Copyrighted Material
-For Private Use Only-

Cover art by Jon Lomberg
Cover Art by Jon Lomberg

The New Nuclear Nations

Leonard S. Spector

A Carnegie Endowment Book
Vintage Books
A Division of Random House
New York

A Vintage Original, November 1985
First Edition
Copyright (D 1985 by The Carnegic Endowment for International Peace.
All rights reserved under International and Pan-American
Copyright Conventions. Published in the United States
by Random House, Inc., New York, and simultaneously
in Canada by Random House of Canada Limited, Toronto.
Library of Congress Cataloging in Publication Data
Spector, Leonard S.
The new nuclear nations.
- A Carnegie Endowment book."
- A Vintage special.
Includes index.
1. Nuclear weapons. 2. World politics 1975-1985.
1. Title.
U264.S634 1985 355X25119 85-40322
ISBN 0-394-74189-7 (pbk.)
Manufactured in the United States of America
Cover art by Jon Lomberg.

Chapter II
Nuclear Netherworlds

As for the offers submitted to Iraq, the Chief of Armaments told us they were interested in SAM-7, mortars, AH-1s armed with TOWs, plutonium, [and] Glauco's rocket system.
Telex from Glauco Partel and Ivan Galileos, Rome, to Eugene Bartholomeus, Sydney, Australia (March 30, 1982), disclosed in Italian court files, mid-1984.

A netherworld of covert dealings in nuclear technology and materials has existed since the dawn of the atomic age. It began in the 1940s with Soviet atom bomb spies and continues today with nuclear blackmarketeers and conspiring generals. From mid 1984 to mid-1985, a wealth of new information has come to light providing a closer look than ever before at parts of it and at the challenge of attempting to police it.

At the center of all the machinations is a single goal: possession of the knowledge or the materials necessary for the development of nuclear weapons. So far, the hidden principals have been national governments; individuals and corporations have served only as instruments of national nuclear programs-often at great profit to themselves.

The national governments have been forced to practice secrecy and deception to evade international controls and diplomatic actions aimed at halting the spread of nuclear weapons and sometimes to avert domestic opposition as well. The private parties abetting these national endeavors have found similar ways to avoid prosecution or to shield themselves from the adverse publicity that would follow disclosure of their part in suspected nuclear-weapons programs. Outright smuggling; the quiet exploitation of loopholes in export controls; the careful selection of above-board, supposedly peaceful, civilian nuclear technologies to advance undisclosed military objectives; and secret nuclear-weapons research-and-development efforts are all part of the nuclear netherworld.

These activities are recent manifestations of a long-standing but rarely acknowledged pattern. As early as 1942, some three years before the United States tested the first atom bomb, Manhattan Project officials had begun negotiating a series of secret agreements with the Belgian Congo, India, Brazil, and other nations in a surreptitious attempt to obtain a postwar monopoly of the world's known uranium resources.' Wartime nuclear spying by the Soviet Union in the United States, Canada, and Britain, and French advances toward the, bomb in the 1950s, under the veil of its civil nuclear program, are more widely known examples of clandestine nuclear activities involving the principal nuclear powers. 2 Today's emerging nuclear states have followed their example. Israel secretly acquired the Dimona reactor from France in the late 1950s, and only when the United States discovered the partially completed nuclear plant in 1960 did Israel admit its existence. It is now thought to be the backbone of Israel's presumed nuclear- weapons program. In addition, the Central Intelligence Agency (CIA) has long believed that during the mid-1960s, Jerusalem diverted more than 100 kilograms (220 pounds) of highly enriched uranium, enough for four rudimentary weapons, from a private U.S. processing plant at Apollo, Pennsylvania. 3 In 1968, 200 metric tons (220 short tons) of uranium concentrate disappeared while in transit on the Mediterranean Sea from Antwerp, Belgium, to Genoa, Italy. American and European officials are convinced that the material - essential for the long-term operation of the Dimona reactor - wound up in Israel (4).

In the early 1970s, India quietly decided to use nuclear equipment and materials imported from Canada and the United States, ostensibly for civilian research and training, to build nuclear explosives. Although New Delhi -later claimed that its May 1974 nuclear detonation-its only one to date-was the test of a "peaceful nuclear explosive," Ottawa and Washington considered the diversion of their exports a violation of the terms under which they had provided them.

Similarly, from 1970 to 1975, as South Africa built its pilot enrichment plant at Valindaba, obtaining needed equipment from the United States and other Western nuclear suppliers (and possibly covert aid from West German firms), it repeatedly declared the exclusively peaceful aims of the installation (5). But in 1977, when a Soviet spy satellite observed Pretoria readying a nuclear test site in the Kalahari desert and the United States confirmed the finding, the conclusion was that the plant, South Africa's exclusive source of weapons-usable material, was being put to hidden military ends. (6)

In the mid-1970s, a new danger emerged as a number of nations, thought to be pursuing nuclear weapons capabilities, openly sought to purchase sensitive reprocessing and enrichment plants, which would give them direct access to nuclear-weapons material. In 1975 and 1976, for example, South Korea and Pakistan both tried to buy large-scale reprocessing plants from France; the deals reached varying stages of completion before being canceled. Also in 1975, Brazil concluded a deal with West Germany for the transfer of both reprocessing and enrichment technology. In 1976, Iraq ordered the large Osiraq research reactor from France and radiochemistry laboratories from Italy; by June 1981, when Osiraq was destroyed by Israel, Iraq was negotiating with Rome for a larger reactor and a reprocessing installation.

Although all of these transactions were defended as necessary for nuclear-energy development and would have been covered by International Atomic Energy Agency (IAEA) safeguards, there is solid reason for concern that these facilities were, in fact, being acquired for more suspect purposes. As a result, in 1976, all of the major nuclear supplier states agreed to "exercise restraint" in transferring such plants, and several, including the United States, France, and West Germany, largely ruled out such sales altogether. Indeed, none has been consummated since.

Today the emerging nuclear states are pursuing two new strategies for developing the installations and technologies needed for a nuclear-weapons option: reliance on indigenous resources and resort to covert, and often illegal, nuclear purchases from more advanced (usually Western) states. India's new reactors and reprocessing units, Brazil's classified uranium en- research, Israel's possible enlargement of the Dimona reactor, and North Korea's recently disclosed construction of a large research reactor-the keys to each of those countries' nuclear-weapons potential--all appear to be essentially indigenous enterprises. Pakistan's Kahuta enrichment plant and Argentina's Pilcaniyeu enrichment facility and Ezeiza reprocessing plant, each of which could provide direct access to nuclear-weapons material, are indigenous to a degree but rely, in part, on foreign technology and equipment acquired piece by piece through "back-channel" or clandestine means. In 1982, moreover, Syria may have tried to purchase complete nuclear weapons and Iraq, plutonium, from an Italian arms smuggling ring purporting to offer these for sale.

During the past year, seven prosecutions for nuclear smuggling to Pakistan were completed in trial or appellate courts in the Netherlands, West Germany, the United States, and Canada, and an American businessman was indicted in Los Angeles for similar illegal exports to Israel. Extensive data on alleged trafficking in nuclear arms and plutonium in the Middle East were disclosed in Italian court files. Congressional and press investigations in the United States brought to light South Africa's hiring of American nuclear power plant operators, in apparent violation of U.S. regulations -one of many cases highlighting illicit commerce in nuclear skills. Swiss, Italian, U.S., and West German firms' quiet assistance to Argentina's enrichment and reprocessing efforts was also reported. And a Swedish technical journal exposed the details of Sweden's secret nuclear-weapons research program during the 1960s and early 1970s. These developments are the subject of this chapter.

Troubling as these events are, they do not suggest the existence of a full-scale nuclear black-market comparable to those in conventional arms or illicit drugs. So far, the number of countries that have turned to covert nuclear commerce remains relatively small, and only a few transactions have involved the most-sensitive nuclear equipment or materials. Some purchases, though injurious to international non-proliferation efforts, are perfectly. legal under today's rules governing nuclear commerce, since what is bought and sold are hard-to- regulate, "dual-use" items with wide application in non-nuclear industrial settings. Most deals, moreover, are pursued through ad hoc networks set up by the purchasing country, not through independent smuggling rings like those in other blackmarkets.

Nevertheless, one clear lesson emerges from all these events. Despite the significant tightening of various controls since the early 1970s, including those covering dual-use commodities, today's international non- proliferation efforts have not been sufficient to eradicate the nuclear netherworld or to deter new entrants, governmental and private, from operating within it. Largely for fear of offending allies and potential friends in regions of East-West competition, the United States and its European partners, the governments best placed to take action, have been unwilling to deal forcefully with governments that violate existing standards of nuclear conduct, directly or through intermediaries. Yet without such decisive action, the spread of nuclear arms will remain a continuing menace.

Pakistani Prosecutions

Between March 1977 and April 1980, Albrecht Migule, head of the tiny West German engineering firm CES Kalthoff GmbH, dispatched sixty-two truckloads of equipment to Pakistan. Purchased from firms throughout West Germany, which apparently knew, nothing of the ultimate destination, the equipment was - assembled near the town of Dera Ghazi Khan, under the supervision of Migule's West German associates, into a complete installation for transforming natural uranium into uranium hexafluoride, a critical step in the process of improving it to weapons-grade uranium. Migule was eventually prosecuted in West Germany for illegally exporting the plant. As a West German official later testified at his trial, "What Migule sent .the Pakistani atom bomb program was like yeast to a cake." (7)

Migule signed three related contracts between November 1976 and March 1977 to provide the installation for a total of $6 million (DM 15 million).(8) By this time Pakistan was already developing a gas-centrifuge-type uranium- enrichment plant, the dauntingly complex installation necessary to produce weapons-usable, highly enriched uranium from the uranium hexafluoride manufactured in Migule's plant .

At the center of this effort was Dr. Abdul Qadir Khan, a German-trained Pakistani metallurgist. Between 1972 and 1975, Khan worked in the Netherlands for a subcontractor of URENCO, the British-Dutch- German uranium- enrichment consortium. While there, he had been given access to sensitive data concerning the URENCO gas- centrifuge enrichment plant at Almelo, Holland. After a series of incidents in mid-1975 suggested that Khan might be sending classified enrichment technology to Pakistan, Dutch authorities banned him from further enrichment- related work. Within months, Khan fled to Pakistan where he was placed in charge of the centrifuge enrichment program. In 1984, Pakistan's principal enrichment facility, located at Kahuta, was named after him.

A 1979 Dutch government report on the "Khan Affair" concluded that Khan had provided Pakistan with "essential gas centrifuge know-how." Khan will never be prosecuted for his espionage through 1975, however, even in absentia, because it is virtually impossible to establish what information he actually transferred. (9) Nevertheless, it is now quite clear that

throughout the latter half of the 1970s, Khan and a network of other Pakistani operatives conspired to obtain outside help for Pakistan's enrichment program, principally from firms and individuals in the West. (10)

In December 1976, just after Migule signed his first contract for the uranium hexafluoride plant, Khan wrote a colleague at his former place of employment the URENCO subcontractor, FDO (11) - asking for information on the technique of chemically etching tiny holes on "precision disks," a classified process used in the URENCO gas centrifuge project. ' The following summer, he solicited from a second former FDO colleague classified data on special shock absorbers for the high-speed centrifuges. Both requests were re fused and ultimately reported by FDO to the Dutch authorities. They formed the basis for Khan's indictment in 1983.

What FDO did not report, however, was its sale to Pakistan of specially designed, computerized, measuring machinery in 1977. The apparently unique items had been fabricated originally for use at the URENCO enrichment plant at Almelo, but URENCO backed out of the purchase. This left FDO, a small research firm, with a costly piece of equipment on its hands, until, as one Dutch official put it, like a "gift from heaven" aides to A.Q. Khan purchased the item. Although the equipment itself was not mentioned on the Dutch export-control list, the machinery contained computers and other devices that were. In 1983, FDO's director and sales manager were indicted for failing to obtain an export license for the sale.

Roughly coinciding with these activities were Pakistani efforts to purchase 6,500 specially hardened steel tubes from Van Doorne Transmissie, a Dutch manufacturer of auto transmissions partly owned by the Dutch government. In centrifuge-enrichment plants, tubes of this type apparently serve as the actual centrifuges, the tall spinning drums in which uranium hexafluoride gas is swirled at high speed so that the desired uranium isotope, uranium-235, can be concentrated. According to Dutch officials, the blueprints for the tubes were supplied to Van Doome Transmissie by purchasing agents at the Pakistani Embassy in Brussels.

Van Doorne made and exported the tubes from 1976 until late 1979. Reviewing the matter during its 1979 investigation into Khan's earlier activities at FDO, the Dutch government reported that Van Doorne had made the exports despite repeated oral and written warnings not to do so. These tubes were recognizable as parts of a gas centrifuge system, a fact of which the management of the company was also apparently aware, since, during a visit which an official of the Ministry of Economic Affairs paid to the company in 1977, he was, he states, informed by the managing director that the tubes were going to be sent to Pakistan and would be used in a gas centrifuge project . (12)

In 1983, Van Doorne Transmissie was indicted for failure to obtain a license for the export of hardened steel tubes "especially designed and prepared for use in gas centrifuges. "(13)

Khan had another contact in the Netherlands, Henk Slebos. 14 Throughout the late 1970s and early 1980s, Slebos served as a purchasing agent for the Pakistani nuclear program, forwarding such items as 10,000 steel balls- which could be used for the bases of the Kahuta centrifuges-and various aluminum components Dutch officials believed were intended for the plant. Slebos persisted in his export activities despite repeated warnings from the Dutch Foreign and Economics Ministries. The Dutch authorities were, however, powerless to block his actions because the items exported, which were unspecialized and in widespread commercial use, were not on the Dutch export-control list.

In late 1983, Slebos tried to export a high-speed oscilloscope manufactured in the United States by the Tektronics Corporation. The addressee was a company known as Assaf Electrical Establishment in the United Arab Emirates. Dutch authorities later discovered correspondence that showed the ultimate destination of the equipment to be Pakistan and the purchaser, a General Aziz, believed to be associated with the Pakistani nuclear program. (15)

High-speed oscilloscopes are on the Dutch export control list. Slebos did not obtain the special license required for such exports, however, but merely relabeled the equipment and marked it on his customs declaration as a type authorized for export.

At this point, Slebos's luck ran out, but only, it seems, because Dutch customs agents, in the midst of a labor dispute, were engaged in a "work-to-rules" slowdown, under which they were methodically opening every package being sent abroad rather than merely the usual token number. The mislabeled oscilloscope was discovered, and Slebos was arrested in early 1984. He was subsequently charged with exporting controlled goods without the required license.

Khan's network also reached North America. From 1977 to 1980, he obtained the aid of Abdul Aziz Khan, a Pakistani-born electrical engineer with Canadian citizenship. According to correspondence between the two, which was later seized by the Canadian government, Abdul Aziz Khan visited Pakistan on several occasions to work on A.Q. Khan's project. 16 In mid 1980, Abdul Aziz Khan and two accomplices were useful in helping A. Q. Khan overcome the problem of acquiring high-frequency electrical inverters. Pakistan. had previously obtained the devices, needed to regulate the speed of fast-spinning centrifuges, from Britain and the United States, but in 1978 and 1979, the two governments persuaded the companies manufacturing the dual-use item to stop further sales to Islamabad.

In August 1980, Abdul Aziz Khan and his two accomplices were arrested at Montreal's Mirabel Airport for attempting to export nineteen boxes of inverter components marked with the innocuous label "condensers and resistors." This was apparently the tenth of a series of shipments, nine of which had eluded the authorities. According to Canadian court records, two Pakistani officials linked to the country's nuclear program had come to Canada in the summer of 1980 with a list of the specific items needed. Operating through two small Montreal electrical-equipment stores, Abdul Aziz Khan and his friends purchased the components from unsuspecting U.S. and Canadian manufacturers. The items were subsequently repackaged and shipped to Pakistan or to front companies in the Middle East for transshipment. The Canadian government indicted all three for exporting inverters without the required special export license. (17)

Pakistan's covert purchasing activities extended beyond the acquisition of commodities necessary for uranium enrichment. From the mid-1970s, Pakistan had also been seeking to build installations that would enable it to produce plutonium, an alternative to highly enriched uranium for use in nuclear weapons. Pakistan already had a sizeable nuclear power reactor at Karachi, which could produce significant quantities of plutonium in its spent fuel, and it was working on a reprocessing plant for extracting pure plutonium from other spent fuel constituents. In September 1981, the nation began producing its own fuel for the Canadian supplied Karachi reactor, a plant, unlike the enrichment plant A.Q. Khan was building at Kahuta and Migule's uranium hexafluoride plant near Dera Ghazi Khan, subject to IAEA inspection. But once Pakistan began introducing its own fuel into the reactor, rather than the Canadian fuel the Agency had inventoried on arrival, IAEA found it much more difficult to verify that spent fuel was not being secretly diverted from the facility. Between late 1981 and early 1983, when IAEA safeguards on the installation were tightened considerably, U.S. and other observers feared Pakistan might be secretly withdrawing plutonium-bearing spent fuel from the plant.

On October 31, 1981, U.S. Customs agents at New York's Kennedy Airport seized a 5,000-pound shipment of zirconium metal marked "mountain climbing equipment" just as it was being placed aboard a Pakistan Airlines jet as baggage. Zirconium, which is used to sheathe nuclear-reactor fuel, was an essential part of Pakistan's indigenous fuel-production program. The purchaser of the material, whom authorities were unable to locate, was reported to be Dr. Sarfraz Mir, a retired Pakistani army colonel and close friend of Pakistani President Zia ul-Haq. (18) 18 Mir's company, S. J. Enterprises (apparently yet another Pakistani front corporation), purchased the material from the manufacturer, Teledyne Wah Chang, which reportedly alerted authorities to the order. The actual attempt to export the metal was made by Albert A. Goldberg, head of the National Tronics Company, who had been penalized twice before for making illegal exports and whose export privileges had -been suspended for three years by the U.S. Commerce Department in 1976.

On November 22, 1981, in an action that foreshadowed the disposition of all subsequent prosecutions for nuclear smuggling involving Pakistan, the United States decided not to indict either Goldberg or Mir for criminal acts but to permit the Commerce Department to impose civil penalties against them. These penalties, which could have included substantial fines, were limited to the indefinite revocation of both men's export privileges. At the time, the United States was attempting to respond to the Soviet invasion of Afghanistan by strengthening ties to Pakistan with a mammoth $3.2 billion military and economic aid package. The State Department hailed the case as a major non-proliferation success. (19)

Whatever the deterrent effect of the Goldberg-Mir penalty, Pakistan's smuggling activities in the United States and Europe appear to have continued and, indeed, to have taken a still more ominous course as the nation began to seek the components used in nuclear weapons themselves; previous efforts had involved merely the facilities necessary to manufacture weapons-grade nuclear materials. Reportedly, these new efforts included unsuccessful attempts to obtain precision-shaped, high-explosive charges (known as lenses) and special steel plates for the non-nuclear triggers of atomic weapons, (20) as well as perfectly sculpted, hollow, steel spheres to serve as the internal support structure for the bomb .(21) Pakistan is also reported to have obtained atomic-weapon design information from China. (22)

In June 1984, one Nazir Ahmed Vaid was arrested along with two accomplices by U.S. Customs officers for attempting to smuggle fifty KN-22 krytrons -highspeed electronic switches used in atomic-weapon triggers- out of the United States. Although the devices have other high-technology uses, such as triggering lasers, Pakistan's modest level of technological achievement, the relatively large number of krytrons sought, and evidence that has subsequently emerged linking the purchase to the Pakistani nuclear program leave little room for doubt that the devices were intended for nuclear weapons. Indeed, the episode is widely viewed as the "smoking gun," removing any question as to Pakistan's intention to acquire nuclear arms.

Eight months earlier, on October 18, 1983, Vaid, who ran a Karachi-based freight-forwarding company with a branch in Houston, had attempted to purchase the switches in person from their Massachusetts manufacturer, the Electro-Optics Division of E G & G, a major U.S. corporation. Vaid claimed he was buying the devices for the University of Islamabad.

E G & G officials, startled by this unparalleled attempt to buy the items over-the-counter, advised Vaid that a license would be needed to export the krytrons. (Vaid is said to have been wearing Pakistani-style clothing at the time.) Vaid then attempted to order the parts from E G & G by telephone and was turned down again. Finally, on October 31, 1983, he ordered them through a small Houston electrical supply store, Electrotex. Under the direction of U.S. authorities, whom E G & G officials had contacted after Vaid's initial visit, E G & G supplied the items to Electrotex. Law enforcement officials tracked the shipment until Vaid consigned it to a freight forwarder on June 22, 1984 for export to Pakistan-without a license. Vaid was immediately indicted for violating U.S. export laws. (23) (Two Electrotex employees arrested with him were not charged.)

Between mid-1984 and mid-1985, the criminal justice systems of two West European nations, Canada, and the United States acted on the cases of Migule, A.Q. Khan, FDO, Van Doorne Transmissie, Henk Slebos, Abdul Aziz Khan, and Vaid. on March 6, 1985, nine years after he had signed the first contract On the deal, Albrecht Migule was convicted of causing substantial harm to West German foreign relations by illegally exporting a uranium hexafluoride plant to Pakistan. At his trial he claimed that he had never actually exported the elements of a uranium hexafluoride plant, but only miscellaneous la - oratory equipment. (24) Because the customs records id not resolve the matter, the prosecution was force to rely on the three contracts Migule had signed in 1976 and 1977, providing for-the construction of a hexafluoride facility. But Migule asserted these were only drafts that he never executed; the real contracts, which he presented to the court shortly before trial, were, according to him, three others that had been signed at approximately the same time as the first three, but that promised delivery of legally exportable laboratory equipment. Migule's associates who had worked for him in Pakistan also gave conflicting testimony on what had been built near Dera Ghazi Khan.

As the trial proceeded, Migule's detailed invoices allowed West German government experts to demonstrate that parts for a hexafluoride facility were actually being shipped. In addition, an August 1977 progress report, which Migule prepared at a time when he claimed he was filling the second, innocent set of contracts, specifically referred to the steps he had taken in fulfillment of the first. As this part of his case fell apart, Migule gradually acknowledged that his exports might be usable to produce uranium hexafluoride, although he denied that this had been his intent, and he went so far as to correct a drawing made by one witness of the layout of the various parts of the hexafluoride installation-without challenging that the parts had in fact been built.

This did not end the matter, however. The prosecution still had to prove that the export had caused "substantial harm to the foreign relations of the Federal Republic of Germany," a key portion of German export law that had never before been applied and whose exact meaning had never been determined. Migule's lawyers claimed that the government had to show specific instances where West German foreign relations had suffered, such as expressions of disapproval by other states. In what must have been a painful moment, West German non-proliferation officials pointed to an acid statement by U.S. Senator Alan Cranston, for many years an outspoken critic of German nuclear-export practices, condemning Bonn for having allowed the hexafluoride plant to slip through its export-control net.

The prosecution also argued more broadly 'that Migule's exports undermined Germany's ability to comply with its obligation under the Non-Proliferation Treaty (NPT) and thus automatically caused substantial harm to t ' he nation's foreign relations. Under the NPT, nuclear-supplier countries have pledged to make nuclear exports only if recipient states agree to place the exported item under IAEA inspection. Supplier countries' nuclear-export licensing systems are set up to ensure that this stipulation has been met before individual exports are allowed. In bypassing the licensing process, Migule allowed Pakistan to obtain an unsafeguarded installation from Germany, directly undercutting the NPT pledge.

In the end; the court, a low-ranking provincial tribunal, accepted both the prosecution's version of the facts and its precedent-setting interpretation of Germany's federal export-control statute. Migule was found guilty on March 6, 1985, and received an eight month prison term (out of a possible maximum of three years) and a DM 30,000 (roughly $10,000) fine. He will not have to serve the prison term, however, if he completes five years of probation, and his fine is undoubtedly a small fraction of the profit on his $6 million deal. Migule has also appealed the conviction and will, in effect, be able to have the entire case retried, including the government's interpretation 1 of the export law. Migule's accomplices, who supervised the construction of the plant, were not indicted because the export of know-how does not require a license under German law.

An Amsterdam court tried A. Q. Khan in absentia in October 1983 for soliciting classified centrifuge enrichment technology from two of his former colleagues at FDO. Khan was not represented at the trial, and the prosecution, in possession of the letters he had written the two and facing no rebuttal of its assertion that the information sought was properly classified, had little difficulty establishing his guilt. On November 14, 1983, he was given the maximum sentence of four years in prison, a sentence that could not be carried out unless he returned to the Netherlands. (25)

Trials in absentia are not uncommon under Dutch law, which permits absent defendants to be represented by counsel, although the privilege is rarely exercised. One critical procedural requirement, however, is that defendants be formally advised of the charges against them and of their impending trial. In 1981, when the indictment against Khan was filed, he was in Pakistan, and his place of residence was apparently known to Dutch prosecuting officials- The normal practice, however, is to serve defendants in foreign countries through their governments; accordingly the Dutch embassy in Islamabad forwarded the summons of the Amsterdam court to officials there. No reply was received. until the day after Khan's conviction, when the Dutch were advised that the Pakistani government was unable to serve Khan because no specialized legal treaty had been signed between the two nations.

Khan promptly appealed., Through his newly retained Dutch counsel and in personal interviews he asserted that his trial had been unfair because he had-' been unaware of the case and argued that, in any event, the information he had solicited was in the public domain. On March 28, 1985, a Dutch appeals court overturned the conviction on the procedural ground that he had not been served proper notice of the proceedings against him.

It is uncertain whether Khan will be retried. Now that he has appeared through counsel and it is clear that he cannot be served through the Pakistani government, he can be sent notice of the proceedings by registered mail and will be unable to raise his procedural defense again. Nevertheless, while the Netherlands and its two partners in the URENCO gas-centrifuge enrichment consortium, Great Britain and West Germany, still have good reason to pursue Khan, countervailing factors may lead to the case being dropped. The desire to avoid injuring relations with Pakistan and the danger that Khan might be acquitted if he could show that the data he sought was publicly available are substantial points weighing against retrial. But whether or not he is retried, it appears highly unlikely that. he will suffer any penalty for his actions in the Netherlands.

Nor does it appear that FDO and its former director and sales manager will be penalized for supplying custom- made computerized measuring equipment for Pakistan's enrichment effort. These defendants were acquitted on November 13, 1984, after the prosecution was unable to establish that the devices they exported were encompassed by the list of items requiring special export licenses-documents FDO had failed to obtain. The court rejected the prosecution's argument that because restricted computer and other technology were embedded in the exported equipment, FDO's exports also required a special permit.

One problem the prosecution faced was that the case was tried some seven years after the acts in question occurred. In the interim, computer items that had been tightly controlled in 1977, such as floppy disks and relatively small memory chips, were in such widespread use that the prosecution had difficulty demonstrating the seriousness of FDO's offense.

Because the legal theories in the case could have wider significance, the Dutch government may appeal the court decision. The prosecution, it may be noted, was seeking only monetary penalties of 2,500 guilders ($700) from the individual defendants and 7,500 guilders ($2,100) from the company.

The outcome was little different in the prosecution of Van Doorne Transmissie for the export of 6,500 centrifuge tubes to Pakistan. Once again, because of the intricacies of the Dutch export control list -a list agreed to- in 1976 with other nuclear-supplier countries -the prosecution had to establish that the tubes had been "specially designed and prepared" for use in gas centrifuges. Although a Van Doorne official had apparently told a Dutch Economics Ministry aide that he knew the tubes were destined for use in such a facility, at the trial the company argued that the tubes were not specially designed for that use but could serve other purposes as well, such as in aircraft landing gear and auto transmissions. When the government's key technical witness conceded that such uses could not be ruled out, the prosecution's case fell apart. Van Doorne Transmissie was acquitted, and no appeal by the prosecution is planned. The prosecution, it may be noted, had sought $6,000 fines and one-month suspended sentences for the two company officials indicted. The Nuclear Suppliers Guidelines and the corresponding Dutch nuclear export control list have since been upgraded to identify by their technical specifications hardened steel tubes whose export is prohibited without a license.

In 1985, Dutch officials did succeed in convicting Henk Slebos for the export of a high-speed oscilloscope. In comparison to their previous nuclear prosecutions, the prosecutors' work was made easier by the fact that the item was specifically listed on the export-control list. The offense charged was exporting the item without the necessary special license. Slebos was sentenced to a year in prison but is free on bond while he pursues an appeal. (26)

Like their Dutch counterparts Canadian prosecutors had only limited success in their case against Abdul Aziz Khan and his accomplices. (27) Here, proof was required that the defendants had exported inverters especially designed and prepared for use in an enrichment plant. Because the defendants had exported not complete inverters but various electronic parts, the prosecution had first to show that these added up to inverters and then to establish that they were intended for nuclear uses (the point that undid the Van Doorne Transmissie case).

A set of rulings prevented the introduction of evidence-which the court said was hearsay-linking A. Q ~ Khan to Pakistan's centrifuge-enrichment program and the two Pakistanis who came to Canada with lists of the needed parts to the Pakistan Atomic Energy Commission .21 In the end, Abdul Aziz Khan was acquitted on July 13, 1984, and the two others convicted only of lesser charges, which carried a fine of $3,000 for each.

In the United States, authorities were able to convict Nazir Vaid for the attempted illegal export of fifty krytrons. The key to the case was a series of subpoenaed cablegrams between him and two Pakistanis based in Karachi, S.A. Butt and Colonel Umar Din Dar, which made clear that Vaid was aware he would be exporting a strictly controlled item. Confronted with this evidence, Vaid pleaded guilty. (29)

Although the prosecutor had declared at Vaid's arraignment that he believed Vaid to be a Pakistani agent assisting the Pakistani nuclear-weapons effort, this element of the case was dropped, apparently because the prosecution felt it lacked evidence to back up the claim. When Vaid was sentenced, in October 1984, the judge and the prosecutor adopted the stance that he was an overeager businessman attempting to accommodate a customer. In the words of the prosecutor at the sentencing hearing, "We never claimed Mr. Vaid was an enemy of the country or that Pakistan was an enemy, far from it." On this basis, Vaid was sentenced to two years in prison, but the judge suspended the jail term to the several months Vaid had served while awaiting trial. Vaid, who lives in Pakistan, was deported there on November 10.

As it turned out, the prosecution did possess evidence linking Vaid to the Pakistan atomic-weapons effort, but it was apparently unaware of the fact. The cables between Vaid, S. A. Butt and Colonel Dar did not specify who the individuals were, but Butt at least was well known to the non-proliferation community--including the State Department officials supposedly tracking the prosecution-as the kingpin in a number of the clandestine Pakistani purchasing networks operating in Europe during the 1970s. And, as investigative journalist Seymour Hersh discovered, letters taken from Vaid at the time of his arrest by the Customs Bureau identified Butt as "Director (S & P)" (i.e., Supply and Procurement) of the Pakistan AEC and Dar as a "Senior Executive Officer" there. The letters did not, however, mention any contraband, and when Customs checked Butt and Dar on the Treasury Department computer, no entries were found. Thus the importance of the letters was never recognized and Vaid, like the vast majority of defendants in the various Pakistani prosecutions, largely escaped punishment.

Indeed, the total time likely to be served by the defendants in all of these cases appears to be 15 months, and the total amount of the fines levied, about $16,000. The Migule deal alone, by contrast, involved $6 million.

Still, the prosecutions probably had some overall deterrent value in Western Europe and North America. Few businessmen enjoy being taken to court, even if the odds for acquittal are high. The publicity attending the prosecutions no doubt assisted non-proliferation officials in their informal, and at least partially successful, efforts to persuade sellers of nuclear commodities to refrain from providing assistance to the Pakistani nuclear program. (30)

It must be recognized that these cases dealt with only a fraction of Pakistan's overall effort to obtain key nuclear commodities from the West. Numerous other companies and individuals who are believed to have provided critically important assistance or served as intermediaries escaped prosecution altogether. These include Belgonucleaire in Belgium; Emerson Electric, Weargate, W. Canning Engineering, and Source Reliance International, Ltd. in Britain; St. Gobain Techniques Nouvelles in France; Alcom in Italy; several unnamed Dutch corporations noted in the "Khan Affair" report; CORA Engineering and Vakuum Apparat Teknik in Switzerland; various unnamed Turkish intermediaries and American electronics companies mentioned in several press accounts; and Leybold-Heraeus, Letfeld, Aluminiumwerken, Team Industries, and Migule's associates in West Germany. (31)

Pakistan itself, which has been the hidden hand behind all of these clandestine affairs and continues to enjoy their fruits, has similarly escaped punishment. Today, Migule's plant is generally believed to be producing uranium hexafluoride near Dera Ghazi Khan, and Van Doome Transmissie's hardened steel tubes are thought to be spinning at the Kahuta enrichment plant, regulated by British and American inverters and FDO measuring devices. Yet no aggrieved nuclear supplier country is known to have demanded that Pakistan return or close down any of its illegally obtained equipment. 32

As West German officials pointed out in the Migule trial, Pakistan's circumvention of supplier-country export controls amounts, in effect, to an external assault on the Non-Proliferation Treaty, which mandates these controls. (Pakistan is not a party to the pact.) Yet, as much as the United States and its European trading partners value the treaty, not one is known to have condemned Pakistan's exploits in any international forum. Nor was Pakistan's ascendency to the IAEA Board of Governors for a two-year term in 1982 protested in any way, although it must be recognized that board membership is traditionally a matter for the states of a particular region to decide.

Aggrieved European and North American governments have also shied away from imposing any sanctions against Pakistan for its illicit activities. Indeed, the day after the Dutch lower court convicted A. Q. Khan of espionage, the chief of the Dutch foreign aid program happened to be in Islamabad -with an offer of increased economic assistance previously approved by the Dutch government. Similarly, the Goldberg-Mir zirconium seizure in 1981 and the Vaid arrest in 1984 had little or no impact on the U.S. congressional votes approving increased aid to Pakistan that followed on the heels of the two events. Increasing Western assistance to Pakistan has been the rule for over a decade (even before the Soviet invasion of Afghanistan), except for two brief U.S. cutoffs in 1978 and 1979 . (33)

Many factors have contributed to the supplier states' policy of making criminal prosecution the sole punitive response to Pakistani smuggling. The Reagan administration and many observers believe that strengthening ties to Islamabad and providing increased military support are the only way to obtain leverage for influencing Pakistan's nuclear program and providing Pakistan with an alternative to nuclear weapons for protecting its security. Threats of sanctions and public condemnation undermine this approach, which may be keeping Pakistan from building complete nuclear weapons, even if it has not stopped Pakistani nuclear smuggling. Undoubtedly the most important reason for rejecting sanctions, however, has been the desire to avoid seriously injuring relations with a state opposed to the Soviet Union in a region of increasing strategic importance to the West. Although the matter is admittedly complex, during a recent interview a senior Western European non-proliferation official may have summarized the outlook of the Western nuclear-supplier states in a single phrase. Asked whether more aggressive measures should be taken against Pakistan, he replied, "Why should we do that? They're 1 our friends." (34)

Israeli Indictment

On May 16, 1985, federal prosecutors in Los Angeles indicted Richard K. Sinyth, owner of a small aviation consulting and export brokerage firm, for illegally exporting 810 high-speed electronic switches to Israel between 1980 and 1983. The switches, known as krytrons, are used in the triggering mechanisms. of U.S. atomic weapons. (35)

Israel is widely believed to possess, a small nuclear arsenal, thought to consist of perhaps 20 to 25 atomic weapons .36 Although the krytrons have non-nuclear uses in advanced electronics applications, such as triggering lasers, revelation of the exports led to immediate speculation that the devices were intended for use in the Israeli nuclear-weapons program. The U.S. State Department normally scrutinizes export requests for krytrons on a case-by- case basis, and export approvals are rare because of concern that the devices may be diverted to use in nuclear arms. Indeed, State Department officials have indicated informally that in view of Israel's nuclear activities, a request for a license to export the items to Israel would not have been granted.

According to the indictment, Sinyth purchased the krytrons through his U.S.-based concern, Milco International, Inc., and then made fifteen shipments of ten to fifty krytrons each to an Israeli firm, Heli Trading Company, without obtaining the required special export license. Heli is a purchasing agent for several Israeli defense contractors. (37) The maximum penalties for the charges against Sinyth are two years imprisonment, $100,000 fine for each of the 15 export violations, and a fine of $10,000 for each of 15 false statements made in mislabeling the exports as innocuous commodities.

As word of the pending indictment leaked, the Israeli Defense Ministry acknowledged that it was in possession of the krytrons but declared that they had been used only for research and development of conventional weapons, such as laser range-finders and testing equipment, not for nuclear purposes. (38) Israeli officials also claimed to be unaware that the devices had been obtained in violation of U.S. export laws. One source familiar with the Los Angeles investigation that led to the indictment was quoted as saying, however, that it was "clear on the face of things that the Israeli Defense Ministry knew it was receiving restricted goods without having followed the prescribed rules. (39)

According to the U.S. State Department, by May 17, at the request of the United States, Israel had agreed to return those krytrons that were not in use, said to be a large proportion of the total, and to provide the United States "formal written assurances that the devices [Israel kept] will not be used for nuclear purposes." (40) State Department spokesmen also revealed that the United States was seeking physical proof of the whereabouts of the krytrons. "Washington," said the spokesman, "wants Israel either to put the devices on a table for U.S. officials to count or else take the Americans to the specific locations, whether nuclear or non-nuclear, where the devices are stored or in use. (41)

The affair bears many similarities to the Vaid case.(see Pakistani Prosecutions) In the Israeli case, however, the goods got through fifteen times. Israeli claims that the devices were being used for non-nuclear purposes match assertions made by Vaid that the krytrons going to Pakistan were intended for innocent research at the University of Islaniabad. However, Israel's high-tech military research-and-development program, unmatched in Pakistan, gives its claims a greater element of credibility. Moreover, the number of krytrons at issue is far more than needed for Israel's presumed nuclear arsenal, even if that arsenal is considerably larger than previously thought. (42)

But it is highly unlikely that Israel would admit the devices were being used in nuclear weapons in any case, given the country's longstanding policy of not commenting on its reported possession of nuclear arms. Past Israeli statements to the United States on nuclear matters have sometimes been inaccurate. When Israel used U.S. jets to destroy the Osiraq reactor outside Baghdad in 1981, for example, it met the accusation that it had violated agreements with Washington not to use the aircraft offensively by arguing that Baghdad was on the brink of manufacturing nuclear arms in a secret chamber beneath the reactor. After the first political shock over the attack had worn off, Prime Minister Menachem Begin acknowledged that no such chamber existed. (43)

On balance, given the particular suitability of the krytrons to nuclear-weapons use, the fact that U.S. export regulations are predicated on the assumption that foreign recipients intend to use the devices for this purpose, and the presumed existence of Israeli nuclear weapons, there is reason for concern that Israel may have used a number of the krytrons for nuclear ends. (44)

The U.S. executive branch and congressional officials apparently learned of the krytron-smuggling operation in January 1985. Since that time, ironically, Washington has agreed to establish a free-trade zone with Israel, eliminating tariffs on Israeli goods, and to allow it to participate in the research-and-development phase of the Strategic Defense Initiative. U.S. aid to Israel was also substantially increased, and previously agreed-to transfers of advanced U.S. electronics technology for the Lavi fighter were apparently unaffected. (45)

Recent U.S. legislation would prohibit U.S. economic and military aid for Israel or any other nation that violated U.S. export laws in order to obtain equipment that the President determined would materially assist it to make nuclear weapons. (46) Thus, while Israel has not been penalized for its apparent past circumvention of U.S. controls, any future repetition may seriously injure its relations with the United States

On August 20, 1985, Richard K. Sinyth failed to appear for his trial on the krytron-smuggling charges. Law enforcement officials reported he had vanished more than a week earlier and feared he had fled the United States. (47) It thus appears likely that many questions about the affair will remain unanswered.

Syrian Scam; Iraqi Interest

If Pakistan, Israel, and others have been able to circumvent nuclear-supplier-country controls to obtain equipment, a criminal proceeding now being conducted in Italy shows that, at the more basic level of safeguarding nuclear weapons and weapons-usable plutonium, worldwide non-proliferation controls have been more successful.

On June 5, 1984, Italian prosecutors filed a fifty-two page indictment in Trent, charging thirty-eight individuals from three countries with participating in an extensive drug-and-arms smuggling ring. The defendents, who included eminent members of the Italian Socialist Party, a ranking Italian counter-intelligence official, and movie star Rossano Brazzi, were said to have imported morphine base from Turkey through Bulgaria for processing in Europe and to have sold armored cars, tanks, helicopters, and small arms to a number of third world nations, including Somalia, Thailand, and Iran.

The most startling charge in the indictment was that a group of five lesser-known figures had conspired to sell three complete nuclear weapons to Syria and nuclear-weapons-usable plutonium to Iraq. Neither of these deals was alleged to have been consummated, however.

Deposited with the indictment was a 6,000-page dossier compiled by. the investigating magistrate, Judge Carlo Palermo. Although the judge did not release the dossier to the press, over the past year various news organizations have obtained parts or all of it, which has permitted a more detailed look at the basis for these extraordinary allegations. 48

A key intermediary in the offers of both the three atomic weapons and the 33.9 kilograms of plutonium was Glauco Partel, a 55-year-old Italian engineer specializing in rocket technology, based in Rome. Under questioning, Partel told Judge Palermo he had been an operative for Western intelligence services, including the U.S. National Security Agency, for fifteen years and had been rewarded by them in 1980 with an authorization to sell conventional arms on the unofficial, back-channel market. Palermo found that his story partly checked out but his activities went beyond those (49) with any official sanction .

Partel appears to have served as broker for representatives of various arms suppliers with whom he maintained on- going relations, assisting them to find buyers for their wares. His u ' usual practice seems to have been to identify and contact a set of middlemen. trusted by the potential buyers and then serve as go between. The offers to sell the atomic weapons and the plutonium were separate, although both reached a peak of activity in the first half of 1982. According to' statements by Partel and documents in his possession, seized by the Italian authorities, the atomic bomb offer was initiated in early 1982 by Eugene Bartholomeus of Sydney, Australia, who, Partel said, had ties to Ameri can intelligence services. Bartholomeus also appears to have been the source of conventional arms in a number of deals that Partel brokered. Over the following months, at the behest of Bartholomeus, Partel contacted several middlemen with close ties to various Middle Eastern states.

Partel claims that the three atomic weapons never existed. From the outset, he told Judge Palermo and subsequent interviewers, the deal was a sham, a ploy orchestrated by Bartholomeus on behalf of U.S. intelligence interests, partly. in order to learn which states in the Middle East might be interested in acquiring blackmarket nuclear arms. A more important U.S. goal, he claims, was to elicit expressions of Syrian interest as an indirect means of cautioning Israel against invading Lebanon: presumably, a leak to Israeli intelligence services that Syria was seeking atomic arms would put Israel on notice that there was a limit to what it would tolerate. Only Bartholomeus and Partel were in on the scam, however; other participants believed in the deal -whose credibility was supported by the pair's extensive trading in conventional arms.

While there is every reason to believe Partel's assertion that the weapons never existed, his claim that the offer was an intelligence ploy to forestall the Israeli invasion of Lebanon appears farfetched, although the timing would be correct; it was at about this juncture that the United States became aware that Israel might be contemplating such an invasion. (50) A more likely motive for the subterfuge was that Partel and Bartholomeus intended to use the A-bombs as bait to attract buyers, who would then be told that bombs were unavailable but that Partel and company could provide a variety of conventional arms instead. Such an explanation would hardly have placed Partel in a favorable light before Judge Palermo, however, so it would not be surprising if he embellished the facts to help legitimize his activities.

Whatever the underlying motive, on January 7, 1982, Partel sent Antonio Giovanelli, an Italian businessman with close ties to the Syrian government, the following telex offering the three bombs:

Terms of offer: Power of 3 units same as Japanese original ones (20kt). Delivery will be effected in buyers country. The 3 units are being assembled now and will be ready for delivery in the first week of March. Testing [without a detonation] will be effected in a neutral country of Europe at [sic] the presence of the seller and buyer reps/experts. Selling people will arrange for transportation. This unit have [sic] not been offered elsewhere. Net price for the 3-units delivered is USD 924 millions. Deposit of 462. millions. On delivery the balance will be paid. (51)

Other details of the weapons offer, according to Judge Palermo, were contained in an undated telex in Partel's files, which specified that the bombs weighed 90 kilograms (198 pounds) each, contained 40 kilograms (88 pounds) of uranium-235 (sufficient for a Hiroshima-size weapon), and measured 41" X 11". (52) The telex continues, "The whole unit is covered by lead easing for safety. Can be used only from aircraft. (53)

The cable specified payment terms more or less consistent with those in the telex to Giovanelli. The buyer was to deposit the full purchase price into an escrow account at the Trade Development Bank, Luxembourg, under the code "GEM,,, Attention Glatic Partel, Passport H063381. After a meeting between buyer and seller at the bank, the buyer would either withdraw his funds or use them to open an "irrevocable, transferrable, and divisible several times" letter credit for the full purchase price. The buyer would then send three experts to the seller's production plant, "where they will follow all steps of production. which takes six weeks for completion." At this point Partel would draw down 60 percent of the total, "After completion 'and testing," the document concludes, "the 3 units will be handed over under the most strict safety measures and upon delivery the balance of 40% of the value of the letter of credit will be paid .... Countries to be absolutely excluded from this deal: Israel-South Africa. (54)

After the initial offer to Giovanelli, the progress of the A-bomb sale becomes murky. When interrogated in 1983, Partel stated that he made contact with Syrian representatives (whether through Giovanelli or others is not clear) who expressed interest in acquiring the weapons, a point he would not confirm explicitly in interviews conducted during 1984 and 1985. In any event, the operation was compromised almost immediately. According to statements given to Palermo by both Partel and Giovanelli, soon after the latter received Partel's offer, he informed Israeli intelligence officials of it, either for reasons of conscience, or, as Partel suggests, in the hope of selling the weapons to Israel instead.

Subsequent Israeli behind-the-scenes intervention may have contributed to Partel's inability to find buyers prepared to pursue the deal beyond the preliminary stages. Nevertheless, telexes from- Partel's files, grouped by Judge Palermo under the heading, "A-bombs, Arabs," appear to indicate that Partel did, in fact, bring one, and possibly more, groups into serious negotiations for the weapons.

Most chilling are descriptions of Partel's visit to Iraq in March 1982 with an associate and co-defendant in the case, Egyptian-born Ivan Galileos. In a series of cables dated March 30 to Bartholomeus in Sydney, Partel linked the Palestine Liberation Organization (PLO) to the "three toys," a term considered by Palermo and others who have examined his dossier to be a code for the three atom. bombs. Partel also stated that earnest money for their purchase had already been deposited in Europe:

TLX No. 1

This is the first of a series of TLXS reporting on our trip to Baghdad presenting the general picture of the whole situation.

It was apparent that the first operation-the supply of some specific armaments to Iraq-is the security key for the men involved in the operation of the three toys. By the successful completion of the first operation the men in question are to receave [sic] the official protection of the Iraqi services throughout the Middle East and Europe (not only with regard to Israel but also with nonfriendly Arab countries). In other words, the two deals were interconnected. We were introduced at the top Iraqi military and political levels by the PLO men responsible for the three toys and by the Iraqi services ....

TLX No. 5 Re 3 Toys

Funds have been transferred to Europe to let know [sic] that they are ready and willing to carry out the operation.

After we left Iraq, one of the men involved in this phase and who was present in Baghdad, has gone to Beirut to get latest instructions, while another one is still in Iraq. It is forseen [sic] that a meeting will take place in Rome ... and soon after we will be able to give you all their proposals and comments in order for seller to be prepared and discuss same in Lux .... (55)

While not precisely confirming these various details, Partel did state in a recent interview that he had contact in Lebanon with Arabs representing interests that he would not disclose, who were seeking Saudi backing to purchase the three bombs. In addition to these dealings, he appears to have offered the weapons to others; in a telex to Bartholomeus, he claimed to have arranged a preliminary "summit" meeting with unnamed buyers' representatives in Rome in February 1982. Offers were sent to L. B. Albert the same month and to a Mr. Ghada in Lebanon in April 1982. There is no indication whether the summit took place, nor is there evidence of interest in the weapons by the other parties.

Partel's offer of 33.9 kilograms of plutonium to Iraq presents a somewhat less convoluted but no less troubling picture. In recent interviews, Partel declared that, unlike the atomic weapons, this material was available for sale through Ricardo Aeschbach, who was living in Switzerland in early 1982 when the deal was launched. Aeschbach, according to Partel, was to obtain the 33.9 kilograms of the material-enough for several nuclear weapons-from France.

In the wake of Israel's destruction of the French supplied Osiraq reactor in June 1981, he said, Iraq had continued its program of purchasing uranium on world markets (56). In the course of looking for sellers, Iraq learned that France might be willing to provide the plutonium if it could do so without publicity. Partel believed that after secret direct negotiations between the two states, the deal was agreed to in principle. The French side then enlisted Aeschbach, with whom it already had ties, to set up a clandestine channel for the transfer.

Through Antonio Giovanelli (the man who had compromised the A-bomb sale), Aeschbach was put in contact with Partel and with Carlo Bertoncini, an individual with close ties to the Iraqi embassy in Italy, and co-defendant in the pending case.

Here again, Partel's explanation is not credible. By 1980, France had become quite cautious in its nuclear dealings with Iraq, insisting, for example, that Iraq allow French technicians to observe activities at Osiraq for ten years. Moreover, while France had agreed, despite U.S. and Israeli protests, to supply weapons-usable highly-enriched- uranium fuel for the plant, it had declared that it would limit the total quantity of the material provided to Iraq at any one time in order to reduce the possibility of the material's being diverted for military ends. For France to provide Iraq with plutonium through back channels, where, unlike the Osiraq plant, it would not have been subject to IAEA inspection, would have run directly counter to these decisions and to France's policy of making nuclear transfers only under IAEA supervision.

Whether or not French or other sources were, in fact, prepared to provide the plutonium, Partel offered the material to Iraq during a trip to Baghdad with his associate, Ivan Galileos, in March 1982, a visit Bertoncini had helped set up. In his cable to Bartholemeus on the visit, Partel stated explicitly that Iraq was interested in obtaining plutonium:

As for the offers submitted to Iraq, the Chief of Armaments told us they are interested in SAM-7 [surface-to-air missiles], mortars, AH-1s [Cobra helicopters] armed with TOWS [wire-guided anti- tank missiles], plutonium, Glauco's secret rocket system. (57)

During a 1985 interview, Partel emphatically confirmed that in March 1982, senior Iraqi military figures expressed interest in obtaining plutonium. He mentioned specifically the "Chief of Staff' and "Chief of the Services," though he did not give their names.

Partel claimed that Iraq pursued the chances of acquiring plutonium at a second meeting on the subject in Rome in May 1982, where he discussed price, "so much per kilo," with a "general of the Iraqi army in charge of missiles." This statement is supported by a May 7, 1982 cable from Aeschbach (via Giovanelli) to Partel, which stipulates a price for the plutonium of $60,572,000, including transport and crew, with a $3.5 million returnable deposit for the aircraft, and declares, "from our side we are ready to carry out the transport as soon as the deposit is made in the bank." (Transportation arrangements included flying a Lear jet to Baghdad under a commercial jet airliner to avoid detection.) In a second telex on the same date, Partel replied that a meeting was set in a Zurich bank on Monday, May 10.

At this point, the deal seems to have foundered, Aeschbach complaining in a May 10 telex that Partel had missed the agreed-to deadlines. In August, according to Partel, Aeschbach and Bertoncini went to Baghdad to try to get the operation back on track. There, however, Iraqi experts asked for technical data, which the two were unable to provide. In a telephone interview, Bertoncini confirmed this account, stating that after Aeschbach had met with Iraqi officials in August 1982, Aeschbach told him they wanted to see samples of the material before proceeding further.

The deal then collapsed for good. Partel says that France had decided not to go through with it and informed Iraq, which then terminated relations with Partel, Bertoncini, and Aeschbach. At least as likely is that Iraq lost interest when it became clear that the group could not deliver samples, much less the entire 33.9 kilograms of material. It is also possible that the plutonium sale was compromised from the start by Giovanelli, and Iraq realized it could not conclude the deal without a high risk of exposure.

Drawing conclusions is not easy when the only firsthand testimony is from men whose stock in trade is deception. Nevertheless, the following points appear to be supported by the evidence: In 1982, atomic' weapons and plutonium-almost certainly nonexistent - were offered for sale on the black market by individuals actively involved in illicit sales of conventional arms. Prices were set - $308 million for an aircraft-deliverable, 20-kiloton atom bomb and $60.5 million for 33.9 kilograms of plutonium ($1.78 million per kilogram) -and business arrangements established for step-by-step negotiations on the transactions. A number of buyers connected with Arab states or organizations, including the PLO, appear to have expressed at least superficial interest in purchasing the three "toys," and it is possible that as much as $800 million was moved to a bank in Europe to further negotiations on the sale.

There is also evidence (in the form of recent statements by Partel and Bertoncini and earlier telexes) that during part of 1982, Iraqi officials actively pursued negotiations to purchase plutonium from Partel and his cohorts. But this evidence is by no means conclusive; nor is it clear what level of the Iraqi government sanctioned these talks. It is worth noting that purchase of the material, which plainly would not have been subject to IAEA safeguards, would have violated Iraq's obligation under the Non-Proliferation Treaty to maintain its entire nuclear program under IAEA inspection.

There is a positive lesson to be learned from these events. First, Partel's network was ultimately exposed -by Giovanelli's disclosures to Israeli intelligence and by Judge Palerino's investigation- showing that the international community is by no means defenseless against would-be nuclear blackmarketeers. (58) Second, the conspirators-and their clients-were unable to obtain nuclear weapons or their essential components, proving that the national and international security measures covering these commodities are substantial barriers against their diversion. As Partel cabled a buyer when one of his potential nuclear deals fell through, "We have lost time/energy/money for nothing." (59) It is to be hoped that no clandestine purveyor of nuclear destruction will ever be able to say more.

South African Sojourners

In late January 1985, South Africa's Electricity Supply Commission (ESCOM) admitted that it had hired U.S. nuclear reactor operators and other skilled personnel to help run its only operating nuclear power plant, Koeberg 1. (60 ) Press reports earlier that month alleged that many of the Americans were working in South Africa in violation of U.S. law. (61)

Koeberg I (and its sister unit, Koeberg II, which started up in mid-1985) are subject to IAEA safeguards and are not considered part of South Africa's presumed nuclear-weapons capability, which dates from the late-1970s. (62) Nevertheless, since February 1983, the U.S. Department of Energy has required individuals and corporations to obtain specific authorization before assisting in the start-up, operation, maintenance, or shut-down of a nuclear power reactor in countries like South Africa, which are considered high proliferation risks. (63) Prior to that date, such activities, but not work at more sensitive enrichment or reprocessing plants, was generally authorized, and no special case-by-case review was required. The new regulations, which permit closer scrutiny of all U.S. nuclear assistance, also provide a legal basis for the curtailment of such assistance, augmenting the embargo of U.S. fuel and reactor sales under the 1978 Nuclear Non-Proliferation Act to many of the nations on the U.S. Department of Energy list.

None of the forty-three Americans employed at Koeberg had obtained a special Energy Department authorization. A subsequent investigation by Department of Energy officials in early 1985 uncovered that many of them were working under contracts signed prior to the 1983 tightening of regulations or were performing unregulated tasks and therefore not acting illegally. At least twenty-five others, however, were found to be assisting South Africa without the requisite authorization. (64)

Department of Energy officials claimed to have become aware of the potential violations of U.S. law, which can carry a maximum penalty of ten years in prison and a $10 thousand fine, in November 1984, when the department was informed that five more Americans might be on the verge of leaving the United States to work at Koeberg. (65) In December, in response to a U.S. request, ESCOM notified its American employees of the Department of Energy's regulations, and a number of them contacted the department to get the necessary authorization to continue their work. (66)

Responding to press accounts in January 1985 that it might be employing U.S. nationals illegally, an ESCOM spokesman said, "The relationship of the American operators and/or other staff that we have and the State Department obviously is one between them privately. It does not actually affect ESCOM directly. (67) Although this may be technically true, ESCOM was apparently aware of the regulations when it recruited the Americans and acknowledged as much during the Department of Energy investigation. (68) Indeed, some thirteen special authorizations had been granted to U.S. companies seeking ESCOM contracts for maintenance and safety work at Koeberg in September 1983, and ten more requests were said to be pending. (69) Moreover, the February 1983 DOE restrictions were widely publicized in the general and trade press. (70)

In addition, ESCOM apparently took extraordinary steps to recruit U.S. personnel, who were considered to be particularly well trained . (71) According to C. H. Dean, Jr., Chairman of the Tennessee Valley Authority, between 1980 and late 1984 some twenty-one former TVA employees left to work for ESCOM. In correspondence with a congressional committee investigating the matter, he stated,

While it has not yet been verified, there is some indication that several [out of eleven] of the employees who left in 1984 were recruited by a former TVA employee who went to South Africa in 1980. The recruiter is reputed to have received a $5,000 bonus for each licensed individual hired. In addition, there is some indication that the compensation package offered by ESCOM included a salary range of $90,000 to $200,000 and fringe benefits such as profit sharing, quarters' allowances, relocation bonuses, free roundtrip fares to the United States for. leave, and the payment of South African taxes. (72)

In fact, ESCOM actually intensified its U.S. recruiting efforts after the February 1983 tightening of U.S. controls. In late 1983, South African nuclear regulators disqualified numerous South African reactor personnel from certain jobs at the Koeberg site, apparently because their accumulated radiation doses exceeded permissible limits. With the second Koeberg reactor expected to begin operating in the near future, ESCOM's manpower shortage became acute, requiring it to look overseas for qualified personnel. In 1984, ESCOM recruiters made two trips to the United States, double the previous year's effort, apparently with considerable success: of the thirty-three Americans at Koeberg interviewed by the U.S. Department of Energy, nearly half were recruited during this period. (73)

Although there is insufficient evidence for concluding that ESCOM encouraged the twenty-five reactor operators and other technicians to circumvent U.S. laws, it at least turned a blind eye to the issue and, judging by its statements to date, had no qualms about employing the individuals even though a violation of U.S. law might be involved. (Once the issue threatened to become public, however, ESCOM assigned the Americans to non-nuclear tasks.)

It is also possible that a secret ESCOM "strategic fund" may have been used to pay the Americans it hired. This fund, amounting to more than $3.65 million and apparently intended, in part, to assist ESCOM in purchasing nuclear fuel in Europe after U. S. exports had been embargoed, has been the subject of a number of press stories in South Africa. ESCOM ultimately denied that any of the fund had been used for recruitment purposes in the United States, but its very existence raises the possibility that additional secret activities have been undertaken.(74)

Despite these developments, including the insouciant stand of ESCOM toward its employees' compliance with U. S. law, Department of Energy officials hinted that they were favorably disposed toward granting the special authorizations needed to permit U.S. personnel to continue working at the Koeberg facility. This stance would be consistent with the Reagan administration's earlier readiness to aid the project. (75) In July 1985, however, all requests were rejected. Nonnuclear concerns over South Africa's incursion into Botswana (which led Washington to recall its ambassador to Pretoria) and the declaration of a state of emergency soon after because of racial unrest undoubtedly contributed to the U.S. decision not to approve did for the South African nuclear program.

The entire episode points once again to the existence of a nuclear underground, this time involving services and skills rather than hardware. While it is worth emphasizing that the particular assistance being given to South Africa was not directly contributing to its nuclear-weapons capability, it must be said that the -aid allowed South Africa to evade economically painful nuclear controls imposed by a foreign government in response to its maintenance of that capability.

It remains to be seen whether the individuals involved knowingly violated U.S. nuclear-export regulations. All of them have told the U.S. Department of Energy investigators that they were not aware of the need to obtain special authorization. But it is hard to imagine any individual involved in the nuclear industry being unaware of the long-standing belief that South Africa is secretly developing nuclear weapons or of the fact that South Africa has been the subject of U.S. nuclear-trade sanctions. Moreover, if TVA's information is correct, the Americans accepted very attractive inducements to work in South Africa.

All in all, the episode, like the case of the engineers who helped Albrecht Migule build his uranium hexafluoride plant in Pakistan, reveals the existence of a pool of trained nuclear experts, whose activities, whether innocent or conscious, can significantly undermine the global ground rules that have been established to curb the spread of nuclear arms.

Argentine Assistance

During 1985 new evidence came to light indicating that Argentina was relying on assistance from a number of Western European and American companies in the construction of its two most sensitive nuclear installations, the Pilcaniyeu (pronounced PIL-CAN-EEjAY-00) enrichment plant and the Ezeiza reprocessing installation. Although, in most cases, the assistance does not appear to be a violation of international nonproliferation controls, it has undoubtedly been of considerable value to Argentina in advancing the completion of these two facilities.

Both nuclear projects were initiated by the military regime between 1977 and 1978. Construction of the Pilcaniyeu installation continued in secret for five years. Only in November 1983, on the eve of-the inauguration of Raul Alfonsin, Argentina's first democratically elected leader in over a decade, did Argentine nuclear officials reveal its existence. As they described it at the time, it had the potential, when completed, to produce enough highly enriched uranium for several nuclear weapons annually, although they asserted it had been constructed to produce only non-weapons-grade material. Argentina's natural-uranium/heavy-water power reactors do not, however, use enriched uranium fuel, and although Argentine nuclear officials claimed that they planned to improve the performance of the reactors by fueling them with slightly enriched uranium, no other country currently using such reactors does. Argentina's research reactors do require enriched uranium, but a plant of the size announced could have provided more than sixteen times the amount necessary to serve these small research facilities. Moreover, Argentina has refused to place the plant under IAEA safeguards, which means that if it is configured to produce weapons-grade uranium, Argentina will be free to use it for nuclear arms. Finally, the installation was begun at a time of increasing Argentine militarism. (76) All these factors strongly suggest that the plant was intended to give Argentina a nuclear-weapons option.

The disclosure of the installation in 1983 caught the international community, and U.S. officials in particular, by surprise . (77) Although the then head of the Argentine nuclear program, Admiral Carlos Castro Madero, stated that some 15 percent of the plant, which uses the gaseous diffusion method of enrichment, had been purchased abroad, little information emerged as to who provided the equipment or what it consisted of.

In late 1984, a photograph of the installation taken for an Argentine magazine answered the question in part. The name Sulzer showed plainly on the exterior of a cooling unit. (78) Subsequent inquiries by an Amen can Broadcasting Company (ABC) television documentary team revealed that the sign referred to the Swiss firm of SuIzer Brothers. Senior Sulzer Brothers officials in Switzerland said the company had supplied the plant's electric generators in 1981 and, through an Argentine subsidiary, its entire cooling system sometime afterward. It is not clear when the Swiss parent learned of the latter sale.

Because equipment of this kind is used in a host of industrial settings, its export from the nuclear-supplier states is not specifically controlled, even where the equipment is custom-tailored for particular facilities. Cooling equipment is, however, a necessary adjunct to an enrichment plant to remove process heat and the on-site generators were needed during the construction of the plant; thus the sale of the equipment undoubtedly contributed to the project's realization. Personnel from the Argentine subsidiary installed all of the equipment at the site. And while equipment of this type is sometimes bought "off the shelf," it is also possible that the installation of the cooling equipment required SuIzer to perform significant engineering work for the facility-for example to establish the heat balance across the entire installation. The scale of such activities may possibly have meant that SuIzer Brothers - either the Swiss parent or the Argentine subsidiary, or both-were aware that Argentina had embarked on the construction of a secret uranium enrichment installation. In this case, it would be not only SuIzer Brothers' equipment but also its silence that would have abetted Argentine efforts to advance the project. In an interview with Swiss television, however, a spokesman for the Swiss parent stated that Argentine nuclear aides had deceived it at least when they ordered the generators, telling SuIzer that the facility was a relatively innocent plant to produce zirconium, the metal used to sheathe reactor fuel. (79)

Unlike the Pilcaniyeu facility, the Ezeiza reprocessing plant has been an openly announced project since its inception. Although the facility will be under IAEA inspection for the foreseeable future, it will provide Argentina with direct access to weapons-usable plutonium and has been a long-standing proliferation concern of the United States. In the late 1970s, the Carter administration tried unsuccessfully to persuade Argentina to abandon it. Argentina has reserved the right to operate the plant without IAEA oversight whenever the facility processes fuel from unsafeguarded reactors-which Argentina now lacks.

Argentina has defended its decision to limit the safeguards because the plant is not based on imported reprocessing technology. But in 1985 it became clear that the country was receiving significant foreign assistance in completing the project, although it is not clear whether or not the reprocessing technology itself, which is strictly controlled by the nuclear-supplier countries, has been supplied. Two concerns known to have given assistance are the Italian firm of SNIA Techint and the U.S. company, Honeywell.

SNIA Techint S.p.A. of Rome has designed the high-level waste-management facility to be built adjacent to the plutonium separation portion of the plant. This is a critically important part, which will receive and treat the highly radioactive liquid wastes produced in the plutonium-extraction process. Indeed, it appears that the plutonium separation portion cannot be operated without this waste treatment unit, although some technical details remain murky. Since this assistance does not, however, specifically involve the actual process of separating plutonium from spent fuel, it is not subject to strict regulation, and the Italian government cannot be said to have formally authorized any such activities by the Italian corporation. (80)

Nonetheless, U. S. officials have expressed concern that the Rome firm may, in fact, be providing assistance in the design and construction of the portion of the plant performing the plutonium separation. SNIA Techint S.p.A. is well versed in this technology and provided hot-cell laboratories and a mock-up of a reprocessing facility to Iraq in the late 1970s.

The Argentine firm in charge of constructing the Ezeiza reprocessing plant is Techint S.A. of Buenos Aires, a company, as its name implies, related to SNIA Techint. The stock of Techint S.A. of Buenos Aires is owned by a holding company, Techint International. The same holding company also owns, through a subsidiary. (81) half of the stock of SNIA Techint S.p.A. of Rome. To put it simply, the same group of individuals ultimately controls the activities of both companies. In addition, in the course of completing the waste management project, SNIA Techint of Rome personnel have made visits to Argentina, and Argentine nuclear aides have made visits to SNIA Techint offices in Italy. This means it is possible that at the direction of the parent Techint International, the Rome company has provided technical assistance on plutonium separation to its Argentine relation.

Italian officials state that they have examined this issue in depth and concluded that the Argentine and Italian subsidiaries are operating completely independently of each other and only the "financial and economic position" of SNIA Techint of Rome is monitored by its immediate parent company without the Techint holding company (which also owns the Buenos Aires firm) becoming involved. The Italian government also received written assurances from SNIA Techint S. p. A. of Rome that no reprocessing technology, as such, was being exported, and Argentina has denied that it has received detailed technical assistance on reprocessing. (82)

The U.S.-based Honeywell Corporation is believed to have provided microswitches for the plant's central control room in 1981. (83) The switches, which do not require special export approval unless they are destined for nuclear end uses, had been previously exported to Honeywell's Argentine subsidiary, where the became part of the firm's general inventory, until they were purchased for the Ezeiza plant. It is unlikely that Honeywell officials in the United States were aware of, the sale, although the subsidiary may possibly have violated U.S. regulations by failing to obtain U.S. approval before selling the items for use in a nuclear plant.

At least one West German firm is reportedly engaged in assisting the construction of the plutonium separation portion of the Ezeiza installation. (84) West German officials have denied that such activities, which would require specific export authorization, have taken place but acknowledge that West German firms have provided unregulated dual-use items for the plant (i.e., items with both nuclear and non-nuclear uses), including the remote manipulators essential to permit the handling of radioactive materials behind heavy shielding. (85)

These transfers belong to the dark world of nuclear commerce. At the very least, Argentina has quietly exploited loopholes in the non-proliferation controls governing exports of equipment from nuclear-supplier countries. While these activities would by definition not be illegal, they demonstrate a willingness on the part of major corporations to disregard, for the sake of profit, the proliferation potential of the installations whose completion they advance. And if, as alleged, Argentina has received critical plutonium-separation technology from Germany (or possibly from Italy) without the required license, or if it obtained SuIzer Brothers assistance for Pilcaniyeu under false pretenses, Argentina has joined the ranks of Pakistan, Israel, and others in using subterfuge and deception to bypass the technology-transfer controls of the nuclear supplier states.

Swedish Secrets

According to information revealed for the first time in 1985, Sweden conducted an extensive nuclear weapons development program for nearly three decades beginning in 1945. Between 1958 and 1968, the program was pursued under the leadership of a handful of top civilians in apparent contravention of a series of parliamentary votes that prohibited research on nuclear arms and declared that Sweden would postpone a decision on the development of nuclear weapons. From 1968 to the termination of the program in 1972, the research activities were carried on despite Sweden's explicit renunciation of nuclear arms and the nation's 1970 ratification of the Non-Proliferation Treaty. (86) The fact that no complete nuclear weapons were ever produced or tested and that Sweden rejected nuclear arms after they were within its grasp is testimony to the strength of the nation's non-proliferation commitment.

The details of the Swedish nuclear weapons program were disclosed for the first time in a lengthy investigative report, published in the Swedish technical journal Ny Teknik in April 1985. The article, based on newly declassified documents and interviews with key participants, provides an invaluable case history of a secret nuclear-weapons research effort, revealing step-by-step how such programs are conducted and disguised. Of greatest importance are the article's revelations describing the development of high-confidence nuclear-weapons designs without nuclear testing, (87) the deliberate structuring of a civilian nuclear program to serve military objectives, and the use of misleading declarations of nuclear policy to mask ongoing nuclear-weapons research.

From its early stages, Ny Teknik reports, the Swedish nuclear-weapons effort followed parallel tracks. One was a classified nuclear-weapons research-and-development program, known initially as the "L-program," housed within the Institute for Defense Research (Forsvarets Fokskningsanstalt) under the direct control of Sweden's supreme military commander. The other was the secret partial structuring of Sweden's civilian nuclear program under the Atomic Energy Corporation (Ab Atomenergie) to meet military objectives set by the L-program.

Between 1945 and 1958, the year in which the Swedish parliament prohibited all nuclear-weapons research except "protective" activities, researchers in the L-program at the Institute for Defense Research completed extensive work on nuclear arms. At least two designs for Nagasaki-size (20 kiloton) nuclear weapons were prepared. The weapons were to weigh 1,300 pounds and 650 pounds respectively and were streamlined aerodynamically, the larger measuring 3.5 meters in length (11.6') and 0.6 meters (2') in diameter, and the smaller presumably of proportionately reduced size. Thus, even without nuclear testing, the designs were far advanced in comparison to the original 5 ton (10,000 pound) U.S. plutonium bomb used on Nagasaki, which was a stubby ovoid, 10' long and 5 5' in diameter. According to the report, Defense Research Institute planners also developed three different weapons-production programs-for two, five, and ten Nagasaki-size bombs per year-and had prepare detailed technical, economic, and geographic studies for implementing them.

L-program specialists also selected a nuclear-weapons test site, Ny Teknik found, and had obtained 10 grams of metallic plutonium (the form used in weapons) from Great Britain in 1957. Plans to build a military plutonium laboratory for extracting 30 grams of plutonium per day from spent fuel (enough for about one weapon per year) and for fabricating the bomb's plutonium cores had also been drawn up. In 1958, the report states, construction of the facility was quietly approved.

Paralleling these efforts were the closely coordinated activities and construction program of the civilian Atomic Energy Corporation. Under a secret agreement, unearthed by Ny Teknik, the Corporation and the L-program agreed in 1950 to conduct their activities according to a joint plan and to exchange data, personnel, and, nuclear materials as necessary.

In 1953 the chief of research at the civilian Atomic Energy Corporation prepared a blueprint for nuclear development in Sweden for the Defense Research Institute. Revealingly, the study was entitled, "Concerning the Preconditions for the Production of Atomic Bombs in Sweden." The report set out a detailed nuclear program, based on natural-uranium/heavy-water reactors, selected for their superior plutonium-producing capabilities and because they would allow Sweden to be largely independent of external sources of nuclear materials. The report called for the production of ten Nagasaki-size bombs in ten years. The author was Sigvard Eklund, the man who would later serve for twenty years as the head of the IAEA.

In 1954, Sweden's first experimental nuclear plants, began operating under the aegis of the Atomic Energy Corporation. These included a small research reactor (using French fuel and Norwegian heavy water), a pilot uranium mine, and a reprocessing laboratory. The laboratory, which produced the nation's first small samples of plutonium the same year, was undoubtedly placed on such a fast track because of military priorities.

By this time, a specialized group within the Corporation, funded secretly in the military budget, was doing the research needed to ensure that the reactors and other installations to be built would also meet military specifications should a decision be made later to manufacture nuclear arms. By 1958, it was decided that plutonium for the nation's initial nuclear weapons was to be produced from the spent fuel of Sweden's first sizable civilian reactor, Agesta, to be built by the Atomic Energy Corporation. Agesta was completed in 1963 and through 1974 produced steam for district heating and electricity. The United States supplied enriched uranium for the plant (to improve its performance) in the 1960s under terms ruling out the use for military purposes of any plutonium produced with the material. (88)

In 1958, following several years of public debate over whether Sweden should build nuclear weaponsa debate that split the ruling Social Democratic Party -the Swedish parliament voted to postpone a decision on the matter, ruling that in the meantime, no basic research on nuclear weapons would be permitted, although "protective research" at the Defense Research Institute could be expanded to allow Sweden to protect itself in the event of a nuclear attack and to help maintain a nuclear option until further decisions Were made. (89) Thereafter, in a series of actions between 1959 and 1965, the parliament explicitly rejected the efforts of the Defense Ministry to expand this mandate. (90) In 1961, it voted down a proposal that the ,,protective research directive" be expanded to include basic weapons research, and in 1962, it rebuffed a bid by the Supreme Commander for the formation of a special committee to investigate the technical and economic aspects of a Swedish nuclear-weapons program. (91)

It was after the 1958 parliamentary decision to halt all but "protective" weapons research, however, that the weapons program was most aggressively pursued. The Ny Teknik report shows with what extraordinary latitude the nuclear weapons group at the Defense Research Institute interpreted this ruling and pursued an intensive program to master virtually every element of nuclear weapons manufacture. According to Ny Teknik, the group's actions were approved by a small group of top civilian officials, including Olaf Palme -Sweden's current Prime Minister- and secret appropriations were -provided annually between 1958 and 1965. (Palme, facing a reelection campaign, now denies that he was aware of the details, and some people in Sweden believe the Ny Teknik article was timed to embarrass him.) The degree to which the Swedish parliament may also have been informed remains unclear, but recent editorials in Swedish newspapers show that the public believed nuclear-weapons research had been banned by the 1958 vote. (92)

Through 1965, under the guise of protective research, the weapons-development group designed and fabricated all of the key -components for atom bombs, (93) mastered the techniques for turning and casting plutonium (triggering a U.S. inquiry into the whereabouts of plutonium it had supplied for civilian research), and conducted numerous full-scale tests of model atomic bombs using dummy cores of non-weapons material. By this point, the weapons program, whose staff numbered more than 300, was also beginning to experiment with models of very small nuclear weapons, seeming to indicate that considerable progress toward miniaturization of such arms is possible even without nuclear tests. (These findings in the Ny Teknik report are supported by documentary evidence or by participants' statements and are not known to have been challenged.)

Intimate coordination with Sweden's civilian nuclear program continued. In 1959, to ensure adequate supplies of uranium for weapons unrestricted by external non-proliferation controls, the government, at the urging of the weapons-development group, directed the Atomic Energy Corporation to develop a uranium production center at Ranstad. The corporation had opposed the concept as contrary to its plans for commercializing nuclear power, because uranium from Ranstad would be 70 percent more expensive than that imported from abroad. Publicly, the facility was justified as necessary for emergency preparedness.

During this period, plans were also laid for the construction of a reprocessing plant at Sarmas capable of extracting enough plutonium for twenty nuclear weapons annually, the highest of the nuclear-weapons production options being considered by Defense Research Institute planners. The facility was formally proposed by the Atomic Energy Corporation, but its specifications had been determined by the weapons group. Although land purchases for the installation began in 1963, funding for the plant was never approved, apparently because it could not be justified on civilian nuclear-energy terms and the government was unwilling to go forward with a purely military nuclear project of this size in view of the 1958 parliamentary decision.

During the first half of the 1960s, as the Atomic Energy Corporation designed Sweden's next civilian reactor, Marviken, the secret military research group within the Corporation conducted extensive design analyses to make sure that the facility would meet the military goal of producing plutonium for ten nuclear weapons per year. According to Ny Teknik, at least one feature of the reactor, the capability for refueling without the need to shut down, was a military requisite aimed at increasing the plant's efficiency as a plutonium producer. In the end, delays in completing the plant led the military to abandon it as a prime source of plutonium and to begin' designing a more efficient, purely military production reactor. The Marviken reactor was completed in 1968, but never operated-for safety reasons, among others. The military reactor apparently did not get beyond the design stage.

By 1965, Sweden was well on the way toward possessing a nuclear weapons capability of perhaps three nuclear weapons per year, based on the design work completed, the small-scale reprocessing laboratories, civilian and military, and the Agesta reactor. 94 At this point, support for nuclear arms among Sweden's military began to waver, partly because the growing strategic arsenals of the superpowers diminished the significance of Sweden's modest nuclear potential and because the mounting costs of nuclear arms threatened to undermine Sweden's conventional defense. (95) Although there was no political consensus for renunciation of nuclear arms, scientists were moving in this direction and the momentum of the nuclear-weapons program was lost.

A national consensus to reject nuclear arms. reached in 1968, and in March parliament overwhelmingly adopted a firm non-nuclear policy. (96) Shortly afterward, the weapons-research team was reduced from 300 to 30, uranium production at Ranstad was mothballed, land acquisition at Sannas was halted, the military research activities at the Atomic Energy Corporation were terminated, and weapons-component fabrication laboratories were shut down.

Nevertheless, until 1972, two years after Sweden had ratified the Non-Proliferation Treaty, some weapons work apparently continued. The activities culminated in 1971 and early 1972 in a series of ten tests to examine the impact of explosions on weapons-grade plutonium.

Although some press accounts of the Ny Teknik report initially characterized these experiments as "nuclear tests", it is now accepted that, because each used only 5 to 10 grams of plutonium, they did not produce the self-sustaining nuclear chain reaction seen in nuclear explosions. Rather, the tests used conventional high explosives to smash plutonium pellets against solid targets, probably made of uranium, so that the impact of shock waves on the plutonium could be determined. The plutonium core of a nuclear weapon is subjected to similar shock waves when the weapon is detonated, and there is little doubt that the purpose of the tests was to provide data on the construction of nuclear arms. In the words of one participant quoted by Ny Teknik, "that was as far as we were able to get without exploding a finished core. " (97) After the tests, the remnants of the weapons-development effort were dismantled, and apparently Sweden has conducted no further work in the field.

The disclosures in the Ny Teknik article offer several invaluable insights into the development of undeclared nuclear programs-the motive force at the very heart of the nuclear netherworld.

First, Sweden's accomplishments provide a standard, or benchmark for estimating the accomplishments of roughly similar nuclear weapons efforts. Without a nuclear detonation Sweden was able to design an atom bomb and through extensive testing of its non-nuclear components, gain considerable confidence that the weapon would work. The Swedish weapons designs of the late 1950s would have been easily deliverable by virtually any advanced ground attack aircraft available today. Further miniaturization was being attempted by 1965, again apparently without the need for nuclear testing.

Although precise comparisons may not be valid, it is fair to assume that Israel and South Africa could have accomplished at least this much-certainly by 1985 -if as many believe they have been actively pursuing nuclear arms. India, especially after its 1974 nuclear test, is surely capable of similar achievements. And Pakistan, which has reportedly been working on designing atomic weapons since at least 1981 and which may have received Chinese assistance in this area, is also probably capable of developing deliverable weapons without conducting a test. (98)

The scale of the Swedish program is also instructive. During the 1950s, Swedish defense planners considered the production of as few as two nuclear weapons per year militarily significant, and Sweden's maximum program called for the annual production of only ten to twenty such weapons. Israel's, South Africa's, and India's known nuclear-weapons-material production capabilities fall within this range, as would Argentina's if plants now under construction are completed. Pakistan also seems to fit the mold, although less is known about the size of its production capabilities. The correspondence between what is now known about Swedish weapons planning and the scale of the other covert activity makes clear that programs of this size can have a military purpose.

Similar lessons may be drawn about the scale of individual installations. The key to the early phase of the Swedish nuclear-weapons production plan was a large reprocessing laboratory, where presumably much of the plutonium-separation work would be done by hand, using remote-controlled manipulators to move spent fuel through the plutonium separation process in a series of lead-shielded rooms known as hot cells. A more highly automated reprocessing plant able to handle larger quantities of spent fuel was not deemed necessary until a later stage of the weapons program. This underscores the proliferation potential of reprocessing laboratories, if they are large enough, and indicates that the development or acquisition of ostensibly civilian reprocessing laboratories and hot-cell capabilities over the years in Israel, Argentina, Pakistan, Taiwan, and Iraq must be taken into account in assessing the undeclared nuclear-weapons potential of these countries. (99)

Another major lesson is the way in which civilian nuclear programs can be used to mask military objectives. In the Swedish case, the deception was only partial, because the fact that a nuclear weapons option was embedded in Sweden's civil program appears to have been widely understood, even if many of the details about the degree of integration between the civil and the military were not known publicly. (100)

Nevertheless, the Swedish case reveals the tricks of the trade which, when observed elsewhere, can be read as tip-offs that a civil nuclear program may be disguising the pursuit of a nuclear-weapons option. The most obvious indicators are the selection of heavy-water reactors, early interest in reprocessing (plutonium separation), efforts to achieve independence from outside nuclear suppliers who might attach non-proliferation controls to their nuclear transfers, and the development of demonstrably uneconomic civilian nuclear projects to achieve that independence and/or to provide access to nuclear-weapons material (as seen in the Ranstad uranium center and the Sannas reprocessing plant proposal). There is a similar pattern in India's nuclear development, which led to a single nuclear test in 1974. Argentina, too, seems to fit the mold, particularly in its long-standing efforts to attain nuclear independence and its pursuit of enrichment and reprocessing projects. Though publicly justified in terms of the nation's energy program, these projects have little economic basis. Argentina's nuclear history, so similar to that of two nations whose programs are now known to have been deliberately structured to provide nuclear-explosive capabilities, is added reason for suspecting its motivation, particularly during the tenure of its former military leaders.

The fact that Sweden, a nation with an international reputation for rectitude, pursued an active nuclear-weapons research program at a time when it was widely believed that the nation had formally prohibited such activities underscores the need for healthy skepticism when weighing the peaceful claims of other states whenever there is reason to believe that they might be on the path to nuclear arms.

It is also possible that Sweden may have engaged in deception that violated its non-proliferation commitments to the United States and others, although there is still not enough evidence to make a definitive judgment. During the mid-1960s, the United States supplied Sweden with enriched uranium for the Agesta reactor, heavy water for use in other installations, and quantities of plutonium for civilian, research., The agreement between the two nations for nuclear cooperation, under which the materials were transferred' prohibited the use of this plutonium, or any plutonium produced in any Swedish reactor when U.S. heavy water or enriched uranium was present, for nuclear-weapons research of any kind. According to the Ny Teknik report, some of the plutonium produced in Agesta was extracted and taken to the L-program plutonium laboratory at Ursvik in the late 1960s, eventually triggering a U.S. inquiry on how it was being used. Apparently the United States concern was satisfied, but whether a violation of the agreement took place is not known.

Similarly, if Sweden used U.S.-origin plutonium or that of any other nation imposing comparable nonproliferation controls for its secret 1972 explosives tests, it may have violated these restrictions. British officials state that all the plutonium provided by Great Britain was, in fact, restricted to peaceful uses, but the plutonium Sweden seems to have obtained from France may not have been. In May 1985, the United States delayed nuclear fuel shipments to Sweden pending an investigation into whether U.S.-origin material had been involved in the 1972 tests. Swedish officials denied that it had. (101)

It is also conceivable, although again no specific evidence is available, that Sweden violated the Non-Proliferation Treaty between 1970 and 1972. The 1972 experiments with explosives, if they did not entail nuclear explosions, did not violate the letter of the treaty, which prohibits the "manufacture" of nuclear explosives, but they may have violated the spirit of the pact. And if Sweden continued during this period to manufacture nuclear-weapons components, this could have been a violation under the U.S. interpretation of the term "manufacture." In the course of negotiating the pact, U.S. representatives were asked what would constitute "manufacture" of a nuclear weapon or other nuclear explosive device. In reply, the representatives stated in part:

Some general observations can be made with respect to the question of whether or not a specific activity constitutes prohibited manufacture under the proposed treaty. For example, facts indicating that the purpose of a particular activity was the acquisition of a nuclear explosive device would tend to show non-compliance. (Thus the construction of an experimental or prototype nuclear explosive device would be covered by the term "manufacture" as would be the production of components which could only have relevance to a nuclear explosive device.) (102)

This interpretation was apparently the basis on which the treaty was ratified by the U.S. Senate. (103)

Whether or not Sweden may have violated IAEA safeguards is an additional question. Under the NPT, all Swedish nuclear material is supposed to be declared to the. Agency. Apparently the plutonium used in the 1971-72 tests was not.

But the most important lesson to be learned from Sweden's nuclear past is that the nation ultimately rejected nuclear weapons. Although more than a decade of public and behind-the-scenes debate over the issue was required, national consensus was eventually obtained and a decision to forego nuclear arms implemented. For the most part, such debate has been absent in virtually every country of proliferation concern today.

The Future

The activities of the nuclear netherworld reveal a frightening history of intrigue, duplicity, and unprincipled profiteering. But the picture is not entirely bleak.

Certain smuggling networks have been rolled up; export controls have been tightened; and, to a certain extent at least, Pakistan and possibly other illicit importer states may be feeling a pinch as their access to needed nuclear goods is slowly constricted. Moreover, the number of nations operating to date in the nuclear underground is relatively small, and international controls have clearly prevented the development of a true nuclear blackmarket. Certainly Sweden's renunciation of nuclear arms, after having pursued them for so long, has made an important contribution to non-proliferation by demonstrating that it can be in a nation's selfinterest to forego this option.

In the nuclear supplier states, the legal framework and political consensus for fully controlling dual-use commodities have yet to be developed. The problem will not be easily solved as emerging nuclear states increasingly shift their clandestine purchasing away from large-scale controlled equipment toward the acquisition of widely used, unregulated components of such hardware. Similarly, lack of agreement to apply existing rules aggressively has allowed continued nuclear assistance on installations that may be destined for military use.

The realities of international politics have induced the United States and others to stay their hands in dealing with governments whose minions patronize the nuclear underground. Meanwhile, clandestine activities are continuing. Pakistan, according to U.S. government sources, is still purchasing, through faceless intermediaries, commodities to further its nuclear-weapons quest. Possibly North Korea has done the same, although how it obtained the necessary items for its "research" reactor project remains unknown. (see Chapter III Introduction) And further Israeli smuggling from the United States, involving the production of short-range missiles, has been alleged, (1O4) although there has been no official reaction as yet.

Chapter III: Asia

In principle, we are opposed to the idea of becoming A nuclear power. We could have done so for the past 10 or 11 years, but we have not. If we decided to become a nuclear power, it would only take a few weeks or a few months.
- Indian Prime Minister Rajiv Gandhi, June 5, 1985

The danger of proliferation in Asia increased substantially between mid-1984 and mid-1985. In response to Pakistan's continuing efforts to master nuclear weaponry, India, which conducted a single nuclear test in 1974, announced that it was reconsidering its policy against acquiring nuclear weapons. (1) Hints by Prime Minister Rajiv Gandhi in June 1985 that India may already have manufactured the components needed for nuclear weapons and India's announcement in August that a newly commissioned research reactor would give it a major new source of weapons-usable plutonium not subject to international controls provide further reason for concern.

Today Pakistan stands on the threshold of becoming a nuclear weapon state, and there are some observers, who believe Pakistan has actually manufactured sever- nuclear weapons or their easily assembled components. In late 1984, Pakistani President Zia ul-Haq apparently agreed to level off Pakistan's nuclear program without manufacturing weapons-grade uranium. But the July 1985 report that Pakistan had successfully tested a non-nuclear "triggering package" for an atom-bomb, plus India's concern that Pakistan may have already stockpiled weapons material, make it unlikely that nuclear tensions in the area will ease in the near future.

Another disturbing report, recently confirmed by the U.S. State Department, is that North Korea is building a sizeable, indigenously-designed research reactor at Yong Byon (2), the site of its only other known nuclear installation, a small Soviet-supplied research reactor. Little information has been made public about the indigenous plant, which is likely to cause alarm in South Korea (3). So far, the North, which is not a party to the NPT, has not indicated whether it intends to put the facility under IAEA safeguards, which are applied to the Soviet plant.

The type of reactor being built has not been revealed. However, North Korea has no domestic capability to produce enriched uranium, which would be needed for a light-water/enriched-uranium type research reactor, nor is it capable of producing heavy water, needed for a natural-uranium/heavy-water type research reactor. And there have been no published reports that it is receiving either of these commodities for the reactor from abroad.

This leaves the possibility that it is building a third type of research reactor, using natural uranium and graphite, commodities it might be able to produce more easily than heavy water or enriched uranium. If so,, there would be added cause for concern that the facility was part of a nuclear weapons effort, since the use of a sizeable graphite reactor for ordinary research and training is highly unusual. Conversely, because a graphite reactor can be quite efficient in producing plutonium and is relatively simple to build, it is considered the reactor of choice for clandestine nuclear weapons development programs. (4)

Although South Korea has apparently abandoned the nuclear-weapons program it initiated in the early 1970s, it made unsuccessful overtures to Canada during the past year to institute a joint program involving the separation of weapons- usable plutonium: (5) With the capture of a North Korean commando unit attempting to sabotage South Korea's CANDU 3 nuclear power plant at Wolsting in August 1983, and, two months later, the killing of four members of the South Korean cabinet by a. bomb planted by North Korean agents, tensions between Seoul and Pyongyang have remained high, despite some progress in recent bilateral talks. (6) Largely because of its unwillingness to jeopardize U.S. security guarantees, however, South Korea is not believed to be reconsidering its renunciation of nuclear arms.

In Taiwan, confidence in U.S. security ties may not be as strong as it is in South Korea, but there is no evidence of a reawakening of the Taiwanese nuclear weapons program of the early 1970s. (7)

While the United States has continued to supply Taipei with defensive arms, it has sharply limited the types of weapons provided and their sophistication, and the People's Republic of China continues to apply pressure to restrict such sales. President Reagan's trip to China in April 1984, when he invoked Beijing as a strategic counter to Moscow, demonstrated the growing importance to Washington of Sino-American relations.

Taiwanese leaders realize that this means reduced long-term U.S. military support for Taiwan. (8) To counter this, they have to make the most of U.S. aid, while developing Taiwan's conventional arms industry, with particular emphasis on air-defense capabilities. Taiwanese President Chiang Ching-kuo told the German weekly Der Spiegel in May 1983 that his nation would not build atomic weapons, although it had the scientific and technological capability to do so. (9) Similarly, in November 1983, the Taiwanese press reported that Taiwan's cabinet, in a formal response to a question from a legislator, stated that the "government of the Republic of China under its established policy would never develop nuclear weapons .... There is no reason to develop nuclear weapons to 'kill our own people.' However, . . . the government has spared no effort to develop high technology defense industries in this country." (10) No evidence has emerged since to contradict these declarations.

While Japan unquestionably has the capacity to produce nuclear weapons and is developing sizeable reprocessing and enrichment capabilities, the strong domestic opposition to nuclear weapons resulting from its history as the first victim of nuclear attack excludes it as a proliferation threat at least for the present. It is conceivable, however, that were North or South Korea or Taiwan to acquire such weapons, or were ties with the U.S. to falter, Japanese attitudes might change rapidly.

The activities of the People's Republic of China are also critically important to the risk of further proliferation in the region. China has been a declared nuclear weapon state since 1964. Through the 1970s its policy was not to oppose the spread of such weapons to additional countries, which it saw as a means of diminishing the power of the United States and the Soviet Union.. Indeed, during the early 1980s, there were repeated reports that China had made sales of nuclear materials to such countries as South Africa, India, and Argentina, without requiring them to place the imports under IAEA inspection. (11) From 1982 to 1984, several press accounts, usually quoting unnamed Reagan administration sources, reported that China was aiding Pakistan's efforts to acquire nuclear arms by providing it with nuclear-weapons design information, weapons-grade uranium, and technical assistance in completing an unsafeguarded uranium-enrichment plant, potentially capable of producing weapons-grade uranium. (12) Against this background, the Reagan administration suspended the talks it had initiated with Beijing in mid-1981 on an agreement for cooperation in the field of nuclear energy.

In mid-1983, however, China began a dramatic shift in its posture, and the talks were renewed. In January 1984, China joined the IAEA and subsequently advised the United States that in the future it would require IAEA safeguards on its nuclear exports. In, a series of statements beginning that month, top Chinese leaders declared China's firm commitment to non-proliferation. (13) Based on these developments, a nuclear trade agreement was initialed in Beijing during President Reagan's April 1984 visit.

The administration did not submit the accord for the required review by Congress until July 1985, over a year later, reportedly because of concerns that Chinese nuclear aid to Pakistan was continuing. (14) After further talks, however' China agreed to implement its non-proliferation policy, according to U.S. officials, "in a manner consistent with basic non-proliferation practices common to the United States and other suppliers," an official U.S. circumlocution apparently intended to signify that any Chinese assistance to Pakistan had been halted. (15) China has repeatedly denied that anything beyond normal scientific exchanges took place'. (16)

The U.S.-China agreement, the first U.S. nuclear accord with a nuclear-weapons state that is not an ally, is now undergoing congressional review. (17) Two provisions have been especially controversial. One is an article, required by U.S. law, granting the United States the right to veto the extraction of plutonium from spent fuel produced through the use of a U.S.-supplied reactor or U.S.-supplied fresh fuel. Some believe the wording of the provision is so complex as to leave the U.S. veto right in doubt. (18)

The second and perhaps more fundamental area of controversy is that the agreement gives the United States only limited means to verify that China is abiding by the agreement's restrictions on the use of U.S. nuclear imports- including the restriction that they be used only for non-military purposes and that no plutonium extraction take place without U.S. approval. Unlike U.S. agreements with non-nuclear weapon states, the U.S. -China agreement does not provide for IAEA safeguards to be applied to these imports, nor, it appears, will the U.S. be able to safeguard them itself by using standard techniques for the accounting of nuclear materials. The agreement provides for only informal visits and exchanges of information, not the formal procedures necessary for effective inspections. The limited nature of U.S. verification rights is highlighted by the fact that during 1984 and 1985 China signed nuclear trade pacts with Brazil, Argentina, and Japan under which it agreed to accept IAEA safeguards on nuclear imports from those countries. China apparently refused to grant the United States similar safeguards because both are nuclear weapon states. (19)

India

Between mid-1984 and mid-1985, there was increasing evidence that India is moving toward building an undeclared nuclear arsenal. India demonstrated its ability to produce nuclear arms in its May 18, 1974, nuclear test but has since refrained from conducting further tests and, apparently, from building complete weapons.

Spurred by growing concerns over Pakistan's emerging nuclear weapons capability, India's new prime minister, Rajiv Gandhi, declared in a series of statements during the spring and summer of 1985 that he was reconsidering India's policy against nuclear arming. He also hinted that India might already have prepared components for nuclear weapons, allowing it to deploy nuclear arms rapidly if a decision were made to do so.

In early August, Indian nuclear aides, announcing the inauguration of a large, indigenously-built research reactor, emphasized that the plant would give India the capacity to produce plutonium not subject to external non-proliferation controls. Indian and Pakistani observers saw the announcement as an implicit warning to Pakistan that if India were pressed to build nuclear arms, its nuclear might would quickly overshadow that of its western neighbor. With overall Indo-Pakistani relations remaining tense despite renewed bilateral talks, and with Pakistan, according to mid-1985 reports, still pursuing its efforts to master nuclear weaponry despite President Zia's reported pledge not to manufacture nuclear-weapons material, there is reason to fear that India will continue its apparent move toward acquiring an undeclared nuclear force.

India's 1974 test was authorized by then Prime Minister Indira Gandhi. Gandhi was ousted in India's 1977 general elections but returned to power in January 1980. From that point until her assassination in October 1984, she tended to minimize the Pakistani nuclear threat in her public utterances while quietly enlarging India's actual capability to produce nuclear weapons. In a July 1984 interview, several months after Pakistan announced that it was capable of enriching uranium, a key step toward developing nuclear arms, Mrs. Gandhi said, "I don't think we can do anything about it except to try to talk people out of stockpiling or making nuclear weapons. (1) Similarly, in a statement to the Indian parliament a month later, she declared that India did not intend to alter its commitment to a peaceful nuclear policy despite reports of Pakistan's attempts to acquire nuclear weapons because "Pakistan having a nuclear bomb will not make much difference . . . . " (2)

Notwithstanding these mild pronouncements, in March 1980 Mrs. Gandhi announced that she would "not hesitate from carrying out nuclear explosions ... or whatever is necessary in the national interest," (3) and in 1981 and again in 1983 it was reported that India was maintaining a nuclear test site in a state of readiness. (4) Mrs. Gandhi also promoted Dr. Raja Ramanna and Dr. P.K. Iyengar to key nuclear policy-making positions. The two men had played central roles in developing the nuclear device for the 1974 test and are thought to favor India's development of nuclear arms. (5)

Despite strong pressure from the United States, Mrs. Gandhi refused to place under IAEA safeguards a number of India's nuclear installations, including the plants used to produce the plutonium for the 1974 test and a power reactor (Madras 1), commissioned in 1983. The latter facility, which can produce 50 to 60 kilograms (110- 132 pounds) of weapons-grade plutonium annually, sat idle for two years because India did not have a sufficient domestic supply of heavy water to run the plant and was unwilling to buy the material abroad because this would have required placing the plant under IAEA oversight. (6) Mrs. Gandhi also pursued completion of the Dhruva reactor, the Tarapur reprocessing plant, and the Trombay reprocessing plant. These various facilities form the backbone of India's nuclear weapons potential today. She also supported increasing expenditures for India's space program, which may ultimately permit construction of intermediate-range nuclear missiles.

Despite the generally accepted view that since the 1974 test India has not built nuclear weapons, several recent reports suggest that in 1980 Mrs. Gandhi may have secretly resumed nuclear-weapons design activities and the fabrication of components for such weapons. (7) As the history of Sweden's nuclear-weapons development efforts described in Chapter II shows, such efforts could easily be carried out in secret; they would also have been consistent with the reports that India was readying a test site.

In the summer of 1984, Indo-Pakistani nuclear tensions worsened after the arrest of three Pakistani nationals in the United States who were attempting to smuggle parts for atomic weapons to Pakistan and reports that China had given Pakistan nuclearweapons-design information . (8) Overall relations between the two also deteriorated, as India postponed the bilateral talks scheduled for July after accusing Pakistan of 'fomenting unrest by Sikh nationalist groups in the Punjab region and of making incursions into Kashmir which led to a series of border clashes. (9)

The situation worsened in mid-September when press reports, based on a U.S. intelligence briefing to a congressional committee, stated, that Mrs. Gandhi was being urged by military advisers to launch a preemptive attack against Pakistan's key nuclear installation, the Kahuta enrichment plant. (10) The plan was apparently rejected for fear that Indian nuclear installations and off-shore oil rigs would be vulnerable to counterattack. Indeed, in October, Pakistan's foreign minister stated that it would have "no alternative but to retaliate" if India attacked Kahuta. (11)

In mid-October, Mrs. Gandhi departed from her earlier muted statements and declared in an address to a group of army commanders that the Pakistani nuclear program was "a qualitatively new phenomenon in our security environment," which must add a "new dimension" to India's defense planning. (12) Her remarks followed a series of developments that seemed to threaten Indian security interests. These included rumor -quickly denied in Washington-that the United States had offered to extend a nuclear "umbrella" to Pakistan; a statement by the U.S. ambassador to Pakistan that the United States would not be neutral if India attacked the country; and the disclosure that Pakistan was seeking an airborne early-warning radar system from the United States, a system India believed would significantly enhance the power of the advanced U.S. aircraft Pakistan was already receiving. As India protested these developments to the United States, some Indian press accounts used the term "war clouds" in describing the overall state of Indo-Pakistani affairs. (13)

On October 31, Indira Gandhi was assassinated by Sikh members of her bodyguard, and her son, Rajiv, was appointed interim prime minister. He was confirmed in the post after his landslide victory in the December 28, 1984, general election. In the immediate aftermath of the assassination, he said little of Pakistan's nuclear program and seemed to welcome a major peace initiative by Pakistani President Zia ul-Haq. (14) Nevertheless, throughout the election campaign and afterwards, he continued to complain of Pakistani aid to Sikh extremists and of the country's build-up of conventional arms with U.S. aid. Skirmishing between Indian and Pakistani troops in Kashmir continued. (15) And despite the recommencement of formal bilateral talks in April, relations between the two nations have remained troubled.`

Within a month of his election, Gandhi began a reexamination of India's response to Pakistan's nuclear weapons program. In interviews with the foreign press on January 30 and February 21, he stated that India would not build nuclear weapons even if Pakistan took this step. (17) But in a separate January 31 discussion with reporters from an Indian fortnightly, he declared that "it will be a point of no return on the subcontinent if someone has nuclear weapons. We will have to review our policy to see how we are going to counter that imbalance. " (18)

By April, the latter theme was plainly dominant in government pronouncements, and Gandhi, apparently laying the groundwork for his trip to Washington in June, began to criticize the United States for failing to arrest its ally's nuclear ambitions. When asked by the Financial Times on April 4 whether he thought Pakistan had nuclear weapons and whether Washington had done enough to stop it, he replied,

I think [the Americans] must make their policy with regard to Pakistan very clear. Are they going to let Pakistan make a nuclear bomb for example: every indication today is that they are.

[The Pakistanis] are very close to one if they haven't got one and nothing is being done to stop it. And once it's there it'll be a fait accompli. We'll be landed with a bomb in the subcontinent. (19)

In a similar vein, on April 16, the annual report of India's defense ministry declared that "Pakistan's relentless pursuit of [al nuclear weapons capability has added a new dimension to India's security environment" and went on to describe Pakistan as India's principal security concern. (20)

In May, Gandhi gave further hints that India was reconsidering its rejection of nuclear arms. Addressing a convention of his political party, Congress (1), on May 4, he stated that India believes Pakistan "is developing a nuclear weapon," and we are looking into various aspects of this to see what action we should take. (21) Gandhi also criticized other "powers" for not preventing Pakistan's nuclearization.

Gandhi subsequently denied that he had intended to suggest that India was considering nuclear arming, (22) but in a key interview with Le Monde on June 5, he left little doubt that such consideration was taking place. Asked, "Do you think Pakistan already has nuclear weapons? If so, what will be your response?" he replied,

Yes, we think that they are very close to having one or that they already have one. In fact, more than one .... We, for our part, have not yet taken any decision. But we are thinking about it. You must understand that for India it is very worrying that Pakistan should have a nuclear weapon. Islamabad has already attacked us three times. The fact that they had the bomb would therefore change all the rules of the game. We must therefore think about this seriously.

He went on to hint that India had already made all the components for nuclear arms, permitting their rapid assembly once a decision to fabricate them was taken:

In principle we are opposed to the idea of becoming a nuclear power. We could have done so for the past 10 or 11 years, but we have not. If we decided to become a nuclear power, it would take a few weeks or a few months.

Q. Are you contemplating this?

A. Not yet ....

Q. ... Will you or will you not take the decision to produce nuclear weapons?

A. We have not yet reached a decision, but we have already worked on it. (23)

The strategic implications of Gandhi's comments were not lost on Pakistani observers. (24)

In a series of subsequent interviews in France and the United States during June, Gandhi appeared to pull back from these comments, stressing that while he could not rule out India's opting for nuclear arms to meet the Pakistani challenge, he was reluctant to take this step . He did not repeat the claim that India could produce nuclear arms quickly, but emphasized that India had not built a second device in addition to the one tested in 1974 and had not continued its nuclear explosives program after that date. (25) He reiterated his complaint, however, that the United States had taken a soft line toward Pakistan's nuclear arming.

During Gandhi's five-day visit to Washington in mid-June, Reagan administration officials apparently persuaded him that the United States was working strenuously to forestall the Pakistani nuclear-weapons program. In addition to noting American efforts to cut Pakistan's supplies of nuclear equipment, they may have stressed Pakistani President Zia's pledge not to produce nuclear-weapons-grade material at the Kahuta enrichment plant- a pledge apparently made at U.S. urging. (26) Whatever the details, on his return to New Delhi, Gandhi stated that "I am fairly satisfied that the United States will do everything it can" to keep Pakistan from becoming a nuclear power. (27)

During Gandhi's visit, the Reagan administration announced that it had decided to authorize the sale of advanced military technology and weaponry to India, part of a broad effort to wean New Delhi away from its long-standing ties to the Soviet Union which Gandhi had reaffirmed during a six-day visit to Moscow in May (28). During that visit, in addition to confirming their promise to send India their newest fighter, the MiG-29, Soviet leaders gave India one billion rubles of new credits and offered to cooperate in the fields of defense technology and nuclear energy-including technology for nuclear submarines. As in the case of Pakistan, the U.S.-Soviet competition to win India's favor will complicate either great power's efforts to curb further Indian proliferation. (29)

Washington's assurances that it could be counted on to slow Pakistan's progress toward nuclear arms were badly undercut in early July when ABC News, citing U.S. intelligence sources, reported that Pakistan had successfully detonated the non-nuclear portion of an atomic bomb. (30) Such a test, which presumably involved setting off specially shaped "lenses" of conventional high explosives to compress a hollow sphere of metal comparable to an A-bomb core, is a major milestone on the way to manufacturing nuclear arms. (31)

A statement by Pakistani President Zia in a joint interview with the London-based Saudi Gazette and the magazine Arabia on July 15, five days after the ABC report, intensified Indian nuclear concerns still farther after it was misquoted in the Indian press. According to Indian accounts, Zia stated in the interview that Pakistan had the right to manufacture nuclear weapons and would not bow to U.S. pressure to abandon this option. (32) (The interviews do not contain these statements, but only Zia's reiteration that Pakistan, like India, Israel, and South Africa had not signed the Non-Proliferation Treaty despite U.S. urging. (33)) Seeming to reply to these developments, Gandhi on July 17 said in an interview of his own that "a Pakistani bomb will be a weapon of blackmail," and that while India opposed making nuclear arms, "national security cannot be compromised." (34)

Undoubtedly fueled by these events and unimpressed by Pakistan's July 20 offer of a treaty renouncing nuclear arms, a convention of India's leading rightwing opposition party (the Bharatiya Janata Party) adopted a resolution on July 22 demanding that the Gandhi government "take immediate steps to develop our own nuclear bomb. (35) Three days later, Gandhi, responding to a question in an "agitated" parliament, appeared to escalate his rhetoric one more notch by declaring that India was "preparing to meet, the nuclear threat from Pakistan. (36) The same day Indian Atomic Energy Commission Chairman Raja Ramanna declared to a group of reporters that India could develop a "suitable nuclear-weapons delivery system," and that the country's growing nuclear self-sufficiency meant that "if anyone tries to twist our hand, we could flex our muscles, too." (37)

On August 8, following further parliamentary calls for action, (38) Ramanna announced the start-up of the Dhruva research reactor. Rather than merely extolling India's achievement in commissioning one of the world's largest civilian non-power reactors, the announcement carefully underscored the fact that the plant would give India the ability to produce plutonium from domestic technology and fuel -a none-too-subtle declaration that India had a new source of nuclear-weapons material free from all non-proliferation controls. (39) The announcement, which was widely publicized in Pakistan, appeared to be a warning that India's considerable nuclear capabilities would guarantee it superiority in any nuclear arms race. (40) Indeed, shortly before the announcement, India's minister of state for external affairs, Khurshed Alam Khan, told the lower house of parliament that if Pakistan thought it was ahead in nuclear technology, "let it remain in that illusion. (41) Khan went on to give the most militant statement of India's nuclear posture to date, declaring that "we will reply stone by stone" to developments in Pakistan. (42)

These pronouncements indicate that the bilateral Indo-Pakistani talks held in July did little to curb India's apprehensions. (43) While for the moment, as one Indian editorial put it, Gandhi is "keeping his nuclear powder dry," during the. summer of 1985 he appeared increasingly sensitive to mounting domestic pressure for nuclear arms, and at least one report suggested that he had already authorized Ramanna to prepare a deliverable nuclear weapon. (44)

Despite the commissioning of the Dhruva plant, for the moment India seems to have little if any plutonium it can use for nuclear arms without violating prior international undertakings. (45) India has been extract ing plutonium from spent fuel produced in the Rajasthan nuclear power station since 1982 at its Tarapur reprocessing plant, for example, but this material is covered by IAEA safeguards. India cannot use the recovered plutonium for nuclear explosives without violating its pledge to the Agency and fates a high risk that Agency inspections would catch any surreptitious violations. As India has not begun to separate plutonium produced in spent fuel from its other nuclear power plants at Tarapur and Madras, no plutonium is currently available from these sources. (46)

India has a second, unsafeguarded reprocessing plant at Trombay which operated from 1966 to 1974, when it was closed for decontamination. It apparently reopened in late 1983 or early 1984. (47) This facility can reprocess fuel from the Dhruva reactor and from India's only other sizeable research reactor, the Canadian-supplied CIRUS plant, which began full operation in 1964. Since the Dhruva plant has only recently been inaugurated, it has produced little if any spent fuel, leaving the plutonium produced in the CIRUS reactor the only material possibly available for Indian nuclear arms today. Currently, India may have between 100 and 150 kilograms of plutonium from this source, enough for 12 to 30 Nagasaki-size weapons, using the standard of 5 to 8 kilograms per device. (48)

Canada transferred the CIRUS reactor to India in the early 1960s with the condition that it be used - exclusively for peaceful purposes, although Canada did not require India to subject the plant to external inspection. India cannot therefore legally use plutonium from CIRUS for weapons; whether or not it is honoring this commitment is unknown. India did use CIRUS plutonium for its 1974 test. In 1971, Canada advised India that its peaceful-use restrriction on the CIRUS reactor meant that plutonium from the plant could not be used for nuclear explosives of any kind. India asserted, however, that the bilateral agreement did not preclude use of CIRUS plutonium for "peaceful nu clear explosives." At the time, the United States and the Soviet Union had programs to examine the use of nuclear detonations for civil excavation and rock-fracturing projects. In 1974, India held fast to its interpretation, declaring that its test was a "peaceful nuclear experiment" permitted under the Canadian transfer agreement. Canada charged that India had violated the accord, and terminated all nuclear cooperation with New Delhi. Despite this rift, Canada considers the CIRUS agreement to be in force today, although India may not. If it is, even New Delhi's liberal interpretation would preclude the use of CIRUS plutonium for nuclear arms. (The CIRUS reactor also contained U.S.-supplied heavy water, also subject to a peacefull use-only guarantee. If this material is still in the reactor, it, too, would preclude the use of CIRUS plutonium for nuclear arms.)

Indian nuclear aides may have provided a clue to their view of the matter when they announced the start-up of the Dhruva reactor. By emphasizing that the facility gave India a totally indigenous source of plutonium, they were plainly contrasting the facility to the CIRUS reactor and perhaps tacitly acknowledging the continuing applicability of the Canadian restrictions. On the other hand, Rajiv Gandhi's statement to Le Monde that India could manufacture nuclear weapons in a matter of weeks or months suggests that India considers itself free to use its stockpile of CIRUS plutonium for weapons.

Even if India denies these legal constraints on its existing stock of plutonium, New Delhi's current plans apparently call for the use of at least 50 kilograms of the material to fuel an experimental breeder reactor, due to start up in late 1985. (49) If these plans are implemented, India's remaining plutonium would appear to leave it with a very limited near-term nuclear weapons capability.

As plutonium from India's various new installations becomes available, however, this capability could grow very substantially. The 100-megawatt Dhruva reactor, two-and-a-half times larger than CIRUS, will be able to produce about 25 kilograms of plutonium per year, while the twin unsafeguarded Madras reactors (the second of which began low-power tests in mid-1985) can each produce between 50 and 60 kilograms of weapons-quality plutonium annually. (Madras 1 and 11 fuel could be reprocessed at the Tarapur reprocessing plant since India has retained the right to operate the facility without IAEA monitoring when spent fuel from unsafeguarded reactors is being processed; thus there would be no restriction on using plutonium from Madras 1 and 11 for weapons.) This would give India the capability to produce between 15 and 29 weapons annually, beginning in 1986, using the standard 5 to 8 kilograms of material per weapon and assuming CIRUS plutonium were not available. Additional unsafeguarded power reactors and reprocessing plants are under construction.

India is also reported to be pursuing research on centrifuge and laser enrichment of uranium at the Bhabha Atomic Research Center, site- of the CIRUS land Dhruva reactors. (50) In mid-1984, the head of the Bhabha Center, Dr. P.K. lyengar, hinted that India had mastered the basics of enrichment, without specifying the process involved, when he stated, "The question is only one of industrial scale transformation and installing [the] facility which takes five years." (51)

India has a range of advanced aircraft capable of delivering nuclear weapons, and a sizable space program that could provide it with an intermediate-range ballistic-missile capability in the 1990s.(52) It is possible that India is looking to the development of a missile-based nuclear force over the long term to guarantee its predominance in South Asia and achieve major power status. For the time being, however, Pakistan is India's chief nuclear concern, and it can be addressed with a far less sophisticated deterrent based on aircraft delivery. ( India's aircraft able to deliver nuclear weapons are listed at the end of this section.)

While India has become increasingly autonomous in the nuclear sector, during the past year it upgraded nuclear trade relations with three advanced nuclear-supplier countries: the Soviet Union, France, and Japan. (53) Its role as an emerging nuclear supplier in its own right was highlighted in May 1985, when an Indian Atomic Energy Commission delegation made a week-long visit to Vietnam, during which India offered to strengthen cooperation in the nuclear field. (54)

The future course of India's nuclear decision-making remains highly uncertain. If Pakistan detonates a nuclear device or declares itself to be a nuclear power without a test, it is hard to see how Rajiv Gandhi could withstand pressures to take comparable steps. But it is unlikely that Pakistan will go this far, since it would jeopardize its increasingly important relations with the United States. Nor, after eleven years of restraint, is India likely to take such action without provocation. Either step could damage India's long-standing links with Moscow and its improving ties to Washington. Moreover, under U.S. law, a test could trigger a cut-off of U.S. economic aid and possibly jeopardize multilateral aid as well. (55)

On the other hand, India has been unwilling to accept a Pakistani offer for mutual nuclear inspections or formal mutual renunciation of nuclear arms, arguing that these could not ensure that Pakistan had not already stockpiled nuclear-weapons material. Since Pakistan would be at an equal disadvantage in this regard, it seems more likely that India has rejected these offers because of its desire to avoid restrictions on its right to develop a nuclear deterrent directed against China at some future time.

In theory, a de facto nuclear freeze, under which Pakistan refrained from producing weapons-grade uranium at its Kahuta enrichment plant while India limited its production of separated plutonium, might bring a measure of stability to their nuclear relationship without requiring either to give up its underlying nuclear-weapons capability. This seems to be the approach the United States is encouraging and Pakistan has declared it is pursuing unilaterally. With new revelations of Pakistan's continuing efforts to master nuclear weaponry, however, Indian officials appear unconvinced that its program has levelled off.

In the short term, apart from the possibility of a preemptive strike against Kahuta, the key uncertainty about India's nuclear policies is how far Rajiv Gandhi is moving toward building a de facto nuclear deterrent. With nuclear projects pursued by his mother now on-line or nearing completion, the infrastructure for a substantial nuclear arsenal (based on aircraft-deliverable atomic bombs) is largely in place. What remains unresolved is whether Gandhi is standing by India's previously announced policy of not building nuclear weapons or whether he has already, or may soon, authorize the preparation of the non-nuclear components for nuclear arms and the building of a stockpile of separated plutonium so as to acquire a "bomb in the basement" capability. (Indeed Mrs. Gandhi may have taken this step. (56)

Rajiv Gandhi's readiness over the summer of 1985 to respond with increasing assertiveness to each new development in the Pakistani program is good reason to fear that he is embracing the latter stance, if only to protect himself against domestic criticism in the event that Pakistan becomes an overt nuclear-weapon state. Though its contours remain amorphous, a race on the subcontinent to build undeclared nuclear arsenals may now be under way. But it may still be possible, if both sides exercise restraint, to regain a nuclear equilibrium before nuclear arming takes place.

Map and Tables

Map of India.

Power Reactors - Operating or under Construction, Table 1, Table 2, Table 3

Power Reactors: Sources and Notes, Page 1, Page 2.

Selected Combat Aircraft Able to Deliver Illustrative 1300 lb Nuclear Bomb.

Selected Combat Aircraft Able to Deliver Illustrative 1300 lb Nuclear Bomb, Sources and Notes.

Transfers of Aircraft Able to Deliver Illustrative 1,300 lb Nuclear Bomb".

Pakistan

Pakistan is today at the threshold of becoming a nuclear-weapons state. Some observers believe it already possesses all of the components needed to assemble a number of nuclear weapons. (1) The more widely held view, however, is that Pakistan has probably completed the installations and mastered the technology essential for manufacturing nuclear arms, but remains just short of the goal because it does not yet have enough nuclear. weapons material. (2) Most observers believe Pakistan could develop reliable nuclear weapons without a test, particularly if, as alleged, it has received nuclear-weapons design information from China . (3) In late 1984 Pakistani President Zia ul-Haq, apparently at U.S. urging , pledged that Pakistan would henceforth not produce weapons-grade material at its nuclear installations. This commitment, if adhered to, could keep Pakistan a short, but important step away from nuclear arms (or at least keep any small stockpile it may now have from growing).

The centerpiece of the Pakistani nuclear-weapons effort is a gas-centrifuge uranium-enrichment plant at Kahuta, believed to be based on designs obtained illegally from the Netherlands in the mid-1970s and on equipment purchased clandestinely in Western Europe, Canada, and the United States. (4) Such a facility can be used, in theory, to produce weapons-usable highly enriched uranium, i.e., uranium in which the concentration of the most easily split type of uranium atom, uranium- 235, has been increased to 90 percent or more. In natural, unimproved uranium, uranium235 occurs at a concentration of only 0.7 percent (see appendix A). The Kahuta plant is not subject to IAEA inspection or any other non-proliferation controls, leaving Pakistan free to use its output for nuclear arms without violating any international undertakings.

Pakistan also possesses a sizable nuclear reactor, the Karachi Nuclear Power Plant, and a small research reactor. A pilot-scale reprocessing facility known as the New Labs is reportedly near completion. (5) The facility could extract weapons-usable plutonium from the reactors' spent fuel, giving Pakistan a second route to nuclear arms. However, the reactors are under IAEA "safeguards" so that any plutonium produced in them could not legally be used for nuclear explosives, and any surreptitious movement of their spent fuel to the unsafeguarded New Labs plant would probably be detected. (6) In August 1984, Pakistan announced that it was producing high-quality graphite. If further purified, this could allow it to build a simple unsafeguarded graphite reactor, which could be used to produce plutonium. (7) For now, however, it appears that Pakistawis concentrating on the enriched-uranium route to nuclear arming.

By mid-1984, Pakistan was apparently producing enriched uranium at the Kahuta plant. In February 1984, Dr. Abdul Qadir Khan, head of the Pakistani enrichment program, announced that Pakistan had succeeded in producing enriched uranium for the first time but did not specify the level attained or the quantities involved. (8) Two weeks later President Zia confirmed the development, again giving few details. (9) In a June speech, however, Senator Alan Cranston claimed that Pakistan had completed construction of 1000 centrifuge units at Kahuta - enough to produce 15 kilograms of highly enriched uranium annually, possibly sufficient for a single weapon, using standard estimates of 15 to 25 kilograms of material per device. (10) Barely a month later, an unnamed "senior" Reagan administration official was quoted as saying that Pakistan had in fact produced weapons-grade material. (11) Other U.S. officials questioned this judgement, but did not dispute that at least part of the Kahuta facility was operating. (12)

Other summer 1984 press reports citing anonymous U.S. officials stated that China, long familiar with the uranium-enrichment process, was helping Pakistan run the facility. (13) Chinese aid for the plant was first reported in February 1983 and lent credibility to reports that the plant was running. (14) One account also alleged that China had given Pakistan highly enriched uranium, possibly enough for a single nuclear bomb. (15)

In mid-1984 new evidence came to light indicating that Pakistan was designing and building the non-nuclear parts for nuclear arms. Pakistan has reportedly been engaged in such activities since 1981. (16) In June, it was reported that China had provided Pakistan with a nuclear-weapons design, that of the device detonated in its fourth test. (17)

In July, three Pakistanis were arrested in Houston, Texas, while attempting to smuggle high-speed electronic switches used to trigger nuclear weapons to Pakistan. (18) Although the switches, known as krytrons, have other high-technology applications, evidence uncovered by U.S. authorities demonstrated that the devices were destined for the Pakistani nuclear program. (19)

Apparently in response to these developments - including Pakistan's brazen attempt to violate U.S. export laws President Reagan sent a letter to President Zia in September, expressing strong U.S. concern over Pakistan's continuing nuclear activities and, it appears, urging Zia not to produce weapons-grade uranium at the Kahuta plant. (20) Since 1981, Pakistan has been the beneficiary of a $3.2 billion U.S. economic and military assistance program, which included the sale of 40 advanced F-16 aircraft, all aimed at bolstering Pakistan against Soviet pressures from neighboring Afghanistan.

The U.S. law authorizing the aid package, which is to run through 1987, specified that the aid would be terminated if Pakistan detonated a nuclear device.(21) In addition, the Reagan administration warned General Zia separately in 1982 that U.S. aid would be jeopardized if Pakistan began to extract plutonium from spent fuel at its unsafeguarded New Labs reprocessing plant, a step that would be a telltale sign that Pakistan had secretly violated IAEA safeguards, which covered all of its known sources of spent fuel. (22) Apparently not until the Reagan letter of September 1984, however, did the administration suggest that Pakistan's producing 1 weapons-grade uranium might similarly put U.S. aid at risk. Perhaps U.S officials did not believe until about this time, that the Kahuta plant would function as planned.

The Reagan Jetter came at a moment of increasing Pakistani dependency on the United States. Between late August and early October, some 40,000 additional Soviet troops were deployed in Afghanistan, many close to the Pakistani border. At the end of September, Soviet planes bombed Afghan refugee camps in Pakistan and began repeatedly violating Pakistani air space, triggering concern that even more aggressive acts were in the offing. (23)

Relations with India were also deteriorating amid reports that Indian military planners had urged Prime Minister Indira Gandhi to launch a pre-emptive attack against the Kahuta enrichment plant. (24) On October 12 Pakistan declared that, if attacked, it would retaliate, and the Indian press used the term "war clouds" to describe the state of Indo-Pakistani affairs. (25)

In early October, Pakistan had sought U.S. authorization to purchase Hawkeye early-warning radar planes, ostensibly to help meet the increasing Soviet challenge but no doubt to address the threat from India as well. Shortly afterwards, U.S. Ambassador Deane Hinton indicated Washington's sensitivity to Pakistan's plight by declaring that the United States 11 would be responsive very quickly to contingencies from the west [i.e. from Afghanistan] ... and if the contingency is from the east [i.e. from India] we will not be neutral, if there is an act committed by anybody of flagrant aggression." (26) Congress, too, appeared willing to continue U.S. support, the Senate rejecting an amendment by Senator Alan Cranston to the continuing resolution appropriating monies for fiscal year 1985 that would have cut off aid to Pakistan unless the President certified that it was not "developing a nuclear explosive device or acquiring ... technology, material, or equipment for the purpose of manufacturing or detonating a nuclear explosive device." (27) On November 16, Pakistani Foreign Minister Yaqub Ali Khan, on a visit to Washington, reportedly gave the White House a reply to Reagan's September letter. Apparently Pakistan indicated its readiness to adopt the U.S. -recommended approach insofar as Kahuta was concerned. (28)

Also in November, in the wake of Indira Gandhi's assassination, General Zia sought to patch up relations with India by launching a "peace offensive. (29) Though only modestly successful in reducing tensions between the two countries, the initiative, together with efforts to ease frictions with the United States, may have been part of a larger effort by President Zia to ease external pressures on his regime.

In February 1985, Zia announced that Pakistan had been able to produce a small quantity of low-enriched uranium, in his words "to acquire enrichment to a grade where it is necessary to run a nuclear plant." Zia stressed, however, that "it is less than 5 percent, and enriched uranium for weapons-grade is about 90 percent. " (30) Though Zia correctly stated that the material produced could not be used for nuclear arms, the ability to produce it amounted to a declaration that Pakistan was, in principle, capable of producing weapons-grade uranium. Owing to the peculiarities of the enrichment process, the hardest part of producing highly enriched uranium is upgrading natural uranium to the range of three to five percent uranium-235; upgrading this material to the high-enriched level requires far fewer steps.

Of more immediate interest is that low-enriched uranium fueled reactors use 3-percent enriched uranium, not 5-percent material. This suggests that Zia's emphasis on the fact that enrichment levels were below 5 percent - a point he carefully reiterated in a subsequent interview on March I- was intended as a signal that Pakistan was abiding by the U.S.-proposed ceiling. (31)

Unfortunately, in the March 1 interview, Zia called his credibility into question when he misleadingly stated that Pakistan's earlier effort to obtain krytrons from the United States was entirely innocent and unrelated to Pakistan's nuclear efforts: "We're trying to find timing switches for revolving lights, you know, the type you put on ambulances. For search lights, ambulances, lights affixed to the tops of buildings to warn aircraft of the buildings' height." As for the three men who were arrested for smuggling, Zia said, "And incidently, they were arrested, but they were let off. So, you see, they were not stealing anything." (32) But correspondence between the key defendant in the case and parties in Pakistan made it clear that the devices had been intended for the Pakistani nuclear program. (See Chapter 11, Pakistani Prosecutions.)

Despite the uncertainties raised by these comments, two reports in March stated that Pakistan had, in fact, limited activities at Kahuta, laying off some 200 workers and assigning Dr. A. Q. Khan some non-nuclear duties. (33) In this ambiguous climate, on March 13, the Reagan administration agreed to supply Pakistan with sophisticated air- to-air missiles to help it fend off continuing violations of its air space by Soviet and Afghan aircraft. (34)

Congressional support also remained firm. Acting in the wake of increasing, Soviet pressure on Pakistan to curtail its aid to Afghan insurgents, the Senate Foreign Relations Committee and a key House subcommittee voted in late March to continue U.S. economic and military assistance under the $3.2 billion aid package approved in 1981. (35) Each adopted an amendment, however, that would later be enacted into law, seeking to restrain the Pakistani nuclear program without terminating U.S. assistance immediately. (36) The Senate committee voted that aid to Pakistan could continue as long as the President determined that it did not yet "possess" a nuclear explosive device - a relatively mild formulation that did not require Pakistan to cease any on-going efforts to develop nuclear arms, as long as no complete weapons were assembled. The House subcommittee, responding to the July 1984 krytron affair, adopted an amendment to suspend aid to any non- nuclear weapon state that violated U.S. export laws in order to obtain equipment for the manufacture of a nuclear explosive device. (37) The provision gave the President broad discretion to waive its application, however. (38)

Through June and July Pakistani officials began a major campaign to rebut India's allegations, intensified during Prime Minister Rajiv Gandhi's visits to Moscow, Paris, and Washington, that Pakistan was developing nuclear arms. The campaign not only professed Pakistan's entirely peaceful nuclear intentions but took the offensive. President Zia and his top aides accused India of provoking a nuclear arms race by building -up its nuclear capabilities and by refusing Pakistan's repeated offers of mutual nuclear inspections and bilateral non-proliferation pledges. India's provocative announcement in early August that with the commissioning of the Dhruva reactor it had gained the capability to produce plutonium without foreign assistance provided new grist for Pakistan's nuclear apologists, one top Pakistani aide calling it "yet another major step" on India's nuclear march. (40)

The effectiveness of this campaign was severely undercut by developments in July and August. On July 11, ABC News reported that Pakistan had passed a major new milestone in its quest for the bomb by successfully testing the non-nuclear triggering package for a nuclear weapon. (41) If this report was accurate and Pakistan now possesses sufficient nuclear-weapons material for one or more weapons, as some believe, then it has become a de facto nuclear weapon state. Even if it has upheld its reported understanding with the United States not to produce weapons-grade uranium, this new accomplishment would still represent a significant advance, which would leave the country only a short, though important, step from nuclear arms.

According to the ABC account, Pakistan used a U.S. -made krytron to detonate the triggering package. Presumably the item had been obtained through another network than that of the men who were arrested in July 1984, since all of the krytrons they Were trying to export were seized. This suggests that Pakistani smuggling operations in the United States have been more extensive than previously realized and further undercuts the credibility of President Zia's assertions in March that Pakistan's motives for acquiring krytrons were entirely innocent.

Notwithstanding these developments, which were apparently known to U.S. officials for several weeks prior to their disclosure, on July 11 - the very day of the ABC report-the Reagan administration announced that it was rushing anti- aircraft missiles to Pakistan to bolster its defense against increasing Soviet and Afghan air incursions. (42) The episode graphically highlights the seemingly intractable dilemma confronting U.S. decision-makers as they seek both to contain Soviet adventurism in Afghanistan by strengthening a regional ally and to restrain the Pakistani nuclear program, which threatens to aggravate the incipient South Asian nuclear arms race and, by its example, to undermine U.S. non-proliferation efforts worldwide.

In August, yet another report of Pakistan's pursuit of nuclear armaments surfaced. According to U.S. officials, the Pakistani Army had been attempting to purchase a small U.S.-made flash X-ray machine from the Hewlett-Packard company. (43) Flash X-ray machines are used to take split-second photos through solid materials of very rapid processes and are used in nuclear-weapons development programs to observe dummy nuclear cores as they undergo compression following the detonation of a nuclear-weapon triggering package. (44) In effect, such machines, which played a key role in the Swedish and French nuclear programs, permit verification that the triggering mechanism is working as designed. (45)

The machine being sought by Pakistan was too small to be used for this purpose and was intended to calibrate artillery guns. Nevertheless U.S. officials be came concerned because the training to be provided for the small machine was the same as that needed to operate a larger flash X-ray apparatus that Pakistan had obtained from Sweden in 1982. Washington had intervened in that instance and persuaded Sweden not to supply the operating manuals and spare parts needed to run the larger machine, which was apparently sitting idle. U.S. officials became suspicious of Pakistan's purpose when they learned that two individuals who were scheduled to come to the United States for instruction were employees of the, Pakistan Atomic Energy Commission. (It may be noted that there are alternatives to using a flash X-ray machine for verifying the effectiveness of nuclear-weapons triggering packages. Thus Pakistan's lack of this equipment does not mean that it could not have conducted the triggering package test revealed in July and judged it a success. (46)

One factor that may slow Pakistan's nuclear efforts in the future is the loss of assistance from China. In July, President Reagan sent the long-pending U.S, China nuclear agreement to Congress for review. It is widely believed that officials held up, the pact because of concerns over continuing Chinese nuclear aid to Pakistan. During the intervening months since the accord was initialed in April 1984, Washington is believed to have negotiated an understanding with Beijing confirming that China will not assist non-nuclear weapon states to develop nuclear arms and, in particular, that it will terminate aid to Pakistan. (47) The details of the new understanding and of the assistance China is said to have provided Pakistan have not been publicly disclosed; China has denied providing any aid.

The future course of Pakistan's nuclear program remains unclear. It is possible that Pakistan is already a de facto nuclear weapon state, but it is more likely that it has everything necessary for a small number of nuclear arms except highly enriched uranium. In principle, it could produce this at the Kahuta plant, but it is not doing so, either because the plant is not yet capable of improving uranium to this level or because Pakistan is adhering to an understanding with the United States not to take this step.

Given President Zia's increasing dependence on the United States for arms and political support as he confronts the Soviets in Afghanistan, there is reason to hope that he will, in fact, refrain from producing weapons-grade uranium as he has declared, and, if some weapons material is already available, that he will refrain from conducting a nuclear test. Judging from the experience of the past year, however, Zia seems likely to forego the development of other elements of a nuclear-weapons program. Thus, at best, Pakistan will remain one short step from a "bomb in the basement," with a stockpile of low-enriched uranium that could be rapidly upgraded. This might leave Pakistan months away from nuclear arms, while India, as its plutonium stocks grow, could have a number of undeclared complete weapons or ready-made components that could be assembled in weeks or less.

While this stand-off would be far short of an all-out nuclear arms race, the situation would be a fragile one. Each new revelation that Pakistan has mastered some aspect of weapons-making has tended to excite passions in India, pushing it ever closer to an overt weapons program. At the same time, India's repeated hints that it is moving toward nuclear arming, and its recent muscle-flexing in announcing that it had a new source of plutonium free from non-proliferation controls, have exacerbated Pakistan's already strong sense of nuclear inferiority and undoubtedly increased domestic pressures on President Zia to abandon the few remaining restraints on Pakistan's nuclear activities.

India's continuing rejection of President Zia's offer for mutual nuclear inspections and the inability of the two nations to reach agreement on less controversial issues in their bilateral talks make it unlikely that nuclear tensions between them will soon ease. It remains to be seen whether a nuclear equilibrium can be established below the level of nuclear arming.

Map and Tables

Map of Pakistan.

Pakistan Power Reactors Operating or Under Construction, Table, Sources and Notes.

Combat Aircraft Able to Deliver Illustrative 1300 lb Nuclear Bomb: Table. Sources and Notes.

Timetable of Transfers of Selected Combat Aircraft Able to Deliver Illustrative 1300 lb Nuclear Bomb.

Iraq

Iraq's nuclear program has been dormant since Israel's 1981 bombing raid destroyed its centerpiece, the large French-supplied Osiraq research reactor outside Baghdad. Although France has agreed in principle to help rebuild the facility under strict non-proliferation controls, the matter remains under negotiation. (1) Similarly, implementation of the Soviet Union's offer to build a nuclear power plant in Iraq is still at the site-selection stage. (2) Iraq's continued possession of 12.5 kilograms of French-supplied highly enriched uranium fuel, possibly enough for one carefully designed nuclear weapon, could, however, present a potential proliferation threat. (3) And Iraq appears to have remained interested in acquiring nuclear arms. In 1982, according to a report now supported by new evidence discussed in Chapter II, it expressed interest in purchasing plutonium from an Italian arms smuggling ring, Iraq is a.party to the Non-Proliferation Treaty, but its 1984-85 violations (4) of the 1925 Geneva Protocol against using lethal chemical weapons, to which it had agreed, raise fundamental questions as to the strength of its other arms-control commitments.

There is a growing consensus, based on strong circumstantial evidence, that the hidden objective behind Iraq's expansion of its nuclear sector in the mid-1970s was the gradual acquisition of a nuclear-weapons capability (5). A new analysis of one key element of this circumstantial case, discussed below, provides strong confirmation that this was, in fact, Baghdad's goal. During the past year Iraq also became the third nation to attack an enemy nuclear-reactor site - and apparently the first to do so for the purpose of inflicting economic injury. The belief that Iraq intended to use nuclear equipment obtained in the late 1970s to pursue development of a nuclear weapons capability is based on several factors.