THE DCI'S ROLE IN PRODUCING STRATEGIC INTELLIGENCE ESTIMATES

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP93T01132R000100020034-5
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 151
• Document Creation Date: December 27, 2016
• Document Release Date: May 4, 2012
• Sequence Number: 34
• Publication Date: January 1, 1980
• Content Type: MISC

Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
, THE DCI'S ROLE IN PRODUCING
STRATEGIC INTELLIGENCE ESTIMATES
By
VICTORIA S. PRICE
January, 1980
This document is the final report of a study sponsored by
the Center for Advanced Research, Naval War College,
Newport, Rhode Island under Contract Number N00140-78-M-7082
The views contained herein are those of the author and
publication of this study by the Naval War College does
not constitute endorsement thereof by the Naval War -
College, the Department of the Navy, nor the U.S. Govern-
ment.
Further reproduction of this final report by agencies of
the U.S. Government must be approved by the President,
Naval War College. All rights to the reproduction of this
final report outside of the U.S. Government reside with
the author. Reproduction by nongovernmental agencies or
individuals, without the written consent of the author,
is prohibited. However, the content is open to citation
and other reference in accordance with accepted academic
research practice.
DISTRIBUTION STATEMENT
A; UNLIMITED
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
TABLE OF CONTENTS
I. OVERVIEW - Page 1
II. EVOLUTION OF THE STRATEGIC ESTIMATIVE PROCESS: 1947-1957 - Page 10
A. Deficiencies in the Original Concept of Coordinating NIEs - Page 10
B. Definition of the DCI'S and CIA's Role by General W. Bedell
Smith - Page 14
C. Allen Dulles and the Modern Strategic NIEs - Page 19
D. Trends Since 1957 - Page 26
III. THE DCI'S MANAGEMENT OF STRATEGIC NIES AT THE DRAFT STAGE - Page 31
A. Management Direction to CIA Production - Page 33
B. Management Attention to Department of Defense Coordination - Page 50
IV. SUBSTANTIVE INPUT OF DCIs TO THE DRAFT STRATEGIC NIEs - Page 60
V. THE DCI'S LEADERSHIP ROLES AT USIB OR NFIB - Page 73
A. Support of the CIA's Position - Page 74
B. Encouragement of Competitive Debate - Page 78
C. Final Adjudication of Views - Page 86
VI. THE DCI'S ROLE IN PROMOTING INDEPENDENT REVIEW - Page 96
A. Allen Dulles, the Princeton Consultants and Strategic
Advisory Panel - Page 96
B. The 1960s - Page 98
C. James Schlesinger and the Intelligence Community Staff Page 100
D. The President's Foreign Intelligence Advisory Board and the
1976 A-B Team Experiment - Page 101
VII. ASSESSMENT OF THE PERFORMANCE OF DCIs - Page 106
A. Summary of Roles Played by DCIs - Page 106
B. Mutually Supporting and Conflicting Roles: Manager versus
Prophet - Page 115
C. Steps Not Taken by DCIs - Page 119
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 1 -
I. OVERVIEW
Since the position of the Director of Central Intelligence (DCI)
was created in 1946 and the CIA in 1947, a major responsibility
has been to produce coordinated estimates of the likely future
capabilities and intentions of foreign countries. The most polished
and prestigious of the CIA's finished intelligence evaluations are
the National Intelligence Estimates, called NIEs. The NIEs have
existed in their present form since 1950 when the CIA's Office of
National Estimates and Board of Estimates were created. Drafted
mainly by CIA staff, the NIEs include the input, judgements and
dissents of intelligence officials from other agencies, primarily
the Defense and State Departments.
After coordinative meetings at the CIA with representatives
from these agencies to discuss drafts, the NIEs are reviewed by
the DCI and presented to the United States Intelligence Board (USIB),
renamed the National Foreign Intelligence Board (NFIB) in 1977.
Chaired by the DCI, NFIB consists of the heads of other intelligence
agencies, and has broad responsibilities for coordinating the
activities of the intelligence community.* After joint discussion,
the NIEs are approved by the DCI and issued, in his name, to the
President.
*NFIB was called USIB from 1958 to 1977, and before that the
Intelligence Advisory Committee (IAC). Chaired by the DCI,
NFIB consists of the Deputy Director of CIA (representing the
CIA), directors of the Defense Intelligence Agency (DIA), the
National Security Agency (NSA), the State Department's Bureau
of Intelligence Research (INR), and the heads of the intelligence
sections of the FBI, Treasury Department, and the former AEC.
Since 1961, the heads of Army, Air Force and Naval Intelligence
have not been official members, but attend meetings, debate
and may dissent from the NIEs.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
2
The NIEs are intended to inform the President, National
Security Council and other top-level officials about upcoming for-
eign events, both short and ions' term, of national importance.
For example, some NIEs predict political-economic trends in various
trouble spots around the world. These are meant to lead to timely
estimates, which may be termed "crisis prediction" NIEs, predicting
major crises affecting US foreign policy, such as the outbreak of
war, political coups or sudden oil shortages in foreign countries.
Then, NIEs termed "operation evaluation" NIEs keep abreast of
military and political events during conflict situations, such an
during the Korean and Vietnam wars.'
Special National Intelligence
Estimates, called SNIEs, are produced to respond to specific,
short-term questions posed by the White House, such as during the
Cuban missile crisis in 1962.
Another group of NIEs'is what this author terms the "strategic
NIEs." These are annually produced evaluations of Soviet, and now
Chinese, military strength.* Strategic NIEs primarily address
questions of military force structure, technology and new weapons
for an upcoming five year period, of which analysis of strategic
policy and intentions is an integral part. From a single, broad
estimate on Soviet military capabilities and intentions in 1950,
a series of separate estimates on Soviet military strength grew
up in the late 1950s. Designated by the number "11", the Soviet
*The author uses the term "strategic" for both intercontinental
and intracontinental arms.
narlaccifian in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 3 -
NIE series has consisted of individual estimates on conventional
arms, offensive nuclear weapons, defensive weapons, nuclear tech-
nology, strategic doctrine and other subjects. Presently, the
most important Soviet strategic NIE is numbered 11-3-8, combining
considerations of intercontinental offensive and defensive weaponry.
The Soviet strategic NIEs are the focus of this study.
Controversy About the Soviet Strategic NIEs.
From the beginning, the Soviet strategic NIEs have been the
object of considerable controversy within and outside the intelli-
gence community. The debate has been heated, protracted and inter-
departmental. In producing these NIEz, the CIA often has been pitted
against the Defense Department's intelligence agencies, such as the
Defense Intelligence Agency (DIA) created in 1961 and Army, Air
Force and Naval Intelligence. The CIA is supposed to include these
agencies in all phases of drafting and, compared to other NIEs ex-
cept during wartime, the Pentagon's involvement has been heavy.
Ultimately, the DCI is responsible for ensuring that the strategic
NIEs represent a coordinated view, i.e., that all intelligence is
considered and all opinions voiced, before formal approval at NFIE.
More than other NIEs, the debate about the strategic NIEs re-
volves around the process of coordinated production, aspects of
which various participants claim have caused major, predictive
inaccuracies. For example, controversy has existed over the large
role of the CIA in collecting and analyzing purely military
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
4
intelligence, and its domination over the drafting process. Ques-
tions are raised about how to involve the Pentagon agencies more
in the early stages of drafting, and whether their dissents have
been adequately heard and reflected in the final NIEs. Accusations
abound as to the biases involved in institutional judgements. For
example, the three military services often have been charged with
basing excessive estimates of Soviet weaponry on their own bud-
getary priorities, a concern expressed by both Presidents Truman and
Eisenhower about the NIEs.2 Beginning with the Nixon Administration,
CIA analysts have been criticized for a "liberal bias" which caused
them to systematically underestimate Soviet strategic goals.3 As
one participant observed:
The truth is that the DCI, since his author-
ity over the intelligence process is at least
ambiguous, has an uphill struggle to make
a sophisticated appreciation of a certain
range of issue prevail in the national in-
telligence product over against the parochial
views and interests of departments....4
In the early 1970s, such criticism led to a demand for review of
the strategic NIEs by outEide experts, culminating in the A-B Team
experiment in competitive analysis in 1976, described later.
The central purpose of this study is to provide better under-
standing of one element in the production of coordinated strategic
intelligence estimates, namely, the role of the Director of Central
Intelligence. How has the role of the DCI in producing the strate-
gic NIEs historically evolved? How have different DCIs observed
their functions? What involvement have DCIs had in drafting the
nni-laccifiinri in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
5
strategic estimates, encouraging interdepartmental competition, and
resolving debate at USIB or NFIB? Can a feasible and appropriate
role for the DCI be prescribed for the future?
Complexity in Defining the DCI's Role
The difficulty in defining the role of the DCI in producins
the strategic NIEs is inherent to the nature of this country's in-
telligence apparatus. As the Murphy Commission concluded in 1975:
The United States set up what might be termed a
'mixed' system in an attempt to combine the best
features of decentralization and centralization
while avoiding the obvious weaknesses of each.
The compromises involved in the establishment
of a 'mixed' system have given rise to difficulties
over the years.5
Our intelligence system is decentralized in the sense that the
CIA was never intended to have (and never achieved) amonopoly
over intelligence collection and analysis. Other agencies, pri-
marily those of the Defense Department for strategic estimating,
are supposed to make intelligence and evaluative inputs tc the
NIEs as well. The advantages of decentralization are considered
to be efficiency and improved analysis, if each agency specializes
in collecting and analyzing intelligence within its own area
of expertise, and multiple advocacy and competition in estimating.6
The system is centralized in the position of the DCI, who
has three broad roles to play. He manages the CIA, including
its covert activities; he acts as chief intelligence advisor
to the President; and he coordinates the activities of the
intelligence community. Over the years, the DCI's authorization
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 ?,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
6
for coordinative tasks has expanded steadily. In addition tc
approving the NIEs, he is responsible for focusing the entire
community on issues critical to the White House, efficiently
dividing and overseeing collection tasks, and since 1971, re-
viewing and presenting a consolidated intelligence program
budget to Congress. To help him, the DCI may devise whatever
coordinative bodies he wishes within the CIA. However, the
central coordinating institution of the intelligence community
has been USIB and its subcommittees, and now NFIB.
If intelligence coordination is represented in the person
of the DCI and institutionally by USIB,* the NIEs are its substan-
tive embodiment. The NIEs are meant to be the DCI's document.
However, no other intelligence product so readily exemplifies the
potential conflicts between the DCI's three jobs. For example,
as manager of the CIA, is it not natural for the DCI to expand
its intelligence collection and analytical functions, even though
such steps might duplicate or weaken the programs of other agencies?
As the man responsible for CIA morale and most familiar with its
work, should the DCI let the intelligence professionals work out
their differences or mainly promote the CIA's position within the
final NIEs? As chief intelligence advisor to the President,
should the DCI represent his own opinions in the draft or final
NIEs, identify himself with a minority or, after encouraging dis-
sent, resolve the debate by making intelligence judgements. What
happens if the DCI does not have the technical background to
*Since NFIB was called USIB for almost 20 years, the author uses
the acronym USIB throughout most of this study, unless under
a specific DCI it was called something else.
im,,ninecifiori in Part - Sanitized CODV Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
7
understand the strategic issues at hand? As coordinator of the
intelligence community, how should the DCI resolve what the
Senate Select Committee on Intelligence has identified as "the per-
sistant conceptual dilemma...of consensus versus competition."7
In the interest of presenting a timely, concise document to the
President, how far should the DCI promote competition within the
NIEs?
Research Design and Methods
This author contends that such questions historically have not
been subject tc Presidential direction. Few Presidents have ad-
dressed the judgemental issues facing the DCI in producing the
NIEs. In contrast, more Presidential attention has focused on
the DCI's role in coordinating intelligence collection and, now,
budgets. In an attempt to explore intelligence production issues,
in the following pages the actual roles assumed by successive
DCIs for the strategic NIEs are described in depth.
In Chapter II the evolution of the strategic NIE process
under DCIs Admiral Roscoe Hillenkoetter (1947-1950), General Walter
Bedell Smith (1950-1952) and Allen W. Dulles (1953-1961) is traced.
The author concludes that the role of the DCI and CIA in the
strategic estimative process was originally ill-defined and un-
realistic, and that DCI's Smith and Dulles took the initiative in
strengthening the DCI's personal role and the CIA's internal
stratesic work. Then, trends since 1957 increasing the decision-
making burden on the DCI are described.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 8 -
In Chapters III, IV, V and VI, the attitudes and behavior of
six DCIs in producing the strategic NIEs are compared systematically.
The DCIs selected by the author for study are: Allen W. Dulles
(1953-1961); John A. McCone (1961-1965); Richard Helms (1966-1972);
James R. Schlesinger (1973); William E. Colby (1973-1975); and
Admiral Stansfield Turner (1977-present). Two DCIs not examined
here are Admiral William Raborn (mid-1965 to mid-1966) and George
Bush (1976), primarily because of their short tenures.* Although
James Schlesinger remained in office barely five months, he is
included because most interviewees reported that his influence on
the strategic estimative process was large.
In these chapters, seven possible roles of the DCI are ex-
amined. Two pertain to the mechanisms of the NIE drafting process:
management direction to CIA production, and management attention
to coordination with the Department of Defense intelligence agencies.
Another involves the DCI's personal, substantive input to the draft
NIEs. Three roles focus on a DCI's leadership posture during USIB
deliberations: support of the CIA's position; encouragement of
competitive debate; and final adjudication of views. The last is
the role DCIs played in promoting independent, outside review of
the strategic NIEs. The reader should bear in mind that the purpose
of these chapters is descriptive not evaluative.
The author concludes that the approach of DCIs to the strategic
NIEs has been inconsistent. Some DCIs have emphasized some roles,
while others neglected them. In Chapter VII, these differences
*DCI George Bush's Important role in the 1976 A-B Team experiment
in competitive analysis is raised briefly in Chapter VI.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
9
are summarized according to a DCI's individual background and
concepts of his job, and in light of historical events. The
author also examines potential conflicts between roles exercised
by DCIs, addressing the question of whether a DCI can be effectively
both "manager and prophet" for the strategic intelligence estimates.8
This study is based on four principal sources. The first is
governmental reports, primarily unclassified, examining aspects
of the strategic estimative process. These include reorganization
commission reports, recent work of the Senate Select Committee on
Intelligence, and internal CIA studies. Second axe scholarly
studies and journalistic accounts of the strategic NIEs themselves.
These are used as a means by which to document historically the
competing positions of different agencies in critical intelligence
debates, such as the famous "missile gap" debate from 1957-1961.
The third source is unclassified NiEs, mainly non-strategic.
Since these sources rarely focused on the activities of DCIs,
the major portion of this study relies on over 50 interviews con-
ducted by this author in 1978 and 1979. Interviewees include past
and present CIA officials, two former DCIs, Pentagon intelligence
officials, and NSC and Congressional staff. Interviewees are
listed at the close of the study. Current CIA officials are not cited
by name within the footnotes.
neclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 10 -
II. EVOLUTION OF THE STRATEGIC ESTIMATIVE
PROCESS: 1947 - 1957
The National Security Act of 1947 creating the CIA directed
it "to correlate and evaluate intelligence relating to the national
security and to provide for its appropriate dissemination."1
In his memoirs, President Truman wrote that he desired a national
intelligence organization which had access to all information,
headed by a Director who could speak authoritatively for the
whole community.2 He wanted to reduce the separate, sometimes
conflicting and seemingly biased intelligence reports reaching
his desk, and asked the DCI to produce one analytical, predictive
document combining these.3 Also created in 1947 was the Intelli-
gence Advisory Committee (IAC), the predecessor of USIB and NFIB.
Chaired by the DCI and consisting of all intelligence agency
heads, IAC was meant to serve as a coordinating body for the
entire intelligence community.
A. Deficiencies in the Original Concept of Coordinating NIEs
The task of producing coordinated estimates fell to Rear
Admiral Roscoe Hillenkoetter, a former head of Naval Intelligence
who served as DCI from 1947 to mid-1950, and the CIA's new Office
of Research and Evaluation (ORE). In retrospect, the original
concept of how they were to do this was highly unrealistic, in
assigning a subordinate role to the Pentagon and prescribing a
dominant role to the fledgling and ill-equipped CIA.
In theory, the 1947 Act said that the "departments and other
agencies of Government shall continue to collect, evaluate,
correlate and disseminate departmental intelligence," to which
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 "
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
the CIA was intended to have complete access.4 Then, it became
the task of a small team within ORE to examine this work, both
raw intelligence and finished reports, and generate new predictions
if necessary. Its analysis was summarized in special estimates
called ORES, to which the concurrence or dissent of Defense Depart-
ment agencies was sought. The CIA was authorized to carry out re-
search and analysis "not presently performed" by other departments,
but this was not considered one of its primary estimative functions.5
In short, the CIA was not designed to duplicate or supercede
the Pentagon's military intelligence collection, research and
analytical functions. Rather, as Sherman Kent then wrote: "Its
job is what might be called policing the professional competence
of the departmental outfits and continually pushing departmental
frailties back into departmental laps."6 In 1949, Kent did not
believe that the CIA even should do its own descriptive and evalua-
tive studies of purely military subjects, since he argued that it
could not compete effectively with the Pentagon. In fact, he worried
that if attempted, the Pentagon soon would dominate the CIA.7
Immediately, the CIA's effort to produce coordinated judge-
ments was undercut by several factcrs. First, as the Senate
Select Committee on Intelligence later noted, the Army, Navy, Air
Force and Joint Chiefs of Staff (JCS)* refused to provide the
new agency with much raw intelligence since they feared encroach-
ment on their traditional estimative rcle.8 Second, the CIA had
*The Joint Chiefs of Staff's intelligence arm was called the Joint
Intelligence Committee (JIC) or sometimes J-2. Its head was a mem-
ber of IAC, and then USIB, until DIA was created in 1961.
rlarlaccifiPri in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 12 -
little independent capability to analyze military data even if
forthcoming. For example, the small estimative team within ORE
were generalists and the CIA had only one scientist to examine
Soviet technical developments.9 Thus, it often analyzed only
finished reports from the Pentagon. For example, in 1948 the CIA
evaluated an Air Force estimate discussing Soviet troop move-
ments in Germany, which suggested that the Soviets were prepared
to go to war at any time. The CIA went over the document line
by line, and reportedly labeled the estimate "mistaken."10 The
CIA said that it had evaluated the information differently.
Third, the Pentagon intelligence agencies often refused to
cooperate in drafting and reviewing the OREs, and sometimes with-
held their final concurrence. The first estimate of Soviet military
objectives and capabilities, intended to be an annual affair,
took two years to complete in March, 1948, and ORE considered it
mainly a CIA document.11
A sampling of the estimates prepared in
the late 1940s indicates that some never were issued formally, since
the Pentagon refused to concur.12
Further, IAC reportedly spent
most time reviewing collection programs and intelligence requirements,
rather than acting as a high-level deliberative and approval
body for the estimates.13 In 1949, one study group concluded:
"The principle of the authoritative NIE does not yet have established
acceptance in the government. Each department still depends more
or less on its intelligence estimates and establishes its plans
and policies accordingly."14
By 1950, it was asserted that "ORE
did little more than produce its own analyses"15 and "90% of its
work" was current, non-predictive intelligence.16
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 13 -
Finally, the CIA's intended authority over the strategic
estimative process was not advanced, or aggressively defined, by
DCT HillenkoettOr. In theory, the DCI's only formal responsibility
for the coordinated estimates was to approve and issue them. Thus,
he had considerable latitude to establish an active, personal role.
However, although Admiral Hillenkoetter daily briefed President
Truman, he did not have sufficient rank or stature wjthin the
military intelligence community to elicit their full cooperation.17
For example, a CTA memorandum in 1950 stated that "the Intelligence
Advisory Committee had assumed an advisory role to the NSC and
functioned as a supervisory body for the DCI--contrary to the initial
intention. ,,18
DCI Hillenkoetter also seams to have been uncertain about
how far the CIA should go in making firm predictions of foreign
military intentions. For example, in the aftermath of the North
Korean attack on South Korea of June 10, 1950, Hillenkoetter
staunchly defended the CIA by asserting that top-level officials
were in possession of the pertinent data before the invasion, and
the CIA subsequently was exonerated in this respect.19
However,
he reportedly argued that the CIA could not be responsible for
the "evaluation" of enemy intent.20
A year later, he told Congress
that the CIA similarly predicted the location and capabilities
of Chinese troops in Manchuria before their December 1950 invasion,
but that "to predict the intentions of the enemy, real or potential,
you would need a crystal ball."'" In contrast, Hillenkoetter
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 14 -
was accused in the press fcr "empire-building" in current intelli-
gence matters,22 and he frequently sent President Truman personal
memoi summarizinc "raw" CIA agents reperts.23
B. Definition of the DCI's and CIA's Role by General W. Bedell Smith
By 1950, five separate studies called for a greater "centraliza-
tion of authority" in the DCI's hands for long-term, predictive
estimates.24
General Bedell Smith arrived as DCI in the late
1950s determined to assert his prerogative for this.25 As a fcur-
star general, formerly General Eisenhower's Chief of Staff and
Amhassador tc Russia, he was well-equipped to do so. DCI Smith
took four steps. First, he created the Office of National Estimates
(ONE) and Board of Estimates in CIA in 1950, whose sole purpose
was to produce speculative-evaluative NIEs. Second, he launched
the CIA on the path of acquiring an independent capability for
analysis of military data. Third, he forcefully elicited the
cooperation cf the Pentagon in preparing NIEs on military subjects.
Fourth, he considered the NIEe his estimates, which should reflect
a clear-cut, hopefully unified opinion for the President, and his
own views when they existed. For this purpose, he turned IAC intc
a body in which he supervised formal deliberations on the NIEs.
The ONE was designed as an analytical unit tc draft NIEs.
To do this, General Smith asked historian William Langer, ONE's
first director, to recruit the best talent available. The ONE
deliberately was kept small, ranging over the years from 30-75
professionals. Its role was to write the NIEs, relying entirely
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
4.44
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 15 -
on the work generated by other CIA components and other agencies
to do so. The Board of Estimates was created to review the draft
????
NIEs, and consider the comments and criticisms of other agencies
before submitting them to the DCI and IAC for final approval.
The Board always consisted of about a dozen men, including retired
military officers. The first nine included five historians, one
lawyer, and three former military officers, of whom one formerly
was Eisenhower's G-3. An outside panel of experts, known as
the Princeton Consultants, was established to assist the Board and
DCI in reviewing the NIEs.
General Smith initially believed that the three military ser-
vices should collect and review all military and strategic data,
which ONE then should coordinate and evaluate as a basis for
drafting the NIEs. If inadequate, "his reaction was not to do
it yourself, but to get others to do it."26 In practice, a short-
age of hard intelligence on Soviet troops, bombers, airfields and
military plans resulted in the Pentagon making large demands on
the CIA's clandestine division, the Directorate for Plans (DDP).
As a former DDP Soviet specialist writes, from 1949-1954, the DDP
"operated almost solely as an instrument of the Department of
Defense and its theater commands. Their need was great."27
Further, it soon became evident that for ONE to draft NIEs pre-
dicting the capabilities of Russian heavy bombers, nuclear pro-
duction, and eventually missiles and warheads, the CIA had to have
its own capability for scientific, technical and economic analysis.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 16 -
Prirarily, this was because General Smith became convinced that
the CIA could not rely on the Pentagon to generate and submit
such-intelligence analyses to it.26
Smith thus waged a bureaucratic battle to establish the CIA's
independence in military and eventually strategic research.
For example, in an agreement with the State Department in 1951, the
CIA's newly formed Office of Research and Reports (ORE) was granted
exclusive responsibility for economic research on the Soviet bloc.29
The Office of Scientific Intelligence (OST) established in 1.949
moved slowly into weapons technology research, although an August
1952 agreement with the Defense Department restricted it to basic
scientific research and pilot projectE. The Office of Current
Intelligence (CCI) created in 1951 first specialized in political
intelligence, but soon had a military division. In early 1952,
these three units (ORR, OSI and OCI) were grouped together along
with ONE in the CIA's Intelligence Directorate (DDI).
Through the strength of his professional rank and personality,
General Smith induced the military services and JCS to contribute
written, draft material to ONE for consideration. After the
role of the Pentagon was established as one of regular, formal in-
put to the draft NIEs, a second problem was to encourage officers
of sufficient rank and knowledgeability to participate in
the deliberations at CIA and, third, to encourage them to elaborate
their positions.30
General Smith believed that considerations of
Soviet military hardware should be accompanied by predictions of
Soviet military strategy. "Almost immediately, the military
1
challenged ONE on the nature of the estimates, demandinc, that
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 17 -
they be factual and descriptive."31
General Smith prevailed,
and in so doing hoped to reduce the Pentagon's tendency to over-
estimate Soviet weapons based on individual service budgetary
priorities.32
General Smith played a strong role in deciding what military
issues should be the subjects of NIEs. In theory, NIEs were
produced following a specific request from the NSC, and the NSC
was also supposed to have a hand in deciding the "terms of reference"
of an NIE (i.e., the specific questions posed by an NIE). However,
from the beginning, General Smith and the CIA often took the
initative. During Smith's tenure, the military-oriented NIEs on
ongoing operations in the Korean war and Soviet military actions
in Eastern Europe took precedence over what this author has termed
the Soviet strategic NIEs. Smith was keenly interested in the
former, for which he posed specific questions and wanted "quick
and authoriative" answers.33
It was typical for Smith to phone
his colleagues at the Pentagon and say something like: "'I want
you to get over here and decide whether Russia is going to invade
Poland.'"34
Prior to submitting the NIEs to IAC, he read them
thoroughly and voiced any personal disagreements privately to
ONE or Pentagon officials.35
Then, the NIEs might be edited by ONE
to more clearly reflect these.
General Smith's "extensive contacts at the senior military
level and his pervasive prestige" enabled him to turn IAC into a
deliberative body over which he firmly presided.36 In 1950, the
CIA's General Counsel had told Smith that:
npclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 18 -
IAC's inflated role had diminished the DCI's
ability to demand departmental cooperation for
the CIA's national estimates responsibility.
Houston advised that the DCI would have to
exert more specific direction over the depart-
mental agencies, if coordinated national
intelligence production was to be achieved.
Smith acted on Houston's advice and informed
the members of IAC that he would not submit
to their direction.37
Smith believed that by encouraging competitive discussion of the
NIEs at IAC, the Pentagon's intelligence heads would take a more
active interest in them. He directed the debate along lines
of thinking he saw as most productive. No formal vote was taken
on the NIEs, and it was up to him to open or close discussion
as he so chose. His approval of the NIEs was meant to be final.
After IAC, "he insured that they were presented to Truman without
contrary briefings being given him by other agencies as so often
had happened before."38
Almost immediately, General Smith addressed the issue of
how much and what kind of military dissent to permit, for example,
in the NIEs on the Korean war. In theory the right to dissent
from the main text of an NIE in a brief footnote existed from
the beginning. The NIE was supposed to provide an "exposure of
information and judgements, reflecting both the consensus and
the differences within the intelligence community."39 In practice,
this did not pertain to the CIA, since the Board of Estimates
worked to reach a unified position and a dissenting faction within
the CIA was never permitted a footnote. Further, Smith was intol-
erant of military dissent which appeared to him poorly reasoned
or biased. Thus, one of his major aims before and during IAC
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 19 -
meetings was tc encourage the Pentagon to state their dissents
clearly and precisely and, failing that, to overrule them."
In this sense, Smith played a role which this author later terms
as "adjudicatory." He decided whether a dissent reflected a
real predictive uncertainty, or was discrete and well-reasoned
enough, to be represented in a final NIE.
C. Allen Dulles and the Mcdern Strategic NIEs
When Allen Dulles became the first civilian DCI in early
1953 under President Eisenhcwer, the DCI had the license to shape
the NIEs as he saw fit. Regarded as his document, he might in-
fluence what the NIEs examined, represent his personal opinion
in them, or seek a consensus-oriented NIE or one reflecting dis-
senting views. As chairman of IAC, he could act as leader of
debate on N:ES, expanding or limiting discussion as he inclined.
Final decisions at IAC were meant to be authoritative and binding.
However, compared to other NIEs, the Soviet strategic NIEs
were primitive and the DCI's role in this process had not taken
shape. Primarily, this was because the DCI as yet did not have
to preside over intense, intricate conflicts between the CIA ane
Pentagon reflected in the strategic estimates. Further, the drafting
process was still evolving. In the first half of the 1950s, the
Pentagon dominated estimating on Soviet strategic weapons, since
the CIA's capability for independent work was fledgling.
npriaccifipri in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 20 -
By 1953, the CIA issued annually two major NIEs related to
Soviet strategic strensth. One NIE, numbered 11-2,* assesssed
the Soviet's "gross capability" to attack the US and focused on
Soviet nuclear development. What was to become the more contro-
versial estimate was NIE 11-4, a comprehensive "soup to nuts" docu-
ment addressing Soviet political, economic and military capabilities
to pursue its goals throughout the world. NIE 11-4 included a
military chapter, sometimes chapters, on Soviet conventional arms
and manpower, and medium and long-range delivery vehicles. In
the late 1950s, separate NIEs were initiated to cover these and
other topics.
In drafting the military portions of NIE 11-4, ONE is said
to have relied chiefly on the Pentagon's written input and judge-
ments on Soviet hardware. That is, until late 1956, the Army, Navy
and Air Force were asked tc draft individual, uncoordinated es-
timates pertaining to their special expertise, e.g., the Air Force
for Soviet bombers and missiles. Then, ONE reviewed, edited and
polished these into a standard NIE form. Because the CIA's in-
dependent research and analytic work was limited, ONE did not
draft the strategic estimates from scratch, as it did for other
NIEs. In preparing these, one CIA analyst asserts: "The ONE
*Henceforth in this study, specific NIEs for specific years will
be referred to by number, with the last two digits referring to
the year of production, for e:tample, NIE 11-4-58.
**For example, arcund 1955 a separate NYE, numbered 11-3, was in-
itiated to address Soviet air defense systems. In 1958, NIE 11-8
was created to cover Soviet strategic attack forces, and it is this
NIE which became one of the most significant and controversial.
NIE 11-4 evolved into an estimate discussing Soviet strategic
doctrine, NIE 11-5 focused on Soviet missile capacities and,
NIE 11-15 addressed Soviet and Warsaw Pact conventional forces.
In 1974, NIEs 11-3 and 11-8 were combined. There never has been
a complete NIE 11-1 to 11-15 series produced for any given year.
Rather, the number assigned tc each tcpic persists, and some
NIEs in the series are produced only occasionally.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 21 -
accepted fairly uncritically the inputs made by the Pentagon."41
An Army intelligence official repozts: "The CIA generally took the
Pentagon's viet,;13 in NIE 11-4.?42
The ONE analysts did not favor this procedure. In the early
1950s, ONE had experimented in writing its own first draft state-
.
ment of Soviet strategic aims.43 In 1955, ONE's chief urged the
staff to write an entire first draft of the relevant military
sections.44
Further, not all ONE staff and Board members agreed
with the Pentagon's figures. Differences existed, but have
been described as "similar cleavages...to those existing through-
out the defense establishment? as opposed to a unified, independent
CIA position.45
Nc2netheless, the first half of the 1950s was
characterized by a slowly evolving CIA role in drafting, in which
it could not match the resources and skill of the three military
services.
This story is best illustrated by the now famous "bomber
gap" overestimates of Soviet long-range manned bombers, presented
in NIE 11-4 from the spring of 1955 to the fall of 1956. Basically,
these predictions represented an Air Force judgement, unchallenged
and thus supported by the CIA, and sharply contested by the
Army and Navy. The "bomber gap" predictions were contained in
the main text of the NIEs, not in footnotes. As one author writes,
the NIES of 1955 and 1956 "provided a range of possible production
curves, the choice among them being largely a matter of in-
stitutional commitment and interest."46
For example, in NIE 11-4-55
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 22 -
the CIA and Air Force predicted a range of 30-80 Bison bombers in
place, and 500-800 Bisons by mid-1960.47 The Army and Navy ob-
jected, and their estimate later proved the most accurate. By
1961, the Russians had built only 180 long-range bombers, the
majority of which were Bears, not Bisons.48
Allen Dulles upheld the "bomber gap" estimates in 1955 and
1956, even though "the input of strategic intelligence into the
defense debate in 1955-1957 was extremely damaging to the Adminis-
tration" whose stated goal was to hold the defense budget constant.49
As Dulles later wrote:
In 1954...there was evidence that the Soviet Union
was producing long-range intercontinental heavy
bombers comparable to our B-52s. At first, every
indication...pointed to the conclusion that
the Russians were adopting this weapon as a major
element of their offense and planned to produce
heavy bombers as fast as their economy and tech-
nology permitted. 50
In comparison to Soviet bomber estimates, in January 1956 the
USAF Strategic Air Command had only 41 B-52s in place, although
57 had been completed.51 In response to the "gap" predictions,
President Eisenhower authorized acceleration of the B-52 pro-
duction rate to 12 a month in the spring of 1955, and in May 1956
the rates were accelerated twice again, first to 17 a month and
then 20 a month.52
In predicting Soviet bombers and subsequently ICBMs, the
Air Force did not dominate the NIEs for long. By the end of
1956, the CIA emerged as a rival to Pentagon intelligence and
engaged in its first, heated debate with the Air Force over
the bomber gap. When Dulles presented his annual testimony
Declassified in Part - Sanitized Copy Approved for Release 2012/05104: CIA-RDP93T01132R000100020034-5
?
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 23 -
to Congress on the Soviet strategic threat, the comments offered
by the Air Force officer, who traditionally accompanied the DCI
to answer technical questions, were challenged by the CIA.
Changes that CIA analysts wished to make in editing the officer's
testimony were presented to Dulles, who was receptive.53
The
CIA's position was reflected in its first independent draft of
NIE 11-4 in the spring of 1956, in which its estimates of Soviet
bombers began to slide. In response, President Eisenhower reduced
the B-52 production rate to 15 a month.54
Taking the August
1956 NIE as the baseline for estimates of Soviet heavy bomber
development, the CIA reduced the estimates to 62.5% in the end of
1957, 25% by the end of 1958, and 19% by February 1960.55 The
Air Force dissented from the CIA's view in brief footnotes.
Thus, by 1957 the modern strategic NIE had evolved. The
CIA independently drafted its institutional position as the
main text, and the Pentagon participants were relegated to
footnoted dissents. This form was to prevail for the next two
decades. The CIA's claim to a dominant position within strategic
NIEs resulted from three factors: the U-2 photo reconnaissance
flights; the CIA's independent strategic research and analytic
capability; and the personal decision of Allen Dulles.
First, through the DDP side, the CIA developed, funded,
managed and flew the U-2 plane which took its initial photographs
over Soviet territory in June 1956.56
One of Dulles' priority
projects, as DCI he controlled the complex interagency decision-
making mechanisms for U-2 flights subject to final Presidential
neclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 24 -
approval.57
The CIA developed its own photo interpretation
capacity. From one specialist in 1953, a special office was
created under the DDI which promptly grew to 20 persons and
eventually 1,200.58 By August 1956, Soviet factories building
bombers had been located for the first time and bombers at the
factories and on the airfields were counted. According to CIA
analysts, these numbers did not support the "bomber gap" theory.59
Second, on the DDI side, the CIA's economic, scientific and
current military intelligence work provided ONE with independent
analyses to use in drafting the strategic NIEs. The most sophis-
ticated, novel and prestigious reportedly was the Soviet economic
group called the Economic Research Area (ERA). Based on its
analysis of Soviet materials and industrial capability, ERA
developed new methodologies for estimating production rates,
a speciality of the CIA's ever since . As the Senate Select Committee
on Intelligence describes:
By introducing economic production capacities into
assessments of Soviet strategic capabilities, the
Agency challenged the basic premises of the
military's judgements...Air Force intelligence
based its estimates on knowledge of Soviet tech-
nology and laboratory research, which by 1953 were
well advanced. ORR (of which ERA was a part) based
its estimates of Soviet deployments on Soviet eco-
nomic production capabilities, which were severely
limited as a result of the war. 'Consequently, ORR's
methodology attributed lower strategic deployments,
i.e., long-range bombers and missiles, to the
Russians. 100
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 25 -
Third, by 1957 Allen Dulles ,favored a document in which
the CIA presented its independent view as the main text and
conclusion. This decision he is said to have "agonized over" for
some time after.61
An OSS hero, Dulles was eager to establish the
CIA as the chief intelligence arm of government. Although Dulles
had an independent base in the Administration, he was cautious
about advertising the institutional growth of the CIA, lest the
Pentagon become alarmed unduly. Personally, he did not like to
preside over awkward jurisdictional disputes between the CIA
and military. By the mid-1950s, the CIA already was engaged in a
bitter battle with the Army concerning the latter's clandestine
operations and counterintelligence roles, at the direction
of the NSC.62
Dulles' preferences led to outside criticism.
For example, he was accused of failing to promptly consolidate
CIA's military research components "on the grounds that the
services might interpret such a move as a unilateral attempt by
the Agency to assume large responsibilities in their fields of
primary concern..63 Three times during his tenure the President's
Foreign Intelligence Advisory Board (PFIAB) and its predecessor
called for the DCI to assume a stronger coordinative role for
the entire intelligence community.64
Interestingly, Dulles was not firmly convinced that "it was
in the best national interest" to present competing claims
to the President within the NIEs.65 Further, he personally ques-
tioned the wisdom of allowing a civilian agency clearly to dominate
npclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 26 -
the military agencies in the final strategic estimate. President
Eisenhower does not appear to have helped him resolve either un-
certainty. At that time, it was Eisenhower's known preference
to have disputes ironed out prior to NSC briefings. Also,
he was unsympathetic with "gap" predictions, reportedly inventing
the term "military industrial complex" when criticizing the
Pentagon's tendency to press for growth of defense budget items.66
Nonetheless, these indirect signals did
guidance on how to handle dissents, how
and whether to remain passive or choose
not provide the DCI clear
far to advance uncertainty,
between alternative positions.
Such questions became critical as the "bomber gap" grew into
the "missile gap" NIEs of 1957-1961. Dulles' initial hesitancy
about the CIA's dominant position with the strategic NIEs aptly
reflects the controversy surrounding them ever since.
D. Trends Since 1957
Since 1957, several changes in the strategic estimative
process have increased the management and decisionmaking burden
on the DCI. First, the CIA's internal strategic capability has
grown steadily. Ten years later, the DDI alone employed hun-
dreds of persons involved in strategic work. Of these, 178
were engaged in strategic research primarily related to the NIEs,
and in 1967 were grouped together in the Office
Research (OSR). Plus a sophisticated knowledge
of
of
nology, one of OSR's specialities is to "cost out"
production.
Strategic
advanced tech-
Soviet weapons
In 1979, OSR employed 260 persons, and also studied
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 27 -
Sino-Soviet strategic doctrine.67
For the CIA's total strategic
capability, this number is just the tip of the iceberg. For ex-
ample, the huge Directorate of Science and Technology (DDS&T)
created in 1965 is involved in complex research and technical
programs contributing to the estimative process. Chart A on the
following page describes the CIA's organization in 1972 for strategic
programs.
At the same time, the number of CIA persons drafting and
reviewing the strategic NIEs has been reduced. Whereas ONE em-
ployed a small, sometimes rotating drafting staff, when it was
replaced by the National Intelligence Officers (NI05) in 1973,
one person employing two assistants was responsible for drafting
NIE 11-3-8. Whereas the ONE staff was assisted by the twelve-mem-
ber Board of Estimates which reviewed the NIEs, when abolished
in 1973, no replacement was made until recently. In 1979, a
three-member Senior Review Panel was created to assume some of
the Board's functions. Finally, ongoing review of the NIEs by
outside experts has not been part of the official estimating
process since the early 1960s.
A third change, alluded to earlier, was the emergence of
a distinct institutional perspective within the CIA vis-a-vis the
Pentagon. Often, the CIA assumed an adversary relationship with
the military agencies, based on an assumption of biased military
estimating. As described subsequently, the image was heightened
following the entire community's "missile gap" overestimates
from 1957 to mid-1961. To quote a former Chairman of the Board
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
r-
WAND AND Of f ICE
OF NATIONAL ESTIMATES
- 28 -
CHART A
CIA ORGANIZATION 1972
MATERIAL SECURITY C0AM:IL
DIRECTOR OF
CENTRAL INTELLIGENCE
DEPUTY DIRECTOR OF
CENTRAL INTELLIGENCE
XECUTIVE OMECTOR
COMPTROLLER
MUM, TO ME OINECTOR
Of CIEISTRAL ISITELLIGE MM.
INEELUIGNICE CONESIRsrt
OFFICE Of MAIMING.
PROMMIREING.
/MD INOGETING
LEGIMATIVE COUNSEL
DEPUTY DIRECTOR
IONNETELLIMICE
CENTRAL
REPERENCE SERVICE OPERATIONSCENTER
OMMETC OFFICE Of MIN AND
CONTACT SERVICE IjOSOSISMNIC WELLED
FORMED IMMAOCAST OPMEG OF
RIPORSULTION KRIM. CURRENT INTELLIGENCE
INFORMATION
gtouintalcion STAFF
COMORE%
Off= Of
ECONOMIC MENEARCH
OFFICE Of
STRATEGIC RESEARCH
.1NATKINAL NOTOGRAMM
INTERPRETATION CENTER
REMIGNY
ANALTIMI ENTICE
1
I
I
I
I
J
DEPUTY DIRECTOR
FOR MANI
MINTY DIRECTOR FOR
SCIENCE MINI TECHNOLOGY
DEPUTY DIRECTOR
FOR WINN
[
DELETED
]
OFFICE Of
0:WOUTIR RIEIVICILE
OFFICE Of
MAGICAL SERVICES
OFFICE Of CLINT
OFFICE OF
CONNWIRCATIONS
OFFICE Of RESEARCH
AND DEVELOPMENT
OFFICE OF LOGISTICS
OMNI OF
MEDAL ACTIVITIES
OFFICE OF PERSONNEL
OFFICE Of SCIENTITIC
INTELLIGENCE
OFFICE Of SECURITY
FOREIGN MIME AND]
PACE AIMING CENTER
OFFICE Of TRAINING
OFFICE Of
MEWL PROJECTS
?TINE Of FINANCE
Source: Senate Selecte Committee on Intelligence
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 29 -
of Estimates: "...there is a natural thrust in military intelli-
gence to maximize threats and oversimplify the intentions of
potential adversaries."68 For its part, the Pentagon intelligence
agencies grew to distrust CIA analysts, who they argued tended
to underestimate the Soviet threat. James Schlesinger summarized
these views in 1976: "In fact, the [CIA's] intelligence directorate
tends to make a particular type of error systematically in close
harmony with the prevailing biases in the intellectual community...
There was an assumption that the Soviet has the same kind of arms
control objectives that they wished to ascribe or persuade
American leaders to adopt."69
Created to be free of bias, the
CIA thus was accused of perpetuating one.
A fourth change was in the nature of the issues debated. The
advent of improved satellite reconnaissance and other detection
means in the 1960s meant that, in the words of one participant,
the world was no longer "opaque."70
However, this prevented
neither mistakes nor quarrels. As Albert Wohlstetter has
documented, both civilian and military estimators "underestimated"
Soviet ICBM, SLBM and heavy bomber build-up in the mid and late
1960s.71
As the SALT treaties loomed large, both the DCI and
CIA were drawn into a qualitative debate about overall Soviet strategic
intentions. As DCI George Bush argued in 1978, perhaps prematurely:
The revolutionary developments in the technologi-
cal means of intelligence in the last 10 years
take care of many loopholes in terms of the pre-
cision of our projects. We don't debate about
a minor missile gap now. There's no debate. We
measure them, we count them, know where they're
npriassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 30 -
sitting, we look at them....There is debate, however,
about the things that can't show on a photograph
or can't be picked up by electronics: political
-factions' debates behind closed doors of politburos,
research laboratories, general staffs. These things
can't be learned by technology.72
Finally, the format of the strategic NIEs has changed markedly.
From a document presenting a dominant consensus and brief foot-
noted dissents, the strategic NIEs in the 1960s became highly
complex documents riddled with many, lengthy footnotes and
appendices. In 1969, the Nixon Administration found this confusing,
and subsequently asked the DCI to experiment with means of
presenting alternative views "up front" in the main text of the
strategic NIEs. Supported by the DCI, this experiment went through
successive stages. Now, the strategic NIEs may present what
are labelled "true alternatives" to the President, and the use
of footnotes has been abandoned.73
Thus, the current defintion of
a coordinated NIE is the opposite of what originally prevailed.
In terms of the NIE's format, the DCI's role has changed from
one of consensus-seeking to presiding over competition.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 31 -
III. THE DCI's MANAGEMENT OF STRATEGIC NIES
AT THE DRAFT STAGE
In the next four chapters, the roles of six DCIs since 1957
in producing the strategic estimates are examined in depth. The
focus is on the personal attitudes and leadership postures adopted
by successive DCIs. These are explored within an informal con-
text; that is, in the absence of formal legislative and Presidential
direction, DCIs have had the liberty to assume or ignore any of
the roles described. The author's conclusions for six DCIs and
seven possible roles are summarized in Chart B on the following
page. The chart is referred to in subsequent chapters.
These chapters are written from the DCI's perspective and
are mainly descriptive. The author is interested in the his-
torical pattern of behavior, establishing what six men have done,
or not done, to ensure that coordinated NIEs were produced.
Whether these actions resulted in improved NIEs is an evaluative
question, raised later only tangentially. For example, a DCI
may have initiated a reorganization which was later criticized
or reversed. Or, a DCI may have inserted his personal judgement
or sided with a minority view within an NIE, which in retrospect
proved inaccurate. Rather than scoring a DCI on whether he was
right or wrong, we seek to establish whether a DCI did anything at
all. How did DCIs regard the strategic NIEs, what roles did
they believe they should pursue, what tasks did they find difficult
or conflicting, what steps were ignored?
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
DRAFT
STAGE
OF
NIEs
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
CHART B
INDIVIDUAL ROLES ASSUMED BY DCIs FOR PRODUCTION OF STRATEGIC NIEs
(X - played a role; XX - viewed as major role by DCI; 0 - played little or no role)
ROLE IN PRODUCTION
1. Management Direction
to CIA Production 0 X
2. Management Attention
to DOD Coordination 0 0
DULLES McCONE
(1953-61) (1961-65)
3. Personal, Substantive
Input to Draft 0
4.
FINAL
DELIBERA-
TIONS AT
USIB OF 5.
NFIB
Support of CIA's
position X
(from '57 on)
Encouragement of
Competitive Debate 0
0
6. Final Adjudication
of Views
DRAFT OR 7. Promote Independent
FINAL (Outside) Review
NIEs
DCIs
HELMS SCHLESINGER COLBY TURNER
(1966-72) (1973) (1973-75) (1977- )
0 0
(W.H. after
0
0
XX
XX
0
X XX
0
(tenure too
brief)
indicates DCI responsive to Presidential or White House preferences.
XX
XX
XX
XX
(but mainly
prior to
NFIB in 3.)
0 0
(W.H. Bush
in '75) (1976)
(XX)
narinccifipri in Part - Sanitized Coov Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 33 -
This chapter discusses the actions of DCIs in managing the
estimative process prior to USIB consideration, at the time when
the NIEs are drafted by the CIA with input from the Pentagon.
It concerns the mechanics of the drafting process, for example,
who wrote the NIEs, how they were written, and how the Pentagon
intelligence agencies contributed.
A. Management Direction to CIA Production
The CIA's three functional Directorates have been described
in this manner: "The DDI is a production outfit and can run
itself, the DDS&T spends money, but the DDP always involves people
problems."1 For most years located in the DDI, the Board of
Estimates and ONE did "run itself." After the policy decisions
of DCIs Smith and Dulles previously described, their leaders
established the traditions and operational steps for producing
NIEs which persisted until 1973. Until then, DCIs remained in
a supportive role.
For example, DCIs Allen Dulles, John McCone and Richard Helms
did not review or initiate organizational innovations, dictate or
modify drafting and review procedures, or personally select the
drafting team, although they did approve Board appointments. Like-
wise, these DCIs did not decide when the strategic NIEs were
drafted, instigate innovations in style and format, or review
the theoretical underpinnings of earlier concepts. In part, this
reflects their satisfaction with the estimative process as it
was being run and their attention to other CIA programs. Even
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 "
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
- 34 -
John McCone who, compared to DCIs Dulles and Helms, was keenly
interested in management questions, reportedly spent "90% of
2
his time on clandestine (DDP) programs." However, as indicated
in Chart B, John McCone is given an "X" because of his role in
a breakthrough relating to highly classified, US strategic
planning data available to CIA estimators.
Beginning in 1973, a trend was established for a more active
management of estimative work by the DCI. DCIs James Schlesinger,
Williman Colby and Stansfield Turner believed that organizational and
drafting innovations were needed, and that management was a prime
function of the DCI. After Colby abolished the Board of Estimates
and ONE in 1973, both he and Admiral Turner personally hired
the National Intelligence Officers (NI0s), designed new drafting
techniques, and monitored the NIOs' progress by reading successive
draft NIEs.
Some observers claim that it is inappropriate for the DCI
to concentrate the power, resources and intellect of his office
to managing a handful of men who draft the strategic NIEs. Since
these persons originally were intended to be uniquely independent
but, in turn, totally dependent on external staff who collect
and analyze strategic intelligence, perhaps the DCI should turn
his attention to the rest of the CIA's strategic work. For example,
which strategic functions should be grouped together in the
DDO,* DDS&T and DDI and what are the optimal means by which
*The clandestine Deputy Directorate of Plans (DDP) was renamed the
Deputy Directorate of Operations (DDO) in 1973.
in Part - Sanitized Com/ Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 35 -
the three Directorates share work with one another? What is the
appropriate balance between human intelligence gathering (HUMINT),
electronic, communications and signals intelligence collection
(ELINT, COMMINT and SIGINT), and the political analysis of all
three.* How much should CIA drafters rely on CIA as opposed to
Pentagon work in producing the strategic NIEs?
Because of the enormity of these questions and the competing
claims of different functional units, this author's research
has focused primarily on the CIA's estimates staff and its coordina-
tion with the Defense Department intelligence agencies. Un-
less a DCI initiated organizational innovations in the CIA's
strategic intelligence work primarily for the short-term impact
on the NIES in mind, such changes are not discussed in this
study. Below, the DCI's management direction to CIA estimative
work before and after 1973 are described separately.
1. Before 1973
The early independence of the Board of Estimates and ONE
staff from the DCI is explained by the uniqueness of their
functions and the aggressiveness, small number and stability over
time of the individuals involved. The Board and ONE were designed
to offer the DCI an independent, collective analytical judgement
for his final review. Elite units of talented professionals,
*Sometimes analytical work also is referred to as HUMINT. In the
past decade, the question of whether the CIA is long on technological
collection programs and technical analysis, and short on political
analysis, has been raised by at least three DCIs--Schlesinger,
Colby and Bush. It continues to be a focus of outside criticism.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 36\ -
they were supposed to be isolated from the daily pressures and
competing demands of other CIA units. There were only four
Chairman of the Board and one, Sherman Kent, served from 1952-1967.
The Board's membership of approximately twelve was meant to be
well-rounded, and include one or more retired military officers.
The total ONE typically was about 50 professionals, variously
organized on a geographic and functional basis. The Chief of
ONE handled most administrative matters for the estimates staff,
in order to free the Board members for substantive work. In the
mid-1960s, it became common practice for ONE staff to be elevated
to Board positions.
The Board and ONE developed the format for the NIEs. For ex-
ample, all NIEs consisted of an opening, brief "Summary," followed
by tight paragraphs of "Conclusions" and a presentation of the
background and evidence. In the 1950s Sherman Kent developed
a lexicon of probabilities, which was a "system of carefully
graded verbal measures of certainty.0 An executive summary was
prepared according to a President's interest or request. For
example, for the strategic NIEs, President Johnson requested a
special one-page summary and President Carter is said to read up
to 15 pages.4 The Soviet strategic NIEs included a carefully
worded statement of overall Soviet strategic emphasis and inten-
tions, which beginning in 1969 was included in the executive
summary, and since 1974 constitutes a separate chapter of NIE 11-3-8.
im,,,Inecifiori in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 37
In the late 1950s, separate estimates in the Soviet "11" series
were initiated by ONE, reportedly in part to facilitate the input
of new CIA units to the draft.5 In the 1960s, the strategic
NIEs expanded from 30 to, by 1970, over 100 pages. Presently,
NIE 11-3-8 consists of an executive summary, and two volumes
of about 125 pages each. The first is a summary volume and the
second a documentary volume, containing detailed supporting
data, charts, graphs and tables.
As the format and timing of the strategic NIEs became
routinized, there was little need for the DCI to dictate in ad-
vance when they should be written and what topics to consider.*
For example, the "terms of reference" of the strategic NIEs was
circulated by ONE in late winter, a first draft completed in late
spring, followed by lengthy interagency coordinative meetings at
the CIA, which hopefully culminated in a final draft in October for
DCI and USIB approval. This schedule anticipated Presidential
consideration of the NIEs in preparing his defense budget. Generally,
each strategic NIE updated last year's predictions and inventoried
new weapons, for the next five years. As described later, in the
1970s DCIs sometimes decided at USIB whether to include new, highly
controversial subjects in the current or next year's strategic
NIEs.
The Board of Estimates early established two traditions which
have persisted with the DCI's support. First is to draft the NIEs
from the point of view of a foreign country, with no specific
*In contrast, DCIs often have been involved in the "tasking" and
terms of reference of non-strategic NIEs, such as during potential
crisis situations.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 "
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 38 -
reference to US policy options, military force levels, or conflict
scenarios. This practice is meant to free estimators from bias
that might result if they are drawn into debate on US military
programs, and thereby also increase the credibility of the NIEs.
Direct comparisons are considered a "Net Assessment" rather than NIE
function, and have been primarily the job of the Defense Department.6
Reportedly, in 1962 CIA estimators were so concerned with preserving
this distinction that they were reluctant to gain access to highly
classified US strategic planning data, lest the NIEs then appear
less independently written or unbiased.7
DCIs have reinforced this tradition. For example, Allen Dulles
systematically avoided statements that gave the appearance that
the CIA was involved in foreign policymaking. His colleagues des-
cribe how he refused to raise' even the implications of NIEs for
the US strategic arsenal, even when specifically asked by Congressional
and NSC members.8 Dulles never talked publicly about "gaps." "The
DCI's job was to provide the facts, period."9 John McCone is said
to be less of a "neutralist" or "purist" in this regard.10 "Mine
is not a policy job," he once said "but when asked I'll give my
opinion."11 In 1969, National Security Advisor Henry Kissinger
argued that rigid adherence to this tradition decreased the NIEs'
policy relevance. However, DCI Richard Helms held his ground,
and resisted the White House's initial proposal for involving
CIA estimators in SALT I verification predictions.12
The agree-
ment subsequently established remains in effect, with a distinction
drawn between predicting Soviet potential to violate the treaty
and verifying a substantial violation, which Admiral Turner considers
a policy matter.
im,,ninecifiori in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 39 -
However, White House interest in direct US-USSR force comparisons
in the NIEs has continued. In the 1970s, the strategic NIEs in-
cluded a brief numerical summary of current and predicted US force
levels in an appendix, and some requests were made for US-USSR conflict
scenarios. Recently, President Carter asked DCI Turner to offer
so-called "red-blue" comparisons between Soviet ("red") and US
("blue") forces in the main text of the NIEs.13 One CIA analyst
said that the issue of how to best present these "has not been
resolved. ,,14
A second practice established by the Board of Estimates is
to reach a unified CIA position within draft NIEs. That is, a faction
within the CIA does not have the option of taking a footnote.
In effect, the Board and not the DCI monitored disputes between
CIA estimators. One Board member "chaired" each NIE from the
beginning, and it was up to him to see that all CIA viewpoints
were aired before reaching a conclusion. Typically, this occurred
before the Pentagon agencies were called in to discuss the draft.15
A unified CIA opinion is, in part, inherent in the nature of es-
timative work. As one author notes, there is also a bureaucratic
imperative: a unified view reinforced the CIA's draft and reduced
the likelihood that military dissenters would seek alliances and
gain momentum.16
No DCI has reversed this practice. For example, DCIs Dulles
and McCone did not sit in on Board of Estimates meetings or monitor
drafts in progress to observe how differences were handled. Richard
Helms told this author that this is not an appropriate function
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 -
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 40 -
of the DCI, since it might reduce his authority or bias his
consideration of the final product.17 Even now, when the NIOs have
been given the license by Admiral Turner to explore all alterna-
tives, they consider it "logical for the CIA to pull together
at some point" and CIA differences that might persist are not
formally identified within the strategic NIEs.17a Earlier, several
observers argued that such differences occasionally were significant.
For example, during the 1957-1961 missile gap estimates a few
ONE staff are said to have believed that the Soviets could deploy
few ICBMs by the early 1960s, as opposed to the several hundreds
officially predicted in the NIEs.18 In the 1960s, one author con-
tends that on technical issues ONE generally took the position of
OSR and its predecessors, and the DDS&T. However, "on questions
involving a higher...political content, agreement was less certain."19
The lack of management direction to the estimates team by
DCIs Allen Dulles and John McCone reflects not only its aggressive-
ness and talent; it also reflects the faith these DCIs had in the
original set-up. This is demonstrated and was maintained by the
strong personal relationships between the DCIs and various Board
members. Dulles and McCone respected the Board, invited Sherman
Kent to morning staff meetings, discussed estimates individually
with analysts, and sometimes proposed new Board members.20 The
Board and ONE enjoyed a certain institutional independence from
the DDI. For example, they hired their own staff, and made
personnel transfers from other DDI units without the DDI's approval.
neclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 41 -
Organizational charts show that in the mid-1960s the Board was
elevated out of the DDI into a staff positon to the DCI.21 How-
ever, Kent's stature was equal to that of any DDI and this step
was significant on paper only.
John McCone initiated several functional reorganizations
within CIA which were heralded as appropriate rationalizations
of the CIA's growing strategic work, such as the creation of the
DDS&T in 1965.* However, of more direct, short-term impact on
the strategic NIEs was a breakthrough McCone precipitated in
gaining CIA access to highly classified US strategic planning docu-
ments generated by the Pentagon. McCone believed that "the DCI
had to be informed of US strategic capabilities in order to give
adequate intelligence support to the President."22
During the Cuban
missile crisis in October 1962, McCone acted as a key participant in
White House deliberations and the CIA estimators were asked to
draft several NIEs. At that time, McCone obtained the data he
requested on US force dispositions.
This was the wedge he needed. Following the crisis,
with encouragement from (Defense Secretary] McNamara,
he continued to make the requests. By the mid-1960s,
the DDI was procuring information on US strategic
planning on a regular basis. Consistent access to
this data increased the Agency's information base
considerably and further established the CIA's claim
to strategic research.23
*Illustrative of the continuing competition between various Directorates,
in 1965 the DDI criticized the DDS&T's taking of the Office of
Scientific Intelligence (OSI) on that grounds of weakening the DDI's
estimative work, and later reclaimed OSI. Photo interpretation
work has moved back and forth from the DDI and DDS&T.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
- 42 -
Actually, the CIA's access was not always "consistent" or easy.
For example, the Navy remained hesitant about providing the
CIA with some data. In the early 1970s, it refused to provide
CIA analysts with some requests. Reportedly, Richard Helms then
intervened on their behalf.24
Richard Helms long tenure as DCI, from mid-1966 to early 1973,
was characterized by passivity vis-a-vis the estimates team. CIA
analysts report that Helms maintained his distance from the Board
of Estimates, developed few personal relationships there, and was
little interested in estimative mechanics.25
Helms' DDIs were
forceful personalities who exerted a stronger administrative in-
fluence on facets of that Directorate's work. On the other hand,
Helms defended the Board and ONE'S right to be left alone to do
their work. "To some extent the Office was shielded by Richard
Helms. During a period in which the CIA was regularly under attack,
Helms found that the prestige and credibility of his analysts...were
great assets...."26
Unfortunately, during Helms' tenure the prestige of the CIA
estimators declined. As the Senate Select Committee on Intelligence
reports, the charge frequently was echoed that the Board and ONE
had become "insulated" and in-bred, and that Helms did not infuse
it with new blood.27 For example, the Board of Estimates gradually
changed from a broadly-based group in which individual members
had considerable stature and independent outside ties, to a place
of tenure for ONE staff.28
The last two Chairman previously had
narinccifipri in Part - Sanitized Coov Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 43 -
been Board members and before that ONE staff. Helms reportedly made
one Board appointment which was regarded as politically motivated,
and by 1973 there was no military representation on the Board.29
Pentagon intelligence staff privately argued that after new ONE
staff were hired, including former DIA employees, they soon acquired
a "CIA perspective."30
Helms' role in creating the Office of Strategic Research in
1967 was minimal. Helms approved the reorganization, proposed to
him by the DDI with the backing of Defense Secretary Robert McNamara.31
The consolidation of several strategic components in OSR meant that
the NIE staff could turn to one place within the CIA for finished
strategic analyses. OSR's stature was greatly enhanced by the
fact that Secretary McNamara immediately began to solicit its advice,
through special memoranda and reports as well as the NIEs. As one
author summarized:
In the 1960s, a period of disillusionment with military
estimating and one in which the civilians were
ascendant in the NIE process, the NIEs were found to
be useful counterweights to the excessive claims of
the military. McNamara started off expecting to
be able to use the DIA as his source of intelligence,
but he became increasingly dissatisfied with it.
As a consequence he turned more and more to the CIA
for estimates on the Vietnam War and Soviet military
capabilities. ONE did a number of papers at his
personal request, and McNamara encouraged the forma-
tion of OSR, a CIA office designed to compete with
the military on military estimates....McNamara
would take the NIE judgement as final, and would
refuse to speculate on any new Soviet capabilities
until they had been properly discussed through
the NIE process.32
According to the same author, the CIA estimates team relied in-
creasingly on OSR, compared to other sources, in preparing the
NIEs.33
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 "
6 14.,1,4
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 44 -
2. After 1973
Beginning with James Schlesinger in 1973, DCIs have demonstrated
more interest in managing the estimates function. During his five
month tenure as DCI in early 1973, Schlesinger took several steps
to revitalize the NIE process. For example, he examined options
for reorganizing the estimates staff, debated these with Board
members and finally determined to abolish the Board.34 He argued
that the strategic NIEs had become cumbersome, and urged stylistic
modifications to reduce their length and complexity.34a
For example,
he promoted the use of Interagency Intelligence Memorandum (IIMs),
to deal in-depth with highly technical issues, such as Soviet
civil defense and defense spending. The IIMs are coordinated in-
formally with the Pentagon and do not require USIB approval. Their
conclusions then may be summarized in the strategic NIEs.
To improve estimative work, Schlesinger arrived as DCI with
a "mandate for change" in the rest of CIA, following his report
in 1971, as Deputy Director of the Office of Management and Budget,
on intelligence community reforms. For example, in the report
Schlesinger had challenged the notion that "our hypotheses about
foreign intentions, capabilities, and activities have improved com-
mensurately in scope and quality as more data comes in from modern
collection methods" and was determined to upgrade analytical work
within CIA.35
As one colleague observed, if Schlesinger had re-
mained as DCI "he would have taken the place apart."36
William Colby accomplished the first reorganization of the es-
timates function within CIA in over twenty years, and then assumed
direct responsibility for it. Although several of his reforms have
in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 45 -
been criticized ever since, DCI Colby is given a "XX" in Chart
B for his personal role in them. Colby replaced the Board of
Estimates and ONE staff with 11 National Intelligence Officers,
called NIOs. Most NIOs had geographic responsibilities, e.g., Latin
America, and several had functional tasks, e.g., strategic in-
telligence and world economic trends.*
Colby had three rationale for this move. First, he hoped
to improve the NIEs, and respond to outside criticism, by creating
a drafting system which relied more heavily on the direct input of
non-CIA agencies. To accomplish this, the NIOs were located as a
staff function to the DCI and assigned only one staff assistant.
In theory, anyone could contribute draft material to the NIO, who
had both management and production responsibilities. That is,
the NIO was responsible for seeing that a draft was produced but
he did not necessarily have to write it himself. Rather, he
"managed" the work of other persons, for example, in DIA, whom
he might ask to write first drafts. Changes in drafting procedures
which resulted are discussed below in Section B.
Second, DCI Colby sought to sharpen the lines of responsibil-
ity by abolishing the Board of Estimates. As Colby told this
author, the Board tended "to isolate the DCI from the true believer."37
He testified to the Senate Select Committee on Intelligence:
I have some reservations at the ivory tower kind of
problem that you get out of a board, which is too
separated from the rough and tumble of the real
world....I think there is a tendency to be in-
stitutionally committed to an approach and to an
*Presently, there are additional "functional" NIOs for Warning,
Nuclear Proliferation and Arms Control.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
? ?
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
- 46 -
appraisal of a situation and to begin to interpret
new events against the light of a predetermined
approach toward thoseevents...I like the idea of in-
dividual total responsibility, and then you don't
get any fuzz about how there was a vote....That
really is my main problem with the Board, that it
diffuses responsibility....38
Colby's third rationale pertained, broadly, to the CIA as a
whole, the reorganization of which Colby said he had "just begun."39
The NIOs were supposed to be the pinnacle of a system more geograph-
ically focused to provide better intelligence. In addition to
estimating, the NIOs were meant to be the DCI's "eyes and ears on
the world" to provide him with prompt, up-to-date intelligence
answers and survey broad intelligence requirements." To aid them,
Colby became the first DCI to attempt to "force the DDO to interact
with other elements of the Agency," which his new Management
by Objectives system was supposed to facilitate.*41 Like Schlesinger,
Colby sought to increase the CIA's capacity to analyze foreign
political changes. For example, he created the first Office of
Political Research (subsequently abolished) in the DDI and by
firing the head of the DDO's Office of Counterintelligence, Colby
sought to refocus the DDO's Soviet work on Soviet political in-
tentions.42 As Colby recently said:
We used to go through a great deal of effort to dis-
cover the number of [military] divisions in Eastern
Europe, but we now get this information without the
use of agents or defectors. We do have to worry about
political estimating, a political judgement of the
forces at work in countries which are hidden behind
closed doors. 3
*Traditionally, the contact between DDO and NIE staff was limited
to rare personal visits of DDO officials to the Board of Estimates,
and circulation of covert agents reports "graded" in terms of re-
liability by the DDO. The Board of Estimates housed these in its
"all-source" reading room. Occasionally, the Board and ONE received
important agents reports which some DDI units did not receive, such
as those of Soviet military intelligence officer Oleg Penkovsky in
1961. Colby told this author that he once brought together a dozen
CIA country specialists, in the same room, to discover that most
never had met.
im,,ninecifiori in Part - Sanitized CODV Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 47 -
William Colby assumed an active relationship with his NIOs.
In personally hiring each one, he sought to bring in new persons.*
Of tfie first eleven, eight were new to the estimates process, one
was from ONE, another from the Board, and a third (the NIO for
Strategic Intelligence) had been with ONE in the 1950s and early
1960s. The NIOs were elevated to a staff relationship with the
DCI, and directly responsible to him, although administratively
supervised by a "chief" NIO. Colby met frequently with individual
NIOs, read successive drafts, listened to each NIO's analytical
problems, and tried to remain informed about coordinative difficulties
In addition, he turned to the NIOs for their views on broad ques-
tions, for example: "Are we collecting enough? Are we processing
raw data properly? Are we spending too much money on it? Are we
organized right to do the job?"45
When Admiral Stansfield Turner became DCI in early 1977, he
sought to refine Colby's managerial innovations and retains a keen
interest in management issues. For example, to reduce the intellect-
ual burden on the NIOs, he assigned them only one function, to
draft NIEs. To reduce the administrative responsibility of the
DCI, he appointed a director and small staff to oversee the NIOs on
a full-time basis. After several organizational experiments, Turner
relocated the estimates function back within the DDI, which was
renamed the National Foreign Assessment Center (NFAC). The first
head of NFAC, called the Director of National Foreign Assessment
(D/NFA), spent most of his time on estimative work. Chart C on the
following page describes the composition of NFAC in 1979. In late
*The NIO who wrote the NIE on Soviet Bloc conventional forces was
an active Admiral.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Chart C
National Foreign Assessment Center- 1979
Senior
Review Panel
Director
Deputy Director
Associate Director-Management National intelligence
Associate Director-Substantive Support Officers
Administrative
Planning Staffs
601005
Coordination
Staff
Current Reporting
Group
Office of Office of
Central imagery
Reference Analysis
Office of Office of
Economic Geographic Si
Research Cartographic
Research
CIA Operations
Center
-
Publications
Presentations Group
Office of Office of
Scientific Weapons
Intelligence Intelligence
Office of Office of
Regional & Strategic
Political Research
Analysis
Requirements &
Evaluation Staff
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 49 -
1979, the newly appointed head of NFAC was asked to reevaluate
the NIO set-up, deciding whether further organizational innovations
were needed. Reorganization now is in progress.
in early 1979, Admiral Turner reconstituted the collective
NIE review function within CIA, a step advocated by several govern-
mental study groups. For example, in 1976 the Senate Select Committe
on Intelligence recommended that "the Director establish a board
to include senior outside advisors to review the NIEs."46 Turner
hired a three-man Senior Review Panel, which included an eminent
academic in strategic affairs, a former Ambassador, and a four-
star general. The Panel serves both the DCI and the NI0s, and
reviews IIMs as well as NIEs, at several stages during the
drafting process. However, unlike the Board of Estimates, the Panel
informally advises as opposed to collectively approving the NIEs.
In 1979, questions were raised as to whether the Panel had either
the prestige or authority to function effectively within CIA.47
Admiral Turner has promoted innovations in drafting procedures.
For example, each NIO has a list of expert consultants he may
use in drafting. Several years ago, a consultant was hired to
write the entire first draft of NIE 11-4 on Soviet strategic doc-
trine, and presently this NIE is drafted by the NIO for Soviet Bloc
affairs. Much flexibility and diversity prevails in the drafting
of non-strategic NIEs. For example, each NIO follows different
and highly individualistic drafting procedures,* and may use
varying formats.43 The NIOs are encouraged to prepare NIEs when
*For example, one NIO may write a first draft himself before calling
in other agencies for discussion; another may ask a CIA team and/or
other agencies to draft assigned portions of an NIE, which he then
edits; and a third might solicit "think" papers from outsiders
which he uses in preparing an NIE himself.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 50 -
they see the need to do so, as opposed to routinely addressing
topics on an annual basis. This pertains to NIE 11-4 as well.
In total, the number of NIEs has been reduced from about 50 a
year, to 20-25 a year. Like Colby, Admiral Turner maintains an
ongoing, personal relationship with his NIOs throughout the
drafting process, reviewing and monitoring their work during
successive stages.
B. Management Attention to Department of Defense Coordination
This section explores the DCI's role in the mechanics of
coordinating the CIA draft with Department of Defense (DOD) in-
telligence agencies, prior to USIB consideration. The Pentagon
agencies discussed are Army, Navy and Air Force Intelligence and
the Defense Intelligence Agency (DIA) created in 1961.*
As indicated in Chart B, most DCIs focused little attention
on the means by which the Pentagon might best contribute to the
draft strategic NIEs. After the mid-1950s, the Board of Estimates
and ONE took the lead in defining the timetable and mechanisms
for the Pentagon's involvement. The three military services and
DIA exercised their prerogative in deciding whom to send to CIA
coordinative meetings, what kind of internal work to generate
parallel to the CIA's, and how far to advance disagreements in
the CIA draft. Beginning in 1969, Presidents urged changes in
the format of presenting military dissents in the strategic NIEs.
With the DCI's support, the CIA estimates staff gradually implemented
*The input of the National Security Agency (NSA) is not discussed
here. Before 1961, the Joint Chiefs of Staff J-2 intelligence arm
participated in the drafting and had a seat on USIB. The DIA re-
placed it.
r)..r.iaccifiart in Part - Sanitized Copv Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 51 -
these changes. The exceptions to this pattern of traditional non-
involvement of DCIs are William Colby and Stansfield Turner. By
different means, both sought to enhance the Pentagon's direct
input and improve interagency coordination in the drafting stage.
Both are given a "XX" in Chart B.
1. Before 1973
As described earlier, General Bedell Smith was the first DCI
to elicit the cooperation of the Pentagon in drafting NIEs. Later,
Allen Dulles gave the CIA what has been termed the "power of the
draft" for strategic estimates, by supporting its inclination to
prepare independently a first draft.49 However, he considered it
inappropriate to prescribe optimal coordinating tactics to Defense
Secretaries. Thus, the Board of Estimates and ONE adopted a pat-
tern for coordinating the Pentagon's response whichpin several
respects, remains similar today.
First, early in the year ONE circulated the terms of reference
of the forthcoming NIE, and invited interagency comments. Then,
it continued an ongoing survey of raw intelligence and finished
analyses on strategic issues from different CIA and DOD units.
It often solicited additional work from CIA, DOD, consultants,
and special USIB subcommittees. These subcommittes are interagency
groups, often initiated by the DCI, assigned specific topics for
study. For example, a USIB subcommittee important to drafting
NIE 11-8 was the Guided Missiles and Astronautics Intelligence
Committee (GMAIC)* created by Allen Dulles in 1956, to "track the
*This subcommittee was first called the Guided Missiles Intelligence
Committee (GMIC).
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
- 52 -
rabbit to its lair."5? The ONE had complete freedom to decide what
work to collect, consider and integrate into its draft. That
is, no DCI established guidelines to assist it in choosing, re-
jecting, or balancing different sources. ONE then completed a
first draft, submitted it to the Board for approval, and circulated
it within the Pentagon and other agencies for review.
The next stage of interagency coordination was, and is, diffi-
cult and protracted. Intelligence agency representatives were in-
vited to CIA to discuss the draft. These meetings went on throughout
the summer and fall, and sometimes were reconvened after DCI and
USIB review. They were chaired by a Board member and later an
NIO, and included CIA representatives and often more than a doz-
en Pentagon staff, but not the DCI. Here the draft strategic NIEs
were heatedly debated, and here the CIA's "power of the draft" was
keenly experienced by the DOD representatives. The purpose was
to explore and redefine all areas of agreement and, failing that,
to agree on how to disagree, prior to submitting the final draft
to the DCI and then USIB. Since the CIA was well prepared with
a completed document in hand, the Pentagon representatives were
put in the position of reacting to, or effectively challenging, what
already was written. Further, ONE did most subsequent redrafting.
As one observer notes: "In the draft ONE would resolve...ambiguity
according to its own views, using style and structure as well as
argument. "51
in Part - Sanitized CODV APproved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 53 -
The DOD representatives often arrived at CIA for discussion
ill-prepared to influence ONE or make concrete rebuttals. For
example, during the late 1950s and the 1960s, they did not come
with completed analyses parallel or competitive to ONE'S, which
could be integrated into the draft. Rather, the representatives
were familiar with their own agencies intelligence studies, esti-
mates, budgetary and posture statements, the parameters and time
frames of which might differ from the CIA's draft.52 Thus, they
relied on verbal means to influence ONE and, from time to time,
called in different intelligence specialists to advance technical
arguments. Traditional interservice rivalries and budget competi-
tion make it impossible for the military representatives to form
stable coalitions to use as bargaining tools with ONE. For ex-
ample, in the early 1960s the Army and Navy typically lined up
against the Air Force in predicting Soviet ICBMs in NIE 11-8;
but the Army and Navy often disagreed on Soviet SLBMs in NIE 11-8;
and on Soviet ABM potential in NIE 11-3.53
The creation of DIA in 1961 did not alter this seemingly ran-
dom approach to the draft strategic estimates. The DIA was es-
tablished, with the backing of President Kennedy, to work on
strategic intelligence matters pertaining to the three services'
tactical work, and speak with a unified voice for them. However,
the DIA was not "functionally geared to the production of con-
vincing estimates."54 Earlier, Allen Dulles had written: "Two
powerful and well-financed agencies such as DIA and CIA will become
rivals and competitors. Some of this could be healthy; too much
of it could be both expensive and dangerous."55 Actually, in the
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 54 -
beginning DCI John McCone worried that DIA might exert too much
pressure to conform on the three military services, and a CIA
group set about reviewingits programs.56 However, soon McCone
complained to McNamara that DIA's work was "irrelevant and redundant"57
and both men reportedly lost interest in ways to overcome what
more than one observer has called DIA's "inherent bureaucratic
weaknesses" in competing with the military service intelligence
agencies."
If the CIA had the "power of the draft," the three military
services had the "power of the footnote." Footnotes represented
these issues for which a compromise through rewording and restructuring
could not be reached, or to which the Pentagon would not acquiesce.
In the late 1950s and 1960s, they typically were specific. For
example, CIA and Air Force numerical predictions of Soviet ICBM
deployment over the next five years might differ by a 100 or more,
or the two might disagree on the characteristics of a Soviet missile
such as the SS-9 versus the SS-11. Then, the Air Force had the option
of taking a footnote at the bottom of the page, saying: "Air Force
Intelligence disagrees....It believes...." Until 1969, footnotes
were accepted as a clear-cut, useful method of registering dis-
approval. The CIA did not deny their use, and no DCI or other top-
level official sought to alter the system.
The most clear indication of tacit acceptance of the footnote
system was its pervasive use. Throughout the 1960s, the strategic
NIEs were rife with numerous and complex footnotes to the main
text, which was supposed to represent the consensus. For example,
various Pentagon agencies disagreed with the CIA's position on a
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 55 -
Soviet ABM system, MIRV capability, and other aspects of new delivery
vehicles and warheads. Sometimes, a majority of pages had foot-
notes,. For example, in NIE 11-8-65, the dissents are said to have
been larger than the text itself, and thus the evidence for dissents
eventually was presented in appendices prepared by the Pentagon.59
On occasion, DIA took an independent position "contrary to the views
of all the service agencies."" For example, in the above-mentioned
NIE, an upper limit of Soviet ICBMs was predicted only in a foot-
note supported by the three military services, but DIA dissented
from the footnote.61 After reading NIE 11-8-68, Henry Kissinger
asserted that the document made little sense, since the text was
"bland," "inconclusive" and refuted or "taken back" in footnotes.62
By the early 1970s, the Pentagon began taking footnotes to less
specific issues. For example, in the opening pages of the NIE,
it might footnote what one participant called "the basic tone"
or broad implications of the NIE.63
The first effort to rationalize the footnoting system came
from President Nixon. In early 1970, Dr. Kissinger requested DCI
Richard Helms to experiment with new means of presenting dissent
in the NIEs. Kissinger argued that a fuller sampling of alternative
views and supporting evidence in the main text would be useful to
the White House." As in a lawyers brief, he wanted the reasons
for rejecting various options officially recorded. In other words,
the CIA (and by implication the DCI) was not required to reach
a dominant consensus on highly controversial issues, but offer
the policymaker "facts, not opinions" which he then might arbitrate.65
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 56 -
Helms agreed, but left it to the Board of Estimates and ONE to
implement the request. In NIE 11-8-70, a few alternative views
were presented "up front" in the main text, although footnotes
were still used for some time. To illustrate the stylistic diffi-
culty in implementation, this NIE was over 100 pages, which was
over twice as long as its predecessors, and one footnote was a
page long analysis.66 When James Schlesinger became DCI, he said
he wanted alternative positions presented even if the estimators
"had to cut the page in two."67
The White House also offered the Pentagon a license to more
systematically compete in the drafting of strategic NIEs. In 1970,
a special Directorate of Estimates (DE) was created in DIA, headed
by the highly qualified Army General Daniel Graham. Graham began
producing estimates called DIEs, focusing on narrow but "the more
contentious topics," such as Soviet MIRVing of a single missile,
which were important to the NIE debate. "In some ways the DIEs have
served as a dry run for the NIEs--used to create a more rigorous
DIA position on controversial issues."" They were prepared before
the NIE coordinative meetings and, importantly, sent to the White
House. Henry Kissinger also stepped outside the NIE process by
urging defense intelligence officials, as well as non-intelligence
officials such as the Director of Defense Research and Engineering
(DDR&E), to help draft the White House's more policy-oriented
National Security Study Memorandum (NSSMs). In so doing, they
could incorporate selective material from the NIEs. Soon DCI
Helms found himself in debate with the DDR&E at the White House,
in which NIE conclusions were reopened."
in Part - Sanitized CODV Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 57 -
After 1973
William Colby was the first DCI to take the lead in altering
existing patterns of DOD coordination. His 1973 NIO reorganization
was meant to increase the direct input of the Pentagon to the
first draft. He told his NIOs to set up procedures by which
this might be accomplished. The following year, DIA was asked
to prepare a first draft of one out of five chapters of the com-
bined NIE 11-3-8, and the State Department's INR was given the
chapter on Soviet strategic doctrine. In 1975, DIA drafted two
chapters,
Air Force
with this
including the one
assisted.70 Reportedly, DIA officials were "delighted"
on Soviet doctrine, in which the
opportunity.71 Presently, DIA still drafts at least
one out of five chapters of NIE 11-3-8.
One participant described this new technique as the "bottom
up" approach to coordination.72 DCI Colby hoped it would in-
crease the personal stake that DIA, and the services if they par-
ticipated, had in the draft, and heighten interest and skill in
sustained estimative work. Further, it was meant to introduce
a more "competitive element" into the drafting, by removing "a
certainty that a particular agency will have a decisive say in
a particular estimate."73 However, the real impact of this inno-
vation now is debated. First, observers note that the outside
contributions were weak stylistically, and subsequently edited
or rewritten by the NIO for Strategic Intelligence.74 Second,
the DIA's interest and service cooperation may have declined
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 "
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 58 -
recently.75 Third, NIE 11-3-8 still is written mainly by one
Nb, including its executive summary. The estimators themselves
assert that one benefit is a common data base and shared knowledge.76
While CIA staff have acquired specialized knowledge from the
military, e.g., on how missiles systems are operated, military
officials have broadened their grasp on a range of topics.
DCI Colby supported his innovations by assuming a personal,
coordinative role. Primarily, he did this by listening to Pentagon
representatives who came to his office to outline their grievances
about the draft, and then reporting these complaints to the Nb.
The then head of DIA, General Graham, argues that Colby was "highly
accessible" to requests for hearings, and that coordination in
the drafting improved during his tenure as DCI.77 Colby also ad-
vanced the presentation of alternative views in the main text
of the strategic NIEs, with President Ford's support. By 1976,
italics were used to highlight differences. For example, one
paragraph might present the majority view, followed by a paragraph
in italics representing another view. Footnotes were used for less
controversial, discrete, or individual dissents.
Admiral Turner plays an even stronger personal role in managing
competition during the drafting. Like DCIs Schlesinger and Colby,
he supports the trend of incorporating alternative positions in an
NIE's main text, a policy for which President Carter indicated
his agreement.78 However, Turner has extended this concept by
virtually eliminating the use of footnotes. He directed the
head of the National Foreign Assessment Center and his NIOs to
consider all alternative views "regardless of their origin,"
im,,,J?,ifinf-1 in Part - Sanitized COCA/ Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 59 -
and eventually narrow them down to a few.79 By discarding foot-
notes, Turner hopes that the Pentagon will explore and delineate
their differences more concretely, reaching agreement on which
are the most significant. The implication is that the Pentagon
should weight their dissents internally, advancing and documenting
those which reflect a major predictive uncertainty, and abandoning
those less central to the current year's strategic arguments.
Alternatives are presented in the NIEs as "parallel texts."
That is, different positions are alternatively developed, para-
graph by paragraph, throughout the text.
Like General Bedell Smith over 25 years earlier, Admiral
Turner personally intervenes to assist CIA staff in eliciting
from the Pentagon well-reasoned, agreed-upon, objections. He
rigorously examines successive drafts, and then takes the initiative
in raising strategic topics directly with military officials.
For example, if he believes that a Pentagon option is inadequately
expressed or lacks technical documentation, he calls up the
Pentagon analyst involved and says so." If, after further re-
vision, Turner still has questions about the position advanced,
he is likely to bring these to the attention of the relevant in-
telligence head, for example, the Director of the Office of Naval
Intelligence. Often this occurs during regular meetings between
all the intelligence heads in the DCI's office. If he still has
reservations about the wording or logic of the alternative, he
may call the head of the military service involved, for example,
the Chief of Naval Operations, and ask his opinion. By going up
the chain of command, Admiral Turner is interested in discovering
whether the dissenting view is widely shared by Navy officials.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 60 -
IV. SUBSTANTIVE INPUT OF DCIs TO
THE DRAFT STRATEGIC NIES
This chapter explores the role of DCIs in contributing sub-
stantively to the draft strategic NIEs. This author considers "sub-
stantive" any personal input a DCI makes, in his view, to improve
the content or increase the validity of a specific NIE. Substantive
inputs occur at several levels. For example, a DCI may edit
successive drafts for clarity, offer his personal views to es-
timators, devise new ways of evaluating technical issues, eliminate
spurious arguments, or represent his personal judgements in the
final draft.
For strategic NIEs, the substantive input of DCIs usually
has been considered a function of their technical background
and interest in strategic affairs. As the Senate Select Committee
on Intelligence reported, before 1976 "only three DCIs attempted
to address their primary attention to the quality of intelligence
production: Walter Bedell Smith, John McCone, and James
Schlesinger."' This was, in part, a "function of their background."
This chapter supports this conclusion. As indicated in Chart B,
out of six DCIs, the three who did make substantive input--John
McCone, James Schlesinger and Admiral Turner--had previous tech-
nical training and/or familiarity with strategic issues.
This chapter also examines a DCI's substantive involvement
as a function of his personal attitudes and administrative style.
Here, all six DCIs have behaved differently. For example, John
McCone frequently interacted with the CIA estimates team, by
asking questions, debating and reviewing its work. However, on
occasions when he differed sharply with it, he permitted the
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 61 -
staff view to go forward to USIB before reaching a final decision.
In contrast, Admiral Turner reads and edits successive drafts
from the beginning, and at times plays a major role in determining
how and which competitive ideas are expressed prior to NFIB con-
sideration. For this reason, DCI Turner's substantive role in
drafting is what the author later describes as "adjudicatory,"
and he is given a "XX" for this function. James Schlesinger is
also given a "XX" for his unique intellectual contribution. On
the following pages, the six DCIs are described in sequence and
occasional differences in behavior for strategic and non-strategic
NIEs are noted.
Allen Dulles
Like Bedell Smith, Allen Dulles spoke of the NIEs as "his"
estimates. As chief intelligence advisor to the President, Dulles
considered it his prerogative to insert personal judgements into
the NIEs, a view which the Board of Estimates supported.2 His
attitude of possessiveness was heightened by his personal role
in briefing President Eisenhower, his brother as Secretary of
State, and other top officials, on the NIEs. During the Eisenhower
years, all NIEs were linked formally to the NSC process. For ex-
ample, NIEs appeared on the weekly agenda, and were utilized and
appended to policy papers prepared by the NSC's Planning Board.
Regular NSCmeetingswere opened by the DCI's briefing on major
intelligence issues, including new NIEs, which one participant
said "was never less than 15 minutes long."3 In retrospect, one
Pentagon official argues that Dulles' briefings represented the
"most effective and timely" means of reaching policymakers.4
im,,ninecifiori in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 62 -
However, in practice Dulles' substantive input to strategic
and non-strategic NIEs was limited by the fact that he paid
little attention to the text of the draft NIEs. Rather, Dulles
relied on informal and verbal means to influence the CIA es-
timates team. For example, during the year estimators were kept
informed of his concerns in the course of meet-ingsnot primarily
devoted to the NIEs, such as the DCI's morning staff meetings and
current intelligence briefings. Sometimes, Dulles called junior
analysts in ONE, to offer his own view or highlight new intelligence.
Then, the Board and ONE were left alone to do their work. Allen
Dulles did not read successive drafts, and his first formal review
of NIEs occurred when the final draft was completed.5 Then, he was
briefed on the draft by Board members just prior to NSC or USIB
meetings. At this time, Dulles focused primarily on the summary
and conclusions of the NIE.
His concern here was threefold. First, his colleagues re-
port that he was intolerant of what he considered wordy, imprecise
or pedantic sentences, and might make editorial changes to en-
hance their concreteness.6 Second, he listened and queried the
estimators, until he felt comfortable in presenting the NIEs to
the NSC or USIB the following day. Typically, a staff member
then was selected to accompany him to meetings. Third, for
non-strategic NIEs, Dulles occasionally modified the conclusions
to reflect new intelligence to which he was privy. Estimators
report that, in several instances, they were saved from embarass-
ment by the DCI's intervention at the last minute.7 However, they
reported no instances in which the content of strategic NIEs was
so altered.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 63 -
If Dulles influenced the estimates team by informal means,
they, in turn, had to systematically convince him of the appro-
priateness of their strategic predictions. This took place
throughout the year. One Board member argues that Allen Dulles
"worried over each strategic estimate."8 One of ONE'S directors
asserts that "language describing qualifying and accurately
quantifying the Soviet threat came from CIA draftsmen, and had
to be fought through up to the DDI and DCI decision levels."9
In the late 1950s, the CIA's internal research on Soviet production
capabilities was vitally important to its prediction of an on-
going Soviet ICBM build-up, since little hard data from the Soviet
Union was available. For example, the press reported that the
200 U-2 flights over the USSR up to May 1960 detected only two
operational ICBM launchers in Soviet territory. 10.
John McCone
John McCone's attitudes about the strategic NIEs were curiously
mixed. Board of Estimates members say that he initially did not
regard the NIEs as "his" estimates. Rather, he considered them
a product of good staff work for which the DCI should adopt a
"hands-off" policy.11 Further, McCone was highly critical of the
quality of past NIEs. When McCone arrived as DCI in late 1961,
he privately rebuked both the CIA and Air Force for their roles
in promoting the missile gap theory over the last four years.12
He then witnessed the rapid explosion of this myth, following
the first satellite reconnaissance evaluated in August 1961.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 64 -
For example, between late 1960 and late 1961 the CIA halved its
predictions of Soviet ICBMs for mid-1963, from 400 to 200 ICBM5.13
Actually, by mid-1963 it was estimated that the Soviets had built
only 100 ICBM5.14
Nonetheless, once the drafting process was in motion, McCone
found it difficult to stay on the sidelines, and Board of Estimate
members contend that they consistently urged him to become more
involved. As the former Chairman of the AEC, an engineer and
successful corporate executive, McCone was well-equipped to do
so.* As one colleague reports, he wanted the NIEs to be "right"
and so utilized in policymaking.15 He had "the inquiring, skeptical
turn of mind of the good intelligence officer" and found technical
intelligence "endlessly fascinating. u16
McCone's subsequent involvement in drafting has been de-
scribed as "enterprenaurial," although he was not consistently
or equally aggressive for each NIE.17 Throughout the year, McCone
kept up-to-date on new intelligence, analytical problems and
community-wide disagreements by demanding "instant, full briefings
on anything that caught his attention.018 He frequently called
up Board members to ask questions, give advice and, occasionally,
argue his viewpoint. When the final draft reached his desk, he
examined it thoroughly, in the words of one aide, "going over
each line as if it were a corporate mortgage.H19 Then, it was
not surprising for him to call Sherman Kent at home, early in
the morning, and say something like: "'On page 20, you say this....
Can you prove it.,n20 McCone was impatient with what he considered
imprecise and ambiguous statments, and sometimes edited the draft
himself.
*Less remembered is the fact that McCone's own engineering firm was
involved in shipbuilding and aircraft production in WWII, and after
that he became Under Secretary of the Air Force.
narlaccifiPci in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part- Sanitized Copy Approved forRelease2012/05/04 : CIA-RDP93T01132R000100020034-5
- 65 -
On a few occasions, DCI McCone differed sharply with the CIA
draft. He then had a choice: either let the NIE go' forward to
USIB, where it could be reviewed again; or change the NIE to re-
flect his view.* McCone found the first option most appropriate,
and stopped short of inserting his view to the draft prior to
USIB debate. The best example of this is his behavior during the
now infamous NIE, in September 1962, on the likelihood that USSR
might place offensive nuclear weapons in Cuba. Although this NIE
was not part of the "11" strategic series, it partly hinged on an
ongoing debate about Soviet difficulties in perfecting their ICBM
guidance system.21 In a rare community-wide agreement, this NIE
predicted that "it would be incompatible with Soviet policy to
introduce strategic missiles in Cuba."22
For some time, McCone had believed otherwise. Following in-
telligence reports on an increased Soviet presence in Cuba over
the summer of 1962, McCone warned President Kennedy in August that
"something new and different" was going on.23 Before leaving on
his honeymoon trip to France in early September, McCone privately
said that he was "very worried" and requested an NIE.24 The
first draft did not support McCone's hunch. While abroad, McCone
supplied the Board with a barrage of cables pressing his claims,
and asked that the draft be reconsidered. With no dissents, it again
concluded that "the Soviets simply would not do anything so un-
characteristic, provocative and unrewarding," but Sherman Kent
advised McCone to alter the NIE if he wished.25 McCone did not,
believing that the NIE's arguments were sufficiently well-reasoned
to remain inviolable. A U-2 flight over Cuba on October 14 proved
that McCone's prediction was correct.
*At that time, a third option later employed by William Colby and
Admiral Turner, to present the DCI's opinion separately in the
NIE, was not considered compatible with the NIE process.
narinccifipri in Part - Sanitized CoPv Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 66 -
Richard Helms
Richard Helms' concept of the DCI's role in the drafting process
was well-defined. Helms told this author that the DCI should dis-
tance himself from the drafting, lest early association with differ-
ent viewpoints bias him in favor of one or another.26 The DCI
should preserve his freedom to make final, substantive decisions
just before or at USIB, by not becoming an advocate too early.
Helms considered the NIEs "the Director's piece of paper" and thus
the DCI "had the right to intervene or change them" if deemed
necessary. 27
DCI Helms did not exercise this option for the draft strategic
NIEs. Helms had risen up through the ranks of the DDP in CIA, and
his primary interest was political intelligence. When he arrived
as DCI in mid-1966, a unique technical competency was required
of the DCI. For example, one controversy concerned the "mysterious
footprints"of Soviet MIRVs, evidence which the entire intelligence
community interpreted differently. 28 Some participants questioned
Helms' basic interest in strategic matters, and one Pentagon
official described Helms, compared to other DCIs, as "the most
dilatory in terms of substance."29 Helms himself has said that
many of the issues at hand ultimately were "unknowable" and his
instincts came into play only in "the final phase of production."3?
Helms did review the final drafts of the strategic NIEs, with
two concerns in mind. First, he tried to keep the process moving
in an orderly, timely fashion, and second, he preferred to have
interagency disputs resolved prior tp USIB consideration. Thus,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 67 -
he reviewed drafts for internal consistency and clarity, and if
he ascertained that an NIE was so deficient or that the consensus
offered masked realdivisions, he promptly sent it back to ONE
for revision.31 After 1969, the Board and ONE were left alone to
administer changes in the presentation of dissent in the strategic
NIEs, mandated by the White House. Thereafter, the DCI experienced
more difficulty in sending them to USIB on time. Sometimes, Helms
had to make the decision to go forward with the NIE as it was.
The only documented instance in which Helms personally inter-
vened in a strategic draft before USIB followed a request from
Defense Secretary Melvin Laird. In NIE 11-8-69, the CIA included
a summary paragraph which read: "We consider it highly unlikely
that they [the Soviet Union] will attempt within the period of
this estimate to achieve a first strike capability..."32 Just be-
fore the scheduled USIB meeting, one of Laird's assistants asked
Helms to revise the controversial statement since it "contradicted
the public position of the Secretary."33 Helms deleted the entire
paragraph, although the State Department's representative on USIB
reintroduced it as a footnoted dissent.
James Schlesinger
During his brief tenure as DCI in the beginning of 1973, James
Schlesinger was involved intimately in drafting the strategic NIEs,
primarily on substantive grounds. A highly trained economist, former
strategic analyst at the RAND Corporation and former AEC Chairman,
Schlesinger was qualified to address complex strategic issues.
Earlier, he had criticized the CIA for possible "bias" in assessing
nariaccifipn in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 68 -
the Soviet threat, and as DCI appeared willing to take on CIA analysts
on their own ground.34 For example, the then Chairman of the Board
of Estimates reports that Schlesinger frequently debated past
NIEs with him, pointing out sentences and phrasings that, in the
DCI 'S view, were misleading or inaccurate.35 As the Senate Select
Committee on Intelligence concluded: "Schlesinger's overall ob-
jectives were to maximize his role as Director of Central Intelli-
gence rather than as head of the Agency and to improve the quality
of the intelligence product."5
One of Schlesinger's major contributions was in strategic
methodology. CIA analysts explain that he was highly creative
and helpful, for example, in devising new methods by which to
count Soviet MIRVs and techniques to ascertain the characteristics
of new Soviet cruise missiles.37 Schlesinger enjoyed reading
early drafts, and then often summoned analysts to his office to
draw charts or demonstrate how a problem might be approached mathe-
matically. Describing Schlesinger, William Colby said that he
was "the one DCI" who grasped the strategic debate intuitively,
and would have elevated the current state of art of strategic
analysis. 38
William Colby
William Colby also involved himself deeply in the draft process
but, for the strategic NIEs, not on a substantive basis. Like
Richard Helms, Colby rose up through the ranks of the CIA and had
little specific expertise in strategic topics. Colby told this
author that he did not have the technical knowledgeability to make
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 ?,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 69 -
personal input to the draft.39 Rather, he saw the primary role
of the DCI as effectively managihg the competitive process, the
chief virtues of which were thoroughness, patience, and open-minded-
ness as opposed to intervention or arbitration. Colby presided
over the strategic NIEs at a time when, in the words of one CIA
analyst, the "consensus had fallen apart."" He hoped to improve
the quality of the NIEs by increasing their representativeness.
Like McCone, DCI Colby viewed the draft as a prelude to USIB
debate and endorsement. Colby read successive drafts with the
hope of better grasping the strategic debate and ensuring that
minority views were considered by his NI0s. As described earlier,
Pentagon officials were granted individual hearings with the DCI
during the drafting, after which Colby raised their complaints with
the relevant Nb. For example, DIA officials argued their new es-
timates in his office and Air Force intelligence chief Major
General George Keegan presented him with numerous, documented
dissents, as outlined in the next chapter of this study. At the
drafting stage, Colby was observed to be more aggressive in offering
advice and editing the non-strategic, compared to strategic NIEs.41
However, when he disagreed with the draft, Colby typically waited
until USIB to register his views.
Admiral Stansfield Turner
Admiral Turner systematically makes substantive input to the
strategic NIEs. Like McCone and Schlesinger, his Naval training
provides a technical background in strategic issues to make him
npclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 70 -
useful in critiquing NIEs. As DCI, he was "highly dissatisfied
with a large number of national intelligence estimates" and ex-
ercised his mandate to play a role in improving them.42 This role
is well-defined and highly personal. As Admiral Turner recently
argued, the NIE process is ultimately a one-man system. After an
estimate is prepared by the community, "the one-man system comes
in, because I decide, I sign for it, I vouch for it."43
Admiral Turner takes the time to stay in close touch through-
out the drafting process, primarily by devoting attention to each
successive draft. As one aide reports, Turner is involved in
the draft strategic estimates "from beginning to end," from the
first one prepared by the NIO and Pentagon to successive redrafts
after interagency coordinative meetings. His "critiques occur all
throughout the process" and his approach is "socratic."44 Turner
carefully examines the drafts in his office, underlines the parts
he believes need improvement, and suggests ways of doing so. For
example, he may ask whether a line of reasoning is relevant, out-
line alternative ways of handling a point of ambiguity, or state
how he disagrees. The draft is sent back to the NIO and a dialogue
begins. This frequently entails discussions between the DCI and
junior analysts as well, and, as examined earlier, between the DCI
and Pentagon officials. Turner then reads the next draft to ascertain
if the changes make any difference. If he considers these in-
adequate, he may redraft paragraphs or entire sections himself.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
- 71. -
Admiral Turner is given a "XX" in Chart B for another reason.
This is the personal role he sometimes plays in deciding which
views are advanced as major alternatives in the text, decisions
which occur prior to NFIB review. The author describes these
decisions as adjudicatory, and Turner is not the first DCI who has
personally arbitrated between views. However, unlike other DCIs,
Turner does not use NFIB as a final debating forum and discourages
the use of footnotes as the ultimate recourse of dissenters. Thus,
his choices at this time may determine the final content of the
NIEs.
Two examples illustrate Admiral Turner's different handling
of alternative views within draft NIEs. Recently, alternative
positions were expressed in a draft of NIE 11-4, on how the Soviets
might view US behavior in a particular conflict scenario.45 The
disagreement was explicit and intricate, based on interpretations
of recent Soviet publications on strategic doctrine. Admiral
Turner reviewed the evidence on both sides, and finally agreed to
go forward with the CIA's view. This he slightly modified, re-
placing the phasing "Soviet intentions" with "Soviet expectations."
On the other hand, Admiral Turner upheld the presentation of
alternative positions in a recent NIE 11-3-8. At issue was the
strategic implications of the Soviet Backfire bomber. The CIA
and Air Force disagreed on the potential range of the aircraft by
at least 1,000 miles." Both had examined the question at some
length, hiring professional contractors to construct experiments
in Part - Sanitized CODV Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 72 -
to test the aircraft's potential range. The evidence on both
sides appeared adequate and, therefore, the dispute was not re-
solved within the NIE. Instead, the NIE went forward with the
two opinions intact.
A third option employed by Admiral Turner was first used by
William Colby. In cases where the DCI disagrees with both sides
or strongly endorses one view, Turner may attach a covering memo
to the NIE or insert a statement into the NIE, which explains this.
That is, the DCI separately states his opinion in a letter or
paragraph to effect that "The DCI thinks..." Turner also uses
this technique to explain how decisions were reached within the
NIE. The author does not regard this as adjudicatory.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 73 -
V. THE DCI's LEADERSHIP ROLES AT
USIB OR NFIB
This chapter examines the varying approaches of DCIs to the
strategic NIEs during the final phase of endorsement by the USIB,
or now NFIB. Here again, a DCI's posture during USIB deliberations
reveals his attitudes about the NIEs, his role as CIA director
versus coordinator of the intelligence community, and what kind
of document best serves the President. Three possible roles are
discussed; first, support or advocacy of the CIA's position; second
encouragement of competitive debate and a multiplicity of views;
and third, final adjudication of alternative views, i.e., personal
decisions which resolve conflict or otherwise change the NIE's sub-
stantive content.
As indicated in Chart B, the author concludes that the first
role was emphasized by Allen Dulles, vithin an historical context
that excluded other roles on his part. Competitive debate at USIB
was encouraged by DCIs John McCone, James Schlesinger and William
Colby. Final adjudication was a role exercised occasionally by
DCIs John McCone and Richard Helms. As just described, DCI
Stansfield Turner also acts as an adjudicator, but primarily out-
side the context of NFIB.
This chapter focuses on USIB because, since IAC in General
Bedell Smith's time, it was meant to be a high level forum for NIE
debate and final endorsement. Composed of the heads of the in-
telligence agencies, USIB is intended to have the flexibility to
reexamine disagreements and the decisionmaking authority to resolve
them once and for all. As Chairman, the DCI is supposed to play a
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .?
Declassified in Part- Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 74 -
community-wide coordinative role, as highlighted in the late 1960s
by the appointment to USIB of the Deputy Director of CIA (the DDCI)
to formally represent the CIA's views. However, the DCI and USIB
members frequently have been criticized for acting only as a rubber
stamp for the draft NIEs. For example, one opinion is that "most
DCIs have been reluctant to engage in confrontation with members
of the USIB over substantive findings.'1 Another is that USIB
members lack specific expertise and "were reluctant to quibble...with
the preparatory work done by its,staff."2
This study's findings do not support the view that most DCIs
and USIB members have been passive during USIB meetings. Historically,
USIB often has been a major debating ground for the strategic NIEs.
USIB discussion sometimes has been lengthy and intense, and on
occasion the content of the NIEs altered as a result. Most critical
to this study, some DCIs have exerted individual leadership over
the process. Contrary to the opinion that "the power balance of
the USIB did not vary greatly from that involved in the inter-agency
review," the intervention of some DCIs upset this balance.3
A. Support of the CIA's Position
Only one DCI examined here, Allen Dulles, saw his primary role
at USIB as personally supporting the CIA's position within the stra-
tegic NIEs. Dulles' role in promoting the CIA's view must be placed
within the context of the missile gap debate of 1957-1961. Dulles'
earlier decision to permit the CIA to draft the strategic NIEs
according to its independent view was a policy decision which
nariaccifipn in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 75 -
subsequently elevated the CIA's role in the estimative process.
It also reflected an interest in enhanced competition within the
NIEs,-since the CIA's view was presented in addition to what it
considered exaggerated and offsetting claims of the three military
services. In the late 1950s, Dulles recognized that he thereby
was lending support to NIEs which offered competing claims to
President Eisenhower, a concern which continued to plague him.4
"When asked to answer the single most important intelligence ques-
tion of the period, the intelligence community provided a series
of contradictory estimates and gave the impression that the Soviet
ICBM programme was far more expeditious and productive than it in
fact turned out to be."5
The missile gap estimates thus became the classic case of
disunity within the intelligence community. The only point of
agreement was that there might be a "gap." For example, in
NIE 11-8-60 the CIA predicted that the Soviets would build 400
ICBMs by mid-1963; in footnotes, the Army and Navy predicted 200
ICBMs, and the Air Force 700 ICBMs, by mid-1963; the State Depart-
ment and JCS were in-between the CIA and Air Force.6 To some
observers, the CIA's position represented an attempt for institutional
compromise rather than a logical extension of concrete analysis.
As General Daniel Graham later wrote of the missile gap years:
"Curiously,...the CIA position was generally--and sometimes precisely--
halfway between the Army-Navy view and the Air Force view. To the
military analysts this smacked of a political rather than a care-
fully reasoned, objective position."7
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 76 -
Nonetheless, once having decided to assert the CIA's individual
perspective within the strategic NIEs, Allen Dulles' instinct
was to support "his" analysts publicly. Dulles reasoned that pre-
USIB debate on the draft gave sufficient hearing to all parties
and, after this, the institutional independence, prestige and
morale of the CIA was on the line'. Personally, Dulles was dis-
inclined to participate in prolonged debate at USIB, and he did not
vigorously advocate the CIA's position during the oral discussion.
Rather, he promoted the position of CIA analysts by presiding over
USIB deliberations in a manner which did not permit reopening of
previous decisions. Further, he personally briefed the President
on the final NIEs and, in one instance, attached a memorandum
to NIE 11-8 personally endorsing the CIA's position. Parenthetically,
Dulles' colleagues report that as the missile gap debate intensified,
the CIA drafters had to work hard to convince him of their position
and that the U-2 flights were targeted to close the CIA's un-
certainty.8
As Chairman of USIB, Allen Dulles is said to have been firmly
in charge of its proceedings. While his attitude about discussion
was more permissive than DCI Smith's before him, to the Pentagon
participants the outcome was a foregone conclusion.8 By the end
of 1960, one Presidential study group complained that USIB had
become primarily a "deliberative body" neglecting its broader
tasks of intelligence community coordination.1? Still, deliberations
followed a certain format. The DCI summarized the NIE, and then
each military service and the JCS representative presented their
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 77 -
dissents, for example, the Army argued its case for Soviet emphasis
on medium-range ballistic missiles and inadequacies in ICBM trans-
portation facilities.11 Generally, individual positions were well-
entrenched. For example, when the Soviets ceased ICBM testing from
April 1958 to March 1959, the CIA argued that this indicated
performance trouble while the Air Force maintained precisely the
opposite.12 After listening to each side, Dulles patiently re-
stated his affirmation of the draft NIE and closed the meeting. No
redrafting was done. Dulles then scheduled a meeting with President
Eisenhower to brief him on the NIE. One Pentagon official
said that Dulles' briefings effectively foreclosed any end-running
to the White House.13
The missile gap predictions advanced by Allen Dulles to
President Eisenhower were anything but popular with the Administration.
One observer wrote that Eisenhower considered these NIEs "probably
invalid" and he did not respond as promptly as he had during the
bomber gap debate.14 For example, Eisenhower deferred authoriza-
tion of increases in first generation Atlas and Titan missiles
until the first Polaris and second generation Minuteman missiles
were available and evaluated. During Kennedy's successful presiden-
tial campaign on a strategic gap platform, the DCI was repeatedly
pressed by journalists to give his personal opinion. Dulles up-
held the NIEs as valid intelligence projections in a case where
sufficient doubt ekisted. When NIE 11-8-60 was presented to
President-elect Kennedy, Dulles attached a personal memorandum
explicity stating his support of the CIA's position in the main
text of the NIE.15
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05;04 : CIA-RDP93T01132R000100020034-5
- 78 -
That subsequent DCIs did npt see their primary role at USIB to
promote the CIA's position resulted, in part, from the missile gap
experience. In mid-1963, it was estimated that the Soviets had
constructed only 100 ICBMs, proving that the entire intelligence
community earlier had been wrong. However, by this time the CIA's
position within the strategic NIEs was no longer on the ascen-
dant, but was dominant. Moreover, a distinct institutional perspective
vis-a-vis the Pentagon had emerged following the missile gap over-
estimates. As General Graham later wrote: "To put it bluntly,
there is a considerable body of opinion...which regard threat
estimates prepared by the military as being self-serving, budget-
oriented and generally inflated....It stemmed from a series
of bad overestimates [which] have hung like albatrossess around
the necks of military intelligence officers ever since."16 Since
the DCI did not have to throw his weight consistently on the
side of the CIA in order to protect its institutional view and
independence, he has been free to assume other roles at USIB.
B. Encouragement of Competitive Debate
Of the five subsequent DCIs examined here, three--DCIs John
McCone, James Schlesinger and William Colby--actively promoted
competitive debate on the final draft of the strategic NIEs. All
three considered USIB as an appropriate, higher-level sounding
board or second stage for the presentation of new ideas and a
reopening of old dissents. They personally participated in debate
and acted as a court of last resort for intelligence officials
Darf - Raniti7ed CODV Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 79 -
dissatisfied with the draft. Under McCone and Colby, USIB meetings
were lengthy and intense. Both DCIs reserved their judgement and
were willing to alter the strategic NIEs as a result. Although
Schlesinger's tenure was brief, interviewees describe his attitudes
as being similar to McCone's. For his philosophy of supporting con-
sistent debate at USIB, Colby is given an "XX" in Chart B. In
comparison, the inclination of DCIs Richard Helms and Stansfield
Turner has been not to use USIB (or NFIB in Turner's case) as a
forum for debate. As documented earlier, Admiral Turner encourages
competition within the final NIEs chiefly by efforts to improve
coordination at the draft stage. Below, these five DCIs are described
In sequence.
John McCone
In response to criticism about Allen Dulles' inattention to
community-wide coordination, President Eisenhower once said: "I'd
rather have Allen as my chief intelligence officer with his limita-
tions than anyone else I know."17 In contrast, President Kennedy
explicitly directed John McCone to strongly assert his role as in-
telligence community coordinator.18 McCone accomplished this, in
part, through USIB.
DCI McCone believed that USIB should serve as a prominent
debating field in which the full multiplicity of views could be
expressed, within the context of seeking his approval of the NIEs.
One of the first battles McCone waged to support this view was
with Defense Secretary McNamara, over the composition of USIB after
1961. As one observer noted: "The intention of the Administration
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
80 -
had been to charge the DIA with the resolution of conflicting
service estimates and to make its presentation to the United States
Intelligence Board (USIB) stand as the final judgement of the
Pentagon."19 After DIA was created, McNamara proposed that it re-
place not only the JCS representative but also the heads of Army,
Air Force and Naval Intelligence on USIB. They, and the DCI com-
plained. John McCone reportedly argued with McNamara that this
step might weaken the DCI's ability to hear multiple, conflicting
views, and lead to "watered down" estimates.2? Thus, a compromise
was arranged whereby the three military service intelligence chiefs
were invited to USIB meetings. While they lost their official member-
ship or "vote," they retained an equal right to argue and dissent.
In effect, each service had a "right to appeal past the DIA" to
USIB and its Chairman.21
McCone demonstrated an interest in listening to these appeals,
if well-reasoned and thoughtfully presented. Significantly, McCone
became the first modern DCI to be willing to support a service
position over that of the CIA's. As one author wrote: "McCone
showed that he could be convinced by good arguments, and gained a
reputation for objectivity by being ready to overrule the CIA on
the United States Intelligence Board."22 According to CIA es-
timators, this happened in the case of several NIEs assessing progress
during the Vietnam war, for which McCone was known to be "hawkish"
in disputing the effectiveness of a gradualistic approach to bombing
of North Vietnam.23 During USIB debate on one such NIE, McCone told
CIA analysts to redraft the estimate after talking to military
commanders "who know what is going on."24
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 81 -
A critical example of McCone's support of competition with-
in the strategic NIEs is NIE 11-3-63, regarding the possibility
that the Soviets were deploying an ABM system around the so-called
Tallin Line. The potentiality of an "ABM gap" was one of the most
hotly contested strategic issues in the first half of the 1960s.
In 1963, a major division existed between the CIA, supported by
the Navy and State Department, and the Army, supported by the Air
Force and DIA. The CIA maintained that the observed installations
around the Tallin Line did not represent an early ABM system, but
rather were relegated to anti-aircraft defense. The opposite view
was a more pessimistic or "hard assessment" which viewed the new
installations as the beginning of a full-fledged anti-missile system. 25
At USIB, McCone was impressed by the Army's line of reasoning. 26
After the Army's initial presentation, McCone scheduled another meeting
for elaboration. As described in the following section, McCone then
assumed an adjudicatory role and officially sided with the Army's
view, as the final text of NIE 11-3-63.27 After this, he hired the
Army general delivering the presentation, who later became head of
the CIA's new DDS&T.
McCone was impatient with debate at USIB that he considered
repetitious, once he had made up his mind. For example, in the
early 1960s he is said to have been intolerant of the Air Force's
views on Soviet ICBM delployment and often cut short debate on NIE
11-8. In 1963, he sided with the CIA's prediction of 300-600
ICBMs by mid-1967 and considered the Air Force's predicted level
of 700-800 by that date much too high.28 Albert Wohlstetter documented
how, in the succeeding years, the Air Force estimates gradually con-
verged with those of the CIA. "In 1965 and 1966 [convergence] was
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 82 -
complete in the near term and in 1967 it became complete for all
years."29 In 1974, Wohlstetter wrote that for the years of his
study, 1962-1-969, the intelligence community had "systematically
underestimated" Soviet heavy bomber, ICBM and SLBM deployment.30
Richard Helms
Richard Helms' tenure as DCI was divided in two parts: from
1966 to 1968 he served under President Johnson and from 1969 to 1972
he served under President Nixon. Helms' inclinations were the
same in both periods. He believed that drafting strategic NIEs
was the job of professional analysts, who should be able to reach
agreement prior to USIB. Helms did not enjoy listening to debate
on highly technical issues, and argued that USIB was not the proper
place to reopen past grievances.31 Rather, USIB meetings should
lead to timely approval. President Johnson asked that the strategic
NIEs be reduced to a one-page executive summary, which does not
suggest much encouragement to the expression of competitive views.
In his Administration, USIB meetings reportedly were some of the
shortest in length hitherto, typically lasting only forty-five
minutes.
During the Nixon Administration, competition gradually in-
creased. As the SALT I treaty loomed large, Pentagon dissenters
were less willing to relinquish their arguments without a fight,
and both the White House and Defense Secretary Laird indirectly
encouraged competition by criticizing the strategic NIEs. Helms
thus was faced with potentially long and angry USIB meetings. Es-
pecially irritating was to be assured that disagreements had been
?cnniti7p4innnvApproved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 83 -
resolved, only to have them "pop right back at USIB."33 Helms told
this author that he favored enhanced competition within the NIEs,
that estimating was a "give and take" process in which it was
conceivable that different parties might exaggerate their claims
in order to emphasize a point, and that CIA estimators sometimes
regarded him as too "hard line."34
However, his handling of competition was simple. He listened
to competing argument only long enough to determine that it had
not been resolved. Then, he closed the meeting, and typically sent
the NIE back to ONE for redrafting. Participants say that he
signaled the end of the meeting by removing his glasses.35 After
1970, successive redrafting sessions, followed by USIB meetings,
were required. As a consequence, several strategic NIEs were not
officially approved until the spring of the following year.36
James Schlesinger
DCI Schlesinger enjoyed competitive debate at USIB and actively
participated in it. He considered USIB a place where intelligence
chiefs with broad responsibilities and interests could meet in an
informal atmosphere. His colleagues describe how he sat back in
his chair, puffed on his pipe, and listened. Then, he might in-
itiate a lengthy dialectic with the intelligence chiefs, during
which new options might be considered.37
Like McCone, Schlesinger was impatient with assertions
but respected stimulating and documentable analysis. As described
earlier, one of his major concerns was to keep competition in
manageable proportions. For example, he layered the strategic
estimating process by making some issues, such as Soviet civil
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 84 -
defense, the subject of IIMs, then summarized in the NIEs. Much
discussion concerned how to best present major alternative positions,
and stylistically control the complexity of the strategic NIEs.
During Schelsinger's tenure, raw data was contained mostly in appendices.
William Colby
William Colby utilized USIB as a court of last resort, where
appeals to the draft NIE from the Pentagon could be heard. Colby
considered himself primarily a manager of competitive debate. In
the words of one aide, he preferred open, group discussion at USIB
as a means of hearing dissent, rather than private meetings, and
believed that USIB was a place where all could actively participate
in endorsing the draft.38
During Colby's tenure, different Pentagon officials challenged
the CIA's basic assumptions about Soviet behavior within the
strategic NIEs. Notably under attack was the CIA's premise,
summarized in a paragraph of the executive summary, that the USSR
was not seeking strategic superiority, or a first strike capability,
vis-a-vis the United States. For example, General Daniel Graham
as head of DIA called this "the pablum statement" and, as noted
earlier, began footnoting "the basic tone" of the NIEs.38a Ob-
jections also were offered to the CIA's format for handling dissent.
For example, as head of Air Force Intelligence, General George
Keegan objected to placing some dissent in footnotes rather than
in the main text, the presentation of supporting evidence in
appendices, the exclusion of critical dissents from the executive
summary, and the elimination of pivotal issues, such as Soviet
defense spending, from NIEs.38
nna-Inecifiarl in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 85 -
At USIB, Colby listened to several hours of debate on one sec-
tion of an NIE, and scheduled other meetings to continue. To
follow-up meetings, Colby invited the Pentagon intelligence chiefs
to bring along staff and written documentation. General George
Keegan now came to USIB meetings armed with new Air Force trans-
lations of Soviet strategic writings in order to demonstrate that
CIA estimators had misinterpreted evidence about Soviet inten-
tions.40
To support his claims that the Soviets rapidly were
deploying new missiles and undertaking a massive civil defense
effort, Keegan brought along satellite photos, engineering reports
and Air Force estimates of Soviet defense costs. They were passed
around the table. According to General Graham, Colby gave Keegan,
"who tends to round things off to the nearest million," every
opportunity to present his case.41
For each major argument, this
never took less than one-half hour. Then, Colby often authorized
marginal changes at USIB. For example, Keegan was permitted to
take additional footnotes, rewrite existing footnotes, present new
evidence in an appendix, or substitute wording in the text.42
By 1975, Pentagon criticism of the CIA draft NIE 11-3-8 had
reached such a pitch that the DCI took an unprecedented step. Prior
to the scheduled USIB meeting, Colby held a special conference
of all intelligence agency heads and staff "to ensure that they
were informed of all difference of opinion" prior to reviewing
the draft itself.42a
This meeting was the first, acknowledged con-
cession to military officials who complained that it was difficult
to comprehensively express their divergences from the CIA draft,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .?
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 86 -
while at the same time having to "chop up, edit and revise" the
text.43 This meeting reportedly lasted over six hours, and the
DCI was present the entire time.
Admiral Stansfield Turner
Admiral Turner's attitude about competitive debate at NFIB
is similar to that of Richard Helms, although unlike Helms, he
spends considerable time on competitive issues at the draft stage.
In the words of one aide, Admiral Turner prefers to "work text" in
his office rather than subjecting the draft to lengthy oral debate.44
Turner believes that most issues of dissent successfully can be re-
solved during the drafting process. At NFIB, the heads of the
intelligence agencies then should be in a position to discuss and
work over the NIEs' conclusions, rather than the entire text. Re-
portedly, NFIB meetings for both strategic and non-strategic NIEs
now take place approximately once a month, and last from one to one-
and-a-half hours. After the meetings, Turner is said to spend more
time than past DCIS in reviewing, editing and polishing the final
documents.45
C. Final Adjudication of Views
The extent to which DCIs have inserted their personal judge-
ments in the final strategic NIEs now has become a subject of
controversy. For example, the Senate Select Committee on Intelligence
chose several cases to investigate regarding DCI Richard Helms'
alleged role in altering final NIEs, to illustrate what it labels
neclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 87 -
"the constant tension...between the DCI...and the agencies."46
Presently, Admiral Turner's critics maintain that he "tinkers" with
the estimates and "has been more demanding and more pre-emptive
than any Director of Central Intelligence in recent times."47
The author regards as "adjudicatory," decisions made at the
time of USIB which involve a choice between existing options and,
consequently, alter what the draft NIE says. For example, a DCI
may arbitrate between competing views by deciding that a new one
should dominate or existing one be eliminated; or, a DCI may find all
views insufficient and draw up a new solution reflecting his personal
judgements. The author does not regard as adjudicatory, actions in
which a DCI permits additional footnotes, inserts his own view
separately in the final NIEs, or assists all parties in reaching a
conclusion by offering his personal opinions. Likewise, upholding
the CIA's position is not considered here as adjudicatory, unless
there is evidence that the DCI judged all alternatives equally and
impartially, and/or modified the CIA text.
An adjudicatory role for the DCI was implied in the original
concept of the NIEs and USIB's predecessor body. The IAC, and then
USIB, has been called the "Supreme Court" or "Board of Directors"
of the intelligence community, although there was never any formal
vote-taking.48 As Chairman, the DCI was more equal than the other
members in his right to participate. General Bedell Smith made
it clear that, as part of the DCI's role in coordinating and approving
the NIEs, he was free to select or eliminate views, or present his
own judgements in the final NIEs. No President explicitly ruled
out this role for the DCI although, as described earlier in the
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
-88 -
Nixon Administration, Henry Kissinger implied that he himself
should act as chief intelligence advisor and coordinator in ad-
judicating competing claims.
Nonetheless, most DCIs since General Smith have approached this
role for the strategic NIEs with caution and exercised it on rare
occasions. As seen, it ran against DCI Dulles' personal and po-
litical inclinations to act aggressively as a public advocate for
one or another view. Further, he did not separate himself from
the CIA as an institution on strategic matters and defended it as
necessary.* Of the five subsequent DCIs examined here, one, William
Colby, argues that adjudication is not an appropriate role for the
DCI to play. As indicated in Chart B, three DCIs--John McCone,
Richard Helms and Stansfield Turner--on occasion made adjudicatory
decisions, but their approach has varied.** Moreover, Richard
Helms told this author that he did not consider himself an adjudica-
tor and, in describing Admiral Turner's role, one aide said that
"he would not call himself that."49
Below, occasional differences
in approach to the non-strategic NIEs are noted.
John McCone
Initially, DCI John McCone did not believe that the DCI's
role should be adjudicative, but regarded the NIEs as a product of
good staff work. In the case of the 1962 Cuban missile estimate,
*As noted earlier, Allen Dulles was more aggressive in presenting
his personal views in the non-strategic NIEs.
**Because of Schlesinger's brief tenure as DCI, it is impossible to
determine whether he played an adjudicative role at USIB. The
reader should bear in mind that Admiral Turner's adjudicatory decisions
primarily are made prior to NFIB meetings.
narinccifipri in Part - Sanitized Coov Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 89 -
he criticized and conjoled estimators to no avail and then per-
mitted the NIE to go forward as it was. However, in this case,
McCone's personal relationship and accessibility to President
Kennedy were secure and he expressed his personal views directly
to the President.50
When McCone did play an adjudicatory role,
he was careful to
exercise it within the context of USIB. Adjudication is precisely
what occurred in 1963 during the USIB debate on potential Soviet
ABM capability. McCone listened intently to each side's views on
the question of whether the Tallin Line installations represented
the beginnings of an anti-missile or anti-aircraft system. He
called for an elaboration of the Army's view of the former possibility,
and eventually sided
with it. The text of NIE 11-3-63 then was
rewritten to present the Army's predictions that the Soviet intent
was to deploy an ABM system around the Tallin Line. Since there
was no provision for the CIA to take footnotes to an NIE, CIA
analysts say that, in this instance, "the specificity of their view
was lost."51
The less pessimistic assessment, which regarded the
new installations for bomber defense, was represented in footnotes
taken by the Navy and State Department.
This case is the most clear-cut example of adjudication at
USIB that this author
could find. It is interesting that, at the
time, McCone's action was not challenged by CIA
grounds of being inappropriate or heavy-handed.
that this is because McCone's USIB decision was
seemingly deliberative. In retrospect, for the
11-3-63, McCone's view proved inaccurate.52
In
estimators on the
One can speculate
highly visible and
time frame of NIE
the following year,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 90 -
CIA convinced him that he had been wrong and its position in the
"ABM gap" debate once again assumed the text of NIE 11-3. McCone's
1963 decision opened Pandora's box, however, and subsequently "some
of the technical panels serving the USIB felt so deeply that the
CIA was dangerously underestimat*ng the ABM danger that they took
their views directly to the White House."53
Richard Helms
DCI Helms is described by his contemporaries as acting as a
strong adjudicator at USIB. For example, in discussing Helms'
handling of the non-strategic NIEs, William Colby said that it was
typical for Helms to listen and then say: "'I disagree; I think we
should say this...."54 Then, the NIE would be sent back to ONE
for rewriting to incorporate the DCI's opinion. Richard Helms once
said of an NIE on Cambodia: "...in the end, I want a good paper
on the subject even if I have to make the controversial judge-
ments myself..55 To others, Richard Helms had judicial qualities.
For example, one aide said that.he was the last DCI for whom all
USIB members rose to their feet when he entered the room or closed
the meeting.56 He gave the appearance of being beyond the fray and,
above all, decisive. One author describes Helms' handling of the
Vietnam NIEs in this manner: "Throughout the paper wars, Helms
was a bureaucratic general."57
On the other hand, Helms rarely was involved in substantive
debate on strategic matters, and this author could find no in-
stances in which Helms overruled the CIA at USIB on strategic matters.
On the contrary, if he ascertained that USIB debate was inconclusive,
narlaccifian in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 91 -
he typically sent the NIE back to ONE for redrafing to incorporate
a Pentagon view. He was reluctant to redraft himself and did not
permit any Pentagon participant to do so. The best example of this
is his handling of NIE 11-8-68. Following USIB approval in June
1969, Dr. Kissinger "asked that [the NIE] be rewritten to provide
more evidence supporting the DCI's judgement" that the Soviet MIRVing
of the SS-9 was not imminent, a conclusion with which Kissinger was
known to disagree.58 The Board of Estimates rewrote the NIE, pro-
viding additional argument that the "triplet" under discussion
was not a MIRV. In investigating this case history, the Senate
Select Committee on Intelligence concluded that Helms had not at-
tempted to influence the outcome.59
However, this author concludes that Helms did act, at times, in
an adjudicatory capacity, which differed from the approach of either
DCIs McCone or Turner. Helms' adjudicatory role was management-
oriented, exemplified by decisiveness and brevity in handling
the strategic NIEs, and motivated by an interest to produce the
NIEs on time. If the DCI determined that the discussion was going
to be protracted and ultimately irresolvable by more ONE redrafting,
he "approved" the existing version and sent it to the White House.
This led to complaints of arbitrary behavior. For example, one
Pentagon official said that the status of dissents "was left
hanging in the air" and that he was uncertain about the content
of NIEs until receiving the final, printed version.60
Another official complained that Helms sometimes resolved
debate by eliminating a heatedly contested point from the NIE
altogether, i.e., overruling a dissent.61 For example, General
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
1
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 92 -
Keegan was not allowed to include his views on Soviet civil de-
fense measures in the summary of NIE 11-3 until the early 19708.62
In 1970, DIA completed a study, presented at USIB, of a new approach
to estimating the size of the Soviet defense budget. Reportedly,
this study was aimed at demonstrating the CIA was grossly under-
estimating the size of the Soviet commitment to new strategic weapons.
At USIB, the DCI agreed that the report was interesting and worth-
while, but asked DIA to publish its views outside the current NIE.63
James Schlesinger
Because of Schlesinger's brief tenure, it was impossible for
this author to determine whether he played an adjudicatory role,
as indicated in Chart B. If he had remained as DCI, it is highly
probable that he would have so intervened. This speculation flows
from his heavy substantive involvement in drafting, encouragement
of debate at USIB, open mind about the credibility of CIA work, and
impetus to keep competition within reasonable bounds in order to
present a stylistically coherent document to the policymaker. Ad-
judication usually is considered the strongest role a DCI might
play in producing the strategic NIEs. In this author's view, DCI
Schlesinger would have found this role intellectually challenging
and necessary to resolve issues of sustained competition in the
strategic NIEs.
William Colby
William Colby told this author that it was not the DCI's job
to arbitrate between conflicting views in the strategic NIEs.64
Two convictions seem to underlie this concept: first, that the
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 93 -
DCI's main role is to open competitive debate at USIB, which
during Colby's tenure reached its highest pitch; second, that the
subject matter was technically too complex for the DCI to effectively
intervene, without appearing to be arbitrarily choosing sides. As
one observer noted, Colby's political instincts were on the "humanist
left" but he did not impose these on USIB members.65
That is not to say that Colby was passive during USIB delibera-
tions. As documented earlier, he often requested that additional
evidence be presented to shed light on a difficult issue. After
listening to extensive debate, Colby sometimes summed up the
discussion by offering broad, even philosophical observations of
his own, without dictating their inclusion in the NIE. For example,
Colby believed that conventional "theater balance" considerations
tended to be lost in the shuffle of complex Soviet strategic pro-
grams, and that Soviet technological gains might have a political
impact on less developed countries.66 Colby's willingness to
marshall diverse evidence at USIB gave the appearance of impartiality
in weighing arguments, even if the draft NIE remained unchanged.
For example, in the case of the "Particle X" debate, Colby heard
extensive argument before siding with the prevailing scientific
(and CIA) viewpoint, after which General Keegan was permitted to
take a footnote.67
Unlike Helms, Colby did not refer all re-
drafting back to the CIA. Also, he did not redraft himself. As
described earlier, Colby typically resolved debate by authorizing
marginal changes or additions to the draft NIE at USIB, such as a
new footnote, which then was written by the dissenting member.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 94 -
DCI Colby behaved more aggressively during USIB debate on
the non-strategic NIEs, for which he sometimes held views distinct
from the CIA's-or a Pentagon agency. In such cases, Colby represented
these separately in the text of the NIE, following USIB deliberations.
That is, he added his own opinion in a separate sentence or para-
graph, rather than redrafting the NIE to reflect this. For example,
in one NIE on Latin America, Colby took one position, CIA and the
State Department's INR another, and DIA and one military service
a third." During Colby's tenure, the Deputy Director of CIA,
General Vernon Walters, also employed this option in instances when
he sided with the Pentagon. For example, in one NIE on Italy, the
DDCI sided with DIA during USIB debate. Colby sent the estimate
back to his NIO for drafting to incorporate a joint DDCI-DIA alterna-
tive."
Stansfield Turner
This author credits Admiral Turner with having played the
strongest adjudicatory role of any DCI. The difference between DCI
Turner and other DCIs is a matter or degree and timing. As the New
York Times reported, "what has changed, it appears, is Admiral Turner's
involvement in what he describes as restructuring and redrafting"
prior to NFIB meetings.70
Turner says: "'I am not bashful about
that. I end up telling them this section has to be redrafted. You
won't find many sentences I personally penned. Mostly it is be-
cause they didn't bring out two views strongly. Another way, I look
at the outline, the concept at the beginning, and I restructure
that, saying, You are asking the question wrong.',71
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 95 -
One intelligence official describes DCI Turner's adjudicatory
role as highly "personalized."72
After reviewing successive drafts,
Turner may make one of four decisions about what an NIE says.
First, he may "support" the CIA's view by rejecting an alternative
proposed by a Pentagon agency, eventually eliminating it from the
draft. As documented earlier, this happened in a recent NIE 11-4.
Second, Turner may "overrule" the CIA by agreeing that a Pentagon
alternative is of sufficient merit to warrant equal status to
that of the CIA's, for example, in the case of the Soviet Backfire
bomber already described.
Third, Admiral Turner may reject all options and draw up a
new one. For example, the above-mentioned New York Times article
reported that this happened in a 1977 NIE on the balance of strength
between North and South Korea following withdrawal of US troops in
the South. "Admiral Turner concluded, contrary to the original
estimate, that withdrawal would substantially diminish the deterrent
balance on the peninsula."73
Fourth, Turner may insert his personal
view separately, or alongside of other, within the NIEs.74
Admiral Turner has been criticized for not using the joint,
high-level forum of NFIB for final deliberations. Because he prefers
dealing with Pentagon officials individually throughout the draft
stage, his role is less visible than that of DCIs McCone or Colby
and critics advance charges of political motivation. As one in-
telligence community official complains: "Turner has 'asked the
community to redo the estimates or has rewritten them and sent
them on without further reference to the National Foreign Intelligence
Board, or he has sent them back to convince, cajole or bully the
other participants into alternative estimates."75
, Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 96 -
VI. THE DCI's ROLE IN PROMOTING
INDEPENDENT REVIEW
Independent review of the final strategic NIEs is discussed
here for three reasons. First, outside review of the NIEs originally
was considered a wholesome ingredient of the process. Second,
some DCIs actively promoted such review. Third, the independent,
parallel analysis of the outside "B Team" to the inside "A Team's"
NIE 11-3-8-76 reportedly had a sizeable impact on that and sub-
sequent strategic NIEs, as described below.
By "independent" review, this author means review by any
group which does not play an official part in drafting the NIEs
and is not employed directly by the CIA. Review of the strategic
NIEs may occur at the draft stage, or just before or after USIB
endorsement. It may be systematic, e.g., each year's NIE 11-3-8,
or periodic. As indicated in Chart B, only two DCIs--Allen Dulles
and James Schlesinger--are regarded by this author as having en-
couraged independent review of the strategic NIEs. William Colby
is a borderline example since, with reservations, he agreed to
the concept of competitive review before leaving the office. How-
ever, it was DCI George Bush who guided the 1976 A-B Team experiment,
with initial enthusiasm, to conclusion. In contrast, DCIS John
McCone, Richard Helms, and Stansfield Turner demonstrated little
interest in establishing ongoing mechanisms for independent review.
A. Allen Dulles, the Princeton Consultants and Strategic Advisory
Panel
Almost from day one of the NIEs under General Bedell Smith, a
group of distinguished outsiders known as the Princeton consultants
met near Princeton University to examine the estimates. Originally,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part-Sanitized Copy Approved forRelease2012/05/04 CIA-RDP93T01132R000100020034-5
- 97 -
the panel was self-supporting and composed largely of men with
wide, practical experience in public affairs "as a corrective
to what [the DDCI] regarded as the disproportionate number of
academics on the Board (of Estimates)." The consultants first
met at the home of the DDCI, William Jackson, and then other
locations around Princeton. Throughout the 1950s, they gathered
regularly for two full days, four times a year. An additional half
day prior to formal meetings was spent at Princeton reading the
NIEs. Typically, the consultants reviewed NIEs just prior to IAC
or USIB meetings, although earlier drafts sometimes were examined.
The primary way in which Allen Dulles promoted the involvement
of the Princeton consultants was by his presence. As DDCI in 1952,
Dulles attended every meeting of the panel and was instrumental
in proposing new members. Throughout his tenure as DCI, Dulles
continued this practice, often leaving busy meetings at the CIA
to fly up to Princeton.2 Dulles was accompanied by several Board
of Estimates members and ONE staff. The discussion was lively and
intense, and Dulles is described by one panel member as being "keenly
interested and always eager to learn."3 Face-to-face contact be-
tween CIA officials and outside experts was considered natural
and invaluable during the 1950s and, since the Princeton consul-
tants issued no formal reports, constituted its chief means of
communicating with the CIA.
Although the Princeton consultants spent the majority of
time on non-strategic NIEs, they contributed to the strategic
NIEs as well. Panel members were useful in devising precise lan-
guage by which to communicate the Soviet threat. Individuals
occasionally took it upon themselves to examine Pentagon numerical
npciaRsified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 98 -
predictions in depth. For example, one member said that he looked
into figures on Soviet coastal artillery which had appeared con-
sistently in each NIE, and discovered that they were the same ones
utilized during WWII.4 Nonetheless, by the second half of the
decade, questions about Soviet technology became so complex that
other mechanisms were devised to review these NIEs.
In the late 1950s, Allen Dulles established the Strategic
Advisory Panel "as a device for resolving some of the competing
claims on Soviet missile activity."*5 Composed of ten members, the
panel included CIA consultants and recognized aerospace technology
experts, and was regarded as independent of the CIA. It met twice
a year for several days, after having been briefed on the strategic
NIEs by experts involved in drafting, usually just before the
NIEs were issued. The panel then formed its own judgements, which
it presented separately to the DCI and the President's security
advisers. During "the 'missile gap' days the panel's expertise had
been found useful in sorting out a lot of technical arguments and
removing much of the excess speculation."6
B. The 1960s
Neither DCIs John McCone nor Richard Helms actively promoted
independent review of the strategic NIEs. During the early 1960s,
the Princeton consultants and the Strategic Advisory Panel fell
into disuse. In both cases, insiders contended that their own
growing expertise in the technical aspects of strategic estimating
*The Strategic Advisory Panel originally was known as the Hyland
Panel for its first chairman, a Vice President of Hughes Aircraft.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 99 -
made the contribution of outsiders less insightful. In the be-
ginning of McCone's tenure, the Princeton consultants met twice
a year at a retreat in Virginia, but the DCI was not present and
work centered primarily on the non-strategic NIEs.7 Eventually, the
group dissolved. CIA analysts came to believe that the Strategic
Advisory Panel had become too "hawkish" and its intervention "ir-
ritating."8 It was disbanded in the early 1960s, and its re-
constitution in 1967 was ignored by Helms and intelligence pro-
fessionals.
During most of the 1960s, review of the strategic NIEs was
conducted entirely by those persons who wrote them. For example,
from time to time individual Board of Estimates members initiated
post-mortems on critical NIEs or groups of NIEs, for internal
circulation.9 Growth took place in the influence and competency
of USIB's many subcommittees, which some observers consider the
real, working-level heart of the USIB process.10
Of the fifteen
subcommittees eventually created, at least half concerned themselves
exclusively with strategic matters, such as the Joint Atomic Energy
Intelligence Committee and the Committee on Imagery Requirements
and Exploitation. Composed of CIA and Pentagon specialists, they
contributed substantively to drafting of the strategic NIEs. One
Pentagon official complained that both DCIs McCone and Helms "stacked
the deck" in appointing CIA officials to chair these subcommittees)
During the Nixon Administration, National Security Advisor
Henry Kissinger utilized the NSC to review the strategic NIEs. After
registering dissatisfaction with the conclusions reached in NIE 11-8-68
predicting delays in the Soviet MIRVing program, Kissinger created
a "MIRV Panel" under the NSC to study the question in depth. Conducted
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 100 -
in the spring of 1969, this interagency review included CIA and
Pentagon DDR&E officials, who soon became the chief combatants. It
involved "a stream of intensely focussed studies" on Soviet MIRV
capabilities, "reached no conclusions, but exhaustively laid out
data and identified areas of disagreements."12
Subsequently, Kissinger
created an NSC "Verification Panel" to study issues related to SALT
verification, which immediately expanded beyond a focus on the
strategic NIEs. When Kissinger became Secretary of State in
1973, the NSC's review of specific NIEs ceased.
C. James Schlesinger and the Intelligence Community Staff
DCI Schlesinger's contribution to an independent review in-
volved the Intelligence Community (IC) Staff, which now serves
the DCI. Since the IC Staff is funded out of a budget separate from
the CIA's, it qualifies in this author's definition of "independent"
even though personnel are drawn from the CIA and Pentagon. Since
1971, the IC Staff assisted the DCI in administrative housekeeping
tasks related to his role as coordinator of the intelligence community.
In late 1972, a Product Review Division (PRD) was created. The
PRD was supposed to regularly appraise intelligence studies and es-
timates, "testing them for objectivity, balance, and responsiveness."13
Schlesinger was the first DCI to provide the IC Staff with
muscle, by emphasizing its responsibility for community-wide
budgeting, authorized by President Nixon in 1971. In encouraging
the PRD's independent work, "Schlesinger altered the composition
of the IC Staff by increasing the number of non-Agency personnel.
In this way, he hoped to facilitate the Staff's contacts with the
other components of the community.?14
The PRD commenced in-depth,
case studies of past NIEs, choosing several recent "crisis prediction?
I Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 101 -
NIEs for special emphasis--e.g., NIEs on the 1973 Middle East
war, the Cyprus crisis of 1974, the Indian nuclear detonation, and
the Mayaguez incident. Its .conclusions had implications for
the intensifying strategic debate. For example, one study con-
tended that, in 1973, the "intelligence community was disposed to
believe that the Arabs were unlikely to resort to war against
Israel because to do so would be 'irrational,' in light of relative
Arab-Israeli military capability."15 In evaluating the 1974
Cyprus crisis, the PRD argued that CIA analysts again were prey to
"the perhaps subsconscious conviction (and hope) that, ultimately,
reason and rationality will prevail, that apparently irrational
moves (the Arab attack, the Greek-sponsored coup) will not be made
by essentially rational men."16
Precisely this complaint was
echoed by individuals charging CIA analysts with a "bias of systematic
optimism. ,,17
D. The President's Foreign Intelligence Advisory Board and
the 1976 A-B Team Experiment
Among its several functions, PFIAB historically has demon-
strated a strong interest in the strategic NIEs. Chief among its
assets as an outside, independent reviewer has been the prestige
and experience of its members, and its inclination to look at issues
from what has been termed the "grass roots level."18 Up to the
time it was abolished in 1977, PFIAB frequently called intelligence
professionals, including DDP personnel, junior analysts and dissident
officials, to testify before it. PFIAB also solicited raw intelligence
reports and drafts of finished products to evaluate. Although PFIAB
had no staff of its own, it assembled special panels of members
and consultants to examine selected issues, reporting directly to
the President.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
(
1
L
I
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 102 -
The 1976 A-B Team "experiment in competitive analysis stemmed
from PFIAB's opinion that the NIEs had been underestimating the
process of Soviet strategic weapons."19 Individual PFIAB members
gradually had arrived at this conclusion after examining strategic
NIEs of the past decade and listening to the testimony of dissenting
Pentagon officials, such as Generals Daniel Graham and George Keegan.
In early 1975, several PFIAB members complained to DCI Colby about
the suggestion, in NIE 11-3-8-74, that the Soviets were seeking
only "rough parity" with the US. They presented Colby with their
own draft statement of Soviet policy and asked him to initiate an
outside review.20 Colby listened, but defended the NIE on the
grounds that it represented the best combined judgement of the
intelligence community, and appeared reluctant to go further.21
At the same time, President Ford requested PFIAB to ascertain why
the strategic NIEs presented such a disturbing number of dissents.
In mid-1975, General Keegan was invited to the White House to
make his case, before President Ford and PFIAB, concerning the
significance of Soviet civil defense measures and new guided missiles.22
Other Pentagon officials also testified at this meeting. In August
1975, PFIAB's Chairman proposed that the President authorize the
NSC to implement a "competitive analysis."23
DCI Colby responded
with a proposal that the PFIAB first examine an "applicable NIE then
underway and thereafter determine what specific course of action
to take. The PFIAB found weaknesses in that NIE and, after having
made further investigations of its own, again proposed [in April
1976] an experiment in 'competitive analysis.',24 In June 1976,
President Ford asked DCI George Bush to carry out the experiment.
Bush is described as being initially "enthusiastic" and considered
an effective review one of his main tasks as DCI.25
I , Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 103 -
Two teams were set up to prepare separate versions of NIE
11-3-8-76. The A Team was composed of the regular CIA analysts,
and additional consultants. The B Team consisted of seven outside
experts and pest government officials, headed by Dr. Richard Pipes
of Harvard University. It was chosen by the DCI and the President's
deputy assistant for national security. A CIA analyst was desig-
nated by Bush to provide the B Team with the same raw material
utilized by the CIA in preparing the NIEs.
According to the New York Times, both teams were assigned
thrbe topic* for intense study; thil accuracy of Soviet guided
miNNIIes; qv. penetrability of Soviet air defense by low-level
bombers; and overall Soviet strategic objectives.26
On security
grounds, the Navy eliminated a fourth topic from consideration.
As related by members of both teams, this story contended that
there was a "stand-off" on Soviet missile accuracy, and "both teams
influenced each other" regarding Soviet air defense.27 Greatest
dispute arose over the assessment of Soviet strategic aims, with the
B Team challenging the CIA's view that the Soviets were not seeking
strategic superiority over the US. The discussions were described
by one participant as being "absolutely bloody" with the B Team
accusing the CIA of dealing in "faulty assumptions, faulty use of
intelligence and faulty exploitation of available intelligence."28
The final version of NIE 11-3-8-76 presented to President-elect
Carter reportedly was influenced by the pessimism of the B Team.
According to the same New York Times story, DCI George Bush spoke
of "changed perceptions" within the CIA regarding Soviet behavior
as a result of the parallel analysis, and said the "worrisome
signs included newly-developed guided missiles, a vast program of
in Part - Sanitized CODV Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 104 -
underground shelters, and a continuing build-up of air defenses."29
CIA participants referred to the NIE as being "'more somber."30
A high-level Pentagon official described the estimate as being
"more than somber--it was very grim. It flatly states the judge-
ment that the Soviet Union is seeking superiority over United States
forces. The flat judgement that that is the aim of the Soviet Union
is a majority view in the estimate. The questions begin on when
they will achieve it.'"31 The Senate Select Committee on Intelligence
immediately investigated the experiment, after which one member,
Senator Daniel Moynihan, wrote that the B Team's notion "that the
Soviets intend to surpass the United States in strategic arms and
are in the process of doing so, has gone from heresy to respect-
ability, if not orthodoxy.,32
In early 1977, PFIAB proposed that the entire estimative
procedure be revised and that outsiders, like themselves, be brought
into the process once again.33 As the Senate Select Committee
on Intelligence concluded, the concept of the review of NIEs by
outside experts was a "legitimate one." However, its value in
this case had been undercut by three factors: first, the B Team
"reflected the views of only one segment of the spectrum of opinion;"
second, it "spent much of its effort on criticizing much earlier
NIEs rather than...producing alternative estimates;" third, "de-
tails concerning these highly classified questions leaked to the
press, where these appeared in garbled and one-sided form."34 DCI
George Bush, who favored the experiment at the outset, concluded
that "never again" should an outside review be attempted in the
same manner.35
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 105 -
By the end of 1977, President Carter had abolished PFIAB and,
under DCI Stansfield Turner, independent review of the NIEs has
not been sustained. Rather, Admiral Turner has encouraged increased
input by consultants to the drafting of the strategic NIEs. For
example, one consultant to the NIO for Strategic Intelligence is
known to have views described as hard-line.36 Admiral Turner also
became the first DCI to release some NIEs, primarily those delaing
with energy issues, to the public domain. Admiral Turner's three-
man Senior Review Panel now assists him in reviewing the NIEs.
However, this group is employed full-time by the CIA and is not
meant to be "independent" by this author's definition.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 106 -
VII. ASSESSMENT OF THE PERFORMANCE OF DCIs
If this study has demonstrated nothing else, it is that DCIs
have been inconsistent in their approach to producing the strate-
gic NIEs. Some have emphasized some roles, and others neglected them.
Then, what kind of DCI should a President choose if he desires
changes in the strategic intelligence estimative process? Which
functions are "ideal?" Can a DCI appropriately be both "manager
and prophet?" What tasks have not been performed?
Below, the performance of seven DCIs in producing the
strategic NIEs is assessed. In addition to the six DCIs selected
by this author for extensive study, General Bedell Smith is in-
cluded because of his strong role in the estimative process, even
though the modern strategic NIE did not emerge until the mid-1950s.*
The author summarizes the roles exercised by DCIs according to two
criteria: their individual background, interest, and concept of
the DCI's job; and historical trends. Then, separate functions are
examined to determine which are mutually supporting and which
are potentially conflicting, or biasing. Finally, the author
presents her conclusions about roles underemphasized by DCIs.
A. Summary of Roles Played by DCIs
Seven isolated roles of the DCI in producing the strategic
estimates have been described in this study. Three pertain to the
NIE drafting process: (1) management direction to CIA production;
(2) management attention to Pentagon coordination; and (3) personal,
*The role of General W. Bedell Smith (1950-1952) is discussed in
Chapter II. The other DCIs include: Allen W. Dulles (1953-1961);
John A. McCone (1961-1965); Richard Helms (1966-1972); James
R. Schlesinger (1973); William E. Colby (1973-1975); and Admiral
Stansfield Turner (1977 to present).
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 107 -
substantive input. Three involve leadership at the United States
Intelligence Board
(USIB), now called the National Foreign
Intelligence Board (NFIB): (4) support of the CIA's position; (5)
encouragement Of competitive debate; and (6) final adjudication of
views. The last role is: (7) promotion of independent review of
the stratetgic NIEs.
Personal Background and Concepts
Looking at these roles as a function of a DCI's personal back-
ground and interests, the following observations are made.*
(1) Management direction to the CIA estimates team was exerted
by five DCIs with diverse backgrounds--General Bedell Smith, John
McCone, James Schlesinger, William Colby and Admiral Stansfield Turner.
After General Smith created tha Board of Estimates and ONE within
CIA in late 1950, no charges were initiated in this set-up for 23
years and two DCIs (Dulles and Helms) with a combined tenure of 15
years devoted little attention to management. The first director
to consider management reform and leadership the DCI's primary role
was William Colby in 1973, an "inside" director who was familiar
with a range of CIA programs. Since 1973, the National Intelligence
Officer (NI0s) who draft the NIEs are personally selected by the DCI.
*The "military" directors include four-star General Bedell Smith
and four-star Admiral Stansfield Turner. The "inside" directors
are William Colby and Richard Helms, both of whom had risen up
through the ranks of the clandestine, or DDP, side of the
CIA. In addition, Colby had served briefly as Executive Director-
Comptroller of CIA and as DDP. "Outside" DCIs include John McCone
and James Schlesinger, both of whom had previous government careers
including Chairman of the AEC, and the first was a successful private
businessman and the second had been a strategic scholar at the RAND'
Corporation. For lack of a better term, Allen Dulles is called an
"inside-outside" DCI, since he was an OSS verteran and served briefly
as DDP and as DDCI, but had an independent professional and political
base. Not examined here is Admiral William Reborn (mid-1965 to mid-1966),
a "military" DCI, and George Bush (1976), an "outside" DCI.
nariaccifipn in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 108 -
(2) Management attention to coordinating the draft strategic
NIEs with Pentagon intelligence agencies, such as DIA and Army, Air
Force and Naval Intelligence, was a focus of only three DCIs--General
Smith, William Colby and Admiral Turner. Two are "military" directors
with previous, high-ranking careers, who both designed a new for-
mat for Pentagon interaction and raised estimative issues directly
with individual Pentagon officials. In comparison, William Colby
created a new mechanism for increased Pentagon input but, while
responsive, did not take initiative in seeking Pentagon comments at
the draft stage. To the author, this suggests that, for good or ill,
"military" DCIs are more aggressive in eliciting direct Pentagon
interaction.
In contrast, Dulles considered it inappropriate for the DCI
to dictate coordinating procedures to the Pentagon, and Helms was
passive except when issues were raised by the White House or Defense
Secretary. Throughout McCone's tenure, Defense Secretary McNamara
supported ongoing CIA mechanisms for drafting the strategic NIEs.
(3) Four DCIs made personal, substantive input to the draft
strategic NIEs, by assisting the CIA team in designing and answering
strategic questions and/or presenting their personal opinions in
the draft. All four--General Smith, John McCone, James Schlesinger
and Admiral Turner--had previous professional training, technical
familiarity and jobs in complex strategic issues. In contrast, Allen
Dulles and two "inside" directors (Helms and Colby) had most ex-
perience in the DPP or clandestine side of the CIA and, admittedly
in Colby's case, little technical competency in strategic matters.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 109 -
Instead, these three made substantive input to the non-strategic NIEs.
This suggests that, if he prefers a DCI who can make substantive
contribution to the draft strategic NIEs, the President choose
a person with the technical background to do so.*
(4) At USIB, now NFIB, the leadership concepts of DCIs varied
greatly. Only one DCI, Allen Dulles, considered his primary role
to support the CIA's view in the strategic NIEs. This resulted from
Dulles' decision in the mid-1950s to permit the CIA to prepare an
independent, institutional position within the draft strategic es-
timates, which then became the main text of the NIE. His decision
was made within an historical context of encouraging competition to
the Pentagon's intelligence estimates. At USIB, Dulles did not
aggressively advocate the CIA's views, but rather presided over
final deliberations in a manner which reinforced earlier decisions
made in the draft.
(5) All other DCIs have assumed other roles at USIB. Four DCIs
with diverse backgrounds--General Smith, John McCone, James
Schlesinger and William Colby--actively encouraged competitive de-
bate at USIB, or the Intelligence Advisory Committee (IAC) in Smith's
case. Under these four, debate was lively and intense, and USIB was
considered an appropriate, higher level forum for the reopening of
old dissents or presentation of new ideas. In the case of the
"military" director (General Smith), debate was viewed as a means
of promoting Pentagon involvement in the NIE process. The two
"outside" DCIs (McCone and Schlesinger) were interested in hearing
a multiplicity of opinion. However, competitive debate at USIB was
*There has never been an "inside" DCI selected from the CIA's
DDS&T or DDI directorates.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 110 -
encouraged, most consistently and at length on strategic issues,
by William Colby, who argued that his primary role at USIB was to
manage competition. This suggests that an "inside" DCI can rise
abovi internal CIA demands and politics to assume a higher, coordina-
tive role at USIB.
In contrast, two very different DCIs--Richard Helms and Admiral
Turner--did not promote competitive debate at USIB, or NFIB in Turner's
case. Neither enjoyed lengthy discussion or found it useful. They
believed that issues of dissent should be ironed out beforehand.
Consequently, final deliberations were short. However, the similiarity
ends here. The "inside" DCI (Helms) did not personally intervene
at most levels of the draft stage. The "military" director (Admiral
Turner) is heavily involved in drafting, and encourages competition
by attempting to increase coordination with the Pentagon through-
out the drafting process.
(6) Four DCIs of diverse backgrounds--General Smith, John McCone,
Richard Helms, and Admiral Turner--acted as final adjudicators of
conflicting views and a fifth--James Schlesinger--probably would have
if he had remained longer in office. According to this author's
definition, adjudication implies a final choice between competing
positions which results in a new position gaining dominance, or
some views being eliminated. Primarily, this occurs when a DCI
represents his own opinion as the final text, overrules the CIA's
view in favor of a Pentagon view, or eliminates a Pentagon dissent
from the NIE. This author found instances, albeit rare, of each.
With the exception of Helms, the other four DCIs (if Schlesinger
is included) appear to have considered adjudication a natural means
to resolve conflict. All four previously had devoted management
attention to drafting procedures and made substantive input to
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
draft NIEs. Three, with the exception of Turner, made adjudicatory
decisions after hearing extensive debate at USIB. In contrast,
Turner's adjudicatory role is exercised primarily at the draft
stage just prior to NFIB meetings.
This suggests a natural flow, or escalating pattern, to the
DCI's involvement in producing the strategic NIES. If a DCI makes
management decisions regarding drafting procedures, and has the
technical background to make substantive input to the draft, he
also may encourage competition and end up making final, intelligence
judgements in the NIEs. That is, some resolution of conflict is
sought.
That DCI Colby did not make adjudicatory decisions on strategic
matters at USIB, but rather resolved conflict through compromise
by adding footnoted dissents to the NIE, bears out this hypothesis.
Colby, an "inside" DCI without strategic background, was mainly a
manager and not substantively involved in drafting. Richard Helms
is an exception. Helms occasionally made adjudicatory decisions at
USIB, partly in hopes of keeping the strategic NIEs on a timely
schedule, without previously having been involved in management
or substance. Moreover, Helms' single role is difficult to ex-
plain, since he typically is described by observers as being a strong,
thoroughly professional leader of the CIA bureaucracy. This suggests
that for strategic matters, Helms tried to adhere to a philosophy
of allowing the professional analysts to argue out their differences.
Finally, this author found no evidence that either "inside"
director (Helms or Colby) or the "inside-outside" director (Dulles)
ever overruled the CIA, whereas there is evidence that one "outside"
DCI (McCone) and one "military" DCI (Admiral Turner) did so. This
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
(
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 112 -
suggests that, bureaucratically, it is easier for a non-insider to
make adjudicatory decisions against the CIA, at least on strategic
issues.*
- (7) The last role, promoting independent review of the strategic
NIEs, was performed by only three DCIs examined--General Smith, Allen
Dulles and James Schlesinger. Essentially, there was no independent
review of the strategic NIEs by outside experts not employed by
the CIA from 1961-1973, and 1977 to the present. If one includes
an "outside" director not studied here, DCI George Bush, who guided
the 1976 A-B Team experiment in competitive analysis to conclusion,
it appears that "non-inside" directors are most enthusiastic about
the notion of independent review.
These observations are summarized in Chart D on the following
page. In addition, Chart D outlines the functions of DCIs in pro-
ducing strategic NIEs according to concepts of the three broad jobs
authorized for the DCI: manager of the CIA; coordinator of intelli-
gence community affairs; and chief intelligence advisor to the
President. This chart is based on an assumption that separate NIE
production roles, e.g., encouraging competitive debate, can be
readily classified as demonstrating emphasis on one of three over-
all assignments of the DCI, e.g., coordinator of the intelligence
community. As Chart D indicates, a majority of DCIs emphasized at
least two broad roles of the DCI in producing specific NIEs. Four
DCIs--General Bedell Smith, John McCone, James Schlesinger and
Admiral Stansfield Turner--played all three roles.
*As documented in this study, Dulles, Helms and Colby were more
aggressive in criticizing CIA work in the non-strategic NIEs.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 "
/ Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
OVERALL ROLE. AS DCI
A. Manager of CIA
- 113 -
CHART D
BROAD ROLES EMPHASIZED BY SEVEN DCIs FOR STRATEGIC NIES
SEPARATE PRODUCTION ROLES*
1. Management Direction to CIA
Drafting
4. Support of CIA Position
at USIB
7 DCIs
Smith, McCone,
Schlesinger, Colby
& Turner (5)
Dulles (1)
BACKGROUND**
Diverse
(2 military, 2 out-
side, 1 inside)
One DCI
("inside-outside")
B. Coordinator of In-
telligence Community
2. Management Attention to DOD
Coordination in Drafting
5. Encouragement of Competi-
tive Debate at USIB
Smith, Colby
& Turner (3)
Smith, McCone,
Schlesinger, &
Colby (4)
Mainly Military
(1 military, & 1
inside)
Diverse
1 military, 2 outside
& 1 inside)
C. Chief Intelligence
Advisor to President
3. Personal, Substantive
Input to Draft
6. Final Adjudication of views
(prior to or at USIB or
NFIB)
7. Promotion of Independent
Review
*The seven production roles are
this study.
/*An "inside" DCI had a previous
or government career; a "milit
outside" DCI.
Smith, McCone
Schlesinger,&
Turner (4)
Smith, McCone,
Helms, probably
Schlesinger &
Turner (4-5)
All with technical
background, no inside
(2 military, & 2 out-
side)
Smith, Dulles,
and Schlesinger
(3) (if include
Bush as 8th DCI:
(4)
Same as above
(plus 1 inside)
Diverse
(if include Bush,
mainly outside: 2 out-
side, 1 "inside-out-
side," &1 military
numbered according to the order in which they were discussed earlier in
CIA career; and "outside" DCI had a previous private business, academic,
ary" DCI is a military officer. Allen Dulles is considered an "inside-
noriaccifipn in Part - Sanitized Copy Approved for Release 2012/05/04 CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 114 -
Historical Trends
The activities of DCIs also may be assessed in light of his-
torical hindsight. As indicated in Chart B earlier, the last three
DCIs examinedin this study--James Schlesinger, William Colby and
Stansfield Turner--have been involved more actively in the strategic
NIEs than their predecessors. For example, compared to the long
tenures of Allen Dulles and Richard Helms, recent DCIs exercised more
control over different phases of production. Compared to John McCone,
there has been more theorizing and follow-through of late, concerning
the appropriate leadership role of the DCI. This is indicated by
the number of "XX" in Chart B, beginning in 1973.
To this author, the recent trend towards increased DCI inter-
vention in the production process is one response to outside criticism
of the quality and validity of the strategic NIEs. In 1969, National
Security Advisor Henry Kissinger and Defense Secretary Laird were
not the first top-level officials to criticize the strategic NIEs.
President Eisenhower himself was known to doubt the accuracy of
"gap" predictions in the 1950s. As documented in this study, what
is different about the criticism of the first half of the 1970s is
that it gathered momentum, supported by dissident Pentagon officials anc
outside groups such as the Presidents Foreign Intelligence Advisory
Board (PFIAB), and was directed mainly against CIA estimators. It
culminated in the A-B Team competitive review in 1976. Richard Helms
did not respond by exercising more leadership over different phases
of production, although he frequently directed the CIA estimates
team to incorporate dissenting views in the strategic NIEs. For
' Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 "
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 115 -
the most part, Helms' tenure as DCI was characterized by passivity
towards the strategic estimative process. When he did intervene at
the last moment, his decisions sometimes were criticized by individual
Pentagon officials as "arbitrary."
Beginning with James Schlesinger in 1973, recent DCIs felt the
need to play a stronger role during the drafting stage. The one
trait they share is increased management direction over the CIA es-
timates team. Indeed, they overhauled it completely. In choosing
the NIO system for drafting NIEs, they then assumed more direct
responsibility for it. In the absence of the collective review
and supervisory functions formerly exercised by the Board of Estimates,
DCIs now handpick the NI0s, and personally monitor and review their
work. In addition, two DCIs examined since 1973--William Colby and
Admiral Turner--have devoted more attention to means they believe
will improve coordination with the Pentagon intelligence agencies
during the drafting stage. They became the first DCIs to do so
since General Bedell Smith in 1950.
Here the similarities end. In 1973, William Colby became "the
man for his time" by extending competitive debate to the open forum
of USIB, considering USIB as a court of last resort and sounding
board for dissenting members. In 1977, Admiral Turner reversed this
trend. Turner's actions have led to outside criticism that he has
"personalized" the decisionmaking process, making adjudicatory decisions
in a less open, or more private, atmosphere.
B. Mutually Supporting and Conflicting Roles: Manager versus Prophet
A useful way to approach the question of the "ideal" role or
roles of the DCI in producing strategic intelligence estimates is to
ask which functions are mutually supporting and which potentially
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 116 -
are conflicting, or biasing. Assuming that a President prefers
a DCI who can help resolve conflict in the strategic NIEs, which
roles best ensure that a DCI's intervention will not lead to es-
timates biased by his personal view or one institutional view?
Should a DCI be both "manager and prophet?"
In this author's opinion, it is impossible to eliminate potential
bias altogether if a DCI is expressly selected on the basis of playing
a strong substantive and adjudicatory, or "prophet" role. However,
some functions are potentially more biasing than others, some are
conceivably mutually supportive, and in both cases it is partly a
matter of how decisions appear to others. Mutually supporting roles
are those in which the performance of one enhances performance at
the next stage, or those in which a balance is struck, for example,
between drafting functions and leadership at USIB.
Most questions of personal or institutional bias on the DCI's
part are more complex than simply advocating the CIA's institutional
viewpoint. Allen Dulles played this role at USIB; but this auto-
matically precluded him from exercising other roles, and all other
DCIs avoided this pitfall. Usually, the subject of bias is raised
by critics when and if a DCI personally selects the CIA drafters,
argues his own views before them, individually reviews and approves
their work, and inserts his personal opinions in the draft or makes
other adjudicatory decisions. Stated simplistically, how can a
DCI both advocate one view in the draft NIE and also act impartially
as an adjudicator at USIB? Because of potential conflicts along
the way, should a DCI concentrate on designing an "optimal" manage-
ment structure for writing, coordinating and reviewing the draft
NIEs and then adopt a "hands-off" policy until final USIB deliberations:
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 117 -
Two extremes of administrative behavior have been described
in this study--that of DCIs Richard Helms and Admiral Turner.
On the one hand, Helms contends that the DCI should be little in-
volved in drafting lest this "bias" his final decisions at USIB. If
the DCI becomes an advocate too early, this will limit his freedom
later. Instead, the CIA should be left alone to do its work, and
the DCI intervene only in the final phase as deemed necessary. The
difficulty with Helms' approach is that his adjudicatory role at
USIB appeared to some ill-informed and accomplished in a substantive
vacumn, and often eliminated Pentagon dissents. When Helms did
not "adjudicate," he sent the draft NIE back to the CIA drafters
to represent more strongly the Pentagon's alternatives. Neither
action necessarily resulted in an unbiased resolution of conflict.
Critics also complained that, by inaction, Helms allowed the member-
ship quality of the Board of Estimates to deteriorate.
DCI Helms' approach is almost a case of approach-avoidance.
This suggests that a DCI's involvement in drafting is supportive of,
or consistent with, execution of subsequent roles at NFIB. Drafting
the strategic NIEs takes at least six months, while formal USIB or
NFIB debate is a matter of hours, or at most days, and under Helms
was less. Once a draft is written, it becomes difficult to change.
In order to be sufficiently informed about competitive strategic
debate at NFIB, a DCI cannot ignore what goes on during the lengthy
drafting sessions.
For example, at a minimum, a DCI might review the CIA's organi-
zational set-up for estimative work to ascertain if it is operating
as intended, appoint an estimates team with which he can work
comfortably, monitor and review successive drafts so that he is
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 118 -
informed about the boundaries of major analytical disputes, and
listen to the arguments of dissenting Pentagon representatives
about coordinative procedures. This "management" approach to draftin;
is exemplified by DCI William Colby, who then went on to manage
highly technical, argumentative debate at USIB.
On the other hand, Admiral Turner has taken all these steps,
and more, and he is criticized for bias in being both personal ad-
vocate and final adjudicator. In this author's opinion, these
accusations result, partly, from poor timing. Admiral Turner pays
too much attention to the drafting of the strategic NIEs compared
to leadership at NFIB meetings. Because of an imbalance between his
involvement at these two stages, DCI Turner not only opens himself
to criticism but also deprives himself of the opportunity of hearing
high-level, group debate. That Turner does not fully utilize NFIB
is an admission that, while brief, oral discussion at NFIB meetings
between diverse intelligence chiefs with their own axes to grind
may be a cumbersome and time-consuming way to complete complex
analytical work.
Admiral Turner's approach implies that a DCI's decisions might
appear less biased, or be less biased, if he reserves his adjudicatory
role until after open, competitive debate at NFIB, conducted jointly
in the presence of all intelligence agencies. A DCI still might exert
management direction over the drafting process and make personal, sub-
stantive input beforehand.
For example, a DCI might read and edit successive drafts, re-
shape the questions asked, offer his personal advice, advance his
own opinions alongside others, discuss strategic issues with CIA
and Pentagon estimators, and satisfy himself that all views are
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 119 -
incorporated. However, he could stop short of rewriting the NIEs
to reflect his final judgements. This approach was executed
successfully by DCI John McCone who, after initial hesitation, made
substantive input to the draft but was willing to change the NIEs
as a result of USIB debate. In essence, McCone was both manager
and prophet, and was not criticized at the time for exercising his
prerogative to play these roles.
C. Management Steps Not Taken by DCIs
In this author's view, systematic attention to management and
coordination at the draft stage of the strategic NIEs is the sine
qua non of improved intelligence estimates. Throughout this study,
we have been interested primarily in whether a DCI devoted any
attention to these issues, as opposed to evaluating reorganizations
and other changes actually intitiated. Below, the author presents
her conclusions about the strategic NIE process, derived from a
year of study. These recommendations focus on mechanisms a DCI
might devise to elicit a multiplicity of views before final intelli-
gence judgements are made. Three criteria govern their selection:
they must be easy to administer; they should reduce the likelihood
of institutional or personal bias; and they must not pose, in the
DCI's view, a security risk. All proposals build on changes authorized
in recent years, rather than sharply reversing them.*
(1) This author applauds the notion that the DCI appoint the
CIA estimators, or NI0s. Since the DCI must work closely with them,
he should select them and they, in turn, be responsible to him.
*Since reorganization of the estimates function is ongoing in CIA,
some of these proposals soon may be moot.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 120 -
The NIO system now has been in effect for seven years, so that
the earlier criticism of inexperience can be laid aside. However,
recent DCIs have been negligent in designing additional internal
review mechanisms. DCI Colby abolished the Board of Estimates be-
cause he believed its collective approval responsibility isolated
the DCI from "the true believer." Moreover, a majority of members
had been on the Board for many years, and thus the question of
stagnation and perpetuating bias was raised by outsiders. Almost
immediately, however, the cry went up for some collective review
procedure in the CIA.
This author contends that the DCI and NIOs need an ongoing
mechanism for internal review of the NIEs, since the present system
relies so heavily on the work of a few individuals. Admiral Turner's
three-man Senior Review Panel created in 1979 partly answers this
demand, and fulfills several preferred qualifications. First,
members are selected from outside the CIA, but work in the CIA on a
daily basis. Second, membership is broadly based, or diverse. Third,
the Panel advises but does not consent or "approve" in the same way
that the Board of Estimates did. Fourth, it serves both the NIOs
and the DCI. Finally, this author believes that member's terms of
office should be fixed, in order to infuse the Panel with new personnel
on a systematic basis.
However, observers question whether the newly created Senior
Review Panel has sufficient prestige and responsibility to effectively
challenge the draft. Consideration might be given to expanding its
membership by two or three, including strategic experts. Some in-
dividuals might be hired on a part-time basis, so that they could
nna-Inecifiarl in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 121 -
continue their current outside careers. The Panel might also
conduct collective meetings among the NI0s, who now rarely meet to-
gether. In this manner, both Panel members and NIOs could benefit by
sharing experiences and problems common to estimative work. The
Panel might also adopt internally the principles of the 1976 A-B Team
experiment in competitive analysis, as described below.
(2) There has been too rigid adherence to the principle that
the CIA present a unified opinion in the draft and final NIEs.
Individual CIA estimators never were permitted to take footnotes, and
even now CIA estimators usually pull together in supporting a major
CIA position. The rationale for past practices is inherent in the
nature of estimating, and bureaucracy. However, the DCI might
authorize periodic competitive or parallel reviews within the CIA,
under the auspices of the Senior Review Panel. For example, the
Panel might choose several CIA analysts and consultants to assist
it in reviewing selected aspects of an ongoing strategic NIE.
Such experiments would provide the Panel with additional clout and
experience, without necessarily expanded approval authority. Much
like the Pentagon's war gaming, internal competitive analysis might
ensure that minority views in the CIA were aired along the way.
(3) From time to time, the DCI might sit in on coordinative
meetings between CIA and Pentagon drafters. To this author's knowledge,
no DCI has ever done so, although DCIs have called in dissident
Pentagon officials to hear their complaints. By never observing
first-hand, the DCI has lost an opportunity to witness coordinative
methods and hear dissent from what has been termed the "bottom up."
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 122 -
(4) This author supports the continuance of a system whereby
Pentagon intelligence officials, and sometimes consultants, prepare
first drafts of selected chapters. By this time, sufficient expertise
in *riting should have been accumulated by non-CIA analysts to over-
come earlier criticims on this score. On a periodic basis, the
portion assigned to individual agencies, including the CIA, might be
rotated. Rotation need not be annual, since such a move would in-
crease the administrative burden on the NIO, who is ultimately
responsible. Periodic rotation reflects another internal approach
to competitive or parallel drafting. It creates both a precedent
and history for mulitple advocacy.
(5) In its day, the footnote system in the strategic NIEs was
much abused. Both recent Presidents and DCIs have preferred an
NIE format which represents major differences of opinion as al-
ternative positions, or parallel texts. The advantage of this format
is that all participants are encouraged to express their objections
clearly, advancing only those which reflect a significant, documentable
predictive uncertainty. Presently, there is more stylistic clarity
and cohesion in the strategic NIEs than in the past. This new
format also allows the DCI the option of advancing his personal
opinion as an alternative separately, or alongside of, other views
in the NIEs. However, exclusive reliance on this format has the
disadvantage of eliminating altogether a view strongly held by one
individual or a distinct minority, which readers might want to
bear in mind. This author believes that the NIOs should permit
the use of footnotes in occasional instances. The DCI and NFIB
members might overrule footnotes later on a case-by-case basis, if
they so chose.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 123 -
(6) Ongoing independent or outside review of the strategic
NIEs is a complex issue. Major difficulties are the administrative
burden on the DCI and CIA staff, potential security leaks and,
as the Senate Select Committee on Intelligence noted, the possibility
that reviewers are biased. However, because of the large impact
of the 1976 A-B Team experiment on subsequent perceptions in the
strategic NIEs, this author agrees with the former PFIAB's recommen-
dation that external review once again be a part of the NIE process.
Because of the immense security risk, external reviews should be
authorized only periodically, such as once every three years. Un-
less otherwise directed by the President, the time, conditions and
reviewers should be authorized by the DCI with the assistance of
another governmental body, such as the NSC or Presidential study
group.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
FOOTNOTES
CHAPTER I: OVERVIEW
1. See, Richard K. Betts, "Analysis, War, and Decision: Why Intelligence
Failures Are Inevitable," World Politics 31, no. 1 (Oct. 1978), P. 64.
2. Harry S. Truman, Memoirs, Vol II: Years of Trial and Hope (NY:
Doubleday, 1956), P. 58. DIA was created in 1961, in part, "to pro-
vide more objective intelligence than that being produced by the
service intelligence components," and flowed from a study group es-
tablished by President Eisenhower, known as the Kirkpatrick Commission.
(See, United States Senate, Select Committee To Study Governmental
Operations With Respect to Intelligence Activities, known as the
Senate Select Committee on Intelligence, Final Report, Supplementary
Detailed Staff Reports on Foreign and Military Intelligence, Book IV,
Washington, DC: GPO, April 1976, p. "7-6, fn. 10).
3. The origin of the term "liberal" appears to have been Henry Kissinger.
(See, Lawrence Freedman, U.S. Intelligence and the Soviet Strategic
Threat, Boulder, Colorado: Westview Press, 1977, p. 47). This view
also was expressed by James Schlesinger. (See, Senate Select Committee
on Intelligence, Final Report, Foreign and Military Intelligence,
Book I, Washington, DC: GPO, April 1976, pp.76-77).*
4. John Huizenga, former Chairman of the Board of Estimates, in testi-
mony to the SSCI. See, SSCI, Book I, p. 76.
5. Report of the Commission on the Organization of the Government
for the Conduct of Foreign Policy, known as the Murphy Commission
Report, Appendices, Volume 7 (Washington, DC: GPO, June 1975),
Appendix U: Intelligence Functions Analysis, p. 11.
6. See, for example, Alexander George, "The Case for Multiple Advocacy
in Making Foreign Policy," American Political Science Review, Vol. 66
(Sept. 1972).
7. SSCI, Book I, op. cit., p. 272.
8. This term is drawn from Benjamin Schemmer, "The Slow Murder of the
American Intelligence Community," Armed Forces Journal International,
March 1979, p. 53. See also Schemmer, rThe Intelligence Community's
Case Against Turner," Washington Post, "Outlook," Sunday, April
8, 1979, p. D-5.
*Henceforth, the Senate Select Committee on Intelligence is abbreviated
by SSCI, and all references are to its Final Report of April 1976
unless otherwise indicated.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
CHAPTER II: EVOLUTION OF THE STRATEGIC ESTIMATIVE PROCESS
1. National Security Act of 1947, Section 102(d), subparagraph (3).
2. Harry S. Truman, Memoirs, Vol. II, op. cit., p. 58.
3. SSCI, Final Report, Book I, op. cit., p. 73.
4. National Security Act of 1947, Section 102(d), subparagrpah (3).
5. Presidential Directive, January 22, 1946, establishing the National
Intelligence Authority, Central Intelligence Group and position of
the DCI. See, SSCI, Book IV, "History of the CIA." by Anne Karalekas
(henceforth referred to as "History") pp. 113-114.
6. Sherman Kent, Strategic Intelligence for American World Policy
(Princeton: Princeton University Press, Paperback Edition, 1966)
pp. 64-66. This book was published first in 1949.
7. Ibid., p. 101.
8. SSCI, Book IV, "History," pp 13 and 24.
9. Military estimates work was done in ORE's Global Survey Division, mainly
by three men. They were generalists in the sense of having little
Soviet expertise. However, Ludwell Montague, an historian, had been
the Army's Executive Secretary of the JCS' Joint Intelligence Committee,
and DeForest Van Slyck, a lawyer, had WWII intelligence experience
with the Air Force. Interview with DeForest Van Slyck, 5/2/78.
See also: SSCI, Book IV, "History," p. 13; Ray S. Cline, Secrets,
Spies and Scholars (Washington DC: Acropolis Books Ltd., 1976)
p. 105.
10. The Eberstadt Task Force on Military Intelligence, of The Hoover
Commission, which reviewed both the Air Force's and CIA's es-
timates in the spring of 1948. Quoted in The Washington Post,
October 11, 1948.
11. SSCI, Book IV, "History," p. 13, and interview with DeForest Van Slyck,
5/2/78.
12. Declassified Documents Reference System (Washington, DC: Carrollton
Press, 1977 to present), which ihis author read in its entirety.
See, for example, DpRS, #77-261C, ORE #27-48, "Possible Developments
in China," Nov. 2, 1948.
13. SSCI, Book IV, "History," p. 25.
14. Study on the CIA and National Organization for Intelligence, known as
the Dulles-Jackson-Correa Survey, commissioned by the NSC in 1948
and completed in Jan. 1949. See, SSCI, Book IV, "History," p. 16.
nni-laccifiinri in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
-3-
15. SSCI, Book Iv, "History," p. 17.
16. Ibid., p. 16.
17. Ibid., p. 11, and interviews with Lawrence Houston (3/22/78) and Walter
FREFzheimer (4/13/78). Admiral Hillenkoetter's White House briefings
occurred after 1949 and represented a victory for the CIA, since the
State Department earlier had challenged its right to channel intelli-
gence directly to the President.
18. Memorandum from CIA General Counsel, Lawrence Houston, to General W.
Bedell Smith, August 1959, quoted in SSCI, Book IV, "History,"
p. 18.
19. Congress conducted full-scale hearings of the "surprise" invasions in
mid-1951, and agreed that the intelligence data had been available,
and circulated. See, for example, Richard K. Betts, "Analysis, War
and Decision," op. city, p. 63.
20. Admiral Hillenkoetter testified to the Senate Foreign Appropriations
Committee on the following day. (Interview with Walter Phorzheimer,
4/6/79, who helped prepare the testimony). The quote is drawn from
the Norfolk Ledger Dispatch, May 25, 1951.
21. Washington Evening Star, May 19, 1951, about Admiral Hillenkoetter's
testimony to Congress of that day.
22. Washington Post, October 1, 1950, p. 5.
23. Declassified Documents Reference System, 2E. cit., #77-259C (Memo of
June 25, 1948), 77-264D (Memo of June 30, 194YT7 77-261C Memo of
Nov. 23, 1949), and 77-259E (Memo of Dec. 10, 1948). The memos in-
cluded reference to the CIA's grading of agent (source) and information
reliability. Hillenkoetter hardly was the last DCI to do this.
24. SSCI, Book IV, "History," p. 18 and fn. 11. "Centralization of Authority"
was a term used frequently by Hanson Baldwin, who covered intelligence
activities for the New York Times. See, for example, Hanson Baldwin,
in Armed Forces, October 18, 1947.
25. 1enera1 Smith reportedly asked President Truman to call Langer, and
he and other academicans were recruited on the basis of Smith's
concern that WWIII was "imminent." (Interviews with Jack Maury,
2/9/79, and John Bross, 5/29/79). When Smith arrived, he reportedly
found no current, coordinated estimate on military operations in
Korea, which precipitated one of his famous battles with Gen. Douglas
MacArthur. (See, SSCI, Book IV, "History," p. 104).
26. Washington Post, Feb. 25, 1951.
27. Harry T. Rositzhe, The CIA's Secret Operations (NY: Reader's Digest
Press, 1977), p. 20,. According to Rositzhe, the Pentagon's demands
were specific and forcefully stated, and led to several of the ill-
fated drops in the Ukraine.
narinccifipri in Part - Sanitized CoPv Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5 ?,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
4
28. Interview with past and current CIA official (B), 5/30/79.
29. For this, see, SSCI, Book IV, "History," pp. 20-23.
30. Cline, Secrets, Spies and Scholars, op. cit., p. 120. Also inter-
views with DeForest Van Slyck (5/2/78), Paul Sorel (9/1/78), Ray
Cline (8/20/78) and John Huizenga (9/13/78). The CIA did not want
"professional meeting-goers."
31. SSCI, Book IV, "History," p. 19.
32. Interview with past and current CIA official (B), 5/30/79.
33. Cline, Secrets, op. cit., pp. 111 and 121.
34. Interview with John Bross, 5/29/79.
35. Ibid.
36. SSCI, Book IV, "History," p. 18.
37. Ibid.
38. Cline, Secrets, op. cit., p. 111.
39. Quoted in Freedman, U.S. Intelligence and the Soviet Streategic Threat,
op. cit., p. 35.
40. The press reports that: "In the 35 to 40 problems in which the CIA
has had to furnish an estimate since General Smith arrived...there
have been no dissents...." (See, Providence Bulletin, May 25, 1951).
However, this author found a 3 page-long dissent by the Air Force
in NIE-12, "Consequences of the Early Employment of Chinese Nationalist
Forces in Korea," Dec. 27, 1950 (Secret, Declassified Aug. 2, 1977).
John Bross says that Smith's main concern "was to get out the product."
(Interview, 5/29/79). Hal Ford reported one instance in which Smith
initially overruled a Navy dissent pertaining to the effects of a
Naval blockade of China during the Korean war, but reinstated the
footnote following a complaint from CINCPAC. (Interview, 12/7/78).
41. Interview with past and current CIA official (A), 5/30/79.
42. Interview with Gen. Daniel Graham, 5/9/79.
41. Cline, Secrets, op. cit., p. 123.
44. Interview with past and current CIA official (A), 5/30/79.
45. Freedman, op. cit., p. 35.
46. Colin S. Gray, "'Gap' Prediction and America's Defense: Arms and
Race Behavior in the Eisenhower Years," Orbis, Vol XVI-1, Spring
1972, p. 263.
47. Ibid., p. 261, and Freedman, op. cit., p. 67.
nariaccifipn in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- -
48. Freedman, op. cit., p. 67.
49. Gray, op. cit., p. 263.
50. Allen W. Dulles, The Craft of Intelligence (NY: Harper and Row, 1963)
p. 149.
51. Gray, op. cit., p. 262.
52. Ibid.
53. Interview with past and current CIA official (A), 5/30/79.
54. Gray, op. cit., p. 262.
55. Stuart Symington, "Where the Missile Gap Went," The Reporter, Feb.
15, 1962, P. 22. Symington earlier had conducted Senate hearings
on the bomber and missile gap predictions.
56. Interviews with Richard Bissell (5/26/78), Jack Maury (2/9/79), and
John Bross (5/29/79). See also, David Wise and Thomas Ross, The
U-2 Affair (NY: Random House, 1964). Bross says that in the early
1950s, the Air Force rejected one plan which would have been imple-
mented under its auspices, allegedly because it was too complicated
to put a man in the aircraft. Interviewees agree that, subsequently,
only a handful of persons outside of the CIA knew about development
plans, approved by President Eisenhower in 1954. Because of the
secrecy surrounding its development and use, the CIA provided the
administrative framework for all aspects of the U-2. In the late
1950s, coordination with the Air Force was informal, consisting of
a small Air Force contingent at CIA.
57. Following review by a formal interagency task force under Dulles,
all flights had to.be approved by President Eisenhower. Eisenhower
gave only one blanket approval, for the first ten days. After that,
individual flights were reviewed in a meeting at the White House con-
sisting of the President, the Secretary of State, the Chairman of the
JCS, Eisenhower's special advisor Gen. Andrew Goodpaster, the NSC
Secretary, and Allen Dulles. Bissell suggests that John Foster
Dulles also had a final veto power. (Interview with Richard Bissell,
5/26/78).
58. See, Freedman, op. cit., pp. 32-33 (who cites a figure of 3,000); and
Thomas Powers, The Man Who Kept the Secrets: Richard Helms and the
CIA (NY: Alfred Knopf, 1979) p. 97 (who uses the 1,200 figure).
The office is called the National Photographic Interpretation Center
and remains a sub-unit of CIA.
59. Freedman, op. cit., p. 67; Gray, op. cit., p. 260; and James C. Dick,
"The Strategic Arms Race, 1957-1961: Who Opened a Missile Gap?"
Journal of Politics, XXXIV (Nov. 1972), pp. 1067-9.
60. SSCI, Book IV, "History," p. 56.
61. Interview with past and current CIA official (A), 5/30/79.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 ?,
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
6
62. An NSC Directive in 1951 had directed the CIA to implement an "agreed
activities" arrangement with the Pentagon, whereby the Pentagon would
turn over its peacetime covert operations to the CIA. In 1954, the
Doolittle Commission concluded that both the CIA and Pentagon were
"dragging their feet." See, Declassified Documents Reference System,
op. cit., "Report on the Covert Activities of the CIA," The Doolittle
Report, Sept. 4, 1954, #78-139C and 140B (Top Secret, Declass. in 1976).
63. SSCI, Book IV, "History," p. 60. Dulles consolidated the ERA's four
divisions into a Military Economic Branch in 1958, two years after
it was proposed. Actually, the OSI continued to expand its work
into areas earlier restricted by an agreement with DOD. In 1956, Dulles
also created the interagency Guided Missiles Intelligence Committee
under IAC, with the agreement of the Defense Secretary, but over the
objection of the JCS and services who wanted exclusive control.
(See, SSCI, Book IV, p. 61). Thus, some of the criticism is ill-
founded.
64. SSCI, Book IV, "History," pp. 62-63.
65. Interview with.past and current CIA official (A), 5/30/79.
66. Interview with Gen. George Keegan, 2/10/79. Keegan also said that
he was present at a meeting between President Eisenhower and President-
elect Kennedy in which Eisenhower said: "You must never let the
Pentagon and military intelligence put your party at risk."
67. The personnel figures were cited in an interview with past and current
CIA official (B), 5/30/79. Actually, OSR's work has not expanded
steadily in some areas. For example, after 1967 OSR commenced work
on some "friendly" countries and also Net assessments, which McNamara
approved but was opposed by the services. Helms reportedly was
embarrassed, and asked OSR to stop. (Interview with Gen. Daniel
Graham, 5/9/79). At least one author maintains that Colby restricted
OSR to current intelligence work. (Paul W. Blackstock, "The
Intelligence Community Under the Nixon Administration," Armed Forces
and Society, Vol. I, No. 2 Feb. 1975, p. 239).
68. John Huizenga, in testimony to the SSCI, 1/26/76, quoted in SSCI,
Book I, p. 76. All testimony of officials to the SSCI remains
classified.
69. James Schlesinger, testimony to the SSCI, 2/2/76, quoted in SSCI,
Book I, p. 76.
70. Interview with Andrew Marshall, 2/9/79.
71. Albert Wohlstetter originally published his findings in three articles
in Foreign Policy, entitled: "Is there a Strategic Arms Race?"
(No. 15, Summer 1974); "Rivals, But No Race," (No. 16, Fall 1974);
and "How to Confuse Ourselves" (No. 20, Fall 1975). A more ex-
tended version was published by the United States Strategic
Institute, Levends of the Strategic Arms Race, USSI Report 75-1.
Wohlstetter did not use the NIEs for his data, but rather relied on
the predictions within the Annual Posture Statements, and Defense
Program and Budget Statements, of the Secretary of Defense.
r)..r.iaccifiart in Part - Sanitized Copv Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
-7-
72. See, "Bush, Cline Colby and Helms on the CIA," The Washington
Quarterly: A Review of Strategic and International Studies, Vol. 1
No. 4 (Autumn 1978), published by the Center for Strategic and
International Studies, Georgetown University, p. 28.
73. Interview with past and current CIA official (A), 2/9/79.
CHAPTER III: THE DCI'S MANAGEMENT OF STRATEGIC NIEs AT THE DRAFT STAGE
.1 ?
SSCI, Final Report, Book IV, "History of the CIA," by Anne Karalekas,
op. cit., p. 74, fn. 7.
2. Ibid. This official was identified later to the author as Walt Elders,
McCone's executive assistant.
3. Freedman, U.S. Intelligence and the Soviet Strategic Threat, op. cit.,
p. 74.
4. Interview with William Hyland, 11/17/78 (on President Johnson); Inter-
view with past and current CIA official (A), 2/9/79 (on President
Carter).
5. Interview with Anne Karalekas, 10/13/78.
6. For example, See Ray Cline, Secrets, Spies and Scholars, op. cit.,
pp. 140-144. As an ONE staff member, Cline participated in the
first Net Assessment at the Pentagon, an undertaking reporting
directly to a special NSC subcommittee. Under Kennedy and Johnson,
Net Assessments were done in the systems analysis shop under the
Office of the Secretary of Defense. Under Nixon, there was a Net
Assessment unit on the NSC staff, which later was moved back to OSD.
7. SSCI, Book IV, "History," p. 80 and 81 (fn. 14).
8. Interview with past and current CIA official (A), 2/9/79, about Dulles'
briefings to Congress. Tales about Dulles' refusal to discuss
policy implications at NSC, referring the questions to his brother,
are legendary.
9. Powers, The Man Who Kept the Secrets, op. cit., p. 162.
10. Ibid.
11. Ibid.
12. See, for example, John D. Leacacos, "Kissinger's Apparat," Foreign
Policy 5 (Winter 1970), p. 22. Also: Interview with Andrew
Marshall, 2/9/79; and Freedman, op. cit., pp. 46-47.
13. Interview with past and current CIA official (A), 2/9/79.
14. Ibid., 5/30/79.
15. Freedman, op. cit., pp. 37-39. However, Sherman Kent told this author
that, in the early 1950s, disagreements occasionally were carried
forward to IAC. (Interview, 5/2/79).
nprdaccifien in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 ,?
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 8 -
16 Freedman, op. cit., p. 38.
17. Interview with Richard Helms, 6/7/79.
17a. Interview with past and current CIA official (A), 5/30/79.
18. Interview with Andrew Marshall, 2/9/79.
19. Freedman, op. cit., p. 36.
20. For example, interviews with Ray Cline (5/2/78), Robert Amory (9/13/78),
John Huizenga (10/26/78), and Williard Matthias (10/25/78). However
they agree that there was a qualititative difference in the two DCIs'
relationship with the Board. Dulles is described as "enjoying" Board
members, whereas McCone "respected" them. Under McCone, the relation-
ship was more formal and sometimes characterized by mutual distance.
21. There is some disagreement about when and if this occurred. Cline
wrote that it occurred under McCone, and may have weakened estimative
work. (Cline, Secrets, op. c4., p. 200). Later, Cline said that
the transfer dia-EUE73ccur until after he left as DDI, and Kent
persuaded Helms that the move was a good thing, since Kent might have
more independence. (Interview with Ray Cline, 8/20/78). John
Huizenga denies that this move ever took place (Interview, 10/26/78).
22. SSCI, Book IV, "History," p. 81.
23. Ibid.
24. Interview with past and current CIA official (B), 5/30/79, and con-
firmed by Gen. Daniel Graham, interview of 12/7/78. The SSCI also
concluded that the Pentagon had not always provided the CIA with
access to US strategic planning data. (See, SSCI, Final Report,
Book I, op. cit., p. 268).
25. For this and the following, interview with R. Jack Smith (the then
DDI), 9/20/78. Also, interview with past and current CIA official
(C), 12/15/78.
26. Freedman, op. cit., p. 52.
27. SSCI, Book IV, "History," p. 86. Also, interviews with Keith Clark
(9/19/78) and Jack Smith (9/20/78), and undenied by other Board
members. Clark maintains that ONE was not as in-bred as the Board
in the early 1970s.
28. Interview with Keith Clark, 9/19/78, regarding the one Board appointee,
who was an Admiral. By 1973, there were several vacancies on the
Board, in part because Schlesinger's intended reorganization was
known.
29. Interview with Keith Clark (9/19/78) and Hal Ford (12/7/78).
30. Interview with Tom Lattimer, 5/10/79.
31. Interview with past and current CIA official (B), 5/30/79.
npriacsified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
-9-
32. Freedman, op. cit., p. 43. M?NaMarals requestsfor work from ONEf
OSR and OE lce of Economic Research, which did bomb damage
assessments during the Vietnam war) have been well documented.
(See, for example, SSCI, Book IV, "History," pp. 80-81).
33. Freedman, 92. cit., p. 36. Also interview with Fritz Ermath, 10/26/78,
who later joined OSR, first as a RAND consultant. Ray Cline goes
even further, and criticizes the creation of the DDS&T as well.
Cline maintains, thereby, "that CIA advocacy of its own scientific
collection techniques became mixed up with its objective analysis
of all scientific and technical developments. The appearance of
objectivity was hard to maintain when analysis and collection were
supervised by the same staff." (Cline, Secrets, op. cit., p. 200).
34. Interview with John Huizenga (9/13/78) and Keith Clark (9/19/78).
Colby often is regarded as the "executor of Schlesinger's will."
However, both interviewees confirmed that Schlesinger had not made
up his mind about the precise form of reorganization.
34a. Interviews with Keith Clark (11/3/78), John Huizenga (10/26/78).
William Hyland (11/17/78) and Hal Ford (12/7/78). Clark said that
Schlesinger wanted the NIEs short enough "to make any NSC member
ashamed of not having read them." Actually, Schlesinger's approach
clashed with that of Henry Kissinger who, in effect, wanted longer
NIEs.
35. Office of Management and Budget, "A Review of the Intelligence Community,
March 10, 1971, known as the Schlesinger Report, p. 10a. See also,
Blackstock, "The Intelligence Community Under the Nixon Administration,"
op. cit., p. 232.
36. Interview with Daniel Graham, 5/9/79.
37. Interview with William Colby, 5/29/79.
38. William Colby, testimony to SSCI, 12/11/75, cited in SSCI, Book I, p.
75, fn. 17. Earlier,diffusion of responsibility was viewed as
an advantage, protecting individuals from outside pressure.
39. Interview with William Colby, 5/29/79.
40. Interviews with John Huizenga (9/13/78) and Sam Hooskinson (11/9/78).
As Colby told Keith Clark, "I want to multiply myself by twelve."
(Interview with Clark, 9/19/78).
41. SSCI, Book IV, "History," p. 86. Admiral Turner has continued the
MBO system, although its impact on DDO and DDI interaction remains
limited. (Interview, current CIA official (H), 9/25/79).
42. Interview with William Colby (11/17/78) and Jack Maury (8/24/78). See
also, Honorable Men: My Life in the CIA (NY: Simon and Schuster, 1978)
pp. 364-365. Colby belived that Angleton's suspicions about the DDO's
Soviet desk sources were inhibiting effective political intelligence
gathering.
narlaccifiRci in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 10 -
43. See, "Bush, Cline, Colby and Helms on the CIA," The Washington
Quarterly, Autumn 1978, op. cit., p. 29.
44. Interview with past and current CIA officials (A) (2/9/78) and (C)
(2/15/78). They agree that as the pace of the Senate investigations
of the CIA increased, the attention Colby could devote to each NIE
declined.
45. U.S. Congress, Subcommittee on Priorities and Economy in Government
of the Joint Economic Committee, Allocation of Resources in the
Soviet Union and China-1975 (Washington, DC: GPO, 1976), P. 54.
46. SSCI, Book I, p. 434.
47. Interviews with Tom Lattimer (5/10/79), Sam Hooskinson (11/9/78)
and Richard Betts (5/3/79). Reportedly, one Panel member said
that he didn't have enough to do.
48. Interview with current CIA officials (D) and (J), 12/15/78.
49. Interview with Ray Cline, 11/8/78.
50. Interview with past and current CIA official (A), 5/30/79.
51. Freedman, op. cit., p. 37.
52. The Pentagon always has prepared several Soviet strategic estimates
relating to different, U.S. joint defense planning parameters. For
example, in the 1960s these included the National Intelligence
Projections for Planning (NIPP), a ten-year projection, to which
McNamara encouraged CIA input. The JCS annually produced several
estimates ranging from 2-15 years. Under McNamara, the Pentagon
also prepared a separate "worst-case" Soviet estimate, based on
his interest in keeping this kind of estimating out of the strategic
NIEs. This was called the "Greater Than Expected Threat" estimate.
As excellent summary of the problem is provided by Gen. Daniel Graham,
"Estimating the Threat: A Soldier's Job," Army, April 1973, pp. 14-18.
53. Several interviews suggested that alliances were more or less
"automatic" to provide offsetting predictions. For example, inter-
views with Helmut Sonnenfeldt (11/30/78), Williard Matthias (10/28/78)
and William Hyland (11/17/78).
54. Freedman, op. cit., p. 25.
55. Allen Dulles, The Craft of Intelligence, op. cit., p. 52.
56. This was accomplished in McCone's new office, for National Intelligence
Programs Evaluation, established in 1963 to improve interaction
between components of the intelligence community. See, Thomas Powers,
The Man Who Kept The Secrets, op. cit., p. 160.
i7. Ibid.
nprdaccifien in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
58. Patrick J. McGarvey, "DIA: Intelligence to Please," in Morton
Halperin and Arnold Kanter, eds., Readings in American Foreign
Policy. A Bureaucratic Perspective (Boston: Little, Brown, 1973),
p. 324. See also: Report to the President and Secretary of
Defense on the Department of Defense by the Blue Ribbon Defense
P.anel, known as the Fitzhugh Report (Washington, D.C.: GPO, July,
1973); and Gen. Daniel Graham, "US Intelligence at the Crossroads,"
United States Strategic Review, 1976, pp. 3-17; and Edgar Ulsamer,
wMiritary Intelligence: Streamlined, Centralized and Civilianized,"
Air Force, August 1976.
59. Interview with William Hyland, 11/17/78.
60. Freedman, op. cit., p. 38. Gen Graham says this did not begin until
the mid-1960s. (Interview, 12/7/78).
61. Daniel Graham, in Allocation of Resources in the Soviet Union and
China-1975, op. cit., p. 97.
62. Freedman, op. cit., p. 49. Also, interview with Andrew Marshall,
2/9/79.
63. Interview with Gen. Daniel Graham. 12/7/78. Graham says that foot-
notes to the "tone" of the NIEs also were encouraged by Schlesinger
when he became Defense Secretary in mid 1973. John Huizenga said
this kind of footnoting occurred when the Pentagon could not
document its dissents precisely. (Interview, 9/13/78).
64. Interview with William Hyland, 11/17/78.
65. Interviews with Hal Ford (12/7/78) and Tom Lattimer (5/10/79).
66. Freedman, op. cit., p. 50.
67. Interview with Gen. Daniel Graham, 5/9/79.
68. Freedman, op. cit., p. 45.
69. Interview with William Hyland, 11/17/78. See also, Laurence E.
Lynn, in The Murphy Commission Report, op. cit., Volume 7,
Appendix U, p. 47.
70. Interviews with past and current CIA official (A) (2/9/79), Major
Gen. George Keegan (10/11/79), and Andrew Marshall (2/9/79). See
also, SSCI, Book I, p. 75.
71. Interview with Tom Lattimer, 5/10/79.
72. Interview with past and current CIA official (A), 2/9/79.
73. Freedman, op. cit., p. 55.
74. Interview with past and current CIA official (A), 5/30/79.
narlaccifian in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 12 -
75. Interview with Andrew Marshall (2/9/79) and Tom Lattimer (5/10/79)
who says "DIA's interest goes up and down."
76. Interview with past and current CIA official (A) (2/9/79 and 5/30/79),
and Andrew Marshall (2/9/79), who both agree as well that this method
helps eliminate "red herrings" and serves as a means to check minor
arithmetic errors.
77. Interview with Gen. Daniel Graham, 5/9/79. Also, Graham, "US
Intelligence at the Crossroads," op. cit., p. 11.
78. Interview with past and current CIA official (A), 5/30/79.
79. Ibid. Also, interview with past and current CIA official (C), 12/15/78.
80. This paragraph is derived from interview with current CIA official
(E), 9/25/79.
CHAPTER IV: SUBSTANTIVE INPUT OF DCIs TO THE DRAFT STRATEGIC NIEs*
1. SSCI, Final Report, Book I, op. cit., p. 258.
2. For example, interviews with Ray Cline, 5/1/78 and 11/8/78. Although
DCI's review was considered part of the process, Cline also said that
it worried Sherman Kent, lest this link the NIEs too closely to
policy considerations.
3. Interview with Bromley Smith, 9/12/78. Also, interview with Robert
Amory, 9/13/78, who as Dulles' DDI was a member of the NSC's Planning
Board.
4. Interview with John Bross, 5/29/79, who was quoting Ernie May, later
a member of The Murphy Commission.
5. While interviewees agree on this point, they vary in describing how
much of the NIE Dulles might read at this time. To this author, this
suggests inconsistent attention to different NIEs by the DCI. For
example, one interviewee said that Dulles "flipped through the pages"
(Jack Smith, 9/20/78); but Sherman Kent said that "he knew what was
in every estimate. The important ones he studied very carefully.
He never talked about one which he hadn't read." (Interview 5/2/78).
Stories of how Dulles kept Kent waiting into the early evening and,
then, sometimes became distracted during the NIE discussions, are
legendary.
Che following analysis is derived from multiple interviewees. Unless
:here is a difference of opinion, only the principal source or sources
is cited.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 13 -
6. Interviews with Ray Cline (5/1/78), and Keith Clark (9/19/78 and
11/3/78). Cline said that, in his heart, Dulles preferred using his
own words in the NIEs and, at this time, might dictate a new sentence
or paragraph to Kent. If there was quibbling among Board members
over words, Dulles might say: "I'm the boss" and insist on his
own wording. Clark said that Dulles believed that the NIEs should
offer firm predictions and/or give odds on the likelihood of occurrenc(
"'You're paid to estimate,'" he once told the Board.
7. Interviews with Ray Cline (5/1/78), Paul Sorel (9/1/78) and Sherman
Kent (5/2/78). Kent said that Dulles could be "genially bull-headed"
about a non-strategic NIE and, on occasion, insisted that his judge-
ment be represented. Cline concurs. Borel said that once Dulles
rewrote an entire SNIE and the Board dissented from it.
8. Interview with Deforest Van Slyck, 5/2/78.
9. Cline, Secrets, Spies and Scholars, op. cit., p. 143.
10. Hanson Baldwin, New York Times, May 6, 1970.
11. Interview with Ray Cline, 11/8/78. Cline says that McCone did not
want to say: "This is my estimate." He believed that "the NIE
system was bigger than he was," and would have preferred to take a
footnote himself.
12. Interviews with Gen. Daniel Graham (12/7/78) and William Hyland
(11/17/78). Graham says that McCone once told General LeMay at a cock-
tail party: "'Air Force Intelligence is the laughing stock of
Washington.'" As AEC Chairman, McCone had been on the low side of
the missile gap predictions, basing this on the AEC's estimate of
Soviet fissionable material.
13. Freedman, U.S. Intelligence and the Soviet Strategic Threat, op. cit.,
pp. 75-77. Since the CIA's estimates kept changing in different
drafts during this period, these figures merely provide one indica-
tion of downward revision.
14. Ibid., p. 101.
15. Interview with John Bross, 5/29/79.
16. Cline, Secrets, op. cit., .p. 192.
17. Interview with John Bross, 5/29/79. Interviewees disagree on McCone's
level of involvement, which to this author indicates inconsistent
attention to different NIEs, and McCone's preference for keeping his
personal views out of the draft. Most interviewees indicated a high
degree of serious interest on McCone's part. However, three said
that McCone's policy was "not to get involved" (Sam Hooskinson,
11/9/78), that he "never interfered with the process," (Williard
Matthias, 10/25/78, and that "he was not very interested" (William
Hyland, 11/17/78).
18. Cline, Secrets, op. cit., pp. 192-193.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 14 -
19. Interview with Keith Clark, 11/3/78.
20. InterviewwithJohn Bross, 5/29/79.
21. Interview with-Jack Maury, 2/9/79. Oleg Penkovsky's intelligence
suggested that the Soviets were having trouble with their ICBMs. Maury
said Penkovsky once observed: "'They couldn't hit a bull in the ass
with a balalaika.'"
22. Klaus Knorr, "Failures in National Intelligence Estimates: The Case
of the Cuban Missiles," World Politics XVI (April 1964), quoting
the Stennis Report investigating the Cuban military build-up.
23. Arthur Schlesinger, A Thousand Days: John F. Kennedy in the White
House (Boston: Houghton Mifflin, 1966), p. 798.
24. Interview with John Bross, 5/29/79. McCone made this statement at
Bross' home.
25. Thomas Powers, The Man Who Kept The Secrets, op. cit., p. 162. Also,
interview with Ray Cline, 11/8/78.
26. Interview with Richard Helms, 6/7/79.
27. Richard Helms, testimony to the SSCI, 1/10/76, cited in SSCI, Book I,
p. 79.
28. Interview with Richard Helms, 6/7/79; and Freedman, op. cit., p. 139.
29. Interview with Gen. Daniel Graham, 5/9/79. This view was confirmed
in interviews with John Huizenga (9/13/78), Ray Cline (11/8/78),
Hal Ford (12/7/78) and CIA official (I) (9/25/79). They agree that
Helms felt ambiguous about the strategic NIEs, in part because he
believed that they were not that influential and he didn't have
that much to contribute.
30. Interview with Richard Helms, 6/7/79.
31. Interview with William Colby (5/29/79) and John Bross (5/29/79).
32. Draft NIE 11-8-69, approved by the Board of Estimates prior to the
USIB meeting of 8/28/69, cited in SSCI, Book I, p. 78. The SSCI
investigated Helms' role in deleting this paragraph from the NIE.
33. Freedman, op cit., p. 133. According to John Bross (5/29/79) and
John Huizenga 10/26/78), the Helms-Laird battle was not cut-and-
dry. They say Helms called Laird first, invited him to send repre-
sentatives to the USIB meeting to discuss the paragraph, and sub-
sequently modified the wording. Laird told Helms that he was concerned
with the implications of the paragraph, should the Soviet get wind
of it.
34. The possibility of CIA bias was implied in the Schlesinger Report,
op. cit., 1971, pp. 10-11. According to Gen. Daniel Graham, Schlesinger
had midle several remarks about CIA bias privately, before assuming
office. However, Schlesinger's more famous comments were made after
he left as DCI, in testimony to the SSCI in 1976.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
35. Interview with Sam Hooskinson, 11/9/78, who was present.
36. SSCI, Book IV, "History," op. cit., p. 85.
37. Interview with William Colby (5/29/79) and Sam Hooskinson (11/9/78).
Hooskinson also called Schlesinger a "redrafter."
38. Interview with William Colby, 5/29/79.
39. Ibid.
40. Interview with past and current CIA official (C), 12/15/78.
41. Interview with William Colby, 5/29/79. One interviewee said that
Colby believed that the NIOs were there "to advise him on his views"
(Sam Hooskinson, 11/9/78).
42. David Binder, quoting a Pentagon intelligence official, in "CIA Head
Accused of Tailoring Estimates to Policy; He Denies It," New York
Times, Monday, November 6, 1978. Binder said that he based this
article on interviews with five CIA officials and the DCI. (Interview,
2/9/79).
43. Binder, NYT, Nov. 6, 1978, quoting Admiral Turner.
44. Interview with current CIA official (E), 9/25/79.
45. This paragraph is derived entirely from interview with past and
current CIA official (A), 5/30/79.
46. Ibid.
CHAPTER V: THE DCI'S LEADERSHIP ROLES AT USIB OR NFIB.
1. SSCI, Final Report, op. cit., p. 76.
2. Freedman, U.S. Intelligence and the Soviet Strategic Threat, op. cit.,
p. 40.
3. Ibid., p. 39.
4. Interview with past and current CIA official (A), 5/30/79.
5. Freedman, op. cit., pp. 66-67.
6. Stuart Symington, "Where the Missile Gap Went," The Reporter, Feb. 15,
1962, p. 25.
7. Daniel Graham, "The Intelligence Mythology of Washington, Strategic
Review, op. cit. p. 61.
8. Interview with Gen. Daniel Graham, 5/7/79. Graham said that Army
Intelligence wanted to target U-2 flights on medium-range missile sites,
but the CIA kept searching for ICBM sites. He said this was true in
the case of the ill-fated Gary Powers flight of May 1, 1960 about
which the Army had argued: "You won't find anything." See also,
Everett Allen, "Lack of Information led US to Overestimate Missile Lag,"
St. Louis Post-Dispatch, Jan. 27, 1965.
neclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 16 -
9. Dulles' posture at USIB was described in interviews with John Huizenga
(9/13/78 and 10/26/78), Jack Smith (9/20/78), Gen. Daniel Graham
(12/7/78), Ray Cline (8/20/78), Hal Ford (12/7/78) and past and
current CIA official (C) (12/15/78). Interviews agree that Dulles
listened to lengthy debate but sometimes grew impatient towards the
end. They agree that Dulles' "heart was in favor of an unspoken
compromise" and he defended the CIA view only as necessary. How-
ever, on non-strategic NIEs he occasionally was impatient with the
CIA as well, and sometimes encouraged the Pentagon to express its
views. For example, this happened in the case of a SNIE during the
1956 Suez crisis. (Interview with John Bross, 4/14/78).
10. The Joint Study Group Report on Foreign Intelligence Activities of
the United States Government, known as the Kirkpatrick Report,
December 15, 1960, p. 14. See, Declassified Documents Reference
System, op. cit., #77-270D. The report also asserts that Pentagon
members on USIB were not permitted to appeal its decisions without
the approval of the Defense Secretary.
11. Interview with Gen. Daniel Graham, 5/9/79.
12. Freedman, op. cit., p. 70.
13. Interview with Bromley Smith, 9/12/78, who was quoting General
Andrew Goodpaster, Eisenhower's Chief of Staff.
14. Colin Gray, "'Gap' Predictions and America's Defense," Orbis, op. cit.
p. 269.
15. Interview with past and current CIA official (A), 5/30/79. This was
confirmed in an interview with Charles V. Murphy, who also said that
during the campaign Nixon was "upset" and asked Dulles for his views.
16. Graham, "Estimating the Threat," Army, op. cit., p. 15. Several in-
terviewees said that, subsequently, CIA officials blamed the Air
Force for influencing the CIA's positions during the missile gap
period. See, for example, Allen Dulles, Harper's Magazine, April
1963, in which he writes that the CIA responded to "great pressure"
from the Defense Department. See also, Edgar M. Bottome, The Missile
Gap: A Study of the Formulation of Military and Political Policy (New
Jersey: Fairleigh Dickinson University Press, 1971).
17. Quoted in SSCI, Final Report, Book IV, "History," op. cit., p. 62.
i
The time and original source of Eisenhower's comment s unclear.
18. Letter from President John F. Kennedy to John McCone, Jan. 16, 1962,
quoted in SSCI, Book IV, "History," p. 73.
19. Freedman, op. cit., pp. 22-23.
20. Interviews with Gen. Daniel Graham (12/7/78), Gen. George Keegan
(2/10/79), and John Huizenga (9/13/78). McNamara privately criticized
McCone for interfering in internal DOD matters, and resented his
requests for help.
21. Freedman, op. cit, p. 23.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
22. Ibid.,p. 17.
23. Interview with Hal Ford (12/7/78) and William Hyland (11/17/78).
24. Interview with Williad Matthias, 12/25/28. This dispute concerned
a 1963 NIE on "Prospects in South Vietnam."
25. Freedman,op. cit., pp. 90-94. The question of whether the Tallin
Line installations could be "upgraded" into an ABM system reopened
in the late 1960s.
26. Interviews with past and current CIA official (A) (2/9/79 and 5/30/79)
and Gen. Daniel Graham (5/9/79).
27. Ibid. The Army general was Carl Duckett.
28. Interview with Gen. George Keegan, 2/10/78. The numbers are drawn
from Defense Secretary Robert McNamara's Posture Statement for Fiscal
Year 1964, Feb. 1963, p. 29.
29. Freedman, op. cit., p. 108. See Albert Wohlstetter, "How to Confuse
Ourselves,"- Foreign Policy, op. cit., pp. 185-187. Wolstetter explains
this by the "power of consensus and "pressure for conformity."
30. Wohlstetter, "Is There A Strategic Arms Race," Foreign Policy, op. cit.,
p. 5. The major focus of Wohlstetter's work was long-range ICBM projec-
tions. He explains that, in the short term, official ICBM projections
merely tried to keep up with and "finally touched reality," but in the
long-term "forecasts got worse and not better. To say. 'there was no
systematic learning' puts it mildly." (See, "How to Confuse Ourselves,"
op. cit., p. 180.)
31. Interviews with Hal Ford (12/7/78), John Huizenga (10/26/78), Ray
Cline (11/8/78), and current CIA official (I) (9/25/79).
32. Vincent Marchetti and John Marks, The CIA and the Cult of Intelligence
(NY: Alfred Knopf, 1974), p. 86.
33. Interview with Gen. Daniel Graham, 5/7/79, regarding DIA's attempt to
coordinate the services' estimates. Helms sometimes cut short debate
by banging the gavel. (Interviews with Hal Ford, 12/7/78, and CIA
official (I), 9/25/79).
34. Interview with Richard Helms, 6/7/79.
35. Interview with Gen. Daniel Graham, 5/9/79.
36. Interview with Gen. George Keegan, 10/11/79. Delays in obtaining
USIB approval also affected DCI Colby following his 1973 NIO
reorganization. Interviewees say this partly resulted from stylistic
difficulties in writing the NIEs.
37. Interview with Gen. Daniel Graham, 5/9/79.
38. Interview with past and current CIA official (A), 5/30/79).
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 18
38a. Interview with Gen. Daniel Graham, 12/7/78.
39. Interview with Gen. George Keegan, 2/10/79 and 10/11/79. Keegan
then considered Schlesinger's reforms "non-competitive."
40. Ibid. See also, William T. Lee, Understanding the Soviet Military
Threat: How CIA Estimates Went Astray (NY: National Strategy
Information Center, Agenda Paper Nov. 6, 1977) p. 36. Lee says that
in the early 1970s Air Force Intelligence translated nine basic books
on Soviet military and strategic doctrine. Previously, the only
translated Soviet book on this subject was that of the late Marshal
Sokolovskiy, Military Strategy. In the mid-1960s, Keegan had
challenged McNamarafs reliance on one translation of this book (by
Thomas Wolfe of the RAND Corporation), claiming that it was full of
inaccuracies. (Interview, 2/10/79).
41. Interview with Gen. Daniel Graham, 5/9/79.
42. Interview with Gen. George Keegan, 2/10/79. Neither Colby nor Keegan
were particularly satisfied with this method of resolving dissent.
One interviewee said that Colby worried about excessive footnoting.
(Sam Hooskinson, 11/9/78).
42a. Interview with past and current CIA official (A), 5/30/79.
43. Ibid.
44. Ibid.
45. Ibid.
46. SSCI, Book I, p. 79.
47. David Binder, quoting a Pentagon intelligence official, NYT, Nov. 6,
1978.
48. See, Blackstock, "The Intelligence Community Under the Nixon Admin-
istration," op. cit., p. 236, and The Murphy Commission Report, op. cit.,
Volume 7, Appendix U, pp. 11 and 44. Although the DCI carries more
weight than the other USIB members, one author concludes here that
"the DCI's chief tool is persuasion, and the normal outcome when
disputes occur is often no more than a partially satisfactory compro-
mise." (William J. Barnds, p. 12).
49. Interview with Richard Helms (6/7/79) and past and current CIA official
(A) (5/30/79).
50. There is some dispute about whether McCone presented his views to
Kennedy from Europe. Ray Cline says that McCone told Kent that he
would write a memo to Kennedy, but never did so. Rather, McCone's DDCI
went to see Kennedy. (Interview, 11/8/78). Another interviewee said
that McCone "went to the White House more than once with his own views"
on other NIEs. (William Hyland, 11/17/78).
51. Interview with past and current CIA official (A), 5/30/79. He also
said that McCone "felt strongly" about the Soviet's ABM potential and
"sharply modified" the CIA text. (Interview, 2/9/79).
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 19 -
52. For example, see Freedman, op. cit., p. 93.
53. Ibid.
54. Interview with William Colby, 5/29/79. However, on other matters
Colby wrote: "Helms had far too exquisite a sense of political
realities to get into a fight he knew he couldn't win, and which
could hurt both CIA and himself." (See, Colby, Honorable Men, op. cit.
p. 330).
55. SSCI, Book I, p. 80, fn. 29.
56. Interview with CIA official (I), 9/25/79.
57. Powers, The Man Who Kept The Secrets, op. cit., p. 177.
58. See, for example, SSCI, Book I, p. 77 and Laurence Lynn in The Murphy
Commission Report, op. cit., Volume 7, Appendix U, p. 48.
59. SSCI, Book I, p. 78 and fn. 25.
60. Interview with Gen. George Keegan, 10/11/79.
61. Interview with Hal Ford, 12/7/78, pertaining to the National Security
Agency.
62. Interview with Gen. Daniel Graham (5/9/79) and Dick Anderson (5/10/79).
63. Interview with Gen. Daniel Graham, 5/9/79.
64. Interview with William Colby, 5/29/79.
65. Interview with Gen. Daniel Graham (5/9/79)and John Bross (5/29/79).
66. Interview with William Colby (5/29/79) and Gen. Daniel Graham (5/9/79).
67. Interview with William Colby, 5/29/79.
68. Ibid.
69. Interview with Gen. Daniel Graham, 5/9/79.
70. David Binder, NYT, Nov. 6, 1978.
71. Ibid., in an interview with Admiral Turner.
72. Binder, draft of NYT article (provided to the author), prepared
Oct. 12, 1978.
73. Binder, NYT, Nov. 6, 1978.
74. Interviews with current CIA official (E) (9/25/79) and past and current
CIA official (A) (5/30/79).
75. Ibid.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 20 -
CHAPTER VI. THE DCI'S ROLE IN PROMOTING INDEPENDENT REVIEW
1. SSCI, Final Report, Book IV, "History," op. cit., p. 19.
2. Interviews with Klaus Knorr (5/30/79), Hal Ford (12/7/78), Paul
Sorel (9/1/78) and Gen. Daniel Graham (12/7/79).
3. Interview with Klaus Knorr, 5/30/79.
4. Ibid.
5. Freedman, U.S. Intelligence and the Soviet Strategic Threat, op. cit.,
p. 40.
6. Ibid.
7. Interview with Klaus Knorr (5/30/79) and Ray Cline (5/2/78).
8. Freedman, op. cit., p. 40.
9. SSCI, Book I, testimony of Ed Proctor (former DDI) of 3/1/76, p. 276,
fn. 46. These studies were self-initiated and remain classified. Also,
interviews with Ray Cline (8/20/78) and current CIA official (K) 5/30/79).
10. See, for example, Blackstock, "The Intelligence Community Under the
Nixon Administration," op. cit., p. 241.
11. Interview with George Keegan, 10/11/79. Other subcommittee chairman
are automatically appointed by the Defense Secretary, such as for
GMIC's successor.
12. Freedman, op. cit., p. 50. See also, John Newhouse, Cold Dawn: The
Story of SALT (NY: Holt, Rinehart and Winston, 1973).
13. SSCI, Book I, p. 276.
14. SSCI, Book IV, "History," p. 85.
15. SSCI, Book I, p. 271.
16. Ibid.
17. Lee, Understanding The Soviet Military Threat, op. cit., in a Foreword
by Eugene V. Rostow, p. 1. This is also referred to as "mirror imaging."
18. Interview with anonymous past CIA official, 4/5/79. See also, The
Murphy Commission Report, op. cit., Vol. 7, Appendix U, p. 75, iNI
interview with William Casey PFIAB member in late 1976), 5/7/79.
19. SSCI, Subcommittee on Collection, Production, and Quality, The
National Intelligence Estimates A-B Team Episode Concerning Soviet
Strategic Capability and Objectives (Washington, DC: GPO, Feb. 16,1978),
p. 2.
20. Interview with William Colby, 5/29/79.
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
- 21 -
21. Interview with Gen. Daniel Graham, 5/7/79. Graham said that, throughout
1975, PFIAB asked Colby several times to authorize an outside review,
but Colby hesitated on security grounds.
22. Interview with Gen. George Keegan, 2/10/79.
23. SSCI, The National Intelligence Estimates A-B Team Episode, op.
p. 2.
24. Ibid.
25. Interview with William Colby (5/29/79 and Fritz Ermath (10/26/79).
However, the experiment soon became an administrative nightmare and
the effort to preserve secrecy failed. William Casey described Bush's
attitude at the end of the year as "not too positive." Interview,
5/7/79.
26. David Binder, "New CIA Estimate Finds Soviet Seeks Superiority in Arms,"
NYT, Dec. 26, 1976. The article is reproduced in Lee, Understanding
Soviet Military Threat, op. cit., p. 63-69. Apparently, it is
this article which later caused controversy.
27. David Binder, draft of NYT article (provided to the author), prepared
Dec. 23, 1976. Binder Figed his story on interviews with Bush and
other participants.
28. Ibid. (draft).
29. Binder, NYT, Dec. 26, 1976.
30. Ibid.
31. Ibid.
32. SSCI, The National Intelligence Estimates A-B Team Episode, op. cit.,
"Separate Views of Senator Daniel P. Moynihan," p. 9.
33. Binder, NYT, Dec. 26, 1976.
34. SSCI, The National Intelligence Estimates A-B Team Episode, op. cit.,
p. 2. The Senate also reported that the press account of the experiment's
results "appeared in garbled and one-sided form." (p. 2.).
35. Interview with Keith Clark, 11/3/78. Also, interviews with William
Casey (5/7/79) and Gen. Daniel Graham (5/9/79).
36. Interview with current CIA official (F), 9/25/79.
neclassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
LIST OF INTERVIEWEES*
Robert Amory
William Bader
Richard Betts
David Binder
Richard Bissell
Paul Borel
James Billington
John Bross
Deputy Director for Intelligence (DDI), CIA,
1951-1961 (9/13/78)
STAT
Staff, Office of Strategic Research, CIA, 1974-1977;
Staff, House Committee on Intelligence, 1978-present
(5/10/79)
Task Force Leader, Senate Select Committee on
Intelligence, 1974-1977; Office of the Secretary of
Defense, 1978; Staff Director, Senate Foreign
Relations Committee, 1979-present (12/5/78)
Staff, National Security Council and Senate Select
Committee on Intelligence, 1975-1977 (9/29/78,
11/9/78, and 5/3/79)
Editor, New York Times, Washington office, writer on
Soviet bloc and strategic affairs (2/9/79 and 5/29/79)
Special Assistant to the DCI, 1953-1954; Director,
Development Projects Office, CIA , (for the U-2),
1954-1958; Deputy Director for Plans (DDP), 1958 to
early 1962 (5/26/78)
Office of Research and Evaluation, CIA, 1947-1950;
Director of Administration, Office of National
Estimates, CIA, 1950-1962; Assistant to the Deputy
Director for Intelligence, CIA, 1962-1966 (9/1/78)
Staff, Office of National Estimates, CIA, 1954-1957
(10/13/78) .
Eastern Europe Division, Directorate for Plans, CIA,
1950s; Comptroller, CIA, 1962-1964; Deputy to the
DCI for intelligence community coordination (head
of National Intelligence Programs Evaluation),
1964-1971 (2/14/78 and 5/29/79)
William Bundy Staff, Office of National Estimates and National
Security Council liaison, CIA, 1950-1961 (10/11/78)
William Casey
Bruce Clark
Under-Secretary of State, 1973-1974; Member, Murphy
Commission, 1974-1975; Member, President's Foreign
Intelligence Advisory Board, 1976-1977 (5/7/79)
Staff, Basic Intelligence Division of the Office of
Research and Reports, CIA, 1950s; Chief, Military
Intelligence Branch, Office of Current Intelligence,
and Military Intelligence Division, CIA, 1960-1967;
Director, Office of Strategic Research, CIA, 1967-
1973; Director, National Foreign Assessment Center
(D/NFA), CIA, late 1979 to present
*The author lists mainly former positions held relevant to the strategic NIE
process. Date(s) of interview is provided for all interviewees except current
CIA officials. For them, dates are cited in the footnotes and on a separate
page.
narinccifipri in Part - Sanitized Com Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Keith Clark
Ray Cline
William Colby
John DesPres
Hal Ford
Lieut. Gen.
Daniel Graham
Richard Helms
Sam Hooskinson
- 2 -
Staff, Office of National Estimates, CIA, 1950-1963;
Member, Board of Estimates, CIA, 1963-1973; National
Intelligence Officer (Western Europe), CIA, 1973-1978
(9/19/78 and 11/3/78)
Staff Director and other assignments in the Office
of National Estimates, CIA, 1950s; Deputy Director
for Intelligence, CIA, 1962-1966; Director, Bureau
of Intelligence and Research, State Department,
1969-1975 (5/1/78, 8/20/78 and 11/8/78)
Station Chief, Deputy Director of Pacification,
Director of Phoenix Program, in Vietnam, CIA,
1959-1972; Executive Director-Comptroller of CIA,
1972; Deputy Director for Plans (DDP), early 1973;
Director of CIA, officially from September 4, 1973
to January 30, 1976 (11/17/78 and 5/29/79)
National Intelligence Officer (Nuclear Proliferation),
CIA, at present
RAND Consultant and Staff, Office of Strategic
Research, CIA, 1973-1976; Intelligence Community
Staff and B Team liaison, 1976-1977; Staff, National
Security Council, 1978 to present (10/26/78)
STAT
Staff, Office of National Estimates, CIA, 1950-1973;
Senior Staff, Senate Select Committee on Intelligence,
1973 to present (12/7/78)
Special Assistant to the Director of the National STAT
Foreign Assessment Center, CIA, 1977-1979; Assistant
to the DCI, at present
Army intelligence officer, 1950s; Staff, Office of
National Estimates, CIA, 1963-1965; Director, Defense
Estimates, Defense Intelligence Agency, 1970-1973;
Assistant to the Director of Central Intelligence,
1973; Director, Defense Intelligence Agency, 1973-1978
(12/7/78 and 5/9/79)
Chief of Operations, Directorate for Plans (DDP), CIA,
1952-1961; Deputy Director for Plans (DDP), 1962-1965;
Deputy Director of CIA, mid-1965 to mid-1966; Direc-
tor of CIA, officially from June 30, 1966 to
February 2, 1973 (6/7/79)
National Intelligence Officer (Middle East) CIA,
1974-1977; Director, Intelligence Subcommittee,
National Security Council, 1978-1979 (11/9/78)
Lawrence Houston General Counsel of CIA (and its predecessor body),
1945-1973 (3/22/78)
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
- 3 -
John Huizenga
William Hyland
Anne Karalekas
Maj. Gen.
George Keegan
Sherman Kent
James King
Klaus Knorr
Richard Leyman
Andrew Marshall
Williard Matthias
Staff, Office of National Estimates, CIA, 1952-1964;
Member, Board of Estimates, 1964-1966; Deputy
Chairman, Board of Estimates,1967-1971; Chairman,
Board of Estimates, 1971-1973 (9/13/78 and 10161S
S TAT
Center for the Study of Intelligence, CIA, at presen-
Staff, Office of National Estimates, CIA, 1960-1968;
Deputy Assistant for National Security Affairs,
White House, 1969-1973; Director, Bureau of Intelli-
gence and Research, 1973-1978 (11/17/78)
State Dept.,
Staff, Senate Select Committee on Intelligence,
1974-1977 (10/13/78)
Air Force Intelligence officer, 1950s and 1960s;
Assistant Chief of Staff for Intelligence, Air
Force, 1972-1977. (2/10/79 and telephone con-
versation 10/11/79)
Deputy Chairman, Board of Estimates, CIA, 1950-1951;
Chairman, Board of Estimates, CIA, 1952-1967 (5/2/78)
Assistant to the Director of the National Foreign
Assessment Center (for academic liaison), CIA,
1977 to present
Member, Princeton Consultants, 1950-early 1960s;
Member, Senior Review Panel, National Foreign Assess-
ment Center, CIA, 1979 to present
STAT
Staff, Office of Current Intelligence, CIA, 1958-1970;
Staff, National Security Council, 1970-1975; Principal
Deputy to the Assistant Secretary of Intelligence
(and head of DIA), Department of Defense, 1975-1977;
Staff Director, House Intelligence Committee, 1977
to present (5/10/79)
Intelligence staff, CIA, since 1950; Director, Office
of Current Intelligence in 1970s and deputy to the
DCI for National Intelligence, 1976-1977; presently,
deputy to the Director of the National Foreign Assess-
ment Center, and National Intelligence Officer (for
Warning), CIA
National Security Council staff and head of NSC Net
Assessment, 1969-1973; Director, Office of Net
Assessment, Office of Secretary of Defense, 1973
to present (2/9/79)
Staff, Office of Research and Evaluation, CIA, 1947-
1950; Staff, Office of National Estimates, CIA,
1950-1963, Member, Board of Estimates, 1963-1973
(10/25/78)
narinccifipri in Part - Sanitized Com/ Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
Jack Maury
Charles Murphy
William Parmeutter
Walter Phorzheimer
Helmut Sonnenfeldt
Bromley Smith
R. Jack Smith
Howard Stoertz
4
Staff, Office of National Estimates, CIA, 1950-1951;
Chief, Soviet Division, Directorate for Plans,
CIA, 1954-1962; Assistant to the Defense Secretary,
and the Navy Secretary, 1974-present (8/24/78 and
2/9/79)
Former Air Force Intelligence officer, journalist
and author on strategic affairs (5/3/78)
National Intelligence Officer (Africa), CIA, 1973
to present
Legislative Counsel, CIA (and its predecessor body)
1945-1956; Curator, Historical Intelligence (CIA)
Collection, 1956-1974 (4/14/78 and 4/16/79). The
author gratefully acknowledges the use of Walter
Phorzheimer's personal clipping files in preparing
this study
Assistant to the Deputy Director of CIA, at present
Defense and State Department analyst, , 1960s; Staff
National Security Council, 1969-1977 (10/25/78
and 11/30/78)
Director, Office of Current Intelligence, CIA,
1953-1961 (8/8/78)
National Security Council staff, 1954-1956, and
Executive Secretary, Operations Coordinating Board,
1957-1961; Executive Secretary, National Security
Council, 1961-1968 (3/9/78 and 9/12/78)
Staff, Office of National Estimates, CIA, 1950-1959;
Member, Board of Estimates, 1959-1962; Director,
Office of Current Intelligence, 1962-1966; Deputy
Director for Intelligence, CIA, 1966-1972 (9/20/78)
Staff, Office of National Estimates, CIA, 1952-1965;
Chief, Photo Interpretation Section, the Office of
Strategic Research, SALT delegation representative
and special assistant to the DCI on SALT verifica-
tion, CIA, 1965-1973; National Intelligence Officer
(Strategic Intelligence), CIA, 1973-present
Staff, Office of Research and Evaluation, CIA,
1947-1950; Member, Board of Estimates, CIA, 1950-
1960 (5/2/78)
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04 : CIA-RDP93T01132R000100020034-5
STAT
STAT
STAT
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5
CODE TO CURRENT CIA OFFICIALS INTERVIEWED
A Howard Stoertz, 2/9/79 and 5/30/79
B - Bruce Clark, 5/30/79
C -
Richard Leyman, 12/15/78
D -
12/15/78 (brief interview)
STAT
E -
)/25/79
STAT
F
James King,
5/30/79
5/30/79
(and other conversations)
G--
5/30/79
STAT
H -
H-
9/25/79
STAT
I -
John DesPres, 9/25/79
(brief interview)
-
William Parmeutter,
12/15/78
K -
,
9/25/79
STAT
STAT
Declassified in Part - Sanitized Copy Approved for Release 2012/05/04: CIA-RDP93T01132R000100020034-5 .