinking of starting a family. But a promise was a promise. The year after publishing “Extraordinary Sex Ratios” in
Science
he and Christine were back again in the forest and would be taking Romilda and Godofredo to Berkshire. Evolution hadn’t selected him to behave this way toward nonkin, he was now more than ever certain. But lying in his hammock listening to the beat of the wings of an iridescent dragonfly and a seriema pealing its dawn call like cracked bells from the hills, he knew it was in his lonely heart, still.
56

 

A letter from Alice Price to her son George, November 6, 1963

 
No Easy Way
 

I
n the spring of 1957 Senator Hubert Humphrey had introduced a bill offering income tax credit for tuition paid to institutions of higher learning. George wrote a letter of thanks. It was all about education, he thought, and the bill was a step in the right direction. Still, really, intervention was needed much earlier. Why not exempt the first six thousand dollars received for teaching in grade school? After all, the greatest weapon against the Soviets was undoubtedly the first-grade teacher. “I belong to the ‘as the twig is bent, the tree grows’ school,” George explained.
1

He had struck up a correspondence with his old senator from Minnesota, and there were further suggestions to help win the Cold War: Why not send every Russian two pairs of shoes if their government would free Hungary? Or ten million dollars worth of polio vaccine if the Hungarian police would stop torturing prisoners? Or what about radio messages broadcast in Russian extolling the advantages of butter over guns? And could the senator please send a copy of his “U.S. Foreign Policy and Disarmament” speech to Congress in April? And the hearings of the Johnson Preparedness Subcommittee? And had he seen the piece by Donald Harrington yesterday in the
Times
? “Of course! Why didn’t I think of that myself?”: an international court of law to help restrain the Chinese in Formosa. Could the senator perhaps “plant the idea tactfully within Mr. Dulles’s mind?” George would very much appreciate it.
2

The senator replied politely that he would welcome his advice and counsel. George’s views were “eminently sane,” the idea about creating a strong UN police force just along the lines he had imagined. Price’s overanxious manner might have struck the senator as somewhat odd, but his replies made George exceedingly proud. Nothing felt better than a man of stature recognizing his potential.

Increasingly, writing popular science articles became the perfect vehicle for gaining notoriety. It was the journalist’s prerogative, after all, to get in touch with famous people. And so, between “The Physics of Bowling,” “U.S. Begins Search for Beings in Other Worlds,” and “What We’re Learning from Animals” (all for
Popular Science Monthly
), he drove to his old hero Claude Shannon’s home to write a profile on him for IBM’s magazine,
THINK
—the two “paddling around” his lake in “glorious” silence. Between “Achievements of American Science,” “The Real Threat from Red China” (government brainwashing), and “How to Hatch an Egghead” (motivate him), he contacted the Nobel laureate Hermann J. Muller, to ask his views about evolution on other planets. On each occasion the excuse developed into a conversation, with George trying his very best to impress. Muller, who was considered by many to be the greatest living geneticist, remembered “Science and the Supernatural” and complimented him on it; George replied with quotes from Poe and Keats, and his thoughts about people reading too much science fiction. Muller thanked him kindly, and the correspondence ended at that. It was a pattern: the brushup, the exchange, the titillation.
3

If IBM had been stupid enough not to hire him, George was not about to give up and roll over. On the strength of the piece in
Life
he had been offered a contract by Harper & Brothers, and was hard at work on a book now that he hoped would help to save the world.

It was right about then that Skinner came along.

 

 

Adored as a messiah and abhorred as a menace, Harvard professor Burrhus Frederic Skinner was the most influential psychologist in America. He had an elongated face and an unsettling smile. The leader of the “behaviorists,” who likened man to machine, he was reviled by Freudians and humanists. Some called him “totalitarian,” others “Orwellian” others thought his ideas man’s only hope.
4

George was intrigued. Early Christian thinkers thought it was the “soul” that set man apart from animal: God-given, immaterial, impalpable, otherworldly. But what if man were just an animal, firmly planted in the natural world? And what if men and animals were not all that different from machines?

Such thoughts had been afoot since the Diderots and La Metries of the eighteenth century, and even earlier, with Hobbes. Now, after Darwin and the rise of psychology and genetics, Skinner was arguing that freedom and free will were no more than comfortable illusions. For him autonomy was a “feel-good” invention, morality a sinister sham. The belief in an “inner man” was like the belief in God, a superstition, nothing but a symptom of humanity’s failure to understand a complex world. Whether he liked it or not, man was already controlled by external influences, some haphazard, others evil, others merely in step with “convention.” Where things went bad was when man made a fetish of individual freedom, seeking to give life to the internal “soul” at the expense of orderly society. In reality, Skinner preached, it was environments and not people, actions and not feelings, that needed to be changed. The “behavioral technology” called “operant conditioning” was civilization’s hope for deliverance. Where moral arguments had failed, it could create a world where man would refrain from polluting, from overpopulating, from rioting, from hating, even from waging war. George was a “as the twig bends” man. Visiting Skinner’s Harvard lab on the pretext of writing a story, he soon befriended “Fred.”
5

The lab was a scene out of wild science fiction. Pigeons playing Ping-Pong, rats balancing balls on their noses; there was even the pigeon-guided missile, the “Pelican.” But these were animals. What about humans: Could they be similarly programmed? And was it punishment or reward that would be more effective in controlling behavior?

Skinner was convinced that it was reward, and had designed the “Teaching Machine” to prove it. A question was posed on a screen: If the child answered correctly he’d be immediately rewarded, not with a grain of corn but with a printed statement of approval—just as satisfying and effective. Walking George through the lab, Skinner claimed that kids could be taught arithmetic just as rats were taught circus acts and pigeons Ping-Pong.
6

George was captivated. “Very quietly,” he wrote for
THINK
later that month, “almost unnoticed amid the fanfare over thermonuclear weapons, earth satellites and moon probes, an important new invention has made its appearance.” To Skinner he offered: “No doubt the most immediate application for such machines would be to teach reading and writing to the illiterate masses of Asia and Africa.” For the last hour he had been sketching out a way to make “a very cheap machine using an acoustic phonograph with spring motor, plus a paper disc, with controls to automatically position the phonograph pick-up at the beginning of the question, stop the phonograph at the end of questions, do the same for the answers, repeat questions and answers when desired, and skip questions previously answered correctly.” It could be mass-produced at just five dollars a pop.
7

However useful Skinner’s machines, they got George wondering about larger questions. Was man in control of his destiny? Was freedom really just a fantasy, or worse—a trick? Skinner argued that man believes in free will only so that he can take credit for his “good” behavior. But George had been free to leave his family, and was taking no credit for good behavior. Clearly there were traits buried deep in his nature, things that had not necessarily been learned. The problem was how to manage them, especially self-centeredness and egoism, the most entrenched natural trait of all.

He translated personal deficiencies into public affairs, the easier, perhaps, to flee them. Writing drafts of his book for Harper, George remembered Thucydides’ famous description of Athens falling to Sparta on account of selfishness. De Tocqueville too, he recalled, had lamented how through vain self-absorption and greed men “lose sight of the close connection that exists between the private fortune of each of them and the prosperity of all.”
8
He might be a bad family man, George thought, but he cared about his nation. True self-interest lay in strengthening the community, in devotion to the country, in paying a personal price for the good of all. Russia had done it. China had done it. Could America do it too? Could limits be put in a democracy on the individual pursuit of happiness? Skinner was hopeful, and George tended to agree. Americans just needed to be taught that personal sacrifice meant communal reward.

 

 

Meanwhile, Harper had reneged on its contract, finding some of his suggestions “brilliant” and “unexpected,” but the overall book unsalable. With a change of title, and the help of a talented literary agent, George secured a new publisher and advance. It was April 1958, and Doubleday would want
No Easy Way
no later than July.
9

He was living now in the West Village in New York City. The small loft at 88 Bedford Street was just down the way from the historic town home of Edna St. Vincent Millay, whose poetry Thomas Hardy called America’s second attraction, after skyscrapers. Jack Kerouac, Allen Ginsberg, and William S. Burroughs were regulars in the coffeehouses. Dylan Thomas had collapsed and died a few years earlier drinking at the White Horse Tavern, just a few blocks away, and in a few years a young Bob Dylan would show up, playing “neo-ethnic” songs in the coffeehouses in the style of Woody Guthrie. There were writers and poets and artists and “beats,” and, dapper in his bow tie and crew cut, George was an anomaly. But he liked the idea that he was an “ex-chemist.” In a strange way, bohemia appealed to him. There was a kind of prurient satisfaction in being a straight arrow in a world of chaos. Proudly he had “writer” stamped in his passport.
10

Still, he could not escape his past. Julia was after him for alimony, and his money was disappearing fast. He had taken another technical-editing job with IBM, this time in Poughkeepsie, and was only getting home on weekends. The Doubleday deadline was killing him. Besides, the world was changing faster than he could write about it: He had started off calling for armament, but now Russia’s position on
disarmament
seemed much more honorable. Should America compete? Should it withdraw? Could the Russians be trusted? He was confused. He hardly had time to eat. The world around him was spinning.
11

Loosening his bow tie he popped them in, one by one: iproniazid, Dexedrine, ephedrine, Seconal; uppers, barbiturates, psychostimulants. Everyone else around him seemed to be taking them, so what the hell? The effects, he soon learned, were less than exhilarating. “I’ve spent most of the day so far lying down,” he wrote to his psychiatrist, Dr. Nathan Kline. The drugs had taken away his panic but had left him tranquil and sedate. He wasn’t getting anything done. He was loafing. He felt neither pain nor joy. Worryingly, he was taking “some measures” to try to release adrenaline from his adrenals; he just couldn’t quite remember whether it gets through the blood-brain barrier. Did Kline?
12

Then, as the winter of 1960 rolled in, a tumor was discovered in his throat. At first his internist thought, Hodgkins. But an old Manhattan Project friend who was chief of X-ray at Memorial Sloan-Kettering sent him to a cancer expert who diagnosed a nonmalignant thyroid tumor. Recovering from the operation in March, George was handed a new type of medicine: He had a thyroid imbalance, and would need to take tablets from now on if he wanted to stay alive and healthy.
13

 

 

There was an instruction manual for a GE electricity and magnetism kit to finish, and another one on gyros and accelerometers for Sperry-Marine. There was a new love interest, Joan, a teacher at Bennington College for Women in Vermont. There was Alice, still with her Japanese roomers, but growing old and infirm. There was his estranged lighting-expert-inventor brother, Edison, who had taken over Display, renaming it Edison Price Lighting Company Incorporated and operating out of 409 East Sixtieth Street. There was Julia, seeking contested alimony in domestic relations court. There were his daughters, whom he hadn’t seen for quite some time. There were desperate letters to friends asking for loans. There was his health. There were the drugs. And then, on April 10, 1960, there was a letter from Fred Schneider of the Advanced Systems Development Division at IBM. Regarding his old article in
Fortune
, IBM was beginning to invest in computer-aided design (CAD). Would George be interested in joining the project?
14

That fall a surgeon friend from his days in Minnesota wrote a kind note to say that his wife had a bad thyroid, too. Don Ferguson was now at the University of Chicago, and really, George, the thyroid was no big deal. He’d be fine. Still, didn’t he think that he was wasting himself on those damn manuals? After all, he had the kind of mind that needed to be applied to real scientific problems.
15

In fact George had been working on two scientific papers, with the hope of once again gaining a university position. The first was about the fallacies of random neural networks as they pertained to the organization of the brain, the second a theory of the function of the hymen.
16
Of course he had absolutely no training in either of these matters. But he had gone through the literature thoroughly and was planning to send the papers to
Science
all the same. He needed one big breakthrough—one truth, he thought, just one. IBM had toyed with him before, and by now he had lost interest in his Design Machine. Still, until the papers were accepted and made him a name, he’d need to take a job to settle his finances.

Other books

The Protectors by Dowell, Trey
The Night I Got Lucky by Laura Caldwell
Cowboy Angels by Paul McAuley
Cold Comfort by Quentin Bates
Beastkeeper by Cat Hellisen
Unto All Men by Caldwell, Taylor
Trail of Lies by Margaret Daley