I met Chandrasekhar twice. The first was at a seminar at Oxford University arranged by Michael Atiyah, who had invited Irving Segal to give a talk about his (then) recent work on a model predicting a non-expanding universe. Atiyah had persuaded Chandrasekhar to attend, and the latter sat uneasily at the front. Segal started with the assumption that space-time is a poset (here I saw Chandra wince) with the partial order x< y meaning that events at y can be influenced by events at x. He then added more structure, bit by bit, and was able to announce the main theorem, that space-time was isomorphic to R ×SU(2,C). To explain the red shifts, he made the claim that all coordinates (related by a diffeomorphism) are physically allowed, and so was able to introduce a reparametrisation of the time R by tau=t/(1+t²), where tau is the time observed by us, and t is the physical time. The red-shift then appears as a quadratic curve, rather than a line, as in Hubble's law. The quasars are then near objects, and so no outlandish model is needed to explain their luminosity. Segal then showed a graph in which the red-shifts of about 8 quasars are roughly fitted to a quadratic curve, the other axis being (his version of the) distance. At this point, the roll of blackboards he was using returned to the first one, on which was written the word "poset". Segal ended his talk with the quip "I see that space is compact in Oxford, in accordance with my model".
Atiyah then invited Chandrasekhar, as the century's most eminent astrophysicist, to open the questions. Chandra complained that gravity was nowhere mentioned, and that there were no dynamical laws in the theory. He said that he did not think that one could justify the model by fitting half a dozen points to a quadratic curve. Segal replied that one could still put a metric on the space-time of the model, and study equations of motion of the metric and other fields; and he did not agree about his fit to the data; surely, physics progresses by setting up a clear model, with clear predictions; one then asks the experimentallists for their results, and does a best fit using modern statistical methods? Chandra gave a ripost, at Atiyah's invitation: "I don't know where you got your data, but if it is from the dissident Italian group that I think it is, let me tell you, Segal, that I have been editor of the Astronomical Journal for 30 years, and in that time, not one, NOT ONE, paper from that group has ever appeared in my journal". At which point Atiyah looked at his watch and announced breezily "Time for tea".
Soon after that, I saw Chandrasekhar at the Research Institute for Mathematical Research, Kyoto. He came up and said, that he was sorry for his outburst at Oxford. I assured him that I completely agreed with him.
Since then, as I understand it, experiment has disproved that quasars are near objects; they have been observed being passed on the earth's side by drifting galaxies that are known to be very far away.
Go to my HOME PAGE for more links.
© by Ray Streater, 17/12/00.