Sam Taylor, a nuclear engineering undergraduate in my intro nonlinear dynamics class alerted me to a wonderful (Web of Stories) video series of Sam Schweber interviewing Freeman Dyson.
While I was at the Institute in 1975-1978, I was Freeman's very young friend. What brought us together was that the time I had been a very bad dancer in a very charming performance of Blake's "Songs of Innocence and of Experience". 
Here Freeman is recorded in 1998, looking even more youthful then when I knew him. He was the only physicist who would hear about my 1976 work on universality in chaos. It was the most important thing I accomplished as IAS Fellow, but -fore example- Tullio Regge would not hear of it. Dyson was and remained the only physicist I knew who wanted to learn things regardless of whatever he had done them himself. So he became and remained my only hero and role model in this profession.
Shin'ichirō Tomonaga (and, for me, Tom Kinoshita's war years) enters, unexpected, out of the ashes:
And meanwhile, of course, we'd heard about Tomonaga's work which was, I think, also in the spring of '48, when Tomonaga sent his first two papers from Japan, and these came as an absolute total surprise, that somebody in the rubble of Tokyo was actually able to do physics. I hadn't heard of Tomonaga previously and he wrote to Oppenheimer from Tokyo and Oppenheimer sent a copy of the papers to Hans Bethe in Cornell, and so we saw them there. And these two papers of Tomonaga, it was called, I don't remember... on the many... This was the new Japanese journal "The Progress of Theoretical Physics" which was published in English. It was published on brown paper which was all they had in Tokyo at the time, and it was just like a voice from the deep.
I mean we thought of Japan as being a total ruin and there was this man who had somehow or other kept physics alive all through the war, and there it was. And he'd in fact done all this long before Schwinger and essentially arrived at the same results as Schwinger three or four years earlier, with more or less the same techniques.
[Schweber] You stay in Cornell for two years and then you go to the Institute. Do you want to say a few words [...] about life at the university there, and what made you decide to [...] accept an invitation to the Institute in '53?
This was a hard choice, because I was spiritually much more at home at Cornell. Cornell is a much warmer place. It's a real community, partly because of Hans. I mean Hans made it like that, but even without Hans - it's a place which commands enormous loyalty. I mean the friends that we made at Cornell 40 years ago, a lot of them are still there. These people just never leave, including Hans himself, who's now been there for 60 years. And so I felt very much at home there and sort of spiritually I still feel more at home in Ithaca than I do in Princeton. So there were these strong forces keeping me at Cornell. Cornell had always been my vision of America, whereas Princeton is not. Princeton is definitely an alien growth in America. Ithaca is the real thing. So from that point of view I would have preferred to stay in Ithaca, and also I love the people there.
About PhDs:
But I hated the PhD system, and that was what - I felt basically out of tune with the main job I had at Cornell, which was to train PhD students. The whole PhD system to me is an abomination. I don't have a PhD myself, I feel myself very lucky I didn't have to go through it. I think it's a gross distortion of the educational process. What happens when I'm responsible for a PhD student, the student is condemned to work on a single problem in order to write a thesis, for maybe two or three years. But my attention span is much shorter than that. I like to work on something intensively for maybe one year or less, get it done with and then go on to something else. So my style just doesn't fit this PhD cycle. What would happen, a PhD student would want to go on working on a problem for two or three years, but I would lose interest before he was finished. And so there was a basic mismatch between the way I like to do physics and this straightjacket which was imposed on the students. And so I found it was very frustrating [...]; all the PhD students had these same constraints imposed on them, which I basically disapprove of. I just don't like the system. I think it is an evil system and it has ruined many lives.About understanding Schwinger:
That was another of the Columbian experiments done by Kusch and Foley, where they measured the magnetic moment of the electron very precisely, and they found that it was not the Dirac value, but differed from the Dirac value by a certain finite amount, which then Schwinger was able to calculate. [...] It involved things that Schwinger called Green's Functions, which again I found rather incomprehensible, but Schwinger was very obscure when he described the things he was doing. He loved to make the - as Oppenheimer said, most people when they explain something, they're telling you how to do it; but when Schwinger explains something he's telling you that only he can do it! That was roughly the way it was, I mean, so from Schwinger one only had the impression that this was so difficult and so elaborate a way of calculating, only Schwinger could possibly do it. [...] in Ann Arbor in June of '48 I learned the stuff from Schwinger himself. I listened to his lectures in the morning and I spent the afternoons working very hard, just simply going through the Schwinger lectures step by step and really understanding what he'd been saying, which was very hard work because he just had this wonderfully baroque style of lecturing in which everything was dressed up to be as complicated as possible, and the answer somehow came out miraculously at the end.
There were two problems which Cécile and I took along for Feynman to deal with... [...], and the scattering of light by light; especially the scattering of light by light had been a sort of a notoriously difficult problem.[...] It was a formidable calculation and it wasn't at all clear that we could do it by the new methods of Feynman. And it was something we had to get straightened out before this whole theory was really complete. So we asked Feynman about this, and he said, 'Let's see about that.' And he sat down and he just worked through it; in about three quarters of an hour he'd done the whole thing and it was the most amazing performance. He just - with his lightning calculations, it all came out very beautifully. It turned out that the third order effect was zero and the fourth order effect was finite and everything worked exactly the way we wanted.
When I entered graduate school, theoretical particle physics was
the 1965 Bjorken-Drell "Relativistic Quantum Fields", i.e. Feynman diagrams. Feynman diagrams and Feynman
diagrams only. But my academic grandfathers were Feynman and Tomonaga (via
Tom Kinoshita), Schwinger (via Tung-Mow Yan), with Freeman my academic godfather. I wrote a book about it. Feynman reviewed it. Were it not for Tung-Mow, I would have never understood the Schwinger tale.
