THE MANNER OF PRODUCTION OF SOVIET MILITARY ESTIMATES

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP79R00967A001400030026-6
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 23
• Document Creation Date: December 20, 2016
• Document Release Date: November 13, 2006
• Sequence Number: 26
• Publication Date: March 5, 1971
• Content Type: MF

Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
S-E- C-R-E-T
SENSITIVE
C E N T R A L I N T E L L I G E N C E A G E N C Y
OFFICE OF NATIONAL ESTIMATES
DRAFT 5 March 1971
MEMORANDUM FOR: The Director, Office of National Estimates
SUBJECT : The Manner of Production of Soviet Military
Estimates
1. This memorandum gives my view of the advantages and
disadvantages of the task force method of writing the military
NIEs that was tried this year, and makes recommendations for
the writing of these estimates in the future.
2. This evaluation is complicated by the fact that two
things were changed this year -- the length and detail in the
NIE, as well as the method of drafting, and two things changed
about the method of drafting -- it was taken from the ONE
Staff and given to the DDT/DDS&T, and the latter created a
task force for the production of the draft, The following
evaluation recommends that next year's NIE also have more
detail, and address the method of drafting in this light.
3. The general recommendation and three approaches to
carrying it out are given on pages 14 to 23.
..e, sic Staff
GROUP 1
Excluded from automatic
downgrading and
S-E-C-R-E.T declassification
~SENSITIVS~ ..I1'.
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
S-E-C -R-E.-T
SENSITIVE
TI MANVER AND PRODUCTION OF SOVIET MILITARY ESTIMATES
1. The manner of putting out the Soviet military estimates
this year resulted in a generally good, but not outstanding,
product.
A. The a roach of the estimate was generalwell
received by the consumer, so far as can be determined.
1. NIE 11-8-70 apparently has been well received by
the NSC Staff. It lays out the evidences the line of
analysis, the conclusions, and the alternative views in
some detail in a satisfactory manner. The length of the
estimate has not been considered too great.
2. The concept of an explicit estimate laying things
out in detail is accepted as good even by those who dis-
agree with some of the aspects of it. For instance, DIA
objects to the alternative numerical projections, as being
confusing to the planner, but they wish to keep the explicit
presentation of evidence and analysis.
3. The way in which alternative views are explained
at length makes a more "honest" estimate. There is no
need to waffle a position to get agreement. This is
generally agreed to be the biggest "plus" of the new approach.
S-E-C-R-E-T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
S-E?C-R-E-T
SENSITIVE
B. The estimates are not, however, as good as they could
or should be.
1. The basic text was generally written by research
specialists* who were not uniformly good writers.
2. The length of the estimates, the exigencies of
seemingly endless coordination,, and the human failure of
those who worked on the estimates to maintain the same
high level of effort at all times resulted in some un-
evenness in the final product. Consequently some parts
are unnecessarily wordy,, obscure, and repetitious.
1 There is general agreement, however, among those producing
the estimate that the production process worked slowly
and inefficiently.
A. The reason for this was initially the confusion of
how to go about the new garoach.
1. Part of the time taken in the production of the
estimates was a result of the fact that the procedures
had to be worked out by the task forces,, who were un-
familiar with producing an estimate and were not prepared
to do so.
S-E-C-R-E..T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
iwf
SAE-C-.R-E-T
SENSITIVE
2. There was also initial confusion as to how responsi-
bilities and authority would be divided between ONE and
the task force and what the organizing concepts of the
estimate should be.
B. The setting upof task forces to write the estimates
proved to be an Inefficient utilization of CIA resources.
1. The driving concept was to give the writing of
the NIE to those who dealt with the subjects on a day-to-
day basis, to the research specialists, These people
generally knew their subjects well, but they were generally
not able to express their views clearly and succinctly in
writing. Research reports go through an editorial process.
The drafts for the NIEs did not -+- and showed it.
2. The task force set up to put these pieces together
into an VIE was made up of people who were unfamiliar with
how to organize and write an VIE and how to get it through
the Board of National Estimates. The task force staffs
set up to write the estimate were no more used to the
problems, and duplicated the ONE Staff.
6-E-C-R-E-T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
SLEl CL t"EwT
SENSITIVE
3, While the task force was struggling with these
unfamiliar problems,, the ONE Staff which was used to
dealing with them was not used and was reduced in size
by three people.
4. The ONE Staff could perfectly well have written
a more detailed estimate as required; it had not done so
previously _.. not because it could not,, but because the
direction of the Board chairman was that it should not.
C. The task force this year worked best when a large
number of human resources were invested to make up for their
inexperience, It broke down when insufficient task force
resources were marshalled.
1. For HIE 11-8 a GS-17 or -18 was used full time as
a task force chairman. He had at least two peoples in
the GS-14-16 range working for him full time, Once
produced by the task force$ it took the entire time of
the Board chairman and two people from the ONE Staff to
move it through coordination. It was a good job largely
because of the ability and drive of the BNE chairman,, and
because of the large amount of manpower that went into it.
It succeeded almost in spite of the task force approachs
not because of it.
M 5 -
S,-E.. C.*RME-T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
V11 i~~
S-E-C-R-E-T
SENSITIVE
2. NIE 11-3 was moving slowly and awkwardly until the
chairman was put to work on it full time, other GS-13-
GS-15s were working on it full time, and a former ONE Staff
man was given the job of rewriting it.
3. ME 11-14 was not successfully produced by the
task force largely because there was not -- in any real
sense -?? a task force. For all intents and purposes, there
was one GS-15 and his assistant working on the estimate
for the first five months. It was clear that the task
force chairman was overburdened; he could not call upon
others in his organization because they were working full
time on producing inputs to NSSMs,
D. The task forces this year did not make best use of
the resources of the US intelligence community.
1. The task forces made very little use of the inputs
by non-CIA. components of USIB. The quality of those inputs
may not have been uniformly high, but they certainly were
underutilized.
.. 6 ..
S-E- C-R-E"?T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
i%ae %W10
5-E- CMR-E-T
2. The task forces generally did not make full use of
the USIB committees -- GMAIC, SIC., and JAEIC -- which were
set up to deal with technical issues, preferring to rely
almost exclusively on CIA. resources.
E. A task force approach has a great deal of merit in
dealing with specific technical problems, but not with an
estimate as a whole.
1. The idea of individual technical parts being
written by experts is a necessary starting point for any
estimate. In the past; GMAIC, JAEIC, and SIC wrote
contributions to those parts of the NIE for which they
were responsible.
2. A task force -- in which the persons who deal with
the subject on a day-to-day basis get together and define
the facts methods of analysis? agreed conclusions? and the
bases for differences is a very valuable and useful
forum in which analysts can mutually enlighten intelligence
problems.
3. No one person or set of persons is expert on all
points., however. Thus the best procedure for utilizing
S-E-C-R-E-T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
S-E-C.-R-E'-T
SENSITIVE
experts in several fields is to set up several task forces,
each dealing with the subject on which they are knowledge-
able. Thus a task force is appropriately set up on ICBM
characteristics and capabilities,, on ICBM deploymentp on
bomber capabilities, etc. Insofar as feasible these should
use groups already in existence on these subjects.
4. But these groups of experts need overall direction
and their end products need to be put together into a total
final product. An NIE, in the last analysis$ has to be
more than the sum of its parts, more than a number of
individual task force reports stapled together. In the
past? the guidance to the groups of experts to ensure that
the parts all contribute to the whole has been furnished
by ONE. And it was the task of ONE to make the whole more
than the sum of its parts.
F. A major reason for the long production process this
year was that the task force took an inordinately long time to
put the parts together into a whole that was acceptable to the
Board of National Estimates.
1, The task forces for NIE 11-8 and HIE 11-3 went
through three drafts before they produced one acceptable
-8-
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
S-E-CR-E-rT
SENSITIVE
to the BNE. The task force for NIE li.'l4 never did produce
an acceptable draft. The task force took three months to
do this in the case of NIE 11-8; 4 months in the case of
DIE 11-3; and tried for five months in the case of VIE 11-14.
2. Generally speaking, the product did not become
acceptable until the task force gave the drafting to a
generalist, who did not follow the information on a day-
today basis, and who was used to writing a product with
unity.. coherence, and proper emphasis.
as NIE 11-8 was largely written by the task force
chairma::% or by his staff man., or by members of the
BNE and the ONE Staff.
b, NIE 11-3 didn't get off the ground until it
was put into some shape by a former ONE Staff man who
had to be specially detailed to do it,
C. VIE 11-1+ was written by the ONE Staff,
S-E-C RRE-.T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
S-E'C- RRE-T
`SENSITIVE
III. But even if the task force were to work efficiently, there
are more fundamental systematic problems with the task
force approach used,
A. The basic problem with the manner of produciRS the
estimate was that setting up a DDIJDDS&T task force violated one
of the first principles of administration namely, that both
authority and responsibility should clearly rest in one place.
(See diagram attached.)
1. The directions of the DCI were that the Board
chairman of the estimate was to be responsible for the
estimate. In saying this., he was only restating a NSCID
which has long been agreed to and proven in practice.
The directive did not, however, define the authority of
the Board chairman,
2. The DCI directive then gave the job of producing
the estimate to the DDI and DDS&T who set up a task force
to do the actual job on each estimate. The relative
responsibilities and authority of the Board chairman and
the task force chairman were never defined. In fact,,
however, the task force chairman was responsible to, and
was paid by OSR and the MI or OSI and the DDS&T.
S-E-C-R-E,oT
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
i;wi
S-E'( C~^-7R-E-T
SENSITIVE
The chart below represents the organization of the
task force and its relation to the Board chairman..
CURRENT TASK FORCE APPROACH
DDI/DDS&T
OSR/OSI
BNE
Chairman
ONE Staff
Ad Hoc Staff TASK FORM - - GMAIC, SIC, JAEIC
to write NIE Chairman
DIA, State, NSA, AEC, Etc.
OSR
Vice Chairman
iS
OSI/FNSAC
Vice Chairman
The ONE Staff and the non-CIA USIB agencies and USIB committees
are in dashed lines, as they were never made a part of the
production process.
S-E-C-R?E-T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
)Swi *M0
S-E-C*-R-E-T
3. As a consequence the Board chairman did not have
effective authority over the task force chairman,., who was
in a position of being able to decide which instructions
he was'going to follow., and which not. This was an
impossible administrative situation.
4. The problem of administrative authority and responsi-
bility was then compounded on the staff level, The task
force chairman set up his own staff to support his activi-
ties. thereby duplicating the ONE staff, The Board chair-
man, however., was used to working with the ONE Staff and
continued to do so, This led to rather unproductive debates
over the authority and responsibilities of the respective
staffs, The debates were as often as not resolved by
having both staffs do a job, with the decision as to which
product to use put off until the product was reviewed,
B. The task force approach also confused the position of
the Director as heed of USIB with his position as head of CIA.
1, OW was created to support the Director as head of
USIB in his responsibility for national estimates? The
creation of ONE was a recognition of the validity of that
distinction. And this is why ONE is in the Office of the
Director, not under the DDI.
8.E-GAR-E-T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
S-.EM C??R-EaT
SENSITIVE
2. If one accepts the principle that the NIEs should
be produced by the DDI and the DDS&T representing CIA as
one of the member agencies of USIB, there is no logical
reason why the NIEs should not be produced by some other,
presumably coequal, member of USIB such as DIA.
3. USIB committees such as GMAIC, SIC and JAEIC were
unilaterally bypassed in the productl.on of the estimates.
The task force relied upon the CIA component in each case?
rather than upon USIB machinery as a whole.
C. The tack force em roach did not sufficiently recognize
that the production of an estimate is a job for the generalist.
1. In the final analysis, the Soviet military NIEs bad
this year to be put together by generalist ..- not a person
working on a subject on a days-to-day basis.
2. The reason for this development is that the writing
of an WE is primarily a problem of conceptualization and
expression, not a problem of research,, An I1 needs to be
written by scneone used to thinking about the whol.. problem
of the estimate, by someone used to writings not by someone
whose primary job is to do research on parts of the problem.
BE -C RRE-T
SENSITIVE
Approved For Release 2006/11/13: CIA-RDP79R00967AO01400030026-6
 Approved For Release 2006/11/13: CIA-RDP79R00967A001400030026-6
NNW
S-E~C~R-E