DISSENTING VIEWS TO THE REPORT OF THE WORKING GROUP ON MILITARY ESTIMATES

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP79R00971A000400050028-8
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 4
• Document Creation Date: December 20, 2016
• Document Release Date: October 13, 2006
• Sequence Number: 28
• Publication Date: March 26, 1970
• Content Type: MF

Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8
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
26 March 1970
MEMORANDUM FOR THE BOARD
SUBJECT: Dissenting Views to the Report of the Working
Group on Military Estimates
In my view, the proposals outlined by Mr. Clark's report
constitute an insufficient departure from present practices.
I propose a fourfold approach to the problem along the following
lines:
I. A continuing book on Soviet/EE military strengths and
capabilities should be maintained on a looseleaf basis. It
would deal with what is known and what can be estimated with a
relatively high degree of probability -- 60% or better. Principal
staffing responsibility would rest with OSR. Anyone in the
community could propose changes; these would be studied and
formulated by OSR; the ONE Board and Staff would review them;
if there was doubt or dissent, there would follow the usual
coordination process; final approval would be accorded by the
USIB (though this could be done in writing, rather than in formal
meeting).
GROUP 1
Excluded from automatic
S-E-C-R-E-T downgrading and
noouME;~ITNo. -----G declassification
NLJ i-Ii'l~d& :..;N CLASS. I
I ~ DECLALS'i!f ?Ei)
CLASS. (>iANGED 10: TS S C
NEXT {REVIEW DATE:
AU-[H: HR 70-2
CD
F7
Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8
 Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8
The theory behind this is that these data are of sufficient
importance to be "national intelligence" and therefore binding, but
often of such detail that their inclusion in estimates of the type
we do now dilutes both consideration of important (often technical)
questions relevant to them and consideration of the broader questions
of military policy which those estimates also address. Moreover,
different analysts are properly involved, and in the existing
process we tend to have compromise negotiators not really competent
to handle either. Finally what we know and what we estimate and
what we postulate and what we guess tend to be insufficiently
delineated and convey an impression of much greater clarity and
certainty of knowledge and judgment than exists.
II. An annual or.semi-annual summary of Soviet/EE military
capabilities should be published for senior policy makers. It
would summarize the information in the loose,e ,f book. It would
also state clearly what we know and what we don't know. It would,
where possible, interpret these capabilities in terms of capacity
to do, or not to do, certain things -- such as intervene in far-
distant areas, destroy certain types of targets, etc.
Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8
 Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8
III. An annual estimate of Soviet military policy should be
prepared. This paper should consider the various strategies and
options open to the USSR and take account of such evidence as
there is (doctrines, developments in research and testing, etc.,)
relevant to these various options. It can, and should, state the
outside limits of Soviet policies (such as "a clear military superi-
ority"), identify such policies as seem clear, and express views
(even if there are dissenters) on what courses of action at the
point of writing seem most likely. While the general dimensions
of future force strengths might be discussed, the paper should
avoid specific projections on a year-by-year basis of specific
force goals.
IV. In addition to the above, we should do individual estimates
on specific problems when necessary and appropriate. The SS-9 seems
to me to be such a subject. The object should not necessarily be
to arrive at specific answers to all questions; the object should
be to shed light. If there are anomalies or uncertainties or
sheer ignorance, they should be openly stated. If there are
varying interpretations of the evidence, they should be disclosed.
If there are important points regarding which the methodology,
rather than the evidence, is crucial, the methodology should be
Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8
 Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8
explained. In short, this method of dealing with current, critical
issues will do much more to clarify the problem, and hopefully to
prevent misuse of estimates, than continuation of the present
omnibus system in which crucial elements of the estimating process
are lost under the pressure of time.
25X1
Approved For Release 2006/10/13: CIA-RDP79R00971AO00400050028-8