Site Map

CHURCH COMMITTEE REPORTS
XII. CIA PRODUCTION OF FINISHED INTELLIGENCE
The main purpose of the intelligence system of the United States is
to provide the President, his chief advisers, and the Congress in appropruaA:
e ways with the 'best information about activities abroad ,that can
be obtained. It is not surprising, therefore, that the quality of finished
intelligence produced by the intelligence agencies has been a source
of continuing concern and controversy. Policymakers are understandably
seldom satisfied with the intelligence they receive, for they want
and need intelligence which eliminates uncertainties and ensures successful
policy decisions. Since such perfection is unattainable, however,
the realistic question is how to evaluate and improve the quality
of our finished intelligence. This is an extremely complicated and
difficultareJa. The simple answer is that 'there are no objective criteria
or standards that can·be universally applied. In the end, the assessment
by policymakers of the value and quality of our finished intelligence
is necessarily subjective. There is a record of steadily improved quality
over the years, bUlt the need. for a higher level of performance is accepted,
both at the policy level and among the intelligence agencies of
the U.S. Government.
The Committee's examination of the production of finished intelligence
focused on the CIA and within it, the Directorate of Intelligence
(DDI). This is by no means the whole of national intelligence,
but it is the core element in the production of finished national
intelligence. The CIA's Directorate of Intelligence is by far the best
analytical organization for the production of finished intelligence
within the Government, but it does have shortcomings. The CIA for
its part has, in the view of the Committee, made creditable efforts to
improve the quality of finshed intelligence, although much remains to
be done.
Because the provision of the best possible fact and predictive analysis
ito our policymakers is the most important mission of our intelligence
system. the problems of the production of finished intelligence
will require the most searching 'and systematic examinaJt~on'by a future
oversight committee. The preliminary work of the Select Committee
in this area is based on interviews and hearings, as well as documents
from the Intelligence Community Staff concerning their post-mortems
of past intelligence failures. Because of the complexity and difficulty
of the subiect matter, the examination of the Select Committee can only
be regarded as a 'beginning, only broadly indicative of the problems
involved, and suggestive of the areas which will require more thorough
and compr~hensivellItitention in the future.
Although the provision of intelligence analysis to policymakers is
the major purpose of the intelligence mission, the production of intelligence
has been referred to as the "st~pchild of the community." 1
It is an area which has been overshadowed by the glamour of clandestine
activities and the lure of exotic technical collection systems. Yet
(257)
258
the basic rationale ror intelligence operations is the provision of inrormation
to the people who need it in order to do their jobs-~he
President and other senior officials rei'ponsible ror the formulatIOn
and implementation of foreign policy.
The Pearl Harbor experience, which so heavily influenced the establishment
of the Central Intelligence Agency in 1947~ pointed to
the need for the collection, coordination, and analysis of all national
intelligence in a centralized fashion, so that policymakers could be
assured of receiving all the information they ne0ded, when they
needed it. Finished intelligence represents the "payoff" of investment
in the plethora of collection activities.
The CIA and its predecessor body, the Central Intelligence Group,
were established to rectify the duplication and biases that existed in
the intelligence production of the State Department and the military
services. By reviewing and analyzing the data collected by these departments,
the CIA was to provide senior government officials with
high-quality, objective intelligence. In practice, however, the CIA
has given precedence to independent collection and production, becoming
a competing department in the dissemination of information.
Historically, the departments resisted providing their data to the
Agency and thereby prevented the CIA from fulfilling its designated
role in the production of "coordinated" intelligence. Moreover, individual
Directors of Central Intelligence have not been consistent
advocates of the Agency's intelligence production function. For the
DCIs, the demands of administering an organization with thousands
of employees and in particular, the requirements of supervising
clandestine operations encroached on the mtended priority of intelligence
produotion. Only three DCIs attempted to address their primary
attention to the quality of intelligence production: Walter
Bedell Smith, John McCone,and James Schlesinger. In each case, the
DCI's attitude was a function of his background, his relative strength
as Director, and the particular demands of his time in office.
In recent years, however, and particularly with the introduction of
advanced technical collection systems, the requirement for bringing
together the vast quantities of information into useable analytic forms
has become the primary concern of the intelligence community.
In the course of its investigation, certain problems and issues in the
area of the production of finished intelligence in the CIA have come
to the attention of the Committee. The Committee believes these problems
deserve immediate attention by both the executive branch and
future. c0!1gre~sional ov~rsig~t bod.ies. These problems bear directly on
th~ pnonty gIven to filllshed mtellIgence by policymakers. Other issues
r~Ise~ here, such as the personnel system of the DDI and the organIzatIOnal
structure of intelligence production, are really functions of
the larger issue of priorities.
Briefly defined, the production of intelligence is the process whereby
~he d~ta collected by the intelligence community is transformed into
mtellIgence reports and studies that are relevant to the concerns of
seni.or po~icymakers. I!1telligence production involves many tasks. It
begms WIth the collatIOn and evaluation of incoming "raw" intelligence
repQrting-direct from the collectors, whether from open
sources, the clandestine service, or signals intercepts and other means
1 Office of M~nageme~tand Budget, "A Review of the Intelligence Community,"
3/10/71, (hereInafter Clted as the Schlesinger Report), p. 11.
of technical collection. The significance of new reporting is analyzed,
often in relation to intelligence already available on the subject. The
preparation of "finished" intelligence reports-the outcome of the production
process-thus entails the evaluation and analysis of the full
range of raw reporting from a variety of collection means.
Production of finished intelligence is done within the intelligence
community by the Central Intelligence Agency, the Defense Intelligence
Agency (DIA), and the State Department's Bureau of Intelligence
and Research (INR). Within the CIA (which is responsible
for the production of "national intelligence"), both the Intelligence
Directorate and the Directorate of Science and Technology (DDS&T)
produce finished intelligence. The Select Committee has focused on the
DDI, although the issues and problems cited are applicable in varying
degrees to the other production elements as wp,ll.
A. EVOLUTION OF THE CIA's INTELLIGENCE DIRECTORATE
The scope of the DDI mission is global. It covers the affairs of any
foreign country from the standpoint of politics, economics, defense,
geography, cartography and biography. Scientific reporting is largely
the responsibility of the Directorate of Science and Technology.
The Directorate of Intelligence was formally established on January
2, 1952. Specifically, the intelligence activities which the DDI
originally administered were:
a. Production of finished intelligence by the Offices of National
Estimates (ONE), Current Intelligence (OCI) , Research
and Reports (ORR), and Scientific Intelligence
(OSI).
b. Collection of essentially overt information by the Divisions
of the Office of Operations (00) : Foreign Broadcast
Information (FBID), Foreign Documents (FDD), and
Contacts (CD).
c. Dissemination, storage and retrieval of unevaluated intelligence
information and basic reference documentation
by the Office of Collection and Dissemination (OCD).
d. Coordination of intelligence collection by the Office of
Intelligence Coordination (01C) .
In the twenty-three years since its founding, the Intelligence Directorate
has gone through a number of reorganizations stimulated
by advice from external panels, changing international circumstances,
shifting requirements for finished intelligence production, and reduced
resources with which to perform its mission.2 Changes in the
first few years were fairly rare. In 1954, the Ole was abolished, and
in 1963 the Office of Scientific Intelligence was transferred to a new
Directorate for Science and Technology.
1. Intelligence Production
Estimative Intelliqence.-Producing National Intelligence Estimates
(NTEs) was the function of the Office of National Estimates
• The information contained in this section on the evolution of the DDI is derivt>
d primarily from a CIA paper prepared for the Select Committee by the
Office of the DDI, "The Directorate of Intelligence: A Brief Description". (Hereinafter
cited as "The Directorate of Intelligence.") December 1975.
260
which was in the Intelligence Directorate until 1966, when it became
a staff under the direction of the Director of Central Intelligence.
This move was made, in part, to emphasize that the NIEs were the
product of the entire intelligence community rather than a single
agency. ONE was abolished in 1973 and its responsibilities were transferred
to the newly formed National Intelligence Officers attached to
the Office of the DCI. With this move, much of the work of producing
draft estimates reverted to the production offices of the Intelligence
Directorate.
Current I ntelligence.-Primary responsibilities for producing current
intelligence remains where it has been since the Directorate was
established-in the Office of Current Intelligence. Originally, OCI
was responsible for all current intelligence reporting except economic.
At present, however, it eoncentrates on current political reporting,
leaving the preparation of reports on economic, military. geographic
and scientific developments to the research offices responsible for these
matters. OCI coordinates and consolidates this specialized reporting
on all subjects for presentation in its daily intelligence publications.
Basic lntelliqence.-Production of basic intelligence was stimulated
primarily by the realiz'ation in World War II that the U.S. Government
had too little information about many of the foreign countries
with which it was required to deal. The Basic Intelligence Division
(BID) or ORR was charged with responsibility for coordinating the
production of "factual intelligence . . . of a fundamental and more
or less permanent nature on all foreign countries." Because of the scope
of the subject matter, the production of this type of intelli~nce
required a cooperative effort involving the resources and capabilities
of several departments and agencies of the Federal Government. The
product of this government-wide effort was known as the National
Intelligence Surveys (NIS).
In 1955. BID became a sepamte office, the Office of Basic Intelligence
(OBI). This was in line with recommendations made in May 1955 by
the Task Force on Intelligence Activities.3 The elevation of Basic
Intelligence to Office status was an acknowledgment of the importance
that the Agency and the rest of the national security apparatus
attached to the NIS Program.
The early years of OBI were devoted mostly to the coordination
of this program. Many of the chapters were written by other elements
of CIA or by other government agencies on a contractual basis. In
1961, OBI took over responsibility for the production of the political
sections of the NIS from the State Department's Bureau of Intelligence
and Research when State claimed that it no longer had the resources
to do this work. OBI delegated the task of producing these sections
to OCI in 1962. In 1965, the geographic research function was
transferred from the Office of Research and Reports, creating the Office
of Basic and Geographic Intelligence (OBGI). The NISs continued to
be published until 1974 when the program was terminated because of
lack of resources. At this time, OBGI became the Office of Geographic
and Cartographic Research.
Military lntelligence.-Until the mid 1950's, the production of intelligence
on military matters had been considered the primary respon-
• The Clark Task Force, headed by Gen. Mark Clark, of the Hoover Commission.
For members of the task force, see Hearings, Vol. 4, p.1l2-13.
261
sibility of the Department of Defense. But the "bomber gap" and later
the "missile gap" controversies gave CIA a role in foreign military
research, an involvement which has continued and expanded. In 1960
the DDI created an ad hoc Guided Missiles Task Force to foster the
collection of information on Soviet guided missiles and to produ<X\ intelligence
on their manufacture and deployment. The Task Force was
abolished in 1961 and a Military Research Area was established in
ORR. As a result of increasing demands for CIA analysis of military
developments, a new Office of Strategic Research was established in
1967 by consolidating the Military-Economic Research Area of ORR
and the Military Division of OCL The scope and focus of responsibilities
of OSR have increased over the years and in 1973 a new component
for research in Soviet 'and Chinese strategic policy and military
doctrine was added.
Geographic Intelligence.-The Geographic Research Area (GRA)
of the Office of Research and Reports (ORR) originally had the responsibility
for geographic intelligence production. The GRA was
transferred in 1965 to the Office of Basic Intelligence changing its title
to the Office of Basic and Geographic Intelligence (OBGI). In 1974,
OBGI became the Office of Geographic and Cartographic Research
when the Kational Intelligence Survey (NTS) Program was
abandoned.
Economic Intelligence.-Activity in this area remains the responsibility
of the organization that succeeded the Office of Research and
Reports in 1967: the Office of Economic Research. In earlier years, the
Agency conctmtrated its economic research J.argely on the Communist
states. In recent years, however, the Department of State has dropped
much of its intelligence production on the non-Communist areas, leaving
this job to the Agency. OER has also expanded its research into
such subject areas as international energy supplies and international
trade. Today it is the largest research office in the Intelligence
Directorate.
Biographic Intelligence.-The Hoover Commission Report of 1949
recommended dividing the responsibility for biographic intelligence
production within the Community to prevent costly duplicaton. As a
result, the foreign political personality files mamtained by OCD
were transferred to State. In 1961, however, the Bureau of Intelligence
and Research claimed it no longer had the resources to provide
this service and the responsibility for reporting on foreign political
personalities and, subsequently, for all non-military biographic intelligence
reporting was transferred to CIA. The task was taken over by
OCD's successor organization, now the Central Reference Service.
In-Depth Political Research.-In-depth foreign political intelligence
reporting has not been, until recently, represented in the Office
structure of the Intelligence Directorate. Originally, whatever efforts
were made in this field were concentrated in OCr. In 1962, a modest
step toward increased foreign political research was taken with the
establishment of a Special Research Staff (SRS) in the Office of the
Deputy Director for Intelligence. In recent years, however, the diminished
role of State's Bureau of Intelligence and Research in intelligence
community affairs, a perceived need for more sophisticated work
in this field by CIA, and the appearance of new methods of political re262
search, including computer applications, encouraged the Directorate to
invest more resources in this area. Accordingly, an Office of Political
Research (OPR) was established in 1974. It incorporated the Special
Research Staff, some people from OCI and the then disbanding Office
of National Estimates.
Round-the-Clock WatchjAlert.-The Cuban Missile Crisis of the
fall of 1962 clearly spotlighted the need for a single Directorate facility
for round-the-clock receipt of intelligence information and for
a center in which the expertise of all its offices could be .rallied in
crisis situations. In March 19m~, the DDI set up a Special Study
Group on DDI Organizational Tasks to study this and other problems.
One of the results of its work was the establishment of an operations
center under the administrative direction of the Office of Current Intelligence
(OCI). Over the next ten years, the Operations Center grew in
size and capability, largely as a result of the Vietnam War. In 1974, it
was separated from OCI and renamed the CIA Operations Center, a
title warranted by the fact that all Directorates of the Agency now
maintain permanent duty officers within the Center. Today, the CIA
Operations Center provides the mechanism and facilities with which
the full information resources of CIA can be mobilized to work in
concert with the community in foreign crisis situations.
iJ. fntelligel1ce Oollection
At its founding in 1952, the Intelligence Directorate inherited the
Office of Operations (00) from the then Directorate of Plans-today's
Operations Directorate. 00 ,,'as composed of three main elements: the
Contact Division, the Foreign Broadcast Information Division, and
the Forei,gn Documents Division. The rationale for including these
components in the Intelligence Directorate was that their work was
essentially overt and thus inappropriately situated within the Clandestine
SerVIce.
The Domestic Contact Service originated in the Central Intelligence
Group in 1946 as an outgrowth of the 'World 'War II effort tq
insure that all domestic sources of information on foreign activities
were contacted by the Government. It was initially placed in 00 to
keep its essentially overt work separate from the clandestine activity
of the other major collection organizations. It maintained this separate
status after the founding of CIA, but in 1951 joined the Directorate
of Plans. This arrangenwnt lasted for only one year, however,
as the 00 and its Contact Division (CD) was moved to the Intelligence
Directorate in 1952. By 1953, CD was a network of offices
in 15 maior cities and several smaller residencies established across
the U.S. "With the abolition of 00 in 1965, CD became an independent
office known as the Domestic Contact Service (DCS) and continued
in that status until the appointment of William Colby as DCI. In
Inn, he decided that maintaining the separation of overt and covert
collection elements was less important than the goal of consolidation
of a11 human collection capabilities in the Operations Directorate. Accordingly,
the DCS was transferred to the Clandestine Service and
renamed the Domestic Collection Division.
The Foreign Broadcast Information Division (FBID) had been
founded by the Federal Communications Commission in 1940. With
the advent of World 'Val' II, it was absorbed by the Office of War
263
Information and. shortly thereafter, bf>came one of the original elements
of the OSS. At the end of the war, it was briefly administered
by the Department of the Army hefore joining the Central Intelligence
Group in 1946. It was formally included in the Agency's Directorate
of Plans at its founding ill December 1950 and remained there
as part of 00 until its transfer to the Intelligence Directorate in 1952.
By then it had established the \yorldwide net\york of broadcast monitoring
bureaus which-with some alterations in location-it operates
today. FBID received the status of an independent office and was renamed
the Foreign Broadcast Information Service with the dissolution
of the Office of Operations in 1965.
s. Information Processing
Between the collection and production phases of the intelligence
process there is an activity known as "information processing." Information
processing involves special skills or equipment to convert
certain kinds of raw information into a form usable by intelligence
analysts who are producing finished int€lligence. It includes things
like photointerpretation and translations of fore.ign documents as
well as the receipt, dissemination, indexing, storage, and retrieval of
the great volumes of data which must be available to the production
offices if they are to do their analytical work.
InfO/motion Di88eminatiol1. Storage al1d Retriel'al.-0ne of the
original offices of the Central Intelligence Group, thp Office of Collection
& Dissemination (OCD). began this work in 1948 whpn it introduced
business machines to improve reference, liaison and document
security services. Ultimately, this Office became CIA's own departmental
library and centralized document service. Its steady growth in
size and capabilities was given a boost in 1954, when responsibility
for the procurement of forpign documents was transferred to OCD
from the Dppartment of State. Other specialized collections also
became a part of the holdings of that office, including those of motion
picture film and photography. The systems of storage and retrieval
developed by OCD were 11l1nsnally effective for that time and the Officp
began to gain recognition throughout the intelligence community.
In 1955, OCD \yas renamed the Office of Central Reference to
more accurately reflect its Agency-wide responsibilities. In 1967, OCR
WtlS renamed the Central Reference Service (CRS). Today, ems can
offpr intelligence analysts thronghont the community some of the most
sophisticated information storage and retrieval systems to be fonnd
anywhere in the world.
Photograplu'e Intl'r)iTetation.-CL\'s work with photographic interpretation
began in ] f);j2 and was initially centprl'd in tIl(' Gl'ographic
Hl'sparch Area, ORR. In ]0;jR. a npw Photographic Intelligpncp Cpntl'r
(PIC) \yas created hy fusin~ tIll' Photo Intplligl'ncp Diyision of
OI{R with thp Statistical Branch of OCR. Tllp nl'w Cpnter was gi\'pn
office-It'H'J status and the rpsponsibility for producing photographic
intelJi~encpand prm'iding relatl'<l selTicps for CIA and the rl'st of the
Intelligence Community. In l!H)l PIC \yas fnrtlwr p]p\·atp(] to J)('COI11P
the Xational Photographic Intprpretation ('pntpr (XPIC). This Center
was staffed by fornlPr members of PIC and DL\ ppl'sonnel cIptaile(]
to XPIC: All personnel were functionally lllldPr the Director, XPIC,
who contmuecI to n'I)Ort to the DDT.
An interagency study conducted in 1967 concluded that NPIC's
national intelligence responsibilities had grown so substantially that
departmental imagery analysis requirements were not being adequately
served. Accordingly, the DDI established an Imagery Analysis
Service (lAS) as a separate office of the Directorate to deal exclusively
with the photo intelligence requirements of CIA. In 1973, it was
decided that NPIC would be more appropriately placed in the Directorate
of Science and Technology with other elements dealing with
reconnaissance at the national level.
Translation Services.-The Foreign Documents Division (FDD) of
the Office of Operations (00) had its origin in the Army and Navy's
\Vashington Document Center. Founded in Hl44, it was a repository
for captured Japanese and German records. It was absorbed by the
Central Intelligence Group in 1946 and, during the late forties, evolved
from a repository into an exploiter of all foreign language documents
coming into the community. It joined the Central Intelligence Agency
as part of 00 in the Directorate of Plans. With the transfer of 00
to the Intelligence Directorate in 1952, FDD continued to expand its
work into the field of document exploitation, concentrating increasingly
on materials received from the communist countries. In 1964,
it was separated from 00 to become part of the Office of Central
Reference (OCR). This arrangement lasted only three years, however,
as FDD was transferred again to become part of FBIS in 1967. The
intent of this move was to combine the Directorate's efforts to exploit
foreign media-radio and press-in a single service and to concentrate
its major assets in terms of foreign language capabilities. FDD
remains in FBIS to this day, providing translation services for the
Agency, the community, and to a lesser degree, for the Government
and the general public.
DIRECTORATE OF INTELLIGENCE
[DiRECTOR OF CENTRAL INTELLIGENCE I
DEPUTY DIRECTOR FOR INTQLUGCNCc
etA OpcrntlOt\S
Centc"
CollecllQn Guld,3nctt
.'l.nd
Asscs:::'rncnlS Sl::lll
COMlRe:X.
Stan
ElIccutlVQI
ond
MaNlQcmcnl Sta.lrs
BEJ Ece-r"::"n,c
nc:::.e~~ch
ltT'\UC'I.~ry
An,:,l)'SIS
.scrVI(;:c
orru:cof
Pol.I'C"'l1
ncs\J,lrch
all,cco'
Gcogr.:'l~hICand
C."lfl.::>£;riJP..hlC.
Rcsc.;).tch
~OIe,g"
BrODd~t
InlormabOt\
5<:(.."c:.
011_01
c"""'"
~
B. THE INTELLIGENCE DIRECTORATE TODAY
In FY 1976, the DDI had a relatively small share of the Agency's
budget and personnel. Resources allocated to intelligence production
have represented a relatively steady percentage of the intelligence
budget over the years. Intelligence production is a people-intensive
activity, requiring relatively little in the way of supplies, equipment,
structures, and operational funding. The Intelligence Director
spends approximately 75 percent of its budget on salaries. Of the positions
in the DDI, 74 percent are classified as professional and 26
percent as clerical. Of the total, 54 percent are directly involved in
"intelligence production" (researching data, analyzing information
and writing reports), 28 percent are tasked with "intelligence processing"
(performing reference land retrieval functions, preparing
publications, or providing other support services), and 18% are involved
in "intelligence collection" (monitoring overt foreign radio
broadcasts and publications).4
The most important group of DDI products consists of the daily
intelligence publications, designed "to alert the foreign affairs community
to SIgnificant developments abroad and to analyze specific
problems or broadly-based trends in the international arena." 5 These
include the President's Daily Brief; the National Intelligence Daily,
prepared for Cabinet and sub-Cabinet level consumers; and the National
Intelligence Bulletin, distributed more broadly to the defense
and foreign affairs communities. The DDI issues a number of weekly
periodicals on specialized subjects, prepared in the research offices of
the directorate.
The DDI also produces in-depth and analytical studies on a periodic
or one-time basis. These are monographs on particular problems; some
are DDI-initiated, others respond to specific requests of the policymakers
or their staffs. In addition, DDI analysts usually provide the
bulk of the staff work for the National Intelligence Estimates (NIEs),
which are prepared under the auspices of the National Intelligence
Officers (NIOs).6
The Intelligence Directorate also performs a variety of coordinating
and analytical services in providing intelligence support to policymaking.
Most National Security Council (NSC) meetings begin with
an assessment of the current situation given by the DCI, and prepared
by DDI analysts. The DCI, similarly supported by DDI personnel,
also participates in an array of interagency policy groups (e.g., the 40
Committee, the Senior Review Group, the Washington Special Action
Group, and the Strategic Arms Limitation Talks [SALT] Verification
Panel). The DCI's representatives are involved in lower-level
interdepartmental groups, including geographic area groups, functional
area groups, and ad hoc groups.
Analysts from DDI frequently contribute to the preparation of
National Security Study Memoranda (NSSMs), which are usually
• "The Directorate of Intelligence," p. 4.
•1ma., p. 2.
• lind., p. 2.
207-932 0 - 76 - 1B
266
drafted by interagency groups under the direction of the NSC staff.
Often ~t NSSM: will include an intelligence assessment of the problem
at hand as an annex to the memo itself; this might also be summarized
in the text.
Three examples illustrate how the DDI contributes such intelligence
support. .A SALT support staff has been assembled in CIA to coonlinate
SALT-related activities of production offices in the DDI and
DDS&T. The staff serves as the point of contact to respond to intelligence
requirements generated by the NSC staff, the Verification
Panel, and the U.S. SALT delegation. The staff relies on the analytical
offices of the CIA for substantive intelli,g-ence.
In another case, after the 1973 )Iiddle East war, the DDI was asked
to examine all aspects of possible Sinai withdrawal lines on the basis
of political, military, geographic, and ethnic considerations. Eight
alternative lines were prepared for the Sinai, a number of which Secretal"
y of State Henry Kissinger used in mediating the negotiations
between Egypt and Israel.
Finally, the DDI provided assessments to the policy groups who
prepared U.S. positions for the Law of the Sea Conference in 1975,
including descriptions of the strategic straits under discussion, analysis
of each country's undersea mineral resources, and information
about political positions the participating countries would be likely
to take!
TIlE ISSUES
The Select Committee began its examination of intelligence production
by considering the relationship between intelligence and policYl
and the limits of intelligence. These considerations served to highlight
cel1:ain problems in production which the Committee feels deserve further
attention by both the executive branch and congressional oversight
bodies. These problems bear on the key issues of quality, timeliness
and relevance of finished intelligence. They derive in large part
from the nature of presidential leadership and the particular emphasis
and preoccupations of successive Directors of Central Intelligence. In
the past, the national leadership has used the CIA more for operational
purposes than for its analytic capabilities. Other concerns derive from
the structure of the analytical personnel system, the intelligence culture
and the nature of the intelligence process, the overload of the system,
the preoccupation with current events, and the lack of sufficient
quality control and consumer guidance and evaluation.
C. THE RELATIOXSIIIP BETWEEX INTELLIGENCE AND POLICY
The relationship between intelligence and policy is a delicate and
carefully balanced one. One witness told the Select Committee that
there is a "natural tension" between the bvo and that
if the policy-intelligence relationship is to work, there must
be mutual respect, trust, civility, and also a certain distance.
Intelligence people must provide honest and best judgments
and avoid intrusion on decisionmaking or attempts to influence
it. Policymakers must assume the integrity of the intelli-
7 Staff summary of briefing given by Edward Proctor (DDI), 4/24/75.
267
gence provided and avoid attempts to get materials suited to
their tastes.8
In recent years there has been a tendency on the part of high officials,
including Presidents and Secretaries of State, to call for both raw
reporting and finished intelligence to flow upwards through separate
channels, rather than through a centralized analytical comp'onen~.This
has resulted in many cases in consumers doing the work of mtelh~ence
analysts. Presidents and Secretaries of State have ~ll to.o often relIshed
the role of "crisis managers", moving from one senous Issue to another
and sacrificing analysis and considered judgment in the pressure. of
events. In between crises, their attention is turned to other pressmg
matters, and careful long-range analysis tends to be set aside.
By circumventing the available analytical pl'Ocess, the cons,umers
of intelligence may not only be depriving themsel\'es of the skIll~ of
intelligence professionals; they may also be sacrificing necessary tIme
and useful objectivity. In making his own intelligence judgment based
on the large volume of often conflicting reports and undigested 1':'1w
intelligence instead of on a well-considered finished piece of mtelligence
analysis, a high official may be seeking conclusions more
favorable to his policy preferences than the situation may in fact
warrant.
The essential questions about the intelligence product concern its
usefulness to the policymakers for whom it is intended. Does intelligence
address the right questions? Does it deliver the kinds of information
and insights policymakers need in order to make foreign policy
decisions? Is it timely? Is it presented and disseminated in the manner
and format most useful to the consumers? 'Will they read it in other
than crisis situations? The answers to these questions are by no means
simple. Still, the Select Committee- believes they are deserving of
examination-and periodic reexamination-in the interests of maintaining
an effective intelligence service.
While intelligence analysts have a very good record in the area
of technical assessment (e.g., hard data on foreign military hardware),
the record is weaker in qualitative judgments, trend forecasti~
lg, a~d political estimatin~. 'Vhile analysts may be able to
furmsh faIrly complete and rehable reporting on tangible factors
such as numbers and make-up of Soviet strategic missile forces, they
ar~ n?t as good at assessing such intangibles as why the Soviets are
bUIldmg such a force. The problem pertains to other issues, too, for
example, in analyzing the likely negotiating stance of a particular
country in economic negotiations of interest to the 1.Tnited States.
In particular, sOJ.lle policymakers feel that intelligence analysts
have not been especIally helpful to policvmakers on the more subtle
questions of political. eC?IlOmic, and military intentions of foreign
groups and .leaders. The VIeW fro~ the ton is. of course, very different
from the VIew held by analysts III the departments and agencies or
I~ the field. Too often analysts are not willin!! to address suchquestions
~Iredlv. A.na!vsts tend to believe that policymakers want answers
mstead of. IllsIghts. Some consumers argue that intelligence analysts
lack suffiCIent awareness of the real nature of the national security
8 John Huizpnga te~timony, 1/26/76, p. 14.
decisionmaking process-how it really 'Yorks, where and how intelligence
fits in, and what kinds of information are importanU
On the other hand, the Select Committee is concerned that analysts
are not always kept sufficiently informed, in a timely fashion, of U.S.
policies and activities which affect their analyses and estimates. The
Committee is concerned that the secrecy and compartmentation surrounding
security policy decisionmaking affects the relevance and
quality of intelligence analysis. The analysts in the DDI may not
always be aware of ,,-hat a key foreign leader has told high-level
American policymakers in private, and so they may be missing crucial
information on a particular nation's intentions in a given situation.
The Select Committee's study of covert action has revealed that on
a number of occasions in the past intelligence analysts were not told
what U.S. covert operators ,,-ere doing abroad, an omission which could
seriously affect the accuracy of intelligence assessments. Likewise,
because of security compartmentation, DDI analysts sometimes did
not know about particular U.S. strategic weapons R&D programs,
and so were not able to assess completely the reasons for countermeasures
that were being taken in the development of Soviet strategic
forces. .
D. THE LnIITs OE' IXTELLIGEXCE
Clearly what is needed is a realistic understanding by both producers
and eonsumers about the limits of intelligence: what it can
and cannot do. As a former senior analyst explained to the Select
Committee,'O what intelligence can do is to follow the behavior of
foreign leaders and groups o,-er a long prriod of time in order to get
a sense of the parameters within ,,-hich their policies move. American
policymakers are not then likely to be greatly surprised by foreign
behavior even though intelligence analysts might not be able to predict
rrecise intentions at any giwn moment with respect to a given situahon.
Nor can analysts be expected to predict human ewnts when often
the actors themselves do not know in advance what they will do. As
the Schlesinger Report said:
In a world of perfect information, there would be no uncertainties
about the present and future intentions, capabilities,
and activities of foreign powers. Information, however,
is bound to be imperfect for the most part. Consequently,
the intelligrnce community can at best reduce the uncertainties
and constmet plausible hypotheses about these
factors on the basis of what continues to be partial and often
conflicting evidence.l1
~o expect more may be to cOllrt (lisappointnlPnt. Despite this reco~ni!
lOn on the part of many poJicvmakers, if anal:vsis is not correct, there
IS often the charge of an "intelligencp failnre." Good inte11igence or accurate
pn'dietions cannot insure ag'ainst bad policy, in any case. For
example, as the current Drputy Director for Intplligcncr maintains,
the pessimistiC' CIA estimatps on Viptnam had little or no effpct on
U.S. policy decisions then'. Vietnam Illay han bern a policy failure.
9 Staff summary of An(lr('w Marshall int('rri('\\,. 2/10/76.
'0 Huizenga, 1/26/76, p. 24.
11 Schlesinger Report, p. lOa.
269
It was not an intelligence failure. 12 Similarly, the United States hall
intelligence on the possibility of a Turkish invasion of Cyprus in
1974. The problem of taking effective action to prevent such an invasion
was a policy question and not an intelligence failure.
E. TIlE PERSO:SNEL SYSTEM
To some extent, problems in the quality of the analytical performance
of the intelligl:'nce community are simply in the nature of things.
The collection function lends itself to technical and managerial approaches,
"'hile the analytical job is more dependent on the intangibles
of brainpower. In the final analysis, the intelligence product can only
be as good as the people who produce it.
The CIA prides itself on the qualifications of its analysts. The
Agency's exemption from Civil Service constraints-unlike the DIA,
for example-has enabled the DDI to attract the best analysts in the
community. Nevertlwless, those in the highest positions in the CIA
have traditionally come from the operations side of the Agency.
The Agency's promotion system is structured ill such a way that the
most outstanding lower-level people are singled out for advancement
into managerial positions. Such a system "'orks well for the purposes
of the Directorate of Operations (DDO), where the skills nece~sary
for good management are essentially the same as those required of a
good case officer, But when applied to the DDI, that system encourages
the best analysts to assume supervisory positions, reducing the time
available to utili7.e their analvtical skills.
Although the CIA has several hundred "supergrade" positions H_
and very few government agencies are permitted so high a numberthere
are virtna.llv no "supergrade" slots which involve only, or even
primarily, analytic responsibilities. The Agency maintains that DDI
supervisors are indeed analysts, since they review and critique the
work of junior analysts. In this view, superTisory positions amplify
the analytical capabilities of senior personnel. Thus, there is not
"supervision" in tlH' usuul s('nse by DDI supervisors; they are viewed
as participants in tIl(' analytical process.14
The Office of Natioual Estimates was the only place where a regular
ItlTllngf'ment for high-Ie\'el analysts existed, but that office was
abolished in 1973. Todav only the DDl's Office of Political Research
(OPR) has been able to 'retaill several supergrade staffers who do only
analysis (ont of a staff of abont 4() to ,rj() analysts.) The OPR, created
only in UI74, is treated by the nDT as an elitf' group. Much of its ,york
is intf'l'disciplinary in natnre. The emphasis is placed on keeping:
OPR analysts out of the everyday rontine of requests for current
intelligence work which can he performed bv other offices in the
directorate.15 .
Some analysts complain that the pf'rsonnel system has fostered too
much bureaucratic "layering." and that there are too many people
writing reports about reports. The effects are predictable. In the words
of former' DCI and Secretary of Df'fense .Tames Schlesinger, "If you've
12 Staff f'lummary of Edward Proctor interview, 5/16/75.
13 John Clarke testimony, 2/4/76, p. 37.
H Proctor (Staff !'ummary), 3/1/76.
"'Ibid.
270
got too much specialization and pigeonholing of people, you get the
kind of people lJl the intelligence game who don t mind being pigeonholed,
and the entire U.S. intellIgence establishment is too much
bureaucratized." 11; The Intelligence Community (IC) staff, in its
post-mortems of major U.S. intelligence failUl'es, has pointed in all
cases to the shortage of talented personnel. As the fonner deputy head
of the IC staff pomted out to the Select Committee in his testimony,
"giving people more flexibility in pay scale and so forth doesn't always
guarantee that they hire the right people." 17
F. RECRUITMENT AXD TRAINING OF ANALYSTS
The Agency tends to bring analysts in early in their professional
life, emphasizing lifetime careers in intelligence work and the development
of institutional commitment. There has traditionally been
minimal lateral entry of established analysts and experts into the
profession at middle and upper levels (more in DDS&T than in
DD1.) 18 This might be characterized as the "craft guild" approach
to intelligence, 'where recruits are brought in to serve their apprenticeships
within the ranks of the profession.10
Specialized analytical training for intelligence analysts is qu~te
limited. The CIA's Office of Training (OTR) has a program III
methodology and research techniques and a variety of mid-career
courses and senior seminars. About 25% of the DDI personnel who
receive in-house training are in management and executive developmerut
courses. Various DDl offices sponsor courses on speoific skills
such 'a,.<; computers and st'atistics.20 For the most part in the past ,the
Agency-run courses available were oriented toward developing skills
necessary for clandestine p.divity. According to Dr. Schlesinger:
Within the CIA, most of the training effort in the past has
gone into training operators rather than training analysts.21
The Agency maintains there is now an increased emphasis on the
development of sophisticated analytical skills and nnderstanding.
Most of the suhst'antive training for intBlligence analysts Itake.q place
outside the Agency, bath in academic ,institutions and in other government
departments. Of the total number of DDl personnel par1ticipa'ting
in such ex,ternal training in FY 1975, about one quarter were
involved in training courses longer than 6 weeks in duration.
G. THE INTELUGEXCE CULTL"HE AND AXALYTICAL BIAS
There isa set of problems stemming from whaJt might be called
rtheintelligence "cul'ture"---'a particular outlook sometimes attributed
to the analysts which tends to affect the overall quality of judgment
reflected in their work. Although the problem of preconcelJ<tions is
one of the most intractable in intelligence analysis, it clearly is one
16 James Schlesinj:(er testimony, 2/2/76, p. 72.
17 Clarke, 2/5/76, p. 31'.
18 In FY 1975. 18 analysts out of 105 hired from outside the CIA by the DDI
were at OS-12 to 15.
1. ~Iarshall (Staff summary). 2/101711
2<l Proctor (Staff IiHlmmary), 3/1/76.
21 Schlesinger, 2/2176, p. 27.
271
of the most critical, and has been a focal point of the IC staff postmortems.
As one former senior official told the Select Committee, "By
and large, good intelligence product-ion should be as free as possible
from ideological biases, and the higher the degree of ideological bias,
the greater will be the blind spots." 22
Among the examples of analytical/intellectual bias and preconceptionsare
the following: In 1962, some CIA analysts judged that the
Soviets would not put missiles into Cuba because such a move would
be "aberrationa1." 23 In 1973 most of the intelligence commlmi'ty 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 capabilities.24
The same mechanism operated-the inability to foresee critical
e\'en1ts, in the face of mounting evidence, to the contrary-during the
Cyprus crisis in the summer of 1974. According to the IC Staff postmortem
of ,that episode, the CIA analysts were again prey to:
the perhaps subconscious conviction (and hope) that, ultimately,
reason and rationality will prevail, that apparently
irrational moves (the Arab attack, the Greek-sponsored roup)
will not be made by essentially rational men.25
The charge is frequently made that intelligence estimates issued by
the Defense Depar1:ment and the military sen-ices are not wholly objective,
since those groups have particular depar1:mental interests and
programs to acl\·ocate. By contrast, the CIA is supposed to be free
from such bias. But although the DDI is not in the position of having
to defend budgetary items or particular weapons systems, in the view
of o~:her parts of the intelligence community, there has been a tendency
for a CIA institutional bias to develop over time. The Committee notes
that some observers have pointed to a CIA "line" on certain issues.25a
H. THE NATURE OF THE PRODUCTION PROCESS: CONSENSUS VERSUS
COMl'ETITIO~
The nature of the production process can itself undel'mine the
quality of the produet. That process is consensus-oriented, varying in
degree from the formal United States Intelligence Board (USIB)
coonlina:t,ion invoh-ed in producing aNational Intelligence Estimate 26
to the less stI'llctur-ed daily analyst-to-analyst coordination, which
take.'3 place at the working level. For the monographs produced on an
irregular basis by the Intelligence DirecJtorate's research offices, the
bulk of the coordination effort is between these offices, although occasionally
such coordination will cross directorate lines. land less frequently
it will involve going outside the Agency. An analyst from the
DDI may meet with his opposite numbers in State or DIA prior to
'2 Ibid.
22 Huizenga. 1/26/76, p. 25.
.. Ie Stat'( post-mortem on 1973 Middle East war (January 1974), p. 14.
., Ie Staff post-mortem on 1974 Cyprus crisis, p. iv.
25a See Chapter V, pp. 76-77.
c" Prior to the President's February 1976 reorganization of the intelligence
community, thl' USIB approved all National Intelligence Estimates. See the chapter
of this report on "The Director of Central Intelligence" (pp. 74 ff.) for a
fuller d;scussion of the estimates coordination process.
272
publishing llil article in their mutual field. 21 Th.e coordination pr0re3S,
however necessary and desirable, may tend to produce a "reinforcing
consensus," whereby divergent views of individual analysts can become
"submerged in a sea of conventional collective wIsdom," arid
d~ubts ~r disagreements can simply disappear in the face of mutually
remforcmg agreements.28
Although the purpose of coordination is "to assure that the facts
nnd judgments presented tlwrein arc as comprehensive, objective, and
accurate as possible," 29 it sometimes has the unfortunate side-effect
of blurring both the form and content of the product. The NIEs
have been criticized, on occasion, for this. The estimates undergo the
most formal coordination process, one which is integral to policy consensus-
building. Some consumers complain that finished intelligence
frequently lacks clarity, especially clarity of judgment, and that it is
often presenteel in waffly or "delphic" forms, \vithout attribution of
views. Opposing views are not ahvays clearly articulated.•Judgments
on (lifficult subjects are sometimes hedged. or represent the outcome of
compromise, and are couched in fuzzy. imprecise terms. Yet intelligence
consumers increasingly maintain that they want a more clearly
spelled out distinction between different interpretations, \vith judgments
as to relative probabilities.
In fact, the issue of consensus versus competition in analysis represents
a persistent conceptual dilemma for thr intelligence community.
Policymakers tend to want one "answer" to an intelligence question,
but at thr same time they do not want anything to be hidden from
thrm. Consumer needs can change drastically in a short period of time,
and the same policymakrrs may need different kinds of intelligence for
different kinds of situations.
Some memhrrs of the intelligence and foreign policy communities
today argue that the consrnsus approach to intplligrncc production
has improperly comp to substitutp for competing centers of analysis
which could deliver more and diffrrent internretations on the critical
Iluestions on which only partial data is availa-ble. This conceptual conflict
should be closely examinrd by the successor oversight committee.
I. TIlE "CURREXT EYEXTS" SYNDROME
The task of producing current intelligf'nce-analyzing day-to-day
eTents for quick dissemination-today occu~ies much of the resources
of the DDI. Responding to the growing demands for information of
currf'nt concern by policymakers for more coverage of more topics, the
DDI has of nf'cessity resorted to a "currrnt evrnts" anproach to much
of its research. The;e is less interf'st in and fewer resol1rces have been
devoted to in-depth analysis of problems with long-rang-e importance
to policymakers. The Directorate has had to devote considerable resources
in order to keep up on a dav-to-day basis with events as ~hey
happen. To some extent. analysts fer! they must compete for tImeliness
with the considl'rahle amount of raw reporting which reaches
consumers,
27 "The Dirpctoratp of Intpllig-encp." Annex A, p. 2.
2Il Ie fHaff po~t-mortpm on the 1973 :\1iddle-Ea~t War. p. 18.
29 "The Directorate of Intelligence." Annex A, p. 1.
273
According to some o1>seners, this syn(lrome has had an unfavorable
impact on the quality of crisis warning and the recognition of longer
term trends. The "current events" approach has fostered thl' problem
of "incremental analysis," the tendency to focus myopically on the
latest piece of information without systematic consideration of an
accumulated body of integrated evidence. Analysts in their haste to
compile the day's traffic, tend to lose sight of underlying factors and
relationships.30
For example, the 1966 Cunningham Report points out that the
CIA's sinologists \vere so immersed in the large volume of daily
FlUS 31 and other source reports on Communist China in the early
1960s that they failed to consider adequately the broader question of
the slowly developing Sino-Sm'iet dispute.32
The Intelligence Directorate is now turning more attention to
such increasingly important long-term (and inter-disciplinary) problems
as world food balances, raw material supplies, population pressures
and pollution of the environment. Ne\'ertheless, the nnI itself
feels that an e\'en greater effort should be made in these areas. "Such
matters have not been the focus of national security interest in the
past, but they clearly will be within the next ten years and this Directorate
should be building its capacity to analyze and report in these
fields." 33
J. INNOVATION
The CIA is thought by many observers to be technologically one
of the most innovative research centers in the country, and it allocates
considerable funds to continue the search for new technology. But
despite recent increases, the intelligenee community still expends relatively
little effort on R&n in the analytical field-in contrast to intensive
effort in new and costly collection methods.
The analytic community has suffered from the secrecy that surrounds
the work of the intelligence community as a whole. This
insulation is recognized to have had a detrimental effed on the quality
of analysis. The Agency recognizes the need for condueting a free
exchange with academics, contractors, and eOllsultants. For example,
in FY 1976, 17 analysts were on leave at private institutions with
an additional 14 ppople in various Government programs (e.g., the
State Department senior seminar, or the Congressional Fellows
program) .34
Some nnI offiees lIave panels of eonsultants (outsiders) to review
major papers, and outside speakers are on oceasion brought in for
speeial seminars. There have been efforts like the one made by OPR
to arrange for onr-year sabbatieals for visiting academics during
which the visitor eould produee both government and public papers.
Such efforts have been only partially successful.
30 See IC staff pORt-mortemR on Middle EaRt war and CypmR crisis.
31 The Forei~n Rrondcnst Information Senice. run hy the Intelligence Directorah'.
monitors foreign media and open source material and puhlishes daily
RUryeVS hy area.
32 CIA Inspector General. "Forei~n Intelligence Collection Requirements,"
Decemher 1966 (The Cunninghnm Report). pp. VII-13, 14.
33 "The Directorate of Intelligence," p. 12.
M Proctor (Staff summary), 3/1/76.
274
The question of CIA relations with academics and private groups
like foundations and research organizations is a controversial one.35
The Committee nows the desirability of a more open attitude on both
sides, one which both recognizes the legitimacy of the analytic work
of the intelligence community and refrains from the secret use of
academics and others for operational purposes.
K. OVERLOAD ON ANALYSTS AND CONSUMERS
Few observers would dispute the fact that as consumer demands
have grown and the amount of data collected has burgeoned, the
analysts' work load has become a serious problem. But Wn years ago
the Cunningham Report expressed the concern that:
In the long run it is not the crude question of work load which
matters most, nor even the point that each item uses up customers'
time and atwntion which cannot be given to any other
item, so that each of our products must receive steadily less.
·What matters most is the question whether this quantity of
information is degrading the quality of all our work. 36
And the 1971 Schlesinger Report said that it was "not at all clear that
our hypotheses about foreign intentions, capabilities, and activities
have improved commensurately in scope and quality as more data
comes in from modern collection methods." 37
Yet today the intelligence establishment remains structured in such
a way that collection guides production, rather than vice versa; available
data and "the impetus of technology" tend to govern what is
produced.38 To be sure. much of the proliferation in data collected has
proven invaluable to the analytic effort. Technical collection systems
have provided "hard" data, e.g., on missile silos which have contributed
to the generally acknowledged high quality of CIA assessments
of Soviet and Chinese strategic forces.
In 1971, thE' Schlesinger Report said, "It has become commonplace
to translate product criticism into demands for enlarged collection
efforts. Seldom does anyone ask if a fmther reduction in uncrrtainty,
however small, is worth its cost." 39 The community's heavy emphasis
on collertion is itself cletrimental to correcting product problems, said
the report. for each department or agency sees the mainwnance and
expansion of collection ea rabil itirs as the route to survival and strength
,,-ithin the community. Thrre is a "strong presumption" that additional
data collection rather than improve(l analysis will provide answers to
particular intelligence problems.40
Analysts naturally attempt to read all the rrlevant raw cl.nta reports
on the snbjects they are working on, for fear of missing an important
piece of information. The Cnnningham Report referrecl to this as the
" SE'f' ChRntf'r X of thi~ rE'port on thE' CIA'~ relation~ with these groups in support
of intelligence collection and covert action.
"" f'unningham Report, p. VIIJ-13.
31 S("hle~in!!erReport, p. lOa.
3ll Ihid.. p. lOa.
39 I"hid .. p. 11.
00 Ibid., p. 11.
"jigsaw theory" of intelligence-that one little scrap might be the
missing piece.41 The present trend within the DDI is to reduce the
amount of raw data coming to analysts by more effective screening
processes.
In the opinion of one intelligence community official, analysts in
the future are going to have to rely to a greater extent than heretofore
on others' judgments. The collectors themselves may have to
present their output in summary form, with some means of highlighting
important inforlllation,42 despite the community's sensitivity to
the distinction between "raw" and "finished" intelligence reporting.
On the other hand, consumers tend to treat the intelligence product
as a free good. Instead of articulating priorities, they demand information
about everything, and the demand exceeds the supply. And
analysts, perhaps for fear of being accused of an "intelligence failure,"
feel that they have to cover every possible topic, with little regard
for its relevance to U.S. foreign policy interests. The community
must part with the notion that it has to beat the newspapers in reporting
coups in remote areas of the world if what happens in those
areas is only of marginal interest to U.S. policymakers. In this regard,
there are serious efforts being made by DDI to focus analysis on major
areas of importance to the United States.
The community has looked increasingly to the advent of automated
information-handling systems to solve the problems of systems
overload, but the impact of computerization is not yet clear. In 1966
the Cunningham Report warned that "great technological advances
in storage and retrieval" of information can do mOre harm than good
if "drastically higher standards" for what is to be stored and retrieved
are not instituted}'
It has often been pointed out that not only are analysts swamped
with information, but the consumers also are inundated ,vith intelligence
reporting, both "finished" and "raw." The volume of papPI'
degrades the overall effectiveness of the pJ'oduct, since there is simply
too much to read, from too many sources. In addition to the daily DDI
publications and the various DDI Offices' specialized weeklies and
other memoranda, a variety of other intelligence publications, regularly
cross the desks of seJiior GoveI'llment officials. As former DCI
Richard Helms has told the Selpet Committee:
It seems to me that one of the things that's tended to happen
is that almost every agency has got to have its national publication.
In other words, it's got to have a publication that
arrives in the White House every mOI'lling.44
Policymakers receive DIA's Defense Intelligence Notices (DINs),
produced on particular snbjeets as the occasion demands-sometimes
several per day on a given topic. NSA semIs out a daily SIGINT Summary,
which is not classed as finished intelligence. And a consid-
"Cunnin/{ham Report. p. VII-19.
•• Rtllff ~ummllrv of Ri('Iwrn Rhr:vo('k interview. 2/10/76.
.. f'llnninl!"ham Rf'Port, P. VJI-12 (footnote) .
.. Richard Helms testimony, 1/30/76, p. 29.
276
erable amount of raw reporting of clandestine human source intelligence
is routinely distributed to consumers on the NSC staff, at the
Departments of State and Defense, and in the military services.
This glut of paper raisE's a number of issues which the Select Committee
feels deserve further attention. The proliferation of departmental
publications tends to undermine the centralized nature of the
system for the production of national intelligence. It contributes to
confusion rather than clarity in the decisionmaking process, since
different publications often present different conclusions. Often the
reasons for the differences are only clear to a sophisticated intplligence
analyst. And direct reporting from the collectors usually arrives before
the analytical reporting can, preempting the analysts' work in
evaluating the data.
L. QUALITY CONTROL
In 1972 a "Product Review Division" (PRD) was established
within the IC Staff. It has the task of rE'gularly appraising intelligence
articles and studies, "testing them for objectivity, balance, and responsiveness."
45 TIl(' Intelligence Directorate has no formal or independent
system for quality control, depending instead upon its regular review
and coordination process.46
~fost of PRD's attention to date has been directed to the conduct of
communitywide post-mortems on particular crises-for example, the
1973 Middle East war. the Cyprus crisis in 1974. the Indian nuclear
(letonation. and the Mayaguez incident. The Division was involved in
changing the old daily Oentral Intelligence Bulletin from a CIA publication
into a community publication (now called the National Intelligence
Bulletin). PRD participated in discussions leading to the
transformation of the old 'Vatch Committee into the DCI's Special
Assistant for 'Warning, ,,-ith a Strategic 'Warning Staff.
PRD has not yet been significantlv involved in the development of
new analyticalmcthods, in resource allocation for production elements,
or in training or recruitment issues. ('ontact with the consumers of
the intelligence product has been on an irregular basis (mostly for
post-mortems), although PRD is currentlv at work, through the NIOs,
collecting consumer reactions on particular papers of concern to the
FSIB.
The Division has no authority to order changes in the management
of production which might affect the quality of the product; rather,
it has been in the position of making recommendations to the USIB
and encouraging their implementation.
M. COXSU)lER GrIDAXCE AND EVALUATION
The DDI manages its production plannin.,! by compiling a Quartprly
Production and a Quarterly Research Schedule, outlining those
finished intelligence studies slated for publication in the following
three months as well as projects which support other intelligence
efforts. but which may not be published. The quarterly schedules are
prepared by DDI's Executive Staff based on inputs received from
'" Rhryock (staff summary), 2/10/76.
'" Proctor (staff summary), 3/1/76.
277
each office within the Intplligrnce Dil'rctoratr, and tIl(' .\ssociatr DDI
reviews them to ensure that tIl(' planned projects are responsiye to
con"unwr nerds.47
'Yhile tlwre is no formal or in~titntionalizNIrr\'iew bv COllsnllwrs of
the qual'trrly s(']ll'(l\lles, there al'e freqtll'nt Dirpctol'Ht~-lp\'pl contacts
with policymakers who pxpress an intpl'est in intelligellC'p infol'mation
anll assrssnwnts on pal'ticnlar forpi,!.m policy issues.
Eyaluation of the intrlligence prolluet bv the conSllIllers tlll'mseln's
is virtually nonexistent. TI;e KSC Intelligence Committee, which was
supposed to IwrfOl'lll that functioll, \nts largply inaetin> allll has now
bern aholished in tIll' President's n'organization plan. Harely, if ryer,
do high oRicials take the time to re\'ie\y the product carefnlly \\'ith the
analysts aIHI explain to them hmy tIll' product could ]w impro\'rd and
made more \lseful to policymakers. The intelligence community, tlwn,
by defanlt, pyaluates its O\\'n prrfonnancr \yithont the 1)('lIefit of any
real feedback. One former seniOl' analyst tolll the Splpct Committee:
I belieye t!trrr ought to he rpq\liremrnts on thp policy side to
rpspolHl by COJllmpnt or othrl'\\'ise to major intelligence products,
oln-iously not thp wholr [low of stuff, and I think that
there ought to he a responsibility at an appropriate Ipyel. say
at an Assistant Spcrptarv Ipyel. to do this, and at the XSC
le\·pl. This kind of recogliition, thr sense of participation in a
serious procpss is, I think, the best thing that can IX' done for
analysts.48
X. TIlE COXGHESSlOXAI~ ROLE
Congress does not at prpsent recpi \"P National Intell igence Estimates,
although some of the pstimati\'r material is presentplI to the
Congress in occasional hriefings bv intplligence officials. In thp past,
the Senatr Foreign Relations and Arnwd Sen-ices Committres received
the National Intelligence Daily, which could be cut off at executiYe
will, and has been on some occasions, most recrntly in .January
1976.49 In 197f), the DDI began publishing a daily Intelliaell(,(, Oheckli8t
specifically tailored to what it prrcri \'ed to be the intelligpnce nepds
of the Congress.
With the resurgence of an active congressional role in the foreign
and national security policymaking PI'OCPSS comps the nped for members
to recpive high oualitv. reliablr, allll timply information on which
to base congressional decisions and actions. Access to the bpst a\'ailable
intelligence product should bp insisted upon by the legislative
branch. Precisely what kiIHls of intelligence the Congress requires to
bptter perform its constitutional responsibilities rpmains to be worked
out brtween the two branches of government, but the Select Committee
believes that the need for information and the right to it is clear.
• 7 "The Directorate of Intellig-ence." p. 8.
.. Huizeng-a, 1/26/76. p. 23.
•• Laurence Stem, "CIA Stops Sending Daily Report to Hill," Washington
Po,~t, 2/4/76.
 

Go to Next Page