How to Keep the Bomb From Iran

PREVENTING THE UNTHINKABLE

The ongoing crisis with Tehran is not the first time Washington has had to face a hostile government attempting to develop nuclear weapons. Nor is it likely to be the last. Yet the reasoning of U.S. officials now struggling to deal with Iran's nuclear ambitions is clouded by a kind of historical amnesia, which leads to both creeping fatalism about the United States' ability to keep Iran from getting the bomb and excessive optimism about the United States' ability to contain Iran if it does become a nuclear power. Proliferation fatalism and deterrence optimism reinforce each other in a disturbing way. As nuclear proliferation comes to be seen as inevitable, wishful thinking can make its consequences seem less severe, and if faith in deterrence grows, incentives to combat proliferation diminish.

A U.S. official in the executive branch anonymously told The New York Times in March 2006, "The reality is that most of us think the Iranians are probably going to get a weapon, or the technology to make one, sooner or later." Such proliferation fatalists argue that over the long term, it may be impossible to stop Iran -- or other states for that matter -- from getting the bomb. Given the spread of nuclear technology and know-how, and the right of parties to the Nuclear Nonproliferation Treaty (NPT) to enrich uranium and separate plutonium, the argument goes, any foreign government determined to acquire nuclear weapons will eventually do so. Moreover, the 1981 Israeli attack on the Osirak nuclear reactor in Iraq may have delayed Iraq's progress, but similar air strikes are unlikely to disable Iran's capacities, since its uranium-enrichment facilities can be hidden underground or widely dispersed. Imposing economic sanctions through the UN Security Council is clearly a preferable option. But as Washington learned with India and Pakistan in the 1980s and 1990s, sanctions only increase the costs of going nuclear; they do not reduce the ability of a determined government to get the bomb.

Faced with only unattractive options to stem proliferation, some Bush administration officials are reluctantly preparing to live with a nuclear Iran. Military planners and intelligence officers have reportedly been tasked with developing strategies to deter Tehran if negotiations fail. Washington officials cry that the sky is falling whenever they face the prospect of a hostile state's getting the bomb, yet they seem to find solace in the recollection that deterrence and containment did work to maintain the peace during the Cold War. So why worry that the latest crop of rogue regimes might prove less deterrable than the Soviet Union and China? The Bush administration already appears to have adopted this logic with respect to North Korea. According to The New York Times, administration officials privately predict that deterrence will work against Pyongyang: "The North Koreans know that a missile attack on the United States would result in the vaporization of Pyongyang," the paper quoted an official as saying. And if deterrence can work with Kim Jong Il, why not with Ayatollah Ali Khamenei? "Iran is just one instance of the [proliferation] problem, and in Iran's case, containment might work," argues Brent Scowcroft, who was national security adviser to President George H. W. Bush.

But both deterrence optimism and proliferation fatalism are wrong-headed. Deterrence optimism is based on mistaken nostalgia and a faulty analogy. Although deterrence did work with the Soviet Union and China, there were many close calls; maintaining nuclear peace during the Cold War was far more difficult and uncertain than U.S. officials and the American public seem to remember today. Furthermore, a nuclear Iran would look a lot less like the totalitarian Soviet Union and the People's Republic of China and a lot more like Pakistan, Iran's unstable neighbor -- a far more frightening prospect. Fatalism about nuclear proliferation is equally unwarranted. Although the United States did fail to prevent its major Cold War rivals from developing nuclear arsenals, many other countries curbed their own nuclear ambitions. After flirting with nuclear programs in the 1960s, West Germany and Japan decided that following the NPT and relying on the protection of the U.S. nuclear umbrella would bring them greater security in the future; South Korea and Taiwan gave up covert nuclear programs when the United States threatened to sever security relations with them; North Korea froze its plutonium production in the 1990s; and Libya dismantled its nascent nuclear program in 2003.

Given these facts, Washington should work harder to prevent the unthinkable rather than accept what falsely appears to be inevitable. The lesson to be drawn from the history of nonproliferation is not that all states eyeing the bomb eventually get it but that nonproliferation efforts succeed when the United States and other global actors help satisfy whatever concerns drove a state to want nuclear weapons in the first place. Governments typically pursue nuclear power for one of three reasons: to protect themselves against an external security threat, to satisfy the parochial interests of domestic actors, or to acquire an important status symbol. Iran is, mostly, a classic case of a state that wants nuclear weapons to dissuade an attack. It sits in a perennially unstable region, has long faced a belligerent Iraq, and now wants to stand up to Washington's calls for regime change in Tehran. Any viable solution to Tehran's appetite for nuclear weapons will therefore require that Washington learn to coexist peacefully with Iran's deeply problematic government. U.S. officials should not assume that Iran will go nuclear no matter what and draw up plans for containing it when it does. Nor should Washington rely exclusively on UN sanctions, which might not work. Instead, the U.S. government must dig into its diplomatic toolbox and offer -- in conjunction with China, Russia, and the EU-3 (France, Germany, and the United Kingdom) -- contingent security guarantees to Tehran.

DELUSIONS OF DETERRENCE

The nuclear monopoly the United States enjoyed at the end of World War II did not last long. Nonproliferation discussions in the United Nations soon after the war came to naught because the Soviet Union understandably distrusted any plan that gave the United States a monopoly on the scientific knowledge and engineering experience needed to build a nuclear weapon. As Cold War hostilities grew, first President Harry Truman and then President Dwight Eisenhower considered launching attacks against the Soviet Union to prevent it from developing a nuclear arsenal. Moscow had tested its first atomic bomb in 1949, but it was the prospect of the Soviets' amassing a large H-bomb arsenal that particularly alarmed Eisenhower. In 1953, he asked Secretary of State John Foster Dulles if "our duty to future generations did not require us to initiate war at the most propitious time that we could designate." Eisenhower eventually rejected the idea, however, because he feared the Red Army would respond by invading U.S. allies in Europe. Even if the United States did emerge victorious from such a conflict, Eisenhower told his advisers in 1954, "the colossal job of occupying the territories of a defeated enemy would be far beyond the resources of the United States at the conclusion of this war."

As the Soviet nuclear arsenal expanded, it triggered a chain reaction. The United Kingdom and France raced to develop their own nuclear weapons (which they first detonated in 1952 and 1960, respectively), partly as an independent deterrent to Soviet aggression in Europe but also as a symbol of their continuing great-power status. That U.S. allies developed such capacities did not much concern Washington, but the U.S. government became deeply worried that China under Mao Zedong might acquire its own bomb. Still, the Kennedy administration rejected plans to launch a preventive air strike on Chinese nuclear facilities in 1963 for fear that it would spark a major war and because the Soviets had rejected Washington's secret request for their assistance.

It is common today to look back nostalgically on those years as "the long peace." But this oversimplifies the challenges of the Cold War. Nuclear weapons did seem to have a sobering influence on the great powers, but that effect was neither automatic nor foolproof. Both the Soviet and the Chinese governments originally hoped that having the bomb would allow them to engage in more aggressive policies with impunity. Moscow repeatedly threatened West Berlin in the late 1950s and early 1960s, for example, confident that its growing arsenal would dissuade the United States from coming to West Germany's defense. Soviet Premier Nikita Khrushchev also believed that if the Soviet Union could place nuclear weapons in Cuba, the United States, once faced with the fait accompli, would be deterred by the Soviet arsenal from attacking Fidel Castro's regime.

What could be called dangerous learning by "trial and terror" also characterized relations with China. Mao appears to have genuinely believed that nuclear weapons were "paper tigers" and that China could survive any large-scale nuclear war. Beijing's foreign policy certainly did not turn moderate after its 1964 nuclear tests. Mao ordered military ambushes of Soviet armed forces on the disputed Chinese-Soviet border in March 1969, instructing Chinese generals not to worry about Moscow's response because "we, too, have atomic bombs." Soviet leaders retaliated against Chinese units along the border and threatened a preventive nuclear strike against China's nuclear facilities. Mao eventually accepted a negotiated settlement of the territorial dispute, but only after evacuating the Chinese leadership to the countryside and putting China's nuclear arsenal on alert.

A REGIME IS BORN

The frightening crises of the 1960s led U.S. and Soviet leaders to understand that nuclear weapons guaranteed only a precarious peace. Increasingly, the two superpowers pursued bilateral arms control measures -- such as the Strategic Arms Limitation Talks and the Anti-Ballistic Missile Treaty -- to try to manage their nuclear relationship. They also recognized that a new multilateral approach was needed to stop the spread of nuclear weapons.

In March 1963, President John F. Kennedy told the press that he was "haunted" by the fear that by the 1970s the United States would "face a world in which 15 or 20 or 25 nations" possessed nuclear weapons. Five years of negotiations later, the United States, the Soviet Union, the United Kingdom, and 59 non-nuclear-weapons states signed the NPT. Under the terms of the treaty, states possessing nuclear weapons agreed not to transfer weapons or knowledge about how to build them to their friends and allies. (This commitment effectively ended Washington's hope of supplying West Germany and other NATO powers with "a multilateral force" of nuclear weapons, a prospect that had deeply troubled Moscow.) They also undertook "to work in good faith" toward the eventual elimination of nuclear weapons. The non-nuclear-weapons states, for their part, agreed not to seek nuclear weapons and to cooperate with inspectors from the International Atomic Energy Agency (IAEA) to allow monitoring of their peaceful nuclear research and energy facilities. The idea behind this "I won't if you won't" provision was to reduce the security threats, potential or real, that non-nuclear-weapons states posed to one another. The treaty also guaranteed that non-nuclear-weapons states in good standing would gain the full benefits of peaceful nuclear energy production, creating a "sovereign right," Iran has since argued, for any such state to develop a full nuclear-fuel production cycle of its own. The broad ambition behind the NPT was to slow down proliferation by reducing the demand for nuclear weapons. By both providing some assurance that states subscribing to the treaty would not develop nuclear bombs and creating, through the IAEA, a system to detect their efforts if they did, the NPT assuaged the security concerns of many states. It also reduced the bomb's appeal as a status symbol by creating an international norm according to which "responsible" states followed NPT commitments and only "rogue" states did not. And by offering hope that the nuclear states would take significant steps toward eventual disarmament, the treaty made it easier for nonnuclear governments to justify their own self-restraint to their domestic constituencies.

The NPT system proved reasonably successful for quite a long while. Although they are less discussed than the failures, the nonproliferation successes -- the nuclear dogs that did not bark -- are more numerous. Many non-nuclear-weapons states did continue to develop nuclear energy facilities after the NPT was signed, and some -- such as Japan, with its massive plutonium stockpile -- kept nuclear materials and continued their nuclear research in case the NPT regime fell apart. (Uncertainty about the treaty was so strong at first that Japan and other nonnuclear states insisted that they be allowed to review and renew their membership every five years.) But the NPT and U.S. security guarantees eventually reduced those countries' interest in proliferation. Other U.S. allies were caught cheating -- most notably South Korea in the 1970s and Taiwan in the 1980s -- but they ended suspected military-related activities when Washington confronted them and threatened to withdraw its security assistance. Egypt sought nuclear weapons in the early 1960s, but it signed the NPT in 1968 and ratified it in 1979 after striking a peace deal with Israel that reduced its national security concerns. Belarus, Kazakhstan, and Ukraine were nuclear powers from the moment of their independence, having inherited arsenals when the Soviet Union collapsed in 1991. But they soon handed over the weapons to Russia in exchange for economic assistance, highly limited security assurances from the United States, and a chance to join the NPT in good standing. The NPT has been enough of a success that at the 1995 NPT Review Conference, all 178 states that had ratified it agreed to extend it permanently.

PERILS OF PROLIFERATION

A few outliers have bucked the system, however, and it is their actions that have bred the fatalism about proliferation that now dominates in Washington. Israel has never officially admitted to possessing nuclear weapons, but it is widely known to have constructed (with France's help) a small arsenal in the 1970s. South Africa secretly built seven nuclear devices under the apartheid regime in the 1980s (but unilaterally destroyed them well before a black-majority-rule government took over in 1994). India and Pakistan developed nuclear capabilities in the late 1980s and came out of the closet with them in May 1998. Iraq had been inching along, too, and after the 1981 Israeli air strike on its Osirak reactor, it started an underground gaseous diffusion facility to produce bomb-grade uranium, which was belatedly discovered and destroyed by UN inspectors after the 1991 Gulf War.

A number of political and military developments since the 1990s have further weakened the nonproliferation regime. The Pakistani scientist A. Q. Khan, among others, began secretly selling uranium-enrichment capabilities and even bomb designs to potential proliferators. The emergence of new nuclear states has threatened those states' neighbors, and the United States itself is increasingly seen as a security threat by some potential proliferators. Some states -- Iran in particular -- insist that they have a "right" to develop nuclear-fuel-production capabilities, which would get them uncomfortably close to developing nuclear bombs if they were subsequently to quit the treaty. In 1999, the U.S. Senate also dealt the regime a blow by voting against ratification of the Comprehensive Test Ban Treaty despite the Clinton administration's promise to ratify it during the 1995 NPT conference as proof of the U.S. commitment to eventual disarmament.

Most important, some new nuclear states have proved to be particularly risky actors. Consider the unsettling case of Pakistan. Islamabad has been dangerously lax since its 1998 nuclear tests, exercising weak control over its military personnel, intelligence officials, and scientists who have access to nuclear weapons, materials, and technology. Soon after the 1998 tests, Pakistani military planners developed more belligerent strategies against India. Dusting off an old plan, in the winter of 1999, Pakistani infantry units disguised as mujahideen snuck into Indian-held Kashmir. The incursion sparked the 1999 Kargil War, in which over 1,000 soldiers were killed on both sides before Pakistani forces reluctantly withdrew. According to U.S. and Indian intelligence, before the fighting ended, the Pakistani military had started to ready its nuclear-capable missiles for potential use. But when President Bill Clinton raised the possibility that this had happened with Pakistani Prime Minister Nawaz Sharif, he displayed a disturbing lack of knowledge about what his own military was doing. Similarly, Pakistani leaders gave important nuclear command-and-control responsibilities to the notorious Inter-Services Intelligence (ISI), which has intimate ties to both the Taliban and jihadist groups fighting in Kashmir. Doing so was a recipe for trouble, raising the risks that a rogue faction could steal a weapon or give it to terrorists. According to credible reports, during the Kargil War, Pakistani military planners and the ISI considered hiding Pakistan's nuclear weapons in western Afghanistan to protect them from a potential preemptive attack by India; they even contacted Taliban officials to explore the option. Islamabad has also exercised incredibly loose control over Pakistani nuclear scientists. After the 9/11 attacks, it was discovered that a number of individual scientists -- including Sultan Bashiruddin Mahmood, a senior official of the Pakistan Atomic Energy Commission (PAEC) -- had met with Osama bin Laden in Afghanistan and discussed techniques for developing nuclear weapons and other weapons of mass destruction. In April 2002, Pakistani President Pervez Musharraf admitted that PAEC scientists had been in contact with al Qaeda but claimed that "the scientists involved had only very superficial knowledge." Most proliferation experts also believe that senior Pakistani military officers were involved in many, if not all, of the deals in which A. Q. Khan and his associates sold nuclear centrifuge components to Iran and Libya, offered to help Saddam Hussein build a bomb just before the 1991 Gulf War, and provided North Korea with uranium-enrichment technology.

THE MOST DANGEROUS GAME

Dealing with a nuclear Iran in the near future would be more like dealing with Pakistan than with nuclearized democracies such as Israel and India or even nuclear totalitarian states such as the Soviet Union and China. Not only does Iranian President Mahmoud Ahmadinejad spew belligerent anti-Israel and Holocaust-denying statements, but the Iranian government as a whole continues to nurture revolutionary ambitions toward Iran's conservative Sunni neighbors and to support Hezbollah and other terrorist organizations. Tehran, like Islamabad, would be unlikely to maintain centralized control over its nuclear weapons or materials. In order to deter Tehran from giving nuclear weapons to terrorists, in January 2006 the French government announced that it would respond to nuclear terrorism with a nuclear strike of its own against any state that had served as the terrorists' accomplice. But this "attribution deterrence" posture glosses over the difficult question of what to do if the source of nuclear materials for a terrorist bomb is uncertain. It also ignores the possibility that Tehran, once in possession of nuclear weapons, would feel emboldened to engage in aggressive naval actions against tankers in the Persian Gulf or to assist terrorist attacks as it did with the Hezbollah bombing of the U.S. barracks at the Khobar Towers in Saudi Arabia in 1996.

There is no reason to assume that, even if they wanted to, central political authorities in Tehran could completely control the details of nuclear operations by the Islamic Revolutionary Guard Corps. The IRGC recruits young "true believers" to join its ranks, subjects them to ideological indoctrination (but not psychological-stability testing), and -- as the IAEA discovered when it inspected Iran's centrifuge facilities in 2003 -- gives IRGC units responsibility for securing production sites for nuclear materials. The IRGC is known to have ties to terrorist organizations, which means that Iran's nuclear facilities, like its chemical weapons programs, are under the ostensible control of the organization that manages Tehran's contacts with foreign terrorists. It is misguided simply to hope that eventual regime change in Tehran would end the nuclear danger because, in the words of one Bush administration official, who spoke to The New York Times anonymously, Washington would then "have a different relationship with a different Iranian government." This wish assumes that another Iranian revolution would end gently, with an orderly transfer of power, rather than in chaos and with the control of nuclear weapons left unclear.

THE REASONS WHY

If Iran must not be allowed to go nuclear, what then can be done to stop it? A U.S. military strike on Iran today should be avoided for the same prudent reasons that led Eisenhower and Kennedy to choose diplomacy and arms control over preventive war in their dealings with the Soviet Union and China. Even if U.S. intelligence services were confident that they had identified all major nuclear-related sites in Iran (they are not) and the Pentagon could hit all the targets, the United States would expose itself (especially its bases in the Middle East and U.S troops in Afghanistan and Iraq), and its allies, to the possibility of severe retaliation. When asked about possible U.S. air strikes in August 2004, Iranian Defense Minister Ali Shamkhani said, "You may be surprised to know that the U.S. military presence near us is not power for the United States because this power may under certain circumstances become a hostage in our hands. ... The United States is not the only power present in the region. We are also present from Khost to Kandahar in Afghanistan and we are present in the Gulf and can be present in Iraq." Iran might also support attacks by terrorist groups in Europe or the United States. Bush administration officials have sought to give some teeth to the threat of a military attack by hinting that Israel might strike on Washington's behalf. The Pentagon notified Congress in April 2005 of its intention to sell conventional GBU-28 "bunker-buster" bombs to Israel, and President George W. Bush reasserted Washington's commitment to "support Israel if her security is threatened." But an Israeli air strike on Iran's nuclear facilities would do no more good than a U.S. one: it could not destroy all the facilities and thus would leave Tehran to resume its uranium-enrichment program at surviving sites and would give Iran strong incentives to retaliate against U.S. forces in the Middle East. Muslim sentiment throughout the world would be all the more inflamed, encouraging terrorist responses against the West.

With no viable military option at hand, the only way for Washington to move forward is to give Tehran good reason to relinquish its pursuit of nuclear weapons. That, in turn, requires understanding why Tehran wants them in the first place. Iran's nuclear energy program began in the 1960s under the shah, but even he wanted to create a breakout option to get the bomb quickly if necessary. One of his senior energy advisers once recalled, "The shah told me that he does not want the bomb yet, but if anyone in the neighborhood has it, we must be ready to have it." At first, Ayatollah Ruhollah Khomeini objected to nuclear weapons and other weapons of mass destruction on religious grounds, but the mullahs abandoned such restraint after Saddam ordered chemical attacks on Iranian forces during the Iran-Iraq War. As former Iranian President Hashemi Rafsanjani, then the speaker of Iran's Parliament, noted in 1988, the conflict with Saddam showed that "the moral teachings of the world are not very effective when war reaches a serious stage," and so Iranians must "fully equip ourselves in the defensive and offensive use of chemical, bacteriological, and radiological weapons." Tehran began purchasing centrifuge components from A. Q. Khan's network in 1987 and received, according to the IAEA, documents on how to cast enriched uranium into the form needed for nuclear weapons. Iran's nuclear-development efforts were further accelerated when, after the 1991 Gulf War, UN inspectors discovered and disclosed that Iraq had been just one or two years away from developing nuclear weapons of its own.

The end of Saddam's rule in 2003 significantly reduced the security threat to Tehran. But by then the United States had already taken Iraq's place, Washington having made it clear that it wanted regime change in Iran, too. In his January 2002 State of the Union address, President Bush had denounced the governments of Iran, Iraq, and North Korea as members of an "axis of evil" with ties to international terrorism. Increasingly, Bush administration spokespeople were advocating "preemption" to counter proliferation. After the fall of Baghdad, an unidentified senior U.S. official told a Los Angeles Times reporter that Tehran should "take a number," hinting that it was next in line for regime change. It did not help that the 2002 Nuclear Posture Review, which was leaked to the press, listed Iran as one of the states to be considered as a potential target by U.S. nuclear war planners. When asked, in April 2006, whether the Pentagon was considering a potential preventive nuclear strike against Iranian nuclear facilities, President Bush pointedly replied, "All options are on the table."

In the meantime, Iran's program has advanced. The last official U.S. intelligence estimate given to Congress, in February 2006, vaguely stated that if Iran "continues on its current path [it] will likely have the capacity to produce a nuclear weapon within the next decade" -- an estimate that has since been widely interpreted to mean five to ten years. Last April, Tehran began operating a cascade of 164 uranium-enrichment centrifuges at Natanz. According to the State Department, it will take over 13 years for an experimental cascade of this size to produce enough highly enriched uranium for even a single nuclear weapon. But without an arms control agreement, Iran is free to construct more centrifuge cascades at Natanz, and without intrusive IAEA inspections in place, Iran could build a covert enrichment facility elsewhere. What was once a proliferation problem is now a proliferation crisis.

AGREED FRAMEWORK IN FARSI

The depth of Tehran's security concerns is precisely the reason that, despite the Bush administration's hopes, Libya cannot be a model for how to deal with Iran now. Libyan President Muammar al-Qaddafi finally relinquished the pursuit of nuclear weapons in 2003 in exchange for both an end to trade sanctions and positive economic incentives. But Tripoli was always a very different foe from Tehran. For one thing, the Libyans turned out to be the gang that could not proliferate straight. For years, Qaddafi reportedly tried but failed to purchase complete nuclear weapons directly from China, India, and Pakistan. When he did purchase 20 centrifuges and components for another 200 from A. Q. Khan in 1997, he could not get enough of the machines assembled in the right way. In the late 1990s, moreover, as Qaddafi's regime was becoming more concerned with domestic threats -- economic stagnation and the rise of jihadist insurgents -- than it was with external ones, its nuclear program began to turn into a liability. Tehran today is in a very different position: it is much closer to being able to develop weapons, and it continues to have serious external security reasons for wanting them.

A better source of inspiration for handling Iran would be the 1994 Agreed Framework that the United States struck with North Korea. The Bush administration has severely criticized the deal, but it contained several elements that could prove useful for solving the Iranian nuclear crisis.

After the North Koreans were caught violating their NPT commitments in early 1993 (they were covertly removing nuclear materials from the Yongbyon reactor), they threatened to withdraw from the treaty. Declaring that "North Korea cannot be allowed to develop a nuclear bomb," President Clinton threatened an air strike on the Yongbyon reactor site if the North Koreans took further steps to reprocess plutonium. In June 1994, as the Pentagon was reinforcing military units on the Korean Peninsula and briefing Clinton on war preparations, Pyongyang froze its plutonium production, agreed to let IAEA inspectors monitor the reactor site, and entered into bilateral negotiations with a view to eventually eliminating its nuclear capability. It is unclear whether North Korea blinked out of fear of military intervention, because of concerns about economic sanctions, or because Washington's proposal held out the promise of security guarantees and normalized relations. But the talks produced the October 1994 Agreed Framework, under which North Korea agreed to eventually dismantle its reactors, remain in the NPT, and implement full IAEA safeguards. In exchange, the United States promised to provide it with limited oil supplies, construct two peaceful light-water reactors for energy production, "move toward full normalization of political and economic relations," and extend "formal assurances to [North Korea] against the threat or use of nuclear weapons by the U.S."

By 2002, however, the Agreed Framework had broken down, not only because Pyongyang was suspected of cheating but also because it believed that the United States, by delaying construction of the light-water reactors and failing to start normalizing relations, had not honored its side of the bargain. When confronted with evidence of its secret uranium program, in November 2002, Pyongyang took advantage of the fact that the U.S. military was tied down in preparations for the invasion of Iraq and withdrew from the NPT, kicked out the inspectors, and started reprocessing plutonium. Pyongyang is now thought to have six to eight nuclear weapons, to be producing more plutonium in the Yongbyon reactor, and to be constructing a larger one.

President Bush famously promised, in his 2002 State of the Union address, that the United States "will not permit the world's most dangerous regimes to threaten us with the world's most destructive weapons." Yet when North Korea kicked out the IAEA inspectors, Secretary of State Colin Powell proclaimed that the situation was "not a crisis," and Bush repeatedly declared that the United States had "no intention of invading North Korea." Deputy Secretary of State Richard Armitage quickly underscored the position: "The president has no hostile intentions and no plans to invade. That's an indication that North Korea can have the regime that [it] want[s] to have." The point was not lost on Tehran.

The 1994 Agreed Framework thus serves as a reminder of what to do, and its failure as a warning about what to avoid. If Washington is to offer security assurances to Tehran, it would be wise to do so soon (making the assurances contingent on Tehran's not developing nuclear weapons), rather than offering them too late, as it did with North Korea (and thus making them contingent on Tehran's getting rid of any existing nuclear weapons). As with North Korea, any deal with Iran must be structured in a series of steps, each offering a package of economic benefits (light-water reactors, aircraft parts, or status at the World Trade Organization) in exchange for constraints placed on Iran's future nuclear development.

Both Washington and Tehran will need to make major compromises. The Bush administration has said that a condition of any deal must be that "not a single centrifuge can spin" in Iran. But it might have to soften its stance. Allowing Tehran to maintain its experimental 164-centrifuge cascade, which poses no immediate danger and yet is an important status symbol for the Iranian regime, could help Tehran save face and sell a deal with Washington to its domestic constituencies by allowing it to claim that the arrangement protects Iran's "sovereign right" to have a full nuclear fuel cycle. One way to do this would be to draw a line between research on uranium enrichment (which would be allowed) and significant production of enriched uranium (which would be prohibited). In exchange, Tehran would have to accept verifiable safeguards on all its enrichment operations, permit throughout the country the more intrusive type of inspections required by the Additional Protocol of the IAEA, supply the IAEA with full documentation about suspected past violations, and freeze the construction of more centrifuges and heavy-water reactors that could produce plutonium.

History, particularly that of U.S.­North Korean relations, suggests that such agreements are just the start of serious negotiations. Even if a deal is struck, delays and backsliding should be expected. To limit their impact and keep them from leading to the agreement's dissolution, it would be necessary for Washington to both keep its promises and maintain credible threats that it would impose sanctions or even use limited force against Iran if Tehran violated its commitments.

Most important, however, would be a reduction in the security threat that the United States poses to Iran. Given the need for Washington to have a credible deterrent against, say, terrorist attacks sponsored by Iran, it would be ill advised to offer Tehran a blanket security guarantee. But more limited guarantees, such as a commitment not to use nuclear weapons and other commitments of the type offered North Korea under the Agreed Framework, could be effective today. They would reassure Tehran and pave the way toward the eventual normalization of U.S.-Iranian relations while signaling to other states that nuclear weapons are not the be all and end all of security. None of this will happen, however, if U.S. officials keep threatening to topple the Iranian government. In any final settlement, Tehran will need to agree to freeze its nuclear program and end its support for terrorism, and Washington -- along with China, Russia, and the EU-3 -- must issue a joint security guarantee that respects Iran's political sovereignty, thus committing the United States to promote democracy only by peaceful means. Peaceful coexistence does not require friendly relations, but it does mean exercising mutual restraint. Relinquishing the threat of regime change by force is a necessary and acceptable price for the United States to pay to stop Tehran from getting the bomb.