INTELLIGENCE REQUIREMENTS: AN ORGANIZATIONAL PROCESS PROBLEM

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP83M00171R002300070002-4
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 45
• Document Creation Date: December 12, 2016
• Document Release Date: April 23, 2002
• Sequence Number: 2
• Publication Date: August 28, 1974
• Content Type: MF

-'Approved For Release 20021051S GKTRDP83MOO171 R00230007
Return fa 14, Reg
B-111259/74
Copy
2 8 AUG 1974
MEMORANDUM FOR: General Graham
SUBJECT : Intelligence Requirements: An Organizational
Process Problem
1. On 21 August 1974 you requested that I review the
attached material and provide you with
a. an analysis of the "requirements problem," and
b. a recommended course of action to fix the
problem. This was to be accomplished prior to your departure
as D/DCI/IC.
2. Based upon the information provided to me and
discussions I have had with others on the IC Staff who have
authored papers on the subject, what follows is my reaction
to the problem of rationalizing the requirements, collection
tasking and production processes of the Intelligence Community
in a more systematic way.
3. In all of the material I have read, I am most
influenced by the arguments outlined in the IC Staff study
of August 1973 entitled The US1B Committee Survey Task Group
Report. The Task Group found that:
"The issuance of collection 'requirements' and
'guidance' by most committees tends to be an ineffective
use of committee time and resources for the following
reasons:
This document may be downgraded
SECRET when Attachment 1 is
detached.
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
SECRET
 Approved For Release 2002/05/01$iGWP83M00171 R002300070002-4
B-111259/7
a. Such requirements are generally uncoordinated
with respect to any integrated collection guidance system or
objective strategy.
b. Most substantive intelligence requirements
lack a common prioritization base and/or have no prioritization
indication at all.
c. For the most part, these requirements do not
actually drive collection programs and are not structured
for collector performance evaluation.
d. They are often viewed as merely necessary
outputs to show that 'collection requirements' exist,
although it is unofficially recognized that they have little
or no operative impact upon collection system decision
behavior and actions."
In addition to these conclusions, I am persuaded by a series
of DCI Letters of Instruction to USIB Committee Chairmen
that have been issued over the past several weeks. In each
instance, the DCI has called for the USIB Committee Chairman
to pay close attention to the requirements problem and to
work closely with the D/DCI/IC and D/DCI/NIO in resolving
requirements issues.
4. These documents led me to conclude the following:
a. The USIB Committee Study goes to the heart of
the matter and m fix the problem in the long run;
b. The individual DCI letters of Instruction
appear to be a piecemeal approach and postpones "fixing" the
problem until some future date.
5. I do not intend to summarize or repeat arguments or
proposals m .ae by otters ww have stuuiec this problem, but
rather to recommend some affirmative specific action, that
in my view, would remedy the problem as it has been character-
ized.
6. In making this recommendation to you, I am imposing
the following criterions:
This document may be downgraded
SECRET when Attachment 3 is
detached.
Approved For Release 2002/05/01 : OA-RDP83M00171 R002300070002-4
SECRET
 SECRET
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
B-111259/74
1A
1A
a. ensure that the design for the Committee
structure is simple enough to be understood by all;
b. ensure that the structure is flexiable enough
to be changed;
c. ensure that the structure is responsive to the
DCI's needs, as reflected by the ICI?'s; and
d. ensure that the design is such that one can
evaluate it at the end of a two year period.
7. Recommend the following actions be taken:
a. Adopt Alternative A of the USIB Committee
Survey Task Group Report of August 1973 for reasons outlined
in the document on page 10 (see attachment 1).
b. Coordinate with the D/DCI/NIO, memorandum for
USIB Principals, which would implement Alternative A above,
entitled Establishment of USIB Committees to Support the
NIA's (see attachment 2).
8. This is another view of a complex Issue. I hope it
is of some help to you.
Captain, U A
DCI/IC/CS
Attachments:
1. USIB Committee Report
2. Draft Memo to USIB
3. Background Package on Requirements
DCI/IC/CS/S (28 Aug 74)
Distribution:
1 - Addee
2 - AD/DCI/IC (hold)
/3 - IC/Registry
4- CS Subj.
5 -, CS Chrono
6 - OChrono
25X1A
This document may be down-
graded SECRET when Attachment
1 is detached.
3
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
SECRET
 Approved (D Release 2002/05/01 : CIA-RDP83M0ID 1 R002300070002-4
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/Q5/b4- M00171 R002300070002-4
DRAFT
28 August 1974
MEi IORANDUM FOR USIB PRINCIPALS
SUBJECT: Establishment of USIB Committees to Support the NIOs
1. During the several months since I established the NIOs,
there have been some disconnects in actions the NIOs have taken
because of the lack of a USIB structure to deal with the subject.
The result has been that several of the NIOs have created informal
"USIB" committees.
2. I propose to formalize these efforts and to establish
ten new USIB committees to assist the NIOs in the coordination
of Community substantive and operational matters. These com-
mittees will serve two prime purposes:
a. They will insure a collaborative approach to
problems.
b. They will keep USIB principals informed of
Community actions in the NIO area.
3. I propose creation of a USIB committee for each geographic
area covered by an NIO. These committees would be chaired by the
appropriate NIO with membership to consist of one representative
Approved For Release 2d02(O5/01 aPA!,rDP83M00 J
_71R0023000 i r C L PY__ _
700 ex 21 11
 Approved For Release 2002/05/01: CIA-RDP83M00171 R002300070002-4
from CIA, DIA, INR, and IC Staff,plus military service representation
as recommended by the Director, DIA. Each committee will have only
a two year charter to operate, continuation subject to my approval.
Separate committees would be established for:
Soviet Union/Eastern Europe
PRC
Southeast Asia
Western Europe
Middle East
0
4. I further propose to have a USIB committee for each functional
subject for which an NIO has been named, as follows:
a. Economic Intelligence Committee
To remain as presently constituted, with the addition
of the NIO for Economics as chairman and an IC Staff officer
as member.
b. Energy Committee
To be chaired by the NIO for Energy, with membership
to consist of one representative from CIA, DIA, INR, Treasury
and IC Staff.
2 -
Approved For Release 2' 0 Y0010 ;:' C-fA2 DP83M00171 R002300070002-4
 Approved For Release 2002/05/01 CIA-RDP83M00171 R002300070002-4
c. Strategic Programs Committee
To be chaired by the NIO for Strategic Programs with
membership to consist of one representative from CIA, DIA,
INR, AEC, Army, Navy and Air Force. In addition, the chairmen
of DMAIC, JAEIC, SIC and IC Staff will be ex officio members.
. Conventional Forces Committee
To be chaired by the NIO for Conventional Forces, with
membership to consist of one representative from DIA, CIA, NSA,
INR, and IC Staff and sub-committee chairmen from Army, Navy.
and Air Force. The DIA representative will be the Deputy
Chairman. Three sub-coimrittees are proposed as follows :
(1) General Purpose Ground Forces Sub-Committee
Chairman to be provided by the ACSI, Army, with membership
to consist of one representative from CIA, DIA, NSA, NPIC
and IC Staff.
(2) General Purpose Naval Forces Sub-Committee
Chairman to be provided by the DNI, with membership to
consist of one representative from CIA, DIA, NSA, NPIC
and IC Staff.
Approved For Release 200 5iO1 '.~CtA=FfDP'83M00171R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
(3) General Purpose Air Forces Sub-Coirrnittee
Chairman to be provided by the ACSI, Air Force, with
membership to consist of one representative from CIA,
DIA, NSA, NPIC and IC Staff.
5.' I would like to have your comments on this proposal by
W. E. Colby
Chairman, USIB
Approved For Release 20 /05/01: CIA-RDP83M00171 R002300070002-4
 Approve Release 2002/05/01 : CIA-RDP83M071 R002300070002-4
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 CIA-RDP83M00171 R002300070002-4
lja
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002Or%~K;{1= 3M001718002300070002-4
28 August 1974
MEIORA,N'DUUM FOR USIB PRINCIPALS
SUBJECT: Establishment of USIB Committees to Support the NIOs
1. During the several months since I established the NIOs,
there have been some disconnects in actions the NIOs have taken
because of the lack of a USIB structure to deal with the subject.
The result has been that several of the NIOs have created informal
"USIB" committees.
2. i propose to formalize these efforts and to establish
ten new USIB committees to assist the NIOs in the coordination
of Community substantive and operational matters. These com-
mittees will serve two prime purposes:
a. They will insure a collaborative approach to
problems.
b. They will keep USIB principals informed of
Community actions in the NIO area.
3. I propose creation of a USIB committee for each geographic
area covered by an NIO. These committees would be chaired by the
appropriate NIO with membership to consist of one representative
Approved For Release 20 2 05 1- : IA),OPVM0017 30023000700W
 ~v J . ^, .a ti~i ai
Approved For Release 2002/05/01 CIA-RDP83M00171 R002300070002-4
from CIA, DIA, INR, and IC Staff,plus military service representation
as recommended by the Director, DIA. Each committee will have only
a two year charter to operate, continuation subject to my approval.
Separate committees would be established for:
Soviet Union/Eastern Europe
PRC
Southeast Asia
Western Europe
Middle East
0
Latin America
a. Economic Intelligence Committee
To remain as presently constituted, with the addition
of the NIO for Economics as chairman and an IC Staff officer
as member.
b. Energy Committee
To be chaired by the NIO for Energy, with membership
to consist of one representative from CIA, DIA, INR, Treasury
and IC Staff.
subject for hich an NIO has been named, as follows
4. I further propose to have a USIB committee for each functional
-2-
Approved For Release 2(1b /b510 TP83M001 7R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
c. Strategic Programs Committee
To be chaired by the NIO for Strategic Programs with
membership to consist of one representative from CIA, DIA,
INR, AEC, Army, Navy and Air Force. In addition, the chairmen
of GMAIC, JAEIC, SIC and IC Staff will be ex officio members.
d. Conventional Forces Committee
To be chaired by the NIO for Conventional Forces, with
membership to consist of one representative from DIA, CIA, NSA,
INR, and IC Staff and sub-committee chairmen from Army, Navy
and Air Force. The DIA representative will be the Deputy
Chairman. Three sub-committees are proposed as follows:
(1) General Purpose Ground Forces Sub-Committee
Chairman to be provided by the ACSI, Army, with membership
to consist of one representative from CIA, DIA, NSA, NPIC
and IC Staff.
(2) General Purpose Naval Forces Sub-Committee
Chairman to be provided by the DNI, with membership to
consist of one representative from CIA, DIA, NSA, NPIC
and IC Staff.
A 1-:
i
Approved For Release 20 f05/O : rIAA-I4[*83M00171 R002300070002-4
 Approved For Release 2002/05/01 CIA-RDP83M00171 R002300070002-4
(3) General Purpose Air Forces Sub-Committee
Chairman to be provided by the ACSI,,Air Force, with
membership to consist of one representative from CIA,
DIA, NSA, NPIC and IC Staff.
5." I would like to have your comments on this proposal by
W. E. Colby
Chairman, USIB
Approved For Release 200 4qy/Q,1, ,CIA-RDP83M00171 8002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 is
Approved For Release 2002/05/01 : CIA-RDP83M00171 R02 600706Q2-4
DCI Executive Committee
1. Mr. Colby has described the role of the National Intelligence
Officers as being his alter ego in their specific areas of geographic
or substantive concern. This charge connotes interest in and certain
responsibility for identifying and 4br stating requirements, guiding
collection by pertinent disciplines, being aware of and coordinating
analytical, production and estimative capabilities, and evaluating the
quality and responsiveness toward satisfaction of substantive objectives.
Some concern. for adequacy of resource levels is evident also.
2. It is obvious that the skeleton staffs which comprise each
NIO cannot perform all of these functions directly or fully. Instead,
their role must be one of awareness, coordination, and influence
among the various elements and structures of the Intelligence Community
which have varying forms of responsibility for these actions. All of
these functions are performed 'in one degree or another by the individual
agencies, departments and offices making up the Intelligence Community.
The U.S.. Intelligence Board is the organizational capstone on which all
'of) the Community elements are represented. The Committee structure
of the USIB comprises the various subject and functional arms in which
all Community elements participate, as appropriate, to conduct the staff
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00l71 R002300070002-4
3. The Intelligence Resources Advisory Committee is the Community
resources management body supporting the DCI. The `sub-structure of
the IRAC is less diversified than that of the USIB, involving only a
general Working Group, an R&D Council and an Information Working
Group. Thus, there is heavy dependence by the IRAC for certain resource
management actions which are staffed by the USIB Committees in support
of IRAC or USIB or both. This relationship of USIB Committees to IRAC
has not been specifically set forth in the DCI directives (DCIDs) which
constitute the charters for the individual USIB Committees, but has been
conveyed orally.
4. The IC Staff represents yet another structure under the DCI
for the performance of certain functions and coordination actions beyond
the purview or capability. of the NIOs and the USIB Committees. The
IC Staff functions principally in the area of planning, resource program
development and review collection and processing assessment, product
review and the compilation and analysis of community management
information. These functions of necessity interface with the NIOs, the
USIB Committees, all Community organizations and certain other organi-
zations and activities which are not directly included as members of the
intelligence community.
5. The responsibilities and functions of the USIB and its .
committees are reasonably well documented. The responsibilities and
Approved For Release 2002/05/01 : CIA-RDP83M00l71 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
functions of IRAC and its sub-structure are less well documented.
There is no specific documentation on the responsibilities and functions
of the NIO structure or the IC Staff.
6. Throughout the organizations and committees making up the
Intelligence Community management superstructure, there are several
fundamental fuci Lions for which specific authority or coordination
responsibility is obscure. Some of these functions are overall require-
ments management, overall collection management, resource effectiveness
evaluations, and intelligence value assessments. None of these broad
functions appears to fall within the responsibility of any of the existing
community staff or committee elements, although partial responsibility
is assigned to one or another of these elements.
7. One solution to these problems would be the creation of
additional USIB Committees to deal with them on a full-time basis.
Alternatively, recognition should be given to existing Committees which
deal with parts of these problems (SIGINT, COMIREX, HUAIINT, CCPC,
etc.) and the capabilities and responsibilities of the IC Staff and NIOs.
8. This suggests the need for a DCI executive coordination
committee which would be responsible for bringing together all the
functional capabilities in an interrelated and coordinated manner for
the DCI. Such a DCI Executive Committee should have represented
on it individuals responsible for the broad functions charged to USIB,
Approved For Release 2002/05/01 : CIA-RDP83M00l71 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00l71 R002300070002-4
IRAC, the NIOs and the IC Staff and should visibly include both Defense
and DCI-level responsibilities. Candidates for membership would be:
D/DCI/IC
D/DCI/NIO
Chairman, IRAC Working Group
A Senior Representative from DIA
,$SIGINT Committee Chairman
COMVIIREX Chairman
HUMINT Committee Chairman
Requirements Committee Chairman (if formed)
Other USIB and IRAC Committee/ Working Group Chairmen
on ad hoc basis
Individual NIOs on ad hoc basis
9. The Community-wide nature of the Committee's authority
and responsibility indicates the D/DCI/IC as Chairman, and the need
for Defense visibility indicates the senior representative from DIA
(representing both the SecDef and JCS) as Vice Chairman. The
Executive Secretary of the Committee should logically be the Chief of
the Coordination Staff on the IC Staff.
'
/'. _`i. 1...i: .. 1. ., :_
Approved For Release 2002/05/01 : CIA-RDP83M00l71 R002300070002-4
STAT
 Approved For Release 2002/05/01 : CIA-RDP83MOOl71 R002
DCI/IC 74-069
19 May 1974
MEMORANDUM FOR: D rector of Central Intelligence
SUBJECT: USIB Committees Supporting NIOs
1. You will recall that sometime ago you instructed me to
reactivate the proposition to set up new USIB committees for the
purpose of support to the NIOs. Attached is a draft of a proposition
to do that.
?2. I discussed this with George Carver who had reservations
about making the change at this time. George sees this as an
eventual solution but would prefer a less structured approach at
this time. Some of the NIOs have de facto USIB committees assist-
ing them now and others will probably accumulate such groups as they
go. George further expressed his doubt that Admiral de Poix would
support establishing the committees.; I discussed it with de Poix
and that was a fact; de Poix is strongly opposed to the idea.
3. In view of the above, I believe we should let the issue
lie dormant for awhile, perhaps bringing it up this fall. I continue
to believe that the structure would be useful for two primary reasons:
First, it would provide a tasking mechanism for the NIOs which would
help ease the strain of carrying the KIQ/KEP load; secondly, it would
tend to solve some of the institutional problems in the NIO business
at levels lower than USIB itself.
4. Other USIB committee structure issues that are being addressed
now are:
- The continuation of the Human Sources Committee.
- Review of the SIGINT Committee struture.
- The dissolution of the CCPC.
ILLEGIB
Approved or Release : A-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
25X1A
be addressing at the next USIB.
- A recommendation byl to divest SIC of the responsi-
bility for covering existing weapons systems (aircraft, BW-CW, etc.)
in order to concentrate on the more purely scientific problems such
as lasers. This involves a recommendation to form a weapons technology
committee fed by JAEIC and GMAIC to address existent systems. There
is a good deal of merit in this proposition in my view.
- The expansion of the Security Committee charter which-you will
Daniel Graham
Lieutenant General, USA
D/DCI/IC
Attachment
DRAFT Memo for USIB
Principals, 29 Apr 74
25X1/
2
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 CIA-RDP83IV100171' R00230007-0002-4
The Intelligence Requirements Process
Introduction
The term "intelligence requirements" is perhaps the most frequently
used and also the most frequently misused term in the lexicon of the intelligence
community. "Requirements" can cover the spectrum from broad statements
of intelligence production needs to the specific information sought in the tasking
of an individual collector or technical sensor. In every instance, a stated
requirement is used as the basis for creating or operating a collection resource;
for processing, , analyzing or exploiting the collected data; and for synthesizing,
evaluating and reporting single or multiple source information in a product
which can range from a spot report to an in-depth study or a national estimate..
The full range of requirements reflect the insatiable appetites of the
intelligence consumers, the intelligence managers and the intelligence analysts.
Most, if not all, requirements are stated without regard for satisfaction capa-
bility , feasibility, complexity or cost. Few, if any, stated requirements
are ever fully satisfied; those which may be satisfied usually give rise to
further and more detailed statements of need or desire or to regeneration of
the requirement for up-dating purposes. Thus, the volume of requirements
continues to multiply in proportion to the satisfaction achieved.
Apart from relatively superficial "validation" procedures, no particular
effort is made to challenge or reject requirements statements, nor is the
originator ever informed of the feasibility, complexity or cost of meeting a
stated requirement. In this aspect, the requirements process can be viewed
as a "huge juggernaut with no brakes and few effective control mechanisms.
Prioritizing or priordering of requirements is another imponderable
which detracts from the effectiveness of the process in most instances. The
plethora of requirements originators and the wide range of responsibilities
they represent adds confusion and conflict to judgments of priority.
Process versus System
So far, I have referred to the requirements "process" . It would be
"e ble to refer to it as the "system" . However , a system by definition
is `c, re - ularly in teraodn or li:?:'i'?:, .Tlv.en z;r?oup Of lie:`P.::i for ing a unified
~
whole" with the connotation of an oroanizat_on serving a common purpose and
under specific leadership. Except for small segments of the present process
which are systematized, the total process does not appear to meet the "system"
criteria. To make the process a system should perhaps be one of our basic
objectives.
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00'171 R002300070002-4
Elements of the Process
11 The requirements process can best be understood when viewed in
terms of se 1uential steps which become separable elements in the process.
In broad terms these are:
Objectives or Goals -- general statements, usually reflecting end
results; relatively few in number and preferably in some order of priority
or emphasis; correspond to what are otherwise described as First Order
Requirements. Existing example is DCID 1/2 and corresponding JCS JSOP
Annex A.
Statements of Requirements or Information Needs -- Expansion of
general objectives into more specific descriptions of information needed to
support intelligence analysis and product; not directed at any specific
collection, processing or analysis discipline; not in priority order except
as. derived from general objectives; correspond to Second Order Requirements.
Existing examples are Key Intelligence Questions, Defense Intelligence
C~ z:1
Requirements, Essential Elements of Information (EEI) stated in Unified
Command war and contingency plans.
Guidance (to collectors, processors, analysts) -- probably the least-
requirements or i foration
defined element of the process; involves directing K ._~
needs to one or more collection disciplines based on judgment of most likely
sources to provide data in timely and useable form; also serves as manage-
ment mechanism for processors and analysts; determination of essentiality
of one collection source over others is part of guidance. Existing examples
are SORS mission guidance to Directors, NRO and NSA; COMIREX guidance
to NRO, Intelligence Guidance for COMINT Programming (IGCP).
Tasking -- A further detailing of requirements or information need
statements into specific tasks to be performed by individual collectors or
sensors in consideration of guidance provided; equates to Third Order
Requirements in detail of observables, circuits to be covered or targets
to be photographed. This element is best carried out by the collection
resource manager who can marry the data needs with the technical or
access capability of the collector or sensor. Examples are NSA SIGINT
system tasking, technical tasking of overhead systems, or specific intelligence
tasks leiried on Ui, R T collectors .
Interrelationships of the Process
In an ideal world, the elements of the process identified above
should provide for requirements development to flow in an orderly pro-
gression of sequential steps, each detailed statement or action at any level
2
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP831V(00171 R002300070002-4
being relatable to something at the next higher level. Unfortunately, this
does not now occur, and it is difficult if not impossible in many cases to
trace tasks, guidance and information needs back to requirements and broad
objectives. This is true for two principal reasons:
- many originators of requirements have direct access to collection
and processing systems without review by any central authority, and
In the imagery collection and exploitation endeavor, the requirement
and guidance procedure is much more orderly as the result of a central
authority (COMIREX).which receives all requirements, prioritizes them,
and provides for collection and processing action in accordance with system
capabilities. In a sense, the comparison of the SIGINT system to the imagery
system is unfair and unrealistic. Imagery acquisition involves a relatively
few systems constrained only by vehicle availability and weather. Success
(requirement satisfaction) is a "yes" or "no" proposition- Target denial or
target security measures are not significant factors inhibiting collection.
SIGINT collection, on the other hand, is a highly complex mixture of human
and technical systems operating in a deliberately non-cooperative environment
with success attributable to educated technical tasking, tedious monitoring,
fortuitousness and technical competence in processing. Few requirements
are e _r sr co o'_e`el.y ~: Gnd- " ?i? V.
-a . .. s
.- ._ '"si.eC
a?
It is reasonable to conclude, therefore, that the major problems
confronting the intelligence community in requirements management pertain
to the SIGINT system.
Approved For Release 2002/05/01 : CIA-RDP83M00l71 R002300070002-4
- there is no central authority or structured mechanism through
which requirements can pass for validation, association and the provision
of guidance.
For example, the centralization of the SIGINT system over the past
several years and the provision by NSA of direct service in place of direct
support necessitates all commands making known to NSA their SIGINT require-
ments. Commanders are afforded the opportunity to originate and state these
requirements directly to NSA, keeping DIA informed. But no challenge is
offered to the validity or relative priority of such requirements, and there
is no procedure whereby they are associated with corresponding or possibly
conflicting requirements levied on the SIGINT systems from other sources,
either other commands or national (Washington-level) authorities. The
total of all requirements thus received by NSA clearly exceeds the resource
and technical capability of the SIGINT system to respond in an effective and
timely manner . This results in the claim 'that NSA is failing to fulfill many
requirements.
 Approved For Release 2002/05/01: CIA-RDP83M00{171 R002300070002-4
When is a Requirement not a Requirement?
There is one school of thought in the intelligence community that
a stated requirement which exceeds our technical or human capability to
satisfy should not be "validated" and levied for collection or processing
action. I reject this vier.. Any requirement for substantive response
should be stated and accepted whether or not a capability exists to operate
against it. The lack of a current capability could, for example, cause
necessary research and development to be undertaken toward creation of
a capability., At the same time, we should recognize that some of our most
pressing requirements are not likely to ever be satisfied. That fact, however,
is not sufficient to deny that the requirement exists.
This brings into play the need for prioritizing and the careful
provision of guidance in the application of resources toward requirements
satisfaction. The objectives or goals element of the requirements process
provides a general priority framework within which second and third order
requirements can and should be fit. To do so effectively requires, among
other things, that all stated requirements (second order) be reviewed by a
central authority who is able to associate and priorder all statements. This
having been done, the further provision of guidance incident to conveying
the requirements into particular discipline areas can include recommendations
on the emphasis of resource application consistent with overall priorities
and essentiality of the source. This procedure would recognize that some
lower priority requirements would receive no effort in order that appropriate
effort is applied to higher priority needs. These determinations should
be made by the central authority responsible for providing guidance in
each discipline area. Such a procedure is already in effect for the relatively
simple imagery discipline and needs to be developed for the SIGINT discipline.
To Make the Process a System
An analysis of the requirements process and its application to the
principal disciplines of imagery, SIGINT and HUMINT indicates that the
process lacks systematic organizational structure in the first two elements--
objectives or goals and requirements or information needs . The statements
in both of these elements need to be associated, the second being derivative
from and an expansion on the first. Both sets of statements need to be
1~''a c f I ^.^Tli to ,)e aoplied.
'; o V Or`_." 11"res CL'
It is suggested that there should be a USIB Requirements Committee
established for these purposes.
The application of stated requirements to particular collection and
processing disciplines should be accomplished by the respective committees
of USIB--CO1`IIREX, SIGINT and HU;,IINT--in the form of guidance to be
4
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 CIA-RDP83M00c71 R002300070002-4
utilized by appropriate program managers in the allocation of resources
and the assign :lent of operational tasking. Obviously, these procedures
v ill take differing forms depending on the detail and nature of guidance
to be provided for tasking purposes.
It seems clear that the SIGINT guidance area is the most complex.
Past efforts to systematize this area have been hampered by two major
shortcomings:
a lack of detailed ?knowledge on the part of the SIGINT
Committee cdneerning resources allocated and capabilities existing within
the SIGINT system to respond to requirements and guidance;
- a lack of centralized access to all requirements levied on the
SIGINT systems by various originators.
There is evidence that the new SIGINT Requirements Management
System (SIRE) being developed by NSA, if shared with the intelligence
community and the' SIGINT Committee, could go a long way toward alleviating
the first problem area. The nature and degree of assistance to be derived
from SIRE needs to be negotiated with NSA and developed for community
application. There appears to be a willingness to do this.
The "other requirements" problem is a USIB and SIGINT Committee
matter involving all members, but particularly the DIA, IVIilitary Service
and Treasury members. While substantial progress can be made on this
problem within the SIGINT Committee, an even more effective system could
evolve with. the assistance of a USIB Requirements Committee.
Recommendations
It is recommended that:
- serious consideration be given to forming a USIB Requirements
Committee to fulfill the functions discussed above;
- the SIGINT Committee Ad Hoc Review Group give particular
attention to a Committee structure or sub-structure which will move toward
more effective sysi:ematization of requireiT tints for which SIGT?.T is judged
- _ J
~ iJ r s i:
%O be an essential source, 1 Gi these ing iZgilir~ ... emen_S, and -'_e
provision of guidance to the SIGINT program manager.
Approved For Release 2002/05/01 : CIA-RDP83M00l71 R002300070002-4
 Approved For Release 2002/05/01 CIA-RDP83M00171 R002300070002-4
Addendum
There is an obvious interface necessary between action recommended
for the SIGINT Committee and tic, Nation al SIGINT Plan under development by
Director, NSA. Requirements for which SIGINT contributions are essential,
resources allocated, system capabilities and an assessment of responsiveness
and satisfaction are all ingredients to be considered in the Plan.
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
DCI/IC 74-034
n
3 June 97-
25X1A
- 'ORAMDU T
zii
r,~I~~ti rQ2:
25X1 SUBJECT: Piece on Requirements ("The Intelligence
Require men is Process" previously distributed)
25X1
25X1
i. I intend to work Mac's paper up into a paper to be coordinated
t,ith key USIB agencies. is getting at a fundamental question
that has to be answered in order for USIB and for us to do the intelli-
gence job correctly. We are still making resource decisions ,without an
audit trail back to the fundamental requirements to be served. The
KIQ/KEP process is going to help, but as of now it is primarily useful
in pointing up the. lack of a proper requirements system in the total
intelligence process. Even with the KIQ/KEP process operating on?a
Steady basis, we still have USIB failing to address its responsibilities
effectively in the requirements area.
3
2. 1 have some problems t,iith Mac's approach to the ordering of
requi regents. To me, a first-order requirement is one that states a
fundamental need for intelligence as seen by the user. OCID 1/2 and
JSOP Annex A don't quite meet this cri ter ion. They are more on the
order of broad guidance and prioritization of effort. The iKIQs as
presently constituted don't quite fit the criterion either. Some do;
s oie don't. Examples of first-order requirements would be:
- Monitor the development of Chinese advance ,ieapons programs.
- Keep Soviet SSBNs under surveillance.
Provide forecasts of worldwide basic food commodity avail-
aility, etc.
Perhaps I am speaking of a vet t'2 r set of tthe old P -Ilvs.
.,..=mot r .. r cL y.LS
Approved For Release 2 [0 /01 : CIA-RDP83IM00171 R02300070002-~4
0
 25X1A
25X1
Approved For Release 20(#~"/d&1oi l CIA-RDP83M00171 R002300070002-4
3. As far as the KIQs are concerned, it may be that they should
all, in fact, be second-order requirements rather than first, since
first-order requirements tend to be so broad that precision in resource
expenditure against them is not feasible. Second=order requirements
should be a breakdown of first-order requirements into sub-questions
that must be answered in order to meet the first-order requirements.
In looking over the list of KIQs presently on the books, a number of
them fit in this category.
4. The third order of requirements is as Mac has stated, those
specific elements of information which allow you to answer the second-
order questions.
5. Naturally'enough, Mac is not delving deep into the parallel
problems of requirements levied on human and overt sources. This
needs to be fleshed out in the paper. ould appreciate the views
of would
on this matter.
6. When we float the paper to the Pentagon we need also to
address the separation of peacetime and wartime requirements. A good
deal of confusion arising across the board (not just with SIGINT) in
the requirements business is the admixture of wartime and contingency
requirements with the peacetime requirements. Once we have established
a requirements hierarchy and system for the peacetime requirements
process, we should ask our Pentagon brethren to establish a parallel
set for wartime contingencies invol the major commands as required.
Such an approach will help I I efforts in the tactical/
nationa-1' intelligence inte ace game.
7: I would like to get together on this problem Wednesday after-
noon
(It mi
ht b
4$
.
g
e useful for the 10:30 meeting on Thursday with the
DCI and the NIOs reference the KIQs.)
DANIEL 0. L-K-3, USAF
Daniel 0. Graham
Lieutenant General, USA
D/DCI/IC
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 ApprIoved For Rel4se10b2/05%01 CIA-RDP~3M00171 R002300070002-4
DCI/IC 74-083
3 June 1974
MEMORANDUM FOR: Admiral Harvey
SUBJECT Some Thoughts on How to Manage
Collection System Trade-Offs
1. In theory, and to a considerable extent in practice, the
SIGINT, COMIREX, and HUMINT Committees of USIB provide
coordinated requirements and collection guidance to NSA, the NRO,,
and the HUMINT community respectively. What is perceived to be
lacking is a community mechanism to coordinate collection between
and among these three separate collection systems.
2. The ability of the DCI and program managers to make
rational judgments on SIGINT /Imagery/HUMINT trade-offs will
depend heavily on a more effective system of evaluation than now
exists but which, hopefully, will evolve from KEP. Thus, in my
view, whatever mechanism is developed to staff out such judgments
should be the same mechanism that makes the key judgments in the
KEP process. This points inevitably to the NIOs.
3. What will happen in practice is that each NIO, after an
accumulation of experience and knowledge based on the data pro-
duced by KEP, will say I need X amount of SIGINT, Y amount of
Imagery, and Z amount of HUMINT to meet the needs of my bailiwick.
George Carver will add up all the X's, Y's, and Z's and present the
totals separately to each appropriate USIB Committee. These will
be considered the highest priority, rock bottom requirements for
each collection system. The Committee will be responsible for
determining what additional requirements will have to be added to
meet "other" legitimate needs of customers. This process will
require brokering not only with the program managers of three
collection systems but also consultation among the three committee
chairmen and finally clearance through IRAC and USIB.
Approved For Relegggs 2002Rh/ 1
I r' 4DP83M00171 R106360616602=4
~`~ ~' ~
 Approved For Rele 1,, J1-FDP83M00171 R002300070002-4
L
4. It may be argued that the NIOs are not adequately staffed
to assume this new function. In my view, it is inextricably bound
up with the function they have already been assigned with respect
to KEP. And it should take very few, if any, more people to do
the former in addition to the latter.
5. The IC Staff, as now constituted, is not equipped to do this
job. The essential ingredient, which only the NIOs are possessed
of, is the capacity to make substantive judgments on what kinds of
collection systems are producing the required types of intelligence
to meet our highest priority needs. The IC Staff can, however,
provide the resource cost data which will necessarily be required
by the NIOs as an input to their substantive judgments.
6. A super-committee, consisting of the chairman of the three
cognate committees, would not be in a position to make objective and
independent judgments, although this might be a useful staff element
under strong NIO guidance and direction as proposed in paragraph
3 above.
cc: AD/DCI/IC
D/MPRRG/IC
Approved For Relea `-204 02/05Yb'9- `~iA 3P83M00171 R002300070002-4
 CONFIDENTIAL
Appr~ved For Release 2002/05/01 : CIA-RDP43M00l71 R002300070002-4
4 June 1974
MEMORANDUM FOR
SUBJECT Random Thoughts on Coordination
of Collection Disciplines
1. Pursuant to last Friday's Staff meeting discussion,
John Dickey and I have discussed the problem and come up with
a few thoughts.
2. It has been suggested in several forums (including
General Allen's discourse at the USIB Meeting of 30 May) that
there be a mechanism under the USIB to coordinate the applica-
bility of the various collection systems to a particular target.
The question has been raised as to the best means of acquiring
data; i. e. from HUMINT, SIGINT or Photography. How does the
community determine which system is best suited for collecting
specific information? In any particular situation a decision is
needed as to whether they are totally independent of one another,
redundant or mutually exclusive.
3. By the nature of the systems mentioned much information is
unique to the specific discipline. On the other hand, under tight
budgets, tasking of more than one discipline may not be cost effective.
Other than the moribund CCPC no mechanism exists to sort the
pieces. Even during its active days the CCPC approach tended to
be problem oriented and narrow in scope even though valuable in
specific cases.
4. One could argue that the KIQ/KEP innovation will provide the
answers to this question. On the other hand, the KIQs are also
narrow in their application and serve mainly to highlight those
program managers who believe their resources can be applied to the
satisfaction of the stated question or deficiency. But who determines
which program manager should proceed in collecting and reporting.
CONFIDENTIAL EX-2, IMPDET
Approved For Release 2.002/05/01 : CIA-RDP83M00l71 R002300070002-4
 Approved For Rele 2 OTLtIA-RDP83MOO171 R002300070002-4
(.
on the deficiency? The NIO structure currently is not geared to
such technical determinations. The committees of USIB germane
to this problem deal only in their own disciplines. While the
committee system provides for inter-Committee and inter-Agency
coordination it does not, for many reasons, work a good share of
the time.
5. Currently, the SIGINT Committee is engaged in developing
a SIGINT Requirements Mechanism to interface with NSA as the
Program Manager. The proposed concept for this mechanism is
in its early stages of development. It considers DCID 1/2 as a
general starting point to translate requirements into SIGINT terms.
Discussion of this proposal has indicated that the general nature of
DCID 1/2 creates difficulties in the translation process. In addition
questions have been raised as to the applicability of DCID 1/2 to the
HUMINT and Photographic requirements mechanism. Here again
the overall question of collection by requirements highlights the need
for a coordinated approach to all. disciplines.
6. There is precedent within the USIB structure for organizing
appropriate chairmen of pertinent USIB committees to address cross
discipline programs. This precedent could serve as a basis for a
Collection Coordination Committee (CCC) to address the applicability
of all collection disciplines to an intelligence requirement. Such a.
committee could be composed of the Chairmen of the Human Sources
Committee, SIGINT and COMIREX, and be chaired by the D/ DCI/NIO.
The function of this committee would be to review and determine
which collection discipline is best suited for a particular task. It
may be that all three, SIGINT, HUMINT, and Photography will be
required. On the other hand it may prove that only one will do the
job. This committee should not in of itself develop requirements
and/or priorities, but should apply those developed by other committees
to cross discipline tasks. That is to say SIGINT, CO1dIR..EX_and
Human Sources Committees should continue to develop requirements
end priorities. The Coordinating Committee on the other hand would,
for example, determine that a particular subject justifies higher
priority for Photo than for S I GINT. This Committee could also
support IRAC on cross-program matters.
- 2 -
CONFIDENTIAL
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 CONFIDENTIAL
Approved For Release 2002/05/01 : CIA-RDP~3M00171 R002300070002-4
7. The complexity and breadth of the overall problem is so
great that there is a real danger of developing a mechanism too
unwieldy to do the job. Thus , some consideration should be
given to having a "floating membership" for specific items, not
dissimilar to Plummer and Potter's de facto membership on
USIB for certain agenda items. Thus Chairmen of other Committees,
IC Division Chiefs and the NIO's might sit on the CCC for certain
items as appropriate.
8. In the past one problem with a Committee of Chairman has
been that they are all CIA. This potential stumbling block may be
avoided by the CCC with the D/DCI/NIO Chairmanship, the SIGINT
Chairman's non CIA affiliation and the possibility of an HSC
Chairmanship from outside the Agency.
9. The CCC concept provides the best available balance between
having a Committee with the clout necessary to do this job and the
expertise needed to do it right.
10. In our review we considered various approaches to this
problem including a larger committee with representation from
several USIB Committees. We discourage such an approach as
unwieldy and unable to focus on the stated problems.
Executive Secretary, USIB/IRAC
- 3 -
Approved For R 1 a i'2DO 051I4lLCIA-RDP83M00171R002300070002-4
 Approved For I4~ d' 400 5/b1 ' A-RDP83M00171 R002300070002-4
Thoughts on Rationalizing the
Requirements and Collection Tasking Processes
The President, in his letter of 22 March, stated: "I am
particularly concerned that the link between substantive
intelligence needs and intelligence resources be clearly under-
stood and evaluated". The most difficult problem facing the
DCI in meeting this charge is that there is no such link. The
main reason the Intelligence Community could never have had
an effective evaluation process (assuming it wanted one) is the
tremendously wide gulf between the establishment of require-
ments and priorities on the one hand and the allocation of
resources and evaluation on the other: never the twain meet.
This is a pervasive problem. It 'is reflected in attitudes, thought
processes and even organizationally, with one group--USIB--
concerned with "substance" and another- -IRAC--concerned with
resources.
The KEP, of course, was designed mainly to bridge this
gulf and, in effect, to provide the link which so rightly concerns
the President. But even when operational, the KEP will address
only the tip of the iceberg, and there are major problems ahead
in making it work at all.
Beneath this overall problem are other closely related
obstacles to effective community management. One is the nature
of the requirements process. has written on that
aspect. Another, which threatens a success of the KEP itself,
is the absence on the collection side of the community of any
capability to review the total collection task, which, bad enough
in itself, leads to still another lack: there is no structure
allowing for an integrated interface between the production and
collection sides of the community.
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 ii ;- ~(
Approved For ReleasasZd0litj~t0 CifP~=RDP(83M00171 R002300070002-4
Looking at the KEP alone for a moment, this void on the
collection side is perhaps the single most serious obstacle to
successful implementation. Recognizing the fact that all twelve
NIO's cannot effectively provide for the essential interface with
collection managers, the IC Staff has provided a KEP monitor
to perform this task, among others. This has worked fairly
well in the relatively simple lob of preparing the Baseline Reports
in the pilot run. It is highly unlikely that it will work even for
the Performance Reports in the pilot run and almost certain that
it will not for a fully operating, steady-state KEP.
The larger problem, of course, is the lack of any meaningful
overview of all collection efforts. Combined with the lack of a
systematic formulation of requirements, and in the context of the
overall separation between "substance" and resources/evaluation,
one can hardly avoid the conclusion that as presently structured
the Intelligence Community "can't get there from here". Without
a soundly-based set of requirements linked to.collection tasks,
production output and the associated costs, it is difficult to see
how the DCI can even know in any coherent way what the community
is doing, why it is doing it, and how the resources are being used--
much less evaluate overall performance.
Clearly some structural changes are needed.
has addressed the requirements dimension and recommended an
overall Requirements Committee. A good proposal but more is
needed. As I see it, there are several approaches one could take:
(1) Establish a DCI Executive Committee responsible
for reviewing and validating community-wide requirements
and the resultant collection tasking. The community would
report directly to the DCI and be composed as follows:
D/DCI/IC - Chairman
Senior Representative from DIA - Vice Chairman
D/DCI/NIO
Chairman, SIGINT Committee
Chairman, COMIREX
Chairman, Human Sources Committee
Chairman, Requirements Committee (if formed)
(2) Establish a USIB Requirements and Collection
Committee (same responsibilities as in 1 above) composed
as &.olldws:
Approved For Re gg, SQ2AU/q1- A-RDP83M00171 R002300070002-4
i/ i~1-3t.)L!i i 11 :i-
 { CON 1- ID 14 l,A
Applved For Release 2002/05/01 : CIA-RDP(83M00171 R002300070002-4
D/DCI/IC - Chairman
Senior Representative from DLA - Vice Chairman
D/DCI/NIO
Chairman, SIGINT Committee
Chairman, COMIREX
Chairman, Human Sources Committee
Chairman, Requirements Committee (if formed)
(3) Establish two separate committees, one on Requirements
and one on Collection, composed as follows:
Requirements Committee Collection Committee
Chairman designated by Chairman designated by
DCI with USIB advice DCI with USIB advice
Members representing
USIB members/Mil Depts
Members representing
USIB members/Mil Depts
(4) Rely on the NIO's per
memo on this subject.
Attachment: (1)
(2)
Requirements paper
paper on "Collection
System Trade-offs"
suggestion in his
Approved For Re)e e;~6?2~Orr.~/ i . -RDP83M00171R002300070002-4
 Approved For Release 2002/05/01 : CI.A-RD43M00171 R00230.0070002-4
25X1
25X1
25X1
25X1
25X1
MEMORANDUM FOR:
SUBJECT A USIB Requirements Committee and Related
Questions
1. What follows is a review and analysis of the
ttrequirements problem" based on the material which you
gave tol and me on 3 June.
2. memorandum argues persuasively
for a "centra authority or structured mechanism through
which requirements can pass for validation, association
and the provision of guidance." The real crunch in
requirements management, as he sees it, is the SIGINT
system. Those who originate requirements have direct
access to parts of this s stem without review by any central
authority, recoutitiends two Lhings: first,
the forma ti equirements Committee and, second,
a Committee structure or sub-structure to deal specifically
with requirements levied on the SIGINT system. It appears
that the specific problem of uncoordinated requirements
levied on the SIGINT system could be solved by a USIB
Requirements Committee. However, if a USIB Requirements
Committee were formed, it might then be unnecessary to
form a Committee structure or sub-structure specifically
to process SIGINT requirements. The most efficient course
of action is perhaps to decide first whether there should
be a USIB Requirements Committee. If this is decided
affirmatively, SIGINT requirements could be handled as part
of all other requirements levied on collection and processing
systems.
3. memorandum goes deeper, perhaps,
than the formation of a USIB Requirements Committee. This
becomes clear when his memorandum is read in the context
of the literature which has been collecting since August
1973 about USIB Committee structure. It becomes clearer
still when one thinks about who would staff a Requirements
Committee, how this Committee would be structurally related
to the other USIB Committees as presently constituted, and
particularly how a Requirements Committee would satisfy
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Release 2002/05/01 : CIA-RDP~3M00171 R002300070002-4
growing intelligence community dissatisfaction with the
present system. At worst, the addition of a Requirements
Committee would be a makeshift piece of patchwork grafted
on an already cumbersome structure. At best, a Requirements
Committee would provide a short-term solution to a set of
nasty problems. In this light, General Graham's memorandum
to Mr. Colby, dated 22 February 1974, on "Re-examination
of the USIB Committee Structure" goes right to the heart
of the matter. General Graham referred to the IC study,
"The USIB Committee Survey Task Group Report," upon which
Mr. Colby deferred action pending the establishment of the
NIO Office. General Graham reviewed the various problems
raised about USIB Committees and observed that "the
situation calls for more than a piece-meal approach." The
addition of a Requirements Committee appears to be a
"piece-meal approach" to the general problem.
4. Your attention is also called to a memorandum from
Mr. Colby to USIB Principals, dated 29 April 1974, regarding
the "Establishment of.USIB Committees to Support the NIOs."
Mr. Colby proposed, in short, the creation of a USIB
committee for each geographic area covered by an NIO. If
such committees were added to the present structure without
a major reorganization, the result would most likely be a
really complex and virtually unmanageable entity.
5. Perhaps several birds can be killed with one stone.
The IC study recommended that USIB Committees be reorganized
along geographic lines. Their suggested Committees are
listed below in the left-hand column. The geographic areas
covered by the present NIOs are listed below in the
right-hand column.
IC Study Suggestions
for USIB Committees
USSR and Warsaw Pact
People's Republic of ChInn
sia
Southeast Asia
Middle East and Africa
Latin America
Strategic Weapons and SALT
General Purpose Forces and MBFR
Economic Intelligence Committee
Watch Committee
Security Committee
Present NIOs
USSR and East Europe
China
Western Europe
OPacific
Southeast Asia
Middle East
Latin America
Strategic Programs
Conventional Forces
Economics
Special Activities
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Apprroved For Release 2002/05/01 : CIA-RDI'83M00171 R002300070002-4
This almost complete correspondence is very suggestive.
If the plan proposed in the IC study were adopted, each
NIO could be the chairman of the corresponding USIB
Committee. This would implement Mr. Colby's plan as
stated in his 29 April memorandum.
6. The IC study also recommended the establishment
of support groups as follows: The IREX Support Group;
The SIGII,T Support Group; The Human Sources Support Group;
and The Foreign. Science and Technology Group. This
recommendation could connect with the requirements issue
in the following ways. The NIOs could function as the
USIB Requirements Committee, subject to the review of
the USIB. All requirements would be processed through
this Requirements Committee. The establishment of
priorities and the precising of requirements from first-
order to second-order would be done by the Committee.
The levying and further precising of collection and
processing requirements could be done .through the four
Support Groups.
7. The advantages of such a system are as follows.
First, the USIB Committees could be staffed along intelligence
community. lines, insuring the full participation of member
agencies at the national level. Second, the Requirements
Committee of NIOs would insure the coordination of
requirements levied on the community by intelligence
consumers. Third, this Requirements Committee could
serve as the focal point of communication between
intelligence consumers, on the one hand, and the collection
and processing systems, on the other hand. Fourth, priorities
in requirements could be settled or negotiated directly
with intelligence consumers. Fifth, interaction with
collection and processing systems could be handled by means
of the four Support Groups. The Support Groups could
function as brokers of requirements from the Requirements
Committee to the collection and processing systems. Further,
these Support Groups could translate second-order into
third-order requirements. Finally, the interaction between
the four Support Groups and actual collection and processing
systems could concentrate on translating third-order into
fourth-order requirements. This picture would give.clear
sense to the different orders of requirements and, if
implemented, would provide a central authority and mechanism
for translating and processing the needs of the intelligence
consumer to all elements of the community which work to
satisfy those needs.
Approved For Release 2002/05/01 : CIA-RDP83M00171 R002300070002-4
 Approved For Rey
>0 / "1 16 -RDP~3M00171 R002300070002-4
DCI/IC 74-090
7 June 1974
SUBJECT Intelligence Requirements Process
1. The matrix of judgments about intelligence "requirements"
has tended to operate with several unspoken assumptions. There is
a need to define and include formally some additional criteria in
order to arrive at a more precise and economical application of
resources. The intelligence community cannot make some of the
key judgments alone, and perhaps should not try to make them inde-
pendently.
2. For about 25 years a great proportion of our judgments
have been subtly but perceptibly influenced by the idea that, given
our power and will to exert influence and "leadership" in world
affairs, the U. S. would be most likely to become involved in any
situation of much significance. We therefore felt the need to maintain
a very large data base in order to support the widest array of options
for action, should policy and decision makers choose to involve
themselves and the nation in a given subject. It seems safe to say
now that there is a general recognition that our relative ability to
act has declined, and our will to do so is not lightly to be assumed.
We will increasingly tend to examine a situation carefully before
deciding whether or not we must become involved, or could advan-
tageously take some action. And when we decide to act, we are more
likely now to do so in a political or economic policy sense rather than
in the kinds of material ways we often launched some years ago.
3. The definition of the factors that need to be cranked more
formally into "requirements" judgments needs some study, but I
suggest there are perhaps three elements: the probability of occurrence
of the situation to which the requirement is linked; the relative impact
on U. S. national interests of that situation; and the relative importance
to the type of response the U. S. seems most likely to make, of the
specific information addressed in the requirement.
Approved For Releaser2.QQ2/0/O1y:I lj{ ,-l 83MOOl 71 0 .;
 Approved For Release 2002/05/01 CIA-RDP80M00171 ROO2300070002-4
4. I think it is in the interest of the intelligence community to
ask for a much more specific involvement of the Executive Branch
and perhaps even the Congress in making judgments about general
and longer range requirements and priorities, otherwise the com-
munity must assume responsibility for making assumptions about
U. S. domestic attitudes and policies which are outside its purview.
5. One can view the requirements process from a slightly
different perspective in order to integrate these other factors.
6. The current intelligence operation is on the one hand like
running an underground newspaper. On the other hand, the collection
and "requirements" side, it is the most directly responsive to policy
and decision makers, operating on current problems and in highly
specific terms. These intelligence operations support the situation
we are involved in, whether by choice of the nation's leadership, or
as a consequence of actions of others involving U. S. interests clearly.
The element that has been removed here is anticipation. We do not
need to guess what the U. S. needs will be, we are being told by those
responsible for guiding the enterprise. The peripheral considerations
also tend to be narrow. We can see much more clearly the probable
direction of events and can more concisely allocate resources for the
short term. But this is not the real problem area when we speak of
defining requirements, it is the longer term question dealing with
anticipated needs.
. 7. The intelligence analyst/report producer is an encyclopedic
scholar in his field. He wants to maintain the widest possible data
base, and he tends to be highly situation oriented by nature: the
soldier prepares for the worst battle he can envision; the economist
for crash and inflation; the political scientist for apathy or revolution,
etc. , etc. The analyst says, to cope with X situation, we would need
to know so and so. The intelligence analyst/producer should be
asked to make only one principal set of judgments. ? Assuming the
occurrence of the situation he is equipped to deal with, what is
critical to know and what is advantageous ? This listing and priority
ordering establishes what might be called outline of the optimum
data base. Being situational in nature, this could embrace wartime,
peacetime, political, economic, sociological, and other kinds of
problems.
8. The next level of judgment to which the intelligence community
can substantially contribute is an estimate of the probability of occur-
rence of given situations. The general trend of the actions of others
2
Approved For Rely?p;2/pVU 1' DP83M00171 R002300070002-4
 Appro'ed For R $e X00 g5/O 1 !PdA-RDP80M00171 R002300070002-4
gives us clues: knowledge of what is under consideration in the
centers of power, potential power, or potentially disruptive
influences provide tip-offs. Judgments in this area of national
interest are currently being reflected in documents such as the
Perspectives paper, KIQs, requests for NSSMs or study papers,
and the selection of topics for other in-depth intelligence reports.
9. A key related level of judgment at this point takes the
intelligence community into domestic politics: will U. S. leadership
or public pressures select involvement if such-and-such a situation
develops? To what degree? Or do we have to be very well informed
just to prove that we are very well informed, although we really
don't think the U. S. needs to or will do much about it? It is on this
imponderable that the intelligence community should demand
greater policy guidance, pinpointing the problem and its effect on
resource allocation. The alternative, now being practiced, is
application of simple budgetary constraints, leaving the substantive
judgments to the technicians, who then bear the burdens should
national leadership take an unforeseen tack.
10. Most of the requirements documents I have seen do a fair
job of relating the substance of a given informational need to several
layers of increasingly generalized objectives and goals. Many of
them can be "justified" or "validated" by rational discussion if
certain sets of assumptions are shared, but most often the underlying
assumptions are ill-defined and not necessarily common to the
community as a whole.
11. Our fundamental objective should be to design a requirements
system which states all the assumptions and assigns more clearly
the responsibility for making concise and explicit levels of judgment.
An agency or department may best be able to judge how crucial
individual items may be to the performance of a given mission, if
and when it is called on to perform in explicit situations. But it
cannot always be sure that national leadership will assign it the same
mission it plans for itself. The intelligence community should be able
to comer an opinion about the probability or trend of international
situations. The community may also be able to venture an opinion
about U. S. involvement, but it should request the executive and/or
Congress to offer an explicit indication or confirmation of the
probability of a major U. S. involvement or response.
Approved For Rel a ~.GA2I 5JQ h :;GIE;1 RDP83M00171 R002300070002-4
 Apprived For R "te 120k ~t L
~85ki 61A-RDP63M00171 R002300070002-4
. 12. Collection resources should be developed to meet the
more probable involvements. Subsequent decisions about the
employment of collection resources against these requirements
might best be made by. a high level intelligence community group
including representatives of the major program managers, supported
by NIO and USIB Committee expertise. This group should rarely
have to meet, since ongoing evaluations by NIOs and USIB Committees
would indicate desirable shifts in use of collection resources
,,,to topic and situation, as would also the KIQ/KEP process.
according
13. It would seem that the really basic and broad statement of
requirements would require review and endorsement only about
once every two years. NSC action (i. e., tasking levied by NSC
level officials, even if not always as the formal group) would give
specific focus to level of operation and tasking in the interim
periods. In this connection, there may be room for improvement
in the technique for handling collection tasking. Policy makers
quite often find a means of tasking program managers or sub-
elements directly, making coherent management of resources
difficult, and identification of the total workload subject to question
when combined community planning is involved.
Approved For Rele 62' 6 , DP83M00171 R002300070002-4