Sigma Xi FORUM 1995
          "Vannevar Bush II: Science for the 21st Century"
(Why Should Federal Dollars Be Spent to Support Scientific Research)
                          March 2-3, 1995

"Science and Society in the Decades Ahead"
     Frank Press (Former President, National Academy of Sciences;
     currently Chair of the National Research Council Committee on
     the "Future of Science," which is coordinated through the
     National Academy of Sciences, the National Academy of
     Engineers, and the National Institute of Medicine - also known
     as the "Vannevar Bush II Panel")

Why is the U.S. looking at a "revolutionary" approach to the future
of science, rather than the evolutionary approach taken after the
Vannevar Bush report in 1945?  V.Bush I was certainly very
successful, especially given how it was crafted in an era of turmoil
and world strife.  However:

1) The political situation in the world has changed with the
     cessation of the cold war.
2) The major threat to the U.S. is now that of "rogue nations," some
     with nuclear capabilities.
3) The industrial and commercial competition world-wide has changed.
     There is an increasing emphasis on products with short "life-
     cycles."  This approach places the emphasis on incremental
     improvements rather than evolutionary new products.  (Hence,
     there is a shift to more emphasis on applied research than
4) Traditional problems (e.g., population, diseases, etc.) have
     taken on new dimensions in this global environment.
5) There is a need to re-think institutional structures.  (e.g.,
     Industry has re-engineered its management and manufacturing
     structure in recent years, cutting out whole layers of middle
     management.  Should universities do the same?  Should
     government also be re-structured?)
6) There are extraordinary opportunities in many scientific fields.

The "V.Bush II" study (aka the NRC Study on Criteria for Federal
Support of Research and Development) was mandated by Congress and is
bi-partisan.  Among the charges are to determine:

     a) appropriate criteria for apportioning R&D funds.
     b) the appropriate balance between different types of research
          institutions (e.g., universities, national labs, etc.).
     c) appropriate objectivity in the allocation process.
     d) appropriate means for funding "cross-cutting" processes.
     e) appropriate use of results from research.
     f) appropriate criteria for accountability.

The NRC "V.Bush II" panel consists of 17 members drawn from a
diverse group of constituencies (including a university president, a
former senator, bench scientists, etc.)

Questions being considered by the panel:
     1) What did we do "right" to get us where we are?
        What should we "preserve?"
          a) The peer review process?
          b) The lack of a centralized "Ministry of Science"
          c) The "tie-in" between graduate education and research?
          d) Our outstanding manufacturing capabilities (for
               scientific research instruments, etc.)?
     2) What criteria should be used for allocation of research?
          a) National Needs/National Problems?
          b) Does creation of new knowledge "pay for itself?"
          c) Does serendipity really work?
          d) Does rapid progress in a field, or cross-disciplinary
               efforts among several fields, have a priority?
     3) How can we improve technology transfer?
     4) How can we compensate industry for industrial investments in
          research? (e.g., tax credits, etc.)
     5) What should be the balance between national labs,
          universities, other research entities?
     6) How can we make federal funds go further?
     7) Should there be a "Department of Science" in the U.S.
          government?  Or should science stay decentralized?
     8) What should be the future of graduate training?
     9) What is the appropriate government role in technology
          research with industry?  (e.g., the Advanced Technology
          Program, the Technology Reinvestment Program the Small
          Business Program)
    10) Should we re-engineer Congress & its oversight of research?
          There are currently 26 committees deciding allocation of
          funds for various areas of scientific research.
    11) Is there a better way to make the "case" for continuing
          government support of science?

The committee is interested in hearing from the scientific community
at large.  They can be reached electronically via their e-mail
address (r-study@nas.edu).

"Bohr, Pasteur, and Edison: Models for Science"
     Donald Stokes (Woodrow Wilson School for Politics & Public
     Affairs, Princeton)

The German "University" provided the model for the research
university in the 20th century.

Prior to World War II, funding for research at universities was
provided primarily by private funds, student tuition, etc., but not
from federal sources.  Government research was done by government
labs such as the Smithsonian, the Geological Survey, etc.  (The one
difference was the establishment of the Agricultural Experiment
Stations by the Hatch Act in 1887.)

Vannevar Bush changed this (through the implementation by Franklin
D. Roosevelt).  Bush also proposed a "National Research Foundation."

There are two major aphorisms from Bush's report:

"Basic research is performed without thought of practical ends."
"Basic research is the pacemaker of technological innovation."

The "National Research Foundation" concept failed, but Bush's
ideology triumphed.  (NRF failed in that it fragmented in the first
five years into the Atomic Energy Commission, the Department of
Defense, and the National Institutes of Health.)  The creation of
the National Science Foundation (NSF) in 1950 represented a "gutted-
out" version of the NRF.

Another major event in the development of federal funding for
research was the launching of Sputnik, which ushered in the "golden
age" for American science.  Virtually every field of science

Basic vs. applied research -- can the concept be defined in terms of
quadrants?  e.g.:

                 Are there considerations of use of the research?

                               NO             YES
                       |               |                |
                       |  Pure, Basic  | Use-inspired   |
                   YES |  Research     | Basic Research |
  Is there             |  (e.g. Bohr)  | (e.g. Edison)  |
  a quest for          |               |                |
  fundamental          +---------------+----------------+
  understanding?       |               |                |
                       |  (Hopefully   | Purely Applied |
                    NO |     empty)    | Research       |
                       |               | (e.g. Pasteur) |
                       |               |                |

          [However, even though Pasteur was interested purely in the
          applied aspects of his research, he ended up founding an
          entire field of science (microbiology) from his applied

A truism - technology is very much driven by scientific discoveries.
(However, increasingly, science is becoming driven by technological

"Answering The Questions That Have Not Been Asked"
     Neal Lane (Director, NSF.  Note: Lane's original talk was
     "Federal Support of Science."  However, probably due to the
     uncertainly of federal scientific funding in the new Republican
     Congress at the time of the Forum, Lane gave this alternate

In the last five years, there have been major changes & adjustments
due to the world political situation.  What does this mean for the
future of universities?  Times are changing... (Collision forces,
the tension between research and teaching, the crumbling research

Basic research - the concept of "unasked questions."  What is the
role of NSF in this pursuit?

     Eugene Ionesco - "It is not the answer, but the question which

"Strategic research" should really be called "research in strategic

Today - all major players in scientific research are re-examining
their roles:

     1) National labs - their role w/respect to industry
     2) Universities - what should they be in the 21st century?

NSF will continue to support basic research, but the emphases will
change.  There is a need to facilitate change and new ideas at
universities and colleges.

Teaching and research should not be separated universities.  This is
becoming a more and more contentious political issue.

A recent series of three articles in the Washington Post addressed
the future of scientists and universities.  It mentioned that 65% of
faculty time is spent writing proposals for funding, and that most
of the research is done by graduate students, technicians and post-
docs.  The implication is that most of the best researchers are not
only not teaching, but also not doing research (which is being

"New Partnerships Between Government and Industry"
     Graham R. Mitchell (Asst. Secretary, U.S. Dept. Commerce)

Global Commerce is driven by rapidly-moving technology.  (e.g.,
advances in information technology - driven by increases in
communications technology.)

Scientific ages:         Stone Age
                         Iron Age
                         Bronze Age
                         Steel Age
                         Steam Age
                         Electricity Age
                         Nuclear Age
                         Space Age
                         Information Age
                         Biotechnology Age (coming)

The "life cycle" span for each succeeding age is getting shorter.

Funding for research by industry exceeds the combined total from
federal and civilian sources.

The administration's strategy for growth is 4-pronged:
   1) People (middle class bill of rights, etc.)
   2) Investment (deficit reduction, etc.)
   3) Innovation (regulatory reform, etc.)
   4) Open markets (NAFTA, GATT, APEC, etc.)

The private sector should be viewed as a "customer."  There should
     Incentives for savings and capital formation
     Regulatory reform
     Product liability reform, etc.

One major problem - there are different priorities for different

Chart - utilization of research can be viewed as an "Innovation

                       |                             .
                       |                           .
                    D  |                        .
                    O  |                     .
                    L  | ...........       .
                    L  |            .     .
                    A  |             .   .
                    R  |              . .
                    s  |
                          Basic     Development   Commercial
                         Research    & Start-up    Operation

The "Basic Research" component represents $26 billion of federal
funding.  The "Commercial Operation" represents more than $100
billion (mostly industry).  The "Development & Start-up" period is
often viewed as the "Valley of Death" in the Innovation Pipeline.

"Health Science Research in the 21st Century"
     Kenneth Shine (President, Institute of Medicine, NAS)

Health science research has benefited significantly from scientific
achievements in all fields.

However - now 14% of the gross national product goes to health costs
alone (with no indication that health in general has improved).

Suggestion - six critical issues:

1) There is a transition from a "war on disease" to a "campaign for
     health."  (e.g., looking at environmental impacts on health.)
2) Science in general (and health science in particular) has been
     reductionist (i.e., from organismal down to the cellular and
     molecular level).  The future will see a move back to full
     integration at the organismal/individual/population level.
3) U.S. medicine has used science, but has not been practiced
     scientifically.  New technologies (e.g., data systems) will
     become of increasing importance.  (e.g., Telemedicine, remote
     diagnosis and surgery, virtual reality.)
4) Behavioral and social sciences will be increasingly important to
     the health sciences, particularly for preventative health
     practices (e.g., by influencing/changing behavior patterns).
5) Health science research will be increasingly concerned with
     ethical issues.  (e.g., If the immune response system can be
     suppressed, there is a possibility that animal organs can be
     transplanted to humans.  This raises all sorts of ethical
6) Technology transfer - once one has the new technology, when and
     how often should it be used?  (e.g., Will a technology be used
     just because it exists?)

Closing observations:

The "Evil Empire" (i.e., the USSR) is now gone.  It is time to
tackle the "3 P's & a D."  (That is - Poverty, Pollution, Poor Jobs,
and Disease.)  There is no longer simply a "war on disease."

Any list of goals for the new science should include "A Healthy and
Human Society."

"Reinventing the Research University"
     Kumar Patel (UCLA; President of Sigma Xi)

What will be the future "shape" of research universities?  Since WW
II, there has been a mutually beneficial tie-in between research and

(Proceedings will soon be available from the June 1994 conference in
California on "Reinventing the Research University")

Changes:  1) Expectations will be tied to economic growth
          2) The face of graduate education will change
          3) There will be funding issues (even at the national labs)

Recommendations?  (from conference)

     1) Administrators and faculty should be pro-active in planning.
     2) Universities need to establish better links with industry
          and government.
     3) Universities need to address the changes required, such as
          the "Renard" structure.  (e.g., the tenure system)
     4) Universities need to develop better skills at dealing with
          intellectual property rights.
     5) Universities need better links between research and
          undergraduate education.
     6) There is a need for new approaches to make universities (and
          faculty) more accountable to their constituencies.

"Science and Society: the Troubled Frontier"
     Sidney Drell (Stanford Linear Accelerator Center)

1) Public and government officials do not understand science.
2) Science is an engine of change, but society views it as a hazard.
     (There is a general mistrust of science.)
3) The idea of partnership between industry and government (as
     envisioned by V. Bush) has seriously eroded.  (e.g., there is
     the perception of "over-regulation" by government)
4) There are serious mismatches between the planning timescale of
     resources offered by society/government and the timescale of
     scientific research.

Question - "Did the A-bomb create false expectations of the promise
(and rate of progress) of science???"

There is a problem in that externally set agendas for short-term
solutions of strategic goals (e.g., the A-bomb, cancer, AIDS) may
have unrealistic expectations.

Science should become part of the American curriculum, equal to
reading, writing and arithmetic.  (This is increasingly important in
this technological age.)

Industry can be one of our most powerful allies.

"How Graduate Education Must Be Changed"
     Phillip Griffiths (Chair, NAS Study on Graduate Education)

The NAS panel is considering the graduate education process for
scientists and engineers.  They have conducted lots of interviews,
and have received over a hundred written responses.  The report is
still in the review stage.

Major findings:

     Myth: Most Ph.D.'s go on to careers in academia.
     Fact: About half of Ph.D.'s go off to jobs where research is
               not the primary activity.

     Myth: There is a high unemployment and underemployment of
     Fact: Unemployment is approximately 2.1%, which is about 1/3 of
               the national unemployment rate.  However,
               expectations of a first position (and "first
               employment") is a problem.

     Myth: The number of Ph.D.'s in science and engineering is
               increasing exponentially.
     Fact: Growth of Ph.D.'s is approximately 2-3% each year, with
               most of this occurring in the non-U.S. citizen
               category.  However, growth is twice that of the
               general population.

FACTS: 1) The time to degree and time to employment has increased in
               all fields (up to 40%).
       2) There are more and more Ph.D.'s taking post-doctoral
               positions as their first job.

Some "deluded" solutions:
     1) "Close down some Ph.D. programs" (especially at schools
               other than one's own!)
     2)  Re-design the Ph.D. to be more "vocational."

New scientists and engineers must be more flexible, as they may end
up in non-academic positions, or may be involved in team approaches
to research (as opposed to the traditional "go it alone").

How do we change the system?
 1) We can change the way we reward academic research.
 2) We can change the way we conduct research and graduate
     education.  (The current model is that the Professor runs the
     laboratory, and the research is paramount.)

 1) Make graduate programs more flexible, and provide more
     opportunities for graduate students.  For example:
          a) For some students, a Master's may be sufficient.
          b) There may be a need for a "Two Master's" option.
          c) Ph.D. - should prepare for more than just a career in
               traditional research.
          d) "Enrich" the Ph.D. with a related Master's program or
               with additional formal coursework.
     The idea is to broaden careers, and provide more options.
 2) There should be better information and guidance for graduate
     students.  A national database on employment opportunities and
     trends can be established.  Departments can help by monitoring
     the fates of their own graduate students.
 3) The time to degree could be "tightened."  e.g.,
          Phase I   -  2 years of coursework
          Phase II  -  2-3 years of research training
          Phase III -  "post-doc" phase (for polishing skills)
 4) New kinds of education grants (i.e., education training grants).
     Similar to the Pew Charitable Trust "Training the Future
     Professoriate" program, but with a research component and an
     intent to lessen the time to degree.

The GUIR (Government-University-Industry Roundtable) might be a good
vehicle for implementing this type of change in graduate education.

"The Washington Perspective"
     Martha Krebs (Director, Office of Energy Research, U.S. Dept.

Clinton & Gore (1992/93) had a science agenda:
     1) Technology for economic growth
     2) A National Science & Technology Council
     3) A Climate Change Action Plan
     4) The National Information Infrastructure (NII) program
     5) Technology for a sustainable future
     6) "Science in the National Interest" (released 8/94)

What does the new Republican Agenda for 1995 mean?
     1) Fundamental science
     2) Strategic research
     3) Industrial partnerships
     4) National labs
There are stresses among all of these objectives.

The federal budget process: "The President proposes, Congress

Krebs' view of Energy research priorities:
     1) Science Facilities Initiative
     2) Renewing High Energy Physics
     3) Energy and the Environment

     Current federal funding for basic research:
          NIH  --  $6.31 billion
          NSF  --  $2.1  billion
          NASA --  $1.8  billion
          DOE  --  $2.8  billion
          DOD  --  $1.21 billion

"What can Social Science Tell Us about Solving Societal Ills?"
     Neil J. Smelser (Director, Center for Advanced Study in
     Behavioral Sciences)

How can scientific research deal with social problems?
     1) A continuously changing panoply exists due to increasing
          technology.  (For instance, there are new ways to cheat
          and get around technology, such as credit card fraud.)
     2) Intensification of various environmental problems.  (Is this
     3) There is a continuation of problems related to international
          inequality and differential development.
     4) The appearance of "our" types of social problems in other
          parts of the world.  (e.g., divorces, the spread of health
          problems, etc.)
     5) Many social problems will be created in areas other than
          where they appear, but will be under the jurisdiction of
          the home country.  (e.g., Conditions of factory workers in
          Asia that are manufacturing products for an overseas
     6) Social problems are becoming less and less localized, and
          more and more tried in the court of public opinion.  (In
          particular, by the international press)
     7) Expectations (for growth and democracy) may increase world-
          wide social problems.

In the social sciences, early on (in the 1930's) they took two
lessons from the natural sciences:
     1) Social problems are identifiable object themes, which can be
          analyzed and solved by research. A solution can be by a
          "social invention" (e.g., by law or by an organization).
          This was the "biological science" model, but now has been
          largely discounted.
     2) Social problems are a kind of "disease," which can be solved
          with a pathological approach.  (e.g., can "ills" be solved
          by a social "vaccine?")  This was the "medical" model.

Neither approach (which involve "simple solutions") has been
successful.  Nonetheless, expectations are high, and when these
solutions fails they are subject to disappointment and ridicule.

An alternative theory: Social Problems represent a set of empirical
assertions that are embedded in a set of social context accepted by
a larger "whole."  This approach requires a "normity" or "value"
structure to identify social problems.  These would also be subject
to chance.

Sometimes social problems only become problems as values and public
morality changes.  (e.g., child abuse is now considered a problem,
although an old adage is "spare the rod, spoil the child")  These
types of problems might be considered as "pre-existing" conditions.

In fact, a "social problem" is a political process, not a "thing."
(Splinter groups often lobby for their own political agendas - for
instance, the NRA doesn't think assault weapons are a social

Where are the points of entry, where knowledge can be brought to

1) Studying the process itself?  Where can knowledge be brought to
2) Arbitrating empirical claims.  (Often, spiritual debates turn
     into factual debates.)  To participate in the solution, social
     scientists should be neutral.
3) Identify and establish causal claims and relationships.  (Analyze
     the results of actual interventions.)
4) Cumulative results should yield different ways of looking at or
     defining phenomena.

"Should the National Laboratories Exist in 2005?"
     Katherine Gillman (Special Assistant for Defense Conversion,
     Office of Science and Technology Policy)

There are several hundred National Labs at the moment, ranging in
size from very large (e.g., NASA) to very small (e.g., a one
scientist lab with a small support staff)

The three largest national labs are DOD, DOE and NASA.  (NIH is
behind these in terms of "in-house" work).

Of the $23-24 billion in R&D at the national labs, the "big 3" have
$15 billion.  All are rich in facilities, and all were shaped by the
Cold War era.  All will probably still exist in 2005.

However - for what purposes, and at what size, should they exist in
the future?

The National Science & Technology Committee (NSTC) oversees the
federal R&D budget, and recently initiated an inter-agency review (a
"meta-review) of national labs.  NSTC was looking at the role of
evolving needs of a post-Cold War era.  Currently, the reviews of
the "big 3" are in to the committee.

The Department of Energy Review Task Force (chaired by Bob Galvin of
Motorola) looked at nine multi-program laboratories.  Their
conclusions were:

     1) National Labs are essential.
     2) Energy is the essence of DOE, and the energy mission is a
          public need.
     3) The traditional missions (power, security, etc.) are the
          right ones.
     4) The National Lab system of governance is "broken" and needs
          to be fixed.  Red tape is to the detriment of scientific
     5) The National Lab system is over-sized for its mission,
          including excess capacity to design nuclear weapons.
     6) There is a need for a better mission for the individual
          labs.  The "core competencies" seem to be the same for
          many labs.  (e.g., Should the Lawrence Radiation
          Laboratory and Los Alamos both continue to exist?)

     Energy demands will continue to increase dramatically in the
     near future, especially in developing countries.  The
     environmental implications need to be considered.  For example:
          1) Clean sources of energy are need to replace fossil
          2) Computer Simulation Modeling needs to be improved.
          3) Remediation problems need better science and
               technology, particularly for radioactive sites.
               (Note: the DOE labs don't agree with this point in
               the Task Force Report.)

     Labs have a tremendous potential and expertise for science.

     How about industrial technologies?  There is a strong push by
     DOE to cooperate with industry, but the Task Force said that
     this should not be pursued unless the project(s) is directly
     related to the lab's mission.  (This is another point of
     contention between the DOE labs and the Task Force.)

     Governance?  Thousand of directives and orders take up an
     incredible amount of time (and paperwork).  There is no dispute
     over the problem, but the solutions are debatable.

The Department of Defense Review was conducted "in house."  DOD has
about 30 laboratories with lots of pass-through funds, but in-house
efforts are still substantial.  (About the same amount of money as
DOE, but split over more labs.)


     1) The over-riding mission of DOD is National Security.
     2) Others (e.g., industry) can do some of the research, but
          there are clear reasons why DOD should conduct in-house
          work.  (Among others, they can coordinate R&D efforts,
          help services to be "smart" buyers, and provide special
     3) A major difficult is that the inflexibility of the civil
          service system does not allow for effective down-sizing.
          (Even with the proposed 4% reduction to the end of the
          century, this is merely covered by attrition and does not
          allow for building.)
     4) DOD labs are not major players in environmental science, but
          they are key players in some areas.  (e.g., the Naval
          Research Lab)
     5) Many labs may have "dual-use" opportunities, particularly in
          construction (e.g, the Army Corps of Engineers), and in
          aeronautics (especially aircraft engines).

The National Aeronautic and Space Administration Review indicated
that NASA had a huge research budget, second only to DOD.  Of this,
about $4 billion is used for R&D at NASA's research centers.  The
Aeronautics Centers (i.e., Ames, Langley, Lewis) are mostly R&D,
while the Space Centers (e.g., Houston) have R&D activities mixed
with operations.

     1) The declining budget defines NASA's future.  (Projections
          are a decline from $14 billion to $11.5 billion in the
          next five years.)  Therefore, downsizing is important and
          NASA must concentrate on focusing its mission.  There is
          some redundancy in Centers, and some have expanded into
          areas never envisioned originally.
     2) NASA should draw in more technology from outside sources.
     3) NASA should try to downsize, within civil service
     4) NASA should work with DOD to coordinate and close redundant
     5) There are excessive layers of organization in the management
          structure.  NASA should yield some program
          responsibilities to the Centers (and eliminate some
          central management).

"How Should Government Decide Who Gets the Money?"
     Lewis Branscomb (Director, Science, Technology and Public
     Policy, John F. Kennedy School of Government, Harvard)

What are the boundaries of research?  All of the basic research,
plus much of the applied efforts.  Allocation of funds should be
looked at in the context of the research, not that of the investor.

The National Information Infrastructure and the Intelligent
Transportation System programs are two examples of where government
can coordinate highly creative work.

The primary value of science is its propensity to induce change.
(The ability to pose questions is very important, as opposed to
"problem solving" research such as engineering research.)  The
acquisition of scientific insights to inform technological choice is
a key bridge (both in government and industry).

There is a difference between "opportunity-driven" (i.e.,
exploratory scientific research) and "need-driven" research.  For
opportunity-driven research, peer review is critical and the
diffusion mechanism is "horizontal."  For need-driven research,
dissemination is vertical in nature.

How does one allocate funds for opportunity-driven research in
different fields?  There are different possible approaches:

     1) One can use the "pressure" of non-solicited submitted
          proposals.  (e.g., one can work up a ratio of physics
          proposals as opposed to plant biology proposals.) This
          approach probably will not work, however.
     2) One can use "national goals."  However, these are more
          practical and better used for "need-driven" research
     3) One can ask the scientists themselves.  Unfortunately, the
          National Science Board doesn't have the legitimacy to make
          these allocations.  (It would need an active charge from
          the White House.)

Closing thoughts:
     1) How do we categorize research, and the contexts in which
          research is performed?  What is the balance needed between
          "exploratory" research and "need-based" research?  This
          would determine the amount of discretion to be granted to
          a P.I.
     2) How should we categorize the intent of the sponsor of the
          research?  This would be more complex with "exploratory"
     3) How should the investment property (intellectual property
          rights) of "exploratory" research be invested?

For politicians, we need a policy that is simple and intuitive.