Proud but Humble, Departing DMS Director Reflects on Three Productive YearsJuly 3, 2002
"I think of NSF as a provider of venture capital for learning and discovery," says Philippe Tondeur, whose three-year term as director of the Division of Mathematical Sciences at the National Science Foundation ends in July. "It's our responsibility to take intelligent risks." Tondeur spoke to SIAM News in the spring about what he sees as his accomplishments at NSF. He leaves behind a "balanced portfolio"---including Focused Research Grants, new mathematical institutes, a significant expansion in the number of DMS-funded postdocs, and bigger grants for mathematical scientists.
Nearing the end of his three-year term as director of the National Science Foundation's Division of Mathematical Sciences, Philippe Tondeur was the guest of honor at a May 15 reception at the National Academy of Sciences. Among those called on to speak about Tondeur's accomplishments was General William Odom, former director of the National Security Agency and, more recently, chair of the committee that produced the 1998 "Odom report."*
On May 15 friends and colleagues of Philippe Tondeur (center) gathered at the National Academy of Sciences to thank him for his outstanding work as director of the National Science Foundation's Division of Mathematical Sciences. With him here are his wife, Claire (second from left), Margaret Wright, whose phone call three years earlier, according to Tondeur, provided the needed spark to get him thinking about a position in Washington, and reception organizers Gilbert Strang (left) and David Eisenbud (right), representing SIAM and the American Mathematical Society.
"Everyone knows that Washington can be a disorienting place," Odom said. Tondeur, on arriving in Washington almost three years earlier to succeed Don Lewis as DMS director, had called Odom "with some questions," after which, Odom reported, he pretty much "disappeared from my radar for a while."
Odom followed events at NSF during Tondeur's three years and supported him at some key points during that time. At the May reception, he had high praise for his "protégé": "Philippe has done an extraordinary job. . . . I'm happy for the mathematics community, and for the country. What Philippe has done is a giant step forward."
Even as he seemed to disappear, Tondeur was hard at work, from his first day in Washington, figuring out how the system works. And with no clear idea yet about what had to be done, he told SIAM News in a late spring interview, it was in the Odom report that he found his job description. Setting the tone in the report's preface, Odom had warned that the report carried a mixed message: "The U.S. mathematics community holds a dominant position in the world, but several adverse trends are undermining it."
As stated in the report, U.S. leadership in mathematics "depends very substantially on immigrants who had their mathematical training elsewhere. . . . Young Americans do not see careers in the mathematical sciences as attractive. Funding for graduate study is scarce and ungenerous. . . . Students wrongly believe that jobs that call for mathematical training are scarce and poorly paid."
Tondeur, who was born and educated in Switzerland (receiving a PhD from the University of Zurich in 1961) and frequently expresses his gratitude for the opportunities he has had in the U.S., had found his charge in the Odom report and in several sessions with Don Lewis. One of the main, critical needs, he told SIAM News, is "to make the mathematical sciences as a discipline attractive to young people."
How successful did he think he'd been? He was ready with statistics about some of the programs that either were introduced or grew significantly under his leadership: In his first year at NSF, DMS made four CAREER awards, he pointed out; the number had grown to 11 by his second year, and to 20 by his third. At approximately $300,000 per award in FY 2002, the cost to DMS is $6 million. At this point, asking to be reminded of the total DMS budgets for those years, SIAM News was turning the discussion to the unprecedented growth in funding for DMS during Tondeur's term: An increase of $15 million during his first year brought the total to $121 million; with increases of $30 million in each of the subsequent two years, he would be leaving a division funded at $181 million.
"I think my successor could double the budget again, and then we'd be in the right ballpark," he said. With a disarming combination of humbleness and pride, he pointed out that the actual gain is even more than reflected in the budget numbers, as NSF has identified the mathematical sciences as a priority area. Indeed, NSF director Rita Colwell, herself a strong believer in the dependence of all the sciences on mathematics, was among those present at the May 15 reception; like Odom, she pointed to the rich legacy Tondeur leaves for the community, and to her continued commitment to increase support for the mathematical sciences.
The additional funding, Tondeur told SIAM News, "brings with it a huge
responsibility"---the community needs to "use the money to change the world. . . . We have to pick up the challenge."
Recruited by Margaret Wright in 1999 to apply for the DMS directorship, Tondeur had never spent time at a federal agency or even thought about taking such a job. At the time of Wright's call, though, he was already in another position to which he had never aspired---chair of the Department of Mathematics at the University of Illinois, where he had been a member of the faculty since 1968. Describing himself as grateful to Jesse Delia, the dean at Illinois who pulled him out of his life as a "standard math professor," he spoke to SIAM News of "sweeping changes" under way at Illinois at the time of his departure. Along with an influx of postdocs (from almost none when Tondeur took over to 18 today), the department was hiring a large number of new faculty members and---in a project with the sort of symbolic resonance that Tondeur seems to enjoy---ripping out walls.
The walls---sacrificed to create a large seminar room from smaller rooms---were to have parallels in his activities at NSF. Most obviously, the NSF programs of which Tondeur is particularly proud share with seminar rooms the function of bringing people together. VIGRE (Vertical Integration of Research and Education in the Mathematical Sciences), for one, is really "about mentoring," he pointed out to SIAM News, adding that mentoring is of course one of the primary responsibilities of faculty. But in "forcing faculty to think about the basis of their intellectual mission," he believes that the program, which was instituted under the leadership of Don Lewis, has had a significant impact.
At its inception, VIGRE, which accounts for approximately 10% of the DMS budget, was not intended mainly to increase numbers, Tondeur said, although a by-product of the program has been increases in the populations of DMS-funded graduate students (by 18.5% in the first year of VIGRE) and postdocs (with the approximately 150 VIGRE postdocs supported so far, the number of U.S. postdoc opportunities has essentially doubled). The impact at the undergraduate level has been large, Tondeur said. The REU (Research Experiences for Undergraduates) program, from 360 student participants his first year at NSF, now involves 900 students. By comparison, he offered the number of students who receive bachelor's degrees in mathematics each year: about 11,000. In the fall of 2002, he continued, 2800 students will begin graduate study in the mathematical sciences; half of them will be foreign. Of the entire group, the number that will go on to receive PhDs is about 1000---"too small a number" and unsatisfactory also in being "part of a decreasing trend." He anticipates larger numbers of PhDs by the academic year 2004-05---"and we need them."
Perhaps inevitably controversial, given its size and novel approach, VIGRE has not won the approval of everyone in the mathematics community. Criticisms have included the restriction to U.S. citizens for graduate students and postdocs funded under the program, a perception of the program as further separating the "haves" among institutions from the "have-nots," and the question of continuity---how can VIGRE institutions plan for the future under a program with a finite time horizon?
Complementing VIGRE is the Focused Research Group (FRG) program, initiated almost the day of Tondeur's arrival at NSF, with a solicitation drafted at his request by a few of the DMS program directors. The philosophy of the FRG program, he said, is that "together you will do more than you would separately-the whole is greater than the sum of the parts." In 2000, 12 groups were funded (in 21 separate proposals); 10 groups (18 proposals) were funded in 2001, and "there will be more in 2002," Tondeur said of what he views as an "immensely successful program."
"I think of this as venture capital," Tondeur said of the three-year FRG awards, usually about $1 million. The money is all for research, he pointed out. "I think we're really fostering excellence with FRGs; it's significant support."
The National Science Board's philosophy, Tondeur explained, is that "if you're going to fund people, fund them well." Seeing FRGs as a way to carry out the NSB mandate, he pointed out that FRG researchers aren't simply getting good results---"they're also empowering others to achieve greater things." Asked to name a few of the research areas covered by the program, he obliged with a list that would be interesting to a SIAM audience: dynamics of thin viscous films, analytic properties of discrete groups, epitaxial growth, L-functions, optimal transport, computational conformal mapping, and topological methods in data analysis. Many of the FRGs are interdisciplinary, although that's not a requirement, he pointed out.
Another important step in fostering collaboration among mathematical scientists has been the creation of several new institutes that will complement the IMA, MSRI, and their newer counterpart, the Institute for Pure and Applied Mathematics at UCLA. At the time of the SIAM News interview with Tondeur, work was under way on cooperative agreements with three new institutes---the American Institute of Mathematics Research Conference Center, in Palo Alto; the Mathematical Biosciences Institute, at Ohio State University, Columbus; and the Statistical and Applied Mathematical Sciences Institute, at Duke University, North Carolina State University, the University of North Carolina at Chapel Hill, and the National Institute of Statistical Sciences in Research Triangle Park. The new institutes became operational on July 1. Tondeur was also an important player in the startup of another large-scale institute, in which NSF plays a significant but very different role: the joint Canada-U.S. Banff International Research Station.
One final program that can be mentioned here, in biomathematics, is the first DMS collaboration with the National Institutes of Health (specifically the National Institute of General Medical Sciences). DMS program directors Keith Crank and Michael Steuerwalt are now running the program, although Tondeur's intention is to hire a program director in biomathematics (the position had not been filled at press time). One detail mentioned by Tondeur about the NIGMS collaboration: Grants will be bigger than usual for the mathematical sciences.
"I took the job to change things," Tondeur said of his three years at NSF. He could have been referring to his term as chair at Illinois as well, where an iconoclastic side of his nature emerged in some nontraditional fund-raising necessitated by the mathematics department's wall demolition project.
That same aspect of the very dignified and, on first impression, formal Tondeur was well in evidence during his time in Washington. One of the things he especially liked about being at NSF, he told SIAM News, was "getting to think bigger." It was by taking a nontraditional approach, he learned, that he could bring about "a paradigm change in the way math is funded." One activity he mentioned to SIAM News in this connection grew out of his thinking about ways to get the attention of the Administration. Hoping to enlist the commercial and financial sectors in his quest to establish the importance of mathematics, he organized, with the help of Odom and Phillip Griffiths of the Institute for Advanced Study, a dinner at IAS for several industry CEOs and other prominent people. As the keynote speaker, Tondeur delivered, in Odom's words, a "star performance."
People who know Tondeur invariably mention his optimism, a quality that came through clearly in the interview with SIAM News. He believes, for example, that several of the "millennial" problems in the mathematical sciences will be solved in his lifetime (he's 69). At the time of the interview, a proof of the Poincaré conjecture was circulating; experts had not yet delivered a verdict (the rumors turned out to be premature). For his part, Tondeur was closely monitoring developments, ready not just to appreciate news of an impressive breakthrough but to send out announcements if warranted.
Consistently seeking opportunities to spread the word about the importance of mathematics, he once introduced Lynn Simarski, one of Rita Colwell's speech writers, to this reporter as "one of the most important people at NSF."
This is high praise, for he spoke several times to SIAM News of the "amazing talent, creativity, and dedication to public service" of the people he has worked with at NSF. His esteem for his "incredibly good" DMS colleagues is whole-heartedly reciprocated.
"His passion engaged everyone around him," Deborah Lockhart said at the May reception, speaking for all the program directors. Recently returned from a 12-day trip to Russia in the company of Tondeur and other DMS staff, Lockhart told the audience of arriving somewhere in Russia at the point of total exhaustion, while Tondeur was "fresh and ready to go out and meet people." This is pretty much the way things were during his entire three years at NSF, she said. "He charged up the people who worked for him."---GRC
*The "Odom report," Report of the Senior Assessment Panel of the International Assessment of the U.S. Mathematical Sciences, can be found at www.nsf.gov/pubs/1998/nsf9895/start.htm. Information about all the programs mentioned in this article can be found at www.nsf.gov/mps/divisions/dms/.