TEXTS RELATING TO THE CONCEPTION OF NATIONAL INTELLIGENCE ESTIMATES AND HOW THEY SHOULD BE PRODUCED

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP79R00971A000100020002-2
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 17
• Document Creation Date: December 20, 2016
• Document Release Date: October 13, 2006
• Sequence Number: 2
• Publication Date: August 19, 1969
• Content Type: MFR

Approved For Release 2006/10113 :CIA-RDP79R00971A000100020002-2
19 August 1969
StT&7ECI': Texts relating tc the Conception of National
Intelligence Estimates and How They Should be
Produced.
i:xrc~iive i3egistzy
,~~J_J,
Attached is a compilation of hic~tarical texts relating
the subject. Having b~~en put to some trouble to collect these
texts from many scattered sources, I think it ~rorth~rhile to
record them in one document. I have added comment on particula...~
texts, and in general at the end of the series.
LD'I~WELL L. MGPZTAGtTE
Board of National Est~.mates
G#tOUP 1
Excluded from autorr~Gt:i.L
downgrading aiad
declassification
Approved For Release 2006/10113 :CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113 :CIA-RDP79R00971A000100020002-2
S-~ -C-K-E-T
TEXTS RELATING TO THE PRODUCTION OF NATIONAL INTELLIGENCE ESTII~IA'I'~.'S
=Lo Prig, Gen. John Magruder, Deputy Director, OSS, to
Col, C. R. Peck, Exocut3ve Secretary, J??S, 3D Ju1.y ?g~3.
A vicious circle Yeas r;hus 'been created fre~in which there is
no escape unless, first, the functions of each service are
clearly defi~ted, and second, a centralized joint agency is
established with authority to cantrvl and. coordinate all
intelligence services a,rid to create a joint operating age~2c;~
competent to analyze, synthesize, a~:d integrate int~:lligencc
material. from all sources.
Maj. C,en. Williaya J, i.~novan, Director, OSS, to the
President, 1 November 19Y~+. ^'
TYze Central :Intelligence Service sha11 perform the ...
final. evaluation, synthesis, and dissemination within +he
Government of the intelligence required to enable the
Government to determine policies with respect to national
planning and security in peace and *aar, and the advancement
of bread national policy.
3. Donovan to the President, 26 December lg~-4.
The er~d product of ir~-telligence activity must be a complete
synthesized estimate upon which policy with respect to tYie
national security can sa~~`e1y be based.
~+. JIC X39/5, 1 January _19+5
The Central 1:ntelligence Agency shall accomplish the synthesis
of departmental intelligence relating to the national security
and t?~e appropriate dissemination within the Govertvnen.t of t~:~:
resulting strategic cnd national policy inte113gence.
Approved For Release 2006/10113 :CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
5.
(JIC 239%5 was the compromise reached in the JIC after a
six-weeks deadlock over Donovan's proposal to the President.
The controversy, however, was over control of the prospective
CTA, not over its functions. This text, which I drafted, was
intended only to be a briefer expression of item 2. --
Ludwell Mvntaguej
William H. Jackson to the Director, OSS, July 19+5?
The best intelligence opinion in England is thus obtained
from the heads of all important intelligence agencies having
an interest in the subject. This opinion, drafted in the
first instance by the JIC staff, itself representative of
various intelligence services and agencies, is based on alA
information available in London in any military or civil
agency or department. A JIC appreciation is important,
then, because it expresses an agreed view of the most res-
ponsible intelligence experts based on a1.I. available infor-
mation. Against this obvious advantage, there may be a
lesser disadvantage.... The result sometimes appears to
be a compromise which represents no one's view, least of
all ~hat7 of the intelligence agency which should know
most about the subject....
b. Ferdinand Eberstadt to_the Secretary of the Navy (Forrestal),
22 October 19 5.
The collection of information can and should be made +hrough
the military services and other departments and agencies oi'
the government.... All information so collected should be
available to the Central Intelligence Agency. Its compila-
tion, analysis, evaluation, and dissemination, however,
particularly as relating to matters of national security,
should be coordinated by the Central Intelligence Agency.
7. The President's Letter to the Secretaries of State, War,
and Navy, 22 January 19 6.
The Director of Central Intelligence shall accomplish the
correlation and evaluation of intelligence relating to t;~e
national security, and the appropriate dissemination within
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 ApprovedrrFor Release 2006110/13: CIA-RDP79R00971A000100020002-2
ttie Govertrment of the resulting strategic and r-atio~aa3 i;olicy
intelligence. In so doing, full use shall be made of the staff'
and facilities of your Departments.
(Ad~airal Sorters, who dz~a:~?ted Lhis text for President ~rtunan,
told me that "correla.tion and evaluation" was substituted for
"synthesis" without intent to alter the meaning of JIC 239r~1
(item ~? above). The President did not understand the word
"synthesis" ar_d thought it to have a derogatory connotation,
as in "synthetic". -- Y~udwell Montague )
3. Ludwell Montague to Whe: P.s~:istant Director, Research ar...d
Evaluation, 29 January 19+7. This text is included as an
elaboration oY ?the original intention inadequately expressed
in items ~+ and 7. It should be added that the original con-
weption included a preliirinary coordination with IAC repre-
sentatives at the ORS: level and final coordination at an IAC
meeting, with power of decision vested in the 1~T., the purpose
of the meeting being final determination of concurrence or
dissent. -- L.M.)
'i'he true primary mission of ORE is clear in the light of the
President's Letter and of NIA. Directives No. 1 and No. 2. It
is to produce strategic and national policy intelligence
through the correlation, evaluation, and final synthesis of
all intelligence ini'ormati.on and finished intelligence avail-
able in the State, "'~Iar, and Navy Departments and other Federal
agencies. By "strategic and national policy intelligence"
should be understood that intelligence required at the highest
policy making and planning level a.s a basis for the determina-
tion of national policy and strategy in the broadest sense.
It relates to those issues which are of collective concern to
the State, War, and Navy Departments, or, conversely, which
are not the exclusive concern of any of them. In this concept
ORE has no occasion to duplicate or complete with departmental
intelligence agencies -- rather it i.s charged to make full use
of them and of their product --? but ORE does have the function
of fine? evaluation and final synthesis, The departmental
agencies are tributary to it. Its own contribution is the
added value provided by authoritative final interpretation a~ad
snythesis for the benefit, primarily, of the high authorities
Approved For Release 2006!10113 : CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
,,^,-F-C-R-E-T
whom it servES, and, incidentally, of the cortri'buting
agencies. xt must be supported and manned in such away
as to assure that it does speak with recognized authority.
9~ Ludwell Montague to the Assistant Director, Reports and
Estimates, 17 July 1y 7. I include this text because I
bel"' ieve it to have been the inspiration for item ? 1, which
in turn appears to have 6~een the earliest conception of a
Board of Natio~n4~ and~cousd noteescapelfrum. thatrstultis
a component
fying context. -- Z?~~?
The essence of this concept is that the ~GlobSele ~~~ men
Group? should be composed of a few carefully
of broad intelligence experience (rather than part~Crued
specialization) and of proved insight, who, being inevit"~
of operational responsibilities, have the freedomo(ponder
bF1~ denied to all cn.iefs of operational units
the broader aspects and less obvious implications of the
.developing international situation, to consult the mast
expert opinion with respect to trends thus perceived, to
effect the ate auidanceeto the staffsnandsbranchestconive
the appropr g
cerned,
l+ 96 July 19+7 ( cf . items
10. T.he National Security Act of 19 7,
end 7 .
:Ct sha11 be the duty of tine Agency... to correlate and
evaluate intelligence relating to the national security,
and provide for the appropriate dissemination of such
intelligence using where appropriate existing agencies
and facilities.
('iris text is a classic example of how literal meaning is
changed by coma errors to something unintended.)
-E-C-R-E-T
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 Approved For Release 2006110/13: CIA-RDP79R00971A000100020002-2
11. The Eberstadt Report, 15 November 1948 -- i.e., the Report
of the Committee on National Security Organization ("the
Eberstadt Committee"} to the Commission on Organization of
the Executive Branch of the Government ("the Hoover Com-
mission").
The greatest need in CIA is the establishment at a high
level of a small group of highly capable people, freed from
administrative detail, to concentrate upon intelligence
evaluation. The Director and his assistants have had to
devote so large a portion of their tame to administration
that they have been unable to give sufficient time to
analysis and evaluation. A sma11 group of mature men of
the highest talents, having full access to all information,
might well be released completely Prrnn routine and set to
thinking about intelligence only, Many of the greatest
failures in intelligence have not been failures in collec-
tion, but failures in analysing and evaluating correctly
the information available.
(Cf. items !~ above and 12 below. I believe this to have
been the original conception of the Board of National
Estimates. -- L.M.}
12. The Dulles - Jackson - Correa Report to the NSC,
"~
1 January 19 g
There shou3_d be created in the Central Intelligence Agency
a small. Estimates Division which would draw upon and review
the specialized intelligence product of the departmental
agencies in order to prepare coordinated national intelli-
gence estimates. Under the leadership of the Director of
Central Intelligence, these estimates should be submitted
for discussion and approval by the reconstituted Intelli-
gence Advisory Committee whose members should assume col-
lective responsibility for them.
(This text picks up the Eberstadt Report's "small group"
in CIA, but the supporting discussion shows that the basic
conception was derived from W. H. Jackson's admiring report
in 1945 on the British JIC -- item 5 above. Entirely missing;
Approved For Release 2006110/13: CIA-RDP79R00971A000100020002-2
 `?"-'"'"`"~pprove~F'or?Re1-ease 2006/10113: CIA-RDP79R00971A000100020002-2
is the idea of independent, authoritative eva.~.uation, which
Eberstadt's repart shows that he understood. Instead, the
emphasis is on coordination and "collective responsibility",
In 19+9 the quoted words stood for the idea that the IAC
was advisory to the NSC rather than to the DCT, acid that in
it the DCT was only one among equals. -- i,.M.)
13. Ludwell Montague to the Assistant Director RBeE, 11 February
1 9 with reference to item 12 ,
This is the method of f:~nal coordination originally intended
(until June 19+6) and, effectively implemented with proper
understanding and procedural safeguards, would be a much
more efficient method than that now in force. An important
reservation must be noted, however. These who developed the
orig
Lion
CIA's estimates unit was
intended expressly to overcome both these things by pro-
ducing authoritative final estimates with full cognizance
aF departmental views, but without subservience to depart-
mental prejudices. IAC review was expected to eliminate
apparent differences susceptible to adjustment through
discussion, but not to gloss over real divergences of
informed opinion. The resultant estimate would still be
essentially that of CIA, with notations of concurrence
or dissent, the latter being limited to real, substantial,
and well defined issues. The Report conveys no assurance
that this vital aspect of the matter is truly understood.
It contains one incidental. reference to the Function of
the Estimates Division in countering departmental bias,
'but many passages which suggest that to the authors
"national" intelligence is merely "coordinated" intelli-
gence and coordinated intelligence merely joint intelli-
gence.
11+~. NSC j0, 1 July 19+9, (NSC 50 consisted of the cottments
and recommenns of the Secretaries of State and Defense
with regard to the recendations of the Du3.les Report.
Tire recar,~et~d~af~3 one c~i' he :~e~~i?etaries were substantially
adopted by the NSC.)
Approved Far Release 2006/10!13 : CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
We do not believe that tree Director and the IftC should
be bound by the concept of collective responsibility,
because this would inevitably reduce coordinated national
intelligence to the lowest common denominator among the
agencies concerned. A procedure should be adopted whicYi
would permit the Director and the IAC to fulfill their
respective responsibilities to the President and the NSC
.regardless of unanimous agreement, but providing for con-
current submissions of dissent. The CIA, however, should...
refrain as far as possible from competitive activities in
the production of research intelligence estimates.
ry,Te concur in the recommendation that? out of the
present Off ice of Reports and Estimates there should be
created (a) a small estimates division which would draw
upon and review the specialized intelligence product of
the departmental agencies in order to prepare coordinated
national intelligence estimates and (b) a research and
reports division to accomplish central research in, and
coordinated production of, intelligence in recognized
fields of common interest.
(In adopting NSC 50, the National Security Council made
it reasonat>ly clear that it intended the DCI, with the
aid of a small estimates staff devoted solely to that
purpose, to base national intelligence estimates on the
findings of departmental intelligence research (in order
to avoid duplication of research effort), but to exercise
independent judgment in reviewing and evaluating depart-
mental contributions. DS' contumaceously evading the NSC's
injunction to establish such an estimates offices CIA
provoked the cataclysm of October 195a in which the DCI's
:independence of judgment again narrowly escaped subordina-
tion to a ;joint committee system. -- L.M.)
15. National Security Council Directive No. 1, as revised
19 January 1950.
The Director of Central Intelligence shall produce intelli-
gence relating to tree national security, hereafter referred
to as national in+elligence. In so far as practicable, he
shall not duplicate the intelligence activities and researc~~
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
of the various Departffients and Agencies but sha13 make use
of existing intelligence facilities and shall utilize depart-
mental intelligence for such production purposes.
(This text, produced by an L9C subcommittee, is preoccupied
with making the DCI dependent on departmental contributions.
His independence of judgment finds no expression except by
implication in his individual responsibility to "produce"
national intelligence. This passage in the NSCID was not
altered during General vlnith's tenure as DCI. -- L.M.)
lei. Notes dictated by William H. Jackson (DDCI - designate) in
September 1950.
The Act apparently gives the Central Intelligence Agency
the independent right of producing national intelligence.
As a practical matter, such estimates can be written only
with the collaboration of experts in many fields of intel-
ligence and with the cooperation of several departments
and agencies of the Goverrnnent.... An intelligence estimate
of such scope would go beyond the competence of any single
Department or Agency.... The estimate should be compiled
and assembled centrally by an agency whose ob,~ectivity and
disinterestedness are not open to question. Its ultimate
approval should rest upon the collective responsibility of
the highest officials in the various intelligence agencies.
he was unaware that the NSC had ruled aga~.ns a oc tine
of "collective responsibility" when it approved NSC 50 --
see item 14. On 16 October 1950, however, when he presented
these notes to the DCI, Jackson acknowledged that his use
of the term "collective responsibility" was a mistake. He
or Smith substituted "collective ,judgment" for it. This
change certainly resulted from an oral remonstrance by the
General Counsel, Lawrence Houston. It may also have been
influenced by the document which follows. -- L.M.)
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 pra~~rr'~F~'ase-2006/10113: CIA-RDP79R00971A000100020002-2
17. Ludwell Montague to thc~ llDCI, l4 October 1g5O (in response
to his request for a plan for an Office of Estimates).
This plan is "based on the concepts held in 195-1g~+6 and
more recently set forth in the Dulles Report, NSC 54, and
the "Webb Proposals". Qne point must be made absolutely
clear, however, in order to avoid the patent defects of a
joint committee system. It must be understood by all con-
cerned that the Director at his level and the Assistant
Director at his, having heard all the pertinent evidence
and argument, have a power of decision with respect to
the form and content of the estimate, other interested
parties retaining the ra,ght to record divergent views
when these relate to substantial issues and serve to
increase the reader's comprehension of the problem, and
then only,
The plan also presupposes:
a. The establishment of a Research Office in
CIA to provide intelligence research reports in
fields of common concern (e.g., scientific, econo-
mic, geographic).
b. Action to make sure of the availability of
research support S'rom the departmental agencies
adequate to meet the requirements of the Estimates
Office as to both timelitaess and content. This
conditian cannot be met at present.
c. The recruitment of requisite senior personnel
as rapidly as possible. The contemplated Office
cannot tae adequately manned with personnel now in
czA.
d. Thorough indoctrination of the TAC agencies
in the new, cooperative concept, and a new start
in relations with them.... This plan will not work
except an a basis of mutual confidence and coopera-
tion in the national interest.
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
18, 7;AC Minutes, 2O October 190 (IAC-M-1}.
(Genexal Smith read the full text of Mr. Jackson~s Hates an
"The Responsibility of the Central Intelligence Agency for
National Intelligence Estimates ", from which item 15 is
excerpted, with the substitution of "collective judgment"
for "collective responsibility", leaving it ambiguous
whether "ultimate approval" was a function of the DCI or
of the IAC. }
There was general assent at the meeting to ~hi~ statement
of the responsibility of the Central Intelligence Agency far.
national intelligence estimates. General Smith stated that,
in order to discharge this responsibility, he proposed at
the earliest possible time to set up in the Central Intelli-
gence Agency an Office of National Estimates. This division,
in his opinion, would became the heart of the Central Intell~_-
gence Agency sand of the national intelligence machinery.
1g. ONE Draft for the ICI's deport to the NSC on the Implementation
of NSC 50, ll Februaryr 7.952.
The basic concept of bNE is that it has but one mission: to
produce NIE`s in close collaboration with the 7AC agencies.,.,
In the discharge of its mission ONE considers itself an inte-
gral part of a joint production mechanism, of which it serves
as coordinator. rWow; Who said joint?~
The production of national estimates through the colla-
boration of numerous IAC agencies and C7rA. offices has entailed
the development of a complex, at times cumbersome, estimates
machinery. The role of ONE and the 73vard of National Estitna+~s
vis-a-vis that of the other agencies and offices in this
machinery is still in the course of evolution. Over the pa,s?c:
year, however, an effective working relationship has been
achieved.
The new system has resulted in genuine cooperation
among the IAC agencies, which have devoted more of their
resources to national estimates and have taken their pro-
duction far more serious7.y than was the case with the
estimates made by ORE. However, the price which has been
paid for this close cooperation is the almost inevitable
difficulty of producing estimates by the committee method,
Approved Far Release 2006/10!13 : CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
While in the vast majority of cases disagreements have
been successfully ironed out..., there has been an
occasional tendency to dilute or water down the estimates
in an effort to reach agreement where serious conflicts
were involved,
The dissociation of ONE from all other intelligence
functions has permitted undivided attention to estimates
production and has resulted in better estimates being
produced. Yet this separation of ONE from all but the
estimating function also inevitably tends to separate it
froth the unfinished intelligence on which its estimates
are based. Despite its efforts to brief itself independ-
ently and to check on agency contributions, ONE itself
must largely depend upon the quality of the contributions
it receives.
20. Report of the DCI to the NSC on the Implementation of
NSC 50, 23 April 1952?
Specifically, there has been established an Office of
National Estimates to produce intelligence estimates of
national concern.... In its operations this Office
utilizes the resources of the total United States intelli-
gence comanunity.... As far as our intelligence production
is concerned, the Central Intelligence Agency is basically
an assembly plant for information produced by collaborating
organizations of the Government, and its final product is
necessarily dependent upon the quality of the contributions
of these collaborating organizations.
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
_' -E-C -R-E-T
-~-~~~
Comment from the Point of View of 1969
The language of items 15-16 and 18-20 above tends to
obscure the DCI's personal. responsibility for the judgments
contained in national intelligence estimates and to suggest
instead a joint intelligence system based on "collective
responsibility". There is evidence additional to that in
item 19 that even in ONE there was, on that account, consider-
able misunderstanding of General Smith's position.
In the circumstances of the time, General Smith deliberately
refrained from emphasizing his statutory responsibility and con-
sequent authority. That would only have prolonged the deadlock
that had frustrated Admiral Hillenkoetter; Smith had been called
to break that up. Instead, he engaged in a remarkable public
relations effort to enlist the willing cooperation of the members
of the IAC by giving them a sense of effective participation in
the production of national intelligence estimates. But General
Smith had no intention of compromising the personal authority
of the DCI. I remember vividly one meeting of the IAC (11 Nov-
ember 1950) at which General Smith was again laying the butter
on very thick -- a very remarkable performance for a man with
his reputation as a holy terror. One member of the I.AC was so
carried away by his enthusiasm as to use the words "Board of
Directors", an expression which in those days stood for the
doctrine of collective responsibility. Without noticing dir-
ectly, and without raising his voice, General Smith continued
talking, only now he was talking about the statutory responsi-
bilities of the DCI. General Smith did not have to speak on
that subject more than once.
This incident does not appear in the IAC Minutes, of
course, but the Secretary (James ~. Reber) did record it in
a Memorandum for Record as follows:
At the conclusion of the IAC Meeting on 11
November, General Sh~.ith stated that he had
changed his concept on these meetings since
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
his arrival in Washington. In his opinion the
term "Intelligence Advisory Committee" was a
complete misnomer. He felt that this committee
was now the Joint Intelligence Board of the
United States government. He said that this
new concept would place increasingly larger
demands upon the chiefs of the agencies, but
by their working harmoniously around the table
they were raising the value of the intelligence
product. ere came the interru ting reference
to a "Baar~' of Directors"-- L.M_~ He did say,
however, that there was one important difference
between this Board of Directors concept and that
of the Board of Directors of a civilian concern.
In a civilian concern the officers of the company
were bound by the decisions of the Board in
accordance with his own judgment.' The responsi-
bility for National Intelligence Estimates has been
delegated to the DCI regardless of the desires of
the IAC members to share this responsibility.
General Smith was able to ingratiate the IAC without losing
control of the situation (as had happened to Hillenkoetter when
he tried the same approach) because of the strength of his posi-
tion; his superior rank, his personal prestige, the force of his
personality, and the knowledge that he could count on the strong
support of the Presicient and the NSC. No member of the IAC dared
to challenge him as they had challenged Hillenkoetter. They were
glad to accept the consideration that he offered them, and to
retreat over the golden bridge that he had built for them.
Thus the actual situation in General SYnith's day was quite
different from that suggested by the tents quoted -- items 18-20.
No official text sets forth the doctrine of the DCI's per-
sonal responsibility and consequent authority as clearly as does
item 17 above, but it is adequately implicit in the current NSCID
No. 1, dated ~ March 1961+, which reads:
National intelligence is that intelligence
which is required for the formulation of
~ Reber's notes must have been a hasty scrawl. The phrase "in
accordance with his own judgment" is obviously out of place.
It belongs in the last sentence, after "DCI".
Approved For Release 2006/10113: CIA-RDP79R00971A000100020002-2
 Approved For Release 2006110/13: CIA-RDP79R00971A000100020002-2
national security policy, concerns more than one
department or agency, and transcends the exclusive
competence of a single department or agency, The
Director of Central Intelligence shall produce
national intelligence with the support of the U.S.
Intelligence Board. Intelligence sa produced shall
have the concurrence, as appropriate, of the members
of the U.S. Inte111gence Board or sha11 carry a
statement of any substantially differing opinion of
such a member ar of the Intelligence Ghie~ of a
Military Department.
The most striking difference between the situation in 1952
and that in 19bg is that ONE, with the support of OSR, OER, OCI,
OSI, and FMSAC, is no longer dependent on departmental contribu-
tions, as it was in 1952. That difference is the result of a
gradual, almost imperceptible, evolution over a period of lfi
years during which experience proved that departmental contribu-
tions were neither adequate nor reliable, and that independent
research capabilities within CTA were imperatively necessary to
supplement and check on them, and to stimulate departmental re-
search by breaking new ground. (Cf, item 15 above.)
One cannot foresee how far this evolution may go. General
Vandenberg conceived of an omnicompetent central office of research
and evaluation that would render redundant the departmental in-
telligence agencies -- and thereby he set the woods on fire. ORE,
his creation, failed for want of professional competence as well
as because of the inevitable hostility of the IAC agencies. Today
the general professional superiority of the CIA reseaxch offices
and the ONE staff over their counter-parts in the departmental
agencies is obvious. Every authority wants its own intelligence
agency, however, and it is not to be expected that the depart-
mental agencies wi11 dust disappear.
Since 1950 national intelligence estimates have been valued
for two qualities that are complementary rather than mutually
exclusive, although they were conceived to be in opposition until
General Smith combined them. One is the exercise of responsible,
independent, disinterested judgment by a Board of National Estimates
in the evaluation and synthesis of departmental and other contribu-
tions. The other is the fact of coordination: the assurance that
14
Approved For Release 2006!10113 : CIA-RDP79R00971A000100020002-2
 Approved For Release 2006110/13: CIA-RQP79R00971A000100020002-2
all points of view have been seriously considered and that every
relevant intelligence authority in the Government has been re-
quired either to concur in the DCT's estimate as written or else
to dissent in context. In our enthusiasm for the first of these
qualities, gratifyir~ to ourselves, we should not forget the
second.
The highest service that a member of the Board of National
Estimates is called upon to perform is to decide whether a
disputed issue can and should be resolved (or evaded), or whether
it is of such important significance that it should be clarified
by inviting a dissent. Acting, not for himself, but for the DCT,
pending the Id's awn consideration of the problem, he must take
into account both of the values set forth in the preceding para-
graph. There is no simple rule or presumption that supplies an
easy determination. That requires the exercise of informed and
responsible judgment. -- ~udwell Montague.
Approved For Release 2006110/13: CIA-RQP79R00971A000100020002-2
 Approved For Release 2006/10113 :CIA-RDP79R00971A000100020002-2
Q
Approved For Release 2006/10113 :CIA-RDP79R00971A000100020002-2