Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 25
3 FUNDING OPPORTUNITIES
Discussion under this topic encompasses how to motivate faculty to apply for
grants, both as individuals and through cooperative efforts; some extraordinary
sources of grants, including equipment andfacilities grants; and how to enhance
cross-disciplinary collaboration and research.
LOVE S LAB ORS WON: NONCONVENTIONAL
FUNDING FOR THE MATHEMATICAL SCIENCES
James Glimm (Organizer), State University of New York, Stony Brook
OUTSIDE FUNDING
Daniel Goren~tein, Rutgers University
DEVEEOPMENT GRANTS
Calvin C. Moore, University of California
25
OCR for page 26
OCR for page 27
FUNDING OPPORTUNITIES
.
LOVE S LABORS WON:
NONCONVENTIONAL FUNDING FOR THE MATHEMATICAL SCIENCES
James Glimm
State University of New York, Stony Brook
"Love's Labors Won" was chosen as the title for this presentation to remind you that this is not a discussion
of how to apply for a grant. It is really a discussion of how to win a grant. We assume that you are familiar with
the process for obtaining support from the National Science Foundation. Here we are considering the
complement of the National Science Foundation/Division of Mathematical Sciences.
The winning of a grant begins, of course, with the submission of a proposal. So we will consider the
ingredients of a good proposal. First, a good proposal should contain quality science. I will not attempt to advise
you on that issue.
Second, the proposed project should have realistic goals and potential applications. The list of goals should
include objectives that can be achieved in one to three years. Longer-range goals may also be included.
A third ingredient of a good proposal is political support. For example, there may be a technology transfer
component to the proposal. In this case, political support means that one has identified a user group and the
proposal contains a mechanism by which the user may interact with the new science as it is created. That support
could be expressed in the form of letters from the potential users. Stronger support consists of letters from
potential purchasers of the new science who pledge to contribute to the support of the effort.
Finally, the requested level of funding must be realistic. This includes the proposal itself and the costs of
actions that might be based on the proposal.
These ingredients for a good proposal were set forth by a member of the New York State legislative staff
at a symposium I attended on studies of water quality on Long Island. However, Here are some lessons there
for mathematicians, also.
So, how does one win a nonconventional grant? Knowledge of the program is a key ingredient. Each
program has goals, people, and program officers. It is essential that a grant applicant have this information.
Additionally, nonconventional grants usually require cooperation with other disciplines. Often mathematics is
only a part of the project.
The reputation of the proposer is also important. This is related to the quality of the science involved.
Finally, successful nonconventional proposals usually have some level of cost sharing from various areas,
including the university, the state, the community, and the private sector.
I will close with a brief checklist of things to do in preparing for the submission of a nonconventional grant
proposal.
.
1. Make an inventory of your opportunities.
· What are the local industries? What are the technical problems of local industries? Invite the technical
staff to have informal discussions with faculty and students so that you can determine the mathemati-
cal aspects of their problems. Try to place a student on the staff.
What are the local governmental laboratories? (Use the same approach as above.)
27
OCR for page 28
CHAFING TO MATHEMATICAL SCIENCES DEPARTMENT OF TO 1990S
· What are the strong technology and science departments in your university? What kind of mathema-
tics do they use? What courses do they teach? Is there a possibility of joint Ph.D. theses?
· What are the existing contacts from your department?
· What are the areas of mathematical strength in your department? What are the potential applications
of these areas? Based on these strengths, can contacts be created?
· What are the important problems? Where does technology fit in? What mathematics is involved?
2. Identify funding programs.
· University
· Governmental (state, local, and federal)
· Industrial
· Private
Begin with small-scale internal resources.
· Seminar
· Key visitors/technical conference
· Cross-disciplinary students
4. Look for promising problems.
· Establish a research program.
· Establish interdisciplinary collaboration.
Learn the subject from the viewpoint of other disciplines.
Pre-review proposals.
28
OCR for page 29
FUNDING OPPORTUNITIES
OUTSIDE FUNDING
Daniel Gorenstein
Rutgers University
I speak in three distinct capacities: first, as a former department chair, then as a member for many years of
a university committee concerned with priorities in research and graduate education, and finally as the director
of the new NSF Science and Technology Center in Discrete Mathematics and Theoretical Computer Science
(DIMACS).
Let me say first that I have found that, given any reasonable expectation of success, mathematicians are
sufficiently self-motivated to apply for individual research grants. However, especially but not exclusively, the
problem is rather the preparation of an effective research proposal. As chair, I wrote up a detailed set of
guidelines. I then critically read drafts of the proposals of junior faculty (and any senior faculty who asked for
my opinion). On a number of occasions, I got the faculty member to give me a lecture about the proposal so that
I could determine what he/she was really after. I like to think that in at least a few cases, my help made the
difference between being funded or not. This job need not be limited to the chair, but could be assigned to any
experienced grant-getter in the department.
In my committee member role, I looked for ways of increasing faculty incentives to apply for grants. As
I said, this is not so important in mathematics with respect to individual grants, but is important in other fields,
especially the experimental sciences, and also in interdisciplinary or unusual types of grants. Rutgers instituted
a policy of a return of a percentage of overhead to the department. This gave departments a strong interest in
having their faculty seek outside funding. Other areas where the committee affected university policy on
research grants included (a) the total percentage of allowable summer salary (up to 3/9 for someone having more
than one grant), (b) reduced teaching for faculty with academic year support, (c) tuition remission for graduate
students, and (d) reduction in university overhead rates (included as part of the university cost-sharing) in
connection with unusual grants that often have programmatic funding limitations.
My own feeling is that, in the end, such incentives normally play only a marginal role in who seeks external
funding and who gets support. In most cases, the inner drive of the faculty member and the merit of the proposal
itself are the dominant factors.
We come now to extraordinary type grants. Certainly the grant to DIMACS (as our center is called) falls
in this category. I can describe how the center works in any amount of detail you would like to hear, but in these
comments I would prefer to focus on what was involved in putting together a successful proposal. It goes without
saying that without the existing scientific strength in the four participating institutions Rutgers, Princeton,
Bell Labs, and Bellcore-no amount of effort would have availed. But with only 11 centers funded out of 322
proposals received, this was a necessary but far from sufficient condition for success.
The NSF solicitation request stressed four things:
1. Proposed centers should be primarily research-oriented.
2. It should involve university-industrial cooperation to strengthen U.S. competitiveness.
3. The proposal should include a significant educational component.
4. It should also include a "knowledge transfer" component to ensure that the fruits of the center's research
be widely disseminated.
29
OCR for page 30
CHAIRING TO MATHEMATICAL SCJENCES DEPARTMENT OF TO 1990S
I have no doubt that the ultimate success of the Rutgers proposal was in part a result of the considerable
effort made to meet these requirements: first, by involving both Bell Labs and Bellcore as participating
institutions along with Princeton and Rutgers; second, by designing research programs that struck a balance
between basic research and possible industrial applications; and third, by including in our educational
component two pre-college programs of foundational material: one for high school mathematics teachers, the
other for high school students with mathematical ability and interest.
Now comes the task of turning this written proposal into a living reality. With four distinct partners, each
in its own geographical location and each with its own rules of operations and organization (and with the
Center's independent of lice space still a few months off), this has not been easy to achieve. Nevertheless, even
though we have been in full operation for only a very short period, I believe we are on our way to implementing
the proposal effectively.
ADDENDUM: PREPARATION OF NSF PROPOSALS
Over the years, I have had a lot of dealings with NSF proposals. At first, I was primarily a reviewer, but
in recent years, I have helped some of the junior faculty with the preparation of their proposals. This was done
more or less on an ad hoc basis someone would ask me to look over a proposal, and I would make some
suggestions for improving it. Or I would go over referees' reports with someone who did not get funded the
previous year to see if the criticisms that were leveled against the last proposal could be avoided this time around.
This year I decided to be a little more systematic about it, and I have just read through all proposals from
junior faculty who either were not funded last year or are new members of our department, with a view to helping
them improve their proposals. (I would be happy to do the same for any senior faculty member, but feel it is
out of place for me to do that on my own initiative.)
I have been struck all along by the uniformity of criticism which I tend to make of the proposals I read. It
therefore seems a good idea for me to detail some of the general ideas that I have about NSF proposals and the
NSF itself. That is the purpose of this memorandum. Perhaps these comments will be useful to senior faculty
as well, and so I am distributing them to the department as a whole for whatever it is worth.
1. Quality of the proposal versus quality of the individual
I believe mathematicians have a misconception about the weight that the NSF assigns to the quality of the
actual proposal that is submitted. I know many whose proposals are only a page long sometimes less just
a bare statement of the problems. The assumption is that they are really "good enough" mathematicians to get
funded if they simply sign their names to the proposal. The fact is that this is true in many cases; but it primarily
applies to outstanding, well-established mathematicians.
An individual who falls only slightly below this level puts himself/herself in jeopardy by submitting such
a proposal. Once a referee criticizes an actual proposal, even if he says fine things about the person, it makes
it much more difficult to get funded. As far as I can tell, the NSF goes only on the basis of its referee's reports;
that is the sole criterion it uses (at least of ficially) in reaching its decisions. Having had many one-page and one-
paragraph proposals to review over the years, I can assure you that it is often a strain on the reviewer to evaluate
such a proposal. One has to guess what was in the proposer's mind, and many times the reviewer is unwilling
to make that guess the same way as the proposer. I know specific cases of leading mathematicians whose
proposals were criticized and, as a result, were not funded.
Short proposals often suffer from vagueness. Without sufficient detail, it is not always clear what the
30
OCR for page 31
FUNDING OPPORTUN1T~S
proposer really has in mind. The proposer very often leaves out essential aspects of the proposal, assuming that
they are implicit in what he has written, whereas, in fact, a common criticism by NSF reviewers is that a proposal
is "vague."
If one will contrast what goes on in other fields, one will see at once the difference in attitudes toward
proposalpreparation. Iknow many cases where the preparation oftheproposal takes amonth, orlonger, of really
hard work on the part of the proposer.
The NSF has just issued new guidelines on lengths of proposals: a maximum of 15 pages. Presumably they
had been getting proposals of 75 pages or even longer, but l5 pages is not the same as 1 page! For ajunior faculty
member, these comments are important.
Of course, don't get the idea from this that individuals are not also judged by reviewers. Certainly a reviewer
may comment on someone's abilities, past accomplishments, or lack thereof, sometimes favorably, other times
very harshly. But this is something over which one has no control. It is not clear how anything one puts into a
specific proposal can influence a reviewer's judgment of an individual's stature in the field.
2. Significance of the proposal
Most of the `'junior" proposals I have read make little or no attempt to put the problems in any kind of general
context and rarely indicate the impact of a positive (or partial) solution on other problems. It is just taken for
granted that the reviewer, being an expert in the field, understands all this, and will assume that the proposer
does as well. Hence it is presumptuous-perhaps even patronizing for the proposer to elaborate the full
significance of the problem area But the reality is the exact opposite the reviewer, in general, will not make
any such assumption about what the proposer knows, particularly if he doesn't know the proposer at all, and
maybe even if he does. Another problem is that the reviewer may well have a totally different opinion as to what
is "significant" than the proposer has. Maybe no matter what the proposer writes, the reviewer will not regard
it as an important problem (and perhaps he'll be right), but if the proposer makes a strong case as to why he's
planning to work on this particular problem, he may influence (or at least temper) the reviewer's judgment of
its significance. Believe me, for a reviewer to write that "this problem is not very significant" is usually the kiss
of death.
In applied mathematics, especially in interdisciplinary areas, the difficulties are compounded, for one can't
even go on the assumption that the reviewer is, in fact, an expert in the field. NSF has much more trouble finding
appropriate referees in such areas than in well-established clear-cut fields such as "classical harmonic analysis"
or "group theory." So a great deal of thought has to go into this aspect of the proposal. In this connection, it is
possible for a proposer to suggest to the NSF a list of potential reviewers.
3. The likelihood of success
I find that this represents one of the weakest parts of the proposals I read. Many individuals content
themselves with a precise statement of the problems to be attacked and the results obtained so far toward their
solution. That is appropriate for a paper, but totally off the mark for an NSF proposal. Well, I should modify
this. If one has a reasonably substantial record of having solved problems in the past, it may be entirely sufficient
to write the proposal in this way. The referee will most likely assume that the individual has the capability of
solving the problem, or at least obtaining something worthwhile on it. But again, one is taking a gamble. I know
specific cases in my own field in which the reviewers, despite the clear significance of the problems, were unable
to decide from the proposal whether the individuals had any ideas as to how to solve them. These were middle-
level junior people of very high ability, but they did not get funded. It is easy to propose to work on the Riemann
31
OCR for page 32
CHAD{TNG TO MATHEMATICAL SCIENCES DEPARTMENT OF TO 1990S
hypothesis, but I guarantee that despite its significance, not many of us would get funded in that direction!
I think that an essential ingredient of any proposal (especially for junior faculty) is to explain in some
intuitive way what one has in mind. I know that mathematicians, being such an honest group, are reluctant to
make claims about what might or might not work on a given problem. But it's not a question of learning how
to "exaggerate." What is required here is an explanation of the techniques you plan to use, why you think they
have a chance of working, including special cases that you may have solved, examples that you may have
worked out already, partial results you may already have obtained, etc.
One componentofareviewer's evaluation will almost certainly be his judgment of We proposer's chance
of success. Moreover, the question of "success" is extremely important to the NSF. In examining how the NSF
goes about justifying its funding from Congress, one will discover a very detailed discussion (in general terms,
of course, for Congress) of specific problems that have been solved on NSF grants. For example, last year's
mathematics request to Congress referred specifically to Sims and Leon's construction of the "baby monster."
4. Summary
A good proposal should include the following items:
a. Sufficient detail.
b. The significance of the project, and
c. Justification for thinking problems can be solved.
32
OCR for page 33
FUNDING OPPORTUNIT~S
DEVELOPMEN r GRANTS
Calvin C. Moore
University of California
Based on experiences at the Mathematical Sciences Research Institute at Berkeley and in university
administration as a vice president, I would like to share with you some ideas about funding opportunities and
strategies. But first I will discuss the general situation in our profession.
Ph.D. production in the mathematical sciences in the United States has declined steadily to approximately
750 per year. The most alarming part of this number is that half of these degrees are being earned by students
who are not U.S. citizens. There is a general view that we are not renewing the profession at the necessary rate.
We cannot continue to rely on foreign students because of the increasing opportunities these students will have
as opportunities and funding for research increase in their home countries. Currently, we are able to induce a
large number of these students to remain here and contribute to the profession.
I am happy to say that the most recent data indicate at least a bottoming out of the decrease in the proportion
of U.S. mathematical sciences Ph.D. degrees earned by U.S. citizens. Perhaps this is the beginning of a
turnabout. However, we have a very serious problem.
There is a need to renew the profession, not only in the mathematical sciences, but in all disciplines. For
example, the University of California, all campuses, will need to hire 10,200 faculty in the next 16 years. The
recent book by Bowen and Sosa, Prospects for Faculty in the Arts and Sciences, paints a similar picture
nationwide. Additionally, the NSF has projections of the doubling of the need for terminal degree holders in
science and engineering. The NSF also projects that the current level of production will fall short by a factor
of 50 percent. The report Wor1fforce 2000 2 indicates that of the total additions to the work force through the year
2000, only 15 percent will be white males. This has enormous implications. It means that we will have to
capitalize on the talents of women and minorities to a far greater degree than before. In the past one spoke of
affirmative action programs as a rationale based on social justice. Now the rationale is much wider; it is a matter
of survival.
In brief, I am sketching a human resource problem of major proportions that must tee addressed at all levels,
from kindergarten through postdoctoral years. Given this situation, there are excellent opportunities for
funding. I would like to suggest to you a certain type of proposal that has certain analogies with the mathematics
institute proposal. I call these, for lack of a better term, development grants.
This kind of proposal would have a number of components, including graduate student support,
postdoctoral support, and considerable attention to undergraduate programs to increase the flow of students into
graduate school. Also, special attention would be given to enhancing the participation of women and minorities.
The most important component is a well-thought-out plan. In other words, the proposal would be for a type of
mathematics institute within a department.
Bowen, W.G., and J.A. Sosa, Prospectsfor Faculty in the arts and Sciences (Princeton University Press, Princeton, N.J.,
1989).
2 Johnston, W.B., and A.E. Packer, Wor~orce 200kWork and Workers for the 21st Century (Hudson Institute,
Indianapolis, Ind., June 1987).
33
OCR for page 34
CHAFING THE MATHEMATICAL SCIENCES DEPARTMENT OF THE 1990S
Recall that the mathematics institutes have as one of their primary missions the training of postdoctoral
fellows, that is, providing a research experience for them in a rich atmosphere of senior investigators and a flow
of ideas that would enhance their career potential. That is a human resource program at the postdoctoral level.
The proposals I am discussing would be for human resource development at all levels. Precedents for this exist
in the National Science Foundation, in the Computer Sciences Division and in the Biological and Behavioral
Sciences Division. The prospects for the funding of such proposals are enhanced if the application clearly states
the kinds of resources the department is willing to contribute to the project.
These grants address human resource issues. My reading is that such proposals would resonate with current
thinking in the federal government, especially at the NSF. According to Dr. Judith Sunley, there is a set-aside
in the Division of Mathematical Sciences of the NSF for proposals that address increasing the participation of
. . .
women and minorities.
The NSF has proposed to issue a rather remarkable document, Important Notice 107. This notice adds
requirements to ordinary grant proposals. Specifically, each proposal must include a statement specifying the
potential of the proposed research to contribute to the education and development of human resources in science
and engineering at the postdoctoral, graduate, and undergraduate levels. This statement may include, but is not
limited to, the role of the research in student training, course preparation, and seminars, particularly for
undergraduates. Special effectiveness or achievement in the area of producing professional scientists and
engineers from groups currently underrepresented should be described.
This statement is set forth in a document that applies to all proposals submitted to the NSF after it becomes
effective. It is a clear signal of a change in the reward structure. It addresses a problem that was discussed earlier
today: How do we convince faculty to devote their energies to issues that go beyond research and include human
resource development? The NSF has taken an important step. Research proposals are required to address these
issues and combine them with research.
34
OCR for page 35
FUNDING OPPOR~N1TIES
QUESTION-AND-ANSWER SESSION
PARTICIPANT: I noticed that two of you mentioned the fact that our graduate students are the best means of
technological transfer. I would like to amend that. They should be, but they are not. It is clear that we have the
best research establishment in mathematical sciences in the world; yet we have one of the poorest technological
transfer establishments in the world. I suggest, as a possible solution to this, that a portion of the graduate
programs in mathematical sciences should be changed to address that issue.
DR. GLIMM: Graduate students are the best means of technology transfer if they are transferred. However, in
many cases they pursue academic careers. I think you are raising the following questions: Should our graduate
education be broader? Should we expect a certain fraction of our students to have industrial laboratory careers
and an appropriate education? Affirmative answers to these questions would assure that the technology would
be transferred through those students. Perhaps we were neglecting that kind of employment consideration in
. .
c eslgn~ng our programs.
DR. GORENSTEIN: For a certain percentage of students, that is very appropriate. Maybe we have neglected
it. But there is still core mathematics. The best way to train people for that is to let them prove theorems. I don't
know the proper balance.
PARTICIPANT: Technology transfer has happened more in statistics than in core mathematics. One of the
reasons is that, historically, statistics has recruited into graduate programs students with undergraduate careers
in application areas. These persons pursue careers in industry and government more than mathematicians do.
It is not easy for statistics to recruit these people; I am sure it is more difficult for mathematics to achieve, but
that appears to be a source of technology transfer people.
PARTICIPANT: I was struck by Dr. Gorenstein's comment about departments at Rutgers getting a return of
indirect costs. I have the task of trying to elevate the level of funding in my department. I am curious as to what
fraction of indirect costs comes back to a typical mathematics department.
PARTICIPANT: About 5 percent.
PARTICIPANT: The University of Maryland gives back 25 percent.
DR. GLIMM: In addition to the routine recovery, I think unusual situations sometimes allow for a larger
recovery.
PARTICIPANT: Just a comment to follow up-that is not the sole measure of how well a department does.
Many universities reward departments that bring in a lot of money in other ways. They may increase their basic
budgets, for example. We get 30 percent return, but our basic budget is about one-quarter or less of what we
need to operate. So it is just a matter of how the money is passed to the department.
35
OCR for page 36
CHASING TO MATHEMATICAL SCIENCES DEPARTMENT OF TO 1990S
DR. GORENSTEIN: Princeton University gives back zero. However, from my experience, the computer
science department is treated very well atPrinceton. Every faculty member has a secretary. So one mustexamine
the way in which a university operates vis-a-vis the department.
PARTICIPANT: I know that in hiring senior faculty members, particularly in the physical sciences, it is
becoming more common for administrations to agree to return a fraction of grant overhead generated by that
person as start-up costs. It seems to me that if one is hiring a very strong faculty member in mathematics, one
could get the same kind of agreement for start-up costs from the dean that the physics chair is getting for his
people.
DR. MOORE: The issue of start-up costs is a concern from the perspective of provosts and vice presidents. It
is, on average, $100,000. That is averaged over all fields, including the humanities. For senior people in the
physical sciences, it can go as high as $1 million. This is a form of subsidy to those departments. You are quite
right that mathematics departments have not entered into the sweepstakes to the degree that they might.
36
Representative terms from entire chapter:
funding opportunities
