Skip to main content
SearchLoginLogin or Signup

On “A conversation with Sir David Cox”

Published onApr 27, 2023
On “A conversation with Sir David Cox”
·
key-enterThis Pub is a Commentary on

In 1993 I had the honor and pleasure not only of spending connected time with David Cox, but being able to, almost required to, ask him nearly anything. The conversation was written up for the interview series in Statistical Science (Reid, 1994). It was referenced and even perhaps read anew after David’s passing, which made me uneasy, for reasons I’ll try to describe. This also led me to ponder more about the community of statistical science and our oral and written history.

In 1992, Rob Kass became the third editor of Statistical Science, and I became an associate editor. Although the journal was just 6 years old, it had already established itself as a high-quality review journal under the leadership of Morrie DeGroot and then Carl Morris. Each issue included an interview with one or two well-known statisticians. As I recall, Rob and I thought that while the interviews were very popular, their content tended to the ‘lightweight’ side. In our youthful arrogance we agreed that a more scholarly approach to interviews would be valuable. I don’t think we were quite sure what we meant by that, but we had some exchanges around the notion of more emphasis on a ‘history of ideas,’ and less on the personal history of the interviewee. In this vein I approached David with a request to interview him, aware that he had previously fended off similar requests. In the email I wrote “would you be as reluctant if the interview … simply concentrated on some of your most influential ideas?” and I attached a list of things I promised not to include, such as “photographs of a vaguely familiar youth in short trousers” or “admiring ‘ahs’ about how busy you are…”

I suggested that we structure the interview around a set of seven statistical topics and I sent some brief questions on these in advance. We spent a memorable (for me) 2 days at Nuffield College in the warden’s office, talking for several hours each day. The interview was recorded on cassette tape and transcribed at the Statistical Science office at Carnegie-Mellon University. And then I really got to work—it is the most-revised paper I have written. It went through six drafts, the last two incorporating helpful comments from Mary Thompson, Rob Kass, and Peter McCullagh. With trepidation, a nearly final version was sent to David; his relatively minor comments were incorporated into the published version. I didn’t change what he had said, beyond removing the ‘ahs’ and ‘hmms’ and a certain amount of repetition. I did reorganize the material substantially, as our conversation had rambled around the various topics in a somewhat incoherent way. When it was published, some colleagues who knew him commented that I had captured him well, although I myself find that many interviews in Statistical Science seem to capture their subject well, through some alchemy of the interview process.

After David’s passing, I had several exchanges with a science journalist from The Wall Street Journal (Hagerty, 2022). He had done his homework thoroughly, and asked me many questions related to the material in the interview. David’s obituary in The Times (2022) also included some of the text verbatim. A few things flying around Twitter also linked to the interview, and the net effect of this was to make me very uneasy, because it seemed that in the absence of anything more definitive, the interview was being interpreted as an authoritative account of the man, and I knew how far short it fell of anything like that.

When we recorded the conversation in October 1993 David was 69, a widely respected giant in the field, still one or two years from retirement. I tried against all my nature to become a ‘nosy journalist’ for the occasion, and a few personal details, especially about his early years as an academic, worked their way in. But my focus was on learning as much as I could, on behalf of the community, about David’s current thinking on matters statistical—I had no real thought to any potential future life of the article. It was simply a snapshot, albeit a privileged one, at an interesting moment in time. I was so enamored of the notion of a ‘conversation of ideas’ that I overlooked many aspects that would have interested readers, and would still: a friend later commented “[the interview] is notable for the lack of personal details. …  I have no idea what he read, looked at or listened to.” It is my hope that a proper biographical account may appear in the future, by a proper historian and writer.

Interviews in print might in 2023 seem unnecessary—we have easy access to recorded lectures, interviews, and podcasts, something that was unforeseen 30 years ago. A forward-looking exception is the American Statistical Association (ASA) series of video interviews launched by Harry Posten at the University of Connecticut around 1990, which continues today. As valuable as this resource is, when I look at the ASA video interview with David from 1994, it strikes me as quaint, partly due to the technology having been superseded, partly because the novelty of such a format at the time made the interviewers and interviewee slightly nervous, but I think mainly because it is, by its nature, a largely unedited stream. In contrast, a written interview can perhaps strike a better balance between capturing the voice of the subject and providing a more thoughtful overview of key themes.

At an event in London in November 2022 to celebrate “Fifty years of the Cox model,” it was noted that David had “good taste in friends.” And indeed, the self-selected audience was an exceptionally congenial group, happy to honor David and reminisce about his place in their lives, and appreciative of the work presented. That got me thinking about David’s role in the community of science in a slightly different way. He was famously generous with his time and ideas, agreeing to meet people almost nonstop, especially while traveling. I recall an arrangement for him to meet one of my colleagues at the airport, en route to his flight home. He welcomed a steady stream of visitors to Oxford, sent comments on vast numbers of papers, and more. Farewell’s (2022) obituary has a lovely snapshot of this. But more than that, he created a community of colleagues around the world who followed his lead, I think, and tried to pay forward his generosity and keen interest in all matters statistical. By virtue of his influence and leadership, I think he was instrumental in increasing the civility of our academic exchanges. And in doing so perhaps helped to make the community of statistical scientists happier and more generous.

A quote from our 1994 conversation was used for the cover of the April 2022 issue of Significance that featured a tribute to his work:

In a sense, the only thing that matters is if you can look back when you have reached a vast, vast, vast age and say, “Have I done something reasonably in accord with my capability?” If you can say yes, okay. My feeling is that in one sense, I’ve done that…

Neither of us at that time could have predicted that David had some 29 years of productive working life ahead of him; that papers 226–385 were still to be written. The quote above went on:

… in the tangible sense of books and papers, I’ve done more than I would have expected. In another sense I feel very dissatisfied: there are all sorts of problems that I nearly solved, and gave up, or errors of judgment in doing a little something and not taking it far enough…”

This seems to me to capture the essence of an academic life, well lived.


Disclosure Statement

Nancy Reid has no financial or non-financial disclosures to share for this article.


References

Farewell, V. T. (2022). Obituary: Sir David Cox. Pharmaceutical Statistics, 21(2), 507–511. https://onlinelibrary.wiley.com/doi/10.1002/pst.2201

Hagerty, J. R. (2022, January 28). British Statistician won global acclaim for his methods. The Wall Street Journal . https://www.wsj.com/articles/british-statistician-won-global-acclaim-for-his-methods-11643382040

Reid, N. (1994). A conversation with Sir David Cox. Statistical Science, 9(3), 439–455. https://www.jstor.org/stable/2246357#metadata_info_tab_contents

The Times. (2022, January 29). Professor Sir David Cox. https://www.thetimes.co.uk/article/professor-sir-david-cox-7skkzp3bn


©2023 Nancy Reid. This article is licensed under a Creative Commons Attribution (CC BY 4.0) International license, except where otherwise indicated with respect to particular material included in the article.

Connections
1 of 19
Another Commentary on Remembering David Cox
Another Commentary on Remembering David Cox
Another Commentary on Remembering David Cox
Comments
0
comment
No comments here
Why not start the discussion?