Foreign Relations of the United States, 1969–1976, Volume E–14, Part 2, Documents on Arms Control and Nonproliferation, 1973–1976

177. Memorandum From the President’s Assistant for National Security Affairs (Scowcroft) to Secretary of State Kissinger¹

Washington, undated.

SUBJECT

• PNE Verification Panel Meeting, March 5, 1976

There will be a meeting of the Verification Panel on Friday, March 5, at 9:30 a.m. to review remaining issues in the PNE negotiations in an effort to reach agreement on new instructions, and to discuss options with respect to the TTB/PNE agreements recognizing that there is virtually no prospect that we could complete their ratification by their March 31 effective date.

Remaining PNE Issues

The Soviets recently accepted our long-standing position that all individual PNEs must be restricted to a maximum yield of 150 KT under the PNE Treaty. This effectively revises the “ground rules” under which these negotiations were begun in the summer of 1974, and represents a major Soviet concession. In addition, there is clear agreement that observers will be permitted for PNE salvos with aggregate yields greater than 150 KT. As a result, the two sides are considerably closer to an agreement, and this memorandum proposes new instructions on the remaining issues in an effort to bring the talks to a speedy conclusion. In view of the March 31, 1976 effective date of the TTBT, there is considerable urgency in this effort.

The VPWG on Test Bans has completed an analysis of these issues (Tab C) and, after extensive lobbying, a near consensus has been reached on a new position on the most important issues, which should facilitate our reaching an agreement. There are tactical differences on how far we should move initially in two of the remaining issues, but there is general agreement on the flexibility that we have on each issue. Some of the agencies may back away somewhat in the VP meeting from the VPWG consensus and argue for no movement at all at this time, preferring to wait for further Soviet concessions. It is clear, however, that we have accomplished our two major objectives—no individual PNEs above 150 KT and observers for large PNE salvos. As a result we should move quickly to conclude the agreement, giving ground where necessary on the remaining issues.

The chart at Tab B provides a useful précis of U.S., Soviet and intermediate positions on each of the major outstanding issues. The proposed position taken in the draft NSDM attached to this memorandum is highlighted on this chart. There are 24 other issues according to the Delegation’s latest count (their summary cable is at Tab E), but most of these can probably be resolved through backstopping once we break the log jam on the major issues.

Duration of the Agreement. Our present position is that the duration of the PNE agreement should be five years, as in the TTBT, with a strong linkage between the PNE verification provisions and the TTB threshold. The Soviets prefer a 15-year duration and have tried to decouple the treaties as much as possible. All agencies agree that we could accept a compromise position of a 15-year duration with a strong, explicit linkage to the TTBT and a provision for review at five-year intervals. OSD and JCS, however, would prefer that this be authorized only as a fall back at a later point.

The intermediate position is much closer to our present stance than the Soviet position and it seems unlikely that we would accomplish much with such a concession. Therefore, I recommend that we accept the Defense position on this issue to gain leverage in going to their fall back on the other tactical issue they have raised, the aggregate yield limit.

Observers Below 150 KT. The two sides agree that to reduce uncertainties it would be useful to permit observers for individual PNEs in the 100–150 KT yield range; however, the Soviets insist that this should be at the discretion of the side carrying out the PNE, while we argue that this should be an absolute right of the verifying side. All agencies would prefer that we maintain our present position. There was little support for raising the threshold to 125 KT for these observers since at that level, the Soviets could effectively preclude observer participation by only slightly misstating the planned PNE yield—e.g., 124 instead of 150 KT [1 line not declassified]. Surprisingly, most agencies would prefer to drop the requirement altogether rather than move to 125 KT.

Emplacement of Yield Measurement Sensors (SLIFER). This is a controversial issue that has led to considerable irritation on the part of our Delegation, since it appears that we will have to back away from an ad referendum agreement they had achieved on the basis of earlier guidance. This controversy reflects the uncertainties in utilizing newly developed equipment such as the SLIFER, for which relatively little data are available, in a geologically different and somewhat hostile environment.

The issue arises from the fact that in addition to gathering yield data, SLIFER also can acquire electro-magnetic pulse (EMP) data that can reveal details of device design. Consequently, while accepting SLIFER as the basic tool for on-site PNE yield measurement, the Soviets have endeavored to keep the end of the SLIFER cable as far away from their nuclear device as possible. Curiously, this doesn’t really eliminate the EMP data, although it does degrade the SLIFER accuracy.

After extensive discussion, the two sides reached agreement, on an ad referendum basis, on a formula that related the distance of the end of the SLIFER from the bottom of the device canister to the planned yield of the device. (This distance would be 3.5W1/3 meters below 20 KT and 3W1/3 above that level, where W is the yield in KT.) This formula was actually somewhat better than the one we had authorized at the time. The ERDA laboratories thoroughly examined the ad referendum agreement, however, and found that some of their earlier projections were inaccurate. As a result, ERDA, supported by Defense, asked that we inform the Delegation that their ad referendum agreement was unacceptable. They proposed an entirely new formulation in its place which would shorten the canister as a function of yield for yields below 50 KT and keep the end of the SLIFER cable a constant distance of one meter above the top of the canister.

We asked the Delegation for its views on such a proposal and they were alarmed, noting that the SLIFER would be much closer to the higher yield devices than with the ad referendum agreement. In addition, this might cause the Soviets to reinstate their request for a switch on the SLIFER cable to eliminate the EMP signal. ERDA had been quite concerned that the switch either might not work or would be highly susceptible to “bugging.” On the other hand, at lower yields, the SLIFER would be somewhat further away with the ERDA approach, since with the ad referendum agreement, the SLIFER cable actually would overlap a standard 10 meter canister for lower yields.

ACDA feels strongly (and this is my personal view) that no modification to the ad referendum agreement is needed since the cheating scenarios at low individual PNE yields are so implausible. For example, in one cheating scenario, the Soviets would announce a 400–500 KT salvo involving 20–30 devices with announced yields of about 20 KT which would all actually yield less than 5–10 KT. This might produce enough yield uncertainty to carry out a collateral event on the order of 300 KT or so (our teleseismic net would still record about a 400 KT salvo). However, such a collateral event would have a significant risk of detection through on-site monitoring activity or through satellite photography and, since the experiment would have to be hidden beneath other shots, it would be difficult to get good diagnostics from the test. In addition, the Soviets would have to explain how such a large number of proven PNE devices had simultaneously failed. (We have agreed the PNE device development must be done at the weapons test site.) Another factor is that the TTBT uncertainty is such that one could test at the weapons test site at almost the same yield with considerably less risk of detection and far better diagnostics.

To some degree, the SLIFER experience reflects a lack of verification experience in the ERDA laboratories. They tend to look at each technique in isolation and to pursue “absolute” rather than “adequate” verification as a goal. Unfortunately, the other agencies, whose best people are frequently assigned to non-PNE matters, defer to ERDA’s technical expertise. Finally, after much argument in the VPWG, we have reached agreement on a compromise formulation that would retain the ad referendum agreement above [50] KT and adopt the ERDA formulation below 50 KT. ERDA will reluctantly accept this approach although they would still prefer their proposal for all yields.

[3 lines not declassified] From a negotiability standpoint it is a step backward, but we hope that it will be negotiable in the context of the forward movement discussed below on the aggregate yield limit and on-site seismic net threshold.

On-site Seismic Network. Our present position is that we should be permitted to deploy on-site seismometers for any PNE salvos above 300 KT, whereas the Soviets insist on 700 KT. The on-site seismic network is intended to detect either an initial seismic signal or cavity collapse signal of a collateral event somewhat distant from an announced project. It can only detect the cavity collapse of a close-in collateral event (such as one beneath the array) and cannot estimate its yield. It is very difficult to quantify the contribution of this technique to our overall verification capability, but for this reason it may have considerable deterrent value. All agencies agree that we can compromise and raise this threshold to 500 KT.

Aggregate Yield Limit on PNE Salvos. Our position is that PNE salvo yields should be limited to 1000 KT, whereas the Soviets prefer 2000 KT. Naturally, the higher the salvo yield for a given uncertainty, the higher the yield of a potentially successful clandestine event. [9 lines not declassified]

In its first cut at the options paper, ERDA concluded that we could accept the Soviet position of 2000 KT. Subsequently, they have recommended 1500 KT and the other agencies concur in this figure. DOD would prefer to propose 1250 KT initially before falling back to 1500 KT, and they have informed us that 1500 KT is the maximum yield they could accept.

We believe it is important to present as forthcoming a position as possible on this issue. Having made a major concession on the exceptions issue, the Soviets are badly in need of a tangible US concession to report to their authorities, and the 1500 KT aggregate limit is the only one we can offer that is readily communicable at the policy level. In addition, a significant concession in this area together with the 500 KT seismic net threshold should ameliorate the problems caused by our “falling forward” on the SLIFER problem and assist in negotiating our preferred position on the duration and observer issues as well.

Status of High-Yield Test Program

An issue which could play an important role in the ratification decision is ERDA’s progress in completing its high-yield test program. They indicated to me in a recent briefing that any significant meteorological or technical problems at the Nevada Test Site would prevent them from accomplishing this goal and that a 60–90 day extension of the TTBT would be helpful in that event. As of March 1 everything is still on schedule and ERDA is confident that barring technical or meteorological problems, all high-yield tests will be completed by March 31. Within the last two weeks, two tests have been carried out and both “went very well” according to ERDA. [less than 1 line not declassified]:

—[less than 1 line not declassified]

—[less than 1 line not declassified]

[1 paragraph (6 lines) not declassified]

[1 paragraph (11 lines) not declassified]

Ratification Problem

As you know, even if we completed the PNE agreement by mid-March there would be no realistic prospect of completing ratification of either the TTB or PNE agreements by March 31. The VPWG has prepared an options paper on how we might proceed in this event (Tab D). If the PNE agreement has been concluded, all agencies would probably agree to abide by the TTB/PNE provisions pending ratification of the agreements, assuming Soviet acceptance of such an approach. However, if we have been unable to resolve the PNE issues during the next few weeks, there will be pressures to extend the effective date by several months with no interim restrictions on testing or, possibly, to abandon the TTB/PNE agreements altogether.

ACDA has pointed out that the Vienna Convention on the Law of Treaties requires that once we have signed a treaty, we must abide by its provisions until ratification is completed. The situation is complicated by the fact that the Vienna Convention itself is not yet in force although we have apparently been acting in accordance with its provisions. In addition, we do not know how the Soviets plan to interpret this Convention prior to its entry into force.

It seems clear, however, that while the Vienna Convention may have some effect if a PNE Agreement has been signed, it should probably be ignored altogether if we have not concluded a PNE Agreement by March 31. Article III of the TTBT calls for conclusion of such an agreement, and we could not consider ourselves bound by the TTBT if the other side is free to conduct weapons tests in the guise of PNEs.

The VPWG paper contains two sets of options corresponding to whether a PNE Agreement has or has not been concluded. In the first case, with both agreements in hand, the obvious course would be to reach agreement with the Soviets that both sides will fully comply with the TTB/PNE provisions pending exchange of instruments of ratification. This would include going ahead with data exchange and observers, if large PNEs are carried out during this period. If the Soviets balked at the PNE conditions we could suggest a moratorium on PNEs during ratification. If, in addition, the Soviets refused to go ahead with the data exchange, we could simply extend the effective date until ratification is completed with no interim restrictions.

It appears likely that most of the agencies would agree with the above approach—including ERDA and DOD—and it would be useful to confirm this during the VP meeting. The other situation, with no PNE Agreements, is considerably more difficult, however. In this case our actions would obviously be influenced by whether we had reached a serious impasse in the PNE negotiations and whether ERDA had run into trouble in completing its high-yield test effort.

Even though we expect to be able to conclude the PNE Agreement before March 31, the Soviets might unexpectedly balk at any time over what appears to be secondary issues, such as the number of observers permitted or other observer-related issues. Consequently, it would be useful to discuss this contingency and try to get some feeling for the agencies’ views.

The VPWG paper presents several options for this no-PNE Agreement case, all of which involve prior consultation with the Soviets:

—Full compliance with the TTBT, moratorium on PNEs. State and, possibly, ACDA can be expected to support this position, and it might be agreeable to the Soviets as well. However, they have a relatively small PNE salvo (four shots with an aggregate yield of less than 100 KT) in readiness along the route of the planned Kama-Pechora Canal and might object to a lengthy moratorium. (You might ask CIA to review the Soviet PNE plans.)

—Compliance with TTBT (perhaps less data exchange) and 100 KT ceiling on PNEs.

This would accommodate the Soviet PNE effort noted above, but there would appear to be no point in offering this concession to them at the outset.

—Extend the TTBT effective date with no interim weapons test or PNE restrictions.

[6 lines not declassified] A longer delay of six to twelve months would be needed to carry out further high-yield tests since additional test holes would have to be drilled. (Neither programmatic approval nor funding is currently available for this purpose.) Additional tests of this nature are described in the appendix to the VPWG paper, but all of them would be related to increasing confidence in present warheads rather than testing new systems.

For this reason, the shorter extension period might be preferable in that it would prevent the Soviets from testing any new systems or concepts they may have in mind. [4 lines not declassified]

Conduct of the Meeting

There are no new significant intelligence developments in this area, and you should go directly to a discussion of the PNE issues, with the objective of reaching agreement on the consensus highlighted at Tab B and reflected in the attached draft NSDM. You should stress that the Delegation is experiencing serious morale problems and that it is in everyone’s interest to conclude the agreement as quickly as possible, now that we have accomplished our primary objective. The discussion on the PNE issues should be aimed at quickly affirming the consensus on each issue in order to avoid digression into lengthy discussion of cheating scenarios. In short, the desired consensus would reaffirm our position on the duration and “observers below 150 KT” issues, but authorize a higher aggregate yield limit of 1500 KT and high seismic network threshold of 500 KT in return for Soviet acceptance of our modified SLIFER spacing rule.

You should then discuss the ratification issues, noting that the issue would be much easier to handle if we are able to quickly complete the PNE Agreement. In conclusion you might encourage everyone to work as cooperatively as possible in order to resolve the other PNE issues through the backstopping process.

Brent Scowcroft

1. Summary: In advance of a March 5 Verification Panel meeting on peaceful nuclear explosions, Scowcroft briefed Kissinger on remaining issues in the PNE negotiations and presented ratification options for the TTB/PNE agreements.

   Source: Ford Library, National Security Council, Institutional Files—Meetings, Box 6, Verification Panel Meeting, 3/5/76, PNE (1). Secret. Printed from a copy that bears Scowcroft’s typed signature. Tab A is a draft of NSDM 321, the final version of which is Document 181. Tabs B and E are not attached and not found. Tab C, a Verification Panel Working Group paper entitled “TTBT/PNE Negotiations: An Examination of Several Remaining Issues,” and Tab D, a Verification Panel Working Group paper entitled “Possible Options if the TTBT and PNE Treaty are Not Ratified by March 31, 1976,” are attached but not published. All brackets are in the original except those indicating text that remains classified and “[50]”, added for clarity. Under a March 4 note, Marcum sent Scowcroft talking points in advance of the meeting, commenting: “I hope you were not too surprised by the length of the VP memorandum for the Secretary. With three topics to cover and no recent policy level review, I felt that a detailed treatment would be useful.” (Ford Library, National Security Council, Institutional Files—Meetings, Box 6, Verification Panel Meeting, 3/5/76, PNE) The minutes of the March 5 Verification Panel meeting are ibid., Box 22, Verification Panel Minutes (Drafts), March 1976.