CRYPTOGRAPHY

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP88-01070R000100150010-8
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 12
• Document Creation Date: December 20, 2016
• Document Release Date: May 17, 2007
• Sequence Number: 10
• Publication Date: April 21, 1982
• Content Type: OPEN SOURCE

Approved For Release 2007/05/18: CIA-RDP88-01070R000100150010-8
RADIO TV REPORTS, INC.
PROGRAM The MacNeil-Lehrer Report STATION WETA-TV
PBS Network
DATE April 21, 1982 -7 ?'lek~ CITY Washington, D.C.
SUBJECT Cryptography
ROBERT MACNEIL: In recent months, the Reagan Admin-
istration has been trying to clamp down on the flow of sensitive
information to the Soviet Union. The Administration sees what
it calls a hemorrhage of scientific and technological secrets
to the Soviets and other unfriendly countries. The efforts to
stop it have caused anxiety in the academic community, particu-
larly the veiled threat of government censorship of scientific
information.
Twice in recent months, Admiral Pobby Inman, Deputy
Director of the CIA, has told scientists that if they didntt
practice voluntary restraint in publishing scientific informa-
tion, the government might impose it. Many scientists say such
restraint would violate their academic freedom and stifle re-
search. They talk of the risks of importing Soviet-style secrecy
into American life.
Tonight, how can this country protect national security
and yet maintain an open society?
JIM LEHRER: Robin, surprisingly, the area where the
conflict has been most severe and most lengthy has been crypto-
graphy, the science of codes, making them and breaking them.
For centuries, it had been the exclusive, secretly exclusive
domain of soldiers and spies. There was no need for civilians
to be tinkering around with cryptography. So few did. But along
came the computer. Business and industry store much information
on computers, information they want to keep private, most particu-
larly from competitors. So suddenly there was a civilian need
for cryptography, from computer codes that couldn't be broken.
Cryptology researchers in the private domain went to work. Soon
the new advances worried the government, particularly the National
OFFICES IN: WASHINGTON D.C. ? NEW YORK ? LOS ANGELES ? CHICAGO ? DETROIT ? AND OTHER PRINCIPAL CITIES
Material supplied by Radio N Reports, Inc. may be used for file and reference purposes only. It may not be reproduced, sold or publicly demonstrated or exhibited.
Approved For Release 2007/05/18: CIA-RDP88-010708000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
2
Security Agency, the super-secret intelligence agency which does
the coding and decoding for the United States Government.
In the late '70s, Robby Inman was head of NSA, and he
declared the unrestrained public discussion of cryptographic
research threatened our national security, by making it easier
for the Soviets and others to develop better codes and to break
ours. The private cryptography community and the government have
been sparring over the issue ever since.
MACNEIL: The cryptography issue actually surfaced four
years ago, when a computer scientist applied for a patent for a
coding device he'd invented. The National Security Agency slapped
a secrecy order on his invention, but he fought it and he won.
That led to the formation of a panel to consider the
national security aspects in cryptography work. The panel recom-
mended that scientists submit their research to the NSA before
publication.
The scientist whose work started the fuss is George
DeVita of the University of Wisconsin at Milwaukee.
Why did you fight the agency's order to keep your in-
vention secret?
GEORGE DEVITA: Well, Robin, basically I fought it be-
cause I thought it was unecessary to clamp any kind of secrecy on
work that's fundamentally a mathematical subject. Its principles
are very difficult to really control and to hide. Basically, the
encryption algorithms that we tend to devise in the academic
community are based on relatively simple mathematical principles
and electrical engineering principles. So I found it difficult
to see how such principles could in fact be kept under wraps.
But more importantly, I felt that encryption was indis-
pensable to the protection of data, particularly as the society
gets more computerized. It has become practically impossible to
keep plugging up loopholes in computer systems that keep cropping
up every now and then. And I feel that encryption really is the
only effective method that we can use to protect that data.
P.IACNEIL: Things like your bank record or health records
or things like that, which are kept on computers, could be --
could be stolen or looked at by people we didn't want to unless
they were encoded. Is that it?
DEVITA: Yes. The computerization of society, Robin,
has proceeded at an extremely rapid rate. We have thousands of
data bases concerning medical records, credit records, our pur-
chasing habits, electronic mail that is coming in rather shortly.
All of these data bases and communication networks that are being
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
established are basically vulnerable to all kinds of attacks. And
the only really effective method that we can use to protect data
while it is being stored on disks or while it's traveling on the
electronic wires is to encrypt these messages or this data.
MACNEIL: How do you know that your particular device
was not -- would not have damaged national security if, say, the
Soviets had obtained it?
DEVITA: Well, basically, I find the argument that re-
search will lead to any kind of damage or harm to the mission of
NSA, and hence national security in general, I find it very diffi-
cult to really believe. Because encryption is a very difficult
subject matter to control. We're not the only ones who do research
on encryption, first of all.
Second, the kind of principles that we engage in are
not likely to lead to any kind of accidental discovery of codes
that NSA uses, because we don't really know what NSA uses, nor
do we want to, for example, know what NSA uses.
And as for the codes leading to any kinds of denial of
intelligence gathering to NSA, I'd say the probabiligy of that is
rather low. Because, in principle, most of the systems are not
unbreakable. In fact, they're breakable. So I find the arguments
that we have suddenly devised techniques that would deny NSA the
intelligence-gathering capability, I find them very difficult to
believe. In fact, historically, every important code has been
broken.
MACNEIL: Now, you were a member of this panel set up
after the incident with your device, which then recommended --
and it's been the practice ever since -- that people in your field
would submit scientific papers and findings to the NSA before they
publish them, and they do it voluntarily. But you, alone on that
panel, disagreed and opposed that. Why did you disagree?
DEVITA: For many reasons. Again, one of the reasons
is that I felt that encryption is an extremely important techno-
logy for civilian applications. But other reasons were reasons
of fears that that decision would lead to more formal restraints
on academic freedom that I didn't think should be instituted,
slowly, voluntarily initially, and subsequently, perhaps, with
legislation. I felt the decision was a rather unwise one at
the time.
MACNEIL:
Now,
that's been working for a year. And
scientists, unlike
you,
have been submitting their papers to the
NSA. Has any harm
come
of that?
DEVITA:
Well,
it's not clear to me how it has worked
or not worked, because -- again, I don't know what NSA in fact
does get or does not get. I believe that the issues of sending
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
materials to NSA are not really the real issues here. I believe
that many people would send things to NSA voluntarily, but I don't
believe that really calls for any kind of system to be instituted
for formal prior review at all. I think NSA can easily ask most
authors to send them their papers, and they probably would. I do.
MACNEIL: You do it anyway.
DEVITA: I would do it anyway. I don't find any reason
for any kind of formal mechanism to have NSA request our papers.
They can simply get on our mailing list and they can get the papers
ahead of time.
MACNEIL: Is it possible, is it practicable for the gov-
ernment to restrict the dissemination of this kind of information
in the cryptology field, these devices? You have two of them in
front of you here.
DEVITA: No, not at all. And that's one of the key
arguments that I try to make in this whole debate about protecting
this kind of technology.
Here we have an encryption device which is actually the
government-approved data encryption standard. This particular
board happens to be an encryption and decryption device. This
is, of course, controlled. You cannot export this without a
license from the Commerce Department.
On the other hand, here we have a general purpose com-
puter which can be also made into an encryption device rather
easily. All you would need is a program.
So, effectively, these two board are almost identical.
DEVITA: That's correct. And the argument is that we
cannot control this technology because microprocessors, which is
really this chip at the top of the hoard, are abundant and inex-
pensive. And furthermore, we're not the only ones who make these
microprocessors. The Japanese and the Germans and the British,
everybody else is making these microprocessors. So it would be
very difficult to really control the materials that go into making
crypto systems.
MACNEIL: Well, thank you.
LEHRER: Also on that panel which devised the voluntary
review plan was Daniel Schwartz. Mr. Schwartz was then the gen-
eral counsel for the National Security Agency. Since last Octo-
ber, he's been in the private practice of law here in Washington.
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
5
Mr. Schwartz, has the voluntary system worked, in your
opinion?
DAN IFL SCHWARTZ: It has so far. It has basically been
in effect, in a formal sense, for about six months. But there
has been quite a long period of time in which researchers in this
field have been submitting papers.
LEHRER: How many papers have been submitted, in rough
SCHWARTZ: Since the initiation of the program, some 35
papers have been submitted. But there were papers submitted
before that. The problem was that they were not routinely sub-
mitted by everyone doing writing in the field.
LEHRER: Well, of these 35, were any of them killed by
SCHWARTZ: None of -- all of the papers have gone for-
ward through their publishing process, at least have not been
restrained by NSA in any way. A few of the papers have raised
some problems, and an accord has been reached with the authors
in each case for minor changes in the papers.
LEHRER: But without exception, all 35 did go ahead.
They were published. Is that right?
SCHWARTZ: They haven't actually been published...
LFHRFR: I know what you mean. They went through the
process and they were cleared.
All right. If there hasn't been any need to kill any
papers or to prohibit the publication of any of the papers, doesn't
that prove that the concern over the cryptography and the publi-
cation of these papers was not founded on fact?
SCHWARTZ: No, for two reasons. First of all, a six-
month period is a very short period, particularly in a science
like this. The study group recommended that the results of this
voluntary process be reviewed after two years to see what kind
of results had occurred and how NSA had dealt with the academic
community. I would say six months is simply too short to make
that judgment.
Secondly, there have been some changes that have been
made where there was some feeling, in fact, that it would be dis-
advantageous to the national security to have the papers go as
submitted.
LEHRER: I mean were these serious problems? And I
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
realize you can't tell me what they were. I wouldn't probably
understand it if you did. But I mean were they serious problems,
that if those papers had not been changed, that there would have
been serious harm to the national security? Or were they little
things?
SCHWARTZ: I don't know how you make that kind of a
judgment. You can't really make that kind of a judgment until
you see what the results are. Very often, particularly, in the
field of cryptology, it is impossible to trace the result. YOU
really don't know who's reading what, and you really don't know
what kind of effect it has.
Part of the problem in this whole field is one that
Dr. DeVita mentioned. As he said, he doesn't know what NSA does
He doesn't know what the U.S. Government does. The problem is
that most scientists in this area have no real idea whether what
they're writing on might have an impact on the national security.
Part of the interest in establishing this kind of pro-
gram was to give scientists an opportunity, and on a voluntary
basis, to determine whether, in fact, there might be some impact
in what they were writing. They wouldn't independently know that.
LEHRER: What about Mr. DeVita's point that it's really
impossible to control the flow of this kind of information on
crytography. It's a very -- you know, his example of the two
boards there, that they're identical, you just have to make a
minor change. And one board is available anywhere; the other
one is restricted.
SCHWARTZ: There is no intent, I think, here, realis-
tically, to get a perfect system. What one would be seeking, at
best, is some incremental advantage, some additional help in
trying to Iimit the outpouring of technical data that might be
of assistance to our actual or potential adversaries. It's very
hard to stem it completely, as we know. But the hope would be
that in various -- in various ways, that this could be limited.
LFHRFR: When Admiral Inman made his statement -- I
think it was in 1979 -- saying that the free discussion, or the
open discussion of cryptography was endangering the national
security, did he have a specific incident in mind, that some-
thing had happened to prove his point? Or was it a fear based
on what he was seeing that might happen?
SCHWARTZ: Two things were happening during that per-
iod. One was, there was an increased amount of research, and
therefore publication, in the field of cryptography. This is
an inevitable development. The concern was that that would
raise problems, potentially. And there were incidents where,
in fact, it did raise fairly serious concerns.
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
The second thing that happened, however, was that there
have been -- in part because the National Security Agency was such
a secret and highly classified agency -- a real lack of a dialogue
between that agency and the scientific community. And part of what
Admiral Inman was calling for at that time, and part of what this
study group provided, was a vehicle to have a dialogue to address
the problem.
LEHRFR: And you think that's on its way, that's in pro-
SCHWARTZ: Well, so far, it is on its way. There has
been, as far as one can tell, a fairly good response. And we
will just have to see in the long term.
LEHRER: Thank you.
MACNEIL: Now a view from the Reagan Administration.
Stephen Bryen is Deputy Assistant Secretary for International
Economic Trade and Security Policy at the Defense Department.
Mr. Secretary, Admiral Inman warned that if some volun-
tary restraint did not happen, government -- and he mentioned the
Pentagon specifically -- might be forced to step in. Do you
agree with that?
SECRETARY STEPHEN BRYEN: Well, I think what we might
have to do, if we can't work a system of voluntary restraint, is
consider how we manage our own in-house research and development,
and particularly the contract research that goes out to universi-
ties and scientific organizations around the country.
MACNEIL.: You mean if they didn't voluntarily restrict
publication of information, you would bring that research inside
the Pentagon, so to speak.
SECRETARY BRYEN: It's a possibility, but I think a
slight possibility, because, in fact, I expect we will have very
good cooperation from our scientific community. We are engaged
now in the beginning, only the beginning, of a dialogue on this
subject. I think we can become more precise about the things
we desire. And I believe that the scientific community is going
to be very responsive.
MACNEIL: Are you doing some planning in the Defense
Department right now for some ways of the government restricting
this dissemination of the scientific community does not cooperate?
SECRETARY BRYEN: Only in a very limited sense. We
are beginning to try and specify what particular programs we
should be extremely careful about, in terms of how we deal on
the outside. I think in the past...
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
8
MACNEIL: Do you mean we, Pentagon, or we, American?
SECRETARY BRYEN: We, Pentagon.
There are some sensitive programs that we have that
were fairly loosely managed in the past, and we're trying to
tighten those up a bit. I doubt that anyone on the outside would
even see where the tightening occurs, but we will know.
MACNEIL: Do you care to mention an example?
SECRETARY RRYEN: One of the programs that we are quite
concerned about is the very-high-speed integrated ciruit program
we have underway in the Pentagon. We don't mean to tighten it up
to the point where it becomes an entirely in-house effort, because
I think we would fail at that. But we do mean to he very careful
about the very specific circuits that apply to military hardware.
Now to the p o i n t of what the need is for a l l this res-
triction. I read a figure today, and I've forgotten what the
source of it was, but an estimate that only one percent of the
so-called hemorrhage of sensitive information to, say, the Soviet
Union could he traceable to the scientific community and publi-
cation, and that most of it comes from other sources, spying and
so on.
SECRETARY BRYEN: The answer is we don't know the an-
swer. We have tried, ourselves, to come to grips with how much
is transferred through open literature, how much through trade
conference, how much through legal sales -- and quite a lot was
in the past transferred through legal sales -- how much through
illegal acquisition. And we simply don't know today what the
proportions are.
What we do know is that everywhere we look we find a
plethora of examples of transfers of very sensitive technology
that have gone to the Soviet bloc, have been turned around and
used in Soviet military hardware. That much we do know.
I must say it's a very difficult area to be precise
in, but we are trying to find more of the answer. Because,
obviously, how we use our own resources will depend on the out-
come.
MACNEIL: Can some parts of science be placed off
limits, on national security grounds, when they have become vital
in civilian life? Take the example of cryptography, as we've
just heard from Professor DeVita.
SECRETARY BRYEN: I don't think so. I think some very
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
special areas that relate directly to military applications can
be assigned a special status and category. Put I should mention
for the record that in many cases today civilian technology is
actually well in advance of the kind of technology that's em-
bodied in the weapons systems that defend this nation. That's
one of the great problems that we face today. It's something
that the Soviets are well aware of and are exploiting.
MACNEIL: You mean that, unwittingly, a scientist like
Professor DeVita could invent something that was way ahead of
anything you were actually using or planning to use.
SECRETARY BRYEN: Absolutely.
MACNEIL: And not know it.
SECRETARY BRYEN: Possibly not know it.
MACNEIL: I see. V!eII, thank you.
LFHRFR: A second and different overview now from
William Carey, Executive flirector of the American Association
for the Advancement of Science, the largest general science group
in the country, one vitally involved in this issue of scientific
information an national security.
Mr. Carey, how is the scientific community reacting
to the Inman call for voluntary restraint?
WILLIAM CAREY: I think with consternation and dis-
belief. The shock that was administered to science by Admiral
Inman's remarks was considerable.
It is not that science, in any way, is against the
national security. That is not the problem. Science has sup-
ported the national security in the last 35 years through three
hot wars and one very long cold war, and it will continue to do
so. The issue arises over the imposition of censorship, whether
compulsory or voluntary.
LEHRER: There is a difference, though, isn't there,
between -- or do you not see a difference between voluntary cen-
sorship and compulsory censorship?
CAREY: Of course there is a difference. In the one
case, the voluntary censorship, science cooperates with the
national security authorities. In the other case, science has
no choice.
LFHRFR: Vlell, let's take the voluntary first. What
is the basic objection to vo l untar i l y going along with Pryen and
Schwartz? Schwartz is a former employee, I should say.
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
CAREY: The objection is the generality in which the
government side has presented or stated what the problem is.
When Admiral Inman, who is a superb public servant, by the way,
spelled out a long laundry list of areas of technology and ap-
plied science and general science to make his point, he did not
discriminate among fields of science or among the technological
intensities of technology versus research. And it makes a very
big difference.
He has now proceed to think it over a little more.
And what he is now telling the Congress is that 70 percent or
more of the so-called hemorrhage -- which I believe is now re-
ferred to as leakage -- comes from high technology which is em-
bedded, in many ways, in American manufacturing, goods or pro-
ducts, or in process technology.
Now, I don't think that very many people in the scien-
tific community would argue with that proposition. But the re-
maining 30 percent of the problem comes down applied research,
mainly in the industrial area, where the information seems to
leak out through business deals, through legal and illegal ar-
rangements.
And then you get down to science, the discovery pro-
cess, the searching process, the creative process. And it may
very well be that we're talking about one or two percent of the
problem that the national security people are concerned about,
one or two percent on a scale of 100 percent, defining the entire
problem. And when government shakes a finger at scientists and
says, "If you don't submit to a voluntary system of bringing to
government information, your ideas, what are you going to do
research on, if you don't send us copies of your publication or
your preprints before you allow them to go into the system, then
there' II he a public outcry and Congress will get after you and
make it worse," well, the difficulty is that the way science
works is -- it's like billiards. Your idea caroms off my think-
ing and off the thinking of a third scientist and a fourth. And
in each collision, and in the total collision, the idea matures.
It's changed. It's altered. And this is the way discovery
works, the way knowledge advances. And finally, out of this
collision, there comes an idea, a new idea, a new piece of in-
formation that fits into the jigsaw of knowledge.
And that's what would be impaired if the arteries of
communication were clamped off for very serious reasons, coming
from the government side. You clamp those arteries off, you're
going to shut off the blood, you're going to slow down science,
and you're going to invite second-rate science, which will not
keep our national defense ahead of the Soviets. That's the prob-
lem.
LEHRER: Thank you.
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
PIACNEIL: Mr. Secretary, what's your answer to that?
SECRETARY BRYEN: Well, I think our answer is as fol-
lows: We agree entirely that a completely broad-brush government
approach is not necessary, is not called for, is not desirable.
I agree fully to the notion that we want to keep our scientists
as unfettered as we possibly can.
I think we have a responsibility, therefore, to try
and be precise about the areas where we have concern, as happened
in the area of cryptology, as will, I hope, happen in the area
of very-high-speed integrated circuits, and other areas that are
identified in precise terms.
What we need to then have is a dialogue with our uni-
versity people and with our research organizations so that we can
work out reasonable solutions with the least amount of, even
voluntary, restrictions.
MACNEIL: What's wrong with that, Mr. DeVita?
DEVITA: Well, I think it's very difficult, really,
to assess the damage that's done to a scientific effort when
you withhold even a small number results. I think that the vol-
untary system really envisions withholding some results when NSA
objects to them.
MACNEIL: Is that what it does envision, Mr. Secretary?
SECRETARY BRYEN: Well, I don't know if it does. I
think that there are certain results that we would rather not
publish that could bring harm to this country. We have a pretty
good idea what they are by now. I think our scientific community
that works on these problems has a pretty good idea of what they
are.
MACNEIL: Well, Mr. Schwartz, you've thought about this
a lot. What do you say -- I think Mr. DeVita and Mr. Carey have
made the same point, that, in his image, the billiard game would
be slowed down. You wouldn't have the number of collisions that
produce new ideas if you stop publishing even a few.
SCHWARTZ: First of all, in practice, what happens, at
least in the field of cryptology, is that there is a review. And
so far, there has been very little effect on the final published
work.
Secondly, I think it's important to stand back and look
at what's really asked for here. What was done in the field of
cryptology, and I think what Admiral Inman asked for, was a dia-
logue, recognizing that there are very important concerns on both
sides. In the cryptology, with that study group, in fact, the
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
 Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8
original proposals that Admiral Inman had suggested were com-
pletely rejected. But the people on the study group -- and I
was the only government representative. The people on the study
group came up with an entirely different proposal that is now
being tried.
MACNEIL: Can I just interrupt? We just have a few
seconds. I wanted to get Mr. Carey's view.
Mr. Carey, is the scientific community going to cooper-
ate the way Secretary Pryen hopes it will?
CARFY: The scientific community is going to do its
best to find an accommodation, to find a balance. I think that's
going to be a very difficult, very responsible process. The
National Academy of Sciences is working on that. We are setting
up a panel. It has begun to meet.
MACNEIL: But you think it's going to be very difficult.
CAREY: I t ' s g o i n g to be d i f f i c u l t , but I t h i n k we w i l l
make some ground, we will gain some ground, and come into some
kind of understanding. We have to.
Approved For Release 2007/05/18: CIA-RDP88-0107OR000100150010-8