APPROVED FOR RELEASE 1994
CIA HISTORICAL REVIEW PROGRAM
18 SEPT 95
Evolution and role of "the most broadly controlling document in the field of requirements."
PRIORITY NATIONAL INTELLIGENCE OBJECTIVES
Ludwell L. Montague
Clyde Heffter, in the "Fresh Look at Collection Requirements" which he takes in a recent issue of the Studies,1 notes the "conspicuous hiatus" between such high-level guidance documents as Director of Central Intelligence Directive 1/3, Priority National Intelligence Objectives, and the collection requirements actually produced at the working level, particularly with respect to the question of determining relative priorities among such requirements. He invites discussion of the problem of "how to formulate needs and priorities in such a way as to facilitate the satisfaction of needs in a degree roughly proportionate to their priorities, through the most effective use of the collection means available."
In the nature of the case, collectors are likely to be more keenly aware of this problem than people working in other phases of the intelligence process, but its existence and gravity should be of concern to researchers and estimators as well, for it is their work that ultimately suffers from any diffusion and misdirection of the collection effort. The hiatus between general guidelines and practical requirements that Mr. Heffter points out is real, and its consequences are serious. He has considered it from the collectors' viewpoint. The purpose of this article is to complement his analysis with an examination from the other side of the gap-specifically, to describe the development of the PNIO concept and to review what the PNIO's are and are not intended to be. Conclusions as to what is wanting for the determination of practical priorities are substantially the same from either point of view.
Evolution of the PNIO's
From the outset it was understood that the responsibility of the Director of Central Intelligence for the coordination of U.S. foreign intelligence activities included a responsibility to provide authoritative guidance for intelligence collection and production from a national (as distinguished from departmental) point of view. To this end, National Security Council Intelligence Directive 4, adopted by the NSC in December 1947, prescribed two specific duties:
(1) To prepare "a comprehensive outline of national intelligence objectives [generally] applicable to foreign countries and areas."
(2) To select, on a current basis, the sections and items of this outline having priority interest.
By "comprehensive outline" the drafters of NSCID 4 meant an integration of such then existing departmental documents as the Army's Index Guide and the Navy's Monograph Index. What they had in mind has actually been accomplished by the preparation of the National Intelligence Survey outline (NIS Standard Instructions, June 1951). However, the publication of DCID 1/2 (15 September 1958) was considered necessary to meet the formal requirement for a "comprehensive outline" of national intelligence objectives.
The 1947 directive had the fault of prescribing a method rather than a mission. Manifestly, national intelligence objectives have never been determined by the selection of "sections and items" from a "comprehensive outline." They are no longer required to be in NSCID 1, of 15 September 1958, whose subparagraph 3b (1) is the present-day survivor of the original NSCID 4.
The fact is that no priority national intelligence objectives were formulated until 1950, and that their provenance then was unrelated to NSCID 4. In May 1950 the Joint Intelligence Committee produced JIC 452/7, "Critical Intelligence Objectives of the Department of Defense with Respect to the USSR." This document identified as critical intelligence objectives five generalized aspects of Soviet military capabilities. In September its text, with the addition of two highly generalized references to political warfare, was adopted as DCID 4/2, "Priority National Intelligence Objectives." In June 1952 this DCID was amended to cover explicitly not only the USSR but also "its Satellites (including Communist China)."
The preoccupation of DCID 4/2 with Soviet military capabilities was a natural consequence of its origin and of the circumstances of the time, the shooting war then in progress in Korea. In August 1953, however, an armistice having been signed, the adequacy of the DCID as priority guidance for a national intelligence effort was questioned. The Board of National Estimates was directed to study the problem and to propose a suitable revision. Its study, in consultation with research and collection personnel throughout the Agency, extended over a period of ten months, followed by six months of inter-agency coordination.
It was represented to the Board that the almost exclusively military character of DCID 4/2 resulted in claims of priority for the collection of any desired item of military information over any other information, no matter how significant the latter might be in relation to the national security. Such claims were plainly out of consonance with the current estimate (NIE 99, October 1953) that, for the near term at least, the Kremlin would probably avoid military action with identifiable Bloc forces, that the active threat to U.S. security was likely to be a vigorous Communist political warfare campaign designed to undermine the Western power position, and that there was danger of a weakening of the unity of the Free World. They were also plainly out of consonance with NSC 162/2, Basic National Security Policy (October 1953), which emphasized a need for intelligence on the capabilities and intentions of friendly and neutral states as well as of the Soviet Bloc.
The Board concluded that the list of priority national intelligence objectives must be expanded to cover at least the most significant of these non-military concerns, and that there must also be some discrimination between military objectives of greater and of lesser consequence. This expansion of the list and need for discrimination within it led to the development of three general categories of priority within the listing. A single list in absolute order of priority was considered infeasible and also undesirable, as likely to introduce self-defeating rigidity into the system.
The revised DCID proposed by the Board of National Estimates and adopted by the Director of Central Intelligence with the concurrence of the Intelligence Advisory Committee (DCID 4/4, December 1954) was the prototype of the present DCID 1/3 (4 January 1961). The differences between the two represent only those adjustments normally to be expected as responsive to developments in the situation.
Criteria for PNIO Selection
The DCID has an annex that sets forth certain criteria to govern the selection of priority national intelligence objectives. A gloss on these criteria is in order at this point.
It is stipulated, first, that the PNIO's should be directly related to the intelligence required in the formulation and execution of national security policy. Through its role in the preparation of national intelligence estimates, the Board of National Estimates is cognizant of the intelligence requirements of the NSC and its subordinate policy boards. It is also cognizant of the most critical problems inherent in the estimates required to meet their needs. Its identification of these substantive problems as priority national intelligence objectives can provide a basis for identifying priority research and collection requirements, but of course does not in itself define such requirements.
Second, since the bulk of the intelligence required in the formulation and execution of national security policy will be the product of routine intelligence collection and research, the PNIO's should be limited to the critical problems which require special attention and effort. This principle should be axiomatic. There is, however, constant pressure to make the listing more inclusive, with a consequent danger of its becoming so nearly all-inclusive as to deprive the word "priority" of meaning. This pressure, which apparently springs from a desire to get everyone's favorite topic listed as a priority objective in order to insure that it will not be neglected, has to be resisted.
Third, in order to afford a stable basis for intelligence planning, the DCID should be designed to remain valid over an extended period. This consideration requires the exclusion of topics of momentarily urgent, but transitory, interest, which will require and receive ad hoc treatment in any case. The present practice is to review and revise the DCID annually, the process sometimes extending the period between revisions to as much as eighteen months.
Fourth, since broad generalities are of little practical use, the PNIO's should be specific enough to provide discernible guidance for the allocation of research and collection resources, but not so specific as to constitute in themselves research and collection requirements. The application of this criterion presents the greatest difficulty in the formulation of PNIO's and is the source of complaints from those collection personnel who refuse to accept them, with Mr. Heffter, as "a constitution which requires both laws and courts to interpret it." The criterion has served on the one hand to rule out the kind of generality found in the 1950-52 DCID 4/2, and on the other to keep the PNIO in rather broad terms, especially in comparison with specific collection requirements-that is, to maintain its character as the statement of a critical substantive intelligence problem rather than an itemizing of the essential elements of information needed for its solution.
Role of the PNIO's in Guiding Research and Collection
The function of the PNIO's as stated in the DCID, is to serve as a guide for the coordination of intelligence collection and production. They are intended to be only the first step in a process beginning with a need for information felt at the national policy planning level and extending to the servicing of specific collection requirements in the field.
In this first step, the Board of National Estimates, with the advice of other Agency offices and in coordination with USIB representatives, identifies the critical substantive problems inherent in the general body of intelligence required for purposes of national security policy. This is as far as estimators can properly go in relation to the total problem. The identification and formulation of collection requirements related to these priority national intelligence objectives requires analysis by research personnel to determine the elements of information essential to a solution of the problem, the elements already available or readily obtainable through research, the additional information obtainable through routine collection and the residual information of such critical importance as to warrant a priority collection effort.
Obviously, not every bit of information somehow related to a priority national intelligence objective will be required with equal urgency. Many are procurable by routine means. It is therefore necessary that research personnel exercise discrimination and restraint in formulating collection requirements, claiming priority for only those aspects of a priority objective that actually do require a priority collection effort. As Mr. Heffter has pointed out, the criterion here is not a possible incidental relationship of the collection requirement to a PNIO, but the importance (the essentiality) of the desired information for a solution of the critical problem designated in the PNIO and its unavailability from other sources. If research analysts present unjustified claims for priority collection, citing some not cogent relationship to a PNIO, collectors must exercise their own judgment and authority in rejecting them.
If a particular system of intelligence collection is unable to satisfy all of the legitimate requirements levied upon it, determinations have to be made as to which requirements will be accorded priority. In this operational context, however, priority can never be determined solely by reference to the PNIO's. One requirement related to a PNIO and certified by a responsible research agency to be a really essential element of information, being well suited to the particular mode of collection, may consequently be accorded the desired priority. Another such requirement may be totally unsuited to that mode of collection and therefore unworthy of any consideration whatever, no matter what the PNIO to which it is related. All sorts of gradations are possible between these two extremes. In these circumstances collection officers must assume the responsibility for deciding between the importunate claimants for their services. Their decisions may be informed and guided by the PNIO's and other instruments that Mr. Heffter cites, but they must be made primarily in terms of the collector's expert professional knowledge.
Problems such as these are inherent in the administration of intelligence research and collection. No reformulation of the PNIO's could obviate them-unless, indeed, the PNIO's were to be transformed into a community-wide listing of coordinated collection requirements in an absolute order of priority. Even if this were done, something like 'the present PNIO's would then have to be reinvented to guide the coordinators of collection requirements. The problem lies, not in the PNIO's, however imperfect they may be, but in the gap between them and the scramble to obtain priority for individual collection requirements.
What Can Be Done About It?
In 1954 the Board of National Estimates was keenly aware that the formulation of PNIO's was only a small part of the total problem. It recommended that the then Special Assistant to the Director for Planning and Coordination be directed to review existing procedures for the development and coordination of collection requirements in relation to the PNIO's, and to propose improvements. The Special Assistant made such a study and concluded that no action was advisable. Like Mr. Heffter, he considered that a single community-wide mechanism for coordinating collection requirements, assigning priorities to them, and allocating particular collection resources to their service would be a Rube Goldberg contraption, more a hindrance than a help. The Board of National Estimates would heartily agree. It had not meant to propose the invention of such a machine, but it had hoped that serious study of the subject might bear such fruit as a more general understanding of mutual responsibilities and more systematic procedures for cooperation in the common cause.
For six years, however, the gap has remained, and collectors as well as estimators evidently find it to be not a Good Thing. And now Mr. Heffter comes forward with some constructive suggestions and a welcome invitation to professional discussion of the problem. Rejecting as impractical the idea of a community-wide coordination of collection requirements in priority order, he suggests that the situation could be alleviated if more systematic use were made of the findings of the several USIB subcommittees under their assigned authority, in their respective fields, "to recommend ... intelligence objectives within the over-all national intelligence objectives, establish relative priorities on substantive needs, review the scope and effectiveness of collection and production efforts to meet these objectives, and make the necessary substantive recommendations to the departments and agencies concerned." This would be precisely the kind of implementation of the PNIO's which the Board of National Estimates has advocated for many years.
More important than any procedural proposals, however, is Mr. Heffter's recognition of the fundamental need for a truly professional doctrine and discipline in relation to this subject. The professional discussion which he seeks to stimulate is a necessary step toward the satisfaction of that need.
It is now time for someone to join the discussion from the viewpoint of the research components of the community.
1 IV 4, p. 43 ff
Top of page