29. Memorandum From the President’s Assistant for National Security Affairs (Kissinger) to President Nixon1


  • SALT MIRV Limitations

Limitations on MIRV appear to be the critical issue in SALT and could determine whether there is any SALT agreement for your meeting with General Secretary Brezhnev. Your guidance is required on how to proceed on this question.

The Current Situation

Our intelligence indicates that the Soviets are pressing forward with strategic programs that will provide a MIRV capability. Unlimited deployment of MIRVs will greatly enhance the strategic posture of the Soviet Union and eventually could place us in a disadvantageous position. However, unlimited U.S. MIRV deployment can also threaten the Soviet Union. Possibly for this reason, Soviet representatives have urged that we give priority attention to MIRV limits and have implied that agreement could be reached by the time of the summit meetings.

MIRV Options

All the agencies in the National Security Council agree that we should seek to limit Soviet MIRV at the expense of limiting our own freedom of action to some extent. You need to decide which of the possible MIRV measures discussed in the Verification Panel outlined below should be advanced at this stage.2

OPTION 1: We could seek a total MIRV freeze which would bar the Soviets from any acquisition of MIRV technology or capability.

—Under this approach the Soviet Union could neither test nor deploy any MIRVs. This could preserve our strategic advantages and greatly enhance the survivability of our ICBMs. It would be verifiable with high confidence.

—Our MIRV technology and deployments would be frozen. We would halt at 350 Minuteman MIRV missiles rather than the 550 we plan and 416 Poseidon MIRV missiles instead of the planned 496. More [Page 92] important, we would undoubtedly have to forego new MIRV missile systems such as Trident.

A basic political question is whether—six weeks before the Summit—we should put forward such a one-sided approach in the face of what appears to be a Soviet effort to reach an early agreement on MIRVs.

If we do, there is likely to be an equally one-sided Soviet counter proposal: for example, that we tear down our MIRVs. The result will be a deadlock. It would be doubtful that Brezhnev could resolve this in Washington without full politburo decision.

OPTION 2: At the other end of the spectrum, a minimum MIRV agreement would be only to ban MIRV on heavy ICBMs—i.e., the Soviet SS–9. This, too, is clearly in our interest since the SS–9 poses the most massive MIRV threat. This option would not limit our options since we do not plan to deploy very large ICBMs.

—The Soviets would be free to develop both SLBM and ICBM MIRVs for their light missiles. They could deploy MIRVs on all 1100 light ICBMs.

—There would be some verification risks: MIRVs on light ICBMs could be transferred illegally to heavy ICBMs that the Soviets could, thereby, gain an unexpected heavy MIRV capability in a short period of time. Even without this, however, the Soviets could gain a substantial counterforce capability against our Minuteman ICBM force by MIRVing their light ICBMs. Nonetheless, this type of agreement has advantages over no limits on Soviet MIRVs.

OPTION 3: In between these approaches we could propose a freeze on all land based ICBM MIRVs.

—We would stop at 350 Minuteman MIRVs and the Soviets would have none.

—Both sides could have SLBM MIRVs. To balance our MIRV advantage, the Soviets could retain roughly their current advantage in the number of strategic missiles.

—This approach would make a major contribution to strategic stability and could have minimal verification risks. It would require that we stop short of MIRVing 200 additional Minuteman ICBM as we now plan. Some believe that this is too great a price to pay to stop all Soviet ICBM MIRV deployments.

Even if this approach proves unacceptable, as is quite possible, it has the added advantage of permitting you to move in either of the other directions—towards a total MIRV freeze or a more moderate numerical limit on the types of ICBMs that can be MIRVed.


The State Department supports the total freeze proposal mainly to demonstrate to the Congress that we made the effort to block the Soviet [Page 93] MIRV program. The Defense Department is opposed to putting our Trident missile program in jeopardy.

The Defense Department favors a variant on the heavy ICBM MIRV ban in which both sides would also agree to deploy no more than 550 light ICBMs with MIRVs. However, because of verification difficulties we would have to assume that all 1100 Soviet missiles are eventually MIRVed. The Defense Department would also recommend we seek reductions in overall levels—a feature that could be a part of the other approaches as well.

The third approach—the ICBM MIRV freeze—could be an advantageous starting point. It is not a totally maximalist position that might convince the Soviets we are not serious. But it also gives up little and does not jeopardize Trident—which could be the result of advancing a total MIRV freeze.

Realistically, heavy ICBM MIRV ban with equal MIRV deployment for light ICBMs, may be the best we can accomplish. The question now is, should we try to use Soviet desire to retain their current numbers in overall strategic forces to gain the real strategic advantages of maintaining our edge in MIRVs by an ICBM MIRV freeze. If this failed, we could agree to equal MIRVs on light ICBMs if we can get equal overall numerical limit on strategic forces.

Your Decision

That you authorize proposing one of the MIRV limitation options outlined in this memo to the Soviet Union.

Option 1. Total MIRV Freeze. Soviets get no MIRVs. U.S. stops all MIRVs and foregoes Trident.

Option 2. Heavy ICBM MIRV ban. Each side MIRVs equal number of light ICBMs.

Option 3. ICBM MIRV Freeze. Soviets get no ICBM MIRV, but retain numerical edge while we retain MIRV advantage. Trident is unaffected.3

  1. Source: National Archives, Nixon Presidential Materials, NSC Files, Box 889, SALT, SALT TWO–I–Geneva, April 1973. Top Secret.
  2. The last Verification Panel meeting occurred on April 30; minutes are ibid., NSC Institutional Files (H–Files), Box H–108, Verification Panel Meeting Minutes, 3–15–72 to 6–4–74, Originals.
  3. Nixon signed his approval of option 3.