Foreign Relations of the United States, 1989–1992, Volume XXXI, START I, 1989–1991

1. Memorandum From the Vice President’s Assistant for National Security Affairs (Gregg) and the Vice President’s Deputy Assistant for National Security Affairs (Watson) to Vice President Bush¹

Washington, March 18, 1988

SUBJECT

• START: Much Tougher than INF

The elimination of all INF missiles means there is no requirement to collect intelligence on, and monitor numbers of missiles, ALCMs, non-deployed missiles, mobile missiles, test-firings, or research and development of permitted new types of weapons. Nor is there the requirement to agree on counting rules that allow us to indirectly (and imprecisely) count warheads. A START agreement will require us to do all these. START will build on INF but will have to go much beyond it in detail, intrusiveness, and difficulty if it is to effectively constrain Soviet military capability. Because we have to assume they intend to cheat monitoring becomes all the more important.

Obviously, the deeper the reductions of strategic arms the greater would be the negative effect on us of cheating. Secondly, Soviet cheating negates the effect of reductions.

The attached letter to the President from Ambassador Anne Armstrong came to us yesterday. In it she expresses the Foreign Intelligence Advisory Board’s serious misgivings about our ability to monitor the START agreement. She may be seeking an appointment with you and wanted to be sure you knew her concerns.

Her letter coincides with a vigorous interagency work program to devise monitoring provisions for a START agreement. Attached are two sensitive arms control papers which are still in draft.²

Of all the increased monitoring difficulties mentioned in our first paragraph monitoring non-deployed missiles and mobile ICBMs will be the most difficult. We would like in the following paragraphs to summarize the various monitoring approaches being considered.

First, the task is to develop methods to detect and monitor limits on legal missiles and launchers, deny illegal ones access to legal test and maintenance facilities; detect illegal missiles and launchers; and barring this, make it much more difficult and costly for the Soviets to cheat (by manufacturing, maintaining, flight testing, transporting and manning them).

Simply put, the combination of approaches would be designed to capture Soviet systems from “cradle to grave.”

Several approaches are emerging to accomplish this:

— Initial Data Exchange of numbers, locations and performance characteristics of treaty limited items (TLI).

— Periodic Updates of information.

— Baseline Inspections at entry into force.

— Accountability: provisions to determine changes of accountability during manufacturing, testing, if damaged, or if destroyed.

— Tagging: Each treaty limited item would have a tag (a fragile piece of plastic with randomly distributed colored plastic chips embedded in it) glued to it. An optical reader would read the tag anytime a TLI moved thus accounting for the item. This is a unique and high tech proposal that is still in development by Sandia National Laboratory.

— Suspect Site Inspections: Each side would conduct short-notice inspections at locations where they suspect covert production, deployment, storage or repair of TLI. This concept was discarded in the INF Treaty but will certainly be essential if START is to be ratified. Given the open and closed nature of our societies, our problem is to strike a balance between effective verification and protecting sensitive U.S. facilities. As you can imagine, this will probably be the most controversial START provision within the USG—as it is in the chemical weapons negotiations.

— Restricted Deployment Areas: Missiles and launchers would be limited to a specific geographic area (such as their Main Operating Bases—MOBs, training areas, and maintenance facilities). Provisions for dispersal areas will be problematic: we need effective verification and operational flexibility.

— Perimeter/Portal Monitoring: Entry and exit of garrison areas will be through special portals that weigh, measure, photograph, and count rail-launcher cars. As the data would be transmitted to a remote facility, there would be no routine presence of inspectors at the garrison site.

— National Technical Means: Backing all the above techniques up would be our NTM. Cooperative Measures (such as openair display of TLI) would be sought.

— Notification: Sides would notify each other of movements in advance, and of the routes upon completion.

This is a short description of possible provisions. There are far more detailed limitations and rules in the draft text.

There is another lesson here which we believe is important: we can not accept an approach that says monitoring problems are too difficult. With good minds and intense study and research solutions can be found. We’re not yet convinced that the approaches above will solve all our START monitoring concerns, nor are we convinced they will give us high confidence. But the effort is necessary if we are to be serious about arms control as a contributing element of our national security. As you know, many think the problems in monitoring a chemical weapons ban are too difficult to overcome. The US experience in START may show us that monitoring difficulties can be tackled and solved.

The draft GRIP compartment papers are attached if you want more technical details.

Attachment

Letter From the Chairman of the President’s Intelligence Advisory Board (Armstrong) to President Reagan³

Washington, February 5, 1988

Dear Mr. President:

In recent weeks your Foreign Intelligence Advisory Board has focused its attention on the intelligence issues related to the INF treaty and START negotiations. Although we have some concern about the Intelligence Community’s ability to monitor Soviet compliance with the INF treaty, the Board believes the total ban on the intermediate and shorter range missiles and the prohibition on testing such weapons make Soviet cheating more difficult and less likely over time.

In the case of a possible START treaty, however, the Board has serious misgivings. Our concerns center on the Government’s ability to monitor the agreement as well as its effort to assess the strategic impact of such an accord. We believe that the negotiating process may be ahead of our resolving key intelligence uncertainties. If these uncertainties are not lessened markedly, they could permit the Soviets to attain strategically decisive advantages in a post-START world.

There appear to be greater incentives for the Soviets to cheat under a possible START agreement and many more opportunities for them to do so. We are struck, for example, by the problem of accounting for warheads given the “counting rules” outlined in the US/Soviet Joint Summit Statement.⁴ Several Soviet missile systems will be counted as carrying fewer warheads than they actually can. In time of crisis or war, the Soviets could, undetected by our National Technical Means and unconstrained by whatever peacetime on-site inspections might have previously occurred, deploy these missiles armed to their full capacity. Combined with counting rules that underestimate the Air-Launched Cruise Missile capacity of Soviet strategic aircraft, the United States could face nearly half again as many Soviet warheads as permitted under the treaty. These additional warheads could be deployed in times of crisis with little or no prior testing. Such deployments might well provide the Soviets a militarily and politically significant advantage in a world of deep reductions.

[2 paragraphs (18 lines) not declassified]

On-site inspections cannot overcome these uncertainties. They will not tell us confidently of Soviet warhead deployments when we most need to know, nor can they solve issues of clandestine production and deployment. Although the details of the suspect site inspections agreed to in principle in the US/Soviet Joint Summit Statement need to be carefully worked out, there are limits to the compliance gains that can be hoped for from any inspection regime. Notifying the Soviets that a particular installation is a suspect site risks the intelligence sources and methods that brought the Soviet installation to U.S. attention. It also places sensitive U.S. installations at risk by opening them to reciprocal Soviet inspections. In either case, there will be an understandable reluctance on the part of the United States to demand inspections of Soviet suspect sites. We have reservations whether satellites or inspectors can confidently be expected to pierce in a timely way a determined Soviet effort to conceal illegal missile production and storage.

[1 paragraph (10 lines) not declassified]

Similarly, we worry that Soviet military capabilities that seem less central to the strategic balance today, such as Soviet Anti-Submarine Warfare and Air Defense, may weigh much heavier if nuclear weapons are sharply reduced. Finally, we believe that these specific concerns should be addressed in light of a general assessment of where Soviet strategy is heading. The key question is: from the Soviet perspective, regardless of what other objectives they might have, how does the START treaty advance their military goals?

In our opinion, these and similar intelligence uncertainties require immediate and focused review.

The issues raised above fall within the Board’s charter to provide you our best advice on matters of intelligence. By your careful appointments, however, the Board comprises a unique repository of experience in national security affairs. It includes a former Secretary of State, a former Secretary of Defense, and two former National Security Advisors among other former Cabinet members. We hesitate to step beyond our formal mandate. But if the issue warrants, we believe it our obligation to make known to you other concerns which may fall outside the letter of the Board’s governing executive order.

One such concern arises directly from the possible cheating or breakout scenarios mentioned above. We believe detailed analysis by the Secretary of Defense and the Joint Chiefs is required to determine what the net effect on the strategic balance might be under these scenarios. For different aspects of a START accord, at what level of intelligence uncertainty need we assume the Soviets may have gained a militarily significant advantage? At what level of intelligence uncertainty do the unknowns become dangerously destabilizing in times of crisis?

A second area of concern is the disposition of the remaining U.S. strategic forces under a START regime of 50 percent reductions. Have the appropriate studies been undertaken to determine what force modernization or restructuring will be required to enhance the survivability and deterrent effect of the remaining forces? Will such forces meet the objectives and criteria set out in National Security Decision Directive 13,⁵ “Nuclear Weapons Employment Policy”? And have these requirements been put up against the budget constraints faced by the Department of Defense for the foreseeable future as well as the enormous political difficulties facing any attempt to restructure or modernize U.S. strategic forces?

Mr. President, we have not raised these concerns to frustrate efforts to reach an agreement. We do so to ensure that whatever shortcomings exist with the proposed treaty be fully understood so that they can be addressed, thereby increasing the soundness of a START accord.

As always, your Board stands ready to assist you in any manner you deem appropriate to address the issues raised in this letter.

Sincerely,

Anne L. Armstrong
Chairman⁶

1. Source: George H.W. Bush Library, Bush Vice Presidential Records, Office of National Security Affairs, Donald P. Gregg Files, Subject Files, OA/ID 19863–001, START [1]. Top Secret; Sensitive; Eyes Only. Watson initialed the memorandum on Gregg’s behalf. Bush wrote in the top right-hand margin of the memorandum: “good paper. Sam: see question on page 2 of Anne’s letter ?? also p. 3 GB 3–19.”
2. Attached but not printed are two papers drafted by the Arms Control Support Group: GRIP 34 H (Mobile ICBMs), dated March 12, 1988; and GRIP 59A (Suspect Site Inspections), dated March 7, 1988.
3. Top Secret.
4. The Washington Summit Joint Statement is printed in Foreign Relations, 1981–1988, vol. XI, START I, Document 255.
5. Scheduled for publication in Foreign Relations, 1981–1989, volume XLIII, National Security Policy, 1981–1984, Part 1.
6. Bush wrote at the bottom of the letter: “Why are we worse off even with the problems she raises if both sides verifiably destroy X number of weapons. The verification problems, I take it, already exist so why are we diminished if 50% cuts are made.”