Joel Greenhouse writes:
I saw your recent paper on Feller [see here and, for a more fanciful theory, here]. Looks like it was fun to write. I recently wrote a paper that asks an orthogonal question to yours. Why during the 1950-1960′s did Jerry Cornfield become a Bayesian? It appeared in Statistics in Medicine – “On becoming a Bayesian: Early correspondences between J. Cornfield and L. J. Savage.”
In his paper, Greenhouse writes:
Jerome Cornfield was arguably the leading proponent for the use of Bayesian methods in biostatistics during the 1960s. Prior to 1963, however, Cornfield had no publications in the area of Bayesian statistics. At a time when frequentist methods were the dominant influence on statistical practice, Cornfield went against the mainstream and embraced Bayes. . . . Cornfield’s interest in Bayesian methods began prior to 1961 and that the clarity of his Bayesian outlook began to take shape following Birnbaum’s ASA paper on the likelihood prin- ciple and his subsequent discussions with Savage.
This is interesting to me because I find Savage to be the least convincing of early modern Bayesian writers. I find Lindley, Good, and Box all to be much more persuasive.
Anyway, here’s what I wrote in reply to Greenhouse: Thanks for sending–this is fascinating. I always thought Cornfield worked in agricultural experiments, but I guess that was because of his name.
The sad thing is: this is no joke!
Anyway, Joel generously replied:
If it makes you feel better during WWII he worked for the Bureau of Labor Statistics and solved a famous a linear programming problem. called the Diet problem which had to do with food, which is pretty close to agriculture.
I’m always glad to hear of others who are interested in the history of statistics.