MEMORANDUM FOR: MR. TAYLOR MR. DIRKS MR. CARVER MR. DONNALLEY(Sanitized) FROM RICHARD LEHMAN

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP91M00696R000700130014-1
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 33
• Document Creation Date: December 16, 2016
• Document Release Date: May 19, 2005
• Sequence Number: 14
• Publication Date: August 15, 1975
• Content Type: MF

Approved For Relea '2005/06/06: CIA-RDP91 00696R00& 0130014-1
15 August 1975
MEMORANDUM FOR: Mr. Taylor
Mr. Dirks
Mr. Carver
25X1
25X1
1. Herewith Draft #5 (I think) of the new
structure. It incorporates all the suggestions
from yesterday plus a few others that became
obvious. For instance the Boards and Committees
paragraph broke down into recommendations.
2. In addition Ed Proctor gave me the attached
comment. To meet his points I have included an
additional paragraph on page 14 and a more extended
discussion of NSA, including a recommendation. I
have also finessed NPIC by dropping reference to it.
He is quite right on this. NPIC would start tasking
NRO rather than the producers as at present, and that
would be disastrous. The paper doesn't suffer from
the change.
3. There are also attached three substitute pages
for "Role of the DCI".
25X1
Richard Lehman
Director
Strategic Research
Attachments:
As Stated
Approved For Release 2G 57 O ~IA-RDP91 M00696R000700130014-1
 Approved For Relea'4 2005/06/06 : CIA-RDP91 M00696R00M0130014-1
A central question has not been considered in the arrangements
proposed in the paper:
America's foreign problems of growing importance and frequency
that intelligence is being called upon to address are not in the
military field. Yet most of the sources unique to intelligence
are under the management of the Department of Defense (e.g., NSA
and NRO). This together with the traditional and perhaps archaic
outlook of the managers of these programs allows these resources
to be used to satisfy some of the most insignificant needs of
the military at the expense of some of the most important non-
military, intelligence problems. Continuation of OoD management
of NSA will tend to continue this imbalance. Subordination of
NSA to the DCI and separation of the functions of the DIFTISA(civilian)
Commander
and the/CCS (military) would go a long way toward fining this.
I also shutter at the idea of consolidating NPIC and NRO to
consciously create another NSA. It will become inbred and as
self-serving as NSA has.
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
 Approved For ReleaW2005/06/06 00696ROOOW130(,14-Iehman
i 15 August 1975
The Role of the DCI
1. In any discussion of the organization and
management of the Community, the DCI--what he is,
what he does, what he is supposed to do--is the central
issue. This paper focusses on the DCI's general
responsibilities and powers as an introduction to
the more detailed papers that follow.
Statutory Background
2. The National Security Act in essence sets
up the DCI primarily to produce national intelligence,
? r
although it subsequently has been interpreted to permit
him to callect and to conduct covert action. It impli-
citly makes him the leader of something that has come
to be called the "Intelligence Community". It,does not,
however, specify his functions (beyond "correlate and
evaluate") nor does it provide him with specific author-
ities over the agencies that make up the Community.
3. The President's letter of November 1971 made
explicit some of the responsibilities that were only
implicit in the Act. In so doing it increased the
DCI's responsibilities without increasing his powers.
He was directed to:
Approved For Release 2005/06/06 -RDP91 M00696R000700130014-1
SftRET
 Approved For ReleaW2005/06/06: CIA-RDP91 M00696R0000130014-1
JLt''uT
The continuing review and assessment of collection
results; and the establishment of requirements for
new systems; product review, which includes both
the final shpaing of the intelligence product to
match the needs of the national consumer and a
continuing evaluation of the product against those
needs; and inspection. Note that all of these
except inspection are interdependent and operate
at the interfaces between the various communities.
Great Responsibilities
5. It is apparent that the DCI is a member in some
sense of all the communities. It is also apparent that
he wears three hats--as Presidential adviser, as head
of "the Community" (Chairman of USIB, IRAC, and EXCOM),
Approved For Release 2005/06/06: IH TM00696R000700130014-1
 Approved For Relea 2005/06/06 : C J 0696R000 0130014-1
9-UE1
that concern more than one agency. He advises
0MB on the Community budget. For the Congress,
he provides intelligence, defends the Community
budget, and advises on all foreign intelligence
matters.
c. The DCI as Director of CIA. As DCIA, the
DCI is a line officer administering a large inde-
pendent agency under the NSC. He is a producer
of intelligence for the mechanisms over which
he presides wearing his other two hats. Quite
distinct from these roles, he has a specialized
line function as the agent of,,the President, or
the NSC, in the implementation of foreign policy
decisions through covert action and confidential
communication with foreign governments. For the
Congress, this DCI too is a source of foreign
intelligence. It expects him to present and defend
CIA's budget, and to account for its performance.
He is required to inform the Congress of covert
action programs; whether he will have to defend
them is not yet established.
Approved For Release 2005/06/06 :~ i R9fM00696R000700130014-1
 1
Approved For Relea '2005/06/06: ST r,_1M00696ROOWaW?OR1e'h_
man
8/15/75
The preceding papers have discussed in some
detail the state of American intelligence today. They
have identified a number of serious problems and, more
generally, support a conclusion that the present situa-
tion is unworkable and unacceptable. Current political
developments suggest that the National Security Act
of 1947 will be rewritten; our analysis of the Act and
the intelligence structure it established convinces us
that it ought to be extensively revised. In this paper
we seek to draw all the threads together and to present
recommendations for a national intelligence structure
that will meet the requirements of the next quarter-
century. We are unanimous in our support of these recom-
mendations, but wish to emphasize that we arrived at this
position through an exhaustive examination of the various
ways in which the system might be structured.
The Political Imperatives
Our recommendations grow out of a common view of
the political and bureaucratic environment in which
intelligence must function in the mid-70's. The public,
the Congress, and the President have concerns and interests
that constrain in general terms the problem that confronts
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
SEC ET
 Approved For Relee 2005/06/06 : CIASEIO~flJ696R0O' '00130014-1
1. Public attitudes toward the problems posed
by an intelligence service in a free society
are, of course, not homogeneous, or even
mutually consistent. Much of the public
gives scant consideration at all to in-
telligence problems. Furthermore, the most
articulate segments of the public are often
not fully representative of public attitudes.
To the limited extent that generalizations
are meaningful, "the public" probably:
--Wants the benefits and protections of
a strong intelligence structure, but has
little sophisticated understanding of
what that desire implies in specific terms.
--Is confused by a number of the issues
which are currently the focus of both
press and Congressional attention--covert
action, proprietaries, domestic collection,
etc.
--Wants to be reassured that U.S. intelli-
gence is not a "rogue elephant", but is.
both accountable to and effectively
controlled by the public's elected repre-
sentatives, the President, the Congress,
or both.
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
-2-
 n r&
Approved For Releft* 2005/06/0 t 61 M00696R00`dW130014-1
2. Congressional Attitudes. At this writing,
Congress also speaks with a muultiplicity
of voices. To the extent that we can 'make
generalizations about Congressional attitudes,
they appear to include the following:
--A desire for a strong intelligence system,
or at least for the benefits of a strong
intelligence system.
--Inadequate understanding of the structural
requirements for attaining these benefits,
and especially of the degree of secrecy
essential if they are to be attained.
--A desire for a "correlation and evaluation"
entity independent of any Cabinet department,
especially of the Defense Department and
the military services.
--A recognition of the need for at least
some clandestine collection operations,
but without--so far--a matching willing-
ness to face up to the secrecy requirements
thereof.
--A recognition of the need for covert
action in some contexts; Congress also
wants--or thinks it wants--a larger voice
Approved For Release 2005/06/06 : CIA-tbP91 M00696R000700130014-1
 Approved For Rele` 20051 JJP: A-RDP91 M00696R00 100130014-1
in the approval of such actions, but
has not yet recognized the responsibility
such a voice entails or the need for dis-
cretion it imposes.
--A desire for greater access to the intell-
gence product, although the constitutional
implications of and obligations imposed
by receiving classified intelligence are
also matters Congress must ultimately face.
3. The Presidency. In discussing "the President's"
attitudes, a distinction has to be.drawn between the
abstract needs of the office and the concrete attitudes
of any specific incumbent. The former--especially as
perceived by persons never likely to hold that office--
may not always square with the latter. Nonetheless
any President will probably:
--Want a strong intelligence system, includ-
ing a strong, flexible and responsive covert
action capability.
--Want reassurance that that system is
under control--meaning his control and not
anyone else's.
--Want the system run efficiently, with
due regard for budgetary considerations.
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
SECRET
 cFrP
Approved For Relea 1005/06`10g'.UTIP91 M00696R000F 0130014-1
--Want the intelligence system and its
activities not to be a source of political
difficulty or embarrassment.
--Want independent advice, particularly
in time of crisis, from capable people
primarily loyal to the Presidency and
independent of the Departments that execute
policy.
--Need a system that can function well
both in peace and in war, although the
problems here involved--e.g., the national/
tactical question--have not been thought
through clearly.
4. A Given President. The specific attitude of any
particular President will be very much shaped by
his own personality, working style and confidence
in his immediate associates (or lack thereof). Here,
most generalizations are of little value. Given the
formidable pressures and obstacles involved in being
elected President, however, there is one generaliza-
tion which probably has some validity. The holders
of this office are likely to be strong-minded men
inclined to place a high premium on loyalty in their
subordinates, certainly in subordinates who enjoy
Approved For Release 2005/06/06 : CIA-I DP91 M00696R000700130014-1
 Approved For Relea'2005/06/06 : Cl" g1-M 0696R000W0130014-1
their confidence. No President is likely to be
charitably disposed toward, to make extensive
use of, or to support any intelligence organization--
or head thereof--that does not clearly acknowledge
the primacy of its, or his, responsibilities to
the Executive Branch and the President.
5. This President has an additional requirement.
He has already suffered political embarassment through
revelation of past intelligence activities today
considered by many to be unacceptable. These are
not of his making, a fact that makes it both easier
and more necessary for him "do something" about
intelligence, to show that he is responsive to the
public and Congressional mood. He also has an
opportunity. His predecessors saw to varying degrees
a need for structural reform in the intelligence
system, reform they were unable to carry out without
amending the National Security Act. This they
were unwilling to do. Now, however, the Act is
going to be reopened by Congress in any case.
We believe these Congressional and Presidential
imperatives are not irreconcilable in any fundamental
way. We believe they can best be met by an independent
DCI presiding over an Intelligence Community shaped by
the necessity to balance national and departmental needs.
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
 Approved For ReleaVav'2005/06/06: CIA-RDP91 M00696R000M0130014-1
SECRET
The President, in meeting Congressional requirements
for fundamental reforms in the oversight of intelligence,
can at the same time meet the Executive requirement for
fundamental improvements in its management. This paper
deals with these internal improvements, a subsequent one
with external reforms.
A Stronger DCI
One common thread that runs through the preceding
papers is that of responsibility without authority.
This is reflected in the difficulties faced by the DCI
in trying to enforce or even to make rational decisions
across the entire range of intelligence management. In
the production field he finds it extremely difficult to
establish the primacy of national intelligence, contain-
ing as it does a coherent'presentation of alternative
positions, over uncoordinated departmental views. In
the management of collection he lacks the power to enforce
a systematic approach that can serve efficiently national
needs while not neglecting departmental ones, both across
and within collection systems. In resource management
the budgetary weight of the Department of Defense makes
it impossible for him to allocate resources rationally
or to deal effectively with duplication and obsolescence.
-7-
Approved For Release 2005/06/06 - RD~~ M00696R000700130014-1
UPI
 Approved For Relea2005/06/06 : MTTT00696R00130014-1
His position in a number of areas, most notably that
of crisis management, is being eroded by encroachment
of the military. In short, responsibility without
authority is more than a cliche'.
To provide the authority of the DCI needs we believe
two essential conditions must be met. The first is a
point that has often been met in these pages but will
bear repeating. The DCI need not be a close friend and
confidant of the President, but he must have the
President's confidence and support. Especially, he
must have--and be seen to have--regular, frequent per-
sonal access to the Oval Office. Gen. Smith was able
to be effective as DCI where Adm. Hillenkoetter was not
largely because the members of the Intelligence Advisory
Committee knew he had a weekly appointment with the
President.
Presidential support, however, is not enough. It
is reasonable to expect that the Secretaries of Defense
and State will similarly have the Presidential ear, and
can outweigh the DCI unless he is able to act within a
framework that provides him stronger statutory authority.
The main girder of this framework should be resource
management. The stronger the DCI's voice in the alloca-
tion of funds, the easier it will be for him to impose
rationality in other aspects of his job. We therefore
Approved For Release 2005/06/06 : CIA-R8AM00696R000700130014-1
rLT
 Approved For RelefiW 2005/06/06 : ~l 00696R00D 0130014-1
? 1~ ~ I i : r
recommend that a large segment of the intelligence budget
now appropriated to Defense be instead appropriated to
the DCI for further allocation to the various program
managers. This would include the NRP, CCP, and some
elements of the GDIP.*
This does not mean vesting operational control of
these programs in the DCI. We have considered and
rejected the concept of a unitary command structure
for intelligence, either under an independent Director
of National Intelligence or embedded in Defense or State.
Rather, we are proposing a new concept of the DCI, one
that would exchange his present powers (variously to
command, advise, and persuade) for more effective and
Perhaps less conspicuous management powers at key points
in the system.
We have earlier identified "communities" of resource
managers, collectors, producers, and users of intelligence.
In simplest terms these communities are inter-linked as
follows: funds flow from resource manager to collector
Such an arrangement was established for the Poverty
Program in the 1960's. Funds were appropriated to the
Director of the Office of Economic Opportunity but
then delegated to the Department of Labor for actual
program operation.
-9-
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
 Approved For Relea 22005/06/06: ON-- 00696R00 130014-1
and producer; raw intelligence flows from collector to
producer; finished intelligence from producer to user; user
then determines whether his needs have been met and states
new needs to resource manager and to producer; producer
states new requirement to collector, or resource manager
provides funds to develop new collection capabilities.
We propose that the new DCI sit astride this system,
controlling these linkages rather than exercising line
authority over any of the communities.
--We have already stated that funds should flow through
and be in broad categories allocated by him.
--He and his staff should carry out and integrate
the collection management functions now assigned
to COMIREX, the SIGINT and Human Sources
Committees, and the Collection Guidance and
Assessment Staff now in DDI.
--He should continue to coordinate ("correlate and
evaluate") finished intelligence production as
he now does.
--He should seek consumer reaction to his product,
evaluate it, and through this process identify
gaps to be filled by tasking existing collection
systems or by developing new ones.
Each of these functions is closely related to and dependent
on the other three. (See attached sketch) The DCI must
have a strong integrated staff if he is to make effective
Apar ed f of slep aQW/W0@pIRP*,Wgo@Rp RQ0130014-1
 Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
25X1 SECRET
D-G-I
25X1
SECRET
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
 Approved For ReleaW2005/06/06 : gft--=PMtM00696R0MW130014-1
Li
.The DCI and the Community
CIA. A DCI who could effectively regulate the
linkages among the various communities would have
acquired greatly strengthened management powers at a
time when there are political pressures to weaken him.
Thus there must be a balancing decrease in his line
authority over CIA, and we so recommend for this and
a number of other reasons.
We recognize that a separation of the DCI from
direct management of CIA has been suggested many times
before, and as many times rejected. It was argued
that:
--The National Security Act would have to be
changed.
--The President could no longer look to one
man for intelligence and covert action.
--The DCI would need a substantial staff.
We submit that the first two of these reasons are no
longer valid. The third is obviously true, but not
necessarily a reason for maintaining the status quo.
. On the other hand, the reasons for such a separa-
tion are stronger than before.
--Both the 1947 Act and the President's
letter of November 1971 give the DCI
Approved For Release 2005/06/06 : ClA]FDP91 M00696R000700130014-1
 Approved For Releaan'2005/06/06: C~ I5.91 M00696R000400130014-1
important responsibilities in the Community
as a whole. His ability to exercise these
responsibilities has been compromised by
his role as head of the CIA, both externally
in the Community and internally within the
Agency.
--CIA continues to be widely criticized. A
DCI not closely identified with it would be
far more politically acceptable and available
as the senior national intelligence officer.
Indeed, a President would find it easier to
give a DCI the access and confidence upon
which his power must ultimately rest if the
DCI were not himself considered an intelli-
gence operator.
--Present arrangements already require the
DCI to carry a number of very complex
responsibilities; if we increase further
his overall management and budgetary role,
we must reduce his management span in other
ways.
These reasons made a compelling case for-an officer we
will call Director-General for Intelligence, replacing
the DCI as Presidential adviser and leader of the
Approved For Release 2005/06/06 : CIA-F t P91 M00696R000700130014-1
SECR Ei
 Approved For Relea` 2005/06/06 :. t L ET 00696ROOON0130014-1
Community; and a Director/Foreign Intelligence Agency,
separately appointed and confirmed, replacing him as
Director of CIA (thus renamed). The D/FIA would be
responsible to the NSC. We believe the DGI should be a
statutory member of the NSC, both to increase his status
relative to State and Defense and to clarify his relation-
ship to the D/FIA.*
Defense. Another common thread running through
these papers is the adversary relationship of CIA and
Defense over a broad range of issues. It is here that
the balancing of national and departmental interests
becomes most difficult. On the one hand, the existence
of an intelligence organization (CIA) not subject to
the control of any other line department or agency
within the USG is essential; on the other the Department
of Defense, charged with responsibility for defending
the nation, requires a measure of control over important
collection, processing, and other intelligence activities
which also contribute in major ways to the solutions of
problems faced by CIA. These two facts are both the
basis for many of the problems which have characterized
the overall management of American intelligence since
World War II and the reason why workable solutions-.to
these problems are so difficult to develop.
Hereafter, in speaking of the future, we will use the
terms DGI and FIA; in speaking of the present and
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
past, we will use DCI and CIA.
 Approved For Relea2005/06/06 : 00696R000-d0130014-1
ot L I
--For a number of reasons, primarily the high
cost of technical collection systems and
their overlap with tactical intelligence,
the bulk of the national foreign intelligence
program budget involves Defense Department
funds, controlled by that Department, plus
personnel and physical assets also belonging
to it. The President's letter of November
1971, however, makes the DCI in some way
responsible for this budget, including funds
for tactical intelligence.
--Moreover, the foreign problems most likely
to preoccupy the national leadership, and
therefore national intelligence, in the
next few decades are political and economic.
Yet the military orientation of the,
technical collection systems tends to
favor relatively minor military require-
ments over major non-military ones.
--In a wartime situation, however, the military
services' need for certain types of intelli-
gence will be paramount, and neither they
nor their civilian chiefs will be comfortable
with any arrangement which does not give
them control over the assets providing this
needed support.
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
nn,r"('
 Approved For Rele 2005/06/06 : G9M lo0696R000~D0130014-1
t ~_
--In peacetime, the military services are
geared toward insuring that the intelli-
gence needs of major US force commanders
are met.
--The wartime/peacetime problem is compli-
.sated by an ambiguity inherent in the
Defense Department's notion of the
"National Command Authorities," a
concept which includes the President
and the Secretary of Defense, but
does not include other officers of
the government such as the Secretary
of State and the DCI.
--There is an understandable resistance
in the Defense Department, and particularly
in the uniformed military, to the concept
that at any group--especially a group
of "civilians"--should provide indepen-
dent analysis to the President which
affect decisions regarding U.S. military
forces.
Given these attitudes, a proposal to transfer-'sub-
stantial funds and authority from Defense to the DGI
would obviously meet bitter resistance. Defense would
Approved For Release 2005/06/06 : 044 ROP91 M00696R000700130014-1
 Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
Nqw~
T
have to be convinced that it would derive considerable
benefit from the arrangement. We believe there would
in fact be mutual advantage in an agreement-between the
DGI and the Secretary of Defense, ratified in statute,
that both defines and greatly changes their respective
roles in the management of intelligence.
In reaching such an agreement, it is important to
see that the root'of the problem lies in a failure
to recognize the impact of planning for war on practice
.in peace. The role of the Secretary of Defense in war-
time is very clearly established and is embodied in the
NCA concept. The role of the DCI in war, on the
other hand, is fuzzy indeed. This causes bureau-
cratic guerrilla warfare cross a wide front. There
is much skirmishing for authorities, access, systems,
resources--Defense because it will need them in war
CIA because it needs them in peace. The result
has been uneasy compromise, duplication, and rivalry;
we still do not have a truly national intelligence
system. Moreover, at the onset of war, or at various
undefined points in a major crisis, national intelli-
gence assets would be transferred piecemeal to
Defense control under chaotic conditions. The nation
would not be well served.
-16-
Approved For Release 2005/06/06 CIA-RDP91 M00696R000700130014-1
 Approved For ReIL a 2005/06/06 : CIA-RD 1 M00696R0 700130014-1
SEC1 ..
If, however, we consider the question from the
wartime end rather than, as we have since 1947,
from the peacetime one, A "Gordian knot" formula
becomes apparent. The National Security Act of
1975 might read more or less as follows:
The DGI shall be a member of the National
Security Council responsible to the President,
except that in the event of major hostilities
he shall be responsible to the President through
the Secretary of Defense, unless the President
directs otherwise. When he is subordinate to the
Secretary of Defense he shall retain the right
to render substantive assessments independently
to the President.
Such a formulation would tend to cause the interests
of the Secretary of Defense and DGI to converge where
they are now adversary. The Secretary would be more
interested in seeing that the DGI built a strong
intelligence system in peacetime, while the DGI would
be more concerned that the system be designed to meet
Defense's needs in peace or war. The DGI would be
de facto a part of the National Command System, and his
relationship to the National Command Authority would
be clearly established. In the event of war, the
Approved For Release 2005/06/06 : CTz-?#bP91 M00696R000700130014-1
RE
 Approved For Releede 2005/06/06: CIA-RDP91 M00696R0W00130014-1
SECRET
entire system, including the DGI, would move to Defense
as a unit with far less disruption of internal command
mechanisms than would take place under present under-
standings. The door would be open to develop a more
coherent system, with a unitary budget, in peace. At the
same time, the Congress could be assured that the peace-
time DGI was in fact independent of the Department of
Defense.
Out of this arrangement Defense would gain as well
as lose. The same disagreements that have prevented
development of a true national intelligence system have
also severely handicapped development of the military
system. With the DGI clearly responsible for both war-
time support of the military and for rational organiza-
tion of that support in peace (in collaboration with DOD)
a serious problem for military planners is removed.
Defense can also expect national intelligence production
to be more responsive to its needs. (This does not
mean less objective). Finally, Defense would be relieved
of several nagging administrative and budgetary headaches.
The extent to which the intelligence structure
can be rationalized and its management strengthened
depends directly on the degree to which the DGI-Defense
relationship can be clarified and made compatible.
- is-
Approved For Release 2005/06/06 : CIA RDP91 M00696R000700130014-1
(ZcT
 Approved For Relea3 2005/06/06: CI R V 00696R0007"40130014-1
GR
State. The DCI's relationship with the Secretary
of State is less complex than that with the Secretary
of Defense. (We speak here of the general relationship,
not of the unusual situation created by the dual respon-
sibilities of Dr. Kissinger). It is also less trouble-
some, but there are nonetheless a number of important
and persistent problems.
--As Defense resists independent intelligence
assessment and reporting on matters affect-
ing the military, State resists on matters
affecting diplomacy. On the other hand, the
DCI needs State support. to strengthen the
civilian hand in intelligence assessment.
--Tie most important single source of
political and economic intelligence is
Foreign Service reporting. State does not
consider this to be intelligence and will
not accept any linkage between it and
intelligence requirement mechanisms.
-19-
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
25X1
SEC [:T
 Approved For Relea'2005/06/06 : CI/'FOP f U f0696R000M0130014-1
--Covert action is, or should be, the subject
of close coordination with State both in
Washington and in the field.
--The Intelligence Community must work with
State through INR, but INR has little
influence over the operational arms of State
that control most matters of vital importance
to intelligence.
Some of these problems would undoubtedly yield to
the increased general authorities we propose for the DGI.
There does not exist, however, any mechanism by which
the entire range of Community-State relationships can
be regulated at a senior level. We believe there should
be an arrangement whereby the Under Secretary of State
for Political Affairs is charged with these matters
in the Department and the DGI is charged with coordina-
tion between him and the Community elements concerned.
NRO. A DGI armed with budgetary powers and a
defined relationship with Defense will be in a position
to manage technical collection more efficiently, to
25X1
Approved For Release 2005/06/06 : CIA RDP91 M00696R000700130014-1
20-
 Approved For Relea5'2005/06/0SEmr 1 M00696R000 130014-1
make more sensible choices, and to respond more flexibly
to new requirements. Better arrangements will be needed,
however, to link him with technical program managers.
The NRO in its current form is an anomalous patchwork
cobbled together in considerable bureaucratic strife;
it cannot persist in its current form. The element
of competition within the NRP is not as important in
the present and future as it was in the past and the
problems of coordinating within a structure designed
for competition are becoming increasingly difficult.
Second, the need for military commanders to derive
direct support from satellite collection resources is
becoming increasingly important, but a policy allowing
each military service to pursue its own satellite
collection programs would be prohibitive in cost, in-
efficient, and unnecessary. The current NRO organization
with the Under Secretary of the Air Force as director
is not well suited to meet this problem.
We recommend that the NRO be reorganized as a
integrated operating organization under the D/FIA, jointly
staffed by FIA and Defense. This would create an organi-
zation in some ways analogous to NSA, which has under
NSCID #6 a clear line of command over virtually of the
CCP. It would remain, however, subject to the broad
guidance of an EXCOM chaired by the DGI. (We note that
an NRO under CIA would balance an NSA under Defense.)
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
-21-
 Approved For Rel a 2005/06 P91 M00696ROOCf700130014-1
NSA. The strengths of NSA are also its weaknesses.
Unitary organization coupled with physical separation
produces a self-contained (indeed, introverted) organi-
zation isolated from and resistant to legitimate external
interests in its business. NSA is the hair shirt of any
DCI seeking to exert any authority over it or even to
extract the information he needs to form any judgment
as to its effectiveness and responsiveness to national
needs.
For reasons valid in the past but less so today
NSA is dominated by the military. It is controlled by
Defense, many of its personnel are in uniform, and its
field intercept organization is run by the services.
Military influence must be reduced if NSA is to
.be fully responsive to the growing political and economic
needs of national intelligence.
We recognize, however,
that NSA remains so entangled with Defense that fundamental
change will be most difficult.
We therefore would take a cautious approach to the
handling of NSA. In addition to funding the CCP through
the DGI, we recommend only that NSA be placed under the
guidance of an EXCOM chaired by the DGI. The EXCOM,
25X1
Approved For Release 2005/06/06 : C A DP91 M00696R000700130014-1
 Approved For Re1[e 2005/06/01SEt 1 M00696Rd700130014-1
however, would be changed as a matter of high priority
with developing a program to "civilianize" NSA and
studying the possibility of separating the functions
of a civilian NSA from those of the military Combined
Cryptologic Service, both with a view to the eventual
resubordination of NSA either to the D/FIA or to the
NSC,parallel to FIA.
Recommendations
In summary, our recommendations are as follows:
1. Amend the Act to create a DGI separate from
the FIA and to establish a working relationship between
him and the Secretary of Defense along the lines suggested
on Page 17. Make him a member of the NSC.
2. Provide him with a staff capable of perform-
ing at least the "linkage" functions outlined on page 10
above, and with an inspection group as proposed in our
paper on external controls. (There are of course many
other staff functions that could be assigned, such as admini-
stration of a Community-wide intelligence career service.)*
* What elements of the present CIA he should take with
him is a complex question that must ultimately be
addressed. There are strong arguments in favor of a
small staff limited to coordination, but there are
also strong arguments for assigning to the DGI the
production function.
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
Crr?T
 Approved For Rele 2005/06/06 RM00696ROOOM0130014-1
3. Charge him with preparation of a total
intelligence budget covering the CIAP, NRP, CCP, and
leave responsibility for GDIP in Defense. Appropriate
funds for the programs covered by his budget to him for
allocation according to guidelines jointly agreed with
OMB and Congress. This would make it possible to
abolish IRAC (retaining its useful R&D Council) and to
eliminate the office of ASD/I.
4. Charge the DGI with planning the transfer of
national intelligence assets to the Department of Defense
in war and with using these assets to meet the needs of
Defense in peace and in war. Charge Defense with coope-
rating in this endeavour by providing access, staff
support, and'quality personnel. Charge the DGI with
establishing a National/Tactical Planning Board, on
which the U&S Commands would be represented, as the
regulating mechanism for this program.
5. Create a new D/FIA appointed by the President
and confirmed by Congress. Place under him the present
CIA minus the DGI's staff. He would be under law re-
sponsible to the NSC but in practice would report
through the DGI. (The extent to which he should be
legally required to do so is an unsettled question).
25X1
Approved For Release 2005/06/091 M00696R000700130014-1
 Approved For Rel a 2005/06/06 : IA-RT 00696R0tl00130014-1
6. Reconstitute EXCOM with the DGI in the chair
and the Deputy Secretary of Defense and a senior White
House official, e.g. the Science Advisor if that Office
were reestablished, as members. Charge it with broad
budgetary and policy guidance over NRO,and NSA. 25X1
In particular charge it with carrying out the studies
suggested on page 23.
7. Reorganize NRO as an integrated organization
reporting to the D/FIA and jointly staffed by FIA and
Defense.
8. Establish an Intelligence Coordinating Committee
to regulate relations between the intelligence system
and State (except for substantive production). This
'Committee would be chaired by the DGI and would include
the Deputy Secretary for Political Affairs as its
principal member.
9. Reconstitute USIB as an Intelligence Production
Board under the DGI as chair, with its membership reduced
to include only the major production organizations. The
Board would retain the present substantive responsibilities
of USIB. All other functions of USIB not otherwise re-
assigned in these recommendations would become responsi-
bilities of the DGI.
Approved For Release 2005/06/ ~-GPA; 91 M00696R000700130014-1
 Approved For Rel We 2005/06/06: CIA-RDP 1 M00696R09 00130014-1
10. Make the DGI Chairman of NSCIC. (The only
way to get some consumer response is to give the
interested party control).
These changes add up to a relatively "clean" arrangement,
given the complexity of the matters involved. We believe
they would greatly improve the management of US intelli-
gence. We are fully aware that these changes are
revolutionary as change goes in the bureaucratic world,
and that they will meet strong resistance in many quarters.
In particular, the ability of the DCI to meet military
needs has not been tested and will--be suspect. Neverthe-
less, these are traumatic times. They create both the
need and the ooportunity for radical change.
Approved For Release 2005/06/Ca j 91 M00696R000700130014-1
 Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1
COLLECTORS
COLLfCTi h
WANARET
PpRE}SOURCEyy
iti~1 ~rRaV
ENT 1 17-14+`'i
PRODUCERS
USERS
Approved For Release 2005/06/06 : CIA-RDP91 M00696R000700130014-1