ARMS CONTROL COMPLIANCE AND INTELLIGENCE

• Collection: FOIA Collection
• Collection: Declassified Articles from Studies in Intelligence: The IC’s Journal for the Intelligence Professional
• Document Number (FOIA) /ESDN (CREST): 0006002219
• Document Page Count: 8
• Document Release Date: July 30, 2014
• Case Number: F-2013-02322

Approved for Release: 2014/07/29 C06002219The impact of analysisArms Control Complianceand Intelligence(b)(3)(c)most intelligence professionals have anopinion on the extent to which intelli-gence analysis is considered in, or influ-ences, policy decisions. This opinion is usuallysomething like "if intelligence supports policy, ithas an impact; if not, intelligence is ignored by thepolicymakers." Maybe this puts the case too na-kedly, and certainly the extent to which this judg-ment holds depends largely on the administrationand the individual policymaker. It could even beargued that how our analysis is used is largelyirrelevant to how we do our work, so long as ourconclusions are not subject to policy influence andso long as we create a paper trail for use, particu-larly when the Congress begins to debate adminis-tration policy. But most of us want to know whatimpact our work is having.It is often difficult to determine to what useintelligence analysis is put. A US decision to takesome action, or refrain from some action, dependson many factors?cost, reaction of our allies, prac-tical feasibility?so even careful consideration ofintelligence judgments is only part of the equation.There is, however, one area in which the contrastis more starkly drawn. This is in the area of armscontrol.The DCI and Arms ControlThe Director of thtelligence (DCI) is en-gaged in  (b)(1) preparation oUS negotiating positions and bottom lines;   (b)(1)   monitoring with nationaltechnical means of verification, and developmentof compliance judgments. Most of the decisionsregarding treaty provisions are made by the7r(b)(3)(n)National Security Council (NSC) and are therefore"policy" decisions. Nevertheless, the DCI is aninfluential player in these decisions when theyhave an intelligence dimension, as do all decisionsrelated to verification. All this is in addition to theDCI's traditional role of protecting intelligencesources and methods.There also is an arms control role for the DCI inhis relationship with the Congress, and this role issubstantively much the same as when supportingthe executive branch. The House and Senate over-sight committees and the Senate Foreign RelationsCommittee (SFRC) are the Congressional ele-ments most often interested in the progress ofarms control negotiations and the capability of theIntelligence Community (IC) to collect data andmake assessments concerning Soviet compliancewith treaties being negotiated, as well as treaties ineffect. The oversight committees maintain a con-tinuing interest; the SFRC interest grows as negoti-ations approach their conclusion, and it peaksduring the treaty ratification hearings. Becausefinal treaty provisions may not always allow maxi-mum feasible monitoring, the Congress is especial-ly concerned about the DCI's efforts in this regard.That is, was the administration, and especially theNSC and the president, told of IC monitoringcapabilities and concerns as the negotiations pro-ceeded, was the case for greater monitorabilitymade strongly, and what were the trade-offs whenother considerations prevailed?In the early stages of preparing US negotiatingpositions, and during the negotiations themselves,it is the responsibility of the DCI representativesto protect and enhance monitoring capabilitiesthrough appropriate treaty provisions and to pro-vide judgments to the policymakers on how wellApproved for Release: 2014/07/29 C06002219gbakretStretre   (b)(3)(n)Approved for Release: 2014/07/29 C06002219we can monitor the provisions being negotiated.The function here is more than merely advisory,because it is often necessary to argue for provi-sions that enhance monitoring.Even though policy officials often nod in thedirection of "effective verification," political andresource considerations frequently get in the wayof best-case monitoring provisions. Therefore, be-cause there is never 100-percent monitoring confi-dence, and usually much less, there can exist astate of tension between the DCI and other ele-ments of the government.While monitorability always remains the principalDCI concern, as negotiations proceed other as-pects of the prospective treaty also become impor-tant. For instance, in this new era of burgeoningon-site activities, DCI representatives must payclose attention to on-site insoection (OSH orovi-sions.  Intelligence(b)(1)A treaty goes into effect after the Senate hasprovided its advice and consent, and the instru-ments of ratification are exchanged between theparties. At this point, the IC role becomes opera-tional. Using intelligence resources, the traditionalIC treaty monitoring function is to follow Sovietactivities pertaining to treaty provisions and toassess their meaning. National technical means ofverification are important resources in this regard,but other resources also contribute.OSI is a new feature of treaty monitoring. Startifirst with the INF treaty, OSI has become astandard part of the - ? a ? itcurrent fle: .tiations.(b)(1)Monitoring and VerificationOne of the lessons an intelligence officer mustlearn when first becoming engaged in arms controlactivities is the distinction between "monitoring"and "verification." Monitoring is the use of intelli-gence collection and analysis to determine whatthe Soviets are doing regarding activities associat-ed with arms control treaty provisions. Verifica-tion is the judgment of whether or not the Sovietactivities are in compliance with the treaties. Mon-itoring is collection of data and analysis of whatthe data mean. Verification is a political judgmentfor political purposes. It may have some relation-ship to the intelligence analysis results, but it alsoay not. Monitoring is performed by the IC;erification judgments are made by policymakers,primarily the NSC staff, with the advice of allagencies involved.ight be argued that the definitions present adistinction without a difference. This presumes(b)(1)gtis?rt8Approved for Release: 2014/07/29 C06002219inteigerApproved for Release: 2014/07/29 C06002219that policymakers hear what the intelligence ana-lysts say, believe what they hear, and use what theyare told without regard to politics or ideology. Thefocus of this article is narrowed to the ThresholdTest Ban Treaty (TTBT), which was signed inMoscow on 3 July 1974.A second nuclear threshold treaty, the PeacefulNuclear Explosions Treaty (PNET), was signed on28 May 1976. It exists only to close loopholes inthe TTBT, and it is not central to this article.History of the TreatiesThe principal provision of the TTBT is that eachparty undertakes not to conduct underground testsof nuclear weapons with yields greater than 150kilotons. The treaty addresses only undergroundtests, because an earlier treaty, the Limited TestBan Treaty, prohibits all nuclear tests that are notunderground.Both sides recognized two difficulties associatedwith the TTBT. The first has to do with the largeuncertainties in nuclear yield estimates obtainedthrough national technical means. These uncer-tainties have been a factor in why this treaty hasnot been ratified. The second difficulty is thatphysicists and engineers who design and buildnuclear weapons cannot with certainty estimatethe precise yield that will occur when the weaponis detonated. Even after the fact, nuclear weaponspecialists cannot determine the yield to betterthan 10-percent accuracy.In the context of this second difficulty, the sidesreached an understanding that was made part ofthe public record. The statement of understandingis that:Both Parties will make every effort to complywith all the provisions of the TTB Treaty.However, there are technical uncertaintiesassociated with predicting the precise yieldsof nuclear weapon tests. These uncertaintiesmay result in slight, unintended breaches ofthe 150-kiloton threshold. Therefore, the twosides have discussed this problem and agreed9geecet(b)(3)(n)that: (I) one or two slight, unintendedbreaches per year would not be considered aviolation of the Treaty; (2) such breacheswould be cause for concern, however, and, atthe request of either Party, would be thesubject for consultations.The US also stated that while it would not consid-er such slight, unintentional breaches as viola-tions, it would nevertheless carefully review eachsuch breach to ensure that it is not part of ageneral attempt to exceed the confines of thetreaty.There is still sufficient imprecision in these state-ments to allow reinterpretation of the definition ofcompliance. For instance, no one knows what ismeant by a "slight breach," and the interpretationis likely to vary considerably for reasons of phys-ics, application, and politics. Another obviouslyimportant loophole is the word "unintentional."The TTBT was not submitted to the Senate foradvice and consent associated with ratificationuntil July 1976, after the PNET was signed. For avariety of political reasons, these treaties still havenot been ratified, even though President Reaganresubmitted them to the Senate in January 1987.1Nevertheless, each party has indicated its intent tocomply with the treaty pending ratification, eachparty has accused the other of violating the treaty,and both parties claim not to have violated thetreaty.A number of related bilateral and unilateral occur-rences have taken place since these treaties weresigned. Comprehensive Test Ban negotiationswere started and broken off, the Soviet Uniondeclared a moratorium on nuclear testing thatlasted for 18 months, and the two sides beganNuclear Testing Experts' Meetings that lasted overa year, and which have led to the Nuclear TestingTalks, full-scale negotiations aimed at rewritingthe verification protocols so that the treaties canThe treaties with their new protocols were submitted to theSenate in June 1990 for advice and consent to ratification,which was given in September 1990. The treaties entered intoforce in December 1990.Approved for Release: 2014/07/29 C06002219?Serctegeseret?,_ (b)(3)(n)  Approved for Release: 2014/07/29 C06002219be ratified. Meanwhile, the IC continues to assessSoviet nuclear testing and the US administrationcontinues to decide which tests are violations.Teleseismic MonitoringThe only monitoring capability of any conse-quence for the TTBT is the US ability to deter-mine the yield of each and every Soviet nucleartest. The only way to do this at present is throughthe use of long-range seismic measuring instru-ments, called teleseismic monitoring. Workers inearthquake research discovered that shocks to theearth cause disturbances that radiate in all direc-tions and travel for thousands of kilometers. Thisalso is true for shocks originating from nuclearexplosions. The same kinds of instruments areused for locating and measuring the strength ofearthquakes and nuclear tests.There are three steps in teleseismic monitoring.First, an explosion must be detected. There is acontinuous background of natural and man-madedisturbances, referred to as "noise," that makesdetection difficult for low-level explosions. But foryields above about 10 kilotons detection is quiteroutine.The second step is to discriminate between nuclearexplosions and earthquakes or chemical explo-sions. There are certain differences in detail in thesignals received from earthquakes and those fromexplosions, so that such discrimination is straight-forward. Chemical explosions can be eliminatedon practical grounds, again for explosions aboveabout 10 kilotons. The mass of chemical explo-sives required for such large events is just too greatto be assembled without detection by some othersource.The final step is the estimation of yield of thenuclear event. While the uncertainty for detectionand discrimination is practically nil at large mag-nitudes of interest, particularly near and above the150-kiloton threshold, the same cannot be said foryield estimation.(b)(1)StocrtIntelligence(b)(1)The Soviet Nuclear Test ProgramThe USSR conducted its first nuclear weapons teston 29 August 1949 at the Semipalatinsk NuclearWeapons Proving Ground. A second nuclear testsite was established at Novaya Zemlya, with thefirst nuclear test held there on 21 September 1955.All tests were conducted on the surface, underwa-ter, or in the atmosphere, until II October 1961,when the Soviets conducted their first under-ground nuclear test at Semipalatinsk. After theLimited Test Ban Treaty (LTBT) went into effectin October 1963 banning all tests in the atmo-sphere, outer space, and underwater, the Soviettest program moved completely underground.(b)(1)10Approved for Release: 2014/07/29 C06002219IntoIligerApproved for Release: 2014/07/29 C06002219gebrat(b)(3)(n)(b)(1)The Compliance ReportSince 1984, the President has each year beenresponding to Public Law 99-145, which, in Sec-tion 1002, requires that:Not later than December 1, 1984, and notlater than December 1 of each following year,the President shall submit to the Congress a(b)(1)11 bug_Approved for Release: 2014/07/29 C06002219SNcet(b)(3)(n)Approved for Release: 2014/07/29 C06002219 Intelligencereport (in both classified and unclassified ver-sions) containing, with respect to the compli-ance of the Soviet Union with its arms controlcommitments, the findings of the Presidentand any additional information necessary tokeep the Congress currently informed.The report is an interagency product. It contains,for each treaty, a section describing related Sovietactivities, a history of compliance evaluation,compliance analysis, and a finding. The finding isdebated among the agencies of the government,but the final wording is decided by the NSC, and itcontains the political judgment regarding compli-ance. In the I December 1988 report, the Presi-dent found the Soviets to be in "likely" violation.This is consistent with the judgment made in  previous years  (b)(1)(b)(1)12Approved for Release: 2014/07/29 C06002219/nielligeAPproved for Release: 2014/07/29 C0600221914bwetp(b)(3)(n)(b)(1)13Approved for Release: 2014/07/29 C06002219Approved for Release: 2014/07/29 C06002219 Intelligence(b)(1)This article is classified SbS4kTgesKt 14Approved for Release: 2014/07/29 C06002219(b)(3)(n)