University of California • Berkeley

Regional Oral History Office University of California

The Bancroft Library Berkeley, California

David Blackwell

AN ORAL HISTORY WITH DAVID BLACKWELL

Interviews conducted by

Nadine Wilmot

in 2002 and 2003

Since 1954 the Regional Oral History Office has been interviewing leading participants in or well-placed witnesses to major events in the development of Northern California, the West, and the nation. Oral History is a method of collecting historical information through tape-recorded interviews between a narrator with firsthand knowledge of historically significant events and a well-informed interviewer, with the goal of preserving substantive additions to the historical record. The tape recording is transcribed, lightly edited for continuity and clarity, and reviewed by the interviewee. The corrected manuscript is indexed, bound with photographs and illustrative materials, and placed in The Bancroft Library at the University of California, Berkeley, and in other research collections for scholarly use. Because it is primary material, oral history is not intended to present the final, verified, or complete narrative of events. It is a spoken account, offered by the interviewee in response to questioning, and as such it is reflective, partisan, deeply involved, and irreplaceable.

*********************************

All uses of this manuscript are covered by a legal agreement between The Regents of the University of California and David Blackwell, dated April 11, 2002. The manuscript is thereby made available for research purposes. All literary rights in the manuscript, including the right to publish, are reserved to The Bancroft Library of the University of California, Berkeley. No part of the manuscript may be quoted for publication without the written permission of the Director of The Bancroft Library of the University of California, Berkeley.

Requests for permission to quote for publication should be addressed to the Regional Oral History Office, The Bancroft Library, Mail Code 6000, University of California, Berkeley, 94720-6000, and should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

David Blackwell, "An Oral History with David Blackwell," conducted by Nadine Wilmot in 2002 and 2003, Regional Oral History Office, The Bancroft Library, University of California, Berkeley, 2003.

Copy No. J-

David Blackwell, 1962

TABLE OF CONTENTS— DAVID BLACKWELL

Introduction by Albert H. Bowker i

Interview History iii

Interview 1 : April 1 1 , 2002 1

[Minidisc 1] 1

Early years and family in Centralia, Illinois 1

College years at University of Illinois, Champaign-Urbana 8

Interview 2: April 18, 2002 17

[Minidisc 2] 17

More on early years in Centralia and undergraduate education 17

Graduate school with dissertation advisor Joseph Doob 22

Obtaining a Rosenwald fellowship and going to the Institute for Advanced Study 25

What is the meaning of denied access? 33

Looking for work 34

Teaching at Southern University and Clark Atlanta University 36

Interview 3: April 25, 2002 41

[Minidisc 3] 41

More on the Institute for Advanced Study 41

Howard University 43

Teaching and mathematics department faculty 46

Marriage and family 53

Interview 4: May 2, 2002 57

[Minidisc 4] 57

Howard University, curriculm and environment 57

Meeting Abe Girshick 62

Community of statisticians 66

Rao-Blackwell theorem 67

Bayesian approach 69

Summers at the Rand Corporation 72

More on Centralia and family 145

Northtown and Southtown 147

Grandchildren 149

Civil Rights Movement 150

Interview 8: September 1 1 , 2002 1 53

[Minidisc 8] 153

Affirmative action at UC Berkeley 153

Howard versus UC Berkeley, a teaching institution versus a research institution 154

The roots of being anti-war 155

How having a family and children impacted professional choices 156

Interview 9: November 13, 2003, and Interview 10: January 29, 2003 inserted in text

[Minidisc 9]

INTRODUCTION— David Blackwell

It is a pleasure to be asked to write an introduction to David Blackwell 's oral history. I suppose I owe this invitation to the fact that I have known him longer than anyone else in the Berkeley community. During World War II, I worked with Abe Girshick in a military research group, and it was Abe who had interested David in mathematical statistics after his graduate work with J. L. Doob and his early papers in probability. Through Girshick I came to know of David.

When we met in 1946, David was living in Washington D.C., where my parents also lived. My frequent visits to the city often included a meal at the Blackwell home, and I still remember in those early days driving with him from Washington to an Institute of Mathematical Statistics meeting at Cornell in 1946.

In January of 1947 I accepted a position at Stanford, where I started to organize a separate department of statistics. (The mathematics department at Stanford, unlike the one in Berkeley, did not want to include statistics.) Girshick joined the department, and David became a frequent short-term visitor and also spent the 1950-1951 academic year in residence. During the 1950s he spent several summers at the Rand Corporation, and became acquainted with Kenneth Arrow, another early star in statistics at Stanford.

His Berkeley connection was established when, in 1954, he came to Berkeley from Howard University as a visiting professor. The following year he accepted a position as Professor of Statistics. I had become chancellor at Berkeley in 1971, and this was a wonderful opportunity to renew our friendship.

David has been an ideal faculty member. First and foremost, he is an excellent classroom performer. His classes have been both rigorous and popular, and he is also listed as the supervisor of nearly fifty PhD dissertations. He has done his share of administrative and committee work: a term as department chair, Director of California Study Abroad Center, Faculty Research Lecturer and chair of the committee to select others, and member of the committee that recommended my own appointment as chancellor.

David's recognition is not limited to his work as a teacher. He has done fundamental work in game theory, Bayesian inference, and information theory. He was a principal developer of Bellman's dynamic programming ideas. He founded the theory of comparison of experiments, later developed by his colleague Lucien Le Cam to become one of the central tenets of mathematical statistics. With Arrow and Girshick he developed the backward induction method of solving sequential decision problems to solve the problem of determining Bayes solutions to such problems. He is the author of the classic book, Theory of Games and Statistical Decisions, along with some eighty papers in professional journals.

Although I do not want to wander too far from his academic and scholarly contributions, I cannot refrain from mentioning at this point that in many ways in his career was deeply influenced by the fact that he is an African American who had to overcome great educational and professional obstacles. He grew up in modest circumstances in Centralia, Illinois, a home town of railroad employees. It is to the credit of his family and his teachers that they recognized his remarkable ability and encouraged him to seek higher education. He was accepted by the University of Illinois, although there was some reluctance to have him pursue advanced study there. After obtaining his doctorate and spending a year at the Institute for Advanced Study, he applied to traditionally black universities, teaching first at Southern University and Clark, before going to Howard and only much later to Berkeley. In fact, the first proposal to hire him at Berkeley was defeated on racist grounds.

Despite the limitations that some tried to place on him, David has constantly forged ahead. I note that we are the same age, and he completed his PhD in 1941, the year that I graduated from college with all the

11

privileges and encouragement that I accepted as a matter of course! Recognition has come to him over the years. He has served as president of the Institute of Mathematical Statistics, the Bernoulli Society, and the International Association for Statistics in the Physical Sciences. He has given a number of prestigious named lectures, and has a dozen honorary degrees. In 1965 he became the first African American to be elected to the National Academy of Sciences and was elected to the Academy of Arts and Sciences in 1968.

In recent years universities have actively sought diversity in faculty ranks, and being an African American is not as much of a disadvantage as it once was. But David's remarkable career was established at a time when prejudice ruled in academia. We salute him for his persistence and human courage as well as for his great scholarly achievements.

Albert H. Bowker November 2003

Ill

INTERVIEW HISTORY— David Blackwell

David Blackwell was born in 1919 in Centralia, Illinois. He went on to become a great mathematical thinker and a professor who is known for his dedication to teaching. Blackwell has made fundamental contributions to the areas of probability theory, mathematical statistics, set theory and logic, and of course, game theory, to name a few. His numerous achievements, honors, and affiliations are well documented, so I won't do that here. I would, however, refer the reader to the preface and body of Statistics, Probability, and Game Theory: Papers in Honor of David Blackwell (1996, Institute of Mathematical Sciences, Lecture Notes-Monograph Series, Volume 30) created by his friends and colleagues, Thomas S. Ferguson, Lloyd S. Shapley, and James B. McQueen. This volume is excellent for a concise overview of his professional contributions and some insight into his intellectual milieu and continuum for the layperson. There are also two excellent short interviews, one in Mathematical People — Profiles and Interviews (1985, pages 17-32, eds. Donald J. Albers and G.L. Alexanderson, Contemporary Books, Inc.) and the other in the very first issue of Statistical Science (1986, vol. 1, pages 40-53) that were my first points of reference in preparing for this interview.

Who are you calling a first? When Dr. Blackwell came to UC Berkeley in 1954 after a decade at Howard University in Washington D.C., he became, we think, the first African American ladder rank faculty person systemwide. What does it mean to be "first," as David Blackwell was in so many arenas? As Dr. Blackwell himself says in his interview, "You have to be careful when you talk about firsts." This interview sheds some light on this much lauded but often mysterious experience.

It must be said here that this interview is a conversation between a theoretical mathematician and someone with little background or training in math and statistics. When I suggested to Dr. Blackwell at the outset that we bring in a second co-interviewer who was more fluent in game theory and mathematical statistics than I, he declined. This makes for an exchange between us that is sometimes productive and sometimes rudimentary, peppered with my almost comical misunderstandings when it comes to the subjects of game theory and statistics. I think, though, that Dr. Blackwell and I did the best we could given this barrier.

The interview began in spring of 2002 and concluded in January of 2003. In total, there were seven full interview sessions in addition to a series of three smaller recorded follow up conversations. The follow up conversations have been appended as a follow up interview and integrated into the body of the interview as noted. All interviews took place in Dr. Blackwell 's office on the third floor of Evans Hall on UC Berkeley's campus. This interview was recorded on minidisc. Dr. Blackwell declined my proposal to video tape the interview. The transcript was reviewed by Dr. Blackwell for accuracy of names and dates.

David Blackwell's interview is the first in the African American Faculty and Senior Staff Oral History Project series, which explores the experiences of African American faculty and senior staff at UC Berkeley as part of the broader history of the University of California and its commitment to access and diversity. This series is grounded in the premise that higher education is one of the primary strategies for gaining social equality — access to employment and income — for historically disadvantaged communities. Moreover, the university, comprised of its students and faculty and administration, with all of its intellectual and financial resources operates as a critical touchstone in processes of systemic social change. Therefore, the university functions not simply as an educational institution, but also as a significant site of historic and future potential for imagining and crafting opportunity for ethnic and racial groups formerly excluded from higher education. This project recognizes that the University of California, as California's premier public educational institution, plays a significant role hi the socio-economic mobility of all of California's residents. The story that we hope will emerge from this project is a story of California — it's people and one of its most important public institutions.

IV

This interview was conducted under the auspices of the Regional Oral History Office. The Regional Oral History Office was established in 1954 to augment through recorded oral memoirs the library's materials on the history of California and the West. Copies of all interviews are available for research use in the Bancroft Library and in the UCLA Department of Special Collections. The office is under the direction of Richard Candida Smith, Director, and the administrative direction of Charles B. Faulhaber, James D. Hart Director of the Bancroft Library, University of California, Berkeley.

Nadine Wilmot, Interviewer/Editor Regional Oral History Office Berkeley, California December 2003

Regional Oral History Office Room 486 The Bancroft Library

University of California Berkeley, California 94720

BIOGRAPHICAL INFORMATION (Please write clearly. Use black ink.)

Your full name

Date of birth Father's full name

Occupation PCu&sKrAel k 0

, M( °> Birthplace

Birthplace

Mother's full name

Occupation /f AU K tAA-A

— —

Birthplace

Your spouse/partner /)> M H fM fi £** Occupation

<XA

Your children

^ }Dt)\fi<D " , fltti,, fa

Where did you grow up? Present community

Education A /?, /4/frl; IA- /^.

--

Occupation(s)

* A 1

o ' tl

Birthplace /f />tW>T/; /

Areas of expertise

I\A

M. n

Other interests or activities

Organizations in which you are active_

SIGNATURE

DATE:

: l/

INTERVIEW 1: APRIL 11, 2002

[Minidisc 1]

Wilmot: Well, to begin, I was looking at this photo from when you were a very young person when you were in school.

Blackwell: Yes.

Wilmot: How old were you there?

Blackwell: Maybe eleven.

Wilmot: You were in sixth grade.

Blackwell: I'm not sure.

Wilmot: I'm looking at the expression on your face in that photo; it's just this wonderful expression. So, I guess to begin, I wanted to ask you, where were you born?

Blackwell: I was born in Centralia, Illinois. That's a small town in southern Illinois, about 12,000 people.

Wilmot: When were you born?

Blackwell: I was born in 1919. 1 grew up in Centralia. All my elementary and high school schooling was in Centralia.

Wilmot: What kind of a place was Centralia?

Blackwell: Most of the people worked for the Illinois Central Railroad. I believe, in fact, Centralia was named because it's on the Illinois Central Railroad. It's a junction point for the railroad; and my father and the fathers of most of my boyhood friends all worked for the Illinois Central Railroad. It really was a railroad town.

Wilmot: Can you tell me your parents' names?

Blackwell: My father's name was Grover Blackwell. My mother's name was Mabel Johnson Blackwell.

Wilmot: What did your father do for the railroad?

Blackwell: He was what's a called a hostler. He took the engines from the engineer and put them into a roundhouse. Then he took them out of the roundhouse and delivered them to the engineer. I believe he had that job since he was eighteen years old, all his working life. He loved it.

Wilmot: Did you ever go to his place of work? Was it a place where young people would go? Blackwell: Very often!

Wilmot: Yes?

Blackwell: Yes, he used to take me on rides in the steam locomotive, and even let me pull the throttle once or twice.

Wilmot: Wow! That must have been so exciting. Blackwell: Yes.

Wilmot: Because that's like a fantasy for a young boy and then suddenly it's reality.

Blackwell: I still have a great affection for steam locomotives. So for me, they're just big and

beautiful and wonderful. You know they're noisy, and they're smelly, and they shake the ground — oh boy, it's an impressive thing to watch a steam locomotive.

Wilmot: Yes, I agree. Now can you tell me what your mom did?

Blackwell: She didn't have any outside activities, she was a housewife. She raised me and my two brothers and my sister, and looked after the house.

Wilmot: Wow, sounds like a big job in and of itself.

Blackwell: Well, she had a couple of rental properties that she managed. She was pretty active.

Wilmot: Were your parents active in any organizations?

Blackwell: My mother was active in a church. My father wasn't active in any organizations.

Wilmot: Did he have any professional affiliations as a hostler, or—?

Blackwell: I don't think so. No, none that I remember. He liked his job so much that when the working day was reduced from seven days a week to six days a week, he used to go down to the shop on his day off just to make sure that his replacement was taking care of those steam locomotives.

Wilmot: Yes, making sure they were okay.

Blackwell: [laughing] That's right!

Wilmot: It sounded like he almost viewed them like they were, like, alive.

Blackwell: Oh yeah!

Wilmot: He took really good care of them.

Blackwell: Well, you know that each steam locomotive sounds different. My father and all the other railroad men could tell you the number of a steam locomotive just by listening to it. Sometimes, they could even tell you the name of the engineer who was driving it, because they would have different whistles and different sounds.

Wilmot: So when you say "your father and all the other railroad men," that was your friends' fathers as well?

Blackwell: Yeah.

Wilmot: Were they also hostlers?

Blackwell: No, there were only three hostlers. Each hostler worked an eight-hour shift, because there were only three hostlers. And a fourth man who would replace one of them if he was absent.

Wilmot: Sounds like a very finely tuned system.

Blackwell: Yes.

Wilmot: Did your family then know the other two hostlers?

Blackwell: Oh, very well! Yes.

Wilmot: Did they have children your age that you remember?

Blackwell: No.

Wilmot: The house you grew up in, were you in the same house from the time that you were very small until the time that you went away to school?

Blackwell: Yes. The house I was bom in my parents continued to live in, until they died, always the same house.

Wilmot: What was that house like?

Blackwell: Oh, just an ordinary two-story house on what seems to me now like a lot of land. Our house extended from one street to the next street. So it was, maybe, on a lot, maybe, two hundred and fifty feet deep by, maybe, a hundred feet wide or something like that. So we had quite a large lot. But the house, as I say, was just an ordinary frame house.

Wilmot: Were there lots of trees and gardens on that lot?

Blackwell: Oh yeah, a tremendous number of trees. My mother's family owned the entire square block in which I lived. Now, that sounds a lot more impressive than it was because they were not by any means well off. But they just happened to own that whole square block. Next to our house, in a lot, on one side was my grandmother's garden and on the other side of my grandmother's garden was my Uncle Fred's lot. And then, going the other way were a couple of rental houses that my mother owned, and then a lot of vacant space where there was a garden. My uncle grew corn there. Now, as I say, that sounds like a lot. But we were just poor, working people. But my mother had inherited that land.

Wilmot: Sounds like you were surrounded by family, which is really nice.

Blackwell: Oh yes! And across the street my another uncle and aunt lived, and my grandmother lived upstairs across the street. That whole area was sort of known as "Johnson territory." My mother's maiden name was Johnson. And I remember, once my wife and I were visiting there and she started walking around and she ran into a fellow. And he said, "I know you, you're one of them Johnsons." [laughter]

Wilmot: Funny. So your mom's family, the Johnsons, where did they come from?

Blackwell: Her father was born in Cleveland, Ohio. No, in Columbus, Ohio. Then he became a school teacher in Weakley County, Tennessee. My grandmother was one of his pupils there, and they got married. Then he moved to Desoto, Missouri, and then he moved to Centralia, Illinois, and founded a grocery store there. It was quite successful, apparently, so he acquired a fair amount of property, I'd say a couple of square blocks all together.

Wilmot: And your father's family?

Blackwell: I don't know much about my father's family, [pause] They came from Weakley County, Tennessee, but I don't know very much about them at all.

Wilmot: Okay, do you know the story of how your parents met and came to be together in Centralia?

Blackwell: I know what my mother said. She claimed that my father came up to her one Sunday after church and claimed to have met her before, [laughter] So that's how they met.

Wilmot: Sounds like it was effective.

Blackwell: Well, yes.

Wilmot: A very effective approach.

Blackwell: Probably it's been used many, many times. "Haven't I met you somewhere?"

Wilmot: Yes, for sure, it's a good one. So was the neighborhood that you lived in, was it primarily African American, or was it mixed?

Blackwell: No, it was quite mixed. I would say primarily white. But a fair number of African

Americans — well, you see from this picture, there aren't many black students in that picture. But the neighborhood was somewhat blacker than this picture would give you the impression of. I'd say 30 percent black.

Wilmot: Did you find that mixture was reflected in your playmates and your social life? Blackwell: No, my playmates were mostly black.

Wilmot: So, you went to school in Centralia public schools. And that's where, I assume, from what I've read, you really discovered that you love math. So I wanted to talk to you a little bit about what your education was like there, and who you remember there.

Blackwell:

Wilmot:

Blackwell:

Wilmot:

Blackwell:

Wilmot:

Blackwell:

[Wilmot: Blackwell:

Wilmot:

Blackwell:

Wilmot:

[pause] Oh, I just had a standard education. I got skipped a couple of times so I was in elementary school only six years instead of the usual eight. But it was in high school that I found something that I really liked. I really liked geometry. I had always been pretty good at math and somewhat interested in it, but geometry really excited me.

But I certainly never thought of making a career in mathematics. In fact, I had decided that I was going to be an elementary school teacher. The reason for that was rather simple. I grew up in the Depression, and getting a job was very important. When I was in high school, one of my father's friends promised me a teaching job if I would finish college. He was on the school board in Pulaski County in southern Illinois. So that just settled it for me right there. By the end of my freshman or sophomore year in high school, I already knew that I was going to go to college, and graduate, and then be an elementary school teacher.

Do you feel like the students you went to school with, your classmates, did they have similar aspirations?

Yes. In high school, two of the fellows who were in the same mathematics club with me, and with whom I worked quite a bit, went on to college. They both became electrical engineers. One of my black classmates — playmates — went on to college and he became a college history professor. Yes, there were — several of the boys that I associated with had aspirations to go to college. As I've said, there was just never any doubt that I was going to go to college.

And that was the attitude of your family as well? Oh yes.

From what I've read, it sounds like your family was very committed to your achieving at that level, to going to college.

There was just no question about it, that's right. I don't know how that happened. Even though neither of my parents had gone to college, they just assumed that their children were going to go to college.

Why do you think that was?

I don't know, but my parents' friends shared the attitude: they wanted their children to go to college, and many of them did.]

That was their commitment.

Yes.

Now, your father, he was educated to what level?

i

1 . Bracketed material, here and in the following pages, comes from a follow-up interview on November 13, 2002, recorded on minidisc 9 and inserted during the editing process.

6

Blackwell: Oh, only to about the fourth grade, [pause] He was not an educated man, at all. He read very little, for instance. My mother dropped out of high school after her sophomore year. But she was reasonably well educated for her day. I got many letters from her over the years, she never misspelled a word, for example. She — .

Wilmot: Well, there are some that say also that education was better. What you got in school was better than what you get now.

Blackwell: I think that's right, yes. Yes, it was certainly more rigorous in many ways.

Wilmot: So they both were like, you are going to school, definitely, no ifs, ands, or buts. Blackwell: That's right.

Wilmot: Can I go back to the experience you said of being skipped several times? Now, what grades did you skip?

Blackwell: I think I spent only half a year in each of the first four grades, as I remember it, it was sort of like that.

Wilmot: That's just a blur, you moved really quickly.

Blackwell: Yes, they were basing it on my reading skills. Somehow, I learned to read very early. And so, if I could read at a fourth grade level, they put me in the fourth grade.

Wilmot: So, you were always in school, then, with people that were older than you?

Blackwell: That's right. That's right, [pause] Yes, a year or two older, yes.

Wilmot: Wow.

Blackwell: I'd never thought of that before, but it was true! [laughing]

Wilmot: Sometimes that's what this is for. It gives you an opportunity to think about things you never thought about before, it's very funny. So, that must have been a very interesting experience for you. To just, you know, always be with people who were older than you but those were your peers and who you socialized with. Did you socialize with your — ?

Blackwell: Yes.

Wilmot: That must have been really interesting, [pause] I was wondering, since you mentioned that geometry — actually, I wanted to ask you a question about reading. Do you remember which books — if you remember any books that absolutely transported you when you were in that elementary school time. It may be that books that transport you don't come until a little later, but I'm wondering what you remember?

Blackwell: I remember the first book I ever read, and that was before I went to school. It was called The Gold Hunters, by James Oliver Curwood. It was a book about the far north, and traveling with sled dogs over the snow. And I remember the second book that I tried to read. It was, I believe, by a woman writer named Gene Stratton Porter. And it was a

novel about love and romance. And I didn't understand a bit of it! And I stopped reading it after the first few pages, [laughs] Because I was just not ready for that at all. But I always liked reading a lot. I read a lot of fairy stories when I was a kid. I read a lot of Tarzan books. I don't know that they influenced me especially, I just enjoyed reading them.

Wilmot: Yes, I mean, I think it's just when you're young that's an imaginary place that you go. Blackwell: Yes.

Wilmot: So you were reading before you went to school, so did you learn at home?

Blackwell: I learned to read in my uncle's grocery store by looking at seed packages. There was this big stand that had seed packages on it. There would be a picture of tomatoes and a sign that said tomatoes. Well, I knew what tomatoes were, so that must say tomatoes. So just by looking at pictures like that, I learned to read.

Wilmot: You learned what all the letters sounded like, and — ? Blackwell: Well, sure, if you get enough experiences like that you just sort of figure it out.

Wilmot: I bet seed package designers had no idea what they were doing with these useful

educational tools, so that's great. Did you spend a lot of time in your uncle's grocery store?

Blackwell: Oh, yes. Yes, as I say, that whole area sort of belonged to our family. So, I spent about as much tune across the street as I did at home.

Wilmot: So, you mentioned that geometry was the place where it really was exciting for you academically. Do you remember who your geometry teacher was?

Blackwell: Oh, certainly. Her name was Luther, Caroline Luther. She also taught German as well as geometry, and — she sort of kept up with me after I left high school. She invited me to come and visit her a couple of times after I was in college. And two of the other boys that I worked with in high school were also in that same geometry class, and she paid attention to them, too. There was another high school math teacher that influenced me a lot. His name was Huck, Raymond Huck. There was a mathematics magazine that was partly for high school mathematics students, and it had a problems section. Mr. Huck encouraged us to try to solve the problems there, and sometimes we did. I solved a couple of them, and Mr. Huck wrote up my solution and mailed it into the mathematics magazine. And a couple of times, my name appeared in the magazine and once even my solution — really Mr. Huck's write-up of my solution — appeared in the magazine. I was very pleased with that.

Wilmot: Really exciting, I mean, you were in high school! Blackwell: Yes.

Wilmot: Were there any other teachers that you can think of besides Mr. Huck and — ?

Blackwell: Oh, I can think of several other teachers, yes. There was an English teacher, Grace Seiler, who encouraged me to go to college. In fact, she wanted me to go to her alma mater, which was DePauw University in Greencastle, Indiana. But that was a little far away for me. I got a chance to go to [University of] Illinois, so I chose Illinois. She was very interested in me. I say, "they were interested in me;" they had a general interest in the students, not just me particularly. They kept up with students after they left.

Wilmot: In general, did you feel like you really had access to good education and exciting, caring teachers? It sounds like maybe you did.

Blackwell: I had an absolutely fabulous high school education. When I went to college, I was a

semester up on most of the college students there. There were things that they studied in college algebra I had already studied in high school. Things that they studied in freshman English I had already studied in high school. High school was harder for me than college!

Wilmot: Oh, my goodness. Do you remember the process of choosing, or applying, to college and thinking about the different places you could possibly go?

Blackwell: There was never any doubt in my mind, I wanted to go to the University of Illinois, and I intended to go there.

Wilmot: Did you apply anywhere else?

Blackwell: At the urging of this high school teacher, I applied to DePauw University in

Greencastle. But, you see, in those days, any graduate of a high school in the state of Illinois was eligible for admission to the University of Illinois. There wasn't the big hassles, all the "Can I get admitted?" that there is now. There was no question about it, if you were a high school graduate, you were in. So, application was just a matter of saying that I'd like to go.

Wilmot: Wow, that is different.

Blackwell: Of course a fair number of students flunked out, too. Wilmot: Of college or of high school?

Blackwell: Of college. In the fraternity that I was in, in college, about a third of the freshmen would flunk out. And that was not atypical.

Wilmot: So you applied and were accepted to the University of Champaign-Urbana, Illinois, and that's where you went. Do you remember what it was like arriving on that campus the first day? And did your parents come with you?

Blackwell: My parents did not come with me. Certainly, I remember very well what it was like. I came on the train, with suitcase, walked a mile and a half with my suitcase from the train station to the campus. I was heading for some university office where I was going to inquire about housing. A fellow saw me from across the street and said, "Hey, young fellow, come here, come here!" And I walked across the street, it turned out that he was a member of a black fraternity there and he invited me to come and live in that black

fraternity house. And, I did. So that was a big first day for me. That was an important encounter.

Wilmot: How far was your university from your home town?

Blackwell: A hundred miles.

Wilmot: And how long was that on the train?

Blackwell: Oh, three hours, something like that. Two or three hours.

Wilmot: What was that train ride like?

Blackwell: Don't you know what a train ride is like? It was no different from trains now. Well, of course, a steam locomotive is different than a diesel but — since my father worked for Illinois Central Railroad, I had a pass. So, I never had to pay anything to ride on the train.

Wilmot: So, you met this very important person on your first walk outside the train station with one suitcase in hand?

Blackwell: Yes.

Wilmot: Wow! And can I ask you, I've been wanting to ask you, this fraternity, which one was it?

Blackwell: Alpha Phi Alpha. Ever hear of that?

Wilmot: Yes.

Blackwell: Good. Yes, I lived in the Alpha house all the time that I was in school at Illinois.

Wilmot: So that's—

Blackwell: Six years.

Wilmot: And when did you pledge?

Blackwell: Right away.

Wilmot: Were any of your fraternity brothers also involved in math and statistics?

Blackwell: No. I was the only math major in my fraternity house.

Wilmot: Do you remember what areas your fraternity brothers were involved in academically?

Blackwell: Vaguely. Three of them were philosophy majors. You wouldn't find that nowadays. But, in those days, jobs were scarce. So, you didn't especially choose your major with reference to getting a job. You were more inclined to choose as your major something you'd like. Well, these guys chose philosophy, some chose physical education, some

10

chose pre-law. My closest friend chose bacteriology. I don't know what his original plans were in choosing bacteriology, but he actually went on to get a PhD in bacteriology. But when he started, he wasn't thinking of getting a PhD any more than I was thinking of getting a PhD.

Wihnot: What was his name?

Blackwell: Arthur Webb. He was from Chicago. Practically all the people in my fraternity house were from Chicago.

Wihnot: So maybe you were the only one from a small town a hundred miles away? Blackwell: That's right, yes.

Wihnot: Can you tell me, as an organization, your fraternity — what kinds of things did you do as an organization on campus?

Blackwell: Mostly we supported each other. For instance, there was a big file of essays in our

fraternity house on various subjects. And each essay had a history. Some of them were "A" essays, and some of them were "B" essays — [laughter] That is, essays that had been graded "A," some graded "B," and some graded "C." And students would go and pick one of those essays and copy it if it was appropriate for their assignment. If you'd been doing "C" work, you didn't pick an "A" essay, you picked a "B" essay.

Wihnot: Yes, so you supported each other.

Blackwell: Yes, and we supported each other with jobs. We, the Alphas, owned a certain sorority, the Pi Beta Phi sorority in this sense; all the waiters in the Pi Beta Phi sorority were Alphas. And when one Alpha would graduate, he would turn over his job as a waiter in the Pi Phi house to another Alpha. And the Kappas, another black fraternity, had a fraternity that they "owned" in that sense. They had all the waiting jobs in that fraternity, and passed them down from one to another. And there were various other jobs that got passed down that way. So, hi that sense, the Alphas helped each other through school.

Wihnot: Do you know how now they have work-study jobs, work-study jobs for people who come to college?

Blackwell: Yes.

Wihnot: At that time, were there also positions available on campus, too?

Blackwell: Yes, and that was very important for me and many of the other students. It was called NYA, National Youth Authority. That was the equivalent for college students of the WPA, the Works Progress Administration. These were all Roosevelt's ideas for helping people through the Depression. That made a big difference for me and many other students.

Wilmot: Do you remember what kinds of jobs you held during that time?

11

Blackwell: Certainly. MyNYAjob was working in an entomology laboratory. And the one thing I had to do when there was nothing else to do was go through all the vials of bugs and keep them filled with alcohol. The bugs were preserved in alcohol, but the alcohol would evaporate, so it was my job to go through and fill those vials with alcohol. Some of the NYA students would drink some of the alcohol, [chuckles]

Wilmot: That's very high-proof alcohol.

Blackwell: I think so, yes, I think it was 1 80 proof or something like that

Wilmot: Yes, that's in almost medicinal levels, that's not for internal consumption.

Blackwell: Well.

Wilmot: Oh my goodness. Well, do you remember, were other members of your fraternity also employed through the NYA jobs?

Blackwell: Oh yes. Yes, NYA helped a lot of us get through college. Wilmot: And socially, what kinds of things did your fraternity do?

Blackwell: Oh. Our fraternity sponsored a — used to give a dance or a party a couple of times a

year, and we would play basketball against the Kappas once a year. But that was about it.

Wilmot: Did you continue to act as a welcoming committee for new freshmen?

Blackwell: Oh, certainly. Every fall we would go down and meet the train coming from Chicago and try to spot incoming freshmen who weren't sure where they were going to live. And we would try to get them to come and live at the Alpha house and the Kappas would try to come and get them to live at the Kappa house.

Wilmot: So what was that like? How did the Kappas and the Alphas try to bring people into the organization?

Blackwell: Well, as I say, you would go down and meet the trains. And if a black fellow was getting off and looked bewildered, you'd go up to him. "May I help you?"

Wilmot: Yes, so it was a very critical first moment depending if he met a Kappa or an Alpha first.

Blackwell: That's right. Of course, most of them were from Chicago and they already knew where they were going.

Wilmot: So, did the Kappas and the Alphas work together around different projects? Blackwell: I don't remember anything that we did together.

Wilmot: Well, you were there six years and you stayed in the Alpha house for six years. Were you a leader in the fraternity?

12

Blackwell: Not especially. I was in the Alpha house for six years, but it couldn't stay the same house all that time. We were there for three years, and then we were forced out and didn't have a house for one year, and then we got a new house for the last two years or something like that.

[Wilmot: How did this happen?

Blackwell: Our lease expired. We were leasing only the second floor of a two-story building, and the owner found a new tenant for the entire building.]

Wilmot: Okay. So, you arrive on campus and you find a community and a place to live all at once. And you started going to classes?

Blackwell: Well, yeah!

Wilmot: And how was that? How were your classes when you first started there?

Blackwell: Well, I remember one significant thing that happened. You had to take a foreign

language, and I had studied Spanish in high school and I planned to continue that. But one of my fraternity brothers told me — this was within two or three days after I got there — that I should take German instead of Spanish. Because, he said, "If you're pretty good at mathematics, you may want to go on for a PhD. And Spanish won't help you, but you'll be required to read German, so you should take German." That was extraordinarily good advice from just one of my fraternity brothers who was two or three years ahead of me and could look that far ahead.

[Wilmot: German is important for math?

Blackwell: It was very important in those days for mathematics. You had to pass exams in French and German in the PhD, and many of the books and papers you studied were in German.]

Wilmot: So he helped you make some good choices? Blackwell: He certainly helped me make that very good choice, yes.

Wilmot: So you went to college and you were planning on being an elementary school teacher. How did you choose your course load? Did you choose education classes or — ?

Blackwell: No, education courses had a bad reputation in those days. They were joke courses and — well, you put off taking them as long as you could. So I didn't say, "I'm not going to take them," I just kept postponing it. And I never did get around to it. [chuckles] No, I knew that in order to be an elementary school teacher I would have to take them. And, as I said, I planned to take them, but not right now.

Wilmot: And instead, you started getting into mathematics and statistics courses?

Blackwell: I took every undergraduate mathematics course that was offered. Just because you had to major in something, mathematics was easy, I liked it — so that was that.

13

Wilmot: And did you take any of your literature classes while you were in undergraduate school? Blackwell: Oh, yes, I took whatever you were required to take.

Wilmot: Now, you're here in college with college being easier for you than high school. And

what was that like, being surrounded by people who were a little bit older than you but who were challenged academically by something that was, for you, very easy?

Blackwell: [pause] I don't remember very much about that, especially. Academically, I spent more time with the other math majors than I did with my fraternity brothers.

Wilmot: Okay. Blackwell: And, of course, I was challenged just as much as they were.

Wilmot: Okay, my question just came from when we were talking earlier about high school and about how you found college to be easier and I was just coming from that perspective. But, okay, that kind of gets into the question of socially, what was college like for you? Like, were you one of the people who was really socializing? Or you were, were you more academically focused?

Blackwell: Well, I was almost completely academically focused and economically focused. I had to make a living! I had to — after my freshman year, I got practically no support from my parents. I found out that my father had been borrowing money to send me to college. So, I told him that I could manage, and I did. But that meant that supporting myself was an important — a big part of my activity.

Wilmot: Did you do that primarily through the entomology laboratory job?

Blackwell: Primarily through that, and meal jobs — washing dishes and waiting tables at the Pi Phi house.

Wilmot: Did the institution help you at all, did your university help you at all, financially? Blackwell: Well, of course the entomology lab job was a university job. Wilmot: Were you definitely responsible for paying your full tuition?

Blackwell: Oh! I went there on a scholarship. And the scholarship was thirty-five dollars a

semester. That was the tuition. It sounds absurdly low, but it was something in those days.

Wilmot: How did you win that scholarship?

Blackwell: By taking an examination. The state of Illinois, at that time, offered one scholarship in each county. There were 102 counties in Illinois. You took an examination and whoever scored the highest on that examination got the scholarship. So I won the Marion County scholarship that year in high school.

14

Wilmot: Yes, that's an important thing. And that reduced tuition to thirty-five dollars per semester.

Blackwell: That paid the tuition which was thirty-five dollars a semester.

Wilmot: Okay, but you were responsible for all of your — . Blackwell: Living expenses and textbooks and so on, sure.

Wilmot: So that was you, okay. When you were working with the entomology department, what kind of job was that? Was that a place you felt like you were learning anything while you were there? Or was it just work?

Blackwell: No! Just work, friendly atmosphere, but just work.

Wilmot: Also, while you were washing dishes, was that an experience where you worked with other people who you in a friendly atmosphere?

Blackwell: I worked in a friendly atmosphere, with my fraternity brothers — . And working for the woman who ran the kitchen.

Wilmot: So let's move more into the academic area. Blackwell: Okay.

Wilmot: If you can tell me what it was like encountering this community of mathematicians. You said that you mostly were, kind of, moving in a math circle at that point, socially. So, if you can talk a little bit about who your peers were.

Blackwell: Well, it's not easy to remember. There was one fellow named Abe Charnes that I

learned a lot from. We took several classes together. But, somehow, he always knew more than I did. So, I learned a lot from talking to him. It's all sort of vague in my mind. There was a room for undergraduates in the mathematics library. And we used to go into that room and discuss problems with each other. But, as I say, I have the idea that I learned more from them than they did from me. Maybe they'd say that they learned something from me. But I know that I learned a lot from them.

Wilmot: You mean from the other undergraduates? Blackwell: The other undergraduates, yes. [pause]

I remember that — this is going to be a little technical — Abe showed me how to negate a mathematical statement by changing every "for every" sign to a "there exists" sign, and every "there exists" sign to a "for every" sign and then negating the inner proposition. I had been doing that informally, but sometimes, in a complicated thing I would make a mistake. But he showed me a way to do it where you never made any mistakes. This general rule — and various things like that I learned from him. I saw him a few years ago, by the way, he is a retired mathematics professor at the University of Texas, I think it is.

15

Wilmot: So he went on to be a professor?

Blackwell: Yes. Also he, as an undergraduate student, had started accumulating a mathematics library. And that really impressed me.

Wilmot: Did you go and start building your own as well?

Blackwell: In a weak sort of way, but I sort of remembered that. As I say, though, at that time, I hadn't really determined that mathematics was my career. It's just that I liked it.

Wilmot: If you were to name a turning point, then, where you realized that you were going to get your PhD?

Blackwell: Let me tell you sort of what happened. I had a four-year scholarship. I completed my undergraduate work in three years. So I decided to use that fourth year of my scholarship to go on and get a master's degree. Then, while I was working on that master's degree, I was encouraged to apply for a fellowship or a teaching assistantship to go on for a PhD. So my focus was gradually shifting and I did apply. And I got a fellowship. So then, that settled it. Then, I was going to go on for a PhD.

But let me tell you something very funny that happened. See, there were two kinds of awards, fellowships and teaching assistantships. They paid the same amount of money, but for a fellowship you didn't have to do any teaching. So they were the preferred award. And there were maybe three fellowships and twenty teaching assistantships every year. But you submitted a single application. We all submitted applications. One of the other graduate students told me that I was going to get one of the fellowships. I said, "How do you know that?" He said, "Well, you're good enough to be supported and they're not going to put you in a classroom!" Because I was black, of course. He was right, sure enough I did get one of the three fellowships. And I'm sure that a partial consideration was, "Well, we need to support this fellow, and we can't put him in a classroom, so let's give him a fellowship."

Wilmot: Whoa. Blackwell: So it was an actual advantage to be black.

Wilmot: Yes, I also wonder, once you had a fellowship, did that change your life on campus in terms of working to support yourself economically or was it still very much an onward struggle?

Blackwell: Yes, after I got the fellowship I didn't have to work any more. Yes, that's right.

Wilmot: Okay, so that must have just left a lot more room for the intellectual life on campus? Blackwell: Somewhat more, yes. Yes, at that time, of course, all my work was in mathematics.

Wilmot: I guess I want to ask you some questions about the institution. Since you were there for six years, I want to know more about the University of Illinois, at Champaign-Urbana.

Blackwell: Okay.

16

Wilmot: Was it a large student body and was it diverse?

Blackwell: Would you say that again?

Wilmot: Sure, I did that thing where I talk too softly, huh?

Blackwell: Yes.

Wilmot: First I wanted to ask you if this was a large student body?

Blackwell: Oh, I believe about 12,000, which is small compared to UC Berkeley.

Wilmot: Yes, but maybe it seemed larger, larger to you than where you'd come from.

Blackwell: Oh, much larger, yes, certainly.

Wilmot: And was this a — if you remember — was this a diverse student body? Was it primarily white, or were there people from other backgrounds there as well?

Blackwell: Well, it was overwhelmingly white. There were about a hundred black students out of 12,000. So, yes. All the black students knew each other; it was that small. And there were rather few foreign students or rather few non-white students. Yes, it was just overwhelmingly white.

Wilmot: And as far as your instructors and professors and also the administration, did you have the opportunity to interact with administrative people very frequently in terms of — ?

Blackwell: Not very much, no. Wilmot: Okay.

Blackwell: But the interactions I did have were friendly, [pause] I interacted much more with students than I did with professors or administrators.

Wilmot: Well, I think I'd like to spend a lot of time on that the next time we get together. Blackwell: Okay.

Wilmot: I think that for today, I would like to end. Is that okay? Blackwell: Yes, sure.

[end of interview]

David Blackwell's parents, Grover Blackwell and Mabel Johnson Blackwell, Centralia, Illinois, Spieth Studios, 1917

17

INTERVIEW 2: APRIL 18, 2002

[Minidisc 2]

Wilmot: It's the morning of April 18. I'm here with Dr. David Blackwell. Dr. Blackwell, I had a couple of questions to follow up from our last conversation.

Blackwell: Okay.

Wilmot: Our last conversation, we focused mostly on your early childhood and there were some things that you raised that made me think of more questions. So I wanted to go back to them.

Blackwell: Sure.

Wilmot: The first one was really about your father again, and his work as a hostler. Blackwell: Yes.

Wilmot: Which, by the way, when I found it in your bio-bib, I had to go to the dictionary. And it said two things; one of the things it said was taking care of engines, and the other one it said was an innkeeper. So, it was good when you clarified for me what that was.

Blackwell: So he was an innkeeper for engines, for steam locomotives, [humorously]

Wilmot: Right, I was wondering, if there are three people who had shifts, which shift did your father work, or did it change a lot?

Blackwell: It varied. The three shifts were three to eleven in the afternoon and evening, eleven to seven overnight, and seven a.m. to three p.m. Those were the shifts. And my father had all three of them at one time or another.

Wilmot: Okay, and were the other hostlers also African American?

Blackwell: Yes, yes, they were all black. I think that the Illinois Central Railroad had very little mixing in this sense: some jobs were black and some jobs were white, and you just didn't go from one to the other.

Wilmot: Did your father share with you how he came to have that job, how he came to become a hostler?

Blackwell: No, I don't think there was ever any idea that I might become a hostler. There probably was an idea that my younger brother might become one. I think that he probably wanted to do that.

Wilmot: My question was more around did your father — was he ever anything else besides a hostler? Did he have another career besides that?

Blackwell: No, no.

Wilmot: And you said your younger brother was perhaps more interested in becoming a hostler?

18

Blackwell: Yes, and he did get a job working for the Illinois Central Railroad when he was about sixteen or seventeen, and worked there all his working life.

Wilmot: Can you tell me his name?

Blackwell: Yes, everybody called him Skeet. His real name is Johnson Wesley.

Wilmot: Can you tell me the names of your other siblings?

Blackwell: Yes, my other brother is named Joe. And my sister is named Elizabeth Louella.

Wilmot: So it's Joe, Skeet, and Elizabeth Louella.

Blackwell: Yes, Joe is a lawyer in Cleveland. And "Betty Lou," as we call her, is an elementary school teacher in Chicago.

Wilmot: And, I had a question also, just to follow up on our last conversation, about — you'd mentioned that your mother went to church. And I was wondering, first, which denomination? And was your whole family going to church or was it just your mom?

Blackwell: Roughly speaking, the women went to church and the men didn't, [laughs] I don't think my father ever went to church. Nor did my uncles. But my mother did, and my aunt did, and my grandmother did.

Wilmot: Which denomination was it? Blackwell: Baptist.

Wilmot: And I wanted to ask you also your grandparents' names.

Blackwell: My mother's father's name was David S. Johnson. My mother's mother's name was Elizabeth Beckett. My father's father's name was Dabney Blackwell, and my father's mother's name was Ida White Blackwell.

Wilmot: Was your family unique in Centralia as a black family owning land? Were there other families who were owning land?

Blackwell: Many black families owned their own homes. I don't remember other black families owning a substantial amount of land other than their homes.

Wilmot: It strikes me that in Centralia, you lived independent of white people as well as very closely with them in a highly integrated environment. Does that sound correct? Does that sound true?

Blackwell: I'm not sure what you mean by "independent" of white people.

Wilmot: It seemed to me that, because you lived in something like a family compound, like you were surrounded by your family. And also because I read in an interview with you that you said your grandfather didn't send your uncle to school because he didn't want him to be —

19

Blackwell: He didn't want him to be discriminated against, that's right. But, of course, that was in the generation before mine. I have no doubt that I was discriminated against in certain ways; but I never felt it.

Wilmot: Did you have a sense that your family was aware of different social-political

movements like Marcus Garvey's UNIA, or W.E.B. DuBois' Niagara movement?

Blackwell: No, my family was not very political. I believe that they routinely voted Republican, but not out of particularly strong political conviction. Practically everybody that I grew up with was Republican. In fact, at election time, the Republicans used to come around and pay everybody five dollars to go and vote. Because they knew they would vote Republican if they voted! [laughs]

Wilmot: Also during the twenties and thirties when you were coming up, I believe there was a sense of — and it was by no means mainstream, but there was a growing sense of Pan- Africanism, meaning that people were thinking in terms of the Caribbean, there was a growing sense of that. I was wondering, was that a sense that was reflected at all in your community when you were growing up?

Blackwell: No, I suspect that was more in big cities like Chicago and St. Louis, no, we didn't feel that at all.

Wilmot: You've mentioned several times the effects of the Depression in shaping your choices and decisions, and just your outlook. And I was wondering if you could elaborate more on that and also if you can tell me about how your family fared with the Depression.

Blackwell: Well, my immediate family was lucky, in that my father held on to his job throughout the Depression. My uncles didn't do so well. They had a grocery store that went bankrupt in the Depression and they had to find other work. So, we were not directly affected by it. That is, my father and mother and I, and my brothers were not directly affected by it. But many people around us were.

Wilmot: Do you remember what kind of work your uncles found?

Blackwell: Yes, one of them went to St. Louis and got a job in a packing house, I believe. And,

another one went to Chicago and then somewhere in Wisconsin. And I don't remember exactly what kind of job he got, but they had to leave Centralia to find work.

Wilmot: And the uncle who ran the grocery store where you used to spend a lot of time, can you tell me his name?

Blackwell: Yes, his name was David S. Johnson. He was named for his father.

Wilmot: For much of the Depression, because you went to school in 1935, you were in school. Blackwell: Yes.

Wilmot: And I was wondering, for your sisters and brothers, were they working during that

time? Or were they much younger than you? I guess I should start by asking if you were oldest.

20

Blackwell: Oh yes, yes, they were younger than I. They were in elementary and high school during the Depression.

Wilmot: When we spoke last time, you told me there were a hundred black students in a campus of 12,000. And I wanted to ask you, did you have the sense that you were one of the first groups of black students on campus?

Blackwell: I'm not sure exactly what the question means, but I'll say this: I had the impression that there had been about a hundred black students on campus for many years. It's just a changing group; but I think it stayed about the same size. My fraternity house had been there for a good many years and so had the Kappa house.

Wilmot: As far as being part of a group of a hundred people on a campus of 12,000, what did that mean for you in terms of how you approached your work academically and your campus life?

Blackwell: I'm not sure exactly what that means. There we were, mostly living in two black

fraternity houses, one black sorority house, and a few other houses, and we all knew each other.

Wilmot: You mentioned the story of arriving on the train and not knowing anyone and there was a fraternity brother to greet you. And that was one reason why you became a member of the Alphas. There were probably other reasons that you joined the Alphas and I wanted to ask you about them, because the Alphas have a very specific historical legacy and tradition. Were there other reasons?

Blackwell: No, there was no more to it than that. I had never heard of black fraternities until this fellow called me across the street. And I didn't know where I was going to live and he suggested I live there and that was great with me. It was right across the street from the campus.

Wilmot: When you think of pledging processes today, do you feel like it was much more rigorous or less rigorous than what people do today?

Blackwell: Looking back on it now, the whole process looks a little ridiculous. And, of course,

there would be no paddling nowadays. Do you know about paddling? [laughs] That was just routine in those days.

Wilmot: I suspect it may still be routine.

Blackwell: Well, I'm not proud of my fraternity if it still is routine. I don't remember very much about it.

Wilmot: Did you provide social services to the community outside of campus, as an Alpha? Blackwell: No. We were just busy trying to make it and trying to have a good time and —

I got to know a fair amount, indirectly, about Chicago, because, as I said before, most of my fraternity brothers were from Chicago. One of the things that surprised me was how corrupt a city Chicago was and how they, essentially, accepted this. For instance, certain

21

jobs had a price. If you wanted to get a job in the Chicago school system, that would cost you a certain amount of money. If you wanted to be a policeman, that would cost you a certain amount of money. They simply accepted this as the way things were.

Wilmot: That was very different from where you were from. Blackwell: As far as I know, yes, I certainly never heard of anything like that in Centralia.

Wilmot: Did everyone who was in your fraternity live in that house with you? Or were there people who were outside of the house as well?

Blackwell: Most of them lived in the house. A few of them lived at home in Champaign-Urbana, but most of them lived in the fraternity house, yes.

Wilmot: Do you remember having discussions, together as a group, about the university, and your experience there?

Blackwell: We certainly had bull sessions, but I don't remember what they were about. I remember more that they were heated than what they were about, [laughs]

Wilmot: What's a "bull session?"

Blackwell: Oh, just where a group of fellows get together, maybe in somebody's bedroom, and just start arguing about this and that. You learn a lot that way, though.

Wilmot: Did you also talk about politics? Just the government and what was going on in the world?

Blackwell: I suspect that some of the sessions were about that, but I was never interested.

Wilmot: Can you tell me the name of the black sorority that was on campus as well? You said there was one sorority.

Blackwell: Yes, Alpha Kappa Alpha.

Wilmot: How often did you get on a train and go visit your family? While you were at graduate school.

Blackwell: Just about every week. I think I told you, that was how I got my laundry done a lot of the time. I didn't tell you that? Every Wednesday, I would go down and get on a train and ride to Centralia and deliver my dirty laundry and pick up some clean laundry, and then get on the train and come back. So, I spent virtually every Wednesday night on the train.

Wilmot: And did your family come visit you on campus?

Blackwell: From time to time, yes. There used to be a state high school track meet and my brother would come up every year for that.

22

Wilmot: While you were at Urbana-Champaign, were you involved with any academic committees or groups, other than your fraternity?

Blackwell: Yes, I belonged to the mathematics club there.

Wilmot: I understand you were the president.

Blackwell: I think so, but I didn't have any duties. I suspect it was really run by the faculty sponsor.

Wilmot: Who was that?

Blackwell: I think it was Professor Bailey, but I'm not sure.

Wilmot: Your peers in the math department, did you have the sense that they were also having multiple jobs in the way that you did?

Blackwell: Some, yes, but mostly, no. I have the impression they were mostly supported by their parents.

Wilmot: And were they all from the community surrounding the university as well? And had they all got a similar education as you?

Blackwell: Well, let's see. One was from Chicago, one was from Kansas City, Missouri. And, I don't know where the others were from. No, we were just all undergraduate math majors there together. And then, later, in graduate school, of course, the graduate students in mathematics came from all over.

Wilmot: Did you have any guidance in selecting your courses? Other than the guidance that was offered to you by your fraternity brothers?

Blackwell: Yeah, let's see. I don't think so. As I told you, as an undergraduate I took every

mathematics course that was available to me. And as a graduate student, the program was pretty well set. So, I didn't have many options.

Wilmot: So, did you need assistance in navigating the courses and requirements? And if you had that need, was there someone there to help you?

Blackwell: I discussed things with the other students, as I've mentioned. I got much more help from students than from faculty in choosing courses and in learning things.

Wilmot: Did you have an academic advisor prior to finding Joseph Doob?

Blackwell: I don't think so. I'm sure there was someone who signed the programs of the students, but I think that was a pro-forma thing. I think you just showed up saying what you wanted to take, and he checked if it was reasonable and it allowed you to make reasonable progress towards graduation, he'd sign it.

Wilmot: Do you remember when Jerzy Neyman came to address the University of Illinois in 1937?

23

Blackwell: I remember it very well, yes. Wilmot: Can you tell me about that?

Blackwell: One of my mathematics teachers at Illinois was Professor Crathorne, C-R-A-T-H-O-R- N-E. He — this is not quite on the subject, but I'll get there. He was very helpful to me. I took calculus from him, and I took a course in probability from him. And he gave me a couple of jobs. One of the jobs was bookkeeping for him. He was treasurer of some church there. So, it was my job to keep books for him. Another job was going through an algebra book that he was writing and doing all the problems and making sure the answers were correct.

And, of course, I appreciated the jobs and he paid me a certain amount for them. But looking back on it, I realized that that was just his way of helping me. He was treasurer of the church; he didn't need me to keep books. And, well, the other job, checking all the answers, was of some use but what he really wanted to do was to help me.

And it was he who invited Jerzy Neyman to come and speak. So, he told his class, "Be sure and be there." And he told me, in particular to be sure and be there. And, at some point either before or after the lecture, Professor Crathorne introduced me to Neyman. And I hadn't understood much of what Neyman was saying but, in any case, I was honored to be introduced to this very distinguished man. It never occurred to me that I would have any later contact with him, but that incident stuck in my mind.

Wilmot: It sounds like Crathorne was a good mentor. It sounds like he was he was thinking about you.

Blackwell: Yes he was. I didn't appreciate it at that time. It was only when I look back on it that I realize how he helped me.

Wilmot: And Neyman came and spoke.

Blackwell: Yes, he spoke about his work and what was going on in statistics. As I say, I understood very, very little of it. But that was, I believe, before Neyman came to Berkeley.

Wilmot: How did you come to work with Joseph Doob as your academic advisor?

Blackwell: There was a student there named Don Kibbey, and I was talking to him one day. And he asked me who I was going to work with. And I said that I didn't know but it was time I was deciding. And he simply said, "Why don't you work with Joe Doob? I'm working with him. He's a nice guy and he might take you." So, I trusted Don Kibbey, and I just went and asked Joe Doob if I could work with him and he said yes. And that was it. It wasn't complicated at all.

Wilmot: Did you think he knew about you as a student? Was it a small enough department that he knew who you were and knew a little bit about your work?

Blackwell: He probably knew something about me, yes. It was a small enough department that all the professors knew or knew about all the graduate students, yes.

24

Wilmot: So, as your advisor, what was his job?

Blackwell: Why, his job was to give me a thesis topic and guide me in writing it. Wilmot: And is that what your working relationship was like?

Blackwell: Yes, the first thing he did was to give me some papers to read. These were papers in

measure theory, and most of them were in German. Paul Halmos was there, he had been a student of Joe Doob's, and he had read these same papers a year or two before, and was very anxious to talk about them. And I wanted to learn about them. So, I talked to Paul Halmos and he essentially gave me a course in measure theory. And, later, he wrote a book on measure theory. I think I received the first version of Paul Halmos' book on measure theory.

Wilmot: I want to back up a little bit and ask you a little more about Joseph Doob's background and special interest area in terms of academics. What was his background and training?

Blackwell: He got his degree, his PhD, at Harvard in 1932. But his thesis was in analytic functions that had very little to do — had nothing to do — with probability theory. He learned probability entirely on his own and really helped to lay the foundations of the subject. I was very lucky to have him as an advisor. The things that he told me to read and the things that he wrote were just fundamental in the future of the subject of probability.

Wilmot: When you think of what he conveyed to you about ways to approach the work, can you think of—?

Blackwell: [pause] I couldn't even begin to describe it. There's just a certain cleanness about his work and — he asked the right questions.

Wilmot: Do you feel that you took that away with you, in terms of your approach? From your experience with him?

Blackwell: Yes. The first few papers that I wrote, the main question I asked myself was, What will Joe Doob think of this? And I tried to say things that he would think were interesting.

Wilmot: Did he have other advisees besides yourself?

Blackwell: Yes, Paul Halmos whom I mentioned to you. And Warren Ambrose was one of his advisees.

Now, in 1941 — Joe Doob was going to go to the Institute for Advanced Study in Princeton in 1941-42. Paul Halmos had finished, and Warren Ambrose had finished, and they were going to go to the institute with him. Now, there I was. So it was important to Joe Doob that I should finish. Because he didn't want to leave me hanging there; at the same time, he was going to go to the institute. So he sort of pushed me to finish my thesis so that he would be free to go to the institute without worrying about me. And I have always thought that he approved my thesis a little sooner than he would have if he'd been staying there.

Wilmot: Yes, he was like, hurry up and come on.

25

Blackwell: That's, hurry up and finish your thesis, so I can go to the institute without worrying

about you. And so, he not only pushed me to finish my thesis and approved it but — and this is something you were going to touch on a little later — got a Rosenwald Fellowship for me so that I could go to the institute, also. When I say, "He got it," I'd never heard of such things. He told me to apply for it. He got the necessary information, and he sort of suggested what I should say. So, Joe Doob really just got me this Rosenwald Fellowship. So, I did finish my thesis and we all went to the institute together. So, in 1941-42, 1 was at the institute, Joe Doob was at the institute, Paul Halmos was at the institute, and Warren Ambrose was at the institute.

Wilmot: I have a few more questions for you before we arrive at the institute. Blackwell: Okay.

Wilmot: This is just a list that I found on the internet for your other members at the institute these two years. I wanted to ask you about your thesis on Markov chains. I wanted to ask you first, what was the inquiry and conclusions of your thesis?

Blackwell: Let's see. Well, one part was — can you describe all the Markov chains that are

idempotent? Now, I have to tell you what that means. That means that given where you are today, where you expect to be two days from today is exactly the same as where you expect to be one day from today. Your future two days from now looks exactly like your future one day from now. And, it was known for finite state chains what all the idempotent chains are, and Joe Doob suggested it would be interesting to explore that question in general, and see if I could describe them all. And, I more or less did.

Well, that was an interesting, fairly small question. But then there was another big question. Joe had discovered that Martingales going backward converge. And if you apply that to finite state Markov chains, you get an interesting theorem about convergence of products of Markov matrices. And he suggested that I describe what you get there. And, again, I more or less described that, not completely, but I made a fair amount of progress on that.

So that, the first topic, was a sort of small problem that you could complete and then forget about. But the second one was one that had big implications and that was worth studying a lot. I later published a small paper on each one of those topics.

But they were just Joe Doob's ideas, I was just exploring some questions that he had raised.

Wilmot: Were there any other topics that you were interested in pursuing as a thesis, or were you quite happy to work with those ideas?

Blackwell: Oh, I was quite happy to work with those and they took all my time during my last year at Illinois. And they were a big part of what I worked on during my year at the Institute for Advanced Study.

Wilmot: This is a question from the non-statistics background person. What was the potential real-world application of your thesis?

26

Blackwell: [laughs uproariously] None. Mostly, you don't do mathematics because of its

application to the real world. You do it because it's mathematically interesting. Of course, you're pleased if someone happens to find a real-world application. But that's not why you do it.

Wilmot: As your time at graduate school drew to a close, how did you begin to think about your next steps in terms of career with Joseph Doob off to the institute? What were your concerns and priorities?

Blackwell: I was pretty sure I would get a job in one of these 105 black colleges. I was interested to see which one. [chuckles]

During the three or four months, during the summer between the Institute for Advanced Study and my job at Southern, I got a job working for the Office of Price Administration. I bet that's not in my bio-bib, or whatever. Anyway, I'll tell you a story about that.

I had a very good friend at Illinois named Dunbar MacLaurin. He was a black economics student who got a PhD at about the same time I did. And he got a job in Washington, working for the Office of Price Administration. So, he was there while I was at the Institute for Advanced Study. Now, at that time, young people were being drafted to go into the army. MacLaurin didn't want to go into the army; I didn't want to go into the army. He wrote to me and said, "If you want to be protected from the draft, come down here and get a job working for OPA. That will be an essential occupation and you'll be protected from the draft."

So I came down to Washington, D.C. MacLaurin arranged for me to be interviewed by somebody at OPA. And a few weeks later I got a job at the Office of Price Administration. And, of course, I took it because that would protect me from the draft. But, shortly after I arrived at the Office of Price Administration, MacLaurin, who had claimed to have a job that was exempt from the draft, his boss was drafted! [laughter] So that told both MacLaurin and me that our jobs were not exempt from the draft. So, that settled it for me. I had already accepted a job at Southern, but I had planned, I suppose, to try to postpone it so that I could keep this government job. But when I found out that that wasn't going to keep me from the draft, then I just resigned that OPA job and went on to Southern. So, for about three months I worked for the Office of Price Administration.

Wilmot: What did you do there? Blackwell: I don't remember much about it. Wilmot: Maybe you had a desk?

Blackwell: Oh, I certainly did have a desk. And, I remember a few of my colleagues. But it's not clear what we did. [laughter]

Wilmot: Okay, I kind of imagine it was some kind of number-crunching, but I'm not sure. Certainly from the title —

27

Blackwell: Well, now, I'm going to tell you something and it's not going to be right; but it's going to be fairly close to being right. The director of the office of the Office of Price Administration was a man named Leon Henderson. And he needed an economic advisor, somebody just to help him make decisions. There was a Harvard economics professor that he wanted, I think his name was Richard Gilbert. So Leon wanted to bring Richard Gilbert to come to Washington to be his economic advisor. So you had to ask the question, "What would be a suitable salary for Richard Gilbert?" Turned out that in order to pay Richard Gilbert that salary, he would have to be the director of a division. So they created a division for Richard Gilbert called the research division. That's the division I was in. [laughter]

Wilmot: I think I understand better why you were not sure what you were doing there.

Blackwell: Yeah. I'm not at all sure that I have the details right. But it's a good story, and it's approximately right.

Wilmot: Well, that's the best. And, you know, feel at liberty any time you have an approximately right, good story, put it on the table. That's good.

Blackwell: As long as I say, make it clear, it's approximately right but I'll get the details wrong. For instance, I'm not sure about the name Leon Henderson or about Richard Gilbert, but something like that happened.

Wilmot: That's pretty good, that's pretty good.

I wanted to go back a little bit. We're kind of slowly inching up on your time at the Institute for Advanced Study, but I just wanted to talk a little bit more. In 1941, the U.S. mobilized to enter World War II.

Blackwell: Yes.

Wilmot: And you talked a little bit about how you experienced these events and trying to find a job for shelter from the draft. Because I guess, you know, being a graduate student or an academic or a professor wasn't considered an essential occupation, or was it?

Blackwell: I'm not sure, but I know that I had the impression that working for OPA would be more essential, but I was wrong.

Wilmot: I actually heard this story from Russ Ellis. He said that you were somehow excluded from the draft.

Blackwell: Oh. This is later, in 1943 or early 1944 they started scraping the bottom of the barrel, as they called it, in drafting people. So, they called me to be drafted. And I had to go to Chicago and take some kind of examination. And I did, and among other things you had to go to a psychiatrist. And he asked me a few questions, wrote some things down. But before that happened, I had mentioned to one of my students, well, I had mentioned, maybe in my class, that I was going to be drafted or something, and didn't want to be.

So, a student came into me, and said, "Prof, I understand that you've been called for the draft." I said, "Yeah." "And that you're worried about it." "Yeah, I am worried about it."

28

He said, "You don't need to worry. Because," he said, "I was drafted and I went into the army and examined and I was rejected for psychoneurosis. I am one, I know one when I see one, and you are one." [laughter] "You don't have to worry!" One of my students told me that, and Nadine, he was right! [laughter] I showed up and was examined by the psychiatrist among other people. He didn't make any decision right then, but a few weeks later I was notified by mail that I had been rejected and the reason was what they called "anxiety neurosis." So I was just very pleased with that.

Wilmot: I can imagine. I'm not sure what it means, "anxiety neurosis" or "psychoneurosis," I'm not sure what those designations mean. But does that mean high-strung? That could mean anything.

Blackwell: Yes, I'm not sure what it meant. But, anyway, the psychiatrist rejected me and he was right to reject me because I would not have been suitable for the army.

Wilmot: Right, I understand. Well, I want to ask you then, did either of your two brothers end up being drafted or fighting?

Blackwell: No. Skeet was not drafted. I've forgotten how that happened. But Joe was younger, and by the time Joe was ready to go to college, the war was over and the G.I. Bill had been passed. And he volunteered, and spent two or three years in the army. And then used the GI. Bill; and that supported him all through college. Both Joe and my cousin John did the same thing. They graduated from Centralia High School in the same year and volunteered for the army, and then went to college under the GI. Bill.

Wilmot: I have a fairly large question, which is — and it's very general and maybe it's something that you'll answer in small pieces — but how then did you experience World War II? For example, do you remember when the Japanese bombed Pearl Harbor? And how did you experience these events that were happening around you?

Blackwell: Yes, I certainly can't answer that question, [pause] I was just not politically aware in those days. I almost never read a newspaper, for example. I was much more interested in mathematics and athletics and girls.

Wilmot: What were your expectations when you went there, to the LAS?

Blackwell: Oh, just that I would continue studying the topics, my thesis topics, and continue to talk to Joe Doob and Paul Halmos and Warren Ambrose.

Wilmot: Was this the first time you had traveled outside of Illinois? Blackwell: Oh no. It was the first time I had lived outside of Illinois, certainly. Wilmot: What did your family think of this opportunity for you?

Blackwell: I don't think they had any opinions. They were assured that what I was doing was reasonable and just glad that I came back to Centralia for a visit.

Wilmot: I read in one of your interviews that you had noted that your mother had more, kind of, questions about the path you were on than your father.

29

Blackwell: Yes, she wanted me to study business somehow. But she accepted that I was going to make a reasonable living doing what I was doing.

Wilmot: You mentioned that during your time at the Institute, you basically kind of continued the work that you had started in your thesis, the two problems that you'd been really working on and focusing on in graduate school under the advisorship of Joseph Doob. Did you start to branch out into other areas as well? Could you describe the other primary math and statistical themes you explored while you were there?

Blackwell: Yes, one of the things that I got interested in while I was there was game theory. Von Neumann was there, and he had proved the basic theorem in game theory, and that was sort of in the atmosphere; so I got interested in that.

And also, I got mildly interested in statistics at that time. There was a fellow named Sam Wilks at Princeton University who gave the course in statistics, graduate course in statistics, there. And he was a pretty important man. He was the editor of the leading statistics journal. And I sat in on Sam Wilks's statistics course while I was at the Institute for Advanced Study. And one of the main ideas that I got from that course was that I wished that I could understand statistics and the way statisticians thought, but it was too hard. But I knew there was something interesting there and I wished I could understand it.

Wilmot: What do you think appealed to you about it at that time?

Blackwell: Well, the questions that statisticians ask are interesting. "Given this data, does that say that this treatment is better than that treatment?" That's clearly an important question. And it was also clear that it was what statisticians were trying to answer. And Sam Wilks proved lots of theorems that, as I say, I couldn't quite adjust to my way of thinking. But, as I say, I knew that I would like to understand it.

Wilmot: Which faculty members did you work with, while you were there at the institute? My understanding is that there is a standing group of faculty members who are there.

Blackwell: Yes. I didn't work with any of them. I had one or two conversations with Von Neumann. And I had just one conversation of one minute duration with a very distinguished logician named Godel. But except for that, I didn't interact with the resident professors. But I did with other young people there. Again, as in my graduate days, I learned more from my contemporaries than I did from the higher-ups, so to speak.

I see you have a list here of people who were at the Institute at that time.

Yes, Paul Halmos, Shizuo Kakutani. He was a young Japanese mathematician at the institute at that time. I learned a lot from him. He had also worked in Markov chains. But he understood them a lot better than I did, and I really learned a lot from him. He's at Yale, by the way, he retired some years ago. I think he went to Yale right after the war, and stayed there all his career.

Gerhard Kalisch, I worked quite a bit with him and learned quite a bit from him and we were friends. We used to go for walks together. I once admired a tie that he was

30

wearing, a beautiful brown tie. He took it off and gave it to me. [demonstrates] He got a job at UCLA and taught there until he retired. I used to see him occasionally.

George Mackey was there, yes. I learned a fair amount from him at the time. Several years later, he and I discovered the same theorem at about the same time. And it was interesting to me because the way he came at the theorem had absolutely nothing to do with the way I came at the theorem. Somehow we both found the same theorem.

Wilmot: Can you describe the theorem and then describe your different approaches to it?

Blackwell: I can — [pause] I can describe the theorem and I can describe what led me to it, I think, [pause] Well, I'll first — I don't know if I can describe it or not in a way that you'll understand, [wondering] It says that two countably generated sigma fields of Borel sets that have the same atoms are identical. Now, wait a minute, and I'll see if I can find the paper that says that, [rifles through papers with no success] I'm sorry, Nadine, my files are not in very good shape.

Then Dorothy Maharam. Wilmot: That's the one woman.

Blackwell: Yes. Again, I learned a fair amount from Dorothy Maharam. She is also interested in measure theory. And knew a lot more about it than I did. She and her husband, Arthur Stone, were both at the institute that year. I haven't seen them for three or four years, but they come to Berkeley from time to time, and always come by my office when they come. And again, I've had mathematical contact with Arthur Stone over the years. I asked him a question once; and he remembered the question and asked the question of somebody in England; and that person answered the question. I don't remember exactly what it was, but I remember how sort of indirect this was.

And Jimmie Savage was at the institute that year! Well, he and I have a number of interactions over the years.

Wilmot: Didn't you work together at Rand as well?

Blackwell: Yes, yes, yes, we did, among other things. Well, those were the people at the institute that I've worked most closely with. But I tell you again, it's sort of my contemporaries that I worked with.

Wilmot: You have an interesting way of always framing it as you were learning from them and, I mean, I know you were learning from them. But I'm sure there was a lot of times when they were learning from you, as well.

Blackwell: Well, you'd have to ask them about that, [laughs]

Wilmot: It seems to me also that, just from your describing where one other person talked to his friend in London and the response came back to you, that you were suddenly in a much larger, more international network of people who were thinking about math and statistics once you were at that institute.

31

Blackwell: Well, mathematics is completely international. Yes, yes. Of the people that I met at the institute, let's see. Halmos is actually originally from Hungary. Kakutani is from Japan. Mackey is from Texas. And Maharam is from the East Coast somewhere. Jimmie Savage is — not sure, I think from Michigan. Arthur Stone is from England, so they — . [pause]

Nadine, I'm glad you're interested in all these ramblings, but I still don't know why. [laughs] But that's all right.

Wilmot: Okay, well, I'll try and share with you more. I think it's important to have perspective on the larger framework.

Blackwell: Okay, sure.

Wilmot: I want to ask you, this is kind of a general question, but what was life like as an IAS

member? And the reason that I'm asking is because I just kind of imagine this group of young people having this incredible opportunity to share and exchange ideas. Is that basically what it was like there?

Blackwell: Yes. You had a small office and a library and rooms where you could gather and talk, go to the blackboards and — you just bump into each other, and for instance, I remember one day I was reading something in this book, Paul Levy's book. And I puzzled over it, and puzzled over it, I couldn't understand it. And a fellow name Ralph Phillips was sitting a couple desks away and I just went over and asked Ralph whether he could understand this passage. Let's see, is his name on here? [on the list] Well, Ralph Phillips was there, and I went over and asked him, "Ralph, I can't follow this reasoning, can you explain it?" Now, he, although not a probabilist, started reading it and he explained it to me. Well, that was the sort of casual atmosphere that there was. We were all there trying to learn and if you thought somebody could help you, you did, you asked.

Wilmot: This book, let's see if I can pronounce this correctly, Theorie de L 'Addition des Variables Aleatoires, it's from Paul Levy.

Blackwell: Sounds French to me.

Wilmot: I don't know what that means in English, though.

Blackwell: The Theory of Addition of Random Variables, [pause] And many of these are papers that I grew up on, the papers that are cited in this book, [looks through book]

Wilmot: Let me start here, while you were at the institute, the director there, was it [Abraham] Flexner or [Frank] Aydelotte?

Blackwell: I believe it was Aydelotte.

Wilmot: Okay. In the interview, it kind of mentions that Princeton had had a real problem with your being there, and that the director had to really go to bat for you. And had staunchly gone to bat for you. Can you tell me that story?

32

Blackwell: I can tell you the story as I understand it. I didn't find out about it until years later. There was a custom that all members of the institute would be made honorary faculty members at Princeton. And so when I was invited to become a member of the institute, that meant that I would be appointed an honorary faculty member at Princeton. Well, the president of Princeton did not want any black honorary faculty members at Princeton. And as I understand it, he notified the director of the institute and there was a big fuss over this. And several of the professors in the institute complained about it and threatened to disconnect the institute from Princeton unless I was accepted. And I guess it wasn't a big thing, so the president of Princeton backed down.

I never knew anything about that. Of course, it was all settled before I got there. And I was just welcomed cordially along with everybody else. It was only much later that I found out that there had been all of this to-do.

Wilmot: Did you learn who were the faculty who threatened to sever ties with Princeton?

Blackwell: One of them, I understand, was Oswald Veblen, who was a professor of mathematics at the institute, a geometry man. When I arrived, he was very cordial to me [laughs], but didn't act as if there was anything special about me, so I didn't know anything about that, then.

Wilmot: You didn't know the other names of the people who stood for your — ?

Blackwell: No, it seems to me that I once did know something. But, no, I've forgotten. I don't know who would know anything about it now.

Wilmot: How did you hear about this? You said someone told you years later, who told you? How'd you hear about it?

Blackwell: I don't know. I have the impression that someone just mentioned it under the impression that, of course I knew all about it, but I don't know.

Wilmot: But you really didn't know all about it.

Blackwell: No, I didn't know anything about it.

Wilmot: Other than that experience, which kind of happened—

Blackwell: — off-stage!

Wilmot: — off-stage. Was the institute an environment that felt free from racial prejudice?

Blackwell: Yeah, [pause] Yes, I never had any problems.

Wilmot: And was it a place where no one questioned your right to be at the table?

Blackwell: That's right.

33

Wilmot: Outside of the institute, in academic circles, when you encountered people who

questioned your abilities from their own racial prejudice, how did you address that? How did you deal with that?

Blackwell: I've never had that kind of incident. I've been denied access to things because I was black, but not among my colleagues or the people I associate with.

[Wilmot: I want to go back to that question of being denied access, and ask you what that meant?

Blackwell: [pause] Well, there have certainly been incidents in my life where I was "denied access." My memory on this is somewhat vague, but there was a meeting of the American Mathematical Society somewhere in Virginia, and I and two white mathematicians drove from Washington, B.C. down to the meeting. And I was not allowed to stay in the dormitory where the other mathematicians were staying, and where the meetings were to be held, I think. So, I simply turned around and drove back to Washington, D.C., and missed that meeting. That's certainly one case where I was, quote, "denied access," unquote, and there may have been other incidents like that. Maybe two or three. I remember some correspondence I had with the secretary of the American Math Society over such incidents. I don't remember any others, but there may have been others. But again, my mathematical colleagues were quite sympathetic. They gave me the impression, at least, that they wished it had not happened.

Wilmot: Is there a way to bring issues of structural racism into probability? Is there a way to theorize these things in math?

Blackwell: [laughing] I don't know how. Structural racism — [musing] — that's a new term to me.

Wilmot: Okay. When I say structural racism, I guess what I'm speaking to is practices of discrimination that are embedded in institutional societal practice.

Blackwell: I understand. I just gave you an instance of that. I just thought of another one that

happened to me and my wife in Washington. I may have told you about this one. There was a play that my wife and I wanted to see. So, we went to the downtown theater in Washington and stood in line for tickets. And when we got to the head of the line, the sales clerk said that she couldn't sell us tickets. So, I stood there for a few minutes sort of arguing with her, and the people behind started a chant, "Sell him the tickets! Sell him the tickets!" But she couldn't. She said she would lose her job if she did, she wasn't supposed to do that. So, we finally left. So, there is, in what you call this institutional racism, something that is stronger than the particular people that are involved in it.

Wilmot: Yeah, it's not about hurt feelings. It's about denied access. It's much bigger than that.

For example, one time, when you and I were speaking, and you mentioned the death penalty. And I was thinking about — you know how there's this whole thing right now where they're reviewing the death penalty, and thinking about how often it's been wrong, basically. Is there like a place where you could take probability into that kind of analysis and theorizing, given the different factors that go into making those kinds of decisions?

34

Blackwell: Well, people do do informal probability calculations. The governor of Illinois probably said that since this one innocent fellow came within two weeks of being executed, probably there are other innocent people that have been executed. That's kind of probabilistic reasoning, which we all do, every day. And of course, probability concepts enter into our criminal justice system, generally. Preponderance of evidence is a probabilistic concept. Beyond a reasonable doubt is a probabilistic concept. So, probabilistic reasoning enters into these things.

Wilmot: I have another question about the institute. You were telling me, you said, "Oh, we

talked mostly about math. We were talking about math." And if it's possible to say what were the hot issues to talk about in math at that time, like where were yourself and your colleagues focused, around math? Like what issues were the ones that you were talking about all the time? Which were the ones that kept you awake all night?

Blackwell: No. Issues is not the appropriate word at all [laughing]. What problems, what concepts, what ideas, [pause] I was still learning. For me, I was mostly learning about probability, real variables, point set topology. Some of my colleagues and friends were learning about Hilbert's space, and I learned a little about it just from listening to them, but. . .oh, nothing big. We were each doing our own research and if you came across a problem that you thought one of your colleagues might know something about, you'd ask him about it. There were a lot of us around there, just beyond the PhD. We were just getting started. That was a big year for me.

Wilmot: When you say it was a big year for you, I imagine you in this kind of accelerated learning space, and just really open and really delighted.

Blackwell: Yeah, I was with all — well, there were essentially two kinds of people there. There were the really great mathematicians and all these fresh new PhDs. So, we didn't have much contact with the really great mathematicians. They gave lectures from time to tune, you'd learn something from them. But you had a lot of contact with the other young people. I learned a tremendous amount from them.]

Wilmot: I have a question for you about applying for work after your dissertation and after the institute — which is that I'm wondering if you'll tell me the story of how you looked for work?

Blackwell: Well, I did two things. I wrote those 105 letters. And I went on an automobile tour of about thirty colleges. And just drove up to an institution and looked for the mathematics department and went in and introduced myself to the head of the math department and told him I was looking for a job. Crazy way to do things! But I didn't know any better, [laughs] And, mostly I got cordial receptions but it was made clear to me that — well, often, I was referred to the president of the institution because the head of the math department didn't make appointments to the math department; the president of the institution made appointments. So, sometimes I got to see the president. And out of all this activity came three job offers.

Wilmot: And all 105 applications you sent out to historically black colleges? Blackwell: Yes.

35

Wilmot: And where did you tour?

Blackwell: On the East Coast. I started at Morgan, and went to Howard, and on down the East Coast. I don't remember the names of all the colleges that I visited. But I was having fun, just getting to see what the country was like and what the black colleges were like.

Wilmot: When you stopped in to visit at Howard, did you, did you have — that was eventually where you ended up for ten years. You met with the math department there?

Blackwell: I thought I told you that story, but I'll tell you again. Wilmot: I don't remember you telling me that story.

Blackwell: All right, yes, I showed up at Howard and went in to talk to the head of the math department. He was very polite. I believe he was a Jamaican, by the way. I don't remember his name. As I say, he was very polite. Asked me a few questions, but it was clear to me, even at the time, that he wasn't really interested at all.

However, another man who was not head of the department was sitting at his desk over there, listening to me. And after the head of the department had finished talking to me, he asked me one or two questions. And then, that was that. Well, two years later, the head of the Howard math department retired and this other man became head of the math department. He wrote to me and invited me to come to Howard for an interview. He remembered that conversation, and then he offered me a job, which I took.

Wilmot: Do you remember his name?

Blackwell: Oh, very well. His name was Woodard, Dudley Woodard. Let me tell you a story about Woodard. I've told this story before, but maybe not to you.

Wilmot: Not to me.

Blackwell: Not to you, okay. For many years, Woodard was dean of the College of Letters and Sciences at Howard. At the age of forty-five, he decided to go and work for a PhD in mathematics. And he did, at the University of Pennsylvania, studied there for three years, and wrote his thesis. And then, he came back to Howard, not as dean and not as head of the math department, but as a professor of mathematics. So that was a two-level demotion. And if you know how authoritarian Howard University was at that time, you will recognize that that was a big decision that he made to give up the deanship to go to work for a PhD in mathematics to come back for a professorship. But he did.

That was the man who was sitting in the other chair. And I was really impressed with Woodard. He was a serious mathematician. He persuaded the dean to set aside a special room for a mathematics library at Howard. And he started a mathematics seminar at Howard. I would give talks and he would give talks and graduate students would give talks. As I say, he was a serious mathematician and a serious man.

Wilmot: Well, it sounds like it to make a choice like he did, to make a choice like he did it sounds like he was very serious. He really made, it was definitely a commitment. And it sounds like he was very passionate about teaching and math.

36

Blackwell: That's right.

Wilmot: When you went on this tour of the different schools, it sounds like you were kind of investigating and exploring the math departments. What did you want to find? What were you looking for?

Blackwell: I wanted to find a job! [laughter] I'm sorry!

Wilmot: But, I mean in terms of your criteria when you were kind of exploring what it would be like to be at this institution. What were the criteria? What were you looking for hi terms of institutions?

Blackwell: I didn't know. I'd go to all these places and find out what it was like and apply for jobs, and then of all the offers I'd got, then it would be time to think about which one I wanted. But I was just trying to find out, that's all. I didn't have criteria in mind.

Wilmot: Out of all of the hundred and five applications, when you got back, how many affirmative responses?

Blackwell: Eventually, I got three offers. One was from Southern, one was from Clark College, and one was from West Virginia State. But as I remember it, as soon as I got the offer from Southern, I accepted it and didn't worry about maybe getting other offers.

Wilmot: Because that was where you wanted to go, or it was the first offer you got?

Blackwell: It was just the first offer I got. [laughs]

Wilmot: Southern. That's in Louisiana?

Blackwell: Yes. State university in Louisiana, and at that time, of course, all black.

Wilmot: And was that in Baton Rouge?

Blackwell: Yes, the post office was Scotlandville; but it really is in greater Baton Rouge.

Wilmot: Now, were these one-term, one-year positions, that were offered to you, or did you make them one-year positions?

Blackwell: I thmk they were one-term positions, presumably renewable. I believe, at that time, in the black colleges there was no such thing as tenure except maybe at Howard. All positions were one-term renewable at the pleasure of the president, but normally renewed. I'm sure I could have stayed at Southern forever on one-year positions.

Wilmot: When you arrived there, what did you find in terms of institutional support for the math department?

Blackwell: Why, there was a man who was the head of the math department, he simply told me what courses I was to teach. Institutional support is a fancy term, [laughs] As I remember, the textbooks were already chosen. It was a friendly atmosphere.

37

Wilmot: How many other math professors were there?

Blackwell: I think nobody except me and the head of the department.

Wilmot: Wow, so were you teaching many classes?

Blackwell: I believe I taught fifteen hours, yes, I believe it's something like that.

Wilmot: And this was your first time in the classroom?

Blackwell: Yes.

Wilmot: Wow, how was that?

Blackwell: It was fun for me, but it was hard for the students. I don't remember much about it, but I'm pretty sure that I made the courses more difficult than they should have been.

[Wilmot: How?

Blackwell: I think that I tried to make arithmetic challenging, instead of just a drill.] Wilmot: Do you remember what course you were teaching?

Blackwell: No. It seems to me that they were mostly freshmen courses, though. The advanced courses were taught by the head of the department.

Wilmot: Were most of the students just sort of checking in at basic math, or did you find students who were really excited about pursuing more advanced ideas?

Blackwell: I remember one student who showed real mathematical talent. I think he went on to become business manager at Southern University, something like that eventually happened to him. But he's the only one I remember especially as being talented.

Wilmot: This was your first time, also, living in the South.

Blackwell: Yes.

Wilmot: How was that for you?

Blackwell: It was interesting, and I didn't know quite what to make of it.

I did one thing that if I'd been wiser I would not have done. The first time I got on a streetcar in New Orleans — there's a little board that you plugged into the top of the seat, and on the front of it said "White" and on the back of it, it said "Colored." The idea of it was that if the board was here and all the Colored seats were taken, and the next row above was vacant, you moved the separation board up one row and then sat there, and vice-versa. I thought that board was rather funny. And when I got off the streetcar or the bus, I took the board with me. [laughs]

Wilmot: What did you do with it?

38

Blackwell: I don't know, maybe took it back to my room, posted it for a while. So I, of course, accepted segregation but I didn't take it very seriously. Just another one of those silly customs.

Wilmot: Did you have the sense that other people around you were taking it very seriously? Blackwell: I don't remember.

By the way, that reminds me of an interesting experience that I had. I wanted some mathematics books, and of course, they were not at Southern. So I went over to Louisiana State University to use their library. Of course that was a segregated institution. And I went to the mathematics department just to find out how I would go about it. There were two people on the faculty in mathematics at Louisiana State University that had been my fellow students at Illinois. So, I ran into them. One of them sort of avoided me and made it very clear that he didn't want to be associated with me there. The other one was extremely helpful and took me in, introduced me to the librarian, and asked that I be allowed all the facilities and shown the courtesies and so on. And that was that.

But the interesting thing to me was that the man who refused to have anything to do with me had grown up in the North, and the man who was so courteous and helpful was from Mississippi. I suppose the northern man was trying to bend over backwards, making sure he didn't want to be associated with black people. I imagine the Mississippian wasn't worried because he knew how far he could go. I guess that was it, anyway.

Wilmot: Things don't always turn out as you expect. Blackwell: Right.

Wilmot: For sure, they're definitely not always very clear, "if p then q" kind of lines. It was also, I understand that when you went to go teach at Southern, it was also the first time that you'd been in an educational institution that was all black.

Blackwell: That's right, [pause]

Yes, I learned a lot about black people during that first year. For instance, there's what's sometimes called the Negro National Anthem, "Lift Every Voice and Sing." I had never heard that until I went to Southern! Just a lot of things like that, were just completely new to me. It was a learning experience for me. By the way, that Southern choir was one of the most, well, the most beautiful singing group I have ever heard.

Wilmot: Many people have described teaching to be this really transformative process, basically because it forces the teacher to learn as much as the students. Did you find that to be true? And not so much in terms of the subject matter, but in terms of how to communicate it to another human being in a way that they can hear it.

Blackwell: I have found that I always learn something when I'm teaching. And if you're trying to teach something to someone else, you need to understand it really well because you want to make it as easy as possible. And that means you have to be able to look at it

39

from all directions. Yeah, I've always found that even if I'm teaching a course for the tenth time, I still learn something every time I teach it. And I think I've become a little better teacher because of that.

Wilmot: Can you give me an example of this? Blackwell: No, I can't

Wilmot: Why did you leave Southern University and go to Clark in Atlanta?

Blackwell: It just sounded like a more interesting place to be. There was not only Clark, but there was Morehouse and Atlanta University all in the same region.

Wilmot: Was there more than two people in the math department there?

Blackwell: No. At Clark I got to teach advanced mathematics courses, or some, because the head of the math department was on leave. But I think if I had stayed, I would have been teaching only elementary courses at Clark, too.

But there was a cooperative atmosphere there. Clark, and Morehouse, and Morris Brown and Atlanta University cooperated and I had students from all four institutions in my classes. And there was joint seminar, that was run by me, during the year I was there and a man from Morehouse and a man from Morris Brown, and a man from Atlanta University. I made the right decision from that point of view. It was a richer experience than Southern.

Wilmot: In the seminar, you were teaching advanced math?

Blackwell: Yes, each one of the four of us would give a talk, and graduate students from Atlanta University would give talks, yes.

Wilmot: That sounds very exciting.

Blackwell: One of the fellows, the man from Morehouse, was an extraordinarily able teacher. I

learned a lot about teaching from him. By the way, my wife, whom I had not met at the time, had also studied mathematics with him. His name is Dansby, Claude Dansby. And there's a building at Morehouse now that's named after him. I was glad to see that, that he was really recognized as a distinguished teacher.

Wilmot: Do you remember the names of the other people who were part of that seminar?

Blackwell: Yes, Joe Pierce was from Atlanta University and Herbert Boggs was from Morris Brown.

Wilmot: So there was definitely a community of sorts.

Blackwell: Yes. Boggs introduced me to my wife, by the way. [laughs] Yes, it was a community, all right.

Wilmot: So it was in Atlanta that you met Mrs. Blackwell.

40

Blackwell: Yes.

Wilmot: I want to ask you, as far as these one year instructorships at Southern University and Clark College, did they contribute to your knowledge about how to work within an academic institution? You were a fledgling teacher.

Blackwell: Oh, I'm sure they did. As I say, I'm sure I was a considerably better teacher after that first year at Southern. You get a feeling for what to expect of your students and how to present things.

Wilmot: I have to check in with you about time, because we've been talking for about two hours and I can talk for the whole day. I know that you have commitments, would you like to stop now?

Blackwell: Yes.

Wilmot: Okay, we're signing off for today, [end of interview]

41

INTERVIEW 3: APRIL 25, 2002

[Minidisc 3]

Wilmot: Thank you. Okay, I think we're all right. So, good morning.

Blackwell: Good morning.

Wilmot: I had some follow-up questions from our last conversation.

Blackwell: All right.

Wilmot: And actually, one of them goes back to one of our first conversations. I wanted to ask you — and this goes back to Centralia, Illinois — when you applied for schools, for college, did you ever think of going to a black college at that time?

Blackwell: No. I had always assumed that I would go to the University of Illinois. And the only other college that I thought of was the college in Indiana where Ms. Seiler got me a scholarship. I've forgotten the name of that college now, but I told you last tune. [DePauw University]

Where I was growing up, for most people there were only two colleges. The University of Illinois and Southern Illinois University at Carbondale. And if you thought you could make it at Illinois, you went there. And if you weren't sure you could make it at Illinois, you went to Carbondale.

Wilmot: So it seemed that there were just two different levels of college education? Blackwell: Just two colleges that most people considered.

Wilmot: My next questions are about the Institute for Advanced Study at Princeton, my next follow-up questions. While you were at the institute, World War II was taking place. And my understanding from talking to you is that everyone at the institute was from other places. Not all of them, but many of them were actually from Europe where the war was happening. And I wanted to ask you, what was the atmosphere at the institute at that time around these issues?

Blackwell: I don't remember very much about the atmosphere. There was some conversation about the war, but mostly it was about mathematics.

Wilmot: The other thing that I learned about the Institute for Advanced Study is that while it was really — I think there's actually a quote from Abe Flexner, from Mr. Flexner who was a founder — he said that basically it was established with the idea that people could come and push the boundaries of the known.

Blackwell: Sure.

Wilmot: Just push the boundaries of the known. That was it. It's a really amazing idea to me that it was just a place where people could just think their thoughts and push their ideas. And at the same time, it was a place that was founded during a time when many Jewish

42

intellectuals were fleeing or coming from hostile environments in Europe and finding a place at the institute. And I wonder if you have — what your experience was around this.

Blackwell: I didn't really — . For me, the institute was a place where a lot of young mathematicians got together, and we talked mostly mathematics. It was a very friendly place. But it was pretty isolated, mostly, from what was going on in Europe.

Wilmot: Also, when you were at the institute, you told me you had a brief conversation with one of the faculty members there, Godel. My understanding is that he had this argument that was based on math for the existence of God.

Blackwell: I didn't know anything about that. I remember my conversation with Godel. I said something that, when I look back on it, was really kind of dumb. I said to him that it seemed to me that there are some propositions which, if they were undecideable, must be true. And he sort of shrugged his shoulders, "Oh, yes, yes, yes, we know that." And, of course, now that I look back on it, of course, every logician knew that very well, but it was just new to me.

Wilmot: Well, I think there's that moment oftentimes when you have a conversation with someone and you feel that you've made a great discovery and you share that with someone and you realize that it's been in the world for many years.

Blackwell: Yes.

Wilmot: I'm just going to take a minute and listen to us talk here on the CD, I'm a little

concerned about making sure it's working well. Hello. Dr. Blackwell, could you say some words?

Blackwell: Okay, what would you like for me to say? [laughs]

Wilmot: Well, I think whatever you think is important to say is the point of our conversation, so — . I would actually like to ask you the question about Godel again because I'm worried that it caught everything else we talked about but I think I may have turned it down about at that time. So if I could ask it one more time, I'm sorry. What was the conversation with Dr. Godel like?

Blackwell: Well, we were walking along and I said to him that it seemed to me there were some propositions which, if they were undecideable, must be true. He said, "Yes, yes, we know that." [laughs]

Wilmot: And you felt like you were telling him something new.

Blackwell: Well, I was telling him something that was interesting to me. And as I later found out— of course, that was page one in what logicians knew about things.

Wilmot: So what do you think of that possibility of using the language of math or logic to theorize the existence of God?

43

Blackwell: [pause] There you've got me. I'm not religious, it wouldn't occur to me even to ask that question. [Nadine hands Godel info] Wow. Nadine, I'm not going to have any comment on this, [laughs]

Wilmot: I think I was more trying to verify that I was actually in the same ballpark with the kind of work that you do, if that makes any sense. Just in terms of the tools, that's all. There were some other faculty member at IAS who we were discussing earlier, and in addition to Godel and Von Neumann, was there anyone else who you remember interacting with?

Blackwell: No. I had some interaction with Sam Wilks at Princeton, not a lot. But he taught the main statistics course there. I sat in on that course and had a few conversations with him. In fact, Sam Wilks recommended me for some job in Washington, I've forgotten now what it was. There was a war going on, and some of the people at the institute and at Princeton were very active in government work. Von Neumann was one, Sam Wilks was another one. I've forgotten now what it was that Sam Wilks recommended me for.

Wilmot: We talked a little bit about that. Did you have any interaction with Albert Einstein while he was there and you were there?

Blackwell: No, just a few polite courtesies, that's all. Nothing serious.

Wilmot: When we were talking about this last week, there was someone you mentioned who was in the year — who didn't appear on my list but was there.

Blackwell: It may have been Ralph Phillips.

Wilmot: Yes, I think he was in that year preceding you.

Blackwell: Yes, Ralph Phillips. Then he must have been just visiting the institute for a day or two or something, because I certainly had that conversation with him.

Wilmot: There also may have just been some overlap in the years, in the academic years. Was there anyone else in that previous year that you recalled talking to or working with closely?

Blackwell: Well, [Paul] Erdos, of course, was probably also there in '41, '42, but he's not listed there.

Wilmot: No, he isn't.

Blackwell: He's not listed here, but he was certainly at the institute often during that year. And Paul Halmos, he's probably listed. Yes, he's listed. And I don't see any of the others.

Wilmot: Okay, I would like to move on to Howard. You told me this really interesting story

about how when you were doing your drive, it was like a drive through the country to propose yourself to the different math departments at the different colleges. And you were at Howard, and you interviewed with the director of the department and there was someone else listening.

44

Blackwell: That's right, I've forgotten the name of the head of the department, but Professor

Woodard was sitting at the next desk and listening. And at the end of the interview, he asked me a couple of questions, and I left. And a year or two later, the head of the department retired and Woodard became head of the department. And right away wrote to me and asked me if I was interested in coming to Howard, invited me up for an interview. And subsequently, offered me a job.

Wilmot: Who else was in the math department when you arrived there?

Blackwell: Woodard was the head of the department. Elbert Cox was also in the math department. There were a couple of young instructors in the math department. Woodard and Cox were the only senior faculty members in the department at that time. By the way, Cox was the first black man to get a PhD in mathematics; and Woodard was the second black man to get a PhD in mathematics. So they were the first two and they were the only senior math members at Howard.

Wilmot: When you arrived at Howard, what was the state of the math department? And I should qualify that question by saying in terms of the kinds of resources that you had to operate with, and the kind of intellectual level of the work there.

Blackwell: Well, there was a math library. It was — [pause] I don't remember going outside of it very often. Mostly, I was interested in mathematical statistics at that time. And, of course, the Howard library carried the leading math stat journal.

But I went outside the university quite a bit for mathematical contacts. My friend, Abe Girshick, for instance, was a statistician at the Department of Agriculture. And, I was a consultant at the Operations Research Office in Washington, and there was some interesting mathematics going on there.

Wilmot: Operations Research Office? Blackwell: Yes.

Wilmot: I'll have to ask you some more questions about that.

Blackwell: Well, I won't be able to tell you very much because I don't remember very much. I was reminded of it by your list, because you had the name Nicholas Smith on there. He was the head of the group at the Operations Research Office that I worked with.

Wilmot: Do you remember what kind of work you were doing there? Blackwell: Mostly game theory and optimization theory.

Wilmot: And may I ask, what were the potential applications, around what kind of applications?

Blackwell: Mostly having to do with wars and fighting. I doubt that we were very helpful, but it

was mathematically interesting, anyway. Yeah, one of my good mathematics ideas came out of thinking about wars and how to fight them. Just a minute, [tape interruption]

45

Anyway, thinking about that kind of conflict led to what I call "approachability." Other people have called it "approachability theory." I just introduced the concept of a set being approachable by one player or excludable by another player. And it's been used by a number of people working hi game theory. The point is that it came out of my thinking about wars in connection with my work at the Operations Research Office.

Wilmot: So, to what extent did you feel like this, the problem you were working with, could translate into a real, into an application in a war or a battle?

Blackwell: I didn't think that it could, and I certainly don't think now that it could, very well. For me, it all went the other way. If you can look at a real situation and translate that into something that's interesting mathematically, that's sort of what I like to do. Not going the other way, taking mathematics and interpreting it to the real world. But taking something from the real world and see if it suggested some interesting mathematics.

Of course if you're going to make it Interesting mathematically, you have to change it and simplify it. So probably the changed/simplified situation won't correspond to anything in the real world at all. So, what you're doing is not practically useful. So, I don't think this work on a multi-component attrition game is militarily useful, but it's mathematically interesting.

Wilmot: When you were working for the Operations Research Office, did you get the sense that people outside of your immediate statistics unit found your work valuable? Valuable is the wrong word. Useful. Or that they understood it?

Blackwell: I don't think very many people outside of the group knew what we were doing at all. [chuckles] For example, I don't think that this particular paper, "On multi-component attraction games" [Naval Research Logistics Quarterly, 1954], has ever been cited by anybody else. As far as I know, it hasn't been. But it led me to a couple of theorems that have been cited.

Wilmot: Which were these theorems?

Blackwell: Oh, a theorem on approachability, let's see — . [shuffles though papers] This is the paper, "An Analog of the Minimax Theorem for Vector Payoff." Let's see, here I introduce the concept of approachability. Yes, "approachable" and "excludable." Now those concepts of approachable sets and excludable sets have been used by several people in game theory.

Wilmot: Okay. I think you've really opened up a place where I may need to develop more

questions around this. So, I'm going to return to Howard now, because that's where I'd like to talk today. But this is very interesting, this is very interesting. I wonder how you had tune to consult while you were also teaching at Howard full-time, that's pretty amazing.

Blackwell: Consulting was just a part of my mathematical activity. While you're teaching, you're also learning and you're — . My consulting was just talking to other people. I think I helped them and taught them something, but I sure learned a lot from talking with them. The fact that I got paid for it was just something extra; I would have done it for nothing because it was part of my mathematical activity.

46

Wilmot: Were you teaching a full complement of courses at that time?

Blackwell: Yes, I think the standard teaching load at Howard was twelve semester hours. And yes, that what I taught. That's what I taught, I think, most of the time when I was at Howard. Maybe when I was head it was reduced to nine, or something like that. Yes, I had a substantial teaching load.

Wilmot: And when you went to Howard, did you find that there were — that the math department, and you may think of this because at the time you were coming from Southern, and also from Clarke. Did you feel that the math department had resources?

Blackwell: The main mathematical activity at Howard when I went there was the activity

sponsored by Woodard, who I mentioned earlier. Woodard had a mathematics seminar at which, well, we met once a week. And he would speak, and I would speak. Sometimes Professor Cox would speak and sometimes, one of the graduate students would speak, mainly about topology, because that was Woodard's main interest.

Wilmot: Topology? Blackwell: Yes. But as I said before, I went outside Howard for a lot of mathematical activity.

Wilmot: I was trying to speak to financial support of the math department, how well the university kind of gave the math department financial support.

Blackwell: Well, I — . They paid our salaries! [laughs] Howard had, what I thought, and still think is a very good travel policy. Howard would pay your expenses to attend one professional meeting a year, whether you gave a paper there or not. Just so you could go there and find out what was going on. Except for that, I don't remember much support, but you don't need other support.

Wilmot: I understand. Was the math department housed in the College of — ? Blackwell: It was in the College of Letters and Science.

Wilmot: Did you find students at Howard who were really excited about math? That first year?

Blackwell: Now, most of the students that I taught were, of course, not mathematics majors. And I had many very bright students, yes, at Howard. And there were not many graduate students, but among them were some who were just as good as the graduate students I've encountered here at Berkeley.

Wilmot: What classes did you teach?

Blackwell: I would say that a typical course load for me would be a freshman course — . No, maybe two pre-calculus courses, one calculus course, and one more advanced course which might be a senior level course or it might be a graduate level course. That would be a typical teaching load.

Wilmot: So that's four classes.

47

Blackwell: Four classes, each one three units, so maybe twelve units all together. Wilmot: Okay, were there courses that you particularly enjoyed teaching?

Blackwell: Yes, I think the course that I liked best was an upper-division analysis course. But I liked teaching them all.

Wilmot: While you were at Howard, did you remain active in your fraternity, the Alphas?

Blackwell: You know, I don't think I was. I had never thought about that, but no, I don't remember being active at all at Howard.

Wilmot: I understand that you became chair of the math department within three years after you arrived there at age twenty-eight?

Blackwell: Something like that, yes.

Wilmot: And that you remained chair until you left Howard in 1954. Blackwell: Yes.

Wilmot: And I wanted to ask you, first about tenure at Howard, and then more about the

chairship. So those are my questions. But first about tenure, how soon did you get tenure at Howard and what was that process like for you?

Blackwell: I think I got tenure after three years. But it was completely painless. I didn't ask for it or have anything to do with it. I was just sure that I would get tenure eventually, and I was surprised by how quickly it happened, but I — . It was nothing like the same somewhat painful process that it is here, stress-inducing. No, it just was not like that at all. There were people at Howard who had been there for ten or fifteen years and who did not have tenure. But we were pretty sure that they would remain there as long as their teaching continued to be satisfactory, and it was. So, mostly, getting tenure was a question of how much research you did. And, I was just pretty sure that I was going to do enough research that I would get tenure, and as I say, I was surprised that it came within two or three years.

Wilmot: And they just let you know, "By the way, you have tenure now." Blackwell: Yeah, that's right!

Wilmot: Okay, good. So as chair at the — I mean, I must say, twenty-eight is a young — . I know times are different, but to me twenty-eight is a young age to be chair. My understanding is that you became chair when Dr. Woodard retired?

Blackwell: That's right.

Wilmot: So he kind of made a decision to step away?

Blackwell: Well, he had reached, maybe had reached mandatory retirement age. I think that's probably the way it was.

48

Wilmot: Blackwell:

Wilmot:

Blackwell:

Wilmot:

Blackwell:

Wilmot: Blackwell:

Wilmot:

Blackwell: Wilmot:

And Cox was still there?

Cox? Yes.

Okay. So what major projects and initiatives did you undertake as chair?

I don't remember. I would have said that I didn't do anything, but one of the later chairs told me that I had done some things when I was chair, but — I don't remember. I think my idea was just to continue running the department, continue letting the department run the way Woodard had let it run. I thought it was a pretty satisfactory department.

So, the position of chair is more of a custodial kind of position than a "visioning" position?

No. My main job as chair was to make sure that all the classes ran. See, this was right after World War II, in 1947, '48, '49. A lot of students were coming in. So, we had more students in the math department than the regular faculty could teach. So, it was my job to find visiting faculty to teach those classes. And that was my main activity as chair was just to make sure there was a teacher in front of every class. And there were certain people out in the community that I called on to come and teach at Howard. And occasionally, there was a vacancy in the department or the dean approved a new position, so it was my job to go out and find someone for that new position. And I did that, I initiated two or three appointments while I was there.

Where did you look for the ones who were not full-time instructors? There were two appointments that I have here, but when you were just trying to get people in front of the classes, where did you look?

There were people who applied for positions at Howard. One of them was a full-time teacher at another institution.

So you can borrow people from other institutions?

Yes. See, there was a teacher's college right across the street from Howard, and sometimes we would get people from the teacher's college across the street to come in and teach.

As far as curriculum, when you became chair, did you make any changes in math major requirements? Or were there priorities that you felt needed to be more front and center academically for students?

I'm pretty sure that I made no changes.

Okay. The two people that I understand that you brought, that were brought on while you were chair, were William W.S. Claytor and Jeremiah Certaine.

That's right. Yes.

Were you instrumental in bringing them to Howard's math department?

49

Blackwell: Yes, indeed. Claytor was an extremely able man. He had been Woodard's student. And he went on to get a PhD under the same man that Woodard had studied under, J.R. Klein. There was some kind of tension between Claytor and Woodard. And Woodard would never appoint Claytor. Woodard appointed me when, looking back on it, I think he probably should have appointed Claytor. But when a vacancy came and I was chairman, I did appoint Claytor, got Claytor appointed.

Claytor had been at the University of Michigan as a post-doctoral fellow for several years. And I believe a serious attempt was made to appoint him there, but at that time, it was very hard for a black man to get an appointment in a white university. So, that did not go through, but a lot of people were in favor of it. Claytor was one of the best lecturers I ever heard in my life — very clear and well-organized.

Wilmot: What was his area of expertise?

Blackwell: Topology. Same as Woodard, yes.

Wilmot: Was it very easy to bring him on?

Blackwell: Yeah, in the sense that there was no trouble getting him appointed.

Wilmot: And in terms of getting funding for his position?

Blackwell: Well, the funding came first. You would ask the dean for a new position and if the dean gave it to you, then you would go out and look for somebody.

Wilmot: It says here that not counting theses, Dr. Claytor was the first African American to

publish mathematics research, and that he was awarded a Rosenwald fellowship. And that was how he did his post-doctoral studies at the University of Michigan.

Blackwell: Oh, Claytor got a Rosenwald fellowship! I also got a Rosenwald fellowship. Wilmot: I know.

Blackwell: Ahh, yes, I had forgotten this! In 1941, Paul Halmos told me, "There's this black mathematician who is up at Chanute Field, which is only thirty miles north of the University of Illinois. Let's go up and meet him." So I remember now, Paul Halmos drove me up and I met Claytor there. [Looking at brief bio of Claytor] I hadn't seen this before.

Wilmot: Did you find his ideas — his mathematical inquiry very engaging or interesting? Did you have the opportunity to engage with him intellectually or were the areas of work so different?

Blackwell: Yeah! We had many conversations — mathematical. Claytor was a perfectionist in a certain way. He was very interested in music and he — let's see, he had a device for reproducing music. He had some fancy record player, and he would adjust it to get the tone just right. And he would finally adjust it to the point where he couldn't make any adjustments that would improve it as far as he could hear. But then he borrowed equipment from the physics department and let the equipment listen to it! [laughter]

50

And he kept adjusting it until the equipment couldn't hear any changes when he tried to improve it. That was the kind of perfection that he was looking for.

Wilmot: I wonder what kind of equipment it is that can do that.

Blackwell: I don't know. Also, we used to play golf together. Again, Claytor was a perfectionist. He had all these rules that he tried to follow in adjusting his drive and making sure he did everything just right. He was a pretty good golfer, but — . Excuse me for a minute, I'll be right back, [leaves room and returns]

Wilmot: You know what, Dr. Blackwell, it's very interesting that they, in this writing it says they think you met — or they've written that you'd met Dr. Claytor at an army base at Chanute Field, about twenty miles from Urbana, Illinois.

Blackwell: Yeah, that's what I was telling you!

Wilmot: So, Jeremiah Certaine, how did you meet Dr. Certaine?

Blackwell: I don't know, I'm a little vague about — . I think he simply applied for a job at Howard and he came down for an interview, and we appointed him.

What I remember more about him is how he left. There was a man named J. Ernest Wilkins, you probably — I don't know, did you ever hear of him? No, well, he got a PhD in mathematics at the age of nineteen, or something like that. Really precocious, very good mathematician, black fellow from Chicago. And Wilkins was working for some company, I believe it was the American Optical Company up in New York. And I had been trying to get him to come to Howard.

Wilkins came down to give a lecture at Howard and I talked to him about coming to Howard and he wasn't particularly interested. And he left, and very soon after that Jerry Certaine resigned and went to work for the American Optical Company. So what Wilkins had done, when he came to Howard, was to steal Jerry Certaine from us!

Wilmot: Yes, he recruited him. Blackwell: He recruited him, that's right, [laughter] Wilmot: Did Jerry Certaine work with you?

Blackwell: Not especially closely. I don't think Jerry was at Howard more than two or three years. Does it say how long he was there?

Wilmot: It says 1 947 to 1 95 1 . Blackwell: Yes.

Wilmot: So again, it seems like that first year that you became chair, you brought him, you brought them both right on to deal with that shortage of instructors.

Blackwell: Yes, that GI. influx was a big thing.

51

Wilmot: Did you find that more people were interested in hard sciences and mathematics as a result?

Blackwell: I don't know if they were more interested but they certainly worked harder. The

students in 1947 and '48 were a few years older than the typical students. They were quite serious and hard-working. It was a pleasure to teach them.

By the way, the building facilities were so over-crowded that we had to extend teaching hours. The first classes started at seven o'clock in the morning, and we had Saturday classes. Then, after a few years, the enrollments dwindled a bit, facilities expanded, so we could give up Saturday classes, and seven o'clock in the morning classes. But I remember one fellow came to me and said, "May I have a seven o'clock class?" Now, why do you think he wanted a seven o'clock class?

Wilmot: I really don't know. Blackwell: Parking. Traffic. Wilmot: Oh!

Blackwell: Traffic was light driving to work at that hour in the morning, and parking was easy at that hour in the morning.

Wilmot: Did you oblige? Blackwell: To some extent. I think the dean didn't want us to have seven o'clock classes.

Wilmot: So, I think also because as a chair you had taken on more administrative

responsibilities, did you interact with other parts of the College of Letters and Science? hi particular, what I'm trying to say, and I'm saying it in my usual roundabout way, is how did you navigate the administrative workings of the university to get what you needed for the department?

Blackwell: Well, Howard was a very authoritarian system. You just worked through your dean, you asked your dean for what you thought you needed and he would decide, yes or no. There wasn't a lot of political maneuvering, at least not that I knew about.

Wilmot: Who was your dean?

Blackwell: J. St. Clair Price was the dean.

Wilmot: Okay, did you get along well with him?

Blackwell: Yes.

Wilmot: Can I ask, what was his academic background or training?

Blackwell: I think his PhD was in education, but I'm not sure.

52

Wilmot: Did you have the opportunity to interact with president of the university, Mordecai Johnson?

Blackwell: No.

Wilmot: Did you interact with Frank Snowden?

Blackwell: Just a bit, yes. He was on a floor below me in Douglass Hall. How'd you happen to ask about him?

Wilmot: Well, to be honest, I was looking through this book, and I was trying to find deans. I didn't find your dean, the College of Letters and Science, and I was just trying to find any dean who was there at that time and he was the dean of the College of Liberal Arts.

Blackwell: Ah, yes. But I think he became dean after I left. Yes, he was just professor of classics when I was there.

Wilmot: So, you said your offices and your department were based in Douglass Hall? Blackwell: Yes.

Wilmot: And you mentioned that you shared that space with a number of other people, including John Hope Franklin?

Blackwell: Yes, and E. Franklin Frazier, and Rayford Logan, and Sterling Brown, and Alain Locke, and — there were a number of very distinguished people there.

Wilmot: Did you all have lunch sometimes? Or just, did a couple of you get together and have lunch across disciplines?

Blackwell: Not very often, no. No, we would just have informal conversations in the hallway, that's all.

Wilmot: Were there any of those that you just mentioned who you formed friendships with?

Blackwell: No, the main ones that I was friendly with were the other people in the math department.

Wilmot: And did you spend time together as individuals, or with their families?

Blackwell: Both. Claytor and I and our wives, and another fellow named John Doggett and his wife all used to get together socially. And Claytor and John Doggett and I would go out and play golf together.

Wilmot: Who was John Doggett?

Blackwell: John Doggett was a young instructor at Howard. He got his master's degree at Howard, and just stayed on as an instructor. But he got interested in computing and learned how to write programs for a computer. At that time, computers were just becoming popular. Lots of people wanted to use them, but very few people knew how to program for them.

53

So, John Doggett got a job at the National Bureau of Standards as a programmer, paying nearly twice what he was being paid at Howard. So he left and went to the National Bureau of Standards. And he didn't stay there very long because some private institution hired him at a much higher salary than that!

Wilmot: Wow. Now, during this tune I understand your wife, Mrs. Blackwell — was she working while you were teaching and consulting with the Office of Operations?

Blackwell: No, she didn't have a job while we were at Howard. But during that period, we had seven children, so it's not quite right to say she didn't have a job! [laughs]

Wilmot: Can you tell me when your first child was born?

Blackwell: Yes, September, 1945.

Wilmot: That was a year after you came to Howard.

Blackwell: Yes, we got married right after I came to Howard.

Wilmot: And you told me that you met, you actually met through — . Oh, wait, I had it written down.

Blackwell: Well, we were introduced by — Wilmot: — Joe Pierce?

Blackwell: No, Herbert Boggs. You want me to tell you the circumstances? Okay. I was to teach an elementary physics course in the summer of 1944. And Ann came to register for the physics course, thinking that an old friend of her family's named Eagleson was going to teach the course. Well, Eagleson had taken the summer off, he, sort of, owned the physics course, and asked me if I would teach it and I had agreed to.

So, Ann showed up to register for this physics course under Eagleson. And Boggs and I were registering people, and when she found out that Eagleson wasn't going to teach it, she started to leave. But Boggs persuaded her to stay and introduced me to her and told her that I would be a reasonable teacher. So, she took the course. But if hadn't been for Boggs's intervention, she would have gone elsewhere because she was interested in the course mainly because Eagleson was going to teach it.

Wilmot: So she was also a mathematician, you told me?

Blackwell: No, she was a chemistry student at that time. In fact, she was teaching chemistry. Wilmot: And was it easy for you to go from math to physics and teach physics?

Blackwell: No, it was hard. And I think the way I taught the course, it was mostly mathematics

anyway. As I remember it, the students helped me a lot in doing the experiments. I got a few of the students who were pretty good at setting up the equipment and making the experiments go, so that's the way that all worked out.

54

Wilmot: Okay. So, your first child, can you tell me the name of your first child? I'm jumping a little bit to focus on family history.

Blackwell: Yes, that's all right. Her name is Ann Louise, my wife's name was Ann Elizabeth, and my first child was named for her Aunt Louise. I'll just go on down the list. Our second child is a daughter, Julia Madison. Julia was my wife's sister, and Madison was my wife's maiden name. Then, David Harold, named for me. Then, Ruth. I'm not sure where the name Ruth came from. Then, Grover, named for my father. Then Vera. I think we just liked that name, I'm not sure where that came from. Then Hugo. I liked that name. An old friend of my father's was named Hugo, hi fact, the man who had promised me a job teaching in the elementary schools in southern Illinois was named Hugo. And, there was a mathematical friend of mine named Hugo Steinhaus. Anyway, the name Hugo came up. Those seven children were born in about a nine-year period, something like that. Then there was a gap of several years, and my youngest daughter Sara was born then. I'm not sure where the name Sara came from.

Wilmot: And she was born here at Berkeley? Blackwell: Yes, yes, she was born here at Berkeley. Wilmot: Thank you for sharing that with me.

Blackwell: Oh, you're welcome. Sara is the one who told me [when she learned of this oral history], "Ooh, you get to make things up!" [laughs]

Wilmot: Well, you are totally at liberty to make anything up you like. What I really appreciated was the last time we met and you said, "Well, this is approximately so, you understand, it's approximately so!" I thought that was very good. Goodness, so you were really building and raising your family during those Howard years.

Blackwell: Yes, those were busy years.

Wilmot: And where did you live in D.C., your family, where did you all live?

Blackwell: The first year we rented a house. The second — . Then for four or five years we lived in an apartment building in northeast Washington. And for the last two to three years we lived in a house that we bought in Washington. Again, in northeast Washington.

Wilmot: I'm not so familiar with Washington D.C.'s neighborhoods; I'm having trouble placing what kind of neighborhood it was.

Blackwell: Well, the house was near the Maryland line. As I remember, about a thirty-minute commute from Howard University.

Wilmot: Did you drive in? Blackwell: Oh, yes. [laugh]

Wilmot: I have to get my transportation questions in.

55

Blackwell: Yes.

Wilmot: Well, I have a question for you which is — . And it's one of those questions that you may wish to throw out. But I wanted to ask you what a typical day was like while you were teaching? And I'm thinking when you became chair, I'm thinking around that time.

Blackwell: Nadine, I don't remember that, [laughs] It's possible that all my teaching was on

Monday, Wednesday, and Friday and on Tuesday and Thursday I went to the Operations Research Office. That's possibly the way it was.

Wilmot: And maybe may have taught a 7:00 a.m. class on Saturdays.

Blackwell: No. [laughs]

Wilmot: I understand that it was while you were at Howard that you met Abe Girshick?

Blackwell: Abe Girshick? Yes.

Wilmot: I understand that that was also the time when you kind of fully made that turn.

Blackwell: Into statistics, that's right.

Wilmot: Can you describe that meeting for me?

Blackwell: Sure.

[end of interview]

56

The Blackwell family on the front steps of their barn in Philo, Calfornia. Top row: Hugo, Tootsie, Sara, Grover Middle row: Ann, Julia, Moon, Vera Front row: David and Ann

57

INTERVIEW 4: MAY 2, 2002

[Minidisc 4]

Wilmot: Good morning, it's May 2 and I'm here with David Blackwell and this is interview

number four. Last time when we stopped we were talking about your years at Howard. You were beginning to tell me about Abe Girshick. I wanted to ask you a few more questions about Howard.

Blackwell: Sure.

Wilmot: Do you remember your first day on Howard's campus when you first arrived there?

Blackwell: No, I don't remember when I arrived there to take up the job, but I remember the first day I visited Howard, I told you about that earlier.

Wilmot: You did, you did. This is kind of a small question, but what kind of car was that first car you bought for visiting all the colleges on the East Coast?

Blackwell: Yes, that was a 1939 Buick Century. It was a fairly big, fancy car. In my economic position, I shouldn't have had a car like that, but I did. I got that car while I was a graduate student at Illinois.

Wilmot: I was in touch with the archivist and finally he got back to me from Howard and he gave me this list of instructors and professors who were there while you were. And, you told me about most of the professors who were at Howard's math department the years you were there.

Blackwell: Yes. Some of them I don't remember myself anymore. Do you have any questions?

Wilmot: Yes, I do have some questions. I was just wondering if there was anyone there who you — in particular, Professor Butcher?

Blackwell: Yes, George Butcher was a Howard University graduate who went on to get a PhD at the University of Pennsylvania, studying with J.R. Klein. The same one who had been the thesis director for Woodard and Claytor. So Butcher was the third in that series. He came back to Howard, yes, hi the 1950s and stayed there until he retired. He still lives in Washington, D.C.; he's long since retired, though.

Wilmot: What was his area of math?

Blackwell: Topology.

Wilmot: Topology?

Blackwell: Yes.

Wilmot: They all were topologists.

Blackwell: Yes, as I say, they all studied topology under the same man, J.R. Klein at the University of Pennsylvania.

58

Wilmot:

Blackwell: Wilmot:

Blackwell:

Wilmot:

Blackwell:

Wilmot:

Blackwell: Wilmot:

Blackwell:

Wilmot:

Blackwell:

Wilmot:

Blackwell:

And are there any of those instructors that you remember and wanted to say anything about? You talked to me already about Doggett.

Yes, Doggett and Claytor and Certaine.

Okay. You told me that you found them, many of them you found from the teacher's college across the street?

None of these came from the teacher's college across the street. These were people who had regular appointments at Howard, but almost every year, I had to go out and find other people for temporary appointments and they're not listed here.

Okay, they were supplementing the regular appointments. Yes.

Can you tell me, what about a person made you think they might be a likely instructor? Like, good at math, good at teaching math?

Most of them I got because they were already teaching math. One of them — I don't remember much about that.

Was there a sense that people were happy to have instructor positions in Howard's math department?

Oh, I think so, yes. Yes, I think that some of them came to expect that they would be called and they were rather hoping for an overflow of students so that it would happen.

Yes, for the job. Also, I think in one of your previous interviews, you described Howard University as every black scholar's dream.

Yes.

Can you speak to that?

Well, it was just — . That's where many of the major black scholars were. That's where E. Franklin Frazier was; that's where Alain Locke was; Sterling Brown was there, and — . Well, there was a ranking among black colleges that just everybody accepted, and Howard was at the top, and I think rightly so.

You've also described Howard as a very authoritarian environment, what does that mean?

Why, it means that you did what your boss said. The head of the math department really determined what courses the other people taught and determined what appointments to recommend. The dean of the Letters and Sciences approved, or disapproved, everything that the head of the department said. An instructor wouldn't dream of going over the head of the math department to the dean. The head of the math department wouldn't dream of appealing the decision of the dean to the vice president, just very hierarchical.

59

Wilmot: Okay, I'm going to jump just a bit, and before I go back to Howard, was that different from Berkeley?

Blackwell: Well, yes. Professors here are much freer than they were at Howard. For example, at

Howard the head of the department determined the textbooks, and everybody — if there were several sections, they all had to use the same textbook. But here, if there are two different sections of the same course, each instructor chooses his own textbook.

Wilmot: Granted, I realize that you were at Howard in really a different era, almost ten years

difference, so I can imagine that there'd be different — . I mean of course they were very different places.

Blackwell: I suspect that it is quite different now, yes.

Wilmot: So in that very authoritarian environment, you were the chair. You were the authority, you were the one who was putting down the rules — at Howard?

Blackwell: Yes.

Wilmot: Right. When you thought about basic math standards for undergrads, that was primarily who you were teaching, right? Undergrads?

Blackwell: Yes.

Wilmot: When you thought about basic math standards for undergrads, what was your frame of reference? What were you thinking about making sure people had before they went out into the world?

Blackwell: We had a fairly standard curriculum. Calculus came in the second year and upper- division courses came in the junior and senior year. And that was not very different from the University of Illinois. Nowadays, of course, calculus comes in the freshman year.

Wilmot: Were basic math standards higher then than they are now, do you think?

Blackwell: Oh no, basic math standards were lower then than they are now. Actually, even then, I would say that Howard University seniors took courses that would be given at the junior level here at Berkeley. So that Howard was about a year behind Berkeley. Level for level, though, the Howard students were just as good as the Berkeley students, it's just they were at a slightly different level.

Wilmot: How would you talk about a career in math with students at Howard?

Blackwell: I don't think I ever talked about a career in math at Howard. As I remember it, most of the math majors expected to be either high school math teachers or to get a job in the civil service in Washington. I don't remember, but I don't think that any of them had particular problems getting a job.

Wilmot: So you wouldn't say, "You, too, can become a professor of mathematics, like myself."

60

Blackwell: No, and I don't think most of them were interested in that.

Wilmot: It's a very uncommon occupation, becoming a professor of math, deciding to teach and think about, and be in a world of ideas.

Blackwell: Yes, and certainly when I started college it's not what I expected to do, but my focus just gradually shifted.

Wilmot: I think you also said you made classes much harder than they had to be. And you talked about this, I think, mostly when you were at Southern. You said that you made classes much harder than they had to be, as a new teacher. And by the time you were at Howard, you were no longer a new teacher, but do you still think you were making classes harder than they had to be?

Blackwell: No, I think I learned in one or two years what the appropriate level was.

Wilmot: As a result of your time at Howard, do you think that black educational institutions are very important?

Blackwell: Yes, I do. I think there were, and probably still are, many students, many black students, who are slightly uncomfortable around white students and they feel more willing to express themselves among other black students.

Wilmot: You were at Howard at a very, as far as I understand from looking at your work, at a

very formative time in your work. And I know that much of your mathematical activity occurred outside of Howard. But in terms of your own development for yourself and for your family, do you think that that was a very important institution for you to be at? And the fact that it was all black was very important?

Blackwell: The fact that it was all black was not very important, but that was certainly a very

important time for me, yes. My interests were basically formed at that time, while I was at Howard. It was not a handicap, being at a black school, at all.

Wilmot: In terms of the role that historically black colleges and universities currently play within the black community, can you talk a little bit about that?

Blackwell: I've been away from that a long time and things have changed so much that I don't really have any useful opinions on that.

Wilmot: You can always volunteer opinions, you know. They don't have to be useful.

Blackwell: I know, [laughs]

Wilmot: I want to ask you, you said that you still go back to Howard on occasion.

Blackwell: Yes.

Wilmot: Yourself and your family, or maybe just yourself?

61

Blackwell: No one in my family has been back since my daughter Sara moved from Washington. But, I go back to Washington on business occasionally, every few years, and whenever I go I always make a special effort to visit Howard, see my old friend Jim Donaldson, who was head of the math department for many years and is now dean of Letters and Sciences at Howard.

Wilmot: Did any of your children go to Howard?

Blackwell: No.

Wilmot: Did any of your children go to historically black colleges?

Blackwell: No, they're all Californians.

Wilmot: Did you want them to?

Blackwell: No, I left that pretty much up to them. I don't think they would have listened to me, anyway.

Wilmot: You've always said that your family raised you with the attitude that definitely you're going to college; college was definite.

Blackwell: Yes.

Wilmot: And did you raise your children in the same way, with the same attitude?

Blackwell: Well, yes, certainly, it was assumed that they would all go to college and they all did. Not all right away.

Wilmot: You know how at UC Berkeley there was a loyalty oath controversy? Blackwell: Yes.

Wilmot: And I was wondering if there was ever a corresponding atmosphere at Howard

University where people were concerned about being accused of anti-American beliefs?

Blackwell: There was very little of that at Howard. No, there were a few people on the faculty who were said to be Communists, but no one bothered them and that was just something interesting about a person, if he was a Communist. My dates are a little vague, I'm not sure just when [Joseph] McCarthy flourished. I guess it was around 1950, wasn't it? No, I don't remember that touching Howard, especially.

Wilmot: Yes, I think it was 1950 through mid-1950s. And I know that the loyalty oath here was 1949. It's funny because Al Bowker, in his memoir, in his oral history, says Stanford benefited from Berkeley's loyalty issues because key people left and went to Stanford. But we'll talk about that another time. Yes, and I think it was also during that time — I think it was also about the time when W.E.B. DuBois was — I guess he was targeted as a Communist and he chose to go away. People could basically have their political beliefs and weren't censured in the same way that they may have been had they, if they were at Berkeley?

62

Blackwell: That's right.

Wilmot: I wonder. Because my sense is that Howard also received some federal money, and maybe that wasn't the, kind of, crux of the issue as to whether or not you received federal money, but I kind of wonder about that.

Blackwell: It seems to me that Howard was just beginning to receive a lot of federal research

money at the time I left, around 1953 or '54. Before that, I don't think federal research money was a big part of Howard's budget. Of course, Howard is a kind of federal university, you know. It gets its biggest appropriation from the federal government.

Wilmot: Were you instrumental in bringing some of the research money to Howard?

Blackwell: Just before I left, I did have a small research contract, yes. But I don't think it amounted to much.

[Wilmot: Tell me about the research contract.

Blackwell: It was for the summer of 1953, to do research in game theory. One of my papers that I like best, "On Optimal Systems," has a footnote saying that it was written under the contract]

Wilmot: While you were at Howard, do you remember hearing about the loyalty oath at UC Berkeley? From your friends in statistics?

Blackwell: Vaguely. Again, I was just not very political at that time.

Wilmot: I understand that it was while you were at Howard that you met Abe Girshick and fully made that turn into statistics away from — . Not away from math, but it was somewhat of a departure. And I wanted to ask you how you met Abe?

Blackwell: How I met Abe? When I was in Washington at Howard, I looked all around

Washington, D.C., for mathematical activities. And I went to a lecture, sponsored by the Washington chapter of the American Statistical Association, by Abe Girshick. And he spoke about Wald's work in sequential analysis and mentioned this equation called Wald's Equation. Well, that equation surprised me, it was new to me and I didn't even— I didn't think it was true.

So I went home, and thought I had found an example showing that it wasn't true. And I wrote to Girshick, he was at the Department of Agriculture, explaining my counter example. Well, instead of just dismissing it, because my counter-example was wrong by the way, he invited me to have lunch with him.

Wilmot: So it was wrong but interesting in the way it was wrong.

Blackwell: Well, apparently so. Anyway, he invited me to have lunch with him, and I did, and we discussed it. And he got me interested in sequential analysis, and had various problems that he was working on, and I started working on some problems with him.

Wilmot: Can you tell me what Wald's equation is?

63

Blackwell: If you're in a gambling house where every gamble is fair, there are systems that you can use that are certain to make money. For instance, if you and I keep betting on a fair coin. I get a dollar every time it comes down heads, you get a dollar from me every time it comes down tails. If we keep tossing that coin, sometimes I'll be ahead and sometimes you'll be ahead. So eventually, there will come a time when I'm ahead. If I just wait until that time comes and then stop, then I stop while I'm ahead; I've won money from you. So, as I said, there are systems where you are certain to win money.

However, before you are ahead, you may be behind by a certain amount. Wald's equation says, among other things, that you could be a large amount behind before you win. In fact, the amount you're behind before you win has an expectation of "infinity." Or, yes — you expect to be an infinite amount behind before you win, that's a consequence of Wald's equation. And, as I say, that seemed quite remarkable to me. Wald didn't put the equation in that way. The way he put it was, if you don't expect to be an infinite amount behind, there is no winning system, the gain is fair. So that's what the equation says, and I eventually proved it under more general circumstances than Wald did. I extended Wald's equation, but I started out by not believing it.

Wilmot: So you were saying you'll only be a finite amount behind?

Blackwell: That's right, the expected amount — [pause] Yes, I first thought there could be winning systems where you'd be only a finite expected amount behind, yes.

Wilmot: So, in 1946, when you wrote this paper, "On an Equation of Wald," was that when you expanded Wald's equation?

Blackwell: Yes, that's right. And, as I say, my interest — . I first heard about Wald's equation in this lecture by Abe Girshick.

Wilmot: I have a question about "On an Equation of Wald," but I want to get back to — I'm torn, because I want to get back to Abe Girshick but I also want to ask you about this equation of Wald, can I ask you about the equation of Wald?

Blackwell: Yes.

Wilmot: So, how did you expand his theorem, his equation? What did it — how did you expand it? In what ways? What did your paper say about Wald's equation?

Blackwell: Wald had talked mainly, maybe only, about systems where you make the same bet over and over. I extended it to the case where you could make many different bets, just as long as they were all fair. You didn't have to stick to the same game. In the case Wald considered, I found a new proof of his theorem. So I added something, all right, to it.

Wilmot: Yes. So let me just try and ask one question, which is about — . hi extending it to other games, were all those games still heads and tails games?

Blackwell: No, it could be any game as long as it's fair.

Wilmot: And when you say "fair," that word is confusing to me.

64

Blackwell: That's a technical concept, but it means that if you play it many times, your average gain will be near zero.

Wilmot: The average score of the game or the average game? Blackwell: Your average winnings will be near zero.

Wilmot: You know, it's hard because I think when you translate the mathematical terms into that gambling scenario, which is what happens often — . But I think that it's hard, even though in some ways it's often the most suitable kind of metaphor for talking about this, it still doesn't quite match up to what goes on for me.

Blackwell: Nadine, just think about Reno or Las Vegas and a casino, and you'll be thinking in the right way. You go in, there are all these games. You can bet on this one. You can bet on that one, and so on. Now it happens that most of them are slightly unfair, they have a slight bias in favor of the house. But now let's imagine that instead, they were all exactly fair. Now you could go in and you could still try out various systems. And the point is, if you have a system where the expected amount that you're behind is finite, then it must be fair. If each game is fair, either the system is fair or you can expect to be an infinite amount behind before you win.

As I say, the only technically difficult thing is the idea of "expected income," and that's — . Well, if you study elementary probability, you learn about expected value.

Wilmot: Did you ever take these systems to an actual casino? Blackwell: No. I hate to lose money, [chuckles]

Wilmot: What was the response in the mathematical world to this paper, "On an Equation of Wald?"

Blackwell: You never know that. I think I've done only a couple of things that got any kind of noticeable response, and that was not one of them.

Wilmot: No letters from fans of Wald who said, "That's wrong," or "That's right"? Blackwell: No, no. [laughs]

Wilmot: Okay, so you met Mr. Girshick and you wrote him. He invited you to have lunch after reviewing, after seeing your contest to Wald's equation?

Blackwell: Yes.

Wilmot: And what was that like?

Blackwell: I don't remember in detail. We had a conversation, he mentioned some problems; I went away and thought about them and we had another conversation. I think he was going to go to Rand. Let's see, what was it? I'm sure it was through Abe Girshick that I got invited to go to Rand as a consultant. Maybe it was at the time he left the U.S.

65

Department of Agriculture to go to Rand. My memory is a little vague, but it was right around that time.

Wilmot: I have some questions on Rand I wanted to ask you quickly. Abe Girshick, I know he worked at the Department of Agriculture. What was his professional work then, his work at the Department of Agriculture? What was he using statistics to do there?

Blackwell: I don't remember.

Wilmot: Did you get the sense that he was seeking out mathematical activity outside of his work place?

Blackwell: Oh, he was certainly doing mathematical statistics and welcomed the chance at talking with someone, sure. Just as I did.

Wilmot: Right, can you tell me his training? What was his training?

Blackwell: Yes, he got his master's degree in statistics at Columbia University, and later his PhD., also from Columbia, working under Abraham Wald, by the way.

Wilmot: Oh, very interesting! Blackwell: Well, mathematical statistics was a fairly small group at that time.

Wilmot: I guess, for me, it just shows me his investment in trying to work with Wald's equation and engaging you around it. So, it just shows me something.

Blackwell: Investment?

Wilmot: Yes, when I say that I mean his interest, his avid interest, because he was a student of Wald's.

Blackwell: During the war, during World War II, there was a group at Columbia University called the Statistical Research Group. I was not a part of the group. In fact, I didn't know anything about it. But Abe Girshick was, and Wald was, and Jimmie Savage was, and several other people were. So that's how Abe got started, I guess, doing statistical research. And while he was at the Department of Agriculture, he simply continued that.

Wilmot: One thing that Al Bowker has said that sometimes, while you weren't part of that group, sometimes members of the group would talk with you about problems they were working on. Do you remember this?

Blackwell: Vaguely, yes. I certainly talked not only to Abe Girshick but to Jimmie Savage.

Wilmot: I have in that group, a Paulson?

Blackwell: Ed Paulson.

Wilmot: And Solomon.

66

Blackwell: And Herb Solomon, yes.

Wilmot: And Bowker.

Blackwell: Yes.

Wilmot: Do you remember what kinds of problems they were working on?

Blackwell: No, I don't

Wilmot: So what did that mean for you to leave what you had known around mathematics and move into statistics? In terms of your work at Howard, did that change anything at Howard?

Blackwell: No, it — . My boss at Howard, Woodard, certainly encouraged me in this. He was just glad to see me active, doing things.

Wilmot: While you were at Howard, were there graduate students who were also very interested in mathematical statistics?

Blackwell: No. They were mostly interested in algebra or topology. I think I taught statistics,

mathematical statistics, just once while I was at Howard. I was still learning while 1 was at Howard. I wouldn't have known enough about statistics to teach several courses in it.

Wilmot: Can you tell me if this is true? My understanding of statistics is that it's a field that grew out of actuarial and agricultural applications. And, so — .

Blackwell: Well, those are certainly two of the sources. It also grew out of biological and medical problems. Statistical problems arise in many different fields, in engineering—

Wilmot: I understand that it came of age — I mean, when you became interested in statistics, it was in some ways still a very young discipline. Would you say that's true?

Blackwell: No, mathematically, it was a rapidly developing discipline. People were looking for the proper mathematical foundations for statistics. Neyman, and R.A. Fisher, and Wald all made really essential contributions to understanding what the proper mathematical framework for statistics is. So, I was lucky in that I happened to get into statistics at a time when it was very rapidly advancing.

Wilmot: I've been learning about systems theory. And a group of people who really focused— they focused all of many different disciplines on trying to posit different systems or systems theory for different either naturally occurring phenomenon or social behavior, so many different things. And as I understand, Von Neumann was very active in that group or was one of the key members in that group, in terms of people who were thinking about that. And it was almost like this commitment to truth, and making sense of what seemed very, kind of, illogical and bringing it to a systems theory place and making it truthful and real, using, and sometimes — often, mathematical statistical tools.

Blackwell: I'm not sure just what you're talking about. Von Neumann had very wide interests. I, myself, have never gone very far outside probability, game theory, and statistics.

67

Wilmot: Were you aware of other people in your circle who were pushing statistics in other directions, that had more to do with — ?

Blackwell: My own interests have always been to look at a problem, and try to find some

interesting mathematics in it. Never mind if it helps in solving the actual problem, but just is there any interesting mathematics there?

Wilmot: I understand. So when you— in that case, when you thought about applications I understand that you were more focused on the theory. Is that true?

Blackwell: Yes.

Wilmot: I should ask, did you think about applications of your work?

Blackwell: No. I doubt if the things that I've done have any applications. As I say, my own

tendency is to look at a real problem and see if it suggests any interesting mathematical problems. Now, usually, to get an interesting mathematical problem you have to change and simplify and distort the real situation a lot. So that the mathematics that you do isn't helpful in the real situation. It's helpful in the very simplified situation that you created.

Wilmot: I understand that, but I also see that in that time during the growth of statistics as a

discipline, or during that time of a lot of activity, there were a lot of people who actually saw real world applications and funded people to work with the idea that there were real world applications. And I learned about that mostly in your Neyman book. Primarily it was agricultural, but I understand that that's different than what you were doing, but I'm just thinking about — .

Blackwell: No, it's there, and it's big, but I'm not the person to talk about it.

Wilmot: So, for you, when you thought about game theory in particular, it was not a vehicle, in and of itself to take you to an application?

Blackwell: That's right. [Wilmot: When did you develop the Rao-Blackwell theorem?

Blackwell: Well, let me begin by giving you a little history. Up until about 1940, statisticians had studied how to deal with samples. But the samples were always of a fixed size. You'd go out and ask two hundred people what they think about this or you look at fifty specimens of something and take measurements, and then you'd try to decide what to do. Wald, about 1940, started studying a situation where you keep sampling until you decide that you've seen enough and then you stop. Well, it turned out that it wasn't clear how to adapt certain concepts that had been developed for a fixed sample to this kind of sampling, which was called sequential sampling. For instance, if you want to estimate the proportion of people with a certain characteristic in a population, you take a sample and you look at the proportion in your sample with that characteristic, and that estimates the proportion in the population. And it has a property of being unbiased. That means that if you do this many times, the average value of your estimate is near the true value. But it turns out that if you don't decide in advance how big a sample you are going to take, that using the sample average as an estimate of the population average is not

68

unbiased always; it can give you biased results. So, statisticians started looking at how can you get unbiased results with sequential sampling where you don't decide in advance how big a sample you are going to take. Now, the simplest case is where you are trying to estimate a proportion — is it 40 percent or is it 90 percent?

Three statisticians — Abe Girshick, Fred Mosteller, and Jimmie Savage — found a way to get an unbiased estimate from a sequential sample. But their formula was rather mysterious; people didn't really understand it. They could prove that it worked; but it wasn't clear what was going on. So I was one of those who was trying to understand their estimate, trying to understand why it worked. And I was walking along one day and all at once, the idea popped into my mind that, It's a conditional expectation. That's what their estimate is.

Wilmot: You were literally just walking along?

Blackwell: Yeah, in a casual way. And all at once that idea popped into my mind. Well, that's really all there is to this — the theorem says that if given a sample, you calculate the conditional expectation of any unbiased estimate, you get another unbiased estimate which is better than the original one you had. And of course, that told people how to do unbiased estimation in sequential sampling, so people paid attention to it. But that's really all there was to it. I was just trying to understand their estimate and I was able to explain it.

Wilmot: Around what year was this?

Blackwell: I probably did it in 1946; it was published in 1947. Now, two years before that, though, in 1945, [C.R.] Rao published his thesis. And the same result that I had was one of many results in his thesis.

Wilmot: Where was he?

Blackwell: He was a student of R.A. Fisher's in Cambridge, England. He, himself, was originally from India. So, because in his thesis, the result was buried among several other results, people hadn't paid much attention to it, and I didn't know anything about it. But, when I rediscovered it, and used the rediscovery to explain the Girshick-Mosteller-Savage estimate, people paid attention to it. [pause] But once I discovered it, anybody could prove it. Any competent statistician could prove that yes, it's correct, if you take the conditional expectation, you do get an another unbiased estimate, and it's better than the one you started with. So, it's not a deep thing mathematically; it was just a new idea.

Wilmot: Were you in communication with Rao at all?

Blackwell: Not until three or four years after that.

Wilmot: After 1947?

Blackwell: After 1947, yes.

Wilmot: What was that like? What kind of communication did you have with each other?

69

Blackwell: I don't remember what our first meeting was like. I do know, though, that he's not

especially happy that my name is attached to the theorem. And he shouldn't be, because he has the priority by two years. It's just that somehow when I did it, it got publicity.

Wilmot: So did you get a lot of responses to the paper about this theorem?

Blackwell: A fair amount. As I say, there were maybe twenty citations in the first couple of years. And for me that's big! [laughter] Most of the things that I do get zero or one citation.

Wilmot: As far as the work that you pursued after that, did it lead into more discoveries, as the result of this? Or was it the building block in other work that you did later?

Blackwell: No. No, I don't think I have — . I may never have cited that paper after that.

Wilmot: It's interesting to me too, when you describe Rao, because Rao was a student of Fisher and you worked closely with Jerzy Neyman, who —

Blackwell: — yeah, Neyman and Fisher were antagonists in a way.

When I first started working in statistics, the mathematical community was pretty small. And we all knew each other or knew of each other. I was lucky to come along and get interested in statistics at the time when a lot of new things were being discovered and evaluated and explored. So twenty years after that, all the things that were easy to discover and understand and explain had been discovered and understood and explained. It's harder to find new, interesting things.]

Wilmot: So, you said that you also spent some time at Rand, and that Jimmie Savage had, kind of, probably worked on bringing you there? Or no, Abe Girshick had also, had gone to Rand?

Blackwell: Yes, Abe Girshick and I worked together quite a bit at Rand, on various problems. And Jimmie Savage was there for at least one summer that I remember.

I remember one big influence that Jimmie Savage had on me. He converted me to Bayesianism, so to speak.

Wilmot: You became a Bayesian. Blackwell: Yes, shall I tell you about that? Wilmot: I would love to hear about that.

Blackwell: Well, a Rand economist came in one day to talk to me while I was visiting there. And he said, "I need a number. I need to know the probability of a major war within the next five years." And he explained to me why he needed to know that number and it made a lot of sense. But, I turned him off. I said, "The concept of probability makes sense only in a long sequence of events under identical conditions." And the occurrence of a war in the next five years is a unique phenomenon and the probability is either zero or one and we won't know for five years. And he looked at me, and he said, "Thank you." He said

70

Wilmot:

Blackwell:

Wilmot:

Blackwell:

Wilmot: Blackwell:

that he had spoken with several other statisticians and they'd all told him the same thing, and he left.

That conversation bothered me. The man had asked me a serious, reasonable question and I had given him a kind of flip answer, off-hand, "Ohh." And a couple weeks later, Jimmie Savage showed up at Rand. I went in to welcome him to Rand, and I told him about this conversation I'd had with this economist. Then Jimmie started explaining to me about the subjective theory of probability. And how, of course, the economist's question made perfectly good sense and I'd been wrong to turn him off in that frivolous way.

What Jimmie said made a lot of sense to me because I had been thinking about the probability of single events all my life. And to me, that was the natural way to think about probability. I just regretted that it wasn't the correct way. And now Jimmie was explaining to me that it was the correct way. And, for me, that was a very important intellectual shift. The idea that probability does not apply just to events that occur under identical conditions but that the concept of probability applies to single events, unique events. By the way, the statistician who made this approach popular among statisticians was Abraham Wald.

But Wald knew that applying probabilities to single events would not be respectable among statisticians. So he didn't call them probabilities, he called them weight functions. However, in giving a general name to his approach, he did call them Bayes solutions. He called his solutions obtained by putting probabilities on single events, he called them Bayes solutions. And the Bayes approach has become fairly popular among statisticians, but it has gone far beyond statistics. Engineers, economists, and business people all use the Bayes approach very much. And it all grew out of Wald 's work in sequential analysis.

Why do you think that approach is more popular with businessmen and economists? What about those disciplines make it more interested in subjective, looking at subjective probability? If that makes sense.

Well, I think that it's just that they are applied, and real people face single events, that you can't embed in a context of large sequence of events under identical conditions. Every event is unique.

So when you're tackling a problem using a Bayesian approach, how do you factor in those unique circumstances that affect the probability of something happening?

That's a big question, and a lot of people have studied that. It's no different from how you do it in everyday life. You make judgments all the time. Is it likely to rain? Should I buy this car or that one? You know, all kinds of uncertainties. And you're just asking, how do you decide on which is more likely?

So it kind of makes it a more textured and nuanced inquiry? With more dimensions? We're just talking about uncertainty. You understand uncertainty as well as I do.

71

Wilmot: But in terms of that original problem, what's the likelihood of having another world war in the next five years, approaching it from a "frequentist" approach would mean that you would say the chances are one or zero. Is that correct?

Blackwell: The frequentist would say that that concept doesn't really make sense. You'd have to put it in the context of, "If our sequence of identical worlds under the same conditions, and in some of them war would break out and in some of them it wouldn't," and you'd have to count the frequency. Nobody wants to do that. So that's why frequentists would simply say that that question didn't make sense.

Wilmot: Whereas Bayesians would say, "Not only did that question make sense — ".

Blackwell: But it's an important question and you have to take account of everything you know, and how you take account of everything is not clear.

Wilmot: That's the part where I'm lost, is like how you take account of everything. That's the part where I'm trying to understand the approach.

Blackwell: I don't know that.

Wilmot: Okay, [laughter] Can you tell me a little bit about Jimmie Savage, who brought the Bayesian approach. Did he also study with Wald? Was he also a student of Wald's?

Blackwell: No. He got interested in statistics seriously, I guess, when he went with the Statistical Research Group. But I first met Jimmie at the Institute for Advanced Study in 1941. He had just finished his PhD, I guess, at Michigan. I think his work was in geometry. But again, he was vaguely interested in statistics, and he and I both sat in on Sam Wilks' lectures that year. Then I ran into him a couple of summers later at Brown University. The government was trying to get all mathematicians, and other people, too, interested in war work. And there was a big session at Brown in applied mathematics. I went there for that session, and so did Jimmie Savage. We had a good many contacts during that summer. I remember something that Jimmie said that I thought was funny. He said, "Blackwell's theorem is: every suitcase can be closed." [laughs] You see, I had said that once. We were in a room and somebody was trying to pack a suitcase and put a lot of stuff in it. And I simply said, "Every suitcase can be closed." And Jimmie remembered that and quoted it back to me a month later as a theorem.

Wilmot: That's funny, so there wasn't a whole deeper statistical meaning in it. Blackwell: That's right, [laughter]

Wilmot: Okay, what does that mean? I think sometimes that's true, actually. Well, in any event — .

Blackwell: Have you ever had a suitcase that you couldn't close?

Wilmot: No. I work on it. I also travel very light, I'm one of the lightest travelers ever. Blackwell: There you go.

72

Wilmot: Do you remember there being, when you attended this conference at Brown University where you saw Jimmie Savage, was there an excitement about the work that the military was proposing? Wartime work for mathematicians and statisticians?

Blackwell: I didn't feel that. Again, I was just there to learn mathematics. And one of the people that I had a fair amount of contact with that summer was Will Feller. He was a very good friend of my thesis advisor, Joe Doob. He was at Brown at that time. My memories are pretty vague, but — .

Wilmot: That's okay. So what kinds of things were you working on while you were at Rand?

Blackwell: The most significant thing that I did at Rand — well, there were two. One was duels, [laughs] Do you believe that?

Wilmot: I do, I've read about your duel theory.

Blackwell: Yeah, and that lead to an interesting class of games. You see, I thought of a duel as a game. There are two people approaching each other, and each one has a gun with one bullet in it. And you can fire anytime you please, but to make it interesting, after you fire you can't run away. You have to keep approaching. Let's say you're on a treadmill, you're fixed. The roads are moving this way and it's going to keep moving. And you can decide when to fire. Now, you have to think about two things, you see. The closer — the longer you wait, the better chance you will have of hitting him. On the other hand, he might fire first. The question is, when should you fire? Well, that was the first problem that I posed and that has a fairly easy solution. And the other was something called "A Comparison of Experiments." Here I got the idea from some other people at Rand.

And again, it had a kind of a military origin. They called it "Comparison of Reconnaissances." Yes, you're going to play a game, but before you play the game you have a chance to get some information about what the enemy is doing. So you have, maybe, two spying systems. And the question is, which one to use? But then, I made it more interesting. The guns are silent. So there you are, if you know that he had already fired and missed, you would wait until you got right upon him. On the other hand, once you get pretty close, if he hasn't fired you'd better fire. So now the question is, when should you fire? And I solved that problem. And that got expanded into a general theory that other people did called "games of timing." And those games of timing all grew out of this simple duel. So, that was one of the things I worked on at Rand.

Now, it could happen that one system will give you more information than the other under all circumstances, in which case it's clearly a better system. And these fellows, [Henri F.] Bohnenblust, [Seymour] Sherman, and [Lloyd S.] Shapley, developed a great theory for this in the case where there were only two hypotheses. And then I looked at their work, and thought of another, even stronger way to compare two systems. And I asked the question whether these two methods were equivalent.

In one direction it was clear. It was clear that my method was stronger than theirs. But the question was, was it really stronger? And I didn't think it was really stronger, and I was able to prove that no, that in a case of just two alternatives the two systems — the two comparisons are the same. And then other people proved that, in general — the two

73

systems, the two methods of comparison are the same. The main interest is not in military things at all, but the problem grew out of thinking about military problems.

Wilmot: Did you have a sense, once you solved these problems, that there were people who understood your work and could, maybe, apply it?

Blackwell: Oh, there were certainly people who understood it and were interested in it. I don't think there have been — I'm not aware of any applications.

Wilmot: For the solution to the dueling problem, was that actually a number? A time when should fire? Or was it a calculation?

Blackwell: Ah. In the case where the guns are noisy, you calculate an actual time when you should fire, yes. But in the case where the guns are silent, there is no one time you should fire, there are probability distributions over firing times. And so the solution is to find the probability distributions.

Wilmot: And for the problem on reconnaissance?

Blackwell: Here, it was just a yes/no question. Are these two methods of comparison the same? Are they equivalent? And the answer turned out to be yes, they are.

Wilmot: So while you are at Rand, was it a place where you came with your own problems and developed problems? Or was it a place where people say, "Here are some problems, can you work on them?"

Blackwell: No, it was closer to "Here are some problems, let's work on those." In a general context of anything that has to do with warfare is interesting. So, you start thinking about conflicts in general. As I say, a duel is clearly a very special kind of fighting, and this comparison of experiments was put in terms of comparisons of reconnaissances in a game theory context. And, of course, games are clearly conflict situations. So, that meant that Rand would be interested in them.

Wilmot: I'm trying to think of other kinds of things like warfare or games where you can

basically have determined, specific variables and winners and losers that could make it a good arena for this kind of analysis.

Blackwell: Well, economists have all kinds of problems like that. There are whole journals devoted to that. This one called Games and Economic Behavior.

Wilmot: Yes. I have a copy of this one that's dedicated to you, actually, I think. Blackwell: Oh, that's right, there was one.

Wilrnot: So who was deciding what kinds of problems that the group of people at Rand would work on? I understand that it was everything that was warfare, but maybe there were more specific problems that people brought in and said, "Work on this."

Blackwell: Well, Rand had a regular staff of permanent people who had problems they were

working on. And you were welcome to help them work on their problems or find your

74

own problems. The atmosphere at Rand was extremely informal. I was working with a man named Olaf Helmer. I worked quite a bit with him, and one day, he told me that we couldn't meet that morning because he had to go to a meeting of group leaders, or something like that it was called.

And I said, "Olaf, I didn't know you were a group leader. Who is in your group?" And he said, "You are, for one!" [laughs] So that's how informally it was organized, see. I had no idea who my boss was, there.

Wilmot: Wow, that sounds like a very interesting atmosphere. Blackwell: Oh, it was a great atmosphere.

Wilmot: So, it was a place where you were more focused on warfare and not so much thinking about economic industry, engineering — . It was mostly just warfare?

Blackwell: Yes. But, of course, the military is interested in all kinds of things. I mean, name something that doesn't have an application to warfare.

Wilmot: My first thought was agriculture, then I had to withdraw it. Blackwell: Yes.

Wilmot: Do you remember who — sorry?

Blackwell: I was just thinking, one of the things that Olaf Helmer, my boss, was interested in was what he called Delphi. The idea was, how can you take a group of experts, and ask each one what he thinks about something, and then combine all the opinions and come out with an opinion that is more reliable than that of any individual expert. For instance, if you wanted to know how likely a war is in the next five years, get five experts to talk about it and tell you what they think and then try to combine their individual opinions into one opinion. Olaf was interested in finding methods of combining expert opinions.

One example he was interested in looking at was horse races. Every day, in the newspapers, several horse racing experts would write their opinions. You know, who's a good idea for this race, and who's a good longshot for this race, and so on. So, Olaf tried to see if he could find a consensus among the experts that would be more reliable than any one expert opinion. He wasn't able to do very well on that, but I remember that someone wondered why he was subscribing to all these daily racing forms! [laughs]

By the way, one thing that we did on that that I remember was this. We tried to see if we could combine expert mathematical opinions into an overall mathematical opinion that was more reliable than any one expert. So, some of us got together and devised a long mathematical questionnaire. Now, these were questions that could be answered if you had enough time. But, we gave the experts only a very short amount of time on each question. "Ms. Wilmot, just off-hand, what do you think about this, and this, this, and this." And, of course, we, having lots of time, knew what the answers were. So we tried to combine all the individual opinions into a group opinion that was more reliable than any expert. We had various ways of doing it. And the most reliable consensus that we had was able to get the right answer 8 1 percent of the time! However, it turned out that

75

there was one individual who was right 80 percent of the time. So — alas, you didn't really need a consensus, all you needed was to ask Lloyd Shapley because he was almost as reliable as the best consensus we could find, [laughter]

Wilmot: When you say "we," who is "we?"

Blackwell: Olaf Helmer, and I, and Abe Girshick, and a fellow named Alex Mood, and maybe there were two or three others.

Wilmot: So, instead you just found the one expert.

Blackwell: Found the one expert, yes.

Wilmot: And rendered all the rest of you, kind of — .

Blackwell: Superfluous!

Wilmot: Was that your first time in California?

Blackwell: Yes.

Wilmot: Did you stay pretty close to Rand, or did you kind of get out a little?

Blackwell: I stayed pretty close. I think we went down to Tijuana a few times, but — . Sometimes, I came to Rand just for the summer and my family stayed in Washington. But once I was at Rand for six months and my family moved out.

Wilmot: Do you remember them liking California?

Blackwell: Yes, my children were pretty small, but — .

Wilmot: I have a couple of more questions on the Bayesian approach.

Blackwell: Yes.

Wilmot: I read an interview with you where it said that you were surprised that that school, that approach was still flourishing. Because, initially, when you, kind of, first embraced it, it seemed to you that it was dwindling and there weren't very many people who were adherents to that approach.

Blackwell: Yes. Now, statisticians seemed to have advanced beyond both the Bayesian and the frequentist approach. Such questions are not much discussed among statisticians anymore. And as I say, engineers, and economists, and biologists have pretty much gone to the Bayes approach. Especially engineers and economists.

Wilmot: My understanding, though, is that Berkeley has always been the stronghold for a frequentist approach.

Blackwell: That's right, and still is.

76

Wilmot: In fact, when I was trying to research colleagues who I could speak to about you to learn about the Bayesian approach, there were very few people who I could find who were also Bayesian in their approach.

Blackwell: I think there's just one. A fellow named Kjell Doksum is a Bayesian, but he and I haven't worked together at all. But, I think he's the only one.

Wilmot: And I'm kind of just going to jump ahead a little bit, but I wonder what that was like for you when you came to Berkeley, and everyone was frequentist and you had this Bayesian approach?

Blackwell: No problem whatever! [chuckles] When I taught statistics, I taught it in a Bayesian way, and other people taught it in a frequentist way.

Wilmot: In particular, I was wondering about when you interacted with Jerzy Neyman. And he was not Bayesian at all. Did that ever come up between you two, did you ever talk about it?

Blackwell: No. And he was not completely anti-Bayesian. But, no, he and I never discussed it.

Wilmot: I'm going to ask a lot more questions about Neyman later on, but not now. For me,

coming from a place where both of these things are fairly new ideas for me, I think the thing I'm trying to understand is when you take a frequentist approach or a Bayesian approach, do you end up at different places as a result of those approaches? Or do you travel different roads and arrive at the same place?

Blackwell: Usually, you arrive at the same place. If there's a lot of data, a lot of information, then you will almost certainly arrive at about the same place, yes. But when there's very little information, then — . Well, when there's very little information, a Bayesian would say that the frequentists don't arrive anywhere. And the Bayesians have a lot of uncertainty, but — .

Wilmot: Okay. I wanted to just look at some of these problems. How are we doing for time? Are you okay for time?

Blackwell: Yes, I can stay a little longer.

Wilmot: Okay, good. Blackwell: Yes.

Wilmot: I guess I should start off by asking if there's any on this list here that you want to talk about? Because for me, a lot of these are the things I don't know about, I tend to pick on the basis of language and, like, what has the most beautiful language associated with it.

Blackwell: Yes, all right. Well, you have marked them in yellow. That's with Richard Bellman, and "Some Two Person Games Involving Bluffing." Yes, Dick Bellman and I noticed that in a lot of games, card games, for instance, in our theoretical, continuous version, no bluffing was needed. Whereas in actual card games, people do bluff all the time. So, we asked, why, theoretically, did it come out that often bluffing is not needed? And, it

77

turned out that if you have a large number of cards, that you can replace bluffing by something that — that you don't need to bluff.

And let me give you an example. Suppose that when you get — . Suppose that no matter what card you get, you should take one act with probability one-half and the other act with probability one-half, bluffing, you see. And what you can do is this, you can just say — . You can divide the cards mentally into two groups and say, "If I get a card in this group, I'll bluff, if I get a card in the other group, I won't bluff." Now, since you don't know which card I got, to you that looks just like bluffing. That's all. And we showed, mathematically, that you can replace bluffing by this splitting the cards into two groups and deciding to take one act if you see a card in this group, another act if you see a card in the other group.

In other words, you let the cards you're dealt do the bluffing for you. So, formally, it doesn't look like bluffing because bluffing involves you doing one act of probability with a certain probability and another act with a certain probability in the same circumstance. Well, you're not doing that, but it has the same effect.

Wilmot: And that was with Richard Bellman?

Blackwell: That was with Richard Bellman.

Wilmot: Where was Bellman?

Blackwell: Bellman was also at Rand.

Wilmot: So, when two people are working on something together, do you just kind of trade back and forth the ideas? What is that process of working together on something like?

Blackwell: Somebody asks the other person a question, and then you both start thinking about it. And the next day, maybe I talk to you or you talk to me. Whoever has an idea talks about it, and it goes back and forth like that, and sometimes you're in the same room together, talking about it together. But, as often, you're thinking about it separately and then coming in to say, to tell the other person what you have discovered. And finally, you either give it up as hopeless, or you decide, well, "We found something