STUDIES IN INTELLIGENCE VOL. 11 NO. 1

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78T03194A000200050001-8
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 255
• Document Creation Date: December 16, 2016
• Document Release Date: February 22, 2005
• Sequence Number: 1
• Publication Date: January 1, 1967
• Content Type: SINT

ApOr~c2 . d For Release 2005/03/15: CIA-RDP78TO319 c00?200050001-8.
rKA
5J;11O.i'PS IN
I T E L L I :L I I N t :
E, I
Approved For Release 2005/03/15 : CIA-RDP78T03194A000200050001-8
 Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
),~ t to e- P,
Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
 Approved For Rele
r - T03194A000200050001-8
T P7
STUDIES
in
INTELLIGENCE
VOL. 11 NO. 1 WINTER 1967
CENTRAL INTELLIGENCE AGENCY
OFFICE OF TRAINING
Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
SECRET N? 1582
 SECRET Approved For Release 2005/03/1 : CIA-RDP78T03194A00020 -
I
All opinions expressed in the Studies are those of the
authors. They do not necessarily represent the official
views of the Central Intelligence Agency or any other
component of the intelligence community.
STUDIES IN INTELLIGENCE
EDITORIAL POLICY
Articles for the Studies in Intelligence may
be written on any theoretical, doctrinal, oper-
ational, or historical aspect of intelligence.
The final responsibility for accepting or re-
jecting an article rests with the Editorial
Board.
This material contains information affecting the National Defense
of the United States within the meaning of the espionage laws Title
18, USC, Sees. 793 and 794, the transmission or revelation of which
to an unauthorized person is prohibited by law.
The criterion for publication is whether or
not, in the opinion of the Board, the article
makes a contribution to the literature of in-
telligence.
EDITOR
PHILIP K. EDWARDS
EDITORIAL BOARD
SHERMAN KENT, Chairman
GROUP 1
Excluded from automatic
downgrading and
declassification
DONALD F. CHAMBERLAIN
LAWRENCE R. HOUSTON
WILLIAM N. MORELL
JOHN H. RICHARDSON
R. J. SMITH
Additional members of the Board are
drawn from other CIA components.
Approved For Release 2005/03/150 CIA-RDP78TO3194A000200050001-8
25X1
SECRET I SECRET
 Approved For Release 2005/03/15 : Cl -RDP78TO3194A000200050001-8
SECRET SECRET
1 -1
25X1
CONTENTS
CONTRIBUTIONS
Intelligence for the Policy Chiefs ......... James P. Hanrahan
Keeping the President and his top advisers current. SECRET
Page
1
Contributions to the Studies or communications to the editors may
come from any member of the intelligence community or, upon in-
vitation, from persons outside. Manuscripts should be submitted
directly to the Editor, Studies in Intelligence, Room ID 27 Langley
I I and need not be coordinated or submitted through chan-
nels. They should be typed in duplicate, double-spaced, the original
on bond paper. Footnotes should be inserted in the body of the text
following the line in which the reference occurs. Articles may be
classified through Secret.
Community Progress in Information Handling . Zane Thornton 13
Outlook after a push from the President's Board. SECRET
Automation for Information Control ... ...... Paul A. Borel 25
Review of possibilities and some applications in CIA.
CONFIDENTIAL
The Kidnaping of the Lunik ......... Sydney Wesley Finer 33
Clandestine operation to get Soviet space industry markings.
SECRET
For inclusion on the regular Studies distribution list call our office
dissemination center or the responsible OCR desk, call
For
back issues and on other questions call the Office of the Editor,
A New Kind of Air Targeting ........... William A. Tidwell 55
Disrupting the base sanctuaries of insurgency.
CONFIDENTIAL
A Staff Agent's Second Thoughts ......... Louis Boifeuillette 61
More problems than immediately perceived in deep cover.
SECRET
Which Way Did They Go? ...... ... .... Takemi Miyagi 67
Follow-up on the Japanese spies of World War II.
SECRET
All copies of each issue beginning Summer 1964 are numbered serially
and subject to recall.
The Illustrious Career of Arkadiy Harting
Rita T. Kroncnbitter 71
From informer to socialite chief of the Okhrana abroad.
CONFIDENTIAL
Cumulated Grouping of Articles ....
In Volumes VI through X. CONFIDENTIAL
Intelligence in Recent Public Literature
The V-weapons. OFFICIAL USE 93
Two German spy books. OFFICIAL USE ......... .. 99
Soviet intelligence and security: 2nd bibliography .. i
UNCLASSIFIED
SECRET
Approved For Release 2005/03/15 : Cl
25X1
 Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
 SECREApproved For Release 2005/03/15 : CIA -RDP78TO3194A000200050001-8
The needs of the President and
his top advisers, and how these
needs are met.
SECRET
No Foreign Dissern
INTELLIGENCE FOR THE POLICY CHIEFS
James P. I-Ianrahan 1
In this discussion of intelligence needs at the top national level and
some specific ways in which they are filled, I shall be speaking from
the perspective of CIA's Deputy Director for Intelligence. I will not
attempt to speak for the other organizations of the Washington in-
telligence community or pretend to be presenting the whole picture.
First it will be useful to say who the people are that are served by
what we call national, as opposed to departmental, intelligence. We
start with the President, of course. But we must take into account
certain members of his personal staff and in particular his special
assistant handling national security affairs and his staff. Next come the
heads of departments, in particular State and Defense, the military
chiefs, and the heads of independent agencies dealing with foreign
affairs. Then there are numerous interagency bodies established for
the purpose of recommending policy; the Committee of Principals
on disarmament is an example. And at the senior level are also the
regional proconsuls, such as Ambassador Lodge in Vietnam and
Ambassador Bunker in Santo Domingo, who have been delegated
extraordinary authority.
But in the end the buck stops at the President's desk, and the advent
of the nuclear age has greatly multiplied the number of things he
must decide personally. He has almost become, in Richard Neu-
stadt's words, "a decision machine." His decisions in international
affairs are influenced by many people and institutions, but in particu-
lar by those just mentioned.
The requirements for intelligence at this national level are particu-
larly fascinating because they are so kaleidoscopic. They change
with the men, they change with the times, they change with the
bureaucratic structure, they change with each policy decision. As a
result, it is possible to generalize only most broadly on the needs of
'Adapted from a paper prepared for presentation at the Intelligence Methods
Conference, London, September 1966.
Approved For Release 2005/03/15 : CIA DP78TO3194A000200050001-8 MORI/HRP PAGES 1-12
SECRET SECRET 1
 Approved FEo?Or &gecdse 2Q08/03/15 : C
the senior policy maker. He certainly must be provided, if possible,
with what he thinks he needs to know. He sometimes should be
provided with things the intelligence people think he should know.
Often he must be given material which in the beginning neither he
nor the intelligence officer realized would be needed-material gen-
crated by the interaction between the two as they work together.
Lines of Contact
The most direct way of finding out what the senior policy maker
needs is to ask him. Fortunately, all DCIs have had regular direct
access to the President and have not been reluctant to ask what he
wants. Meetings in person or talks between the two by phone are
more frequent than most people, including Washington political in-
siders, realize. Mr. McCone, for example, met every morning with
)President Johnson throughout the first weeks of his administration
to deliver an early morning intelligence brief.
There is of course a limit on access to the President and the time
he has available. But we are in frequent touch with the other senior
policy makers, who not only know their own needs but have a pretty
good idea of the President's. Then communication and rapport with
,-he President's immediate staff are of great importance. These men
close to him are in the best position to make his needs known. At
present they usually do this by telephoning the Director or his Deputy
for Intelligence.
We are constantly receiving requests for information and analysis
from the White House staffers who handle national security affairs,
and it is an advantage that some of our former officers have served
or are serving on this staff. For example, when Mr. Komer received
his special assignment to concentrate on South Vietnamese problems
we asked him how, as a former member of the Office of National Esti-
mates, he felt we could best meet his needs. Ile asked for a periodic
summary of economic and pacification developments in South Viet-
nam, information that tends to get buried in the welter of military
reporting, and we now have such a weekly publication tailored espe-
cially for him.
Moving from the White House to the Pentagon, the Agency has an
intelligence officer serving in the office of Secretary McNamara. He
is attuned to the Secretary's needs and levies many requirements on
us for him. These supplement those that come directly from Mr.
McNamara through his frequent meetings with the Director.
Approved For Release 2005/03/15 : C
SECRET
IA-RDP78T03?64W62&%D 6001-8
Over at State we have a new mechanism called the Senior Inter-
departmental Group, chaired by the Under Secretary of State and
comprising top representation from agencies concerned with foreign
affairs, including the DCI. The SIG is responsible for insuring that
foreign policy problems requiring interdepartmental attention receive
systematic consideration. It stands at the apex of a series of Inter-
departmental Regional Groups chaired by the Assistant Secretary of
State for each region. Intelligence is represented on each of th,se
groups, too. They thrash out new regional policy recommendations
which then move on through the Senior Group to the Secretary. In
essence, the new system attempts to apply in Washington the country-
team approach of a large American embassy abroad. We expect
these groups to become particularly important in the slower-moving
policy problems; the big, Class-A flaps tend to bypass any set in-
stitutional framework, generating their own high-level task forces
responsive directly to the President.
Outside the departments there are the several statutory or ad hoc
committees with special tasks in the field of foreign affairs. Intelligence
is represented on many of these bodies, for example on the Economic
Defense Advisory Committee concerned with Western multilateral
trade to Communist countries and on the Advisory Committee on
Export Policy handling U.S. unilateral controls.'
Last but by no means least, to discover the needs of the policy
maker there is always the "old boy" net: people we have known, gone
to school with, worked with, played with, fought with, and whom
we are now in contact with either on the policy level or in intelligence
components. To take one good example, one of our representatives
eight years ago at the first Intelligence Methods Conference, William
P. Bundy, is now Assistant Secretary of State for Far Eastern Affairs.
From these people, because they know us and we know them., ve?
get a constant stream of suggestions as to the needs of the men above
them, and we usually hear quickly when what we produce fails to
meet those needs-so that we can try again.
Tailoring
How do the needs of the senior policy maker, these "national" re
quirements, differ from departmental requirements? To my mind
they can be distinguished in two ways: first, if they involve more
'See Sherman R. Abrahamson's "Intelligence for Economic Defense" in StudiFa
VIII 2, p. 33 IF.
-RDP78T03194A000200050001-8
SECRET
 Approved FF1`&666F~ 6/03/15 : C
than one department's interests and it is either difficult or plain im-
possible to separate out each department's responsibility; second, if
they are so critical that the judgment of more than one department
is desired. More simply, you might say that when any of the people
or groups we have been talking about asks you something, you know
it is a national requirement because they are all involved in the making
of national policy. It is almost impossible today to identify a na-
tional policy matter that lies wholly within the sphere of one
department.
What level of detail does the policy maker require? No clear-cut
answer can be given. In the Cuban missile crisis one did not have
to be clairvoyant to know the President was himself handling all
the details of the naval quarantine and that he personally wanted
to know the exact location of every Soviet merchant ship that might
he bound for Cuba. We did not wait to be asked, we simply sent
the information on as fast as we obtained it. At certain points in
the Laotian crisis in the spring of 1961 also, it became obvious that,
as Ambassador Winthrop Brown put it, the President was the "Lao-
tian desk officer." And everyone knows how greedy for information
an area desk can be.
There are some other maximums. Anytime the lives of a country's
nationals, civilian or military, are endangered in foreign countries, the
highest level wants to know about it quickly and in as much detail
as possible. Communist kidnapings in Latin America, helicopter
shootdowns in the Berlin area, or for that matter shootdowns any-
where-in all these cases the President wants to get the complete word.
These days when he must spend a great deal of time with the Viet-
namese war, we have found it wise to err on the side of giving too
much in this field rather than too little.
Beyond these cases where it is obvious that you shoot the works,
there are only rules of thumb. We have come, fortunately or no,
a long way since the good old days of the one-page precis so favored
by General Marshall. If we are specifically asked for something by
a senior policy maker and no length is mentioned, we write as much
as we think required to do the job, no more. Then we ask someone
to review it and cut it in half for us. If this cannot be done-or even
if it can-we put a summary up front.
If we have not been asked specifically but feel it desperately im-
portant to get something across to the senior policy maker, brevity is
the overriding virtue. Conclusions and judgments are the nub; argu-
mentation can come later. If his appetite is whetted, if he wants to
Approved For Release 2005/03/15 Cl
4 SECRET
-RDP78T0W94A0b200'850001-8
know more, or if he violently disagrees, we expect to pick up some
feedback somewhere along the line so that we can follow through with
more detail as necessary.
It is here that the regularly scheduled publication, the daily or the
weekly, comes into play. By and large we find that such publica-
tions prepared for senior policy makers should hit the high spots. It is
not necessary for them to carry all the classified news that's fit to
print. They should serve rather as an alert to any developments
which might directly or indirectly affect the nation's security. In the
course of preparing them every bit of information the intelligence
officer can get his hands on is reviewed, but it is then put through
a very fine screening. If the policy maker wants more on a given sub-
ject or if the intelligence officer thinks the policy maker needs more-
a separate memorandum or paper is written.
Communication Hazards
There are always difficulties in maintaining contact with the polic,,
maker. One difficult situation is when he is on the road-how to get
to him in an emergency, how to keep up his continuity of information.
We have partly solved this one through a system of briefing cables
tailored specifically for the high-level traveler. They consist in 1:he
main of a synopsis from our daily publication supplemented by mate-
rial in which the traveler may have a special interest because of th ,
area he is visiting or the people he is meeting.
Sooner or later, a period seems to come when the demands on th,.,
time of the senior policy maker are so enormous as to preclude our
getting through to him in any way at all. In these circumstances we
can only wait for an opening and hope he may be able to take a
quick look at our regularly scheduled intelligence publications. In
these we note the things that he really should not miss even if he iti
spending 100% of his time on Vietnam or the Dominican Republic.
When Mr. Kennedy became President, he brought with him a deep
interest in foreign affairs, a voracious appetite for reading, a retentive
memory, and above all a different style of doing things. Our publicc-
tions in January 1961 simply did not fit his needs. Our primary daily
publication was the Central Intelligence Bulletin. It had been e,:-
pressly asked for by President Truman. Then it was specially
adapted to meet President Eisenhower's needs, and although we had
tried to alter it further it did not suit President Kennedy's style and
he did not read it.
RDP78T03194A000200050001-8
SECRET 5
 Approved For ~eI.eas~h20t05/03/15 :
or o icy
We were thus without a daily link or any periodic link with which
to carry out our critical alerting function. We bent every effort to
restore contact. Finally we succeeded, adopting a new publication
different in style, classification, format, and length but not different
in fundamental concept-a medium whereby we present to the Presi-
dent in the tersest possible form what he should know about the
play of the world for that day, particularly as it impinges on U.S. na-
tional security interests. This publication became the President's
alone, leaving the Bulletin to serve readers at the next level down.
There remains one other basic problem of communication with the
policy maker. That is that the desk-level intelligence analyst, the
fellow at the heart of the process, is never going to have all the clues
to what is making the high-level world go 'round. He does not sit
in on the National Security Council sessions. The Director, who does,
cannot for various reasons-the need-to-know principle, the sheer
physical impossibility of spreading the correct word and feel down
far enough-fully communicate it to the analyst. I submit, however,
that the analyst is not thereby relieved of his responsibility to keep
track of developments in national policy. The daily press and the
favored columnists are excellent sources. If the President or the Sec-
retary of State delivers a speech on foreign policy, it will be revealing
and should be read. I suspect that the percentage of intelligence
analysts who read such speeches is still far from 100%. You hear
the argument that the less one knows about policy the more objective
one's analysis is. But the counterargument that you cannot produce
intelligence in a vacuum, cannot recognize threats to U.S. policy in-
terests unless you know what those interests are, seems to me over-
riding.
So on the question of requirements for intelligence at the national
level, we might summarize as follows: In large and complex govern-
ments, there are no simple ways to determine the full range of the
policy maker's needs. They change as situations emerge, develop, and
subside. Communication-free and easy contact in an atmosphere of
confidence-is essential to the smooth working of the intelligence-
policy relationship. Mechanisms can be established to speed the
flow of intelligence up and requirements down, and these mechanisms
are essential. But nothing is so valuable as an effective person-to-
person relationship. In our country all policy authority and decision
rest ultimately in one man. It is he that intelligence must serve.
Approved For Rele$5/03/15 :
IA-RDP78TP031 4A000050001-8
Now we turn to how we go about filling the policy maker's needs,
however expressed or divined. This is a discussion of technique, and
form, and formula. Again let me stress that I am not saying, "This
is the way to do it," but "This is the way we in CIA are doing it."
We do it both by working in concert with other members of the in-
telligence community and by preparing unilateral reports.
The scope of the information we process is determined by the
nature of the information that comes in and by the range of national
security interests it impinges on. The form in which it is processed
is determined by the requirements of the consumers, in particular
the quite personal requirements and preferences of the President.
From the beginning almost twenty years ago, the DCI has considered
his role to be that of the President's number-one intelligence officer,
responsible for seeing to it that the President is kept unexceptionably
informed and directing the work of the entire intelligence community
to that end.
In the Kennedy and Johnson administrations, the White House has
generally preferred to deal with big problems by calling together
the top policy makers, putting all the available information on the
table, and then discussing possible courses of U.S. policy and action.
This method of operating places a premium on rapid intelligence
support. "Rapid" does not necessarily imply crash assessments,
thoughts formulated on the run. It is more often a matter of re-
shaping or resynthesizing for the occasion the assessments we have
already published in our regular production routine. I want to
underscore the importance of a deep and stable base of day-to-day
intelligence production. This is what enables us to respond quickly
to big and little flaps, whatever the subject or area.
Regular Production
The routine production base includes three "national" intelligence
publications representing the coordinated views of the intelligence
community and dealing respectively with the past, present, and future.
The past, so to speak, is represented by the National Intelligence
Survey, an agreed-upon basic compendium of factual detail and his-
torical development. The future is represented by the National In-
telligence Estimate, containing the best thinking the community can
put forward on a given problem for future U.S. policy. The present
is represented by the Central Intelligence Bulletin, the daily which
brings current developments to the attention of high-level readers in
brief form.
1A-RDP7%TO kj 4A000200050001-8
 r P li Chiefs For Policy Chiefs SECRET
SECRET ApprovedF~or Ie1~ase 2005/03/15: CIA-RDP78T03194A000200050001-8
The procedure for coordinating the evaluations made in the Bulletin
among the agencies of the intelligence community may be of interest.
Each day the items are drafted in the CIA Office of Current Intelli-
gence, often with help from analysts in CIA's economic, scientific,
and technical research components, and circulated to the community
by secure communications channels. They are reviewed by the com-
petent desks and branches within CIA, in the Defense Intelligence
Agency, and in the Bureau of Intelligence and Research at the State
Department, whose representatives then meet in the afternoon, bring-
ing such changes, additions, or deletions as the desks may have sug-
gested. An agreed version is hammered out, footnotes being used,
as in national estimates, to register any sharp dissent. By six o'clock
in the evening the draft Bulletin constitutes agreed national current
intelligence. Before the publication reaches its readers at the open-
ing of business the next morning, however, it has to he updated. We
in CIA make the changes unilaterally, so marking them. The Bulletin's
reporting on Vietnam, for example, will incorporate information re-
ceived up to 4:30 in the morning; this is not an hour conducive to
formal coordination.
Besides coordinating these community publications we produce
others under the CIA imprint, some of which may also be coordinated
with other agencies. A weekly world roundup reviews current re-
porting in a little deeper perspective, and one or two special annexes
accompanying it usually treat some current problem in a fairly com-
prehensive way. Then there are regular publications for particular
purposes, such as a daily Vietnam situation report, the weekly Vietnam
report I mentioned, a weekly tailored to the needs and agenda of the
new Senior Interdepartmental Group, and monthly compilations on
shipping to North Vietnam and Cuba.
Special Publications
A problem common to these regular issuances is created by the con-
flicting demands of classification and dissemination. We want to
serve as broadly as possible everyone in the government requiring
intelligence information for the performance of his duties. On the
other hand, we want to be able to publish information of the most
restrictive classifications. We tightly limited the dissemination of
i:he Central Intelligence Bulletin from its inception in order to make
its content as comprehensive as possible. But new collection mech-
anisms with highly compartmentalized reporting systems now supply
information which cannot go even to all recipients of the Bulletin.
There are valid reasons for the restrictions, but they make it im-
possible to serve the Director and the President adequately with
normal publications.
We are therefore forced to create new and ever more tightly con-
trolled special publications for these readers. They are prepared by
a very small number of senior officers and go outside the Agency in
only a very few copies. Their content is governed by the concept that
there can be no piece of information so highly classified or so sensi-
tive that it cannot be passed to the President. The main one is the
President's Daily Brief. It generally follows the lines of the Bulletin,
but it contains added material too sensitive for the wider audience
and is written in a more spritely style, with less concern for citing
the evidence underlying the judgments expressed.
Inevitably, some such publications become more widely known and
get into such demand that their dissemination creeps up, no matter
how hard we fight it. At this point, lest the added circulation destroy
their purpose, we put sensitive information on a separate page included
only in the copies of the prime recipients.
The trouble with regular publications, in addition to the classifica-
tion problem, is that they tend to have fixed deadlines, format, and
dissemination schedules and hence suffer in flexibility and timeliness.
As a result, we have been turning increasingly to individual intelli-
gence memoranda to meet many of our responsibilities. Then we
can let the requirements of the particular case dictate the deadline,
the format, and the distribution, as well as the classification.
For the CIA research components one of the most important de-
velopments in recent years has been a sharp increase in the servicing
of policy makers with memoranda and longer reports devoted to
particular policy issues. This reflects both a more sensitive apprecia-
tion on our part of precisely what kinds of intelligence are required
and a growing awareness among policy officials that intelligence can
be responsive and helpful on some of the more troublesome questions
underlying their decisions. A few of the economic studies done
recently in support of policy decisions have been on the effects of
economic sanctions against South Africa, the logistic situation of the
Communist forces in Vietnam, the effectiveness of U.S. bombing there,
the consequences of certain proposed actions in the Zambia-Rhodesia
Approved For Release 2005/03/15 :
A-RDP78T03194A000200050001-8
SECRET
SECRET 9
 Fo
licsy 20b51s03/15 : Cl
Approved For e 2 Poleae
crisis, and the implications of change in U.S. economic policy toward
the Communist world.
From scientific and technical research come, for example, special
memoranda concerning foreign military research and development,
especially in the USSR and Communist China, for consumers such
as the President's Scientific Advisor and Advisory Board, the Presi-
dent's Foreign Intelligence Advisory Board, and the Director for
Defense Research and Engineering in the Department of Defense.
These officials have an important role in determining the direction
U.S. military research and development must take to counter the
Soviet and Chinese threat. They often require more detail than is
presented in the standard National Intelligence Estimate, or they
require very specific answers to equally specific technical questions.
Such memoranda are often accompanied by a briefing.
The intelligence memorandum originally prepared in answer to a
specific request from a senior policy maker also tends to generate
additional, self-initiated memoranda either to update the first response
or to insure that the recipient, in concentrating on a narrow aspect of
a problem, doesn't overlook something else that is germane. Finally,
in servicing such requests from the policy maker you build up over
a period of time an intuitive sense of what he is going to ask, and
you anticipate it.
The Operations Center
Another way we endeavor to insure that we are providing timely
and useful intelligence support is to know what is going on with U.S.
operational forces. We have found that our top customer regularly
expects a full picture of any crisis situation, particularly where U.S.
forces are involved or may become involved. To be able to marry
the kinds of data wanted on U.S. operations with the customary in-
telligence on foreign activities and developments, the intelligence pro-
ducers need regular inputs not only from the intelligence collectors
but from the operators. We need immediate access to the operational
people in National Military Command Center in the Pentagon. We
need to know the directives State is about to send to embassies in
crisis situations.
To deal with this problem, we have recently expanded our former
Watch Office into an Operations Center. The Center continues to
have the watch office function of filtering incoming information and
Approved For Release 2005/03/15 C
10 SECRET
-RDP78Td:f{94Ar0i66001-8
alerting the proper people as necessary. Outside of normal office
hours it is directed by an experienced generalist of senior rank.. It
has teleprinter service from the Foreign Broadcast Information Serv-
ice and from the National Security Agency. It has secure teleprinter
and voice communications with the White House, Pentagon, and
State Department, and through these switchboards with American
military and governmental outposts all over the world. The amount
of information received and screened in the Center is now running
in excess of a million items a year.
The Operations Center maintains up-to-date briefing information
on critical situations and areas in a special situation room. When
there is a major flap, a task force with representatives from all of
the components involved can be pulled into the Center to operate
there on a 24-hour basis if necessary. (At one period we had four
task forces going-on Vietnam, the Dominican Republic, Indonesia,
and Kashmir. I must say it got a little crowded in there.) During
the Dominican crisis, the Director called for situation reports every
hour on the hour, around the clock. To a certain degree Vietnam
reporting now remains in the same category.
The point is, of course, that the policy makers have gone tactical
in their concerns, and apparently this is the way it will be whenever
the United States is engaged in a fast-moving potentially dangerous
situation. At such times the President and his top cabinet officers
become involved in day-by-day and hour-by-hour operational planning,
down to the selection of targets and the deployment and commitment
of troops. This is because of the world-wide political implications
of tactical decisions today, and it is made possible by the capabilities
of modern communications systems. The situation room in the White
House is manned by seven of our experienced watch officers borrowed
from the Operations Center, who are no longer completely unnerved
to find the President peering over their shoulder at almost any hour.
Fund of Confidence
In summary, we might say that in a system to support the senior
policy maker two ingredients are essential-a good production base
and a readiness to adapt it as necessary. One must be alert to the
changing needs of the policy maker, and be ready to meet then.
Above all, there must be a pool of experienced intelligence officers,
both generalists and specialists, with continuity in their jobs and ob-
SECRET
 For Policy Chiefs
Approved For Release 2005/03/15 : C
jectivity in their outlook. Ted Sorensen wrote in Decision-Making in
the White House:
No President, of Course, pays attention to all the information he receives,
nor can he possibly remember it all. What he actually considers and re-
tains may well be the key to what he decides, and these in turn may depend
on his confidence in the source and on the manner in which the facts are
presented. He is certain to regard some officials and periodicals with more
respect than others. He is certain to find himself able to communicate more
easily with some staff members than with others. He is certain to find that
some reports or briefing books have a higher reliability than others.
We want the policy maker to be confident that in asking us for in-
telligence, he is getting as knowledgeable, pertinent, unbiased, and
up-to-date a presentation as it is possible to provide.
Approved For Release 2005/03/1
SECRET
L-RDP78T03194A000200050001-8
Status and prospects in a command
performance directed toward an inte-
grated data system.
SECRET
No Foreign Dissem
COMMUNITY PROGRESS IN INFORMATION
HANDLING
Zane Thornton
Information handling methods occupy a pervasive position in
the whole administrative framework of the U.S. intelligence
community. ... The systems problems of intelligence in-
formation access will continue to be of the most difficult type?
heightening the importance of great improvements in the
depth of understanding and of skills in tackling the wide
variety of such problems which confront all levels of Govern-
ment personnel concerned with access to the national intelli-
gence base.
In these words addressed to the President on 15 June 1965, his
Foreign Intelligence Advisory Board focused attention on a problem
that has assumed alarming proportions in the past decade. The con-
spicuous proximate cause has ben the information explosion created
by new collection technology and increasingly massive use of that
technology. But the volumetric burst has also been accompanied by
a significant diversification in the types of data that must be digested
and an increased need for interchange of information among intelli-
gence agencies with overlapping responsibilities. During the same
period the speed and power of the new weapons systems and the
critical potential of each policy decision have generated requirements
from outside the community, from national command authorities and
policy makers, for rapid delivery of large quantities of data. These
have perhaps put more strain on the community's information-han-
dling systems than its internal requirements have.
Advances in the technology of information processing, which have
great potential for curing this ill, have so far been applied in such
a way as often only to increase distress. Automatic data processing
is painfully superimposed on manual processing operations, a com-
puter used to duplicate an existing system instead of to make possible
a better system. In many applications the computer becomes little
CIA-RDP78T03194A000200050001-8
SECRET
 Comp XiddfP6HQL 6aY90A6V03/15 : Cl
more than a large, fast typewriter, its versatile capability almost un-
used. The analyst becomes skeptical of innovation, and the conflict
between the manual and the "automated" system degrades both.
Presidential Spur
The community, though aware of its need for better information
systems, has not been able to make satisfactory progress toward
getting them, either community-wide or in individual agencies. The
reasons for this are complex, but two deficiencies appear to stand
out. One is the shortage of systems-oriented personnel skilled in
information technology. Here intelligence must compete with the
rest of the government and with industry for a talent whose supply
is significantly less than the demand. The other is the lack of an
adequate central mechanism for coordinating and managing the com-
munity's effort. The agencies have not even been able to reach
agreement on what information systems are required, particularly
those which cut across agency and functional lines.
In this situation the community needed an authoritative push from
outside, and the PFIAB memorandum has sparked at least some
preliminary coordinated study and experimentation. It contained
three recommendations:
Recommendation No. 1. "That selected personnel among the
departments and agencies making up the U.S. intelligence com-
munity be provided specialized training and advanced studies
at a university center or centers where systems thinking and
systems skills are understood and imparted, and which at the
same time possess adequate background in conventional bibli-
ography and other more classical approaches to literature and
information management."
Recommendation No. 2. "That the Technical Information Process-
ing System (TIPS) project, now under way within the National
Security Agency, be expanded to include participation by other
member agencies of the intelligence community in an experi-
mental operating system constituting a first step toward inter-
agency (and interbuilding) information handling. Since results
should be sought from the experiment as promptly as feasible,
the participation of other agencies should be achieved by
September 1965; the capability for extensive handling of the
Russian biography problem should be available in the com-
munity-wide system by the summer of 1966; and by the summer
Approved For Release 2005/03/15 : Cl
SECRET
-RDP78T0691W$ dd626do8rdo6yrh Handling
of 1967 it should he possible to exchange outputs from various
mechanized sources in the fashion pioneered by the TIPS project."
Recommendation No. 3. "That there be established a Panel,
under the joint sponsorship of the Special Assistant to the Presi-
dent for Science and Technology and the President's Foreign
Intelligence Advisory Board, having responsibility for: (a.) pro-
viding guidance to the intelligence community in forwarding
of methods and facilities for information handling and access;
(b) evaluating in technical terms the true meaning of the
enormous and somewhat heterogeneous growth of the intelli-
gence community's information pool."
The President approved all three recommendations and instructed
the Director of Central Intelligence to carry out the first two. These
were passed for action to the USIB Committee on Documentation, and
CODIB established two ad hoc working groups to develop plans for
the systems training program and the expanded Tn'S experiment re-
spectively. This paper reviews the conclusions of the two working,
groups and considers prospects for an integrated community informa-
tion system.
It is emphasized that the plans herein described are only the prod-
uct of preliminary work. The proposals for expanding TIPS have been
approved by the USIB for implementation, but the indications a.re.
that the systems training plan will not be approved in present form
It is nevertheless worthy of notice for its conceptual approach to
the problem.
Systems Training
The PFIAB, in recommending specialized training for community
personnel in the information-handling field, suggested for the purpose
"university centers" (such as the Library School of the University
of Chicago) where both systems thinking and conventional library
methods are understood and taught. The CODIB working group
responsible for developing this recommendation read its spirit as a
general call for action and did not view the specific suggestion as
restrictive. Its proposal therefore goes beyond the scope of the
literal recommendation and presents a broader solution to the systems
training problem.
The working group recommended, in a report of 13 May 19136,
the establishment of an Intelligence Systems Institute to develop the
skills necessary for the building, operation, use, and management of
RDP78T03194A000200050001-8
SECRET'
 Cc pFovUdl o?r'e1ease 2885%03/15: CIA-
information systems. The Institute would be central to a compre-
hensive program of classroom instruction, laboratory work, field trips,
individual student projects, on-the-job training, and courses in aca-
demic and commercial institutions. It would be wholly controlled by
the intelligence community so that instruction in basic intelligence
techniques and processes could be interwoven with the instruction
in systems disciplines. Systems subjects would be taught in the con-
text of the real-world community environment in which the informa-
tion-handling systems must function.
There would be four different curricula for four kinds of trainees-
builders, users, operators, and executives concerned with information
systems. The builder is the one who analyzes the processing require-
ment and pertinent aspects of its environment, designs a system to
meet the requirement, constructs the system, and subsequently modifies
it in accordance with operational experience. The user is the cus-
tomer, the researcher or analyst whom the system serves. Operators
are those with the various skills required to run the system-indexers,
microphotographcrs, machine operators, etc. The "executive" means
any senior official whose responsibilities call for knowledge of the
system.
The builder would be the principal target of the Institute, and
his needs would be the chief determinant of the overall proposed
program. Ile must master more disciplines than the user, operator,
or executive to qualify as an effective member of the intelligence
systems team. His course would be five months long, whereas the
user's would last one month, the operator's two or four weeks, and
the executive's one week.
All four types of course would have some elective and some re-
quired subjects, varying with the background, experience, and assign-
ments of individuals. The builder's course would have five segments,
as follows:
Basic Fields-mathematics and statistics.
EDP Technology-computing equipment, basic and advanced
computer programming, data and storage structures, theory
of formal mechanized languages and design of special pro-
gramming languages, artificial intelligence, and electronic
communication.
Document Reduction and Handling Technology-media, dimen-
sion considerations and standards, and automated delivery and
read-out systems.
Approved For Release 2005/03/15 : Cl
SECRET
4DP78T0319&&b~ 000gDcg c-'h?n Handling
Systems Development Techniques-systems concept, system de-
velopment cycle, requirements and feasibility study, analysis
and design, implementation, performance evaluation, resources
for development, and personnel roles in development.
Intelligence Information Systems-introduction to the intelligence
community, intelligence data bases, intelligence analysis tech-
niques and their susceptibility to automation, and intelligence
and information processing systems and methods.
The courses for users, operators, and executives would be assembled
from appropriate portions of the builder's course, augmented by spe-
cial emphasis on problems unique to each.
The proposed Institute would not eliminate reliance on govern-
ment, industry, and university training in aspects of information-
handling systems. The facilities of the computer industry, for ex-
ample, would still be used to train community personnel in the opera-
tion and maintenance of hardware, and systems management personnel
would continue taking government or university management courses.
We would still look to the universities for courses in mathematics,
electronics, operations research, and other disciplines required in
systems work.
The main function of the Institute would be to teach basic systems
disciplines in the context of their application in the intelligence
community. By carefully blending subject matter it would form a
bridge between the theories of systems and their practical application
to live intelligence processing problems. As an important by-product,
it might foster interagency cooperation in the development of sys-
tems through the personal relationships established in its student
body. Students from all agencies would graduate with a greater
understanding of their counterparts' problems and a realization that
unilateral action is not an adequate approach to most important corn-
munity problems.
The working group recommended that the Institute be established
as an independent activity having a full-time, professional director
who would report to an interagency board. It would receive house-
keeping and logistic support from a member agency designated by
the USIB. The director would get technical advice and assistance
from a group comprising representatives of CIA's Office of Training,
the Defense Intelligence School, the National Cryptologic School,
and other training organizations as appropriate.
DP78T03194A000200050001-8
SECRET
 Community Information Hrr~I'/lq/15 : CIA-
Approved For Release UUU ~lT.S
CIA-
As we hinted above, this concept of an Intelligence Systems Institute,
however attractive, may be for the present an impractical one. The
proposal raises questions like the following. Can a truly qualified
information systems designer or analyst be produced in a five-month
training course? Could the Institute attract and retain the caliber
of systems educators needed to make it successful? Could it keep
up with the high rate of change in the technology of information
processing? Could it avoid conflicting with or duplicating other train-
ing programs? Is the Institute the most economical way to satisfy
the community's systems training requirement?
It doesn't appear now that these questions can be answered in a
way that would justify the early establishment of the Institute as
proposed. The exercise of developing the proposal has nevertheless
been useful. At least it will have focused community attention on the
requirement for systems training, and in the process it may have moved
us closer to a common understanding.
Experimental System: Design
The CODIB working group assigned the action on the PFIAB's
second recommendation, for broadening NSA's TIPS into an inter-
agency experiment, formulated the concept of a Community On-Line
Intelligence System (COINS) and set forth the details of its plan in a
report dated 25 May 1966. Tips, on which it is based, is an experi-
mental retrieval system rooted in NSA's Univac 490 remote-access
system and designed to handle thirteen different formatted files con-
taining approximately 38 million characters of information of interest
only to the cryptologic community. The pilot version has as its main
objective giving NSA technicians experience in developing and utiliz-
ing an on-line, near-real-time retrieval system. Follow-on develop-
ment, already under way, may lead within three years to a third-
generation system capable of handling approximately 500 million
characters. The ultimate goal is establishment of a Sigint Command
and Control Complex.
The COINS concept is not concerned with the purpose or content
of Tips but with its multiple-file, remote-access, and immediate-
retrieval features. While not intended by any means to solve all
interagency problems of information retrieval and exchange, the ex-
perimental system will give each participating agency a capability for
remote interrogation of its own files and selected files of other par-
ticipating agencies. It will operate on formatted files, as opposed to
Approved For Release 2005/03/15 : C
SECRET
DP78TO319~R6&J~ 0& 'bon Handling
continuous text materials, but it may include formatted indices of
continuous text materials, biographic dossiers, finished intelligence
publications, microfilm or video images, and documents.
The COINS network will have four computer centers and one remote-
enquiry station connected by communications lines and a store-and-
forward switching computer operated by DIA. The equipment will
be heterogeneous: NSA and NPIC will each use a Univac 494 com-
puter system with KG-13/IIN-9 crypto devices and modulator-de-
modulators feeding through 2400-bit-per-second secure data links to
the communications switch. CIA will use an IBM 360 Mod 50 com-
puter system with the same crypto and terminal devices. DIA will
use the II3M 1410 computer in its Intelligence Support and Indications
Center and will also maintain an IBM 7740 as the communications
switch for the network. State Department will use only a standard
100-word-per-minute teletype and KW-7 crypto devices to tie into
the communications switch.
The selection of intelligence data to be made available in the net-
work has been heavily influenced by PFIAB's specification of Russian
biographies for the first experimentation. Following is a partial list
of the data registers of various agencies to be used initially:
Soviet Personalities and Organizations
Soviet Scientific and Technical Personalities
Soviet Military Personalities
Soviet Airfields
Target Briefs
Air Order of Battle
Missile Order of Battle
Radar Order of Battle
Air Defense Order of Battle
Viet Nam Activities
Chinese Communist Location Dictionary
Chinese Communist Organizational Directory
Soviet Biographic Dossier Index
Soviet Elite Dossier
Soviet Elite Travel
In the initial phases, participants will be restricted to file-oriented
interrogations, as opposed to query by subject matter. This device
to relieve the urgency of some software development problems will
require the using analyst to have a very thorough knowledge of the
file he wants to interrogate. At a minimum, he must know its name.
the names of its fields, and what data elements are in each field.
SECRET
 SECRET C p rr,ibled-F @rr l@@sU?@&(Y3/15 : CIA-JDP78T031~4~YUC~"doO I~Pf~'11?8t'On Handling SECRET
Participating agencies will thus have to exchange extensive file de-
scriptions and give analysts training in them.
Significance of the I'xpcrirnent
The real significance of COINS lies in the amount of coordinated
developmental and technical work required to put the system into
operation. All of the hardware must be assembled and connected
through cryptographic, communications, and data terminal equipment
to the central communications switch. The completed network must
meet the technical criteria for handling transmissions to the Top
Secret Special Intelligence level. But the hardware aspects, complex
as they are, are the least of the problems.
Software development will he complicated because of the different
kinds of hardware being used and the fact that there is no real com-
monality among the computer programs which the individual agencies
already use to process their own files. The COINS software will have
to compensate for both sets of dissimilarities. It will include a user's
language to enable the analyst to query the system, an acceptor pro-
gram analyzing incoming messages to determine whether the re-
quester is authorized to interrogate a particular file and whether the
file is currently available, translate programs to convert incoming
interrogations into the right form for the computer in question, pro-
grams for store-and-forward switching, and programs to maintain
chronological logs of activity at each terminal and at the communica-
tions switch.
Another series of problems turns on the files of intelligence data
to be used. At least initially, a file can be accepted only if it is in
machinable form, useful to an analyst, and offered by its owner for
others' use. Preliminary indications are that the number of files satis-
fying these criteria will certainly not overload the experimental system
in its early phases. Then when the files have been selected it is
necessary to attempt standardizing the data elements and codes in
them to make them usable without other costly conversion programs.
Many difficulties and delays have been encountered in trying to estab-
lish general community-wide standards in this respect, and it may be
necessary to set up special codes and standards just for the COINS
experiment. Finally, there is the task of maintaining the files once
they have been incorporated into COINS. So far only exploratory work
is under way on this problem,
Approved For Release 2005/03/15 : Cl
SECRET
In any experiment, evaluation of the results is a crucial step. The
nature of COINS will make a (Ina] evaluation necessary--a technical
critique to determine whether the network and its individual com-
ponents have met specifications, and an operational review to find
out how useful the system is to the analyst. The technical evaluation,
in addition to appraising hardware and software performance, should
judge the practicality, in the sense of economy and effectiveness, of
an intelligence information-handling network made up of several dis-
similar computer systems patched together with special software
packages and conversion routines.
The operational evaluation will probably be more difficult to per-
form and will undoubtedly be more important in terms of impact on
the community's future course. Although ultimately influenced. by
technical performance, it will focus directly on the user's appraisal of
the system as a method for retrieving items from the national intelli-
gence data base. The results of this evaluation will determine whether
there should be a follow-on to the experiment and, if so, what char-
acter it should have.
COINS could be one of the most important experiments ever under-
taken by the intelligence community. At its worst, it could prove
that it is nonsense to talk about an integrated community Whose mem-
bers practice a policy of maximum exchange of information in the
interests of producing the best possible national intelligence. At its
best, it could help make the integrated community a reality, providing
a base on which to build succeeding generations of information-
handling systems to link and service that community.
The Third Recommendation
The PFIAB memorandum of 15 June 1965 is a paper of potentially
even greater importance to the intelligence community than the con-
crete developments we have described suggest. Because the systems
concept which the PFIAB has set loose in the community is (in its
purest form) no respecter of arbitrary boundaries, parochial inter-
ests, preconceived solutions, or conventional wisdom, one cannot help
wondering what its eventual impact will be. Will the memorandum
turn out to have been a one-shot affair whose effects wore off after
a minimal response to each of its recommendations? Or a thin-edge-
of-the-wedge which ultimately engendered organizational and func-
tional changes in the intelligence system?
DP78T03194A000200050001-8
 SECRET Community Information Handlin
Approved For Release 2005/03/15 : CIA- DP78TO319 ~~~'d2d 0 d1'c 'on Handling
The more specific question is this: will the community's information-
handling mechanism toward which we are now taking a first step be a
tightly-knit, unified system responsive to a single authority or a loose
confederation of several semiautonomous systems each responsive to
a different master? The PTIAB intent is ambiguous in this respect;
the experiment the Board called for in Recommendation No. 2 can
be viewed as one merely for developing information exchange tech-
niques rather than the preliminary foundation of a large, truly inte-
grated system. Perhaps such ambiguity is inevitable at this stage.
The general USIB philosophy of joint operation by consensus has
precluded the establishment of a central body with the authority to
set community-wide information system goals and direct their ac-
complishment. CODIB, viewed as a forum for the exchange of ideas
and information to assist individual agencies in developing separate
systems and then to encourage improved communication among these
as steps in the evolution of community networks, is a useful instru-
ment, reflecting the often assiduous part-time efforts of its members
and the good will of their parent agencies. But from the viewpoint
of the systems engineer, who wants early agreed decisions and vigorous
follow-through in order to build effectively, economically, and in time
to meet the problem, the Committee has wholly inadequate resources
and authority. It must labor for months or even years to achieve
only a partial solution to well-defined difficulties which all members
agree should be solved by joint action. On ill-defined matters char-
acterized by divided opinion and vested interest, progress often ap-
proximates zero.
In these circumstances the most important of the PFIAB recom-
mendations, in this writer's opinion, may be the third, for a mechanism
providing guidance and direction to the community's information-
handling systems. If the Guidance and Evaluation Panel established
in response to this recommendation is vigorous and competent in its
probing of the community's present capabilities and future needs, and
if it can see to it that its recommendations are acted on, then it may
become the nucleus of a central body that will give sense and purpose
to the community's development of information systems. If en-
trenched departmental prerogatives make a voluntary joint solution
of community-wide information-handling problems too slow and in-
efficient, there may be no alternative but to impose the solution from
outside through such an instrument of the President's Board.
Approved For Release 2005/03/15 :
SECRET
The PFIAB has identified a planning and management gap in the
intelligence community. It has left sufficient latitude for the com-
munity to fill that gap by voluntary joint action if it can. At
the
same time, it has laid a possible foundation for bridging the void
externally if necessary.
-RDP78TO3194A000200050001-8
SECRET
 Approved For Release 2005/03/15 : CIA-RDP78T03a944MM~01-8
o Foreign Dissem
A positive but undeluded view
of ADP for intelligence refer-
ence services.
AUTOMATION FOR INFORMATION CONTROL
Paul A. Borel '
Last July the Saturday Review, in a special issue on the automation
revolution, equated the computer with the atomic bomb as a tech-
nological development formidable enough to make a turning point in
human history. Some months earlier a Newsweek report entitled
"Good-by to Gutenberg" gave readers a glimpse of other things to
come in the field of information technology: a photosensitive crystal
the size of a sugar lump that is capable of containing images of
100,000 pages; a lensless photographic system which could lead to
three-dimensional home television; a no-contact, no-pressure printing
technique that can write on sand, print a message on a pizza, or put
a trademark on a raw egg yolk.
Marshall McLuhan, in Understanding
Media-the Extensions of Alan, predicts that books and newspapers
will in time no longer exist, that publishing will give way to an active
servicing of the human mind through research packages done to suit
individual needs.
Spume and Substance
Over the last 20 years we have learned not to depend on such
extravagant promises to get us out of our practical difficulties. I do
not doubt that amazing developments will continue to take place. I
do doubt that we can count on them for early and revolutionary solu-
tions to our data-handling problems. The problems created by the
exploding mass of intelligence information have a habit of staying
well in front of innovations.
In CIA we are currently upgrading our computer facilities with
third-generation hardware, infinitely superior to our initial gear. Yet
the contribution of the computer to the task of producing intelligence
Adapted from Section IV and Appendix C of the author's presentation to the
London Intelligence Methods Conference, September [966. The full paper, Con-
trolling Intelligence Information, 51 pp., covers trends in information, their impact
upon intelligence, and the controls used, including the use and promise of ad-
vanced information processing systems.
' 'L ONFIDENTIAL
25
 CONFIDENTIAL Automation Automation CONFIDENTIAL
Approved For Release 2005/03/15 : CI -RDP78T03194A000200050001-8
is still both specialized and very limited. We have secure telephone
communications; but these arc far from ideal, with few instruments
and high costs. Great strides have been made in our printing estab-
lishment. Still, the lapse between preparation of copy and its avail-
ability to the reader can be measured in weeks rather than days for
non-priority items. Reproduction techniques have shown major gains.
But material received in such poor quality that it cannot be microfilmed
runs in some categories as high as 20%. We have improved our
means of instructing reporters. Yet 50% of the titles in some report
series have to be rewritten to reflect the content properly.
This experience gives ground for caution against any wholesale
abandonment of the workable (if less than satisfactory) old in favor
of the glamorous but untried new. Nevertheless, this is a time of im-
portant new developments in practical means for information handling,
and intelligence should pay more attention to what is going on in this
field outside. As never before, we have opportunities to capitalize
on the work and ingenuity of others to relieve some of our own prob-
lems. Much of the work done outside is solid and relevant. We
ought to use it, pick-a-back, whenever we can.
Active State of the Art
Let me mention a few such outside developmental activities touch-
ing the library science field. Two programs are being carried out
in the academic community at large. One, named Intrex, for "in-
formation transfer experiments," has been called a step toward a Bial-
a-thought world. It is setting up an experimental laboratory to test
ways of giving professors and students instant access to information.
Xerography, film projection, and telephone communication between
computer and user are planned. Basically, the experiments will at-
tempt, first, to automate and rationalize the functions of libraries and,
second, to develop a computer-based information transfer network.
Another program, under an organization called Educom, the Inter-
university Communications Council representing over 30 universities
in 20 states, is evaluating the significance for higher education gen-
erally of electronic hardware (computers, light pens, graphic dis-
plays), and software (computer programs).
A number of individual university libraries have forward-looking
programs, Washington State and Florida Atlantic to name only two.
The latter has the distinction of being the first in the United States
to have introduced data-processing methods and techniques into its
Approved For Release 2005/03/15 : C
CONFIDENTIAL
operations at its very beginning. Washington State, on the other
hand, is converting from traditional library methods to a totally on-line
system which offers multiple remote access to a single library record.
Sharing the time of the university computer (an IBM 360/67), it
will be able to reduce typing substantially, eliminate duplicate manual
riles, and give complete control of each item's location and status in
the library.
The value of these projects to us is that they are comparable in size
to those which intelligence libraries may undertake. While much
valuable information has been published about ways to automate the
Library of Congress, the sheer size of its holdings makes many of the
parameters of that undertaking inapplicable for us.
The publications of professional engineers, documentation specialists,
and experts in various aspects of the information-handling :industry
are also increasingly solid and relevant. Some particularly useful
books and articles are listed in the bibliography. These make evident
my point that outside the intelligence community there is much
wisdom and talent which we have neither tapped sufficiently nor used
effectively because we are ill organized to do so.
CIA Applications
Certainly CIA has had in the past no organization worthy of the
name to identify this outside work and relate it to our own improve-
ment programs. This gap has now been filled with the organization
of our Intelligence Sciences Laboratory, which is acquiring its own
computer and associated equipment to provide an experimental en-
vironment closely approximating actual operations. Illustrative of its
prospective areas of activity are on-line analytic processing, pattern
recognition, language and text processing, and speech and audio ma-
nipulation. We will thus better bridge the work done outside and our
own EDP-related operations.
While CIA pioneered much automatic information processing with
its punched-card equipment, our experience with general-purpose Com-
puter operations is short of six years. In those years we have very
considerably expanded our use of these machines. We now have a
major computer center providing counterintelligence and operational
support, another serving intelligence production, and a third devoted
to imagery analysis. In the last three years we have reached the
point where we see computer applications in almost every element
of the intelligence cycle, from the management of collection require-
ments to the printing of finished intelligence.
DP78T03194A000200050001-8
CONFIDENTIAL
 Approved For Release 2005/03/15 : CIAIRDP78TO3194A000200050001-8
Some of these applications are initiated by the analyst charged with
intelligence production. Where the task data is numeric (like mili-
tary-economic costing or agricultural production and soil moisture
statistics) or in simple standard format and the data preparation is
done by the EDP staffs, such applications, especially in the economic
area, have given high-yield products which require no great invest-
ment of effort from the analyst. A wide range of other applications
of this type remain to be tried, including computer control of Soviet
scientific and technical literature. Work of this kind, involving pri-
marily collaboration between an analyst and an applications expert
from our central computer facility, can be characterized as special
projects.
A much more ambitious application is our current attempt to change
fundamentally the present method of doing business in the Agency's
Office of Central Reference-our Project Chive. In one degree or
another this general project will change the way hundreds of analysts
are now working.
The Chive Project
The need for Chive arose from developments over the past nearly
20 years, during which we evolved a number of special reference
services to support the production analyst. The multiplicity of classi-
fications, of indexing tools used for control, and of formats employed
in collection, dissemination, storage, and retrieval made it increasingly
difficult to meet customer needs. The problem of heterogeneity was
compounded by the increase in volume of data received and, with
the passage of time, the volume in file. Moreover, the intelligence
production expected of the analyst today is characterized by greater
sophistication and shorter deadlines. Project Chive is designed to
help him meet that challenge.
There are many unique features of the Chive approach. The project
team is an integrated group drawn from production, reference, and
computer components of the Agency, and it includes contract per-
sonnel as well. Experienced operators of our information systems
have been given training in advanced techniques and placed in charge.
The prospective user of the system is drawn in as active participant. A
single important geographic area, China, has been selected for first
application of the system, and even here it will be conducted as a
pilot operation in parallel with the old system to permit experimenta-
tion before it carries the whole load.
Approved For Release 2005/03/15 : Cl
CONFIDENTIAL
From the user's point of view, the system should provide a number
of advantages, most immediately:
All-source retrieval from a file system covering every type of
printed document, including maps and photos, at whatever
classification level.
Single-point retrieval service organized by geographic area, as
opposed to the old multiplicity of indices and registers.
Literature searches that turn up all books, documents, reports, etc.,
that bear on a subject in question.
Information searches that turn up facts in answer to specific ques-
tions, facts concerning foreign personalities, organizations, in-
stallations, and activities.
Counts, whether of Algerian students in the USSR, public ap-
pearances of the Chinese leaders, or CIA intelligence reports
on Haiti, and the trend of changes in such counts.
Detection of redundancies and inconsistencies in the system store.
Less matter-of-course but not at all visionary are services like the
following:
Automatic inference-making-manipulating the wide variety of
stored facts about people, institutions, and activities to produce
new hypotheses about their character and connections. The
variety of problems to which such a capability might be applied
would depend on the ingenuity of the intelligence analyst.
Machine-assisted language translation, both machine translation
of Russian documents and machine conversion of oral transla-
tions from other languages into printed documents in English.
Analyst referral service from a central directory or "profile" of
human sources with expert knowledge in special subjects.
Remote querying which will enable users to interrogate and main-
tain from their own offices special-purpose files in the central
system.
Half a Loaf
We see the development of this improved system as extending over
ten years and many difficulties. Only because computer technology,
capability, and capacity are what they are today and will be tomorrow
do we dare count upon the success of the project.
Even so, there are risks. In this costly field you try to reduce the
risks, but after you have done all you can they are still considerable.
DP78T03194A000200050001-8
CONFIDENTIAL
 CONFIDENTIAL
Automation
Approved For Release 2005/03/15,
We recently had a series of meetings with Dick Brandon of Brandon
Applied systems during which he reviewed with us the experience of
others in using computers. He said that of 16,000 installations
equipped with 27,000 systems in the United States today, 40% are un-
successful in the use of their computers. This means that 6,500
organizations are not deriving economic benefit from them or arc not
achieving their objectives. In 90% of these cases, schedules and
budgets have been exceeded. The main reason for this, in Mr.
Brandon's view, is that the people using the machines are 'way behind
the technology. They are not capable of utilizing the machines'
capability.
My own rule of thumb in the application of machines to non-
numeric problems is this: expect half as much in twice the time at
twice the cost. If you get it you can count yourself lucky.
Books
Planning Conference on Information Transfer Experiments, Massachusetts
Institute of Technology, 1965. Intrex: Report of a Planning Conference
on Information Transfer Experiments, 3 September 1965. Edited by
Carl F. Overhage and R. Joyce Harman. Cambridge, Massachusetts:
MIT Press, 1965. 276 pp.
King, Gilbert W., and others. Automation and the Library of Congress.
Washington: Library of Congress, 1963. 88 pp.
Licklider, J. C. R. Libraries of the Future. Cambridge: MIT Press, 1965.
219 pp.
Bourne, Charles P. Methods of Information Handling. New York: J. Wiley,
1963. 241 pp.
Greenberger, Martin (Ed.). Computers and the World of the Future. Cam-
bridge: MIT Press, 1962. 340 pp.
Feigenbaum, Edward Albert, and Julian Feldman (Eds.). Computers and
Thought. New York: McGraw-Hill, 1963. 535 pp.
Borko, Harold (Ed.). Computer Applications in the Behavioral Sciences.
Englewood Cliffs, New Jersey: Prentice-Hall, 1962. 633 pp.
U.S. National Science Foundation, Office of Science Information Service.
Current Research and Development in Scientific Documentation. No. 14.
Washington, 1966. 662 pp.
Conference on Libraries and Automation, Airlie Foundation, 1963. Libraries
and Automation; Proceedings, edited by Barbara Evans Markuson.
Washington: Library of Congress, 1964. 268 pp.
U.S. National Academy of Sciences, National Research Council, Automatic
Language Processing Advisory Committee. Language and Machines:
Computers in Translation and Linguistics. Washington, October 1966.
American Documentation Institute. Annual Review of Information Science
and Technology, Vol. 1, edited by Carlos A. Cuadra. New York: Inter-
science Publishers, John Wiley & Sons, 1966. 389 pp.
Approved For Release 2005/03/15
CONFIDENTIAL
gg t matton
CIA-RDP78TO3194A000200050001-8
Articles
Fiock, L. it., Jr. "Seven Deadly Sins in EDP." In Harvard Business Re-
view pp. 88-96, May/June 1962.
Flood, Merrill M. "The Systems Approach to Library Planning." In Li-
brary Quarterly 34:326-338, October 1964.
Knox, William T. "The New Look in Information Systems," Washington:
Office of Science and Technology, Executive Office of the President,
1966. 18 pp.
Salton, Gerard. "Progress in Automatic Information Retrieval." Reprint
from IEEE Spectrum vol. 2, no. 8:90-103, August 1965.
Shapiro, E. 13. "An Evaluation of Computers in Text Editing." Menlo
Park, California: Stanford Research Institute, 1966. 31 pp. (SRI
project 5849).
Swanson, Don R. "On Improving Communications Among Scientists." In
Library Quarterly 36:79-87, April 1966.
Schwartz, Judah L. "Computers and the Polieynuiking Community." Liver-
more, California: University of California, Lawrence Radiation Labora-
tory, 1966. 19 pp. (AEC contract no. W-7405-eng-48). (UCRL-
14887; TID-4500, UC-32).
John McCarthy and others. "Information." In Scientific American, Sep-
tember 1966 (special issue), pp. 65-260.
U.S. Federal Council for Science and Technology, Committee on Scientific
and Technical Information. "Recommendations for National Docu-
ment handling Systems in Science and Technology." Washington, 1965.
20 pp. Appendix A, "A Background Study," vol. 1. Washington, 1965.
"The New Computerized Age." In Saturday Review, pp. 15-37, 23 July
1966.
IA-RDP78TO3194A000200050001-8
CONFIDENTIAL
 Approved For Release 2005/03/15: CIA-RDP78T03i94A00020?58&01-8
reign issem
a ouie upper-stage space vehicle.
THE KIDNAPING OF THE LUNIK
Sydney Wesley Finer
A number of years ago the Soviet Union toured several countries
with an exhibition of its industrial and economic achievements. There
were the standard displays of industrial machinery, soft goods, and
models of power stations and nuclear equipment. Of greater interest
were apparent models of the Sputnik and Lunik space vehicles. U.S.
intelligence twice gained extended access to the Lunik, the second
time by borrowing it overnight and returning it before the Soviets
missed it. This is the story of the borrowing, which required the
efforts of many people and close cooperation between covert and
overt intelligence components.
On View Abroad
The Soviets had carefully prepared for this exhibition tour; most
of the display material was shipped to each stop well in advance.
But as their technicians were busily assembling the various items in
one exhibition hall they received a call informing them that another
crate had arrived. They apparently had not expected this item and
had no idea what it was, because the first truck they dispatched was
too small to handle the crate and they had to send a second.
The late shipment turned out to be the last-stage Lunik space
vehicle, lying on its side in a cabin-like crate approximately 20 feet
long and 11 feet wide with a roof about 14 feet high at the peak.
It was unpacked and placed on a pedestal. It had been freshly
painted, and three inspection windows cut in the nose section per-
mitted a view of the payload instrument package with its antenna.
It was presumably a mock-up made especially for the exhibition; the
Soviets would not be so foolish as to expose a real production item
of such advanced equipment to the prying eyes of imperialist
intelligence.
Or would they? A number of analysts in the U.S. community
suspected that they might, and an operation was laid on to find out.
After the exhibition closed at this location, a group of intelligence
officers had unrestricted access to the Lunik for some 24 hours. They
ApprW,or
Release 2005/03/15: CIA-RDP78T03194A000M 3Y IP8PAGES 33-39
 Approved For Reel . s6u2ti05/03/15 :
jjich
found that it was indeed a productio ruitem fr
enh ree ved
arid most electrical and electronic < l
t he of probable per
from the viewpoint
y examined it th suorourernghelrryts.
formance, taking mea dctcrrninirrg its structural character
istics and wirier r format cstimatirr~r cu Yin< ~.e, and so forth.'
3 A tew \
but
been copied from the "t"" durnrg this operation , ut not with
sufficient detail or precision to pcrnnt a definrt~stem used. was
the producer or deterrniuatirr of thes~ F
ccess or a factory
l
rcr a
therefore decided to try to get urot
team.
For the ultimate c nitribntioce of this iufonu;ition and a sketch of the Lunik see
Space Rce, hs' .1lbert ll. WI 'elon and Sidney N. Gray
"intelligence for the a
Approved For Release 2005/03/15 : C
SECRET
The Lunik
IA-RDP78T03194A000200050001-8
lions (timid 1'rnhh'ms
As the cs hihition moved from one city to another, an intercepted
shipping m ii ifest showed an item called "nxxlels of astrornouric a,r
pauatus" ys'hom, dimensions were approxinrrtely those of* the L.mil'
crate. "1'hi:; oformnation was sent to the CIA Station nearest the
destination ssath a request to try to arrange secure ;recess if the Ltniif
should On the hasis of our experience at trade lairs anei.
other exhibitions, we in factory markings preferred access heferc the
opening of ann exhibition to the alternatives of examining it ~duleiii
the exhibition hall or after it had left the grounds for anothc=-
dcstination
Soon the Lunik crate did arrive and was taken to the exhibition
grounds. 'I he physical situation at the grounds, however, rnlcd on:
access to it icrior to the show's opening. 't'hen during the slims th,
Soviets prcvidcd their own 2'f-hour guard for the displays, so ihcr,
was no pcssihility of marking a surreptitious night visit. 'Ibis lit
only one charnce: to get to it at some point after it left the exllibitior
grounds.
In the nrc.urtime our four-man team of specialists from tire' Jo.ni
Factory NI ii kings Center had arrived. A'Ve brought along our spc
cialized photographic gear- and basic tools. We each went out an,I
bought a complete set of local clothes, evcrythiug from the skin ont
\V ,c held a "cries of meetings with Station personnel over the c,uu-s('
of a week mutually defining capabilities and regrrirenieuts, laviia_..
plans for < c r'ss and escape, and determining what additional t.,11111,
merit we vonld need. The Station photographed the Lunik crate
repeatedly v.o we would get a better idea of its coustrirctiori. 'I'hc
photograptrs showed that the sides and ends were bolted togetlr. i
from within, the only way to get inside was through the roof. ~A'
therefore bought more tools and equipment-la;L GODI HEY, Jn. WI1.I ,IAM N. MOREl.r
JOHN H. RICHAnDSON
Additional members of the Board are
drawn from other CIA components.
SECRET "' `. i SECRET"
 Approved For Release 2005/03/15 : CI ,-RDP78TO3194A000200050001-8
CONTENTS
CONTRIBUTIONS
Contributions to the Studies or communications to the editors may
come from any member of the intelligence community or, upon in-
vitation, from persons outside. Manuscripts should be submitted
directly to the Editor, Studies in Intelligence, Room 1D 27 Langley
I I and need not be coordinated or submitted through chan-
nels. They should be typed in duplicate, double-spaced, the original
on bond paper. Footnotes should be inserted in the body of the text
following the line in which the reference occurs. Articles may be
classified through Secret.
DISTRIBUTION
For inclusion on the regular Studies distribution list call your office
dissemination center or the responsible Central Reference Service desk,
For back issues and on other questions call the Office of the
Editor,
All copies of each issue beginning Summer 1964 are numbered serially
and subject to recall.
25X1
A Theorem for Prediction .................... Jack Zlotnick
A mathematical model based on probability theory may aid
intelligence analysts. SECRET
Page
1
On the Soviet Nuclear Scent .......... Henry S. Lowenhaupt 13
Detective work on the USSR's early atomic effort centers on
shadowing Gustav Hertz and his fellow scientists. SECRET
Aerial Photography for Agriculture ....... William R. Gasser 31
Experimental methodology for crop estimation by overhead
reconnaissance. SECRET
The Metal Traces Test
William J. Maximov and Edward Scrutchings 37
The gun-toting guerrilla betrayed by chemistry. CON-
FIDENTIAL
Roderick "Steve" Hall .................... Anthony Quibble 45
An Alpine tragedy from World War H. OFFICIAL USE
Communications to the Editors ............................ 79
From counterintelligencers on their toes. CONFIDENTIAL
The Sherlock Holmes of the Revolution ... Rita T. Kronenbitter 83
Burtzev, one-man counterintelligence bureau for the Russian
revolutionaries. CONFIDENTIAL
Passport to Death ........ Contributed by Walter Pforzheimer 101
Marta Hari's last application.
Intelligence in Recent Public Literature. OFFICIAL USE
The German scientists ................................ 103
Contemporary issues ................................. 106
Notes from World War II ............................ 110
In Memoriam ........................................... 113
Desmond FitzGerald
Approved For Release 2005/03/15 Cl1
 Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
 I
Approved For Release 2005/03/15 : CIA
Experimental application of prob-
ability mathematics to predictive in-
telligence estimates reveals a disci-
plinary potential.
SECRET
No Foreign Dissem
A THEOREM FOR PREDICTION
Jack Zlotnick
THE STUDIES IN INTELLIGENCE AWARD
An annual award of $500 is offered for the most significant contribu-
tion to the literature of intelligence submitted for publication in the
Studies. The prize may be divided if the two or more best articles
submitted are judged to be of equal merit, or it may be withheld if
no article is deemed sufficiently outstanding.
Except as may be otherwise announced from year to year, articles
on any subject within the range of the Studies' purview, as defined in
its masthead, will be considered for the award. They will be judged
primarily on substantive originality and soundness, secondarily on
literary qualities. Members of the Studies editorial board and staff
are of course excluded from the competition.
Awards are normally announced in the first issue (Winter) of each
volume for articles published during the preceding calendar year. The
editorial board will welcome readers' nominations for awards, but re-
serves to itself exclusive competence in the decision.
Philosophy, wrote critic and educator Mortimer Adler, is the process
of entertaining any idea as merely possible. This act of tentative
acceptance is the good beginning in intelligence analysis. The de-
sirable end is a correct evaluation of the several hypotheses' com-
parative merits.
Seldom is the evidence so determinative as to clinch the case for
a single hypothesis. Usually, as it accumulates, it only changes the
position of one hypothesis or another on the probability scale. Sur-
prise attack is more likely or less likely today than it was a week
ago; a Sino-Soviet break in diplomatic relations is more probable or
less probable now than before; it is becoming more doubtful or less
doubtful that the Labor government's position against pound devalua-
tion can withstand the next speculative run on sterling.
Since intelligence judgments are so often probabilistic, does it follow
that the mathematical theory of probability offers intelligence valid
pointers on logical method? Promising research with relevance to
this question, some of it government-financed, has been done by
psychology faculties in university laboratories. The main aim of
the psychologists has been to compare intuitive judgments about
hypotheses with the results that would be given by a mathematical
model based on probability theory. Borrowing from these experi-
ments, CIA's Office of Current Intelligence in the summer of 1967
designed a mathematical simulation of predictive intelligence analysis
in crisis situations of recent history.
The mathematical model derives from an equation, familiar to
students of probability theory, named after Reverend Thomas Bayes,
who first formulated it in the eighteenth centur
The foll
i
y.
ow
ng ex-
position of Bayes' Theorem does not require mathematical sophisti..
cation of the reader; it assumes only that his learning blockages do
not include an ingrained antipathy to any kind of numerative idea.
Approved For Release 2005/03/15 C1 -RDP78T03194A000200050001-8
MORI/HRP 1-12
 Rave SECRET Approved For Rease 2005/03/15: CIA-~2DP78T03194602Oe`~~01-8 SECRET
Bayes' Theorem
A good entry point for the discussion is the concept of probability
as it is used in mathematics. In the absence of certainty, the proba-
bility that an event will occur (or has occurred, if past occurrence is
the matter at issue) has a decimal or fractional value between zero
and one. Thus the probability is .7 that a red poker chip will be
picked in a random drawing from a box containing ten chips, seven
red and three blue. A rational gambler would give no more than
$7 for a raffle ticket that paid $10 upon the random drawing of
a red chip from the box.
In the idiom of wagers, the term odds is often used instead of
probability. The odds favoring the random selection of a red chip
over the random selection of a blue one set the probability of the
first event against the probability of the second. The odds of seven
to three in this case are represented mathematically as the fraction
obtained by dividing the .7 probability of drawing a red poker chip
by the .3 probability of drawing a blue one.
New evidence affects a gambler's estimate of probabilities or odds.
Suppose there are two large boxes filled with red and blue poker
chips. In one the ratio of red chips to blue is 60 to 40; in the other
it is 40 to 60. One of the boxes is set before a gambler, but he is
not told which. He can therefore give no better than even money
that its color mix is predominantly red or blue. Allow him to draw
some of the chips, however, and he will then make a more confident
choice between the two color-mix possibilities. The more chips he
draws, the better the odds he will offer in favor of this choice.
This is precisely the setting of recent laboratory experiments at the
University of Michigan and other centers. College students, serving
as the test subjects, were required to give their gambler's judgments
of the odds after successive drawings of poker chips, and these judg-
ments were compared with the odds obtained by using Bayes' Theorem.
In more simplified notation than is commonly used in the textbooks,
the equation of Bayes' Theorem can be written:
R=PL
R, standing for revised odds, represents the odds favoring one hy-
pothesis over another after consideration of the latest evidence (in
this case, the color of the poker chip most recently drawn). P stands
for the prior odds, those prevailing before this evidence turned up.
L, the weight of the evidence that changes the odds, stands for like-
lihood ratio (referred to sometimes in the literature as Bayes' Factor).
Approved For Release 2005/03/15 CIA
RDP78T03194A000200050001-8
2 SECRET
SECRET
The likelihood ratio compares the probabilities of the occurrence of
an event under alternative hypotheses. Suppose the evidence in the
poker chip experiment is the selection of a red chip on the first
drawing. There is a .6 probability of this happening under the hy-
pothesis that 60 percent of the chips in the box are red. There is
only a .4 probability of its happening under the hypothesis that the
drawing is from the other box, where only 40 percent of the chips
are red. So the likelihood ratio for the occurrence of this red draw-
ing is .6 divided by .4, or %Z.
The prior odds, P-here 1/1 for even money-are multiplied. by this
L to get the revised odds after the first drawing. The revised odds
then become the prior odds on the second drawing, and so on. Sup-
pose the gambler draws 12 red and 8 blue poker chips in the :first
20 drawings, replacing the chip in the box after each drawing. Cal-
culation will show that he could give better than 5 to 1 odds in
favor of the hypothesis that he has been drawing from the box with
the 60-40 red-blue color mix. If the first hundred drawings are 56
red and 44 blue he could give well over 100 to 1 odds in favor of this
hypothesis.
Significance for Intelligence
He could and he would if he reasoned like a mathematician and
had the capital to finance many wagers of this sort. Otherwise he
would probably shrink from the degree of certainty implied by such
high odds. The students in the University of Michigan experiments
did give more confident odds the more drawings they had to go on.
They did not, however, move as far from their original one to one
odds as Bayes' Theorem would have justified. They did not, in
other words, make the most of their inconclusive data. Like intelli-
gence estimators in some parallel situations, they hesitated to move
very far very fast from prior norms.
Similar overly conservative estimates were obtained in University of
Michigan experiments simulating intelligence analysis. A set of six
hypotheses was set before the test subjects-five of different imminent
war situations and a sixth of peace. A scenario of events provided
successive increments of evidence bearing on these hypotheses. For
each increment the test subject gave five likelihood ratios expressing
his opinion of how much more likely the event would be under each
of the war hypotheses than under the peace hypothesis.
 SECRET Approved For4 'a e2 I03/15 : CIA-RDP78T0A94 b49W50001-8 SECRET
The test subjects of course differed among themselves in their
judgment of the proper likelihood ratios. But the most noteworthy
feature of the experiment was that their conclusions were not con-
sistent with their own readings of the evidence. Like the subjects
in the poker chip experiments, those working with intelligence sce-
narios were very conservative in their final estimates. When their
likelihood ratios implied, according to Bayes' Theorem, odds of 19
to 1 in favor of a war hypothesis, their own blend of intuition and
reasoning resulted typically in odds of 2 to 1. When the scenario
was changed and mathematical calculations would have given 19 to 1
odds favoring peace, they came up with odds in the neighborhood
of 6 to 1.
What Bayes' Theorem thus does for intelligence is to offer a mathe-
matical test for internally consistent analysis. The rigor of mathe-
matical logic is no indispensable aid when analysis is largely deduc-
tive, proceeding from such general propositions as "The USSR ap-
preciates how dangerously provocative would be its shipment of stra-
tegic missiles to Cuba." The instructed intellect's naked eye, so to
speak, is keen enough to follow the thread of deductive thought and
to detect the more tenuous strands of the argument.
The case for mathematical assistance is stronger when analysis
is more a process of inductive inference, proceeding not from a few
general propositions but from many particulars. Mere verbal ex-
position is then less likely to ensure against fallacy and non-sequitur.
Intelligence on such occasions is well advised by Francis Bacon's
injunction that "the mind itself be from the very outset not left to
take its own course but guided at every step; and the business be done
as if by machinery." Bayes' Theorem is the kind of mechanistic aid
to the intellect that Bacon here idealized.
Using this aid, the intelligence analyst does not address himself
directly to the merits of hypotheses. His procedures for estimation
require him to postulate, not debate, the truth of opposing hypotheses.
Bayes' Theorem thus helps him get around one of his most troublesome
pitfalls-his human tendency to hold fast to his prior estimate when
uncommitted opinion would go along with a change. And it helps
spare the estimator the labor of fighting other biases besides his own.
The Reliability Problem
In the university experiments the test subjects were in no doubt
about the color of each chip they drew; nor did they have to question
Approved For Release 2005/03/15. CPA-RDP78T03194A000200050001-8
SECRET SECRET
the evidence set before them in the intelligence scenarios. The CIA
experiment, however, incorporated a probability element to reflect
the frequent uncertainties in the workaday intelligence world about
the accuracy of reports from the field. The result was a modification
of the Bayesian equation.
The modified equation was worked out by analogizing from the
poker chip experiments. Suppose that the test subject, instead of
drawing poker chips out of the box himself, turns his back and gets
his information, sometimes accurate and sometimes not, from an
assistant. Suppose also that he has some reasonable basis for estimat-
ing the probability of correct reporting, perhaps the assistant's past
record.
Call this probability of correct reporting the reliability rating.
A 30 percent reliability rating would mean that 30 percent of the
reports with such a rating are true, in the rater's opinion, and the
other 70 percent are false.
False reports are of two kinds. One is bereft of any corresponding
fact, the utter fabrication for example. Such a report would be the
assistant's announcement of a red poker chip when he had actually
picked nothing at all out of the box. If the report has a probability
of being false in this sense, the required modification of the equation
is only to make the reliability rating (r) an exponent of the likelihood
ratio:
R=PLr
The second kind of false report is one which deliberately or in-
nocently confuses one event with another, for example the assistant's
announcement of a red chip when it was in fact blue. For reports
estimated to have a probability of being false in this sense, the
required modification of the equation becomes perhaps too involved
to explain in a non-mathematical journal, but the mathematics is not
really difficult.
The problem of the reliability rating does not enter into all ap-
praisals of evidence. Reliability ratings are unimportant for much
of the evidence received through technical collection. Nor are they
necessary in intelligence appraisals of propaganda evidence, provided
the analysis turns on the reasons why statements were made rather
than on their truth or falsity. But the problem may well loom. large
in the event of garbles from technical collection and in the evaluation
of reports received from human sources; and so the analyst must
be at special pains to understand the very restricted meaning of the
 Approved For Release 2005/03/15: CVRDP78T0JT?6Jb626N50001-8
rating. It is in no way affected by the content of a report but
represents only an appraisal of source reliability, insofar as one can
be made on the basis of such considerations as the amount of cloud
cover in photography or the past record of clandestine human re-
porters. The pitfall to skirt with utmost care is the reliability rating
that is nothing better than the analyst's prejudgment about the hy-
potheses. If his P or his R, in other words, affects his r, the analyst
can find himself in a circular rut from which no mathematics can
rescue him.
In real-life intelligence analysis perhaps no analyst can altogether
separate his biases about the hypotheses from his appraisals of source
reliability. When the credibility of some item of evidence is crucial
for final conclusions, therefore, the analyst had best take a detour
around the reliability issue. A case in point is the Cuban refugee
report that alleges the sighting of strategic missiles near Havana.
The intelligence estimator examining the hypothesis of imminent
strategic missile shipments from the USSR to Cuba can hardly assign
a reliability rating to this report. If he did, he would probably be
putting into his analysis a judgment about credibility that is pre-
cisely the answer he wants to get out of his analysis.
To exclude altogether this refugee report and others like it from
his body of evidence, however, would put the estimator into the
untenable position of giving no more weight to a hundred such re-
ports than one. His recourse is to appraise such reports much as
he appraises propaganda evidence, eschewing judgment about truth
or falsity. His likelihood ratio then represents only his opinion of
how much more likely it is that unsubstantiated evidence of this sort
would appear under the hypothesis of strategic missile shipments
than under another hypothesis. This way out of the difficulty is ad-
mittedly not the most elegant of solutions, and possibilities of other
methodological options are being explored.
The Cuban Missile Estimate
One test of the CIA mathematical model, a simulation of analysis
just before the Cuban missile crisis, has been completed. Two in-
telligence exercises were simulated. One is an estimative study in
mid-September 1962, when a National Intelligence Estimate on Cuba
was in fact published. The other is an estimative review as of
three weeks later.
Approved For Release 2005/03/1
6 SECRET
The analysis sets up two mutually exclusive hypotheses. Hypothe??
sis one is that the USSR will soon ship strategic missiles (MRBM.,
IRBM, or ICBM) to Cuba. Hypothesis two is that the USSR will
not go so far as to ship strategic missiles, despite the sharp upsurge
of military aid to Havana in the summer of 1962. The task calls for
estimation of the odds favoring hypothesis one over hypothesis two.
The background of the missile crisis reaches back at least to Feb-
ruary 1960, when a visit to Cuba by Soviet First Deputy Premier
Mikoyan ended the year of Soviet reserve that followed Castro's
seizure of power. In the wake of Mikoyan's visit, several economic
assistance agreements were signed and Soviet deliveries of armaments
commenced, giving the Cubans armored, artillery, anti-aircraft, and
anti-tank capabilities appropriate for defensive and internal security
purposes. The Soviets withheld the obsolescent IL-28 jet light bomb-
ers and more advanced weapons that it was supplying to other
countries.
Up to 1962 more than 200 agent and refugee reports alleged the
presence of missiles in Cuba. Aerial photography failed to confirm
any of these reports. The Soviet Union to this point had not shipped
strategic missiles to any foreign country, Communist or non-Com-
munist.
This background information is useful only for establishing reason-
able starting odds. As of January 1962, one to ten odds are postulated
in favor of hypothesis one (in everyday parlance, ten to one against
it). The mathematical analysis then proceeds to determine and apply
likelihood ratios and reliability ratings for the evidence appearing from
January 1962 on. This process, carried out in 1967 with the 1962
evidence, produces three to one odds as of mid-September 1962
against Soviet emplacement of strategic missiles in Cuba.
The mathematical calculations of 1967 thus support the estimate
published in 1962. They also show, however, that the odds are shift-
ing rapidly in favor of the strategic missile hypothesis. The fall in
odds against the hypothesis accelerates, and by the end of the first
week in October enough new evidence is in hand to make strategic
missile emplacements an even money bet.
The Shape of Evidence
A pioneering experiment is often as interesting for the problems
encountered as for the results achieved. The principal technical
problem encountered in this test trial with Bayesian method was the
 Approved For Release br03/15 : CIA
identification of units of evidence. In a poker chip experiment there
is no doubt about the unit of evidence; it is the drawing of a poker
chip of a particular color. The intelligence analyst, however, receives
many reports of events. Can he make each report rather than each
event his unit of evidence?
The answer is no; at least it is negative for the mathematical model
used in the Cuba test. Other models may be developed, but this
particular one can tell only the significance that events, not reports,
have for hypotheses. To take reports as units of evidence would
overweight events on which volume of reporting is high and under-
weight possibly more significant events on which it is low.
Several reports about the same event are therefore treated in effect
as one, and volume of reporting is reflected only in the analyst's re-
liability ratings. These ratings represent the probability in the
analyst's mind, in the light of all the reports available to him, that his
evidence is accurate.
But an event, like an atom, is made up of smaller particles, and
the analyst needs to have a working rule of reason to guide him in
his segmentation of the evidence. The rule is to combine items of
evidence so clearly associated in content that separate appraisals would
virtually be double counting. Successive photography showing prog-
ress in the construction of a surface-to-air missile site can be taken
as a single unit of evidence on the operational status of the site.
Broadcasts on the same propaganda theme can logically be counted
as one unit of evidence rather than entered broadcast by broadcast
into the mathematical processing. The following two extracts from
the simulated Cuba analysis illustrate the burden on the analyst to
combine his evidence fairly. The italicized head names a unit of
evidence; the relevant reports are then described; the unit is ap-
praised and given a likelihood ratio, an estimate of how much more
(or less) likely it is that the unit would appear if hypothesis one
(strategic missiles) is true than if hypothesis two (no strategic mis-
siles), is right.
Cuban-Soviet Frictions: On 26 March, veteran Cuban Communist Anibal
Escalante was ousted from party leadership. Soviet press commentary
in April endorsed the removal of Escalante but also called for an end to
divisions among Cuban revolutionaries. The commentary emphasized the
virtues of collective leadership. The intimation of the commentary was
that the USSR was disturbed by the setback suffered by its proteges in
Havana.
Approved For Release 2005/03/15
8 SECRET
Bayes' Theorem SECRET
DP78T03194A000200050001-8
A June report from clandestine services, originating with a usually
reliable Paris source, has Castro saying privately that he wanted to stay
independent of the "men of Moscow." Castro reportedly said he felt sur-
rounded by orthodox Communists who would resort to anything to obtain
control in Cuba, "even a temporary arrangement with Washington."
Fidel Castro's brother Raul, deputy premier and minister of armed
forces, arrived in Moscow on 2 July. He was met at the airport by
Marshal Malinovsky, the Soviet defense minister. Raul departed on 17
July without fanfare or final communique. This lack of red-carpet fare-
well suggested he did not get what he wanted out of the Soviets.
Simulated Mid-September 1962 Appraisal: These indications of fric-
tions hardly put Cuba in the character of the most reliable of Soviet
allies. The frictions are evaluated as unlikely, given the assumption that
the Soviets are about to ship strategic missiles to Cuba. On the other
hand, the frictions seem little more likely under an alternative hypothesis
that assumed sharply expanded military aid of any other sort. The evi-
dence, therefore, carries only slight diagnostic value for contradicting the
hypothesis of imminent strategic missile shipments to Cuba.
Likelihood Ratio: 1 to 1.2
Reliability Rating: .8
Hints of New Cuban Capabilities: A clandestine services report in July,
sourced to a fairly reliable Cuban businessman with good contacts: among
Castro adherents, described Cuban naval officers as pessimistic about
Cuban capabilities to resist a new invasion. Cuban army officers were said
to agree but to feel that the principal danger would be over by September.
In another report, a knowledgeable Cuban was quoted as saying that
the US was afraid to interfere with Soviet-flag vessels but "in September
the Americans will also respect the Cuban flag."
At one point, the Cuban (Che Guevara according to one account) re-
ferred to the NATO nations as a belt of bases surrounding the Soviet
Union. He was reportedly "livid" as he added that "in September Cuba is
going to be the buckle in this belt."
Simulated Mid-September 1962 Appraisal: The allusion to NATO bases
suggests a development consistent with the assumption of strategic missile
installations in Cuba. The allusion could also have been expressed, al-
though less probably, given the assumption of expanded military aid to
Cuba that stopped short of strategic missile emplacements.
The accuracy of the reports is open to question.
Likelihood Ratio: 1.5 to 1
Reliability Rating: .5
As these two extracts indicate, the telescoping of reports sharply
reduces the number of units of evidence available for mathematical
processing. The reduction gravely complicates the analyst's task.
The reason is that Bayesian analysis takes off from starting odds which
#RDP78T03194A000200050001-8
SECRET 9
 Approved ForpI elease 2005/03/15 : G 1-RDP78T0A4 OO MM0001-8
may be more intuitive than grounded in evidence. If many units of poring over details of evidence and for the resulting higher level of
evidence are available, these should in time outweigh the influence explicitness in his working materials. Should the tabulation of relevant
of the starting odds. The rub comes when there are not many units evidence be embarrassingly short, both analyst and reader are alerted to
of evidence. The prospect is then that starting odds rather than the weakness of the evidential base and to the pivotal position of a
evidence will constitute the predominating influence on the final odds. priori judgment in the estimate.
The Cuba test suggests that this problem will bedevil intelligence To argue for evidence, however, is to knock on an open door. Every-
more often than not. Intelligence collection during the Cuban mili- one would like to appeal to the verdict of evidence. The deep skepti-
tary buildup was massive, but the evidence touched on comparatively cisms are not about the virtues of evidence but about the practicality
few subjects. The opportunities to increase or reduce the starting of representing evidence with mathematical precision. It is one thing
odds of ten to one against the strategic missile hypothesis did not, to work with probabilities of drawing a red poker chip from a box
therefore, come thick and fast, and an analyst would want to offer with a given color mix of chips. Is it not quite another thing to work
his choice of hypothesis with considerable reserve. Perhaps the best with likelihood ratios and reliability ratings that are personal opinions
he could do would be to say how much the evidence had shifted the about the probabilities? The underlying data in the one case are nu-
odds since the starting date of his analysis. While this interpreta- merical counts, and all the experts are agreed on the rules for assigning
tion might not justify confident predictions, it could alert policy- probability values to such data. In the other case, the probabilities
makers to the implications of recent developments. are subjective judgments and tentative besides. If the intelligence
analyst says that an event is twice as likely to happen if one hypothesis
Critique is true than if another hypothesis is true, does he really want that figure
Working with the Bayesian model, intelligence is not a blend of to be taken literally? And if he says the chances are only four out of
deduction, insight, and inference from the body of evidence as a whole. ten that a source is reporting accurately, does he want precisely this
It is a sequence of explicit judgments on discrete units of evidence. opinion about the source and no other to count in the basis of his
Bayesian analysis can carry conviction only if the evidence itself per- final conclusions?
suades. The analysis cannot apply the additional dialectic leverage The question is almost its own answer. The likelihood ratios and
of well-reasoned generalization cast in finely finished phrase. reliability ratings do no more than suggest roughly how the analyst is
This necessity to work with a hard base of evidence limits the weighing evidence in his own mind. Mathematical processing :in real..
prospective usefulness of Bayesian method. Current evidence in many life intelligence analysis ought not, therefore, to restrict itself to one
situations carries little weight for longer-term estimation. Even for set of likelihood ratios and reliability ratings. It should rather involve
short-term prediction, the base of available evidence may be too small several passes over the evidence with different sets of figures.
a foundation to support by itself the estimative structure that intelli- The processing would thus show the sensitivity of final conclusions
gence must often put together for the high councils of government.
A forecast of foreign reaction to a postulated course of U.S. action prob- to variation in appraisals of the evidence. Suppose one or two mixes
ably has some evidence to go on but not much, at least not until the of likelihood ratios and reliability ratings led to a conclusion that con-
United States gets nearer the decision to take the postulated action. I tradicted those given by the other passes over the evidence or that
Would it be worth while, then, for estimates of the future to include contradicted the intelligence consensus reached by conventional analy
interpretive tabulation of all units of evidence as in the Cuban sis. It should then be incumbent on the analysts to determine the
such an simulation? There is much to be said for requiring such a tabulation
reason for this contradictory conclusion. They might decide in the
7P7 I
in all cases. Bayesian method is helpful not only for its rules to assure end to rule against it on the ground that it was based on unreasonable
valid induction but also for its duress on the analyst to separate fact weighting of the evidence. But if they felt the weighting was not
from opinion. Even if the analyst does not follow through with math- beyond the bounds of reason, they might decide to rethink the whole
ematical processing, his analysis should be the better for his labor in subject.
Approved For Release 2005/03/15 -CI = -RDP78T03194A000200050001-8
10 SECRET SECRET
11
 oa es ? neorem
Approved For Release 2005/03/15
Mathematical processing will not become an alternative to present
methods of intelligence analysis. It will become a reliability check
on present methods. It will help show the plausibility of conclusions
which the intelligence analyst would not otherwise recognize as com-
patible with the evidence and his own inner logic. It will tell the
analyst: if you interpret the evidence in this way, then here is the con-
clusion you should probably reach. Often the mathematics will be
persuasive.
Traces of the borrowed German sci-
entists combine with other scraps of
information to throw light on the
USSR's early atomic program.
SECRET
No Foreign Disseirn
ON THE SOVIET NUCLEAR SCENT
Henry S. Lowenhaupt
As World War II in Europe ended, the German nuclear scientists,
handicapped by insufficient coordination and paltry official backing,
were nevertheless only just short of achieving a self-sustaining chain
reaction in a heavy-water-moderated pile. They had elaborated most
aspects of reactor theory; they knew the best arrangement for the
lattice of fuel elements; they had gained experience in the production
and casting of metallic uranium. They had prepared detailed designs
for two pilot plants for the industrial production of heavy water. They
had also experimented with several methods of isotope separation for
concentrating the fissile U-235, especially the gas centrifuge method,
though none of these had by any means reached the production stage.
In short, they had a body of know-how, experimental machines, and
basic materials unique outside the United States and Britain.
U.S. and UK forces moved aggressively to prevent the proliferation
of this nucleus of nuclear capability. They promptly seized the sci-
entists and materials in their own zones of occupation and snatched
some from the agreed zones of France and the USSR ahead of their
advancing armies. They even destroyed by air attack the Auer Com-
pany plant, in the prospective Soviet zone, that had produced the
uranium metal for the German program. They interned near London
the ten ranking scientists, led by Professors Otto Hahn and Werner
Heisenberg, most directly concerned with the program, and only after
Hiroshima did they release them under such conditions that they would
not want to go to the USSR.1
Scientists Eastbound
Yet the sweep could not be clean. In June 1945 British intelligence
reported that Dr. Nicolaus Riehl of the Auer Company had left Ger-
Approved For Release 2005/03/15
SECRET
'The story of the German effort and its denouement is well told in David
Irving's The Virus Home (London, 1967), reviewed on page 103 of this issue.
-RDP78TO3194A000200050001-8 MORI/HRP PAGES
SECRET 13-29 13
 SECRET v r Soviet Nuclear Start SECRET
Approved 'or U9ds~~269103/15 : Cf RDP78TO3194A000200050001-8
many for the USSR along with
six others who had worked with
him on the manufacture of ura-
nium metal. 'Chen four days after
Hiroshima word came from Lon-
don that Professor Gustav Hertz
had flown to Moscow four weeks
previously and Professor Adolf
Thiessen was in a Soviet camp
with eighteen fellow workers
awaiting transportation to Russia.
Both Hertz and Thiessen, though
not immediately involved in the
German atomic program, were
prominent and technically com-
petent scientists who could com-
mand the loyalty of other scien-
tists. Hertz, a Nobel Prize win-
ner in atomic physics, had been
chief of the famous Siemens-Ilalske Laboratories since 1934 and had
discovered the gaseous diffusion method of separating isotopes.
Thiessen had directed the Kaiser Wilhelm Institute for Physical Chem-
istry and had published an impressive string of important research
papers.
From this point U.S. and UK intelligence had the task of trying to
follow the incipient Soviet atomic effort, and it was largely the early
results of this pursuit, as decribed below, that encouraged the U.S.
Air Force to mount a watch for the first Soviet test explosion two years
before it was expected.2 G-2, OSS, and their British counterparts,
under the direction of the two nations' atomic authorities,3 began with
a vigorous campaign to discover which Germans had been recruited
for this effort and which Russians were doing the recruiting. The
See Northrup and Rock, "The Detection of Joe I," Studies X 4, p. 23 if.
e The intelligence analysis and the general direction of the collection effort in
the nuclear field were vested, on the U.S. side, in General Groves' "Manhattan
Engineering District" until its dissolution in January 1947, when these functions
passed to CIA. In Britain they were performed through 1952 by it section of the
Ministry of Supply and, after its formation, the British Atomic Energy Authority.
The Supply section was staffed in part by Secret Intelligence Service officers
under the leadership of Lt. Comdr. Eric Welsh. See The Virus House, cited in
footnote 1 above, for Welsh's role in atomic intelligence to the end of 1945.
task wa; complicated by the fact that the Russians were recruiting
German and Austrian scientists and technicians for all sores of pro-
grains; th,- numbers ran to many hundreds. By the end of the year,
however, it was clear that for atomic work well over a hundred tech-
nicians v--re being grouped around a few rather good scientists. as
leaders.
In addition to Riehl, Hertz, and Thiessen, the group leaders inclu dr'd:
Baron Manfred von Ardenne, Germany's foremost cyclotron constrtic,-
tor; Pronesor Max Vollmer, an outstanding physical c:11ucunist; .und
Dr. Ilan,; Born of the Kaiser-Wilhelm Institute for Brain Research, who
had been working on the biophysics of radiation. As for the Russian
recruiters: at Leipzig there was a General "Katchkatchian' aided a
Major "Yr:issin"; a Colonel "K. K." Kikoin at Karlshorst had hersu,ic.ed
Hertz to go; and a Lt. Colonel "Kargin" had handled negotiations with
Vollmer. A General Ivanov," who had had to do with recruitment
in Vicnn i, turned out to be none other than General Meshik, 1,avnauliy
Beriya's right-hand man .4
Many :tf the German scientists were well enough know; that tlu:!ir
specialtiu s and skill could be assessed. The intelligence reporting also
tended to sort them into groups under the respective leaders. But [his
did not toll its what each group was to work on in the USSR ;.nd where
they wer ' to do the work; and that was what we needed to know.
Russian security was initially well below its subsequent standards.
By February 1946 the Strategic Services Unit, successor to OSS, was
able to retiort from an agent in the East Zone of Gcrntany that
Baron von Ardenne's presumably cyclotron-centered group -,vent to t]a'
Crimea in the summer of 1945 and then in October was established
in one of the small communities between Anaklia and Poti on the c,;st
shore of the Black Sea, about 120 kilometers north of the Turkish
border. Another agent reported that Thiessen, IIertz, and Vollmer,
as well a;, Von Ardenne, were on this stretch of the Black Sea coast
between `yukhumi and Poti--in ancient Colchis, Where the A;-gon:nuls
found the (;olden Fleece. They had reportedly not done any work
up to the beginning of November 1945, as housing and hthoratorics
were still under construction. The biophysicists under Born, as i,ve;ll
as Riehl's Ancr Company group, were left unaccounted for.
' P. Ya. 11 shik was executed on 23 December 1953. As Minister of Intern i1
Affairs of Vic Ukrainian SSR, lie was charged with being an active participant
in the coup attempted by Beriya.
Approved For Release 2005/03/15 : CIA-RDP78TO3194A000200050001-8
14 SECRET SECRET 1,7
 Novi t Nuclear
Approved For Release 2 ` %t03/15 : CI
The Russians rounded out their atomic recruitment early in 1946
by assembling a group of German scientists under Dr. Heinz Pose,
who had worked on nuclear reactor physics at Ronneburg under the
German Bureau of Standards. This particular group had been con-
sidered inferior by their more renowned fellows, but in fact they had
shown Heisenberg an error in his calculations and thus put the pro-
gram on the right track towards a working reactor. We had no infor-
mation on where the Russians stationed these reactor specialists.
Letters and Defectors
At about this time U.S. and UK intelligence stumbled onto the inter-
ception of letters from the expatriated scientists as a source of informa-
tion about their locations and activities which in the end proved far
more fruitful than the alternative of penetrating institutes in East
Germany. An intercepted letter dated 18 March 1946 from Hertz to
his son disclosed the identity of the Russian go-between in Germany as
Lt. Colonel "Cedenko," 46 Wassersportallee, Berlin-Gruenau. Then
in August and September there was a change in Russian personnel
and their address, for Lt. Colonel "Yelan" and Lt. Petrochenko at
Buntzelstrasse 11, Gruenau, were handling the mail.
In October Riehl wrote from Elektrostal-a small town about 60
kilometers east of Moscow. Later his location there was confirmed by
a March 1947 letter postmarked Moscow from Mrs. Blobel, his secre-
tary, which indicated that biophysicists Born and Karl Zimmer, as well
as the Auer Company people, were living 60 kilometers from Moscow.
The implication was that the processing of uranium ore and the study
of biological effects were being organized in or near Elektrostal while
theoretical and experimental work was going on down by the Black Sea.
The Russians had always maintained a security wall between them-
selves and the East Germans; but after four German atomic scientists
who had been to the USSR for job interviews returned to East Ger-
many and defected to the West in early 1947 .the rules were tightened
tip. From then on no East German was ever told anything about
German atomic scientists in Russia. All letters from the scientists were
strictly censored and bore without exception the return address Post
Box 1037P Main Post Office, Moscow.
The Russian assessment was correct: these defectors did possess
information of value to us. For instance, Dr. Adolf Krebs had first
had interviews in Germany with Colonel Professor "Alexandrow" and
a Professor Leipunski. The former was clearly Professor Simon Peter
Approved For Release 2005/03/15 : Cl
SECRET
DP78T0319AM101Y0`A&M-0118
Alexandrov, who represented the USSR at the Bikini "Crossroads" tests
in 1946 and in UN discussions on atomic energy in 1947, the latter
presumably was A. I. Leipunski, a well-known Russian nuclear physi
cist. When Krebs was then flown to Moscow (without his consent)
he learned that the German groups worked as an independent organi-
zation under the supervision of
General "Sawiniaki," whose staff
of several generals included a
General "Krawtschenko." Un-
garbled, the boss must be Colonel
General Avram Pavlovich Zaven-
yagin, the builder of Magnito-
gorsk in the Urals and the Norilsk
Nickel Combine in far northern
Siberia; he was reportedly head
of the secret Ninth Chief Direc-
torate of the MVD and had a
General Kravchenko as assistant.
Thus the MVD continued into
1947 to play a significant role in
the Soviet atomic energy pro-
gram, even though this had been
reorganized in late 1945 as the
First Chief Directorate attached
to the Council of Ministers under Colonel General Boris Lvov:ich Van-
nikov, who had managed Russia's munitions production during the wa.r.
Krebs also reported: that the Hertz group was working on isotope
separation problems at Sukhumi; that the Von Ardenne and Thiessen
groups were also there, as we had thought; that Dr. Vollmer and sev-
eral assistants were working at Sukhumi on heavy water production
methods, that Dr. Riehl and his group at Elektrostal were turning out
uranium metal on a production scale; and that Dr. Patzschke, a
former director of the Joachimsthal uranium mine in Czechoslovakia,,
was head of a group prospecting for uranium ore near Tashkent in
Central Asia. The Pose group was presumably somewhere east of
the Urals, since in May and June of 1.945 this territory had been
surveyed, Krebs had heard, as to its suitability for their reactor work.
The news that the Vollmer group was working on heavy water came
as a surprise: by this time it was known that a group of Germans
under I)r. P. Herold from the former I. G. Farhen Leona ? iant
 Soviet Nuclear Start
Approved For Release 2005/03/15 : di
Merseberg in East Germany were continuing their wartime research
on methods for the industrial production of heavy water at the Karpov
Institute in Moscow. But the Leuna group was administered quite
separately from the Post Box 1037P groups, presumably because the
Karpov Institute was a research and design facility of the Ministry
of Chemical Industry, while the 1037P scientists were administratively
under the MVD.
Uranium Production; Isotopes
The year 1947 brought the first real confirmation of the thin infor-
mation we had about the manufacture of uranium. The UK had
managed to learn in 1946 that one ten-ton freight car of uranium ore
was being consigned from the Jachymov (Joachimsthal) area of
Czechoslovakia to Elektrostal every ten days. The UK had also learned
that the Russians were requiring the former Bitterfeld plant of I. G.
Farben to set up the production of highly pure metallic calcium at
30 tons per month, enough for the manufacture (by oxide reduction)
of 60 tons of uranium metal. Penetration sources had furnished the
specifications on the amounts of impurities allowable in the calcium;
these conclusively indicated that it was for atomic use somewhere.
It remained for the covert collection arm of CIA to acquire a bill
of lading for three freight-car loads of calcium from Bitterfeld con-
signed to Post Box 3 Elektrostal, Moscow Oblast. This proved beyond
question that at Elektrostal there was a uranium factory making the
metal in quantity, using methods worked out at least in part by the
Auer group under Riehl. Indeed it also forced the conclusion that
the Russians were at least attempting to build somewhere a large
reactor to produce plutonium for nuclear weapons.
Shadowing the German scientists in Russia, largely through mail
intercept, had thus produced information which could form the basis
for detailed debriefing when one of them came to the West, while
penetration attempts had run squarely into Russian security. It was
decided to make thorough preparations, mainly by mail analysis, for
the day when the nuclear scientists might return to an area from which
they could be defected, even though that day might be years away.
Later, in 1951, this concept was extended to all German scientists- in
the USSR under a program called Operation Dragon The work
-RDP78T03194A000`2D6b&&fff-8
cepted most of the letters while the detailed collation of the data
was performed by UK analysts.
Meanwhile atomic collection proceeded on a broad front. In 1918
former prisoners of war began to return from the USSR to West Ger-
many, and it was soon learned that a number of them had helped
construct two institutes in the Sukhumi area, one under Professor
Hertz near the village of Agudzeri, the other near that of Sinop, name-
sake of the Turkish city. The year 1949, if it surprised us with the
Soviets' first atomic test, showing that their plutonium production was
much farther along than we had suspected, also brought the first of
two Russian defections which helped the analytical picture immensely.
The first defector was a scientist nicknamed "Gong" who had worked
in 1947 at the Institute of General and Inorganic Chemistry under a
Professor Dmitriy A. Petrov on a way to make porous metal mem-
branes for the separation of uranium isotopes by gaseous diffusion. A
prize of 100,000 rubles had been promised for the correct solution of
this problem. In the course of his work Gong had that summer visited
Special Laboratory No. 3, located in west Moscow. Here lie had
spoken to Professor Isaac Konstantinovich Kikoin, Deputy Director of
the Laboratory and a corresponding member of the Academy of Sci-
ences. Gong was positive that Special Laboratory No. 3 worked on
the separation of isotopes by the diffusion method and on other
physical-chemical processes. He had also heard of a Special Labora-
tory No. 1, location not known to him, and of Special Laboratory No. 2,
under the direction of Academician Alikhanov in Moscow. All three
Special Laboratories were intimately tied to the First Chief Directorate
with respect to work priorities, supplies, security, etc.
Thus it became clear that the Colonel "K. K." Kikoin who in 1945
had recruited Hertz for work on isotope separation methods was the
person responsible in Moscow for gaseous diffusion research for the
Soviet atomic energy program.
Research papers had been published by Gong's boss, Professor
Petrov, in 1947 5 and 1948 on the subject of "skeleton catalysts." The
method of preparing these catalysts was just that reported by Gong for
barrier membranes. Interestingly enough, moreover, the pores in the
"catalysts" were of a size reasonably correct for a membrane to sepa-
rate out U-235 by gaseous diffusion.
settled into a routine in which the Army Security Agency inter 6 "Investigation of the Structure of the Copper Skeleton
(U.S.) y y Kefeli and S. L. LeIchuk; Dok. An. 57, No. 6, 1947.
Approved For Release 2005/03/15 : CIA-RDP78T03194A000200050001-8
SECRET SECRET
 Soviet Nuclear Start
Approved For Release 2005/03/15 :
Procurement Abroad
In 1950 the second Russian defector, "Icarus," proved of even more
value. As a Colonel of the MVD concerned with supplies, fir.t in the
Moscow office of the First Chief Directorate and later at Wismut AG
in Saxony,6 he knew personally many of the Russians involved in
the atomic energy program in Moscow and in Berlin. He was aware
that General Meshik was in charge of personnel and security for the
whole program. He knew that Lt. Colonel (fnu) Sidenko (the
"Cedcnko" who handled the letters intercepted in early 1946) had
been the representative of the Ninth Directorate of the MVD in Berlin
in 1945 and that he had been replaced (by August 1946, our intercepts
had shown) by Lt. Colonel Elyan (not Yelan, as we had it), who
eventually had returned to Moscow to work for the First Chief Direc-
torate under one Dorofeyev, chief of its Supply Directorate. Icarus
also reported that a man named Panin ran a warehouse under
Dorofeyev known as Post Box 200, Moscow.
Now that we had the correct Russian spelling of the names of the
atomic representatives in Berlin, as well as their addresses, it seemed
useful to investigate their activities in depth. It soon developed that
the Berlin atomic office was always in two sections at separate locations:
one handled mail, packages, etc., for the German scientists; the other
was concerned with special procurement for the Soviet program.
Both sections changed personnel and location approximately every
year and a half. Through some rather clever intelligence work
against these offices, CIA covert collection was to show in 1952-1953
that they expedited the procurement of several million square feet
of very fine nickel wire mesh per year and that at least one shipment
of this mesh was flown from Tewa/Neustadt to Panin's warehouse at
Post Box 200, Moscow. This clearly established by administrative
procedures that the ultimate user was the Soviet atomic program.
The technical specifications and amounts of the mesh suggested porous
barrier for U-235 separation as the only possible use in an atomic
program.
An attempt to learn whether the Bitterfeld plant shipped other
atomic materials than the calcium revealed that all shipments now
bore only the Moscow address of the main offices of cusimz, the Chief
Directorate of Soviet Property Abroad of the Ministry of Foreign
Wismut AG (Bismuth Inc.) was the cover name for the vast Soviet-run
uranium mining operation in East Germany.
Approved For Release 2005/03/
SECRET
RDP78T031M600'01608 4-8
Trade. All carried nine- and twelve-digit order numbers and five-
digit transport numbers. Surely numbers as complicated as these
should have character in the cryptographic sense.
For background purposes, we studied documentation on equipment
ordered by the Soviet commercial mission, Amtorg, in New York.
Unfortunately, by about the time we understood the ordering system
the Russians decided to tighten it up, so that this work was nullified.
However, the reporting on Amtorg (by the CIA domestic collection
organization) showed that a P. M. Sidenko had had a tour of duty
with the mission between December 1946 and June 1948. This man,
presumably the same Lt. Colonel Sidenko who was at the Berlin
atomic office in 1945 and through July 1946, arrived in the United
States during the same month that brought the departure of Anai:oli.
Yakovlev, head of the atomic espionage chain involving Harry Gold
and Klaus Fuchs.
Others working with Sidenko on procurement were soon identified:
Nikolai L. Artemiev, who visited a plant making geiger counters in
November 1946 and who tried in June 1947 to purchase helium leak.-
detectors used in U.S. U-235 plants; Nikolai S. Sventitsky, co-author
of an article on spectroscopy, Artemiev's replacement; and N. N.
Izvekov, who was interested in all sorts of manufactures, from heavy
construction machinery to fine-woven wire mesh "for electronic equip-
ment." Some three million dollars worth of goods purchased by the
Sidenko group was identified as apparently for the Soviet atomic
program; it included the machinery for a complete plant for extract-
ing radium from uranium ore wastes. Sventitsky joined Artemiev in
London in January 1948 when Sidenko returned to Russia.
Into the Fifties
With respect to the German atomic scientists in Russia the early
1950's was a period of continued information collection and analytical
consolidation. Letter intercepts by the hundreds were collected and
results collated. Not only the main groupings but interrelationships
within groups were studied, with a view to the eventual recruitment of
adequate representatives of each group when they were allowed. to
return to Germany. In trying to determine who was in the Von
Ardenne group at Sukhumi, for instance, it was noted that letters (all
severely censored and postmarked Post Box 1037P Moscow) mention-
ing the accidental death of a small child, from playing with matches,
came from Becker, Felicitas Jahn, D. Lehmann, Gerhard Mueller
,
CIA-RDP78T03194A000200050001-8
SECRET
 SECRET Soviet Nuclear Start
Approved For Release 2005/03/15 :
Liselotte Steenbeck, Fran Wittstadt, and Dr. Froelich; that an out-
break of scarlet fever was referred to by Felicitas Jahn, Liselotte
Steenbeck and Frau Schrottke; that "on Saturday the Bernhardts
visited the Schrottkes and on Sunday the Schrottkes visited the
Bernhardts"; that the "bull in a china shop" complained about by
Bergengruen was identified by Felicitas Jahn as Helmut Hepp.
Such studies had resulted in the identification of the seven distinct
groups. The Hertz group was still located in the Sukhumi area, by
the town of Agudzeri. The Von Ardenne and Thiessen groups were
together at Sinop in the same area. Vollmer's group, no longer with
Hertz's, had moved to Moscow, and a POW returnee who had been
used in the electronics program confirmed that it was working on
heavy water production processes. The Riehl uranium specialists
continued at Elektrostal, and Riehl had been awarded a Stalin Prize
and made a Hero of the Soviet Union after the success of the test
explosion he had helped make possible in August 1949. The loca-
tion of Pose's reactor group posed a problem; likewise that of the
biophysicists under Dr. Born, for they had left Elektrostal in 1948.
Some rather clever analysis by the Directorate of Scientific Intel-
ligence in the UK in 1951 succeeded in narrowing down the location
of the Born group to within 20 miles of the town of Kyshtym in the
southern Urals. The Kyshtym area was the site of the nuclear re-
actor which had made the plutonium for the first Soviet atomic device,
and the placing of a biophysics group near a reactor site made good
The British detection was done as follows.
The letters from the Born group described topography, scenery,
weather, and temperatures strongly suggesting the hilly country of
the Urals. In fact, the heavily censored letters spent so much time
on the weather that it was decided to see what could be done with
this information. So the weather as described by known members of
the group on a given day was compared with weather charts of the
USSR for that day, and the irregular portions of the USSR having
such weather were highlighted. Once some dozen of these weather
overlays had been compiled, it was clear that only one area was com-
mon to them all. This was a stretch of the Urals some 100 to 200
miles north and south of Sverdlovsk, with a very slight balance of
probability toward the north.
Approved
-RDP78T031S64'A&ffi '5b t1-8
Now an analysis was made of a train trip from Sukhumi to the
Born group which a man named Rintelen reported in an intercepted
letter: "After the first long train journey, we had an opportunity on
the 10th of December from morning till evening to buy warm
clothes, travel by underground and bus and to sit in good cafes... .
In the evening we traveled on again and arrived on the 12th of De-
cember in the next large town from here [i.e. from the location of
the Born group]. The following evening we traveled a further five
hours by train and on the 14th of December we arrived here after
a two-hour bus journey . . .
Rintelen's pleasant stop on the 10th of December must have been
in Moscow, for it alone of Soviet cities possessed an underground
railroad at that time. There were three trains leaving Moscow on
the evening of the 10th for Perm (then Molotov), a likely "large
town" on the Moscow side of the Urals. Two were scheduled to
arrive early in the morning and one in the evening of the 12th. Why
would Rintelen lay over a day in Perm? An evening train heading
for the north Urals left there at 1620 on the 12th, arriving at Kizel
the five prescribed hours later, so if Kizel had been his destination
he should have taken it. Similarly he would not have had to lay
over had he been going to the eastern side of the Urals north of
Sverdlovsk, say to the Nizhniy Tagil area, for he would have taken
from Moscow one of the two trains that get to Perm in the morning
so as to catch the 1150 for Nizhniy Tagil and arrive there near mid-
night on the same day, the 12th. Thus the north Urals did not appear
a likely destination, and the "large town" of the layover must therefore
be Sverdlovsk (Chelyabinsk lying outside the area defined by the
weather information).
The three trains leaving Moscow on the evening of the 1.0th were
scheduled to reach Sverdlovsk on the evening of the 12th (at ;[520,
1609, and 1702 respectively). Five trains left Sverdlovsk for various
destinations after 1800, so all these appeared unlikely to have beer
Rintelen's. The two trains per day to Kamyshlov, five hours away,
left Sverdlovsk at 1300 and 1525 and so would probably have re-
quired a layover, but Kamyshlov, well east of even the foothills of the
Urals, was quite unlikely on geographic grounds. One last midafter-
noon train, however, left Sverdlovsk southbound at 1420 and five
hours later arrived at Kyshtym. Rintelen would have had to stay
in Sverdlovsk overnight to catch this, and in midwinter at latitude
56?N, the ride from 2:30 to 7:30 p.m. might well have seemed to be
For Release 2005/03/4 CIA-RDP78T03194A000200050001-8
SECRET SECRET
 S vieL clear ~t
Approved I-or Release ZOd9/03/15: CIA
an evening one. Thus by elimination his destination, and the posi-
tion of the Born group, lay some 20-30 miles (the two-hour bus ride)
from Kyshtym in the south-central Urals.
The Reactor Specialists
The Pose group was located in a similar manner. Evidence from
intercepted letters had put it a three-hour bus ride from Moscow..
't'hus it was not now in the Urals, as Krebs had guessed in 1947..
Several U.S. analysts, studying the intercepted mail, gleaned the
additional information that it was two and a half hours by train
from Moscow, that the members had good swimming in a river, and
that there was a great deal of building activity in new suburbs around
them. After a study of maps and railroad timetables the Maloyaro-
slavets area southwest of the capital was suggested as a possibility.
(1K analysts, spurred by this hypothesis, surveyed their much larger
volume of intercept and were able to add that (a) the return trains
from Moscow did not "fit well," (h) there was a local market-town a
lhalf-hour bus ride away, and (c) the "nearest big hospital was 15
km. away."
Railroad timetables showed that Ohnino station, 15 km. northeast
of Maloyaroslavets, was 2 hours and 30 minutes by train from Moscow.
That was also where the road and the railroad crossed the Protva
river on the way to Maloyaroslavets. The train took 16 minutes to get
from Ohnino station to this good-sized town; a bus would probably
take half an hour. The only morning train from Obnino to Moscow
left at 0750. The possible return trains left Moscow at 1300, 1440,
:rnd 1630, giving scarcely more time there than the round trip con-
sumed and so not "fitting well." Some ten other localities were two
and a half hours by train from Moscow, but few were near rivers which
might have good swimming. Of those that were, several, like
Mozhaysk, were large towns in themselves; others had excellent evening
train service. Ohnino station thus remained the only likely place.
In August 1953 attach( photographs from the railroad looking north-
west from the bridge over the Provta river showed several large build-
ings under construction and a completed large stack with blower house
such as is usually required for a nuclear reactor. Photointerpretive
measurement done by comparison of these with wartime German
aerial photography showed that the stack was almost 210 feet high.
Olhrnino was thus the location of a probable nuclear establishment
containing a reactor. '1 'he Pose group moved to Sukhumi in 1952.
Approved For Release 2005/0311$
RDP78T034A4-6&16t5ob11-8
In 1954 tht, ;ussians publicized the initial operation of the first atonuii