Church Committee Reports at American Buddha Online Library

Site Map

CHURCH COMMITTEE REPORTS

V. THE DIRECTOR OF CENTRAL INTELLIGENCE
ISSUES
In January 1946, President Truman established by Presidential
Directive the Xational Intelligence Authority under the direction of
the Director of Central Intelligence (DCI). The Directive authorized
the Director of Central Intelligence to plan, develop and coordinate
the foreign intelligence activities of the United States Government."
That same year, the Joint Congressional Committee on the Investigation
of the Pearl Harbor Attack described how the military
services in Washington had failed to bring all the intelligence together
about Japanese plans and int~ntions and then concluded that
"operational and intelligence work requires centralization of authority
and clear-cut allocation of responsibility." 2
Subsequently, in 1947, Congress passed the National Security Act
giving the DCI responsibility for "coordinating the intelligence activities
of the several Government departments and agencies in the
interest of national security." 3 Concurrently, the President designated
the Director of Central Intelligence as his principal foreign intelligence
adviser and established an Intelligence Advisory Committee
(later reconstituted as the United States Intelligence Board) to "advise"
the DCI in carrying out his responsibilities.4
The precise roles and responsibilities of the DCI, however, were not
cle;trly spelled out. For fear of distracting attention from the principal
objective of the 1947 National Security Act-to unify the armed
services-the White House did not delineate the DCI's functions in
any detail. 5 The Congressional debates also failed to address the extent
1 Presidential Direetive, 1/22/46, Federal Register, Vol. II, pp. 1337, 1339.
• Joint Committee on the Investigation of the Pearl Harbor Attack, Report,
pursuant to S. Con. Res. 27,7/20/46, 79th Cong., 2nd Sess., p. 254.
3 Section 102, National Security Act of 1947, 61 Statutes-at-Iarge 497-499. Provisions
of Section 102 are col1ifiel1 at 50 U.S.C. 403.
• National Security Council Intelligence Directive (NSCID) No.1, 12/12/47.
The Intelligence Advisory Committee was chaired by the DCI, and was composed
of representatives from the Departments of State, Army. Navy, and Air Force,
the Joint Chiefs of Stafl'. and the Atomic Energy Commission. In 1957, the President's
Foreign Intelligence Advisory Board recommended that the Intelligence
Advisory Committee be merged with the United States Communications Intelligence
Board to perform the overall intelligenee coordinating function more effectively.
Consequently, the United States Intelligence Board (USIB) was establishl'd
in 1958.
Under President Ford's Executive Order No. 119;)5, 2/18/76, USIB was dissolved,
but the DCI was given responsibility to "establish such committees of collectors,
producers and users of intelligence to assist in his conduct of his responsibilities."
• Draft Legislative History of the CIA, prepared by the Office of Legislative
Counsel, CIA, July, 1967; and Organizational History at OIA, 1950-1953, prepared
by the CIA, p. 27.
(71)
72
of DCI authority over the intelligence community. Rather, congressional
committees were interested in whether the DCI's primary
responsibility would be to the military services or whether he would
report directly to the National Security Council (NSC) and the
PresidenL" But the problems facing the DCI were obvious from the
beginning. According to a 1948 memorandum by the CIA's General
Counsel:
In its performance of the intelligence functions outlined
in the National Security Act, the primary difficulty experienced
by CIA has been in certain weakness of language
in paragraph 102 (d) concerning the meaning of coordination
of in~lligence activities. Where the Act sta~s "it
shall be the duty of the Agency ... to advise the National
Security Council ... [and] to make recommendations to the
National Security Council for the coordination of such in~lligence
activities;" it has been strongly argued that this places
on the Director a responsibility merely to obtain cooperation
among the intelligence agencies. This weakness of language
and the ensuing controversy might have been eliminated by
the insertion after the phrase "it shall be the duty of the
Agency," the following words: "and the Director is hereby
empowered," or some other such phrase indicating the intent
of Congress that the Director was to have a controlling voice
in the coordination, subject to the direction of the National
Security Council.7
Under Senate Resolution 21, the Select Committee has undertaken
for the first time since 1947 a studv of the manner in which the
successive Directors of Central In~lligence have carried out their
responsibilities, in an effort to determine: (1) whether the Del's
assigned resnonsibilities are proper and sufficient; (2) whether the
DCI has sufficient authority to carry out these responsibilities; (3)
'whether the DCI should continue as Director of the Central Intelligence
Agency, if he is to playa leadership role for the entire intelligence
community; and (4) whether Congress should enact more
explicit or different definitions of the Del's responsibilities.
• Hearings before the Senate Armed Services Committee on S. 758, pp. 173-176,
and Hearings before the House CDmmittee on Expenditures in the Executive
Departments on H.R. 2139 (1947). During the House hearings, Representative
Hale Boggs commented:
"I can see ... even if this hill becomps law, as pre~entlv set np, a great deal of
room for confusion on intelligence matters. Here we have the Director of the
Central Intelligence Agency, responsible to the National Security Council, and
yet the Director is not a member of that Council, but he has to I!et all of his
information down throug-h the chair of the Secretary of National Defense, and
all the other agencies of Government in addition to our national defense agencies.
. . . I ju·.t ennnot quite .•ee how the man is noing to carry out his functions
there without a nreat deal of confu.<Jion, and really more opportunity to put the
blame on somebody else than thpre is now."
~f'£'rptaryof the Navy Jamps Forrestal replipd:
"Well. if you have an organization, Mr. Boggs, in which men have to rely
nnon placing the blame, ... you cannot run any organization, and it goes to
the root really of this whole question. This thing will work, ·and I have Imid
from the beninning it would only work, if the components want it to work."
[Emnhasis added.l
7 Memorandum from Lawrence R. Houston to the Director, 5/7/48.
73
INTRODUCTION
The Pearl Harbor intelligence failure was the primary motivation
for establi,hing a Director of Central Intelligence. President Truman
desired a national intelligence organization which had access to all
information and would be headed by a Director who could speak authoritatively
for the whole community and could insure that the community's
operation served the foreign policy needs of the President and
his senior advisers.8 President Truman and subsequent Presidents have
not wanted to rely exclusively on the intelligence judgments of departments
with vested interests in applying intelligence to support a particular
foreign policy or to justify acquiring a new weapons system.
However, the DCI's responsibility to produce national intelligence
and to coordinate intelligence activities has often been at variance with
the particular interests and prerogatives of the other intelligence
community departments and agencies. During the Second World War,
the Department of'State and the military services developed their own
intelligence operations. Despite establishment of the Director of Central
Intelligence in 1946, they have not wanted to give up control over
their own intelligence capabilities. The military services particularly
have argued that they must exercise direct control over peacetime intelligence
activities in order to be prepared to conduct wartime military
operations. The State and Defense Departments have steadfastly opposed
centralized management of the intelligence community under
theDCL
However, over time the actual degree of conflict between the Del's
responsibility to coordinate intelligence activities and the interests of
the other parts of the community has depended on how broadly each
DCI chose to interpret his coordination responsibilities and how he
allocated his time between his three major roles.9 The three roles the
Del plays are: (1) the producer of national intelligence; (2) the
coordinator of intelligence activities; and (3) the Director of the Central
Intelligence Agency.
A. THE PRODUCER OF NATIONAL INTELLIGENCE
As the President's principal foreign intelligence adviser, the DOl's
major responsibility is to produce objective and independent national
8 Harry S. Truman, Memoirs. Vol. II, p. 58.
• QUESTION: When you were DCI. did you fppI that institutionally or functionally
your position was bumping heads with the DOD intelligence apparatus in different
ways or not, and if not, why not, in view of the structure?
:\Ir. SCHLESINGER: 'YeN, historically there hqvA been intervening periods of
open warfare and detente ... Prior to these, one of the problems of the intelligence
community has been the warfare that exists along jurisdictional boundaries, and
this tended to erupt in the period of the 1960's, in particular when they were
introducing a whole set of new technical collection capabilities; that open warfare
was succeeded by a period of true detente, but the problem with such detente is
that it tends to be based on marriage contracts and the principle of good fences
make good neighbors, and that a mutual back-scratching and the like, so that
you do not get effective resource manalrement under those circumstances. (James
Schlesinger, testimony, 2/2/76, pp. ~O.)
207-932 0 - 76 - 6
74
intelligence for senior policymakers.10 In so doing, he draws on a
variety of collection methods and on the resources of the departmental
intelligence organizations as well as CIA analystsY But the DCI
issues national intelligence and is alone responsible for its production.12
The most important national intelligence which the DCI produces
is the Xational Intelligence Estimate (XIE). An NIE pre~ents the
intelligence community's current knowledge of the situation in a
particular country or on a specific topic and then tries to estimate what
is going to happen within a certain period of time. NIEs are prepared
for use by those in the highest policy levels of government and represent
the considered judgment of the entire communityY Major
differences of opinion within the intelligence community are illuminated
in the text or in the footnotes. When an NIE is released, however,
it is the DCI's own national intelligence judgment, in theory free from
departmental or agency biases.H
To carry out this responsibility to produce independent and objective
national intelligence, DCI Walter Bedell Smith established the
Board of National Estimates in 1950. The Board was comprised of
senior government officials, academicians and intelligence officers and
had a small staff known as the Office of National Estimates (ONE).
One member of the Board would be responsible for supervising the
drafting of the estimates by the ONE staff, for reviewing these judgments
collectively for the DCI, and for adjudicating disputes within
the community. 'Vhen the United States Intelligence Board reviewed
an XIE, the DCI could have confidence in the opinions expressed in
the estimate because each estimate reflected the collective jud,gment of
his own Board. According to the former chairman of the Board of
National Estimates, John Huizenga:
The Board of National Estimates in fact functioned as a kind
of buffer. It provided procedures by which the departmental
views could be given a full and fair hearing, while at the same
time ensuring that the DCI's responsibilities to produce intelligence
from a national viewpoint could be upheldY
10 According to NSCID No.1, 2/17/72, national intelligence is that intelligence
required for the formulation of national security policy and concerning more
than one department or agency, It is distinguished from departmental intelligence,
which is that intelligence in support of the mission of a particular
department.
11 Prior to Pre,oiopnt Ford'~ EXPf'lltivp Order No. 119fl5, ?/lS/7fl, the United
States Intelligence Board, composed of representatives from the various agencies
and departments of the intelligence community, formally reviewed the DCI's
national intelligence judgments.
12 Under Pre~;dent Ford'" EXPf'llth·p Ordpr No, 1100'5. 2/11'1/76, thp Del will
have responsibility to "supervise production and dissemination of national intelligence."
13 At present, the DCI briefs the Congress on the judgments contained in his
NIEs. The Congress does not receive the DCI's NIEs on a regular basis.
14 In his role as CIA Director, the DCI also producps current intelligence and
research studies for senior policymakers, These intelligence judgments are prepared
by CIA analysts who are supposed to be free from departmental preferences.
Such current reporting is not formally reviewed by the other members of
thp intelligence community, but is often informally coordinated.
15 John Huizenga testimony, 1/26/76, p. 11.
75
In 1973, Colby replaced the Board and the ONE staff with a new
system of eleven National Intelligence Officers (NIOs). Each NIO has
staff rf'sponsibility to the DCI for intelligence collection and production
activities in his geographical or functional specialty. The NIOs
coordinate the drafting of NIEs within the community. They do not,
however. collectively review the final product for the DC1.16 Director
Colby testified that. he thought the Board of National Estimates tended
to fuzz over differences of opinion and to dilute the DCI's final
intelligence judgments.l1
In the course of its inn'stigation, the Committee concluded that the
most critical problem confronting the DCI in carrying out his responsibility
to produce national intelligence is making certain that his intelligence
judgments are in fact objectire and independent of departmental
and agency biases. Howe\'er, this is often quite difficult. A most
delicate relationship exists between the DCI and senior policymakers.
According to John Huizenga:
There is a natural tension between intelligence and policy,
and the task of the former is to present as a basis for the decisions
of policymakers as realistic as possible a view of forces
and conditions in the external em·ironment. Political leaders
often find the picture presented less than congenial. ...
Thus, a DCI who docs his job well \vill more often than not
be the bearer of bad ne\vs, or at least will make things seem
disagreeable, complicated, and uncertain.... ",Vhen intelligence
people are told, as happened in recent years, that they
were expected to get on the team, then a SOUild intelligencepolicy
relationship has in effect broken down.18
In addition, the DCI must provide intelligence for cabinet officers
who often have vested interests in recBiving information which supports
a particular foreign policy (State Department) or the acquisi-
16 Under the KIO system, the Defense Intelligence Agency (DIA) and the
military services have assumed greater responsibility for the initial drafting of
military estimates. Because 1'\IOs have no separate staff, they must utilize experts
in the community to draft sections of the estimatt's. In 1975, DIA prepared the
first drafts of two chapters of the XIE on Soviet offensive and defensive strategic
forces. Colby contends that as a eonsequenee, anal~'sts throughout the community
felt more involved. (William Colby testimony, 12/11/75.)
17 According to Colby:
"A board? You say why don't you have a board also? I have some reservation
at the ivory tower kind of problt'm that you gt't out of a board whieh is too
st'parated from tht' rough and tumble of the real world. I think there is a tendency
for it to intt'llt'C'tualizt> and then writt' sermons and appreciations....
"I think there is a tendency to beeome institutionally committed to an approach
and to an appraisal of a situation and to begin to interpret new events against
the light of a predetermint'd approach toward those eYents. I think that has bt'en
a bother. I like the idea of an indiYidual total responsibility, one man or woman
totally responsible, and then you don't get any fuzz about how there was a vote,
and therefort' I really didn't like it but I went along and all that sort of thing,
one pt'rson totally respon~iblp, I think. is a good way to do it. That can be the
Director or whateYt'r you set up. But I do like that idpa of separating out and
making one individual totally responsiblt' so there's nobody else to go to, and
there's no way of dumping the responsibility onto somebody else. That really
is my main problem with the board, that it diffuses responsibility, that it does
get out of the main line of tbe movement of material. (Colby, 12/11/75, pp. 36-37.)
1B Huizenga, 1/26/76, pp. 13-14.
76
tion of a new weapon system (Department of Defense).19 The
President and NSC staff want confirmation that their policies are succeeding.
Moreover, each NIE has in the past been formally reviewed
by other members of the intelligence community. Although CIA
analysts have developed expertise on issues of critical importance to
national policymakers, such as Soviet strategic programs, most DCIs
ha,'e been reluctant to engage in a confrontation with members of the
USIB over substantive findings in national intelligence documents.2o
According to .John Huizenga:
The truth is that the DCI, since his authority over the intelligence
process is at least ambiguous, has an uphill struggle
to make a sophisticated appreciation of a certain range of
issues prevail in the national intelligence product over against
the parochial views and interests of departments, and especially
the military departments.21
Finally, the DCI's own analysts in CIA are sometimes accused of
holding an "institutional" bias. According to James Schlesinger:
The intelligence directorate of the CIA has the most competent,
qualified people in it, just in terms of their raw intellectual
capabilities, but this does not mean that they are free
from error. In fact, the intelligence directorate tends to make
a particular type of error systematically in that the intelligence
directorate tends to be in close harmony with the prevailing
biases in the intellectual community, in the university
community, and as the prevailing view chang,es in that
community, it affects the output of the intelligence
directorate.22
In particular, CIA analysts are sometimes viewed as being predisposed
to provide intelligence support for the preferences of the arms
control community. According to Schlesinger:
For many years it was said, for example, that the Air
Force had an institutional bias to raise the level of the Soviet
threat, and one can argue that in many cases that it did and
that was a consequence.
19 According to Huizenga:
"It should IlP recognized that thE' approach of an operating department to intelligE'nce
issues is not invariably disinterested. The Department of State sometimes
has an interest in having intelligence take a certain view of a situation because
it has a heavy investment in an ongoing line of policy, or because the Secretary
has put himself on record as to how to think about a particular problem. In the
Defense Department, intelligence is often seen as the servant of desired policies
and programs. At a minimum there is a strong organizational interest in seeing
to it that the intelligence provides a vigorous appraisal of potential threats. It
is not unfair to say that because of the military leadership's understandable desire
to hedge ap'a;nst the unexpectE'd. to provide capabilitiE's for all conceivable
contingencies there is a natural thrust in military intelligence to maximize
threats and to oversimplify the intentions of potential adversaries. It is also
quite naturally true that military professionals tend to see military power as the
prime determinant of the behavior of states and of the movement of events in
international politics." (Huizenga, 1/26/76, pp. 11-12.)
'" [hid .. p. 11.
21 Ibid., p. 12.
22Schlesinger, 2/2/76, pp. 24-25.
77
But there deYeloped an institutional bias amongst the
analytic fraternity which ran in the opposite direction. There
was an assumption that the Soviets had the same kind of arms
control objectives that they wished to ascribe or persuade
American leaders to adopt, and as a result there was a steady
upswing of Soviet strategic capabilities, and the most serious
problem, it seems to me, or the most amusin~ problem developed
at the close of the cycle when the Sonets had actually
deployed more than 1,000 ICBMs, and the NIEs, as I recall
it, were still saying that they would deploy no more than 1,000
ICBMs because of the prevailing belief in the intelligence
analytic fraternity that the Soviets would level off at 1,000
just as we had.
So one must be careful to balance what I will call the
academic biases amongst the analysts with the operational
biases amongst other elements of the intelligence community.23
Consequently, on the occasions when the DCI does support his own
staff's recommendations over the objection of the other departments,
the objectivity of the national intelligence product may still be undermined
by the bias of CIA analysts.
Recognizing all these difficulties, the Select Committee has investigated
two particularly difficult eases for Director Helms in an effort
to illustrate the problems the DCI confronts in carrying out his responsibility
to produce objective and independent national intelligence.
During the summer and fall of 1969, the White House and then the
Secretary of Defense indirectly pressured the DCI to modify his
judgments on the capability of the new Soviet SS-9 strategic missile
system. The issues under debate were: (1) whether the SS-9 was a
MIRV (Multiple Independently Targeted Re-entry Vehicle) missile;
and (2) whether the Soviets were seeking to achieve a first strike capability.
The intelligence judgments on these points would be critical
in decisions as to whether the United States would deploy its own
MIRV missiles or try to negotiate MIRV limitations in SALT (the
Strategic Arms Limitation Talks), and whether the United States
would deploy an Anti-Ballistic Missile (ARM) system to protect the
United States Minuteman missile force against a Soviet first strike.
On the first issue, in June 1969, the President's Special Adviser for
National Security Affairs, Henry Kissinger, called Director Helms to
the 'White House to discuss an l'Stimate on Soviet strategic forces. Kissinger
and the NSC staff made clear their dew that the new Soviet missile
was a MIRV and asked that Helm's draft be rewritten to provide
more evidence supporting the DCI's judgment that the SS-9 had not
demonstrated a MIRV capability. In response, the Chairman of the
Board of National Estimates rewrote the draft, but he did not change
the conclusion: All seven tests of the SS-9 were MRVs (Multiple Reentry
Vehicles) ; they were certainly not independently guided after
23 Schlesing-er, 2/2/76, pp. 26-27. eTA analysts are also sometimes accused of
being biased in favor of the clandestine intelligence collected by their own agency.
This charge is not, however, supported by a CIA study of what kinds of reporting
eTA analysts themselves find KEY in writing their intelligence memoranda. For
FY 1974, while eTA analysts considered clandestine reporting to be important,
overt Stat~ Departm\'nt reporting on political and economic SUbjects was cited
more frequently as KEY. (Annual DDI Survey, FY 1974.)
78
separation from the launch nhie1e.24 According to testimony by three
Board members, at the time they saw nothing improper in a 'White
House request to redraft the estimate to include more evidence. However,
in this case, they interpreted the \Vhitc House request as a subtle
and indirect effort to alter the Dcrs national intelligence judgment.25
On the second issue, three months .later, Helms decided to delete a
paragraph in the Board of National Estimates' draft on Soviet strategic
forces after an assistant to Secretary of Defense Laird informed
Helms that the statement contradicted the public position of the
Secretary.26
The deleted paragraph read:
\Ve believe that the Soviets recognize the enormous difficulties
of any attempt to achieve strategic sUlwriority of such
order as to significantly alter the strategic balance. Consequently,
,,,e consider it highly unlikely that they will attempt
within the period of this estimate to achieve a first-strike
capability, i.e., a capability to launch a surprise attack against
the U.S. with assurance that the USSR ,,,ould not itself receive
damage it would regard as unacceptable. For one thing, the
Soviets would almost certainly conclude that the cost of
such an undertaking along with all their other military commitments
would be prohibitive. More important, they almost
certainly would consider it impossible to develop and deploy
the combination of offensive and defensive forces necessary
to counter successfully the various elements of U.S. strategic
attack forces. Finally, even if such a project were economically
and technically feasible the Soviets almost certainly would
calculate that the U.S. would detect and match or overmatch
their efforts.27
Subsequently, the State Department representative on the United
States Intelligence Board inserted the deleted paragraph as a footnote.
.. In a memorandum to the USIB representatives, dated 6/16/69, the Director
of the Office of i'ational Estimates, Abbot Smith, stated:
"The Memorandum to Holders of NIE 11-8-68, approved by USIB on 12 June
was discussed at a meeting with Dr. Kissinger and othprs on Saturday. Out of
this meeting came requests for (a) some reordering of the paper; (b) clarification
of some points; and (c) additional argument pro and con about the
MRV-:\IIRV problem. We have accordingly redrafted the paper with the-se requests
in mind. No changes in estimates were asked, nor (we think) have been
made. But the details call for coordination."
See also, staff summary of Carl Duckett interview, 6/13/75.
25 Staff summaries of interviews with John Huizenga, 7/9/75; Abbot Smitb,
8/2/75; Williard Mathias, 7/7/75.
26 Memorandum from Director Helms to USIB Members, 9/4/69, and staff summary
of Abbot Smith interview, 8/2/75.
According to William Baroody, Secretary Laird's Special Assistant:
"I am fairly confident that 1 did not specifically bring pressures to bear on
the Director of Central Intelligence to delete or change any particular paragraph.
~We did discu~~ the differences at the timp betwepn. as these documents refresh
my mpmory, hetween thp DTA roncpru of that narticular paragraph and the CIA
es~mat('." (William Baroody testimony, 2/27/76, p. 4.)
Draft NIE 11-8-69. approved by the Board of National Estimates prior to
the USIB meeting on August 28, 1969.
79
These are stark, and perhaps exceptional, examples of White House
and Defense Department pressures on the DCI, but they illustra~ the
kinds of buffeting with which the DCI must contend. Director Helms
testified:
A national intelligence estimate, at least when I was Director,
was considered to be the Director's piece of paper.
USIB contributed to tile process but anylJody could contribute
to tile process, the estimates staff, individuals in the 'White
House. And the fact that a paragraph or a sentence was
changed or amended after USIB consideration was not
extraordinary. . . .
So this question which seems to have come up about somebody
influencing one aspect or influencing another aspect of
it, the whole process \vas one of influences back and forth,
some in faVOl' of this and some in faVOl' of that. , . .
So that was the system then. I dont know what is the system
now, but on this issue of the first strike capability one
of the things that occurred in connection with that was a
battle royale over whether it was the Agency's job to decide
definitively whether the Soviet Union had its first strike
capability or did not have a first strike capability. And this
became so contentious that it seemed almost impossible to
get it resolved.
I have forgotten just exactly what I decided to do about
the whole thing, but I don't know, I think it was back in '69.
There was a question about certain footprints and MRVs and
things of this kind, and some people felt that they were very
important footprints and other people thought they were
ul11mportant footprints, and there's no question there's a
battle royale about it.
However, it was resolved however. If you felt that there
was pressure to eliminate one thing, there was a manifold
pressure to put in something else.
But anyway, I don't really see an issue here.28
vVhile Helms may not see an issue here, the Committee found that
constant tension exists between the DCI, \vhose responsibility it is to
produce independent and objective national intelligence, and the agencies,
who are required to cooperate in this effort.
A second case investigated by the Select Committee illustrates the
potential problems the DCI confronts in producing relevant national
intelligence for senior policymakers planning highly sensitive military
operations. In April 1970, following Prince Sihanouk's ouster,
United States policymakers decided to initiate a military incursion
into Cambodia to destroy Korth Vietnamese sanctuaries. In making
this decision, these policymakers had to rely on an earlier (February)
NIE and current reporting from the various departments and agencies.
They never received a formal DCI national intelligence estimate
or memorandum on the political conditions inside Cambodia after
Sihanouk's departure or on the possible consequences of such an
American incursion. "'hy? Because Director Helms decided in April
not to send such an estimate to the NSC.
28 Richard Helms testimony, 1/30/76, pp. 59-61.
80
In April 1970, analysts in the Office of National Estimates prepared
a long memorandum entitled "Stocktaking in Indochina: Longer
Term Prospects" which included discussion of the broad question of
future developments in Cambodia, and addressed briefly the question
of possible United States intenention: 29
Nevertheless, the governments of Laos and Cambodia are both
fragile, and the collapse of either under Communist pressure
could haYe a si~nificant adverse psychological and military
impact on the sItuation in South Vietnam.... Because the
events in Cambodia and their impact are harder to predict,
if Hanoi could be der,ied the use of base areas and sanctuaries
in Cambodia, its strategy and objectives in South Vietnam
would be endangered. Hanoi is clearly concerned over such a
prospect. Cambodia, however, has no chance of being able to
accomplish this by itself; to deny base areas and sanctuaries
in Cambodia would require heavy and sustained bombing
and large numbers of foot soldiers which could only be supplied
by the U.S. and South Vietnam. Such an expanded
allied effort could seriously handicap the Communists and
raise the cost to them of prosecuting the war, but, however
successful, it probably would not prevent them from continuing
the struggle in some form. 30
Helms received this draft memorandum 13 days before the planned
United States incursion into Cambodia. Then the day before the incursion
began, Helms decided not to send the memorandum to the
White House. A handwritten note from Helms to the Chairman of
the Board of National Estimates stated: "Let's take a look at this on
June 1, and see if we would keep it or make certain revisions."
The Committee has been unable to pinpoint exactly why Director
Helms made this decision.31 One member of the Board of National
Estimates recalled that Helms would have judged it "most counterproductive"
to send such a negative assessment to the White House.32
George Carver, Director Helms' Special Assistant for Vietnamese Affairs
in 1970, objected to this conclusion that Helms refrained from
sending the memorandum forward because he thought the message
29 DCI Helms encouraged the analysts to prepare such a memorandum for the
White House. On an early draft, Helms commented to Abbot Smith, Chairman
of the Board of National Estimates: "O.K. Let's develop the paper as you suggest
and do our best to coordinate it within the Agency. But in the end I want
a good paper on this subject, even if I have to make the controversial judgments
myself. We owe it to the policymakers I feel." (Richard Helms, 4/7/70.)
30 "Stocktaking in Indochina: LOl.ger Term Prospects," ONE memorandum,
4/17/70, para. 69.
31 Helms told the Committee:
"Unfortunately my memory has become hazy about the reason's for decisions
on the papers you identify.... In a more general way let me try to be help-ful
to you (I will assume that you have or will talk to [George] Carver and that
you will giyp re~sonahle weight to hi~ c~mments. In the first p'are, it is almost
impossible at this late date to recreate all the relevant circumstances and considerations
whirh went into decisions of the kind you are examinin!\" made six
years ago. Secondly, it is dangerous to examine eXhaustively one bead to the
f'xc'u~inn of other he"~s In the necklace." (Telegram from Richard Helms to the
Select Committee, 3/23/76.)
112Staff summary of James Graham interview, 2/5/76.
81
would be unpalatable or distres.'3ing to the White House.33 Rather,
Can'er ~rgued that Helms judged that it would not be ~ppropriate
to send forward a memorandum drafted by analysts who dId not know
about the planned U.S. military operation.
According to Carver's testimony, Helms was told in advance about
the planned incursion under the strict condition that he could not
inform other intel1igence analysts, including the Chairman of the
Board of Xational Estimatrs and the CIA intelligence analysts working
on Indochina questions. Then because the analysts were not informed.
Helms decided not to send fonvard their memorandum on
Indochina.
According to Carver:
He [Helms] thought that it might be unhelpful, it might
indeed look a little fatuous, because the people who had prepared
it and drafted it were not aware that the U.S. was on
the verge of making a major move into Cambodia, hence their
commentary \vas based on the kind of unspoken assumption
that there was going to be no basic ope.rational change in the
situation, as they projected over the weeks and months immediately
ahead.34
Further, Carver speculated that Helms proba:bly felt he would not
be listened to if it were immediately open to the counterattack that the
analysts did not know of the planned operations.34a In effect, Carver
argues that in carrying out the President's restriction on discussing
the planned operations, Helms denied his analysts the very information
he considered necessary for them to have to provide intelligence
judgments for senior policymakers. Helms took this decision even
though the memorandum in question included a judgment on the possible
consequences of United States intervention in Cambodia.
Thus, for whatever combination of reasons, in the spring of 1970
prior to the Cambodia incursion, the DCI did not provide senior
policymakers formal1y \vith a national intelligence memorandum
which argued that the operation would not succeed in thwarting the
Korth Vietnamese effort to achieve control in Indochina.
Six \yceks later, while the Cambodia incursion was still undenvay,
the State Department requested a Special NIE(SNIE) on North
Vietnamese intentions which would include a section on the impact of
the United States intervention in Cambodia. A draft estimate was
prepared and coordinated "'ithin the intelligence community, just as
the incursion \vas ending. The estimate began with a number of caveats
such as : "Considerable difficulties exist in undertaking this analysis at
this time. Operations in Cambodia are continuing and the data on results
to date is, in the nature of things, incomplete and provisionaL"
The draft went on to say that assessing Hanoi's intentions is always a
difficult exercise but "even more complicated in a rapidly moving situ-
33 George Can'er testimony, 3/5/76, p. 30.
" Ibid .. p. 10. Carver told the Committee that his overall judgments were "based
on what I am reasonably convinced is a recollection of a series of conversations,
although I cannot cite to you a specific conversation or give you a Memorandum
for the Record that says that." (Ibid., p. 15.)
". Ibid., pp. 22-23.
82
at.ion, in which there are a number of unknown elements, particularly
wIth respect to n.s. and Allied courses of action." \Vith respect to the
situation in Cambodia, the estimate concluded:
Although cardul analysis of these losses suggests that the
Communist situation is by no means critical, it is necessary to
retain a good deal of caution in judging the lasting impact
of the Cambodian affair on the Communist position in Indochina.
35
Despite all these qualifications, Helms again decided not to send the
estimate to the \Vhite House. \Vhile Helms does not recall the reasons
for his decision, he did tell the Committee:
In my opinion there is no way to insulate the DCI from unpopularity
at the hands of Presidents or policymakers if he
is making assessments which run counter to administrative
policy. That is a built-in hazard of the job. Sensible Presidents
understand this. On the other hand they are human
too, and in my experience they are not about to place their
fate in the hands of any single individual or group of individuals.
In sum, make the intelligence estimates, be sure they
reach the President personally, and use keen judgment as to
the quantity of intelligence paper to which he should be subjected.
One does not want to lose one's audience, and this is
easy to do if one overloads the circuit. No power has yet been
found to force Presidents of the United States to pay attention
on a continuing basis to people and papers when confidence
has been lost in the originator.36
Nevertheless, as John Huizenga testified:
In times of political stress on intelligence, there is more a
question of invisible pressures that might cause people to feel
that they were being leaned upon, even though nobody asked
them to take out some words or add some words ... When
intelligence producers have a general feeling that they are
working in a hostile climate, what really happens is not so
much that they tailor the product to please, although that's
not been unknown, but more likely, they avoid the treatment
of difficult issues.37
In the end, the DCI must depend on his position as the President's
principal intelligence adviser or on his personal relationship with the
President to produce objective and independent national intelligence.38
Organizational arrangements such as the Board of National Estimates
may, nevertheless, help insulate the DCI from pressures; but
35 Draft SNIE 14-3-70.
36 Telegram from Richard Helms to the Select Committee, 3/23/76.
37 Huizenga, 1/26/76, pp. 20-21.
38 John Huizenga testified that "there were very few instances of gross interference."
While "it's fair to say [the Cambodia and SS-9 cases] were gross, partiCUlarly
the SS-9 case," objectivity and independence are difficult to uphold
when political consensus breaks down over foreign policy issues. Huizenga
concluded, "the experil'nce of these years persuade me that we have yet to prove
that we can have in times of deep political division over foreign policy a professional,
independent, objective intelligence system." (Huizenga, 1/26/76, p. 9.)
83
only if they are used. In the cases of the 88-9 and Cambodia, Helms
took the de"cisions without consulting with the Board collectively.
B. COORDINATOR OF INTELLIGENCE ACTIVITIES
1. The Intelligence Process
In theory, the intelligence proces.'S works as follows. The President
and members of the N8C-as the major consumers of foreign intelligence-
define what kinds of information they need. The Director
of Central Intelligence with the advice of other members of the intelligence
community establishes requirements for the collection of different
kinds of intelligence. (An intelligence requirement is defined
as a consumer statement of information need for which the information
is not already at hand.) Resources are allocated both to develop
new collection systems and to operate existing systems to fulfill the
intelligence requirements. The collection agencies-the National Security
Agency (NSA), CIA, DIA, and the military services--11Ulnage
the actual collection of intelligence. Raw intelligence is then assembled
by analysts in CIA, DIA, the State Department, and the military
services and produced as finished intelligence for senior policymakers.
In practice, however, the process is much more complicated. The
following discussion treats the Committee's findings regarding the
means and methods the DCI has used to carry out his responsibility
for coordinating intelligence community activities.
2. jJfanaging Intelligence Oollection
Although the responsibility of the DCI to coordinate the activities
of the intelligence community is most general, the DCIs have tended
to interpret their responsibility narrowly to avoid antagonizing the
other departments and agencies in the intelligence community. While
DCIs have sought to define the general intelligence needs of senior
United States policymakers, they have not actually established intelligence
collection requirements or chosen specific geographical targets.
The individual departments establish their own intelligence collection
requirements to fulfill their perceived national and departmental
needs. For example, DIA compiles the Defense Intelligence Objectives
and Priorities document (DIOP) which is a single statement of intelligence
requirements for use by all DOD intelligence components, in
particular, Defense attaches, DIA production elements, the intelligence
groups of the military services, and the military commands. The
DIOP contains a listing by country of nearly 200 intelligence issues
and assigns a numerical priority from one to eight to each country and
topic. The State Department sends out ad hoc requests for information
from United States missions abroad. Although the Department
does not compile a formal requirements document, Foreign Service
Officer reporting responds to the information needs of the Secretary
of State.
In the absence of authority to establish intelligence requirements,
the DCI relies on issuing general collection guidance to carry out his
coordinating responsibilities. The DCI annually defines United States
substantive intelligence priorities for the coming year in a DCI Directive.
This sets out an elaborate matrix arraying each of 120 countries
against 83 intelligence topics and assigning a numerical priority from
84
1 to 7 for each country and topic combination. Since 1973, the DCI
has also distributed a memorandum called the DCI's "Perspectives"
which defines the major intelligence problems policymakers will face
O\'er the next five years; a memorandum known as the DCI's "Objectives"
which details the general resource management and substantive
intelligence problems the community \\"ill face in the upcoming year;
and the DCI's "Key Intelligence Questions" (KIQs) which identify
topics of particular importance to national policymakers.
All these documents have in the past been reviewed by members of
the intelligence community on USIB, but the DCI cannot compel the
departments and agencies to respond to this guidance. For example,
the Defense Intelligence Objectives and Priorities "express the
"spectrum of Defense intelligence objectives and priorities geared
specifically to approved strategy" derived from the Joint Chiefs of
Staff. But the DIOP does not include a large number of economic,
political and sociological questions which the Defense Department
considers inappropriate for it to cover. Consequently, Defense-controlled
intelligence assets do not give priority to non-military questions
even though such questions are established as priorities in the
DCI's guidance.
In addition, through three intelligence collection committees of the
United States Intelligence Board, DCIs have tried in the past to reconcile
the different departmental requirements and to insure th?t the
interests of the entire community are brought to bear in the intelligence
collectors' operations.39 The Committee on Imagery Requirements and
Exploitation (COMIREX) dealt with photographic reconnaissance.4o
The SIGINT Committee coordinated the collection of signals and
communications intelligence.41 The Human Resources Committee dealt
with overt and clandestine human collection!2
In the collection of overhead photography and signals intelligence,
the DCI through the COMIREX and SIGINT Committees provides
guidance as to targets and amounts of coverage. These Committees also
administer a complex accounting system designed to evaluate how
well, in technical terms, the specific missions have fulfilled the various
national and departmental requirements. Because of the nature of overhead
collection, the whole community can participate in selecting the
targets and in evaluating its success. The operating agency is responsive
solely to requirements and priorities established by the USrB
committees. At the same time, the DCI alone cannot direct which
photographs to take or when to alter the scope of coverage. The role
of the DCI is to make sure that the preferences of the entire community
are taken into account when targets are chosen.
39 Under President Ford's Executive Order No. 11905, these three collection
committees will probably continue under the DCI's responsibility to establish
"such committees of collectors, producers, and users to assist in his conduct of
his rt'spon~ibi~ities."
'" In 1955, Richard Bissell. a Special Assistant to the DCI, set up an informal
Ad-Hoc Requirements Committee (ARC) to coordinate collection requirements
for the U-2 reconnaissance program. l\1ember~hip initially included repre~entatives
of CIA, the Army, Navy, and Air Force. Later representatives of NSA, the
.Toint Chief~ of Staff. fino the State Department were added. Tn 1960. with the
development of a new overhead reconnaissance system, the ARC was supplanted
by a formal USIB Committee, the Committee on Overhead Reconnaissance or
85
For example, prior to the Middle East war in 1973, .the USIB
SIGINT committee recommended that the Middle East he a priority
target for intelligence collection if hostilities broke out, and asked
NSA to evaluate the intelligence collected and to determine appropriate
targets. vVhen the war broke out, NSA implemented this USIB
guidance. Later in the week, the same committee discussed and approved
DIA's recommendation to change the primary target of one
collector. The DCI did not order the changes or direct what intelligence
to collect, but through the USIB mechanism he insured that
the community agreed to the retargeting of the system.
The DCI has been less successful in involving the entire intelligence
community in establishing collection guidance for NSA operations or
for the clandestine operations of CIA's Directorate of Operations.
These collection managers have substantial latitude in choosing which
activities to pursue; and the DCI has not yet established a mechanism
to monitor how well these collectors are fulfilling the DCI's community
guidance.
During 1975, USIB approved a new National SIGINT Requirements
System, an essential feature of which requires USIB to initiate a
formal community review and approval of all SIGINT requirements.
In addition, each requirement must contain a cross reference to pertinent
DCI priorities and specific KIQs. However, this system does
COMOR. COMOR's responsibilities included coordination of collection requirements
for the development and opemtion of all overhead reconnaissance systems.
As these programs grew and the volume of photographs increased, serious problems
of duplication in imagery exploitation prompted the DCI and the Secretary
of Defense to establish a special joint review group. Subsequently, it recommended
the establishment of the :1'\ational Photographic Interpretation Center
(NPIC) and the creation of a new USIB Committee to coordinate both collection
and exploitation of national photographic intelligence. In 1967, COl\IIREX was
established.
41 During 'Vorld 'Var II, the military services controlled all communications
intelligence. After the war, a U.S. Communications Intelligence Board (USCIB)
was established to coordinate CO~nNT activities for the NSC and to advise the
DCI on COMINT issues. However, in 1949 the Secretary of Defense set up a
separate COl\UNT board under the Joint Chiefs of Staff to oversee the military's
COMINT activities, and this arrangement stood for three years, despite the
DCI's objections. In 1952, XSA was established with operational control over
COl\IIXT resources and the Secretary of Defense was given executive authority
over all COl\IINT activities. At the same time, the USCIB was reconstituted
under the chairmanship of the DCI to advise the Director of NSA and the
Secretary of Defense. In 1958, the USeIB was merged with the Intelligence
Advisory Committee to form the United States Intelligence Board. The COMINT
Committee of the USIB was formed soon thereafter; this became the SIGINT
Committee in 1962 when its responsibilities were extended to include ELINT.
.. General Bennett, Director of the Defense Intelligence Agency, proposed in
1970 the establishment of a USIB subcommittee to provide a national-level forum
to coordinate the various human source collection programs, both overt and
clandestine. Following objections from the CIA's Directorate of Operations,
Director Helms decided instead to establish an ad hoc task force to study the
whole range of HU~IINT problems. After a year's study, the task force recommended
the establishment of a VSIB committee on a one-year trial basis. The
President·s Foreign IntelJic:ence ..\dvi~ory Board (PFIABl. in fI separate study,
also endorsed the idea. Subsequently, the Human Sources Committee was accorded
permanent status in .June 1974 and in 1975 its name was changed to the
Human Resources Committee.
86
not vest in the DCI operational authority over NSA and its collection
systems.43 The Directo~ of XSA 'yill ~till detel:mine which spe?i~c
communications to mOllltor and ,vlnch sIgnals to mtercept. In a CrISIS,
the Secretaries of State and Defense ancl the military commanders will
continue to be able to task NSA directly and inf'Orm the DCI and
the SIGINT Committee afterwards.
In contrast to technical intelligence collection where the DCI has
sought expanded community involvement in defining requirements,
DCIs have not been very receptive to Defense Department interests in
reviewing CIA's clandestine intelligence collection. In part, the DCIs
have recognized the difficulty of viewing human collection as a whole,
since it comprises many disparate kinds of collectors, some of which
are not even part of the intelligence community. For example, Foreign
Service Officers do not view themselves as intelligence collectors,
despite the large and valuable contribution FSO reporting makes to the
overall national human intelligence effort. In addition. the CIA's
Clanclestine Service (DDO) has lobbied against a USIB Human
Sources Committee, fearing that it would compromise the secrecy of
their very sensitive operations.44
So DCIs, as Directors of the agency responsible for collecting
intelligence clandestinely, resisted establishment of a permanent
rSIB committee to review human collection until 1974.45 When
established, the Committee was specifically not given responsibility
for reviewing the operational details or internal management of the
individual departments or agencies. In the case of "sensitive" information,
departments and agencies were authorized to withhold information
from the Committee and report directly to the DCI.
It is not surprising, therefore, that the Human Resources Committee
has only just begun to expand community influence over human
collection. The Committee issues a general guidance document called
the Current Intelligence Reporting List (CIRL). Although the military
makes some use of this document, the DDO instructs CIA
Stations that the CIRL is provided only for reference and does not
constitute collection requirements for CIA operations. The Human
.. William Colby testified before the Committee:
"I think it is clear I do not have command authority over the [NSA]. That
is not my authority. On the other hand. the National Security Council Intelligence
Directives do say that I do have the job of telling them what these priorities
are and what the subjects they should be working on are," (William Colby
testimony, 9/29/75, pp. 20-21.)
.. The DCI currently exercises some control over military clandestine operations.
The Chief of Sblti"n in e~ch country is the DCI's "designated representative"
and has responsibility for coordinating- all military clandestine operations.
In the past, the DDO has only objected if the projects were not worth the
risk or duplieated a DDO operation. The Chief of Station rarely undertook to
evaluate whether the military operations could be done openly Or would be
su('('''ssful.
.. While the DCI has final responsibility for the clandestine collection of
intellig-ence, he :"till faces problems in coordinating- the clandestine and technical
collection pro!!rams in his own ageney, Illustrative of this is the reeent establi.~
hment of a National Intellig-ence Officer (NIO) for SneciRI Activities to help
the DCI fo('us DDO operations on three or four central intellig-ence g-.aps, Director
Colby determined that only through a special assistant co'lId he break down
the separate cultures of DDO and teehnical intelligence collection and the barriers
between the intelligence analysts and DDO.
87
Resources Committee has initiated community-wide assessments of
human source reporting in individual countries which emphasize the
ambassador's key role in coordinating human collection activities in
the field. But the Committee has not defined a national systBm for
establishing formal collection requirements for the various human
intelligence agencies.
In summary, the DCI does not have authority to manage any collection
programs outside his own agency. The DCI only issues general
guidance. The departments establish their own intelligence collection
requirements and the collection managers (NSA, DIA, CIA, and
the military services) retain responsibility for determining precisely
which intelligence targets should be covered. President Ford's Executive
Order does not change the DCI role in the management of
intelligence collection activities.
3. Allocating Intelligence Resources
In a 1971 directive, President Nixon asked Director Helms to plan
and review all intelligence activities including tactical intelligence
and the allocation of all resources to rationalize intelligence priorities
within budgetary constraints.46 Since 1971, the DCI has prepared
recommendations to the President for a consolidated national intelligence
program budget. Director Helms, in his first budget recommendations,
proposed a lid on intelligence spending, noting that "we
should rely on cross-program adjustments to assure that national
interests are adequately funded." 47 However, prior to President
Ford's Executive Order, the DCI has had no way to insure authoritatively
that such objectives were realized.
The DCI has. independent budget authority over only his own
agpncy whir>h represents only a smull percentage of the overall
national intelligence budget. As chairman of an Executive Committee
or ExCom for special reconnaissance activities, the DCI has been
involved in the preparation of the program buJget for the development
and management of the major United States technical collection
systems. However, differences of opinion between the DCI and the
other mpmber of the ExCom. thp A"sistant Spcretary of Defense
for Intelligence, were referred to the Secretary of Defense for resolution.
The Secretary of Defense in his budget allocated the remaining
intelligence community resources.
The DCI's role in the Defense inteJJig-ence budget process was in
effect that of an adviser. The DCI's "Perspectives," which analyze
the politicaL economic. and military environment over the next five
years, have had little impact on the formulation of Defense intelligence
resource requirements. According to John Clarke, former Asso-
.. "Announcement Outlining Management Step;; for Improying the Effectiveness
of the Intelligence Community," November 5, 1971, 7 Pres. Docs. p. 1482. Nixon
;;ought to enhance the ro'e of the DCI as community leader and to give the Del
re;;pon;;ihility to coordinate Defen;;e Department technic-al collection operation;;
with other intellie-ence program;;. Nixon';; directive followed a comprehen;;ive
;;tudy of the intelligence community hv the Office of )Ianagement and Budget
(known a;; the Schle;;inger Report) which recommf'nded a fundamental reform
in th" intelligenC'e community's deci;;ionmaking bodies and procedures.
'7 Director of Central Intelligence, National Intelligence Program Memorandum,
FY 1974, p. 44.
88
ciate Deputy to the Director of Central Intelligence for the Intelligence
Community, the "Perspectives" "did not have any great bearing
on the formal guidances that the different departments naving intellIgence
elements used in deciding how much they needed or how many
dollars they required for future years." 48 The military services and
DIA responded to the fiscal gUIdance issued by the 8ecretary of
Defense.
The DCI's small staff of seven professionals in the Resource Review
Office of the Intelligence Community Staff kept a low profile
and spent most of its time gathering information on the various
Defense intelligence activities. They did not provide an independent
assessment of the various programs for the DCI. Consequently, the
DCI rarely had sufficient knowledge or confidence to challenge a
Defense Department recommendation. When the DCI did object, he
generally focused on programs where he thought the Defense Department
was not giving adequate priority to intelligence activities in
which the PreSIdent had a particular mterest.
For example, partly as a result of the intense concern by the NSC
staff, the DCI expenci.ed substantial effort to insure that two Air
Force ships, initially built to operate on the Atlantic missile range
monitoring Cape Canaveral firings, continued to be available to
monitor foreign missile activities. When in 1970-1971, the number of
United States missile tests decreased substantially, the Air Force
proposed that both ships be retired. The DCI, in turn, requested an
intelligence community study which concluded that the ships were
essential for fureign intelligence purposes. Consequently, the DCI
brokered an arrangement for a sharing of the ships' cost within the
Department of Defense. Today, a little under 20 percent of the ship
program is devoted to intelligence needs. The DCI had neither the
authority to direct the retention of these Air Force ships nor sufficient
resources to take over their funding for intelligence purposes to insure
that they were not retired. Nevertheless, the DCI played a definite
role in working out an arrangement whereby at least one ship will
be available until the national intelligence requirement can be met
by another means.49
In practice, the DCI only watched over the shoulder of the Assistant
Secretary of Defense for Intelligence as he reviewed the budget requests
of DIA, NSA, and the military services. If the DCI wished
to raise a particular issue, he had a number of possible forums. He could
set up an ad hoc interagency study group or discuss the question in
the Intelligence Resources Advisory Committee (IRAC).50 He could
highlight resource issues in the annual fall joint OMB-Defense
Department review of the Defense budget or in his December letter to
the President presenting the consolidated national intelligence budget.
However, the groups were only advisory to the DCI and had no
authority over the Secretary of Defense. The joint review and the
.. John Clarke testimony, 2/5/76, pp. 15-16.
.. According to Carl Duckett, the CIA's Deputy Director of Science and
Technology, "frankly we had to fight very hard the last two years to keep the
ships active at all." (Carl Duckett testimony, 11/10/75, pp. 106--107.)
60 IRAC was establi~hed in 1971 to advise the DCI in preparing a consolidated
intelligence program budget for the Prpsident. Members included representatives
from the Departments of State and Defense, OMB, and the CIA. IRAC was
abolished by President Ford's Executive Order of 2/18/76.
89
DCI's letter to the President occurred so late in the Defense Department
bndget cycle that the DCI had little opportunity to effect any
significant changes.
Thus. the DCI's national budget recommendations were for the
most part the aggregate figurf's proposed by the various Defense
agencies. The Del did not provide an independent calculated evaluation
of the entire national intelligence budget. The DCI did not
present the President with broad alternative options for the allocation
8f national intelligence resources. The Del ,ns not able to effect
trade-offs among the different intelligence programs or to reconcile
differences 0\"81' priorities. Finally. the President's decisions on the
intelligence bndget levels Were not based upon the recommendations
of the DCI, but rather upon Defense Department totals. According to
.Tohn Clarke:
I would have to submit that in my judgment I do not think
the Presidents have used the Director's recommendations
,vith respect to the intelligence budgets. There have been few
exceptions where they have solidified behind the Director's
appeal, bnt fundamentally he has looked to the Secretary of
Defense to decide what level of intelligence activities there
should be in the defense budget.st
Because the Secretary of Defense had final authority to allocate
most of the intelligence budget, the DCI either had to "persuade" the
Secretary to allocate Defense intelligence resources according to the
Director's recommendations or take his case directly to the President.
According to .Tames Schlesinger:
... the authority of whoever occupies this post, whatever
it is called comes from the President. ... To the extent that
it is believed that he has the PreRident's ear, he will find
that the agencies or departments will be responsive, and if it
is believed that he does not have the President's ear, they will
he unresponsive.52
But because the DCI must expend substantial political capital in
taking a Defense budget issue to the President, he rarely has sought
Presidential resolution. Over the past five years, the DCI went directly
to the President only twice. Both these issues involved expensive
technical collection systems, and both times the DCI prevailed.
In summary, DCIs have not been able to define priorities for the
allocation of intelligence resources~ither among the different systems
of intelligence collection or among intelligence collection, analysis,
and finished intelligence. Without authority to allocate intelligence
budget resources, DCIs have been unable to insure that unwarranted
duplication and waste are avoided.
4. Key Intelligence Questions
As described above, DCIs have confronted major problems in seeking
to carry out their coordinating responsibilities under the 1947
National Security Act. They have not had authority to establish requirements
for the collection or production of national intelligence.
51 Clarke, 2/5/76, p. 27.
" Schlesinger, 2/2/76, pp. 43, 45.
207-932 0 - 76 - 7
90
They have not been able to institute an effective means to evaluate
how well the community is carrying out their guidance. They have
not had a mechanism to direct the allocation of intelligence resources
to insure that the intelligence needs of national policymakers are met.
To help solve these problems, Director Colby instituted a new intelligence
management system known as the Key Intelligence
Questions (KIQs). Through formation of a limited number of KIQs,
Colby tried to focus collection and production efforts on critical policymaker
needs and to provide a basis for reallocating resources toward
priority issues.53 This section will briefly highlight the resistance which
Colby's new management scheme provoked and the difficulties experienced
in evaluating the overall community efforts.
The KIQ scheme had four stages. First the DCI issued the KIQs.
Then the Xational Intelligence Officers (NIOs) with representatives
from the various collection and production agencies developed a
strategy to answer the individual KIQs. After surveying what information
was currently available to answer the KIQs, the various
agencies made commitments to collect and produce intelligence
reports "against" the various KIQs. At the end of the year, the DCI
evaluated the intelligence community's performance.
The KIQ management process has finished its first full year of
operation and a beginning has been made to provide intelli!!ence consumers
with the opportunity to make known their priorities for intelligence
collection and production. Collection managers have been
brought together in developing a strategy to answer key questions and
analysts have received guidance as to the kinds of reports they should
produce. In addition, the DCI now has before him considerable information
about how the intelligence community is focusing on
intelligence questions which are important to senior national policymakers.
He should be in a better position to show collection and
production managers where they have failed to meet their commitments
to work against individual KIQs or to spend a high percentage
of their resources on KIO-related activities.
However, while the KIQ concept is imaginative. the management
tool has enconntered serious problems. First, the KIQ system does not
solve the DCI's problem of tryinf! to establish priorities in intelligence
collection and produf'tion. Few topics are not included under
one KIQ or another. The KIQs have not yet been meshed with the
existing requirements system. 'While the KiQs are supposed to pstablish
collection and production requirements in lieu of the DCI's Directive
on priorities, both continue to exist today. The Defense Department
has not only continued to issue the D10P but has produced its
own Defense Key Intelligence Questions (DK1Qs) which number oyer
1,000. Instead of providing a means for the DCI to establish priorities
for the intelligence community, the KIQs to date have added another
layer of requirements.
53 In FY 1975, there were 69 KIQs. drafted by the DCl's Nation~l Intelligence
Officers in consultation with the NSC Intelligence Committee working group.
Approximately <me-third of the KIQs dP31t with Sovipt foreilnl policy motivations
and military technology. The other KIQs dealt with such issues as the negoti.
ating position of the Arabs and Israelis, the terrorist threat, etc.
91
8econd, Colby's management scheme has met strong resistance
from the collectIOn and the production agencies. After one year it is
ditticult to identify many intelligence activities that have changed
because of the l\..l~S. '1he l\..ll"l, "trategy .Keports were issued mne
months after tile K1l,,1,s and tended to llst collectlOn and production activities
already under way. The DCI was not in a posItion to direct
the various members of the intelligence community to undertake comnntments
101' dIUel'ent collecCloll eHorts, and tue :::::itrategy 1teports
rarely contained new commitments.
While all agencies participated, DIA and DDO have responded to
the KIQs onlY insofar as they were consistent with their respective
mternal col.ection obJectives. DIA's "Kl<J Collection Performance
Heport" pOlllted out tllat ..the Delense Attache system pnmanly responds
to the DKIQs and J80P IJoint 8trategic Objectives l'lan]
objectives and therefore, responses to KIQs ,,,ill have to maintain consistency
with the two aforementIOned collection guidance vehicles." 54
In fact, DIA writes its "Intelligence Collection J:{equests" and "Continuing
Intelligence Hequirements," and they are then keyed back to
the relevant K1QS, somewhat as an afterthought.55
The Deputy lJIrector of OperatIOns for the CIA issues his "Objectives"
for the collection of clandestine human intelligence. 'While
these are derived from the KIQs, these "Objectives" are in fact the
collection requirements of the Clandestine Service. Since it takes so
long to recruit agents, DDO considers it is not in a position to respond
to specific KIQs dealing with near-term intelligence gaps unless a
source is already in place. Moreover, DDO determined not to deflect
or divert its effort to satisfy KIQs unless the questions happened to
fall within DDO internal objectives.
DIA and DDO invoked the KIQs to justify their operations and
budgets, however they did not appear to be shaping the programs to
meet KIQ objecti'ves. 'Vithout authority to direct resources to answer
the specific Key Intelligence Questions, the DCI had little success in
compelling the major collectors and producers of intelligence to respond
to the KIQs, if they were unwilling. Only NSA has made a
serious effort to insure that their collection requirements are responsive
to the KIQs. In USIB meetings, NSA Director General Allen
argued that the KIQs should be viewed as requirements for the intelligence
community and the KIQ Strategy Reports should provide
more detailed instructions to field elements for collection.56
COlby's new management scheme also failed to establish a workable
e\'aluation process. NIOs provided sllbiective judgments as to how well
the community had answered each KIQ and an assessment of the relatIVe
contribution of each agency. Although NIOs discussed their assessments
with consumers, they had no staff to conduct a systematic and
50 DIA, "KIQ Collection Performance Report," 8/18/75.
5' In FY 1975, only 7 percent of DIA's attache reports responded to
KIQs. Out of 2,111 attache reports against the KIQs only 34 of the 69 were
covered. According- to DIA, military attaches have acce~s to particular types of
information and it would be unfair to assume they had the capability to respond
to all the KIQs.
'" Minutes of USIB meeting, 2/6/75. Approximately 70 percent of NSA's requirements
for FY 1975 were KIQ-related, and about 50 percent of its operations
and maintenance budget could be ascribed to the KIQs.
92
independent review of how well the community had answered the
questions. Furthermore, NIOs did not base their evaluations on any
specific kinds of information, such as all production reports or all raw
intelligence collected on a particular KIQ. They commented on how
well the agencies had carried out their commitments in the Strategy
Reports without asking the collectors for any information about what
activities they undertook or what amount of money had been spent.
They merely took the collector's word that something had or had not
been done. Finally, they did not develop a method to insure that the
judgments of the individual NIOs were consistent with each other.
In addition, the IC Staff aggregated the amount of resources expended
by the various collection and production managers in answering
each KIQ and determined what problems had been encountered.
However, collection and production managers prepared cost estimates
of the activities expended against individual KIQs according to an
imprecisely defined process. And although the IC Staff provided
guidance as to how to do the calculations, the decisions as to how best
to estimate costs were left to the individual agencies. Not surprisingly,
the agencies employed different methods.G7 Consequently, the cost estimates
were not comparable across agencies, and the IC Staff had no
way of making them comparable, since they could not change the different
accounting systems in the various intelligence agencies.38
In summary, the evaluation process did not permit a comparison
of total efforts and results against the KIQs on a community-wide
basis. Colby lacked the necessary tools to use the KIQ management
system to effect resource allocation decisions. The DCI at best was in
a position to shame recalcitrants into action by pointing up stark failures
in a particular agency'R efforts against the KIQs. The KIQ process
was only a surrogate for DCI authority to allocate the intelligence
resources of the community.
Colby's frustrations in trying to direct intelligence community
efforts via the KIQ pro('ess are indicative of the DCI'R limited authority.
Within the present intelligence structure, an effort to get the
DDO and DIA to respond to what the DCI has defined as key policymaker
intelligence questions met considerable resistance. Thus, the
most important issue raised by the KIQ management experience is not
how to refine the process but whether the DCI can really succeed in
directing collection and production activities in the intelligence community
toward critical policymaker needs without greater authority
over the allocation of resources.
Of For example, DIA begins with the assumption that 60 percent of the Defense
attache budget goes for collection. This figure is then multiplied by the
percentage of attache reports which responded to KIOs and the total cost f'Xpended
against the KIQs was calculated to be $1.8 million. In contrast, DDO
calculates cost according to the Ie Staff's recommended formula, which estimates
the number of manhours devoted against the KIQs and multiplies the estimate
by an average production manhour cost.
.. In addition, while the State Department provides cost estimates of INR's
intelligence production costs, it did not submit collection cost statistics, maintaining
that Foreign Service reports were not intelligence collection. So the
evaluation process did not provide a complete picture of intelligence collection on
individual KIQs.
93
5. Pre8ident Ford'8 Executive Order
On February 18, 1976, President Ford announced a reorganization
of the intelligence community to "establish policies to improve the
quality of intelligence needed for national security, to clarify the authority
and responsibilities of the intelligence departments and agencies...."
The major change introduced by the President is the
formation of the Committee on Foreign Intelligence (eFI) chaired
by the DCI and reporting directly to the NSC. The CFI will have
responsibility to: (1) "control budget preparation and resource allocation
for the National Foreign Intelligence Program;" (2) "establish
policy priorities for the collection and production of national
intelligence;" (3) "establish policy for the management of the National
Foreign Intelligence Program;" and (4) "provide guidance
on the relationship between tactical and national intelligence." 59
It is still too soon to pass judgment as to whether the Executive
Order will aid the DCI in his efforts to coordinate the activities of the
intelligence community. By making the DCI chairman of the CFI,
the Executive Order appears to enhance the stature of the DCI by
expanding his role in the allocation of national intelligence resources.
But, as in the case of the Nixon directive in 1971, the DCI appears to
have been given an expanded set of responsibilities without a real
reduction in the authority of other members of the intelligence community
over their own operations. There exist many ambiguities in the
language of the Executive Order, particularly with regard to the role
of the CFT.
The CFI is given responsibility to "control budget preparation and
resource allocation" for national intelligence programs, but the Secretary
of Defense retains responsibility to "direct, fund, and operate
NSA." The CFI is asked to "review and amend" the budget
prior to submission to OMB, as if the CFI will not control the preparation
of the budget but rather would become involved only after the
agencies and departments independently put together their own
budget. Finally, the relationship is not clear between the DCI's responsibility
to "ensure the development and submission of a budget"
and the CFI's responsibility to "control budget preparation."
Moreover, the specific prohibition against DCI and eFI responsibility
for tactical intelligence appears to be a step backward from the
1971 Nixon directive which asked the DCI to plan and review the
allocation of all intelligence resources. While DCls since 1971 have
not become deeply involved in such tactical intelligence questions, they
have reserved the right to become involved; and on several occasions
they have supported efforts to transfer money from the national
Defense Department intelligence budget to the budgets of the military
services, or vice versa. There are, in addition, at least theoretical tradeoffs
to be made between tactical and national intelligence, especially
since the dividing mark between all intelligence operations has become
increasingly blurred with the development of large and expensive
technical collection systems.
•• Executive Order No. 11905. Other members of the OFI will be the Deputy
Secretary of Defense for Intelligence and the Deputy Assistant to the President
for National Security Affairs.
94
C. DIRECTOR OF THE CIA
At the same time the DCI has responsibility for coordinating the
activities of the entire community, he also has direct authority over
the intelligence operations of the CIA. As Director, the DCI runs
covert operations and manages the collection of clandestine human
intelligence (Directorate of Operations); manages the collection of
signals intelligence abroad and allocates resources for the development
and operation of certain technical collection systems (Directorate of
Science and Technology); and produces current intelligence and
finished intelligence memoranda (Directorate of Intelligence).
The fact that the DCls have also directed the operations of the
CIA has 'had 'a vari~y of consequences. First, DCls have tended to
focus most of their attention on CIA operations. The first Directors
were preoccupied with organizing and establishing CIA and with
defining the Agency's role in relation to the other intelligence organizations.
While Allen Dulles and Richard Helms were DCI, each
spent considerable time running covert operations. John McCone
focused on Improving the CIA's intelligence product and developing
new technical collection systems when he was Director. Admiral
Raborn emphasized refining the Agency's budgetary procedures.6o
Second, by having their own capabilities to collect 'and produce
intelligence, DCls have been able to assert their influence over the
intelligence activities of the other members of the intelligence community.
John Clarke, former Associ'ate Deputy to the DCI fur the
Intelligence Community, testified that Helms objected to the suggestion
that CIA get rid of all its SIGINT activities because he needed
"something to keep [his] foot in the door" so he could "look at the
bigger problem." 61 According to Clarke:
... to some degree historically, the Director's involvement
has not only been based upon good, healthy competition
among systems, which I think is good, but the directors have
seen it as an opportunity to give them a voice at the table in
judgments which have importance to their higher role, a
larger role as Director of CI.82
However, this ability to assert influence in turn has had another
consequence: DCls have been accused of not being able to nlay an
objective role as community leader while they have responsibility for
directing one of the community's intelligence agencies. Potential conflict
exists in decisions with respect to ever;y CIA activity. For example,
on each of the two occasions that the DCI went directly to the
President to object to a Defense Department budget recommendation,
the DCI won Presidential support for a CIA-developed technical
collection system. Such DCI advocacy raises the fundamental question
of whether the DCI can indeed be an objective community leader
if he is also Director of the CIA which undertakes research and development
on technical collection systems. According to James
Schlesinger:
There has always been concern and frequently there has
been the reality that the DCI does not overlook all these
eo Colby. 12/11/75, pp. 4-5.
11 Clarke, 2/5/76, p. 59.
a Ibid., pp. 59-60.
95
assets in a balanced way . . . as long as the DCI has special
responsibility for the management of clandestine activities,
that it tends to affect and to some extent contaminate his
ability t() be a spokesman ()f the oommunity as a whole involving
intelligence operations which are regarded as reasonably
innocent from the purview of American life.
Components of the intelligence community other than
the CIA han>, feared that the DCI would be tempted to
expand the authority of the CIA in the collection activities
relative to the other components of the intelligence community.
And there has been some evidence that supp'Jrts such
suspicion....
What I believe is at the present time you have got inconsistent
expectations of the DCI. He's supposed to be the fair
judge amongst the elements of the intelligence community
at the same time that CIA personnel expect him to be a
special advocate for the CIA. You cannot have both roles.63
President Ford's Executive Order seeks in part to reduce the conflict
of interest problem by establishing two Deputies to the DCI, one for
intelligence community affairs and one for CIA operations. The DCI
and his Deputy for community affairs will have offices in downtown
Washington. Nevertheless, t.he DCI will continue to have an office at
CIA headquarters and to have legal responsibility for the operations
of the Agency and at the same time general responsibility for coordinating
the activities of the entire intelligence community.
.. Schlesinger, 2/2/76, pp. 8, 49.

Go to Next Page