292. Telegram From the Delegation to the Nuclear and Space Talks in Geneva to the Department of State1
4167. Subject: START: Is it a treaty out there or is it a mirage?
1. There was considerable disappointment here at the lack of progress at the Washington Shultz-Shevardnadze meeting. As a result, progress here in terms of treaty text has largely been confined to technical language fixes in the MOU and inspection protocol. These are important and require much effort, but do not generally involve the major substantive issues on which the Secretary would be engaged. The Soviets have not shown any particular reactions to the Washington meeting and are working along the same lines as we are. They have recently fallen off a number of previous positions in the JDT. Some of these are fairly important and long-held, such as “equal security” in the preamble. They have also within the past few days inserted the 7-year reductions period and the sublimit of 800 mobile ICBM launchers into the JDT.
Experts Group
2. The discussions in the Experts Group continue to be easily the most interesting of all the START FORA. The intelligence community is paying close attention to these, and I hope the policy agencies are as well. The Soviets have clearly accepted our challenge to address seriously the verification issues and to design a total verification package for mobile ICBMs and SLCMs (and, although we did not ask for it, for ALCMs as well). They are attempting to respond seriously to all our questions, some of which are quite intrusive. How long this discussion can be kept productive with essentially no input from our side is a serious question. When our guidance on those questions is prepared, it is important that it clearly address the extensive proposals the Soviets have made here, adding and substracting as appropriate, and not appear to have ignored what has been happening in the Experts Group. Some of their ideas are beginning to be inserted into the JDT, but most can only be understood by a careful reading of our reports. You will soon receive an important cable which attempts to summarize all this information.
[Page 1312]ALCM Counting
3. No one here understands the formulation we used in Washington—“distinguish between ALCM heavy bombers and systems not constrained by the treaty.” There are several crucial distinctions required by the U.S. position. Some involve distinctions between items which count and items which do not count, while others involve distinctions among items all of which count, but in different ways: ALCM heavy bombers/non-ALCM heavy bombers, nuclear-armed ALCMs/conventionally-armed ALCMs, heavy bombers/non-heavy bombers and heavy bombers which, count/heavy bombers which do not count (because they carry no nuclear weapons).
4. I believe the Soviets will accept our basic approach of a negotiated ALCM counting rule, as opposed to “equipped for.” However, they are not likely to accept a single number for all heavy bombers. The logical compromise, to which we should move promptly, is to negotiate an appropriate number of ALCMs for each separate type of ALCM heavy bomber. We could perhaps still achieve some discounting of U.S. ALCMs in this way, as well as keeping verification simple. Lebedev told me April 12 that the Soviets do not object in principle to the concept of attributing a number of ALCMs to heavy bombers, but only to the number we propose. Roslyakov confirmed this to me April 13 and also indicated some flexibility on ALCM range. I see no softening of their position on conventionally-armed ALCMs, however.
SLCMs
5. There is a ray of hope here. The new Soviet idea, being discussed in the Experts Group, of some sort of OSI at “arming” and “loading” points on land is not too different from an idea being discussed informally in State. I have tried to encourage them to consider their new idea as an alternative to their remote sensing and OSI on ships proposals, rather than merely a supplement. An accurate count of nuclear SLCMs could obviate the need to limit them to certain classes of ships and submarines or conduct intrusive inspecting on ships, both of which would negatively impact on our naval operations. Avoiding the specific association of nuclear SLCMs with certain classes of ships and submarines would also help us maintain our NCND policy. Since NCND is emerging as a major factor in our thinking, I suggest we tell them so—there is no indication here they are aware of its importance to us, although it has begun to appear in the press. In addition, we need to give them a clear answer on their remote sensing technique and the proposal for a demonstration. Being cute about this will be counterproductive and send misleading signals.
6. If the USG is convinced that the declaratory approach is the answer to the SLCM problem, I think we should put it on the table [Page 1313] and begin trying to sell it. I think it would be risky to save it for the endgame, where it could fail catastrophically. In fact, it might be logical to try the declaratory approach now and, if it fails, turn as a fallback to the idea of some sort of verification on land of the production and/or arming process. Another important reason for doing this is that even thoughtful Soviets who are looking for a solution to the SLCM problem believe that we intend to declare which vessels carry nuclear SLCMs. This is a reasonable conclusion for them to draw, given the very sparse discussion thus far of the declaratory approach, plus our heavy emphasis on data exchange.
Mobile ICBMs
7. We have gotten in the Experts Group both a wealth of information on the characteristics of Soviet mobile ICBMs and intelligent ideas on verification. One point which is emerging is that the regime they are describing is not intended to be generic, but is carefully tailored to their own two systems. In fact, they noted the other day that they had described the verification regime for the SS–24 rail-mobile and asked me what verification regime we have in mind for the rail-mobile Peacekeeper.
8. This system-specific approach is different than the traditional one, but does make some sense and could actually be more effective than a generic scheme which tries to apply identical requirements to diverse systems. The point is illustrated by considering the SS–25 vs. the mobile Midgetman. It would be stupid not to use the verification possibilities offered by the alleged 1-to-1 association between SS–25 tels and their garages just because Midgetman will not have such garages. It would be equally silly to think that the only way to use the garages to help in verification is to force similar structures to be built for Midgetman. I hope we will take those factors into account in designing the U.S. verification regime for mobiles. Specifically, we should use the unique features of Soviet systems and also design a package specific to our own mobile ICBMs, in addition to the general constraints which would apply to all mobile ICBMs.
Non-Deployed Systems
9. I hope the rumors are true that we are moving toward having numerical limits on only non-deployed ICBMs of types deployed in a mobile mode. The simplications in the treaty which would follow would be substantial.
10. The other shoe has finally dropped on non-deployed ALCMs and SLCMs. Although it is not yet in the treaty text, Lebedev made clear in the Experts Group that the Soviet proposal is for a numerical limit on non-deployed ALCMs at each base, expressed as a percentage of the number of deployed ALCMs at that base. The proposal for non- [Page 1314] deployed SLCMs is less specific, but also expressed as a percentage of deployed missiles.
11. A minor issue which is becoming annoying here concerns non-deployed heavy bombers. Why don’t we simply allow each side a small number (10–20?) of “non-deployed” heavy bombers to be located at test facilities, educational establishments, etc.? Both sides would understand, and neither would care, that these could not be distinguished, as far as capability is concerned, from “deployed” models. I should think the Air Force would be delighted at the prospect.
12. Thus a sensible grand compromise on all the issues related to non-deployed systems is possible:
—The U.S. drops its numerical limit on non-deployed SLBMs and non-deployed ICBMs associated with fixed launchers (the geographical and the other collateral constraints would remain). The U.S. accepts a small allowance for non-deployed heavy bombers.
—The Soviets drop their constraints on non-deployed SLCMs and non-deployed heavy bomber armament.
13. The Soviet idea of an overall limit on ICBMs and SLBMs which does not distinguish between deployed and non-deployed missiles remains in play here at an informal level and Obukhov wants Kryuchkov and me to continue exploring it (see Geneva 3187).2 Hanmer and Sackett continue to see merit in the idea and you may get some sort of message from the delegation on the subject soon. I explained the idea in some detail in my last message (Geneva 2711).3 If we are unable to make the simplifying compromise suggested above, then I believe this Obukhov/Kryuchkov idea should be considered.
Phasing of Resolutions
14. The Soviets have now proposed that reductions be carried out in two phases of 3½ years each. Unfortunately, they apparently have only an intermediate ceiling in SNDVs and this is obtained by equal percentage reductions from existing levels. I think we can get them to add intermediate ceilings on ballistic missile warheads, heavy ICBMs, etc., but only if Washington provides us a U.S. schedule of reduction counter the Soviet proposal. I would not be [garble] to that the agreed intermediate levels will all be equal nor do I think we need to insist [Page 1315] on this. Sensible reductions which drop more or less in a straight line over seven years, with a few intermediate points, should be sufficient.
Suspect Site Inspections
15. This is a tough political issue, but the Soviets are showing interest in getting it resolved. Informally, they say they hope we can avoid a game of chicken on this problem as occurred in INF. It would be good if we could reach agreement in principle on the May, since the Soviets seem ready to deal.
Armand Hammer Comments
16. Roslyakov told me he and Obukhov had dinner April 10 with Armand Hammer, who was on one of his Washington–Moscow shuttles. Hammer said that he had spoken with the President, who seemed very calm and in command. He contrasted this attitude with Gorbachev, whom he described as a “raging bull.” The President told Hammer that he did not want to rock the boat and upset the right while INF ratification is pending. However once INF ratification is completed, a big push would be made on START.
Follow Up to High Level Meetings
17. There has often been confusion here between ourselves and the Soviets, as well as some internal confusion, after a Shultz-Shevardnadze meeting. This has three aspects:
—Are new proposals or ideas floated at the high level also supposed to be tabled in Geneva?
—Are responses to such proposals and ideas supposed to be made by the other side in Geneva?
—Are U.S. talking points used by the experts also available for our use in Geneva?
I would think that, in general, the answer to all three questions should be “yes,” but there may well be things which one or both sides wish to keep in a special channel. In any case, two procedures would be most helpful:
—Before ending a high-level discussion, get a clear understanding with the Soviets what follow-up each side will make in Geneva and what will be kept away from Geneva.
—Promptly after the meeting, send us (back channel if necessary) an update of our instructions which reflects what has occurred (not just a MEMCON), so that we know how best to take account of the new situation and support the direction in which Washington is moving.
Schedule
18. Now that it is obvious that there will not be a treaty in May, the question which is assuming greater importance here than ALCMs, [Page 1316] SLCMs and mobiles combined is the schedule issue. Probably the only point on which agreement would be found among all four delegations (START and D&S, U.S. and Soviet) is that there should be a civilized recess after the summit. Appeals to humanitarian considerations are generally ineffective, although it should be noted that people came here in January having arranged their lives under the assumption that the negotiations would end one way or the other in late May.
19. The usual objection to a recess is that it would slow down the negotiations. Most people here believe the opposite is the case. If major breakthroughs are not made at the summit, our efforts here this summer will be rather fruitless. If such breakthroughs are made, and if they have the usual delphic character, a period in which both sides sort out in capitals what these breakthroughs mean and how to reflect them in treaty language would expedite progress. In addition, experience shows that new guidance is best prepared when the entire U.S. delegation is in Washington, takes the lead in drafting, and then can coordinate on the spot with the interagency apparatus. For all these reasons, most people here believe we should recess at or just after the summit and reconvene after the 4th of July weekend for the final push.
- Source: Department of State, Central Foreign Policy File, D880749-0777. Secret; Immediate.↩
- In telegram 3187 from NST Geneva, March 21, the Delegation transmitted a memorandum of conversation for Ifft’s March 17 meeting with Kryuchkov. (Department of State, Central Foreign Policy File, D880241-0674)↩
- In telegram 2711 from NST Geneva, March 9, Ifft reported on the progress of the START negotiations at the halfway point between the Washington and Moscow Summits. (Department of State, Central Foreign Policy File, D880205-0058)↩