#### by Manish Bhattacharjee

In David Blackwell, we have that extraordinary combination of exceptional scholarship and superb teaching that all academicians aspire to but rarely achieve. The teaching aspect was manifest at all levels, ranging from a basic introductory course to cutting-edge research. Anyone who has been at Berkeley for any length of time is familiar with the fact that his section of “Stat-2” class routinely had more students than the combined enrollment of all other sections of the same course, which were typically taught by others. My own introduction to his teaching style was also through an undergraduate course in dynamic programming that I took in my second or third semester as a graduate student. His engaging style of explaining a problem and bringing it to life in this class I believe, on reflection, was a decisive influence on my choosing stochastic dynamic programming and its applications to probability as the primary area for my dissertation research. One of my fond memories of how he brought a problem to life in class concerns a story of how, over a period of time, he and Samuel Karlin wrestled with the problem of proving the optimality of the so-called “bold play” strategy in a subfair “red-and-black” game that is an idealized version of roulette.

The *New York Times* obituary of July 17, 2010,
describes him as “a free-ranging problem solver”,
which of course he truly was. His was a mind
constantly in search of new challenges, breaking
new ground in different areas every few years
with astounding regularity. As
Thomas Ferguson
told *UC Berkeley News* in an interview recently,
Blackwell “went from one area to another, and he’d
write a fundamental paper in each.” His seminal
contributions to the areas of statistical inference,
games and decisions, information theory, and
stochastic dynamic programming are well known
and widely acknowledged. He laid the abstract
foundations of a theory of stochastic dynamic
programming that, among other things, led
Ralph Strauch,
in his 1965 doctoral dissertation under
Blackwell’s guidance, to provide the first proof
of
Richard Bellman’s
*principle of optimality*, which
had remained until then a paradigm and just a
principle without a proof!

Each of us who have been fortunate enough to count ourselves to be among his students will no doubt have our personal recollections of him that we will fondly cherish. A recurring theme among such recollections and the lasting impressions they have left on us individually, I believe, would be his mentoring philosophy and style. He encouraged his students to be independent and did not at all mind even if you did not see him for extended periods as a doctoral student under his charge. He trusted that you were trying your level best to work things out yourself and waited until you were ready to ask for his counsel and opinion. A consequence of this, exceptions notwithstanding, was perhaps that his students took a little longer than average to complete their dissertations (although I have no hard data on this), but it made them more likely to learn how to think for themselves.