A. Q. Khan’s Nuclear Hubris

Christopher Clary is a research fellow at the Center for Contemporary Conflict, US Naval Postgraduate School, Monterey, California. The author’s views are his own, and do not necessarily represent those of the US Navy or the US government.

From 1976 to 2004, Dr Abdul Qadeer Khan was at the centre of the global nuclear black market. He was not the first to benefit from the illicit trade in destructive technologies, but he accelerated the consolidation of this marketplace. Khan was caught ultimately in an audacious attempt to provide Libya with a nuclear-weapons programme. Khan had agreed to provide Tripoli with nearly all of the infrastructure and expertise necessary to build a nuclear bomb: unenriched fissile material, enrichment technology, assistance in creating an enrichment facility, technical consultations, training of Libyan scientists and technicians, and a nuclear-weapons design. This was not quite an offer that virtually gifted Libya with the bomb; rather, Khan was working with the Libyans to construct a nuclear programme from the ground up. Khan’s network enabled Tripoli’s nuclear ambitions. Without Khan, the endeavour might have been too difficult to attempt. With Khan, it became possible—if still very challenging.

 

In Greek tragedy, the protagonist’s exaggerated self-confidence (hubris) results in retribution, and the character’s downfall. Khan’s Libyan effort was overreaching in its ambition. To deliver successfully on his offers, he would have to move large parts of his nuclear operation offshore. He would have to rely on old friends from Europe and new companies in Asia to create the critical nuclear components he had promised. He had to do these things overseas because of the magnitude of the Libyan order, and also because his own position at home had eroded since General Pervez Musharraf seized power in a military coup in 1999.

 

Khan was caught. On 4 October 2003, the German-flagged vessel, the BBC China, was diverted to port in Taranto, Italy, in an action co-ordinated by US, British, German, and Italian authorities. Five containers full of sensitive components—components that had been tracked from their point of origin by US and British intelligence agencies—were unloaded in Italy. The United States and Britain had been engaged for months in talks with Libya, seeking to get Tripoli to forgo its weapons of mass destruction (WMD) programmes. The interdiction of the BBC China, combined with other backchannel signals from US and British interlocutors, demonstrated to Libya that its nuclear effort had been heavily compromised. Two months after the interdiction of the BBC China, Libya made a strategic decision to dismantle verifiably, comprehensively, and irreversibly its WMD programmes. In doing so, Libya provided US and international investigators with significant new data that could be used to attack Khan’s network.

 

Khan’s Libyan effort was the exposed strand of his nuclear network: pulling on it unravelled the whole enterprise. Simultaneously, the International Atomic Energy Agency (IAEA) confronted Iran about its undeclared nuclear activities, and received significant new information about Khan’s assistance to Tehran. Together, the Iranian and Libyan revelations led Pakistan to arrest Khan and investigate his actions. Today, nearly three years after the BBC China was searched, we now know much about Khan’s enterprise. This article introduces the publicly available information on the A. Q. Khan nuclear-supplier network. It examines Khan’s career, exploring both his “day job” as director of a Pakistani nuclear-weapons laboratory and his “night job” as head of an illicit proliferation network. It concludes by assessing the significance of Khan’s nuclear network and offering policy recommendations for a post–A. Q. Khan world.

The Rise of Khan

Perhaps an Indian magazine was more correct than it could have known when it described Khan as a cross between Dr Strangelove and “an Islamic James Bond”.¹ Khan had been enmeshed in the European nuclear scene during the 1960s and 1970s. He studied at universities in Germany, the Netherlands, and Belgium. His professors and fellow graduate students were a helpful resource to Khan throughout the 1970s, though many of them cut contact with him after formal government investigations were launched. From 1972 to 1975, Khan worked for a Dutch sub-contractor to the Anglo-Dutch-German uranium enrichment consortium, URENCO.

 

By early autumn 1975, however, the Dutch authorities had grown concerned about the number of suspicious incidents involving Khan. He was removed from work on gas centrifuge development in October and, shortly thereafter, went home to Pakistan on vacation. He never returned to work at URENCO, and resigned his position in March 1976. Khan had been asked by Zulfikar Ali Bhutto, Pakistan’s prime minister, to stay in Pakistan and assist in the nuclear-weapons effort. He brought with him stolen centrifuge designs and, perhaps more importantly, a list of dozens of companies that supplied centrifuge parts and materials (centrifuges can separate enriched uranium, either to fuel a nuclear reactor or to build a nuclear bomb). After a brief stint with the Pakistan Atomic Energy Commission, he moved to the Engineering Research Laboratories, setting up a uranium-enrichment plant in Kahuta, near Islamabad. Within four years of returning home, his progress was significant enough to cause then-president Zia ul-Haq to rename the facility: Khan Research Laboratories (KRL) was born.

 

As head of KRL, A. Q. Khan would report directly to the president. Initially, KRL would be in charge of one component of the larger nuclear puzzle: enriching gaseous uranium hexafluoride into weapons-grade material. The rest of the process—from mining to yellowcake to gasification, and back again from gas to metal to milling and weapons fabrication—was under the control of the Pakistan Atomic Energy Commission. Both organisations competed fiercely for increased resources and enhanced responsibilities. As KRL pushed out, it inevitably bumped into the commission’s area of responsibility, raising the stakes of the bureaucratic infighting. Over time, KRL would gain a sanctioned role in developing delivery vehicles for the nuclear device. It also gained greater (and possibly unsanctioned) involvement in the machining, fabrication, and assembly of a nuclear device.

 

Almost immediately upon his return to Pakistan from the Netherlands, Khan began to gather as many components and as much information as he could from the network he had established during his decades abroad. He contacted former co-workers, inquiring about difficult technical processes and urging them to visit Pakistan, where he could arrange for technical consultations. Pakistani firms or embassy personnel contacted several Dutch firms about purchasing specialised components. Many of these goods were shipped to Pakistan, slipping through the porous export controls of several European countries.

 

This was part of a broader and clear Pakistani strategy. Khan later said, “I took full advantage of the willingness of western companies to do business and decided to make purchases from the open market.”² Khan and his network were working against time. The export-control system was initially ignorant of the threat and then lethargic in reacting against it. Pakistan’s supply of luck was large, but not limitless. People were starting to notice. As governments were pressed domestically and externally to control this trade in nuclear-sensitive goods, they slowly roused themselves to the task.

 

The supplier cartels were battling horizontally and vertically, i.e., against both multilateral measures and internal controls by individual states. Bilaterally and through multilateral organisations, states slowly harmonised export controls to prevent Pakistan and others from seeking and exploiting the weakest national regulations. Simultaneously, state regulators sought to impose control further and further down the chain of production. Initially, Pakistanis were buying entire systems, then they were buying sub-systems, then major components, then materials useful in engineering the components themselves. Their aim evidently was to assemble more and more of the equipment in Pakistan themselves.

 

Khan’s procurement network was paying important dividends. By the mid-1980s, within a decade of Khan leaving his URENCO offices for the last time, Pakistan had enough highly enriched uranium for a weapon.

Khan’s Nuclear Sales

Sometime in the mid-to-late 1980s, Khan appears to have diverted for his own gain the flow of nuclear supplies and know-how: now he was exporting, not just importing, these goods. He was still bringing in material and components for his nuclear enrichment process, but he seems to have been ordering more than Pakistan needed. At the same time, KRL was maturing. KRL scientists published papers starting in 1987 on constructing more difficult centrifuges of maraging steel, rather than the earlier aluminium-based designs (maraging steel is of superior strength and malleability). In 1991, KRL scientists published details of how to etch special grooves into the bottom bearing of the centrifuge to incorporate lubricants. Both trends—over-ordering and technological innovation—left Khan with excess inventory. An anonymous American official marvelled at the accomplishment: “First, he exploits a fragmented market and develops a quite advanced nuclear arsenal. Then he throws the switch, reverses the flow and figures out how to sell the whole kit, right down to the bomb designs, to some of the world’s worst governments.”³ This section introduces the available public information on Khan’s nuclear assistance (or offers) to Iran, Iraq, North Korea, and Libya.

Iran

The first country that Khan sold to was Iran. In 1987, three Iranian officials reportedly met several members of Khan’s network in Dubai, perhaps including an uncle–nephew team of Sri Lankan businessmen, Mohamed Farouq and Buhary Syed Abu Tahir, and a German engineer named Heinz Mebus.⁴ Tahir would gain international notoriety in 2003 when President George W. Bush called him the Khan network’s chief financial officer—though in 1987 he would have been fairly young. An Iranian exile group has claimed that one of the Iranian representatives was then–brigadier-general Mohammad Eslami, at the time in charge of the Iranian Revolutionary Guard’s research centre. If Eslami was present, it would indicate that even at this early stage the co-operation was viewed as having military utility.

 

Khan’s intermediaries apparently presented a one-page handwritten note outlining a five-point, phased nuclear-weapons development plan. Though this was Khan’s first offer, he apparently hit the ground running. According to the IAEA,

 

This document suggests that the offer included the delivery of: a disassembled sample [centrifuge] machine (including drawings, descriptions, and specifications for production); drawings, specifications and calculations for a “complete plant”; and materials for 2000 centrifuge machines. The document also reflects an offer to provide auxiliary vacuum and electric drive equipment and uranium reconversion and casting capabilities.⁵

 

Khan may also have provided Iran with information on circumventing existing export controls. According to one anonymous Pakistani source, “We confided in them about the items needed to construct a nuclear bomb, as well as the makes of equipment, the names of companies, the countries from which they could be procured and how they could be procured.”⁶ The Iranians may have outsmarted Khan, however. IAEA employees reportedly believe that Iran, using Khan’s document as a shopping list, instead went to European, Russian, and Chinese firms to purchase the equipment and technology at lower prices.⁷ Iran’s ability to continue to purchase from Western companies is a key indicator that efforts to improve export controls were only partially successful.

 

Even if the Iranians did not purchase Khan’s “package deal”, they apparently did buy centrifuges, designs, and centrifuge technology. Co-operation began in 1987, though Khan reportedly visited Iran’s Bushehr nuclear facility in February 1986. In 1987, in addition to inadvertently providing a shopping list, Khan apparently provided Iran with designs and sample components for the P-1 aluminium rotor centrifuge. Between 1994 and 1996, Iran received an apparently duplicate set of P-1 designs along with components for five hundred centrifuges.⁸ It seems that these components were from models that Pakistan had used previously to enrich uranium, perhaps explaining most—if not all—of the enriched-uranium contamination found on Iranian equipment.

 

Iran claimed that it had difficulty setting up the centrifuge cascades, and blamed this difficulty on poor-quality components. At least once, in 1997, Khan’s network replaced previously supplied bellows because of their inferior performance.

 

Also between 1994 and 1996, Iran received designs for the more advanced P-2 centrifuge, though Iran claims it did not work on this design until early 2002. Though the Pakistani P-2 uses maraging steel for the spinning rotors that separate different densities of uranium, Iran claims it had difficulty manufacturing those components. Instead, Iran attempted to use a “shorter, sub-critical carbon composite rotor”.⁹ Anonymous IAEA officials have been quoted in the press saying they also suspect that Iran received a nuclear-weapons design from the Khan network.

Iraq

While A. Q. Khan’s nuclear enterprise was making opening entreaties to Iran, it may have been trying to sell to Iran’s western neighbour as well. According to a 6 October 2024 Iraqi intelligence service memo obtained by IAEA investigators in 1995, a Khan intermediary contacted Iraq “regarding the possibility of helping Iraq establish a project to enrich Uranium and manufacture a nuclear weapon”. According to the memo, Khan was prepared to give Iraq project designs for a nuclear bomb and to ensure the delivery of any requirements or materials from western European countries via a company he owned in Dubai. Iraqi intelligence believed Khan’s motive to be the gaining of profits for him and the intermediary.¹⁰

 

In related documents, Iraq notes that there was an upfront cost of $5 million as well as a 10 per cent overhead on all materials acquired by the proliferation network. Iraq apparently thought the opportunity was too good to be true and that it was an elaborate trap by the United States, using its sometimes ally Pakistan.

North Korea

Pakistani officials have discussed, both in the press and with US officials, the nature of Khan’s nuclear assistance to North Korea. Khan, in a signed statement, reportedly accepted responsibility for “supplying old and discarded centrifuge and enrichment machines together with sets of drawings, sketches, technical data and depleted hexafluoride (UF6) gas to North Korea”.¹¹ Khan may also have provided North Korea with a “shopping list” of all the equipment necessary to produce the machines.

 

The timing of the co-operation is somewhat uncertain. Third-hand reports—Khan supposedly told Pakistani investigators who then informed US officials who then leaked the story to the press—have said that Khan first approached North Korea in the late 1980s, but did not begin major shipments until the late 1990s. This coincides with Pakistani statements that the first orders “were placed for the production of components for centrifuge machines” in 1997, with the first shipments occurring a year later.¹²

 

In September 2005, President Musharraf said that Khan exported “probably a dozen” centrifuges to North Korea. He also claimed there was no evidence that Khan had passed “bomb designs to others” besides the Libyans.¹³ A dozen centrifuges would have been insufficient to produce enough highly enriched uranium for a nuclear bomb. However, they could have been used as a template from which copies were made. Musharraf also indicated that Khan might have sent uranium hexafluoride to North Korea.

Libya

In 1997, Khan launched his most ambitious programme of co-operation, with Libya. Unlike Iran, Iraq, or North Korea, Libya had a limited indigenous nuclear infrastructure. In the early to mid-1980s, Libya had shopped around with European, Soviet, and Japanese suppliers for a uranium conversion facility, and eventually received a modular pilot-scale facility from a Japanese firm in 1986. From the late 1970s until the mid-1980s, Libya also received nuclear material, including over two thousand metric tons of uranium yellowcake and relatively small quantities of uranium hexafluoride, and equipment and training from European firms and the Soviet Union. The programme lay relatively dormant throughout the late 1980s, but in July 1995, according to the IAEA, “Libya made the strategic decision to reinvigorate its nuclear activities, including gas centrifuge enrichment.”¹⁴

 

In late 2003, A. Q. Khan’s intermediary, Buhary Syed Abu Tahir, a resident of Malaysia, was questioned by local police after a tip-off from the United States and Britain. He told the police that in 1997, he and Khan met two Libyans in Istanbul who asked Khan to supply centrifuge units to Libya’s nuclear programme. Starting that year, Libya imported twenty complete L-1 aluminium centrifuges from Khan’s network, along with most of the components for an additional two hundred L-1 centrifuges. Significantly, the network apparently was unable or unwilling to provide the aluminium rotors and magnets necessary for these two hundred unassembled units. At least one of these centrifuges had been used previously in Pakistan until 1987. In 2000, Libya imported two test L-2 maraging steel centrifuges from Pakistan. Both of these centrifuges had been used in the Pakistani nuclear programme, and both were contaminated with particles of highly enriched uranium. Libya placed an order for ten thousand additional L-2 centrifuges, with the first deliveries arriving in December 2002.

 

Khan’s network apparently also transferred a total of 1.87 tons of uranium hexafluoride to Libya. Tahir told Malaysian police that a “certain amount” of uranium hexafluoride was shipped onboard a Pakistani plane to Libya in 2001. The fact that Khan was able to transfer nearly two tons of uranium hexafluoride in 2001 would seem to indicate there were serious weaknesses in Pakistan’s material protection, control, and accounting practices—if the uranium came from Pakistan. If the uranium came from North Korea, it would raise questions about the responsible nuclear stewardship of the Pyongyang regime.

 

The complexity—and audacity—of the Libyan centrifuge order has been captured by David Albright, who notes that ten thousand centrifuges with one hundred components each means that a supplier network would have to procure or manufacture over a million components and ship them all to Libya.¹⁵ Some of these components must have been difficult for Khan to procure by his traditional means. After all, in the Iranian, Iraqi, and North Korean cases, Khan had only supplied designs, a few hundred used components, and perhaps quantities of uranium hexafluoride. Libya’s order constituted a problem that was immensely more difficult. Khan and Tahir responded by turning the existing front companies and procurement vehicles into more robust organisations capable of training foreign scientists and manufacturing certain products. Khan’s biggest innovation—and his downfall—was to establish factories in Libya and third-party states to procure, assemble, and manufacture the necessary components and resources.

 

Workshops in Turkey served as European mini-hubs, from which Khan’s network could procure and deliver centrifuge motors, power supplies, and ring magnets from partially within the web of pan-European export controls. Importing subcomponents from Europe and elsewhere, these facilities assembled the motors and frequency converters necessary to spin the centrifuges at the high speeds required to separate different uranium isotopes. Interestingly, a shipment of these components was sent with false end-user certificates to Dubai, and was placed aboard a German-owned vessel, the BBC China, en route to Libya. Onboard, that shipment joined a larger consignment of centrifuge components from Malaysia that would gain notoriety when seized by Western governments in Taranto, Italy, in October 2003. Highlighting the difficulties of interdiction, authorities removed the shipment from Malaysia, while the components from Turkey proceeded to Libya. Libyan authorities were kind enough to hand over the nuclear-related items to international inspectors upon their arrival in Tripoli.

 

In South Africa, Khan’s network was able to draw upon firms and individuals with connections to the now-defunct South African nuclear programme. This latent nuclear knowledge was retained even after South Africa made a strategic decision to abandon its nuclear-weapons programme in the early 1990s. Apparently, Gerhard Wisser, a German national living in South Africa, stumbled into the Khan network. He met Buhary Syed Abu Tahir at a dinner party in Dubai in 1999 and was offered a lucrative commission if he could arrange for the manufacture of “certain pipe work systems”.

 

Wisser had previously done work for the South African nuclear programme, and had also supplied vacuum pumps and other equipment to Pakistan in the 1980s. He contacted an old business associate of his, Johan Meyer, who owned a South African engineering firm, Tradefin, and had also previously worked for the South African nuclear programme. Wisser and Meyer set about creating a massive steel system to feed and withdraw uranium hexafluoride gas into a centrifuge cascade. The massive system, referred to as “the beast” by Meyer, would have been two storeys in height and filled eleven 40-foot shipping containers. Wisser and Meyer also attempted and failed to produce maraging steel rotors for the L-2 centrifuge. They received a specialised lathe from Gulf Technical Industries in Dubai in late 2000, but were unable to acquire the maraging steel necessary to make the rotors. As a result, they returned the lathe to Gulf Technical Industries in December 2001. The lathe was later discovered in Libya.

 

The most publicised facility, however, was located in Shah Alam, Malaysia. The factory, established in 2001, employed only about thirty people. The plant was operated by Scomi Precision Engineering (SCOPE), a subsidiary of Scomi Group Berhad, a Malaysian oil and gas firm. Scomi claims with some credibility that it was unaware it had become part of the nuclear black market. In April 2002, Urs Tinner, son of a longtime Khan associate Friedrich Tinner, began working as a consultant for SCOPE’s factory at Tahir’s request. Tinner arranged for the importation of lathes as well as cutting, turning, and grinding machines. The company made progress in machining some of the components necessary for a centrifuge. Between December 2002 and August 2003, fourteen types of component were manufactured and shipped to Dubai. The capabilities of the plant, however, should not be overstated. After all, it manufactured only fourteen of the approximately one hundred types of component in a centrifuge. As the Malaysian police bluntly state, “As of now, no factory in Malaysia is capable of manufacturing a complete centrifuge unit,” let alone “the construction of hundreds or thousands of centrifuges.”¹⁶

 

There were limits to what the network could acquire abroad. Khan decided to establish a facility within Libya to manufacture components difficult to procure elsewhere, and also to repair centrifuges that were damaged in the development and operation of an enrichment plant. Two British nationals, Peter Griffin and his son Paul, were allegedly involved in the establishment of a workshop called Project Machine Shop 1001. They were accused of purchasing and delivering furnaces and lathes to assist in the manufacture of centrifuge components, and of arranging training in Europe for Libyan personnel. The Griffin family has denied any role in the endeavour, and Peter Griffin successfully sued the Guardian newspaper under Britain’s libel laws.

 

After the interdiction of the BBC China, it became apparent that the Khan network had been compromised. Individuals associated with the network began a mad scramble to destroy evidence of wrongdoing, or possibly to sell it quickly to other interested customers. Centrifuge components, precision tools and parts for lathes, and perhaps seven valuable maraging steel rotors disappeared in 2003, as the network was collapsing. They may have been sold, or held in reserve for a “rainy day”, in the words of IAEA Director-General Mohammad ElBaradei, by members or states associated with the network.¹⁷

 

While the network of non-state actors is the most analytically interesting portion of the Libya story, it may not be the most significant aspect of Khan’s co-operation with Libya. Libya received nuclear-weapons designs from the Khan network—supposedly as “a bonus”—which were transferred inside an Islamabad dry-cleaner bag. Khan reportedly delivered the warhead plans to Libya in late 2001 or 2002. Many of the blueprints, designs, sketches and instructions found in Libya appear to have been copies of copies of copies. If the copies were passed on through middlemen, control of the information may have been irrevocably lost. As one European diplomat remarked, “There is no limitation on a copy machine.”¹⁸

How Much Harm Did Khan Do?

Khan was not selling nuclear weapons. None of his clients has successfully constructed nuclear explosives based on his assistance. What Khan did was to shorten timelines, perhaps dramatically. In the Iranian and North Korean cases, he provided seed technology. He probably could not have acquired centrifuge technology for Pakistan without information from URENCO and other European firms. He passed on that URENCO-based technology to Iran, North Korea, and Libya. These states may have developed centrifuge technology without him. After all, Brazil developed a centrifuge effort quite independently of Khan. He almost certainly shortened the research and development periods of his clients, probably by years rather than months. These states did not have to go through the trial-and-error process of centrifuge design. Instead, they had a validated design that they could seek to duplicate and manufacture on a mass scale.

 

Even successful construction of a centrifuge—or a hundred, or a thousand centrifuges—is only one crucial part of the puzzle. First, uranium hexafluoride must be created by the ton. There is no evidence that Iran, North Korea, or Libya had mastered this step, though Iran and Libya had previously acquired non-Khan foreign assistance on a uranium conversion facility. Second, centrifuges must be constructed into a large cascade of machines, slowly enriching U-235 from U-238. These cascades must operate for months to create sufficient material for a nuclear device. Their operation draws huge quantities of power from the electricity grid. Flawed centrifuges may break, spinning at thousands of rotations per minute, hurling shrapnel into the cascade. Poor-quality uranium hexafluoride will slow the cascade in its functioning. The cascade may also be discovered by outside groups, as occurred with the nascent Iranian facility at Natanz. Finally, once a state has acquired sufficient quantities of fissile material, the latter must be reconverted into uranium metal and machined into a nuclear explosive device. Khan apparently provided, or at least offered, assistance on both reconversion and nuclear-weapons design. None of his clients appears to have gone this far down the nuclear path.

Lessons to Be Learned

This article has highlighted a number of policy challenges that will confront decision-makers for some time. First, the technological piece of the non-proliferation puzzle is growing more complicated. Precision manufacturing capabilities are available in more and more countries. Information about controlled technologies is also more widely obtainable. Globalisation eases travel and commerce. Specialists from the developed world can travel to the developing world with blueprints on a hard disk. They might programme a lathe for a few days, and then return home to their cottage for the weekend. A firm in Malaysia could and did manufacture for a subsidiary in Dubai parts that turned out to be controlled centrifuge components. The subsidiary in Dubai did not know the ultimate consumer for the products, only a front company. The shipping firm carrying the parts from Dubai to Tripoli did not realise it had sensitive cargo. The intelligence agencies that pulled the containers off the ship did not realise there were other containers on the same ship, carrying sensitive components from a Turkish company that were also destined for the same Libyan bomb programme.

 

Globalisation and technological diffusion have complicated the problem, but the old problem has not gone away. European firms and individuals are still selling proscribed technologies to nuclear programmes. This is partly because of increased sophistication by client states—using front companies, false end-user certificates, and other deception techniques. It is partly because of the lethargy of bureaucracies and reliance on out-of-date export-control lists. And it is also partly because high profits can be made from selling controlled technologies.

 

Governments that learn of illicit procurement are confronted with a policy challenge: do they watch or do they act? There is a tension between action and observation. In other words, policymakers aware of the transfer of WMD technology, expertise, or material have the choice of watching that transfer to gain a better understanding of both the suppliers and consumers, or of moving to prevent or interdict the transaction. The principal challenge is that premature action is unlikely to have a decisive effect on either a proliferation network or a procurement effort. Greater understanding of the structure of a network—achievable only through watching—is often necessary to impair its function. Of course, waiting too long could allow a potential threat to mature, risking the security of the United States and other countries.

 

Taking action is also challenging. Intelligence collected about an unscrupulous merchant, for instance, has to be converted into something usable. If the merchant is in the United States, the intelligence might have to meet legal standards, or lead to other evidence that does meet such standards. This might delay action for months. If the merchant operates in a foreign country, the problem is complicated further. Will sharing intelligence with a foreign government endanger the source? Will that government take action? Will that government value its own commercial interests over a hypothetical risk, particularly a risk that is likely to be felt more acutely by the United States than the foreign government?

 

This article has largely ignored the issue of Pakistani state complicity because of the complexity of that issue. Nevertheless, assumptions about culpability have significant implications. Some policymakers intuitively divide the world into good and bad states. Good states do good things and bad states do bad things. Many analysts regard Pakistan as an archetypical “bad state”, admittedly one that gets away with it. The policy recommendations that flow from that conclusion concern changing the nature of the state. They might be punitive, hoping to raise the costs of bad behaviour. For instance, many believe Pakistan should be sanctioned for its proliferation record, particularly if US investigators are denied access to A. Q. Khan. Alternative policy proposals might seek to reform Pakistan, hoping to change the nature of the state from the inside. Most Indian analysts and many others believe that Pakistan’s behaviour will not conform to global norms so long as the military dominates Pakistani politics. These same analysts blame the military for Pakistan’s dismal proliferation record.

 

There is also evidence that we live in a less Manichaean world. If Khan’s network flourished because of failures of Pakistani command and control, the policy solution is to remedy those deficiencies. Policymakers must look more closely at assisting new nuclear states with their safety and security arrangements. This will be tricky. New nuclear states, which prize secrecy so much, are unlikely to share extensive data with outsiders, particularly outsiders who might be direct threats to new nuclear programmes. Assistance with personnel reliability programmes may provide new nuclear states with greater certainty about their positive and negative control. Hypothetically, such confidence might lead states to deploy their nuclear forces at a higher state of readiness. Given the significant other costs of such deployment, however, it seems unlikely that concerns about personnel are the only obstacle to deployed and ready nuclear arsenals. This potential risk must be weighed against the proven risk of what insiders can do when they operate outside state oversight.

The Next A. Q. Khan?

The A. Q. Khan affair was a significant failure for all involved. The Pakistani government provided Khan too much authority, had minimal nuclear oversight, and was slow to react to internal and external warnings that Khan was up to no good. The United States was too slow in realising the growing danger posed by Khan’s activities, despite having intimations of his nuclear trade by the early 1990s. Europe was too slow in policing its own individuals and firms, which supplied Khan from 1976 until 2004. There is enough blame to go around. By outlining those mistakes, and examining the causes of those failures, I hope they will not be repeated. The problem is not going away, however. If anything, globalisation and the diffusion of WMD-relevant technology have only complicated the policy problem. Each new WMD programme could give rise to another potential A. Q. Khan.

 

What indicators or characteristics should concern policymakers and intelligence analysts in the future? Khan had broad financial autonomy. He had great flexibility in how he operated his procurement network. He could acquire and ship components with little external oversight. He had good connections with foreign suppliers. He was not personally screened by intelligence services, nor were his closest associates. He was a proud and greedy man. His wealth was well known, despite his relatively modest government salary. He had strong rivalries with other programme mangers. He jockeyed with them for public esteem, esteem which he coveted deeply. The public esteem also insulated him from political pressure and oversight. His independence was further reinforced by the turbulent nature of Pakistani politics, particularly during the troubled 1990s. Finally, he was deeply proud of being able to defy the discriminatory and Western-led export-control regime. Do these conditions apply elsewhere? If so, the next A. Q. Khan may already be out there.

Endnotes

1. “Pakistan’s Nuclear Bombshell”, India Today, 31 March 1987.

 

2. Quoted in Zahid Malik, Dr. A. Q. Khan and the Islamic Bomb (Islamabad: Hurmat, 1992), p. 75.

 

3. William J. Broad, David E. Sanger, and Raymond Bonner, “A Tale of Nuclear Proliferation”, New York Times, 12 February 2004.

 

4. There is clearly a great deal of uncertainty over the meeting. Most accounts name Farouq, but some are more equivocal on the presence of Tahir and/or Mebus.

 

5. Pierre Goldschmidt, IAEA deputy director-general, “Statement to the IAEA Board of Governors”, 1 March 2005, excerpted at [http://www.iranwatch.org/international/IAEA/iaea-goldschmidt-statement-030105.htm].

 

6. Massoud Ansari, “Nuclear Scientists from Pakistan Admit Helping Iran with Bomb-Making”, Sunday Telegraph (London), 25 January 2004.

 

7. Dafna Linzer, “Iran Was Offered Nuclear Parts”, Washington Post, 27 February 2005.

 

8. IAEA, “Implementation of the NPT Safeguards Agreement in the Islamic Republic of Iran”, report by the director-general to the Board of Governors, GOV/2004/83, 29 November 2004, p. 6.

 

9. Ibid., pp. 10–11. Sub-critical rotors operate below their first natural frequency of resonance. Super-critical rotors operate above that frequency.

 

10. David Albright and Cory Hinderstein, “Documents Indicate A. Q. Khan Offered Nuclear Weapon Designs to Iraq in 1990: Did He Approach Other Countries?” (Washington, D.C.: Institute for Science and International Security, 4 February 2025) [http://www.isis-online.org/publications/southasia/khan_memo.html].

 

11. “Re-imposition of Sanctions Feared”, Dawn (Karachi), 5 February 2004.

 

12. Ibid.

 

13. David E. Sanger, “Pakistan Leader Confirms Nuclear Exports”, New York Times, 13 September 2005.

 

14. IAEA, “Implementation of the NPT Safeguards Agreement of the Socialist People’s Libyan Arab Jamahiriya”, report by the director-general to the Board of Governors, GOV/2004/12, 20 February 2004, pp. 2–4.

 

15. See Victoria Burnett and Stephen Fidler, “Animal Lover, Egoist, and National Hero”, Financial Times (London), 7 April 2004.

 

16. Polis Diraja Malaysia, “Press Release by Inspector-General of Police in Relation to Investigation of Alleged Production of Components for Libya’s Uranium Enrichment Programme”, 20 February 2004.

 

17. William J. Broad and David E. Sanger, “As Nuclear Secrets Emerge, More Are Suspected”, New York Times, 26 December 2004.

 

18. Douglas Frantz and Josh Meyer, “For Sale: Nuclear Expertise”, Los Angeles Times, 22 February 2004.