Proliferation of Weapons of Mass Destruction: Assessing the Risks

August 1993 OTA-ISC-559 NTIS order #PB94-107612 GPO stock #052-003-01335-5 Recommended Citation: U.S. Congress, Office of “kchnology Assessment, Proliferation of Weapons of Mass Destruction: Assessing the Rish, OTA-ISC-559 (Washington, DC: U.S. Government Printing Office, August 1993). Foreword

hroughout history, human beings have been able to annihilate each other without weapons of mass destruction. However, the development of such weapons has greatly reduced the time and effort needed to kill, giving small nations and even subnational groups the ability to destroy lives on a scale that few nations could otherwiseT manage. Such mass killing does not require state-of-the-art technology; the basic technologies underlying chemical, biological, and nuclear weapons date back to World Wars I and II. Now, modern technologies—and the ever-increasing flow of goods, information, and people across national borders-can place these deadly capabilities in many more hands. Occasionally, the United States may directly influence another state’s decision to pursue weapons of mass destruction. More often, nonproliferation efforts of the United States-together with other countries and international institutions such as the United Nations and the International Atomic Energy Agency—must operate indirectly. By establishing a system of obstacles, disincentives, rewards, and international norms or rules of behavior, nonproliferation measures are intended to lessen the desire for and increase the costs of acquiring these weapons. The challenge is to accomplish this objective in a world where states still threaten one another, and where military power is still viewed as the ultimate guarantor of national survival. Even so, several recent international trends offer us hope that proliferation might be slowed or even reversed. OTA has been asked by the Senate Foreign Relations Committee and the Senate Committee on Governmental Affairs, with the endorsement of the House Committee on Foreign Affairs, the House Permanent Select Committee on Intelligence, and the Senate Committee on Banking, Housing, and Urban Affairs, to assist Congress in its efforts to strengthen and broaden U.S. policies to control the proliferation of weapons of mass destruction. This report describes what nuclear, chemical, and biological weapons can do, analyzes the consequences of their spread for the United States and the world, and summarizes technical aspects of monitoring and controlling their production. (A separate background paper analyzes the technologies underlying nuclear, chemical, and biological weapons and delivery systems in greater depth.) This report also explains the array of policy tools that can be used to combat proliferation, identifying tradeoffs and choices that confront policymakers. A forthcoming report will analyze specific sets of nonproliferation policy options in detail. OTA gratefully acknowledges the contributions of many individuals, firms, and government agencies who assisted its research and writing for this report.

Roger C. Herdman, Director A dvisory Panel

James E. Goodby, chairl Robert G. Gough Marvin M. Miller Distinguished Service Professor Senior Member, Technical Staff Senior Research Scientist Carnegie-Mellon University Sandia National Laboratories Department of Nuclear Engineering Massachusetts Institute of Technology James F. Leonard, chair2 Elisa D. Harris3 Executive Director Senior Research Analyst Janne E. Nolan Washington Council on The Brookings Institution Senior Fellow in Foreign Policy Non-Proliferation The Brookings Institution Geoffrey Kemp George Anzelon Senior Associate William C. Potter Associate Division Leader Carnegie Endowment for Director Lawrence Livermore National International Peace Center for Russian and Soviet Studies Laboratory Monterey Institute of Joshua Lederberg4 International Studies Will D. Carpenter Rockefeller University chemical Industry Consultant Barbara Hatch Rosenberg John W. Lewis Division of Natural Sciences Lewis A. Dunn Center for International State University of Assistant Vice President Security and Arms Control New York at Purchase Science Applications Stanford University International Corp. Lawrence Scheinman Lee W. Mercer Associate Director Randall Forsberg Corporate Export Manager Peace Studies Program Executive Director Digital Equipment Corp. Cornell University Institute for Defense and Disarmament Studies Matthew S. Meselson Leonard S. Spector Department of Biochemistry Senior Associate Thomas R. Fox and Molecular Biology Carnegie Endowment for Director Harvard University International Peace Office of National Security Technology Stephen M. Meyer Sergio C. Trindade Pacific Northwest Laboratories President Center for International Studies 2 Massachusetts Institute of SE T International, Ltd. Alan R. Goldhammer Technology Director of Technical Affairs Industrial Biotechnology Gary Milhollin Association Director 1 Resigned Max. 22, 1993. Wisconsin Project on Nuclear 2 P=l mem~r ~ti J- 1 ~ 193; John M. Googin Arms Control Panel chair after June 1, 1993. Senior Staff Consultant 3 Resigned Jan. 29, 1S93.” Martin Marietta Energy 4 ~-officio; Ivkmbr of ~olo~ Systems, Inc. Assessment Advisory Council.

NOTE: OTA appciates and is grateful for the valuable assistance and thoughtful critiques provided by the advisory panel members. The panel does not, however, necessarily approve, disapprove, or endorse this report. OTA assumes full responsibility for the report ~d the accuracy of its contents. - -- Iv —

Preject Staff

Peter Blair Gerald L. Epstein ADMINISTRATIVE STAFF Assistant Director, OTA Project Director Energy, Materials, and International Jacqueline Robinson Boykin Security Division Thomas H. Karas Office Administrator Alan Shaw Jonathan B. Tucker Louise Staley Program Manager Administrative Secretary International Security and CONTRACTORS Commerce Program Dan Fenstermacher

v Additional Reviewers

Steve Fetter Joseph Pilat John Steinbruner University of Maryland Center for National Security Director, Foreign Policy Studies College Park, MD Studies Program Los Alamos National Laboratory The Brookings Institution Thomas W. Graham Los Alamos, NM Washington, DC International Security program, Rockefeller Foundation Lawrence Sequist Victor A. Utgoff New York, NY Office of the Secretary of Defense Deputy Director-Strategy, Forces, Washington, DC and Resources Division Institute for Defense Analyses Alexandria, VA

NOTE: OTA appreciates and is grateful for the valuable assistance and thoughtful critiques provided by the reviewers. l%e panel does not, however, necessarily approve, disapprove, or endorse this report. OTA assumes full responsibility for the report and the accuracy of its contents.

vi 1 contents

1 Introduction and Summary 1 Introduction 2 Major Findings 4 Weapon Characteristics and Comparisons 7 Implications of Proliferation 11 Proliferation Threats and Prospects 12 The Nonproliferation Policy Context 19 Conflicting Objectives 25 Conflicting Approaches 29 Technical Basis for Monitoring and Controlling Proliferation 32

2 Assessing the Risks 45 Weapons of Mass Destruction 46 Weapon Effects Compared 52 Near-Term Proliferation Threats: Suspected Proliferant Nations 63 Implications of Proliferation 69 A New Dimension to Proliferation: Risks From the Breakup of the Soviet Union 75 Appendix 2-A-Sources on Tables Listing Countries of Chemical and Biological Weapon Concern 79

3 Policy Background 83 Imposing Obstacles to Proliferation 84 Disincentives to Proliferants 93 Rewards for Abstention 98 When Nonproliferation Fails 109 Special and Urgent: Limiting Proliferation From the Former Soviet Union 111 Appendix 3-A-Costs and Benefits of Onsite Inspections for Nonproliferation Regimes 113

Index 117

vii Introduction and Summary 1

ince the end of the Cold War, the proliferation of weapons of mass destruction has become much more prominent in U.S. national security and foreign policy planning. Revelations about Iraqi, North Korean, South African,s and Israeli nuclear weapon programs, the possibility of a nuclear arms race in South Asia, and the multidimensional conflicts in the Middle East all point to the immediacy of this problem. Adding a dangerous new twist is the dissolution of the Soviet Union, a superpower armed with nuclear, chemical, and biological weapons whose successor states are wracked by economic crises and political instability. At least three main factors underlie this renewed emphasis on proliferation. First, the reduced military threat from the former Soviet Union has increased the relative importance of lesser powers, especially if armed with weapons of mass destruction. Second, certain international political and technological trends are increasing the threat to international security from proliferation. Third, new opportunities are opening for enhancing the current international regimes designed to stem proliferation. Since at least as far back as the 1960s, when it sponsored the Non-Proliferation Treaty (NPT), the United States has recog- nized that proliferation is a global problem and combating it requires high levels of international cooperation. This country has also exerted unilateral influence, successfully in several cases, to discourage proliferation; it will no doubt continue to do so. Nevertheless, placing priority on nonproliferation will require the further development and enforcement of international norms and behavior supporting that objective. Inter- national conditions today offer significant opportunities for such cooperation. 2 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

INTRODUCTION joining the Nuclear Non-Proliferation Treaty as non-nuclear-weapon states. Moreover, the Frightening as they are, weapons of mass United States and many other nations have destruction-taken here to be nuclear, chemical, forsworn chemical and biological weapons and biological weapons—represent only part of completely, even in retaliation for in-kind the world’s post-Cold War security problems. attack, by joining the Biological Weapons Diffusion of militarily useful advanced technol- Convention (with 125 parties) and the Chem- ogy, continuing conventional arms sales, and the ical Weapons Convention (with more than resurgence of hitherto suppressed regional and 140 signatories). These three treaties codify ethnic rivalries are spurring a broader problem: strong, if not yet universal, international the growth of advanced military capability among norms against weapons of mass destruction. states and sub-national groups that are potentially The states seeking these weapons today are hostile toward each other. Not only are weapons generally pursuing them covertly, attesting of mass destruction and their delivery systems to the reluctance states have to admit to such spreading, but so are advanced conventional developments. Thus, controlling weapons of weapons (e.g., those other than nuclear, chemical, mass destruction may well be feasible despite and biological), along with equipment needed to the dubious track record of past attempts to build a command, control, communication, and regulate or ban other weapons of war. intelligence infrastructure. Even “low-technology’ weapons can produce massive casualties, as This is the first report of OTA’s project on the shown by the Allied fire bomb attacks in World proliferation of weapons of mass destruction. It War II that caused up to 100,000 deaths in Tokyo describes the various weapons of mass destruc- and 200,000 in Dresden. Nevertheless, proliferation, reviews the status of their proliferation tion of nuclear, chemical, and biological weapons worldwide, and discusses possible consequences is of particular concern for at least two reasons: of their spread. It also surveys the range of nonproliferation policy measures, offering a menu The large-scale and indiscriminate nature of of tools from which a coherent nonproliferation their effects—particularly against unpro- policy could be constructed. tected civilians—differentiates mass- A separate background paper examines the destruction from conventional weapons. Mass- technical bases for nuclear, chemical, and biolog- destruction weapons make it possible for a ical weapons, along with their delivery systems, single missile or airplane to kill as many and seeks to identify opportunities to control or to people as thousands of planeloads of con- monitor their production. A forthcoming report ventional weaponry. These weapons can will provide a more complete specification and give small states or subnational groups the analysis of selected nonproliferation policy op- ability to inflict damage that is wholly tions. disproportionate to their conventional military capabilities or to the nature of the conflict in which they are used. 1 Weapons Described Unlike most categories of conventional weap- Nuclear, chemical, and biological weapons are ons, which will likely be considered legiti- commonly lumped together under the term weap- mate instruments of national self-defense for ons of mass destruction, ’ yet their effects, the foreseeable future, weapons of mass relative lethalities, and military applications are destruction engender widespread revulsion. very different. Nuclear weapons, which can be Some 150 nations have renounced nuclear more than a million times more powerful than the weapons, formalizing their commitment by same weight of conventional explosives, create Chapter 1-–Introduction and Summary 3

shock waves, high pressures, flying debris, and 9 Delivery Systems extreme heat—the same mechanisms by which To do their deadly work, agents of mass conventional explosives injure and kill, albeit at destruction have to be integrated into weapons vastly increased scale. Unlike conventional ex- (e.g., an aerial bomb, a ballistic missile warhead, plosives, however, nuclear blasts also create or even a suitcase) and delivered to their targets. neutron and gamma radiation, which can kill or Such weapons can be highly threatening without harm those exposed at the instant of detonation.1 sophisticated delivery systems. A nuclear device In addition, they can generate long-term radioac- planted by a terrorist or commando squad, or tivity in the form offallout, which can spread over delivered by a disguised cargo ship, civil aircraft, an area much greater than that affected by the or even a small pleasure boat, can kill just as many bomb’s immediate effects. In addition to produc- people as one delivered by intercontinental ballis- ing acute illness or death at considerable dis- tic missile; a given quantity of certain lethal tances from the detonation, fallout can also lead microorganisms would probably kill even more to delayed medical problems such as cancer or people if spread effectively by human agents than genetic abnormalities.2 if by a missile. In the cases of rival states Chemical agents are poisons that incapacitate, bordering each other, weapons of mass destruction mounted on even short-range means of delivery injure, or kill through their toxic effects on the can pose a major threat. Nevertheless, states able skin, eyes, lungs, blood, nerves, or other organs. to couple weapons of mass destruction to delivery Some chemical warfare agents can be lethal when systems with longer range or greater ability to vaporized and inhaled in amounts as small as a penetrate defenses can threaten more nations with few milligrams. As potent as chemical agents are, higher levels of destruction, and with greater however, biological agents-disease-causing mi- likelihood of success. At the same time, since croorganisms such as bacteria, rickettsia, and such delivery systems—taken here to be ballistic viruses-can be many times deadlier, pound-for- missiles, cruise missiles, and combat aircraft— pound. Laboratory tests on animals indicate that, generally pose greater technical challenges, they if effectively disseminated and inhaled, 10 grams are more amenable to international controls than of anthrax spores (a form of disease-inducing are less sophisticated delivery systems. bacteria) could produce as many casualties as a Of these three delivery systems, ballistic mis- ton (one million grams) of nerve agent. Toxins-- siles have attracted the most attention, both defined as toxic substances made by biological because they are difficult to defend against and organisms, or their synthetically produced ver- because they appear to be particularly suited for sions—are banned by both the Biological and the weapons of mass destruction. (They generally do Chemical Weapons Conventions. not have much military value in proportion to

* Nuclear weapons detonated at high attitude can also generate powerful radio waves (catled ‘electromagnetic pulse’ that canwreakhavoc with electronic equipmen~ but do not pose a direct human health risk. 2 In principle, nations or groups could develop rudiofogicaf weapons whose effects are similar to those of fallout from a nuclear wezpon (albeit over a far smaller area) without any of the blast effects or extreme temperatures that make nuclear weapons so devastating. Radiological weapons disseminate highly radioactive material over an area using mechanical means or conventional explosives. They resemble chemical weapons much more than nuclear weapons in their effects, since they contaminate territory and poison living organisms but do not destroy physical structures. Conventional attacks on nuclear power plants could be tantamount to radiological warfare, as the accident at Chernobyl suggests. ‘Ihe amount of fallout from such an attack could be massive, far greater than that from a “traditional’ radiological weapon that disperses radioactive material directly. Although there are as yet no documented cases of anyone trying to acquire radiologicalwmpons, the Geneva-based Conference on Disarmarn ent has an ad hoc committee charged with concluding a convention on them. Sweden has proposed that attacks on nuclear facilities be included, while the United States, France, and Germany favor dealing only with traditional radiological weapons. 4 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

their cost when armed with conventional war- U.S. territory, might become targets for weapons heads, although they can have considerable polit- of mass destruction. ical significance.) Combat aircraft also pose a The breakup of the Soviet Union presents potent threat for delivery of mass-destruction immediate threats to the global nonprolifera- weapons. They are much more widely available tion regimes. One possibility is that Ukraine, than missiles, and efforts to control their spread Kazakhstan, or Belarus will renege on their are greatly complicated by the multiple roles that commitments to return the nuclear weapons aircraft play. Cruise missiles and other unmanned stationed within their borders to Russia and, in the aerial vehicles could also be used as delivery case of Ukraine and Kazakhstan, to join the methods, although such vehicles with the range Non-Proliferation Treaty as non-nuclear-weapon and payload of typical combat aircraft or ballistic states. (Belarus has already ratified the WT.) missiles are not yet widely available. Such actions would seriously undermine the nonproliferation regime. Another danger is the MAJOR FINDINGS leakage of nuclear weapon materials or actual weapons to potential proliferants elsewhere in the world if the nuclear custodial system in Russia I The Proliferation Threat itself were to break down. Yet another concern is Those states most actively working to de- the export from former Soviet republics of velop weapons of mass destruction, although equipment, technology or expertise relevant to limited in number, are for the most part producing weapons of mass destruction. located in unstable regions of the world—the Middle East, South Asia, and the Korean peninsula. For at least the next decade, few if any I Prerequisites to Effective of these states will be able to deliver such Nonproliferation Policy weapons more than a thousand kilometers or so in If nonproliferation policy is to succeed, it a reliable and timely manner. Therefore, the must receive substantial international cooper- greatest threat posed by these states is to their ation. Cooperation is necessary because no nation neighbors and to regional stability. Despite their or small group of nations by themselves can current limitations in long-range military delivery prevent proliferation or contain its consequences. systems, however, proliferant states-at least in Cooperation is possible because many countries principle--can threaten any country on earth have come to recognize that the proliferation of using unconventional means (e.g., covert or nuclear, chemical, and biological weapons poses disguised delivery systems such as a ship or a genuine threat to all nations. However, since truck). states will not always agree on nonproliferation Proliferation poses dangers to all nations. It measures, maintaining and acting on an effective poses particular problems for the United consensus will require each participating country States. As a global power, the United States will to give up some freedom to act independently. almost certainly retain allies and vital interests Some analysts argue that containing prolifera- overseas that might be threatened by states tion in the long run will require a far deeper level possessing weapons of mass destruction. Should of international cooperation than has been achieved the United States need to defend its interests with to date, one that builds international institutions military force--whether acting unilaterally or for a much more cooperative global security under multilateral auspices such as those of the regime. Others argue that the international politi- United Nations--U.S. armed forces, and possibly cal system is inevitably anarchic and that as a Chapter 1–Introduction and Summary 5

result, the degree of cooperation needed to Strategies for inhibiting proliferation have contain proliferation cannot be achieved. four broad elements, all of which contribute to Whether or not either of these views proves existing nonproliferation regimes and form the correct, the end of the Cold War has opened up basis for future ones. For the most part, these new opportunities for cooperative nonprolifera- elements are mutually supportive, although as tion policies. One promising sign is the revitaliza- described later in this chapter, tensions between tion of the United Nations Security Council. them can arise. By emphasizing these elements in Progress has also been made with the signing of different proportion, nonproliferation policies can the Chemical Weapons Convention and the strength- be tailored to particular situations. These ele- ening of various multilateral groups that have ments include: formed for the purpose of controlling the spread ● obstacles to impede those working to ac- of proliferation-sensitive technology: the Nuclear quire weapons of mass destruction, ranging Suppliers Group, to control exports of nuclear from protection of weapon-related informa- technology; the Australia Group, to control mation, to export controls, possibly all the way terials useful for chemical and biological weap- to preemptive military attack against produc- ons; and the Missile Technology Control Regime, tion or storage facilities; to restrict traffic in missile systems and missile ● punitive measures to deter or punish prolif- technology. erants, including economic sanctions or If U.S. nonproliferation policy is to succeed, diplomatic isolation imposed on countries the United States must give it high priority. developing these weapons, and on states, With its leadership role in the world community, firms, or individuals who assist in such the United States has the opportunity and the developments; ability to mobilize international nonproliferation ● rewards to increase the attractiveness of efforts. Free of previously overriding Cold War voluntarily forgoing these weapons, such as security concerns, the United States can now attach greater priority to nonproliferation. Doing development assistance (financial or techni- so, however, is not without costs. Nonprolifera- cal) that is tied to nonproliferation; and ● tion may conflict with economic goals, as export global or regional security improvements promotion is balanced against export controls. to reduce the perceived needs for the weap- Promoting openness, transparency, and verifica- ons. tion of nonproliferation commitments, on the one The increasing international flow of techni- hand, conflicts with maintaining the secrecy of cal knowledge, high-technology goods, and national-security or proprietary information, on trained specialists is eroding the ability of the the other. Nonproliferation will also likely con- United States and its allies to withhold technol- flict with other foreign policy objectives such as ogies relevant to producing weapons of mass maintaining relations with individual states. For destruction from states of proliferation con- example, would the United States be willing to cern. Nevertheless, although technical capability sacrifice its relationship with Israel-and possi- is necessary to develop weapons of mass destruc- bly risk Israeli national survival-to pressure that tion, it is not sufficient, and it is certainly not state to give up a nuclear arsenal it believes causal. A host of nontechnical factors such as the essential to its security? How prominently should diplomatic, political, organizational, and eco- nonproliferation figure in U.S. relations with nomic costs and benefits bear on a state’s decision China, a regional power whose cooperation the to pursue such weapons. United States seeks in other diplomatic or eco- In the long run, the most effective nonprolif- nomic arenas? eration measure is to convince states that it is 6 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

in their own best interest to forgo weapons of can be fabricated indigenously by prolif- mass destruction. Reducing the incentives for erants using equipment (e.g., machine tools) seeking such weapons and raising the costs of that also has civilian applications. doing so are both important. External obstacles ● Most of the equipment needed to produce and disincentives can play an important role in chemical weapons has civilian applica- raising the costs of proliferation to a state tions. Moreover, most of the same chemi- considering it, possibly tipping the balance to- cals, or precursors, used to make chemical- wards nonproliferation. Such coercive measures warfare (CW) agents are also used in com- can also buy time for other diplomatic or political mercial products. Some agents (e.g., sulfur measures to forestall the development of weapons mustard and the nerve agent tabun) could be of mass destruction. However, they may not produced with widely available chemical- always be sufficient to stop states determined to industry equipment. The most potent nerve acquire weapons of mass destruction. agents (e.g., sarin, soman, VX) involve a process step--the alkylation of phosphorous— H Technical Aspects of Nonproliferation that is less common, but that nevertheless is Policy used in a handful of commercial products Trying to control proliferation through control- such as some pesticides and fire retardants. ling exports or placing other obstacles in the way . Virtually all the equipment underlying of potential proliferants requires technical analy- production of biological and toxin agents sis of the production pathways for making these has civil applications and has become weapons. Controls will not work if the target state widely available as fermentation technol- can readily make controlled items indigenously, ogy, and the pharmaceutical and biotech- find uncontrolled sources of supply, or develop nology industries more generally, have alternatives. Technical analysis also underpins spread worldwide. Since militarily signifi- measures to monitor the production of weapons of cant quantities of biological agents could be mass destruction, measures that are needed to produced in a short time in small facilities, evaluate potential threats as well as to formulate they could be used offensively without the verification regimes by which states can assure need for long-term stockpiles. Crude dis- each other that they are not pursuing such semination of biological agents in an aerosol weapons. cloud can be performed with commercially available equipment, such as an agricultural ISSUES FOR CONTROLLING PROLIFERATION sprayer mounted on a truck, ship, or airplane. . Obtaining fissionable nuclear weapon ma- However, developing reliable, efficient pro- terial (enriched uranium or plutonium) jectile or missile warheads for precision today remains the greatest single obstacle delivery of organisms over a target requires most countries would face in the pursuit surmounting major technical hurdles. Even of nuclear weapons. For this reason, theft or so, the United States overcame these hurdles black market purchase of nuclear material or by the 1960s. warheads from the former Soviet arsenal— or collaboration between potential prolifer- MONITORING PROLIFERATION AND VERIFYING ants whereby one provides nuclear materials COMPLIANCE WITH NON PROLIFERATION to another-would be extremely troubling. AGREEMENTS Although nuclear material production, weapon ● All facilities for producing weapon-grade fabrication, and testing require specialized nuclear material have unique features equipment, in many cases this equipment amenable to detection by intrusive onsite Chapter 1–Introduction and Summary 7

inspection. Many have distinctive signa- work can be conducted under the guise of tures that are detectable remotely, although defensive preparations. Identifying where facilities needed for some approaches to to look for evidence of biological agent material production might not be readily production is even harder than for chemical detectable. agents. Several suggestive signatures of ● Production of chemical-warfare agents biological weapon production do exist, and, can be detected through analysis of sam- if integrated effectively with each other and ples taken during an onsite inspection. with other sources of intelligence, they However, considerable access to production might make it possible to infer a weapon facilities is required to collect appropriate production capability. However, the evi- samples. Moreover, highly sensitive analyti- dence supporting such an inference may not cal chemistry techniques that are decisive be sufficient to justify claims of treaty under laboratory conditions might be less so violation before the international commu- under some circumstances in the field, par- nity, either because it cannot be publicly ticularly if the proliferant has been produc- released or because public allegations of ing related legitimate chemicals (e.g., pesti- treaty violation typically require a substan- cides) in the same facility and is willing to tially higher burden of proof than intelli- expend time, effort, and resources to mask, gence assessments. Sensitive techniques exist obscure, or otherwise explain away chemical to identify biological or toxin agents if agent production activities. Such efforts, access to a suspect site is made available. while not likely to eliminate grounds for However, such techniques alone do not suspicion, could create ambiguities or other- ensure that an effective onsite inspection wise complicate detection of chemical agent regime can be established to detect produc- production during an inspection. tion of biological or toxin weapons. Identifying where to look for evidence of covert production is probably the greatest WEAPON CHARACTERISTICS AND challenge for monitoring chemical weapon COMPARISONS proliferation, since highly reliable technologies to detect chemical agent production 1 Lethality and Military Utility from outside a facility are not currently One motivation for developing nuclear, chemi- available. Information on plant design and cal, or biological weapons is their ability to purchase of precursor chemicals may sug- destroy or interfere with military targets. More gest a chemical agent production capability, generally, however, these weapons may also be and may therefore lead to challenge inspec- sought for symbolic, deterrent, intimidating, tions under the Chemical Weapons Conven- or terrorist purposes that may not be simply tion. related to their value from a purely military ● Detection of biological and toxin agent perspective. production is particularly challenging be- Nuclear weapons, particularly at large yields cause clandestine production sites need (hundreds of kilotons or higher) are the most not be large or distinctive, because the potent means of mass destruction.3 In addition to equipment involved has legitimate civil- killing tens or hundreds of thousands of people or ian applications, and because offensive more, a nuclear weapon can obliterate the entire

3 One kiloton is the explosive blast generated by 1,000 tons of high explosive. 8 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

physical infrastructure of a large city and contam- clothing, shelters, and decontamination proce- inate a much larger area with radioactive fallout. dures. Although these weapons can contaminate Given this destructive power, nuclear weapons territory, they do not destroy infrastructure. If have been developed for strategic use against a protective measures are taken in time-which nation’s military infrastructure, its economic requires adequate warning-they can dramati- base, and even its population. In addition, the cally reduce casualties, and hence the military and nuclear powers have developed many tactical political implications, of a chemical or biological nuclear weapons for a variety of battlefield attack. Nevertheless, such weapons can still have missions. These weapons are particularly threat- military value against protected troops, since ening to concentrations of military force such as forcing troops to don protective gear impairs their tightly clustered naval groups, port or depot ability to function on the battlefield and lowers facilities, troop concentrations, or massed forces their military effectiveness. Means for penetrat- of tanks and other armored vehicles. ing protective gear would have serious implications for the military utility of chemical or Unlike nuclear weapons, chemical and bio- biological weapons. Although such means have logical agents-if detected-can be defended been examined, they have operational shortcom- against through use of gas masks, protective ings; moreover, defensive equipment is being improved to mitigate that threat. Biological weapons are so potent that under conditions favorable to the attacker, they can kill as many people as comparably sized nuclear weapons, potentially making them extremely dangerous as a strategic or terrorist weapon against dense population centers. However, their characteristics make them particularly difficult to use on the battlefield. Except for some toxins, biological agents act more slowly than chemical or nuclear weapons, taking days or weeks to achieve full effect. In addition, their effects are much harder to control or predict than those of nuclear weapons, since 1) individuals differ markedly in their sensitivity to biological agents; 2) the lethal areas created by such agents, which depend on wind and other weather conditions, are hard to predict (indeed, such agents may even be blown back upon the attacker by an unexpected shift in wind direction during a battlefield engage- ment); and 3) biological agents must be kept alive through the dissemination process and long enough afterward to infect the target personnel, but not so long as to impede future use of the area. On a pound-for-pound basis, chemical weap- Gas masks and protective clothing can shield against ons are much less lethal than either nuclear or chemical and biological weapons, but they impair biological weapons, and correspondingly greater military @activeness. amounts would have to be delivered to have Chapter 1–Introduction and Summary 9

comparable results. Indeed, it may not even be pathogens) and building the munitions and deliv- appropriate to consider them to be weapons of ery systems needed to make those materials ‘‘mass destruction. ’ Yet they can still induce militarily functional. terror, particularly among troops or civilians Since international norms remain and are being without protective gear. strengthened against proliferation of these weap- In some battlefield scenarios, chemical weapons, proliferants may very likely try to acquire ons might be no more effective than the same them secretly. Concealing potential indicators of weight of conventional high-explosive muni- the necessary activities adds to the expense and tions, even when used against unprotected people. difficulty of acquisition. With enough effort and Like biological weapons, their effects depend on resources, the magnitude and scope-and possi- variable factors such as weather and terrain, bly even the existence-of a covert weapon limiting their predictability. Nevertheless, chemi- program might well be successfully concealed; cal weapons do have tactical applications. Persis- the burden will be on the suspecting parties to tent chemical agents cart create local “no-man’ s- detect relevant indicators and to interpret their lands’ in which restrictions would be imposed on meaning accurately. The background paper ex- military operations of either side. Attacks using ties various ‘‘signatures’ that might indicate nonpersistent agents could disrupt enemy de- the presence of a clandestine weapon program. fenses but still permit attacking troops to overrun the territory soon afterwards. Some chemical agents can be used as incapacitants, either in lieu 1 Probability of Use of lethal force or in conjunction with it. Nuclear weapons have been detonated on adversaries only twice-against Hiroshima and ~ Ease of Acquisition Nagasaki in World War II. Biological weapons, Barring a shortcut, such as the direct acquisi- despite their apparent ease of manufacture and tion of nuclear materials usable in weapons, the their devastating effects, have not played a infrastructure required to produce nuclear weapons is considerably more difficult and expensive to develop than that for either biological or chemical weapons. It is also the most amenable to limitation through control of international technology transfers. Mass production of lethal chemical agents requires a greater investment than that of biological weapons, but is not nearly as expensive or challenging as production of nuclear materials. Table 1-1, drawn from the technical analyses in a separate OTA background paper, compares the relative difficulties involved in trying to produce nuclear, biological, or chemical weapons in 4 several categories. Note that the table differenti- Kurdish victims of an Iraqi chemical attack on ates between producing materials (nuclear ma- the Iraqi town of Halabja during the Iran-Iraq terials, chemical warfare agents, and biological war.

4 OTA background paper on tccbnologies underlying weapons of mass destruction in press. 10 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table l-l—Technical Hurdles for Nuclear, Biological, and Chemical Weapon Programs Nuclear Biological Chemical Nuclear materials or lethal agents production Feed materials Uranium ore, oxide widely Potential biological warfare a- Many basic chemicals avail- available; plutonium and gents are readily available lo- able for commercial purposes; partly enriched uranium dis- cally or internationally from nat- only some nerve gas precur- persed through nuclear power ural sources or commercial sup- sors available for purchase, programs, mostly under inter- pliers. but ability to manufacture them national safeguards. is spreading. Scientific and technical Requires wide variety of ex- Sophisticated research and de- Organic chemists and chemi- personnel pertise and skillful systems velopment unnecessary to pro- cal engineers widely available. integration. duce commonly known agents. Industrial microbiological personnel widely available.

Design and engineering Varies with process, but spe- Widely published; basic tech- Widely published. knowledge cific designs for producing niques to produce known agents Some processes tricky (Iraq either of the two bomb-grade not difficult. had difficulty with tabun cyana- nuclear materials can be diffi- tion, succeeded at sarin alkyla- cult to develop: tion; however, sarin quality was . Separation of uranium poor). isotopes to-high- Iy enriched uranium; . Reactor production and chemical processing to produce plutonium. Equipment Varies with different processes, Widely available for commer- Most has legitimate industrial but difficulties can include fab- cial uses. applications. rication, power consumption, Special containment and waste- Alkylation process is somewhat large size, and operational treatment equipment may be difficult and is unusual in civil- complexity: more difficult to assemble, but ian applications. ● Electromagnetic sepa- are not essential to production. Special containment and waste ration equipment can treatment equipment may be be constructed from avail- more difficult to assemble, but able, multiple-use parts; are not essential to production. ● Equipment for other processes is more specialized and difficult to buy or build. prominent role in wartime.5 Chemical weapons, in kind.6 Iraq also used chemical weapons against on the other hand, were heavily used in World its own Kurdish population. Although the threat War I and have been employed several times of Iraqi chemical weapons loomed large over since then in regional conflicts. Most recently, coalition military forces and civilians in sur- Iraq used chemical weapons during the 1980- rounding countries during Operation Desert Storm, 1988 war with Iran, resulting in some 50,000 they were not in fact used during that conflict. Iranian casualties, with Iran belatedly retaliating Nevertheless, if the past is any guide, chemical

S ~ world war II, Japan used biological agents including bubonic plague on an experimermd bk in occupied ~ reportedly MU some hundreda of Chinese civilians but also causing thousands of illnesses among its own troops (see ch. 2). Biological weapons were not used in any other theater of the war. 6‘R@mony of R. James Wocdsey, Director of Central Intelligence, before the Senate Committee on Govemrnental Affairs, Feb. 25, 1993. Chapter 1-Introduction and Summary I 11

Table I-l-( Continued)

Nuclear Biological Chemical Plant construction and Costly and challenging. With advent of biotechnology, Dedicated plant not difficult. operation small-scale facilities now capa- Research reactors or elec- Conversion of existing com- ble of large-scale production. tric power reactors might be mercial chemical plants feasi- converted to plutonium pro- ble but not trivial. duction. Overall cost Cheapest overt production Enough for large arsenal may Arsenal for substantial military route for one bomb per year, cost less than $10 million. capability (hundreds of tons of with no international con- agent) likely to cost tens of trols, is about $200 million; millions of dollars. larger scale clandestine program could cost 10 to 50 times more, and even then not be assured of success or of remaining hidden. Black-market purchase of ready-to-use fissile materials or of complete weapons could be many times cheaper. Weaponization Design and engineering Heavier, less efficient, lower- Principal challenge is maintain- Advanced weapons somewhat

yield designs easier, but all ing the agent’s potency difficult r but workable munition pose significant technical through weapon storage, deliv- designs (e.g., bursting smoke challenges. ery, and dissemination. device) widely published. Broad-area dissemination not difficult; design of weapons that effectively aerosolize agents for precision delivery challenging (but developed by U.S. by ’60s).

Production equipment Much (e.g., machine tools) Must be tightly contained to Relatively simple, closely re- dual-use and widely avail- prevent spread of infection, but lated to standard munitions pro- able. the necessary equipment is not duction equipment. Some overlap with conven- hard to build. tional munitions production equipment.

SOURCE: Office of Technology Assessment, 1993. weapons are considerably more likely to be used that the states now working hardest to develop in the future than either nuclear or biological weapons of mass destruction (see following weapons. section) are for the most part located in unstable regions of the world, where bitter and unresolved IMPLICATIONS OF PROLIFERATION rivalries have erupted into war in the recent past Mass killing of human beings is not new to and hold the prospect of doing so again. Not only warfare, or even to this century. Nevertheless, might future wars lead to the actual use of weapons of mass destruction compress the time weapons of mass destruction, but the deployment and the effort needed to kill. Wars lasting only a of such weapons in these regions could increase few days could now devastate populations, cities, tensions still further. or entire countries in ways that previously took Even if these weapons are not used, they cast months or years. Particularly ominous is the fact shadows that can affect interstate relations and 12 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

international balances of power. A few analysts, however, the existence of the Soviet nuclear pointing to the role that nuclear weapons seem to arsenal strongly tempered U.S. views of which have played in preventing war between the United ‘‘vital national interests’ were worth risking States and the Soviet Union, argue that their nuclear war to defend. If additional countries spread will actually increase international stabil- acquire the means to threaten U.S. allies, U.S. ity. Most, however, consider such a view to be forces overseas, or even U.S. territory with dangerously misguided. The Cold War was not nuclear weapons, the United States will be forced without serious crises and close calls, such as the to reevaluate the conditions under which it is Cuban Missile Crisis. In the Middle East, South willing to risk nuclear attack. Even though it Asia, and the Korean peninsula, hostile powers might retaliate with its own nuclear arsenal, U.S. share common borders, contest core values retaliation may not compensate for U.S. or allied and vital national interests, and lack both the losses. mutual learning experience and the technical Plausible scenarios for the current set of safeguards that have helped the superpowers suspected proliferants to threaten U.S. territory come to live with the mortal threat each poses the other. with nuclear or other weapons of mass destruction Proliferation, therefore, poses real dangers are difficult to devise. None possess missiles or from the point of view of international security aircraft with sufficient range to reach the United and human welfare. Moreover, in addition to its States, nor are potentially hostile powers likely to global consequences, it poses particular problems develop such systems in the next 10 years (see for the United States. As a global power, the next section). Nevertheless, a state that badly United States will almost certainly retain allies wanted to wreak destruction on a U.S. city and vital interests overseas that might be threat- could probably do so, whether it had advanced ened by states possessing weapons of mass delivery systems or not—and whether the destruction. Should the United States need to United States had effective defenses against defend its interests and principles with military such advanced delivery systems or not. force—whether acting unilaterally or under multilateral auspices such as those of the United PROLIFERATION THREATS AND Nations—U.S. armed forces or territory might PROSPECTS become targets for weapons of mass destruction. The threat of nuclear attack is nothing new to Only five countries (the United States, Russia, the United States. Having faced a massive Soviet the United Kingdom, France, and China) ac- nuclear arsenal for decades, the United States has knowledge possessing nuclear weapon stock- 7 shown itself willing at least to contemplate the piles. Three more-Ukraine, Kazakhstan, and loss of many U.S. cities, and millions of Ameri- Belarus--have former Soviet strategic nuclear can lives, to ensure its own survival and the weapons located within their borders (although as survival of the states under its nuclear umbrella. yet they do not control them), and it is not yet (Granted, this posture has always posed problems certain that they will give them up. Only three for many who questioned what “national sur- states (United States, Russia, and Iraq) say they vival’ means in the context of tens, let alone have chemical weapon arsenals, and all of these hundreds or thousands, of nuclear weapons deto- weapons are in the process of, or are slated for, nating on U.S. territory.) At the same time, destruction. No countries admit to active offen-

7 South Africa has acknowledged having assembled six nuclear weapons but says it has since destroyed them. Chapter 1-Introduction and Summary I 13

sive biological weapon programs.8 Few countries, India, which tested a nuclear device in 1974 and therefore, are overtly deploying or preparing to probably has stockpiles of nuclear weapon ma- deploy weapons of mass destruction. The diffi- terial available, but has made no overt moves to culty in assessing the extent of the proliferation develop a nuclear arsenal; and Pakistan, which is threat lies in determining which states are doing cut off from U.S. military aid because the so secretly. Merely counting the states that today President cannot certify that it does not possess a are capable of mounting a program to produce nuclear explosive device.9 None of these coun- weapons of mass destruction inflates the prolifer- tries is a member of the Nuclear Non- ation threat considerably, just as counting only Proliferation Treaty. South Africa has admitted the states acknowledging such production errs in to mounting a nuclear weapon program that the opposite direction. culminated in the construction of six nuclear This report names countries commonly cited in weapons, confirming suspicions that had in- the public literature as having nuclear, chemical, cluded it in this threshold category. However, or biological weapons, or as having programs to stating that it has destroyed those weapons, it has acquire them. Consistent with the unclassified since joined the Nuclear Non-Proliferation Treaty nature of this report, the estimates given here as a non-nuclear-weapon state and opened up its are not based on classified sources and should nuclear facilities to international inspection. Lit- in no way be considered authoritative or tle information has been released so far on the indicative of official U.S. Government assess- results of those inspections, but to date (June ments. Therefore, the tables in this report may 1993) they have not resolved ‘serious questions’ well disagree in some respects with the best that the United States has concerning South intelligence information available to the U.S. Africa’s compliance with its NPT obligations.10 Government, which itself can be uncertain and A few other states reputed to have nuclear incomplete. If an incomplete public understand- weapon programs are apparently not as far ing of the current extent of proliferation poses advanced as the above four: Iran and Libya, both problems for U.S. nonproliferation policy, it NPT members; North Korea, which has given behooves U.S. policymakers to ensure that the and then retracted notice of its intent to withdraw open literature better reflects the actual state of from the NPT, and possibly Algeria, which is not affairs. an NPT member. North Korea appeared to have taken steps to back away from its nuclear weapon 1 Keeping Score program, permitting inspection (after years of Besides the five acknowledged nuclear powers delay) of facilities that clearly seem to have been and the three non-Russian former Soviet repub- intended for nuclear weapon production. How- lics that still have nuclear weapons on their ever, after having been caught attempting to territory, only three “threshold states” appear to mislead international inspectors as to the extent of possess nuclear weapons or have the ability to its nuclear program, it refused to open other deploy them on short notice: Israel, which is suspicious facilities for inspection. Rather than widely believed to have a clandestine arsenal; comply with its commitment under the NPT to

8 Ru~~ia ~ ~~~ tit the Soviet Ution’s Offmive biologic~ w~pon prog~ pe~isf~ titer the U.S.S.R. signed the Biological Weapon Convention banning such work but insists that this program has since been halted. 9 ~ 1~, p~~>s FOr~@ s~~~ W= quoted u d~lm tit P~SW ~ w the p@ for a nuclw weapon. He subst!quen~y retraeted this statement, with the Ministry of Foreign Affairs claiming he had been misquoted. 1° “Adherence to and Compliance with Arms Control Agreements and the President’s Report to Congress on Soviet Noneomplianee with Arms Control Agreements,” prepared by the U.S. Arms Control and Disarmament Agency, Jan. 14, 1993, p. 18. 14 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

cooperate with such inspections, it gave notice of their delivery systems, figure 1-1 may understate its intent to withdraw from the NPT, becoming the number of countries pursuing such systems.11 the first nation ever to do so. To many observers, Part of the discrepancy may be states that are such actions confirm that North Korea not only pursuing delivery systems but not nuclear, chemi- had been pursuing nuclear weapons all along, but cal, or biological weapons, which would not be seeks to preserve the capability to do so. included in figure 1-1; the remainder might Iraq is a special case. Although the 1991 Gulf indicate countries suspected by U.S. intelligence War and its aftermath arrested and reversed Iraq’s of pursuing such weapons but not yet identified in nuclear weapon program before it could come to open sources. fruition, United Nations inspections showed the Most of the states listed in figure 1-1 have program to have been much broader and more bought or developed simple ballistic missiles advanced than Western intelligence agencies had with at least the capability of Scud missiles. All suspected. Few observers doubt that the program have combat aircraft with characteristics that will resume in the absence of the extraordinarily make them candidates for delivering weapons of international monitoring efforts imposed upon mass destruction. None seems to have cruise Iraq by the United Nations Security Council. missiles adapted to this purpose, but the spread of Argentina and Brazil in the past had been applicable technologies makes cruise missiles a thought to be pursuing nuclear programs, albeit threat to be concerned about in the future. ones that were less advanced than those of the Three features stand out in the combined threshold states. In recent years, however, they perspective offered by figure 1-1. First, the have agreed to open up their nuclear facilities to estimate for the current number of states bilateral and international inspection to assure actively pursuing nuclear weapons is small, each other and the rest of the world that they are and smaller than it might have been a few not developing nuclear weapons. years ago. Second, the set of countries trying to Public reports of the extent of chemical and acquire nuclear weapons overlaps consider- biological proliferation differ with each other more ably with the set suspected of having chemical than do assessments of nuclear proliferation. OTA and biological weapon programs. Third, the has reviewed several compilations of states sus- most immediate and serious threats (beyond pected of pursuing chemical or biological weap- the potential threat posed by former Soviet ons; those states appearing in a preponderance of republics) are concentrated in three regions of these lists are identified in figure 1-1, together international rivalry: the Koreas, India- with the states mentioned above as still suspected Pakistan, and the Middle East. of pursuing nuclear weapons. (See ch. 2.) Longer term assessments of the extent of In all, 14 countries are listed in figure 1-1 as proliferation are harder to make, although some widely believed to possess or to be pursuing trends are clear. For example, for ‘at least another nuclear, chemical, or biological weapons. Given decade,” only China, Russia, and possibly Ukraine, official U.S. Government statements that ‘‘more Kazakhstan, and Belarus--all possessing weap- than 25 countries . . . may have or may be ons that have long been capable of being targeted developing” weapons of mass destruction and at the United States—will pose a ballistic missile

11 ~~ony Of R. J-es WOOlsey, Director of Central Intelligence, before the Senate Committee on Gvernmmtd M*S, Feb. w, 1993. Although he gave some information on the activities of some countries, his testimony did not identify all of the states believed by the United States to be pursuing weapons of mass destruction and their delivery systems, much less specify which ones are pursuing which weapons. Chapter I-Introduction and Summary 15

threat to the territory of the United States.12 Figure l-l-Suspected Weapon of Mass Projections of the future number of nuclear, Destruction Programs chemical, or biological powers are more elusive. It is hard to determine even the present status and anticipated progress of existing programs. Even Prolifera;t 1 with the wealth of information that has been nuclear Algeria? weapon India pr~gram uncovered about the Iraqi nuclear weapon pro- Pakistan gram, for example, experts disagree over how \ long it would have taken the Iraqis to assemble a working weapon. Moreover, extrapolating from current trends can be misleading. We have already noted several examples of apparent ‘rollback” in nuclear weapon programs. Further Myanmar (Burma) changes in the world situation, including those Vietnam that nonproliferation policies seek to bring about, \ \ Chinal will affect the extent of future proliferation. ‘/ / Chemical Biological arsenal weapon 1 Fostering Proliferation (probable or development Trends possible) ‘(possible)

Shaded area: also has RISKS FROM THE BREAKUP OF THE SOVIET Scud-type or longer UNION range ballistic missile The breakup of the Soviet Union-and the shakiness of governmental authority in its succes- 1 The Unitd fjtateg, Ru.sda, United Kingdom, France, and China are sor republics-could contribute to proliferation nueiear-weapon states party to the Nuclear Non-Proliferation Treaty and therefore are not considered nuclear “proliferants.” However, problems. The threat is potentially great, but hard China is suspected of pursuing chemical and biolo@al weapons and to predict. One set of problems could arise from is included in this figure for that reason. 2 lmqi ~mgrms have ken reversed under UN SWuritY ~uncil the emergence of Ukraine, Kazakhstan, or Be- Resolution 687. larus as new nuclear powers and the ramifications SOURCE: Tables 2-6, 2-7, 2-8; Office of Technology Assessment, such actions would have on the nuclear nonprolif- 1993. eration regime. Another concerns the future weapons of mass destruction indigenously, per- integrity of Russia itself, or at least of the system haps drawing on facilities that had once contrib- of controls over Russian nuclear weapons and uted to Soviet weapon programs. nuclear weapon materials. A third issue, perhaps the most immediate risk posed by the Soviet breakup, is the possible leakage to potential THE END OF THE COLD WAR proliferants of people, critical information, equip- Apart from the acute crises posed by the ment, materials, or even complete weapons. Such collapse of the Soviet Union, the end of the Cold assistance could be of great value not only to War has the potential for weakening restraints nuclear but also to chemical or biological weapon against proliferation. Countries that had formerly programs. Finally, in the longer term, various enjoyed Soviet or U.S. security guarantees may former Soviet republics may choose to develop now feel more exposed and insecure, increasing

12 Row M. ~tcS, D~tO~ of c~~~ ~te]lig~e, wore tie Semte Gove~en~ ~~ COtitt&, J~. 15, 1992. Sti “W~pO~ Proliferation in the New World Order,” S. Hrg. 102-720, 102d Cong., 2d Sessiou Jan. 15, 1992, p. 7. Britain and France, with submarine-launched ballistic missiles capable of reaching the United States, are not considered to pose a threat. 16 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

the motivation to develop their own weapons of SPREADING TECHNOLOGY AND mass destruction. Moreover, controls that the INDUSTRIALIZATION Soviet Union had formerly exerted over its allies Economic and technological development will, no longer exist, or at least have been considerably in general, enhance national wealth, technical weakened given the reduced role that Russia is skill, and industrial capabilities useful for indige- playing in international affairs. As the United nous production of weapons of mass destruction States reshapes its own security relationships in and their delivery systems. It will also increase recognition of the Cold War’s end, in particular the number of potential foreign suppliers of skill by withdrawing overseas forces, it too may lose and technology to proliferant nations. Conse- some leverage over its allies. quently, it will be increasingly difficult for a small number of industrially advanced countries to control weapon proliferation by denying access to PERSISTENCE OF REGIONAL CONFLICTS key technologies or materials. Outside the sphere of the former Soviet Union, Indeed, the dissemination of technologies the most serious drivers for proliferation of that have at least some relevance to producing weapons of mass destruction remain the seem- weapons of mass destruction might need to be ingly intractable regional conflicts in South Asia not only tolerated but encouraged if popula- and the Middle East, where most of the current tions in developing nations are to improve proliferants are located. In South Asia, India and their health, environment, and standards of Pakistan are unable to resolve their ethnic and living. This is especially true for chemical and territorial dispute over Kashmir, while India biological technologies. Technologies that can also--or perhaps even primarily--feels threat- contribute both to military and civil objectives, ened by China, the nuclear power to the northeast. often referred to as ‘‘dual-use” technologies, are In the Middle East, the current peace process does actually multi-use, providing basic capabilities not promise early resolution of the Arab-Israeli that can be used in a host of applications (e.g., conflict. In addition, the Iraqi invasion of Kuwait computing, metal-forming, and diagnostic test- illustrates that, even independent of Israel, the ing). Controls on some dual-use technologies will Arab and other Islamic countries (e.g., Iran) prove to be infeasible (if the technologies in- would probably continue to arm against one 13 volved have already disseminated too widely) or another. undesirable (if too many non-weapons-related Proliferation of conventional arms, fueled by activities would be constrained as well). these regional conflicts as well as by the glut of These difficulties notwithstanding, export con- military industrial capacity and weapon stock- trols will remain an important nonproliferation piles in the wake of the Cold War, can stimulate tool. For example, although Iraq’s indigenous the quest for weapons of mass destruction as industrial base was more capable than most ‘‘equalizers. ‘‘14At the same time, continued sales outsiders realized, it still had to import much of of high-performance combat aircraft and the the equipment used in its weapon facilities. This spread of missile technology bolster states’ abil- level of importation was made feasible only by ity to deliver weapons of mass destruction. Iraqi oil revenues.

13 fib ~ ~- disputes with ‘lhrkey, a NATO member, bave the potential to involve the United States directly. 14 For &~uSSiO~ of be spread of conventio~ fi~ ~~ology, see TJ,S. Co-ss, ~lce of khoIo~ Assessmen~ G/oh/ /h2.f Trude, O’IA-ISC+60 (Washington, DC: U.S. Gov ernment Frinting Office, June 1991). Chapter 1-Introduction and Summary I 17

RESISTANCE TO DISCRIMINATORY REGIMES 138 at the end of 1989 to 157 by January 1, 1993. A few developing countries, most notably Two of the nuclear weapon states that did not India, object to external attempts to deny them originally join the NPT (China and France) have nuclear and missile-related technologies that are joined in the last two years, as did a hold-out state accepted as legitimate for certain other countries, that has admitted having produced nuclear weap- Nevertheless, most nations of the world have ons outside the NPT (South Africa). Although been willing to live with the two-tiered, nuclear/non- three non-signatories are believed to have actual nuclear structure of the NPT. This factor is not an or potential nuclear weapon capabilities (Israel, issue with the Chemical Weapons Convention or India, and Pakistan), no states have declared the Biological Weapons Convention, both of themselves to be nuclear powers since China in which apply to all states without distinction. 1964. No non-nuclear members of the NPT have ‘‘gone nuclear,’ although a few have been trying. WEAKENED TABOO AGAINST CHEMICAL The deep reductions in nuclear forces under- WEAPON USE taken by the United States and Russia mean that The 1925 Geneva Protocol prohibits the use of both countries are finally making visible progress chemical and biological agents in warfare. This on their NPT obligation to ‘‘pursue negotiations ban was observed in most of the conflicts in good faith on effective measures relating to following its entry into force, including World cessation of the nuclear arms race, ’ even though War II (except the use by Japan, then a non-party, they seem unlikely to seriously pursue the NPT of chemical and biological weapons in China). goal of “general and complete disarmament. ” However, more recent instances of chemical weapon use have weakened this international Progress in superpower nuclear arms reductions norm. In particular, Iraqi use against Iran in the could undermine longstanding complaints about 1980s may have demonstrated to some defense the discriminatory nature of the NPT that might planners that chemical weapons can be a useful otherwise have hurt the treaty’s chances for military tool. renewal when it comes up for extension in 1995. On the other hand, North Korea would deal the nonproliferation regime a serious blow if it I Trends Favoring Nonproliferation Efforts proceeds to withdraw from the NPT

GENERALLY RISING NORM AGAINST PROLIFERATION END OF THE COLD WAR An international consensus seems to be grow- In part, the end of the Cold War has allowed the ing that further proliferation of weapons of mass strengthening of this norm against proliferation. destruction should be stopped, and that chemical Besides fostering a new level of cooperation and biological weapons should be eliminated between the United States and Russia (as succes- completely. Governments around the world have sor to the Soviet Union in the U.N. Security declared renewed commitments to nonprolifera- Council), the cessation of the U.S.-Soviet conflict tion. Strengthened nonproliferation norms might has also made possible changes in national help deter potential proliferants. More impor- priorities and policy emphases. Although nonpro- tantly, they also improve the prospects of liferation policies may continue to conflict with strong, coordinated world action to deter and other policy goals, they need no longer be punish violators. subordinated to Cold War objectives. In addition, The past few years have brought a significant foreign policy and intelligence resources are increase in the number of signatories to the being redirected from Cold War efforts to deal Nuclear Non-Proliferation Treaty, rising from with proliferation. 18 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Should Russia revert to a foreign policy that is and using it to build and operate other facilities seriously threatening to Western interests, non- that are not open to international inspection or proliferation will be set back. Granted, even controls. during the Cold War, the Soviet Union took a strong stance against nuclear proliferation, and an U.N. ACTIONS IN IRAQ anti-Western Russia would likely do the same. Besides reversing Iraqi mass destruction weapon But future efforts to contain the spread of all types programs, recent U.N. Security Council resolu- of weapons of mass destruction will require tions make approving references to international significant Russian-United States cooperation in nonproliferation and disarmament treaties, set- support of nonproliferation norms, not just paral- ting useful precedents in demonstrating interna- lel policies in limited areas (see ch. 3). tional resolve against weapons of mass destruction. The Security Council has also taken on both RECENT REVERSALS OF NATIONAL POLICIES short- and long-term onsite monitoring tasks to Reversals in the nuclear weapon programs of assure that its decisions mandating elimination of Argentina, Brazil, South Africa, and (albeit invol- Iraqi weapons of mass destruction are carried out. untarily) Iraq follow decisions in earlier decades by Sweden, South Korea, and Taiwan to halt CHEMICAL WEAPONS DISARMAMENT AND programs that had seemed directed at nuclear THE CHEMICAL WEAPONS CONVENTION acquisition. Such reversals, however, are them- The two largest chemical weapon powers, the selves reversible: despite some moves by North United States and Russia, have committed them- Korea to open up to international inspection, its selves to the destruction of their chemical arse- subsequent actions have given rise to serious nals, together with their development and produc- doubts. tion facilities. The Iraqi chemical arsenal is being dismantled under U.N. supervision. Most signifi- GROWING COOPERATION IN EXPORT cantly, the Chemical Weapons Convention (CWC), CONTROL REGIMES signed by more than 140 states in early 1993, bans Several multilateral groups have formed to for the first time the development, production, control the export of equipment or materials that and possession of chemical weapons (in addition might be used in the production of weapons of to their use) and reinforces the international norm mass destruction or of missiles. These control against chemical weapons. Regardless of any regimes have been strengthened in the past few doubts that may remain as to whether the CWC’s years, both by covering additional items and by verification regime is adequate to the task and expanding their membership. Particularly notable whether it will be stringently implemented, the is the April 1992 decision of the 27-member Convention strengthens the international consen- Nuclear Suppliers Group to require importers of sus that chemical weapons are illegitimate. If nuclear technology to accept international moni- some nation were to use chemical weapons in the toring (through the International Atomic Energy future, the international community may feel Agency’s system of safeguards) over their entire committed to react more strongly than it did nuclear programs, not just over the particular against Iraq’s use in the 1980s. facilities built with imported technology. This action leaves China as the only supplier of nuclear BIOLOGICAL WEAPONS CONVENTION technology that does not require such ‘full-scope VERIFICATION safeguards’ as a condition of sale. By requiring In the wake of the Russian admission that the full-scope safeguards, exporters prevent states Soviet Union had violated the Biological Weap- from acquiring expertise in safeguarded facilities ons Convention, the United States, the United Chapter 1-Introduction and Summary 19

Kingdom, and Russia have agreed on a program knowledge; adopting export controls; taking dip- to inspect each other’s biological facilities.15 lomatic or other action to stop exports by third Although states that have joined the Convention parties; and acting to stop or discourage experts disagree over the feasibility or desirability of a from providing assistance. Since the United formal verification regime, an international Ad States is not the only source of technology, Hoc Group of Governmental Experts on Verifica- knowledge, or other support that might be tion (VEREX) is considering potential verifica- useful to a proliferant, such measures must be tion measures. imposed multilaterally to be effective. At the In sum, despite some dangerous trends and same time, however, U.S. leadership is neces- many uncertainties, the world community has sary to mobilize effective international cooper- significant new opportunities to curtail, and ation. perhaps roll back, the spread of weapons of mass If, despite these barriers, a proliferant nation destruction. manages to acquire facilities for a weapon program that another country or countries deem to THE NONPROLIFERATION POLICY pose an intolerable security threat, the ultimate CONTEXT recourse might be to impose another kind of The United States and other countries con- obstacle: destroying the facilities by military cerned about the proliferation of weapons of mass attack. However, such an approach is fraught with destruction have instituted measures seeking to legal, political, and operational difficulties, and impose obstacles to the acquisition of weapons of must be considered at most as a last resort. mass destruction; to create disincentives in order to deter states from developing such weapons or EXPORT CONTROLS to persuade them to reverse course; to offer Export controls are intended to block the most benefits to states that agree to forgo such straightforward paths to developing weapons of attempts; and to develop security improvements mass destruction and to raise the price and the to reduce the perceived needs for the weapons. time required for alternate approaches. They can These measures have been implemented to date also provide information valuable for monitoring through three primary mechanisms: national programs to develop weapons of mass destruc- policies and laws governing states’ actions with tion. Export controls will remain an important respect to others that are developing or assisting component of nonproliferation policy for years, in the development of weapons of mass destruc- especially in the nuclear area. However, control tion; multinational agreements to restrict ex- regimes can be defeated if their targets can ports of certain technologies or to take action “invent around” restricted technologies or prod- against those states found to be violating interna- ucts, if controls are attempted on goods that are tional nonproliferation norms; and international too widely available, or if some potential suppli- treaties and institutions open to all states who ers are not included. Moreover, it is very difficult agree to subscribe to their principles. to control the education of scientists and engineers in one country who may later return or I Imposing Obstacles to Proliferation migrate to another to develop weapons of mass Obstacles that can be put in the way of states destruction. trying to acquire weapons of mass destruction In the United States, export controls are estab- include using secrecy to restrict the flow of lished by a number of public laws and regulations

15 ‘CJOint u.s.~.K,~USSi~ Statement on Biological weapons,’ Sept. 14, 1992, reproduced in The Arms Control Reporter (Cambridge, MA: Institute for Defense and Disarmament Studies, 1992), vol. 11 (1992), p. 701 .D.l 20 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

(see ch. 3), and they are also formally or infor- Military action involves very high risk. Opera- mally coordinated with those of other states. They tionally, the attacking country or group of coun- now cover a range of technologies related to tries must contend with questions about the nuclear, chemical, and biological weapons, as well quality of its intelligence, how well the attack can as ballistic missiles and conventional armaments. be executed, and how badly the attack will damage the proliferant state’s weapon program. SANCTIONS AGAINST SUPPLIERS As the Israeli raid showed, military attack is not The United States can impede weapon pro- a permanent solution. Strategically, a potential grams in proliferant states by helping foreign attacker must consider the degree of international governments block aid that their own citizens or backing it is likely to expect and the prospect of corporations may be providing such programs. If retaliation (military, diplomatic, or terrorist). U.S. intelligence uncovers foreign plans to pro- Military action that is not explicitly sanctioned by vide such assistance, the United States can request the government having jurisdiction over the international community risks damaging con- such activities to stop them. In addition, U.S. laws sensus on future cooperative nonproliferation can impose sanctions directly against foreign policies and might even build sympathy for the individuals or companies, including criminal victim of the attack. penalties, debarment from U.S. Government procurement, and denial of access to the U.S. market B Disincentives and Sanctions Against (see ch. 3). Should a foreign government itself be Proliferants aiding proliferation, the United States can take Mechanisms exist in U.S. law—but are not laid diplomatic measures against it such as denial of out in international law—to punish states found to trade preferences, arms transfers, or financial have used weapons of mass destruction or to have assistance. engaged in activities related to their development MILITARY ACTION (see ch. 3). At the international level, enforcement In extreme cases, the United States, another of international nonproliferation commitments nation, or a multinational coalition might feel falls to the United Nations Security Council, compelled to attack facilities, equipment, or which has the authority to respond to ‘‘threats to materials believed to be connected to weapons of international peace and security” by imposing mass destruction. However, if not authorized by measures such as sanctions, severance of travel the United Nations Security Council, such an and communication links, diplomatic isolation, or action would generally be viewed as contrary to even military action under Chapter VII of the international law unless it could be shown to be United Nations charter.16 Actions of the Security required for national self-defense, and unless Council are binding upon all U.N. members. other means short of attack had been exhausted. Security Council enforcement of existing nonpro- (Although they made precisely those arguments, liferation commitments such as the Nuclear the Israelis were unable to convince the world Non-Proliferation Treaty and the 1925 Geneva community that their 1981 raid against Iraq’s Protocol could deter further proliferation and Osirak nuclear reactor was justified.)

16 BY ~emelves, ~~~tio~ org~~tions involved with nonproliferation such as the hNerrMtiOnd Atomic mewAFcy, ~i~Y ~ take no punitive action stronger than expelling members found to have violated their commitments to the organization. However, the IAEA can refer evidence of violations to the United Nations Security Council for further action. It did so for the first time in APIN 1993, when it found North Korea in violation of its safeguards agreement. .—

Chapter 1-Introduction and Summary 21

strengthen global nonproliferation norms.17 Con- versely, inaction in such cases will weaken the nonproliferation regime. Any United Nations efforts to enforce treaty commitments will not directly affect those states that have not acceded to these commitments in the frost place. However, in January 1992, the Secu- rity Council declared the proliferation of weapons of mass destruction to be a threat to international peace and security, opening up at least the possibility of taking action even against proliferants who are not party to global nonproliferation regimes. Within the United States, current laws and regulations to deter or punish proliferants stress The United Nations Security Council, which has economic sanctions. However, other measures primary responsibility for the enforcement of could be taken, many of which serve not only to international nonproliferation obligations, deter further proliferation but to help address the consequences of proliferation if it occurs. These 1 Benefits for Forgoing Weapons of Mass measures include: Destruction ● embarrassment by disseminating intelli- Coercive measures by themselves may not gence or other information exposing illicit always be sufficient to stop states from acquiring activities; weapons of mass destruction. The best hope for ● provision of technical or military assistance nonproliferation in the long term lies in build- to states threatened by weapons of mass ing a consensus among potential proliferants destruction; that they should jointly refrain from acquiring ● development and deployment of active de- these weapons. However, several factors make fenses (e.g., missile or air defenses) or passive such a consensus difficult to achieve. States measures (e.g., gas masks and protective seeking weapons of mass destruction may want clothing) to protect against the weapons; them for military purposes (including intimida- ● diplomatic isolation of proliferants or forma- tion or deterrence), for political influence, for tion of countervailing military alliances; and national pride, or for international status. The ● withdrawal of U.S. security guarantees. presence of nearby nuclear powers is a powerful The effectiveness of these measures will incentive to develop nuclear weapons, and a depend, like other nonproliferation measures, cascading one. (China acquired them because of on the degree of international cooperation the United States and the Soviet Union; India behind them. The presence of strong international because of China; Pakistan worries about India, norms against acquisition and use of these weap- etc.) To forgo weapons of mass destruction, ons will be important to getting that cooperation. potential proliferants must come to see that their

IT me 1925 ~nev~ Protwol bans use, but not development, production or stockpiling, of chtictd ~d ‘ ‘b=teriolo@d” w~om. my states ratifying it reserved the right to retaliate in kind against chemical or biologicalattack or considered it binding only with respect to other signatories. ‘Iherefore, it effectively became a ‘‘nofwst use” agreement. Moreover, no attempts have ever been made to enforce it against violators. Signatones who have since acceded to the Chemical and the Biological Weapons Conventions, which unconditionally ban those weapons, have rescinded their reservations to the Geneva Protocol. .-

22 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

political or military needs can be met in some weapons of others. This deal underlies regional or other way. global arms control arrangements such as the Although it is not likely to sway a determined Nuclear Non-Proliferation Treaty, the Chemical proliferant, financial, technical, and other devel- Weapons Convention, the Biological Weapons opment assistance can be offered to states forgo- Convention, the Latin American Nuclear-Free ing the development of weapons of mass destruc- Zone (Treaty of Tlatelolco), and the South Pacific tion.18 Exemptions from export controls on dual- Nuclear-Free Zone (Treaty of Rarotonga). These use items, or preferential access to international treaties codify the international norms against aid organizations, might also be offered. The weapons of mass destruction and have value for Nuclear Non-Proliferation Treaty (NPT), for in- that reason alone. Beyond that, however, most of stance, promises technical assistance in the peace- them are also associated with verification regimes ful uses of atomic energy, including medical and intended to permit parties to assure each other that agricultural applications. Note, however, that they are in compliance (see box l-A). such assistance can be a double-edged sword, Nonproliferation treaties involve a “free- since familiarity with nuclear technology can rider” problem: states that remain outside the contribute to military as well as peaceful goals. regime can sometimes enjoy the benefit of The Chemical Weapons Convention (CWC) of- reducing the threat to themselves without having fers more in the way of incentives than the NPT, to pay the price of giving up their own weapon 19 promising not only access to chemical technology options. Moreover, the NPT—which permits but also various assurances to parties who find the United States, Russia, Britain, France, and themselves threatened or attacked with chemical China to retain their nuclear arsenals--does not weapons. Members of the CWC envisage that eliminate the potential nuclear threat that member- chemical weapon-related export controls will be states may believe these nations to pose. (It does, relaxed against member-states judged to be in however, commit the nuclear weapons states to compliance. The Biological Weapons Conven- pursue nuclear disarmament and to assist non- tion (BWC) makes similar provisions for promot- nuclear states in their peaceful nuclear programs.) ing transfers of biotechnology to member-states, The long-run success of nonproliferation pol- although these have never been implemented. icy is likely to depend, at least in part, on the reduction of the security threats used to just@ 1 Security Benefits To Reduce the Demand acquisition of weapons of mass destruction. The for Proliferation security problems in each region of proliferation Technical assistance notwithstanding, the cen- concern are different; each will require specially tral bargain of consensual nonproliferation agree- tailored arrangements if parties are to trust one ments is that states give up their own rights to another enough to halt or reverse their military acquire weapons of mass destruction on the competitions. Such arrangements may consist of condition that they will not be needed to deter the combinations of political accommodations, eco-

18 ~tead of s- -y as an ~centive to adopt other nonproliferation policies, development assistance could itself be a nonproliferation measure to the extent that lack of developmen~ economic deprivation, and competition for economic resources are a source of conflict. Similarly, policies that alleviate international tensions resulting from demographic trends, differing political systems, ideology, and resource pressures can also be considered nonproliferation measures. Analysis of such policies, however, goes outside the scope of this assessment. 19 For this r~o% tie Treaty of ‘rlatelolco contains a provision that keeps it ~rn coming into force until all states ill the region beCOme members. Twenty-three of the Treaty’s parties have waived this provision accepting the Treaty’s obligations. Brazil and Chile, which had not previously waived the provisioq and Argentina and Cub% which had not entered the TrtXy at all, have recently said they will join the Treaty or permit it to enter into force for them. .. .

Chapter I-Introduction and Summary 23

Box 1-A-International Nonproliferation Treaties and Their Verification Regimes

Nuclear Non-Proliferation Treaty (NPT) The NPT prohibits all member-states except the five acknowledged nuclear powers from acquiring nuclear weapons. It also requires ail non-nuclear-weapon member-states to implement a safeguards agreement with the International Atomic Energy Agency (IAEA) covering ail nuclear materials that might be useful for weapons. IAEA safeguards are intended to detect, and therefore to deter, the diversion of materials from peaceful nuclear programs to military use, although they cannot by themselves prevent such diversion. Under the NPT, non-nuclear-weapon member-states must declare to the IAEA all facilitiesthat handle nuclear materials, and these facilities then become subject to safeguards. But the IAEA has had little ability to monitor whether states were conducting nuclear weapon activities in undeclaredfacilities. The limitations of this approach became clearafterthe 1991 Gulf War, when Iraq was revealedto have mounted a major nuclear weapon program outside of its declared nuclear program. Although monitoring declared nuclear facilities will continue to be crucial to verifying compliance with the NPT, it addresses only half the problem. Some means must also be found to allay suspicions that nuclear weapon activities might be undertaken in covert or undeclared facilities. The IAEA has always had the formal abilityto undertake “special inspections” of undeclared fadiities if it had reason to suspect illicit activities there. However, it did not exercise this authority until February 1993, when it attempted to inspect suspicious sites in North Korea. (in response, North Korea refused access to IAEA inspectors and announced its withdrawal from the NPT.) To carry out such inspections, the IAEA must be able to receive and act on information identifying suspect facilities, and it must have the backing of the U.N. Security Council in case the target state refuses to cooperate. Since the NPT entered into force in 1970 for a 25-year period, a review conference will be held in 1995 at which member-states must decide whether to extend the treaty, and for how long. consequently, successful extension of the NPT, preferably for an indefinite term, is one of the most important issues facing the nuclear nonproliferation regime. Treaties of Tlatelolco and Rarotonga Both of these nuclear-weapon-free zone treaties create regional organizations to monitor compliance and also require that member states submit to IAEA safeguards. Chemical Weapons Convention (CWC) The newly signed Chemical Weapons Convention bans the development, production, possession, and use of chemical weapons and establishes the most comprehensive verification scheme yet formulated in an international treaty. When it comes into force, it will create a new international institution-the Organization for the Prohibition of Chemical Weapons (OPCW)--that will receive routine declarations from member states and conduct routine inspections of declared chemical facilities. More significantly, it will also have the ability to conduct “challenge inspections” at any site--government or private-suspected of illegal activity. Far more facilities produce, ship, or use chemicals than are involved in peaceful nuclear activities, making the routine notification and inspection activities of the OPCW much more complicatedthan those of the IAEA. Moreover, the CWC’s challenge inspection provisions are much more rigorous than the IAEA’s provisions for ’’special inspections.’’ The final treaty text--and the implementation procedures now being negotiated among treaty signatories-are based on the principal of “managed access,” in which the state being searched has the right to limit the access of treaty inspectors in order to protect information not germane to the treaty. An important challenge in implementing the CWC’s inspection provisions will be balancing the need to monitor treaty compliance with the need to protect proprietary and national-security information unrelated to the CWC.l

1 S* office of T~nology Assessment, The Chernlca/ Wbapons ConventIon: Effecfs on the U.S. Cht?mM /ndustry, OTA-BP-ISC-1O6 (Washington DC, U.S. Government Printing Office, August 1993). (Continued on nextpag8) 24 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

z TIW ~r~ IMNW Of ttw aioiogld Wapons Convention k the “Convention on the Prohibition of the i)eveioprnm~ Production and $tookpiiing of Bactedoiogkai (Bioiogioai) and Toxin Weapons and on Their Destruction.” nomic measures, military confidence-building a potential threat to their neighbors that will make measures, and arms control. They may also it harder to resolve the regional security situation. involve security guarantees provided to regional Some analysts go so far as to assert that a states by powers outside the region (positive transformation of the whole basis of global security assurances), or assurances from extra- security will be required to have any chance of regional powers that military force-or weapons inducing restraint among many of the states that of mass destruction-will not be used against might otherwise turn to weapons of mass destruc- regional states (negative security assurances). tion. Proliferation cannot be controlled, they Regional security measures and nonprolif- argue, unless the international political system is eration policies have to proceed in tandem. fundamentally changed from one in which states States lacking confidence in regional security assure their own security through their military arrangements may be unwilling to forgo develop- forces and alliances, to a “cooperative security” ment of weapons of mass destruction, thus posing regime in which states do not maintain forces Chapter 1-Introduction and Summary 25

sufficient to conduct aggression in the first place. dilemmas. The established nuclear powers may With the Cold War over, these analysts believe it conclude that if additional states are going to is now possible to move towards such a world. develop and deploy nuclear weapons anyway, it would be prudent to minimize the risk that those ~ When Nonproliferation Fails weapons might actually be used. Therefore, they might wish to help the emerging nuclear powers The United States cannot assume that all those develop stabilizing doctrines of deployment and states now acquiring or possessing weapons of mass destruction will soon renounce them, nor deterrence, and implement appropriate technical that future nonproliferation policies will be 100 safeguards against accidental or unauthorized percent effective. It must therefore consider use. However, this would be tantamount to measures to mitigate the consequences of prolif- admitting these states into the nuclear club, eration for U.S. and international security. showing that proliferation can lead not to interna- Modifying U.S. military force structure and tional condemnation but to legitimacy and even operational planning to cope with prolifera- enhanced status. tion is unquestionably an important task for U.S. policy makers. If they prove technically CONFLICTING OBJECTIVES feasible, actions such as improving intelligence Many of the choices to be made in designing capabilities or adopting passive and active de- and implementing nonproliferation policies are fenses might improve U.S. military capabilities between potentially conflicting objectives; that is, without interfering with nonproliferation objec- the extent to which nonproliferation should take tives. Indeed, by lessening the military value of an precedence over other objectives of U.S. policy opponent’s weapons of mass destruction, such when they cannot both be pursued simultane- actions can simultaneously serve to deter an ously. Certainly the end of the Cold War has opponent from acquiring such weapons, and to removed one such conflict, eliminating what had deter or militarily counter their use if acquired been an overriding concern and permitting non- anyway. proliferation to take much greater priority. Yet Other preparations to mitigate the conse- tensions between nonproliferation and other pol- quences of proliferation, however, might exac- icy objectives continue to force tradeoffs. erbate the process of proliferation. For example, the existing nuclear powers might wish to Many conflicts between competing goals are deter or even to prevent chemical or biological mirrored in the organizational structure of the attack by holding out the prospect of using U.S. Government, with particular agencies pursu- nuclear weapons. Giving nuclear weapons this ing missions that at times conflict with each other. mission, however, could increase their perceived With the possible exception of the Arms utility and status, weakening nuclear nonprolifer- Control and Disarmament Agency, whose com- ation efforts. Moreover, advertising a willingness plete agenda in the post-Cold War organiza- to use even conventional force to preempt or to tion of the U.S. Government is still evolving, no respond to proliferation may persuade some single agency has nonproliferation as its pri- countries not to forgo weapons of mass destruc- mary mission. The other agencies that have the tion, but instead to seek them as a counter- greatest roles in nonproliferation policy-the deterrent. Departments of State, Defense, Energy, and Com- Other measures short of military force might merce—are all charged with pursuing other goals lessen the chances that proliferation will lead to that can compete with nonproliferation, some of use of nuclear weapons, but these pose serious which are described below. 26 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Figure 1-2—Potentially Conflicting Objectives

[ Nonproliferation <“ ‘-- “ –—---- ——> Other foreign policy objectives

Exporl controls G- ‘- ‘ -- -’~ Export promotion

Monitoring of others <-- —~ Avoidance of monitoring by others

Nuclear nonproliferation +–––-—-—--- ● Nuclear deterrence

Preventing proliferation <—-–- ‘——--- –—> Preventing use I

SOURCE: Office of Technology Assessment, 1993.

The pairings depicted in figure 1-2 and summa- nuclear and ballistic missile arsenal. Even if rized below are not discrete alternatives, but Israeli weapons of mass destruction are not rather opposite poles of a continuum. Intermedi- themselves deemed to threaten the United States ate positions are certainly possible, but seeking or U.S. interests, however, their implicit accep- one goal will generally imply lessening emphasis tance complicates nonproliferation policy. Other on the other. These choices must be made on a states condemn U.S. policy as reflecting one case-by-case basis, since the appropriate balance standard for friends and another for all other between conflicting objectives varies depending countries, hampering attempts to build interna- on the individual situation. tional consensus behind nonproliferation policies. U.S. policy towards China also illustrates 9 Nonproliferation v. Other Foreign tensions among conflicting objectives. U.S. policymakers have sought to stop Chinese sales of Policy Objectives nuclear and missile technology. At the same time, U.S. relationships with other states involve a they must also take note of China’s overall host of objectives, both generic and country- strategic importance in the Pacific region, its specific. For example, the U.S. Government may growing economic clout, the need to gain China’s wish to maintain favorable relations with other agreement (or at least its acquiescence) in U.N. states, encourage them to support U.S. positions Security Council actions, and the desire to pro- in international fora, restrain their conventional mote human rights and democratization within arms buildups, promote exports, support human China. Threatening to revoke China’s “most-favored- rights, and work towards common environmental nation” (MFN) status potentially provides the goals. Expending limited U.S. influence to stress United States considerable leverage over China, nonproliferation goals may mean losing a target just as threatening to withhold the U.S. commit- state’s cooperation on other matters, or even ment to Israeli security provides leverage over provoking its hostility. Israel. In both cases, however, other factors that Consider the cases of Israel and China. Israel outweighed nonproliferation have so far kept has a very strong, longstanding relationship with these threats from being executed. the United States, one in which nonproliferation has never figured prominently. Ensuring the security of a democratic ally threatened by hostile # Export Controls v. Export Promotion neighbors has outweighed whatever concern the The push to improve U.S. economic perform- United States has had over Israel’s apparent ance, increase jobs, and rectify the trade imbal- _

Chapter 1-Introduction and Summary ! 27

ance makes it a major national priority to increase exports. Tightening the export-control system for nonproliferation purposes may sometimes conflict with this goal. No respectable exporter deliberately seeks the business of those developing weapons of mass destruction, although too many disreputable ones apparently do. However, some exporters may unwittingly assist proliferants if they do not know by whom and for what purpose their products will be used, or if their goods are diverted after sale. Tightening export controls and applying sanc-

tions against foreign violators have economic and Dual-use electronic equipment seized by the U.S. political costs that must be weighed against their Commerce Department’s Office of Export return in international security. These costs may Enforcement while in the process of being exported be deemed worth paying, but they should be illegally to Iran. The equipment was intended for acknowledged. First, controls can somewhat re- Iran’s Ministry of Defense and its Atomic Energy Organization. strict international trade. Although the number of export denials is a small fraction of all interna- grounds. In a nonproliferation context, however, tional transactions, many transactions must be exporting states may seek to restrict the spread of screened in order to detect those that ultimately weapon-related technology without placing un- are denied. Consequently, a wide range of busi- necessary obstacles in the way of an importing nesses must keep informed about and comply state’s legitimate economic and technological with complex regulations and licensing proce- development—a much tougher assignment.20 By dures. Individual companies may find themselves the same token, however, the greater the depend- losing legitimate sales and the other business opportunities that might have followed those ence of a developing country’s civil economy on sales. More seriously in terms of U.S. jobs and imported technology, the more leverage would be exports, U.S. firms may also find that foreign provided by making access to that technology competitors under less stringent controls are contingent upon acceptable nonproliferation be- moving in to take over their markets. Although havior. U.S. policymakers may be willing to hold U.S. Policies governing export controls must ad- firms to a higher standard, such a policy would dress two sets of issues. The frost involves the interfere with U.S. export performance without internal structure, implementation, and enforce- generating any nonproliferation benefits if other ment of U.S. export controls. U.S. export-control countries do not follow suit. procedures have been the source of bureaucratic In addition to their costs to exporters, controls and political controversy for decades, a situation also impose costs on legitimate foreign users of that is likely to be aggravated as nonproliferation advanced technology. During the Cold War, replaces the Cold War as the primary driver of damage to the Soviet civil sector resulting from export control policy. The second set of issues Western export controls was seen as a “fringe involves the coordination of export control poli- benefit” of a policy already justified on security cies among different nations, and the role that

m M. Grmger Morg~ ad ~tchel B. w~erste~ ‘‘controlling the I-I@ Technology Militarization of the D=elopti worlds” J~es Goodby, cd., Bipolarity Revisited: Problems in North-south Security Relations Ajier the Cold War, (Oxford University Press, in press). 28 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

unilateral initiatives play in shaping multilateral weapons are given a prominent role in ensuring consenses. Multilateral control regimes aimed at the security of the United States and its allies-- different weapons of mass destruction and deliv- particularly against threats from non-nuclear ery systems have evolved separately; they have powers-it becomes harder to make the case that differing memberships, procedures, and objec- other countries should not be able to address their tives. Participating governments may wish to security concerns in similar ways. Granted, reten- examine the existing structure of these multilat- tion by the United States of its nuclear arsenal is eral regimes to see if tighter coordination, or very unlikely to be the sole factor inducing consolidation, is desirable or feasible. another state to pursue nuclear weapons. How- ever, U.S. decisions involving continued nuclear 1 Monitoring Others v. Avoiding the weapon development and testing, continued pro- Costs of Being Monitoring duction of nuclear weapon materials, or reliance Nonproliferation regimes are strengthened by on nuclear threats against nonnuclear attack, will empowering international nonproliferation or- certainly influence nuclear nonproliferation norms. ganizations to make intrusive onsite inspections Some argue that in the long run, there is no way of suspect activities (see box l-A). Yet states will to sustain a stable world order in which some not easily accept such inspections unless other states possess nuclear weapons but all others are states do likewise.21 If the United States expects forbidden to acquire them. In such a view, other states to provide access to outside inspec- stopping nuclear proliferation is impossible with- tors, it may have to open itself up to inspection as out a universal prohibition against national nu- well. Such inspections have costs that must be clear arsenals, with all nuclear weapons either weighed against their nonproliferation benefits. placed in the hands of a supranational organiza- In addition to the disruption of normal activities, tion or banned entirely. However, such a world U.S. Government or industrial facilities exposed still seems remote. to foreign inspections must incur costs to protect classified or proprietary information unrelated to 1 Preventing Proliferation. Preventing Use the purpose of the inspection. Even greater costs As discussed above, some measures to reduce might be incurred by failing to protect such the risk that nuclear weapons might be used— information, or by inadvertently disclosing se- measures that would require the acknowledge- crets that might actually aid a proliferant’s own ment of new nuclear powers—would actually weapon programs. conflict with controlling proliferation by other states in the longer term. 1 Nuclear Nonproliferation v. Nuclear This conflict arises only in the case of nuclear Deterrence weapons, since the Nuclear Non-Proliferation One way to reduce the appeal of nuclear Treaty prohibits not their existence but their weapons is to reemphasize the role they play in spread beyond the five acknowledged nuclear international relations. But to do so would mean powers. Although it would be extremely contro- that the nuclear powers must rely on them less, versial, one could imagine a change to the nuclear weakening the credibility and utility of U.S. nonproliferation regime that legitimized nuclear nuclear deterrent threats--especially those in- arsenals in additional states. The challenge would tended to deter military actions short of nuclear be to assure that such a change would not lead to attack. Conversely, to the extent that nuclear further proliferation, and further admissions to the

21 ~ unit~ Nations inspwtioxIs of Iraq, which have the right to go anywhere at any time, are part of a regime @Kxed upon a defeated power and cannot be considered a precedent for inspection provisions that states would accept voluntarily. Chapter 1-Introduction and Summary 29

Figure 1-3—Potentially Conflicting Approaches

Controlling spread of <-—-–”--— -–—–—> Controlling WMD plus weapons of mass conventional weapon destruction (WMD) systems

I Unilateral approach 4- —––-—-—–—–-–—+ International approach

Coercive policies < ‘“- ‘–--— -- ‘-—— —-—> Consensual policies

Targeted approach +—”– ‘-–– – ‘–-----– ~ Universal approach

Collecting information <––--—---— + Using information LI

SOURCE: Office of Technology Assessment, 1993. nuclear club, in future years. In the case of the planner, all such weapons in the hands of poten- Chemical and Biological Weapons Conventions, tial adversaries-mass destruction or conven- which impose global bans, exceptions that legiti- tional-make it more difficult to deter war or to mized the chemical or biological arsenals of one prevail should war erupt. Elements within the or more countries would be totally incompatible Pentagon now define proliferation as “the desta- with the treaties themselves. bilizing spread, especially to countries of concern in key regions, of a wide array of dangerous CONFLICTING APPROACHES military capabilities, supporting capabilities, al- In addition to choices between nonproliferation lied technologies, or know-how,” placing weap- and other policy objectives, balances must also be ons of mass destruction at one end of a consider- 22 struck between conflicting approaches to nonpro- ably broader spectrum of concerns. Such a view liferation policy. Like the preceding set, these argues for an integrated strategy in response. approaches--summarized in figure 1-3--do not Other linkages between weapons of mass represent diametrically opposed positions, but destruction and conventional weapons also exist. rather indicate opposing tensions that must be A given delivery system might be adapted to carry balanced against each other. either class of weapon, linking a state’s conventional military power with its capability to deliver B Scope of Control Efforts weapons of mass destruction. Similarly, one As noted in the beg inning of this chapter, the state’s conventional forces can motivate an oppo- proliferation of weapons of mass destruction nent to seek nuclear, chemical, or biological takes place within the context of the proliferation weapons in response, linking a state’s conven- of military capability more generally. Low- tional military forces with its opponent’s mass- technology weapons, advanced conventional weap- destruction weapons. As stated earlier, some ons, and the command, control, communications, analysts go so far as to argue that weapons of mass and intelligence infrastructure needed to use these destruction cannot be controlled in the long run weapons most effectively are all spreading around without effective limitations on other aspects of the world. From the point of view of a military military power as well.

22 Under s~re~ of Defense for Policy-Transition Worhg pap% “Spezial Transition Ropotiounterproliferation Strategy, ” Counterproliferation Initiative, Feb. 2, 1993. 30 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

A contrasting view notes that weapons of mass such cases, the United States will have to decide destruction differ markedly from other military whether to preserve consensus, at the risk of not systems in terms of international legitimacy. taking what it sees to be appropriate action, or to Whereas the pursuit of nuclear, chemical, and proceed unilaterally, at the risk of disrupting biological weapons can be deemed violations of international consensus. international norms, conventional arms have long Those analysts emphasizing consensus typi- been accepted as legitimate and necessary to a cally stress that proliferation should be treated as state’s self-defense. No objective standards exist a violation of international norms of behavior. to differentiate acceptable levels of conventional They argue that nonproliferation policies with military capability from levels that pose threats to widespread international legitimacy will be much “international peace and security. ” Therefore, more effective than ones viewed primarily as those extending nonproliferation policy to in- furthering the objectives of the United States or clude conventional armed forces cannot avoid any other single power. This approach empha- basing their decisions at least in part on their sizes the role of international institutions, as individual national interests-interests that may opposed to individual states, because such institu- not be shared by other nations. Mobilizing tions provide a greater degree of international effective international nonproliferation efforts support. However, a drawback of this approach is will therefore become that much harder. More- that unless each nation places high priority on over, those who see the proliferation of weapons nonproliferation, each will have reason to of mass destruction as far more dangerous than downplay it at times (see above discussion on the spread of conventional, or even advanced conflicting objectives). Therefore, the cases in conventional, weapons, would not wish to divert which consensus for international action can be effort away from the more serious threat. reached might be considerably fewer than many The linkages between conventional arms and states individually might wish. At the same weapons of mass destruction can give rise to time-and largely for the same reason-nations paradoxes. In the past, one policy measure have traditionally been reluctant to cede authority employed by the United States to reduce a state’s to international bodies that might later act in (e.g., Pakistan’s) motivation to acquire nuclear opposition to their own interests. Thus, even weapons had been to provide it with conventional when an international institution is able to iden- arms (e.g., F-16 aircraft) in order to address its tify a consensus position, it may not be able to do security needs. However, such planes can serve much. multiple roles, conceivably including the delivery Proponents of an ‘‘internationalist’ approach of Pakistan’s purported nuclear weapons. Thus, envision a world in which civilized nations agree conventional arms transfers intended to reduce a on strong norms against the development, acqui- state’s demand for weapons of mass destruction sition, production, threat, or use of weapons of may actually have the effect of increasing its mass destruction, possibly excepting some resid- capability to deliver them. ual nuclear capability in the nuclear weapon states. States unwilling to subscribe to these norms, or found to be violating them, would be 1 Balancing Unilateral and International considered by the others as pariahs. Such norms Approaches can come about if-and only if-a very large International cooperation is necessary for non- number of individual nations see them as compat- proliferation to be successful. However, other ible with their own national interests. In that case, nations will not always agree with the United states may be willing to use international institu- States on either the problem or the solution. In tions for real enforcement, based on information Chapter 1-Introduction and Summary 31

submitted by individual states or acquired di- in the long run, they can help raise the costs of rectly. such programs and so discourage states from Proponents of a more unilateral approach see pursuing them. In the near term, consensual proliferation as a threat to U.S. national interests policies may not be accepted by those states most against which the United States must take its own likely to develop or deploy weapons of mass steps, whatever other nations may think or do. destruction. In the long run, however, the most Although they would agree that nonproliferation effective nonproliferation measures are those in policies are more effective if implemented multi- which states decide that it is in their own best laterally, they argue that the United States should interest to forgo weapons of mass destruction. not restrict its actions to cases where consensus Like the case of unilateral and multilateral can be reached. In this view, international regimes approaches, coercive and consensual measures strong enough to implement a consistent multilat- can be mutually supportive. Yet states may eral approach are unlikely to emerge; at any rate, respond to coercion with defiance, refusing to the United States should not put itself in the join nonproliferation regimes. Conflicts between position of having to rely on them. Through the coercive and consensual measures may become use of ad hoc coalitions, such as that assembled particularly relevant in cases where potential for the Gulf War, the United States could gain the proliferants are in a position to export proliferation- benefit of a group response while still preserving sensitive technology. Punitive measures aimed at some freedom of independent action. The draw- discouraging the development of weapons of back of U.S. action in the absence of international mass destruction in such states may make it backing, however, is that it may antagonize other difficult to elicit their cooperation in forgoing states whose cooperation will be needed to proliferation-sensitive exports. implement a more effective multilateral policy. Moreover, if a proliferant state can portray itself not as an international pariah but instead as a 9 Balancing Targeted and Universal victim of superpower bullying, it can encourage Approaches other countries to withhold support from--or The targets of coercive U.S. nonproliferation even to undermine--U.S. nonproliferation pol- policies such as export controls and sanctions can icy. be chosen in one of two ways. In a targeted approach, the United States applies these measures to specific countries determined to be of 1 Balancing Coercive and Consensual particular proliferation concern. In a universal Policies approach, the target countries are not specified by Nonproliferation policies include both coer- name, but rather consist of all states that meet cive measures, imposed to frustrate efforts by given criteria such as violation of, or refusal to states to develop weapons of mass destruction, join, international nonproliferation treaty regimes. and consensual ones that invite states to voluntar- Given that the motivations for and the conse- ily forgo such developments. Coercive measures quences of developing weapons of mass destruc- tend to be directed against particular states. tion vary greatly from country to country, the Consensual ones, on the other hand, typically targeted approach provides greater flexibility and involve actions—such as joining treaty regimes— discretion for tailoring nonproliferation policy. In that any state is free to take, and they therefore particular, it permits the United States to treat avoid the need to single out targets. In the near states not considered security threats differently term, coercive measures can impede progress from states judged to be particularly dangerous to towards developing weapons of mass destruction; their neighbors or hostile towards U.S. interests. 32 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

For exactly this reason, however, targeted poli- new technologies. They also describe the level of cies are harder than universal ones to justify and effort, commitment, and resources required for implement multilaterally. Identifying some states any state to mount such developments, thereby and not others as causes for concern unavoidably indicating the relative effects of increasing these leads to charges of discrimination and double costs, for example, by export controls. Neverthe- 23 standards. Moreover, the states implementing less, a country-by-country analysis of potential multilateral nonproliferation policies will proba- proliferants’ indigenous technical expertise and bly not agree on who the problem countries are. industrial infrastructure is beyond the scope of this study. So, too, is a political assessment of the 1 Balancing Collection and Use of incentives facing each of these states, or a Intelligence thorough discussion of the many other nontechni- Much of the information available to the U.S. cal factors that would influence their ability to Government pertaining to proliferation is classi- pursue weapons of mass destruction. fied. Acting on classified information+. g., by The bottlenecks or ‘chokepoints’ identified in exposing a state’s actions to international atten- the background paper—steps that are particularly tion, or by shutting down a covert supplier time-consuming or difficult for proliferants to network-risks compromising the sources and master without outside assistance-might be methods by which the information was originally exploited to control proliferation. Conversely, collected, possibly shutting off access to such steps that are relatively easy, or that make use of information in the future. Therefore, tensions widely available know-how and equipment, make exist between collecting intelligence information and making effective use of it. poor candidates for control efforts. It is important A related problem is that to the extent that to understand the extent to which “dual-use” relevant information must remain classified, pub- technologies or products, which also have legiti- lic debate and discussion—and to some extent, mate civil applications, are involved in the international negotiations and actions-will be development of weapons of mass destruction, conducted on the basis of incomplete informa- since both the feasibility of controlling dual-use tion. items and the implications of doing so depend on the extent of their other applications. TECHNICAL BASIS FOR MONITORING AND Monitoring the proliferation of weapons of CONTROLLING PROLIFERATION mass destruction, or conversely monitoring com- The various weapons of mass destruction pliance with nonproliferation agreements, de- addressed in this report are based on very pends on detecting and identifying various indi- different technical principles and require distinct cators or signatures associated with the develop- sets of industrial capabilities. A separate back- ment, production, deployment, or use of weapons ground paper explores the technical pathways by of mass destruction. Unilateral intelligence col- which states might acquire nuclear weapons, lection efforts can seek to exploit these signatures chemical weapons, biological weapons, and de- with the use of remote or covertly placed instru- livery systems. Those analyses are intended to ments; multilateral verification regimes— identify opportunities for monitoring and, if typically operating within the framework of a possible, controlling proliferation, as well as to negotiated treaty-can make provision for states note the potential implications of certain old and to voluntarily open their facilities to cooperative

23 TJM u.S. State Dep~ent’S list of states supportingttzrori!nq for example, is often WCUSd Of refkthg pditkd tdts, rabr - firm intelligence analysis. — .

Chapter 1-Introduction and Summary ! 33

onsite inspection in addition to sanctioning the 1 Nuclear Weapons use of remote instrumentation.24

Both unilateral and cooperative approaches MATERIAL PRODUCTION have their strengths. A cooperative regime might In terms of costs, resources required, and offer direct access to facilities that would be possibility of discovery, the difficulty of ob- difficult to inspect in any other way. However, taining nuclear weapon materials—plutonium strict limitations may be put on that access. or highly enriched uranium—today remains Moreover, since the inspected party knows the the greatest single obstacle most countries type of instrumentation and procedures to be used would face in pursuing nuclear weapons. Even by inspecting parties, it may be able to defeat straightforward methods of producing such ma- those inspections. Intelligence collection efforts terial indigenously (such as building a small conducted outside the framework of a negotiated reactor and a primitive reprocessing facility to agreement would probably not have the degree of produce plutonium and recover it from irradiated access to any specific site that would be provided reactor fuel) would require at least a modest by a cooperative onsite inspection regime, but technological infrastructure and hundreds of mil- they might have other advantages such as breadth lions of dollars to carry out. Moreover, once such of coverage. Moreover, they would not be con- a facility became known, it could generate strained by prenegotiated procedures, and they considerable pressure from regional rivals or the might be able to gather information about sites international community. The costs of a full-scale where onsite inspection would be denied. How- indigenous nuclear weapon program-especially ever, if unilateral intelligence efforts involved if clandestine--can be substantially higher than covert placement of sensors in the territory of the for a program largely aimed at producing just one inspected party, such efforts would probably be or two bombs and carried out in the open. Iraq viewed as a violation of sovereignty, creating spent 10 to 20 times the cost of such a minimal political tensions if detected. program-many billions of dollars-to pursue Unilateral and multilateral approaches are not multiple uranium enrichment technologies, to mutually exclusive. Indeed, they will be most build complex and sometimes redundant facili- effective if used synergistically: unilateral intelli- ties, to keep its efforts secret, and to seek a fairly gence efforts might trigger a challenge inspec- substantial nuclear capability. Few countries of tion. However, many of the signatures dis- proliferation concern can match the resources that cussed below are likely to be ambiguous, if they Iraq devoted to its nuclear weapon program. (Iran, are detected at all. Deciding on appropriate however, probably could.) responses in the face of incomplete or ambiguous Since production of nuclear materials is information will pose great challenges for non- generally the most difficult and expensive part proliferation policy, as will mobilizing effective of producing a nuclear weapon, the leakage of domestic and international support for those significant amounts of weapon-grade material responses. from the former Soviet Union would provide a great advantage to potential proliferants. In-

24 ~ tie .s~at%ic ~ ~on~~l ~r~=~ ~~=n the United stat~ ~d the Soviet Union, e~h side agr~ not to impede the otk side’s ‘‘national technical means of verii7cation, “ in effect legitimizing the collection of intelligence pertinent to the treaty. . -. .

34 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

deed, the possibility of black-market sales of surplus of some 70 tomes of safeguarded, sepa- weapon-usable materials may represent one of the rated reactor-grade plutonium-the type pro- greatest proliferation dangers now being faced. duced by commercial nuclear reactors in normal Even the covert acquisition of low-enriched operation-will likely continue to grow through uranium, which can fuel nuclear reactors but is the 1990s by more than 10 tonnes per year. not directly usable for nuclear weapons, could be Reactor-grade plutonium is more radioactive and advantageous to a proliferant by enhancing the more difficult to handle than weapon-grade capacity of its isotope separation plants. plutonium, which is produced specifically for use This ominous prospect notwithstanding, nu- in nuclear weapons, but it can still be used to clear materials suitable for weapon purposes make a crude nuclear weapon of significant have to date been extremely difficult to obtain (though probably less predictable) yield. Never- from countries that already possess them. theless, the states that have sought nuclear There is no reliable evidence that any militarily weapons have gone to great lengths to produce significant quantities of nuclear weapon material weapon-grade materials--either highly enriched have been smuggled out of the former Soviet uranium or weapon-grade plutonium-rather than Union. The vast majority of nuclear material in reactor-grade plutonium. (Note that some types of nonnuclear weapon states is safeguarded by a nuclear power reactors, including ones in India, comprehensive system of material accountancy South Korea, and North Korea, can produce either and control administered by the International reactor-grade or weapon-grade plutonium, de- Atomic Energy Agency (IAEA). These safe- pending on how they are operated.) guards are not perfect, but they provide high levels of confidence that significant quantities of OTHER TECHNICAL BARRIERS nuclear material have not been diverted from Unlike chemical and biological weapons, whose safeguarded nuclear reactors. Diversion would be lethality is roughly proportional to the amount of more difficult to detect from facilities such as fuel agent dispersed, nuclear weapons will not pro- fabrication plants, uranium enrichment plants, duce any yield at all unless certain conditions are and plutonium reprocessing facilities that process met: a minimum ‘‘critical mass’ of nuclear large quantities of nuclear material in bulk form, materials must be present, and that material must as opposed to handling it only in discrete units be brought together with sufficient speed and such as fuel rods or reactor cores. At present, precision for a nuclear chain reaction to take however, there are no large facilities of this type place. A proliferant must master a series of under comprehensive IAEA safeguards in coun- technical hurdles in order to produce even a single tries of particular proliferation concern.25 At least working weapon. in the short run, the diversion of safeguarded Nuclear weapons are so destructive that they materials poses less of a threat to the nonprolif- place few requirements on the accuracy of deliv- eration regime than the black-market pur- ery systems for any but the most protected targets. chase or covert indigenous production of nu- Most proliferants would likely be able to design clear materials. first-generation nuclear weapons that were small Under current European and Japanese plans for and light enough to be carried by Scud-class reprocessing and limited reuse of plutonium from missiles or small aircraft. Given additional tech- commercial reactor fuel, the current worldwide nical refinement, they might be able to reduce

u Br~l has a m~i~-sized fiel fabrication facility under IAEA safeguards, and South African enrichment facilities are CO- Waler safeguards with South Africa’s announced destruction of its nuclar weapons and its accession to the IWPT Neither state is considered an active proliferation threat at present. Chapter 1-Introduction and Summary 35

, warhead weights to the point where the 500 kg (l,100 pound) delivery threshold originally established by the Missile Technology Control Regime no longer provides a reliable barrier to nuclear- capable ballistic or cruise missiles.26 Although nuclear weapons were first developed 50 years ago and the basic mechanisms are H widely known, much of the detailed design information, and particularly the knowledge gleaned by the nuclear weapons states from decades of design and testing, remains classified. Much of this information can be reconstructed by . . . a dedicated proliferant, but it will take time and Iraqi electromagnetic isotope separation (EMIS) money. Moreover, “weaponizing” a nuclear equipment, uncovered after having been buried in the warhead for reliable missile delivery or long desert to hide it from United Nations inspectors. Iraq’s shelf-life creates additional hurdles that could EMIS program to enrich uranium for nuclear weapons had not been detected by Western intelligence agencies significantly increase the required development prior to the Gulf War. effort. Therefore, having access to key individuals-- such as those from the former Soviet nuclear Treaty (other than the acknowledged nuclear- weapon program--could significantly accelerate weapon states) are not permitted to operate a nuclear program, primarily by steering it away unsafeguarded facilities handling nuclear materi- from unworkable designs. Specific individuals als, the existence of any such facilities would could fill critical gaps in a given country’s probably indicate an illegal weapon program.27 knowledge or experience, adding greatly to the Nuclear tests at kiloton yields or above would likelihood that a program would succeed. probably be detectable by various means, espe- High-performance computers (so-called “su- cially if multiple tests were conducted. However, percomputers ‘‘ in the 1980s) are not required to such tests are not necessary to field a workable design first-generation fission weapons. Thus, weapon with reasonably assured yield. Simi- placing strict limits on their exports would be of larly, the deployment of a small number of minimal importance compared with limiting tech- nuclear weapons might not be easily detected. nologies for nuclear materials production. IMPLICATIONS OF OLD AND NEW MONITORING NUCLEAR PROLIFERATION TECHNOLOGIES Production of nuclear materials provides many Low- and medium-level gas centrifuge tech- signatures and the greatest opportunity for detect- nology for enriching uranium may become in- ing a clandestine nuclear weapon program. Even creasingly attractive to potential proliferants for a so, a large part of the Iraqi program was missed. variety of reasons, including availability of infor- Since members of the Nuclear Non-Proliferation mation about early designs, difficulty of detec-

26 B~~~&.fig is fmu~, tie ~ssile ~c~ology con~l Regime IIOW covers XI&NflCS capable of delivering chfical ~d biological WtXipOllS as well aa those that could be used (o deliver nuclear weapons. Consequently, the payload threshold of 500 kg has been removed. 27 me excqtion t. ~s s~tement would ~ unsafe~d~ facilities dedi~ted to military purposes unrelated to nuclear weapons, suchm naval nuclear propulsion. Such uses are not prohibited by the Nuclear Non-Proliferation Treaty. They fall outside IAEA jurisdiction however, since IAEA safeguards pertain only to peaceful-e. g., nonmilitary-applications of nuclear power. See Ben Sanders and John SimpsoQ Nucfear Submarines and Non-Proliferation: Cause for Concern, PPNN Occasional Paper Two (Southampton England: Centre for International Policy Studies, University of Southampton for the Prograrnme for Promoting Nuclear Non-Proliferation 1988). —

36 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

tion, ease of producing highly enriched uranium, literature and follow from standard chemical and potential availability of equipment from the engineering principles. Unlike nuclear prolifera- former Soviet Union. Modem, state-of-the-art tion, where the mere existence of an unsafe- centrifuges could lead to even smaller, more guarded nuclear facility in an NPT member state efficient, and relatively inexpensive facilities that is often sufficient evidence of intent to produce would be most difficult to detect remotely. weapons, many legitimate chemical facilities In the longer run, laser isotope separation could have the ability to produce chemical agent. techniques and aerodynamic separation may Intent cannot be inferred directly from capa- have serious proliferation potential as means of bility. producing highly enriched uranium for nuclear weapons. Openly pursued by more than a dozen AGENT AND WEAPON PRODUCTION non-nuclear-weapon states, laser enrichment tech- Certain chemical agents such as mustard gas nologies use precisely tuned laser beams to are very simple to produce. Synthesis of nerve selectively energize the uranium-235 isotope agents, however, includes some difficult process most useful for nuclear weapons and separate it steps involving highly corrosive or reactive ma- from the more common uranium-238 isotope. terials. A sophisticated production facility to Laser facilities would be small in size and could make militarily significant quantities of one class enrich uranium to high levels in only a few stages. of nerve agents might cost between $30 and $50 They could therefore prove to be difficult to million, although dispensing with modem waste- detect and control if successfully developed as handling facilities might cut the cost in half. Some part of a clandestine program. Nevertheless, of the equipment needed may have distinctive considerable development work remains to be features, such as corrosion-resistant reactors and done before this method can be made viable or pipes and special ventilation and waste-handling can compete with existing enrichment technolo- equipment, but these can be dispensed with by gies. Even for the advanced industrialized coun- relaxing worker safety and environmental stand- tries, constructing operational facilities will re- ards and by replacing hardware as it corrodes. main very difficult. Some aerodynamic techniques— Moreover, production is easier if a proliferant which use carefully designed gas flows to sepa- country is willing to cut comers on shelf-life, rate the lighter uranium-235 from the heavier seeking only to produce low-quality agent for uranium-238-require fairly sophisticated tech- immediate use. nology to manufacture large numbers of precision Chemical-warfare agents can be produced small-scale components, but they do not other- through a wide variety of alternative routes, but wise pose technical challenges beyond those of relatively few routes are well suited for large- other enrichment approaches. scale production. Just because the United States used a particular production pathway in the past, however, does not mean that proliferant countries B Chemical Weapons would necessarily choose the same process. The technology used to produce chemical In general, commercial pesticide plants lack weapons is much harder to identify unambiguously the precursor chemicals (materials from which as weapons-related than is that for nuclear materi- chemical agents are synthesized), equipment, als production technology, and relevant know- facilities, and safety procedures required for how is much more widely available. Although nerve-agent production. Nevertheless, multipur- production techniques for major chemical weapon pose chemical plants capable of manufacturing agents involve some specialized process steps, organo-phosphorus pesticides or flame retardants detailed examples can be found in the open could be converted in a matter of weeks or months —

Chapter 1--Introduction and Summary 37

to the production of nerve agents. The choice between converting a commercial plant in this manner and building a clandestine production facility would depend on the urgency of a country’s military requirement for a chemical weapon stockpile, its desire to keep the program secret, its level of concern over worker safety and environmental protection, and the existence of embargoes on precursor materials and production equipment. Agent production, however, is several steps removed from an operational chemical weapon capability. The latter requires design and development of effective munitions, filling the munitions before use, and mating them with a suitable Portable gas chromatograph/mass spectrometer (GC/ delivery system. MS) developed to support onsite analysis for the Chemical Weapons Convention. This equipment can MONITORING CHEMICAL WEAPON detect and identify minute quantities of organic PROLIFERATION chemicals controlled by the CWC. Direct detection of chemical warfare agents in samples taken from a production facility would be and are configured in different ways depending a clear indicator of weapon activity, since these on the process involved. Moreover, many of the agents have almost no civil applications.28 How- same chemicals used to make chemical agents are ever, considerable access to production facilities also used to make pharmaceuticals, pesticides, is required to ensure that appropriate samples and other commercial products. Since many have been collected. Moreover, some of the different types of equipment are suitable for substances produced when chemical agents break chemical agent production, plant equipment per down in the environment are also produced when se does not provide a reliable means of distin- legitimate commercial chemicals break down, so guishing between legitimate and illicit activities. detection of final degradation products does not Nevertheless, some potential signatures of chemi- necessarily indicate agent production. Neverthe- cal weapon development and production exist, less, the suite of degradation products associated and a set of multiple indicators taken from many with a given chemical agent production process sources may be highly suggestive of a production would provide a clear signature. capability. Other than the agent itself, or an ensemble of Indicators at suspect locations that may con- degradation products, chemical agent production tribute to such an overall assessment include: has few unequivocal signatures. Moreover, highly visual signatures such as testing munitions and reliable technologies to detect chemical agent delivery systems; distinctive aspects of plant production from outside the site are not currently design and layout, including the use of corrosion- available. Unlike nuclear weapon facilities, which resistant materials and air-purification systems; generally exhibit fairly clear signatures, civilian presence of chemical agents, precursors, or degra- chemical plants have multiple uses, are hundreds dation products in the facility’s production line or of times more numerous than nuclear facilities, waste stream; and biochemical evidence of chem-

Z8 Ni@ogen musmds have some use in cancer chemotherapy, andphosgene and hydrogen cyanide haveindustiid applications. 38 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

ical agent exposure (including that due to acci- duce large numbers of casualties, although it is dental leaks) in plant workers or in plants and much more difficult to develop munitions that animals living in the vicinity of a suspect facility. have a predictable or controllable military effect. Nevertheless, the utility of specific signatures depends on how a given weapon program oper- AGENT AND WEAPON PRODUCTION ates, including the choice of production process The global biotechnology industry is information- and the extent of investment in emission-control intensive rather than capital-intensive. Much of technologies. Detection capabilities that are deci- the data relevant to producing biological agents is sive under laboratory conditions may be rather widely available in the published literature and inconclusive in the field—particularly if the virtually impossible for industrialized states to proliferant has been producing related legitimate withhold from potential proliferants. A wide- chemicals (e.g., organophosphorus pesticides) in spread support infrastructure of equipment manu- the same facility and is willing to expend time, facturers has also arisen to serve the industry. effort, and resources to mask, obscure, or other- Therefore, producing biological agents would wise explain away chemical agent production be relatively easy and inexpensive for any activities. Testing of chemical agents and training nation that has a modestly sophisticated phar- troops in their use might be masked by experi- maceutical industry. Moreover, nearly all the ments with or training for the use of smoke equipment needed for large-scale production of screens. A robust inspection regime must there- pathogens and toxins is dual-use and widely fore comprise an interlocking web of inspections, available on the international market. declarations, notifications, and data fusion and analysis, all of which a cheater must defeat in order to conceal his violations. Focusing monitoring efforts at a single point--even one thought to be a crucial chokepoint--would allow the cheater to focus his efforts on defeating them. Keeping a production program covert forces other tradeoffs. Some of the simplest production pathways might have to be avoided since they use known precursors or involve known production processes. Purchasing equipment from multiple suppliers to avoid detection, or jury-rigging facilities from used equipment, might increase hazards to the workforce and nearby populations.

I Biological Weapons Biological-warfare agents are easier to produce than either nuclear materials or chemical-warfare agents because they require a much smaller and cheaper industrial infrastructure and because the necessary technology and know-how is widely United Nations inspectors assessing the biological available. Moreover, it would not be difficult to weapon potential of Iraqi fermenters and other spread biological agents indiscriminately to pro- bioprocess equipment. Chapter 1-Introduction and Summary 39

One technical hurdle to the production of kilograms can be militarily significant. Since biological weapons is ensuring adequate contain- large amounts of agent can be grown up from a ment and worker safety during agent production freeze-dried seed culture in a period of days to and weapons handling, although the difficulty of weeks, large stockpiles of agent are not required, doing so depends on the level of safety and although some stocks of the munitions to be filled environmental standards. A government that with these agents would be. placed little value on the safety of plant workers There are no signatures that distinguish clearly or the civilian population might well take mini- between the development of offensive biological mal precautions, so that a biological weapon agents and work on defensive vaccines, since production facility would not necessarily be both activities require the same basic know-how equipped with sophisticated high-containment and laboratory techniques at the R&D stage. measures. Another challenge is ‘‘weaponizing’ Moreover, almost all the equipment involved in the agents for successful delivery. Since microbiological and toxin weapon development and bial pathogens and toxins are susceptible to production is dual-use and hence will not typi- environmental stresses such as heat, oxidation, cally indicate weapons activity. Indeed, the and dessication, to be effective they must main- capacity to engage in illegal military activities tain their potency during weapon storage, deliv- is inherent in certain nominally civilian facili- ery, and dissemination. ties. Some legitimate biological facilities can also A supply of standard biological agents for convert rapidly to the production of biological covert sabotage or attacks against broad-area warfare agents, depending on the degree of targets would be relatively easy to produce and sophistication of the plant and on the required disseminate using commercially available equip- scale of production, level of worker safety, and ment, such as agricultural sprayers. In contrast, environmental containment. At the same time, the integration of biological agents into precise, however, legitimate applications of biological or reliable, and effective delivery systems such as toxin agents (e.g., vaccine production and the missile warheads and cluster bombs poses com- clinical use of toxins) are relatively few at plex engineering problems. Nevertheless, the present. With the exception of a few vaccine United States had overcome these problems by production plants, such activities are largely the 1960s and had stockpiled biological warfare confined to sophisticated biomedical facilities not agents. normally found in developing countries, and these facilities generally do not engage in produc- MONITORING BIOLOGICAL WEAPON tion except on a small scale. Moreover, given that PRODUCTION the global biotechnology industry is still in its Detection and monitoring of biological and infancy, the number of legitimate activities-- toxin agent production is a particularly chal- from which the illegitimate ones would have to be lenging task. Even use of biological weapons distinguished-is still relatively small. could in some cases be difficult to verify un- Sensitive analytical techniques such as poly- ambiguously, since outbreaks of disease also take merase chain reaction (PCR) analysis or use of place naturally. Thanks to advances in biotech- monoclinal antibodies can identify trace quanti- nology, including improved fermentation equip- ties of biological agents and might be able to do ment as well as genetic engineering techniques, so even after the termination of illicit activities. biological and toxin agents could be made in However, the existence of such sensitive labora- facilities that are much smaller and less conspicu- tory techniques does not necessarily translate into ous than in the past. Moreover, the extreme a negotiated verification regime that might be potency of such agents means that as little as a few instituted to monitor compliance with the Biolog- 40 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

ical Weapons Convention. Other factors that must radiation). Biological agents might also be genet- be assessed in establishing such a regime include ically modified to make them more difficult to the likelihood of detecting clandestine production detect by immunological means and insusceptible sites, the ability to distinguish prohibited offen- to standard vaccines or antibiotics. At the same sive activities from permitted defensive efforts, time, genetic engineering techniques could be and the risk of divulging sensitive national- used to develop and produce protective vaccines security or proprietary information during inspec- more safely and rapidly. 29 tions of U.S. facilities. Cloning toxin genes in bacteria makes it Because of the difficulty of detecting clandes- possible to produce formerly rare toxins in tine biological and toxin weapon development kilogram quantities. Moreover, molecular engi- and production, effective tracking of such pro- neering techniques could lead to the development grams will require integrating data from many of more stable toxins. Even so, for the foreseeable sources, with a particular emphasis on human future, toxin-warfare agents are unlikely to pro- intelligence (agents, defectors, and whistle- vide dramatic military advantages over existing blowers). Some weaponization signatures (stor- chemical weapons. It is possible that bioregula- age of bulk agents, preparation of aerosol dispens- tors and other natural body chemicals (or syn- ers, field testing, etc.) would probably be easier to thetic analogues thereof) might be developed into detect than production signatures, but many such powerful incapacitants, but means of delivering signatures could be concealed or masked by such agents in a militarily effective manner would legitimate activities such as biopesticide R&D or first have to be devised. Moreover, if warning of use. Production and storage of components for their use were provided, chemical weapon protec- BW munitions might also be masked by activities tive gear would blunt their impact. associated with conventional weapons, such as production of high explosives, bomb casings, or artillery shells. Since excessive secrecy might 1 Delivery Systems itself be indicative of offensive intent, greater Although military delivery systems such as transparency would tend to build confidence in a ballistic missiles, cruise missiles, and combat country’s lack of offensive intentions. aircraft are not essential to deliver weapons of mass destruction, they can do so more rapidly, IMPLICATIONS OF NEW TECHNOLOGY more controllably, and more reliably than rudi- Genetic engineering is unlikely to result in mentary means such as suitcases, car bombs, or ‘‘supergerms" significantly more lethal than the civilian ships or planes. Controlling the spread of wide variety of potentially effective biological advanced delivery systems by no means would agents that already exist, nor is it likely to eliminate the dangers posed by weapons of mass eliminate the fundamental uncertainties associ- destruction, particularly in terrorist applications. ated with the use of microbial pathogens in However, limiting the availability of these deliv- warfare. However, gene-splicing techniques might ery systems would make it harder for states to use facilitate weaponization by rendering microor- weapons of mass destruction for military pur- ganisms more stable during dissemination (e.g., poses, particularly against well-defended, fore- resistant to high temperatures and ultraviolet warned adversaries.

29 mmited stiws m akidy determined that inspection procedures under the Chernieal Weapons Convention which allow the inspected party to negotiate the level of access to be provided to international inspectom, are sufilcient to proteet national-security information and trade secrets. However, it is not necessarily the case that the same inspection procedures would be suitable for the Biologicsl Weapons Convention should a formal verifkation regime be instituted. Chapter 1-Introduction and Summary 41

Unlike nuclear, chemical, or biological weap- for acquiring ballistic missiles-prestige, diversi- ons themselves, which are not traded openly due fying one’s forces, their psychological value as to treaty constraints or international norms, deliv- terror weapons, lack of trained pilots, and tech- ery systems such as aircraft and short-range nology transfer and export opportunities-will antiship cruise missiles are widely available on continue to make missile technology very attrac- international arms markets, Since the late 1980s, tive for several countries of proliferation concern. the United States and other Western industrialized countries have had some success at delegit- BARRIERS TO MISSILE AND AIRCRAFT imizing the sale of longer range ballistic and PROLIFERATION cruise missiles by creating the Missile Technol- The spread of ballistic missiles around the ogy Control Regime (MTCR), the participants in world was greatly facilitated by the export in the which refrain from selling ballistic or cruise 1970s and 1980s of Scud-B missiles from the missiles with ranges over 300 kilometers, or with former Soviet Union. With an increasing number any range if the seller has reason to believe that of countries abiding by the MTCR, the number of they may be used to carry weapons of mass potential missile suppliers has declined dramati- destruction. However, missiles with ranges up to cally. Of the principal missile exporters, only 300 km-and to a lesser extent, up to 600-1,000 North Korea has not agreed to comply. However, km-are already deployed in many Third-World Ukraine poses future export concerns, since it countries. Combat aircraft are possessed by al- contains much of the former Soviet missile most all countries of proliferation concern. Cruise production infrastructure, yet has not agreed to missiles or other unmanned aerial vehicles with comply with the MTCR. Moreover, additional ranges much over 100 km are not yet widespread countries have learned to copy, modify, extend outside the acknowledged nuclear weapon states, the range of, and produce their own missiles, and but large numbers of cruise missiles, including a small number have developed long-range sys- antiship missiles, are available at lesser ranges. tems-often in conjunction with space-launch In terms of payloads that can be carried to programs and foreign technical assistance. Even specified ranges, the combat aircraft of virtually so, MTCR constraints can slow the acquisition by all countries of proliferation concern far surpass developing countries of technologies associated their missile capabilities. However, aircraft and with more advanced missiles-those having ranges missiles have different relative strengths— particularly in their ability to penetrate defenses— s and the two systems are not fully interchangeable. Piloted aircraft have significant advantages over other delivery systems in terms of range, payload, accuracy, damage-assessment capability, and dispersal of chemical or biological agents. They can be used many times, usually even in the presence of significant air defenses. Missiles, however, are harder to defend against, and they offer distinct advantages for a country wishing to deliver a single nuclear weapon to a heavily defended area. Since missiles are not restricted to operating from airfields, they are also easier to hide from United Nations inspector measuring an Iraqi Al- opposing forces. The wide range of motivations Husayn (modified Scud) missile in Baghdad. 42 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

in excess of 1,000 km or guidance errors of less Glonass system, or commercial equivalents. Inex- than roughly 0.3 percent of their range. pensive, commercially available GPS receivers Given the complex set of technologies and are becoming available to provide unprecedented expertise used in advanced aircraft, especially navigational accuracy anywhere in the world. high-performance jet engines, it remains virtually Although GPS receivers would have only limited impossible for developing countries to acquire utility to emerging missile powers for ballistic these systems without assistance. However, no missile guidance, they could be used to reduce internationally binding restrictions limit trade in uncertainty in the launch location of mobile combat aircraft, and such arms transfers continue missiles. to be used as an instrument of foreign policy. Moreover, overcapacity in Western defense in- MONITORING DELlVERY VEHICLES dustries, and the economic difficulties facing Although individual missiles can be very newly independent Soviet republics and Eastern difficult to detect, a program to develop ballistic European states, provide great incentive to de- missiles is much more visible. Test firing and velop arms export markets. Therefore, states can launching ballistic missiles can be readily seen. and probably will continue to acquire high- Development of intermediate and long-range performance aircraft easily without having to ballistic missiles requires extensive flight testing, build them. Moreover, other options short of making it particularly noticeable. Although states buying aircraft or building them from scratch are pursuing both military and civil space technology available to states wishing to acquire or modify may wish to hide their military programs, civilian combat aircraft, such as engaging in licensed space-launch programs are usually considered a production.30 source of national prestige and proudly adver- If they have sufficient payload and range-and tised. if they can be procured despite export controls— Even a purely civilian space-launch program commercially available unmanned aerial vehicles provides technology and know-how useful for can be adapted to deliver weapons of mass ballistic missiles. The most important aspects of destruction without much difficulty. Developing a missile capability for weapons of mass destruc- cruise missiles requires greater technical capabil- tion—range and payload-can usually be in- ity. Even so, technologies for guidance, propul- ferred from a civil program. (A civil space-launch sion, and airframes are becoming increasingly booster does not need to have high accuracy, but accessible, particularly with the spread of li- neither does a missile carrying weapons of mass censed aircraft production arrangements to many destruction for use against populations.) On the parts of the world. The most difficult technical other hand, certain attributes desired for military challenges to developing cruise missiles— applications, such as reliable reentry vehicles, propulsion and guidance--do not pose much of a mobility, and ease of operation in the field, hurdle today. The highest performance engines suggest distinct technical approaches for military are not required for simple cruise missiles, and and civil applications. Although solid-fueled many sources are available for suitable engines. boosters are in some ways more difficult to Guidance requirements can be met by satellite develop and build than liquid-fueled boosters, navigation services such as the U.S. Global they are easier to use in mobile and time-urgent Positioning System (GPS), possibly the Russian applications. Liquid-fueled boosters were the first

30 me ~outti “~ou~ ~~te~ ~o~d tie world ~ve ~en to develop defe~e ~dustri~, inclutig aircraft industries, ~ diSCUSSd h U.S. Congress, OffIce of ‘Ikchnology Assessment, Global Anns Trude, OTA-ISC-460 (T%.shingtou DC: U.S. Government Printing Office, June 1991). Chapter 1--Introduction and Summary 43

used in military applications and are still more detect. Several types of unmanned aerial vehicles common. (The seemingly ubiquitous Scud mis- are being developed and marketed for civil sile and its modifications, such as were launched purposes, and without inspection rights it will be by Iraq against targets in Israel and Saudi Arabia, difficult to discern whether such vehicles have are liquid-fueled.) been converted to military purposes. Therefore, Since combat aircraft are widely accepted as monitoring of delivery systems capable of integral to the military forces even of developing carrying weapons of mass destruction will countries, there is no reason to hide their exis- continue to be an uncertain exercise, having tence. Individual planes, however, can be hidden. most success with missiles and highly capable Moreover, modifications made to aircraft to carry aircraft. Nevertheless, the risk posed by other weapons of mass destruction, or training given to delivery systems cannot be dismissed. The full pilots for their delivery, might be difficult to range of delivery technology must be taken detect without intrusive inspections. into account when evaluating a country’s Of the three delivery systems, cruise missile overall proliferation capabilities and behavior. development and testing will be the hardest to —— —-

Assessing the Risks 2

weapons kill and maim, but those commonly referred 11to as “weapons of mass destruction ’’-nuclear, chemical, and biological-an do so on an especially large scale. What is more, many of these weapons can cause notA just instant death or mutilation, but lingering disease and suffering. This chapter begins with a review of the basic characteristics of the devices of mass destruction. It also identities various ways these weapons might be delivered. The next section of the chapter then compares the destructive effects and possible military uses of these weapons. This overview of the weapons, their effects, and their uses serves two purposes. First, it illustrates why the weapons have been singled out for particular opprobrium and special efforts at control. Second, it indicates why states (and sometimes nonstate organizations) might come to believe that the weapons would be useful additions to their c arsenals. :@ The proliferation of weapons of mass destruction is both a general and a particular problem. In the long term, dealing with the problem will require strengthening international norms against the weapons and fostering a political order that makes them unattractive. In the nearer term, however, proliferation problems are particular: the weapons are spreading to specific nations that have decided, for one reason or another, that the existing international norms against further proliferation should not apply to them. The third part of this chapter identifies states suspected of trying to acquire one or more types of weapon of mass destruction. This discussion leads to the conclusion that the immediate proliferation threats, as now understood, are serious but still limited in scope. Many more nations are economically and technically capable of building weapons of mass destruction than are actually trying to do so. All this is not to suggest

45 46 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

complacency about the dangers of proliferation: Table 2-1 surveys the destructive agents dis- rather, it suggests that national and international cussed in this report. Table 2-2 identifies factors nonproliferation policies actually have some pros- that can affect just how lethal these agents may be pect of containing the problem. when used. To do their deadly work, these agents The fourth part of ch. 2 discusses what differ- of mass destruction have to be incorporated into ences it may make to the international community weapons (e.g., an aerial bomb, a ballistic missile in general, and to the United States in particular, warhead, an artillery shell, or a even a suitcase) when weapons of mass destruction spread. This and then delivered. Table 2-3 lists the kinds of analysis underscores the strong interest in con- weapons that have been, or in principle could be, taining the threat of proliferation that the United designed for nuclear explosives or chemical or States shares with all civilized nations. biological agents. During the Cold War, the With ch. 1 having reviewed trends in the United States and the Soviet Union developed international arena that nonproliferation policies every type of nuclear weapon listed in the 4 must take into account, the final section of this chart. chapter calls special attention to the multifari- The easiest course for nuclear proliferant ous problems posed by the breakup of Soviet nations would be to try to build aerial bombs first, Union. because these need not be as light or compact as other weapon types. The bomb dropped on Hiroshima weighed about 4,400 kg (9,700 lb), but WEAPONS OF MASS DESTRUCTION proliferants should be able to do much better than “Mass destruction” is a relative term. Allied that on their first try.5 The countries currently fire bomb attacks on Dresden during World War suspected of nuclear weapon ambitions also have II killed between 130,000 and 200,000 people ballistic missile programs; even if they succeed in with 1,400 aircraft sorties over 2 days.1 A single developing heavier aerial bombs sooner, they atomic bomb killed about 68,000 people and seem likely to pursue missile-capable nuclear injured another 76,000 in Hiroshima.2 A 1- explosives in the longer run. Iraq appears to have megaton hydrogen bomb exploding over Detroit been trying to make its first nuclear weapon light might kill 470,000 and injure 630,000 more.3 enough for a missile warhead. Thus, a single weapon of mass destruction can do Chemical weapons were frost used extensively damage equivalent to that of hundreds or thou- in World War I. Initially, gaseous agents, such as sands of ‘conventional’ high explosive or incen- chlorine and phosgene, were released from ground- diary weapons. This report addresses the spread based tanks as airborne clouds; later, liquids such of three broad types of weapon meeting that as sulfur mustard were delivered in artillery criterion of killing more with less: nuclear, shells. Aerial bombing and spraying methods biological, and chemical. How do these weapons appeared between the two World Wars. During injure and kill? the Cold War, the United States and the Soviet

1 Science Applications Inc.,EvaZuatims of CoZZareraJ Damuge (La Jolla, CA: SAIC, Nov. 15, 1976), p. 131, 2 Samuel Glasstone and Philip J. DoIan, (eds.), The Eflects of NucZear Weapons, Third Ed”tion (Washington DC: U.S. Department of Defense and U.S. Department of Energy, 1977), p. 544.

3 U.S. Congress, Offke of ‘Rdmology Assessment The Eflects ofNuclear War (Washington, DC: U.S. Gov ernment Printing office, 1979), p. 37. 4 In the aftermath of the Cold War (and even before), both superpowers begsn to withdraw from service most of their so-called tactical nuclear weapons. S The United States and the Soviet Union had already deployed much lighter weapons by the 1950s. — .

Chapter 2–Assessing the Risks 47

Table 2-l—Weapon Agents of Mass Destruction

Type of weapon agent Examples Mechanism Effects on human beings

Nuclear: fission and fusion Hiroshima fission bomb= 12.5 Blast (overpressure) Bleeding and rupture; violent kt (1 kt = 1,000 tons TNT); displacement; blows or crush- fusion bomb, e.g., largest U.S. Thermal radiation ing by debris test = 17 Mt (1 Mt = 1,000,000 tons TNT) Nuclear radiation (immediate) flash burns, blinding, burning or suffocation from building fires Nuclear radiation (delayed effects and fallout effects) Vomiting, diarrhea fever, bleeding, infection, circulatory failure, respiratory failure, brain swelling

Above effects at high doses; contact burns, cataracts, leu- kemia, other cancers, birth de- fects at lower doses

Biological: viruses Venezuelan equine encephalitis Inhaled or ingested infectious A variety of debilitating or poten- diseases tially fatal illnesses

bacteria Anthrax, brucellosis, plague (same) (same)

rickettsiae Q fever, typhus (same) (same)

Toxins: 2 Botulin, ricin, animal venoms Inhaled or ingested poisons A variety of toxic effects, often fatal

Chemical: Blistering (Vesicants) Mustard, Iewisite Skin and tissue destruction on Skin blistering, blindness, po- contact or inhalation tentially fatal lung damage

Choking Chlorine, Phosgene, PFIB Lung damage on inhalation Fluid build-up leading to fatal choking

Blood Cyanogen chloride, hydrogen Blocking of blood oxygen on Anoxia (severe oxygen starva- cyanide inhalation tion of body tissues)

Nerve Tabun (GA), Sarin (GB), Soman Nervous system disruption on Convulsions, paralysis leading (GD), GF, VX contact or inhalation to death

1 Somechemical and biological agents maycauseirritation, illness, or behavior changes, butmaynotnormally be fatal; weapons using th=eagents may incapacitate people for hours, days or weeks, but cannot be accurately said to inflict mass destruction. Other agents can destroy livestock or crops, having great potential for economic warfare but (except for the possibility of causing mass starvation) not leading immediately to widespread human injury. 2 Toxins are nonliving, ~ionous ~hemica~, first produ~d in biological proc~ses. k w= therefore r~onable to cxmsider them to& biological weapons, and they are covered in the international treaty banning biological weapons. However, as toxic chemical (nonliving) substances, they are also categorized as “chemical’ ’weapon agents-andthey are so considered in the Chemical Weapons Convention banning chemical weapons.

SOURCE: Office of Technology Assessment, 1993. 48 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 2-2—Factors Affecting Lethality of Nuclear, Biological, and Chemical Weapons Factor Nuclear Biological Chemical Delivery modes Higher altitude burst increases Aerial spraying produces wider Aerial spraying produces wider lethal area, decreases fallout; lethal area than explosive bomb lethal area than explosive bomb Lower altitude increases cen- or missile warhead or missile warhead tral blast and fallout Explosive dispersion may also Persistent agents can injure or kill agent organisms kill additional victims as they pass through a contaminated area Terrain Open, flat Increases exposure to thermal Maximizes lethal dispersion of Maximizes lethal dispersion of and ionizing radiation agent agent Decreases injuries from debris, collapsing structures

Hilly Redirects blast effects Atmospheric turbulence im- Atmospheric turbulence im- May shield from thermal and pedes even distribution and pedes even distribution and ionizing radiation increases vertical dilution of increases vertical dilution of agent, reducing casualties agent, reducing casualties

City Supplies material for injurious Atmospheric turbulence im- Atmospheric turbulence im- debris pedes even distribution and pedes even distribution and increases vertical dilution of increases vertical dilution of Masonry may shield from blast, agent, reducing casualties agent, reducing casualties debris, and radiation

Building collapses increase in- Buildings partially shelter from Buildings partially shelter from juries agent agent

Wood buildings and petrochem- icals burn, may produce lethal firestorm

Weather Wind, rain patterns may either Wind may blow agent away Wind may blow agent away increase or decrease lethal dis- from or toward targets from or toward targets tribution of radioactive fallout Air temperature and tempera- Air temperature and temperature gradient affect dispersal ture gradient affect dispersal

Union deployed the gamut of chemical delivery the most part have been well concealed.7 But systems from spray tanks6 to chemical warheads apparently weapon designs have included spray- for short-range ballistic missiles, rockets, land tanks, bombs, cluster bombs, and bomblet dis- mines, bombs, and artillery. The Iraqi chemical pensers.8 Like chemical weapons, biological agents arsenal included artillery shells, bombs, and some are best dispersed as low-altitude aerosol clouds. ballistic missile warheads. (Moreover, explosive methods of dispersion may If any live biological weapons have been used destroy the organisms.) Ballistic missile war- in the twentieth century, their characteristics for heads that can effectively generate aerosols are

s Chemical agent is most efficiently delivered as a spray at low altitudes. 7 As noted in table 2-1, from the delivery and effects standpoints, toxins are closer to being chemical than biological weapons. Japan seems to have attempted limited biological agent attacks in China during World War II-apparently with inconclusive effeets. See below, footnote 23. s World Health Organization Hea(th Aspects of Chemical andBiological Weapons (Geneva: World Health Organii@ioq 1970), p. 84. See also Stockholm International Peace Research Institute,The Problem of Chem”cal and Biological Wa#are: Vol II, CB Weapons Ted@ (New Yorlq NY: Humam‘ties Press, 1975), pp. 83-89. Chapter 2–Assessing the Risks 49

Table 2-2--(Continued)

Factor Nuclear Biological Chemical

Weather (continued) High winds disperse farther, High winds disseminate far- but may dilute lethal concen- ther, but may dilute Iethal con- trations sooner centrations sooner Rains may clear air, wash away Rains may dear air, wash deposited agent away some types of deposited agent Sunlight or drying rapidly de- stroys some agents Cold weather prevents evapo- ration, reducing concentrations in air but lengthening period of ground contamination

Defensive measures Shelters offer varying degrees Immunization possible if agents Antidotes for some agents can of protection outside central known in advance, but mas- be effective if administered soon destructive area of bomb sive exposures can overwhelm enough after exposure; limited immunity preventive treatment also pos- With ample preparation and sible for nerve agents warning, evacuation the best Depending on agent, early med- protection ical treatment can reduce mor- Early decontamination and med- tality rate ical treatment can reduce mortality With adequate detection and warning, special masks and With adequate detection and clothing or buildings or vehi- warning, special masks and cles with filtered, positive inter- clothing or sealed buildings or nal air pressure can protect vehicles with filtered, positive effectively internal air pressure can protect effectively Surfaces can be decontaminated Surfaces can be decontaminated With ample preparation and warning, evacuation the best With ample preparation and protection warning, evacuation the best protection

SOURCE: Office of Technology Assessment, 1993. technically challenging to design, although the The biological agents usually considered for United States had done so by the 1960s. Long- warfare have been infectious (multiplying within term storage of missile or artillery warheads filled the infected person) but not contagious (spreading with live or freeze-dried biological or toxin agent from one person to another). A nation contemplat- is difficult (except for anthrax spores); even if ing the military use of contagious agents would refrigerated, most of the organisms have a limited have to consider the following problems: lifetime. Small, unmanned aerial vehicles carry- ● the spread of the disease might be so slow as ing spray tanks might become an appealing to dilute the military impact; option for third-world countries seeking inexpen- ● there would be a risk that the disease would sive weapons of mass destruction.9 spread back to the attacker; thorough vacci-

9 See W. Seth Carus, “ ‘The Poor Man’s Atomic Bomb?’ Biological Weapons in the Middle East” (Washington DC: The Washington Institute for Near East Policy, Policy Papers No. 23, 1991), p. 11. 50 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 2-3--Weaponizing Agents of Mass Destruction: Actual and Possible Methods of Delivery

Weapon Nuclear Biological Chemical Aerial bomb J d d Bomb subminitions d 4 Aerial spray tank d d Ballistic missile warhead, nonseparating d d d Ballistic missile warhead, separating d (poss.) (poss.) Reentry vehical Artillery Shell d d d Rocket Shell d d d Mortar shell d d Cruise missile warhead 4 (poss.) (poss.) Mine (land) d d Mine (sea) d Antiaircraft missile warhead d Torpedo d Transportable Clandestine Bomb 4 (P0ss ) (poss.)

Actual Cases d Theoretical possibility (poss.)

SOURCE: SIPRI, 1975 and Office of Tec+mology Assessment, 1993.

nation of one’s own troops and population delivery. This section introduces the primary would be very difficult; and kinds of vehicles by which these weapons might . many nations not at war with, and perhaps be delivered. Following sections address possible even allied to, the attacker might also suffer uses. As table 2-2 illustrates, properly configured from the epidemic, exposing the attacker to weapons of mass destruction can be delivered by unnecessary sanctions or retaliation. many kinds of military delivery systems, and at A clandestine terrorist might not care about these ranges from a few to thousands of kilometers. problems. A nation with an advanced biotechnol- Depending on the scenario, such weapons can be ogy program might try to create a contagious highly threatening even without sophisticated organism that was both difficult to treat and military delivery systems. A nuclear device susceptible to a vaccine uniquely available to the planted by terrorists or commando squads, or nation’s own population. These conditions would delivered by disguised cargo ships, aircraft, or be difficult to achieve, as well as to implement even small pleasure craft, could kill just as many covertly on a nationwide scale, but cannot be people as one delivered by an ICBM; a given dismissed as impossible (see the OTA back- quantity of lethal microorganisms effectively ground paper on technologies underlying weap- spread by human agents might kill even more than ons of mass destruction, in press). one delivered by missile. Thus, the absence of advanced delivery sys- 9 Means of delivery tems does not mean that states or sub-national How nuclear, biological, or chemical weapons groups could not use weapons of mass destruc- are incorporated into weapons will depend both tion. Even though few proliferant states (with the on their purposes and on the available means of Chapter 2–Assessing the Risks 51

possible exceptions of India, Israel, and China)10 may fly at speeds of 1,000 to 2,000 km/hr or have---or are likely soon to acquire-military more. delivery systems capable of directly reaching the Ballistic missiles carry both fuel and oxidizer United States, unconventional delivery methods and (except for very short-range systems) fly part could still put U.S. territory at risk. U.S. allies of their trajectory outside the atmosphere. They abroad or deployed U.S. forces are already usually reenter the atmosphere hypersonically, at threatened by shorter range systems. In the cases speeds of thousands of km/hr, and carry smaller of rival states bordering one another, nuclear, payloads than aircraft. They are likely to delivery biological, or chemical weapons mounted on weapons less accurately than aircraft, but high even very short-range means of delivery can pose accuracy is unnecessary for delivering nuclear a major threat. weapons on many kinds of missions. Nevertheless, states possessing considerable Cruise missiles or other unpiloted aerial vehi- numbers of advanced longer range systems equipped cles share some characteristics with piloted air- with these weapons can more reliably threaten craft and some with ballistic missiles. Like more nations with higher levels of destruction airplanes, these missiles fly nonballistic trajecto- than those only possessing short-range systems. ries within the atmosphere and are powered Every state currently of proliferation concern has throughout their flight. (Ballistic missiles, in combat aircraft in principle capable of delivering contrast, are powered only at the beginning of weapons of mass destruction; most of those states their trajectories, coasting to their targets once also own or have programs to acquire ballistic their final rocket motor stages burn out.) Unpi- missiles. loted aerial vehicles range from simple, un- Partly for these reasons, the discussions of manned drone aircraft used for target practice, to delivery systems in this OTA report and its short- and medium-range (10 to 100 km) antiship associated background paper deal primarily with missiles widely available around the world, to 11 advanced systems. Another reason is a more highly sophisticated, longer range, autonomously practical one: because advanced systems de- guided missiles such as the U.S. Tomahawk. Like signed to penetrate enemy defenses are in many ballistic missiles, they do not require as extensive cases technically more demanding, there is greater a training and support infrastructure as do piloted hope (than there is for short-range systems) of aircraft. imposing international controls on their further Since cruise missiles can be launched from air, proliferation. sea, and underwater as well as from land, their The three principal types of advanced delivery own range is extended by that of their carrier vehicle are aircraft, ballistic missile, and cruise (airplane, ship, or submarine). Like other aircraft, missile. Aircraft (in the sense used here) are they can be shot down once detected and identi- piloted, air-breathing (usually jet) airplanes; the fied. However, small size (and radar cross- combat aircraft of many of the world’s air forces section), low-altitude flight, and circuitous courses can deliver payloads of several thousand pounds can make them hard to find. Like ballistic to distances of hundreds of kilometers (or more, missiles, they are expendable, eliminating the if they are equipped for aerial refueling) 12; they need to risk a pilot (and avoiding the possibility

10 DW~~ nuclW.weaPn states are not considered nUCleW ‘Profimants. ‘‘ China, although a declared nuclear-weapon state, is suspected of being a chemical and biological weapon proliferant. 11 Backgomd Pqa on technologies underlying weapons of mass deswctio~ in press.

12 conve~~ ~ago ~m~t or long-rmge bomber ~c~t (if avai~ble) cm fly ~ousands of kilometers without refhe~g. 52 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

of pilot error while forfeiting the potential for WEAPON EFFECTS COMPARED pilot improvisation if something goes wrong). With the worldwide availability of high- 1 Destructive Effects precision navigation services such as that pro- Figures 2-1 and 2-2 illustrate some rough vided by the U.S. Global Positioning System estimates for the effects of comparable amounts (GPS), a simple cruise missile can in principle be of chemical, nuclear, and biological weapons. made more accurate than even a sophisticated These are based on somewhat arbitrary assump- ballistic missile. A piloted aircraft, on the other tions, but they do give a basis for relative hand, has abetter chance than a cruise or ballistic comparison of the weapon types. missile of delivering chemical or biological weapons onto mobile military targets and of These comparisons suggest the following gener- adjusting bombing or spraying patterns to the alizations: ● weather. nuclear weapons remain the most massively An analysis comparing the relative advantages destructive weapons that can be built: unlike chemical and biological weapons, nuclear of aircraft and ballistic missiles for nuclear weapons also threaten massive destruction weapon delivery concludes: of property (civilian or military); Ballistic missiles are of principal concern to ● in principle, biological weapons efficiently the degree they are coupled to the delivery of delivered under the right conditions against nuclear and, to a somewhat lesser extent, chemi- unprotected populations would, pound for cal weapons. But advanced-strike aircraft can be pound of weapon, exceed the killing power effective in delivering nuclear weapons and can of nuclear weapons; on the other hand, if be more effective than ballistic missiles for 13 warning is provided, effective civil defense delivering conventional or chemical ordnance. measures are considerably easier to take On the other hand, to deliver weapons at interconti- against chemical and biological weapons nental range, developing and building intercontinen- than against nuclear weapons; tal ballistic missiles may be easier than acquiring ● for maximum physical effect, chemical and long-range bombers and refueling capabili- biological weapons are more efficiently de- 14 ties. livered by aircraft or artillery barrages than Since biological warfare agents are, like chem- by high-speed missiles; missile attacks, how- ical ordnance, best disseminated in an aerosol ever, may be useful as instruments of terror; over a wide area, aircraft and cruise missiles are ● chemical weapons must be delivered in great better for delivering them than are ballistic quantities to approach the potential lethality missiles. In addition, it is more difficult (but not of nuclear and biological weapons; against impossible) to develop ballistic missile warheads well-protected troops or civilians, they will in which live biological agents can survive the be less lethal than even conventional explo- stresses of space flight and atmospheric reentry. sives; and

13 ~nta for ~tmtio~ sec~~ md Arrns Control, Stanford University, Assessing Ballistic Missile Proliferation and Its COnPOl (Stanford, CA: CISAC, November 1991), p. 7. 14 See ~wad ~~~ foreword to Seti Carus, Ballistic Missiles in Modern Conflict (hkw York NY: ~eger, 1~1) P. vii Chapter 2–Assessing the Risks 53

Figure 2-l-Comparing Lethal Areas of Chemical, Biological, and Nuclear Weapons: Missile Delivery on an Overcast Day or Night, With Moderate Wind (Neither Best nor Worst Case)

(All diagrams in Figures 2-1 and 2-2 are to same scale) = Approx no of deaths, assuming Sarin nerve gas, 300 kg, 70 mg-min/m 3 Iii 3,000 to 10,000 unprotected people/km2 - :, ,. 022 km 2 Scale & - -— — — - ——— j xii 60-200 10 km

Anthrax spores, 30 kg, O 1 mg-min/m 3

Atomic bomb, 125 kt TNT-equiv (Hlroshlma-size), area of 5 lb/in overpressure

:....9”” ‘“’” ;. 78 km2 ..,:,,, ..,,. .<,.:” 23,000-80,000 x

..-!4 xiii 570,000-1,900,00

Figure shows the lethal areas of the agents delivered by a Scud-like missile with a maximum payload of 1,000 kg (note that the amount of biological weapon agent assumed would weigh considerably less than this; since the lethality per unit weight is great, the smaller amount considered here would still more than cover a large urban area). The estimates of lethal areas for chemical and biological weapons were prepared using a model that takes account of postulated release height, wind velocity, deposition velocity, height of temperature inversion layer, urban air currents, and residence time in air of the agent. The diagrams show approximate outer contours of areas with sufficient concentrations of agent that 50 percent to 100 percent of the unprotected people would receive fatal doses. Although some people within the defined area would survive, about the same number in the outer, less lethal areas, would die; therefore, the defined areas give approximations of the total number of unprotected people who could be expected to die in each scenario, With ideal (forlethalit y) population densities and weather, the chemical and biological agents could kill more people than shown here; under worse conditions, they might kill many fewer. The atomic weapons (fission and fusion) are assumed to be ah burst for optimum blast and radiation effects, producing little lethal fallout. The lethal area is assumed to be that receiving 5 lb/in2 of overpressure--enough to level wood or unreinforced brick houses. SOURCE Office of Technology Assessment, 1993. 54 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Figure 2-2-Comparing Lethal Areas of Chemical and Biological Weapons: Delivery by Aircraft as Aerosol Line Source

= Approx. no. of deaths, x assuming 3,000-10,000 people / km 2

Scale ~. — - — - _ -— j 10 km Clear, sunny day, Overcast day or light breeze night, moderate wind Clear, calm night

Sarin nerve “gas”

2 ‘“:.::: 2 ~ 74 km -V 8 k m 2 /;:” 78 km A: 300-700 400-800 3,000-8,000 H R H

Anthrax spores

,..

\ / ,.., / 140 km 2 ,,.

km2

130,00-460,000 420,000-1,400,000 1,000,000- xii la xii 3,000,000

Figure shows the lethal areas of single airplane-loads of chemical and biological weapons, assuming a highly efficientf line-source delivery of the killing agents. The figure also assumes that the aircraft has a greater payload than the missile in figure 2-1, delivering 1,000 kg of sarin nerve agent or 100 kg of anthrax spores. (More anthrax would be inefficient in a city attack.) Given these two factors, a single airplane delivering chemical or biological weapons can be considerably more lethal than a single missile. For an anthrax attack, the diagram shows how fatalities could vary greatly under three different weather scenarios. In one case, that of an overcast day or night with moderate wind, maximizing the lethal area would require distributing the agent in a 4.5 km by 34 km area, which would not be appropriate for most cities; therefore, the figure assumes a more rectangular distribution, which would still generate a comparable number of casualties. SOURCE: Office of Technology Assessment, 1993. ——..———

Chapter 2–Assessing the Risks 55

● because they are so dependent on weather applied tactically or strategically. From a strictly and the degree of defensive protection, the military viewpoint, though, their utilities are not consequences of chemical and biological equivalent because the consequences of using weapons are much less predictable than each are different. those of nuclear weapons; nevertheless, even when military utility is questionable, chemi- TACTICAL USES cal and biological weapons may terrorize The tactical uses of weapons of mass destruc- civilian populations (and, particularly for tion may have both direct and indirect purposes. terror uses, the attacker may be able to wait The direct purpose would be to destroy or disable for optimal weather). specific military targets-bases, equipment, or personnel. The indirect purpose would be to I Military Utility compel the enemy to change his operations to A principal-but by no means exclusive— cope with extraordinary threats. For example, the motive for developing countries to acquire weap- threat of nuclear attack might force the dispersal ons of mass destruction is for their potential of large troop or armor concentrations. Chemical military utility. The symbolic, deterrent, or intim- weapon threats might require troop dispersals and idating uses of these weapons may be dispropor- protective measures that reduce combat effective- tionate to their actual effects on opposing military ness, while perhaps overburdening medical serv- forces. 15 On the other hand, political inhibitions ices with injured personnel. Biological weapon on using the weapons may render moot their threats would impose similar burdens. purely military effectiveness. These cautions During the Cold War, the United States and the noted, table 2-4 compares general military uses of Soviet Union each deployed thousands of ‘tacti- the weapons (along with conventional explosive cal’ ‘ nuclear weapons (e.g., artillery, rockets, weapons, for comparison). short-range ballistic missiles, aerial bombs). For The nuclear age has generated special mean- the most part, these weapons would have been ings for the terms strategic and tactical as applied aimed at military forces and installations more or to weapons of mass destruction. The general less directly involved in battle. Their explosive meaning of strategic military action in this yields reportedly ranged from less than 1 kiloton context is the attempted destruction of the mili- (kt) of TNT to several hundred kt.16 Limited use tary infrastructure, economic base, and even the of these weapons in Europe would, at the very population that enables the enemy nation to make least, have forced armies on both sides to alter war. Tactical attacks, on the other hand, are those their tactics to avoid presenting large concentra- more directly engaging the enemy’s frontline tions of troops and armor as targets. Used en military forces, immediate reinforcements, or masse, these tactical weapons might have pro- supply lines. As table 2-4 indicates, nuclear, duced damage to the population and the civilian chemical, and biological weapons can each be infrastructure resembling that to be expected from

15 III some unuti cases, the uses may intend provocation, as opposed to deterrence or intimidation. The apparent intent Of kW’S use of Scud missiles against Israel during the Gulf War was to lure Israel into military retaliation. Iraq might then have persuaded the Arab members of the Coalition to change sides to avoid fighting on the same side as Israel. Thus, even though the Iraqi attack would probably have had little direct military effect its political effect might have been enormous. (Iraq might have increased the chances of Israeli military action had it used chemical warheads on its missiles; on the other hand, Iraq also had to consider the possibility that Israel would respond to such an attack with nuclear weapons, an escalation Saddam Hussein probably wanted to avoid,) 16 See william M. &kin and RiCharci W. FiddhOUse, Nuclear Battlefields: Global Lirk in the Arms Race (Cambridge, m: Ball@r, 1985), pp. 57-58. By way of comparison the bomb that destroyed Hiroshima had a yield of 12.5 kt, Table 2-4-Applications of Weapons of Mass Destruction Compared

Conventional explosives Characteristics (for comparison with WMD) Nuclear Chemical Biological Destructive effects Blast, shrapnel, fire Blast, fire, thermal radiation, Poisoning: skin, lungs, nerv- infectious disease or biochemi- 0 (See table 2-1) prompt ionizing radiation, ra- ous system, or blood cal poisoning dioactive fallout1

Typical militarytargets Military bases and equipment; Similar to targets for conven- infantry concentrations, towed infantry concentrations, air bases, Command-and-control installa- tional munitions (esp. targets artillery, air bases, ships, ports, ships, ports, staging areas, com- o tions (e.g. command posts, ra- hardened against blast) staging areas, command cen- mand centers ters = dars); troop concentrations; Enemy nuclear or other WMD Cmm ships2 facilities u)

Typical missions against Destruction of targets, person- Destruction of targets Unprotected personnel casual- Unprotected personnel casual- military targets nel casualties Personnel casualties ties; disruption of operations by ties; disruption of operations by requiring protective measures requiring protective measures intimidation of personnel or decontamination or decontamination Disruption of operations by re- Demoralization or panic of per- Demoralization or panic of per- quiring dispersal of units. Dis- sonnel sonnel ruption of communications by electromagnetic pulse effects

Drawbacks as military Small lethal radius requires ei- Potential for great "collateral Relatively large quantities re- Protective measures may re- instrument ther many weapons or great damage” quired duce casualties accuracy for most military mis- Risk of retaliation and escalation Protective measures may greatly Most agents degrade quickly sions in kind reduce casualties With persistent spores, contamination of ground that user may Radioactive contamination of Leave buildings and equipment wish to cross or occupy ground that user may wish to reusable by enemy (but per- cross or occupy sistent agents may require Leave buildings and equipment decontamination) reusable by enemy (but persis- ent spores may require de- With persistent agents, chemi- contamination) cal contamination of ground that user may wish to cross or Effects depend on weather and occupy time of day; are delayed, unpredictable, or uncontrollable Table 2-4-Applications of Weapons of Mass Destruction Compared Conventional explosives Characteristics (for comparison with WMD) Nuclear Chemical Biological Destructive effects Blast, shrapnel, fire Blast fire, thermal radiation, Poisoning: skin, lungs, nerv- Infectious disease or biochemi- (See table 2-1) prompt ionizing radiation, ra- ous system, or Mood cal poisoning dioactive fallout1

Typical military targets Military bases and equipment; Similar to targets for conven- Infantry concentrations, towed Infantry concentrations, air bases, Command-and-control installa- tional munitions (esp. targets artillery, air bases, ships, ports, ships, ports, staging areas, com- tions (e.g. command posts, ra- hardened against blast) staging areas, command cen- mand centers ters dars); troop concentrations; Enemy nuclear or other WMD ships2 facilities

Typical missions against Destruction of targets, person- Destruction of targets Unprotected personnel casual- Unprotected personnel casual- military targets nel casualties Personnel casualties ties; disruption of operations by ties; disruption of operations by requiring protective measures requiring protective measures Intimidation of personnel or decontamination or decontamination Disruption of operations by re- Demoralization or panic of per- Demoralization or panic of per- quiring dispersal of units. Dis- sonnel sonnel ruption of communications by electromagnetic pulse effects

Drawbacks as military Small lethal radius requires ei- Potential for great “collateral Relatively large quantities re- Protective measures may re- instrument ther many weapons or great damage” quired duce casualties accuracy for most military mis- Risk of retaliation and escalation Protective measures may greatly Most agents degrade quickly; sions in kind reduce casualties with persistent spores, contamination of ground that user may Radioactive contamination of Leave buildings and equipment wish to cross or occupy ground that user may wish to reusable by enemy (but per- cross or occupy sisent agents may require Leave buildings and equipment decontamination) reusable by enemy (but persis- ent spores may require de- With persistent agents, chemi- contamination) cal contamination of ground that user may wish to cross or Effects depend on weather and occupy time of day; are delayed, unpredictable, or uncontrollable

-1WI 58 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

a‘ ‘strategic” nuclear war.17 Moreover, they were poorly protected throughout, but apparently only deployed in the context of superpower arsenals by 1986 did the Iraqis learn to use their chemical containing thousands more strategic nuclear weap- weapons in coordinated and effective ways, ons aimed at each other’s homeland. preventing Iranian troops from massing and New nuclear powers are likely to have any- counterattacking with conventional forces. In where from one to a few hundred nuclear weap- 1988, Iraq used chemical weapons in an offensive ons, most likely of explosive yields equivalent to mode, weakening Iranian forward positions and a few tens of kilotons of TNT. These new nuclear limiting rear operations. Iraq reportedly also used powers might intend to use their limited numbers a combination of mustard and nerve agents on of weapons against isolated military targets for Kurdish civilian villages and rebel encamp- tactical purposes, or they might seek to achieve ments. 19 maximum economic damage and psychological Pelletiere and Johnson point out that, as was the effects by directly attacking cities. The decision case in World War I, the ratio of deaths to injuries would probably depend on the military and from chemicals seems to have been low in the political context, including whether the adversary Iran-Iraq war, and that therefore chemical weap- or its allies also had nuclear arms, and, if so, how ons should not be thought of as ‘‘a poor man’s many. nuclear weapon. ’20 Anthony Cordesman con- During World War I, both sides used large cludes that although the contribution of chemical amounts of chemical weapons. Japan used chemi- weapons to Iraqi success in any one battle is hard cal and biological weapons against China in to estimate, and although they produced less than World War II. Since then, the world has had some 5 percent of the more than 1 million Iran-Iraq war further experience with use of chemical weapons: casualties, Egypt (reportedly) in Yemen in 1967 and 1968; Nevertheless, [they] had a critical effect on Iraq against Iran during the 1981-1988 war, Iran Iranian military and civilian morale by late 1987, against Iraq, and Iraq against some of its own and during the Iraqi counter-offensives and ‘‘war Kurdish population. During its war with Iran, Iraq of the cities” in 1988. Sheer killing power is not used aerial bombardment and artillery to deliver the key measure of success: it is rather the mustard and nerve agents against Iranian infantry strategic, tactical, and psycho-political impact of and ‘‘human wave’ attacks and against support the use of such weapons. Even when troops are 18 troops and staging areas. Iranian troops were equipped with defensive gear, they often feel they

17 rnd~, full application of the conventional f~epower deployed in Europe might have had consequences nearly as ternbk+which -Y be one reason why NATO allies were willing to rely so heavily on nuclear deterrence against a Soviet attack. 18 See Steven C. Pefletiere and Douglas V. Johnson II, femons Learned: The lnm-lra~ JVar (Carlisle Barracks, PA: Strategic Smdies Institute, U.S. Army War College, 1991), “Appendix B: chemical Weapons,” pp. 97-101; and Anthony H. Cordesq Weapons of Muss Destruction in the Midde East (Ixmdon: Brassey’s (UK), 1991), pp. 85-93. The following discussion is drawn primarily from these sources. 19 Si@ of mW@d, Deme, ~ blood agents were reported found in Kurdish areas in statements of Ro~rt Cook-Deg~ ad of Debo~ Lief-Dienstag and supporting documents (given in U.S. Congress, Senate Committee on Governmental Affairs, Permanent Subcommittee on Investigations, Global Spread of Chen”calandBiological Weapons, Hearings, IOlst Cong,, 1st sess., S. Hrg, 101-744 (WashingtorL DC: U.S. Government Printing OffiIce, 1990), pp. 242-266). There is geneml agreement that Iraq used the mustard and nerve agents. Some authors argue that Iraqis also delivered cyanide on the village of Halabjalq but Corde_ ibid., concludes that Iranians were the likelier source. Gordon Burck argues that the cyanide could have come from ill-manufactured Iraqi nerve agent (“The Geneva Protocol: Selective Enforcement” in Lessons of the Guf War: Me&”ation and Conflict Resolution, AAAS, Proceedingsfiom an Annual Meeting Symposium, Feb, 17, 1990, New Orleans, Louisian& p. 17). Kenneth Timrne- on the other hand, charges that Iraqis were using purposely developed hydrogen cyanide bombs; see The Death Lobby: How the West Armed Iraq (New Yom NY: Houghton Miffl@ 1991) p. 293. The UN Special Commissio% however, did not report finding hydrogen cyanide weapons in the Iraqi arsenat. m S= Pelletiere ~d Johnsoq op. Cit., footnote 18, p. 100, T@J ~port tit ~though 27.3 ~r~nt of ~ American cmlldti~ ill World W~ I were gas-generated, and 31,4 percent of wounds were gas-related, the death rate among gas victims was only 2 percent. Chapter 2–Assessing the Risks 59

Even if chemical weapons did not inflict large-scale casualties, they could seriously interfere with military. operations. These pictures were taken during a 1988 exercise at Eglin Air Force base. On the left, an airman dons cumbersome protective gear. Top right, a simulated casualty is carried away from the airfield. On the bottom right, an aircraft is decontaminated by spraying and scrubbing with neutralizing chemicals.

are defenseless and break and run after limited or days. Thus, an attack on an infantry position losses. Populations which fear chemical attacks 21 might use a volatile agent like GB, while a may well cease to support a conflict. viscous, persistent agent like VX might be On the other hand, chemical weapons used applied to an airbase or a strip of territory. Table against troops in World War I did not appear to 2-5 indicates approximate quantities of those two damage civilian morale. Nor can it be shown that types of agent that might be needed for some chemical weapons clearly affected civilian mo- representative military missions. Although a few rale in the Iran-Iraq war. drops of nerve agent can kill, the fact that Chemical weapons may be used in tactical chemical agents are usually disseminated as a warfare either to kill or terrorize instantly, or to wind-borne aerosol or spray means that many impose operational difficulties on the enemy by tons may be needed to produce many battlefield contaminating key areas or equipment for hours casualties. The military utility of attacks on troops

21 Cordesw OD. cit.. foomote 18. D. ~. ,’ , ,. 60 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 2-5-Quantities of Chemical Weapons for Various Missions

Mission Quantity

Attack an Infantry position: Cover 1.3 km2 of territory with a “surprise dosage” attack of GB 216 240-mm rockets (e.g., (Sarin) to kill approximately 50% of unprotected troops delivered by 18, 12-tube Soviet BM-24 rocket launch- ers) each carrying 8 kg agent (totalling 1728 kg)

Prevent launch of enemy mobile missiles: Contaminate a 25 km2 missile unit operating area with 0.3 tons 8 F-16 bombers each deliv- of VX per sq. km ering 0.9 ton of VX (totalling 7.2 tons)

Immobilize an air base Contaminate a 2 km2 air base with 0.3 tons of VX twice a day for 1 F-16 bomber, 6 sorties three days

Defend a broad front against large-scale attack: Maintain a 300-m deep strip of VX contamination in front of a 65 metric tons of agent deposition defending a 60-km wide area for 3 days livered by approximately 13,000 155-mm artillery rounds Terrorize population: Kill approximately 125,000 unprotected civilians in a densely 8 F-16 bombers each deliv- populated (10,000/km2) city ering 0.9 ton of VX (totalling 7.2 tons) under optimum conditions

SOURCE: Adapted from Victor A, Utgoff, The Cha//enge of Chemlca/ W-pens (New York NY: St. Martin’s Press, 1991), pp. 23S-242. would depend greatly on how effectively they freed defensive positions in long wars of attri- were protected with gas masks, clothing, and tion.24 Another analyst argues that suitable tacti- shelters .22 cal targets for biological weapons might include For biological weapons, there is little docu- reserve combat units, formations massing in mented experience with military use.23 One preparation for an offensive, air force squadrons, analyst speculates that, for surprise attacks or for and rear area support units: repelling immediate attacks by others, biological Thus, it would appear that biological weapons weapons would be too slow and unpredictable to could be militarily useful in situations when be militarily attractive. He argues, however, that immediate results are not required and where the they might be useful on the front lines against danger to friendly forces is minimal. Thus, even

~ acm.icaI w~orl researchers very likely have in the past studied methods of penetratinggas maaks with chemi~ w~~ agm~. 23 Japmrepo~y ~ biologi~ w~~e ag~t China&fore and during World War II, with konchsive reSUltS. Although ~ J~~e Army performed field trials in which bombs canying plague-infested fleas were dropped on at least 11 Chinese cities, the weapons wcxe not reliable and had little military impact-although they claimed an estimated 700 civilian lives. Conlamination ofChinese o territory with plague also caused thousands of unintended casualties among Japanese troops. See “Japan’s Germ Warfare: The U.S. Cover-Up of a War Crime,” BuZletin of Concerned Asian Schofars, vol. 12, October-December 1980, pp. 2-17; John W, Powell “A HiddaI Chapter in History,” Bdletin of the Atmi”c Sciem”sts, vol. 37, No. 8, October 1981, pp. 44-52; and Peter Williams and David Wallace, UnJ”t 73J: Japan’s Secret BioZogicaf Warfare in Worfd Wur H (New York NY: Free Ress, 1989). u ~~ond A, ~~~, “Biologic~ wiu-fm~d tie Third World,” Politics andthe Li~e Sciences, vO1. 9, No. 1, A-t 1990, PP. 59-76. —

Chapter 2–Assessing the Risks 61

if biological warfare has only slight immediate value on the battlefield, it could have consider- . -- able utility when directed at rear units.25 Note that even if not many troops were killed, a sudden epidemic of incapacitating disease could at least temporarily paralyze both logistic and fighting units. It is feasible—if the right weather occurs and can be utilized-for a single aircraft to disseminate high dosages of biological agent over hundreds, or even thousands, of square kilometers by spraying a long line upwind from the target region. This was one nightmare scenario for coalition forces facing Iraq in northern Saudi Arabia during the Gulf War. For comparison of With adequate warning, troops can be protected from biological weapon attack by means of protective the relative lethalities of biological, nuclear, and suits. Soldiers in Saudi Arabia during Operation chemical weapons under somewhat different Desert Shield (prior to Desert Storm) sometimes scenarios, see figures 2-1 and 2-2. trained wearing chemical/biological protective gear. Defending against biological weapons may be difficult. Currently there is no reliable way of STRATEGIC USE quickly detecting their presence or identifying Nuclear—Between World Wars I and II, the them, so soldiers may not take shelter or don military theorist Giulio Douhet and others devel- protective clothing in time. Vaccination requires oped an idea of strategic bombing in which aerial advance knowledge of the infective agents the attacks on key military and economic targets in troops will encounter, the availability of effective the enemy’s homeland would severely diminish vaccines, and sufficient time for the soldiers to his ability to make war. During. World War II, develop immunity. High concentrations of agent strategic bombing evolved in practice into efforts may overcome the immunity even of vaccinated not only to inflict crippling damage on the personnel. enemy’s infrastructure, but to cripple his war According to the commander of the recently effort by demoralizing the population. Although created U.S. Army Chemical and Biological the strictly military and economic effects of the Defense Agency (CBDA), two atom bombs dropped on Japan did not directly affect Japan’s armed forces, the shock of . . . the biological threat has been recently singled the attacks (combined with the fear that more out as the one major threat that still poses the might follow) led to an unconditional surrender ability for catastrophic effects on a theater- deployed force. Desert Storm solidified the per- that might otherwise not have come so soon. ception in our country-in the Congress and During the Cold War, the nuclear standoff be- among our military leadership--that [biological tween the United States and the Soviet Union was warfare] was something that third-world nations sometimes called the ‘‘balance of terror." Al- considered a potential equalizer.26 though both superpowers integrated nuclear weap-

Z5 CaIUs, op. cit., foomote 9, p. 37. 26 Brig. Gen. George Friel, CO mrnanding General, U.S. Army CBDA, quoted in John G. Roos, ‘‘Chem-Bio Defense Agency Will lhckle ‘Last Major Threat to a Deployed Force,’ ‘‘ Armed Forces Journal International, December 1992, p. 10. 62 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

ons into their military forces, the primary role of civilians may kill many more unprotected people the weapons was not to win wars but to back (e.g., thousands) than would equivalent amounts threats. of high explosives. On the other hand, the many This terroristic component to strategic warfare uncertainties involved in dispersing chemical with weapons of mass destruction makes it agents efficiently-as well as the effectiveness of difficult to analyze just what would constitute relatively simple civil defense measures (e.g., ‘‘rational’ or ‘‘irrational’ use by proliferant wearing gas masks and remaining inside living states. When leaders threaten to use the weapons spaces that are sealed off during attack)--could (whether in an initial or a retaliatory attack), they keep casualties relatively low. Contamination of must decide what level of threat will be suffi- certain areas by persistent chemical agents might ciently intimidating. In some cases, conveying slow down industrial activities for days or weeks, the impression that one could assemble a nuclear but for the most part chemical weapons would bomb in 2 weeks might seem enough; in other leave the economic infrastructure of cities intact. cases, the threat to launch a nuclear missile attack Enclosed military facilities are even more likely in 1 hour might not. to be protected and to continue functioning. A leader actually ordering a strategic attack Biological—Like chemical weapons, biolog- must subjectively predict its psychological im- ical weapons would leave the material (as op- pact on the other side’s population and govern- posed to human) economic and military infra- ment, not just calculate the physical effects of the structure relatively untouched.28 Like nuclear weapons on the other side’s war machine. The weapons, they have the potential in modest attacker must also estimate what kind of retalia- amounts (e.g., a few kilograms of agent), properly tion to expect, and whether he would be willing delivered, to kill and disable many thousands of to accept it. Alternatively, the leader may be urban residents and to seriously impair war- ordering a retaliatory attack, either in pure re- supporting activities. On the other hand, biolog- venge or to warn against further escalation of the ical weapons (except for some toxins) act more conflict. In sum, this section’s discussion of the slowly than chemical or nuclear weapons, taking physical suitability of the weapons for strategic days or weeks to achieve full effect. Moreover, warfare is only part of the story. their effects are much harder to predict than those Threatening both population and property, of nuclear weapons: weather, time of day, local nuclear weapons are the most dangerous strategic terrain, and civil defense measures could all act to weapons. While civil defense measures can miti- reduce casualties (as with chemical weapons). gate their effects somewhat, within a certain Unlike nuclear and chemical weapons, the use radius (dependent on the explosive yield) they of biological weapons might not be attributed to promise certain destruction of all but deeply enemy attack, since outbreaks of disease can buried blast shelters. Despite the great uncertain- occur naturally. The problems of protecting ties in calculating the precise consequences of civilian populations against biological attack are nuclear war, the impact of even a ‘‘small’ or similar to those cited above for protecting troops: 27 “limited” nuclear attack would be enormous. immediate detection and protection are likely to Chemical—Medium- to large-scale attacks be difficult, and effective advance vaccination with chemical weapons (e.g., tens of tons) on may be infeasible.

27 See Om, The Efiects of Nuclear War, Op. Cit., fOOmOte 3, p. 4. 28 wl~ fie exception tit &OIO@c~ w~~~ can ~d ~ve ~n developed for apphcation to target fd IXOp$ tith tie aim Of Strate@C reductions of the enemy’s food supply. Moreover, spore-forming organisms such anthrax might require major decontamination efforts, and therefore interfere seriously with normal economic or military activities. Chapter 2–Assessing the Risks 63

NEAR-TERM PROLIFERATION THREATS: classification as a secret document, since the U.S. SUSPECTED PROLIFERANT NATIONS Government has released few of its estimates about the activities of specific countries. B Weapons Intelligence information might tend to confirm At the U.S. State Department’s last count or undermine some of the estimates in the public published in 1992, there were 188 countries in the literature, These details, however, are more impor- world. Five of the world’s nations (United States, tant for the implementation of U.S. unilateral Russia, United Kingdom, France, China) have policies (particularly those involving covert ac- acknowledged owning nuclear weapons. Three tion or certain bilateral international arrange- other states—Ukraine, Belarus, and Kazakhstan— ments) than for the formulation of the broader have on their territory former Soviet strategic policies to be addressed in this report and its 30 nuclear weapons, nominally under control of the sequel. In any case, the broader policies must be Commonwealth of Independent States,29 and it is formulated in the context of publicly available not yet fully certain that all will give them up. information. First, although Congress can author- Since the end of World War II, three states have ize classified activities that may be subject to admitted having chemical weapons (United some oversight, it cannot pass secret legislation. States, Russia, and Iraq). None say they have Second, Congress responds to public pressure, deployed biological weapons, although five (United which in turn derives from publicly available States, Russia, France, the United Kingdom, and information. Third, achieving international con- Canada) admit having had offensive weapon sensus and collective action on proliferation will munition supplies or development programs in require openness. Fourth, multilateral agreements the past. Additional countries are suspected either such as the Nuclear Non-Proliferation Treaty of possessing some of these weapons of mass (NIT), the Biological Weapons Convention (BWC),31 destruction or of trying to acquire them, but many and the Chemical Weapons Convention (CWC) more are not. In sum, the scope of the problem of were not negotiated or implemented in secrecy. proliferation is worrisome but still limited enough The United States will have to choose and carry to encourage hope that it can be contained. out its national policies toward specific countries This section presents data intended to convey on the basis of the best information available, a sense of the general character of the near-term classified and unclassified. Nevertheless, it proliferation problem. It names countries cited in should be understood that both classified and the public literature as having either the weapons, unclassified assessments of foreign weapon pro- or programs to acquire the weapons, of concern to grams will be subject to uncertainties, incom- this report. The arbitrary criteria for including pleteness, lack of integration of available data, or countries are explained in the footnotes to each inadequate interpretation-as the case of the Iraqi list. These lists should be treated with caution nuclear program well illustrates. and should in no way be considered authorita- Table 2-6 summarizes a published estimate of tive or as representing official U.S. Govern- what countries (beyond the five self-acknowl- ment assessments. To have included such assess- edged nuclear powers) are pursuing nuclear ments in this report would have resulted in its weapon programs. Note that some of the countries

29 me wee non-Russian s~tes have at ]east a political veto over launch of the weapons on their territory, but apparently hey do not cmntlY have the technical means to launch them independently. JO OTA repofl on nonproliferation pOllCitX, in prepmation.

31 convention on the PrOhlbltiOn of the Dey,elopment, p~od~~fi~n and stockpiling ofBacterio[og;Cal (Biological) and Toxin Weapons and on Their Destruction. 64 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 2-6--Countries Reportedly Trying to Acquire Nuclear Weapons

Region Country Comment Middle East/ Algeria Possibly interested in nuclear weapons, but currently lacks facilities; has agreed North Africa to IAEA inspection of formerly secret, Chinese-supplied nuclear reactor; not a party to the NPT.

Iran Reportedly pursuing nuclear weapons, but little public evidence of progress; CIA testimony estimated production unlikely before the end of the decade without foreign assistance.

Iraq Massive program uncovered after Gulf War; United Nations has required destruction of most infrastructure, but knowledgeable personnel still in country.

Israel Widely believed to have a clandestine nuclear arsenal of approximately 100 weapons.

South Asia India Exploded a nuclear device in 1974; probably has sufficient materials for several weapons.

Pakistan Undoubtedly has nuclear weapon program, probably successful. U.S. President no longer certifies to Congress that Pakistan does not possess a nuclear device, suggesting high likelihood that it does.

East Asia North Korea Suspicious reactor and reprocessing laboratory; submitted to some IAEA inspections in 1992 and 1993, but refused others; in March 1993, denied IAEA access to suspected reprocessing waste sites and declared its intention to withdraw from NPT (since rescinded).

Latin America Argentina in agreement with Brazil, seems to have ceased weapons program. No disclosure of progress towards weapons, but suspected of having developed clandestine enrichment plant, a key step towards weapons.

Brazil In agreement with Argentina, has apparently ceased weapons program. in 1987, revealed it had developed the ability to enrich uranium. (Brazil has also had a nuclear power submarine program requiring highly enriched uranium fuel.)

Africa South Africa Widely suspected to be very near nuclear-weapon capability, South Africa declared in March 1993 that it had in fact constructed 6 nuclear weapons, but dlsmantledthem in 1990. The South African president promised that South Africa would cooperate fully with the IAEA to assure the world that it was complying with the NPT. Joined NPT in 1991, placed declared weapons grade uranium under IAEA inspection, and presumably dropped nuclear weapon ambitions,

SOURCE: Leonard S. Speotor and Jacqueline R. Smith, Nucfear AmbWons: 7he Spread of Nuc/ear Wapons 1989-19SW (Boulder, CO: Westview Press, 1990) and Nuclear Non-Proliferation Project, “Nuclear Proliferation Status Report July 1992,” (Washington, DC: Carnegie Endowment for International Peace, July 1, 1992). The latter report aiso: names Libya as “presumed to be seeidng N-weapons,” but does not cite evidence of indigenous nuclear weapon facilities; and names Synia as identified by a U.S. officiai as having a “nuckar program with suspicious intentions,” but no suspicious facilities have been publicly cited. on this list now appear to have halted, or even Independent States. Each of the three govern- reversed, their programs. In a class by them- ments has pledged to abide by the START I selves are three republics of the former Soviet agreement and to join the NPT as a non-nuclear- Union—Belarus, Kazakhstan, and Ukraine. weapon state (so far, Belarus has ratified both On the territory of each are former Soviet strategic treaties, Kazakhstan the START I Treaty). Should nuclear weapons. These weapons are nominally any of them fail to abide by that promise, it would under the joint control of the Commonwealth of become a de facto nuclear-weapon state, although Chapter 2–Assessing the Risks 65

Table 2-7-Countries Generally Reported Table 2-8-Countries Generally Reported as as Having Undeclared Offensive Having Undeclared Offensive Chemical Warfare Capabilities Biological Warfare Programs Region CW Capability Region BW Program Middle East Egypt Middle East Iran Iran Iraqa lraqa Israel Israel Libya Libya Syria Syria East Asia China East Asia China North Korea North Korea Taiwan Taiwan SOURCE: Mentioned in at least four of the following six (i.e., two- thirds): David Fair%all, “Eleven countries Defying Ban on Germ Southeast Asia Myanmar (Burma) Weapons,” The Guardian (London), Sept. 5, 1991, p. 1.; Elisa Harris, Vietnam ‘Towards a Comprehensive Strategy. . .,’’ op. cit., p. 129; Seth Carus, “ ‘The Poor Man’s Atomic Bomb’?’ . . .,” op. cit., p. 25; and Harvey J, SOURCE: Gordon Burck and Charles C. Flowerree, /nternationa/ MeGeorge, “Chemical Addiction,” Defense and Foreign Affairs, April Handbook on Chemiml Weapons Proliferation (New York, NY: Green- 1989, p. 17; Russian Federation Foreign Intelligence service, op. cit., wood Press, 1991), pp. 164-171, cite 19 published repofls, from 1985 and U.S. Arms Control and Disarmament Agency, “Adherence to and to 1989, that identify nations suspected by various sources as having Compliance~’th Arms Control Agreements and The President’s Report chemical weapon programs. In addition, a later publication, Elisa D. to Congress on Soviet Noncompliance with Arms Control” (Washing- Harris, “Towards a Comprehensive Strategy for Halting Chemical and ton, DC: ACDA, January 14, 1993). See app. 2-A for the ta~e derived Biological Weapons Proliferation,” Arms Contro/: Contemporary Sacu- from these sources. rity%ky, vol. 12, No, 2, September 1991, p. 129, cites statements of U.S. Government officials listing suspect countries; also added is a IJ.N. ingpctio~ of Iraq found some evidence of offensive biological Russian Federation Foreign Intelligence Service R+ort: A New weapon research, but no stocks of agent. Quiescence of Iraqi Challenge After the Cold War: Proliferation of Weapons of Mass programs probably depends on continued U.N. monitoring. Destruction, JPRS-TND-93-O07. OTA hes listed here the nations mentioned in two-thirds or more of t hese sources publlshed since 1989. See app. 2-A for the table compiled from these sources, Table 2-8 summarizes 6 published lists of a U.N, insp=tions of Iraq found a considerable chemical arsenal; that nations suspected of having undeclared biologi- which has been found is being destroyed. Quiescence of Iraqi programs probably depends on continued U.N. monitonng. cal weapon programs (which may include anything from research on offensive biological weapons to actual stockpiles of munitions). Like the it might face technical difficulties in operating list of chemical weapon suspects, this one arbi- and maintaining the weapons.32 As of this writ- trarily identifies those appearing in two-thirds of ing, Ukrainian delays in ratifying START and the the published reports. (The former Soviet Union NPT have caused the most international concern. had an undeclared offensive biological weapon Table 2-7 names countries appearing in at least program that violated its obligations under the two-thirds of 11 published lists of countries 1972 BWC. The Russian Republic has ostensibly suspected of covertly developing or producing ended this program, but, since doubts remain offensive chemical weapon capabilities. OTA has about whether the program has been totally made no effort to assess the scale of each eliminated, it could be argued that Russia should countrys program, the precise meaning of capa- be on the list.) bility,’ or the evidence on which the allegations Figure 2-3 combines the data in the previous are based. three tables to provide a combined perspective on .

66 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Figure 2-3-Suspected Weapon of hand, proliferation is still limited enough to Mass Destruction Programs encourage hope that it can be contained. On the other hand, it is occurring in places where political conflicts pose a major complication to Prolifera

Shaded area: also has of these missiles. The Missile Technology Con- Scud-type or longer trol Regime has reduced the potential number of range ballistic missile suppliers of missiles. However, additional coun-

This figure shows the considerable overlap among countries tries have learned to copy, modify, extend the of chemical, biological, nuclear, and missile proliferation range of, and produce their own versions of concern. The countries named In the figure are those in Tables previously imported missiles; a few have devel- 2-6,2-7, and 2-8; as the notes to those tables indicate, the lists oped their own long-range systems-often in are compiled from unclassified sources and should not be considered either authoritative or complete. conjunction with space-launch programs and SOURCE: Off&e of Technology Assessment, 1993. foreign technical assistance. Those emerging missile powers that might the states suspected of having or trying to develop have the intent to strike at the United States (e.g., or produce weapons of mass destruction. Three Iran, Iraq, North Korea, Libya) will not be able to features of the problem stand out. First, the field long-range missiles or ICBMs over the next estimate for the current number of potential 10 years, and those that could develop the nuclear proliferants is relatively small--and smaller capability (e.g., Israel, India, Taiwan) are not than it might have been a few years ago. Second, likely to have the intent. It is therefore unlikely the set of countries trying to acquire nuclear that any country (other than China and the former weapons overlaps considerably with the set sus- Soviet republics that already possess inter- pected of having chemical and biological weapon continental ballistic missiles or ICBMs) would programs. Third, the most immediate and serious pose a direct ballistic missile threat to the U.S. threats (beyond the potential threat posed by within the next 10 years. former Soviet republics) are concentrated in three The only developing country that in the next regions of international rivalry: the Koreas, India- decade is likely to be able to threaten U.S. Pakistan, and the Middle East. Thus, on the one territory with ballistic missiles is China, which

33 SM CYrA, op. cit., footnote 11. Chapter 2–Assessing the Risks 167

Table 2-9-Classification of Indigenous Production Capabilities of Ballistic Missiles

Country None lncipient a Intermediatea Advanceda

Middle East Libya x Egypt x > Israel — — — x Syria x Iraq x Iran — x > Saudi Arabia x Yemen x South Asia India — — — x Pakistan x >? East Asia Taiwan — — — )@? North Korea — — x >? South Korea — )(b > Southern Africa South Africa — )(? >? Latin America Argentina — — )(? —> Brazil — — * >

NOTES: a ‘t[n~pient” means some capability to modify existing Scuds, but little else. “Intermediate” means the capability to reverse-engineer Scud-like missiles, to introduce changes, and to make solid-propellant short-range missiles. “Advanced” means capable of making missiles comparable to those producact by the United States in the mid-1960s (including intercontinental ballistic missiles and space launch vehicles). b South Korea COUM be characterized as “Advanced” although it has only demonstrated capabilities for reverse- engineering. Largely because of diplomatic efforts by the United States since the 1970s, Taiwan and South Korea do not appear to be aggressively pursuing either ballistic or space-launch missile programs at the present time, although they would have the technological capability to do so if they chose. Brazil’s space-launch rocket program is in abeyance for financial reasons, but its technological capability gives it missile-making potential. —> Indicates estimated potential for progress over next 10 years. ? Indicates greater uncertainty. SOURCE: Adapted by OTA from Stanford Center for International Security and Arms Control,Assessing Ba//istjc Missi/e Proliferation and ks Control, November 1991, pp. 153, 15. See OTA background paper on technologies underlying weapons of mass destruction, in press.

has long had that capability. Israel and India, both As shown in table 2-9, countries of prolifera- suspected nuclear powers, have space launch tion concern vary widely in their ability to vehicles in principle adaptable as missiles that produce missiles, extend their capabilities, or could deliver weapons to intercontinental ranges. design new types. Whereas several developing Both have also tested ballistic missiles that could nations have essentially no indigenous capability, reach the territory of other nuclear powers (e.g., others match that of the United States in the Russia, in the case of Israel, and China, in the case mid-to-late 1960s. Practically all, however, de- of India), posing an implicit nuclear threat and pend on assistance or at least purchases of possibly provoking counter-threats in return. supplies from abroad; outside the most industri- . .

68 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Figure 2-4--Proliferants’ Delivery Systems: Selected Aircraft and Missiles

Note: Range, payload, and flight profile SU-24 trade off can vary greatly. + Iran Iraq A Ballistic missiles + attack aircraft . Syria

F-4 Mig-29 Egypt + Iran F-16 + Israel + Iran ,- S. Korea Israel Syria Egypt Egypt Pakistan Libya N. Korea Mirage F-1 ~ S. Korea + Iraq

prithvi Scud B Mirage-5 Mirage 200 ) A A..... Scud C “ ‘ +- .+. India Iran Egypt Syria 4 Iran Egypt N. Korea Libya India Egypt 1 [ Libya ~Pakistan‘“ ---.--1 200 600 800 1,000 1,200 1,400 N. Korea. -— Maximum range (km)

This figure shows nominal ranges and payloads of selected aircraft and missile systems of countries (beyond the 5 nuclear-weapon states) suspected of having or trying to acquire weapons of mass destruction, The graph is not Intended to be exhaustive, but only to indicate that each country already possesses aircraft or missile systems of one kind or another that could be adapted to deliver weapons of mass destruction. SOURCE: Office of Technology Assessment, 1993. ally advanced countries, only Israel, India, and using foreign-licensed technology, but many China might be argued to be independent in have been able to import them. Figure 2-4 shows missile design and production.34 For a more some of the types of combat aircraft owned by detailed breakdown of missile possessions and proliferate countries listed in figure 2-3. Scud programs, see the background paper to this report. missiles, the Iraqi-modified Scud, and the Indian Few nations can produce advanced fighter Prithvi missile are included for comparison. aircraft indigenously; some produce them locally — ~ J-e E. Nolaq Trappings Of power: Ballistic Missiles in the Third World (WSshingtom DC: Brookings ~timtio~ 1991), P. 18. me rewewer of the literature on ballistic missile proliferation has concluded: . . . it is possible to show that the rmssile programs of almost all countries have been exaggerated in the literature. A careful scmtiny of the data shows that as of early 1992, only six or seven countries of the [author’s] list of twenty-two had a meanin@d ballistic missile program or capability. That includes India and Israel, the countries with the most developed programs; Brazil, Argen@ and possibly South Africaj which have meaningfd indigenous launcher development programs but not yet operational ballistic missiles; and North Korea and Iraq which possess (in the case of Iraq did possess) the indigenous capability to produce Scuds. All of the other countries have either purchased missiles (mostly Frogs or Scuds or have indigenous efforts which does [sic] not look promising.) Matthias Dembinski, “Ballistic Missile Proliferation and the New World Order: A Critical Survey of the Literature, ” CSIA Discussion paper 92-07, Kennedy School of Oovernmeng Harvard University, July 1992, p. 6. Chapter 2–Assessing the Risks 169

More than 40 developing countries possess The following sections explore the conse- antiship cruise missiles, with ranges typically quences of proliferation from two perspectives. under 150 km.35 So far, there have been no First, considering the broader concerns of the publicly identified programs among proliferant global political system and human welfare, what nations to develop cruise missiles for delivering might be the consequences for the world? Second, weapons of mass destruction. Rather than buy or from the narrower Perspective of U.S. deterrence indigenously develop long-range cruise missiles, and potential uses of force, what might be the proliferant states seeking them will most likely consequences for U.S. foreign and military poli- attempt either to attach warheads to nonmilitary cies? systems (such as small aircraft) or to retrofit missiles originally equipped with conventional warheads (see delivery systems chapter of the 1 The International Community background paper on technologies underlying weapons of mass destruction). NATURE OF WARFARE Since both nuclear and biological weapons Destruction of human beings on a large scale is carry so much destructive potential in such small not new to warfare, or even to this century. packages, they are both suitable for small scale Nevertheless, weapons of mass destruction com- attacks by unconventional methods-e. g., smug- press the amount of time and effort needed to kill. gling and secret emplacement, or delivery by Wars lasting a few hours could now devastate small boat or light aircraft. Politically and techni- populations, cities, or entire countries in ways cally plausible scenarios for the current set of that previously took months or years. Nuclear or suspected proliferants to threaten U.S. territory with nuclear or other weapons of mass destruc- biological wars among proliferant nations may tion, however, are difficult to devise. None have not match the scope of a U.S.-Soviet exchange of missiles or combat aircraft with sufficient range thousands of thermonuclear weapons, but the to reach the United States. But since strategic damage to their people could still be catastrophic. warfare with these weapons would be so much a Even in a conventional war, high-explosive or matter of psychology, it is also difficult to rule it incendiary bombing attacks on nuclear, chemical, out. A state that badly wanted to wreak or biological facilities could release harmful destruction on a U.S. city could probably do so, substances into the environment. In this way, a whether it had advanced delivery systems or countrys own weapons of mass destruction could not (and whether the United States had effec- be turned against it. In a war in which only one tive antiaircraft or antimissile defenses or not). side had and used weapons of mass destruction, the other might retaliate by attacking nuclear IMPLICATIONS OF PROLIFERATION reactors, possibly causing mass casualties (from If more nations do acquire nuclear, chemical, or radioactive fallout) and economic disruption com- biological weapons, what will be the effect on parable to those it had suffered. international security? Since there are important differences in the effects and the military utilities of each of the three types of weapons, the CHANCES OF WAR consequences of their proliferation will not be the Some scholars have argued that, at least under same. They may, however, be interrelated. the right circumstances, further nuclear prolifera-

35 w. se~ cm~, cm~~e ~i~~ile proliferation in rhe 1990$ (w~~gto~ X: Center for Smategic and International Studies, 1992), p. 2; and Eric Arnett, Sea-bunched Cruise Missiles and U.S. Secun-ry (New York, NY: Praeger, 1991), p. 28. 70 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

tion could be a good thing.36 One of these authors INTERSTATE RELATIONS argues that well-managed proliferation could Massively destructive weapons can alter produce a stable order in Europe, but that international balances of power in both positive “Unfortunately, however, any proliferation is and negative ways. A relatively small nation may likely to be mismanaged. ’ He cites four principal gain useful leverage against larger or more dangers: numerous adversaries. France’s primary argu- ● existing nuclear powers might use force to ment for acquiring its nuclear force de frappe was prevent others from getting nuclear weapons that although a French nuclear blow would be (as Israel tried against Iraq); limited in comparison to the damage that the ● new nuclear powers might only be able to Soviet Union could inflict, it might still impose a afford nuclear forces vulnerable to destruc- higher price on aggression than the Soviets would tion by preemptive first strikes, leading to find worthwhile. Israel seems to believe that its instabilities; undeclared nuclear weapons give it an ultimate ● those controlling nuclear weapons might deterrent against invasion from its more numer- believe they could fight and win nuclear ous Arab neighbors. But, while some nations wars; and might use nuclear weapons to deter aggression, ● increasing the number of fingers on the aggressor nations might use them to deter resis- nuclear trigger would increase the probabil- tance. ity that some would use them accidentally or irrationally, or that terrorists would steal COLLECTIVE SECURITY OPERATIONS 37 them. The spread of weapons of mass destruction The same principles would probably apply, in may make it more difficult to organize groups of varying degrees, to chemical and biological nations (whether under U.N. aegis or within weapons. The predominant view amongst most regional security groupings) to respond to acts of scholars-and national governments-is that these aggression. For example, we do not know, if the dangers are not controllable and that proliferation Iraqi Scuds had been known to carry nuclear should be avoided, not accepted. warheads,

36 See Kenneth N, Waltz-, ‘The Spread of Nuclear Weapons: More my Be Better,’ Adelphi Paper 17Z (_I.mdon: International Institute for S@ategic Studies, 1981) and John J. Weltmaq “Nuclear Devolution and World Order,” World Politics, vol. 32, January 1980, pp. 169-193, arguing although that considerable nuclear proliferation is inevitable, regional balances of nuclear power could emerge and be no less stable than the superpower balance; the latter author was only slightly less optimistic in “Managing Nuclear Multipolarity,’ International Secun”ty, winter 1981/82, vol. 6, No. 3, pp. 182-194. See also John Mearsheimer, “Back to the Future: Instability in Europe After the Cold War, ” International Security, summer 1990 (vol. 15, No. 1), pp. 5-56. Mearsheimer advocates that . . . the United States should encourage the limited and carefully managed proliferation of nuclear weapons in Europe. The best hope for avoiding war in post-Cold-War Europe is nuclear deterrence; hence some nuclear proliferation is necessary to compensate for the withdrawal of the Soviet and American nuclear arsenals tiom Centrat Europe. Idealty, as I have argued, nuclear weapons would spread to Germany, but to no other state. (p, 54) He does not explain how proliferation can be “carhlly managed” and confined only to GermanY, or, indeed, only to Europe. Another author argues not so much that nuclear proliferation will be stabilizing, but that it is inevitable, He therefore concludes that the U.S. must learn to adjust to the situation. While it may be possible tofocus some policy efforts on temporarily delaying proliferation to some “unstable or brutal anti-American dictatorships,’ for the most part the U.S. should give up ‘a non-proli.kation system that is becoming less and less viable.” TM Galen Carpenter, “A New proliferation Policy,” The National Interest, summer 1992, pp. 63-72, ST pmap~u~ from Mearsheimer, Op cit., footnote 36, pp. 37-38. Chapter 2–Assessing the Risks 71

● whether Saudi Arabia would have agreed to Pacific affairs to weakening. Japan also might participate in a coalition to drive Iraq from question its own renunciation of the weapons. Kuwait, ● whether other Arab states within range of Erosion of norms Iraqi missiles or aircraft would have joined, Iraq’s use of chemical weapons has already ● whether the United States and European weakened the international taboo attached to nations would have been willing to send them. The frost large-scale use of biological their troops into the region, or weapons would be shocking, the next less so, and ● what role U.S. nuclear capabilities might so on. Moreover, a single successful application have played in building coalition consensus. of a biological weapon might inspire non-state Nor do we know what coalition reactions would terrorists to try the same thing. Although a small have been if Iraq had threatened to respond with nuclear war might mobilize the international nuclear or biological attacks on European cities. community into action to prevent a recurrence, it A more dangerous Iraqi threat might have caused might instead show that outside powers will try to coalition participants to think twice. On the other keep their distance. hand, they may have come to feel all the more strongly that it was better to stop Iraqi aggression Increase of supply sooner, rather than later when its ambitions and More states in the business of making nuclear, power had grown even larger. The United States, chemical, or biological weapons could also mean for its part, might have considered additional more potential suppliers of means of production long-range bomber or cruise missile attacks as or actual weapons to still other parties-perhaps alternatives to large ground-troop concentrations. states, perhaps terrorist groups. Even if proliferant states did not intentionally transfer these goods, they might become targets for illicit FALLING DOMINOES foreign purchasers and smugglers.

Arms races SHORT OF WAR... One likely result of proliferation is more Nuclear, chemical, and biological weapons proliferation. India justifies its nuclear weapon exact a toll, whether possessing states ever use program by pointing to China’s. Pakistan has them or not. Experiences in both the United States tried to keep up with India. Iran may have decided and the former Soviet Union show some of the it must match Iraq’s chemical weapons, as well as costs and risks. try to develop nuclear weapons. Some Arab nations have sought nuclear weapons to counter Increased chances of terrorist theft those of Israel; or, they may have pursued This report does not address the non-state biological weapons as the “poor man’s atomic terrorist uses of weapons of mass destruction.38 bomb. ” If proliferation proceeds, more nations But any state building these weapons must erect that until now have forgone the nuclear option and maintain a formidable security apparatus, may reconsider. For example, if North Korea got both to protect the secrets of the weapons and to nuclear weapons, South Korea would be strongly prevent their falling into unauthorized hands. tempted to follow suit, particularly if it perceived Ineffective or inexperienced governments, espe- U.S. security guarantees and involvement in cially those with relatively unstable regimes, may

38 See U.S. con~ess, offke of Technology Assessment, Technology Against Terrorism: Sfrucwing Securify, OTA-lsC-s 11 ~~~to~ DC: U.S. Government Printing Office, January 1992). . —- -

72 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

not be as successful as the owners of nuclear, competition for its adversaries, feeding regional chemical, or biological weapon facilities have arms races. Nations pay for these arms races at the been so far. Indeed, it is still too early to be certain cost of their peoples’ welfare. that Russia will successfully gain and keep stable, central control over all the weapons of the former Safety and environmental effects on proliferants’ and their neighbors’ populations Soviet Union. The United States and the former Soviet Union face monstrous clean-up operations: radioactive Increased risk from political fragmentation elements and hazardous chemicals contaminate Disintegration of national political authority, the soil, sediments, surface water, and ground- regional secession, or civil war could deliver water at most or all of the sites where nuclear weapons of mass destruction into the hands of weapons were manufactured.39 To complete the groups that, at best, would be poorly equipped to U.S. cleanup could cost hundreds of billions of manage the weapons safely, or at worst, would dollars. Little is known about the public health use them irresponsibly. Again, the republics of consequences if this mess is not cleaned up-as, the former Soviet Union, perhaps including in the former Soviet Union, it seems unlikely to Russia, seem vulnerable to this risk. be. Production-and destruction--of chemical weapons also poses environmental risks. Neither the United States nor Russia has developed Diversion of economic resources politically acceptable plans (let alone built the The start-up costs of a nuclear weapon program facilities) to destroy their chemical weapon stock- are great. Iraq probably spent about $10 billion piles according to the 10-year schedule specified before its efforts were interrupted. A narrower in the CWC.40 program than Iraq’s might cost less, but could still There is little reason to think that developing cost billions. Acquisition programs for chemical nations manufacturing weapons of mass destruc- and biological weapons cost much less. Despite tion will allocate much of their scarce resources the expense, some countries may see weapons of to environmental health and safety. One might mass destruction as substitutes for larger, even take as an indicator the recklessness with which more expensive, conventional forces (the United the Iraqi chemical weapon program handled toxic States decided in the 1960s that nuclear weapons chemicals (as reported by U.N. Special Commis- were a way of getting ‘‘more bang for the buck’ ‘). sion inspectors). At the same time, those in charge of conventional Infectious biological agents eventually die, and forces may feel that spending on weapons of mass toxins are biodegradable. But some spore- destruction diverts resources from more usable forming microorganisms, in particular anthrax military instruments. Nevertheless, in most cases bacteria, can persist in the environment for many the quest for weapons of mass destruction is years. Moreover, biological weapon programs usually embedded in an across-the-board arms themselves can pose a threat to public health, as competition. Each country’s possession of such a apparently happened when anthrax spores were weapon will inevitably increase the stakes of the accidentally released in 1979 from a biological

39 Se. U.S. conge~~, ~lce of ~c~olo~ Assessment, compl~ cleanup: The Environmental kgacy of Nuclear Weapons Production, OTA-O-484 (Washington DC: U.S. Government Printing OffIce, February 1991) for an assessment of the scope of the U.S. problem; the situation in the former Soviet Union is unquestionably fm worse. m See us, ConPess, Offlce of ~holon Ass~sment, Disposal of Chemical Weapons: Alternative Technol+vlw backgro~d PaWrS OTA-BP-O-95 (Washington, DC: OffIce of lk.chnology Assessment June 1992). —— —

Chapter 2–Assessing the Risks 73

weapon research facility in the Soviet city of the calculus of U.S. intervention would have been Sverdlovsk, triggering a deadly epidemic.41 even more different. The nature of U.S. decisions might have I U.S. Political-Military Policies depended in part on whether U.S. leaders believed U.S. military forces are likely to continue to be that the Iraqi rulers would have themselves been called to deter or combat military actions abroad, deterred from escalating to the use of nuclear whether unilaterally or as a member of an weapons by U.S. nuclear retaliatory capabilities. international coalition. Future aggressor states (To the extent that U.S. military forces will be having weapons of mass destruction will change used in conjunction of those of other states, the the context for U.S. decisions about when and dynamics of building and sustaining coalitions in how to threaten or use force. the face of threats from weapons of mass destruction will also be important; this topic is discussed below.) DETERRENCE We now know of one historical case in which the proliferant country hoped to use its nuclear Deterrence oft he United States weapon not to deter U.S. military intervention, The United States acted to protect its national but to cause it. Although South Africa kept its interests against challenges from a Soviet Union nuclear weapon program secret, heavily armed with nuclear, chemical, and possibly biological weapons aimed at U.S. territory, The strategy was that if the situation in U.S. forces abroad, and U.S. allies. Nevertheless, southern Africa were to deteriorate seriously, a confidential indication of the [nuclear] deterrent the risk of direct conflict with the Soviet Union capability would be given to one or more of the clearly constrained U.S. definitions of its national major powers, for example the United States, in interests, its policies for defending those interests, an attempt to persuade them to intervene.42 and its strategies and tactics for managing clashes with Soviet international policies. Thus, South Africa hoped to engage in a kind of Would other, though vastly smaller, nuclear (or reverse nuclear blackmail. biological or chemical) powers be able to deter the United States from regional interventions to Deterrence by the United States protect its interests? Possibly, depending on To some extent, the U.S. and Soviet nuclear whether U.S. leaders perceived the stakes to be arsenals neutralized each other; the two nuclear worth the risks. In the case of Iraq, for example, superpowers never engaged in direct military the United States was concerned about, but not conflict with one another at least in part because deterred by, the known Iraqi chemical arsenal and of the risk of escalation to mutual annihilation. the possibility of a biological weapon threat. The Even in much more one-sided confrontations, the United States would have had a different problem availability of nuclear weapons to the greater if Iraq had had nuclear weapons. If Iraq could power did not deter, for example, the North have credibly threatened to use a few nuclear Vietnamese from engaging the United States or weapons against U.S. cities or those of U.S. allies, the Mujaheddin in Afghanistan from taking on the

41 U.S. ~iciom abut u event WCre _ Officially confirmed by the RUSSkUM in 1992. !k R. Jefiy smith “Ycltsti B1=CS “79 Anthrax On GeMI Warfare Efforts, ’ Washington Post, June 16, 1992, pp. Al, A2. For a Russian Komsomolskaya Pravah report with details on the inciden~ see Foreign Broadcast Information Service, JPRSReport: Prolz~eration, JPRS-TND-92-022, July 10, 1992, pp. 19-24. See also Milton Leitenberg, “Anttuax in Sverdlovsk: New Pieces to the Puzzle,” Arms ControZ Today, April 1992, pp. l&13. AZ presideylt F.W. deWer~ speech to joint session of South Afrkanparliamen4 h- bed from JoharmedmrgRadio South AfiicaNetworlq Mar. 24, 1993 (JPRS-TND-93-009, Mar. 29, 1993, p. 2). 74 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Soviet Union. Nor did the nuclear stand-off deter MILITARY OPERATIONS the superpowers from arming each other’s ene- In preparing for war in Central Europe, U.S. mies in those two conflicts. In these cases, the forces had to take account of the possibility that lesser powers had good reason to believe that the they would confront Warsaw Pact nuclear or nuclear superpowers were very unlikely to use chemical weapons. They could probably learn to their nuclear weapons-both because of the prepare to operate under such threats elsewhere in opprobrium that would come from such a dispro- the world. Even so, having to cope with weapons portionate use of violence and because of the risk of mass destruction would make U.S. foreign of escalation of conflict with the victim’s nuclear- interventions costlier and more difficult. Nuclear armed ally. or biologica143 weapons (to a greater extent than Emerging nuclear powers that avoid direct chemical) would increase the risk of casualties. attacks on the United States may justifiably doubt For Operation Desert Shield (preceding Desert whether the United States would unleash nuclear Storm), the U.S. had to move in large quantities weapons on them for conventionally armed acts of troops and supplies through a few ports and of aggression elsewhere. Thus, U.S. nuclear airfields. An effective nuclear, chemical, or bio- deterrence, already a small factor in such situa- logical threat against vital transportation nodes or tions, might not be much affected by nuclear staging areas would have caused great difficulty proliferation. for the Coalition. (An alternative strategy might U.S. conventional military threats may have have been to rely on still more intensive long- deterred less industrialized countries from attack- range cruise-missile and bombing attacks than ing U.S. interests abroad. Would further prolifera- were used in Desert Storm; this strategy, how- tion of weapons of mass destruction weaken such ever, would still leave the problem of occupying deterrence in the future? The issue in this case is territory on the ground.) not just whether U.S. leaders decide that U.S. During the Cold War, part of U.S. preparedness interests at stake justify deploying conventional in the European theater was based on the assump- forces in the face of the risks to them posed by tion that the United States would retaliate in kind weapons of mass destruction: it is also whether against Soviet nuclear (and possibly against the nation to be deterred would believe that its chemical) attacks; tither, the United States did own threats would counter-deter the United not foreclose the possibility that it would initiate States, leaving itself free to act without fear of the use of nuclear weapons if it were losing a U.S. intervention. For the next several years, such conventional battle. On the other hand, much of a counter-deterrent threat might take the form of the world would probably see U.S. first use of either limited unconventional attacks on U.S. nuclear weapons in the developing world as cities, or somewhat larger, but still limited, grossly disproportionate to any conceivable U.S. attacks on U.S. forces intervening abroad. interests there.44

43 ~ me a- of ~=tive m- of detection that would allow soldiers to don PI’OteCtivt? g= soon emllgh. 44 U.S. decisionmakers may have already internalid such concerns. Ixwis Dunn notes that . . . war games on this subject have fiquently revealed a reluctance of players from the Washington national security elite to use nuclear weapons against third-world countries, even in nxaliation for nuclear use. Contairu”ng Nuclear Proliferation, A&lphi Papers 263 @mdorx Brassey’s for the International Institute for Strategic Studies, 1991), p. 73, note 1-4S. —— .—— —

Chapter 2–Assessing the Risks ! 75

ALLIANCES OR COALITIONS monetary compensation and regional security A continuing theme of the Cold War was the guarantees) on its progress toward non-nuclear West European fear that the superpowers would status. Even if the various republics comply fully fight a ‘‘tactical’ nuclear war in Europe-with with their commitment-in the Lisbon Protocol consequences for them similar to those of a to the START agreement-to forswear nuclear ‘‘strategic’ nuclear exchange for the United weapons, actual removal would take several States. In the case of the U.S.-Soviet contest, years. Should they choose in the meantime to however, the United States shared at least some become nuclear powers themselves, they could risk of nuclear devastation with its allies. In seize these weapons and adapt them to that confrontations with proliferant nuclear powers purpose. Alternatively, they might dismantle the lacking the means to attack the United States, weapons for their fissile materials and then fail to U.S. allies abroad would bear heavier relative control those materials properly. risks and may be reluctant to participate. Emergence of Ukraine or Kazakhstan (Belarus On the other hand, some states facing a nuclear has ratified the NPT) as new nuclear powers adversary might welcome an alliance with a would seriously undermine the nonproliferation nuclear power—if they believed that the adver- regime in several ways. First, depending on world sary would be deterred by the possibility of U.S. reaction, other potential nuclear powers may nuclear retaliation. As noted above, though, such conclude that the political and diplomatic costs of a deterrent threat might not be fully credible. joining the nuclear club are tolerable. Second, the retention of former Soviet nuclear weapons out- A NEW DIMENSION TO PROLIFERATION: side Russia would likely torpedo the ongoing RISKS FROM THE BREAKUP OF THE nuclear arms reductions between the United SOVIET UNION States and Russia. Russian ratification of the The breakup of the Soviet Union—and the START I Treaty was contingent on the other shakiness of governmental authority in the republics ratifying the Treaty, agreeing to imple- splintered republics--could contribute to all mentation measures, and joining the NPT Since categories of the proliferation problem.45 The the NPT links renunciation of nuclear weapons on threat is potentially great, but just how great it the part of the nonnuclear powers with ‘‘effective will be is hard to predict at this writing. The major measures relating to cessation of the nuclear arms dangers include the following. race,‘‘ interrupting the U.S./Russian arms reductions process could have serious repercussions I Seizure of Soviet Weapons by Non- when a conference to renew the NPT convenes in Russian Authorities 1995. Finally, the de facto creation of new nuclear Ukraine and Kazakhstan have agreed in princi- states in Europe would affect regional security ple to ultimate elimination of the strategic nuclear issues and balances of power, possibly triggering weapons on their territories. Even so, the missiles other European states to reevaluate their nonnu- and warheads are still in place. In the case of clear status. Ukraine, as of this writing the government It appears that all Soviet tactical nuclear continues to place various conditions (such as weapons have been pulled into the Russian

45 For ~m &U~~ion of ~esoWces from the former Sotiet Ufion tit co~d ~d nucl~ prol~~tio~ ~ ZhCky Davis ~d JOXE@UI Medal~ Nuclear Proliferation From Russia: Options for Control, Report 92-310 INR (Washington, DC: Congressional Research Semice, Mar. 30, 1992). See also Kurt M. CarnpbeL Ashton B. Carter, Steven E. Miller, and Charles A. ZrakeL Soviet Nuclear Fission: Conrrol of the Nuclear Arsenal in a Disintegrating Soviet Union (Cambridge, MA: Harvard University Center for Science & International Affairs, Studies in International Security, No. 1, November 1991); and Graham Allison et al., Cooperative Denuclearizan”on: From Pledges to Deed (Cambridge, MA: Harvard University Center for Science & International Affairs, Studies in International Security, No. 2, January 1993). . . . .

76 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Federation. The question of whether the Russian 9 Export of Critical Information, Equipment Federation itself will fragment, or whether the or Materials custodial system for the thousands of former The most immediate risk may lie here. The Soviet nuclear weapons and hundreds of tons of major areas of concern are dual-use technologies, nuclear weapon materials (enriched uranium and critical dual-use materials, and fissile materials. plutonium) will break down, still seems unsettled. Russia and the other former Soviet republics face One can imagine either successor states attempt- severe shortages of hard currency. They are trying ing to become nuclear powers, or non-state to establish market systems of production and groups seizing and exploiting weapons or materi- trade, but the legal infrastructure to regulate those als. activities is not yet fully developed. It is possible that some exporting enterprises may be unaware 1 Export of Weapons or of Weapon of the proliferation risks of particular goods; Components others may intentionally take advantage of poorly enforced or corruptly administered export control Press reports indicate the smuggling of all laws. A Ukrainian firm reportedly has already kinds from the former Soviet Union is a growing exported tens of tons of hafnium and zirconium, problem. Despite some rumors, there is as yet no metals on the Nuclear Suppliers Group list of serious evidence that Soviet nuclear weapons restricted dual-use items.46 have been sold to other countries. There have In its need for foreign trade, a government itself been no reports of the export of chemical or may disagree with other nations’ judgments about biological weapons, but the possibility that it which exports constitute a proliferation risk. For might happen cannot yet be entirely excluded. example, Russia has declared its intent to pro- Much will depend on the continued integrity of ceed, over U.S. objections, with sale to India of the Russian nuclear weapon custodial system cryogenic rocket motor technology for space under conditions of economic hardship and politi- launch vehicles. The United States, declaring the cal confusion. sale to be in violation of the Missile Technology Control Regime constraints that the Russians had 1 Emigration of Technical Personnel voluntarily adopted, has suspended U.S. trade There is also no clear evidence yet that former with both the Russian and Indian organizations Soviet technical personnel with knowledge of involved. how to build weapons of mass destruction have Fissile materials might in one way or another emigrated to other countries. There have been be diverted from former Soviet weapon stockpiles reports of some attempts at recruitment. Although or from production facilities. The possibility of a such scientists or technicians might not be essen- breakdown in the Russian custodial system for tial to a third-world country’s weapon program, weapons is mentioned above. A similar break- they might be able to provide useful guidance down in the control of material production about what works and what doesn’t work, thus facilities, leading to theft and export of fissile speeding the development of weapons.

46 S* William c. Potter, “Nuclear Exports From the Former Soviet Uniom What’s New, What’s T?ue,’ Arms Control T*, January/February 1993, p. 3. Chapter 2–Assessing the Risks 77

48 materials, is also conceivable. Outside Russia, weapon programs. Unlike other new prolifer- some important former Soviet production facili- ants, such countries might inherit, rather than ties remain; of particular concern is a fast breeder have to import, some critical weapon technolo- reactor, capable of producing over 100 kg of gies. Given the current economic conditions weapon-grade plutonium per year, at Aktau, throughout the former Soviet Union, new nuclear Kazakhstan.47 weapon programs do not seem to be an immediate threat. Chemical or biological weapons would be I Indigenous Weapon Development easier to develop. Kazakhstan has inherited chem- A longer term possibility is that some former ical and biological weapon facilities from the Soviet republics might utilize their own expertise, former Soviet military complex; Uzbekistan has equipment, or materials to develop indigenous inherited test ranges for both types of weapons.

47 rbid., p. S. StX dso wilti C. Potter, Nuclear Profiles of the Soviet Successor States (Monterey, CA: CIS Nonpmli.femtion ~oJ~4 Monterey Institute of International Studies, April 1993) for more detailed listings of former Soviet nuclear-related facilities. 48 sw CenM htelfigen~ Agency, Directorate of Intelligence, The Dgfense Indusm”es of the Newly Independent States of Eurasia, ~ publication number OSE-93-10001, January 1993. Appendix 2-A Sources on Tables Listing Countries of Chemical and Biological Weapon Concern

ables 2-7 and 2-8 list countries reported in various published sources to have, or to be trying to acquire, chemical or biological warfare capabilities. As indicated in the chapter text, the lists provided there are in no way to be considered authoritative or comprehensive. OTA has merely recorded the countries listed in two-thirds or more of the cited publications. The following tablesT show not only the countries making this arbitrary cut-off line, but also the other countries mentioned in fewer than two-thirds of the sources. In the case of the chemical warfare program list, this appendix also details the sources cited in the International Handbook on Chemical Weapons Proliferation and utilized in the OTA table.

79 — —

80 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

7777777 >7-7

7-777 >77777> v

7777777777 7

777-777 77 *

777777 77777 7 7

777777 77>77 -?

77777-7 777777

77777 777-777 *.+. .—

Appendix 2-A-Sources on Tables Listing Countries of Chemical and Biological Weapon Concern 81

Sources Cited in Table 2-A-1, Proliferation Risks: Chemical Weapon Programs Sources and Commentary from Gordon Burck and Charles C. Flowerree, International Handbook on Chemical Weapons Proliferation (New York, NY: Greenwood Press, 1991), pp. 164-171:

DNI( 1989: Director of Naval intelligence, Rear Admiral Thomas Brooks, prepared testimony for the Seapower and Strategic and Critical Materials Subcommittee, House Armed Services Committee, Feb. 22, 1989, pp. 38-39.

Harris 1989/1 990: Elisa Harris, “Chemical Weapons Proliferation in the Developing World,” WSLBrassey’s Yearbook 1989 (kndon: Brassey’s Defence Publishers, 1989), p. 74.

NY Times 1989: Stephen Engelberg, “Chemical Arms: Third V@rld Trend,” /Vew York Thnes, Jan. 7,1989 (not a complete listing). (Source given as U.S. administration officials; ?—France is not mentioned in the article).

Time 1989: Jill Smolowe, ‘The Search for a Poison Antidote,” Tjme, Jan. 16, 1989, p. 22 (source given as SiPRi).

USN 1989: Joseph L. Gailoway et al., “Bad Chemical Reactions,” U.S. News & Wfx/cf Report, Jan. 16, 1989, p. 30 (sources for a table that also contains other information are given as ArmsContmi Associaiton, Federation of American Scientists, and Senate Armed Services Committee).

Knight 1989: Knight-Ridder news services, James McCartney, “U.S. Sees Threat of Chemicals,” Phi/ade/phia /nquirer, Jan. 8,1989, p. F1 (also named Japan, Netherlands and Switzerland- a historical possessor and two states that probably have never possessed CW weapons, making this list unusually unreliable).

Chi Trib 1989: Thoma Shanker, “West Underwrites Third VWrld’s Chemical Arms,” Chicago Tribune, Apr. 3, 1989, pp. 1,6; and “Lack of Candor Blocks Chemical Arms Treaty,” Apr. 4, 1989, pp. 1, 6 (source given as U.S. Government officials).

McGeorge 1989: Harvy J. McGeorge, “Chem ical Addict ion,” Defense& Fore/gn Affairs, April 1989, pp. 16-19, 32-33.

McCain 1989: Senator John S. McCain, “Proliferation in the 1990s: implications for U.S. Policy and Force Planning,” table 1, Congressional Record, Nov. 1, 1989, p. S14605; “Estimates are Based on a Variet y of Sources, including unclassif ied testimony by CiA Director William H. WWster, Seth Carus, David Goldberg, Elisa D. Harris and others and donot reflect the estimates of the U.S. Government.” Also published as “Proliferation in the 1990s: implications for U.S.Poiicy and Force Planning,” Strategjc Review, summer 1989, p. 11.

Additional sources: Harris 1991: Elisa D. Harris, ‘Towards a Comprehensive Strategy for Haiting Chemical and Biological Weapons Proliferation,” Arms Control: Contemporary Security Po/icy, vol. 12, No. 2, September 1991, p. 129, which cites statements of U.S. Government officials listing suspect countries.

FiS 1993: Russian Federation Forejgn Inkdh’gence Serv/ce Report: A New Cha~~enge After the Cold War: Pru/iferation of 14bapons of Mass Destruction, JPRS-TND-93-O07. — ——

82 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 2-B-l—Proliferation Risks: Biological Weapon Programs Suspected

Carus Harris Guardian McGeorge FIS ACDA Total % Countries (Policy Paper) (USG officials) (London) Def. & FA 1993 1993 Libya Korea, North Iraq Taiwan Syria Soviet Union Israel Iran China Egypt Vietnam Laos Cuba Bulgaria India

SOURCE: Compiled by Office of Ttinology Assessment, 1993, from various source. See text below.

Detailed listing of sources for table 2-A-1: David Fairhall, “Eleven Countries Defying Ban on Germ Weapons,” The Guardian (London), Sept. 5, 1991, p. 1.; Elisa Harris, “Towards a Comprehensive Strategy. . .,” op. cit., p. 129; Seth Carus, “ ‘The Poor Man’s Atomic Bomb’?’. . .,” op. cit., p. 25; and Harvey J. McGeorge, “Chemical Addiction,” Defense and Foreign Affairs, April 1989, p. 17; Russian Federation Foreign Intelligence service, op. cit., and U.S. Arms Control and Disarmamentt Agency, “Adherence to and Compliance with Arms Control Agreements and The President’s Report to Congress on Soviet Noncompliance with Arms Control” (Washington, DC: ACDA, January 14, 1993). -——

Policy Background 3

his chapter surveys the range of policy measures, present and possible, that can be applied to the problem of limiting the spread of weapons of mass destruction. The chapter will also show that if these measures are to haveT a chance of success, meeting two conditions will be increasingly important. First, policymakers must engage the greatest possible international cooperation for nonproliferation. Second, as a prerequisite to obtaining that cooperation, they must act to strengthen international norms, or rules of acceptable behavior, against the acquisition and use of weapons of mass destruction. To meet those two conditions, policymakers must give the goal of nonproliferation higher priority than they did during the Cold War. It is by no means certain that the levels of international cooperation needed to contain proliferation can be achieved. Indeed, some analysts have argued that the inherently anarchic nature of the international political arena will make nonproliferation efforts futile.1 Others agree that the levels of cooperation needed to stop proliferation entail a transformation of international politics, but they believe that with the end of the Cold War, such a transformation has become feasible.2

1 For example, one argued in 1980 that: . . ,unless the system of states undergoes a revolutionary transformatio~ any suggestion that further proliferation can be stopped borders on the absurd. . .In a world of independent states, some proliferation will be inevitable, much as will some war and the threat of it. John J. Weltxnaq “Nuclear Devolution and World Order,” World Politics, vol. 32, January 1980, p. 192-193. Ted Galen Carpenter, “A New Proliferation Policy,” The National Interest, summer 1991, pp. 63-72, argues that nonproliferation policies were not only futile, but even counterproductive. 2 See Randa.LI Forsberg et al., “After the Cold War: A Debate on Cooperative Security,” Boston Review, vol. 17, No. 6, November/December 1992, pp. 7-19. For further analysis of the linkages between nonproliferation policy and global security policy, see Ashton B. Carter, William J. Perry, and John D. Steinbruner, A New Concept of Cooperative Security (_W%shingtonj DC: Brookings Institution 1992). 83 84 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

What follows is a menu from which the . actions to stop or discourage experts from components of a coherent nonproliferation giving assistance. strategy are likely to be chosen. The menu does If a proliferant nation nevertheless manages to not attempt to organize the policy measures acquire or build facilities for a weapon program, discussed into such a strategy, nor to assess another kind of obstacle is still possible, although their feasibility y or promise. The second report of fraught with legal, political, and operational this study will specify and analyze selected difficulties: taking military or other actions to options in greater detail. disrupt or destroy the facilities. Table 3-1 lists the primary international agreements and U.S. national laws that underpin the 9 Secrecy current nonproliferation regimes. The sections Limiting the spread of nuclear-weapon knowl- below summarize the measures already in effect in these regimes and identify measures that could edge through secrecy has been a tool of U.S. intensify or broaden them. The measures are policy since the first weapons were created during discussed under four broad categories: World War II. Today, although the basic principles of nuclear materials production and nuclear ● imposing obstacles to those trying to acquire weapon design and manufacture are well known the weapons, throughout the world, important engineering de- ● imposing disincentives to deter proliferants, tails and technical shortcuts are still classified by ● offering rewards to increase the attractive- the current nuclear powers. (Despite this secrecy ness of voluntarily forgoing the weapons, and policy, nuclear-weapon states have at various ● offering global or regional security improve- times helped other states develop the weapons: ments to reduce perceived needs for the the United States cooperated in the development weapons. of the British nuclear weapon program; the Soviet An additional section addresses the special, Union helped China before the Sine-Soviet split urgent problems posed by the breakup of the in 1959; French nuclear assistance may have Soviet Union. advanced the Israeli weapon program; China reportedly helped Pakistan; Israel reportedly IMPOSING OBSTACLES TO helped South Africa, although in admitting its PROLIFERATION past nuclear weapon program, South Africa has Proliferant nations, particularly the less indus- denied this.) trialized ones, generally need materials, equip- The basic knowledge needed to produce chem- ment, and knowledge from abroad to acquire ical and biological weapon agents is much more weapons of mass destruction.3 Therefore, block- accessible than that for nuclear weapons. Secrecy ing their access to such supplies can hinder their may help protect important details of incorporat- progress. Methods of blocking access might ing the agents into more effective delivery include: systems, but will not be of much use in blocking proliferation of simpler weapons. ● use of secrecy to restrict the flow of knowledge; . export controls adopted by supplier nations; E National and Multilateral Export Controls . diplomatic, military, or other actions to stop The potency of export controls as an obstacle exports by third parties; or to proliferation depends on the degrees of:

s Assuming, of course, that they cannot simply buy complete weapona outright. No state yet is known to have bought a complete nuclear weapo~ but with the breakup of the Soviet Union, this possibility has become a more pressing concern. Chapter 3-Policy Background 85

Table 3-l—Primary Institutional Bases of Current Nonproliferation Regimes

Regime U.S. Legislation Supplier Groups Consensual Treaties

Nuclear Nuclear Non-Proliferation Zangger Committee (Nuclear Ex- lnternational Atomic Energy Agency Act, 1978 (NNPA) porters Committee), 1971 Statute, 1957 Foreign Assistance Act, 1961 London Club (adherents to Nu- Nuclear Non-Proliferation Treaty Export Administration Act clear Suppliers Guidelines), 1976 (NPT), 1970 (EAA), 1979 (1990 version ve- Coordinating Committee on Multi- Treaty of Tlatelolco, 1968 (Latin toed, provisions then sustained lateral Export Controls (CoCom), American nuclear-free zone) by Executive Order, Act later 1949 Treaty of Rarotonga, 1986a (South reinstated as interim measure) Pacific nuclear-free zone) Atomic Energy Act, 1954 Weapons of Mass Destruction Control Act of 1992 Iran-Iraq Non-Proliferation Act of 1992 Freedom Support Act of 1992 (aid to former Soviet republics)

Chemical EAA Australia Group, 1984 Geneva Protocol for the Prohibi- Chemical and Biological Weapons CoCom tion of the Use in War ofAsphyxiat- Control and Warfare Elimina- ing, Poisonous or Other Gases, tion Act, 1991 and of Bacteriological Methods Weapons of Mass Destruction Con- of Warfare, 1925 trol Act of 1992 Chemical Weapons Convention Iran-Iraq Non-Proliferation Act of (CWC), 1993 1992

Biological Australia Group Geneva Protocol, 1925 Chemical and Biological Weapons CoCom Convention on the Prohibition of Control and Warfare Elimina- Bacteriological (Biological) and tion Act, 1991 Toxin Weapons (BiologicalWeap- Weapons of Mass Destruction Con- ons Convention, or BWC), 1975 trol Act of 1992 Iran-Iraq Non-Proliferation Act of 1992

Missiles Arms Export Control Act, 1976 Missile Technology Control None EAA Regime (MTCR), 1987 Missile Technology Control Act, 1990 Weapons of Mass Destruction Con- trol Act of 1992 Iran-Iraq Non-Proliferation Act of 1992 a The IJn/t~ States is not party to this treaty. SOURCE: Congressional Research Service and OTA. For a more comprehensive listing of relevant U.S. legislation, see Zachary S. Davis and Warren H. Donnelly, Non-Proliferation: A Compilation of Basic Dowments on the International, U.S. Statutory, and U.S. Executive Branch components ofNon-Pro/iferat/on Po//cy(Washington, DC: Library of Congress, Congressional Research service, Dec. 18, 1990), CRS Report 91-85 RCO. See also Zachary S. Davis, Non-Pro//ferat/on Regimes: Po/&ies To Contro/ the Spread of Nuclear, Chem/ca/, and Bio/og/ca/ Wsapons and A.f/ssi/es(Washington, DC: Library of Congress, Congressional Research Service, Feb. 18, 1993), CRS Report 93-237-ENR. See also Leonard S. Spectorand Virginie Foran, Preventing kVeapons/+o/iferation; Shoukfthe Regimes be Combhwd?(Muscatine, IA: The Stanley Foundation, 1992). 86 Proliferation of Weapons of Mass Destruction: Assessing the Risks

● proliferants’ dependence on outside resources share to foreign competitors under less stringent (and their ability to work around blockages controls. In such cases, not only do the U.S. firms of those resources); lose business, but other suppliers obviate any ● controllability of supplies having valid civil nonproliferation benefits that the blockage or applications, but also usable for producing delay of sales might have had. weapons of mass destruction; Another cost of controls may be imposed on ● participation in controls by all suppliers; and international development policy: tighter control ● effectiveness of each nation’s monitoring on dual-use technologies may not only hinder and enforcement of controls. weapons proliferation, but it may also stunt the As more nations advance technologically, the peaceful technological advancement of the im- frost three of these factors are likely to decay. This porting countries. On the other hand, if controls is especially the case for chemical and biological are narrowly targeted to countries of serious technologies, which are already widely available. proliferation concern, countries that cooperate In the case of nuclear technologies, although the with nonproliferation regimes should not find number of potential suppliers has been growing, their peaceful development hindered. many states have also been strengthening and broadening their export control policies. There- NATIONAL EXPORT CONTROLS fore, changes in the net availability of technology Export controls have been a major tool of U.S. useful for nuclear weapon programs are hard to nonproliferation policy since the Atomic Energy assess. Act of 1946 (superseded by the Atomic Energy Despite the global spread of technology, export Act of 1954, itself amended several times since controls will remain an important nonprolifera- then). Table 3-2 summarizes U.S. laws and tion policy tool for many years, especially in the regulations directed at restricting exports from the nuclear area. In addition to impeding prolifera- United States (or re-export of U.S.-originated tion, export controls also supply information items) that could contribute to the proliferation of important for detecting and monitoring it. Never- weapons of mass destruction or of missiles. Other theless, both tightening export controls and ap- legislation, likely to be introduced in the 103rd plying sanctions against foreign violators can Congress, would further restrict proliferation- have economic as well as political costs. These related exports. A later table (3-4) describes costs may be deemed worth the return in interna- sanctions established under U.S. law against tional security, but they should be acknowledged. countries or companies that violate export laws First, controls can somewhat restrict international and regulations. trade. Although the number of export denials is a small fraction of all international transactions, many transactions must be screened in order to MULTILATERAL EXPORT CONTROLS detect those that ultimately are denied. Conse- The United States can help limit proliferation quently, a wide range of businesses must keep by controlling its own exports and by trying to informed about and comply with complex regula- block aid from other countries to proliferants. tions and licensing procedures. Individual com- Nevertheless, there are too many possible foreign panies may find themselves losing significant sources of materials, equipment, and knowledge legitimate sales and the other business opportuni- for unilateral U.S. policies to control the problem ties that might have followed those sales. More alone. Imposing restraints on proliferants re- seriously in terms of U.S. jobs and exports, U.S. quires multilateral cooperation to have a firms may also find themselves losing market chance of being effective. The United States has —. —— ..

Chapter 3–Policy Background 187

Table 3-2—U.S. Unilateral Proliferation-Related Export Control Legislation Legislation, Regulation, or Executive Order Description or Comment Atomic Energy Act of 1954 (as amended) Sets guidelines for dissemination and restriction of data relating to nuclear weapons. Provides statutory framework for export controls on nuclear trade.

Nuclear Non-Proliferation Act of 1978 Tightens export controls by requiring IAEA full-scope safeguards as a condition for exports of nuclear fuel and reactors. Seeks to establish U.S. as reliable supplier for nuclear reactors and fuels to nations adhering to nonproliferation policies. Seeks to strengthen international controls over transfer and use of nuclear materials and technology. Directs the President to seek agreement from all exporting nations to require recipients of nuclear technology and materials to accept lnternational Atomic Energy Agency (IAEA) “full-scope” safeguards on all peaceful nuclear activities. Further specifies legal guidelines for regulation of nuclear commerce and technical assistance. Directs the President to publish procedures for the Commerce Department to control U.S. exports of “dual-use” items that could be used for nuclear explosives. Defines jurisdiction of Departments of State, Energy, Defense, and Arms Control and Disarmament Agency over nuclear exports.

Export Administration Act of 1979 and Commerce Department, after insulting with State and Defense, issues Export Executive Order 12735 (Nov. 16, 1990) on Administration Regulations; its Bureau of Export Administration administers Chemical and Biological Weapons Prolif- export licenses on controlled commodities (including nuclear, chemical, or eration biological weapons-related or missile-related, as well as other items controlled for national security or foreign policy purposes). Authority extends primarily over dual-use goods. EAA of 1979, the primary authority for U.S. export controls, expired Sept. 30, 1990; President Bush vetoed successor act but extended export control authority by executive order under emergency power (conferred by the International Emergency Economic Powers Act of 1977). In 1992 Congress passed an interim renewal of the 1979 Act.

Chemical and Biological Weapons Con- Amended EAA to require Secretary of Commerce to establish and maintain “a trol and Warfare Elimination Act of 1991 list of goods and technology that would directly and substantially assist a foreign government or group in acquiring the capability to develop, produce, stockpile, or deliver chemical or biological weapons” if licensing them would be effective, and then keep a list of countries for which exporters must obtain validated export licenses.

Arms Export Control Act of 1976 Authorizes State Department (through its Center for Defense Trade) to control by licenses items (including chemical and biological warfare agents and missiles) covered by International Traffic in Arms Regulations and U.S. Munitions List. In contrast to Export Administration Regulations (above), authority of this act extends mainly over sales of conventional weapons and weapon components. 1 Many other IaWS ~r~s nonproliferation issues; this list only covers the m@r OneS. SOURCE: OTA and Congressional Research Service (see table 3-1.) 88 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 3-3-Current Multilateral Proliferation-Related Export Control Agreements

Agreement Provisions or Comment

Treaty on the Non-Proliferation Nuclear weapon state parties (now Including China, France, Russia, United Kingdom and of Nuclear Weapons (NPT) (en- United States) agree not to transfer nuclear devices to any recipient, nor to assist any tered Into force Mar. 5, 1970) non-nuclear-weapon State to make or acquire them, All state parties agree not to transfer nuclear materials or related equipment to any non-nuclear-weapon state unless the latter will accept International Atomic Energy Agency safeguards (monitoring) over the materials.

Nuclear Suppliers’ Guidelines: To strengthen and better implement NPT export restrictions, seven NPT members who Nuclear Exporters Committee were major nuclear suppliers (the Zangger Committee) agreed Informally in 1971 on a (Zangger Committee) and Lon- list of nuclear technology items, the transfer of which would trigger application of IAEA don Suppliers Group (London safeguards to ensure that the items were not used to develop nuclear explosives. Club) Forming the “London Club,” in 1976,8 more nuclear supplier nations (including France, not then an NPT member) joined those on the Zangger Committee and agreed on a set of Nuclear Suppliers’ Guidelines, under which “trigger list” exports would further require physical security for transferred items, acceptance of safeguards on facilities replicated from London Club member designs, and prohibitions against retransfer of Items to third parties; suppliers also agreed to “exercise restraint” in transfer of nuclear-sensitive facilities, technologies, and weapons-usable materials. Total of 27 nuclear suppliers agreed In April 1992 to an additional list of 65 categories of dual-use items to be controlled. Participating nations have adapted these controls voluntarily. There is no international mechanism for monitoring and enforcement, but a Japanese-administered secretariat in Vienna is now overseeing the application of the dual-use guidelines.

Australia Group Group of Industrialized nations agreed in 1984 to establish national controls on chemical weapon agents and precursor chemicals that could be used to make them. Group, then with 22 members, agreed in March 1992 to add to the control list organisms, toxins, and equipment that might be used to make biological weapons. Has no formal coordination, monitoring, or enforcement, but does have informal agreements to share intelligence and notice of export denials. Eleven other states apply some or all Australia group standards. attempted to enlist other supplier countries in United States laws and regulations provide for nonproliferation export controls. Table 3-3 sum- sanctions (e.g., criminal penalties or government marizes the results of these efforts. procurement embargoes) against U.S. and foreign companies that violate U.S. export regulations. 1 Blocking Exports From Third Countries Short of criminal indictments, the United States may also impose trade sanctions on foreign firms The purpose of sanctions against suppliers to that it believes are violating internationally agreed proliferant nations is primarily deterrence, not 4 export controls. Some of the laws also provide revenge. The hope of the policymakers is that for trade sanctions against foreign individuals or potential suppliers will not want to risk U.S. companies that export items of types restricted by sanctions just to get the business of the prolifer- the United States (whether they are U.S. goods or ants. not) when the parties know that their exports

4 For example, the United States determined in 1992 that the Russian company Glavkosmos was violating the terms of the Missile ‘Ikehnology Control Regime by agreeing to sell cryogenic rocket motors to IndiIx it then imposed U.S. export and import sanctions both on Glavkosmos and on the Indian Space Research Organhtion. Chapter 3-Policy Background 89

Table 3-3-( Continued)

Agreement Provisions or Comment Missile Technology Control Group of supplier nations agreed in 1987 not to transfer complete rocket systems or Regime (MTCR) subsystems, or production facilities for them. Group now consists of 23 states, plus 2 “partners”; other states, including Argentina, Israel, Russia, and China, have separately promised United States that they will abide by MTCR constraints. Members also agree to restrain exports of other components, material, or technology that would be useful in missile production. Applies to missiles of range over 300 km; also applies to any missiles which the member government judges to be intended for use with weapons of mass destruction. Agreement is subject to no formal coordination, monitoring, or enforcement.

Coordinating Committee on Group of U.S. allies in 1949 agreed not to export listed items (including some related to Multilateral Export Controls, missiles and weapons of mass destruction) to Communist countries. (CoCom) Controls have been relaxed after collapse of the Soviet bloc. CoCom is unique among supplier agreements in attempting to establish common standards of enforcement of national export controls among the members; however, it is ill-suited to control proliferation-sensitive technology because the very states that were its targets-Communist and former-Communist states—would have to be members of any nonproliferation export control regime. CoCom might serve as model for other agreements.1

U.S.-foreign bilateral As noted above, in some cases the United States obtains bilateral agreement with arrangements individual nations to abide by supplier group restraints. State Department also issues diplomatic demarches, urging individual foreign governments to impose controls on specific exports of concern discovered by the United States.

~ ~ IJnit~ States General A~unting Office, Export Corrtrok: rnu/ti/atera/ efforts to improve WIfOfCWXWf: /?e/10rf tO the sUbCO/?7~itkW ofl /nterrrafiona/ Economic Policy and Trade, Committee orI Foreign /7e/ations (Washington, DC: GAO/NSlAD-92-167, May 18, 1992). SOURCE: OTA and Congressional Research Service (see tabie 3-1 ). contribute to proliferation. Some other supplier ● denial of most-favored nation trade status, nations have legal sanctions comparable to those ● forfeiture of property and assets, of the United States. ● denial of assistance from international insti- In some cases, the U.S. laws provide for aid or tutions in which the United States partici- trade sanctions (e.g., cutoffs of economic aid, pates, military aid, or nuclear cooperation) against denial of arms transfers from the United countries, rather than just companies or persons, States, that supply the wherewithal for proliferation to ● denial of U.S. Export-Import Bank credits, other countries. Finally, the President may also ● termination of codevelopment and copro- take diplomatic actions to try to punish countries duction agreements, that defy U.S. nonproliferation policies. Table 3-4 ● blocking of international financial transac- surveys U.S. laws that authorize or require tions, sanctions against foreign suppliers. ● suspension of aircraft landing rights, and Other sanctions against suppliers-including ● prohibition of loading and unloading of cargo nations, not just “persons’ ’-contained in legis- from sanctioned countries in U.S. ports. lation proposed in the 102nd Congress and likely The President can already take several of these to be reintroduced in the 103rd ares: actions at his own discretion under his powers to

5 See U.S. Congress, Congressional Research Service, Foreign Affairs and National Defense Divisiou Weapons Nonproliferation Policy and Legislation: Z02nd Congress, 92-429 F (Washingto% DC: Congressional Research Service, July 3, 1991, Updated May 5, 1992). 90 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 3-4--Legislative Bases of U.S. Sanctions Against Suppliers Law Description or Comment Atomic Energy Act Requires cutoff of nuclear cooperation with states that transfer U.S.-supplied nuclear materials or technology without U.S. permission. Requires cutoff of nuclear cooperation with nuclear-weapon states that assist, encourage, or induce a non-nuclear-weapon state to engage in activities that involve nuclear materials and are significant for the making or acquisition of a nuclear explosive device.

Glenn (1977) and Symington (1977) amend- Require President (unless he issues waiver) to cut off economic and military aid ments (sections 669 and 670) to Foreign to countries that supply the wherewithal for enriching uranium or extracting Assistance Act of 1961 (FAA) plutonium from spent nuclear fuel when all the recipient’s nuclear facilities are not under IAEA safeguards.

Chemical and Biological Weapons Con- Requires President (unless he issues waiver) to deny U.S. Government trol and Warfare Elimination Act of 1991 procurement or any U.S. imports from ‘foreign persons’ (individuals or firms) knowingly and materially contributing to chemical or biological weapons proliferation through the export of goods or technologies either covered by the Act, or that would be covered by the Act if they were produced in the United States.

MissileTechnology Control Act, 1990 (Title Denies U.S. Government contracts or export licenses to U.S. or foreign persons XVII of the FY 1991 National Defense who improperly export missiles or major components; Authorization Act, which added a chapter Denies U.S. Government missile-related contracts or export licenses to those to the Arms Export Control Act and sec- who improperly export missile components, materials, or test and production tions to the Export Administration Act of equipment; 1979) Provides for Presidential waivers of sanctions.

Iran-Iraq Nonproliferation Act of 1992 (A Extends sanctions to Iran that already apply to Iraq: a variety of sanctions section of the FY 1993 Defense Authoriza- against individuals, companies, and countries who knowingly assist Iran or tion Act ) Iraq to acquire weapons of mass destruction.

SOURCE: Adapted by OTA from Congressional Research Service, 1992 (See table 3-l), conduct foreign policy. In November 1990, Presi- experts is another. Proliferants may hire the dent Bush vetoed the revised Export Administra- services of foreign experts to work directly on tion Act (since reinstated on an interim basis) that their weapon programs, to advise their own would have mandated sanctions in some circum- personnel, or to train their own experts. Alterna- stances. He nevertheless announced, in an Execu- tively they may send their own scientists and tive Order dealing with chemical and biological engineers abroad for education and training weapon proliferation, that the United States applicable to weapon programs. would implement trade sanctions against viola- Supplier nations have some possibilities for tors of U.S. law. In December 1990, the Bush control over such transfers of expertise. They can administration issued its Enhanced Proliferation enforce secrecy laws that make it illegal for those Control Initiative, which (among other things) with classified knowledge to transfer the infor- formalized the President’s commitment to impose mation. They can make it illegal to aid or abet sanctions without a statutory requirement to do proliferants (e.g., only since the Foreign Trade so. and Payments Act of 1992 have German courts been able to impose prison sentences on German E Hampering Transfer of Expertise engineers abroad whose activities promote the Exports of equipment and blueprints are one development or manufacture of nuclear, biologi- way to transfer weapon technologies; export of cal, or chemical weapons; the U.S. Atomic Chapter 3-Policy Background 91

Energy Act has long prohibited U.S. nationals Such a policy, however, would require putting from sharing nuclear technology with others.) nonproliferation above other concerns about rela- Another way to restrict the outflow of experts tions with those countries-it would amount to is not to punish, but to reward. For example, treating those states as international pariahs. Western nations are attempting to establish science and technology centers to employ some of the unemployed or underemployed former Soviet 9 Forcible Interference weapon scientists who might otherwise be tempted Perhaps the first clear example of a military to emigrate to proliferant countries to work on response to a proliferation threat was the Israeli weapons of mass destruction. bombing of the Iraqi Osirak nuclear reactor in It is much more difficult to limit education of foreigners in disciplines that could in principle be 1981. This step set back the Iraqi nuclear weapon program but did not end it. The bombing did applied to weapon development. For example, nuclear physics, chemical engineering, and bio- cause Iraq do a better job of concealing it. In 1991, chemistry may be useful for nuclear, chemical, or as part of Operation Desert Storm, the U. N.- biological weapon programs, but they also have backed coalition against Iraq attacked and de- fully legitimate civilian applications. Indeed, the stroyed facilities believed to be connected to Iraqi Non-Proliferation Treaty (NPT), the Chemical mass-destruction weapon programs. The U.N. Weapons Convention (CWC), and the Biological Security Council, as part of the cease-free it Weapons Convention (BWC) all stress the obli- imposed on Iraq, required elimination of all such gation of advanced countries to extend peaceful facilities (see below). technical training to less developed countries. Other types of forcible interference besides Brains are multiuse instruments. To bar foreign direct military attack might include: nationals from educational institutions on the grounds that they might someday work on weapons would exact a high cost: it would damage the openness in which scholarship thrives, and it would deprive developing countries of legitimate technical advancement.6 On the other hand, somewhat less draconian policies might be considered. First, governments could record and analyze the subjects of research and study of foreigners to see if suspicious patterns emerge for particular countries. Such a measure might yield additional information about proliferation activities, even if it did not itself serve as a means of control. Second, the citizens of specific countries could be singled out for denial of educational In the wake of Operation Desert Storm, the U.N. services if their countries were suspected of Security Council ordered the destruction of Iraqi developing weapons of mass destruction, or if facilities connected with weapons of mass destruction. their countries failed certain criteria, such as A nuclear facility at Al-Atheer was demolished on May joining and adhering to the NPT or the CWC. 31, 1992.

s Since a sigrifkant proportion of U.S. college engineering teachers are of foreign origiQ U.S. education might also pay a price for such a policy. 92 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

● sabotage of equipment or materials before the victim, ultimately lessening the obstacles to transfer, either on the territory of its supplier his weapon program. or in transit; Whether such an attack were internationally ● military interdiction of equipment or ma- sanctioned or not, it would also risk retaliation or terial; even full-scale war by the target country against ● sabotage of equipment or materials after either the attacking nations or their allies. import; or It seems unlikely that international authorities ● assassination of key personnel (explicitly will ever sanction covert activities like sabotage, forsworn by the United States). let alone assassination, as means of nonprolifera- While the latter measure is not appropriate for the tion. Almost by definition, covert actions are ones United States, it is not unheard-of in international that states are unable or unwilling to defend affairs. Canadian ballistics expert Gerald Bull, before the international community. Countries who helped Iraq design its ‘‘supergun," was may decide to take such measures for the same murdered in Brussels in 1990. Later, when U.N. compelling reasons they would use military force inspectors requested the names of people in the unilaterally. In doing so, however, they risk Iraqi nuclear program, Iraqi officials refused, exposure and loss of credibility as members of an saying they feared these people might be targeted international community that opposes prolifera- for assassination. tion on grounds of the common good. The U.N. Security Council declared in January 1992 that the proliferation of weapons of mass 1 Imposing Obstacles: The Special Case of destruction “. ., constitutes a threat to interna- Iraq tional peace and security. ” This phrasing, refer- Export controls and other nonproliferation ring to a key clause in the United Nations Charter, makes it conceivable that sometime the Council measures-at least as administered in the 1980s— might approve the use of military force to destroy failed to prevent Iraq from deploying and using facilities for producing or storing weapons of chemical weapons or from trying to develop mass destruction. Even so, such authority is likely nuclear and biological weapons as well as indi- to be highly circumscribed, lest states interpret it genously produced ballistic missiles. In the wake as license to attack others with impunity. More- of the war to liberate Kuwait, the United Nations over the necessary steps of deliberation, approval, Security Council undertook to reverse the prolif- and preparation would likely give considerable eration of these weapons to Iraq. As a condition advance notice to the targeted state. Such an of cease-free, the Security Council decided that internationally sanctioned strike would therefore Iraq should: be poorly suited for missions requiring surprise. ● give up all chemical and biological weapons, As a result, states believing their vital interests all stocks of agents, and all related subcompo- to be at stake may decide to take unilateral nents, as well as all related research, devel- military action against some cases of prolifera- opment, support and manufacturing facili- tion. However, if such actions are not sanctioned ties; by the international community-at least after the ● give up all ballistic missiles with a range fact—they risk damaging the international con- greater than 150 km as well as related major sensus on cooperative nonproliferation efforts. parts and repair and production facilities; Nations committing the action may find them- ● agree not to acquire or develop nuclear selves accused of violating international law. weapons or nuclear-weapon-usable material Moreover, an attack may even build sympathy for or any subsystems or components or any —

Chapter 3–Policy Background 93

research, development, support or manufac- -. turing facilities related to them; ● declare the locations, amounts, and types of all the banned items; ● submit to unrestricted U.N. inspections and supervision of the elimination of the banned items; and ● submit to future ongoing monitoring of verification of its compliance with the U.N. conditions.7 The sanctions for noncompliance with the cease-fire agreement are discussed below. The United Nations Special Commission established to oversee Iraqi compliance, along with the In a situation unlikely to be repeated in the case of International Atomic Energy Agency, appears to other potential proliferants, the U.N. Security Council have exposed and seen to the elimination of most required Iraq to submit unilaterally to inspections of of the Iraqi facilities and items covered by the facilities relating to weapons of mass destruction. Iraq resolution. Throughout, Iraq has tried to conceal has frequently attempted to obstruct such inspection. what it could and in other ways obstruct the U.N. Here IAEA inspector David Kay talks with Iraqi military authorities after they denied access to sites at inspections; it has also refused to acknowledge its Falluja in June 1991. obligation to submit to long-term monitoring of its continued compliance with the cease-fire authority, efficiency and universal scope of terms. For its part, Iraq has made it clear that it the convention. . .“; sees the United Nations as a tool of United States ● stressed ‘‘. . . the importance of. . . work on policy to hamstring Iraq, not as a legitimate a Convention on the Universal Prohibition of international authority. Few doubt that, given the Chemical Weapons and universal adherence opportunity, Iraq will attempt to rebuild its thereto”; and programs for weapons of mass destruction. More- ● declared its awareness of”. . . the threat that over, elimination of such programs based on all weapons of mass destruction pose to military conquest probably does not bear peace and security in the area and of the need much promise as a global nonproliferation to work towards the establishment in the measure. Middle East of a zone free of such weap- Nevertheless, United Nations Resolution 687 ons. did establish Security Council positions that conceivably could set precedents for future international cooperation to limit proliferation. In DISINCENTIVES TO PROLIFERANTS imposing the cease-fire conditions on Iraq, the Council: I Economic Sanctions ● noted ‘‘. . . the importance of all States Related to the sanctions against suppliers (see adhering to . . . [the Biological Weapons table 3-4) is a set of sanctions aimed at deterring Convention] and encouraged the forthcom- potential proliferants. Sanctions are one form of ing review conference’ . . . to reinforce the disincentive intended to make acquiring weapons

7 U.N. Security Councit, Resolution 687 (S/RES/687 (1991), Apr. 3, 1991), 94 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

of mass destruction seem less than worthwhile. matic exchanges, he can encourage other nations Should a country move toward acquiring the to restrain their exports. Likewise, he can threaten weapons, or violate provisions of agreements not potential proliferants with economic or other to acquire them, other countries may apply sanctions under his foreign policy powers. For sanctions in an attempt to enforce compliance example, U.S. diplomatic initiatives played a with nonproliferation norms.8 major role in the 1970s in persuading South Korea Current U.S. laws and regulations stress eco- and Taiwan to reverse what seemed to be nascent nomic, rather than other, sanctions toward poten- nuclear weapon programs. On the other hand, the tial proliferants. The Treaty on the Non- effectiveness of legislated export controls and Proliferation of Nuclear Weapons (Non- sanctions depends on conscientious executive proliferation Treaty, or NPT) implies a mild form branch enforcement; moreover, the laws usually of economic sanction by tying cooperation in allow the President to waive sanctions at his 9 civilian nuclear technology for non-nuclear na- discretion. tions to membership in the Treaty. In general, the In addition to his diplomatic responsibilities, multilateral agreements attempting to limit prolif- the President also manages the U.S. intelligence eration do not contain enforcement mechanisms, agencies. Intelligence plays a key role in identi- except for referral to the U.N. Security Council. fying which nations should be subject to In the case of Iraqi weapons of mass destruction, special export limitations, in discovering the however, the United States and the United Na- actual end uses of exported goods, and in tions have gone well beyond the provisions of the monitoring the exports of other nations to multilateral nonproliferation regimes in which potential proliferants. Along with Presidential abstinence is voluntary. management, congressional oversight can help Table 3-5 summarizes legislative bases for set U.S. intelligence priorities in these areas. (See U.S. sanction policies against proliferant nations. box 3-A for discussion of the implications of using intelligence in nonproliferation policy.) Although the above discussions center on the legislative bases for sanctions against suppliers and proliferants, the executive branch 9 Stronger Diplomatic and Military has wide latitude for discretion and leadership Responses (or default) on those matters. In addition, the Beyond the economic sanctions listed above President can act to mobilize international coop- (which could be applied to proliferant nations as eration on nonproliferation. For example, only well as to suppliers), disincentives might include one (the NPT) of the international, proliferation a variety of threatened responses that would make related agreements listed in table 3-3 is a formal owning and using weapons of mass destruction treaty subject to Senate advice and consent; the seem less attractive. The effectiveness of many others are essentially executive agreements. of these threatened actions will depend, like Through executive branch powers to conduct other nonproliferation measures, on the de- foreign aid and trade policies, the President can gree of international cooperation behind them. selectively apply what amounts to export controls The presence of a strong international norm to specific countries. Through bilateral diplo- against acquiring or using the weapons will be

8 The IAEA Statute, the CWC, and the BWC all explicitly invite members to bring treaty violations to the attention of the U.N. Security council. 9 See Carroll J. Doherty, “Foreign Policy Rules Riddled With Presidential Loopholes, ” Congressional Quarterly, Dec. 5, 1992, pp. 3753-3758. Presidents have frequently vetoed legislation that they believed infringed on their foreign policy prerogatives by limiting their discretion. — — — .

Chapter 3-Policy Background I 95

Table 3-5--Legislative Bases for U.S. Sanctions Against Proliferant Countries

Law Description or Comment Nuclear Non-Proliferation Act of 1978 amend- Termination of nuclear exports if nation: ment to the Atomic Energy Act . detonates a nuclear explosive device, . terminates or abrogates IAEA safeguards, . violates an IAEA safeguards agreement, . engages in activities involving nuclear materials and having direct significance for manufacture or acquisition of a nuclear explosive device. Prohibits sales of nuclear reactors and fuel to non-nuclear-weapon states that do not accept IAEA full-scope safeguards on all their nuclear installations.

Glenn-Symington amendments to For- Cutoff of military and economic assistance to nations: eign Assistance Act (FAA), 1976 and 1977 ● receiving wherewithal for enriching uranium or reprocessing plutonium, unless all such facilities and materials are placed under IAEA safeguards, . receiving a nuclear explosive device, or . detonating a nuclear explosive device.

Solarz Amendment to FAA, 1985 Requires President (unless he issues waiver) to cut off aid to any country that illegally exports, or attempts to export, from the United States nuclear wherewithal that would “contribute significantly” to the ability of a country to construct a nuclear device.

Pressler Amendment to the FAA, 1985 In the 1980s, Presidents Reagan and Bush waived (as allowed by congressional amendments to the Act) the requirements of the Foreign Assistance Act to cut off aid to Pakistan because of its nuclear weapons program; In 1985, Congress added an amendment requiring the President to cut off aid to Pakistan unless he declared in writing that “Pakistan does not possess a nuclear explosive device and that the proposed U.S. assistance program will reduce significantly the risk that . . . [it will]”; in 1990, the President stopped such certifications, and aid stopped (although commercial military sales continued).

Chemical and Biological Weapons Con- Requires President, on request of Chairman of House Foreign Affairs Commit- trol and Warfare Elimination Act of 1991 tee or Senate Foreign Relations Committee, to report whether a specified government has used chemical or biological weapons; If use determined, mandates sanctions including: foreign aid cutoff, arms sales and military financing cutoff, cutoff of U.S. Government credit or other financial aid, cutoff of exports of any controlled national security goods and technology; If, within 3 months, President does not certify that country has ceased using the weapons, provided assurance that it will refrain in the future, and allowed outside inspections, additional sanctions are at least three of the following: U.S. opposition to multilateral financial or technical aid, prohibition of U.S. bank loans, ban on all exports (except agricultural), ban on imports originating in the country, downgrading of diplomatic relations, suspension of aviation rights; Presidential waivers of the sanctions are possible. SOURCE: Office of Technology Assessment, 1993.

especially important to getting that cooperation in ● adversaries equipping themselves with com- the cases of those more severe measures. parable weapons or with effective defenses Preparations to carry out such measures maybe against them, ● countervailing military alliances, seen as efforts not only to deter further prolifera- ● diplomatic isolation, tion, but to manage the consequences of prolifera- ● trade embargoes, tion when it occurs (see section below, ‘‘When ● bilateral or multilateral promises to defend Nonproliferation Fails”). Coercive--or threat- or assist victims of aggression or use of ened—responses to proliferant states include: weapons of mass destruction, -—

96 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Box 3-A-intelligence Dilemma

Acquiring weapons of mass destruction is usually a clandestine activity. National intelligence agencies, particularly those of the United States, are Iikely to have the most complete information on who is trying to get what and who is selling what. However, publicly revealing this information increases the chances that the sources supplying it will be shut off. This principle has several implications for formulating nonproliferation policy:

● it increases the temptation to emphasize unilateral or bilateral steps to block specific U.S. exports or foreign transfers, as opposed to multilateral action, which requires broad sharing of information; . it challenges intelligence agencies and policy makers to find ways to share findings with multilateral organizations that monitor proliferation (e.g., the U.N. Special Commission on Iraq and the International Atomic Energy Agency);

● it places a premium on increasing the transparency of international transactions and national weapons programs by means of agreements among nations to report their actions to international bodies;

● it requires establishing the ability of international bodies to synthesize and act upon the data coming from transparency reports, unclassified sources, and individual national intelligence agencies;

● it forces intelligence agencies constantly to develop new sources of information when old sources are compromised by the overt use of their product; and

● it necessitates the development of unclassified sources of information that can be used in international fora.

SOURCE: Office of Technology Assessment.

> ● collective international assistance to victims of aggression or use of weapons of mass destruction, and ● military response to acquisition or use of the weapons. In 1991, U. S., Coalition, and U.N. actions toward Iraq illustrated a range of possible military responses to proliferation. (Although eliminating Iraqi weapons of mass destruction was not a direct cause of the Coalition military intervention, that goal did become an objective of the war and a condition of cease-fire.) As noted in the section above on imposing obstacles, one military re- Under some circumstances, the international community may impose economic sanctions on a sponse to proliferation is to attempt to destroy the proliferant nation. The United Nations imposed an means of production of the weapons before they embargo on Iraq after its invasion of Kuwait. Here, the can be fabricated and deployed. A second is to Navy’s U.S.S. Pratt (in distance) has stopped a Turkish attempt to destroy weapons already built before cargo ship bound for Iraq. The ship was found to they can be used. A third is to employ defensive contain only foodstuff and was allowed to proceed, measures to try to neutralize the weapons (either Chapter 3-Policy Background I 97

destruction weapons. It seems unlikely, however, that there will be more imposed reversals like that of Iraq, which was the byproduct of Iraq’s overwhelming defeat in a war fought for other reasons. Supplier-imposed obstacles and disincentives can significantly raise the costs of acquiring weapons of mass destruction, particularly nuclear weapons. In so doing, they can buy valuable time during which an aspiring proliferant may undergo changes of wealth, policy, or political regime that might arrest its weapon programs. Regional security environments can improve. States can One military response to proliferation is that taken by reassess the cost and worth of weapon programs. coalition forces during Operation Desert Storm: The international community can strengthen the destruction of potential weapon facilities. Here, IAEA consensual nonproliferation regimes--the NPT, inspectors examine the bomb damage to the IRT-5000 the CWC, and the BWC. Even so, in the longer research reactor at Al-Tuwaitha. term, states that remain determined to acquire these weapons will likely be able to do so. The passive measures, e.g., gas masks and protective technical knowledge and skills enabling their suits, or active defenses, such as antiballistic development will continue to spread through missiles). A fourth approach, embodied in Secu- international education, communication, and em- rity Council Resolution 687 and related resolu- igration. Industrial technologies and equipment tions, is to use or threaten military force to coerce useful for military research, development, and the proliferant into surrendering the weapons and production frequently have reasonable civil appli- their means of destruction. Finally, in a step not cations: preventing their spread even to countries yet (July 1993) taken in Iraq, one might force a of proliferation concern will not always be change of governmental regime to one that would feasible.10 voluntarily forswear the weapons of concern. For the longer run, imposed obstacles to E Benefits and Limits of Coercion proliferation may turn out to be surmountable Imposed nonproliferation measures--obstacles hurdles, not impenetrable walls, In the case of and disincentives—may be necessary, and per- India, one analyst argues that although a policy of haps effective, in the short run. In the near term, technology denial did create problems for Indian the proliferation problem seems limited to a nuclear work, handful of countries in Northeast Asia (North . . . the long-term effects of the policy have been Korea), South Asia (India and Pakistan), and the to promote the indigenous development of nu- Middle East. Continuing and strengthening exter- clear and fuel-cycle technologies in the Third nally imposed obstacles may slow the movement World. Technical constraints can buy time but of these countries toward visible arsenals of mass they cannot resolve the proliferation problem or

10 N@~e]ess, tie rates of spread of, and the potential effectiveness of export controlsOIL necessary technolozes for nucleti, che~~, and biological weapons, and for ballistic missiles vary. See Aaron Karp, Controlling Weapons Proliferation in the 1990s: The Role of Export Conrrols (3%enhausen, Germany: Stiftung Wissenschaft und PolitK Forschungsinstitut Fuer Internatiomde Politik und Sicherheit, September 1992). See OTA background paper on technologies umicrlying weapons of mass destruction for discussions of the relevant technologies. 98 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

contain the indigenous forces of technology in the fruits of science to those who genuinely want to Third World.ll use them for peaceful purposes.13 Moreover, coercive attempts to block prolifera- The best chance for nonproliferation in the tion maybe perceived as unfair challenges, not as long term lies in building a consensus among programs to promote international peace and potential proliferants that it is in their interests stability. A Pakistani diplomat has complained to refrain jointly from acquiring the weapons. that U.S. sanctions against Pakistan unfairly single out his country but bring no pressure to bear on India. He went on to threaten that REWARDS FOR ABSTENTION sanctions could have the opposite of their in- Imposing obstacles to proliferation and threat- tended effect: ening to punish potential proliferants are essentially coercive strategies. Another strategy is To add insult to injury, some elements in consensual: offer benefits in exchange for self- Congress are focusing on the perfectly legitimate -restraint. One such trade-support of peaceful commercial sale of military spare parts to Paki- applications of atomic energy in exchange for stan. Achieving this short-sighted objective would cripple the operational functioning of the Pakistan forgoing nuclear weapons-was promised in the armed forces and might impel the government of NPT. Benefits offered in return for consent not to Pakistan to pursue other military purchases and acquire weapons of mass destruction might in- resume development of its nuclear program.12 clude economic inducements, such as: (It should be noted that there are no indications ● financial assistance, that Pakistan has suspended its nuclear program, . technical assistance, and which has very likely already produced weapons, ● exemptions from nonproliferation export so this threat to “resume development’ is disin- controls on dual-use items. genuous.) Another set of benefits could be broadly catego- Insofar as domestic support for nuclear weapon rized as improvements in security that reduce the programs is based on sentiments of national pride perceived need for or appeal of the weapons. and autonomy, coercive measures may actually Security benefits might include: reinforce motivations to persevere. Both Indians and Pakistanis have argued that ● agreement by potential adversaries not yet export-control regimes are mainly an attempt to owning weapons of mass destruction that deny Third World countries access to nuclear and they also will forgo them, other technology needed for peaceful purposes. ● assurances by existing owners of weapons of For example, the President of Pakistan, speaking mass destruction that they will not threaten at the Pakistan Institute of Nuclear Science and to use them, Technology: ● reduction of the role of weapons of mass . . . we believe that the cherished and noble goal destruction in international relations, of nonproliferation must not degenerate into an ● monitoring or confidence-buildi.ng measures essay in evolving a technical fix or a ploy to to help verify that potential adversaries are eternise technological imperialism, denying the forgoing the weapons,

1 I Brahma Chelaney, ‘‘South Asia’s Passage to Nuclear Power, “ International Security, vol. 16, No. 1, summer 1991, p. 53. 12 M S=U N~~ &pUty chief of mission of Embassy of PakistaQ letter to the Washington Post, July 16, 1992, p. A-23. 13 Gh~~@~ tat of s~hof ~y25, 1~, ~m theP&~tun Ti~s of ~y26, 192, pp. 1-2, u mpfited fiJPRS-TND-92-017, June 3, 1992, p. 12. Chapter 3–Policy Background 99

● broader regional or global arms control ingly a political one. The choice will turn on arrangements that reduce conventional weapon whether a nation views the possession of such a threats, capability as being, on balance, in its national 14 ● foreign commitments to come to the defense interest. of or otherwise assist a nation if it is attacked The conclusion that, in the long run, motivations (with or without weapons of mass destruc- are key still holds true. It applies even more tion), strongly to chemical and biological weapons than ● regional security arrangements that more to nuclear weapons, because technologies for the broadly reduce the chances of war with local former are so much more accessible. Ultimately, adversaries, or nonproliferation policies will have to find ways ● global security arrangements that reduce the of showing leaders still desiring weapons of chances of attack from regional or extrar- mass destruction either that their goals can be egional adversaries. met in other ways or that the price of the weapons route is too high. Following a general discussion of the question Factors that make it difficult to persuade some of addressing motivations for proliferation, this nations to forgo weapons of mass destruction chapter section addresses each of these measures include: in turn. ● the perceived value of the weapons, ● double standards applied to those who al- fl Addressing Motivations ready have nuclear weapons and ballistic Persuading potential proliferants of the bene- missiles and those who do not, fits of going without weapons of mass destruction ● domino effects, and has been partially successful in the past. More ● the entrenchment of proliferation in conven- than 150 non-nuclear countries have ratified or tional military rivalries. acceded to the NPT, many of which are techni- The subsections below discuss these factors. cally capable of building nuclear weapons; most that are capable are refraining. The CWC has been PERCEIVED VALUE OF WEAPONS signed by numerous nations that could, but almost For different countries, the appeal of weapons certainly will not, acquire chemical weapons. On of mass destruction may lie in the national pride the other hand, a few countries have refused to or international status they seem to confer, their deny themselves the nuclear option, while one or deterrent value, or their military utility. Nuclear two others (Iraq and possibly North Korea) have weapons in particular have been associated with violated their agreement to abstain. Several have great power status (see table 3-6 for summary of declined initially to join the CWC, although hope nuclear proliferation motives). That the five remains that they can be brought in. Several are permanent members of the U.N. Security Council suspected of violating their BWC obligations. are all nuclear powers, and show no signs of in its 1977 report, Nuclear Proliferation and wanting to renounce that status, must enhance the Safeguards, OTA found that perceived value of the weapons. Nuclear weapons The technical and economic barriers to prolif- played multiple deterrent roles during the Cold eration are declining as accessibility to nuclear War. Their sheer destructive power makes them weapon material becomes more widespread. Con- attractive to military planners. Even so, the sequently, the decision whether or not to acquire apparent commitment of the two largest nuclear a nuclear weapon capability has become increas- powers, the United States and the Soviet Union

14 U.S. Cowess, Oftlce of lkchnology Assessment (Washington DC: U.S. GPO, 1977), P. 11. 100 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Table 3-6-Possible Motivations for Chemical and biological weapons programs Nuclear Proliferation are more likely to be influenced by military power Category Motive motives than by the other factors cited in table Military power Deter nuclear attack 3-6. These weapons do not seem to hold much Redress conventional arms attraction as symbols of international status or asymmetries with rivals Seek military superiority over rivals national pride. Indeed, their possession is usually Anticipate or match nuclear kept secret because of the stigma associated with weapons of rivals them. 15 Nor do leaders of developing nations Intimidate neighbors or rivals Deter intervention by extra- argue that they must pursue these weapons to regional powers enhance technical or industrial development.

International Enhance regional political status On the other hand, Iraq used chemical weapons political status Enhance global status in the Iran-Iraq war with impunity and with some Enhance image of technical military success, albeit against poorly defended prowess troops and undefended civilians. In addition, in Domestic politics National pride or morale 1990 Saddam Hussein attempted to invoke the Satisfy military groups deterrent value of chemical weapons by threaten- Economic improve- Scientific, technological or ing to use them in response to Israeli nuclear ment industrial spinoffs threats or other (undefined) acts of Israeli aggres- SOURCE: Adapted from Stephen Meyer, The Dynamics of Nuclear 16 Pro/iteration (Chicago, IL: University of Chicago Press, 1984), pp. sion. Many in the Arab world defended the Iraqi 46-74. threat against Western criticism. It remains to be seen what lessons potential chemical weapon (and now Russia), to dramatic reductions in their proliferants will draw from the ultimate inability nuclear forces is at least a step in the direction of of chemical weapons to save Iraq from cata- lowering the prominence of nuclear weapons in strophic military defeat. international relations. Some nations may seek chemical or biological As noted in chapter 2, South Africa added a weapons to deter the use of comparable weapons reverse twist to the motive of deterring outside by other nations. Some may see one of those types intervention: it hoped that its threat to use nuclear of weapon as a “poor man’s atomic bomb,” weapons in a southern African conflict would deterring nuclear neighbors, conventionally supe- 17 induce the United States to intervene in its favor rior neighbors, or intervening powers. They to forestall that use. may also simply seek these weapons as instru-

15 Rwmt n~cle~ prol~er~(s ~Ve alSO found it prudent to remain secretive about their weaponprO~u, Wtile Cle@ @@ pfide ~ tie nuclear technology underlying those programs. lb h a spech on Apr. 2, 1990, Hussein declared: Whoever threatens us with the atomic bomb, we will annihilate him witlh the binary chemical. . .we will make the fire eat up half of Israel if it tries to do anything against Iraq. Baghdad INA, translation in FBIS-NEW-9M64, Apr. 3, 1990, p. 36. When Western countries criticized this threa~ many Arab spokesmen came to Hussein’s defense, saying that Britain and the United States were trying to deny Iraq legitimate means of seLf-defense. For example, the Kuwaiti foreign ministry was quoted as saying: Kuwait while deploring this campaign and its exposed intentions, sides with brotherly Iraq in the right to defend its safety and security of its people using all available means. Kuwait KUNA quoting Kuwait News Agency, FBIS-NES-9M69, Apr. 10, 1990, p. 21.

17 see w. se~ c~$ “ ‘The Poor Man’s Atomic Bomb?’ Biological Weapons in the Middle East” (Washington, DC: The Washington Institute for Near East Policy, Policy Papers No. 23, 1991), p. 11. Chapter 3-Policy Background I 101

ments of military advantage for dominating on all their nuclear facilities (fill-scope safe- military rivals. guards), both have refused to join the NPT Among delivery systems, ballistic missiles are because it is discriminatory. On the other hand, perceived as another symbol of technological and the overwhelming majority of the world’s nations military prowess. The chances seem slim of have been willing to accede to the NPT as building an international consensus that the status non-nuclear-weapon states. of current ballistic missile powers should be The international community reached an un- ‘‘grandfathered’ like that of nuclear-weapon usual consensus on the unfitness of Iraq to own states but that further missile proliferation should nuclear weapons. But neither Iraq nor India nor be illegitimate, More likely, although not a most other nations accept what they see as the near-term prospect today, would be a global ban implication that all but the five acknowledged 18 on these delivery systems. If they are instituted nuclear powers are immature nations unqualified at all, voluntary agreements to forgo or reduce to handle the responsibilities of nuclear guardian- ballistic missiles will probably be in the context ship.20 Thus, the nuclear aspirants are not likely of regional security and arms control arrange- to be persuaded by arguments to the effect that ments rather than in a global nonproliferation only grown-ups should have nuclear weapons. regime. A second perception of double standards stems from the variability of past U.S. nonproliferation DOUBLE STANDARDS policies. From the U.S. point of view, failing to One problem in persuading aspiring ballistic make serious efforts to block Israeli acquisition of missile owners to forgo them is that they are being nuclear weapons or to enforce sanctions against asked to accept an international double standard: Pakistan in the 1980s in response to its nuclear the advanced powers now deploying ballistic program reflect the dilemmas of conflicting missiles have the right to do so, but newcomers to policy objectives. From the point of view of some the club are not welcome. 19 Nuclear nonprolifera- other countries, however, it reflects a willingness tion efforts suffer to some extent from the same to look the other way when the proliferant is a problem. India has complained the most vigor- friend of the United States—to select what ously that it is hypocritical of the United States proliferation is acceptable. and the other nuclear powers to deny the rights of The history of double standards, real or per- non-nuclear nations to acquire the weapons with- ceived, will in some cases be an obstacle to out giving up their own. Although Argentina and international consensus on nonproliferation. Fur- Brazil are moving toward participation in the thermore, enhancement of the nuclear nonpro- Treaty of Tlatelolco (making Latin America a liferation regime may see the emergence of a nuclear-free zone) and have accepted Interna- triple standard: a way must be found to deal tional Atomic Energy Agency (IAEA) safeguards with the three undeclared nuclear powers,

18 For a detail~ propos~ of such a b~ see FA.S. Public Intere$t Report, vol. 45, No. 3, May/June 1992, pp. 1-18. SW dso Wton Frye. “Zero Ballistic Missiles,” Foreign Policy, fall 1992, pp. 3-20. 19 in addition, curbs on missile technology are complicated by its relationship to space technology. India, fOr example, tis boti space-launch and ballistic missile programs. It has expressed strong resentment at U.S, attempts to block foreign exports to its space program that might also be useful to its missile program.

ZO ~ a Speech to P&ktani nuclw scientists and engineers, the President of pakistiuI noted tit The bombs that devastated Hiroshima and flattened Nagasaki were not hatched by the “unstable countries” and the “irresponsible minds’ of the Third World. . From text of speech by President Ghulam Ishaq Khan in The Pakisrtzn Times, May 26, 19!92, pp. 1-2, reprinted in JPRS-TND-92-017, June 3, 1992, p. 11. 102 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

India, Pakistan, and Israel. Will they be nuclear option themselves because they enjoyed treated as de facto nuclear states, or will they the nuclear protection of the United States. Not be asked to disarm themselves of weapons they enjoying such an explicit commitment, Israel do not admit having? developed its own nuclear deterrent against its vastly more numerous (and Soviet-armed) Arab DOMINO EFFECTS adversaries. North Korean nuclear nonprolifera- Proliferation occurs in the context of internation negotiations, stalled as of this writing, had tional conflict. China wanted nuclear weapons been taking place in the context of a broader because the United States and the Soviet Union political and military modus vivendi between had them. India pursued them because China had North and South Korea. them. Pakistan, aided by China, developed them Considering the special dangers that weapons to counter the Indian threat. Israel’s adversaries of mass destruction present, it might be desirable want to catch up with Israel. In the latter case, to treat their proliferation separately from other Israel’s adversaries also have pursued chemical, political and security issues. In the current regions and probably biological, weapons in part to try of proliferation concern, this compartmentaliza- counter the Israeli nuclear advantage-as illus- tion may not be possible. Agreements to forgo trated by Saddam Hussein’s April 1990 threats to weapons of mass destruction may depend on use his “binary” weapon. Iran sought chemical complementary agreements to reduce perceived weapons in response to Iraqi attacks during their conventional military threats. war, and it may be seeking nuclear weapons. 1 Economic Incentives To Forgo Weapons In short, some countries might not be talked out of pursuing one kind of weapon of mass destruc- of Mass Destruction tion unless they are convinced that their enemies will verifiably renounce not only that kind, but FINANCIAL AND TECHNICAL ASSISTANCE others as well. (In some cases, even that may not Linking technical or financial assistance to suffice: even if Israel’s adversaries were to nonproliferation began with President Eisen- renounce chemical and biological weapons, it hower’s 1953 Atoms for Peace Plan. That plan seems unlikely that Israel will give up its nuclear proposed the creation of the International Atomic weapons unless its general military security is Energy Agency, whose mission would be to make assured.) peaceful applications of atomic energy globally available while ensuring that nuclear materials CONVENTIONAL MILITARY RIVALRIES were not diverted to weapons. In the NPT, parties Weapons of mass destruction are frequently agree to foster peaceful applications of nuclear seen as potential compensation for inferior con- energy for peaceful purposes, “especially in the ventional military firepower or personnel. This territories of non-nuclear-weapon States Party to was the case for the United States and NATO the Treaty, with due consideration for the needs of during most of the Cold War.21 Some vulnerable the developing areas of the world”; they also countries-such as Germany, Japan, and South undertake to ensure availability of the benefits of Korea--clearly found it easier to forgo the peaceful nuclear explosions to non-nuclear-

ZI me idea that w~tem superiority in conventional military technology-rather than U.S. nuclear w~pom+odd counter W-W Wt numerical advantage was emphasized mainly in the later years of the Cold War, especially as discomfort with the idea of extended nuclear deterrence grew. Until the very end, the United States declined to follow the Soviet example (however disingenuous it might have been) of declaring that it would not be the fiist to use nuclear weapons. Chapter 3-Policy Background 103

weapon states .22 The U.S. Nuclear Non- proliferation Act of 1978 also links assistance to nonproliferation, while recognizing that disseminating peaceful applications of nuclear technology cannot avoid the potential of contributing to weapon applications as well. The Act provides that the United States “. . . shall seek to cooperate with and aid developing countries in meeting their energy needs through the development of [nonnuclear energy] resources and the application of nonnuclear technologies . . .’ and shall seek to encourage other industrialized nations to do the same. The Biological Weapons Convention (BWC) The International Atomic Energy Agency (IAEA) was calls for parties to facilitate exchange of equip- created to help make peaceful applications of nuclear ment, materials, and scientific and technological energy globally available while ensuring that nuclear materials were not diverted to nuclear weapons. IAEA information for the use of biological agents and headquarters are located in the Vienna International toxins for peaceful purposes; parties able to do so Centre, pictured here, also are to cooperate in contributing to the further development and application of scientific discov- eries in the field of biology for prevention of foreign aid policies were keyed closely to block- disease or for other peaceful purposes.23 ing Communist influence in the Third World.) Article XI of the CWC specifies that its One way to do this is to deny aid to countries that provisions will be carried out “. . in a manner do not participate fully in the nonproliferation which avoids hampering the economic or techno- regimes (e.g., refusing to join and adhere to the logical development of States Parties. ” It also NPT, the CWC, or the BWC). Another would be provides that the states “. . undertake to facili- to offer increased aid to induce states to end the tate, and have the right to participate in, the fullest regional arms races that stimulate desires for possible exchange of chemicals, equipment and weapons of mass destruction and to convert scientific and technical information relating to the military efforts to peaceful development pro- development and application of chemistry for grams. (In the case of the former Soviet Union, purposes not prohibited under this Convention. ” discussed in a separate section below, foreign aid A more comprehensive nonproliferation meas- may be directed at stabilizing polities where the ure would be to tie a large portion of international weapons already exist and at reducing incentives development assistance to nonproliferation goals. to export proliferation-sensitive goods and serv- (There is a precedent: during the Cold War, U.S. ices.)

22 pea=~ nuclew explosiom (PNE) were once a major bone of contention in nonproliferation debates, since there is no difference iII principle between a device that could create a peaceful nuclear explosion and one that would create a destructive one. There now appears to be little political support in the world for maintaining the PNE option. 23 me BWC Second Review Conference in 1986 recommended measures for increasing such cooperation. However, the aUthOr of a 1991 book on the BWC asserts that no concrete results have been obtained: The recommendations [of the Second Review Conference] have served only as a formal recognition of the preoccupations of developing countries at the review conference. Barend ter Haar, The Future of Biological Weapons (New York, NY: Praeger, 1991), p. 37 — —

104 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

needed to deter the weapons of others. The non-nuclear-weapon states party to the NPT assure one another that they will not acquire nuclear weapons. Parties to the Treaty for the Prohibition of Nuclear Weapons in Latin America (Latin American Nuclear-Free Zone Treaty, also known as the Treaty of Tlatelolco) agree not to acquire or to permit the presence on their territory of nuclear weapons. 24 States have agreed to forgo biological weapons under the BWC of 1975. All parties to the Chemical Weapons Convention agree to abjure chemical weapons. Bringing the undeclared nuclear weapon states—India, Pakistan, and Israel-into the The opening ceremony of the Signing of the Chemical nuclear nonproliferation regime will be a Weapons Treaty at UNESCO headquarters in Paris delicate task. In these cases, the first steps may was attended by (from left to right) the Foreign have to be measures to cap or freeze nuclear Minister of Germany, the Secretary-General of the weapon programs where they are-to keep the United Nations, the President of France, the Foreign Minister of France, and the Director-General of bombs in the basement, so to speak. Regional UNESCO. confidence-building measures might eventually persuade these nations to roll their nuclear EXEMPTIONS FROM EXPORT CONTROLS weapon programs back, while assuring their U.S. export controls on items that might neighbors that matching weapon programs of 25 contribute to nuclear, chemical, biological, or their own are unnecessary. One delicate ques- missile programs require licenses for export to tion is whether rollback will be possible without specific lists of countries of concern; countries prior formalization of nuclear status. In other not listed as proliferation risks are more likely to words, the bombs might have to come out of the be eligible to receive goods and technology they basements before they can be eliminated. If want. declarations of nuclear weapon possession and steps to eliminate the weapons are not closely 1 Security Benefits linked, the nuclear nonproliferation norm might be weakened. One way of handling this problem MUTUAL AGREEMENTS NOT TO ACQUIRE was recently demonstrated by South Africa, WEAPONS which dismantled its nuclear weapons first, then The central bargain of consensual nonprolifer- afterwards admitted their existence and promised ation agreements is that states give up their own steps to verify that it no longer had them. rights to acquire weapons of mass destruction on Without addressing this question, Pakistan has the condition that such weapons will not be for several years proposed a South Asian nuclear-

24 ‘f’he T~~~ of R~~t~nga creates a nuclem.free zone in tie Souti Pacific. ~ 1993, Waler I-J.N. auspices, a groUp Of eXpWtS k tO Chft a treaty for the ‘denuclearization’ of Africa. This effort seems to have a better chance of success than proposals, previously studied by the United Nations, for a Middle East nuclear-weapon-free zone. See Jon Brook Wolfsthal, “Nuclear-Weapon-Free Zones: Coming of Age?,” Arms Control To&y, vol. 23, No. 2, March 1993, pp. 3-9. 25 For discussion of confidence-building measures and roll-back, see Gregory F. Gk “Nuclear Proliferation Contingency Planning: Ensuring Global Ordedr In a More Proliferated World,” CNSN Paper, vol. 4, No. 2 (McLeaxL VA: The Center for National Security Negotiations (SAID)). Chapter 3-Policy Background 105

free zone; in November 1992 the U.N. General that it would not use nuclear weapons against any Assembly again overwhelmingly endorsed this non-nuclear-weapon state party to the NPT (or a proposal, but India voted against it.26 India has comparably binding international agreement) ex- repeatedly stated that it will consent to limits on cept in the case of an attack on the United States its own nuclear weapon program only in the or its forces or allies by such a state allied to a context of global nuclear disarmament. nuclear-weapon state or associated with one in the Onsite verification measures, such as those attack. provided for under the CWC, maybe necessary to In the NPT, the parties (including nuclear- build sufficient confidence in compliance among weapon states) agreed to ‘‘. . pursue negotiations the participants in mutual nonproliferation re- in good faith on effective measures relating to 27 gimes. Efforts to strengthen verification of cessation of the nuclear arms race at an early date compliance with the NPT or the BWC will have and to nuclear disarmament, and on a treaty on to take into account the difficulties of balancing general and complete disarmament . . .’ The costs and possible benefits from onsite inspec- United States and Russia have each declared that tions. For further discussion of the issues, see the they have removed most tactical nuclear weapons appendix to this chapter. from deployment and that they will destroy most of those; they have also agreed to deep reductions ASSURANCES FROM EXISTING OWNERS in their strategic nuclear forces; they have contin- In 1968 the U.N. Security Council passed a ued a moratorium on nuclear testing; the United resolution recognizing that nuclear aggression or States has ceased production of new nuclear- the threat of nuclear aggression would create a weapon material and Russia has indicated it will situation requiring immediate action by the Secu- do likewise. Other proposed measures for cutting rity Council, notably its permanent members. In back the weapon programs of the nuclear-weapon addition, the United States, the Soviet Union, and states have included a comprehensive nuclear test the United Kingdom declared that they each ban and formal, verified cessation of production intended to seek immediate Security Council of nuclear weapon fissile materials. action to assist any non-nuclear-weapon state If the spirit of cooperation between the two party to the NPT that was the object of nuclear nuclear superpowers continues, even more dra- aggression or threats. Such positive security matic steps are conceivable. The United States assurances could be strengthened in various and Russia might engage in yet another round of ways. For example, the permanent members of nuclear force reductions, this time bringing the Security Council could each promise to seek France, Britain, and China into the process. All Security Council action not only in cases of the nuclear powers might put even their weapons nuclear aggression, but also in cases involving into “trust’ for the United Nations, pledging chemical or biological weapons. Or, the Security never to use them except in a case approved by the Council could formally promise in advance to Security Council. Although the nuclear genie come to the aid of victims of such aggression. may never be fully rebottled, radically new In 1978 the United States issued a policy institutions for containing it are no longer un- statement providing negative security assurances: thinkable.

26 For fimssion of a r~e of South Asian nuclear arms control proposals that would address situations anywhere from denuclti=tion to ambiguity to declared nuclem weapon status, see Steven Philip Coheu “Policy Implications, ‘‘ in Cohe% cd., Nuclear Proliferation in South Asia: the Prospects for Arms Conrrol (Boulder, CO: Westview Press, 1991), pp. 339-371. 27 Such m~ms might also build over-cotildence: successful concealment of violations at (or away from) inspected sites co~d mislmd states to conclude that others are complying with an agreement when they are not. 106 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

There are other examples of assurances from resolved by placing all nuclear weapons in the owners. The United States in 1969 voluntarily hands of a supranational organization and estab- eliminated its own biological weapons; several lishing a universal prohibition against national other nations followed suit. The Convention on nuclear arsenals. However, such a world order the Prohibition of the Development, Production, still seems remote. Some argue the nuclear and Stockpiling of Bacteriological (Biological) proliferation problem cannot be solved unless and Toxin Weapons and their Destruction (Bio- nuclear disarmament is taken even further-to logical Weapons Convention, or BWC) became total elimination of all nuclear weapons.28 effective in 1975. Parties to the Geneva Protocol Threatening to respond to chemical weapon of 1925 agreed not to use ‘‘asphyxiating, poison- attacks with nuclear retaliation would foster the ous, or other gases” or ‘‘bacteriological meth- idea that nuclear weapons are legitimate instru- ods” in warfare. (The United States ratified the ments of war, and that those lacking them are less Protocol in 1975, at that time reserving the right than full players in the international arena. to retaliate with chemical weapons against states Attempting such deterrence would also have the not observing the Protocol; it has rescinded that effect of elevating the perceived significance of reservation, effective with the signing of the chemical weapons, implying that they are in some CWC in January 1993.) The United States and way equivalent to nuclear weapons as instruments Russia have agreed to destroy their stocks of of mass destruction. At the same time, a nation chemical weapons and, under the CWC, agree to contemplating the use of chemical weapons might forgo such weapons permanently. not believe that the United States would actually resort to so disproportionate a response as nuclear REDUCING THE INTERNATIONAL ROLE OF retaliation. WEAPONS OF MASS DESTRUCTION Biological weapons, effectively administered, Nuclear nonproliferation policy entails per- could turn out to kill as many people as nuclear suading non-nuclear countries that they do not weapons. Nevertheless, retaining a nuclear retali- need nuclear weapons. One way to reduce the atory option against their use could have effects appeal of nuclear weapons is to retire them to the akin to those of trying to use nuclear weapons to background of international relations, to dissoci- deter chemical weapon use: that is, trying to apply ate them from perceptions of power and status. nuclear deterrence to biological weapons could This is likely to be a difficult task for those who reinforce the idea that they are the ‘‘poor man’s already possess the weapons. In the U.S. case, atomic bomb,” and it might just as well stimulate de-emphasizing the international role of nuclear as discourage some countries from trying to weapons would logically mean weakening the acquire them. credibility and utility of U.S. nuclear deterrence. That result might in turn prompt calls in some BROADER ARMS CONTROL nations to reconsider their decisions to rely on As noted in ch. 2 and earlier in this discussion U.S. deterrence rather than acquire their own of nonproliferation incentives, nations now sus- nuclear weapons. On the other hand, in a world of pected of seeking one type of weapon of mass generally lower perceived nuclear threat, such destruction are engaged in arms competitions nations may now feel less dependent on U.S. with neighbors seeking not just the same type of nuclear deterrence for their security. In the long weapon, but sometimes other types and some- run, the nuclear deterrence paradox could be times conventional weapons; this is true in the

28 s=, for ~~ple, Jowph Rotblat, et al., (eds.), A Nuclear-Weapon-Free World: Dem”rable? Feasible? (Boulder, CO: %’estview mss, 1993). Chapter 3-Policy Background 107

Middle East, South Asia, and the Koreas. Limit- DEFENSE COMMITMENTS ing external threats of whatever character may U.S. alliance commitments to Germany and reduce incentives for acquiring weapons of mass other NATO countries technically capable of destruction. For example, North and South Korea building nuclear weapons, as well as to Japan and have discussed verification of non-nuclear status South Korea, probably contributed to their deci- in the context of wider arms control arrangements sions not to acquire nuclear weapons. In the between the two sides. future, the United States, alone or in concert with The choices are among trying to negotiate other nuclear powers, might continue to offer a regimes that limit various combinations of: conventional or a nuclear deterrent umbrella to help persuade some countries to forgo the nuclear ● a single type of weapon of mass destructions, option. Offering a credible conventional deter- ● all types, rent, however, may be complicated by the world- ● delivery systems, and wide reduction and return to the United States of ● conventional armaments and troop levels. U.S. military forces due to the Cold War’s end. Casting the arms control net more narrowly may Offering a nuclear umbrella implies maintaining simplify negotiations. Argentina and Brazil seem deployed nuclear forces that could credibly be to be an example of two nations arriving at used in retaliation for a nuclear attack on a third reciprocal decisions not to develop nuclear weap- party .29 It also would expose the United States to ons and to agree to some verification measures for the risk that the state it retaliated against would mutual reassurance. Elsewhere—for example the escalate to a nuclear attack against the United Middle East—the pursuit of weapons of mass States. Another problem with the maintenance of destruction may be too deeply embedded in the such forces is that doing so would underscore the regional security problem and consequent across- special status that nuclear weapons confer, and the-board arms competition: renunciation of nu- may contradict efforts to lower the profile of clear or chemical arms may not come without nuclear weapons in international politics. The reductions in conventional military threats. existence of either conventional or nuclear de- An important stimulus for limiting regional fense commitments by the United States also arms races could be collective agreements by risks persuading some countries of the need to the major suppliers of conventional weapons develop their own nuclear forces as a counter- to restrain their exports. The United Nations has deterrent to external intervention in regional established an international registry of arms affairs.30 transfers, in the hope that greater transparency will lead to greater restraint. The five permanent REGIONAL SECURITY AND ARRANGEMENTS members of the U.N. Security Council, who also The long-run success of nonproliferation ef- account for the great majority of global m forts is likely to depend in part on the reduction of sales, have held talks to discuss the possibility of security threats used to justify acquisition of limiting sales to the Middle East, but have weapons of mass destruction. Some analysts reached no agreement. (In the fall of 1992, China argue that regional conflicts are the ‘‘root cause’ withdrew from these talks to protest the U.S. sale of proliferation, and therefore that settling re- of F-16 aircraft to Taiwan.) gional security problems is a sine qua non for

29 Some analysts arWe that START-reduced U.S. strategic nuclear forces will more than suffke fOr this purpose, while otiers believe tit smaller yield tactical nucleax weapons would be more credible. 30 some ~ve ~W~ tit ~ Ufited States sho~d develop nucl~ w~po~ spec~y ~or~ for limited rnilitaly pUpOSeS h OthelWiSC3 conventional conflicts; however, arguments that the United States is entitled to special, advanced nuclear weapons, while others should have none at all, are not likely to have wide international appeal. 108 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

containing it. The security problems in each safeguards in 1992. It allowed some inspections region of proliferation concern are different; each in 1992, but in March 1993 it denied further IAEA will require specially tailored arrangements if the access and announced its withdrawal from the parties are to trust one another enough to halt or NPT. After discussions with the United States in reverse their military competitions. Such arrange- June 1993, North Korea agreed to postpone its ments may consist of combinations of political NPT withdrawal, but at that time had not yet agreements, economic steps, military confidence- agreed to the special inspections requested by the building measures, and arms controls. IAEA. Middle East—Achieving a nuclear-weapons- Former Soviet Union and Europe— free zone in the Middle East would probably Ukrainian officials have promised in principle to entail extensive peace arrangements between give up the former Soviet nuclear weapons on Israel and its Arab neighbors. The current Middle their territory. But recently they have tied imple- East peace process aims in this direction. Israelis menting that promise to, among other things, the may see their undeclared nuclear weapons as the kinds of security guarantees they have from (or ultimate guarantor of Israeli deterrence against against) their neighbors (chiefly Russia). See elimination of their vulnerably small nation by its chapter 2 for discussions of concerns surrounding more populous neighbors. At least some Arab the breakup of the Soviet Union and see below in states may see Israeli nuclear and other weapons this chapter for a range of policies for limiting of mass destruction as fully sufficient justifica- proliferation from that region. Several European tion for obtaining the same.31 As the Iran-Iraq war states are technically capable of producing nu- and the Iraqi invasion of Kuwait show, disputes clear weapons, but have renounced the right to do among several Middle East states go beyond the so. In the long run, their adherence to their Arab-Israeli conflict. decisions may depend on their trust in regional South Asia—Pakistan and India have both security arrangements in a post-Cold-War world. ethnic and territorial disagreements with one another. China and India, beyond their own GLOBAL SECURITY ARRANGEMENTS territorial disputes, rival one another as regional In many cases, regional groupings have been great powers. As noted above, Pakistan has unable to establish workable security arrange- proposed a South Asian nuclear-free zone, but ments on their own. The United Nations could India insists not only that China would have to step in either instead of, or in support of, regional participate, but that all nuclear powers would organizations-as it has recently in the Middle have to complete nuclear disarmament. East, Cambodia, and Africa. Realization of last- Northeast Asia—Until recently North and ing security arrangements in the other regions South Korea seemed to be making some progress mentioned above will depend on cooperation toward reconciliation between their deeply hos- from extraregional nations. In some cases, agree- tile regimes. At the end of 1991, they signed a ment not to interfere might be enough; in others, “Joint Declaration for a Non-Nuclear Korean agreement to provide collective security32 assur- Peninsula.” In that context, North Korea (a 1985 ances may be necessary. Such cooperation will signatory of the NPT) finally concluded an require that the world’s great powers— overdue agreement with the IAEA for nuclear particularly the permanent members of the U.N.

31 For discussion of the ~d~e &t nuclm problem, see United Nations, Establishment of a Nuclear Weapons-Free ~ne in the Region of the Middle East, Report of the Secretary-General A/45/435, Oetober 1990. 32 ~ ~ ~ontm~ c ~m~wtive security” implies the response of the international eannmnity to aggression by one of its members, not simply an allianm of some states against other states. Chapter 3-Policy Background I 109

Security Counci1 33--work together. For example, recent progress in regional security negotiations, particularly in the Middle East, is partly related to the end of the Cold War and the loss of Soviet patronage for some Arab states. Some analysts argue that the only way to contain proliferation of weapons of mass destruction and other advanced weapons is to change dramatically not only regional security arrangements, but the whole basis of global security. They propose a concept of “cooperative security,” the purpose of which is

. . .to prevent war. . . primarily by preventing the international security arrangements may reduce means for successful aggression from being national incentives to acquire weapons of mass assembled, thus also obviating the need for states destruction. To succeed, both regional and global so threatened to make their own counterprepara- security arrangements will require cooperation among 34 tions. the world’s great powers, particularly the members of These authors argue that proliferation is closely the U.N. Security Council, pictured here deliberating connected to cooperative security: in March 1992 over Iraqi violations of its cease-fire obligations. In order to have any reasonable hope of inducing restraint among the many countries that military spending and for force and investment have the inherent capacity and potential incentive cuts in other regions. Contractions in defense to acquire advanced weapons, the major military industries and control of export sales should be establishments would not only have to subordi- transformed from politically charged national nate their own national forces to international burdens into internationally shared obligations in coalitions. . but also would undoubtedly have to pursuit of the benefit of lower levels of militariza- shrink reciprocally their own forces, levels, and tion everywhere.35 defense industries and would probably have to adopt deployment restrictions embodying the WI-EN NONPROLIFERATION FAILS principles of defensive configuration. They would Some analysts argue that further proliferation also have to radically de- emphasize weapons of of weapons of mass destruction is inevitable and mass destruction in their defense planning. Fortu- that nonproliferation policy is, if anything, coun- nately, historic contractions in military forces and 36 investment of just this sort are taking place terproductive. Others say that although nonpro- throughout North America, Europe, and the liferation policies should continue, it is prudent to former Soviet Union. If carried out cooperatively, plan for at least some further proliferation, and to this contraction can set the standard for reduced be prepared to try to mitigate its consequences for

33 some IE3VC propoaed that the ~t m~“p of the Security Council be enlarged to include other great pow~ ecially ~Y, JapU ad Nti. On the one hanL such an expanded membership would add legitimacy to Seeurity Council actions; on the other, the larger the Council beeomcs, the more unwieldy will be its opcratiom impairing its ability to respond rapidly to emerging crises Should the veto rights now held by the current pcamanent members be extended to new permanent membera, the Council might be less able to achieve the unanimity of its permanent members requked for Council action. ~ ~ton B. ~, William J. Perry, and John D. Steinbruner, op. cit., footnote 2, p.7. 35 Ibid., pp. 3&37. 36 ~ ~m c~ter, op. cit., foomote 1. 110 ! Proliferation of Weapons of Mass Destruction: Assessing the Risks

U.S. and international security .37 Modifying U.S. For example, the current nuclear states could force plans and structures to cope with the implicitly or explicitly acquiesce in the deploy- possible further proliferation of weapons of ment of nuclear weapons by India, Pakistan, mass destruction is unquestionably an impor- Israel, or Ukraine. They could then offer the tant task for U.S. policymakers. Recognizing newcomers to the nuclear club help in developing this fact, the Department of Defense has plans to stabilizing doctrines of deployment and deter- create a new office of Nuclear Security and rence. The help might be technical assistance to Counterproliferation to be headed by an Assistant reduce the vulnerability of their nuclear forces to Secretary of Defense. a disarming first-strike from others. Or, it might As noted above in the section on creating take the form of technology for tightening cen- disincentives for proliferants, some policies can tralized control over the weapons themselves and simultaneously serve the purposes of deterring for preventing unauthorized use, theft, or acci- acquisition of the weapons in the first place and dents. Promoting safer deployment of weapons of of deterring or militarily countering their use mass destruction would be inconsistent with a when nonproliferation fails. It is important to stated goal of a global ban on possession-as in note, however, that some preparations to mitigate the cases of chemical and biological weapons. the consequences of proliferation might also But in the case of nuclear weapons, the policy exacerbate the process of proliferation. might ‘‘grandfather’ nuclear arms deemed to be One can make deterrent counter-threats to irreversibly deployed, as the NPT does those of dissuade the proliferant from using his weapons. the United States, Russia, Britain, France, and Analysts have variously hypothesized that Iraq China. failed to use its chemical weapons against coali- A policy of acceptance might mitigate postpro- tion troops because it feared U.S. retaliation in liferation risks, but it would also tend to encour- kind, U.S. resort to nuclear weapons, or escalation age further proliferation by showing that success- of the conventional attack to the point of eliminat- ful evasion of the obstacles to proliferation can ing the Hussein regime; others suggest Iraq just eventually lead to legitimacy as a member of the calculated that there was no useful application nuclear club.38 Technical assistance on safety available for chemical weapons. Some argue that and security measures could also lead the new the possibility of Israeli nuclear retaliation de- nuclear power to integrate its weapons more terred Iraq from using Scud missiles with chemi- tightly into its military forces, keep them at higher cal warheads against Israel. levels of alert, and think of them as more usable Noncoercive measures to try to manage the instruments of force. And making the weapons consequences of proliferation are also conceiva- more secure from preemptive first strikes from ble. Given a case in which weapons of mass their neighbors would also make them more destruction are deployed despite U.S. wishes to secure from a U.S. or multinational preemptive the contrary, it would be in U.S. interests to strike. minimize the resulting dangers.

37 For discussion of boti nonprowe~tion measures and proliferation mitigation m&WIR s, see Giles, op. cit., foomote 25. 36 @ mer @ ~e problem of encouraging further proliferation would be to permit a one-time-only expansion of the nucl= club. ‘f’he U.N. Security Council could set a deadline for states to declare themselves nuclear-weapon states, after which it would treat all further nuclear proliferation--including any existing but undeclared programs-to be a threat to international peacs jus@ing a Council inspection and eliminau“on program. See David Kay, ‘‘The IAEA-How Can It Be Strengthened?, ’ Woodrow Wilson International Center for Scholars, Conference, “Nuclear Proliferation in the 1990s: Challenges and Opportunities,” Dec. 1-2, 1992. —.

Chapter 3-Policy Background I 111

SPECIAL AND URGENT: LIMITING 9 Maintaining Central Control of Former PROLIFERATION FROM THE FORMER Soviet Weapons and Materials SOVIET UNION ● The United States could lead in the creation Most of the policy tools described so far in this of an international forum to coordinate chapter will be relevant to the republics of the efforts to help denuclearize the former So- former Soviet Union. But, as indicated in ch. 2, viet republics. the breakup of the Soviet Union has led to new ● The United States and other nations can kinds of proliferation risks. The extent to which continue to insist (along with appropriate the former Soviet republics will disseminate carrots and sticks) that Ukraine and Ka- technology, materials, and expertise for produc- zakhstan ratify the START agreement and ing nuclear, chemical, and biological weapons (as the NPT, confirming their non-nuclear status. well as ballistic missiles) is still far from certain. Belarus, which has done so, could be re- Nor is it certain that the former Soviet weapons warded, and the countries providing eco- themselves will remain under firm and responsi- nomic assistance to the former Soviet repub- ble central control, or that the three non-Russian lics could condition all types of aid on republics having some of the weapons within continued progress in promised denucleari- their borders will yield all of them up for zation. elimination. ● Nunn-Lugar funds are supplying equipment The situation in the former Soviet Union is for the secure transport of nuclear weapons only partially amenable to outside influences. to central locations. The money is also to be Nevertheless, the United States and other nations used to help build storage facilities for the can take steps to encourage favorable outcomes. plutonium from dismantled nuclear weapons. In 1991 and 1992, Congress and the administra- ● The United States could also encourage use tion attempted to help limit these risks by of this money to accelerate current Russian budgeting $400 million each year (beginning dismantlement schedules; the United States with the Nunn-Lugar Soviet Threat Reduction could set an example by accelerating its own Act of 1991) to assist former Soviet demilitariza- dismantlement process. tion. Listed below is a range of policy measures ● The United States has agreed to purchase for addressing the risks identified in ch. 2. Some highly enriched uranium from Russian nu- of these measures are already supported by the clear weapons for use, once diluted to lower Nunn-Lugar and Freedom Support Act funds; levels of enrichment, as fuel in nuclear 39 others are possible future steps. power reactors.

39 Several of the additional steps listed below were advocated by Senators Nunn and Lugm in December 1992; s~ Sm NW ad Richd Lugar, “Still a Soviet l%rea~” Washington Post, Dec. 22, 1992, p. A-21. For a comprehensive collection of policy options for dealing with the nuclear risks posed by the breakup of the Soviet UnioU see GTaham AllisorL Ashton B. Carter, Steven E. Miller, and Philip Zelikow, (eds.), Cooperative Denuclean”zacion: From Pledges To Deeds, CSIA Studies in International Security No. 2 (Cambridge, MA: Harvard University, Center for Science and International Affairs, 1993); for discussion of U.S. nongovernmental efforts toward verified dismantlement of nuclear weapons, see Federation of American Scientists and Natural Resources Defense Council, Report of the Fourth lnrernafional Workshop on Nuclear WarheadElim”nution and Nonproliferation, held in wmhingto~ D. C., Feb. 26-27, 1992 (Washington DC: Federation of American Scientists, 1992); see also Christopher Paine and Thomas B. CochrarL “Ver@ing Dismantlement” Arms Control Today, VO1. 22, No. 1, January/February 1992, pp. 15-17. 112 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

● The United States could propose internation- otherwise be tempted to accept foreign weapons 40 ally monitored storage or disposition of work to earn a living. Joint projects between highly enriched uranium and plutonium U.S. and former Soviet laboratories and firms from dismantled nuclear weapons in both the might be another contribution to that goal. former Soviet Union and the United States. Efficient enforcement of laws and regulations ● The United States could urge accelerated may help.41 Overall improvement in the econo- implementation of START strategic nuclear mies of the states that emerged from the former arms reductions; it could ratify promptly the Soviet Union is probably the best hope for START II agreement and also urge rapid discouraging this kind of emigration. implementation of those reductions. ● The United States and others can offer diplomatic support and, where appropriate, ~ Controlling Export of Critical Information, financial assistance to help settle ethnic and Equipment, or Materials regional conflicts and promote regional se- The Russian Government has issued specific curity regimes. regulations on the export of goods that might be used to make weapons of mass destruction. 1 Preventing Export of Weapons or of Information about the regulations of other former Weapon Components Soviet republics is still spotty .42 But former If weapons of mass destruction remain under Soviet military enterprises, new companies, and effective central Russian control, their export local and regional governments are striving to seems unlikely. But the same sorts of civil earn foreign hard currencies through exports of all disorder and governmental breakdown that could kinds. In the transition to a market economy, there lead to weapons or key components falling into is reason to question how effectively controls on illegitimate hands could also foster the sale either declared exports or smuggled goods will be abroad of such goods. With or without coopera- administered. Western governments with greater tion from officials in the former Soviet states, experience in export regulations may be able to U.S. and other foreign intelligence services may offer technical assistance. U.S. and other intelli- be able to help monitor and stop illicit transac- gence agencies may be able to track questionable tions. exports and direct the attention of authorities in the former Soviet republics to specific problems. H Inhibiting Emigration of Technical A private U.S. organization43 is working with Personnel groups in the former Soviet Union on a project on The United States and other nations have “Building Communities of Nonproliferation Spe- supported creation in Moscow and Kiev (Ukraine) cialists in the Former Soviet Union. ’ If such of International Science and Technology Centers, communities can be established, they may con- intended to help establish meaningful, nonmili- tribute to more rigorous implementation of non- tary work for scientists and engineers who might proliferation policies in the former Soviet Union.

40 o~er ~~ IIMy dso open in MI@ (Belarus) tind MIM-AUI OQZ@S@. 41 ]zveSriVa~ ~t~ fomimrew~ tit Russ~ au~orities bl~~ some nucl~pow~ specUStS from traveling to Nofi Korea. kLWhII officials confirmed to kvestiva that border troops were instructed to detail “a certain category” of Russians thought to be “bearers of SCXXCtS. ” Izvestiva, Dee. 22, 1992, p. 2, trans. in FBIS-SOV-%!-246. Dec. 22, 1992, pp. 15-16. 42 For w~t is publicly avti~blc, see Potter, NucZear Profiles ojthe Soviet Successor States, 10C. Cit. inch. 2, footnote 47.

43 ~ as NOq~lifm~tiOn ~j~t of ~ Cen@r for Russi~ ~d EHfi Smdies at tie Monte~y rnst.itute of hlt~tiolld Shldh, with grants from various U.S. foundations. Appendix 3-A Costs and Benefits of Onsite Inspections for Nonproliferation Regimes

Efforts to strengthen verification of compli- judgments about how high a price to pay for ance with the NPT or the BWC will have to take enhanced verification regimes is the fact that into account the difficulties of balancing costs increments of intrusion and expense will not and possible benefits from onsite inspections. necessarily lead to correspondingly higher confi- Costs include the following: dence in compliance. The United States, for ● Costs of inspecting: inspection teams, equip- example, in negotiating the CWC, judged that ment, and operations, whether they are ‘‘anywhere, anytime’ challenge inspections would nationally supplied or work for international not bring sufficient returns in verification to organizations, cost money. justify the costs (primarily the third category of ● Costs of being inspected: personnel of the government or industrial facilities undergo- sm ing inspections have to spend time and 1 / b money preparing those sites to protect classified or proprietary information from exposure to foreign inspectors. The inspections themselves may to bring site operations to a halt, costing more time and money. ● Compromise of nonpertinent informa- “zig- tion: preparations to protect information that can properly be concealed from inspectors may not always be successful or affordable. 4 Officials worry about revealing military or

industrial secrets or losing competitive ad- e’ vantages. Costs of onsite inspection are one issue in assessing The costs in the frost category are not too difficult the net value of verification regimes for to calculate; those in the second category are more nonproliferation agreements. Pictured here is an x-ray difficult to estimate; those in the third are nearly fluorescence spectrometer used to analyze samples impossible to quantify. Further complicating from nuclear inspections in Iraq.

113 114 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

costs above: endangered military and intelligence indeed any kinds of weapons) have generally secrets).1 acknowledged that no verification regime will be Other verification enhancements also impose perfect: a nation that wants to cheat badly enough costs. For example, improved international export can probably get away with it at some level. reporting to monitor flows of dual-use technolo- Therefore, they suggest as a practical standard gies could compromise legitimate competitive that verification measures should be able to detect advantages for some of the companies involved. “militarily significant” violations. The follow- A multilaterally conducted verification regime ing two examples indicate how the Reagan carries yet another risk: that the information administration framed this concept in its support collected by an international organization might of the Intermediate Nuclear Forces Treaty. Based prove useful to potential proliferants within the on Cold War strategic calculations, these example organization. This risk is greatest in the nuclear do not directly pertain to the consequences of field, where significant weapons know-how is violating nonproliferation commitments. But they still difficult to acquire. Thus, in gathering details do illustrate issues important to any arms control about the Iraqi nuclear weapons program, the agreement. IAEA has tried to restrict the information to At that time presidential arms control adviser members of the organization already possessing Paul Nitze said that the administration. . . . nuclear weapons. . . . would consider the standard to be whether or To be weighed against the above costs is the not the Soviet Union could covertly deploy a question of how much verification is enough. force which would be militarily significant and Beyond the issue of the symbolic or psychic whether we could find that out. . .in a timely benefits of various verification measures, poli- fashion, so that we could take offsetting actions cymakers need to judge arms control regime ourselves in a timely manner.2 verification requirements (e.g., those for the In written responses to questions, Secretary of projected Chemical Weapons Convention) in at State George Schultz outlined factors that would least three dimensions: go into a determination of military significance: ● the significance of potential violations, ● the quantitative level and overall threat . the verification measures that would be presented to the United States and NATO; required to deter or detect significant viola- ● qualitative factors, including kinds of weap- tions, and ons and their capabilities; . the tangible and intangible costs of those ● an assessment of the state of readiness and verification measures. training of the cheating force; Inevitably, judgments on these matters will be ● the extent to which other forces available to complex, subjective, and open to debate: no the cheater make cheating forces redundant conclusive technical criteria will be possible. or add significant capability; The discussion below shows why this is so. ● the extent to which existing U.S. or allied Proponents of arms control or disarmament forces could permit an effective counter; agreements for weapons of mass destruction (or, and

1 The Chemical Weapons Convention has the most extensive onsite inspection regime of any international nonproliferation cmnmitment. For an analysis of some of the costs of implementing such a regime, see U.S. Congress, Ofilce of ‘khnology Assessment The Chenu”cal Weapon.t Conven#”on: Effects on the U.S. Chemical hdh.rtry, OTA-BP-ISC-1O6 (Washington DC: U.S. Government Printing ~lce, August 1993). z Ambassador Paul Nitze in U.S. Congress, Senate Committee on Foreign Relations, The fNF Treaty, Hearings, Part 1, Serial Number 100-522, Pt. 1, 1988, p. 301, —....— ———. — . .

Appendix 3-A-Costs and Benefits of Onsite Inspections for Nonproliferation Regimes I 115

. the overall political and military situation significant quantities of agents points out that surrounding discovery of cheating (e.g., estimates might depend on whether chemical relative stability or tension).3 weapons are being used in covert sabotage In other words, even in the limited case of operations, mass destruction of civilian popula- long-range theater nuclear forces, the actual tions, or battlefield situations. The analysts in that number of illegal missiles would be only one study settled on tactical battlefield employment factor in a judgment of the military significance as their base case. They point out that in that of a potential arms control violation. setting, military significance could depend on: In the nonproliferation context, the problem of ● the toxicity of the agent, defining militarily significant levels of violation ● the degree of incapacitation or mortality is even more difficult. For example, while the possession of 10 illicit nuclear weapons might sought, mean nothing between the United States and ● weather conditions, Russia, who each have thousands more, the same ● the degree of protection of the target troops, number might appear overwhelmingly decisive in ● the delivery systems used, and a contest between, say, Iraq and Saudi Arabia. ● the size of the target region. In addition, when weapons of mass destruction These analysts then decided that a possibly are to be employed as instruments of threat or militarily significant attack would be one against terror, how can one determine what is a militarily 10 battalions in a 100 km2 area. In that case, significant capability? In the case of biological significant quantities might be 30 tons of VX weapons, for example, a small quantity of agent nerve agent or 1,000 tons of mustard gas. They (much less than a ton), properly delivered, could caution, however, that ‘‘these values cannot at kill hundreds of thousands of people; if used this point be equated with detection goal quanti- against protected troops, the same weapons might 4 ties associated with treaty monitoring.” have little military effect. Chemical weapons present comparable problems. One systematic attempt to assess militarily

3 Ibid., pp. 470-471. d Mark F. Mullem Kenneth E. Apg and William D. Stanbro, Criteria jior Monitoring a Chenu”caZ Arms Treaty: Implications for the Verification Regime (Los Alamos, NM: Los Alamos National Laboratory Center for National Security Studies, Report No. 13, December 1991), pp. 5-7. —.

Index

Ad Hoc Group of Governmental Experts on Verification (of monitoring delivery vehicles, 42A13 Biological Weapons Convention), 19 proliferant’s delivery systems, 14,68-69 Advanced delivery systems, 34,51-52,69. See also specific proliferation motivations, 101 types by name in suspected proliferant nations, 46, 66-68 Aerodynamic separation techniques, 36 Belarus. See also Former Soviet republics Africa, 64. See also specific coun~ies by name control of weapons in, 111 Agents of mass destruction risks from breakup of Soviet Union, 4, 75 biological weapons, 6,38-39 weapons of mass destruction in, 12, 63-65 chemical weapons, 6, 36-37 Benefits offered to forgo weapons of mass destruction. See lethality of, 46,48-49 Rewarding nonproliferation; Security improvements surveyed, 47 as nonproliferation incentive Aircraft Biological weapons as advanced delivery systems, 3-4,41-42, 51-52 characteristics of, 2-3, 8-9, 47-50 detection of development and testing, 43 delivery of, 6, 39AI0, 50-52 proliferant’s delive~ systems, 68-69 destructive effects of, 4748,52-56 Algeria, 13,64,66 dual-use technologies and, 6,38-40 Approaches to nonproliferation policy environmental effects, 72-73 coercive v. consensurd policies, 31 implications of new technology, 40 intelligence collection v. use, 32 institutional bases of nonproliferation regime, 85 scope of control efforts, 29-30 militarily significant levels of violation for, 115 targeted v. universal approaches, 31-32 military utility of, 56-57,60-61,62-63 unilateral v. international approaches, 30-31 monitoring production, 39-40 Arab nations. See also specific countries by name onsite inspections and, 7 nuclear weapons in the Mideast and, 70, 71, 102, 108 production, 9-11,38-39 regional conflicts and, 16, 109 proliferation motivations, 100-101 Argentina reducing international role of weapons, 106 delivery systems in, 67-69 strategic uses of, 62-63 policy reversals, 18 in suspected proliferant nations, 14-15, 63-66 nuclear weapon program in, 14, 64 tactical uses of, 55-61 and Nuclear Non-Proliferation Treaty, 101 technical basis for monitoring and controlling prolifera- reducing incentives for weapons, 107 tion, 38-40 Arms Control and Disarmament Agency, 25 unconventional delivery systems and, 69 Arms Export Control Act, 85,87 U.S. military O~mtiOnS and, 74 Arms races, 71, 106-107 Biological Weapons Convention (BWC) Atomic Energy Act, 85-87,90,95 institutional bases of nonproliferation regime, 85 Australia Group, 5, 85, 88 international norms and, 2 linking assistance to nonproliferation, 103 Ballistic missiles monitoring compliance, 40 as advanced delivery systems, 3-4, 41-42 proliferation motivations and, 99 detection of development and testing, 42A3 rewarding nonproliferation, 22 indigenous production capabilities for, 67 security benefits as nonproliferation incentive, 104-106 117 118 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

suspected proliferant nations and, 63-66 Civil defense measures, 62-63 technical training obligations and, 91 Classifkxl information toxin bans, 3 implications of, 15, 32 verification regime, 18-19, 22-25, 113 protection of during inspections, 23-24, 113-115 Brazil CoCorn. See Coordinating Committee on Multilateral Export delivery systems in, 67-69 Controls nuclear weapon program in, 14, 64 Coercive policies. See Disincentives and sanctions against policy reversals, 18 proliferants proliferation motivations, 101 Cold War end and implications for proliferation, 15-18 reducing incentives for weapons, 107 Commerce, Department of, 25 Burm& 65-66 Commonwealth of Independent States, 63-64. See also Bush adrmms“ - tration, 90 Former Soviet republics BWC. See Biological Weapons Convention Conflicting nonproliferation policy approaches and objectives. See Approaches to nonproliferation policy; Central Europe, 74. See ako specific countries by name Objectives of nonproliferation policy Chemical and Biological Weapons Control and Warfare Consensual treaties. See International agreements Elimination Act, 85,87,90,95 Consequences of proliferation Chemical weapons characteristics of weapons of mass destruction, 2-3, 7-11, characteristics of, 2-3, 8-9,46-50 46-50 delivery systems, 50-52 Cold War end and implications for proliferation, 15-18 destructive effects of, 52-56 international community, implications for, 11-12, 69-73 dual-use technologies and, 6,36-38 Soviet Union breakup effects, 1,4, 15,75-77, 111-112 institutional bases of nonproliferation regime, 85 Summary, 45-46 Iraqi failure to use against coalition troops, 110 U.S. political-militaxy policies, implications for, 4, 11-12, militarily significant levels of violation for, 115 73-75 military utility of, 56-57,58-60,62 onsite inspections and, 7 weapon effects compared, 52-63 production, 9-11,36-37 Contagious weapon agents, 49-50 proliferation motivations, 100-101 Controlling proliferation. See Monitoring and controlling quantities for various missions, 60 proliferation reducing international role of weapons, 17, 106 Convention on the Prohibition of Bacteriological (Biologi- strategic uses of, 62 cal) and Toxin Weapons. See Biological Weapons in suspected proliferant nations, 14-15,63-66 Convention tactical uses of, 55-61 Conventional weapons technical basis for monitoring and controlling prolifera- compared with weapons of mass destruction, 46, 56-57 tion, 36-38 motivating development of weapons of mass destruction, Chemical Weapons Convention (cWC) 29, 102 chemical weapons disarmament and, 18 linkages to weapons of mass destruction, 29-30 institutional bases of nonproliferation regime, 85 Cooperative security concept, 109 international norms and, 2, 5, 18 Coordinating Committee on Multilateral Export Controls, linking assistance to nonproliferation, 103 85,89 onsite inspections and, 7 Cordesman, Anthony, 58-59 proliferation motivations and, 99 Counterproliferaton, 19-20,25,29,84,91-92, 94-97 rewarding nonproltieration, 22 Covert activities, 7,92. See aZso Intelligence agencies security benefits as nonproliferation incentive, 104-105 Cruise missiles, 3-4,43,51-52,68-69 stockpile destruction specifications, 72 CWC. See Chemical Weapons Convention suspected proliferant nations, 63-66 technical training obligations and, 91 Delivery systems. See also Aircraft; Ballistic missiles; toxin bans, 3 Cruise missiles verification regime, 22-25, 113 barriers to missile and aircraft proliferation, 41-42 China biological weapons and, 39 conflicting U.S. policy objectives with nxpct to, 5, 26 chemical weapons and, 37 delivery systems in, 66-69 comparison of aimraft and missile delivery of chemical lack of requirement of ‘full-scope safeguards’ for nuclear and biological agents, 53-54 exports, 18 monitoring delivery vehicles, 4243 mass destruction weapon programs in, 12, 14, 17, 63-66 nuclear weapons and, 34-35 proliferation motivations, 102 in suspected proliferant nations, 14, 66-69 regional security concerns and, 16, 108 technical basis for monitoring and controlling, 4043 —..

Index I 119

~S of, 3-4,50-52 Exports. See also Export controls unconventional delivery systems, 69 Former Soviet Union, critical information, equipment, or Department of Defense, 25,29, 110 materials, 76-77 Destructive agents. See Agents of mass destruction limiting regional arms races, 107 Destructive effects of weapons, 52-55 Soviet Union, delive~ systems, 66 Deterrence policies, 28,73-74 Soviet Union, weapons or components, 76, 112 Diplomatic responses to proliferation, 97-98 Discriminatory nonproliferation regimes, 17, 32, 101 Failures of nonproliferation, 25, 109-111 Disincentives and sanctions against proliferants Financial assistance andnonproliferation policy, 22,102-104 balancing coercive v, consensual policies, 31 FMle materials. See Nuclear materials. balancing targeted v. universal approaches, 31-32 ForeQn Assistance Act, 85,90,95 coercion benefits and limits, 97-98 Foreign trade. See Export controls; Exports diplomatic and military responses, 5,20-21,94-97 Former Soviet republics economic sanctions, 5, 93-94 control of weapons in, 72, 75-76, 111-112 Dual-use technologies delivery systems in, 66-67 biological weapons and, 6,3840 emigration of technical personnel, 112 chemical weapons and, 6, 36-38 environmental effects of weapons of mass destruction, controls on as obstacle to proliferation, 6, 86 72-73 exports from former Soviet Union and, 76-77 export of critical information, equipment, or materials, technical basis for monitoring and controlling prolifera- 33-34,41-42,76-77, 111-112 tion, 32 export of weapons or components, 76, 112 spread of contributing to proliferation, 16 risks from bzeakup of Soviet Union, 4, 15,75-77, 111-112 weapons in, 14, 63-65, 75-77, 111-112 EAA. See Export Adrrum“ “stration Act France, 12, 17,63,70 East Asia, 64-69. See also specific countries by name Freedom Support Act, 85 Economic issues as incentive to forgo weapons of mass destruction, 22, Gas centrifuge technology, 35-36 102-104 Genetic engineering techniques, biological agents and, 40 costs of producing nuclear materials, 11, 33-34 Geneva Protocol, 17,85, 106 onsite inspection, costs and benefits of, 28, 113-115 Glenn-$m.ington amendments to Foreign Assistance Act, sanctions against proliferants and suppliers, 5, 20, 27, 90,95 88-89,93-95 GPS. See U.S. Global Positioning System Education of foreigners, transfers of expertise and, 91 Effects of weapons of mass destruction. See also Conse- Health and safety effects of producing weapons of mass quences of proliferation destruction, 39,72-73 destructive effects of weapons, 46-50,52-55 Hussein, Saddam, 100, 102, 110 environmental effects, 39, 72-73 military utility of weapons, 55-63 IAEA. See International Atomic Energy Agency strategic uses of weapons, 55, 61-63 India tactical uses of weapons, 55-61 coercion benefits and limits, 97-98 Egypt, 65-67 delivery systems programs, 66-69 Energy, Department of, 25 difficulty of bringing into nuclear nonproliferation regime, Enhanced Proliferation Control Initiative, 90 104 Environmental effects of weapons of mass destruction, 39, proliferation motivations, 71, 102 72-73 purchase of Russian missile technology, 76 Europe, 108. See also specific countries by name regional security concerns and, 16, 108 Executive branch discretion in nonproliferation policies, weapon of mass destruction programs, 13, 17, 64-66, 31-32,94 104-105 Executive Order 12735,87 Infectious weapon agents, 49-50,72-73 Export Mrmms“ “ tration Act, 85, 87,90 Intelligence agencies Export controls collection v. use of intelligence, 32 barriers to missile and aircraft proliferation, 41-42 monitoring of proliferation, 32-33, 112 exemptions from, 104 role of intelligence in nonproliferation policy, 94, % export promotion v. export controls, 26-28 Intermediate Nuclear Forces Treaty, 114 multilateral cooperation in export control regimes, 18 International agreements, 22-25, 31, 84-85, 104-105. See as information sources, 19,86 also specific agreements by name as obstacles to proliferation, 5, 16, 19-20, 84-88 International Atomic Energy Agency Export controls groups. See Multilateral control groups creation of, 102 120 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

Iraqi facilities, monitoring of, 93, 114 seizure of Soviet weapons by non-Russian authorities, nuclear safeguards, 18, 23, 34, 101, 108 75-76 international Atomic Energy Agency Statute, 85 weapons of mass destruction in, 12, 63-65 International community Korean peninsula, 4, 12. See also North Korea; South Korea cooperation among necessary for nonproliferation, 4-5 domino effect and proliferation, 71, 102 Laser isotope separation, 36 double standards and proliferation, 101 Latin America, 64, 67-69. See also specific countries by norm against proliferation, 17 name prerequisites to effective nonproliferation policy, 4-6 Latin American Nuclear-Free Zone Treaty. See Treaty of proliferation costs and risks, 71-73 Tlatelolco proliferation, implications for, 11-12,69-73 Legal sanctions. See Sanctions against suppliers reducing international role of weapons, 106 Mhality of weapon agents and military action against proli.ferants, 20,92 factors affecting, 46,48-49,53-54 International norms nuclear weapons, 34-35 erosion of, 17, 71 weapon characteristics and comparisons, 7-9 importance of for nonproliferation objectives, 1-2, 17,21, weapon effects, 52-55, 61 30,94-96 Libya, 13,65-69 unilateral and international policy approaches, 30 London Club. See Nuclear Suppliers Group International Science and Technology Centers, 112 Long-range delivery systems. See specific types of system by International security arrangements, 22-25, 107-109 name ‘‘Internationalist’ approach to policy, 30-31 Iran Middle East. See also specific countries by name delivery systems in, 67-69 delivery systems in, 66-69 weapon of mass destruction programs in, 13, 64-66 reducing incentives for weapons, 107 regional security concerns, and, 16, 107-108 Iran-Iraq Non-proliferation Act, 85,90 weapon of mass destruction programs in, 4,64-66,70-71, Iran-Iraq WW, 17,58-59,64-66 102, 108 w Militarily si~lcance of treaty violations, 114-115 costs of nuclear weapon program, 33, 72 Military delivexy systems, 50-51. See also Advanced deliv- delivery system programs in, 66-69 ery systems environmental effects of weapon of mass destruction Military operations program, 72 as responses to proliferation, 19-20, 25, 84, 91-92, 94-97 international response to, if nuclear-armed, 70-71 proliferation implications for U.S., 12,74 militaty response to proliferation, illustration of, 91, %-97 U.S. defense commitments as nonproliferation incentive, monitoring of facilities in, 93, 114 107 proliferation motivations, 100 Military rivahies, proliferation motivations and, 102 U.N. Security Council cease-fro agreement, 18,91,92-93 Military utility of weapons of mass destruction, 7-9,55-63. U.S. political-military policies and, 73 See also Strategic uses of weapons of mass destruc- weapons of mass destruction, use by Gulf War, 10,61, 110 tion; Tactical uses of weapons of mass destruction weapon of mass destruction programs in, 12-13, 15,63-66 Missile Technology Control Act, 85,90 Israel Missile lkchnology Control Regime, 35,4142, 66,76, 85, conflicting U.S. policy objectives with nxpect to, 26 89 delivery systems in, 66-69 Missiles. See also Ballistic missiles; Cruise missiles difficulty of bringing into nuclear nonproliferation regime, institutional bases of nonproliferation regimes, 85 104 Monitoring and controlling proliferation. See also Onsite proliferation motivations, 102 inspections regional security concerns, 16, 108 monitoring others v. avoiding costs of being monitored, 28 weapon of mass destruction programs in, 13, 17, 64-66, monitoring proliferation and verifying compliance with 104 agreements, 6-7 technical basis for, 32-43 Japan, 34, 107 Motivations for proliferation domino effect and, 71, 102 Kazakhstan. See also Former Soviet xepublics international double standards and, 101-102 control of weapons in, 111 military rivalries and, 102 indigenous weapons production and development, 77 perceived value of weapons, 99-101 production facilities in, 77 MTCR. See Missile ‘lkchnology Control Regime risks from breakup of Soviet Union, 4 Multi-use technologies. See Dual-use technologies Index 1121

Multilateral export control groups, 5,18,85. See also specific Nuclear Suppliers Guidelines, 85, 88 groups by name Nuclear tests Multilateral export controls, 86-89 detection of, 35 Myanmar, 65-66 lack of need for, 35 Nuclear weapons National export controls, 86-87 ballistic missile programs and, 46 Near-term proliferation threats. See Suspected proliferant characteristics of, 2-3, 7-8, 47-50 nations costs to mount weapon program, 33, 72 Nitze, Paul, 114 delivexy systems, 50-52 Nonproliferation policy. See also Detemence policies; Policy destructive effects of, 52-56 choices and tradeoffs institutional bases of nonproliferation regimes, 84-85 categories of, 5, 83-84 likelihood of use, 9, 28-29,69-70, 110 disincentives and sanctions against proliferants, 5, 20-21, militarily significant levels of violation for, 115 93-98 militaxy utility of, 55-62 failures of nonproliferation, 25, 109-111 noncoercive measures to manage proliferation, 110 former Soviet republic situation, 111-112 production, 9-11, 33-35 imposing obstacles to proliferation, 5, 19-20, 84-93 proliferation motivations, 99-101 institutional bases of nonproliferation regimes, 84-85 reducing international role of weapons, 106 rewards for abstention, 5, 21-22, 98-104 regional conflicts and, 16, 108 security benefits, 5, 22-25, 104-109 secrecy as obstacle to proliferation, 84 SUII_UTIUy, 4-6, 83-84 strategic uses of, 61-62 North AtYica, 64. See also specific countries by name in suspected proli.ferant nations, 14-15, 63-66 North Korea tactical uses of, 55-61 delivery systems in, 66-69 technical basis for monitoring and controlling prolifera- missile exports, 41 tion, 33-36 and Nuclear Non-Proliferation Treaty, 13-14, 108 unconventional delivery systems and, 69 policy reversals, 18 U.S. military operations and, 74 reducing incentives for weapons, 107 regional security concerns, 4, 16, 108 Objectives of nonproliferation policy, conflicts among, weapon of mass destruction programs in, 13-14, 17,64-66 export controls v. export promotion, 26-28 Northeast Asia, 97, 108, See also specific countries by name monitoring others v. avoiding costs of being monitored, 28 NTPT. See Nuclear Non-Proliferation Treaty nuclear deterrence v. nuclear nonproliferation, 28 Nuclear deterrence policies, 28,73-74 preventing proliferation v. preventing use, 28-29 Nuclear Exporters Committee (Za.ngger Committee), 85,88 Obstacles to proliferation Nuclear materials, 6, 10,23, 33-36, 76-77 export controls, 19-20, 84-88 Nuclear Non-Proliferation Act, 85,87,95, 103 forcible interference, 19-20,25,84,91-92,94-97 Nuclear Non-Proliferation Treaty hampering transfers of expertise, 90-91 adherence to by former Soviet republics, 4, 75-76, 111 Iraq and, 92-93 background, 1 sanctions against suppliers, 20, 88-90 ban on unsafeguarded nuclear facilities, 35 StXP3Cy, 84 central bargain of, 22, 104-106 summary, 4-5 disc rirninatoxy nonproliferation regimes issue, 17, 101 Onsite inspections, 105, 113-115, See also Monitoring and and export controls, 88 controlling proliferation increasing membership of, 17 Organization for the Prohibition of Chemical Weapons, 23 institutional bases of nonproliferation regimes, 85 and the International Atomic Energy Agency, 23 Pakistan international norms and, 2, 17 delive~ systems in, 30,67-69 linkage of technical assistance to nonproliferation, 22,98, difficulty of bringing n nuclear nonproliferation regime, 102-103 104 mo@ing to accommodate additional nuclear powers, proliferation motivations, 102 28-29 regional security concerns and, 16, 108 security improvements as nonproliferation incentive, 104- weapon of mass destruction programs in, 13, 17, 64-66, 106 104-105 technical training obligations and, 91 Perceived value of weapons, 99-101 treaty status of suspected proliferant nations, 13-14, 64 Plutonium, material production and, 33-34 veriilcation regime, 22-25, 113 Policy, See Nonproliferation Policy Nuclear Proliferation and Safeguards, 99 Policy choices and tradeoffs Nuclear Suppliers Group, 5, 18, 76,85, 88 conflicting approaches to policy, 29-32 122 I Proliferation of Weapons of Mass Destruction: Assessing the Risks

conflicting objectives of policy, 25-29 security benefits as nonproliferation incentive, 105 introduction, 14 weapon of mass destruction programs in, 4, 64 major findings, 4-6 South Korea policy background, 19-25 delivery systems in, 67-69 Political fragmentation, weapons of mass destruction and, 72 proliferation motivations, 102 Presidential role in sanctions, 94 reducing incentives for weapons, 107 Pressier amendment to Foreign Assistance Act, 95 reduction of security threats, 108 Proprietary information, protection of during inspections, U.S. role in reversing nuclear weapon program, 94 23-2$, 113-115 weapon of mass destruction prognms in, 18 Southeast Asi% 65-66. See also specific countries by nume Regional conflicts, nuclear weapons and, 16, 108 Southern Afica. See akko specific countries by name Regional security arrangements, 24-25, 107-108 Soviet Union, former. See Belarus; Commonwealth of Rewarding nonproliferation, 21-22,98-99, 102-104 Independent States; Former Soviet republics; Ka- Risks of proliferation. See Consequences of proliferation zakhstan; Russia; Ukraine Russia. See also Former Soviet republics ST~T I Txeaty, 64-65,75, 111-112 Cold War end and, 17-18 STMT II agreement, 112 delivery systems in, 14,66 State, Department of, 25 dramatic reductions in deployed nuclear forces of, 17,75, Strategic bombing, 61-62 99-1oo Strategic uses of weapons of mass destruction, 55,61-63 environmental effects of producing weapons of mass Supplier groups. See Multilateral export control groups destruction, 72-73 Suppliers of weapons of mass destruction maintaining control of nuclear weapons and materials in, proliferation implications, 71 72,76, 111-112 sanctions against, 5, 20, 27, 88-90 political stability of, 72,76 Suspected proliferant nations security assurances from, 105, 108 delivery systems in, 66-69 risks from bnmkup of Soviet Union, 4,15,75-77,111-112 proliferation threat and, 4 weapons of mass destruction in, 12, 17, 63, 65 weapon of mass destruction programs in, 12-16, 63-66 Russian Republic. See Russia Sweden, 18 Syria, 65-69 Sanctions against proliferants, 5,27,93-95 Sanctions against suppliers, 5,20,27, 88-90 Tactical uses of weapons of mass destruction, 55-61 Saudi Arabia, 67-69 Taiwan Schultz, George, 114-115 delivery systems in, 66-67 Security benefits as nonproliferation incentive U.S. role in reversing nuclear weapon program, 94 arms control, 106-107 weapon of mass destruction programs in, 18, 65-66 assurances from existing owners of weapons of mass Targeted v. universal policy approaches, 31-32 destruction, 105-106 lkchnical aspects of nonproliferation policy. See also consensual agreements, 104-105 Technical basis for monitoring and controlling prolif- defense commitments, 107 eration global security arrangements, 22-24, 108-109 issues for controlling proliferation, 6 reducing international role of weapon of mass destruction, monitoring proliferation and ver@ng compliance, 6-7 106 spread of technology and industrialization, 16 regional security arrangements, 24-25, 107-108 technical assistance and nonproliferation policy, 102-104 security benefit possibilities, 98-99 Technical basis for monitoring and controlling proliferation. summary, 5 See also Technical aspects of nonproliferation policy Short-range delive~ systems, 51 biological weapons, 3840 Signatures of weapons facilities, 9,32-33,35-40 chemical weapons, 36-38 Solarz amendment to Foreign Assistance Act, 95 delivery systems, 40-43 South Africa nuclear weapons, 33-36 delivery systems in, 67-69 s~v 32-33 formalization of nuclear status, 104 Technical persomel policy reversals, 18 emigration from former Soviet Union, 76, 112 proliferation motives, 73, 100 hampering transfers of expertise, 90-91 weapon of mass destruction programs in, 13, 17, 64-66 nuclear weapon programs and, 35 South Asia Terrorist activity, 50,71-72 delivery systems in, 67-69 Toxins reducing incentives for weapons, 107 barming of, 3 regional- security concerns ad, 16, 108 characteristics of weapon agents, 47 Index 1123

dual-use technologies and, 6,39-40 deterrence, 11-12,73-74 onsite inspections and, 7 dramatic reductions in deployed nuclear forces of, 17, Trade sanctions. See Sanctions against suppliers 99-1oo Transfers of expertise, 76,90-91, 112 environmental effects of weapon of mass destruction Treaty for the Prohibition of Nuclear Weapons in Latin programs, 72-73 America. See Treaty of Tlatelolco implications of proliferation for political-militay policies, Treaty of Rarotonga, 22-25,85 4, 11-12,73-75 Treaty of Tlatelolco, 22-25,85, 101, 104 importance of in reducing international role of weapons, Treaty on the Non-Proliferation of Nuclear Weapons. See 106 Nuclear Non-Proliferation Treaty military operations, 11-12,74 Trends favoring nonproliferation, 17-19 positive security assurances, 105-106 Trends fostering proliferation, 15-17 U.S.-foreign bilateral export control agreements, 89 weapon of mass destruction programs in, 12, 17, 63 Ukraine. See also Former Soviet republics United States legislation. See also specific legislation by control of weapons in, 111 name exports of critical information, equipment, or materials, 76 export controls, 86-87 expoxts of missiles, 41 institutional bases of nonproliferation regimes, 84-85 International Science and Technology Centers, 112 sanctions against proliferant countries, 95 reduction of security threats, 108 sanctions against suppliers, 90 risks from breakup of Soviet Union, 4, 15,75-77,111-112 Universal policy approaches, 31-32 seizure of Soviet weapons by non-Russian authorities, Uranium, material production and, 33-34 75-76 U.S. Global Positioning System, 42, 52 weapons of mass destruction in, 12, 63-65 Uzbekistan, 77 U.N. General Assembly, 105 U.N. Resolution 687,93,97 Verification regimes U.N. Security Council costs and benefits of inspections for, 113-115 international security arrangements, 108-109 international treaties and, 22-25 Iraq cease-fire agreement, 14, 18,91,92-93 monitoring proliferation and veri@ng compliance with on military response to proliferation threats, 20, 92 a~ements, 6-7, 28 multilateral agreement enforcement, 20-21, 94 onsite veri.ilcation measures, 105 positive security assurances, 105 technical basis for monitoring and controlling prolifera- revitalization of, 5 tion, 3243 role of permanent members in influencing the perceived Vietnam, 65-66 value of nuclear weapons, 99 role of permanent members in conventional arms sales, Weapons of mass destruction. See also Biological weapons; 107 Chemical weapons; Delivery systems; Nuclear weap- U.N. Special Commission on Iraq, 93 Unilateral approach to policy v. International Approaches, com~~ed with conventional weapons, 1, 29-30, 46 30-31 ease of acquisition, 9 Unilateral export controls, 86-87 exports from former Soviet Union, 76-77 United Kingdom, 12,63 military utility of, 55-61 positive security assurances from, 105 probability of use, 9-11 United Nations. See U.N. General Assembly, U.N. Resolu- Weapons of Mass Destruction Control Act, 85 tion 687, and U.N, Security Council United States Yemen, 67-69 alliances or coalitions, 75, 107 Cold War end and, 17-18 Zangger Committee. See Nuclear Exporters Committee defense commitments as nonproliferation incentive, 107