With the death of David Blackwell, following the death last year of, both at ninety-one years old, the senior level of statisticians at U.C. Berkeley is gone.
Erich and I were reasonably close to David. For many years we drove him to the joint Berkeley–Stanford colloquia when they were held at Stanford, giving us time for much conversation.
Erich told me that, when David first came to Berkeley, his large family could not find any place that would accommodate them, and they lived for some months camping out in a park. In David’s interview in Statistical Science in 1986, he stated that the early discrimination he faced (before Berkeley) “never bothered me”. Since the early Berkeley experience wasn’t mentioned, it’s not clear that he was similarly unbothered by this early Berkeley discrimination. I know that he was keenly aware of such issues later.
I talked with him from time to time about discrimination, mentioning things that had happened to me both as a woman and a Jew. I had the impression that he thought a lot about discrimination in general (not necessarily against him personally) in later years. I understand his initial unconcern very well. When I was first looking for jobs there were many ads, at least half, that stated “men only”. It didn’t bother me then — it was the way things were. Only later with the rise of feminism did I begin to see things differently. David was usually very unruffled, but I saw a rare strong reaction when I told him how the Georgia flag, which resembles the Confederate flag, bothered me as I saw it flying over a hotel in which I had just stayed in Atlanta. He mentioned that a white beggar on Telegraph Avenue had approached him for money while wearing a Confederate flag costume and how angry he had felt about that.
It must have been somewhat difficult being a Bayesian in the strongly-non-Bayesian Berkeley Statistics Department. I once mentioned to David that I was not very sympathetic to the Bayes approach but did have some interest in empirical Bayes. He noted that he didn’t believe in empirical Bayes and showed that it didn’t make sense when applied to a single inference. I remarked that it made sense in the context of a large number of similar inferences, to which there was no reply. I interpreted his reaction as illustrative of an aspect of his approach to statistics. He liked elegance and simplicity. Issues had to be clear in the very simplest of situations. Empirical Bayes didn’t meet this test. He felt a Bayesian prior was necessary. His ability to clarify issues in simple and elegant ways was presumably what made him such an outstanding lecturer and teacher.
David was a kind and wonderful person, but he was also a very private person, and there was always the feeling of an inner core that couldn’t be penetrated. I urged him many times to write his memoirs, but he never did.