Gunnar Kallén, physicist

Kallén was the leading Scandinavian physicist until his tragic accident in 1968 while flying his own plane from CERN, Geneva, when he was Professor of Physics at Lund. He died in the crash, but during the descent he secured the lives of his two passengers, his wife Gunnel, and Mrs. von Dardel, the wife of a colleague. His wife also died young, a few years later.

Kallén first made his name writing about quantum electrodynamics in the (German language) Encyclopaedia der Physik. Pauli was particularly impressed by this work. Wightman has said that while he was struggling to understand the grammar of quantum field theory, here was Kallén already writing poetry in the language. I met Kallén while a student at the Varenna Summer School in Summer 1958, by Lake Como. Also there were Pauli, Heisenberg, Symanzik, Garding, J. L. Lyons...All the participants were invited (by the Mayor of Como?) to a boat-trip round lake Como one evening; we were to have dinner on the deck. I noticed Pauli and Kallén were alone at a table for three, so I asked if I could join them. They were speaking German, and said yes, what was my language? I said Englisch, so they switched to English, and continued. When we had finished dinner it was still light, and we had returned to Varenna; the Captain said that we would go round the lake again, because we had not had time for coffee. When we returned the second time, it was dark, and as we passed Varenna, the whole mountainside erupted in fireworks, which made out the words COMO SALUTI I FIZICI (perhaps it was gli fizici, with some dud fireworks).

Kallén developed with Arthur Wightman a new technique for finding domains of holomorphy for functions of several complex variables, a thing too hard for mathematicians at the time. These authors were able to characterise exactly those vertex functions (in a general theory of quantised fields) that incorporated the principle of causality in special relativity theory. Their result was a class of functions of three invariants much more general than given by the formula proposed by Schwinger. In spite of Kallén's famous public refutation of Schwinger's formula, documents held at Lund's Physics Dept. show that Kallén's own attempts did not at first give the most general class of functions either. He had the good sense not to publish these.

Kallén and Schwinger met again at the Rochester Conference on Particles and Fields (Rochester, New York State, 28 Aug-1 Sept., 1967). Schwinger had just invented his "sourcery", which is a way round the divergences of quantum electrodynamics by neglecting certain terms corresponding to back reaction. He was a leading speaker at the conference, along with R. P. Feynman, Y. Nambu, S. Mandelstam, A. S. Wightman, J. S. Bell, R. Haag, G. W. Mackey and Y. Nee'man, J. Cronin and others. C. N. Yang and Hideki Yukawa were present, the latter's work being presented by his coauthor, Y. Katayama.

Schwinger illustrated the use of sourcery in computing the decay rates of certain mesons. Kallén asked the question, why is this new, since Ben Lee (and a coworker) had already computed these decay rates some time before. Schwinger replied that he did not know of this work, but it could not be the same, as sourcery had only been invented (by himself) a few months earlier. Kallén insisted on a supplementary question: he said, that if you start with Lee's assumptions about the interaction, they were the same as Schwinger's; the first line of the calculation is the same; going through the calculations, said Kallén, write Lee's working in the left-hand column, and sourcery in the right-hand column. The workings are, line by line, the same in Lee's paper as in Schwinger's lecture. And, the answer is the same. "That may be so", said Schwinger, "but one column is right and the other column is wrong". The laughter did not allow any further questions. Even Irving Segal, who was with me, smiled. This exchange is omitted in the proceedings of the conference [Proc. of the 1967 International Conference on Particles and Fields, Interscience, 1967; eds. C. R. Hagen, G. Guralnik, and V. S. Mathur]. The discussion presented there seems to be a sanitised version, in which Ben Lee gives a measured argument similar to Kallén's, but more polite. His coauthor, by the way, was Nieh.

Kallén wrote a book, Elementary Particle Physics, Addison-Wesley, 1964, which not only widened his own reputation, but helped to give a good name to other theoreticians working on axiomatic theory.