Foreign Relations of the United States, 1969–1976, Volume II, Organization and Management of U.S. Foreign Policy, 1969–1972

211. Memorandum From the Director of the Program Analysis Staff, National Security Council (Lynn) to the President’s Assistant for National Security Affairs (Kissinger)¹

Washington, August 7, 1970.

SUBJECT

• Blue Ribbon Panel Recommendations on Intelligence²

In connection with your lunch meeting today with PFIAB, this memo:

• —summarizes the comments of the Blue Ribbon Panel on Defense intelligence;
• —notes their recommendations for administrative changes;
• —comments on the limits of the Panel’s analysis.

Summary

The Panel paints a gloomy picture of the U.S. Defense intelligence system, marked by effective autonomy of the intelligence elements from the consumers and effective autonomy of the service intelligence components from the two institutions—NSA and DIA—which are supposed to provide a coordinated and unified DOD intelligence service.

As an administrative cure, it proposes centralization of all defense intelligence activities by creating a collection and a production agency with both management and operational control over activities, reporting directly to the Secretary, through an Assistant Secretary of Defense for Intelligence. The present independent service intelligence activities would be abolished.

The Problem

The chief criticisms the Panel makes of the current situation with the defense intelligence community are:

• —NSA and DIA (themselves separate institutions with no common chief short of the Secretary of Defense) don’t really manage the programs supposedly assigned to them. The Services do.
• —Both in SIGINT (the supposed area of NSA concern) and general intelligence (DIA), the Services, not the central all-DOD institution, run the people, the budgets, the resources, the R&D, and the product very largely autonomously.
• —DIA in particular suffers from “too many jobs and too many masters” since it is supposed to be subordinate to both the Secretary and the JCS, whose interests are often in conflict, and since it is supposed to control and coordinate the intelligence components of the individual Services, at the same time that it and its staff are almost wholly dependent on those same Services for both physical and personnel resources and future careers.
• —In particular, DIA has been forced to negotiate away its supposed power as the producer of all finished DOD intelligence, so that even formally it has “shifted from the production of all Defense intelligence to the production of some strategic or that intelligence used at the JCS/OSD national level. The members of the JCS, as chiefs of Service, still maintain current intelligence and estimates capabilities on their respective staffs …”³
• —On the NSA side, the Service cryptologic agencies (SCA’s) are nominally only collection instrumentalities subject to the management and control of NSA. In reality, they are “jealously guarded prerogatives” effectively independent if not dominant of NSA in personnel, budget, facilities, R&D, methods and procedures. Moreover, all Services run substantial separate security and cryptologic efforts outside the purview of activities run through NSA.
• —The lack of coordination with respect to routine intelligence is exacerbated by the existence of “special programs.” These tend to be managed at a high level, nominally or practically free of even the weak coordination to which other efforts are subject. These special programs usually involve the development of a new technical capability and there is a tendency to keep control of the output as well as the operation with the development agency for too long.
• —The result of these divided administrative responsibilities is a divided, uncoordinated product:
• —There are separate map agencies in each Service, separate procedures and regulations for security clearance investigations (and a costly and inefficient refusal to accept each other’s investigations), and separate sets of estimates and reports on the threat, particularly in the “scientific and technical,” i.e., longer-term, area.
• —Each Service produces its own flow of current intelligence and estimates, with the attendant danger that the intelligence produced will be tailored to the special interests of the Services, particularly with respect to manipulating the threat to justify victory for the Service on new weapons systems.

These observations have to do primarily with problems of coordination within the defense intelligence community. The Panel also, although somewhat more cursorily, discusses the problem of the relationship between the defense intelligence community and the consumers:

• —The process of assigning requirements for intelligence collection is conducted almost entirely within the intelligence community with very little meaningful input by consumers.
• —With respect to compartmentalized intelligence (i.e., SI, TK, B, etc.) access to which lies with the managers of the collection systems, the relationships between the various compartments have never been systematically analyzed and there is a tendency to ignore the importance of balancing the need for security against the need for getting the information to the people who need it.
• —The system of writing estimates is said to water down controversy by compromise.
• —Neither on the civilian nor the military side is there a truly professional, career defense intelligence service, except to some extent in NSA, with a resulting bad effect on the process and the product.
• —The Panel notes comments that the system collects much more information than can be processed or evaluated competently and that what is processed often does not reach the people who need and could use it. Regarding the evaluation of the substance of the intelligence as outside its charter, the Panel does not, however, comment on these charges.

Recommendations

Most of the Panel’s recommendations have to do with improving the internal administrative mechanisms for intelligence within DOD. (Incidentally, by its faint praise and its far-reaching recommendations for administrative changes, the Panel clearly implies that it regards as wholly inadequate Laird’s efforts to deal with the problem by giving some central intelligence responsibilities to Froehlke, his Assistant Secretary of Defense for Administration.)

Its recommendations would, in effect, take the Services entirely out of the independent intelligence business and set up a separate defense intelligence service, reporting to the Secretary directly and not through the JCS (or any service chain of command). In detail, the Panel would:

• —Give overall responsibility for defense intelligence matters to the “Deputy Secretary of Defense for Operations.” (A basic recommendation of the Panel, considering the Department as a whole, is to create separate Deputy jobs, for Management of Resources and for Operations.)
• —Establish under him an “Assistant Secretary of Defense for Intelligence” (ASD/I), who would also have the title of “Director of Defense Intelligence” (DDI). This official would:
• —represent Defense on USIB and other interagency intelligence boards;
• —“direct and control all DOD intelligence activities not specifically designated by the Deputy Secretary for Operations (i.e., not the Services) as organic to combatant forces”;
• —have charge of the allocation of resources, the definition of procedures, establishment of requirements, intelligence-related research and development, and access to information;
• —have as his principal subordinates a “Defense Security Command” (DSECC) and a “Defense Intelligence Production Agency” (DIPA).
• —DSECC (which would be a military command) would be the successor-in-interest to NSA and would be the basic collection agency. It would:
• —take over from the Services all collection activities now conducted by the Service cryptologic agencies (but with authority to delegate operational and administrative responsibility as appropriate);
• —include some processing closely related to collection;
• —take over all the functions of NSA and expand those functions to include the “processing, data base maintenance and reporting of all intelligence information.”
• —DIPA would have charge of all intelligence production not organic to combatant forces. It would be the successor to DIA and would:
• —provide all current intelligence, threat assessments, finished ad hoc intelligence, DOD estimates, and DOD inputs to national estimates;
• —manage all defense intelligence production and dissemination including that organic to combatant forces.
• —DSECC and DIPA would each be responsible for planning, evaluation, and review under the ASD/I, of the intelligence activities under their control.
• —Set up a unified map and topographic service, under the Deputy Secretary for Management of Resources.
• —Create professional, career defense intelligence services, with both civilian and military members.

Comments

These recommendations would greatly centralize the defense intelligence process. The Services can be expected to attack them vigorously as based too much on analysis of abstract management relationships and not sufficiently responsive to the practical and specialized needs of the individual parts of the defense establishment. The Panel explicitly recognizes the importance of a certain degree of competition between intelligence providers, but has concluded that the present system carries competition to absurd extremes.

From your point of view—and probably that of PFIAB—the most important limitation of the Panel’s work are:

• —that it focuses almost entirely on management and administrative problems and does not suggest much which is directly related to improving the quality of the product; (although, of course, better management, less parochialism, and more professionalism should improve the product)
• —that its analysis and recommendations are primarily concerned with relationships within the defense intelligence community itself and not with consumers or with non-defense parts of the intelligence community.

These comments are not meant necessarily as criticisms of the Panel—its job was to look at defense intelligence and primarily from a management, not a substantive, point of view. But, the problem is much broader than the Panel’s charter permitted it to consider.

1. Source: National Archives, Nixon Presidential Materials, NSC Files, Agency Files, Box 239, Defense—Blue Ribbon Defense Panel. Top Secret. Sent for information.
2. “Blue Ribbon Defense Panel, Report on National Command and Control Capability and Defense Intelligence,” submitted to the President on July 1, 1970. (Ibid., Box 1324, Unfiled Material—1970) The report was a supplement, prepared by a small part of the Panel, to the Panel’s main report. The Panel, which began its work in July 1969, was established under the chairmanship of Gilbert Fitzhugh to undertake an extensive study of the Defense Department and make recommendations on its organization and management.
3. Ellipsis in the source text.