Aaron Karp

LESSONS OF IRANIAN MISSILE PROGRAMS FOR U.S. NONPROLIFERATION POLICY

by Aaron Karp

Aaron Karp is Senior Faculty Associate with the Graduate Programs in International Studies at Old Dominion University in Norfolk, Virginia. He is the author of Ballistic Missile Proliferation (Oxford University Press, 1996) and numerous articles that have appeared in such journals as International Security, Survival, and International Studies Quarterly.

ecent disclosures about tina, South Korea and Ukraine are Madeleine Albright, both Iran and Iran’s long-range ballistic less effective with remaining prolif- the United States have powerful ad- R 1 But flexibility missile program have dra- erators like North Korea, India and vocates of dialogue. matically reoriented debates over Pakistan. The second group has not is constrained by many factors, of how to deal with missile prolifera- responded to the same combination which Iran’s ballistic missile pro- tion threats. How this new challenge of export controls, sanctions, eco- gram is the most visible. Recent rev- is addressed depends partially on nomic incentives, and security assur- elations of the Shahab missiles technical assessments of Iranian ca- ances that worked so well with the hardly could have come at a worse pability. Even more salient are the first. These countries are more de- moment, just as Iran’s relations with political factors governing Iran’s termined and less amenable to out- the West are beginning to improve ability to acquire that technology. side influence. Iran’s proliferation after 19 years of confrontation. The immediate priority for counter- challenge is another one seemingly While Washington and Tehran take ing Iranian proliferation is finding immune to old nonproliferation for- tentative steps toward rapproche- ways to deal with exports from mulas. Additional sanctions—like ment, neither can expect the other China, Russia, and, to a lesser de- those in the 1998 Iran Missile Pro- to budge on essential premises of gree, North Korea. liferation Sanctions Act, may slow national policy. America will not In addition to the specific prob- Iran down somewhat, but by no more dilute its commitment to nonprolif- lem of Iran’s strategic ambitions, its than a few years eration and Iran will not compromise programs raise two general chal- Secondly, Iran raises the delicate on its strategic interests. lenges for U.S. foreign policy. First, issue of how to build better relations Iranian weapons proliferation is how should the United States handle in general with important regional all but guaranteed to constrain and the most intractable regional prolif- actors who refuse to cooperate on maybe prevent meaningful rap- erators? The mechanisms that key issues. In President Muhammad prochement. The problem is similar worked with countries like Argen- Khatami and Secretary of State to that faced by the United States in

The Nonproliferation Review/Spring-Summer 1998 17 Aaron Karp its efforts to improve relations with IRAN’S NEW SALIENCE Agni, with a new version currently countries like India and Pakistan America’s previous debate over being prepared for flight test, now where proliferation issues top the missile proliferation surrounded the clearly intended to carry nuclear 8 bilateral agenda. Despite its reputa- National Intelligence Estimate NIE- warheads. Pakistan has renewed its tion for revolutionary fervor, though, 95-19 “Emerging Threats to North own long-stymied missile program Iran may be the most flexible of all America During the Next 15 Years,” through the 1,500-km Ghauri, test- 9 of these countries. In both nonpro- issued in November 1995. Although fired on April 6, 1998. liferation and regional foreign this document dealt with global Missile proliferation is a global policy, Iran is emerging as a “test threats, the debate it catalyzed fo- problem, but the focus on Iran is jus- case” for developing new ap- cused almost entirely on the missile tified and useful nevertheless. Al- proaches to deal with these funda- projects of one country—North Ko- though we are accustomed to mental problems of American and rea.3 Since then, North Korea’s treating Iran as a revolutionary state, Western regional relations. In the projects have slowed, losing their an exception in every way, its mis- long run, there may be no alterna- erstwhile urgency. sile projects are typical in some re- tive to dealing directly with Iran it- spects of the technical and political self. Today’s debate over missile pro- liferation and missile defense is be- problems of regional long-range This article analyzes Iran’s mis- ing propelled instead by revelations rocketry in the late-1990s, illustrat- sile programs from the context of ing many key general issues for the over Iran.4 Once again, a regional U.S. nonproliferation policy. It be- rocketry program has become an in- future of missile proliferation. gins by discussing the increased sa- tellectual prism, diffracting issues lience of the Iranian threat on the that go far beyond immediate tech- INDIGENOUS CAPABILITIES U.S. nonproliferation agenda. It then nical matters. Of central importance Iran’s long-range rocketry and traces in detail the technical evolu- are broader questions regarding non- ballistic missile program has long tion of various missiles in the Ira- proliferation objectives, missile de- been one of the great enigmas in the nian arsenal, as well as their fense, and Middle East strategy. spectrum of global proliferation con- limitations. The article next consid- Equally salient are foreign policy cerns. Detailed information is spotty ers the important—yet often over- questions about relations with Iran, and general understanding weak. looked—factor of “soft technology,” engagement of difficult regional ac- What can be concluded is that Iran’s the human and organizational ob- tors, in general, and the prospects for ballistic missile program appears to stacles that have constrained Iran’s policies based on sanctions and ex- be roughly comparable to that of missile programs to date. Finally, port controls. Pakistan, but less advanced than the analysis turns to external factors North Korea and well behind India. and conditions within Iran’s poten- Missile programs elsewhere are Although Iran has been trying to ac- tial suppliers, particularly China and no less provocative. Iraq has aban- quire long-range missiles since the Russia, where new conditions of doned none of its long-range rock- mid-1970s, its efforts have met with growing pluralism are rendering old- etry ambitions, including plans for unprecedented frustration. No other style nonproliferation controls inef- a 3,000-km weapon, sufficient to country has tried so hard and fective. The article concludes by reach Paris.5 Saddam is only wait- achieved so little. providing guidelines for a new U.S. ing for a second chance to scour the nonproliferation strategy regarding international market for useful ma- Since starting under Shah Iran. Like other intractable actors, terials. North Korea’s Taepo-dong Mohammed Reza Pahlavi, Iran has Iran must be addressed through more projects may have subsided, but it created a broad missile development than new sanctions alone. Ulti- appears to have gone back to con- infrastructure.10 Its efforts can be mately, there may be no alternative centrate on the Nodong series.6 divided into four basic programs: a to creating a more balanced politi- Egypt appears to be receiving new solid fuel effort that led only to ar- cal relationship, leading to a tailored installments of technical assistance tillery rockets, a stillborn Scud- “packaged deal” combining new from North Korea, probably support- based program, a cruise missile nonproliferation instruments with ing improved Scuds.7 India contin- program, and most recently a re- economic and political initiatives.2 ues development of its 2,500-km newed intermediate-range ballistic

18 The Nonproliferation Review/Spring-Summer 1998 Aaron Karp missile (IRBM) program. doubtedly has tried to scale-up its many enigmas of Iranian rocketry is Even after some 25 years of work, solid motors, but without visible re- its failure to use this Scud produc- Iran’s greatest strength is manufac- sult. Although the general recipes for tion capability. The most likely ex- turing small-diameter solid motors, advanced solid fuels are well known, planation is that North Korea cannot best suited for use in unguided artil- the manufacture of large fuel grains supply essential manufacturing ma- lery rockets. It has been extremely requires considerable chemical and terials. difficult for Iranian engineers and manufacturing finesse. The Soviet At one time, Iran was interested chemists to move beyond primitive Union required 25 years to master in the North Korean Nodong, usu- double-based fuels to the more mod- this technology, and China took even ally described as a 1,000- to 1,300- ern composite fuels better suited for longer. Iran will require substantial km weapon based on Scud larger weapons. Solid motor experi- foreign assistance to make faster technology. Iranian officials were ence may be based on Chinese as- progress. present for the first flight of the sistance, but the Chinese role appears Nodong in May 1993. Development to have been exaggerated in many THE SCUD DIVERSION of the Nodong slowed to a crawl accounts. Although China obviously Iran’s efforts to develop liquid thereafter, roughly coinciding with is the source of Iran’s cruise missile propulsion have been no more suc- negotiation of the October 1994 capability, none of its large artillery cessful, limited until recently by the Agreed Framework, which halted rockets closely resemble any known weaknesses of its leading technical Pyongyang’s nuclear program. It Chinese counterpart. A major ques- ally, North Korea. After modest also was at this time that North tion is whether Chinese firms have transfers from Syria, Scud missiles Korea’s much more ambitious helped Iran move beyond their frus- and manufacturing wherewithal Taepo-dong missile project was dis- trations with double-based propel- from North Korea began flowing covered, raising the possibility that lants. into the country in 1985. Iran fired Pyongyang had traded a weak The solid fuel program run by roughly 110 of them against Bagh- project for one based on different Iran’s Missile Industries Group has dad during the 1988 War of the Cit- concepts and technology. Whether unveiled numerous prototypes, none ies, seriously frightening the technical problems or a political de- completely satisfactory. Time and Baathist leadership, but not altering cision stymied the Nodong, Iran was again, Iranian engineers have un- the course of the war.12 Others were compelled to find an alternative. veiled new artillery rockets; their fired at anti-Iranian revolution Japanese sources now say that projects betray poor design, excep- Mujahideen forces as recently as North Korea has gone back to the tional redundancy, and limited de- 1994.13 Scud development was an Nodong. This was partially con- ployment at best. Despite its obvious route for Iran; the missile firmed by reports that the Ghauri frustrations, Iran today is a world can be reverse engineered for pro- missile launched by Pakistan in leader in the development of large duction, readily stretched to ranges April actually was an imported artillery rockets, with ranges of 40 of approximately 1,000 km, and Nodong.16 If so, Iran may turn to to 200 km.11 These might not indi- adapted to many specific uses. North Korea again. But there are vidually be very effective delivery North Korea had no inhibitions serious limits to North Korean as- systems, but any technical weakness about helping Iran, but its support sistance. North Korea is even more could be compensated by their low has been disappointing. Iran’s inven- isolated than Iran, cut off even from cost. Deploying large numbers or tory of approximately 210 Scuds imports of raw materials and com- arming them with chemical or bio- apparently has not risen since the last ponents, and its engineering skills logical agents could compensate for consignment of North Korean Scud- are little better. In some respects, its any technical weakness and saturate Cs arrived in the early 1990s.14 Af- technology is inferior to Iran’s—es- any defensive system. ter building a Scud facility with pecially in solid propulsion—giving Iran strong incentives to look else- Despite their significant tactical North Korean help, according to the 17 potential, even the largest artillery U.S. Defense Department, “the Ira- where for assistance. rockets, however, will not serve Ira- nians are now able to produce the nian strategic priorities. Iran un- missile themselves.”15 One of the

The Nonproliferation Review/Spring-Summer 1998 19 Aaron Karp

A CRUISE MISSILE Iran leaves no doubt that it contin- (TMD) system, and from Iranian ex- ALTERNATIVE? ues to view them primarily as naval patriots trying to mobilize opposition With much less visibility than its weapons. As the ranges of follow- to the Islamic government. While ballistic missile projects, Iran also on systems grow and new guidance both sources may exaggerate aspects has been at the forefront of regional technologies are perfected, land-at- of the two new weapons, there is a efforts to develop cruise missiles. So tack missions will become more growing consensus on their general far, its efforts in this area have been likely. A new generation of cruise characteristics. restrained by lack of advanced tech- missiles—such as stretched Styx/ Shahab-3 is reported to be a solid- nology, weak aerodynamic infra- Silkworm versions redesigned for fuel weapon capable of carrying a structure, and a low funding priority. air-breathing propulsion—could 700-kg payload to a range of 1,300 Yet Iranian interest in cruise missile give Iran weapons effective to ranges km. It appears to be totally new Ira- options is unmistakable and, given of approximately 600 to 800 km. nian design. Its characteristics do not the growing availability of relevant Without massive foreign assis- match any known Russian or Chi- dual-use technology like global po- tance, though, Iran will find it ex- nese weapon. The closest counter- sitioning satellites (GPS) for guid- tremely difficult to develop cruise parts are the larger Soviet SS-22 and ance, further progress seems missiles able to reach farther within the smaller SS-23, both eliminated extremely likely.18 the next 15 years. To do so would under the 1987 Intermediate Nuclear require granting much higher na- Forces (INF) Treaty, although Bul- Iran reportedly is extending the range and accuracy of Chinese-sup- tional priority to cruise missiles and garia and Slovakia still have small plied HY-1 Silkworms. Based on levels of financial and engineering numbers of SS-23s. The closest Chi- readily mastered technology from commitment so far reserved for bal- nese counterpart was the DF-25, a the 1950s, the Silkworm and its Rus- listic missiles and other weapons, larger (1,700-km range) solid-fuel like armored vehicles. missile under development in the sian counterpart (the Styx) are to cruise missiles what the Scud is to Cruise missiles must be taken se- mid-1980s. The latter reportedly was ballistic missiles: cheap, easily re- riously by Western planners, even cancelled in the mid-1990s for lack verse-engineered, and well suited to if they are unlikely to overcome the of a requirement from the Chinese People’s Liberation Army.22 modification and mass production.19 range limitations of Scud-type weap- Although this rocket-propelled ons. In lieu of sudden technical The most likely possibility is that cruise missile has a designed range leaps, they are an ancillary system, the Shahab-3 is a domestic Iranian of only 80 km, Western experts be- essentially a low-tech, fallback op- design, but with extensive foreign lieve this can be extended to ranges tion. Other things being equal, Iran assistance to accelerated fuel cast- of 400 km and perhaps more. To will deploy them in conjunction with ing, casing, and nozzle fabrication. develop its cruise missile potential, Scud-type weapons, taking advan- According to Israeli reports, engine Iran can take advantage of poorly tage of their lower cost, greater ac- development for the Shahab-3 is regulated dual-use equipment. Its curacy and more effective delivery nearing completion and the missile Styx/Silkworm modification work is of some warheads, especially chemi- could be flight tested as early as believed to benefit from direct Rus- cal and biological weapons (CBW). 1999-2000, although more work un- sian assistance. Iran also has more doubtedly will be necessary to mas- advanced Chinese supplied C-801/ RISE OF THE SHAHABS ter the complicated problems of fuel 802 anti-ship weapons (resembling casting to insure predictable perfor- In the mid-1990s Iran started two the French Exocet and U.S. Har- mance. The status of other essential new projects to develop long-range poon), which could be modified to subsystems is unknown. ballistic missiles, the Shahab-3 and fly roughly 500 km.20 -4.21 The revelations about these two Shahab-4 appears to be a liquid- Despite their important advan- projects are not without credibility fuel weapon based on and possibly tages, cruise missiles are not yet al- problems. They come almost exclu- copied from the 1950s Soviet SS-4 ternatives to long-range ballistic sively from Israeli sources seeking Sandal (or R-12). The latter could missiles. By concentrating deploy- American financial support for the carry a payload of 1,360 kg to a ment of cruise missiles to its coasts, Arrow theater missile defense range of 1,940 km. Best-known for

20 The Nonproliferation Review/Spring-Summer 1998 Aaron Karp its role in the Cuban missile crisis, conceived, since essential compo- IRBM deployment. If the engine it too was banned by the INF Treaty. nents and sub-systems are unavoid- tests this winter were successful, a A complicated system, it required a ably absent or lag far behind. In lieu prototype flight article of the fleet of 20 vehicles to transport and of reliable suppliers, they rely on Shahab-4 could be ready within a launch.23 Although later simplified improvisation, progressing not year. An unproven, “stove-pipe” through use of semi-storable fuels, smoothly but in fits and jerks inter- weapon could be deployed within it never was easily transported nor rupted by long pauses. Iran will com- two years. A minimally reliable could its vulnerability overcome by plete these weapons if and when weapon (50 percent or less) based adapting it for silo launching. feasible; it is doubtful if project man- on approximately six flight tests The SS-4 might appeal not as a agers specify milestone dates in could be deployed in about five weapon to be deployed but as an terms more rigorous than periods of years. A high reliability weapon (75 educational vehicle, helping Iranian several years. percent or better) would require at engineers master techniques neces- With missiles, as with other weap- least 12 test flights and approxi- sary for other things in the future. ons, Iran resembles not so much a mately eight years to develop and Static engine tests undertaken by determined proliferator as what Brad deploy. Iran in late 1997 mostly likely in- Roberts calls a “dabbler,” pursuing If Iran benefits from sufficient volved a liquid motor (under favor- opportunities as they appear.24 This foreign assistance and a solid motor able conditions, liquid motors can be does not make its programs any of suitable diameter can be mastered, run over and over, while solid mo- easier to stop. Indeed, the opportu- the Shahab-3 could be deployed tors usually must be replaced com- nistic and improvised nature of slightly faster than its liquid-fuel pletely). The SS-4 was powered by Iran’s long-range rocketry makes it stable mate, since less flight-testing the 65,000-kg thrust RD-214, a typi- virtually impossible for outside pow- would be necessary. If it is ready for cal Soviet design of the period and a ers to stop it altogether. But its de- flight test in 1999-2000, an all-out close relative of the Scud’s engine. pendence on foreign technology push might make deployment of a Although its scale far surpasses any- makes it vulnerable to interruption. small quantity at minimum reliabil- thing in Iran’s previous experience, ity possible in 2001. More could be it would be a sensible choice for a HARD TECHNOLOGY deployed with higher reliability in country beginning to master long- 2003. After over 25 years of rocketry range rockets. It also is possible, but research, Iran has a considerable Guidance, the traditional bottle- less likely, that Iran’s liquid propul- technical endowment. The essential neck after propulsion, could be over- sion relies on North Korean technol- equipment to create long-range mis- come if specifications are relaxed to ogy. siles of mid- to late- 1950s vintage permit minimal accuracy. Although Neither Shahab is being managed appears to be largely in hand. If Iran inertial navigation systems (INS) are like a systematic American or NATO decides to give the program higher highly desirable, they are not a sine procurement program. Rather than national priority, making the leap qua non of ballistic missile devel- clear strategic goals or a rigorous from a dabbler to a determined opment. Simple gyro systems can be time schedule, they appear to be proliferator, it could have a long- adapted from other tactical missiles motivated by a general Iranian de- range system sufficiently advanced for ballistic missions. Used in con- sire to have long-range ballistic mis- for flight testing to begin within two junction with radio-command guid- siles and the availability of foreign years, although key technologies still ance, a missile can be flown to assistance. They are essentially op- would have to be mastered. The pac- intermediate ranges with sufficient portunistic undertakings, driven not ing technology no longer appears to accuracy to deliver a nuclear war- by strategy but by access to technol- be propulsion or guidance. It is more head. Although the system is vulner- ogy and resources. It is enough for likely to be warheads and re-entry able to jamming (signals can be Iranian leaders that the weapons vehicles. intercepted from test-flight telemetry could serve a variety of purposes and and countermeasures readied) adroit If minimal performance character- that their development is feasible. signal management can minimize istics are acceptable, there may no such risks. Neither project is systematically longer be a hardware barrier to

The Nonproliferation Review/Spring-Summer 1998 21 Aaron Karp

Warhead problems are much ing and systems integration. It would Eventually, the Iranian government harder to overcome. Despite the ob- have to acquire INS and a highly will recognize that the Missile Tech- vious strategic connection, Iran’s destructive warhead. Overcoming nology Control Regime (MTCR) is nuclear program does not appear to these problems would require at least not the only enemy of the Shahab-3 be accelerating to match its missiles. 15 years and is likely to be impos- and -4; each missile, rather, is the A recent study concludes that “Most sible without considerable foreign worst enemy of the other. It is when experts feel that Iran has all the ba- aid. we see Iran making a choice between sic technology to build a bomb, but them that we should become most only keeps a low to moderate level SOFT TECHNOLOGY alarmed. weapons design and development While selected imports might ac- effort.”25 The round of nuclear tests celerate development and improve Trained Personnel Shortages by India and Pakistan in May 1998 performance, the most difficult bar- Shortfalls of skilled personnel may inspire Iran to pursue its own riers facing Iran’s missile program limit Iran’s ability to absorb foreign nuclear option more aggressively, may have nothing to do with hard- technology, both dual-use and dedi- but the weaknesses of its program ware. They are in realm of so-called cated missile design and production still must be overcome. Unless its “soft technology”: decisionmaking, equipment. Iran’s missile program nuclear program receives massive management expertise, engineering appears to have competent engineers foreign assistance, possibly includ- skills, and finance.26 In rocketry, as on top, but its human resources ing the import of fissile material, Iran in other aspects of its military indus- grower weaker as one goes down the must rely on conventional explosives trial program, such as armored ve- chain of command. Design may be or CBW for many years to come. hicles and artillery, Iran has proven adequate, but development is poor, Iran probably has experimented with its ability to create new prototypes, system integration worse, and pro- chemical warheads for its Scuds, but but it has been less successful in duction extremely difficult. their destructiveness is difficult to bringing these designs to series pro- test and would be unknown until This problem undoubtedly is re- duction. Several factors are worth actual use. sponsible for Iran’s efforts to cre- examining. ate educational centers for missile Another question concerns re-en- training. Two Russian institutions try vehicles (RVs). Unless Iran has Problems in the Decisionmaking are heavily involved: the Bauman a foreign-developed re-entry system, Structure National Technical University in the range of its first IRBMs will be Poor decisionmaking has been Moscow and the Baltic State Tech- limited to approximately 1,000 to especially significant. Iranian politi- nical University in St. Petersburg 1,200 km. Greater ranges involve cal leaders have been unwilling or (before 1990, the Military Mechani- higher re-entry speeds and concomi- unable to streamline the nation’s cal Institute imeni Ustinova). The tant heating and instability, necessi- rocketry projects. Resources con- latter established a joint missile edu- tating more advanced RVs, which tinue to be divided between the cation center in Persepolis under the Iran is not capable of creating rap- Pasdaran (the Islamic Revolution- direction of Iran’s Sanam Industries idly by itself. ary Guard Corps), the Army, and the Group, an arm of the Iranian Defense As for ICBM potential, Iran faces Ministry of Defense.27 With roughly Industries Organization (DIO), enormous technical barriers. Devel- a dozen missile projects under way which reportedly directs the nation’s opment of an IRBM like the Shahab- simultaneously, resources cannot be solid-fuel rocketry program.28 3 or -4 is a vital stepping-stone to concentrated for maximum effect. Closing down these training cen- larger systems, but ICBMs require Because of their unprecedented scale ters must be a high priority, but it far greater levels of support and en- as development projects, this prob- may be difficult. Russia has under- tirely new technologies. Iran would lem will be most severe for the taken symbolic steps, like expelling have to master much higher perfor- Shahab-3 and -4. in November 1997 a highly visible mance specifications and the com- The critical decision for Iran may Iranian caught trying to buy missile pletely foreign engineering problems be to pick a clear preference, some- technology.29 But institutional coop- of engine clustering, multiple stag- thing it has avoided historically. eration continues. Until this can be

22 The Nonproliferation Review/Spring-Summer 1998 Aaron Karp

halted, efforts must be made to as- Unless new funding or new elas- which our American colleagues have sure that theoretical training stays to ticity in government spending can be so nicely informed us of have been a minimum, that the new centers do found, budgets are unlikely to per- considered and we have provided not become vehicles for dispatching mit anything beyond prototypes and detailed explanations.”34 U.S. offi- Russian rocketry and missile ex- very limited deployment of new cials and press revelations leave no perts, and that they do not facilitate weapons. Financial problems will doubt that Russian technical assis- transfers of MTCR-controlled not halt Iranian rocketry. But unless tance continues.35 Although Russian equipment. budgets rise swiftly and consistently, officials are more outspoken, they When bringing Russian experts to finances will slow technical are following a pattern set by China Iran is too sensitive, it still may be progress, delaying deployments by in its aid to Iran and Pakistan. feasible for Iran to subcontract spe- several years. The Chinese and Russian pattern cific tasks to Russian groups for as- of “transfer/deny/pledge but con- sistance. For example, Iran’s Shahid THE NEW POLITICS OF tinue” can be explained in three dif- Hemmat Industrial Group reportedly TECHNOLOGY SUPPLIERS ferent ways, each based on a has enlisted the services of the Rus- After the choices Iran itself different view of the nature of their sian Central Aerohyrodynamic Insti- makes, the next most important national governments. These alter- tute for missile design and questions are the export policies of natives point to the conclusion that development support.30 China and Russia. They are Iran’s nonproliferation policy in Beijing In addition to Russia, there are most important technology suppli- and Moscow is determined not by other sources for human assistance. ers, the only ones capable of trans- official decisionmaking, but by the North Korea already has supported forming Iran into a long-range deteriorating control of the govern- Iran’s missile program and created missile power. Both clearly are sup- ments themselves. much of its Scud missile infrastruc- plying equipment, development as- Proliferation policy has become ture. Transfers probably include par- sistance, and manufacturing part of larger constitutional question tial engineering for the still facilities. Both also are extremely about the future of the Chinese and incomplete Nodong. There still may difficult for Washington to deal with, the Russian states. Does the central be North Korean missile-engineer- sharing a well-known pattern of of- government still call the shots, or do ing cadres there. More support could ficially denying involvement, even- individual exporters? The alternate come following the collapse of the tually promising to halt further explanations for decisionmaking in North Korean state.31 This could re- assistance, but continuing nonethe- China and Russia can be summa- lease a flood of scientists and engi- less. rized according to three possible neers unmatched since the collapse A typical cycle of this pattern be- models: formal authorization, infor- of the Soviet Union in 1991. gan in January 1997 when Israeli mal acquiescence, or contraventions sources revealed that Russia was aid- of policy. Cost ing Iran’s missile program. Russian Finances will remain a serious officials initially denied any involve- Formal Authorization problem for years to come. Iranian ment. Later, they claimed that Ira- China or Russia may be transfer- military spending fell from an esti- nian efforts to acquire Russian ring ballistic missile technology with mated $11.5 billion in 1986 to $3 technological assistance had been explicit and formal authorization billion in 1994 (in constant 1994 thwarted. On October 26, 1997, For- from their central governments, U.S. dollars). Even after rebounding eign Minister Evgeniy Primakov which may even positively encour- to $3.8 billion in 1996, there is vir- said that “There is no basis to ru- age such transfers. If so, their prom- tually no latitude for expensive new mors that the Iranians are receiving ises and assurances to the contrary missiles from Russia.”33 Perhaps the programs.32 Nor is it clear where are disinformation for foreign con- additional funds might come from. most revealing comment came from sumption. The implications are very The prospect of new oil deals with Yuri Koptev, director of disturbing for diplomacy, since the Western petroleum firms is offset by Glavkosmos, who dismissed the integrity of their leaders comes into the unstable price of oil. matter by saying “The 13 cases question.

The Nonproliferation Review/Spring-Summer 1998 23 Aaron Karp

On the other hand, even if there is and firms no longer respond to regu- flout state authority. The tough sanc- a yawning gap between declaratory lations and follow their own incli- tions against Russia mandated by the statements and action policy, the nations. Regional and economic 1998 Iran Missile Proliferation outlook for future control is not so interests have overwhelmed the abil- Sanctions Act, for example, prom- discouraging. Since the formal in- ity of the state to regulate economic ise greater international pain but stitutions of government remain ef- behavior. The weakness of the state little nonproliferation gain; Russia fective, strenuous diplomacy can may be uneven, affecting some sec- simply may lack the ability to act work. Pressure must be strong and tors more than others. This would effectively, regardless of its inten- expensive incentives may be neces- explain why Chinese missile policy tions. sary. The degree of interaction nec- continues to be virtually laissez-faire If the Chinese and Russian cen- essary to control proliferation may although its nuclear export control tral governments no longer run non- 38 have changed, but not the kind of have tightened considerably. proliferation policy, as increasingly interactions. Formal state-to-state If the state is essentially incom- appears to be the case, traditional diplomacy, both bilaterally and petent in nonproliferation and na- nonproliferation mechanisms like through multinational nonprolifera- tional leaders are all but irrelevant, the MTCR, sanctions, and embassy tion regimes alike, still hold the best there is little to be accomplished demarches probably have passed promise of success. through traditional diplomacy. In- their point of marginal returns. The stead, to be effective, measures must same can be said of more imagina- Informal Acquiescence deal directly with the immediate tive bilateral arrangements. Since The governments in Beijing and agents of proliferation, the offend- 1993, the United States has invested Moscow may be unsure of their na- ing firms and laboratories. Some of more than $470 million in the Rus- tional interests and unwilling to act these may be within reach of carrots sian space program. This keeps decisively.36 Their policies may be and sticks. Others have no export Russia’s manned space program essentially passive, based on a blind markets beyond proliferators like alive, but it has been irrelevant to eye, tolerating transfers that serve Iran; they have no Western business, missile-related exports.39 Efforts to some clear and immediate national no reason to cooperate with Wash- end missile-related exports by offer- interests—such as the welfare of key ington. The only solution may be ing compensation in the form of technological sectors—while poten- restructuring or closing them out- space launch contracts, while very tially harming only speculative and right. lucrative for Russian and Chinese long-term security interests.37 space launch agencies, also are un- likely to be effective, since they do If so, the statements and pledges CONCLUSION: not directly compensate missile of their officials may represent genu- IMPLICATIONS FOR U.S. makers and exporters.40 ine intentions, but they are not suf- POLICY ficient to make policy. In this case, The implications of these ques- Even if the MTCR cannot stop the interests of Western nonprolif- tions are so troublesome that West- Iranian ballistic missiles, it is essen- eration policy lie in inducing their ern officials typically avoid even tial to delaying Iranian progress. leaders to make decisions and regu- acknowledging them. Yet it would Meanwhile, existing nonprolifera- larly reiterate them, both in highly be misleading to continue ignoring tion policies, based on government- visible declarations and through of- the mounting evidence that neither to-government cooperation probably ficial regulations and inter-agency Russia nor China can be dealt with should be broadened to deal with the consultations to insure compliance in the traditional manner of ortho- new challenges. Restraining Iranian throughout the government. dox states. Formal regimes like the proliferation may require shifting MTCR and the Wassenaar Arrange- our focus from supplier countries to Contraventions of Policy ment for dual-use technology are individual firms, such as Russian designed to regulate interactions rocket engine-maker NPO Trud, The authority of central govern- between states. They are of dubious NPO Polyus, China Great Wall In- ments in Beijing and Moscow may utility when individual ministries, dustries, China Precision Engineer- have deteriorated to the point that firms, and laboratories are able to ing Corporation, and even Russian policy is unenforceable. Agencies

24 The Nonproliferation Review/Spring-Summer 1998 Aaron Karp

and Chinese government-run labo- creating a political relationship that 7 “U.S. Checking Sale of Missiles,” Interna- tional Herald Tribune, June 22-23, 1996, p. 1; ratories. If so, new mechanisms, tai- goes beyond the perennial issues of Barbara Starr, “CIA Discloses Source of Egyp- lored to deal not with governments proliferation and terrorism, permit- tian ‘Scud-B’ Parts,” Jane’s Defence Weekly, but with recalcitrant ministries and ting discussion of mutually accept- July 9, 1997, p. 5. 8 “India Ready for Lift-Off on Agni Missile firms, will be essential if today’s able solutions for regional security. Production,” Jane’s Defence Weekly, February flow of technology is to be signifi- Although it has not brought the rapid 25, 1998, p. 17. cantly restrained. success that many had hoped for, 9 Zahid Hussain, “Pakistani Missile Can Hit Delhi,” The Times (London), April 7, 1998. Unless sudden leaps can be made Secretary of States Albright’s effort 10 Joseph S. Bermudez, Jr., “Iran’s Missile Devel- opment,” in Harlan Jencks and William C. Pot- toward better Iranian-American re- to create a more balanced bilateral relationship with Iran offers a poten- ter, eds., The International Missile Bazaar lations—a distant hope to be sure— (Boulder: Westview, 1994), Chapter 3. TMD almost certainly will be tially worthwhile alternative to con- 11 A few have been at least partially successful, frontation. especial the Oghab and Nazeat series. Others needed to insure Middle East stabil- never reached operational service, including the ity and preserve American strategic It is tempting to see improved bi- Shahin-1 and -2 and the mysterious Mushak and Zelzal series. The most recently revealed was the options when Iranian IRBMs are lateral relations as a reward for Ira- NP-110, with an estimated range of 130 to 170 deployed. So long as Iran relies on nian concessions, but this would km, described in Bill Gertz, “China Joins Forces indigenous weapons, existing de- only undermine political prospects with Iran on Short-Range Missile,” The Wash- ington Times, June 17, 1997, p. A3. ployment plans for defensive pro- without affecting Iranian strategic 12 W. Seth Carus, Ballistic Missiles in Modern grams like Arrow, the Theater High intentions. Engaging Iran, develop- Conflict (Westport: Praeger, 1991), pp. 8-9, Altitude Area Defense (or its suc- ing a meaningful bilateral dialogue 37. 13 “Iran Fired...,” The Economist, November cessor), and Navy Theater Wide will strong enough to address delicate 12, 1994, p. 6. be adequate to meet Iran’s capabil- subjects like proliferation and rich 14 The Military Balance 1997/98 (London: Ox- ford and the IISS, October 1997), p. 126. ity as it emerges. But if Iran prepares enough to offer useful trade-offs, 15 U.S. Department of Defense, Proliferation: to purchase off-the-shelf weapons should not necessarily be a reward Threat and Response 1997 (Washington, D.C.: such as the Nodong, accelerated to Iran for giving up its strategic pro- U.S. Department of Defense, November 1997), p. 6. TMD deployment may be necessary. grams. Engagement, rather, may be 16 Tim Weiner, “U.S. Says North Korea Helped What must be appreciated more essential if the Iranian challenge is Develop New Pakistani Missile,” The New York widely than it is today, however, is to be controlled. Times, April 11, 1998, p. A3. 17 Reports that Iran’s solid fuel Shahab-3 missile that while TMD may be necessary, is “based on a North Korean design,” cannot be it will not solve the basic problem reconciled with North Korea’s minimal solid fuel expertise and probably should be dismissed. Cited created by Iranian missile ambitions. in Eric Schmitt, “Senate Passes Bill to Punish In the long term, neither export Giving Missile Help to Iran,” The New York Times, May 23, 1998, p. A3. controls nor defenses are likely to 18 W. Seth Carus, Cruise Missile Proliferation in be sufficient. No export control sys- the 1990s (Westport: Praeger and the Center for Strategic and International Studies, 1992); K. tem will stop all Iranian progress. No Scott McMahon and Dennis M. Gormley, Con- defense will stop all Iranian missiles. 1 Peter Sawchyn, “Gulf Policy Experts Call for trolling the Spread of Land-Attack Cruise Mis- As with other successful nonprolif- More Dialogue between US and Iran,” USIS siles (Marina del Ray: American Institute for Wireless File, April 22, 1998. Strategic Cooperation, 1995). eration efforts, control of Iranian 2 On the most important precedent for package 19 But they are not completely comparable. As an weapons of mass destruction and deals see Michael J. Mazarr, North Korea and anonymous reviewer of this article noted, “It is delivery systems almost certainly the Bomb (New York: St. Martin’s Press, 1995), marginally more difficult to produce the Volkhov pp. 226-231. (mod) engine used in this cruise missile than the will come only through a broader 3 Aaron Karp, “The New Politics of Missile Pro- Scud engine (due to finer tolerances for the shells relationship with the proliferator it- liferation,” Arms Control Today 26 (October and corrugates).” 20 Dennis M. Gormley, “Hedging against the self.41 This might take the form of 1996), pp. 10-14. 4 For example, see Ballistic Missile Threat from Cruise-Missile Threat,” Survival 40 (Spring additional “soccer” diplomacy and Iran, Hearing before the Military Research and 1998), pp. 97-98; Alan George, “Iran Uses other cultural exchanges, yielding to Development Subcommittee of the House Na- Styx Technology in Cruise-Missile Develop- tional Security Committee, November 5, 1997. ment,” Flight International, November 16, enhanced trade, and culminating in 5 “Iraq’s Program of Mass Destruction,” Fact 1994. the short term with a strategic dia- Sheet, The White House, November 19, 1997. 21 The first report came from Kenneth logue. The precise form is second- 6 “Japan Plans Ballistic Missile Defence Deci- Timmerman, The Iran Brief, September 9, 1996, sion,” Jane’s Defence Weekly, February 25, pp. 1-2, and October 1, 1996, p. 3, which refer to ary to the general problem of 1998, p. 3. the Zelzal-3. Another early and detailed report was

The Nonproliferation Review/Spring-Summer 1998 25 Aaron Karp

Barbara Opall, “Israelis Say Russia Aids Iran’s U.S. Army War College, Strategic Studies In- Quest for Missiles,” Defense News, February stitute, 1998), p. 12. 10, 1997, p. 1. 38 Symposium on “Export Controls in the 22 According to a report by SIPRI researcher People’s Republic of China,” The Monitor 3/4 Eric Arnett in Jane’s Defence Weekly, Decem- (Fall 1997/Winter 1988), pp. 17-34. ber 4, 1996. 39 Lisa Burgess, “U.S. Ties Business, Foreign 23 Steven Zaloga, “The Missiles of October: Policy,” Defense News, March 16, 1998, p. Soviet ballistic Missile Forces during the Cu- 12. In fairness it should be noted that non- ban Crisis,” Journal of Soviet Military Studies proliferation is but one of the objectives of 3 (June 1990), pp. 307-323. the NASA-Glavkosmos deal, which also facili- 24 Brad Roberts, “Rethinking N+1,” The National tated the withdrawal of Russian troops from Interest 51 (Spring 1988), p. 77. the Baltics and supports Russian involvement 25 Anthony H. Cordesman and Ahmed S. Hashim, in the U.S.-led international space station. Iran: Dilemmas of Dual Containment (Boulder: 40 Tim Weiner, “U.S. Weighs Deal to Halt Mis- Westview, 1997), p. 306. sile-Gear Sales by China,” The New York Times, 26 The concept of soft technology is developed March 19, 1998, p. A8; Bill Gertz, “Clinton more fully in Aaron Karp, Ballistic Missile Pro- Arms Official Cites Secrecy, Refuses to Com- liferation: the Politics and Technics (Oxford: ment on Missile Deal,” The Washington Times, Oxford University Press, 1996), pp. 51-98. March 19, 1998, p. A3. 27 Shahram Chubin, “Arms procurement in Iran,” 41 Mitchell Reiss, Bridled Ambition: Why Coun- in Eric Arnett, ed., Military Capacity and the Risk tries Constrain Their Nuclear Capabilities (Bal- of War (Oxford: Oxford University Press and timore: Johns Hopkins University Press, 1995). SIPRI, 1997), pp. 233-234. 28 Bill Gertz, “Russia Conspiring with Iran on Missiles,” The Washington Times, February 23, 1997, p. A1. 29 Michael R. Gordon, “Russians Order Expul- sion of Iranian Detained as a Missile Spy,” The New York Times, November 18, 1997, p. A10. 30 Bill Gertz, “Russia Disregards Pledge to Curb Iran Missile Output,” The Washington Times, May 22, 1997, p. A3. 31 Amos A. Jordan and Jae H. Ku, “Coping with North Korea,” The Washington Quarterly 21 (Winter 1998), pp. 33-34; Barbara Starr, “N Korean Government ‘Could Collapse by 2001,’” Jane’s Defence Weekly, January 7, 1998, p. 14. 32 World Military Expenditures and Arms Trans- fers 1995 (Washington, D.C.: U.S. Arms Control and Disarmament Agency, April 1996), p. 78; Cordesman and Hashim, Iran: Dilemmas of Dual Containment, pp. 170-175. 33 Primakov quoted in Dina Kraft, “Russian Minister Tells Israelis Moscow Is Not Arming Iran,” Washington Times, October 27, 1997, p. A11. 34 Koptev in Michael R. Gordon and Eric Schmitt, “Missile-Quality Russian Steel Was Nearly Smuggled to Iran,” The New York Times, April 25, 1998, p. A1. 35 CIA Director George Tenet quoted in Bill Gertz, “CIA Warns of Iran Missile Threat; Russian Technology Helps Speed Development, Tenet Says,” Washington Times, January 29, 1998, p. 1; Gordon and Schmitt, “Missile-Quality Rus- sian Steel Was Nearly Smuggled to Iran,” p. A1. 36 Andrew J. Pierre and Dmitri V. Trenin, eds., Russia in the World Arms Trade (Washington, D.C.: Carnegie Endowment, 1997). 37 Stephen J. Blank makes the case that Moscow, seeing no immediate threat of war from potential enemies like China, India or Iran for eight to 10 years, feels no inhibitions over selling them mili- tary technology. Blank, Russia’s Armed Forces on the Brink of Reform (Carlisle, Pennsylvania:

26 The Nonproliferation Review/Spring-Summer 1998