ATTACHED IS A REVISED VERSION OF THE DRAFT RESPONSE TO THE NSC ON THE OBJECTIVITY AND INTEGRITY OF THE ESTIMATIVE PROCESS EXPANDED, REFINED, AND ALIGNED WITH THE BRIEFING.
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP89T01032R000100060017-6
Release Decision:
RIFPUB
Original Classification:
S
Document Page Count:
21
Document Creation Date:
December 23, 2016
Document Release Date:
May 30, 2012
Sequence Number:
17
Case Number:
Publication Date:
May 20, 1987
Content Type:
MEMO
File:
Attachment | Size |
---|---|
CIA-RDP89T01032R000100060017-6.pdf | 822.52 KB |
Body:
Declassified and Approved For Release 2012/05/30 :CIA-RDP89T01032R000100060017-6
~ach~d'_ *~s i~isedA= :tir>< f `ttie 'draft
_ -~snense~~:to~~~5frnn.~ttie~bti~e~J~-.i.#y and rote-
and :aligned :with ,the..br~ef~ ng~~ ~~,l~ed to
give at NFIB on Nay 9th. *Copi~?;;hav~- been .dis-~
tri buted to the ~IUs ~ and SRR ~~~t~ieir' comments
by COB this Thursday. Any cohn~ehts";-you,may have::
points..
are solicited as wel 1. If ~we`-were able to arrive
at a draft with which you were satisfied by the
end of the week, we could distribute it in advance
of NFIB and use the briefing to summarize the
paper and focus the discussion on its principal
"ran~C B. Horton III
Major General, USAF
Attachment:
As stated
Declassified and Approved For Release 2012/05/30 :CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
THE INTEGRITY AND OBJECTIVITY OF THE ESTIMATIVE PROCESS
1. As the President has stated, it is an important goal of the United
States intelligence effort that the integrity and objectivity of the
intelligence process be maintained, demanding critical differentiation
between foreign intelligence and policy advocacy. This is a particularly
important and sensitive goal as applied to the production of national
estimates. It requires critical and constant attention. This is especially
so since it must be considered side by side with another important goal,
emphasized by the SSCI in a draft report last year, that the utility and
relevance of the intelligence product be sustained, demanding cognizance of
policy concerns in the production of foreign intelligence. This dual
objective of assuring policy relevance while assuring against policy
prescription makes two basic demands.
-- First, while input to the estimative process is to be sought from
the policy community, it should be as to the questions to be asked,
and not as to the answers to be provided. And the policy community
should not be the only source of these questions.
-- Second, while the output of the estimative process should outline
the general implications of the intelligence community's judgments
for US interests, it should not cross the line to the advocacy of
particular policy proposals. That must be deferred to the
consumers in the policy community.
1
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
It should be recognized that the search for perfection in the estimative
process inevitably involves a process of tradeoffs between faithful
observance of the full bureaucratic process and the intellectual vigor of
the final product. As more people and more opinions become involved, many
of the sharp edges can be lost; process is observed but the result can be
pallid, the message to the policymaker unenlightening. The Community must
strive to strike a balance between often conflicting goals of timeliness and
proper consideration, between crispness of judgment and due regard for
alternative views, between safe consensus and boldness of insight, between
policy relevance and policy neutrality.
2. The production of national estimates can be broken down into three
basic elements for purposes of focusing on the maintenance of the integrity
and objectivity of the enterprise--personnel, process, and products. Such
an approach facilitates highlighting areas where vulnerabilities might exist
and safeguards might help--leading to an assessment of what we have done and
what more might be done, especially noting where senior officers directly
affect the process.
a. Personnel. The most critical component in ensuring the integrity
and objectivity of the estimative process are the key individuals involved
in it. Without individual integrity and objectivity at all levels,
particularly at the top, no process will work; with it, nearly any process
wi 11 be successful.
2
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
-- The National Intelligence Council is an ideal environment- in which
skilled and objective National Intelligence Officers can manage the
production of estimates. It provides relative freedom from
specific policy agenda and represents--through the different parent
agencies from which NIOs come--some sense of the differing
perspectives of the intelligence community as a whole.
Also important are the Assistant NIOs, who are selected to
complement the NIOs in terms of organizational and substantive
background. Working with the NIOs and ANIOs are the drafters and
agency representatives, both key to shaping the final product.
Over time drafters are drawn from the many agencies of the
community, and all the agencies are represented in the coordination
of each estimate. All of these personnel--NIOs, ANIOs, drafters,
and representatives--need to be of top quality with a good sense of
their charter. Selection and orientation is thus critical, and is
given close attention.
There are balances that each of these players must strike. NIOs
understand that they serve both as representatives of the DCI and
as facilitators and guardians of the community process. ANIOs know
that they serve as alter egos to their NIOs and as counterpoints to
them. Drafters need to be able to utilize their agency background
and at the same time rise above it and take on the community
mantle. Representatives need to ably and responsibly represent
their agencies, while being capable of setting aside parochialism
in the interest of producing the best possible community product.
3
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
And finally there are the intelligence principals, who need to be
involved throughout the process, initially through their
representatives and ultimately through their participation in the
National Foreign Intelligence Board, to bring to bear their broader
perspective and to help protect the intelligence process against
undue policy pressures. A key to the success of this endeavor is
the NIOs keeping the representatives informed, so that they in turn
can keep their principals in the loop. This does not so much
policitize the process as to protect against pressures for
politicization.
b. The Request for the Estimate. Turning to the process itself,
requests for Estimates can come from any executive office, but are most
often generated externally by State, DoD, or NSC. The DCI also generates
many Estimates, although the greatest number come from the individual NIOs.
The NIOs often suggest that an Estimate be written where they have
learned--from their contact with the policymakers--that such an estimate
will be timely and relevant to the Tatter's needs, or where the NIOs
perceive an issue looming that has not yet captured the policymakers'
attention and that could have considerable impact on US interests.
-- The NIO is routinely in contact with other members of the
intelligence and policy community in his area of expertise. He
needs to be well aware of current US policy to anticipate both the
intelligence needs that the policy might spark as well as the
foreign policy repercussions of prospective US action. Those
4
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
estimates--a boon for policy relevance and legitimate support to
the policy process, but carrying with it the danger of skewing
production to topics formulated only to support, and not to
illuminate, policy.
To guard against this, two processes help. First, many Estimates
are generated outside the policy community, to indicate areas where
policy attention is needed, to tell the policymaker what the
intelligence community feels the policymaker needs to hear, not
necessarily what he wants to hear. And second, each request for an
Estimate is reviewed by the NIO and Chairman, NIC and must be
approved by the DCI. Disapproval is infrequent and never because
of incongeniality to the policy community, but may be necessary
because of resources, priorities, or appropriateness--especially if
it is felt to be of a limited interest or too narrow or parochial
in its focus for a national level Estimate. NFIB principal review
of the production plan assures that intelligence community managers
can advise the DCI on topics for Estimates. The NIOs and NFIB
representatives also maintain a continuing dialogue with each
other, and the NIOs with the DCI to ensure that the production plan
remains up to date and relevant.
c. Preparation of Coordination of TORs. Following approval by the DCI
of the proposal for an Estimate, the NIO draws up a Concept Paper and Terms
of Reference (TOR) drawing on suggestions from the intelligence community to
5
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
define the scope and major themes of the Estimate. This provides yet
another opportunity for community review of the need to add, revise, or drop
an estimate.
Usually the o~fice requesting the Estimate will set forth the main
questions which it hopes to see answered. The NIC views any
intelligence questions posed as legitimate and will respond to
them. Answering only those questions posed by the requestor,
without considering the broader context, however, could give a
misleading impression regarding policy implications. The NIO and
his colleagues thus will not necessarily limit the scope of a
proposed Estimate strictly to the questions and issues posed by the
requestor. This independent, yet collegial, preparation of the TOR
within the Intelligence Community is designed to prevent Estimates
being driven by leading or narrowly focused questions from the
policy side answers to which could present a misleadingly narrow
view of the overall problem.
The NIO frequently consults with the requestor to insure that he
understands the specific interests and purposes of the questions
posed, and to insure that the requestor's intelligence needs are
met. The NIO formulates the TOR in complete independence of the
requestor, however. The SSCI in a September 1986 draft study
however specifically urged that NIO's remain in close contact with
policymakers during the Estimative process to ensure that those
issues relevant to policymakers are addressed.
6
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
A critical part of the TOR is the Key Questions. This section sets
forth the few central questions--usually no more than a half
dozen--which the NIO believes to represent the heart of the
intelligence problem. By nature they are as broadly conceived and
as searching as possible. These questions are drawn up to ensure
the drafter will not become so bogged down in detail that he fails
to reach the hard intelligence judgments that were sought in the
TOR.
-- The TOR also includes a retrospective look at the previous Estimate
on the same subject. This process helps remind the Community of
past discussions on the topic and provides opportunity for minority
or alternative views to be tested against subsequent events.
-- The NIO prepares the TOR with the VC/NIC and the C/NIC to insure
reasonable breadth, comprehensiveness, and focus. Then the draft
TOR is sent to the Senior Review Panel (SRP) for its independent
comment on the appropriate breadth, conception and focus of the
paper. The SRP is made up of retired senior diplomats, military
officers, and scholars and reports directly to the DCI; it is part
of his staff--not part of the NIC. Other independent DCI or NIO
reviewing bodies may also have an opportunity in an advisory
capacity to consider the issues of the Estimate.
7
CF~RFT
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Stl:Kt 1
-- The NIO will .make such changes as he believes are necessary based
on SRP commen~s--which are not binding. The NIO is free to take
issue with the SRP on one or another points, but SRP comments are
usually constructive, valuable and are incorporated. The TOR is
then sent to the DCI for approval with a cover memo indicating how
SRP comments were or were not accommodated.
-- The DCI may make any changes or suggestions in the TOR that he
believes are necessary--usually by way of addition. Upon his
approval the TOR is sent to the Intelligence Community principals
preparatory to holding a community representatives meeting to
coordinate the TOR.
-- The Intelligence Community representatives meet with the NIO to
review the TOR and to suggest changes for content, clarity, and
comprehensiveness. After incorporation of these revisions, the
revised TOR is sent out to Community representatives as the formal
coordinated TOR for the Estimate.
-- This process of in-house and Intelligence Community TOR review
almost invariably adds to the scope, breadth and comprehensiveness
of the paper. Rarely, if ever, are questions or issues dropped in
the course of such review, except when deemed outside the purview
of intelligence.
8
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
d. The Drafting Process. The NIO will select a drafter for the
Estimate from within the Community. The drafter thereafter then reports to
the NIO on the Estimate and must be responsive to him to assure independence
and objectivity--as well as efficiency. The drafters should be seeking
ideas and inputs from throughout the community and from outside experts as
well as he works to develop a response to the agreed TOR.
-- The drafting process, however, is sometimes influenced by the
drafter's parent organization biases. The drafter--whatever his
parent organization--will usually submit his draft for review at
least to his own immediate superior and possibly higher reviewing
officials before the draft goes to the NIO. This is not always the
case, and usually is prompted by a desire to improve the writing
through peer review and present the best possible draft to the NIO.
-- This reviewing process by the drafter's parent organization usually
does contribute in terms of clarity, style and accuracy, but it
also can affect the position taken by the drafter. In the real
world, however, any drafter will be in part influenced by his own
bureaucratic culture whether he undergoes his own in-house review
or not.
-- The NIO will review the draft closely and critically. He has
license to make any changes whatsoever he feels are necessary in
the interest of clarity, style, length, comprehensiveness, focus,
9
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
sharpness of judgment, accuracy of judgment, and depth of
judgment. Initial drafts submitted to the NIO often require
extensive work to insure that they are terse and estimative in
character; initial drafts frequently suffer from excessive length
and unwillingness to reach hard judgments. The NIO's own changes
made in the draft reflect a balance of his own best judgment and an
awareness that he will have to coordinate the text through the
Community.
-- The NIO reviews the draft with the VC/NIC or C/NIC for further
review and comment. The draft will then go to the SRP for
independent review and comment. The NIO will accommodate as much
of the SRP suggestions as he believes is appropriate. In most
cases, SRP suggestions for change and improvement are incorporated.
The revised draft is then sent to the DCI with a cover memo
indicating how the SRP comments were accommodated. The DCI will
often request sharpening of judgments, and explanation of expected
dissenting views. This is an opportune and appropriate time for
the DCI to make his input in the drafting process; it should not be
taken to stifle the opportunity for the later expression of
alternative views by the rest of the community. After the DCI's
approval the paper is sent to NFIB representatives for review and
staffing.
10
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
-- At no point during the drafting process does the NIO or the drafter
discuss the draft with the policy community, and the policy
community should not have impact on the drafting process.
e. Draft Coordination. The role of the NIO at the coordination meeting
is a crucial one. He is responsible to the DCI to produce the best possible
estimate he can along the lines agreed upon in the TOR, and is responsible
to the community to assure that significant alternative ~~iews have the
opportunity to be adequately expressed.
-- The Estimate must reflect the best Community judgments but must
also not be allowed to deteriorate down into the "lowest common
denominator" consensus. It must reflect important alternative
views while avoiding the pitfalls of becoming a Christmas tree of
endless possibilities that destroy clarity of the judgment call.
The NIO cannot simply be an "honest broker" among contending
parties; his role involves a greater degree of leadership--more
akin to that of judge and manager who must ultimately arbitrate
conflicts of view and serve as catalyst to ensure judgments are
considered and f ar-reaching. The NIO must avoid foreclosure of
debate or rejection of certain viewpoints by having recourse to
statements such as "the DCI has special interest in this issue"
which can serve to intimidate some representatives. Highly vocal
representatives must not be allowed to carry the day by sheer
forcefulness, but quiet representatives should be encouraged to be
forceful in expressing the views of their respective
organizations.
11
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
~~~~.~
Ultimately, the NIO's role is to force the Community to make hard
judgments while allowing for reflection of significant divergency
of opinion. It involves a great many judgment calls and is a more
complex role then that of merely an impartial arbiter. At the same
time holders of dissenting viewpoints have the obligation to
present a coherent and significant alternative view that enriches
the document--rather than simply seeking to use less bold
adjectives.
-- At no point in this coordination process is the policy community
involved; indeed, while requests have often been made, no member of
the policy community is allowed to attend a coordination session.
f. Final DCI Review. The DCI reviews the Estimate following community
working level coordination prior to sending it to NFIB principals.
The DCI can and sometimes does seek substantive improvement in the
text, especially where the coordination process has served to
excessively water down judgments, or where Key Judgments have been
worked to the point that they diverge from the text. This role is
a sensitive point in the process because he is now making judgments
on the quality and appropriateness of a text already blessed by the
community working level representatives.
12
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
While this late DCI intervention can generate charges of
politicization, there is a mechanism to adjudicate any concerns
that arise. In routine cases, such as moving statements from text
to key judgments, such proposed changes can be brokered informally
prior to NFIB between the NIO and community representatives. More
substantive issues should be and normally are raised at NFIB with
principals, however, to provide a proper airing and to minimize any
question of politicization that might otherwise arise.
f. Final Review in Community Agencies. Following distribution to NFIB
principals, most agencies will submit the Estimate to further in-house
review. Certain problems can arise in the course of this process.
Some Community working level representatives may not fully
represent their agency and their principal at coordination
meetings--because of unresolved differences within the agency,
inability to get the view of the agency's principal, or lack of
empowerment. As a result, a commitment by a working level
representative at the pre-NFIB coordination meetings may not be
supported by the NFIB principal, sometimes over lesser issues. On
the other hand, it also indicates that the principals are prepared
to engage themselves in the process of judgment and not leave it
all to staff.
13
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
Intelligence organizations attached to Departments with
policymaking responsibility face other complications. Their
working level intelligence representatives sometimes commit their
NFIB principal to an intelligence judgment in the estimate that is
seen as unwelcome by the policy side of the department. This
places a special responsibility on the heads of INR, DIA, and the
Service intelligence organizations to insulate the intelligence
process from any undue policy influence.
For whatever reason, we find that even after completion of working
level coordination, and pre-NFIB distribution, some representatives
report back to the NIO that new changes are being sought by their
Agencies. This part of the process, like final DCI review, is
vulnerable to charges of politicization. As in the case of any
late DCI changes, there is a process for adjudication--pre-NFIB
brokering for routine adjustments, and deferral to NFIB hearing for
those of greater significance.
h. NFIB. The National Foreign Intelligence Board represents the final
forum for adjudication. The principals are free to propose changes in the
work of the Community representatives. It falls to the DCI to adjudicate
disagreements at NFIB, to accept footnotes or alternative language, or to
remand the Estimate for further work in selected areas. Finally, it is the
DCI who approves the document. The NFIB process is the best deterrent and
ultimately the best defense-against any high-level pressures to shape
intelligence to policy preferences.
14
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
i. After NFIB, the process continues with a f aithful reflection of the
NFIB recommendations aad the DCI decisions in producing the estimate, and
the appropriate dissemination of the product to those with a need to know,
to include the timely provision of the key judgments to the top levels of
the policy community. Then there is a feedback process that brings us full
circle. That feedback includes, for example, discussions between NIOs and
key members of the policy community regarding apparent disconnects between
the judgments of past estimates and the policies that were eventually
developed, and discussions among the NIOs and the rest of the intelligence
community leading to retrospective analyses as to accuracy of the judgments
of past estimates.
j. The Problem of Evidence. It is obviously desirable that differing
viewpoints be backed by evidence wherever possible. Lack of evidence and
uncertainty needs to be clearly identified in the text whenever judgments
are made that are based more on informed insight than hard facts.
Appendices can serve to buttress arguments for certain positions taken in
the text--at least at the draft stage--but probably not in the published
text. But questions of evidence also cannot be allowed automatically to bar
a judgment simply because evidence is lacking. A position on an issue can
be sometimes sustained even in the absence of much evidence when
intelligence is skimpy, ambiguous, or non-existent. This is especially true
in areas involving clandestine or covert activity by another party. In
short, while it is risky for a drafter to assert a position against existing
evidence or in the absence of any supporting evidence, the drafter and the
community likewise must not be so mesmerized by absence of indicators, or by
15
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
a misplaced faith in the likely spectrum of knowable intelligence available
to them on a certain topic that it leads to automatic rejection of
considered judgments based on some intuitive feel for behavior patterns of
the players involved. Thus, on the evidentary problem the community is
called to walk a fine line between establishing a legal brief and accepting
a position based on circumstantial evidence and intuition in the absence of
any certain knowledge.
k. The Dilemma of Senior Level Review. As noted earlier, review of
draft Estimates by the NFIB principals and their suggested changes can
provoke charges of politicization of the intelligence process. These
individuals, and their subordinate senior managers, in particular are closer
to the policy process than are the intelligence analysts who draft the
Estimates and they bring a different outlook to the intelligence process.
Differing senses of international realities can mistakenly be identified as
a politically partisan view. The benefit of senior level review of
Estimates is that it often brings to bear a broader conceptual view of the
nature of the international order and the character of international
conflict. These views inevitably influence the assessment of the likelihood
of certain events. It is important for all parties involved to distinguish
between a legitimate critique of a narrow, ill-formed or naive conception of
politics or the international order, and the advocacy of preferred political
views on how to deal with the nature of the international conflict. In
short, the higher the level of generalization about how certain situations
will evolve, the more one's own unspoken philosophical sense of how the
world works is involved. Searching senior official critiques of community
16
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
estimative views need not represent "politicization" of the process. And
the senior reviewer must be careful that his critique is not in fact
politically driven or broadly perceived by the community as such.
1. Products. Out of this process come four types of formal estimative
products.
National Intelligence Estimates and Special National Intelligence
Estimates, or NIEs and SNIEs, go through all steps of the process
just outlined. As they bear the NFIB imprimatur and DCI signature,
they have the greatest weight. SNIEs differ from NIEs in that they
go through some of the steps in parallel rather than in series, to
expedite production of high priority requirements, but they still
involve the key players, the NFIB is still the final forum for
adjudication, and the DCI is still the final approval authority.
Interagency Intelligence Memoranda and Interagency Intelligence
Assessments, or IIMs and IIAs, differ from NIEs and SNIEs in that
they are less estimative in nature and do not necessarily go
through NFIB or bear the DCI's signature, but all agencies (in the
case of IIAs at least all those with a substantive interest) still
have the opportunity to coordinate, and agency representatives
still have the opportunity to bring their principal's point of view
to bear in the coordination process. Like the SNIEs, the IIAs go
11
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
through some of their steps in parallel. Where the issues emerge
as important enough, or the controversy surrounding them becomes
great enough, IIMs in particular can be and sometimes are elevated
to NFIB for consideration as an additional safeguard.
It should be ~~oted that there are other products which the NIOs
work with the community to produce, which are not formal estimates
and do not go through the formal coordination process. These are
warning products, generated by discussions at specially convened
meetings between NIOs and community representatives, and then
produced by the NIOs on a particular issue, or collated by the NIO
for Warning from all the NIOs for a global review. The former are
done on an as needed basis; the latter are done monthly, for the
DCI and further distribution. These warning products will reflect
community inputs, and make note of alternative views, but will
still be clearly identified as other than formally coordinated
community products.
Further, the iVIO, as senior staff advisor to the DCI in his area of
expertise, fulfills a number of roles in which he does not
represent the Community. In these roles, especially in producing
written memos done as think pieces for the DCI or wider
distribution to stimulate discussion, it is imperative that he make
explicit the fact that his views are his own, and he is not
speaking on behalf of the Community, even though his views will
have been formed i~ the context of his interaction with that
community. When applicable and known, alternative views may be
reflected in these memos as ~~11.
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Jtl.Kt I
3. In summary, the estimative process does have some vulnerabilities in
terms of maintaining its objectivity and integrity, but it also contains
many safeguards. These safeguards, consistently applied by all the players
in the process, from drafters and agency representatives through NIOs and
the NFIB principals, can maintain the objectivity and integrity we seek. To
summarize these safeguards, they are:
o Careful personnel selection and orientation for each of the key
roles notes above.
o Policy participation properly circumscribed to providing inputs
regarding topics to be addressed and questions to asked, but not
regarding answers to be provided.
o The use of broad Key Questions in the Terms of Reference to guide
and evaluate the drafting process.
o Many levels and sources of input and review throughout the
intelligence community, including by more independent bodies such
as the SRP.
o Thorough community coordination at key steps along the way,
especially of the Terms of Reference and the drafts of estimates.
19
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6
SECRET
o Encouragement of the development of alternative outcomes to reflect
uncertainty, and alternative views to reflect significant
disagreements throughout the process.
o The continued active involvement of intelligence community
principals working through their representatives early on, and
ultimately through their participation at NFIB.
o Self -evaluation of completed estimates and feedback to preparation
of new ones as the cycle continues.
Continued careful implementation of these safeguards should assure that
objectivity and integrity are served well.
20
SECRET
Declassified and Approved For Release 2012/05/30: CIA-RDP89T01032R000100060017-6