by James Gleick
… We have to fight hard—every inch of the way—we can’t slip ever—a slip costs too much… . I’ll be all a women would be to you—I’ll always be your sweetheart & first love—besides a devoted wife—we’ll be proud parents too—we’ll fight to make Donald real—I want him to be like you… . I am proud of you always Richard—your a good husband, and lover, & well, coach, I’ll show you what I mean Sunday.
Your Putzie
False Hopes
Her health continued to fail. “Drink some milk!” Richard wrote in May. Her weight had fallen to eighty-four pounds. She looked like a woman starving.
You are a nice girl. Every time I think about you, I feel good. It must be love. It sounds like a definition of love. It is love. I love you.
I’ll see you in two days.
R. P. F.
More and more they talked of medical tests. They needed optimism. He was near despair. Time passes fast. Maybe we should start looking for another doctor… . Why don’t you drink an extra bottle of milk right now while you are thinking of it.
The scientific knowledge that empowered the physicists seemed to mean nothing on the soft soil of medicine. With the final desperation of the dying, Richard and Arline reached out for slender possibilities. He had heard about a new drug, sulf-something—he was not sure—and had written to researchers in the East, who told him apologetically that studies of sulfabenamide were in the most preliminary stage. The discovery that substances of the sulfonamide family retarded bacterial growth was not yet a decade old. They were destined to prove poor substitutes for true antibiotics.
Now Richard was writing to faraway doctors again. It seemed that Arline was pregnant. After ending the celibacy of their marriage, she had immediately missed her menstrual period. Was it possible? They were frightened and jubilant at once. Richard did not tell his parents, but he told his sister, now a college student. Joan was dazzled at the prospect of becoming an aunt. They talked about names and began making new plans. Yet to Richard it still seemed that Arline was wasting away. He thought he saw symptoms of starvation. Perhaps no rational observer could have construed the cessation of menses at this stage of the disease as a sign of pregnancy, but that was how they construed it. The alternative was so grim. Their doctors saw little reason for hopefulness. The chief physician from the sanatorium in Browns Mills, New Jersey, advised urgently that any pregnancy must be “interrupted”—“have it done by a specialist.” Then a pregnancy test gave a negative result after all. They did not know what to think. A doctor at Los Alamos told Richard that the tests were notoriously unreliable but that they could try again at an Albuquerque laboratory. He thought the laboratory had the necessary rabbits for the Friedman test.
The same doctor said he had heard of a new substance made from mold growths—“streptomicin”?—that seemed to cure tuberculosis in guinea pigs. If it worked, the doctor thought it might soon become widely available. Arline refused to believe the negative pregnancy result. She wrote cryptic remarks about “P.S. 59-to-be.” The same day a nurse wrote Feynman from the sanatorium to say that Arline had been spitting blood. He opened his encyclopedia yet again. Nothing. He drifted through the pages: tuberculosis, tuff, tularemia … Tuff was a kind of volcanic rock; Tunicata an animal group. He wrote Arline another letter. “Tumors you know about & Turkey, the country, also.” Some days she was now too weak even to write back. He grasped his uncertainty. Not knowing was frustration, anguish, and finally his only solace.
“Keep hanging on,” he wrote. “Nothing is certain. We lead a charmed life.”
In the midst of their private turmoil came V-E day and then Richard’s twenty-seventh birthday. Arline had prepared another mail-order surprise: the laboratory was flooded with newspapers—handed about and tacked to walls—proclaiming with banner headlines, “Entire Nation Celebrates Birth of R. P. Feynman!” The war in Europe, having provided so many of the scientists with their moral purpose, had now ended. The bloody circle was closing in the Pacific. They needed no threat of a German or Japanese bomb to urge them onward. Uranium was arriving. There would be one test—one last experiment.
At the Mayo Clinic in Minnesota another kind of experiment was under way, the first clinical trial of streptomycin, a substance that had been discovered nearly two years before, in August 1943. The population participating in the trial: two patients. Both had been near death from tuberculosis when the experiment began in the fall of 1944; both were improving rapidly. Even so, it was not until the next August that the Mayo trial had expanded to as many as thirty patients. The doctors could see lesions healing and lungs clearing. A year after that, the study of streptomycin as an antitubercular agent had become the most extensive research project ever devoted to a drug and a disease. Researchers were treating more than one thousand patients. In 1947 streptomycin was released to the public.
Streptomycin’s discovery, like penicillin’s a few years earlier, had been delayed by medicine’s slow embrace of the scientific method. Physicians had just begun to comprehend the power of controlled experiments repeated thousands of times. The use of statistics to uncover any but the grossest phenomena remained alien. The doctor who first isolated the culture he named Streptomyces griseus, by cultivating some organisms swabbed from the throat of a chicken, had seen the same microbes in a soil sample in 1915 and had recognized even then that they had a tendency to kill disease-causing bacteria. A generation had to pass before medicine systematized its study of such microbes, by screening them, culturing them, and measuring their antibiotic strengths in carefully labeled rows of test tubes.
Nuclear Fear
In its infancy, too, was the branch of science that would have to devote itself to the safety, short-term and long-term, of humans in the presence of nuclear radiation. The sense of miasmic dread that would become part of the cultural response to radioactivity lay in the future. The Manhattan Project’s researchers handled their heavy new substances with a breeziness that bordered on the cavalier. Workers handling plutonium were supposed to wear coveralls, gloves, and a respirator. Even so, some were overexposed. The prototype reactors leaked radioactive material. Scientists occasionally ignored or misread their radiation badges. Critical-mass experiments always flirted with danger, and by later standards the safety precautions were flimsy. Experimentalists assembled perfect shining cubes of uranium into near-critical masses by hand. One man, Harry Daghlian, working alone at night, let slip one cube too many, frantically grabbed at the mound to halt the chain reaction, saw the shimmering blue aura of ionization in the air, and died two weeks later of radiation poisoning. Later Louis Slotin used a screwdriver to prop up a radioactive block and lost his life when the screwdriver slipped. Like so many of these worldly scientists he had performed a faulty kind of risk assessment, unconsciously mis-multiplying a low probability of accident (one in a hundred? one in twenty?) by a high cost (nearly infinite).
To make measurements of a fast reaction, the experimenters designed a test nicknamed the dragon experiment after a coolly ominous comment of Feynman’s that they would be “tickling the tail of a sleeping dragon.” It required someone to drop a slug of uranium hydride through a closely machined ring of the same substance. Gravity would be the agent in achieving supercriticality, and gravity, it was hoped, would carry the slug on through to a safe ending. Feynman himself proposed a safer experiment that would have used an absorber made of boron to turn a supercritical material into a subcritical one. By measuring how rapidly the neutron multiplication died out, it would have been possible to calculate the multiplication rate that would have existed without boron. The arithmetical inference would have served as a shield. It was dubbed the Feynman experiment, and it was not carried out. Time was too short.
Los Alamos hardly posed the most serious new safety challenges, for all its subsequent visibility. These belonged to the vast new factory cities—Oak Ridge, Tennessee, and Hanford, Washington—where plants thrown up across thousands of acres now manufactured uranium and plutonium in b
ulk. Compounds and solutions of these substances were accumulating in metal barrels, glass bottles, and cardboard boxes piled on the cement floors of storerooms. Uranium was combined with oxygen or chlorine and either dissolved in water or kept dry. Workers moved these substances from centrifuges or drying furnaces into cans and hoppers. Much later, large epidemiological studies would overcome obstacles posed by government secrecy and disinformation to show that low-level radiation caused more harm than anyone had imagined. Yet the authorities at the processing plants were overlooking not only this possibility but also a more immediate and calculable threat: the possibility of a runaway, explosive chain reaction.
Feynman had seemed to be everywhere at once as the pace of work accelerated in 1944 and 1945. At Teller’s request he gave a series of lectures on the central issues of bomb design and assembly: the critical-mass calculations for both metal and hydride; the differences between reactions in pile, water boiler, and gadget; how to compute the effects of various tamper materials in reflecting neutrons back into the reactions; how to convert the pure theoretical calculations into the practical realities of the gun method and the implosion method. He became responsible for calculating the way the efficiency of a uranium bomb would depend on the concentration of uranium 235 and for estimating safe amounts of radioactive materials under a variety of conditions. When Bethe had to assign theorists to G Division (Weapon Physics Division—G for gadget) he assigned Feynman to four different groups. Furthermore, he let Oppenheimer know that, as far as the implosion itself was concerned, “It is expected that a considerable fraction of the new work coming in will be carried out by group T-4 (Feynman).” Meanwhile, though Feynman was officially only a consultant to the group handling computation by IBM machines, Bethe decreed that Feynman would now have “complete authority.”
At Oak Ridge, where the first batches of enriched uranium were accumulating, a few officials began to consider some of the problems that might arise. One letter that made its way to Los Alamos from Oak Ridge opened, “Dear Sir, At the present time no provisions have been made in the 9207 Area for stopping reactions resulting from the bringing together by accident of an unsafe quantity of material… .” Would it make sense, asked the writer—a plant superintendent with the Tennessee Eastman Corporation—to install some kind of advanced fire-extinguishing equipment, possibly using special chemicals? Oppenheimer recognized the peril waiting in such questions. He brought in Teller and Emilio Segrè, head of the experimental division’s radioactivity group. Segrè paid an inspection visit, other theorists were assigned, and finally the problem was turned over to Feynman, with his expertise in critical-mass calculations.
As Segrè had discovered, the army’s compartmentalization of information created a perilous combination of circumstances at Oak Ridge. Workers there did not know that the substance they were wheeling about in large bottles of greenish liquid was grist for a bomb. A few officials did know but assumed that they could ensure safety by never assembling any amount close to the critical mass estimated by the physicists. They lacked knowledge that had become second nature to the experts at Los Alamos: that the presence of hydrogen, as in water, slowed neutrons to dangerously effective speeds and so reduced the amount of uranium 235 needed to sustain a reaction. Segrè astounded his Oak Ridge hosts by telling them that their accumulating stores of wet uranium, edging closer to bomb-grade purity, were likely to explode.
Feynman began by retracing Segrè’s steps and found that the problem was even worse than reported. In one place Segrè had been led into the same storeroom twice and had inadvertently noted two batches as though they were accumulating in separate rooms. Through dozens of rooms in a series of buildings Feynman saw drums with 300 gallons, 600 gallons, 3,000 gallons. He made drawings of their precise arrangements on floors of brick or wood; calculated the mutual influence of solid pieces of uranium metal stored in the same room; tracked the layouts of agitators, evaporators, and centrifuges; and met with engineers to study blueprints for plants under construction. He realized that the plant was headed toward a catastrophe. At some point the buildup of uranium would cause a nuclear reaction that would release heat and radioactivity at near-explosive speed. In answer to the Eastman superintendent’s question about extinguishing a reaction, he wrote that dumping cadmium salts or boron into the uranium might help, but that a supercritical reaction could run away too quickly to be halted by chemicals. He considered seemingly remote contingencies: “During centrifuging some peculiar motion of the centrifuge might possibly gather metal together in one lump, possibly near the center.” The nightmare was that two batches, individually safe, might accidentally be combined. He asked what each possible stuck valve or missing supervisor might mean. In a few places he found that the procedures were too conservative. He noted minute details of the operations. “Is CT-1 empty when we drop from WK-1… ? Is P-2 empty when solt’n is transfered … ? Supervisor OK’s solution of P-2’s ppt. Under what circumstances?” Eventually, meeting with senior army officers and company managers, he laid out a detailed program for ensuring safety. He also invented a practical method—using, once again, a variational method to solve an otherwise unsolvable integral equation—that would let engineers make a conservative approximation, on the spot, of the safe levels of bomb material stored in various geometrical layouts. A few people, long afterward, thought he had saved their lives.
Wielding the authority of Los Alamos was an instructive experience. Feynman’s first visit to Oak Ridge was his first ride on an airplane, and the thrill was heightened by his special-priority military status on the flight, with a satchel of secret documents actually strapped to his back under his shirt. Oppenheimer had briefed his young protégé with care. Feynman decided that the plant could not be operated safely by people kept ignorant of the nature of their work, and he insisted that the army allow briefings on basic nuclear physics. Oppenheimer had armed him with a means of handling difficult negotiations:
“You should say: Los Alamos cannot accept the responsibility for the safety of the Oak Ridge plant unless——”
“… You mean me, little Richard, is going to go in there and say——”
“… Yes, little Richard, you go in there and do that.”
John von Neumann may have advised him during their thin-air walks that there could be honor in irresponsibility, but amid the barrels and carboys of the world’s first nuclear hoards, responsibility caught up with him. Lives depended on his methods and judgments. What if his estimates were not conservative enough? The plant designers had taken his calculations as fact. He hovered outside himself, a young man watching, unsure and giddy, while someone carried off an impersonation of an older, more powerful man. As he said, recalling the feeling many years later, he had to grow up fast.
The possibility of death at Oak Ridge tormented him more urgently than the mass slaughter to come. Sometime that spring it struck him that the seedy El Fidel hotel, where he had nonchalantly roomed on his trips to Albuquerque, was a firetrap. He could not stay there any more.
I Will Bide My Time
Hitchhiking back one Sunday night, nearing the unpaved turnoff to Los Alamos, he saw the lights of a carnival shining from a few miles north in Espanola. Years had passed since he and Arline last went to a carnival, and he could not resist. He rode a rickety Ferris wheel and spun about in a machine that whirled metal chairs hanging on chains. He decided not to play the hoop-toss game, with unappealing Christ figures as prizes. He saw some children staring at an airplane device and bought them a ride. It all made him think sadly about Arline. Later he got a lift home with three women. “But they were kind of ugly,” he wrote Arline, “so I remained faithful without even having the fun of exerting will power to do it.”
A week later he rebuked her for some act of weakness and then, miserable, wrote the last letter she would read.
My Wife:
I am always too slow… . I understand at last how sick you are. I understand that this is not the time to ask you to make any
effort to be less of a bother to others… . It is a time to comfort you as you wish to be comforted, not as I think you should wish to be comforted. It is a time to love you in any way that you wish. Whether it be by not seeing you or by holding your hand or whatever.
This time will pass—you will get better. You don’t believe it, but I do. So I will bide my time & yell at you later and now I am your lover devoted to serving you in your hardest moments… .
I am sorry to have failed you, not to have provided the pillar you need to lean upon. Now, I am a man upon whom you can rely, have trust, faith, that I will not make you unhappy any longer when you are so sick. Use me as you will. I am your husband.
I adore a great and patient woman. Forgive me for my slowness to understand.
I am your husband. I love you.
He also wrote to his mother, breaking a long silence. One night he awoke at 3:45 A.M. and could not get back to sleep—he did not know why—so he washed socks until dawn.
His computing team had put everything aside to concentrate on one final problem: the likely energy of the device to be exploded a few weeks hence at Alamogordo in the first and only trial of the atomic bomb. The group’s productivity had risen many times since he took over. He had invented a system for sending three problems through the machine simultaneously. In the annals of computing this was an ancestor to what would later be called parallel processing or pipelining. He made sure that the component operations of an ongoing computation were standardized, so that they could be used with only slight variations in different computations, and he had his team use color-coded cards, with a different color for each problem. The cards circled the room in a multicolored sequence, small batches occasionally having to pass other batches like impatient golfers playing through. He also invented an efficient technique for correcting errors without halting a run. Because a mistake only propagated a certain distance in each cycle, when an error was found it would have tainted only certain cards. Thus he was able to substitute small new card decks that eventually caught up with the main computation.