How to Get a Nuclear Bomb

Three days later, the city of Nagasaki was hit by an even more powerful device—a sophisticated implosion-type bomb built around a softball-sized sphere of plutonium, which crossed the mass-to-surface-area threshold of criticality when it was symmetrically compressed by carefully arrayed explosives. A twenty-two-kiloton blast resulted. Though much of the city was shielded by hills, about 70,000 people died. Quibblers claim that a demonstration offshore, or even above Tokyo harbor, might have induced the Japanese to surrender—and if not, there was another bomb at the ready. But the idea was to terrorize a nation to the maximum extent, and there is nothing like nuking civilians to achieve that effect.

The physicists who had developed these devices understood the potential for miniaturization and a simultaneous escalation in warhead yields, past the twenty-two kilotons of Nagasaki and indeed past a thousand kilotons, into the multimegaton range—the realm, when multiplied, of global suicide. Moreover, they realized that the science involved, however mysterious it seemed to outsiders, had already devolved into mere problems of engineering, the knowledge of which could not be contained. Within a few years humanity would face an objective risk of annihilation—a reality that compelled those who understood it best to go public with the facts. In the months following Japan’s surrender, a group of the men responsible for building the bomb—including Albert Einstein, Robert Oppenheimer, Neils Bohr, and Leo Szilard—created the Federation of American (Atomic) Scientists (FAS), to disseminate nuclear-weapons information. Washington at the time harbored the illusion that America possessed a great secret, and could keep the bomb for itself to drop or not on others. The founders of FAS disagreed. The current vice president of its Strategic Security Proj­ect, an affable scientist named Ivan Oelrich, recently said to me, “The biggest secret about the atomic bomb was whether it would work or not.” But after the United States exploded one, there was no longer any question in the minds of other countries. “They knew that if they did X, Y, and Z, they would have success. So in 1945 the scientists who founded this organization said, ‘Look, there is no secret. Any physicist anywhere can figure out what we did and reproduce it. There is no secret, and there is also no defense.’”

Some of the solutions they proposed may seem quaint. Albert Einstein, for instance, called for the creation of an enlightened world government, complete with the integration of formerly hostile military staffs and the voluntary dismantling of sovereign states. But the founders of FAS were not naive so much as desperate and brave. They said, in essence, If you knew what we know about these devices, you would agree that at any price, the practice of war must end. It was a rare call for radical change by men at the top of their game. But history shows that the future is impossible to predict. There was no exception here. After sixty years there has been no apocalypse, and a nuclear peace has so far endured for all the wrong reasons—an unenlightened standoff between the nuclear powers, each of them restrained from shooting first not by moral qualms, but by the certainty of a devastating response. Moreover, the very lack of defense that worried the scientists in 1945 turned out to be the defense, though treacherous because it required tit-for-tat escalations. But these are latter-day corrections to the concerns of enormously competent men, and their message is equally valid today. Detailed knowledge of nuclear-bomb making has escaped into the public domain, and the use of even a single fission device could pose an existential threat to the West.

Last winter in Moscow I spoke to an experienced Cold War hand, who had skated through the collapse of the Soviet Union, and now occupied a high position in the nuclear bureaucracy of the increasingly assertive Russian state. In his corduroy suit, with his bushy eyebrows and heavy, sometimes glowering face, he looked like an apparatchik from central casting—and he acted like one, too. It was refreshing. He kept poking his finger at me, and accusing Americans of losing perspective over a nuclear Iran. He wanted to do nuclear business with Iran, in electric-power generation. He wanted to do nuclear business with all sorts of countries. He claimed that with one Russian submarine reactor (fueled by high-octane uranium) he could light up all sorts of cities. He meant with electricity. He proposed a scheme to mount such reactors on barges, to be pushed to places like Indonesia and then pulled away whenever the natives ran amok. This way, he said, he could keep his uranium from fueling native bombs. He did not deny the incentives for lesser nations to acquire nuclear devices, but he thought he could handle them, or perhaps he didn’t care. He said, “The Nuclear Non-Proliferation Treaty was the child of Russia and the United States. And this child was raised to fight against other countries, to resist the threat of proliferation. We’re talking about the 1970s. No one thought that proliferation could come from Arab countries, from Africa, from South America. The treaty was aimed at Western Germany, at Japan. It was aimed at dissuading the developed countries from acquiring nuclear weapons—and it worked because they accepted the U.S. and Soviet nuclear umbrellas.”

He was bullying history, but only by a bit. The Nuclear Non-Proliferation Treaty, or NPT, was an effort to preserve the exclusivity of a weapons club whose membership consisted originally of only five: Britain, China, France, the Soviet Union, and the United States. To other countries the treaty promised assistance with nuclear research and power generation in return for commitments to abstain from nuclear arms. It cannot be said to have “worked,” as my Russian friend claimed, but it did help to slow things down. More important, and completely independent of the NPT, were the Cold War alliances that, by offering retaliatory guarantees, eliminated the need for independent nuclear defense capabilities in those nations willing (or forced) to choose sides. But neither the Cold War alliances nor the NPT could counter the natural appeal of these devices—their fast-track, nation-equalizing, don’t-tread-on-me, flat-out-awesome destructive power.

In 1946 Robert Oppenheimer sketched the problem clearly. In an essay titled “The New Weapon,” he wrote: “Atomic explosives vastly increase the power of destruction per dollar spent, per man-hour invested; they profoundly upset the precarious balance between the effort necessary to destroy and the extent of the destruction.” Elaborating, he wrote,

From his perch in Moscow my Russian friend had observed the effect of Oppenheimer’s truths. Continuing with his story of the NPT, he said, “Even as the U.S. and Russia offered our nuclear umbrellas, everyone understood that the weapons could never be used, because of retaliation. For us they were not wealth—they were a burden. At the same time, nuclear technology was becoming even cheaper, more efficient, and it became available to many countries. It became a useful tool especially for weak countries to satisfy their ambitions without much expense. There are no technical barriers, and no barriers to the flow of information, that can prevent it. Once a country has made the decision to become a nuclear-weapons power, it will become one regardless of any guarantees. You needn’t be rich. You needn’t be technically developed. You can be Pakistan, Libya, North Korea, Iran. You can be …” He searched for a country even more absurd in his estimation. He said, “You can be Hungary.” Then he said, “At some point this change occurred. The great powers were stuck with arsenals they could not use. And nuclear weapons became the weapons of the poor.”

It was a simplified view, but not entirely wrong. Certainly the argument can be made that only underdeveloped nations can now afford to use these weapons, not merely because the lives of their citizens seem to be expendable, but also because of the limitations of their nuclear arsenals, which mean that their warheads will incinerate just a few enemy cities, more or less locally, and will not likely frighten Russia and the United States into swapping strikes. The fate of the world is not at stake. Pakistan and India came close to a regional nuclear exchange in 2002, with little risk of igniting a global conflict. The core of that story, however, is that each antagonist had its own cities to protect—particularly its own capital—and each therefore had good reasons for backing down. This was a demonstration rather than a proof, but of special interest because it involved a country as backward as Pakistan. It seems to indicate as a general rule that even a stunted state is deterred by the threat of retaliation, because so long as its leaders have a government, an infrastructure, and indeed a delineated nation (not to mention their individual lives of luxury), they provide rich targets to be smashed and burned in answer to any first strike. At this point it appears that simple calculations of self-preservation should keep fingers off the triggers even in Pyongyang and Tehran. So we should be safe, relatively—but perhaps we are not.

The danger comes from a direction unforeseen in 1945, that this technology might now pass into the hands of the new stateless guerrillas, the jihadists, who offer none of the targets that have underlain our nuclear peace—no permanent infrastructure, no capital city, no country called home. The nuclear threat posed by the jihadists first surfaced in the chaos of post-Soviet Russia in the 1990s, and took full form after the fall of the World Trade Center. With so little to fear of nuclear retaliation, and having already panicked the United States into historic policy blunders, these are the rare people in a position to act.

All this should give you pause long enough to take bearings. You would do well to distinguish between your needs and those of conventional proliferators. Fledgling nuclear-weapons states have little use for just one or two bombs. To assume a convincing posture of counterstrike and deterrence, or simply to exhibit nuclear muscle, they require a significant arsenal that can be renewed and improved and grown across time. This in turn requires that they build large-scale industrial facilities to produce warhead fuel, which cannot be purchased on the international black market in sufficient quantity to sustain a nuclear-weapons production line. Manufacturing high-quality fuel is the most difficult part of any nuclear program; the NPT is meant to interfere primarily at this stage. The construction of everything else is relatively easy.

You could hardly expect to set up the facilities to manufacture nuclear fuel. (Nor could you expect any state, whether Pakistan or North Korea, to risk helping you here, either.) But you would have no need to do so. There is plenty of weapons-grade fissile material in the world today, and more is being produced all the time. Surely you could steal or buy the quantity necessary for a single garage-made device.

You would now need to decide what kind of fuel to pursue. There are really only two choices—plutonium or highly enriched uranium. Plutonium is a man-made element produced by uranium reactors. There are several forms of it, including one purpose-made for bombs. Armies favor plutonium because it can be made to go critical in very small quantities, thereby lending itself to the miniaturization of weapons. Miniaturization has obvious attractions, but it requires a level of engineering sophistication that lies beyond the capabilities of a small terrorist team. And miniaturization is not that important for your purposes. You can operate well enough with a car-sized device locked into a shipping container or loaded into a private airplane behind a couple of dedicated pilots. Furthermore, plutonium has some negatives for an operation like yours. It is not suited for use in a basic cannon-type bomb, and demands instead the explosive symmetry of a Nagasaki-style implosion device. Building an implosion device would introduce complexities you would be better off avoiding, particularly without a place and the time to test the design. And plutonium is difficult to handle—sufficiently radioactive to require shielding, awkward to transport without setting off radiation detectors, and extremely dangerous even in minute quantities if it is breathed in, swallowed, or absorbed through a cut or open wound. There are plenty of people who would willingly die for the chance to nuke the United States, but within the limited pool of technicians who might join your effort, it would be impractical to expect so much. Plutonium might work as the pollutant spread by a dirty bomb, but for your project, plutonium is out.

The alternative is highly enriched uranium, or HEU, the variant of natural uranium that has been refined to contain artificially dense concentrations of the fissionable isotope U‑235. Operationally it is wonderful material—the perfect fuel for a garage-made bomb. During processing, uranium takes the form of an invisible gas, a liquid, a powder, and finally a dull grey metal. It has approximately the toxicity of lead, and would sicken shop workers who happened to swallow traces of it or breathe in its dust, but otherwise it is not immediately dangerous, and indeed is so mildly radioactive that it can be picked up with bare hands and, when lightly shielded, taken past many radiation monitors without setting off alarms. As one physicist in Washington suggested to me, in small masses HEU is so benign that you could sleep with it under your pillow. He warned me, however, that you could not just pile it up in your bed, or anywhere else, because the atoms of U‑235 occasionally split apart spontaneously, and in doing so fire off neutrons, which within a sufficient mass of material could split enough other atoms to cause a chain reaction. Such a reaction would not amount to a military-style nuclear explosion, but it could certainly take out a few city blocks.

I asked the physicist if he wasn’t concerned about giving information to terrorists. He summoned his patience visibly and said, to paraphrase it, This is Boy Scout Nuclear Merit Badge stuff. We continued our discussion. He said that the critical mass of uranium is inversely proportional to the level of enrichment. At the low end of HEU, which is considered to be 20 percent enrichment, nearly a ton would have to be combined before a stockpile could spontaneously ignite. At the high end, which is the “weapons grade” enrichment of 90 percent or greater, about 100 pounds could do the trick. I mentioned that at whatever level of enrichment, the HEU that a terrorist could acquire would by definition be made of units each consisting of less than a critical mass. I asked the physicist to imagine that a terrorist had acquired two bricks of weapons-grade HEU, each weighing fifty pounds: how far apart would he have to keep them? He said that a yard would be enough. I had arrived in Washington from remote mountains along the Turkish border with Iran, where every night hundreds of pack horses are led across the line by Kurdish smugglers, carrying cheap fuel for Turkish cars and opium for the European heroin market. This is the Silk Road revived, and it is one of the prime potential routes for the movement of stolen uranium. With this in mind, I told the physicist I assumed from his measure that the two bricks could be slung on either side of a saddle.

He said, “One on each side should be all right …” He hesitated. “But what is the moderating effect of a horse?”

I had no idea. He said, “Look, if someone’s smart enough to have snuck in and gotten ahold of these two ingots of metal, he’ll be smart enough to negotiate for a second horse.”

It is difficult to get a clear picture here. Turkey is the world’s grand bazaar, and given its geographic position overlooking the Middle East, it is hardly surprising that people have gone looking to sell nuclear goods there. A University of Salzburg database (formerly run by Stanford) that purports to track smuggling activity globally since 1993 lists at least 20 incidents in the vicinity of Istanbul alone. But by including intercepts of all sorts of nuclear materials, that database (like most other treatments in this business) overstates the market’s ability to answer a bomb-maker’s needs. In fact, the marketplace—whether in Istanbul or anywhere else—seems never to have produced what you would require. The closest instance I found in the record dates from 1998, when agents from the Russian Federal Security Bureau (the former KGB, now called the FSB) arrested nuclear workers who were plotting to steal forty pounds of HEU. The enrichment level was never made public—an omission hinting that the uranium may well have been weapons grade. But even then it was less than half of what you would need.

Of course, we don’t know what we don’t know, as we are repetitively reminded. However, the other intercepts have been minor affairs of people caught filching or hawking scraps, often of material that doesn’t pass even the 20 percent mark. For a serious bomb-builder, the reports of “loose nukes” would come to sound like so much background chatter. You will have to obtain the fissile material at the source. If you look through the literature, you’ll soon realize that one of the challenges is the very extent of choices. It turns out that the world is rich with fresh, safe, user-friendly HEU—a global accumulation of over a thousand metric tons (outside of our collective 30,000 nuclear warheads) that is dispersed among hundreds of sites, and separated into nicely transportable, necessarily subcritical packages. The question is how to pick some up. Here again the literature can provide guidance. Although almost all of the HEU is in some manner guarded, there are many countries where it might nonetheless be acquired, and probably nowhere better than Russia.

When post-Soviet Russia came into being, in 1991, it inherited a sprawling state industry that had provided a full range of nuclear services, including medical science, power generation, and ship propulsion—as well as the world’s largest nuclear-weapons arsenal and, almost coincidentally, the world’s largest inventory of surplus plutonium and HEU, maybe 600 metric tons. The physical plant consisted of several dozen research, production, and storage facilities, and especially of ten fenced and guarded nuclear cities, which housed nearly a million people, yet were nominally so secret that they did not appear on maps. But within a few years the industry was obsolete, unable to adapt to the new Russian economy, and in steep decline. The buildings were in disrepair, and morale was low because people were not being paid enough or on time. Worse, the nuclear stockpiles were apparently being neglected. There were stories of guards abandoning their posts to forage for food, and of sheds containing world-ending supplies of HEU protected by padlock only. The question now, some fifteen years later, is why terrorists or criminals apparently did not then take advantage. One explanation is that they were ignorant, incompetent, and distracted. Another is that the defenses were not as weak as they appeared.

In any case, the U.S. government reacted rapidly to a perception of chaos and opportunity in post-Soviet nuclear affairs, and in 1993 launched an ambitious complex of “cooperative” programs with all the former Soviet states to lessen the chance that nuclear weapons might end up in the wrong hands. The programs have blossomed into the largest part of American aid to Russia, amounting so far to several billion dollars. There have been two main efforts. The first, managed by the U.S. Department of Defense, has concentrated on getting Russia to consolidate, secure, and to some degree destroy nuclear warheads, as well as some of the missiles, aircraft, and submarines that carry them. The same programs have facilitated the spectacular denuclear- ization of the former Soviet Union’s outlying nations. But these were the maneuverings of conventional actors following the familiar logic of strike and counterstrike. By comparison the perceived vulnerability of fissile materials in the former Soviet Union has presented the United States with a wilderness of unknowns. Securing these stocks is the second main cooperative effort. The job has been given primarily to the U.S. Department of Energy, and specifically to officials there with experience managing the American nuclear-weapons infrastructure—a group now formed into a semiautonomous agency known as the National Nuclear Safety Administration, or NNSA.

The NNSA sends managers from Washington and technicians from U.S. national laboratories to supervise the local officials. Its frontline agents tend to be hands-on technical people, impatient to pour concrete and get the jobs done. Their impulses lie primarily in an area known as Material Protection, Control and Accounting, or within the NNSA, lovingly, as MPC&A. In brief, this means locking the fissile materials down. Over the years the NNSA has identified approximately 220 buildings at fifty-two sites in Russia that are in dire need of treatment. That’s a lot, and as a result, actually there are two treatments. The first is a stop-gap measure called a “rapid upgrade.” It involves bricking up the warehouse windows, installing stronger locks, fixing the fences, maybe hiring some guards. The second is a long-term fix called a “comprehensive upgrade.” It often involves the full range of Americanized defenses, including crash-resistant fences, bombproof buildings, remote cameras and electronic sensors, bar-coded inventory scanners, advanced locks, well-armed and well-motivated guards, and all sorts of double- and triple-safe procedures.

Such complex constructs require constant care. Agents of the NNSA see evidence already that the Russians are not committed to maintenance and operations, and some complained privately to me that as soon as U.S. funding ends, their elegant MPC&A systems will slip into disrepair. Nonetheless, the NNSA is supposed to wrap up the program, squeeze off the funding, and turn over all the required security upgrades to the Russians by the end of 2008. Knowledgeable observers are skeptical that this schedule can be met. They say that about a third of the identified buildings have yet to be given even rapid upgrades, that these contain about half of Russia’s entire fissile-material stock, and that they sit at some of the most sensitive sites in the country—areas within the closed cities, where new warheads are assembled, and where the NNSA representatives are increasingly seen as meddlers and spies.

The agency’s administrator, a portly former submarine captain and strategic-weapons negotiator named Linton F. Brooks, put it plainly to me. He said, “We are about giving governments the tools to work in those areas where governments have control.” Fine. Brooks is an impressive man, and all the more so for his lack of theatrics. He did not pretend to be winning a war, or even to be fighting one, but more simply to be driving up the costs and complications for would-be nuclear bombers. The NNSA’s job is to shift the odds, and increase the likelihood that its opponents will fail. It cannot dictate to the Russians. It cannot operate with anything like the flexibility of the guerrillas. But of all the U.S. agencies recently engaged to suppress the nuclear threat, it does seem one of the few that may have contributed something real, even if it has to be called MPC&A.

The CIA appears to have added little to the effort. Presumably its people tag along on some missions, but they seem largely just to pursue conventional governmental information—estimating military capacities, or mapping the Russian bureaucracy in order to predict Russian reactions. I spoke to a former high U.S. official who said that during a decade spent securing stockpiles in Russia and receiving countless intelligence briefings, he had never once found information that would have helped him to calibrate the risks specific to a site. Who lives in the neighborhood? Who lives just outside? Who has just arrived? How the hell do any of them survive? What is meant here, tangibly, by organized crime? Is there other crime that counts? Who drives the flashy cars? What are the emotions of the people who do not? How much is known in the street about shipments to and from and between the plants? How much is known about what goes on inside? What do people think about the new wall and fences? What do they feel when they see an American flag? Now start all over again, and tell us about the nuclear technicians, the FSB agents, and the ordinary guards. Tell us about their lovers, their holidays, the furniture they dream of buying at IKEA. Tell us about their inner lives.

The official sighed with resignation. I suppose he felt what many believe, that if the United States is hit someday with an atomic bomb, it will in part be because of Washington’s discomfort with informal realms—because of a blindness to the street, amply demonstrated in recent times, which will have allowed some bomb-builder the maneuvering room to get the job done.

Describing those years in heavily accented English, a Russian plant manager said to me, “All nuclear material was secret. State secret! Anyone stealing nuclear material in Soviet Union was committing state crime. He became state criminal! So there was fear. Real fear. If something got lost somewhere—maybe a piece of paper, or materials, or there was mismatch in balance of plutonium—a person understood that he would be exiled forever.” Hesitating over the right words, he said, “But then when this … change took place, of course people felt more … freedom, I would say.”

He was using the word freedom for the American ear, and making a familiar argument for the extension of foreign aid. The closed cities and nuclear facilities around Ekaterinburg have received a large portion of the U.S. dollars spent on Russian security upgrades, yet they are the subject of continued wariness by the NNSA, and are of still greater concern to independent critics in the United States, who insist that they remain acutely vulnerable to terrorist theft. Take, for example, the closed city of Ozersk, a community of 85,000 people, whose existence was so hushed under the Soviets that it was not allowed a proper name, and was referred to only by its post office box numbers—first No. 40, then No. 65—in Chelyabinsk, an open city forty-four miles away. The nomenclature remains confused. Ozersk is often and mistakenly called Mayak, for its nuclear production area—an industrial complex within the city’s perimeter, a few miles from the residential center. The Mayak Production Association employed 14,500 people as of 2001, and since 1945 has been in the business primarily of processing HEU, plutonium, and tritium for nuclear warheads. Recently it has also thrown the business into reverse, as one of two sites in Russia where fissile material from existing warheads is extracted before shipment to another closed city for blending down. Many tons of the highest-quality weapons-grade HEU and plutonium are present at the site.

And nyet, this is not state secret.

Divulging it is not state crime.

Start with the all-American fact that Mayak is the location of the recently completed “Plutonium Palace,” a heavily fortified $350 million warehouse, which was paid for by the U.S. Congress, and has therefore been heavily publicized. The facility was designed to hold as much as 40 percent of the Russian military’s excess fissile material. For now the storage vaults remain empty because of technical and bureaucratic disputes, as well as, perhaps, a sense, in a Russia on the rise, of not wanting to place nuclear assets quite so far out of reach just yet. Nonetheless, there is little doubt that such a beautiful facility will eventually be used—and that the world will be better off when it is.

For those trying to counter the threat of nuclear terrorism, however, the possibility exists that the Palace is just another Maginot fort, a strong point that can be neatly ignored by the new strategists of war. The fort will neither reduce nor protect the large quantities of weapons-grade materials elsewhere in Ozersk. NNSA technicians have struggled to fill the gaps, installing some cameras and radiation monitors, and strengthening some floors and walls—but only in a few of the many buildings where they believe such work is needed, and only intermittently, as the Russians have allowed.

By perverse logic, therefore, Ozersk comes highly recommended to anyone in pursuit of a bomb. The facilities are tucked away two hours south of Ekaterinburg, down unmarked back roads, on a forested plateau of lakes and small rivers that would seem idyllic were it not for a number of decimated industrial towns and villages. Ozersk’s perimeter is large. It encompasses more than 50 square miles, and includes the city itself, the Mayak facilities, a network of paved and dirt roads, an internal bus service, one large internal checkpoint, multiple railroad lines, burial sites for radioactive waste, multiple radioactive lakes, some radioactive swamps, and a lot of radioactive forest. The main gate is on the north side. It has pens where Interior Ministry troops check vehicles going in or out, and verify people’s papers. The perimeter is marked by twin parallel chain-link barbed-wire fences, separated by a chemically defoliated strip that is apparently not mined, or raked, or checked for tracks even in the snow. The fences are in reasonable repair, but in many places have no road beside them, and they show no signs of being patrolled. If they are monitored remotely, it is safe to assume that they are not monitored well. On the south side of the site, they run through miles of empty forest. You could cross them almost anywhere with little immediate risk, though odds are you would eventually be caught, and the consequences would be severe. And to what advantage? You wouldn’t learn much by walking the streets. You might be able to send agents who could establish residency, but that would require too much time, and offer a poor chance of success. Better to back off and think things through, particularly since the essentials can be known from the outside.

Ozersk is in the nuclear business and nothing else. Its weapons-usable HEU is kept in Mayak as oxidized powder, flat metal pucks, elongated ingots, and finely machined warhead hemispheres. Each form is stored in a different type of steel container. The containers are light, because the shielding they contain is minimal. They are sealed but not locked. They sit in vaults or ordinary storage rooms. In addition to the standard containers, there are larger, brightly colored shipping containers, used to transport the material to and from other nuclear cities and sites, by truck and rail. Empty shipping containers are sometimes kept outdoors. High-resolution satellite photographs freely available on the Internet show them stacked in the yards, and in other ways help to identify the buildings in the complex. It does not matter that the photographs are old. Any of thousands of ordinary workers at Mayak could update them, and also provide information on material-processing schedules, NNSA upgrades, broken cameras, the patterns of the night shifts, and the locations of guards who use narcotics or drink on the job. It would not be difficult to find such an informant—for instance, in the taverns of Chelyabinsk. Afterward the action would have to be fast and accurate. Moving through the forest from the east or south perimeter fences, a raiding party on foot could hit any of the buildings within two hours. This would hardly be a sure thing. Ten years ago would have been a better time to try. But with luck it could still be done.

Or knowledgeable Americans urgently believe so. Some of the best of them work for a richly endowed organization known as the Nuclear Threat Initiative (NTI), which was founded by former Senator Sam Nunn and the CNN mogul Ted Turner. The director of NTI’s efforts in the former Soviet states is a smart, no-nonsense Washington insider named Laura Holgate, who spent years as a senior U.S. official grappling with Russian nuclear-weapons security. When we met, at NTI’s buttoned-up headquarters, in Washington, she repeated a colleague’s anecdote about a certain closed city where informal parking lots have sprung up outside of holes in the fence, because workers prefer not to bother with the gate. But I wanted to know about Ozersk, which seems to be more orderly, perhaps because of the quantities of bomb materials there. I asked about the security upgrades. She answered indirectly. “It’s a very challenging environment to work in. You set up a radiation detector, and it’s constantly going off, mostly false alarms. There’s so much clandestine movement of other stuff—cooking oil, insulation, you name it. So how are you supposed to catch the illicit movement of material as unradioactive as HEU? They put a little box of uranium in the corner, and they’re fine.” Moreover, the Russian government, she implied, can be somewhat dismissive of such matters.

Later, in Russia, I spoke to a modest American technician with a decade of experience in the secret cities of the Urals. He knew enough of Washington to want to remain anonymous, but did not seem interested in diplomatic maneuvering or nuclear policy. He was a detail man, a practitioner. He said, “Some of what the NNSA puts in is worthwhile, maybe. And other stuff seems, ah, not to get used. I took a tour one time, and they had installed a radiation monitor, a portal, for traffic going in and out of the site. So I say, ‘So, how’s the portal working?’

“And this Russian says, ‘Oh, we shut it off most of the time.’

“‘Why?’

“‘Because it’s always going off.’

“‘How come it always goes off?’

“He says, ‘Well … it’s the people on the buses. People go fishing in the lake, and when they catch fish and bring them out on the bus, they set off the radiation monitor. And then we’ve got to respond.’”

The technician laughed at his own story. Just when you think you’ve nabbed a terrorist, what you’ve really nabbed is a radioactive fish. After ten years around Russia’s nuclear facilities, he was not concerned. He said, “But you know, things like radiation portals? That’s just our method of security. They’ve got their own methods, which are probably as effective.”

“You mean human intelligence?”

“Yeah. It’s more people-intensive, the way they do it.”

Others have described Ozersk’s human side in dire terms—children selling drugs in the schools, mobsters in the construction and trucking business, large numbers of unvetted Central Asians arriving to do menial work that before would have been reserved for loyal Slavs. The worst of the lot are the soldiers whose duty is to guard the site. Called “the dregs of the dregs” by some critics, they are second-round conscripts, picked up by the Interior Ministry only after they have been rejected by the army. They in no sense constitute an elite corps, as the ministry sometimes claims. NTI has catalogued a string of known incidents of Mayak guards killing each other, committing suicide, stealing weapons, running away, buying narcotics, drinking on duty, and in one case imbibing a bottle of antifreeze and dying.

The problem for you, in your quest for a bomb, is that even these soldiers will fight. This is less a possibility than a fact. They will fight whether sober or drunk. Their presence at Ozersk means that no raiding party will be able to hit without provoking a noisy response. Still, American specialists have warned that a large and coordinated attack could overwhelm or counter traditional Russian defenses, pointing to the Chechen seizures of a Moscow theater in 2002 and a Beslan school in 2004 as evidence that terrorists within Russia already possess the wherewithal.

If you wanted a bomb, however, you would have to be concerned not only with getting your hands on fissile material but also with getting away. The getaway factor is not often raised in the United States, where the prevailing view is of a Wild West environment in which a nuclear bandit would need only to ride out of town in order to vanish safely. But up close to Ozersk, I kept running across evidence of a powerful and self-confident Russia that seemed more like its old autocratic self, and at least selectively under control.

The nearest international border to Ozersk is with Kazakhstan, only four hours’ drive to the south, but the route is poor, because it runs through Chelyabinsk and crosses border checkpoints using roads that can easily be blocked. A safer plan would be to head southwest, 1,200 miles or more, for the Caspian Sea or the Caucasus, with the goal of smuggling the uranium to Turkey through Azerbaijan, Armenia, Georgia, or northern Iran. The trip, however, would require at least three days just for the escape from Russia—an interval that becomes the minimum head start required before the loss of material is noticed at Mayak. And raiders would be lucky to have three hours. In good conscience, therefore (and perhaps with some relief), a rational bomb-maker would abandon any idea of commando heroics.

But that is not to say you should give up on Ozersk. Possibilities abound for an inside job. A culture of mobsterism and corruption broadens the opportunities for recruiting thieves. Insiders could neutralize any practical defenses, pass undetected through the gates with a load of unshielded HEU, and provide a getaway team with a head start that could be measured in weeks or months—perhaps right up to the time an assembled bomb ignites. Linton Brooks described the possibility of insider theft as the greatest challenge facing the NNSA in Russia today. The solutions, which at best can only be partial, consist of efforts to complicate the task of would-be thieves, requiring them to bring more people into a conspiracy and to operate with larger teams.

An agent with deep experience in these matters said, “You try to make it more difficult by putting in doors that require two people to open. You put in video surveillance. You put in watchers to watch the watchers. You put in accounting systems to bring the facilities from paper ledgers into the bar-code era. But there are very deep-seated cultural issues here. In the Soviet days you were a very trusted person, an elite person, if you were working in these facilities. And now we come along and say, ‘OK, you’re not allowed to go into the vault by yourself anymore.’ That can take a while to sink in. Also, there’s a different perception of rules and regulations. The rule of law is not looked at the same way. There’s more skepticism in Russia. It’s a very complicated problem. Bricking up windows is part of the solution, but it’s not everything.”

It’s not as if the Russians don’t already know how to build strong rooms. I once mentioned to the American technician in the Urals that even ordinary doors there are heavy. He said, “One of the guys at the plant said to me, ‘Jack, I watch American movie last night.’” The technician imitated a Russian accent. “‘And I see something strange. Drug police. They kick in door! In Russia, this not possible!’”

Or maybe in Russia, this not necessary.

In Washington I spoke to one of the many officials in town who, though carrying out their assignments perfectly well, are too nervous about domestic politics to risk being identified. He described an NNSA effort called the Second Line of Defense, which installs radiation monitors at border-crossing points throughout the former Soviet Union. The program is most fully developed in Georgia, a skeletal nation threatened by separatist enclaves and barely able to keep warm through the winters. The official said, “The U.S. Corps of Engineers is working with Customs to build whole new border- crossing facilities in Georgia, and we’re in the process of installing upgraded radiation monitors, in conjunction with that.”

I said, “I guess I can see the logic of the radiation monitors, but why are we building customs stations for the Georgians?”

“It’s a joint venture to try to control smuggling.”

Georgia is one of the most corrupt nations on earth. Many of its politicians are crooks. Its officials routinely steal. Its economy is based almost entirely on black markets. Its people have nothing to survive on if they do not hustle. I said, “Why should we care about ordinary smuggling in Georgia? Cigarettes? Vodka? Fuel? For that matter, narcotics?” What I meant was, If even the ordinary black markets of Georgia are seen as a threat, where does the impulse to impose order end?

He stayed in his lane, but executed a tidy reversal. He said, “It’s a good question. And I don’t know that I have the answer, but the genesis anyway was nuclear. So I guess in conjunction with our new equipment, it makes sense to upgrade the stations.”

Hoping to get some notion of the realities, I went to a model Customs project, a station the Americans call “Red Bridge,” which stands at the main crossing point between Georgia and Azerbaijan. The first improvement there was the construction of a housing compound where for several years before, border guards had been living in tents and cooking over open fires. By the time the U.S. Department of Homeland Security was finished with it, in 2003, the compound consisted of five single-family houses, a barracks, a dining hall, an administration building, a vehicle-maintenance garage, a warehouse for supplies, an armory, various utility buildings, a dog kennel, two water-storage buildings, a sewage plant, an electrical substation, perimeter fences, two guard towers, two helipads, a sports field, a separate soccer field, some new paved roads and parking lots, and, of course, a parade ground. After the official opening, a Customs and Border Protection newsletter asked the CBP program manager, James Kelly, “Is it as grand as it sounds?”

“Indeed it is,” Kelly said. “It was built to Western standards.”

Golly, and that was just Phase One. The United States then shifted its spending down the road to the port of entry itself. Over the following two years, the CBP oversaw the construction of a $2.2 million facility, whose primary purpose would be to “help Georgia become a more effective partner in the worldwide effort to control the passage of terrorists and their weapons.” This time the improvements included a six-lane roadway, inspector booths, cargo-inspection areas, closed-circuit remote-control television cameras, lots of computers, a high-frequency long-distance radio-communication system, and the crowning glory—a beautiful air-conditioned two-story stucco building with spaces for processing the grateful public, as well as detention cells, back offices, a dormitory wing for Georgian customs agents, private sleeping rooms for American officials and other VIPs, and a second-floor patio, apparently for sitting outside on warm evenings.

The day I got there was too chilly for that. A short line of trucks crept in from Azerbaijan, and a few cars passed through in the opposite direction. The new radiation monitors were not yet in place, but within a few weeks they were certain to be. I asked to see the control room, which is meant to be the nerve center of the operation. There was a delay until the key could be found, whereupon the chief showed me in. The room was a windowless refuge, empty except for a few chairs, some computers, and a bank of flat-screen displays. Someone had left a video game on. Someone had left a magazine. Empty cardboard boxes were stacked in a corner. After a while an eager young man came in and demonstrated how a camera on the roof could be swiveled and zoomed.

The anonymous official in Washington tried to explain why such improvements matter. I had asked him why anyone carrying a bomb’s worth of HEU would choose to go through any official gate anywhere. He did not predict that anyone would. Using Georgia as an example he said, “The good news is that there aren’t very many of these border crossings. And to the extent we’ve got them covered, we force people to use horses, I guess.” Horses, mules, donkeys, tractors, dirt bikes, plucky little Ladas—whatever. A bit of off-road travel wouldn’t seem like much of an inconvenience to nuclear terrorists on the move, and American officials know it well. The problem is that U.S. agencies, when pressured by conflicting mandates and forced to work with corrupt and dysfunctional local governments, essentially throw up their hands at the complexity of it all, and abandon the fight in advance.

In this context Red Bridge is not just a customs post, but a premature surrender—and a typical one. Faced with the need to put systems in place that will function day after day to identify unexpected nuclear smugglers, America turns to the uniformed agents of local governments, loads them down with air-conditioned buildings and gadgetry, and then asks them to sit in a closed room watching for information from television cameras and radiation monitors. To make things worse, the failure is not individual but collective, and therefore difficult to correct. It appears to include even the clandestine services, some of which are chasing al-Qaeda around, but none of which shows signs of wanting to lay traplines through the back country on the off chance—highly unlikely in any given location—of snaring two knapsacks’ worth of dull grey metal. Of course it is possible that they are doing this, and being so discreet that for once they leave no evidence of their passage, but if I had to move a load of HEU across international borders, I would gamble that they are not.

It is as if the U.S. government, when looking at a world map for purposes of HEU interdiction, has declared entire regions to be off-limits to anything other than fictions. As noted, those back-country regions are where some of the world’s principal opium routes lie and, for independent reasons, are the areas through which stolen HEU seems most likely to pass. It is generally assumed that for the right price, opium traffickers will provide transportation, lodging, and expert advice to nuclear terrorists moving through. Indeed, the persistence of the drug trade worldwide is often used as an object lesson in the difficulty of trying to stop the smuggling of nuclear materials. A tired joke is that the best way to smuggle an atomic bomb is to put it into a bale of marijuana. For the United States the proximity between the two trades is an unfortunate coincidence, but it could be turned into a fortunate one, through quiet conversations with a few key people. If you wanted a bomb, you should of course be having quiet conversations with those same key people.

Finding them merely requires casual exploration along the preexisting lines of defense, especially along the national boundaries that cross the smuggling routes. The work would involve poking around meekly, sometimes with an amateur translator and guide. The purpose would not be to recruit peasant armies and spies, but to get a feel for the local functioning of power. In most areas there are only two or three people at the top, and they tend to be at once aggressive and benevolent men with interests larger than just the movement of drugs. Their names would quickly become apparent. Some might be dangerous to approach, but most would be hospitable to strangers. On the second or third trip back, a U.S. agent might make it known that if ever a load of HEU showed up, a large reward would be paid.

Some months ago I made the first of two brief trips through the mountains of far-eastern Turkey, in the Kurdish hinterland along the border with Iran. This is the prime smuggling country I mentioned earlier. The local villages may seem sleepy, but they are wide awake. It became apparent to me that the entire region is tightly sewed up, that nothing moves here without notice, and that any sustained activity requires approval. The authority is not the Turkish government. The army is here to fight the sporadic guerrilla war against Kurdish separatists, who occasionally ambush a patrol or plant a mine, then retreat higher into the mountains. The main garrison is in a tiny town called Bascale, which is better known for its heroin labs. When A. Q. Khan was building Pakistan’s nuclear-weapons facilities, some of the more sensitive parts passed through here.

Bascale has a large yellow house in a central compound, which belongs to the chief of a clan that is dominant for miles around, an extended family named Ertosi, which consists of 150,000 to 200,000 people. One afternoon I was invited in for tea. The host was a black-bearded heavyset middle-aged man who had arrived from Ankara for a few days. He was a son of the chief, and so was important, too. He had guards and flunkies around. We sat against the walls in a large bare room, with photographs of dead Ertosis at one end, and a big-screen television at the other. The television was tuned to a banned Kurdish station, broadcast via satellite from London. My translator introduced me as an English teacher on holiday, which threw our host a little off balance, I think. He wore a heavy gold chain and a diamond ring. He had an array of cell phones in front of him on the floor, several of which rang. Between calls, I asked him about the diesel trade. He said a little, but then clammed up when I got into details. To my translator he said, “Why does an English teacher want to know so much?”

Weeks later I returned to the area and sought out a certain subclan leader I had heard about. He was a powerful man, the chief of the 20,000 Ertosis who occupy the most active stretch of the border. I found him in a hamlet high on a mountainside. It was late at night, and the air was cold. We sat among ten of his men in a small stone room around a wood-burning stove. He was a small man, sixty years old, with sharp hooded eyes and a hooked nose. He was dressed like a Kurdish peasant, with a tweed jacket down to his knees. His hands were hardened by manual labor. From the ease with which he gave orders and the deference shown to him by his men, it was clear that he had been laying out law for many years. We avoided mention of narcotics or HEU, and talked instead in a transparent code about the business in “diesel fuel.” He said, “The government cannot control the border. The Kurds just naturally do.” No stranger could cross without his knowledge.

Then comes the problem of assembly. You should take it as a given that no nation would dare to be associated with the construction of an independent atomic bomb. Not North Korea, not Sudan, not Iran. The certainty of retribution after its use far outweighs whatever benefit might be gained. Moreover, you could never trust those governments not to wait until the end and confiscate the goods. So the work will have to be carried out in secret—and in some private machine shop perhaps no larger than a five-car garage. The shop will contain numerically controlled milling machines and lathes, as well as other expensive equipment, and will require a plausible explanation—a front company set up to manufacture, say, transmission components. The best location would seem to be in some Third World city, where governmental control is lax, corruption is rampant, and the noise from the shop will be masked by industrial activities nearby. Take your pick: Mombasa, Karachi, Mumbai, Jakarta, Mexico City, and a finite list of others. You might as well choose Istanbul, because it’s close to where you’d be coming from.

Construction of the bomb would take maybe four months. The size of your technical team would depend on the form of HEU. At the minimum it would consist of a nuclear physicist or engineer, a couple of skilled machinists, an explosives expert, and perhaps an electronics person, for the trigger. The essentials of the work are easy to grasp. A cannon-type bomb employs two HEU hemispheres which when joined together form a single sphere. A sphere is the geometric shape that provides the smallest surface area in relation to mass, lowering the threshold of criticality. The receiving hemisphere is hollow on one side. The other hemisphere, known as the bullet or plug, has a perfectly matched convex surface on the opposing side. It sits in wait at the top of a steel barrel sometimes as long as a stepladder is high. When the time comes to detonate the bomb, a chemical propellant shoots the plug down the barrel. The ideal velocity is 1,000 meters per second, about as fast as a high-speed rifle round. The goal is to slam the masses together before the nuclear reaction has a chance to start. Once they have joined, the HEU can be trusted to do its job, though with a primitive, garage-made device there will be some uncertainty about the explosive yield. A common misperception about terrorist bomb makers is that they would try to imitate a military device, and would therefore require a level of expertise found only in government laboratories—but you would not care whether New York was hit with a ten- or a twenty-kiloton blast.

Even so, the construction of a bomb is not a casual project. The machinery, the noise, and especially the likely presence of team members who are not locals, provide the United States with the last chance of self-defense. A city like Istanbul, which appears from a distance to be anarchic, and is famously resistant to central authority, is up close and in practice a patchwork of tightly knit communities with something of the organic power structures of the borderlands. In even the most chaotic neighborhoods, where industrial shops are mixed among illegal apartment blocks and communities of impoverished newcomers and squatters, it would be difficult to keep the neighbors from asking inconvenient questions. The same is true in Mombasa, Karachi, and every other city where a bomb could conceivably be built. A United States government that could lay traplines in their slums would have a much better chance of stopping a terrorist attack than any amount of naval maneuvering or bureaucratic restructuring can provide.

In the end, if you wanted a bomb and calculated the odds, you would have to admit that they were stacked against you, simply because of how the world works—and that this may be why others like you, if there have been any, have so far not succeeded. You would understand, though, that the odds are not impossible. You would of course have many concerns as you moved ahead. But perhaps the thing that should worry you the least is the American government’s war on terror.