PAKISTAN’S NUCLEAR PROGRAMME AND IMPORTS

Pakistan’s nuclear programme began for purely peaceful purposes. Pakistan produced two power reactors to generate electricity, is currently constructing a third and is planning a fourth. All are, or will be, safeguarded against military diversion by the International Atomic Energy Agency (IAEA) under Pakistan’s facility-specific INFCIRC/66 safeguards agreement. Like India, however, Pakistan has never accepted full-scope safeguards on its entire nuclear industry or signed the Nuclear Non-Proliferation Treaty. In the 1970s Pakistan’s nuclear programme became infused with military objectives. With the exception of the safeguarded reactors, the civilian and military programmes became intertwined. It is difficult to separate the history of one from the other, with the civilian programme falling victim to foreign controls imposed on nuclear exports to Pakistan over concerns about possible diversion to military use. India’s nuclear programme experienced a similar self-induced fate after it used plutonium produced by a Canadian reactor for a nuclear test in 1974. The repercussions from that test were more profound in Pakistan, however, due to its greater dependence on foreign technology and material for its nuclear programme. India, being more self-sufficient, suffered far less from the export control restrictions increasingly imposed by industrialised countries. Pakistan sought other means of pursuing its nuclear objectives. Those means abetted the rise of Abdul Qadeer Khan.

Historical overview of nuclear programme

Origins

A.Q. Khan can be accorded many epithets, including ‘founder of Pakistan’s uranium-
enrichment programme’. However, it is not appropriate to call him, as many do, the ‘father of Pakistan’s bomb’. Two of his countrymen can rightfully claim that title: on the political side, Zulfiqar Ali Bhutto, minister of mineral resources (1958–62), foreign minister (1963–66), president (1971–73) and prime minister (1973–77); and on the technical side, Munir Ahmad Khan, a US-trained scientist who was the chairman of the Pakistan Atomic Energy Commission (PAEC) from 1972 to 1991. Ironically, M.A. Khan directed the Pakistani bomb project between two positions at the IAEA: first between 1957 and 1972 as a staff member; then as a member of the Board of Governors, and even serving as IAEA Board chairman from 1986 to 1987.

PAEC, which traces its origin to 1954, was founded to promote peaceful uses of atomic energy, inspired and assisted by the US ‘Atoms for Peace’ programme. It was under Z.A. Bhutto’s leadership as minister of mineral resources that PAEC set up the Pakistani Institute of Nuclear Sciences and Technology (PINSTECH) in 1960 and sent hundreds of students abroad to obtain degrees in physics and other nuclear-related science disciplines. The first civilian research reactor, PARR-1, in Rawalpindi, became operational in 1965. By that time, since becoming foreign minister in 1963, Z.A. Bhutto had already begun lobbying in earnest to harness nuclear technology for weapons purposes. After the Chinese nuclear test in 1964, he concluded that India would go nuclear and that Pakistan would have to follow suit. He famously declared in a newspaper interview in 1965 that if India developed nuclear weapons, ‘Pakistan will eat grass or leaves, even go hungry’ in order to develop a programme of its own. When named president and chief martial law administrator in December 1971, in the aftermath of a traumatic military defeat by India, one of Z.A. Bhutto’s first priorities was to launch a military nuclear programme. He convened a meeting of several dozen scientists and officials in Multan in January 1972 and asked them to produce a nuclear bomb within five years, putting M.A. Khan in charge of PAEC. On a separate track, Pakistan’s first nuclear power plant, KANUPP (Karachi Nuclear Power Plant, a natural uranium reactor supplied by Canada) went critical in 1971 and was inaugurated by M.A. Khan in 1972.

In March 1974, PAEC set up a group tasked with developing a nuclear device. The programme was simply called ‘Research’ and the team was known as the ‘Wah group’, after the city where it was working, where the Pakistan Ordnance Factories are located. After India’s test on 18 May 1974 of what it called a ‘peaceful nuclear device’, a cabinet meeting the following month confirmed the official launch of Pakistan’s nuclear weapons programme, which was until then a ‘hedging’ option. This explains why, notwithstanding Z.A. Bhutto’s initiation of the programme in 1972, the US State Department later concluded that the ‘nuclear design and development work’ began ‘soon after the Indian test’. That test was the tipping point that transformed the 1972 ‘capability decision’ into a ‘proliferation decision’. Z.A. Bhutto was determined to level the playing field and to demonstrate Pakistan’s technological capabilities.

PAEC began reconnaissance for nuclear testing sites as early as 1976. The next year, the ‘Special Development Works’, a subsidiary of PAEC reporting directly to the chief of army staff (COAS), was set up on orders of (COAS) General Mohammed Zia-ul-Haq (who took executive control in a 5 July 1977 coup but did not assume the presidency until 1978) to prepare potential sites. PAEC’s work on nuclear weapons was coordinated by its Directorate of Technical Development, established in 1974 and run by Dr Samar Mubarakmand, who was and probably remains the most important scientist in Pakistan’s nuclear hierarchy. The directorate was initially located at Chaklala airport. In 1990, this weapons development unit morphed into a separate organisation, the National Development Complex (NDC).

The uranium programme

PAEC initially chose the most efficient pathway to nuclear weapons, the production of weapons-grade plutonium, which was the route India had also taken. Plutonium was considered the most logical path because once a country has a civilian reactor, a chemical reprocessing plant is all that is needed to recover the plutonium produced in the fission process. This is an easier technology than uranium enrichment. This choice seemed obvious given that M.A. Khan was a plutonium expert. However, the plutonium pathway was littered with obstacles. The KANUPP reactor was inefficient and under IAEA safeguards, and Western states soon limited access to the reprocessing technology necessary to extract weapons-usable plutonium from spent fuel from the reactor. Pakistan, therefore, hedged its bets by pursuing the less technically efficient, but more discreet, highly enriched uranium (HEU) route as early as 1974. Two events prompted this decision: the Indian nuclear test, which prompted an overall acceleration of the Pakistani nuclear weapons programme; and the Western restrictions that began to be applied to nuclear exports in September.

At this point, A.Q. Khan appeared on the scene, as a deus ex machina, bringing the gift of centrifuge enrichment technology. He was not a nuclear physicist but a metallurgist by training. In early 1972, as a recent PhD graduate from a Belgian university, he was offered a job at Fysisch Dynamisch Onderzoek, a subsidiary of Verenidge Machine Fabrieken, a company which worked closely with Ultra-Centrifuge Nederland (UCN), the Dutch member of the Urenco uranium enrichment consortium. He gained crucial knowledge of centrifuge operations by working at the Urenco Almelo plant from 1973 to 1975, where his tasks included translating a German report on centrifuge technology (G1 and G2 centrifuge models). On 17 September 1974, he wrote a letter to Z.A. Bhutto, volunteering his services to help Pakistan with the enrichment route to developing fissile material for a nuclear weapon. His offer was no doubt prompted by the Indian test and the bitterness he shared with his countrymen over the 1971 war. Khan may also have been motivated by the restrictions on nuclear exports agreed upon just a few days earlier by Western countries. In any case, his offer came at the right time; the authorities accepted it but asked him to stay longer in the Netherlands, to learn more.

Khan’s offer may have been a catalyst for ‘Project 706’, the codeword for the enrichment project, which was formally launched in October–November 1974 by Sultan Bashiruddin Mahmood under the supervision of M.A. Khan. Pilot installations were set up in Chaklala and Sihala. Z.A. Bhutto approved the enrichment programme and the building of the Kahuta centrifuge plant in February 1975. Kahuta was selected because it was isolated but close to Rawalpindi and its international airport. The army’s Special Works Organisation was brought in to help with construction and purchases. Another army unit was made responsible for the extraction of uranium ore in the regions of Gilgit and Multan.

Realising he was under suspicion from the Dutch government, A.Q. Khan left the Netherlands in December 1975, taking home the centrifuge designs he had stolen (a crime for which in 1983 he was sentenced in absentia to four years in prison by an Amsterdam court, although the sentence was later overturned on appeal). He then went to work for PAEC under the supervision of M.A. Khan and Mahmood. However, he soon expressed dis satisfaction with the slow pace of the programme and threatened to resign. In the spring of 1976, Z.A. Bhutto agreed to give A.Q. Khan control of the centrifuge project, reporting directly to the prime minister. Bhutto also created a three-man ‘coordination committee’ to oversee the project. One member of this committee was Defence Minister Ghulam Ishaq Khan, later, as president from 1988–1993, a key supporter of A.Q. Khan. G.I. Khan’s role was to liaise with the military. The committee, initially named Engineering Research Laboratories (ERL), was inaugurated in July 1976 as an independent entity under A.Q. Khan’s leadership.

Kahuta originally worked on copies of Dutch designs (CNOR and SNOR) made by UCN, and then on German designs (G1 and G2), all of which had been made for Urenco. A.Q. Khan, who had had access to them at the Almelo plant, brought back both sets of designs from the Netherlands. Construction of the centrifuge cascades involved not only the procurement of many spare parts, but also crucial knowledge that Khan tried to obtain from his former European colleagues. The adaptation of the faster-spinning G2 design, capable of enriching more than twice as much uranium per machine than the G1, would prove to be a major technical challenge and required the procurement of large quantities of maraging steel (a high-strength iron alloy used for the G2 rotors; the term derives from ‘martensite age hardening’).

Like his predecessor, Zia directly managed the nuclear programme when he came to power in 1977. He also reinforced ERL’s autonomy. There are various possible explanations for the independence given to ERL. One was Zia’s caution regarding M.A. Khan, who was very close to Z.A. Bhutto. Independence for ERL was likely to induce a brisk competition with PAEC – a healthy process in itself, but also a way for Zia to get information from one on the other. Giving ERL more autonomy may also have been a deliberate strategy to overcome what were seen as the bureaucratic constraints that had hobbled PAEC. Since the policy of direct oversight adopted vis-à-vis PAEC had failed – and PAEC’s own acquisition strategy had not succeeded – it was time for another approach. Finally, ERL autonomy may have been seen as a way to protect the secrecy of the enrichment programme. However, a contrary explanation has also been posited: it is possible that A.Q. Khan was deliberately made the visible face of Pakistan’s programme to act as a ‘decoy’, to shift attention away from PAEC where what was considered the most important work at the time was being conducted.

PAEC’s functions certainly remained central. In addition to the management of nuclear reactors (including fuel fabrication), PAEC retained control of uranium conversion and continued to work on the improvement of enrichment technology. It is noteworthy that 75% of the ERL scientists and engineers were from PAEC. Despite the intense personal and institutional competition within the Pakistani nuclear complex, these were not two entirely separate programmes.

Nevertheless, Kahuta was to become the most visible part of the complex and a ‘symbol of our national defiance’ (in the words of a headline from the 1991 Pakistani Defence Journal). According to A.Q. Khan, ERL succeeded in enriching uranium (in a tiny amount) for the first time in April 1978. The plant was made operational in 1979 and by 1981 was producing substantial quantities of enriched uranium. In May 1981, Zia renamed the laboratories ‘Khan Research Laboratories’ to honour the successful director. The rhetoric used by Pakistanis in general and by A.Q. Khan in particular during the 1980s and 1990s made the mastery of uranium enrichment the quintessential symbol of national pride, scientific and technical modernity, and independence from foreign powers. In 1992, Kahuta was said to have operated some 3,000 centrifuges, mostly of the P-2 design, and was estimated to have a capacity of 9,000–15,000 Separative Work Units (SWU) per year. If these figures are correct, it could produce 45–75kg of HEU a year (enough for two to three warheads, assuming 20kg of HEU is required for a weapon). It is not known how many centrifuges are currently installed and operating at Kahuta.

Unconfirmed news reports from early 2007 indicate Pakistan also developed more advanced centrifuge designs in the mid 1980s, labelled the P-3 and P-4, which resembled other Urenco models. Compared to the P-1, which has a throughput of 1–3 SWU/year and the P-2, which has a throughput of 5 SWU/year, the P-3 has a design throughput of just under 12 SWU/year and the P-4 a throughput of just over 20 SWU/year.

The plutonium programme

The success of the HEU route was important because progress on the plutonium route, which had been initially chosen by PAEC, had slowed due to the actions of Western suppliers increasingly concerned about diversion for non-peaceful purposes. In 1976, Germany cancelled a contract for a heavy-water production plant and Canada terminated its supply of nuclear fuel. In 1978, France withdrew from the construction of a reprocessing plant at Chashma.

PAEC’s plutonium route nevertheless continued in parallel with KRL’s enrichment programme. A heavy-water production facility supplied by Belgium was completed in 1980, with a capacity of 13 tonnes^[1]per year. The unsafeguarded New Labs pilot plant for plutonium reprocessing at the PINSTECH complex near Rawalpindi became operational in the early 1980s. But the only source of irradiated fuel, the KANUPP reactor, was under safeguards. In addition, to produce weapons-grade plutonium, KANUPP would have had to run at low efficiency, thus stretching production time. At 20% efficiency, it would have taken 10 years to produce enough suitable plutonium for one bomb.

The plutonium programme was reinvigorated in the late 1980s, but at a slow pace. A CANDU-type heavy-water-moderated 50MWt reactor at Khushab, built with Chinese assistance and unsafeguarded, began operating in April 1998. Able to produce enough irradiated fuel for about 10–15kg of plutonium per year, it significantly increased Pakistan’s fissile-material production capabilities. In 2000, the New Labs facility began separating plutonium from the Khushab reactor’s irradiated fuel. It has the capacity to separate 10–20kg of plutonium per year, enough for two or three weapons. At Khushab, a new heavy-water production plant was also built, with a capacity recently estimated to be 13 tonnes of heavy water per year.

A second heavy-water reactor is currently being built at Khushab, perhaps as a replacement for the first reactor. According to the US government and some private experts, it resembles the first Khushab reactor and will have a 40–100MWt capacity. Other experts believe its design indicates a 1,000MWt capacity, which could produce enough material for 40–50 weapons a year. In any case, if Khushab-I continues to operate, the addition of Khushab-II would allow Pakistan to increase its rate of plutonium production for weapons by a factor of two to three.

It is conceivable that Pakistan was able to produce plutonium before Khushab-I came on line by diverting fissile material from KANUPP. After Islamabad declared in 1980 that it had begun producing nuclear fuel, the IAEA asked to increase monitoring of the installation, but Pakistan refused. As a result, in 1981 and 1982 the IAEA was unable to conclude that there was no diversion. Having spent 14 years at the IAEA, M.A. Khan was apparently not unduly concerned. Early reprocessing may have taken place at a pilot separation plant built in the early 1970s, or at the New Labs themselves, which reportedly began operating on a limited scale in 1984. Another possibility is that the Pakistanis may have completed the Chashma reprocessing plant after the French cancelled their participation (as Z.A. Bhutto reportedly recommended from his jail cell to M.A. Khan). When the French withdrew in 1978, Pakistan already had 95% of the design plans and, even after the cancellation, French engineers continued with civil construction, and SGN (the French contractor) continued to export equipment to Pakistan, the last SGN technician leaving the site in June 1979. In 1983, a now de-classified US State Department assessment reported that civil works were externally complete. However, it also stated that ‘Pakistan may have difficulty in outfitting the interior of the plant with process equipment and operating it at or near design capacity’. Satellite imagery indicates construction activity in the area resumed between late 2000 and late 2002 and continued steadily through the next four years. Imagery analysis provides circumstantial evidence that the site is intended for reprocessing. It is unknown whether it would be for the original intended capacity of 100 tonnes of spent fuel a year and how close the facility is to being operational. The imagery also depicts what could be a fuel fabrication facility.

Crossing the threshold

According to Pakistani sources, PAEC had its first nuclear weapon design, using uranium-238 (U-238) as a reflector, as early as 1978. The first cold test (triggering the device without use of fissile material) was reportedly carried out on 11 March 1983 by Mubarakmand. In June of that year, the US State Department reported that Pakistan was now able to produce a workable explosive triggering package. This design was intended to be delivered by aircraft. ‘Hot’ testing facilities were completed as early as 1980.

Meanwhile, A.Q. Khan also proceeded with the weaponisation of an HEU device. Many observers argue that his activities in this field were un sanctioned. However, according to A.Q. Khan, Zia had in fact instructed KRL to develop its own weapon in 1982, while keeping the project completely secret. In March 1984, KRL began its own cold tests at Kahuta. In April, A.Q. Khan announced that Pakistan was able to produce HEU.

Pakistan reached the nuclear weapons threshold around 1985–86. The exact date is unclear, and depends on whether one refers to the year enough HEU was produced, or the year when actual weaponisation was achieved. As early as December 1984, A.Q. Khan claimed that KRL was ‘in a position to detonate … a nuclear device on a week’s notice’. A 1986 US National Intelligence Estimate concluded that Pakistan was only ‘two screwdriver turns’ from assembling a weapon and could do so within two weeks of making a decision. The crossing of the threshold became public in early 1987, when A.Q. Khan stated in a press interview that the country was now able to build the bomb. His intent was to warn India, which had massed troops near the border in a military exercise calledBrasstacks, but the statement caused problems with the US administration, which had been assuring Congress there was no sign of a Pakistani bomb (see page 94). According to several reports, including testimony of General Naseerullah Babar, the first weapon assembly took place during the nuclear crisis in the winter of 1989–90, when the escalating insurgency in Kashmir prompted India to contemplate limited military strikes at insurgent training camps across the border. It should be noted, however, that President Musharraf claims that as late as 1999 Pakistan’s nuclear capability ‘was not yet operational’.

Pakistan originally intended to rely on aircraft as delivery vehicles for its nuclear weapons and in the mid-1980s procured 40 F-16 aircraft from the United States that, when locally modified, could be used for that purpose (although the US deliberately did not configure the aircraft for nuclear delivery). From 1985, however, passage of the Pressler Amendment made delivery of additional F-16 aircraft, as well as spare parts, dependent on annual US presidential certification that Pakistan did not possess a nuclear explosive device. That condition became increasingly hard for President Ronald Reagan and then President George H.W. Bush to fulfil, and in 1990 the Pressler Amendment was finally invoked, cutting off military aid and sales to Pakistan, as well as most development assistance. By then, Pakistan had already begun to examine a ballistic missile alternative, prompted additionally by the development of India’s own ballistic programmes. Pakistan thus embarked on a fully fledged pursuit of ballistic missiles from both China and North Korea. A deal reportedly was struck in 1987 with China, which by 1989 had begun supplying missile technology to PAEC.

Not to be outdone, in the early 1990s A.Q. Khan made his own attempts at ballistic-missile procurement. He approached North Korea for liquid-fuelled missiles. This move was fully supported by the Pakistani authorities: Islamabad needed longer-range missiles, and China was beginning to be pressed by Washington to limit its ballistic technology transfers. Pakistani authorities also probably saw an advantage in having two different missile families and another field of intense competition between the two laboratories. Pakistani missile scientists are believed to have inspected the No-dong missile as early as 1992 and were allegedly present during the missile’s flight test in May 1993. At A.Q. Khan’s initiative, then-prime minister Benazir Bhutto initiated an agreement with North Korea in December 1993 to supply No-dong missiles. The deal was reportedly completed two years later and components reportedly began to arrive in 1996 and 1997.

In May 1998, both PAEC and KRL were ready to test nuclear weapons. As the Pakistani leadership was debating the pros and cons of conducting hot tests after India’s campaign, A.Q. Khan claimed that KRL was ‘fully independent in the nuclear field’ and was equally qualified and equipped as PAEC to conduct the tests. After the Defence Committee of the Cabinet ruled in favour of PAEC, he appealed to COAS General Jehangir Karamat, who interceded in his favour. KRL was eventually involved in the testing campaign.

Nuclear imports

Overview

Pakistani nuclear imports from the 1960s to the present range from large, official state-sponsored contracts to a multitude of small-scale operations for critical components or items. Initial imports were complete installations officially designated for civilian use. These included the US-made PARR-1 research reactor and the Canadian-made KANUPP reactor.

In the late 1960s, Pakistan approached the French company SGN for the sale of a reprocessing plant at Chashma, which would use the fuel irradiated at KANUPP. The final contract was signed in October 1974. However, belated non-proliferation concerns led the French to suggest a change in the design (termed ‘co-processing’ rather than ‘reprocessing’) that would make Pakistan unable to produce weapons-grade plutonium. Islamabad’s refusal led to the French withdrawal in June 1978. However, as noted above, Pakistan already had 95% of the drawings, and the French continued civil engineering works. The initial Pakistani–French discussions had taken place before the launch of the military programme, so there may very well have been, at least initially, an honest intention to use the plant for civilian reprocessing, especially since it was to be under IAEA safeguards. M.A. Khan is reported as stating that the plant was not directly intended to produce weapons-grade material, but rather to produce knowledge about how to do so. But there was also a second track: an experimental reprocessing facility was sold by British Nuclear Fuels Ltd (BNFL) in 1971, and installed at PINSTECH. Islamabad then asked the British for a larger plant. When BNFL refused, Islamabad turned to the Belgian firm Belgonucléaire. In the end, it was reportedly the French company SGN which engineered the pilot reprocessing plant, while Belgonucléaire designed the overall building (the ‘New Labs’ themselves) and built a fuel-fabrication laboratory. (The same firm is also reported to have built the Multan heavy-water facility in 1980.)

The launching of the enrichment programme in the mid-1970s signalled the beginning of a vast procurement campaign targeting Western firms. Most of the equipment sought from Europe was for the enrichment process, including components of the centrifuges themselves (high-frequency inverters, high-vacuum valves, scoops pre-forms, and bottom-bearing pre-forms). The shift to P-2 centrifuges, which occurred in the mid-1980s, required the acquisition of large quantities of maraging steel. The procurement network also sought equipment for the plutonium programme (hot cell manipulators, reprocessing equipment, etc.). Beginning in the early 1980s, Pakistani imports included components for the nuclear weapons themselves (high-speed electronic switches – triggered spark gaps or ‘krytrons’ – in particular). Measuring equipment (oscilloscopes and X-ray machines) was also actively sought. Pakistani imports also included nuclear materials (uranium yellowcake, uranium hexafluoride and tritium) and metals (beryllium).

Pakistani imports from the West ranged from full-scale installations to subcomponents. Some of the more significant imports were uranium-conversion facilities (CES Kalthof GmbH, Germany, late 1970s); 6,500 tubes of maraging steel (Van Doorne Transmissie, Netherlands, late 1970s); a yellowcake production unit (Société d’Etudes et de Travaux pour l’Uranium, France, late 1970s); a reprocessing facility (Saint-Gobain Techniques Nouvelles, France and Belgonucléaire, Belgium, late 1970s); a heavy-water production facility (Belgonucléaire, late 1970s); and a tritium production facility (Nukleartechnik GmbH, Germany, late 1980s).

China has been the most important state contributor to the Pakistani nuclear programme, though the extent of its assistance is difficult to assess. Nuclear cooperation with China is one of the most closely held state secrets in Pakistan; otherwise forthcoming officials respond to questions on the topic with polite but firm silence. A nuclear agreement was signed with China in late May 1976, on the occasion of Z.A. Bhutto’s visit to Beijing, his third since 1972. The agreement, which he referred to as one of the most important achievements of his presidency, is sometimes seen as having opened the way for co operation in the military sphere. On the Chinese side, however, the military and civilian agreements may have involved different bureaucracies. It is possible that nuclear cooperation on both fronts proceeded in parallel, during a period when both Pakistan and China had few other friends. Beijing’s assistance in the development of Pakistani nuclear weapons was reportedly acknowledged by former foreign minister Gohar Ayub Khan in 1998. China’s assistance became much more important in the 1980s as Western countries and firms became increasingly cautious vis-à-vis the Pakistani programme. The US State Department believed in 1983 that Beijing may have helped Pakistan overcome some of the difficulties it had in mastering enrichment technology, and that bilateral cooperation was possibly taking place in nuclear-device design. China may have supplied uranium hexafluoride (and, according to unconfirmed reports, possibly HEU for up to two weapons) to Pakistan. It also supplied (no later than 1983) a weapon design that had been tested in 1966. Chinese nuclear and missile assistance to Pakistan became an increasing source of both US–China tension and internal US political debate, although other priorities in US–China relations were sometimes seen to pre-empt proliferation-related sanctions. Chinese government assistance to Pakistan’s nuclear weapons programme apparently ended after the international furore sparked by the 1994–95 transfer of 5,000 ring magnets to KRL from the China Nuclear Energy Industry Corporation. The Clinton administration accepted the claim that senior Chinese government leaders did not know about the transfer, and China pledged in 1996 not to allow any further cooperation with unsafeguarded nuclear facilities. Despite lingering suspicions in some parts of Washington, China’s membership in the Nuclear Suppliers Group (NSG) from 2004 provides an important confirmation of that pledge.

Several African countries were significant suppliers of uranium. A.Q. Khan’s repeated trips to uranium-rich African countries were at least partly in connection with these imports. In particular, Pakistan is known to have imported about 500 metric tonnes of uranium oxide from Niger in 1979 (only 110 of which were declared and safeguarded, because much of it came secretly via Libya).

How the imports network operated

Pakistan’s nuclear programme involved two types of imports: ‘state to state’, that is, sales or exchanges approved by governments on both ends of the transaction, as was the case with many imports from Western countries until the late 1970s and from friendly countries (China and African states) afterwards; and ‘firm to state’: illegal or un authorised sales from Western firms to Pakistan, generally through intermediaries, including both legitimate businesses and front companies.

During the 1970s and 1980s, Western imports were overseen by M.A. Khan, who was, in a sense, the network’s ‘operational commander’. PAEC’s Directorate of Technical Procurement directed the imports. One of the individuals involved was Anwar Ali, now the chairman of PAEC, who played a key procurement role for KRL before A.Q. Khan took over. Another individual who played a critical role was the physicist Siddique (sometimes reported as Sulfikar) Ahmed Butt, whom Z.A. Bhutto asked in 1972 to organise a purchasing network and who for a long time was the ‘tactical commander’ of the imports network. Butt was posted in Belgium in 1975 and in France after 1977. In July 1975, he began coordinating nuclear-related imports, possibly using A.Q. Khan’s ‘address book’, since the latter had by then started giving relevant contacts and information to Islamabad even before returning to his country in December. In August, the Pakistani embassy in Brussels began inquiring about the possibility of buying inverters. The network began operating in earnest in 1976. A.Q. Khan, by then back in Pakistan, and having brought back a list of dozens of companies that supplied centrifuge parts, played a crucial role, one example being the ‘shopping trip’ for enrichment-related equipment he made with his deputy G.D. Alam to Switzerland in 1976. Khan’s early procurement role, however, mostly involved the management of imports related to centrifuge technology. Butt managed both uranium-related and plutonium-related imports, and remained in charge at least until the late 1980s.

The procurement network’s modus operandi included various techniques that enabled its success and longevity:

· Systematically using Pakistani embassies abroad. From the early 1970s until at least the late 1990s, Pakistani embassies around the world, in particular in Europe, were key components of the network, and used diplomatic pouches to send material home.

· Paying above market premium. Direct contracts with Western industrialists were made easier because Pakistan often paid up to 50% more than the market price.

· Keeping one step ahead of export controls. When export controls began to be applied to plutonium reprocessing, the Pakistanis shifted to HEU production. When export controls were reinforced in the late 1970s, they purchased individual components rather than entire units, and developed industrial facilities to manufacture as many parts as possible. Pakistan also sought to import ‘pre-forms’ (unfinished products), which are not necessarily covered by export controls. British engineer and businessman Peter Griffin, who was a regular supplier to Khan for 25 years, boasts that his shipments conformed to whatever export controls were in place at the time.

· Hiding a critical component in a long list of useless material. This ‘needle in a haystack’ tactic, designed to overwhelm Western export controls, was also used later by A.Q. Khan for Pakistani exports.

· Buying a sample and the means to reproduce it. For instance, in 1981–82 Pakistan tried to buy both metal components for nuclear weapons and equipment (such as precision lathes) used to make them from European companies. Once export controls became more stringent, Pakistan did its best to become self-reliant. As A.Q. Khan put it: ‘once the Western propaganda reached its climax and all efforts were made to stop or block even the most harmless items, we said enough and started indigenous production of all sophisticated electronic, electrical and vacuum equipment’.

· Using multiple connections and buyers to look for a given item. To ensure that at least one sale came through, Pakistan often sent two and even three procurement agents or front companies to buy the same product.

· Using front companies. The network created or used front companies in Pakistan, Europe and intermediary countries. Examples in Pakistan include such companies as Al Technique Corporation of Pakistan, Ltd; Allied Trading Co.; ANZ Importers and Exporters; High Technologies, Ltd; Lastech Associates; Machinery Master Enterprises; Maple Engineering Pvt. Ltd Consultants, Importers and Exporters; Modern Engineering Services, Ltd; Orient Importers and Exporters; Technical Services; and the Tempest Trading Company. Other examples in Pakistan include Asiatic Chemical Industries, Ltd; Karachi Silk Mill Co., and Arshad, Amjad & Abid, Ltd. Examples in the United Kingdom included Weargate, Ltd. In Abu Dhabi the network used the services of Khalid Jassam General Trading. One of the most famous Pakistani front companies was the Institute for Industrial Automation, which was the recipient of transfers organised by Henk Slebos, a Dutch metallurgist Khan met in 1964.

· Falsifying the end user. To evade national export controls and internal procedures established by manufacturing companies, the network systematically falsified end-user certificates and forged order forms. South African court documents detail how two South African-based members of the network, Gerhard Wisser and Daniel Geiges, forged order forms for flow meters and other special equipment from Leybold in Germany for Pakistan’s gas centrifuge enrichment plants. They changed the equipment listed on earlier legitimate orders from an innocent company in South Africa in order to circumvent Leybold’s non-proliferation policy. Upon arrival in South Africa, the equipment was exported to Dubai for onward shipping.

· Using multiple intermediaries. Pakistan also increasingly used intermediaries, shipping items to one state, which would then re-export it to Pakistan, using false end-user certificates. The UAE, Turkey and South Africa were among the countries used as transshipmentpoints.

· Enlisting the help of friendly countries. Libya directly helped Pakistan by playing the role of an intermediary for uranium from Niger, and procuring on its behalf. North Korea was reportedly a conduit for some of China’s assistance for Pakistan’s ballistic missiles.

· Involving Pakistani-born foreign nationals. Through financial or ideological incentives, Pakistan enlisted the contribution of foreign nationals of Pakistani origin. A.Q. Khan made extensive use of this method, asking several of his countrymen to come back to Pakistan, collect information, or assist with the procurement of spare parts.

· Making extensive use of personal connections. It is here that A.Q. Khan made possibly his most significant contribution. After returning to Pakistan, he wrote to several former colleagues to get specific technical information and he continued to expand his personal network of accomplices. Key associates of his included long-time acquaintances: Heinz Mebus, a German businessman who had been a college classmate; Henk Slebos; Peter Griffin, who Khan met in London in 1976; Friedrich Tinner, a Swiss engineer Khan met in the 1970s; Gotthard Lerch, a German engineer Khan met in the 1970s; and Abdus Salam, a British national and another personal friend.

This modus operandi resembled that of Iraq in the 1980s, when Baghdad resorted to an increasingly refined imports strategy which heavily relied on multiple fake companies and sought at least two sources for any given material it needed. Iraqi embassies were heavily involved. Several individuals and companies (German and Swiss in particular) were selling to both countries. The Pakistani network, however, appears to have been more centralised. A British intelligence report in 2003 reportedly listed no less than 95 Pakistani organisations and government bodies, including diplomatic posts abroad, which had assisted in the country’s nuclear imports.

It was a ‘grey’ market rather than a ‘black’ one. In the 1970s and 1980s, Pakistan took advantage of the combination of inadequate export controls and the desire of Western firms to sell technology abroad. The first IAEA guidelines published in 1974 were very limited, and it was only in 1978 that the first NSG guidelines were published. National nuclear export controls initially focused on plutonium, in particular on preventing the sale of heavy-water-moderated reactors and reprocessing plants. In the 1970s, the risk of a non-nuclear country developing enrichment activities for military purposes was not taken very seriously, given the technological know-how required.

Many of the components Khan procured were dual use and used in many other industries. The NSG added controls on dual-use goods in 1992 in response to Iraq’s success in procuring dual-use imports for its covert nuclear weapons programme. Many goods useful to nuclear programmes are not included on control lists at all, because their uses are so varied, and were not properly regulated until exporting countries started adopting ‘catch-all’ controls blocking any export suspected to be destined to a nuclear weapons programme, even if it was not on a control list. Falsifying the end user then became the typical evasion technique used by a procurer.

The weakness of export controls and the fatalism of Western suppliers were the strongest factors abetting the import network. Many industrialists reasoned that ‘if we do not do it, others will’ and deliberately violated the law. A wilful naivety and arrogant scepticism about Pakistan’s ability to put sophisticated machinery to military use also played a role, as in the case of UK-based Emerson Electric, which sold specialist high-frequency converters to Pakistan in the 1970s, even though they had been used in the British uranium enrichment programme. Others believed that the Pakistani programme would remain very limited. Several suppliers did not realise they were helping Islamabad to get nuclear weapons – or did not want to know. According to A.Q. Khan: ‘the Western world would never talk about their own hectic persistent effort to sell everything to us … At the time we received many letters and telexes and people chased us with figures and details of equipment they had sold to Almelo, Capenhurst etc. They literally begged us to buy their equipment.’ Some of the main figures involved in Pakistani imports have stated that they consciously endorsed the spread of nuclear technology, arguing that it would make the world more secure. Peter Griffin offered this justification in a 2006 press interview. The network was indiscriminate with regard to its customers, selling enrichment-related equipment and technology services to both Pakistan and India, according to South African court documents.

European firms were particularly targeted due to the extent of A.Q. Khan’s personal contacts on the continent, the existence of liberal trade policies among European Community members (which allowed Pakistan to hide the final destination of a given piece of equipment), and, perhaps most importantly, the lack of commitment shown by many European countries to more stringent non-proliferation efforts, notwithstanding the wake-up call of the Indian test. Until the end of the Cold War, proliferation was not a great concern, and export controls were weak.

There was also strong resistance to perceived American pressure. Some countries, like France and Switzerland, maintained a self-consciously independent political stance. Others, like Germany and the Netherlands, also made the promotion of exports involving precision engineering a national priority. Many European bureaucrats whose jobs involved export controls may have viewed American pressure to stop exports to Pakistan as a covert attempt to obtain a commercial advantage. Finally, the nuclear market was particularly competitive, and many Europeans did not want to see it dominated by the United States. The Netherlands, Germany and the United Kingdom wanted a competitive enrichment market, and banded together to form Urenco in 1971. It took some time before attitudes changed and non-proliferation concerns at the highest political level translated into action at the bureaucratic level.

The example of the French contract for a re processing plant is telling. Most of the French officials and industrialists involved realised that Pakistan wanted nuclear weapons, but they also badly wanted access to the then-promising Pakistani defence and nuclear market. In the mid-1970s, a change in government in France and US pressure after the passage of the Symington Amendment in June 1976 led Paris to alter its position. The Chashma contract was eventually suspended in 1978. But Pakistan had little trouble convincing some French technicians to remain with them for some time (with the knowledge of French authorities, who viewed the technicians as a potential source of intelligence on the Pakistani programme).

It should also be noted that many Pakistani scientists and engineers gained crucial knowledge about the enrichment process through education, training and internships in European firms (sometimes under the aegis of UNESCO programmes). A.Q. Khan was the best known of these, but certainly not the only one. Tens of scientists were trained in Europe, in particular in Belgium and Germany.

The large number of German companies involved in Pakistani nuclear imports can be explained by several factors. Firstly, German expertise in machine tools, engineering and precision mechanics is renowned. Germany was also involved in the nuclear enrichment industry through Urenco. Secondly, not being a nuclear power, Germany’s export controls in this field were even less efficient than those of France and the United Kingdom, for reasons of expertise. (For the same reason, scientists and engineers with nuclear expertise from the latter countries were likely to be involved in national nuclear programmes, removed from the private market.) Thirdly, German nuclear export controls were, for a long time, notoriously weaker than those of other major European states. This reflected Bonn’s strong political desire to promote an export-oriented economy. During the 1980s, a decade in which the United States reportedly demarched the Bonn government 100 times about German exports to Pakistan, only two officials were assigned to review licences for the exports of nuclear materials. There was also, to an extent, a deliberate policy of self-assertiveness in the face of US pressure. A.Q. Khan also had extensive contacts in Germany dating from his stay in Europe. As a result, in 1990, a member of the German Parliament could say that his country’s export controllers’ motto was still: ‘you never hear anything, you never see anything – and, in particular, you never block anything’. In 1989, the German magazine Stern reported that throughout the 1980s, no less than 70 German firms had sold nuclear-related goods to enterprises known to be associated with the Pakistani programme.

Many of these factors were evident in other European countries, such as Belgium, France, the Netherlands or Switzerland. Swiss firms were particularly involved in the late 1970s. During a stage when export controls did not have catch-all clauses, the firms stuck to the letter of the legal restrictions and ignored what should have been obvious proliferation risks that did not meet the exact criteria of the restrictions. As in the case of Germany, Bern’s legalistic stance was ‘part of a strategy to promote the interests of Swiss industry’. Some have perceived a long-standing divergence between the Swiss ideal of neutrality (including freedom of trade) and the concept of non-proliferation. These attitudes may have persisted: as late as 2004, a Malaysian police report about the network was seen in some federal circles in Switzerland as a possible attempt to negatively affect the image of Switzerland abroad and the competitiveness of Swiss exports.

Pakistan also actively sought materials and equipment from the United States, which was itself not consistent in enforcing its own export control policy. As late as 1994, a US General Accounting Office report stated that US export controls were still woefully inadequate. Between 1988 and 1992, more than 80% (650 of 808) of applications for exports of nuclear-related equipment to Pakistan were approved. Although most of these were for items with legitimate civilian uses, three went to sensitive end users. There were also a high number of documented ‘failures’ by Pakistan to import nuclear-related material from the United States. This indicates both that the United States was particularly targeted, and that US export controls were more efficient than elsewhere. Indeed, by the late 1990s, US export control officials were routinely using catch-all provisions to deny licenses to many kinds of otherwise innocuous items for both Pakistan and India.

In the 1980s, however, non-proliferation efforts in industrialised countries were still frustrated by a political desire not to interfere with national commerce, and by a bureaucratic rigidity unable to respond to the loopholes provided by globalisation. In addition, there was no real deterrent; those convicted of violations of export controls were generally given short jail sentences, if any. Moreover, as noted above, the Pakistani imports network progressively ‘went underground’ in developing more subtle techniques designed to get around the tightening of Western export controls. The procurement network also used sophisticated financing mechanisms. Pakistan often used banking channels in the Gulf to purchase goods in western Europe more discreetly. The external support for the Pakistani programme from Libya (beginning in the early to mid-1970s when Z.A. Bhutto befriended Colonel Muammar Gadhafi) and Saudi Arabia (reportedly in exchange for security guarantees through the resultant technology) often seems to have made use of the same channels. Libya was probably financing the Pakistani programme through the Pakistan–Libyan Holding Company, a joint venture set up in late 1978. One particular institution, the Bank of Credit and Commerce International (BCCI, which was closed in 1991), apparently was a key node in the financial network used by Pakistan. The BCCI was originally a Pakistani bank. It was only later restructured into BCCI Société Anonyme (Luxembourg) and BCCI Overseas (Grand Cayman). A US Senate report in 1992 attested to the tangled relationship among BCCI, the Pakistani government and the Pakistani BCCI Foundation, which was created as a means of sheltering BCCI profits from taxation and which in 1981 received tax-free status while Ghulam Ishaq Khan was minister of finance. The foundation used some of the profits it received from Pakistani operations to finance projects the Pakistani government was unable to pay for itself. For example, BCCI provided $10 million in grants in the late 1980s to finance an officially ‘private’ science and technology institute named after Pakistani president G.I. Khan and directed by A.Q. Khan. Around the same time, other BCCI officials arranged nuclear transfers paid for by Pakistani front companies through BCCI-Canada.

Current outlook

Notwithstanding the development of Pakistan’s industrial base over the past 30 years, its nuclear weapons programme is probably not completely self-sufficient. Having based its programme on foreign imports, and disadvantaged by a relatively weak industrial base, Islamabad continues to import foreign components for spare parts and upgrades for the modernisation of its facilities and weapons. Even if Pakistan could produce most parts on its own, it would still have an incentive to seek more advanced, higher-quality components from abroad. Pakistan’s P-2 centrifuges are 20 times less efficient than Urenco’s state-of-the-art TC-21 machines. If Pakistan could obtain more efficient centrifuges through the black market, it would have every reason to do so. Now alert to the problem, Europeans are unlikely to let the technology escape again. The construction of a second, probably much larger, heavy-water-moderated reactor at Khushab, which is currently underway, and the reprocessing plant apparently being built at Chashma and related expansion of the plutonium programme will likely require additional imports.

Evidence that Pakistan continued to try to procure from the nuclear black market even after A.Q. Khan was exposed has surfaced in at least two Western court cases. In one set of cases, Pakistani businessman Humayun Khan and his Israeli associate Asher Karni were found to have sought to illegally export to Pakistan 200 triggered high-speed spark gaps, and ten high-end oscilloscopes. This equipment has medical applications but can also be used to test, develop and detonate nuclear weapons (see page 155). In another case, Rainer Vollmerich, a German businessman, was found to have exported to Pakistan, until 2004, mechanical and electronic equipment with military nuclear use (see page 104).

Recent developments confirm that Pakistan continued to seek nuclear imports even after the exposure of the A.Q. Khan network. This should not come as a surprise, since Khan’s network was only one part of the Pakistani procurement effort. In July 2006, the Russian government revealed Pakistan had been actively searching for nuclear-related technology in the country. In fact, Pakistan had to recreate overseas procurement networks when it learned that some of the middlemen it was using were also procuring for Iran. (Not all reported incidents can be taken at face value, however. The Swiss federal police agency stated that in 2004 it precluded two attempts at exporting aluminium tubes of Russian origin to KRL. Since Pakistan’s P-2s do not use aluminium, the tubes in question apparently have other problematic uses.) According to press reports, in July 2005, a confidential report prepared by European intelligence agencies for the EU reported that Pakistan was still ‘shopping’ for high-grade aluminium, ring magnets and machine tools that could be useful for its nuclear programme: ‘since the beginning of 2004 extensive procurement efforts for the Pakistani nuclear sector have been registered’. The document reportedly said the range of materials and components being bought ‘clearly exceeds’ that required for spare parts and replacements for Pakistan’s nuclear programme. Pakistan’s procurement agents also continue to buy older-generation valves not on the NSG control lists but still serviceable for Pakistan’s purposes. The claim by some Pakistani officials that they need to continue buying abroad because of damage done to key installations by the October 2005 earthquake does not explain why these procurements preceded the earthquake date.

Pakistani nuclear forces: current state and likely evolution

Delivery vehicles

Pakistan’s deterrent capability was originally designed for aircraft delivery (French-origin Mirage-5, US-made F-16 and Chinese A-5 aircraft). Although Pakistan is transitioning to a missile-based force, it would make sense for it to continue to maintain an air-based component. Deliveries of American F-16s for conventional defence could indirectly help Pakistan to maintain a significant air-delivered nuclear capability, if it modifies them for this purpose.

Like the nuclear programme, Pakistan’s missile programme has developed along two distinct tracks. One programme, run by NDC in conjunction with the Pakistan Space and Upper Atmosphere Research Commission and PAEC, has concentrated since the early 1980s on solid fuel-propelled short-range ballistic missiles. The first model, the Hatf 1 (80km range) was flight-tested in 1989 and fielded in 1992. In 1997 Pakistan revealed the Hatf 2a (300km), and the Hatf 3 (290km), and in 1999 the Hatf 4 (750km), also called the Shaheen 1. The latter three missiles are believed to be based on the Chinese M-11, and many analysts believe the Hatf 2a is a renamed Chinese M-11 missile, 34 of which apparently were imported in 1992. A medium-range solid-propellant Shaheen 2 (2,000km), also called Hatf 6, was first tested in March 2005.

Separately, KRL headed a second programme to produce liquid-propellant missiles. With North Korean assistance, KRL developed the Ghauri (1,500km), originally called Hatf 5, beginning about 1993. TheGhauri is basically a North Korean No-dong (itself a scaled-up Scud). North Korea reportedly supplied 12–25 No-dong assembly kits to Pakistan between 1996 and 1997 and North Korean technicians apparently worked in Pakistan on further development of the missile, which was first tested in April 1998. A two-stage Ghauri-2 (2,000km) was tested on 14 April 1999. It, like a reported third version of the Ghauri (2,500–3,500km), is still under development.

Medium-range ballistic missiles entered the Pakistani arsenal only in 2003; the Shaheen-1 is of limited range (650km), and the Ghauri-1 is liquid-fuelled, and, therefore, less reliable and more vulnerable, as well as less accurate. Pakistan’s long-range strike capability (Shaheen-2, Ghauri-2) is still in its infancy. Pakistani planners refer to the Shaheen-2 as the ‘mainstay’ of the country’s future deterrent. TheGhauri-2 would be kept as a redundant or complementary component, especially since it can reach targets located in the centre of India. Some commentators note that the multiplication of missiles and bases make Pakistan a ‘target-rich’ environment and lessens the possibility of a pre-emptive strike.

Warheads and fissile material stockpile

Estimates of the number of warheads in Indian and Pakistani arsenals vary widely, and each country tends to make a worst-case assessment of the other. Based on the best available information, there does not appear to be a significant overall imbalance in favour of either side. Today, Pakistan probably has about 50 warheads. It also has a significant stockpile of fissile material, enough to build a much larger operational arsenal should it so decide. By the end of 2006, Pakistan was likely to have around 1,300–1,500 kg of HEU (enough for 65–75 weapons), after factoring in the 1998 tests. It also has a stockpile of up to 90kg of plutonium, enough for 11–15 weapons, assuming 6–8kg per weapon. By contrast, India has far less HEU, about 100–300kg, but far more separated plutonium, some 500kg (enough for some 100 weapons) after factoring in reactor fuel production and the 1998 tests. Although Pakistan has less strategic depth and fewer nuclear targets than India (which also has to consider Chinese capabilities), there is no obvious ‘Pakistani nuclear inferiority’ in South Asia.

It is reasonable to assume that Pakistan has at least two different basic nuclear weapon designs. Th, e first was, developed by PAEC and was intended to be carried by PAF aircraft. Its yield is reported to be 10–20 kilotons. The second is a 15–25kt HEU warhead of Chinese origin meant to be carried by aircraft or ballistic missiles. This design came from the fourth Chinese test in 1966. By 1983, US intelligence was aware that Pakistan was in possession of this design. In 1998, A.Q. Khan seemed to confirm the design’s origin when he stated that there was no technical need to proceed with hot tests, since Pakistan had a design ‘of proven reliability’. Based on the bomb design A.Q. Khan passed to Libya, the China/KRL warhead is less than one metre in diameter and weighs about 500kg.

There is a debate among analysts as to whether a warhead based on the Chinese design could be carried by Pakistani short-range missiles. For some, it is ‘much too big’. For others, the warhead was made for Chinese M-11 and was thus suitable for its Pakistani version. In any case, it is certainly suitable for longer-range Pakistani missiles.

It is unknown whether the original recipient of the Chinese weapons design was PAEC or KRL, and whether both laboratories worked from the same blueprint. As noted above, both M.A. Khan and A.Q. Khan worked separately on weaponisation, and as a result it is difficult to know ‘who did what’ (especially since PAEC did not know about KRL’s own weapons efforts until 1984). PAEC would have been the logical recipient of the Chinese design. However, documents found in Libya indicate A.Q. Khan was also working on the Chinese blueprint, so he may have been the one who in fact received the design. Several Western officials have stated that ‘all Pakistan’s atom bombs resemble designs that China tested in the late 1960s and passed on to Pakistan decades ago’. One source said that PAEC’s warhead ‘was so similar to Kahuta’s that Khan believed they had stolen his design’ (typical of Khan’s disparagement of anything relating to his rivals at PAEC). In fact, both parties have argued that their own design was passed along to the other one without their consent. This may very well have been true, a result of a conscious decision by Zia, who wanted both laboratories to work as fast as possible to get the weapon. The two designs may have been different. According to an American official, A.Q. Khan’s weapon ‘was inferior in terms of such things as size, power and efficiency’. However, A.Q. Khan disparaged PAEC’s weapon for being ‘bigger’, meaning less practical for delivery by a ballistic missile, especially on long-range missiles that have less throw-weight.

The tests carried out on 28 and 30 May 1998 were of low-yield, HEU fission weapons. The Pakistanis claim that six different nuclear devices ‘of varying designs, sizes and yields’ were tested. They gave different numbers and data as to the characteristics of the tests. Pakistani officials reported the detonation on 28 May of devices of 25kt and 12kt, and of three sub-kiloton devices; PAEC claimed that the total yield was 40–45kt. A.Q. Khan, for his part, claimed that one of the devices tested that day was a boosted fission device, designed to give a yield of 30–35kt, and that the other four other were tactical devices. For the 30 May test, he reported a yield of 15–18kt.

Uncertainties exist, however, as to the number and types of devices actually tested. Seismological data showed that the official number was almost certainly inaccurate. It was probably given for political reasons (one more test than India). The same data indicated very low yields, in the order of 10kt for 28 May and 5kt for 30 May. As far as the 28 May tests are concerned, US intelligence ultimately concluded that probably no more than two devices were detonated, with a total yield of about 6kt. This indicated that the tests were probably a partial failure, although due to a shortage of fissile materials the Pakistanis may have deliberately exploded ‘scaled-down’ versions of their weapons.

Questions also remain regarding the 30 May test. It was done at a different time, in a different location (Kharan) and with a different setup (vertical shaft). It was reportedly ‘a miniaturized device giving a yield which was 60% of the first tests’, a curious description that may have hidden a fizzle, i.e. a partial failure. It is possible that this was
‘A.Q. Khan’s chance to test’. But the most important question is whether or not it was a plutonium test. Initial US reports and a Los Alamos Laboratories analysis based on air sampling reported that this was the case, and although these findings were later disputed by Lawrence Livermore Laboratories, the question remains open. One possibility is that Pakistan had enough plutonium in 1998 to make such devices (dubious, but not impossible – see pages 20–21), in which case it would have been a PAEC test and not a KRL one. Another possibility is that one of the two laboratories had obtained plutonium from a foreign source (China or North Korea). A third possibility is that it may have been a ‘composite’ (uranium–plutonium) warhead. Finally, some authors have speculated that this may have been a ‘joint venture’ with Pyongyang, i.e. a North Korean test, although there is no credible evidence for this hypothesis.

Alert levels

Pakistani nuclear systems are kept in a low-alert form, probably to minimise the temptation of an Indian pre-emptive strike or the risk of un authorised launch. Missiles are not mated with warheads and the physics packages (the fissile cores) are not inserted into the warheads themselves. According to the Defence Ministry, the launch mechanism, the device and other mechanisms are kept at different places. Nuclear safety, physical security and access (maintenance) reasons all argue for separating the fissile cores from the warheads.

The time required for putting weapons on launch readiness is uncertain. According to a Pakistani scientist, assembling the weapons would only take ‘minutes’, while another account suggests that assembly would be done ‘within hours’. According to former COAS Mirza Aslam Beg, referring to the whole process of assembly and mating, ‘there would be a gap of hours, or even days before [a weapon] could be put together’. Consistently, Beg has also stated that the weapon components were stored ‘many miles away’ from delivery systems.

‘Minimum deterrence’ requirements

Pakistan’s force levels rely on a principle of ‘minimum deterrence’, which was first publicly announced in 1999. After the 1998 tests, Islamabad reportedly adopted a long-term development plan for its nuclear force, with a 15–20-year horizon in mind. A five-year plan for 2000–05 was also adopted. The desired level of minimum deterrence has reportedly been precisely defined. In fact, Islamabad stated in 2005 that it had reached this threshold. The exact significance of what minimum deterrence means to Pakistan is unclear and probably evolving, but a few elements can be inferred from official statements.

The Pakistani government has consistently said that it rejects the logic of parity, and that its goal is simply to attain the ability to inflict un acceptable damage on India. The Shaheen-2 and Ghauri-2 missiles will soon give Pakistan the possibility of covering the whole of mainland India. The difficulty of defining unacceptable damage to India is admitted by Pakistani planners: one quasi-official report by General Mahmud Durrani states that ‘because of the difficulty in predicting unacceptable damage, overkill would by necessity be built into the response’. Islamabad will err on the side of caution. It insists that the level of minimum deterrence can change over time, in light of the evolution of the threat. According to Foreign Minister Abdul Sattar in 1999, minimum deterrence ‘cannot be quantified in static numbers. The Indian build-up will necessitate review and reassessment. In order to ensure the survivability and credibility of the deterrent Pakistan will have to maintain, preserve and upgrade its capability.’ In particular, he later argued that the concept of sufficiency will be Pakistan’s guide in order that assets remain survivable against any surprise attack.

Guaranteed unacceptable damage implies survivability even after a first strike by the adversary. The Pakistanis are likely to use as a baseline planning assumption the possibility of an Indian pre-emptive strike coupled with the deployment of limited missile defence by New Delhi. Pakistani concerns about the possibility of an Indian first strike are well known, and have been compounded by the recent reinforcement of the US–India partnership. Despite future imports of US and Chinese conventional weapons, Islamabad fears that it will be outpaced by its adversary. India will probably move faster than its neighbour in improving its conventional equipments and command, control, communications and intelligence (C³I) capabilities (in addition to missile defence). One US expert estimates that New Delhi might even gain a ‘decisive’ edge in South Asia around 2020. Because the India–US nuclear cooperation agreement would free up domestic uranium for weapons purposes, it could conceivably give New Delhi the capability to fabricate an additional 12 HEU weapons or 75 plutonium weapons a year. In April 2006, the National Command Authority (NCA) stated that ‘in view of the fact that the [US–India] Agreement would enable India to produce significant quantities of fissile materials and nuclear weapons from unsafeguarded nuclear reactors, the NCA expressed firm resolve that our credible minimum deterrence requirements will be met’. This means that Islamabad is worried that India will be able to acquire a disarming strike capability.

Islamabad’s first military priority since the 2003 face-off with India has been to improve the conventional balance in South Asia. However, the Pakistani nuclear and ballistic missile force is still growing, and will continue to do so in the coming years. Musharraf stated in July 2005 that the country’s nuclear programme was progressing ‘ten times faster than before’. Hyperbole aside, there are three reasons for this growth. Firstly, Islamabad is undoubtedly not yet satisfied with its ability to inflict unacceptable damage on such a large country as India, especially with the low-yield warheads Pakistan is believed to have. Secondly, the Pakistanis are probably tempted to have flexible response and escalation dominance options, given their first-use doctrine (see below). Thirdly, a larger arsenal will protect Pakistan against the risk of an Indian first strike.

As an example of the type of calculations that Pakistani planners may make to that effect, a former SPD officer, using calculations which seem based on classic Cold War models, wrote that for a set of ten possible targets, a country may need as many as 68–70 warheads (without taking into account the risk of a pre-emptive strike).

It would also be logical for Pakistan to develop weapons with larger yields, using tritium (which significantly ‘boosts’ the weapon’s yield) and plutonium (which gives a better yield-to-weight ratio than HEU), and perhaps develop thermonuclear weapons (which would need to be tested). The construction of a second heavy water reactor at Khushab and the apparent construction of a reprocessing plant at Chashma signal that Pakistan indeed intends to significantly develop its plutonium weapons stockpile. Whether or not Pakistan already has operational plutonium weapons remains unclear: in October 2006, Musharraf flatly stated ‘we do not have a plutonium bomb’.

But guaranteed unacceptable damage to India can also be ensured through the protection of Pakistani assets. Islamabad could conceivably follow the ‘Chinese way’ and hide some of its missiles, which are all road-mobile, in mountainous areas. Baluchistan would be the best option for that (the north being another option, but roads there are few and tortuous). Due to its remoteness from India, Baluchistan could probably only host the long-range Shaheen-2 missiles (range 2,500km with a 1,000kg payload). A maritime component is also being considered. The 2004 report by Durrani states flatly that ‘Pakistan will work towards the development of a triad by giving the Pakistani Navy nuclear capability’. However, Pakistani planners do not refer to a naval nuclear-strike capability as a priority. Pakistan’s navy is a small coastal defence force, and would need major upgrading to support any kind of nuclear programme. Establishing a triad with a naval leg will depend on five factors: the scope of planned Indian missile-defence deployments; the build-up by India of its own maritime component; Pakistan’s confidence in the survivability of its land-based missiles; available resources and technical obstacles; and the navy’s ability to defend its interests. Such a component would rely on Babur cruise missiles (500km), for which a specific weapon may need to be developed.

Pakistan’s priorities in the nuclear domain for the coming 10–20 years are to improve survivability, penetrability and C³I capabilities. Pakistan’s nuclear programme can be expected to continue to be
military-led and needs-driven, with planning based on rational strategic calculations. Forecasts about the future evolution of the Pakistani nuclear arsenal must also take into account the unquantifiable impact of political debate and public opinion, which, to date, is very solidly in support of the nuclear programme, and unquestioning about its details. In particular, no Pakistani leader can afford to appear ‘weak’ vis-à-vis India. This may impact procurement decisions.

Arms control

It is unlikely that Pakistan would be the first of the four Asian nuclear-capable countries to ratify the Comprehensive Nuclear Test Ban Treaty (CTBT), unless Islamabad took the bold decision to do so after a final testing campaign – not unlike France in 1995. It is equally unlikely that Pakistan will be the first to test again. At the same time, if India were to test, Islamabad would probably seize the opportunity for both technical reasons (improving reliability and security, testing new designs) and political ones (settling the score again). Improving the plutonium formulas, testing a small warhead for the Babur missile, and perhaps also testing fusion designs, would be possible objectives of a second Pakistani testing campaign, which is all the more likely to happen since Pakistan, like India, has a small nuclear arsenal, which puts a premium on reliability.

There are thus three scenarios, ranging from the most likely to the least likely: (a) Pakistan resumes testing after an Indian testing campaign, CTBT signature and ratification then becomes an option; (b) Pakistan resumes testing after a critical design flaw is detected in one of its warheads formulas; (c) Pakistan announces that it is joining the CTBT after conducting a final testing campaign.

As noted above, Pakistan has produced a fairly large stockpile of fissile material. If Islamabad refrains from worse-case assumptions, this could make it easier for Pakistan to join a Fissile-Material Cut-off Treaty (FMCT). However, Pakistan will want to avoid any regime that would give a ‘perpetual edge’ to India in this regard. Therefore, Islamabad’s position is that three conditions should be met for any FMCT: (a) stockpile reductions should be progressive, (b) transfers of stockpiles to civilian use should be organised so that states with the largest stockpiles lead the way in a verifiable fashion, and (c) caps on future stocks should reduce asymmetries in existing stocks. Pakistani officials insist that any FMCT should be ‘non-discriminatory’ and ‘universal’.

The US–India nuclear deal is likely to induce caution in Pakistani minds. They are concerned that India’s plutonium stockpile, already bigger than Pakistan’s, will overwhelm Pakistan’s capabilities in the future. For political reasons, but also for strategic ones (the fear of pre-emption), Islamabad does not want to be outpaced by New Delhi. Thus Pakistan could not participate in a FMCT without India doing so as well, and in the process reducing any asymmetry.

All Pakistani reactors, with the exception of Khushab, are placed under IAEA safeguards (see chart, page 19). However, the Pakistani government admits that it would be difficult to separate installations dedicated to civilian use from those dedicated to military use.

Nuclear policy, doctrine and planning

Rationales for nuclear weapons

According to a quasi-official report, ‘Pakistan’s nuclear capability is solely for the purpose of deterrence of aggression and defence of sovereignty’.

As with most if not all nuclear weapons programmes, there is a political component in Pakistan’s drive for nuclear weapons. Z.A. Bhutto wanted Pakistan to ‘walk tall’. Maintaining equality with India was a primary motivation for him. Being the first Muslim nation to be endowed with nuclear weapons was also a matter of pride for the Pakistanis, and to this day the programme remains popular in Islamist circles. Domestically, popular support for the programme helped Z.A. Bhutto consolidated his political base and control of the army, which has always played a dominant role in politics as well as national security. Externally, nuclear status was a way to boost Pakistan’s prestige among Muslim nations, including rich Gulf monarchies.

This rationale continued to exist after the 1977 coup that toppled Z.A. Bhutto. In the late 1970s, one scholar noted, ‘Pakistan’s nuclear weapons programme became synonymous with national sovereignty and national prestige, even when it was run by the very military that had eliminated Pakistan’s best-known populist politician’.

But the primary rationale for the Pakistani bomb was always focused on security. Islamabad’s humiliating loss of East Pakistan in the 1971 war was a key motivation: Pakistan needed the bomb to ensure its very survival. This ‘never again’ rationale was bolstered by two considerations. One was the perceived inevitability of the Indian bomb after the 1974 test. Another was the perceived lack of a credible security guarantee, from China or the United States.

In Washington’s view, its pledge to defend Pakistan under the terms of the 5 March 1959 US–Pakistan Cooperation Agreement was operative only in case of communist aggression. Pakistan saw it differently and sought more comprehensive security guarantees. It was surprised in 1962 when Washington supported India against China. President John F. Kennedy provided a secret guarantee a few weeks after the war, valid only in case of Indian aggression. Islamabad was disappointed in 1965 and 1971 when the United States did not come to its support against India. (In 1965, Pakistan invoked the 1959 accord, but Washington chose to suspend aid to both parties. In 1971, Islamabad saw that the deployment of an US carrier battle group in the Gulf of Bengal did not have any impact on the conflict.) Pakistan came to believe that its security could be sacrificed on the altar of great-power politics. The Pakistanis continued to ask Washington for formal guarantees, in particular in 1980 and 1998, but to no avail. As Musharraf puts it, ‘we knew we could not count on American protection alone’.

Pakistani leaders regard China as their ‘reliable ally’ (to distinguish it from the ‘unreliable’ United States), but they are aware that even China cannot be counted on for help under all circumstances. When Pakistan sought assistance during its 1965 war with India, Beijing urged it to withdraw its forces from Indian territory. In the 1971 conflict, China officially sided with Pakistan but was of limited help in the face of Soviet threats to destroy China’s nascent nuclear force. During the 1999 Kargil crisis, Beijing appears to have privately echoed the US demand that Pakistan withdraw the infiltrators who had taken up positions on the Indian side of the Line of Control in Kashmir. Chinese leaders, who face their own Islamist separatist challenge in Xinjiang, have grown increasingly uncomfortable with the Pakistan-sponsored insurgency in Kashmir.

While Pakistan is the only country in the world allied to both the United States and China, Islamabad does not believe that either of the two would be ready to risk war to support Pakistan in case of new hostilities in South Asia. And, always worried about Indian conventional superiority, Pakistan considers nuclear weapons as a means to compensate for a possible defeat on the battlefield. Washington has made clear in recent years that its one-time policy of providing Pakistan a technological edge to counter India’s numerical superiority is no longer applicable, and in any case India’s military acquisition budget now dwarfs any foreseeable amount of US military aid.

Finally, an added perceived benefit of the nuclear programme – though probably not an initial motive – was that through it, Pakistan could pursue support for the Kashmir insurgency while deterring a massive Indian conventional retaliation. This misguided belief led to the disastrous Kargil adventure.

Nuclear policy

After the 1998 tests, Pakistan proceeded to a full-scope nuclear review, involving the National Defence College, to develop and test ideas and concepts. In late 1998, doctrine and organisation began to be redesigned. According to one source, the country adopted a three-point nuclear policy in early 2001: Islamabad would not be the first to resume nuclear testing, would not engage in a nuclear arms race with any country, and would not export nuclear technology. Another source reports that a new defence policy was adopted in March 2004. This policy reportedly intended to ‘further strengthen the process of institutionalization of control of strategic assets’ and ‘turn all policies and decisions from an invisible secrecy into a solid documentary form following the recent nuclear proliferation scandal.’

Nawaz Sharif, then prime minister, publicly announced a principle of ‘minimum credible deterrence’ in May 1999. Pakistan’s declared nuclear policy today is ‘to deter all forms of external aggression that can endanger our national security’ by maintaining a minimum credible deterrence. ‘Pakistan will not use or threaten to use nuclear weapons against non-nuclear weapon states’ and vows that it is against an open-ended arms race in South Asia. The overarching principles are said to be ‘restraint’ and ‘responsibility’. As noted above, Pakistan insists that it rejects the logic of parity; that is, it does not seek an arsenal equivalent to that of India. More recently, the Pakistani government has been using the expression ‘minimum defensive deterrence’. This semantic change (from ‘credible’ to ‘defensive’) may be a way for Pakistan to differentiate its concept from India’s stated policy of the same name.

According to Durrani’s report, these ideas translate into four objectives: (1) deterrence of all forms of external aggression; (2) building to this effect ‘an effective combination of conventional and strategic forces, at adequate levels within the resources constraints’; (3) avoiding a pre-emptive strike through protection and the threat of retaliation; and (4) stabilising strategic deterrence in South Asia.

Nuclear doctrine

Pakistan has consistently stated that its nuclear weapons are solely intended to deter military aggression. Officials stress that ‘the use of nuclear weapons as a war-fighting tool is not a contemplated doctrine in Pakistani strategic thinking’. A few statements have referred to ‘weapons of mass destruction’ (WMD), suggesting that the Pakistani deterrent may have a role in discouraging chemical or biological attacks. This is all the more likely after India stated in 2003 that its official policy of ‘no-first-use’ would be reviewed in case of a chemical or biological attack. However, Pakistan’s policy is also in line with the ‘negative security assurances’ given by nuclear weapon states: it will not use or threaten to use nuclear weapons against non-nuclear countries. This means, in practice, that Pakistan reserves the right to use nuclear weapons in response to a hypothetical Indian chemical or biological attack.

Pakistan also reserves the right to use nuclear weapons in response to non-strategic attacks by India as a ‘last resort’. There have been consistent statements by Pakistani officials since 1987 about the nuclear threshold: ‘if our existence is threatened’ (A.Q. Khan, 1987); to ‘defend the territorial integrity’ of Pakistan (Nawaz Sharif, 1999); if its ‘national integrity was threatened’ (Musharraf, 2000); ‘if its existence is threatened’ (Foreign Minister Inamul Haq, 2000); ‘only if the very existence of Pakistan as a state is threatened’ (Lt–Gen. Khalid Kidwai, 2001).

Pakistan also threatens nuclear retaliation in case of a preventive or pre-emptive strike. Announcing the 1998 tests, Nawaz Sharif stated that ‘these weapons are to deter aggression, whether nuclear or conventional’. Pakistan told India in 1998 that an attack against its nuclear installations (which are the subject of a non-aggression agreement between the two countries) would elicit ‘swift and massive retaliation with unforeseen consequences’. More precisely, according to Durrani’s report, the policy is ‘deterrence of Pakistan’s adversaries from attempting a counter-force strategy against its strategic assets by effectively securing the strategic assets and threatening nuclear retaliation should such an attempt be made’.

Since the 1998 nuclear tests, Pakistan has made a considerable effort to think through its nuclear doctrine, and to integrate the nuclear dimension into its defence strategy. This is made clear, for instance, by the fact that strategic force commanders are now invited to participate in corps commanders meetings. The definition of potential thresholds has been refined, at least in public statements by Pakistani officials. The most authoritative of these statements are made by SPD personnel, in the form of four thresholds which were first mentioned by Kidwai in late 2001.

· The spatial threshold. The penetration of Indian forces into Pakistani territory on a large scale may elicit a nuclear response. One can imagine that the critical distance would vary according to the location: the threshold could be low in Pakistani Kashmir because of the symbolic value of the region, and also in Punjab, the ‘core’ of Pakistani power. This would be particularly true if the major city of Lahore, located only 30km from the border, were threatened. Many analysts, including some Indians, believe that the Indus Valley, the ‘lifeline’ of Pakistan, is another ‘red line’ that Indian forces should not cross. The capture of key objectives in this crucial northeast–southwest axis (such as Multan, Rahimyar Khan, Sukkur or Hyderabad) might well provoke nuclear retaliation by Pakistan.

· The military threshold. The destruction of a large part of Pakistani land or air forces could lead to a nuclear response if Islamabad believed that it was losing the cohesiveness of its defence and feared imminent defeat. Here Pakistani thinking is identical to the guidelines given to the NATO commanders during the Cold War. This criterion is even more important for the Pakistani army because of its critical role in maintaining the country’s stability. As noted above, an attack on a nuclear installation has also been posited as a threshold.

· The economic threshold. Economic strangulation is also a potential Pakistani red line. This primarily refers to a potential Indian Navy blockade of Karachi (the main port), or the stopping of the Indus water flow (the ‘lifeline’ of Pakistan). It could also refer to the capture of vital arteries such as the Indus.

· The political threshold. Finally, Pakistani planners suggest that a destabilisation of the country by India could also be a nuclear threshold if Islamabad believed that the integrity of the country were at stake. Stated scenarios are political destabilisation or large-scale internal destabilisation. One example would be encouraging the breakaway of one or more Pakistani provinces.

Pakistani planners insist that these thresholds are indicative scenarios and that they should not be viewed in isolation one from another.

Nuclear planning

Like India’s statements, Pakistani rhetoric on nuclear planning is oriented towards massive retaliation. As early as 1987, Zia is reported to have told Prime Minister Rajiv Gandhi: ‘if your forces cross our borders by an inch, we are going to annihilate your cities’. In 2002, Musharraf used similar language. He stated that ‘any incursion by the Indian forces across the Line of Control will unleash a storm that will sweep the enemy’ and he underlined Pakistan’s ability to inflict ‘unbearable damage to the enemy’. The classic expression ‘unacceptable damage’ is widely used in Pakistan. Islamabad believes in the theory of ‘deterrence of the strong by the weak’; that is, the possibility of a smaller country deterring a larger one through the threat of damage incommensurate with the stakes of the conflict. American scholar Peter Lavoy emphasises that escalation dominance operates at all rungs of the military ladder – from low-intensity conflict through to nuclear war.

It is thus assumed that Pakistani nuclear planning would target Indian cities, due to the relatively low number of warheads it is believed to have, and the poor accuracy of most missiles currently in service (see page 23). Musharraf is reported as saying that Pakistan’s aim is to be able to have ‘enough missile capacity to reach anywhere in India and destroy a few cities, if required’. Pakistani analysts regularly mention numbers in the region of a dozen cities. Delhi is probably first and foremost amongst them: because Pakistan has a small number of low-yield warheads, it is likely to have a strategy akin to that of the United Kingdom during the Cold War, which primarily targeted Moscow. It has also been suggested that a deterrence strategy that threatens large populations is justified in Pakistan in Islamic terms as a suitable way of employing ‘terror’ in warfare. However, it is also likely that as its nuclear force grows and evolves, Pakistan will diversify its set of potential targets, as other nuclear-capable countries have done. In a discussion of the ‘pain threshold of the opponent’, Pakistan Army Brigadier Naeem Ahmad Salik identifies possible targets as ‘major population centres, industrial complexes, major military bases, and communication hubs’.

Islamabad will certainly want to avoid an all-or-nothing strategy, both to reserve forces for a second strike and for other strategic and political reasons. It has certainly also developed limited options, on Indian territory but also perhaps on its own territory. Invading Indian forces might be targeted by low-yield weapons. The goal of such a strike might be not only to signal resolve, but also perhaps to force Washington to intervene, politically or even militarily. Technically, a strike on an Indian formation would be feasible without excessive collateral damage: many areas along the border are sparsely populated, and prevailing winds blow eastward.

Being in a situation of perceived conventional inferiority vis-à-vis a mortal enemy, Pakistan’s conception of nuclear planning is close to NATO Cold War thinking, and its employment policy may very well look like flexible response. Despite Islamabad’s rhetoric about a massive retaliation-type response – a choice that some commentators believe would be logical given the lack of strategic depth – it is likely that Pakistani doctrine allows for some limited nuclear options. Pakistani interest in Western concepts of controlled escalation can be traced back to US military training given in the late 1950s, both in Pakistan at the Quetta staff college and in the United States. Given the small size of its force, however, a more appropriate reference might be the French two-rung escalation ladder of a final warning followed if needed by unacceptable damage.

Pakistan's nuclear programme and imports - [707 KB]view this chapter as a PDF document

Nuclear Black Market order form - [225 KB]Download a copy of the Nuclear Black Market order form as a PDF file

Nuclear Black Market Dossier Press Statement - [263 KB]View this document in PDF format

Nuclear Black Market Flyer - [393 KB]

Download the marketing flyer for Nuclear Black Marke