Midsummer 1944: the laboratory's worst season, and the author's account of how a sentence spoken by no one reorganized the war's most valuable institution and capitalized the following half-century. For once — the reader is begged to remain calm — the documents are on his side. —Ed.
Something came back from the measuring men in the last week of June that broke the laboratory's plan of work at one stroke — killed the way the whole shop had meant to do the thing it existed to do, and left standing only the way nobody knew how. What the something was I could not have told you then and decline to learn now; I have kept a clean record, fifty-six years, of not understanding that mountain, and I am too old to spoil a record.
A man did not need the physics; the crisis published itself in the housekeeping. Inside a fortnight the Tech Area was burning lights at three in the morning, cots stood made-up in the T-building corridor and were slept in, and the mess put on a second breakfast at midnight for men coming off shifts that had not existed in May. The motor pool ran the canyon roads at hours previously reserved for skunks. The computing room went onto nights, the wives cranking the machines at government rates — sound Army economics, the wife being the one machine on that mesa that runs better the less it knows what it is for, and cheaper than the hardware by a mile.
The principals you could read at fifty feet. Oppenheimer thinned by the week and lit each cigarette off the last. Bethe wore the face of a man doing long division in his sleep. Teller had downed tools altogether, over some point of precedence or prophecy — nobody could tell me which, and I doubt he could — and sat in a room by himself on the next war while the corridor sweated over this one.
And the corridor was fuller than it had ever been, the mesa being ankle-deep that summer in Europe's professors, the brains the Continent had thrown out with the rest of its trash, and a more grateful cargo you never saw — for about a payroll. Your refugee is humble to the clerk who stamps him and Socrates to every man after; gratitude in that breed keeps about as long as cut flowers. That July you could not walk to the latrine without being corrected in three accents, each certain the other two were fools.
The colloquium, which had been a weekly church, went to twice a week and then to whenever the bell rang, and through all of it I sat where I had sat since '43, at the little table by the blackboard, minuting. Nobody asks the secretary a question; it is the safest chair in any war. My minutes were famous by then — "unusually lucid" was the phrase that had got about — and the crisis put me on the wives' schedule: fair copy at midnight in the annex, the cabinet open, the bottle out.
The third Tuesday in July was the worst colloquium I ever sat. The room was over capacity by half — men in the windows, men in the doorway, ninety men's heat in a room built for fifty — and three of them held the chalk in the first hour and none gave it up willingly. At the ninety-minute mark the sentence happened. Two voices went off at once — a Kraut voice and a Hunky voice, crossing in the middle of it, while a third man went "no, no, no" in strict metronome time — and what arrived at my table had the pitch of a question, the freight of an accusation, and no verb I ever located. The room could not hold it either: half turned to answer, half asked its neighbors what had been said. The man beside me leaned over. "What did he say?" "Which one?" said the man on my other side, and that was the scholarship of the evening. The chairman rapped the urn and called the coffee, the argument stampeded off in a new direction and never came back for its sentence, and in my pad it went down as it arrived: three words and a dash.
Fair copy was midnight work, and at midnight I met the dash. A hole in the minutes is the one item the secretary cannot file: a hole is a question, and questions get answered by looking at the secretary. I poured a finger from the cabinet, read my three words until they stopped meaning even themselves, and then did what I had done once before with a gap in the observations. I took the wreck of the sentence, threw out every part that quarreled with the rest, typed what was left three ways on a flimsy, and struck out two. Reproduce the survivor I cannot, and would not if the censors begged: I built it by shape — short, aimed at the one stretch of blackboard no man's chalk had touched all month, ending on the plainest word in it — and I have never understood one word of my own masterpiece. Then at two in the morning I inspected the work the way the honest would: this time there was no support wire to cross the shadow. I typed the fair copy, filed the carbon, and went to bed a man who had tidied a garble.1
In an ordinary month the minutes went to the file and the few faithful. In a crisis everybody reads everything — fear is the great circulator — and by Wednesday night my fair copy had been up the Hill and back down twice. The deciding meeting was called for the Thursday, of course it was, the reader knows my calendar by now; and on the Thursday morning Pomeroy came down the corridor at a trot, beaming: upstairs wanted their record kept. "Kept the way the colloquium's is kept," he said. "They asked for you." I went up with the good pad.
The biggest room they had was short of chairs anyway; I took my place against the wall, where nobody looks. For the first hour the argument went in a circle and I minuted all four laps. Then Oppenheimer picked the minutes off the table — my minutes — found the page with one finger, and read the question out of them aloud, word for word as I had built it, in that voice of his that could make a freight schedule sound like scripture.
I have watched what happened next perhaps two hundred times since, in rooms where the stakes were merely money: the silence of a nail going in. The circling stopped. A man three seats down said, slowly, "That is the question. That has been the question since June." A second man began answering it and a third interrupted him with a better answer, and the room came into line behind a sentence with no father. Oppenheimer looked up once. "Whose was this?" There is a particular silence made by thirty proud men each declining to admit he cannot remember; it lasts about two seconds and sounds exactly like modesty. "It came from the floor," somebody said, and the floor accepted the attribution, and the meeting walked on into the decision it had been failing to reach all summer.
The year before, over these same minutes, Oppenheimer had told me I had Bethe saying what he meant instead of what he said, and that this was either very valuable or very dangerous, and he would be obliged if I knew which before he had to. From the wall that Thursday I watched him discover it was both, without his ever learning it was either. And I will enter the one true item of the morning: when the room came round behind my sentence I felt, for the length of a heartbeat, whatever it is honest men feel when they have been of use. By the coffee break I was pricing it.
Within the month they took the laboratory apart to the studs and rebuilt it around the answer — new divisions, new letters over the doors, several hundred men repotted in a fortnight, my own section among them. For a week the Tech Area was all handcarts, desks going one way and their owners the other, and a lieutenant stood in the road with a clipboard, directing furniture.2
I claimed nothing, naturally; the amateur's error is the claim that can be tested. In the mess a captain of engineers put it to me straight: had I truly been in the room when the question got asked? "It was a privilege," I said, "just to keep the record of a discussion like that," and wore the modest frown while he carried the story off to improve it. The true bill is short: I never asked the question; I typed it. Noise, sieved at midnight by a man who understood none of it — the accidental insight, which I have sold for fifty years under the trade name of vision; my mother's orange juicer had taught me who presses the button, and that Thursday taught me the man pressing the button can file for the patent.3 Credit is a liability that compounds. The trade is the asset.
Inside a week I heard the question quoted at three tables by men who had been in neither room, no two renderings alike and every one an improvement; your European does not repeat a sentence, he annexes it. Deny a thing gracefully enough and other men will build it a shrine. The memoirs, when they came — 1958, 1969, 1981 — gave my sentence three separate fathers and not one of them me,4 which suited me down to the socks.
One man on the mesa knew what the question was made of. On the Sunday I took the coffee at Fisk's as usual, and was two sentences into the mess-hall version when he got up without a word, lifted the current notebook down from its shelf, and wrote, standing, dating the line first, in that small square hand you could read across a room. Then he turned the book around on the table and let me read it. 23 July. The question they are all repeating was asked by nobody. It is in K.'s minutes. It was never in the room. He closed the book and put it back on the shelf. "Last of the pot?" he said. We finished the coffee on the weather. So the truth about the most useful question of the age is carried at its correct value on exactly one ledger in the world, on a shelf I could have reached across the table and never again dared touch.
The reorganization, meanwhile, had thrown every billet on the mesa into play — reassigned men are reassigned beds, and the beds were file cards behind a counter at the housing office, where I presented myself on the Monday to defend my room. The line was out the door and I stood in it. Behind the counter sat a WAC with every card on the Hill at her fingertips, and she looked up and priced me faster than Anton had. Some catastrophes you smell two days out. The best ones you stand in line for.
The unsigned fair-copy colloquium minutes noticed at No. 4 — the hand the archivists call "unusually lucid" — survive for July 1944, the question among them. The editor has also examined the contemporaneous working notes of three men who attended that Tuesday. None records the question. One records, at the corresponding point in the evening, "two at once — lost it," and another simply "Babel." The sentence exists first, and only, in the one document written by a man alone at midnight. —Ed.
The crisis of that summer was real, and so was the answer: in August 1944 Oppenheimer took the laboratory apart to the studs and rebuilt it around the problem — new divisions, new letters, hundreds of men reassigned inside a month. Historians of administration rank it among the great management strokes of the war, and rightly. The reorganization had been gathering for weeks before the meeting the author describes; the decision had many fathers and needed all of them. Whether it also needed a question is the one point on which the record is silent, minutes excepted. —Ed.
Readers of the author's 1987 Stanford commencement address, "Build Boldly," will recognize the counsel "ask the one question the room keeps walking past," together with its illustration: a young soldier who asked exactly that in a room containing the century's best minds. The Stanford version differs from the account above in a single word, which is the verb. —Ed.
Three memoirs — published in 1958, 1969, and 1981 — credit "the question that clarified everything," and award it three different askers: a senior European colleague, "a young man at the back whose name escapes me," and, in the boldest of the three, the memoirist himself. None mentions the author, whose name appears in the index of all three books only under "minutes, quality of." On this single point, the Papers are the only account consistent with the surviving documents. The editor has verified that sentence twice and prints it under protest. —Ed.