Foreign Relations of the United States, 1964–1968, Volume XXXIII, Organization and Management of Foreign Policy; United Nations

253. Letter From Director of Central Intelligence Helms to the Chairman of the President’s Foreign Intelligence Advisory Board (Clifford)¹

Washington, September 20, 1966.

Dear Clark:

[3 paragraphs (23 lines of source text) not declassified]

The letters from President Kennedy to Mr. McCone² and from President Johnson to Admiral Raborn³ have created a precedent for the issuance of an express Presidential directive to each individual DCI. I believe these letters have helped to clarify the role of the DCI as a coordinator and have served a generally useful purpose. Significance has been attached in some quarters to differences in the language as between the letters issued by President Kennedy and President Johnson. This suggests the need for considerable care in drafting the language of a new letter, if it is decided that a new letter is desirable.⁴

The most important question to decide is the extent and limitations of the responsibilities of the DCI, particularly insofar as they have managerial implications for programs functioning under agencies of the Government other than CIA, particularly under the Department of Defense. Specifically the questions are to what extent the DCI should be held responsible and accountable and given authority:

a. To determine the needs of the Government for information derived from intelligence channels;

b. To determine the scope, character and level of collection and analytical programs and facilities required to meet those needs; and

c. For the efficiency and economy of these programs (now costing in excess of [less than 1 line of source text not declassified] a year).

Under existing arrangements the DCI, supported and advised by USIB, tends to be regarded as primarily responsible for validating and determining the needs of the Government for intelligence, and the Secretary of Defense to be responsible for the control and management of a very large preponderance of programs and facilities calculated to meet these needs. The precise delimitations of authority and responsibility as between the Secretary of Defense and the DCI remain vague and ill-defined although practical working relationships are satisfactory. It may very well be that there is no practical alternative to the present dichotomy and that we should continue to work under current authority, recognizing the somewhat nebulous nature of the DCI’s commitment “to provide effective guidance and coordination,” but recognizing also that there has been very considerable improvement in the coordination of the national intelligence effort and that further progress under something comparable to the existing authority can be reasonably anticipated.

Before expressing my final views on this subject, however, I should appreciate further time for reflection and suggest that this matter be scheduled for further discussion some time towards the end of this year. I am attaching a memorandum on the subject which may help further to clarify the issue.

[Omitted here is discussion of unrelated items.]

Sincerely,

Richard Helms ⁵

Attachment

Washington, undated.

DISCUSSION OF ADEQUACY OF DCI AUTHORITY TO COORDINATE THE U.S. INTELLIGENCE EFFORT

As a minimum, the DCI must have the authority to find out anything he wants to know about all activities which contribute to national intelligence. He must have the ability to form an independent judgment as to whether intelligence programs are generally responsive to national needs. Institutional arrangements must exist which ensure that he can communicate an authoritative view, if necessary to the President and the President’s immediate advisers, as to which programs of the Government are redundant or marginal and what should be done to fill intelligence gaps.

The law and NSCID No. 1⁶ provide, in a general way, this authority and institutional arrangement. The law provides that the Agency, of which the DCI is the head, shall advise the NSC on intelligence matters and recommend ways to improve intelligence coordination. NSCID No. 1 provides that the DCI “shall coordinate the foreign intelligence activities of the United States in accordance with existing law and NSCIDs.” NSCID No. 1 confers other authority on the DCI, generally subject to consultation with USIB, to do various things. For example, the DCI or his representatives “in consultation with the head of the intelligence agency concerned” is authorized to make surveys of departmental intelligence activities.

The two Presidential letters, respectively to Mr. McCone and Admiral Raborn, constitute at least a clarification and probably an extension of the DCI’s authority. Certainly the Kennedy letter constitutes a mandate to the DCI, acting jointly with the heads of departments, to review the activities of all U.S. agencies “with a view to assuring efficiency and effectiveness…” It has been argued that the Johnson letter is weaker because it directs the DCI to coordinate and guide the total United States intelligence effort “in accordance with NSCID No. 1.” Certain of the provisions of NSCID No. 1 imply a certain obligation on the part of the DCI to act, in some respects at least, with the advice and consent of USIB and to deal with heads of agencies and departments through their intelligence representatives. The Johnson letter also uses language which appears to emphasize the DCI’s responsibility for coordinating intelligence output, rather than intelligence programs and activities.

In view of the practice that has developed of giving the DCI a personal mandate from the President, a new letter of authority, emphasizing the President’s concern with the need for effective guidance to the over-all intelligence effort and directing the new DCI to ensure the provision for this guidance, may be desirable.

It remains to be asked whether the Board expects the DCI to be responsible for more than general guidance and coordination for intelligence activities. Do they expect to hold him accountable for the efficiency of all intelligence activities. Do they expect him to be responsible for the elimination of all waste and extravagance in any intelligence program run by the Government.

As of today, there is no central mechanism in the Government for an over-all budgetary or program review of all intelligence activities as a whole. The four basic programs: CIA, and CCP (SIGINT), the CIP (DIA and the Service intelligence programs), and the National Reconnaissance Program are all reviewed separately with somewhat different representation through somewhat different channels. It should be recognized, however, that the DCI’s right to participate in the review and have a say in the formulation of all three DoD programs is now firmly established.

Consideration has been given in the past to the desirability of establishing a National Intelligence Resources Board, to be chaired by the DCI or jointly by him and the Secretary of Defense which would be responsible for the consolidated review and approval of all intelligence programs. There are many practical considerations, however, which suggest that such an arrangement may be unwise or, in any event, premature.

A basic factor affecting the coordination of intelligence activities is the necessary division of authority amongst individual departments and agencies of the Government. It is inevitable and appropriate that heads of departments having responsibilities in the foreign policy fields and commanders of major military commands should have the personnel and facilities required to assemble and analyze the information needed for their parochial and departmental purposes. Information which they legitimately require is also, in most cases, relevant to national decisions which affect “national security.” Moreover, intelligence data of national significance is collected by individuals, diplomats for example, as part of their official responsibility for carrying out normal departmental activities. Intelligence, whether national or departmental, is very often a by-product of some essentially non-intelligence activity controlled and conducted by non-intelligence components of the Government.

It follows that all of the activities and components of the Government which serve national intelligence purposes can never be totally subordinated to the direction, control and management of a single central authority. Added efficiency would be given to the DCI as a coordinator by subordinating NSA and, through NSA, the cryptologic military services, to the DCI. This was the original recommendation of the Brownell report,⁷ but was rejected as impractical by the then DCI (General Bedell Smith). Exhaustive reviews of the constituent responsibilities of the NRO served to emphasize the impracticability of totally subordinating reconnaissance programs, including their support facilities, launch pads, tracking and recovery facilities, etc., to the managerial authority of the DCI. Even if these two large and probably indigestible ingredients were added to the DCI’s personal command, he would still have to cope with the problem of coordinating a number of activities which cannot, under any circumstances, ever be placed under his direct managerial supervision.

The net result of all this suggests that we are stuck with the present concept under which the DCI “guides and coordinates” the community but does not manage or command it. If so, the various institutional arrangements through which the DCI provides guidance and coordination (USIB, intelligence agency program and budget reviews, the NRO, etc.) should be examined to ensure that they provide an adequate basis for the assertion of his influence but do no imply responsibilities which extend beyond the limitations on his authority.

1. Source: Central Intelligence Agency, Executive Registry, Job 80–B01676R, PFIAB 1966. Secret. Drafted by Bross.
2. Dated January 16, 1962; printed in Foreign Relations, 1961–1963, vol. XXV, Document 99.
3. Document 233.
4. [text not declassified]
5. Printed from a copy that indicates Helms signed the original.
6. A copy of NSCID No. 1, January 18, 1961, indicating the revisions incorporated in the revised NSCID No. 1 issued March 4, 1964, is at the Johnson Library, National Security File, Agency File, CIA.
7. Brownell report on communications intelligence, June 13, 1952. (Central Intelligence Agency, Executive Registry, Job 80–B00269R, Box 10, Folder 62, and Box 14, Folder 130)