The Point of No Return

First Pakistan's A.Q. Khan showed that any country could have made a nuclear bomb. Then he showed—not once but three times—why the nuclear trade will never be shut down

See part one of this article: From Atlantic Unbound:

Interviews: "The World in Which We Live" (November 2005)
William Langewiesche discusses A.Q. Khan and the inevitability of nuclear proliferation.

"The Wrath of Khan" (November 2005)
How A. Q. Khan made Pakistan a nuclear power—and showed that the spread of atomic weapons can't be stopped. By William Langewiesche

By 1990 the mastermind behind Pakistan's nuclear bombs, Dr. Abdul Quadeer Khan, was living flamboyantly in Islamabad—indulged by Pakistan's military and civilian leaders; adored by the masses; ensconced in a multitude of luxurious houses; surrounded by bodyguards and sycophants; writing checks to schools, charities, and mosques; lecturing; and continuing to lead the large government laboratory that carried his name, in nearby Kahuta. In addition to running the 10,000 centrifuges and producing the highly enriched uranium necessary for Pakistan's nuclear arsenal, the laboratory had diversified into the design and production of other weapons, and was beginning to work on the problem of nuclear delivery by means of ballistic missiles. Pakistan, which had not yet tested its warheads, continued officially to deny the existence of a nuclear-weapons program, but its denials were sly and patently insincere, like parodies of diplomatic sophistication, not intended to be believed. Particularly since the successful showdown with India three years earlier, during which both Khan and the Pakistani president were alleged to have threatened their Hindu neighbors with annihilation, Khan had been freed from the need to be discreet. In public he had assumed the role he believed he deserved, no longer of just another refugee from the Partition, or of an arriviste in a land of the poor, but, rather, of Khan the Magnificent, a "brilliant scientist" who was wise and progressive, a patriot, and, indeed, the savior of Pakistan. Moreover, all Pakistan seemed to agree.

The fame had unbalanced him. He was subjected to a degree of public acclaim rarely seen in the West—an extreme close to idol worship, which made him hungry for more. Money seems never to have been his obsession, but it did play a role. The nuclear laboratory was nourished by a large and secret budget for which no accounting was required and from which Khan freely drew funds as if they were his own. One might expect that Khan's largesse would have triggered an investigation, but in Pakistan it did not. I have repeatedly asked people there if they ever wondered about the origins of Khan's wealth. One man close to the ruling military regime led by the current dictator, General Pervez Musharraf, tried recently to convince me that Khan's wife, Hendrina, came from a rich Dutch family, and that it was her money he was spending. But most people were straighter with me. They made it clear that my question was naive, and typical of an American abroad; they had not wondered about the origins of Khan's wealth because they had taken it as a given that he was skimming, like everyone else. A Pakistani parliamentarian made the point that some of the highest positions in the government today are held by people who are not merely corrupt and opportunistic but are the very icons of Pakistani criminality—people from families with a known history of murder, extortion, vote-rigging, smuggling, and fraud. He had once complained about this to Musharraf, who had advised him to be more realistic: Pakistan, Musharraf had said, is an imperfect society. The parliamentarian shrugged. Even the army is run like a real-estate racket, expropriating land from ordinary citizens and passing it on to officers for their personal gain. It is not by chance that Islamabad is a city of mansions, and that many of them are inhabited by retired generals. What was Khan's skimming compared with all that? And unlike the generals, who tended to lose every fight they provoked, Khan had delivered on his words.

Still, the idolization was excessive. I went to see another famous Pakistani who had received much of the same. He was Imran Khan, the Oxford-educated scion of a wealthy family, who had captained the greatest-ever Pakistani cricket team, had led it to multiple victories over the Indians, and in 1992 had capped his athletic career with a World Cup. The Subcontinent is so crazy for cricket that it essentially shuts down during important matches. Imran Khan, now fifty-three, is a tall and handsome man whose reputation for integrity—already strong—has been enhanced by his public denunciations of political corruption and by his founding of a large cancer hospital for the poor in Lahore. But that was not the point of my visit. Instead I wanted to talk to him about A. Q. Khan, and more generally about the nature of fame in Pakistan. I said, "It seems so extreme. I understand how important the atomic bomb is to Pakistan. It is important to any country that acquires it. But I'm still wondering what it is about Pakistan that such a cult could be made around one man."

He said, "You have to understand the psyche of the Subcontinent, and not just Pakistan. If you go to India, there is more idol worship there—the worshipping of 'stars'—than you will find anywhere else in the world. I was a Pakistani playing in India, but I've never had such adulation. I mean, in India everything is worshipped. They have idols for everything. Hinduism, you know."

"Okay."

"You will find Indian film stars—all of them—behaving like A. Q. Khan. It's not just the size of the crowds; it's their attitude. Their film stars are like demigods, literally. Any celebrity. Their number-one batsman, for instance: the way he is treated in India is just incredible."

"People want to touch him?"

"More! He would need security. The first time we toured India, even to cross the lobby of the hotel from the lift to the coffee shop we needed security guards. It was that sort of thing. The hotel was surrounded by thousands of people. We had never seen such a spectacle in our lives. And that culture is also in Pakistan—not to the same extent, because Islam challenges it, but don't forget that most people here were converts from Hinduism, and they have retained a lot of these qualities."

"And when you are on the receiving end? How do you keep perspective on yourself?"

He said, "People react so differently. I used to see, on the cricket team, boys coming up from poor backgrounds, and not well educated at all. When fame would hit them, some would not be able to handle it. It would destroy them. For instance, they would turn to alcohol, because that is the trapping of success in Pakistan: it is very expensive to drink alcohol. So they would get into this crowd that would want to be associated with them: the nouveau riche, the old rich, they would invite these boys. A lot of the players would lose balance, and think this was going to last forever. I think lack of education has a lot to do with it, because education allows you to stop living in a small world. What happens if you don't have an education and you get fame is that even in your own inner world you become oversized—this huge star."

"You studied history at Oxford?"

"And political science."

I said, "But A. Q. Khan was educated too."

He said, "Yes, so with A. Q. Khan I don't know. But when I was in England, I used to find that some people at the university who studied science, when they were outside their field they were quite silly—stupid, really. Whereas people who went into the arts or general fields like politics or history, they were much better socially. I remember this one nuclear scientist who was on our team, and I would wonder how he was ever going to function in life. He did well on his exams, and he even got a first, which is very rare at Oxford. But beyond that?" He laughed at the memory. He said, "I found Abdul Quadeer exactly the same. I remember we were once on a TV program together, in a panel of four or five people on a stage in front of a live audience of students. I heard him answering the questions, and I thought, 'This can't be the great Abdul Quadeer!' Because he sounded like a child, really. Someone in the audience stood up and said, 'You are so great, and I don't think you're getting the acknowledgment you deserve.' You know, he pandered to Abdul Quadeer. And instead of saying, 'Who in Pakistan has received more acknowledgment and fame?,' Abdul Quadeer said, 'Yes, you're right. No one bothers about me in this country.' And he went on and on. I was so surprised. Here was this man, in front of these kids, and he was really feeling sorry for himself. It was bizarre. I thought, 'What is going on here?'"

The convenient answer to that question now, years later, is that A. Q. Khan was running amok, that he was an addict of sorts, unable anymore to find sufficient gratification from his activities in Pakistan, and that unbeknownst to either the Pakistani authorities or the American government he was going rogue, selling his nuclear secrets abroad.

There are elements of truth in all that, as there are in most fiction. In Lahore I went to see a former finance minister named Mubashir Hasan, an engineer by training, who in his later years has become a pacifist and one of the rare critics of Pakistan's nuclear-weapons program. Hasan is a thin man with a gentle demeanor. On both days we met, at his ramshackle house in a leafy district of the city, he was dressed like a saint, entirely in white. We discussed the funding of Khan's laboratory over the years, and in general terms the extent to which Saudi Arabia and other countries had contributed to it. He said, "But if you want to find out exactly where the money came from, from what country, and in return for which state secrets, or how the accounts were kept, how much was taken out by which crooks—all that will remain secret for a long time. There is really no way of knowing."

I asked him how Khan could have gotten away with so much for so long. He said, "It is a cultural trait. The Western assumption that law should treat everyone the same way is no longer applicable in this country, in this culture. In Pakistan relationships exist only on an individual level, and as an individual I am entitled to forgive you or penalize you no matter what the law says. It is a feudal culture—or a degenerated feudal culture. That is why there is no law for the elites in Pakistan—why they do whatever they want to do. So your question of why nobody investigated A. Q. Khan? He must have had allies in high places who ignored his activities. You've given us the bomb. All power to you."

But trading in nuclear-weapons technology is more than just a form of misbehavior. To ignore such activities once they are known is in effect to participate in them. The lack of financial trails is inconvenient, but it does not obscure the essential history. A. Q. Khan had allies in high places who, rather than ignoring his activities, were directly involved. In Pakistan this can only mean the generals, including some of those currently in power, and to a certain but unknowable degree Musharraf himself. Hasan mentioned that the country's leaders had without question been given plenty of warning. It turns out that Munir Ahmed Khan, A. Q. Khan's despised rival and the director of the Pakistan Atomic Energy Commission, had been a longtime friend of Hasan's. In the late 1980s Munir Ahmed Khan had repeatedly complained to Hasan that A. Q. Khan was corrupt and, more important, that he was involved in selling Pakistan's nuclear-weapons technology abroad. According to Hasan, Munir Ahmed Khan had taken the same complaints to the generals in charge at the time, and of course nothing had been done.

Hasan used the term "traitorous" to characterize A. Q. Khan's activities. I said, "Can an activity be traitorous when the government itself is complicit, and in a country without effective law? I mean, at what point does such activity in such a place simply become a policy?"

He gave that to me. He said, "You're right. You can be a traitor only if the power is not aware."

Beyond Pakistan, those aware of A. Q. Khan's growing proclivity to export his nuclear wares included a small number of non-proliferation specialists in the intelligence services of the West. These were people sworn to secrecy, and though they were concerned, they remained largely paralyzed so long as their own governments—and particularly the leaders of the United States—placed greater importance on propping up the various Pakistani regimes than on stopping the spread of nuclear weapons. Just outside these circles, however, stood a few unofficial observers who were harder to control, and who kept peering in. The most persistent of them was an obscure American journalist named Mark Hibbs, who is largely unknown to the public, but must rank as one of the greatest reporters at work in the world today.

Hibbs is a legend within the secretive realm of nuclear power. At the age of fifty-four now, he is based in Bonn, Germany, where he lives with his longtime girlfriend when he is not traveling. He travels a lot. With a slight shift in intent he would have made an excellent spy. He looks like one too, with a bearing so ordinary for a middle-aged man that from even a short distance away—in a hotel lobby, in a restaurant, on a European street—he can be hard to identify. His face would be ordinary too, were it not for the exceptional intelligence that animates his features when he speaks and the habit he has of frowning in deep concentration while remembering the events that have shaped his work.

Those events go way back. For more than twenty years Hibbs has been reporting the news for Nucleonics Week and Nuclear Fuel, two publications with ultra-high subscription rates and correspondingly low circulations, now disseminated primarily on the Web. These publications stand among sixty-five similar ones in a McGraw-Hill group called Platts, dedicated to the petrochemical and energy industries. Other Platts titles include Megawatt Daily, Emissions Daily, and Dirty Tankerwire. Subscribers to Nucleonics Week also receive daily "Nuclear News Flashes," which could nicely be shortened to "Nuclear Flashes" if only Platts would lighten up. But Platts will not. Hibbs is its star, but he earns just a modest middle-class salary that allows him to get by. Platts has a lock on him in part because there are so few journalistic outlets for his knowledge. Much of what he writes is of short-term interest at best, and only to regulators and nuclear-energy insiders; he has filed thousands of such service reports over the years. Imbedded among them, however, are several hundred related dispatches—usually inconclusive, yet accurate and precise—that together tell an ongoing story of great consequence to the international political order, and perhaps even to the survival of mankind. That story is about the gradual failure of the United States, Europe, Russia, and China to prevent nuclear arsenals from spreading around the globe, and the ferocious determination to acquire such arsenals that upstart nations, some banding together, continue to show.

It is to Hibbs's advantage that he is not, strictly speaking, a spy. Because he works in the open, without a security clearance, he is not bound by government policies and cannot easily be silenced. There are others outside the government who share his freedom—professors, analysts, and advocates—and a few of them are very good, but none have produced results to equal his. Superficially, what he does seems simple enough: he ferrets out details from a variety of sources, fits them into patterns in his mind, and writes them up. But that process requires unlimited patience, sound technical knowledge, an intense determination to avoid making mistakes, and a sense for the plausible in a world full of lies. These are rare attributes, and in Hibbs they combine. It helps that he is not a crusader, and that although he privately regrets the spread of nuclear weapons, in his reports he takes no sides. It helps as well that publications he writes for do not accept advertisements, and understand that their value to their readers lies in delivering the news even if that news embarrasses the industry or is otherwise impolite.

He told me once that it is very lonely work, and more so even than ordinary writing, because for all his influence, he writes for an audience imbued with secrecy, and therefore rarely hears from his readers. Speaking of a meeting he attended of the UN's International Atomic Energy Agency, the IAEA, he said, "I remember I sat down in the back of the room, and there was a delegate from one of the Western countries who was sitting next to me. I looked at him, and on his lap he had a copy of an article I did on the Iranian centrifuge program. It was a Xerox of the article I did. And the copy was meticulously underlined, and some things were circled, and there were written comments in the margins. So I tapped him on the shoulder and said, 'You know, that's interesting. Can you tell me where you got that article?' He looked at me and said, 'Why do you want to know?' I said, 'Because I'm the author.' So what did he do? He turned white as a sheet, and just ran out of the room! It was the weirdest thing."

Weird perhaps, but standard for Hibbs. That was his point. He said, "Sometimes I feel that what I do is happening in a black box. The readers have security clearances. They read what I write in classified rooms in government offices and companies all over the world, and often they don't like what they're seeing. So how do they react? They go and talk to their buddies, who also have security clearances. Or they send a message to their enemies, and it's the same thing. You could be unleashing a major international crisis and wouldn't even know it, because it's all secret. Once the information is out there, you have no idea what's going on. Most of the time you just don't know. You don't hear about it. The reaction itself becomes classified. The lack of feedback is the really disturbing thing about this job."

Still, as he admitted to me, he has thrived.

It is not a life a person could seek, or even imagine in advance. Hibbs was born in 1951 in a nondescript town in upstate New York. His father was a smalltime accountant, and characteristically tidy. His mother was a housewife, and came from a large working-class Irish family. Hibbs had six uncles and aunts from her side alone, and many of them lived nearby with children of their own. As one of the oldest among the cousins, Hibbs was the focus of his aunts' and uncles' attention. When the Vietnam War came along, they assumed that the right thing to do was to go off and fight.

But Hibbs went off to Cornell instead, and they were nonetheless proud. They believed that the purpose of attending a good university was to learn good manners—how to wear a suit, sip wine, and hold a knife and fork in just the correct style. For Hibbs it did not work out that way. He joined the rebellion of the time, and when he returned home on break, he ate with his fingers and flaunted his long hair. He opposed the war in Vietnam, and wanted America to change. He marched in demonstrations. But he never became as radical as his friends. He simply could not follow along with other people's thinking without at some point considering that perhaps they were wrong. This, of course, was the real benefit of a good education, and it later served him well in deciphering the mysteries of nuclear proliferation. But even now in his family there must be some who do not understand.

In 1973, the last year of the draft, he graduated from Cornell with a degree in literature and history, and drifted to Boston, where for some years he worked as a cartographer, designing maps for public-transportation agencies. In the late 1970s he moved to New York and enrolled in a master's program at Columbia to study international diplomacy. He knew that Israel had nuclear arms, and that India had tested its own device (the "Smiling Buddha" of 1974), but he was not aware of Pakistan's ongoing program to respond in kind. If he thought about nuclear weapons at all, it was in the conventional terms of the Cold War turning hot. In July of 1981 the realities of proliferation were briefly thrust upon him when the Israeli Air Force bombed the French-built Osirak reactor in Iraq, ending a secret attempt by Saddam Hussein to extract plutonium from spent fuel and acquire a nuclear arsenal of his own. But Hibbs did not imagine that he himself would ever be involved in such matters.

He had a talent for language that allowed him eventually to learn German, Dutch, and French, as well as some Russian and a little Chinese. After leaving Columbia he stayed on in New York, working as a freelance consultant and editor, primarily for a German government office, but he found it difficult to earn a living, and so left for Europe, from which he has never permanently returned. For a while he lived in London, doing research for the Financial Times in the energy business—an area previously unknown to him, but of sufficient depth to engage his mind. He moved to Bonn, where he continued the same work and contributed occasionally to Business Week. His writing was concise. He was in no sense yet anything like a spy. Without having thought his career through in advance, he had become a reporter.

In 1986, when Hibbs was thirty-five, the Soviet reactor at Chernobyl, in Ukraine, melted down. Hibbs began to travel to the Soviet Union to report on its nuclear industry, a subject still largely hidden from view. The puzzle fascinated him. He was a fast learner, capable both of grasping physical technicalities and, more important, of navigating the complex political terrain that surrounds the use of nuclear power. His interest in the field expanded worldwide. By then he had found his outlet, too, and was writing for Platts, which soon hired him full-time.

For a while he wrote only about civil nuclear power. To anyone outside the industry it was mind-numbing stuff. But beneath the surface in Europe was action of a different sort: though the Nuclear Non-Proliferation Treaty (NPT) had come into force, and parallel protocols had been agreed upon to restrict the export of materials necessary for the construction of nuclear bombs, enforcement was lax, and individual companies, particularly in Germany, were eagerly doing business with a growing number of nuclear-weapons aspirants. Much of the business was questionable but not obviously illegal; for the most sensitive items it involved dealings with middlemen who worked through front companies and third-country destinations and provided the sellers with usable explanations. Then as now, European attitudes were soft despite official abhorrence of such activities, because in private the nuclear non-proliferation initiatives—accompanied as they were by repeated U.S. scolding—were perceived by many European officials as yet another self-serving American crusade. Along with their resentment of American domination came an understanding that providing nuclear technology (especially in big packages to the Middle East) was a way of gaining European influence in strategically important regions. Hibbs began to write extensively about the activity in 1988, when the German parliament, the Bundestag, finally opened an investigation. For two years government officials and businessmen were called in to explain their dealings, especially with Pakistan. Some sessions were tense. In the end, however, the Bundestag issued a report absolving German companies of involvement in the trade. The report was absurd. Hibbs wrote it up at length but brushed it aside.

Neither he nor his readers needed the Bundestag to tell them what was going on. Pakistan's nuclear-weapons program, though officially denied, was well known to the world by then, as was the existence of its procurement network in Europe. A. Q. Khan had been bragging about himself for years: it was he who had stolen the centrifuge designs from the Dutch-British-German consortium URENCO, in Holland; who had used them to construct a plant in Pakistan for the enrichment of weapons-grade uranium; who had built the necessary procurement network worldwide; and who, he sometimes slyly implied, had followed through with the design for a Pakistani bomb. Indeed, Khan was already such a figure in the West that as far back as 1985 he had been featured in Time magazine. In the European and American press he was portrayed as an evil scientist and a treacherous spy, but in many ways it made sense for Pakistan to have the bomb—as much, at least, as it does for other countries, including the United States. In private Hibbs would probably have disagreed; recently to me he characterized Khan's efforts as "diabolical." But he kept those thoughts to himself, and in print went hard after the nuclear news.

Some of that news continued to be of Pakistan. A few of the most blatant suppliers were prosecuted in Germany and the Netherlands; more were identified but, for lack of proof of intent, they were left to continue the trade. The Pakistani procurement network remained large and robust, providing not only for A. Q. Khan's uranium-enrichment plant but also for a parallel program to acquire plutonium—the alternative material for a fission bomb—which was led by Munir Ahmed Khan, at the head of the Pakistan Atomic Energy Commission. Over the course of the 1990s Hibbs dug through the evidence, forming hypotheses in his mind and asking increasingly precise questions of obscure specialists who knew of him but did not necessarily know what he was after.

I asked him to take me inside his investigative process. He said, "There was a German company that was exporting a certain device to Pakistan that in the worst case could have been for the nuclear-weapons program. All the surface evidence suggested that no, this was the case of a scientist who went to Pakistan and sold some equipment to an innocuous research institution—and that was the end of it. That was the official explanation in Germany. But then I looked at it, and looked at it again from a couple of different points of view, and there was a very small chance that the equipment could have been used for removing tritium, a gas that comes into play in nuclear weapons. I wondered, 'Could you use the equipment for that? I mean, it's pretty far-fetched, but is it possible?' After a while I realized you could use it for that. Then I wondered, 'Is there anything else he did that indicated he had the know-how?' I started talking to people, and just kept asking one question after another. I was operating in a vacuum, and people in the nuclear laboratories in the U.S. weren't supposed to be telling me anything. But I kept asking, 'Could it be this?' and 'Could it be that?' and 'If that's true, could it also be this?' And after several months, when it appeared that the answers were all yes, finally the article I wrote stated that this guy in Germany had exported equipment to Pakistan for the removal of tritium. And I'm told that inside the U.S. laboratories there was a shit storm, because what I wrote matched what they were thinking, and it was all classified."

I said, "But day-to-day what does that work really look like, tangibly?"

He said, "It looks like talking to as many people as you can."

"With narrow questions?"

"Yeah, but also general, generic questions. If you call somebody in a government laboratory and you ask him, 'Do you think this German is exporting equipment for Pakistan's nuclear-weapons program?,' the guy will probably hang up. He'll think, 'What? Nuclear weapons in Pakistan? This is a government laboratory, and this reporter doesn't have a security clearance! Get OFF the phone!' But if you meet the same guy at a conference and you ask him a generic question about the configuration of the machine—'If the piece of equipment is configured like this, could it be used for that, could it be used for this?'—then maybe he will answer. And then if you ask a bunch of different people the same generic question, maybe something will crystallize out of it, and you'll get closer."

Hibbs writes about wisps of smoke. Still, over the years the wisps have added up. A growing number of diplomats and policy analysts now express satisfaction that earlier predictions of a widely nuclearized world have been proved wrong, but it is largely just the timing that was off. Non-proliferation functionaries would earnestly disagree with such a gloomy assessment, as they must because of the business they are in. It is natural that they cite their successes, and necessary that they struggle on. But despite the occasional collapse of nuclear-weapons programs, and the inconveniencing of programs in place, the readers of Platts over the years can have harbored few illusions about the trend. Proliferation is a ratcheting affair that moves in fits and starts, and often slips backwards, but gradually and incrementally progresses. Diplomatic efforts to suppress it are weakened by national jealousies, UN-style dithering, higher geopolitical priorities, the sheer volume of international trade, and, at the most fundamental level, the inability of the Great Powers themselves to disarm. In a world where perhaps ten countries have already acquired the bomb, and another thirty have the capacity to build one relatively fast, rational reasons for choosing that path will now and then arise. If even Pakistan can go nuclear, almost any other nation can as well.

None of this was an argument for giving up on the struggle to restrict nuclear weapons, or for turning away from the non-proliferation treaties and accords as they existed. Indeed, Hibbs's writing made it clear that these international instruments needed to be strengthened. By 1989 smoke was rising on many horizons. Word emerged through Israeli intelligence that Saddam Hussein, having suffered the loss of the Osirak reactor, had reconstituted a nuclear-weapons program and was pushing to build a functioning bomb within two years. Iraq had signed the NPT, and had therefore been subject to twice-yearly inspections by the International Atomic Energy Agency, which is headquartered in Vienna. Such inspections were at that time strangely unobtrusive worldwide. They occurred by invitation only, with itineraries controlled by the host governments and negotiated weeks in advance. In Iraq they had resulted in little more than some good dinners, perhaps a belly-dancing show, and the chance to tour around. By the IAEA's diplomatic standards the country remained clean. This meant either that the report from Israel was false (and probably propaganda) or that a serious failure had occurred at the very core of the NPT. After the 1991 Gulf War it turned out that both were to some degree true.

Meanwhile, Hibbs filed a dispatch—the first of more than 160 that he wrote on proliferation in Iraq—in which he described widespread skepticism among nuclear experts that any Iraqi program could be so close to producing a bomb, and gave the reasons why. In time his skepticism became a point of contention with the mainstream American press, which reflected Washington's tendency to take Saddam at his word and to exaggerate the immediate nuclear danger that he posed. Hibbs reflected a more sober view. As always, his writing was based on the closest attention to detail. He did not doubt that Iraqi engineers could someday build a bomb, and that Saddam Hussein was just the man to use it, but he was aware of the technical difficulties that would first have to be overcome.

The largest difficulty was acquiring sufficient quantities of bomb-grade fuel. As is now known, Iraq had abandoned the hope of extracting plutonium from its civil reactors and, like Pakistan and other countries, had decided instead to pursue weapons built around cores of highly enriched uranium. As a signatory to the NPT it had the right to use lower-grade enriched uranium for power generation and research, but such fuel was monitored, or "safeguarded," by the IAEA, and therefore was difficult to divert into a weapons program without attracting attention. The solution was to construct an entirely separate enrichment program that from start to finish would remain hidden from sight, or at least be deniable. In the mid-1980s—toward the end of the war with Iran, when Saddam was still something of a friend of the United States—Iraq launched a secret initiative to buy or steal the necessary parts, mostly from the West. Hibbs eventually caught on. Writing from Bonn in the summer of 1990, just after Iraq's invasion of Kuwait, he broke the news of an extensive Iraqi procurement effort under way in Europe and the United States, whose purpose was to acquire the materials necessary for a uranium-enrichment centrifuge plant.

Over the months that followed, during Iraq's occupation of Kuwait, Hibbs continued to dig into Iraq's European activities; without drawing the connection explicitly, he described a gray-market procurement network that was remarkably similar to Pakistan's, and in some cases exactly the same. The main suppliers were German, with the Swiss and others, primarily in Northern Europe, pitching in productively. In December of 1990, during the buildup to the Gulf War, the London Sunday Times published the alarming news that Iraq's centrifuge program was merely a year away from enriching enough uranium for a bomb. On Christmas Eve, Hibbs reported that the centrifuge in question was of an early URENCO design, not quite as advanced as the one presumed to have been stolen by A. Q. Khan, but perhaps good enough to do the job—if it could be made to spin at the necessary speeds without falling apart, and then linked to a minimum of several hundred like-spinning clones arranged in a "cascade" through which gassified uranium would flow. The source of the centrifuge design was believed to be a German subcontractor to URENCO, whose engineers seem to have had access to the plans even after leaving the company. Hibbs had traveled to Munich to meet one of the prime suspects, a centrifuge expert named Bruno Stemmler, who had gone to Baghdad as a consultant two years before, and against whom a German criminal investigation had been dropped in frustration. Hibbs got him to talk. Stemmler claimed to have no knowledge of Iraq's suppliers, but he admitted to having met with Iraqi engineers, and he described what he had seen of their centrifuge in detail. He was so untroubled by the contacts he had made that he even told Hibbs about offering the Iraqis advice on modifications that would shorten the enrichment cycle. "But they were not interested,'' Hibbs—clearly reserving judgment—reported that Stemmler said. The Iraqis' lack of interest does seem unlikely, given the personal consequences of failure under Saddam Hussein, but Stemmler's measure of their program later turned out to be for the most part true. In brief, they had managed to build, at best, perhaps a single centrifuge, which they had not fully tested. They must have made progress in the two years following, but according to Stemmler, as reported by Hibbs, the existence of a functioning centrifuge cascade in an operational enrichment plant, which was the premise of the report in the London Sunday Times, remained "completely far-fetched and absurd."

Then came the Western bombing campaign and the short Gulf War, during which Iraq was easily pushed out of Kuwait. As part of the peace settlement new and more assertive nuclear inspections were imposed on the Iraqi regime. The inspections were run by the IAEA and a group created specifically for this purpose, the United Nations Special Commission, or UNSCOM, whose job was to search for weapons of mass destruction in all forms—chemical and biological as well as nuclear. The inspectors had the right to talk to whomever they pleased, and to go wherever they wanted without advance notice. The Iraqis obstructed them, of course, but over the next few years the inspectors uncovered a lot. What they found came as a shock. Though it was true that the Iraqi centrifuge program had never moved beyond the initial testing stages, it was larger and more serious than had previously been known. In intent, at least, this was not merely an experimental effort but an industrial-scale attempt to build nuclear weapons. Furthermore, Iraq had been pursuing an even larger program using electromagnetic machines called calutrons—an enrichment technology declassified by the United States in 1949, and thought to be so obsolete that no one, including Hibbs, had suspected Iraq might choose it. In this case, however, "obsolete" meant easier to acquire. The calutron program had advanced significantly beyond the centrifuge efforts. In the summer of 1992 Hibbs reported that at the start of the Gulf War it had been about three years from producing enough highly enriched uranium to give Saddam his first atomic bomb.

But the greatest shock was that such industrial-scale nuclear programs—using components purchased in the West, and directly under the nose of the IAEA—had gone largely undetected. Indeed, it became apparent that unlike certain nuclear aspirants that had refused to join the NPT (India, Israel, Pakistan, Argentina, and Brazil, among others), thereby exposing their intentions, Iraq had used the treaty as a cover, gaining greater freedom of action than it otherwise would have had. The revelations emerging from Iraq provoked a crisis among non-proliferation specialists and diplomats; particularly in Europe and the United States it was understood that if the treaty was not to collapse, something would have to be done. And something was. Over the years that followed the Gulf War export-control lists were expanded, necessitating a new level of governmental scrutiny (in participating states) prior to the export of "dual use" machines, materials, and components—the often innocuous items that had been identified nonetheless as the building blocks of nuclear-weapons programs. The export-control-list expansion forced the European procurement networks largely underground by eliminating much of the ambiguity that had existed until then, and it required companies and consultants to break their national laws if they wanted to pursue their nuclear-weapons business, questions of conscience aside. In the new context of a now proven danger, some of the bureaucratic resistance to American activism faded away, and cooperation among the various intelligence and law-enforcement agencies increased. Furthermore, in 1997 the NPT was augmented with an "Additional Protocol" that allowed for more-rigorous IAEA scrutiny of declared nuclear facilities in those countries willing to join. There was a twist, which continues to limit the effectiveness of IAEA inspections today: as an international agency whose primary purpose is to promote the civil use of nuclear power, the IAEA was expressly prohibited from developing expertise in nuclear weapons, and could address proliferation only by auditing civil facilities for signs of military diversions—working around the edges of the real question on everyone's mind. Nonetheless, the Additional Protocol, like the other measures taken in the aftermath of the Gulf War, probably helped to buy the world some time.

But even as the advanced nations closed ranks against conventional proliferation—the leakage of nuclear know-how from nations like Germany to nations like Iraq—the new form of proliferation emerged, in which the nuclear upstarts began to help one another. There were reasons for this beyond convenience and greed. They amounted to a moralistic rejection of the discriminatory nuclear order as enshrined in the NPT. In principle the rejection made sense. If all peoples are created equal, why not here, too? This was the same question that until recently had undermined export controls in Europe and elsewhere. The answers were practical rather than principled, and for certain countries they were not good enough. In any event this new form of proliferation lay largely beyond the conventional structures of control. As is now widely known, the pioneer was Pakistan, where A. Q. Khan exploited the connections he had developed in acquiring nuclear weapons and, by neatly diverting the inbound procurement flows, eventually set up a virtual nuclear-weapons market in which countries could buy the entire package, from the necessary machine shops and centrifuges to the blueprints for a bomb.

Without quite knowing what he was looking at, Mark Hibbs wrote about the early signs. In October of 1990, in a report on the Iraqi centrifuge program, he described skepticism among nuclear specialists that Iraq possessed the necessary engineering expertise to mount such an effort, and he mentioned the possibility of Iraq's secret cooperation with Brazil or Pakistan—two countries beyond the IAEA's reach. It later turned out that those particular suspicions were unfounded—that the expertise at play in Iraq was mostly German. But the possibility of third-country proliferation was obviously on people's minds. In November of 1993 Hibbs broke the news that during a meeting the previous month in Baghdad, Saddam's deputy premier, Tariq Aziz, had told IAEA and UNSCOM inspectors that a Pakistani or Indian known simply as "Malik" and residing in Britain had in 1989 arranged for shipments of special high-strength steel for the construction of Iraq's clandestine centrifuges. Hibbs wrote that on the basis of this information British agents had tried to identify the man but had been unable to find him. At the end of his dispatch he mentioned that following the Baghdad meeting the IAEA director, Hans Blix, had reported to UN Secretary-General Boutros Boutros-Ghali that the information provided by Iraq appeared to be mostly complete, though there were still some gaps.

One year later Hibbs returned to the subject, writing that Malik was a Pakistani whose full name was Mazhar Malik and who lived in South London, where he had a small trading company, Development & Technology Enterprises, Ltd. The company appeared on a U.S. Department of Agriculture list as having inquired into the export of food staples and cigarettes, among other materials. Hibbs reported that the high-strength steel had been sent by an Austrian firm through the port of Antwerp, and had traveled aboard two Pakistani ships to Dubai before being trucked to Iraq. Because of lax Austrian export laws at the time, the British agents who had finally tracked Malik down concluded that nothing illegal had transpired, and the IAEA now considered the case to be closed. Hibbs had gotten through by phone, and Malik had denied that he knew anything about the shipments of steel.

Then, in the summer of 1995, the name came up again. Saddam's son-in-law Hussein Kamel, the man in charge of the pre-war nuclear program, had fled to Jordan, and he was talking. Among the documents subsequently found on his chicken farm in Iraq was a top-secret memo written by agents of the Iraqi intelligence service, the Mukhabarat, and addressed to an unknown person in the nuclear-weapons program. The memo described an approach made to the Mukhabarat in October of 1990, during Iraq's occupation of Kuwait, by a Pakistani named Malik, who claimed to be an intermediary for A. Q. Khan, and who offered to assist Iraq with centrifuges and the construction of a bomb. The asking price was $5 million up front, with an additional 10 percent commission to be paid on all materials and components obtained. In translation the memo read:

Top Secret Personal …

We have enclosed for you the following proposal from Pakistani scientist Dr. Abdul Quadeer Khan regarding the possibility of helping Iraq establish a project to enrich uranium and manufacture a nuclear weapon. The above mentioned expressed it as follows:

1. He is prepared to give us project designs for a nuclear bomb.

2. Ensure any requirements or materials from Western European countries via a company he owns in Dubai.

3. Request a preliminary technical meeting to consult on the documents that he will present to us. However, the current circumstances do not allow for an immediate meeting with the above mentioned. There is the possibility of meeting with the intermediary that we have connections and good relations with in Greece.

4. The motive behind this proposal is gaining profits for him and the intermediary.

5. The project has been given the code name of A-B to use in correspondence and consultations.

Please review and give us your opinion on this matter, so that we can take the initial steps to consult with him, according to the notes and instructions that we received from you. Thank you.

The Iraqis suspected that the offer might be a scam or a trap; they asked for a sample of the goods—possibly a component or blueprints. They never received the sample, because the Gulf War then erupted, and the ensuing international inspections swept away any chance for such a deal. Nonetheless, the offer was almost certainly real. After the memo was discovered, Western intelligence agencies and the IAEA kept it secret for nearly a decade, until 2004, presumably because of political pressures, or to allow further information on Khan's activities to be gathered. Neither Hibbs nor any other reporter wrote it up at the time. But Hibbs was catching on to Khan. He was beginning to piece together the signs that Pakistan's nuclear-procurement network had expanded into the business of spreading these weapons around.

Pakistan's sale of nuclear-weapons technology abroad did not require a deliberative process, a chain of command, or a formal commitment to proceed. More likely it took the shape of opportunities that occasionally arose and were acted on by a small circle of friends—the country's military rulers, its co-opted politicians, and, of course, A. Q. Khan and his men. They knew that such activities would provoke the United States, Europe, and other great powers—but they did not think of themselves as bad people, or believe that they were breaking international law. Whatever profits they hoped to gain from these deals would have been as much for the treasury as for their personal accounts—albeit in a country where such distinctions have little meaning. As to questions about the morality of promoting such lethal technology, they had some questions of their own—about the fairness of discriminatory non-proliferation treaties and a world order in which the established nuclear powers seemed determined to "disarm the disarmed." This was the emotional spillover from Pakistan's experience of building a bomb, and it fed a genuine sense of solidarity with all other nuclear aspirants, including even a potential antagonist like Iran.

Indeed, Iran was Pakistan's longest-standing customer. In May of 1991 Mark Hibbs reported in Nucleonics Week on the possibility that Iran had launched a secret uranium-enrichment program in pursuit of nuclear weapons and that over the previous three years A. Q. Khan had made several visits there. Soon after the article was published, Hibbs received a phone call from an American diplomat named Richard Kennedy, who at the time was the U.S. ambassador for non-proliferation, and the chief American representative to the IAEA.

According to Hibbs, Kennedy said, "I've read your last article."

Hibbs said, "Yeah?"

Kennedy said, "You know that thing about A. Q. Khan—that maybe he went to Iran? Can you tell me who told you that?"

Hibbs answered, "No."

"Can I assume it's a European intelligence source?"

"Yes."

"Will you tell me which government it is?"

"No," Hibbs said. "Does it strike your interest?"

Kennedy admitted that it did. He said, "We have a very strong interest in Dr. Khan and the Khan Research Laboratories. We pay very close attention to his work. In fact, our interest in this man is so intense that you can assume if he takes a toilet break and goes to the john, we know about it. We know where he is."

Though it would be politically inconvenient to admit this now, the United States was aware not only of Khan's peddling of nuclear wares to Iran but also of the likely involvement of the army and the government of Pakistan. Hibbs has reported that the U.S. ambassador to Islamabad from 1988 to 1991, Robert B. Oakley, went around the embassy fuming, "They sold that stuff to those bastards!" and believes that Oakley expressed the same emotion more politely at the National Security Council. Oakley, who now works at the National Defense University, in Washington, D.C., does not recall knowing of the sales to Iran when he was ambassador, and says he was not asked to raise the matter with the Pakistani government. For political reasons more than for reasons of national security, these are some of the most closely held secrets in the United States. For the same reasons, the apparent lack of good information is pointed to as yet another U.S. intelligence failure (add it to the implosion of the Soviet Union, to 9/11, to Iraq), when in reality the CIA knew fairly well what was happening, and an awareness of Pakistani actions should count as a U.S. intelligence success.

Not that awareness required great skill; in Pakistan the intent to sell nuclear-weapons technology lay clearly in view. In 1989, for instance, the Khan Research Laboratories held the first international conference in what would be a fifteen-year series of occasional courses and symposia on issues pertaining to uranium enrichment and centrifuges. These meetings, which were widely advertised, amounted to barely disguised promotional affairs, clearly intended to demonstrate Pakistan's expertise to potential nuclear-weapons customers. By the end of the 1990s the Khan Research Laboratories were sending salesmen to international arms shows—in Malaysia, Indonesia, Abu Dhabi, and back home in Karachi—where they set up booths and passed out A. Q. Khan buttons and brochures advertising their conventional and nuclear products. In 2001, on the occasion of its twenty-fifth anniversary, KRL published a proud self-portrait, part of which read:

Keeping pace with the emerging demands of the competitive international defense products market, KRL ventured to offer its expertise, in the shape of services and products, not only to the domestic consumers but also to an international audience of friendly countries … Although a fresh entrant, the participation of KRL from Pakistan was warmly welcomed. KRL has earned credibility not only in South East Asia, but also in the Middle East and West Asia. Its regular participation … has enabled the Laboratories to set up and maintain close cooperation in this vital sector in many countries.

Pakistan's Ministry of Commerce did its part too. In July of 2000 it ran a full-page notice in the English-language Pakistani press that advertised the nuclear-weapons products that Pakistan had to offer: a full line of materials and devices that ended just one step short of a ready-made bomb.

But back to Iran. In the late 1980s there were persistent rumors of secret Pakistani-Iranian nuclear agreements, and as noted, these were systematically denied. In 1991, however, Pakistan's army chief, General Aslam Beg, returned from a trip to Tehran openly advocating the export of nuclear-weapons technology to Iran, and pointing to the several billion dollars' worth of state revenue that might be in the offing. He may even have written an opinion piece in an Urdu-language newspaper expressing his enthusiasm for the idea—though in Pakistan recently he denied this to me, and I was unable to track the piece down. Beg is an anti-American with sympathies for Iran, and he says that he is the target of a Jewish conspiracy of lies. Be that as it may, he was told to keep quiet in the early 1990s, presumably because the transfer of blueprints and centrifuges was already under way.

Hibbs was onto it fast. In November of 1991, having previously written about the unconfirmed visits of A. Q. Khan, he described an unnamed Western government's suspicion that Iran had possibly obtained uranium-enrichment technology from Pakistan, and that this technology appeared to be that of URENCO, the consortium from which Khan had stolen designs. The official reaction in Europe and the United States was "no comment." Hibbs was left to pursue his work alone in his black box.

Unbeknownst to him, the CIA had concluded that the Pakistan-Iran connection had cooled, in part because the centrifuges that Pakistan had sold were castoffs, prone to vibration and inefficient compared with more modern designs. As a result U.S. interest in Khan diminished, and to some extent the trail was allowed to go cold. Hindsight shows that this was a mistake: Khan remained as ambitious as ever, and like a good vendor, he offered improvements to his client. His relations with Iran were solid and all the better because they were out of sight. Throughout that decade, however, as Hibbs occasionally reported, U.S. suspicions remained strong that Iran was continuing to pursue a nuclear-weapons program, with the perhaps unwitting aid of Russia and China, both of which were eager to sell civil nuclear technology to Iran—as they are today.

In 2002 a glimmer of light illuminated Hibbs's black box: a confidential source at the IAEA alerted him to the news that Iran had proceeded so far with a centrifuge program that it was ready to open a production-scale enrichment plant. Hibbs asked if the IAEA had any information on the origin of the design.

His source said, "It's indigenous."

Hibbs didn't believe it. He said, "There's nothing indigenous about a centrifuge program in Iran. There's nothing remotely indigenous about it. It's stolen. Believe me, it's got to be stolen."

Hibbs studied the question, and with the shreds of information he painstakingly formed a picture in his mind. He asked himself: How powerful is this machine? How big is it? How much uranium could it conceivably enrich? How fast? And how long has it taken the Iranians to develop it? He went back to his notes of a decade before, and read all his old files, and finally concluded that it had to be a URENCO design, and was probably from A. Q. Khan. But Hibbs needed some sort of confirmation. With the evidence in hand he went to see another confidential source at a U.S. agency in Washington. They met in a coffee shop. Hibbs is generally diffident, but on this occasion he got right to the point. He said, "Does the U.S. government know where the technology came from?"

His source did not answer right away. Apparently he had not anticipated this question, and he needed time to decide how far to go. The choice was between "no comment" and telling the truth, because only a fool would lie to Mark Hibbs. The man paused for a long while. Finally he said, "Yeah."

Hibbs said, "Where did Iran get it from?"

Again the man paused. "Well, it's the same …" He stopped himself. Earlier that year the United States had leaked word that North Korea had received centrifuge designs and possibly prototypes from Pakistan in return for missile technology, in a state-to-state swap. The leak was directed not against Pakistan but against North Korea, which soon restarted its plutonium-reprocessing facilities and expelled IAEA inspectors. In any case, Hibbs's contact decided to go ahead. He said, "There's only one country that's exporting centrifuge technology."

"Do you mean Pakistan?"

"Yeah."

Hibbs said, "You realize, if I trust you on this and flesh this out and write it, there's going to be a shit storm, and basically it's going to be denied everywhere."

"Yeah."

Hibbs wrote it up, and in January of 2003 his editor put it on page one of Nuclear Fuel, under the drab Platts-style headline "Pakistan Believed Design Data Source For Centrifuges To Be Built By Iran." It was the most important work of Hibbs's career to date—a 2,164-word masterpiece that went to the center of Pakistan's activities and with unerring precision mapped the recent history of nuclear proliferation. The reaction, as Hibbs had predicted, amounted to a chorus of official denials—with various professors chiming in to explain why, for cultural or geopolitical reasons, Pakistan would never have helped the Iranians to arm. But Hibbs stood his ground.

Later he said to me, "There was no comment from the IAEA. I continued to interact with the sources of that story. Throughout 2003 they kept telling me, 'You're not only warm and hot but the IAEA is very angry that you are not letting them control the flow of information. They're onto Pakistan. They know that individuals in Pakistan were deeply implicated in this program. But they can't use the "P" word. No one will say "Pakistan." It's all being discreetly negotiated between the IAEA, the United States, and other countries …'" The problem for the United States was that Pakistan was again now a trusted ally, this time in the effort to destroy al-Qaeda.

I said, "So they wanted you to pipe down."

Hibbs said, "Anyway, we kept working on Pakistan, and more and more bits of the story got confirmed. I kept fingering Pakistan, fingering Pakistan, and pissing off the IAEA and the U.S. government, because at that time they were saying, 'We want to make a deal with these people. We want to make sure it doesn't get out of control.'"

I said, "The story or the activity?"

"The story. They wanted to control it."

Controlling a story once Hibbs starts into it is not an easy trick, particularly because of the dedicated non-proliferationists within the ranks of government, who refuse to submit to higher political agendas and are therefore sometimes willing to talk. The Bush administration did manage to engineer a partial shelter for Pervez Musharraf, allowing him on behalf of the United States to pursue his "war on terrorism," largely against his own people, along the border with Afghanistan. Nonetheless, across the months of 2003, as revelation led to revelation, it became obvious that A. Q. Khan's nuclear empire, which had long been penetrated yet neglected by the West, was at last starting to fall apart.

The trouble over North Korea served as an early warning to Khan—or it could have, had he been wiser and less enamored of himself. Cooperation with the Koreans seems to have dated back to 1992, when Pakistan, having acquired nuclear weapons, cast around for a missile capable of carrying them. Groups of Pakistani engineers and officials made several trips to North Korea to witness test flights of a promising medium-range missile called the NoDong. They later struck a deal. Over the course of the decade North Korea provided Pakistan with prototypes of missiles that were modified and produced at the Khan Research Laboratories, resulting in a successful Pakistani flight just before the tit-for-tat nuclear tests of India and Pakistan in 1998. In return for the missiles Pakistan provided the North Koreans with centrifuge prototypes—the same old URENCO design—and gave them uranium-enrichment and procurement advice. Western intelligence services found out. Behind closed doors in 2000 U.S. officials confronted the Musharraf regime with what they thought was irrefutable evidence (much of it photographic) of the centrifuge trade. The Pakistanis categorically denied that any such activity had taken place. They looked the Americans in the eye and lied, and they did not care that the Americans knew it. The transfers continued. The Americans persisted, some believing that bombs in the hands of Pyongyang would be more dangerous even than bombs in the hands of Baghdad or Tehran. Eventually Musharraf came up with a convenient answer: while admitting to no wrongdoing by Pakistan or himself, or to any consummated transfers of nuclear technology, he quietly pointed at Khan, essentially for being out of control.

His pointing took the form of a mock investigation. Khan was making frequent trips to the Gulf city-state of Dubai, where, like many other rich Pakistanis, he owned a home, and where increasingly he sought medical care for himself and his wife. Dubai had long served as an offshore trans-shipment hub for the Pakistani nuclear-procurement network, and it served just as well now as a center for the business of nuclear distribution. Khan's main collaborator there was a young Sri Lankan named Buhary Syed Abu Tahir—a wholesaler of consumer goods who had warehouses full of televisions and personal computers, and who had supplied air-conditioners to the Khan Research Laboratories before gradually getting involved in the smuggling of nuclear materials. Tahir seems to have been a morally neutral character, friendly to Khan and sympathetic to the aspirations of developing nuclear powers, but motivated primarily by the money to be made as a middleman. Khan did not begrudge him his profits: indeed, he had grown so fond of Tahir that he teased him about his love life and sometimes treated him like a son. But Khan himself remained the great moralist: not averse to personal gain, and delighted to wheel and deal in luxury in Dubai, but convinced that whatever he did he did for Pakistan.

Khan was therefore dumbfounded, upon returning from a short trip to Dubai in 2000, when Musharraf, having called him in for a conversation, told him that he had been under surveillance by Pakistani agents and that there were concerns about financial improprieties. Financial improprieties? In the world of Khan the word had lost all meaning. There was no question of going to prison, but in 2001, just days short of his sixty-fifth birthday, A. Q. Khan was gently relieved of his command, forced to retire with honors from his cherished laboratory, and "promoted" to the position of scientific adviser to Musharraf. This last was a particularly nice touch. There is evidence that the exchanges with North Korea continued for at least another year. When the Bush administration finally decided to go public with its concerns about the North Koreans' nuclear-weapons program, it delayed leaking the intelligence information until late October of 2002, after Congress had given its approval for the U.S. invasion of Iraq. The blundering that fall defies belief: while dragging the United States into a disastrous war in the pursuit of phantom weapons programs in Iraq, the U.S. government condoned the tangible actions of Pakistan—which, as any reader of Hibbs would have known, was delivering nuclear-weapons capabilities into the hands of America's most significant enemies, including regimes with overt connections to Islamist terrorists. Before the attacks on New York and Washington, Musharraf himself had accommodated Osama bin Laden, had supported the Taliban, and had used international jihadis against the Indians in Kashmir and beyond. But times had changed, and by October of 2002 Musharraf was Washington's friend, engaged in trying to suppress the Islamist idea by gunning it down. It was useful that with his move against Khan he had partially protected himself from revelations of Pakistan's trade in nuclear technology. Secretary of State Colin Powell spoke with Musharraf, and afterward, when asked on ABC television about Pakistan's assistance to North Korea, said, "President Musharraf gave me his assurance in that conversation, as he has previously, that Pakistan is not doing anything of that nature … The past is the past. I am more concerned about what is going on now. We have a new relationship with Pakistan."

The past was the past, but it bore a striking resemblance to the present. Khan had been removed from the laboratories, but as the U.S. government must have known full well, he continued to run Pakistan's nuclear networks and to pursue that business all over the world. Khan was vulnerable now, a man positioned to take a fall; but he had become so stupid about himself that apparently he did not believe it, and in any case he was busy. Along with filling the import orders for Pakistan's ongoing nuclear-weapons programs and arranging for exports to Iran and North Korea, he was now occupied with Libya as well. The Libyans had long desired nuclear arms, and like the Saudis, they may have helped to underwrite Pakistan's original uranium-enrichment efforts, in the vague hope that Pakistan would produce a "Muslim" bomb to be shared. That didn't work out. But by the late 1990s, with Pakistan expressing its willingness to make deals, the terms were clear: solidarity aside, it was cash that counted, and foreign governments could simply buy the components necessary to become self-sustaining nuclear powers. Libya decided to proceed. Emissaries from Tripoli met with Khan and Tahir in Istanbul, and later in Casablanca and Dubai, to hash out the details. Libya is a primitive Saharan society, only a half step beyond the traditions of nomadic life, and less capable technologically than any other nuclear aspirant to date. Khan must have said to the Libyans that this did not matter—and given the Europeans' earlier doubts about Pakistan's own competence, he may have believed it. In any case, he offered to equip Libya with a turnkey operation, including all the facilities necessary to enrich uranium and ultimately to build bombs. The asking price was $100 million, which was a bargain for Libya, considering the international muscle that a nuclear arsenal would provide.

When Libya agreed, Khan turned, as usual, to his European suppliers; with their assistance, by 2000 he was installing dozens of centrifuges in a pilot plant on the outskirts of Tripoli. That same year he arranged for the delivery (either through North Korea or directly from Pakistan) of gasified uranium that, if fed in closed loops through the centrifuges, would eventually be sufficient to fuel perhaps a single atomic bomb. The gasified uranium was kept in storage and apparently was never used. This was a pattern throughout the program in Libya, which proved unable to absorb all the equipment it had bought, and left much of it lying unassembled in shipping crates. Nonetheless, Khan pushed forward on multiple fronts. The pilot plant was of course just the start. What Libya ultimately required was a sustained and efficient production line involving thousands of linked centrifuges backed up by the ultra-precise computerized machine tools necessary to make replacement parts and an unending supply of bombs. The machine-tool installations were to be provided primarily by Khan's European suppliers, but with help from certain Turks known to him from the past. The centrifuges themselves were to be manufactured in pieces by various companies around the world, including at least one created solely for that purpose by Tahir and Khan.

In the first week of September 2001, six months after Khan's forced retirement in Pakistan and several days before the terrorist attacks on the United States, the CIA issued a report that Libya had revived its nuclear-weapons program, and mentioned guardedly that such an effort required substantial assistance from abroad. It is known now that British intelligence was tracking the Libyans carefully, and it can be presumed that the CIA had not forgotten about Khan. Three months later, in December, Tahir set up a shop southeast of Kuala Lumpur to manufacture certain narrow-tolerance centrifuge parts for the Libyan contract. The shop was created as a subsidiary of a company owned in large part by the Malaysian prime minister's son, a youngish man named Kamaluddin Abdullah. A spokesman later said that the company's managers had not understood the nature of the parts. A Malaysian police investigation duly cleared Kamaluddin of wrongdoing in 2004, with the acquiescence of the United States. Tahir had married the daughter of a Malaysian diplomat, at a society wedding in 1998 attended by Khan, and she, too, was involved in the venture—though also innocently, she later claimed. The company was called Scomi Precision Engineering. To oversee the production Tahir hired a Swiss engineer named Urs Tinner—the son of a man alleged to be a longtime supplier to Khan, Friedrich Tinner. It has been reported in Switzerland that Urs Tinner was turned by the CIA, and that the parts he was producing for Tahir were intentionally faulty, but this runs counter to the official American assertions of U.S. ignorance and uninvolvement, and it has been impossible to verify. Either way, in December of 2002 centrifuge components began to arrive in Libya in large numbers.

Khan kept flying around, making deals, and enjoying life. Among other, more serious outings, he had taken to visiting Timbuktu, in Mali, where he would arrive in chartered airplanes with groups of his proliferationist friends, and where he funded the construction of a small hotel named the Hendrina Khan, after his wife. Apparently nothing nefarious was going on. In fact, it was such grand fun that one member of his entourage wrote a book about it, titled Timbuktu, which was full of praise for the great A. Q. and offered such helpful advice as not to bother visiting the Bamako zoo. But in 2003 such freedoms were coming to an end. With the IAEA having revealed the existence of secret centrifuges in Iran, and with Mark Hibbs standing on the sidelines repeatedly fingering Pakistan, the political pressures on Musharraf kept mounting. For Khan the collapse came in the second half of 2003, as the result of a complex series of events. The problems started in June, when IAEA inspectors found traces of highly enriched uranium on the surface of ancillary equipment in Iran. Iranian officials, who had been denying U.S. claims that Iran was pursuing nuclear weapons, were put in the awkward position of having to explain away the evidence; they admitted that the equipment was secondhand, and said that it had been imported from another country. In August IAEA inspectors found similar traces at another site, and the Iranians went further toward blaming Pakistan, without quite doing it by name.

Meanwhile, Libya's leader, Muammar Qaddafi, who had decided to normalize relations with the West, was involved in secret talks with Great Britain and the United States about dismantling his programs for weapons of mass destruction. In October of 2003, apparently, the British and the Americans decided to give him a push: they intercepted a ship called the BBC China, which was known to have left Dubai and passed through the Suez Canal carrying nuclear-weapons equipment destined for Libya. The intercept itself was a low-key affair: it amounted to a phone call to the ship's German owners requesting that they divert their vessel to an Italian port for an inspection. The request was backed up by the presence of a U.S. warship, but perhaps unnecessarily, since the ship's owners were innocent businessmen with nothing to hide. In Italy the inspection turned up five containers holding thousands of centrifuge parts manufactured in Malaysia by Scomi Precision Engineering (and packed in wooden crates boldly bearing the company's name), along with other components manufactured in Turkey and paperwork showing that the trans-shipment through Dubai had been arranged by the enterprising Tahir. To the intelligence agents involved none of this would have come as a surprise.

Nor would Khan have been surprised that Pakistan's operations had been penetrated. He had been working under that assumption for decades without slowing down, and the loss of five containers, or even of the Scomi facility in Malaysia, would normally not have fazed him at all. But in a larger sense now he was losing control. In Libya the intercept of the BBC China was seen as an inconvenient event because it amounted to being caught red-handed. It is necessary to discount the Bush administration's subsequent claim that Qaddafi was chastened by the invasion of Iraq, because by late 2003 the lack of Iraqi weapons of mass destruction was already becoming a major public embarrassment, and the U.S. military in Iraq was clearly bogged down. More likely, Qaddafi, far from being impressed by Washington's resolve, was invigorated by the depths of its blunder. Nonetheless, he had decided to do business with the world, particularly with the nearby moralists of Europe, and soon after the BBC China seizure he gave up on his nuclear-weapons program—at least temporarily.

In December of 2003 the Libyans accepted the arrival of IAEA inspectors and others and began to answer questions about Libya's arrangement with Pakistan. They were more forthcoming than the Iranians had been. They provided dates, named names, opened their facilities, and eventually even allowed the equipment to be flown away, under IAEA control, to warehouses in the United States. Their centrifuges were mostly unassembled and incomplete, but they were of the same proven URENCO design as those in Iran—and at least similar to those being built in North Korea, and to those perhaps being considered in Egypt, Syria, Saudi Arabia, and countries beyond. In December the news got out, with mention of the various cross-connections, and was published in the major American papers. How much worse could it get? In January of 2004, having certainly made photocopies first, the Libyans handed over bags from an Islamabad laundry that contained the plans for a Chinese-designed implosion bomb just like Pakistan's. On the margins were handwritten notes in English disparaging the hated Munir Ahmed Khan. Clearly these plans had come from the Khan Research Laboratories, and possibly from A. Q. Khan himself.

Musharraf ducked into the same defensive position he had used before. He launched an official investigation and began hauling in Khan's staff for questioning. The conclusion was known from the start: after nearly six years of warnings, beginning with the snubs Khan had received on the day of the test explosions in 1998 and continuing through his dismissal from the laboratories in 2001, it was finally time for Khan to disappear for good. Before checking out, however, he had one last service to provide to the regime, and if he did it right, technically speaking his life would be spared. On February 4, 2004, after days of persuasion in Islamabad, he appeared on television and absolved everyone but himself of blame. Speaking in English, a language that relatively few Pakistanis understand, he said,

"My dear ladies and gentlemen, Assalam-ul Alaikum, it is with the deepest sense of sorrow, anguish, and regret that I have chosen to appear before you in order to atone for some of the anguish and pain that have been suffered with the people of Pakistan on account of the extremely unfortunate events of the last two months. I am aware of the vital criticality of Pakistan's nuclear program to our national security and the national pride and emotion which it generates in your heart. I am also conscious that any untoward event, incident, or threat to this national security draws the greatest concern in the nation's psyche. It is in this context that the recent international events and their fallout on Pakistan have traumatized the nation.

"I have much to answer for. The recent investigation was ordered with the government of Pakistan consequent to the disturbing disclosures and evidence by some countries to international agencies relating to alleged proliferation activities by certain Pakistanis and foreigners over the last two decades.

"The investigations have established that many of the reported activities did occur, and these were inevitably initiated at my behest.

"In my interviews with the concerned government officials I was confronted with the evidence and findings, and I have voluntarily admitted that much of it is true and accurate.

"My dear brothers and sisters, I have chosen to appear before you to offer my deepest regrets and unqualified apologies to a traumatized nation. I am aware of the high esteem, love, and affection in which you have held me for my services to national security, and I am grateful for all the awards and honor that have been bestowed upon me. However, it pains me to realize in retrospect that my entire lifetime achievements of providing foolproof national security to my nation could have been placed in serious jeopardy on account of my activities which were based in good faith but on errors of judgment related to unauthorized proliferation activities.

"I wish to place on record that those of my subordinates who have accepted their role in the affair were acting in good faith like me on my instructions.

"I also wish to clarify that there was never ever any kind of authorization for these activities by the government.

"I take full responsibility for my actions, and seek your pardon.

"I give an assurance, my dear brothers and sisters, such activities will never take place in the future.

"I also appeal to all citizens of Pakistan in the supreme national interest to refrain from any further speculations and not to politicize this extremely sensitive issue of national security.

"May Allah keep Pakistan safe and secure. Long Live Pakistan!"

Khan then disappeared into his house, perhaps never to reappear. The official Pakistani investigation continued, occasionally producing the news that Khan was to blame. Several of Khan's staff members were placed under house arrest as well, one of whom has not yet been released. Outside Pakistan there was a bit more progress. Tahir was arrested in Malaysia but was not made available to representatives of the IAEA for the first nine months. The Malaysians eventually cleared themselves of wrongdoing. Urs Tinner was arrested in Germany and extradited to Switzerland, where he remains jailed while under investigation. Germany is trying to prosecute one of Khan's alleged German suppliers, a man named Gotthard Lerch, and Holland is trying the same against another of Khan's longtime associates—but both cases are in trouble. South Africa, which turns out to have been an important supplier to the Libyans, is trying halfheartedly to go after some of its own citizens. Meanwhile, Qaddafi has been mumbling about having cut a bad deal with the West, and both Iran and North Korea are rapidly arming, despite various assertions and diplomatic maneuvers that raise the possibility that they might stop. As for the United States, it keeps repeating its exquisite claim of ignorance: that for all those years it knew practically nothing about the activities of the dangerous A. Q. Khan.

But with or without the CIA, readers of Mark Hibbs have long known what was going on. He is a disciplined reporter who sticks closely to the news, but people with a sense for his beat—this world of nuclear secrets—draw as much from reading between his lines. Last September, for instance, under the headline "Pakistan Says Its Role in Probing Khan's Proliferation Is Finished," he led with two apparently simple sentences (one a lovely quote) that are memorable as much for the questions they intentionally left unanswered as for Hibbs's statement of the facts. Hibbs wrote,

The Pakistan government said Aug. 29 that its interrogation of [A. Q.] Khan, former head of Khan Research Laboratories (KRL), about his involvement in proliferating Pakistan's nuclear technology abroad has been completed. "As far as we are concerned, I understand that whatever information that was there has been obtained and has been shared with the relevant countries, and [that] the relevant countries are satisfied with the information," Foreign Ministry spokesman Mohammad Naeem Khan told a weekly press briefing.

The investigation was over? All available information had been obtained? It had been shared? With relevant countries? Which countries were they? Who chose them, anyway? And what about Khan? Since Musharraf had already pardoned him, and he had now confessed everything he knew, what was the reason for his continuing house arrest and his ongoing isolation from Western intelligence and the press? Hibbs did not address these questions for a reason quite obvious to his readers: Pakistan's investigation was a cover-up and a sham—moreover, of a sort possible only in a morally bankrupt and corrupt nation where cowardly and illegitimate rulers, propped up by huge infusions of American dollars and dependent on their soldiers' guns, suppress genuine inquiries because they themselves would be implicated and, in the embarrassment that followed, would be cut off from foreign aid and driven from power by their own people, who almost universally now detest them. The problem for the United States, and conceivably for Europe, is that those doing the driving would most likely be Islamists, who are growing in strength and numbers even as they are being hunted down. The Iranians will probably beat them to it, but if the Islamists in Pakistan took power tomorrow, they would be the first Islamists with nuclear bombs. Hibbs trusted his readers to know all this and more when he wrote about the sham investigation's end. To me in Bonn he simply chuckled and said, "Ah, Pakistan."

But strictly in terms of nuclear proliferation there is the question of what good a genuine investigation would do anyway—as Hibbs understands in full and painful detail. There would be the immediate frustrations that Khan's network was spread around, that it operated in the gray areas of national laws, and that prosecutions would be undermined by political sympathies even if unambiguously illegal acts could be proved. More difficult still, however, is the very nature of such a network, which is not a rigid structure that can be shattered and shut down but, rather, something more like the Internet—a vast and informal web of infinitely flexible connections, capable of reshaping itself with ease, and able therefore to sustain enormous damage without suffering significant loss of efficiency. A. Q. Khan did not create his nuclear-weapons network so much as discover it as a pre-existing condition of the modern world.

Mubashir Hasan, the pacifist and former finance minister in Lahore, told me he worries that Pakistan, like the United States, is the sort of country that would actually use its atomic bombs. He said he had once asked Pakistan's leaders when they thought such use would be justified.

One said, "When we are threatened enough."

"But when will you be threatened enough? If India takes Lahore?"

"We don't know."

"But if you throw a bomb, and India throws two bombs back in return, what then?"

"So what?" the man said. "Then we die."

Hasan was more surprised by this logic than I am. Once a country has a bomb, it must be determined to use it.

In Islamabad I met a smart man close to the military who recognized this risk and the wild-card problem, which is that Pakistan's command-and-control leaves much to be desired. He said, "The best way to fight proliferation is to pursue global disarmament. Fine, great, sure—if you expect that to happen. But you cannot have a world order in which you have five or eight nuclear-weapons states on the one hand, and the rest of the international community on the other. There are many places like Pakistan, poor countries that have legitimate security concerns—every bit as legitimate as yours. And yet you ask them to address those concerns without nuclear weapons—while you have nuclear weapons and you have everything else? It is not a question of what is fair, or right or wrong. It is simply not going to work."

The man was right, of course. A. Q. Khan proved it once, and then three times over again. There will be other Khans in the future. Shifted to America, and stripped of its postcolonial indignation, the point becomes an argument not for standing down from the diplomacy of nuclear non-proliferation but for stoically finding a way to deal in parallel with the reality of a world in which many countries have acquired atomic bombs, and some may use them.