THE ROLE OF THE SRP IN THE POST-BOWIE ERA

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP98S00099R000400660008-6
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 5
• Document Creation Date: December 27, 2016
• Document Release Date: April 26, 2012
• Sequence Number: 8
• Publication Date: October 19, 1979
• Content Type: MEMO

Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
19 October 1979
Memorandum for the Senior Review Panel
Subject: The Role of the SRP in the Post-Bowie Era
Before I become tainted with a knowledge of the
workings of the Senior Review Panel., permit me to offer
some thoughts as a rank outsider. I would be pleased
to discuss any of these"points with you at your convenience.
STAT
ALL PORTIONS ARE
i1NCLASSIFIED
Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6 -r--
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
Some Thoughts on the Senior Review Pr~.nel
Observations
There is a need--in my view, ~~ strong need--for the
enunciation and enforcement of high standards for national
intelligence production.
Stich a quality control function can best be maintained
if it is ultimately independent of responsibility for the actual
production of intelligence.
The standards for judging the quality of intelligence
products should derive from the DCI's and consumer's needs,
rather than from existing methods of intelligence production.
To enforce standards, a formal critical mechanism should
be developed. One does not now exist.
The uniquE talents of the present members of the Senior
Review Panel lie in their broad knowledge of US policy concerns.
To involve the Panel in the detailed mechanics of intelligence
production would be to dilute the effectiveness of these talents.
Re:cornmendat ion s
The Senior Re~~~iew Panel should be an independent.. organi-
~ational entity. It should be responsive to the DCI in his
role as the government's senior intellii~ence advisor. If a
designated representative of the DCI is the de facto audience
of the Panel, regular direct access to the DCI ~i~mself should
still be assured.
The Panel should nog be inserted into the bureaucratic
process of production (e.g., it should not be required to
approve NIE terms of reference before r_he process can proceed).
It should assume a board of directors' outlook rather than
that of line managers.
Under a broad charter for quality control, the Panel
should be free to adapt its concerns and methods of operation
to the needs of the tune. Ir_s greatest impact will come from
the quality of its advice, and the areas of advice should be
unrestricted.
ALL PORTIONS ARE
UNCLASSIFIED
Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
 t
Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
To provide a visible identity for the Panel's activities,
a periodical devoted to evaluation and criticism of. finished
intelligence should be considered.
Discussion
With the minor exception of some Congressional revie;-
of intelligence performance on specific events, there really
is no regular independent evaluation of the quality of finished
national intelligence. There is an almost suffocating internal
review and coordination process during the production of
estimates and other studies, and occasional post-mortems are
held after major studies are completed. But these are self-
evaluations by the people who are responsible for the studies.
There simply is no established mechanism for a comprehensive
look at national intelligence by disinterested parties having
a unified view of what national intelligence should be.
Even discounting the wide variance in audiences for
national intelligence, I think there remains a great disparity
among managers of intelligence production--NlOs, CIA Office
directors, etc.--about the standards for an acceptable
intelligence report. Few would disagree about goals--providing
analysis which illuminates policy issues or whatever--but I
suspect there would be considerable disagreement about how to
tell when we have done that jab well.
The Panel's memorandum of 2 October (Theory and Practice
of NIEs, SNIEs, and Ibis) does a good job of defining what
national intelligence documents are supposed to do, but it
doesn't discuss how to measur,,e progress toward or accomplishment
of the stated goals. It i.s in establishing a mechanism for
relating accomplishment to goals and enforcing standards
that Z believe th~~ Panel could have the most beneficial effect.
What sort of standards? I mean really fundamental ones,
such as:
--Does the document answer the que;;r_ions asked?
--Does i.t address the right policy issues?
--Is it analytically sound?
--Where judgments and projections ar.e involved, is
the basis for the judgment clearly presented?
Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
 s
Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
--Are assumptions cloarly labeled? Is it clear how the
answer would change if the assumptions were wrong?
--Are uncertainties and alternative interpretations
adequately accounted for?
It mi,ht at first appear that such a resort to f.ir.st principles
is a bit childish for a bunch of grizzled intelligence
professionals. I think if you probed the attitudes of various
producers, though, you would find a variety of judgments
as to the adequacy of individual products in meeting such
standards. What I am encouraging is a more uniform control
over the application of standards, so that eventually we
have a closer agreement on the definition of analysis versus
assertion, what constitutes a complete as opposed to a partial
answer, explicit assumptions differing from 'as any fool
knows", etc.
In deriving standards for national intelligence, I think
a hard rethinking of service to the consumer is ir, order.
It has always impressed me that intelligence producers tend
to be arrogant in protecting the consumer from hard thought.
The drive for clarity, succinctness, and simplicity is
frequently self-defeating in obscuring sttbtlties and genuine
complexity. In any event, the standards should stress the
needs of the consumer over the convenience or prejudices
of the producer.
The intelligence trade is unusual among established
scholarly disciplines in lacking a regular critical mechanism.
I believe it suffers from this Lack. There simply is no
peer evaluation, no letters to the editors column where a
continuing dialogue can serve to extend and sharpen
analysis. The new NFAC journal,"Contra", may be a step
in that direction, but it has a pretr_y amorphous shape at
the moment.
I think it would be a mistake j.f the Panel were to
become too closely allied with the NIOs or D/NFAC (as distinct
from DD/NFA). The DCI has abundant sources of qualified
advice about the natty-gritty of intelligence collection,
processing, analysis, and production. 4-Ie has precious
little wisdom available to him on how all of that should be
transfornied into meeting the needs. of the consumer. The
Panel members were selected, I presume, for their proven
record outside the intelligence community; your skills would
be squandered if you were to become overly concerned with
the minutia of intelligence management. I encourage you r_o
stand somewhat above the fray.
Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6
From that preamble, where would I go? Let me approach
it from the standpoint of if I were DCI.
1. I would want a panel which is truly independent of
production offices.
2. The panel should concern itself with improving the
quality of national intelligence. Its focus, then, would be
on the output of the intelligence process, but its authority
should extend to looking as far back into the process as it
thought necessary to explain deficiencies in the output.
3. I would ask the panel to establish and lead the
activities of a formal critical mechanism.
Declassified in Part - Sanitized Copy Approved for Release 2012/04/26 :CIA-RDP98S00099R000400660008-6 ''~!'~