THE TOSHIBA CASE

• Collection: FOIA Collection
• Collection: Declassified Articles from Studies in Intelligence: The IC’s Journal for the Intelligence Professional
• Document Number (FOIA) /ESDN (CREST): 0000621387
• Document Page Count: 6
• Document Release Date: July 30, 2014
• Case Number: F-2013-02322
• Publication Date: June 1, 1990

Approved for Release: 2014/07/29 000621387TITLE:AUTHOR:VOLUME:TheToshiba CaseYEAR:1990(b)(3)(c)34ISSUE: SummerApproved for Release: 2014/07/29 000621387aroved for Release: 2014/07/29 000621387STUDIES ?1INTELLIGENCEA collection of articles on the historical, operational, doctrinal, and theoretical aspects of intelligence.All statements of fact, opinion or analysis expressed in Studies in Intelligence are those ofthe authors. They do not necessarily reflect official positions or views of the CentralIntelligence Agency or any other US Government entity, past or present. Nothing in thecontents should be construed as asserting or implying US Government endorsement of anarticle's factual statements and interpretations.Approved for Release: 2014/07/29 000621387CApproved for Release: 2014/07/29 000621387A personal viewThe Toshiba Case(b)(3)(c)As the Harvard case study suggests, theCIA has faced few analytic issues thathave put as much public and politicalpressure on it as its discovery, analysis, andreporting of the diversions of militarily criticaltechnology to the Warsaw Pact by the Japanesefirm, Toshiba. The study, however, ends in whathindsight shows vas the middle of the case. Whatfollows is an attempt to finish the narrative andto highlight some of the key lessons I learned.Further BackgroundHad the case finished where Harvard ended thestory, it would have been interesting, but not afirefight. In early 1987, Norway had admittedviolations of COCOM. By July 1987, Japan?after several months of foot-dragging?had alsoacknowledged violations. The US House of Rep-resentatives and the US Senate voted to imposetrade sanctions against the companies involved.The House and Senate versions were differentand would require negotiations in the ConferenceCommittee. The administration argued that tradesanctions against the companies were counterpro-ductive, would be perceived in Tokyo as "Jap-bashing," and a better policy would be to workwith our allies to improve COCOM. Administra-tion spokesmen and lobbyists hired by ToshibaCorporation had made progress arguing againstthe sanctions, and the issue appeared to calmdown.As the party was winding down, however, CIAbriefed National Security Council (NSC) andDepartment of State officials in August 1987about new, strong?but not conclusive?evidencelinking Toshiba Corporation, independent of the37activities of its subsidiary, Toshiba Machine,with transfers of significant microelectronic tech-nology to Eastern Europe since at least 1979.Independently reviewing the same evidence, DIAhad also reached this conclusion. This discoverywould be particularly vexing to the Japanesebecause the key to Tokyo's argument againstsanctions was that the Toshiba Corporation hadno knowledge of the illegal sales to the WarsawPact and should not be punished for the action ofa subsidiary. We prepared a paper used by theDeputy Secretary of State and the US Ambassa-dor to brief Japanese officials on the problem.Tokyo denied any wrongdoing by Toshiba Corpo-ration, and many in the administration wanted tobelieve the company, hoping that the issue wouldfade. Many in Congress, however, saw this as aweapon against Japan. The debate with theadministration, Tokyo, Toshiba, and its lobbyistson one side and the Congress on the other con-tinued for almost a year after the Harvard studyends. For most of that year, CIA was in the mid-dle. In the end, the Congress approved and thePresident signed the Trade Bill with sanctionsagainst the firms. The process taught me somelessons.Different AgendasMost consumers want intelligence that supportstheir agenda, and we were not always aware ofwho had what agenda and why. In the machinetool case?detailed in the Harvard study?manyin the Defense Department used and leaked ourinformation to argue for sharp increases in fund-ing for submarine development and construction,as well as antisubmarine technology. Early in theApproved for Release: 2014/07/29 000621387ConfidentialApproved for Release: 2014/07/29 000621387case, State and Defense officials used our analysisas a weapon to persuade Japan and Norway toimprove their export-control systems. As the casedeveloped, some on the Hill and elsewhere usedthe information to stoke the anti-Japanese firealready being fueled by the trade deficit.To support their different agendas, policymakersoften will try to accelerate dissemination of infor-mation or restrict it. Early in the case, for exam-ple, high-level State Department officials wantedto keep pressure on Japan to implement a strongexport-control system. The officials requestedthat we bypass the Bureau of Intelligence andResearch (INR) and the Japan Desk at State inpreparing a demarche based on the intelligence tospeed the process of briefing Japanese officials inWashington and Tokyo. When-State first briefedJapanese officials in Washington on both themachine tool case and on the microelectronicscase, the Japan Desk and INR were excluded. Onthe other hand, once State decided CIA informa-tion was no longer helping and was hurting itscase, these same officials began to insist that allanalysis reflect an Intelligence Community judg-ment, knowing that this would slow the process.Our customers also misanalyzed the agenda ofsome actors. When we briefed the NSC and Stateon the microelectronics case, we noted that Con-gress might use this evidence against the adminis-tration. Nonetheless, I said that we felt obligatedto inform at least our oversight committees. Thesenior NSC person on this issue, however, toldme that Congress was already getting bored withthis problem. As a result, the administrationplayed down the importance of requests we weregetting for Congressional briefings. Because wegave the same briefing to all requesters, Congressknew as much on this issue as the White Houseand was using the information to counter admin-istration arguments.The Mystery of IntelligenceThe wide range in quality and quantity of intelli-gence evidence remains a mystery to manyCaseconsumers.(b)(1)(b)(3)(n) This qual-ity and quantity overcame the preconceived biasof many policymakers. A senior Defense Depart-ment official, for example, when firstapproached by us about Kongsberg's involve-ment, argued that he knew the company and thatit was clean. Our evidence changed his mind.  (b)(1)Consumers also often do not understand the dif-ference between a legal case and an intelligencecase. Even if they understand, they will be likelyto complain when the evidence is clear but itslegal application is not. For example, a DeputyAssistant Secretary of Defense told a senior CIAofficial that CIA should only publish and briefwhat we can prove in court.Withstanding Political PressureThe public and political pressure on us wasintense.(b)(3)(n)High-level officials in Washington and Tokyocomplained about our analysis, our briefings toCongress, and the leaks in the press. In a letter tothe Deputy Director of Central Intelligence(DDCI), an Assistant Secretary of Defenseaccused us of convicting Tos"? .1 the pages of  the Washington Times.  (b)(1)a senior US diplomat accused meof attempting to undermine the US-Japan rela-tionship. Speaking to high-level US officials, aJapanese trade official labeled me the principal"Jap-basher" in Washington.To try to clear the air, key policymakers and theDDCI agreed that I would brief an NSC-chaired38pproved for Release: 2014/07/29 000621387Approved for Release: 2014/07/29 000621387Caseinteragency group on all of our evidence andanalysis. The impact of the actual briefing isdebatable. One official told me it helped clarifykey points, but another told me General ColinPowell left the room unconvinced and concernedabout the thinness of our case. According toanother high-level official, US troops had beensent into countries based on less evidence. In myjudgment, however, the meeting primarilyshowed that I had support at the highest levels atCIA?both the DDCI and the Deputy Directorfor Intelligence attended the briefing?and thatfurther public pressure on CIA was not going tochange our analysis.The impact was clear. At the interagency meet-ings I attended in the weeks following the brief-ing, the debate shifted from putting pressure onthe CIA into changing our analysis to how todevelop a strategy to deal with Congress. In addi-tion, leaks in the press on CIA evidence and39analysis dried up and the policymakers stoppedwriting to the DCI on the subject.ConclusionMany of these lessons will be important in thefuture. Policymakers and the Congress faceincreasingly complex problems; they want clearanswers that help solve problems, not complicatethe issue. But intelligence is almost never thatclear. Increasingly, intelligence is being used tosupport negotiations, monitor international agree-ments and treaties, and aid enforcement arms ofgovernment. Often we will know or have strongevidence about violators but will not be able to"prove" our case in court. When that happens,many of the experiences of the Toshiba episodeand the kind of publicity it inspired are likely tobe repeated.This article is classified  CONFIDENTIAL.Approved for Release: 2014/07/29 000621387Approved for Release: 2014/07/29 000621387?Currfidentiel? CaseEarly - Mid-1986December 1986January 1987February - March 1987April 1987May-July 1987Toshiba?A ChronologyInitial intelligence that Soviets had acquired large number ofsignificant machine tools; demarche, using intelligence, issued toJapan; allegations denied.Using more complete intelligence reporting, Defense UnderSecretary Ikle approaches Japanese Government again; usinginformation supplied by Toshiba, Tokyo rejects demarche.At COCOM high-level meeting in Paris, US announced evidencethat Toshiba and Norwegian firm Kongsberg had supplied theUSSR with machine tools, computer control systems, and tech-nology.Norway begins serious investigation of charges; Japan againdenies involvement.After reports begin to appear in US press, interest in Congressaccelerates and Tokyo initiates investigation.Japan admits violating COCOM; House and Senate approvediffering amendments to Trade Bill calling for sanctions againstToshiba and Kongsberg; Japan revises its export-control proce-dures.August 1987 - September 1987 CIA and DIA uncover strong but not conclusive evidence offurther diversions by Toshiba, involving microelectronics;Tokyo promises a complete investigation.September - October 1987 Japan replies on microelectronics cases, denying any wrongdo-ing.November 1987 In response to Congressional request, CIA briefs Trade BillConference Committee.November 1987 - February 1988 Administration opposes sanctions in Trade Bill; members ofadministration and Congress begin leaking information fromCIA and DIA briefings that bolster their case.March 1988 Administration pressure on CIA to revise assessment or revampbriefing peaks.August 1988 Congress passes Trade Bill with sanctions.Approved for Release: 2014/07/29 000621387