The Nation’s Largest African American Video Oral History Collection Mobile search icon Mobile close search icon

Search Results

Advanced Biography Search
Mobile navigation icon Close mobile navigation icon

William Massey

Mathematician William A. Massey was born in 1956 in Jefferson City, Missouri; the younger of two sons of Richard A. Massey, Sr. and Juliette Massey. Massey attended the public schools of St. Louis, Missouri and high school in University City, a suburb of St. Louis. Upon graduating from University City High School, Massey received a Harvard Book Award and a National Achievement Scholarship. He enrolled at Princeton University in 1973 and encountered his first real introduction to research mathematics in an honor calculus course taught by the late Ralph Fox. Massey wrote his undergraduate senior thesis, “Galois Connections on Local Fields,” under the direction of Bernard Dwork, and graduated from Princeton in 1977 with his A.B. degree in mathematics with honors – magna cum laude, Phi Beta Kappa, and Sigma Xi. Massey was then awarded a Bell Labs Cooperative Research Fellowship for minorities to attend graduate school in the department of mathematics at Stanford University. Massey wrote his doctoral theses, “Non-Stationary Ques,” under the supervision of Joseph Keller, and graduated from Stanford University in 1981 with his Ph.D. degree in mathematics.

In 1981, Massey became a member of the technical staff in the Mathematical Sciences Research Center at Bell Laboratories, a division of Lucent Technologies. His research there included queuing theory, applied probability, stochastic processes, and the performance modeling of telecommunication systems. Massey published over fifty papers in those areas, one of which credits him as the co-author of a U.S. Patent on server staffing. In the area of mentoring, Massey has organized every annual Conference for African American Researchers in the Mathematical Sciences, which he co-founded in 1995. He founded the Council for African American Researchers in the Mathematical Sciences (1996) and is a lifetime member of the National Association of Mathematicians (NAM). In 2001, Massey was named the Edwin S. Wilsey Professor of Operations, Research, and Financial Engineering at Princeton University, making him the first tenured African American mathematician at an Ivy League University.

Massey received the Distinguished Service Award from NAM in 1996 and was invited to give its William W. S. Clayton Lecture. He has given invited lectures at the American Mathematical Society national conference, the Congreso Nacional de la Sociedad Matematica Mexicana, and the Edward Bouchet Conference for African and African American Physicists and Mathematicians that were held in Ghana, Canada, and Germany. The Blackwell-Tapia Prize Committee awarded Massey its 2006 prize and U.S. Black Engineer and Technology magazine honored Massey as the Black Engineer of the Year in 2008.

William A. Massey was interviewed by The HistoryMakers on March 8, 2013.

Accession Number




Interview Date


Last Name


Maker Category

University City High School

Princeton University

Stanford University

Search Occupation Category
First Name


Birth City, State, Country

Jefferson City



Favorite Season




Favorite Vacation Destination

Paris, France

Favorite Quote

Think outside the hypercube.

Bio Photo
Speakers Bureau Region State

New Jersey

Interview Description
Birth Date


Birth Place Term
Speakers Bureau Region City




Favorite Food

Steamed Crab Legs

Short Description

Mathematician William Massey (1956 - ) , co-founder of the Conference for African American Researchers in the Mathematical Sciences, became the first tenured African American mathematician at an Ivy League University when he was named Edwin S. Wilsey Professor of Operations, Research, and Financial Engineering at Princeton University.


Bell Laboratories

Princeton University

Favorite Color


Timing Pairs

Tape: 1 Story: 1 - Slating of William Massey's interview

Tape: 1 Story: 2 - William Massey lists his favorites

Tape: 1 Story: 3 - William Massey describes his mother's family background

Tape: 1 Story: 4 - William Massey describes his mother's education and her career as a teacher

Tape: 1 Story: 5 - William Massey describes his father's family background

Tape: 1 Story: 6 - William Massey talks about his parents' employment in Missouri

Tape: 1 Story: 7 - William Massey talks about his parents' personalities and who he takes after

Tape: 1 Story: 8 - William Massey describes his childhood's neighborhood in St. Louis, Missouri

Tape: 1 Story: 9 - William Massey describes his earliest childhood memory

Tape: 1 Story: 10 - William Massey describes the sights, smells and sounds of growing up

Tape: 1 Story: 11 - William Massey describes his childhood interests

Tape: 2 Story: 1 - William Massey discusses the portrayal of black scientists on television

Tape: 2 Story: 2 - William Massey describes his childhood toys

Tape: 2 Story: 3 - William Massey describes his experience in grade school and his early interest in mathematics and drawing

Tape: 2 Story: 4 - William Massey talks about his childhood friends

Tape: 2 Story: 5 - William Massey talks about the political atmosphere in the 1960s

Tape: 2 Story: 6 - William Massey talks about his experience in a mixed-race schooling system

Tape: 2 Story: 7 - William Massey talks about attending church as a child

Tape: 3 Story: 1 - William Massey describes his involvement with church and in sports while growing up

Tape: 3 Story: 2 - William Massey talks about his training in mathematics in school

Tape: 3 Story: 3 - William Massey discusses his summer jobs, and his high school activities and achievements

Tape: 3 Story: 4 - William Massey describes his decision to attend Princeton University

Tape: 3 Story: 5 - William Massey describes his experience at Princeton University

Tape: 4 Story: 1 - William Massey discusses his concerns about education and violence in the African American community

Tape: 4 Story: 2 - William Massey talks about the misrepresentation of statistics in the media

Tape: 4 Story: 3 - William Massey describes his mathematics coursework at Princeton University

Tape: 4 Story: 4 - William Massey describes his senior thesis on Galois connections on local fields

Tape: 4 Story: 5 - William Massey describes his experience at Bell Labs

Tape: 5 Story: 1 - William Massey talks about African American scientists at Bell Labs in the 1960s

Tape: 5 Story: 2 - William Massey describes the queueing theory and his dissertation research on non-stationary queues

Tape: 5 Story: 3 - William Massey talks about his Ph.D. thesis advisor, Joseph Keller

Tape: 5 Story: 4 - William Massey describes his experience as a doctoral student at Stanford University and his summer experience at Bell Labs

Tape: 5 Story: 5 - William Massey talks about his contemporary generation of African American mathematicians

Tape: 5 Story: 6 - William Massey describes his decision to work at Bell Labs and explains the queueing theory

Tape: 5 Story: 7 - William Massey talks about his research at Bell Labs in the 1980s and 1990s

Tape: 6 Story: 1 - William Massey describes the concept of Jackson networks

Tape: 6 Story: 2 - William Massey talks about other African American mathematicians at Bell Labs and in academia

Tape: 6 Story: 3 - William Massey describes his most significant research contributions at Bell Labs

Tape: 6 Story: 4 - William Massey describes how his research has advanced the theory of dynamic rate queues

Tape: 6 Story: 5 - William Massey describes his involvement in establishing the Conference for African American Researchers in the Mathematical Sciences (CAARMS)

Tape: 6 Story: 6 - William Massey describes his involvement in mentoring students

Tape: 7 Story: 1 - William Massey describes his research in congestion pricing, and his transition from Bell Labs to Princeton University

Tape: 7 Story: 2 - William Massey describes his decision to accept a professorship at Princeton University in 2001

Tape: 7 Story: 3 - William Massey describes his current research in decision-making, at Princeton University

Tape: 7 Story: 4 - William Massey talks about his involvement with the National Association of Mathematicians (NAM) and the African American legacy at Bell Labs

Tape: 7 Story: 5 - William Massey talks about his professional awards

Tape: 8 Story: 1 - William Massey talks about his mentorship of African American students

Tape: 8 Story: 2 - William Massey reflects upon his career's legacy

Tape: 8 Story: 3 - William Massey reflects upon his career's choices

Tape: 8 Story: 4 - William Massey describes his hopes and concerns for the African American community

Tape: 8 Story: 5 - William Massey describes the role of African American organizations in discussing social issues

Tape: 8 Story: 6 - William Massey talks about how he would like to be remembered







William Massey describes his current research in decision-making, at Princeton University
William Massey describes his decision to work at Bell Labs and explains the queueing theory
So, now you continued to do research at a higher level here [Princeton University, Princeton, New Jersey] according to all the paper citations I have on you in this album. I'm not sure it's not comprehensive. But can you just kind of summarize what your research has been here?$$Well, here it was getting into--well, moving over from the world of--well, at Bell Labs [New Jersey], they were called queueing theory performance modeling. So, what I call modeling is that, sort of the deliverable for a model is a forecast, you know. So a good model gives you--enables you to predict what goes on with the actual system. So I didn't do a lot of that. As I got to Princeton, I started moving into the area of decision-making. So, now the deliverable is, instead of a forecast, the deliverable is a policy. And so--and then for communications, there seems to be three different natural types of, well, areas we would develop decision-making policies, and just for alliteration sake since, you know, performance begins with the letter P, these three I called--well, first I just thought there were two, you know, provisioning and pricing. And so, now, if you're not obsessed with using the letter P, then provisioning; another way of saying provisioning will be design, you know; having just enough resources to make your customers happy. Pricing is--would be sort of like control, you know; how to--you use prices and mechanism to control the demand for the services, and then, you know, this gives you a way of--well, you have two things where, on one hand you want to maximum your profit, but on the other hand you don't want to violate the constraints of creating bad service. So you want to keep congestion constrained to be no higher than this level. So what's fun about, you know, being in an executive setting and having Ph.D. students, so you're, you know--so this is kind of the setting I gave to Robert [Hampshire; Massey's student who is now on the faculty of the School of Public Policy and Information Sciences at Heinz College, Carnegie Mellon University, Pittsburgh, Pennsylvania] and I kind of had this nice picture, you know, of performance and then--performance for modeling, but then for issues of decision-making, the two Ps, you know, pricing, provisioning. Then a few months into it, he comes back to me and says, "Shouldn't there be a third P here?" "What do you mean?" "Well, I think there should be one on--but--well, later we were going to call it "prioritization." And you only think of that when you have multiple classes of customers. So you don't assume all the class--customers are the same. You know, they have different needs, they have--they can afford different levels of service; and so, how do you allocate these resources. What's the fairest way to allocate these resources among the different classes of customers? You know, so that's another issue, you know, paper we're still--what we developed, we have the paper but that's--one of my outstanding papers we need to finish up and write up and kick out the door. But we got some from out of the thesis, you know; through collaboration, we got some of the papers, you know, from it. And, so now what I'm doing with Jamal, is that we have these--well, it's an essential object that's called the dynamical system, which is the solution of ordinary differential equations. In the twenty-first century, thanks to the computers, these are easing things to solve. So, if you can formulate some more complicated system in terms of dynamical systems, you almost feel you have a closed form solution. And we used--with Robert I used this to approximate average behavior of these random systems. And then we would control that average behavior, so we'll see average profit through the average revenue. And so what's the strategy that optimizes that? But now, when you look at more stochastic systems--well, I have colleagues who were in finance. They worry a lot about decisions under risk. Because things aren't completely deterministic. There's a certain randomness involved. So there's a risk that occurs. But how do you maximize in the face of that type of risk? And so, it turns out you got to understand things like the variance standard deviation, and it turns out the formulas, we have to approximate those; (unclear) approximate those aren't quite as good as the ones that approximate the mean. So with Jamal, his thesis is developing new techniques. I guess he would say it's involving stuff like skewness approximation, cumulate moments; and give better estimates of the variants. So we could extend this sort of decision-making to, you know, deal with more uncertainty. You know, like, you want to maximize your profit, but you only want to take this level of risk. You know, how do you, you know, how do you do that? And then, now I have a most recent student, Jerome--major move I'm making now is that, up to now, everything has been related to communication, communication services; but I found writing up this, you know, in the act of writing up these papers I've done with Robert or, you know, having do up his thesis, I just realized that, when you look at communication services--okay, so--and, of course, when you're no longer working for a phone company, you know, you feel free to thinking about things outside of telephony. But what's communication services from a business perspective? It's sort of the leasing of shared resources. You know, with your cell phone. You don't buy a radio channel. You know, in effect, you're paying for the leasing of it through the rate of your conversation. And so, that's what we're studying in general, so like Robert is in a department of--I guess it's the School of Public Policy and Management [Carnegie Mellon University, Pittsburgh, Pennsylvania]. And, well, once I made this discovery, I was happy, you know, so I was telling Robert, you know, "You'll be happy to know that a set of tools--," since he's an expert in, you know, he trained to be an expert in queueing theory, "--a set of tools that help you study, you know, the leasing of shared resources, you know, may come in, you know, handy when you're looking at issues of public policy." And so, he's looked at them and applied to areas of transportation. Now, recently, what I've been doing is, a new area a lot of people in operations and research are getting excited by is health care, you know, because, like, in health care, you don't--you go to the hospital, you don't buy a hospital bed; you lease it for the duration of your stay. And then you have a lot more issues of, you know, coordination of different types of resources. So you have this whole elaborate choreograph of resources that all come to bear on your specific, you know, issue. And so, there's a lot of--there's a lot of room for queueing analysis, you know, there, because the problems are a lot more complex than--a lot of problems that's on the communication systems.$$Okay.$Okay. So 1981 when you finished Stanford [University, Palo Alto, California], did you have any doubt that you were going to be working for Bell Labs [New Jersey]? (laughs)$$Oh, no doubt.$$All right.$$Because it just seemed like such a, you know, a congenial environment. Also, I just got a chance to work on exciting problems, because I found that by focusing on this applied area looking at specific time varying queue, I as address the issues--the general theory of Markov processes didn't seem to be (touch it?).$$Okay. So you contributed something new to the field of mathematics--$$Yeah.$$--over in this dissertation?$$Oh, yeah. Just studying--there was a classic queueing model with people who are very familiar with constant rates. And so I developed the sort of approximate or asymptotic theory when you had time-varying rates. And then from this new insight I could show--well, basically show that you could, if you analyze things the old way, you might make a mistake a come in, you know, come with a misleading conclusion. Because you think of--well, okay. So you think of--well, we'll make it simple; where the service rate is constant, so that doesn't change; you just have (unclear) rate. And you think of it, it's like water coming to a bucket at a certain rate, and it goes out--let's make it easy; it goes out at a unit rate. It drains in unit rate, water comes in, okay. Now, if the water rate is always less than the unit rate, then the bucket never fills up. And that's kind of like steady-state behavior. But if the water, incoming rate, exceeds the draining rate, then the bucket will fill up; and then over time, it'll just go all the way up to infinity. So that's sort of the static situation. But what happens if the input rate changes in time? Well, what will happen is that, it may start off being less than the unit rate but later it becomes higher; but later it'll drop back down. And so, the actual level goes up, but it's not going to go all the way up to infinity, it's going to come back down. So the big question is: When does it come back down to zero? And people used to think, Well, this is what's going to happen when the input rate--the first time the input rate is less than the draining rate. Of course, if you try that, you'll realize that's not true. You know, it's sort of--kind of like turning on the bathtub and the water is coming in faster than the draining rate. The minute you turn off the faucet, you know, the water level doesn't drop to zero immediately; it's going to take some time. And so what I showed is that, to talk about stay-state behavior, you have to wait until the time it takes for that part to drain out. And then what I was really surprised about, this is just a couple years ago, I didn't realize that this phenomenon really describes what's been going on with our economy. You know, when people say, you know, after this recession that they--we started a "recovery," the recovery that didn't feel like a recovery? What does that mean? Well, what economists call a recovery is like when you suddenly turn off the faucet and the, you know, the rate at which jobs are disappearing becomes smaller than the rate at which jobs are being created (coughs). That's called a recovery. But this backlog of unemployed people, that doesn't suddenly disappear. So economists may say, when the input rate is less than the draining rate, you know, in terms of, you know, lost jobs, that's a recovery. But everyday people are not going to feel like it's a recovery until, you know, that level of water drops back down to zero. And that's kind of where we are right now. You know, we're getting closer, but we're waiting for that to happen.$$Okay. Okay.