Sir David Cox
David Cox sent me a letter relating to my post of Oct.5, 2013. He has his own theory as to who might have been doing the teasing! I’m posting it here, with his permission:
Dear Deborah
I was interested to see the correspondence about Jeffreys and the possible teasing by Neyman’s associate. It brought a number of things to mind.
- While I am not at all convinced that any teasing was involved, if there was it seems to me much more likely that Jeffreys was doing the teasing. He, correctly surely, disapproved of that definition and was putting up a highly contrived illustration of its misuse.
- In his work he was not writing about a subjective view of probability but about objective degree of belief. He did not disapprove of more physical definitions, such as needed to describe radioactive decay; he preferred to call them chances.
- In assessing his work it is important that the part on probability was perhaps 10% of what he did. He was most famous for The earth (1924) which is said to have started the field of geophysics. (The first edition of his 1939 book on probability was in a series of monographs in physics.) The later book with his wife, Bertha, Methods of mathematical physics is a masterpiece.
- I heard him speak from time to time and met him personally on a couple of occasions. He was superficially very mild and said very little. He was involved in various controversies but, and I am not sure about this, I don’t think they ever degenerated into personal bitterness. He lived to be 98 and, a mark of his determination is that in his early 90’s he cycled in Cambridge having a series of minor accidents. He was stopped only when Bertha removed the tires from his bike. Bertha was a highly respected teacher of mathematics.
- He and R.A.Fisher were not only towering figures in statistics in the first part of the 20th century but surely among the major applied mathematicians of that era in the world.
- Neyman was not at all Germanic, in the sense that one of your correspondents described. He could certainly be autocratic but not in personal manner. While all the others at Berkeley were Professor this or Dr that, he insisted on being called Mr Neyman.
- The remarks [i] about how people addressed one another 50 plus years ago in the UK are broadly accurate, although they were not specific to Cambridge and certainly could be varied. From about age 11 boys in school, students and men in the workplace addressed one another by surname only. Given names were for family and very close friends. Women did use given names or were Miss or Mrs, certainly never Madam unless they were French aristocrats. Thus in 1950 or so I worked with, published with and was very friendly with two physical scientists, R.C. Palmer and S.L. Anderson. I have no idea what their given names were; it was irrelevant. To address someone you did not know by name you used Sir or Madam. It would be very foolish to think that meant unfriendliness or that the current practice of calling absolutely everyone by their given name means universal benevolence.
Best wishes
David
D.R.Cox
Nuffield College
Oxford
UK[i]In comments to this post.
Error Statistics Philosophy 
Dear David: Thanks so much for the comment! I hadn’t thought of the possibility that Jeffreys was just ribbing Neyman with a blatant misuse of his concept of probability, and then needling him further by saying his assistant, Hosiasson, is the one who clued him in.
So my tendency to use “Mayo” is in sync with “how people addressed one another 50 plus years ago in the UK” (males at least).
On Neyman being “Germanic,” or not: When I first met him, he was very old and I was very young. Yet he treated me as an equal, and his manner was anything but formal or dogmatic, not someone to be addressed (if he were German) as Herr Doktor Professor. When I told him I didn’t approve of hypothesis testing, he immediately replied “Then you shouldn’t do it,” in a tone that I believe meant that he was encouraging me to think outside the box. This, by the way, was at a time in which it was still not so nice to tell a Pole that he was German-like.
I must add something about David Cox: I’ve never met the man, but have always admired him, for the clarity of thought that pervades his writings, and his evident view that theory should be used but not abused. He accepted my first real statistics paper, for Biometrika (I’d just published in probability theory until then), which I was/am proud of. I was pleased, upon seeing Deborah’s (er, Mayo’s) work with Cox, to find that he is still active. They don’t make statisticians like D.R. Cox any more, sad to say.
Norm: I very much like your Neyman story, and I can just hear him say it–so in keeping with how he viewed statistics, in my judgment (and how unlike some others).
As a mathematician working on complex problems I am intrigued by Cox’s characterisation of Jeffreys work as ‘objective’ and ‘logical’. My reading of Jeffreys preface is that he regards his approach as making Keynes’ work obsolete, yet he notes that his approach rests on an assumption of simplicity, while it is not clear – for epidemics for example – that this assumption is always justified. It would be good to clarify this. What did he think the scope of his concept was? How would one recognize an area where it could mislead?
Sir David’s comments are as always much appreciated.
The “Germanic” comment might be a reaction to something I said in the earlier thread
“Neyman came from a different tradition closer to the German style in which exact attention to titles and degrees is essential.”
and I just want to stress that I didn’t want to imply that Neyman was Germanic, just that his use of titles was closer to that style.
(In case anyone is wondering, Sir David and I are not (knowingly) related.)
(A very small story of my own is that once I was admitted to Cambridge in 1969 the style of correspondence from the College I applied to changed from “Dear Mr Cox” to “Dear Cox”. I was a little startled at this but soon gathered that far from being a little rude, reverting to what was familiar in secondary school, it was a very small signal that I was now on the inside, albeit as a very junior member of the University.)
Thanks Sir David for a great post!
I feel reminded of Frank Hampel with whom I worked in Zuerich, who always stressed his high esteem for both Fisher and Jeffreys for the way their statistical ideas were informed by their strong experience in top level applied data analysis combined with their competence and leading role in scientific areas other than statistics. He was/is much less keen on Neyman, which I always found a bit unfair.
Should I as a German regular reader of this blog comment on the discussion on what’s Germanic? Well, I’d rather not, but have fun with that!
Christian:
Of course if you can provide insights into “Germanic” please do, but not if you’d rather not.
I’ve come in too late in this story to have an opinion. David is too nice to say that Jeffreys is probably best remembered among geophysicists as the unwavering opponent of continental drift. His prestige held back its acceptance for decades.
Just a word about Jeffreys’ very early work in physics before he became statistical. IN 1919 Jeffreys published a brilliant analysis of how the “classical tests” bear on aspects of the general theory of relativity. Jeffreys essay is remarkable in several respects, not least for his comprehension of the theory at a time when it was difficult even to find a copy of the paper in England. But second, his analysis is entirely a-statistical and uses only the mathematical and logical structure of the theory. In the part of the paper of most interest to me, Jeffreys considers a set of minimal assumptions of the theory that, in combination with the results severally of the tests, entail either instances of a claim of the theory, or a counterexample, depending on the outcome of the tests.
No statistical theory available at the time (or, I would argue, now) captures the relationships Jeffreys identified, or their methodological force.. My guess is that, without crediting it, Eddington drew on Jeffreys paper in the Mathematical Theory of Relativity, and so the expansion of the metric field in parameters associated with tests has been given Eddington’s name, not Jeffreys, as it should have been.
Clark: Thanks for your comment. True, though he claimed to have had a very low prior for Einstein’s theory,when it really came down to it, the appraisal of the evidence was not probabilistic, but rather, along the lines you mention. Of course he was wrong to suppose that the only solution that satisfied the key results was Einstein’s….at least not then.
I don’t have evidence on what he is most remembered for within geophysics, but this is a little exaggerated. First, his contributions to seismology and the use of seismological evidence to infer what is going on subsurface have enduring effects. Second, there were many grounds for being sceptical about continental drift until plate tectonics emerged in the 1960s and many strong sceptics among both geophysicists and geologists. Jeffreys’ prestige was not decisive; even giants rarely hold up entire sciences single-handed for decades. It’s easy to see in retrospect that the doubts were based on misconceptions, but that’s a repeated story in the history of science.
However, it is interesting to note that when Runcorn consulted Fisher on paleomagnetism and the analysis of the data of Hospers (Runcorn’s student) it was with the idea that he could disprove Wegener’s theory Fisher said to him “You will end up by proving continental drift.” (See Joan Fisher Box’s biography P 440) See also Frankel’s book
http://books.google.co.uk/books?id=eYZWLwAYt4kC&pg=PA29&lpg=PA29&dq=Runcorn+Hospers&source=bl&ots=842zvpS2St&sig=rFQ_pc4XRUrDhtnO6ph0vYnYaFw&hl=en&sa=X&ei=gglcUu3JO4zNsgarmoCoAw&ved=0CDEQ6AEwAQ#v=onepage&q=Runcorn%20Hospers&f=false
apkmedium.com