PHYSICS AND THE CLASSIFIED COMMUNITY
Ruth H. Howes
Department of Physics and Astronomy
Ball State University
The goals of a working physicist
include making major scientific discoveries and disseminating results as
widely as possible throughout the community in order to obtain the maximum
recognition for one's work from one's peers. If the results of one's
work are the basis for many other experiments and ideas, one is regarded
as successful.
In doing physics, there are
several accepted norms of behavior. Among them are:
ïShare results with a peer group.
ïEncourage open discussion of methods, materials and
ideas.
ïFamiliarity with existing literature in a field and
care in citing sources in all published work are the hallmarks of a good
physicist.
ïNew ideas are validated by comparison with othersí work.
ïCrediting colleagues with ideas and designs is essential.
ïCarefully delineating uncertainties in work is part
of physics.
Any physicist who does not conform
to these norms (among others, of course, such as making reliable, precise
and repeatable measurements--this list is selected for purposes of this
paper) finds himself isolated from the community. His work is subject
to intense doubt even when it may be physically correct, and his colleagues
are reluctant to talk to him about their work and their ideas.
Members of the uniformed military
have different goals. Their job is to take an order and successfully conduct
a mission. Discussion of orders is out of line, and secrecy is imperative
since its violation may have a direct cost in human lives.
The military establishment traditionally
plays by a very different set of rules from physicists. In part:
ïLoose lips sink ships--talk as little as possible about
your work.
ïThe fewer people who know a secret, the better.
ïDetailed discussion of methods and materials can compromise
sources.
ïSplitting a project into compartments protects the enterprise
from enemy espionage.
ïThe commander always takes both credit and blame for
work of the entire group.
ïDon't bother me with excuses, what is the answer?
Obviously there is an inherent tension
between the military objectives and those of the physics community.
The picture is incomplete without adding in the defense contractors and
national labs who actually conduct much of the research done by the classified
community. In this case, the physicists involved actually do the
traditional type of physics; however their success is measured in patents
obtained, contracts gained and competitive advantage for the organization.
Their rules of behavior might read something like this:
ïTalk only about those aspects of your work which can't
be patented.
ïDon't give information away for free if you can sell
it.
ïBe sure to get credit for what you have produced.
The most cursory consideration
of these varying goals leads one to the realization that physics and classified
work are not happy bedfellows. The problem is certainly as old as
the Manhattan Project. One of the early points of contention between
physicist Robert Oppenheimer and General Leslie Groves concerned technical
seminars. Oppenheimer insisted that all scientific staff members
needed to understand the scope of the project on which they were working.
Groves wished to inform individuals of only those details which they were
required to use in their daily work.
Today, Department of Energy
clearances are given generically and information is released on a need
to know basis. Military clearances are considerably more complex, ranked
secret and top secret. Much information is compartmented data which
means that you have to have a special clearance to know. This is
frustrating when a job overlaps compartments and you are not cleared to
enter one of them.
Working within the classified
community poses a problem for those of us who are accustomed to the freedom
of academia. First of all, academia is intrinsically horizontal.
A full professor, can argue with anyone in the university system.
Faculty opinions are understood to be their own, and professors express
them freely to whomever they desire. The president of a university
may be unhappy with a professor, but he can't fire one for presenting his
ideas.
In the federal bureaucracy,
it is imperative that the national government speak with a single voice.
Any employee of the government may be interpreted as speaking for the government.
Therefore all employees must speak the party line in public whether they
agree with it or not. Arguments are behind closed doors, and any
eruption into public is apt to get someone fired.
There is no such thing as a
good library of classified information. What exists on the data bases
is usually out of date. Information exists in peoples minds and their
safes. To find out what is known about a problem, one must access
a network of friends who may or may not take your phone calls and provide
answers. From time to time, one agency will effectively forbid its
members to talk to members of another agency. The rule of thumb is
that if you cite your sources, you lose them.
The work in progress is swamped
in a formidable bureaucracy. The top management--Congress, the White House
etc. decides that the U.S. needs a position on an issue. An Interagency
Group is formed usually at the assistant secretary level. The IG
in turn forms an Interagency Working Group at the next level down.
The Working Group tasks aspects of the problem out to each of its member
agencies. The agencies involved depend on the problem under consideration:
the State Department, the Department of Defense, the Arms Control and Disarmament
Agency, and the Joint Chiefs are almost always involved with input from
the CIA. NASA joins if the discussion has anything to do with space;
the Department of Energy if it involves nuclear weapons; and the Commerce
Department if it involves technology controls.
The agencies in turn task their
issues out to divisions which in turn assign individual staff members the
task of preparing a draft position. Writing a draft position is a
little like preparing a term paper without benefit of a library.
The draft position is circulated in the division, argued over and changed
to suit the director of the division. The revised draft then circulates
throughout the top bureaucracy of the agency. As modified, it becomes
the agency position in the working group. The working group wrangles over
agency positions and produces a draft which then travels to the IG.
At each step, the draft gets
shorter and shorter. Options drop out and disagreements are resolved.
It is axiomatic that the more power one exercises, the less time one has
to study individual issues. The IG completes a draft which then goes
to the National Security Council for action. The NSC, in turn, sends
the document or a summary of it to the President for signature. By
the time it reaches this level, the position is simple and clear.
There is no room for discussion of confidence levels or error bars.
Technical issues clearly involve
ambiguity. Equally clearly, this process has little tolerance for
this sort of ambiguity.
The ethical guts of the issue
arise when you, a physicist in the system, feel that scientific results
are being distorted by your superior. If you flout the system, you
run the risk of being simply ignored while taking the consequences of your
disobedience at a professional level.
There is no such thing as peer
review within the system. Contracts are let by agencies such as ARPA, the
Advanced Research Projects Agency. ARPA is staffed with bright, young
military officers, many of whom hold advanced degrees in technical areas.
These young people award large sums of money and are responsible for overseeing
the work of many of the best working physicists going. They do a
conscientious job at their work. However, they haven't got a chance
against a senior and successful working physicist, since by definition
these senior people are excellent sellers of their own ideas and very confident
that they are right. Without involving an immoral act, the system
allows scientific blunders of major magnitude to enter the system and even
cost the government millions of dollars. Technical work produced
by and for the government thus varies in quality from the sublime to the
laughable.
The above analysis assumed that
there were no villains in the system. It is easy to see how falsified
results, lucrative job offers to contract monitors, or even over-optimistic
assessments of progress can badly distort the system.
There is probably no better
example of the ethical problems involved in this type of classified work
than the cautionary tale of the X-ray laser. Most of the detail here
comes from the excellent book Teller's War by William Broad (New York:
Simon and Schuster, 1992) which is both very readable and very thorough.
The attached table, which is also based on Broad's data, summarizes the
story.
Essentially, in the early 1980's
Edward Teller, a physicist with a long track record of military research
and development including invention of the hydrogen bomb, used his direct
access to President Reagan to promote the x-ray laser pumped by a nuclear
explosion as the answer to the nation's need for a defense against intercontinental
ballistic missiles. Teller's optimism was based on early tests of
the system which seemed promising. Further analysis of these same
tests showed that the results might not be as encouraging as they seemed
at first.
More conservative physicists
within the system who studied the tests felt that Teller's assessment of
the potential for the x-ray laser were over-optimistic. In particular,
Roy Woodruff, a physicist who was associate director of Livermore National
Laboratory for Nuclear Design, feared that the President would be misled
by Teller's statements if they were not qualified by the judgment of other
physicists in the system. Woodruff at first tried to work within
the system to bring about a reevaluation of the x-ray laser without embarrassment
to Teller, who was after all the founder of Livermore. The system
listened and did nothing.
In March, 1983, President Reagan
committed the United States to a massive research program aimed at the
construction of working defenses against ICBMs. The Strategic Defense
Initiative or SDI was funded heavily, and the x-ray laser formed a major
element in early discussions of defenses. Teller and his deputy Lowell
Wood continued to present optimistic briefings on experimental progress
with the x-ray laser to the highest levels of government.
Meanwhile a variety of experimental
results seemed to Woodruff and others to question the validity of even
the initial experimental results. In particular, the brightness of
the laser was measured by reflecting x-radiation off mirrors to protect
sensors from the effects of the nuclear explosion used to pump the x-ray
laser. New results indicated that the mirrors themselves were radiating
because of their interaction with the effects of a nuclear explosion so
that the brightness measurements for the laser were much higher than they
should have been. In addition, Teller's briefings used preliminary
results from tests and slanted the results in favor of the x-ray laser.
Woodruff became increasingly frustrated with his inability to correct what
he felt was the mistaken impression of the x-ray laser's success that Teller
presented. Teller used his direct access to President Reagan to block
Woodruff's corrections at every turn and was supported by the director
of Livermore, Roger Batzel.
In October, 1985, Woodruff felt
he had no option except to resign his position as associate director in
protest. His resignation was accepted. At this same time, the
press finally broke the story that there were questions about the success
of the x-ray laser. Woodruff found himself facing a major reduction
in salary. He was given no meaningful work and an office the size
of a broom closet, known in-house as Gorky West. Finally in April,
1987, Woodruff filed two protests against his treatment with the University
of California which operates Livermore National Laboratory. A university
panel upheld one grievance, but the General Accounting Office supported
the laboratory and accused Woodruff of taking secret letters home and claiming
membership in Phi Beta Kappa falsely. In May, 1990, Woodruff left
Livermore to work at Los Alamos National Laboratory.
The complex history of the x-ray
laser illustrates three major issues in the ethics of doing physics within
the classified community. First, the data under debate cannot be
presented to the physics community at large for a judgment of their validity.
As long as the data can't be checked, the argument over what they mean
will necessarily be clouded by uncertainty. After all, any physicist
who advances an argument without seeing the data is just presenting a good
guess at what the data show. Because the participants in the debate
can't present data, they say to the public, "Trust me. I'm the smart,
good guy here!" Political decision makers are generally not scientists,
and their reactions to such a debate depends on whom they trust.
President Reagan has never been accused of possessing an in-depth understanding
of nuclear physics. There is no provision for peer review within
the classified system.
Second, a physicist within the
classified community is torn by divided loyalties. Scientists working
on weapons development are loyal to their colleagues and the teams for
which they work. To publicly accuse another member, particularly a leading
member of your own team of making mistakes, damages your team in the vital
game of securing funds for its operation. Thus such an action violates
the principle or loyalty and involves an intrinsic ethical conflict.
Of course the professional consequences for whistle blowers are frequently
severe. Thus the watch dog must face his own divided loyalties as
well as the probable damage to his professional status arising from his
actions.
Last, the history of the x-ray
laser illustrates the difficulty in telling a mistake in science from a
deliberate fraud. Teller had little to gain except prestige from
his position on the x-ray laser. Could he really have misinterpreted
the results of the tests? As in many other case studies in scientific
ethics, this question remains both critical and unanswered.
Issues of scientific ethics
become even more complex within the classified community because of the
intrinsic conflict between the openness demanded by science and the secrecy
needed for military operations. The ordinary processes for scientific
decision making are suspended by the need to keep data from the scientific
community and the discouragement of open debate and argument. Because
these technical issues impact both political power and enormous sums of
money, the temptations to falsify or distort results can be very great.
A system is needed to police the classified technical community, but as
yet, no one has been able to devise one that meets the needs of all the
constituencies involved.
A History of the X-ray Laser
(Based on Teller's War by William Broad; New York:
Simon and Schuster, 1992)
March 1968 Spartan Interceptor tested at Kwajalein
1969
First antimissile debate (Safeguard)
1975
Grand Forks completed ($7 billion)
1976
Grand Forks mothballed
1978
Diablo Hawk--failed test of x-ray laser
1980
Woodruff associate director of Livermore for nuclear design
November 1980 Dauphin-test including Hagelstein's
design
--Wood declares immediate success
--Excalibur-data ambiguous
1981
R Program established to develop x-ray laser
September 1981 High Frontier Group formed
January 1982 Briefing to Reagan
supporting x-ray laser + antimissile defenses
June 1982
R Program officials present conservative briefing to Frieman panel
-6 years & $1 billion to feasibility + decade of engineering development
September 1982 Teller gives private briefing to Reagan:
x-ray laser deployed by 1989 or earlier
December 1982 Teller and Woodruff clash on timetable
February 1983 Joint Chiefs endorse antimissile
defense but not major strategic shift
March 1983 Reagan's
Star Wars Speech
March 1983
Cabra x-ray laser test-failure because data garbled
October 1983 Fletcher Report delivered-$1
billion for x-ray laser but showed itÝhad major problems with deployment;
funding to determine feasibility
December 1983 Romano test-length of rods vs. gain showed
x-ray lasing
One week later, Teller writes very optimistic assessment to Keyworth claiming
quantitative agreement
with theory--no copies to others
Woodruff challenged-qualitative not quantitative-drafted follow-up letter--Batzel
scrapped letter
January 1984 Woodruff sends "results letter" to Keyworth
--Laser's shape and color in quantitative agreement--brightness only qualitative
February 1984 Woodruff meets with Keyworth-Maenchen questions
interaction of sensors with
bright laser beams-glowing reflectors
August 1984 Correo Test by Los Alamos-false
brightness from interaction of sensors with bomb
--Maenchen presents secret theory describing false brightness from interaction
October 1984 Wood presents Super Excalibur to Abrahamson
December 1984 Teller writes Nitze and Macfarlane stating
that they have made a breakthrough
--Woodruff out of loop
Woodruff drafts clarification: more technical breakthroughs + antisatellite
problem + time
Batzel blocks because nothing in Teller's letter violated physics
January 1985 Kerr (Los Alamos) would fire Teller Batzel
and Woodruff defend
Woodruff ordered to increase x-ray laser budget at expense of other weapons
programs
February 1985 Woodruff goes public protesting funding
and wins
March 1985
Bethe and Drell not impressed
March 23, 1985 Cottage test-one sensor modified to look
at brightness problem-Teller hailed as success
April 1985
Wood gives series of briefing extolling x-ray laser
June 1985
Teller visits White House Los Alamos report on glowing beryllium reflectors
September 1985 Teller overrules Woodruff at Pentagon meeting-scientists
leave R program
October 1985 In letter to Batzel,
Woodruff charges Teller with distorting facts to sell x-ray laser
--does not send letter--situation beyond repair
October 31, 1985 Woodruff resigns-publicity on brightness
breaks
December 1985 Goldstone test in spite of bent canister
showed brightness less than expected by factor 10
February 1986 Batzel suggests
reduction in Woodruff's salary--Gorky West days
September 1986 Labquark-focusing seemed to
work
April 1987
Delamar-focusing really edge of annulus
Woodruff files two protests with University
September 1987 University panel upholds second grievance
October 1987 Woodruff story leaks to the press
December 1987 Woodruff promoted to treaty verification
February 1988 Nuckolls appointed Livermore director
supports Teller
July 1988
GAO Report supports Teller's position. Substantial agreement between
Teller and Woodruff
1989
Attacks on Woodruff for false claims of Phi Beta Kappa and secret letters
at home
May 1990 Woodruff
to Los Alamos
DISCUSSION
Some indicated that the Strategic
Defense Initiative was clearly useful only in adding to the offensive capabilities
of the United States while adding little of significance to the defensive
capabilities. Others suggested that these issues were not so obvious,
pointing to the amount of work that was required for the American Physical
Society's Directed Energy Weapons study.
Was the idea of an impenetrable
shield ever a defensible idea? Was the shield concept originally
a physicist's idea or did it come from a politician? These two questions
address the role of physicists as technical advisors. First, there
is the desire among physicists to provide sound technological advice, where
required. It may be that this advice was not solicited or was overlooked
in formulating the impenetrable shield concept. Second, when physicists
do offer advice to elected officials, one would assume that the advice
is based on technical analysis. However, is it possible to be entirely
scientifically objective in delivering such advice, or will political opinions
necessarily color one's analysis? How often are technical opinions
tempered by the desire not to make waves so as to maintain oneís position
of influence? To what extent is an advisor responsible for trying
to purge political bias from technical advice?
How many physicists compromised
their personal beliefs regarding SDI to accept funding from that program?
Does one need to agree with the long term objectives of a program such
as this one in order to pursue funding from it? One could argue that
it is up to our elected government officials to make decisions regarding
objectives, and that scientists are responsible for providing technical
services as requested by the government. On the other hand, even
decisions made by an elected government do not always represent the will
of a majority of the population. Furthermore, it is not clear that
a scientist, who may be familiar with the technical details of a particular
program, should use popular opinion as an excuse to avoid an examination
of the implications of the program.
If a physicist believes money
is being improperly allocated to a program such as SDI, is it ethical to
solicit such funding with the intent of redirecting it to other more palatable
uses? Although such a misdirection is likely to be the result of
the scientist misleading the agency providing the funding, the scientist
may view this misdirection as a service to the country by promoting better
uses of its limited resources. Is this approach undemocratic?
Given that the principle of
openness seems to be fundamental to the academic setting, is there an inherent
problem in the University of California managing a lab such as Livermore
where so much of the research is classified?
It would appear that in the
present state of affairs it is quite easy for technical advice from a physicist
to the government to be altered, ignored, or misrepresented. If the
physicist is employed by the government, there seem to be just two choices:
shut up or resign (and then speak out to the extent that classification
requirements are not compromised). Are there any other choices?
RETURN TO PROCEEDINGS TABLE OF
CONTENTS
RETURN TO MARSHALL THOMSEN'S HOME
PAGE