30. Letter From the President’s Deputy Special Assistant for National Security Affairs (Kaysen) to the Deputy Assistant Secretary of Defense for International Security Affairs (Rowen)0

Dear Henry: Attached is a comment on your draft of the military and related sections of the Basic National Security Policy which you sent to us on 22 May.1 These comments, which I have put together, reflect the views of the relevant people here and have been discussed with Mr. Bundy. The nature of our comments is such as to make it unnecessary at this time to attempt a detailed examination of the language of the document.

I will be glad to answer any questions, or discuss the comments with you at your convenience.

Cordially yours,

Carl Kaysen



  • Draft of Proposed “Military and Related Aspects of Basic National Security Policy”

The fourteen sections of the document (labeled A-N in the longer explanatory part) appear to fall into two quite different parts. Sections E, F, H, and N (corresponding to pages 4-6, 6-12, 14-22, and 30, respectively, of the shorter document) contain a basic revision of military policy and doctrine for both general and local war. Their substantive content deals with problems that are primarily military in content and that fall [Page 103] within the operating responsibility of the Defense Department. The other sections deal with a wide variety of problems ranging from a statement of basic national objectives to alliances and civil defense to arms reduction. These problems are less narrowly military, and they involve the operating responsibilities of the State Department and in some cases other Departments, as well as those of the Defense Department. Further, the statement of a new national policy with respect to many of them involves the resolution of political problems which are not explicitly faced in the present draft.

For these reasons, and also because of the importance we attach to restating our military doctrine as soon as feasible, we suggest that the scope of the document be substantially narrowed to encompass just those issues dealt with in the central paragraphs listed above. When so narrowed, it could take the form of an amendment of the present BNSP, which would be followed by other amendments as soon as possible.


The new views on central war contain two important components. The first is the proposition that our initial recognition of the outbreak of central war should not result in a spasm in which we fire off everything we can in the way of strategic weapons. Rather, we should substitute a doctrine of controlled response, which emphasizes the maintenance of some reserve strategic nuclear striking capacity, and which has as its goal the limitation of damage and the avoidance of a situation in which we are inferior in military strength after the first exchange. For the single goal of “prevailing” in central war, the new doctrine substitutes a spectrum of possible goals running from superiority to stalemate, the appropriate one of which is to be sought in the light of the situation at the time.

We agree on the importance of this change, and the associated changes in the mechanism of command and control which are necessary to put it into practice.

The second component of the discussion of general war is the explicit introduction of second-strike counterforce (countermilitary) capability as an objective to be achieved by our striking force.4 The paper does not seem entirely clear on just what scope is to be given to this objective. On the one hand, it seems clear that a flexible capability for second strike response would include some capacity for hitting military targets, unless the forces were deliberately designed to preclude this, in terms of accuracy, etc. In this sense, stating counterforce capability as an objective adds nothing that is not already implied by the doctrine of flexible [Page 104] response. On the other hand, the inclusion of this capability does represent a significant change from the present doctrine of deterrence. Once we move from that doctrine, we raise the host of problems attendant on the interaction of our decisions with those of the Soviet Union. If we try to increase our second-strike counterforce capability, and by so doing, stimulate them to do the same, are we in fact adding to our security? The draft perhaps deals with this problem by implication, in terms of budget constraints at the present level. It has been argued that counterforce capability that covers more than a part of the enemy’s striking power is at best a transient achievement. Enemy response to our own buildup, plus the real intelligence problems of targetting the force of the other side, make it unclear as to just how much counterforce capability we have at any time and thus, since we cannot really rely on it in choosing our actions, an attempt to build it into our strategy may be dangerous, as well as wasteful. In our judgment, the consequences of the explicit introduction of counterforce capability should be dealt with more directly, and the desired scope of this capability stated more clearly.

The draft proposes that, in local war, we place main, but not sole reliance on non-nuclear weapons. With this we are in hearty agreement. But there are two aspects of this proposition that are unclear. First, the present transition point from conventional to nuclear response—namely, any war involving more than 300-350 thousand troops—is accepted without discussion. We think this point deserves explicit explanation, including the consideration of how desirable it would be to have a larger conventional capability in order to make possible a higher transition point. Similarly, a more explicit treatment of the feasibility of local nuclear war is desirable. At least the alternative of mobilizing more non-nuclear forces might be brought in.

At present, the draft makes no explicit geographic references in its discussion of local war. It would be better to make explicit that it does not deal with the NATO area, and refer to the recent NSC paper on this subject.

The present draft follows the traditional style of BNSP papers. This style, with its heavy reliance on abstract and somewhat dogmatic statement, may be dictated by the nature of the task. The added explanatory section is a commendable innovation. It could perhaps be carried further by modifying it in two ways. First, another annex is needed which indicates in concrete and, where possible, quantitative, terms how far our present force position is from achieving the desired capabilities, and what would have to be done to go the whole way. Second, the basic explanatory material might be treated as a sort of “legislative history,” which provides various examples of how the doctrines would apply in [Page 105] specific circumstances, and thus reduce the scope for partisan exegesis of the abstract language of the NSC directive paper.5
  1. Source: National Archives and Records Administration, RG 218, JCS Records, JMF 3001 (14 Apr 61) Pt. 1, Sec 3. Top Secret.
  2. This May 19 draft is the enclosure to Rowen’s memorandum to McGeorge Bundy, dated May 22. (Kennedy Library, National Security Files, Carl Kaysen Series, BNSP 1/61-5/61)
  3. Printed from a copy that bears this typed signature.
  4. Top Secret.
  5. The most explicit reference to counterforce in the draft paper reads: “Offensive countermilitary attack capabilities, active defenses and passive defenses, supported by warning and reconnaissance systems, should be able to reduce enemy residual military capability at least to levels that will avoid the strategic inferiority of U.S. residual forces, and if consistent with other U.S. wartime objectives, give the U.S. a strategic superiority.”
  6. Other commentaries on the May 19 ISA draft BNSP include those of the JCS and Secretary Rusk. In memorandum JCSM-397-61 to the Secretary of Defense, June 12, the JCS reserved comment on general and local war strategy, but maintained that the draft gave the impression that “our security policy contemplates two entirely separate and distinct sets of forces” for limited and general war, that a precise distinction could not be made between local and general war, and that national security policy should be able to deal with China and the Soviet Union as separate entities as well as a common threat. (National Archives and Records Administration, RG 218, JCS Records, JMF 3001 (14 April 1961), Sec. 2)

    For Rusk’s comments, see Document 35.