I should mention here one of my own peculiarities: I do not like walking uphill. I don't really know why. Some people tell me that I tend to go too fast from impatience and get winded for that reason. I do not mind walking on level ground and I actually enjoy walking down hill. Years ago I bought a German travel guidebook called "One Hundred Downhill Walks in the Alps." Certainly a humorous title.
After the war, on one of these downhill excursions in Frijoles Canyon, I told Fermi how in my last year of high school I was reading popular accounts of the work of Heisenberg, Schrödinger, and De Broglie on the new quantum theory. I learned that the solution of the Schrödinger equation gives levels of hydrogen atoms with a precision of six decimals. I wondered how such an artificially abstracted equation could work to better than one part in a million. A partial differential equation pulled out of thin air, it seemed to me, despite the appearances of derivation by analogies. I was relating this to Fermi, and at once he replied: "It [the Schrödinger equation] has no business being that good, you know, Stan."
He went on to say that in the fall he intended to give a really logical introduction and derivation of quantum theory in his course at the University of Chicago. He apparently worked at it, but told me the next summer, when he returned to Los Alamos, "No, I didn't succeed to my satisfaction in giving a really rational introduction to quantum theory." It is not just a question of axioms as some naive purist might think. The question is why such and no other axiom? Any working algorithm can be axiomatized. How to introduce, justify, tie up or simplify the axioms, historically or conceptually, and how to base them on experiments — that is the problem.
Von Neumann and Fermi were really quite different in personality. Johnny was perhaps broader in his interests than Enrico. He had more specifically expressed interests in other fields, certainly, for example, in ancient history. Fermi did not show any great interest in or liking for the arts. I never remember him discussing music, painting, or literature. Current affairs, politics, yes; history, no. Von Neumann was interested in both. Fermi did not indulge in quotations or allusions, Latin or otherwise, although he liked epigrammatic formulations occasionally. But he did not display a gymnasium or lycée type of education or the resultant mental habits. His overwhelming characteristic was his Latin clarity. Von Neumann did not consciously insist on simplicity; on the contrary, he liked to show clever complications on occasion.
In their lectures to students or scientific gatherings, they demonstrated their different approaches. Johnny did not mind showing off brilliancy or special ingenuity; Fermi, on the contrary, always strived for the utmost simplicity, and when he talked everything appeared in a most natural, direct, bright, clear light. After students had gone home, they were often unable to reconstruct Fermi's dazzlingly simple explanation of some phenomenon or his deceptively simple-looking idea on how to treat a physical problem mathematically. In contrast, von Neumann showed the effects of his sojourns at German universities. He was absolutely devoid of pomposity, but in his language structure he could be complicated, though perfect logic always gave a unique interpretation to his words.
They held high opinions of each other. I remember a discussion of some hydrodynamical problem Fermi had been thinking about. Von Neumann showed a way to consider it, using a formal mathematical technique. Fermi told me later with admiration, "He is really a professional, isn't he!" As for von Neumann, he always took external evidences of success seriously; he was quite impressed by Fermi's Nobel Prize. He also appreciated wistfully other people's ability to get results by intuition or seemingly pure luck, especially by the apparent effortlessness of Fermi's fundamental physics discoveries. After all, Fermi was perhaps the last all-around physicist in the sense that he knew the theory, did original work in many branches, and knew what experiments to suggest and even do himself; he was the last to be great both in theory and as an experimenter.
Niels Bohr, the discoverer of the quantized electron orbits in the atom and a great pioneer of quantum theory, was in Los Alamos for several months. He was not very old. To me at thirty-five, he seemed ancient, even though Bohr in his late fifties was very active and energetic, physically as well as mentally. He walked, skied, and hiked in the Los Alamos mountains. Somehow he seemed the embodiment of wisdom. (Wisdom, perhaps not genius in the sense of Newton or Einstein.) He knew what not to attempt and how much could be done without mathematics, which he left to others. This enormous wisdom is what I liked about him.
Departing from his usual caution about expressing opinions about other people, Fermi remarked once that when Bohr talked he sometimes gave the impression of a Catholic priest celebrating mass. It was an iconoclastic statement, since so many physicists are still under the spell of Bohr.
He had his own kind of genius that made him a great physicist but, to my mind, some of his students were almost benighted by his complementarity philosophy of "one can say this, but on the other hand one can…" or "one cannot say sharply what this means." People without his great sense and intuitive wisdom were led astray and lost the precision and sharpness of their intellectual or scientific approach, in my opinion. But he still has many admirers. Victor F. Weisskopf is one.
It seems to me that as a philosophical guideline, complementarity is essentially negative. It can only console. Whether it can be positively useful other than in philosophical consolations is a question which troubles me.
Bohr's speech was very difficult to understand, and anecdotes about him abound. Most of the time it was impossible to get his exact words. One day a young physicist, Ruby Scherr, was called on the public address system. Here I should explain that every day at periodic intervals the halls of the laboratory resounded with announcements and requests, the most frequent of which was a call for J. J. Gutierrez, who was a supply factotum and jack of all trades. Other calls were requests for the return of such and such an instrument, or even the Sears, Roebuck catalogue. One day among other announcements came one asking Ruby Scherr to please go to Nicholas Baker's office. (Nicholas Baker was the pseudonym of Bohr for security reasons; Fermi's was Farmer.) As Ruby Scherr tells the story, he went to the office, saw several physicists sitting around and obviously listening to a presentation by Bohr. Bohr stopped, mumbled a few incomprehensible sentences in the direction of Scherr, and suddenly ended with a crystal-clear three words: "Guess how much?" Scherr, who had not understood a word of the question, blushed with embarrassment, shook his head shyly and remained silent. After a moment Bohr again in a clear voice said "1041." Whereupon everybody laughed. To this day, Scherr does not know what it was all about.
Another Bohr story illustrates the absentmindedness of scientists: it was well known throughout Los Alamos that Nicholas Baker was Niels Bohr; nevertheless, his true name was never supposed to be mentioned in public. At one colloquium, Weisskopf referred to "the well-known Bohr principle." "Oh excuse me," he fumbled, "the Nicholas Baker principle!" General laughter greeted this security breach.
Not all of us were unduly security conscious. Every scientist, old or young, had in his office a safe where secret documents had to be kept. Indeed, the Project must have had more safes than all the banks in New York. Once in Bohr's office I watched him struggle to open his safe. The safes could be opened with a rather simple combination of three two-digit numbers. He tried and tried for a long time, finally succeeding. He pulled out the drawer and exclaimed delightedly: "I believe I have done enough for the day." The story that Dick Feynman could open safes whose combinations had been forgotten by their owners is true. He apparently listened to the clicks of the tumblers and sometimes he guessed which combinations of digits of numbers like π or e in mathematics, or c, the velocity of light, or h, Planck's constant, had been selected by the owners for the combinations.