Chapter 13

Can the Intelligence Community Keep
Pace with the Threat?
Jeffrey T. Richelson

On November 25, 1991, President George Bush signed National
Security Review Directive 29, "Intelligence Capabilities 19922005." The directive required over twenty government agencies
to specify their intelligence requirements and priorities for the
next thirteen years. Among the highest priorities specified were
intelligence on the proliferation of nuclear, chemical, and biological weapons.!
Few, if any, would disagree with the proposition that monitoring nuclear proliferation activities should be a major task for the
U.s. intelligence community in the years ahead. 2 In addition to
monitoring those countries that have not acknowledged their nuclear weapons capability (e.g., Israel, Pakistan, India), there is
the need to monitor those who, at various times, seek to develop
or acquire such a capability (e.g., North Korea, Iraq, Iran, Libya).
At the same time, many would also point to the post-Gulf
War discoveries concerning the Iraqi nuclear program, as well as
internal intelligence community disputes over Iranian nuclear
intentions, as troubling signs that the U.S. intelligence community may not be able to adequately monitor such nuclear weapons
programs. 3 Among the gaps in U.s.-UN knowledge about the
Iraqi program was the identity of the program's director. In July
1991, five months after the end of the Persian Gulf War, the UN
officials responsible for destroying Iraq's nuclear weapons were

the deputy head of the Iraqi Atomic Energy Commission (IAEC). However. The complex.S. the purpose of another installation (at Furat) that was secretly building centrifuges for enriching uranium to a weapons-grade level was apparently not discovered until after the war. Likewise."8 Further. The Iraqi development of the huge.? In the face of such post-Gulf War discoveries. Although it is impossible. and other Western intelligence agencies for years. Despite their mandate to conduct intrusive inspections. 6 But possibly the biggest surprise concerning the Iraqi nuclear program was its reliance on calutrons. It wasn't until September that the UN inspectors discovered secret Iraqi documents that indicated that the program's director was JaHar Dhia JaHar. 5 Nor was the identity of the program's director the only aspect of the Iraqi program that escaped U. there are reasons to believe that the situation is not as bleak as it might appear. Although the inspectors had been aware of JaHar's existence and position in the IAEC since July.s. to precisely assess how well the U.292 JEFFREY 1'. The significance of the Al Atheer scientific research installation. which was not linked to the Iraqi program until a week before the end of the war.s. "The revelations about Iraq's clandestine nuclear activities also raise questions about the adequacy of nuclear intelligence gathering by the United States and other concerned states. RICHELSON searching for an unidentified "mastermind" who they believed had overall charge of the Iraqi program. a complex of buildings about forty miles south of Baghdad. was misunderstood. the experience of the UN inspection team in Iraq could be taken as another indicator of the great diHiculty of monitoring a foreign nuclear weapons program. they had dismissed the possibility that he was in overall charge of the nuclear weapons program. possibly a foreigner. The belief in an unknown mastermind. one nuclear proliferation expert observed. intelligence community. considered to be an outmoded technology. inefficient devices went undetected by U. the UN inspectors often found themselves involved in a game of hide-and-seek with the Iraqis. and the aid of the U. turned out to be the nerve center of the Iraqi nuclear program. intelligence community might be able to monitor foreign nuclear weapons programs. intelligence. resulted from the perception that none of the Iraqi scientists who had been interro4 gated had a full grasp of the complex program.s. it is possible to examine the challenges faced . at least on an unclassified basis.

"If you identify 10 people and they end up all working at some place labeled a dairy research institute. Reactors discharge heat. it is also vital to monitor what leaves the facilities-specifically. You can remove fissile material. you'd want to go out and find out what they were doing. 9 Such individuals are the one element that are likely to survive a concerted attack on a nuclear weapons program and provide the basis for regenerating a damaged program. Maurizio Zifferero. Thus. the head of the inspection operation of the International Atomic Energy Agency (IAEA). but you cannot take out the know-how. of course. observed. You can destroy instruments. David Kay."l! A second key element is. The function of only some of the facilities is apparent. One key element is the thousands of skilled scientists and technical workers employed in any program. and nuclear test sites. who led the UN inspection teams in Iraq. 12 A crucial question concerning these facilities is exactly what is inside them-how sophisticated is the equipment. uranium mining sites and mills. Secret Iraqi documents seized by the UN inspection team in the summer of 1991 indicated that more than ten thousand nuclear workers. the United States did not discover the role of the Furat centrifuge production facility until after the war. reprocessing facilities that extract pluto- . as noted above. and scientists were involved. centrifuge production facilities." l0 Monitoring the activities of such individuals can provide valuable clues to the course of a nuclear program. Collection Targets Any nation's nuclear weapons program consists of several distinct elements. intelligence community in effectively monitoring such programs and to explore ways in which the probability of successful monitoring can be (or possibly is being) increased. enrichment facilities. noted: "You can remove the equipment. It would be helpful to begin by examining the different collection targets associated with foreign nuclear weapons programs.Lan tfle intelligence Lommumty Keep Face with the Threat? 293 by the U.S. Thus. and what are their capabilities? Just as it is important to investigate what is inside the facilities. communications and emissions. from their appearance or emissions. laboratories. Successful monitoring may help answer some of the questions as to what is inside a facility. the facilities-which may include research and power reactors. weapons assembly and storage sites. technicians.

16 A recent concern is the smuggling of nuclear material stolen from former Soviet and East European nuclear facilities. or groups that supply key technologies and material. Thus. unacknowledged and emerging nuclear statesincluding Pakistan. In addition to people and facilities located within the prolif~ erating nation. And. Africa (uranium ore concentrate).5 into the air. PTB. an investigation by the Italian Service for Information and Military Security (SISMI) identified twenty-two companies that "might have benefited from the financial operations that were 'steered' by the Banca Nazionale del Lavoro (BNL) in Atlanta . In 1990. Pakistan acquired the special steel electronics and processing vessels needed to produce nuclear weapons material from German firms such as NTG. Iraqi agents attempted to illicitly acquire U. Iraq.-made military-grade capacitors. India. France (research reactors). West German investigators uncovered a Pakistani smuggling network that apparently assisted Pakistan's uranium-enrichment efforts by providing specialized containers for the transport and storage of uranium hexafluoride. for nuclear programs. Although the material so far seized by authorities would be of little . and Israel-have shown an inclination and ability to acquire crucial material and technologies by illicit means. Switzerland (metal casings)."]5 Additionally. Israel and Pakistan have sought to illegally acquire krytrons from sources in the United States. r. Iraq received key nuclear-related equipment and material from Britain (plutonium).acilities also must communicate with some related facilities and the nuclear program's nerve center. and the United States (power supply units). sometimes covertly or unknowingly. there are the foreign governments.294 h'''''''H'V T RICHELSO\I niurn emit krypton~R. Finland (copper coils). Thus. corporations. and enrichment plants leak radiation into the surface.S. Thus. South America (uranium ore concentrate). ltaly (plutonium separation utility). 14 Various international banking institutions also playa significant role in the proliferation effort by funding and directing the financial activities of companies involved in proliferation activities. Japan (carbon fiber). in 1989. and Leybold-Heraeus. 13 Cerman corporations have provided key nuclear technology not only to Iraq but also to Pakistan. It also acquired a multimillion-dollar plant to convert natural uranium to gaseous form-······an essential step in enriching it to weapons grade.

communications intelligence (COMINT). imagery intelligence (IMINTL signals intelligence (SIGINT). Thus. 1'7 Just as there are a variety of targets. Collection Methods and Strategies Various means exist to gather intelligence on nuclear weapons programs-including nuclear intelligence (NUCINT). which are sufficiently large to be detectable at long distances. the activity does create a new concern. but with how many warheads they were producing and the characteristics of those warheads. the outflows of hot water from the Savannah River production reactors are easily visible in thermal infrared photographs taken from an altitude of 1. Fuel reprocessing plants can be monitored by the releases of krypton-85 gases.2 kilometers and are so broad that they are also detectable from satellite altitudes. NUCINT sensors are of much lesser value when the questions involve a nation/s nuclear intentions. Such sensors. 20 But NUCINT certainly does not playas great a role in evaluating the nuclear weapons programs of emerging nuclear states as it did in monitoring the Soviet nuclear weapons program. aircraft. 19 Plutonium production reactors can be detected by satellite infrared sensors. can be deployed on spacecraft. In the early 1950s. its efforts to ac- . there are also a number of different collection methods and strategies. ships. The United States was concerned for over forty years not with how the Soviets would attempt to acquire an A-bomb or whether they had succeeded. The constellation of nuclear intelligence aircraft. and human intelligence (HUMINT). and Sound Surveillance System (SOSUS) arrays provided a valuable means of providing answers to such questions. the krypton-85 level was the key to U. small boats. ground stations.Can the intellzgence Communzty Keep Pace wzth the Threat? 295 immediate value in developing atomic weapons. and helicopters. they can even be placed in an attache case.18 The most direct technical means of collecting intelligence on other nations' nuclear weapons programs is through the variety of sensors that fall in the NUCINT category. Naturally.S. which detect radiation and other effects resulting from radioactive sources. estimates of the Soviet Union/s plutonium production and nuclear weapons production rate.

In spite of the massive intelligence-gathering means at the disposition of several nations. suppliers." In particular the report claimed that the CIA was working "at a frenzied pace" to establish human intelligence networks in hard-to-penetrate countries such as Iraq and North Korea. In addition. particularly imagery and signals intelligence. there seem to have been serious deficiencies in the general assessment of Iraq's nuclear program. certain smuggling networks may not rely on any form of interceptable communications in conducting their operations. Further. China's nuclear exports to Argentina and South Africa . research and development.L':Jb JEFFREY L KICHELSON quire nuclear material and technology. an overview that could be extraordinarily difficult to piece together by relying on technical collection. intelligence professionals who have been stymied in their attempts to monitor and curb proliferation through satellite intelligence gathering and export controls are turning anew to developing networks of agents and informants.S. none of which give off signals detectable by such sensors. 21 It has been suggested that such deficiencies were due to an inadequate devotion to human intelligence collection and an overreliance on technical collection. human sources can report on what is inside facilities and can acquire and pass on documents and hardware of great value in helping to understand a foreign nuclear weapons program--for example. Human sources not only can provide information on intentions but also can provide an integrated overview of a program (its people. HUMINT can be applied to foreign nuclear weapons programs to provide a level of understanding that would probably be unavailable from technical collection. At least one penetration of the Chinese government has involved someone with access to information concerning Chinese nuclear relations with Pakistan. 23 Potentially. 24 It is easy to provide examples in which HUMINT was or would have been of tremendous value in evaluating the status of a foreign nuclear weapons program. or production work. That source reported on the following: 1. 22 And it has recently been reported that "U. and progress). including overhead reconnaissance and electronic intercepts. facilities. Human sources can also scoop up soil samples around a· suspected nuclear installation to be examined for traces of uranium hexafluoride. a master list of foreign suppliers.

S. 297 Chinese technicians who helped at a suspected Pakistani bomb-development site Chinese scientific delegations who were spending a substantial amount of time at a centrifuge plant in Kahuta where Pakistani scientists were attempting to produce enriched uranium Pakistani scientists from a secret facility at Wah who showed a nuclear weapons design to some Chinese physicists in late 1982 or early 1983 and who sought Chinese evaluation of whether the design would yield a nuclear blast The triggering mechanism for the Pakistani bomb design that appeared to be very similar to one used by China in its fourth nuclear tesf S Information about the Taiwanese nuclear program was also provided by an informant. which includes a breakdown of the specific function of each unit inside the Tunnel. One American official who has been analyzing Israel's nuclear capability since the late 1960s depicted Vanunu's information. 4. or Machon 2.. Further. This is an enormous operation:'27 It is important to acknowledge the potential value of HUMINT and the value of improving HUMINT collection on proliferation activities. as stunning: "The scope of this is much more extensive than we thought. 26 Aside from the significant new information provided by Iraqi defectors after the Persian Gulf War. This information indicated that the Taiwanese were in the process of building a secret installation that could have been used to obtain plutonium.. who worked in a Taiwanese research institute. Colonel Chang Hsien-Yi. 5.. provided the American intelligence community with the first extensive evidence of Israeli capability to manufacture fusion. weapons. U. 3. In addition to attempts to penetrate the actual or po- . According to The Samson Option: Vanunu's Times interview and his photographs of many of the production units in the Tunnel. Army human intelligence reports on the Pakistani presence at Chinese nuclear test sites and on Iraqi-Chinese discussions about construction of a nuclear reactor also indicate the value that can be played by human sources. or thermonuclear. the Mordechai Vanunu case certainly illustrates the potential value of HUMINT.Can the Intelligence Community Keep Pace with the Threat? 2.

according to one U. Gaseous diffusion plants are reasonably conspicuous because of their large size (and heavy power demand)due to the fact that each passage of uranium hexafluoride through a diffusion barrier increases its enrichment by only a very small percentage. 30 Satellite imagery can show the presence of completed reactors. Local conditions-including the isolation of the country as well as the thoroughness and brutality of the security service-and the nature of a country's relations with the United States may preclude sufficient successful recruitments. Thus. Imagery intelligence can and has provided information valuable in monitoring nuclear weapons programs. Whereas the United States has apparently learned "zip" about the North Korean program from human sources.s. HUMINT successes. an excessive preoccupation with HUMINT may divert attention from enhancing other activities. there have been some significant failures. there is also an obvious value to penetrating portions of the suppliers' network and smuggling operations. imagery and COMINT can provide significant hard information (and a valuable check on HUMINT) on many aspects of foreign nuclear weapons programs.. there is a real danger in viewing HUMINT as the answer to monitoring proliferation activities. as well as facilities under construction and their external characteristics. it has been reported that the CIA networks in East Germany and Cuba were heavily penetrated by those nations' security services. As desirable as it may be. satellite imagery alerted the United States to the existence of a research installation. North Korea. and Iran. and power lines. including technical collection and analysis.298 JEFFREY T RrCHELSON tential nuclear programs. intelligence official. In the 1970s..S. cooling towers. it may be impossible to attain sufficiently reliable human intelligence. which may yield significant benefits. Certainly. railroad . as recruiting a source in the North Korean program. 28 However. Thus. It should not be forgotten that along with notable U. it has learned a great deal from satellite imagery. Recruiting a source in the Pakistani nuclear program will not be as difficult. the United States did have some human intelligence sources reporting on Iraq. probably by several orders of magnitude. But he added: "With North Korea.29 In addition. roads. we had nothing. particularly from nations such as Iraq. Zip.

Similarly. Further photos showed a thirty-megawatt. reconnaissance activities. The conclusion drawn by the CIA was that the North Koreans were planning on burying the pipes between the two facilities and thus concealing their connection from IAEA inspectors. 31 In the winter of 1991. imaging satellites monitoring Iraq have detected uranium-enrichment machinery being moved around on trucks or buried to evade detection by UN inspectors.S. 33 Communications between the different elements of a nuclear program and between the different elements and higher authority are subject to interception by space and other COMINT systems. It has been noted that Pakistani officials regularly use the telephone to relay information about their nation's nuclear weapons program. 34 But clearly. imagery satellites detected North Korean workers digging trenches in frozen ground between Yongbyon's principal facility for reprocessing nuclear fuel and one of two facilities suspected of storing nuclear wastes. possibly to haul away equipment. During 1988-89. in a report circulated to several governments. U. 32 U. and storage areas. technical collection effort directed against Iraq left significant gaps. housing. 35 Thus. In 1980. have made or examined the feasibility of denial and deception efforts to hide aspects of their nuclear weapons programs from U. . the director general of the IAEA. 1979. and South Africa. And it is clear that several nations.S. additional satellite imagery revealed a second large building that seemed perfectly suited for reprocessing plutonium. noted that the North Koreans may have built a "pilot plant" to process plutonium. Hans Blix. U?la incident-it is safe to assume that there are a number of classified examples of the contribution of COMINT. gas-cooled machine that burned natural uranium. Although there are virtually no specific examples in the open literature of the contribution of COMINT to nuclear proliferation intelligence-the only exception being a report of National Security Agency intercepts of South African-Israeli communications after the September 22. the U.Can the Intelligence Community Keep Pace with the Threat? 299 trae](s.S. North Korea.s. U. a satellite returned imagery that indicated that a new reactor was going up at Yongbyon. out of satellite view. graphite-moderated. including Iraq. power grids. imagery satellites also detected trucks pulling up to the reprocessing plant before the arrival of lAEA inspectors in 1992.s. before taking the risky leap and building at Yongbyon.

The tunnels are believed to be part of a program to harden the facility against a possible attack. The increased number of imagery spacecraft in orbit is. a new heavier SICINT spacecraft is scheduled for launch in late 1994. the United States has launched approximately three Advanced KH-lls and two LACROSSE spacecraft. in addition to its visible-light imagery capability. imagery satellites detected the digging of deep tunnels around the nuclear site at Yongbyon in early February 1992. possible plans to build a nuclear reactor that would be camouflaged to defeat satellite surveillance. imagery satellite constellation in the 1987-88 period was quite limited compared with today's constellation---consisting of one KH-ll visible-light electro-optical spacecraft through October 1987 and two KH-lls through November 1988. the circumstances under which the United States can monitor certain activities or their effects-such as heat discharges. 37 At the same time.Likewise. Iraq pursued. They may also be part of a program to hide nuclear weapons components from international inspectors. Since that time. In addition.S. and the LACROSSE is a radar imagery spacecraft. the Advanced KH-ll is capable of producing infrared imagery. . But the new spacecraft have additional capabilities. significant. space SIGINT constellation has increased since 1987-88. it was discovered that Iraq had constructed underground facilities for nuclear weapons development and had buried related power lines. Due to launch failures. it should not be assumed that gaps are inevitable or that denial and deception measures cannot be defeated. nighttime construction. the size of the U.S. and activities taking place under cloud cover-have increased dramatically. 36 In the aftermath of the Persian Gulf War. U. a strict communications security program prohibits Iraqi officials from discussing the nuclear program over the telephone. with China. Further.S. the U. It has also been reported that North Korea was working on aspects of its nuclear weapons program in Bakchon County at a secret underground site designed to avoid satellite detection (and international inspection). As a result. The United States reportedly first became aware of the facilities after a North Korean diplomat defected in May 1991. In addition. in itself. with the launch of two VORTEX and two MAGNUM geosynchronous SIGINT spacecraft.

reconnaissance satellites.Also. technical collection efforts was the former Soviet Union. they certainly did not receive as much attention as they might have and as they are receiving now. 38 Until recently. technical collection systems is a function not only of their number and technical characteristics and capabilities but also of how they are employed-of human decisions concerning the areas and facilities to be covered and the frequency of coverage.S. along . the "take" obtained from constructing and deploying U. Although Iraq. the United States has recently stepped up satellite reconnaissance of Iran's nuclear-related facilities in response to evidence of a "suspicious procurement pattern" of nuclear-related technologies. In the days before the launch of the first U.S. Increased nighttime coverage (employing Advanced KH:-ll and LACROSSE spacecraft) and increased coverage in the presence of cloud cover (relying on LACROSSE spacecraft). by increasing the frequency and area of missions devoted to searching for new facilities. 39 Imagery coverage can be increased in several ways-by scheduling images of targets already on the target list to be produced more frequently. the primary focus of U. and by devoting more attention to anomalous or suspicious facilities. North Korea and other potential proliferators were not ignored in imagery and COMINT collection plans. some believed that it might be imperative to obtain photography of Soviet missile silos while thev were under construction because once construction was completed they would be camouflaged. as well as more frequent searches for new frequencies that might be associated with such programs. 40 Likewise. increased coverage of supplier communications links may reveal valuable new information. the attempts by Iraq and North Korea to construct underground facilities suggest that it may be necessary to detect some nuclear facilities in the construction stage. In addition to increased attention to the communications links of a particular proliferator. COMINT coverage can be increased by more frequent monitoring (andfor processing) of communications links known to be associated with nuclear weapons programs. including more frequent imagery coverage at night and when cloud cover is present.S. Defeating denial and deception measures may depend on a variety of strategies. Thus. Increasing imagery and COMINT coverage means increasing the probability of detecting new facilities and new developments. Although that was not actually a problem.

Aside from the multitude of potential proliferators. Improved analysis is another.. there are a number of suppliers for the resources required. This is particularly true when dealing with an intelligence problem such as proliferation.with coverage of more territory. and most "rational" method may not be the one chosen. increase the chances of detecting construction of underground facilities before they have been completed. is to focus analytic resources on the problem. the logic that "if they were pursuing a nuclear capability. it is necessary to evaluate information against a variety of models of nuclear proliferation. And for any particular path or paths chosen. and since they are not doing X. more attention to the issue would have been justified. Furthermore. given the constraints facing potential proliferators. they would do X. least costly. often including the desire to achieve the capability covertly. But improvements in collection are only one means of enhancing the ability to monitor the activities of emerging nuclear powers. The staff of the Senate Select Committee on Intelligence found only one finished intelligence community assessment of the Iraqi procurement network in the 1983-89 period. analysts had concluded that Iraq was seeking .g. they are not pursuing such a capability" may have little validity. the most direct. A fundamental condition for analytical success. which makes the intelligence problem of monitoring Soviet nuclear weapons developments simple by comparison. there are distinct alternative paths to the production of nuclear weapons. before even the proper analytic strategy. Instead. The analytical component of the proliferation monitoring effort can improve the chances of successful monitoring. Hence. As a result. 41 Clearly. the choice between plutonium and highly enriched uranium). It appears that the surprise that resulted from the discovery that Iraq was employing calutron technology to produce nuclear material stemmed from the dismissal of the method as being obsolete. Analysis Improving the ability to monitor attempts at nuclear proliferation requires more than enhanced collection operations--whether human or technical. paths that are the product of numerous choices (e.

The overall performance of u. It was subsequently discovered that Iraq was pursuing both approaches.~ v (~. the extensive nature of the program resulted in security fences being placed so far from those facilities that the fences went unnoticed.--.•. there is considerable room (and need) for improvement. That pressure. as well as to simply evaluate the meaning of data already collected. at the very least.~v"""". . 44 But as the aftermath of the Persian Gulf War showed.v"". 43 In 1977.S. However.. intelligence community to monitor nuclear proliferation activities. '>-110-. ". ~ ~' •.~ .~ •. Great Britain. then or in the future.. in 1992 the United States was able to persuade China and Argentina to cancel planned transfers of key technologies to Iran. Models of alternative proliferation paths can be (and presumably are) employed to direct collection activities. and West Germany to develop a concerted response to a possibly impending test. Conclusion Although the postwar discoveries concerning the Iraqi nuclear weapons program raise legitimate concerns about the ability of the U. following a tip from the Soviet Union. 'L'-''''"'t" ". Recognizing the variety of paths a nation may pursue to achieve a nuclear weapons capability can also improve the value and efficiency of the collection effort. imagery satellite photographed a suspected nuclear test site in the Kalahari Desert..v •. On the basis of that imagery. 42 Similarly. U. -. they are not cause for despair-for even though it may be desirable for the United States to know everything immediately.. the United States began working with France. Hence. to enrich its uranium through the use of delicate fast-spinning centrifuges that separate the heavier and lighter isotopes of ura~ nium gas.S.7 ·•. may suffice..s. Knowledge of select procurement attempts or other actions. a U.". ¥.. intelligence in the proliferation area can be improved by a number of means. ..>-- / . . this is not generally necessary to put a brake on proliferation. analysts apparently considered it unlikely that some Iraqi facilities detected by satellite reconnaissance were involved in the nuclear weapons program because of a lack of visible security around the facilities.-. may have discouraged a South African test in the Kalahari.. and measures to block such procurement attempts or actions.S.

For in addition to improving HUMINT collection.JV<t JnHKEY L l'(ICHELSON Improving HUMINT collection is one of those means. analysts also need to consider the numerous alternative paths that can be used to produce nuclear weapons. Remarks by Director of Central Intelligence Robert M. it should not be seen as the answer. and financial communications links known to be associated with nuclear weapons programs would increase the probability of discovering relevant information. intelligence should be able to improve the effectiveness of technical collection. for such coIIection may provide information that would be of tremendous value and that is unobtainable from other sources. imagery satellites and the increased number of those satellites in operation. But beyond improvements in sensor capabilities and the number of spacecraft in orbit. p. it is also necessary to improve imagery collection strategies-by devoting more time to (1) searching for new facilities. Notes 1. some improvement would come just from an increase in effort-in other words. More frequent monitoring and processing of nuclear weapons programs. suppliers. A near exclusive focus on improving HUMINT may divert attention from other areas of possible improvement. 2. But it will also be necessary to make better use of the analytical resources already available. Technical collection has already improved due to the increased sensor capabilities of U. However.S. among the most important and necessary improvements are those that lie in the area of analysis. Finally. the hiring of more analysts or the devotion of more analytical time to proliferation problems. 1992. Boston. as well as analysis. The development of an improved broad-area search capability would further enhance imagery capabilities. Increased searches for relevant communications could also produce valuable information. COMINT collection can also be improved in similar ways. and (3) reexamining known nuclear facilities. (2) examining suspicious or anomalous facilities. Better integration of the analytical effort presently being conducted by several distinct intelligence organizations may produce better analysis. As with collection activities. U.S. Massachusetts. In setting intelligence collection requirements and in analyzing data. . Gates to the Association of Former Intelligence Officers. November 14.

Is Building Up a Picture of Vast Iraqi Atom Program. 15. Says It Missed 2 A-Plants. SISMI." New York Times. 1991. Scarborough and Gertz. August 8." Washington Post. "Iraq's Nuclear Quest: Tentacles in Four Continents. "U. October 20. 1990. "Nuclear-Related Plant Discovered in Iraq. Leonard Spector. "U. 1990). Rowan Scarborough and Bill Gertz. October 1.S. A36." 13. AI. A3. pp. "U. November 16. Schmitt. Intelligence. pp. 5. p.Can the Intelligence Community Keep Pace with the Threat? 305 2. pp.s. AI.s. and the Z (Special Projects) Division of Lawrence Livermore National Laboratory. Paul Lewis. May Destroy the Hardware. 1991. "For Sale: Nuclear Contraband." n. A9. 3. "CIA Says Iran Makes Progress on Atom Arms. Strengthening IAEA Safeguards: Lessons from Iraq (Stanford: Center for International Security and Arms Control. pp. 1991. Elaine Sciolino. Jeffrey Smith. 14. "Saddam's Nuclear-Weapons Dream". such as businessmen and travelers. 1991. . 3. R. A44-A4S. Gary Milhollin. ES." 11. October 13. Leonard S. Smith and Frankel. 10. "Saddam's Nuclear-Weapons Dream. Eric Schmitt. 1." New York Times. Cl.s." New York Times. 1989-1990 (Boulder: Westview. "U. "Threats in the Middle East. pp. 16." New York Times. May Destroy the Hardware. December 23. p. the Department of Energy Office of Intelligence. October 8. the Defense Intelligence Agency." New York Times. pp. 1991. pp. intelligence community involved in monitoring foreign nuclear programs are the DCI Joint Atomic Energy Intelligence Committee. "Asia's Nuclear Nightmare: The German Connection.1992." Washington Post.d. A8. A6. 1992. "U. It should be noted that HUMINT comes in two varieties-overt and clandestine. All. "U. the Central Intelligence Agency. 11-14.181-98. p. Spring 1992. 1991. 1991. 1990. June 10. A6. Anthony Fainberg. Aides Discover Atom Arms Center Concealed by Iraq. 10.N. 9. p. Michael Wines." 12. AI." Orbis. 3334. A30. "Subject: BNL Affair-Atlanta Branch. Smith and Frankel. September 22. the Air Force Technical Applications Center. The most prominent organizations in the U.N. but Iraqi Nuclear Workers Remain. November 30. Stanford University. "Saddam's NUclear-Weapons Dream. who may have relevant information. 1991. Spector with Jacqueline R. Steve Call. October 11. November 29. "Iraq's Nuclear Program Shows the Holes in U. Lewis. "U. 29. pp. Smith.N. Overt collection includes interviewing selected individuals. 4. R. Michael Wines. "U. 8. p. pp.N. p. p. Officials Seek Mastermind in Charge of Iraq's Nuclear Effort. Paul Lewis. 18." Washington Times. 1993). Aides Discover Atom Arms Center". C2. July/August 1992. David Albright and Mark Hibbs. 6. AI. 7. Nuclear Ambitions: The Spread of Nuclear Weapons." New York Times. "Iraq's Quest for the Nuclear Grail: What Can We Learn?" Arms Control Today. Says It Missed 2 A-Plants in Iraq." Washington Post. Elaine Sciolino.S. 17." New York Times.N." Washington Post. Jeffrey Smith and Glenn Frankel.

Intentions are detectable by a variety of collection methods. 22.: Pergamon-Brassey's. and should. The Samson Option: Israel's Nuclear Arsenal and American Foreign Policy (New York: Random House. So Do CIA Recruits' Resumes. 377. with communications intelligence being a very significant means. 1985. Seymour M. The standard argument for the benefits of HUMINT is that it can provide information on intentions. C7. "Memorandum for Director of Intelligence. Strengthening IAEA Safeguards." Washington Post. AI. A3. January 12. although intelligence analysts may. SOOth MI Brigade. Tyler. Joanne Omang. p. The unique ability of human sources to acquire documents. Thus. Hersh.s. seek to determine Iran's present nuclear intentions. actions. Under what circumstances the United . RG 341. B2. Hafemeister. Janeway. p. At." New York Times. 21.SON 19. Army Operational Group. Tyler and Joanne Omang. pp. November 9. 1985." Washington Post. 338-88. 26. November 3. Gordon. Frank Von Hippel and Barbara Levi. November 22." Washington Times. which technical collection systems cannot provide. Dagger." Aviation Week and Space Technology. 20. Von Hippel and Levi. October 23. 1992. actions in attempting to limit Iran's acquisition of nuclear technologies--given the nature of that regime and the fact that intentions can change overnight. or soil samples seems to be a far better example of its potentially great value. 1992. AI.S. 27. 1985. Entry 214. D. Files 4-4459 through 4-4592. This is a fairly poor argument for the value of HUMINT. Va. Patrick E. it is not clear that their conclusions should influence U. Subject: (Unci. AS. 1986 (SanitizedlDeclassified). 198. hardware. A15. 24. "Pakistani Use of Chinese Nuclear Weapons Test Facilities. "Controlling Nuclear Weapons at the Source: Verification of a Cut-Off in Production of Plutonium and Highly Enriched Uranium for Nuclear Weapons. pp. Stephen Engleberg and Michael R. knowledge of intentions may have only a limited impact on U. 1985.." November 26. Arms Control Verification: The 1echnologies That Make It Possi/lle (McLean. p." June 19. December 28. Washington. A19. Rodman D.306 JEFFREY T RICHEI.) Determination of Proper Security Classification. Overt HUMINT collection may prove fruitful for some segments of the international suppliers network. In addition. Military Reference Branch. 377. March 23. "Nuclear Pact with China Wins Senate Approval. "Nuclear Exports to China?" Washington Post." Washington Post. David W.S. Fainberg. 1986. pp. 25. 1991). "A Few Spoken Words Sealed Atom Pact. at p. 1991 (SanitizedlDeclassified). A20-21. DCS/O. 20-22. "Help Wanted: Bring Cloak.C. and Peter W. 1954." Washington Post. Jack Anderson and Dale Van Atta. "Taipei Halts Work on Secret Plant to Make Nuclear Bomb Ingredient. "As Intelligence Needs Change. pp. Griffin." in Kosta Tsipis. p. 1992." May 12. James Adams. National Archives and Record Administration. Patrick E. 28. Fulghum. eds. February 9. p. 1986). "Controlling Nuclear Weapons at the Source. pp. U. David A. p." p. "Nuclear Power Plant Development Plans. "China-Iran Nuclear Link Is Reported. "Advanced Arms Spread Defies Remote Detection. and particularly regarding proliferation. 18. 23.

U." Washington Post." p. Decisions concerning the targeting of space imagery and space signals intelligence systems are made by the Dcr Committee on Imagery Requirements and Exploitation (COMIREX) and the SIGINT Overhead Reconnaissance Subcommittee (SORS) of the DCI SIGINT Committee.375. July 29." Presumably.Can the Intelligence Community Keep Pace with the Threat? 307 States would attempt (or has attempted) to recruit clandestine sources in German. 28ff. 128-29. Korea Goes Underground with Nuclear Plants. "South Africa's Sixteen-Year Secret: The Nuclear Bomb. "Iraq's Nuclear Program Shows the Holes in U." Washington Times. "Stasi Files Reveal CIA Two-Timers." Wall Street Journal. 1992. "Journey to Isolation". the United States sharply stepped up its technical collection operations directed at Iraq. 1994). November 15. B15. 1991. A12. Bill Gertz. or other West European corporations makes for interesting speculation. p. Burrows and Robert Windrem. p. "North Koreans Pursue Nuclear Weapons." Washington Times. VI. 38. 32." Newsweek." Blix meant in a location not likely to be the subject of coverage. 39. 29. 1989. Sanger. Michael Breen. 1991. February 21." Washington Times. Italian. "Controlling Nuclear Weapons at the Source. making it easily accessible to all U. All. "Satellite Spots Iraq Burying Atomic Gear. 36. AI. 1993. "Cuban Defector Impeaches CIA Spies. "Journey to Isolation. April 27. Sanger. Von Hippel and Levi. Sciolino. 33. Fialka. pp. Bill Gertz. July 10. . "Journey to Isolation". Army Operational Group. 1992. Bill Gertz. by "out of satellite view. 1989." 35. July 19." New York Times Magazine. A9. 1991.s. June 27. John J. 1988. "U. May 12.S. "N. October 25. Don Oberdorfer. Intelligence. pp. AI. Sanger. 1991. Living by the Sword (Brattleboro. October 7. imagery satellites. Intelligence". "North Korea Digs Tunnels for Nuclear Arms.s. pp. "North Korea May Be Developing Ability to Build Nuclear Weapons. AI. pp. North Korea is located well within 30 to 50 degrees north latitude. p. pp. Korea and the Bomb: High-Tech Hide-andSeek. AID. "N. A16.S. 1988). Jack Anderson and Dale Van Atta. 1993. "Journey to Isolation". pp. Sanger. Jeffrey Smith. March 21. R. "Iraq's Nuclear Program Shows the Holes in U. Jeffrey Smith. (source of quotation). p. A26.S. p.: Amana Books. 28-35. A9. Sciolino." Washington Times. William E. "Saddam's Nuclear Secrets. All. One measure of the value of more intense coverage might be how much additional information the United States acquired about Iraqi nuclear programs due to increased technical collection in the period between the Iraqi invasion and the beginning of the air campaign." Washington Post." New York Times. "Nu_ clear Power Plant Development Plans. AI. p." Washington Post. 31. September 12. R. Critical Mass (New York: Simon and Schuster. Stephen Green. Shows Photos to Argue Iraq Hides Nuclear Material. 1991. 30. All. To be literally "out of satellite view/' the plant would have to be in the Arctic 37." Washington Post. 34. David E. pp. Paul Lewis. After the Iraqi invasion of Kuwait.

"U. 41.S." 44. in 1993. AI. "U.: Ballinger. "South Africa's Sixteen-Year Secret. A30. U. pp. 1991.C. Senate Select Committee on Intelligence." Washington Post. Congress. November 17. it is possible that such a test was planned in 1977 with Israeli help.S. See Smith. 43. p. p. Halted Nuclear Bid by Iran. 292.30H JEFFREY L KICHELSON 40. government and private analysts are skeptical that South Africa had enough fissionable material for a bomb in 1977. Michael Wines. 1993). Steve Coil. A8. 9. Coil. Of course. Mass.S.S. Is Building Up a Picture of Vast Iraqi Atom Program. Halted Nuclear Bid by Iran. The Intelligence Community's Involvement in the Banca Nazionale Del Lavoro (BNL) Affair (Washington.: Government Printing Office. September 27. 1984 (Cambridge." New York Times. 42. p. Leonard S. Nuclear Proliferation Today: The Spread of Nuclear Weapons. And the South Africans confirmed. 1984)." .S. given reports of close Israeli-South African links. 1992. D. "U. Spector. Some U. that two shafts had been dug at the site for a potential nuclear blast.