For nearly 50 years, worries about a nuclear Middle East centered on Israel. Arab leaders resented the fact that Israel was the only atomic power in the region, a resentment heightened by America’s tacit approval of the situation. But they were also pretty certain that Israel (which has never explicitly acknowledged having nuclear weapons) would not drop the bomb except as a very last resort. That is why Egypt and Syria were unafraid to attack Israel during the October 1973 Yom Kippur War. “Israel will not be the first country in the region to use nuclear weapons,” went the Israelis’ coy formula. “Nor will it be the second.”

Today the nuclear game in the region has changed. When the Arab League’s secretary general, Amr Moussa, called for “a Middle East free of nuclear weapons” this past May, it wasn’t Israel that prompted his remarks. He was worried about Iran, whose self-declared ambition to become a nuclear power has been steadily approaching realization.

The anti-Israel statements of the Iranian president, Mahmoud Ahmadinejad, coupled with Iran’s support for Hezbollah and Hamas, might lead you to think that the Arab states would welcome Iran’s nuclear program. After all, the call to wipe the Zionist regime from the map is a longstanding cliché of Arab nationalist rhetoric. But the interests of Shiite non-Arab Iran do not always coincide with those of Arab leaders. A nuclear Iran means, at the very least, a realignment of power dynamics in the Persian Gulf. It could potentially mean much more: a historic shift in the position of the long-subordinated Shiite minority relative to the power and prestige of the Sunni majority, which traditionally dominated the Muslim world. Many Arab Sunnis fear that the moment is ripe for a Shiite rise. Iraq’s Shiite majority has been asserting the right to govern, and the lesson has not been lost on the Shiite majority in Bahrain and the large minorities in Lebanon and Saudi Arabia. King Abdullah of Jordan has warned of a “Shiite crescent” of power stretching from Iran to Lebanon via Iraq and (by proxy) Syria.

But geopolitics is not the only reason Sunni Arab leaders are rattled by the prospect of a nuclear Iran. They also seem to be worried that the Iranians might actually use nuclear weapons if they get them. A nuclear attack on Israel would engulf the whole region. But that is not the only danger: Sunnis in Saudi Arabia and elsewhere fear that the Iranians might just use a nuclear bomb against them. Even as Iran’s defiance of the United States and Israel wins support among some Sunnis, extremist Sunnis have been engaging in the act of takfir, condemning all Shiites as infidels. On the ground in Iraq, Sunni takfiris are putting this theory into practice, aiming at Shiite civilians and killing them indiscriminately. Shiite militias have been responding in kind, and massacres of Sunni civilians are no longer isolated events.

Adding the nuclear ingredient to this volatile mix will certainly produce an arms race. If Iran is going to get the bomb, its neighbors will have no choice but to keep up. North Korea, now protected by its own bomb, has threatened proliferation — and in the Middle East it would find a number of willing buyers. Small principalities with huge U.S. Air Force bases, like Qatar, might choose to rely on an American protective umbrella. But Saudi Arabia, which has always seen Iran as a threatening competitor, will not be willing to place its nuclear security entirely in American hands. Once the Saudis are in the hunt, Egypt will need nuclear weapons to keep it from becoming irrelevant to the regional power balance — and sure enough, last month Gamal Mubarak, President Mubarak’s son and Egypt’s heir apparent, very publicly announced that Egypt should pursue a nuclear program.

Given the increasing instability of the Middle East, nuclear proliferation there is more worrisome than almost anywhere else on earth. As nuclear technology spreads, terrorists will enjoy increasing odds of getting their hands on nuclear weapons. States — including North Korea — might sell bombs or give them to favored proxy allies, the way Iran gave Hezbollah medium-range rockets that Hezbollah used this summer during its war with Israel. Bombing through an intermediary has its advantages: deniability is, after all, the name of the game for a government trying to avoid nuclear retaliation.

Proliferation could also happen in other ways. Imagine a succession crisis in which the Saudi government fragments and control over nuclear weapons, should the Saudis have acquired them, falls into the hands of Saudi elites who are sympathetic to Osama bin Laden, or at least to his ideas. Or Al Qaeda itself could purchase ready-made bombs, a feat technically much less difficult than designing nuclear weapons from scratch. So far, there are few nuclear powers from whom such bombs can be directly bought: as of today, only nine nations in the world belong to the nuclear club. But as more countries get the bomb, tracing the seller will become harder and harder, and the incentive to make a sale will increase.

The prospect of not just one Islamic bomb, but many, inevitably concentrates the mind on how Muslims — whether Shiite or Sunni — might use their nuclear weapons. In the mid-1980’s, when Pakistan became the first Islamic state to go nuclear, it was still possible to avoid asking the awkward question of whether there was something distinctive about Islamic belief or practice that made possession of nuclear technology especially worrisome. Most observers assumed that Islamic states could be deterred from using nuclear force just like other states: by the threat of massive retaliation.

During the last two decades, however, there has been a profound change in the way violence is discussed and deployed in the Muslim world. In particular, we have encountered the rise of suicide bombing. In historic terms, this development is new and unexpected. Suicide bombing has no traditional basis in Islam. As a technique, it was totally absent from the successful Afghan jihad against the Soviet Union. Although suicide bombing as a tool of stateless terrorists was dreamed up a hundred years ago by the European anarchists immortalized in Joseph Conrad’s “Secret Agent,” it became a tool of modern terrorist warfare only in 1983, when Shiite militants blew up the U.S. Marine barracks in Lebanon.

Since then, suicide bombing has spread through the Muslim world with astonishing speed and on a surprising course. The vocabulary of martyrdom and sacrifice, the formal videotaped preconfession of faith, the technological tinkering to increase deadliness — all are now instantly recognizable to every Muslim. And as suicide bombing has penetrated Islamic cultural consciousness, its list of targets has steadily expanded. First the targets were American soldiers, then mostly Israelis, including women and children. From Lebanon and Israel, the technique of suicide bombing moved to Iraq, where the targets have included mosques and shrines, and the intended victims have mostly been Shiite Iraqis. The newest testing ground is Afghanistan, where both the perpetrators and the targets are orthodox Sunni Muslims. Not long ago, a bombing in Lashkar Gah, the capital of Helmand Province, killed Muslims, including women, who were applying to go on pilgrimage to Mecca. Overall, the trend is definitively in the direction of Muslim-on-Muslim violence. By a conservative accounting, more than three times as many Iraqis have been killed by suicide bombings in the last 3 years as have Israelis in the last 10. Suicide bombing has become the archetype of Muslim violence — not just to frightened Westerners but also to Muslims themselves.

What makes suicide bombing especially relevant to the nuclear question is that, by design, it unsettles the theory of deterrence. When the suicide bomber dies in an attack, he means to send the message “You cannot stop me, because I am already willing to die.” To make the challenge to deterrence even more stark, a suicide bomber who blows up a market or a funeral gathering in Iraq or Afghanistan is willing to kill innocent bystanders, including fellow Muslims. According to the prevailing ideology of suicide bombing, these victims are subjected to an involuntary martyrdom that is no less glorious for being unintentional.

So far, the nonstate actors who favor suicide bombing have limited their collateral damage to those standing in the way of their own bombs. But the logic of sacrificing other Muslims against their own wills could be extended to the national level. If an Islamic state or Islamic terrorists used nuclear weapons against Israel, the United States or other Western targets, like London or Madrid, the guaranteed retaliation would cost the lives of thousands and maybe millions of Muslims. But following the logic of suicide bombing, the original bomber might reason that those Muslims would die in God’s grace and that others would live on to fight the jihad. No state in the Muslim world has openly embraced such a view. But after 9/11, we can no longer treat the possibility as fanciful.

Raising the question of Islamic belief and the bomb, however, is not a substitute for strategic analysis of the rational interests of Islamic governments. Like other states, Islamic states act on the basis of ordinary power politics as much as or more than on the basis of religious motivation. Pakistan, which tested a series of warheads in 1998, at the height of tensions with India, has not used its atomic power as a tool of the faithful in a global jihad. The proliferation operation spearheaded by the nuclear scientist — and sometime Pakistani national hero — Dr. Abdul Qadeer Khan appears to have been based on a combination of national interest and greed, not on religious fervor. Khan found buyers in Iran and Libya, but also in decidedly non-Islamic North Korea. (In a twist much stranger than fiction, Saddam Hussein apparently turned down the offer.)

Some observers think that Iran, too, wants the bomb primarily to improve its regional position and protect itself against regime change — not to annihilate Israel. According to this view, Iran’s nuclear push reflects a drive to what is sometimes called national greatness and might more accurately be defined as the ability of a country to thumb its nose at the United States without fear of major repercussions. A televised pageant hastily arranged to celebrate Iran’s atomic program in April of this year featured traditional Persian dancing and colorful local garb intermixed with make-believe vials of enriched uranium. To an Iranian audience accustomed to decoding official symbols, these references were nationalist, not pan-Islamic. (They were also subtly subversive of the mullahs: singing and dancing are not favored forms of expression in the clerical enclave of Qom.)

But at the same time, Ahmadinejad has emphasized Iran’s pan-Islamic aspirations to act on behalf of Muslims everywhere. An emerging nuclear power needs friends. Right now Iran wants to reduce, not promote, division between Sunnis and Shiites — and promoting broader “Islamic” interests by going after Israel is one way to lessen Sunni fears about Iran’s rise. Ahmadinejad has put his money where his mouth is, providing Hezbollah with medium-range missiles — though apparently not chemical warheads — to use against Israel. The nationalist language he has sometimes used at home may be a cover for sincerely held pan-Islamic ends — a version of the old revolutionary strategy of making nationalist claims in order to attract the support of those fellow Iranians who do not respond well to Islamist ideology. That it is convenient for Iran to emphasize Islamic unity does not mean that at least some of its leaders do not believe in it as a motivating goal. 

It is common among foreign-policy realists to suppose that a country acting on nationalist motives is easier to deter than a country moved by religious ones. There is no especially strong evidence for this assumption — plenty of nationalist regimes have done crazy things when they logically should have been deterred — but the claim has a common-sense ring to it. Nationalists care about peoples and states, which need to be alive to prosper. It is a basic tenet of nationalism that there is nothing higher than the nation-state itself, the pinnacle of a people’s self-expression. Religious thinkers, on the other hand, believe almost by definition that there is something in heaven greater than government here on earth. Under the right circumstances, they might sacrifice lives — including their own — to serve the divine will as they interpret it.

We urgently need to know, then, what Islam says about the bomb. Of course there is no single answer to this question. The world’s billion-plus Muslims differ regarding many aspects of their 1,400-year-old religious tradition. Furthermore, nuclear weapons are a relatively new technology, unforeseen by the Prophet and unmentioned in the Koran. Nevertheless, contemporary Muslims are engaged in interpreting their tradition to ascertain how and when nuclear power may be used. Their writings, contained in fatwas and treatises that can be found on the Web and in print, tell a fascinating and disturbing story.

The Islamic discussion of nuclear weapons is profoundly intertwined with a parallel discussion of suicide bombing that is also taking place in the Muslim world. Suicide bombing and nuclear weapons typically kill without discrimination, murdering soldiers or civilians, men or women or children. And using nuclear force against another nuclear power can be suicidal, in the broad sense that retaliation may destroy the nation that attacked first. Beyond these commonalities is the fact that the rise of suicide bombing is driving a historic reconsideration of what might be called the Islamic ethics of violence. To consider Islam and the bomb today must thus inevitably draw us into the complex legal and political thinking of those Muslim authorities who justify the use of force.

The story starts with traditional Islamic law. The Shariah never followed the Roman adage that in war the laws are silent. Because jihad is a pillar of Islam, and because in Islam God’s word takes legal form, the classical scholars devoted considerable care to identifying the laws of jihad. In common with the just-war doctrine developed in Christian Europe, the law of jihad governed when it was permissible to fight and what means could lawfully be adopted once warfare had begun. There were basic ground rules about who was fair game. “A woman was found killed in one of the battles fought by the Messenger of God,” runs a report about the Prophet Muhammad considered reliable and binding by the Muslim scholars. “So the Messenger of God forbade the killing of women and children.” This report was universally understood to prohibit the deliberate killing of noncombatant women and children. Some scholars interpreted it to mean that anyone incapable of warfare should be protected and so extended the ban to the elderly, the infirm and even male peasants, who as a rule did not fight. Muslims living among the enemy were also out of bounds. These rather progressive principles were broadly accepted by the Islamic legal authorities, Sunni and Shiite alike. For well over a thousand years, no one seriously questioned them.

Such black-and-white rules were well suited to the hand-to-hand or horse-to-horse combat characteristic of limited medieval wars. A few quirky challenges did arise, and the Muslim lawyers had to deal with them. The great theologian and jurist al-Ghazali, who wrote in the 11th and 12th centuries and was widely noted for his revival of religious piety and his skepticism of secular philosophy, dealt with the problem of human shields. He ruled that if the enemy drove captured Muslims before him, the Muslim army could still fight back, even if it might mean killing some of those Muslims. The reason he gave was that “we know that the law intends minimizing killing.” There was also the catapult — precursor of artillery and air power — which was capable of sending a burning projectile into a populated city, where the resulting fire might kill women or children. Authorities differed on whether that tactic was permissible. Some disallowed the catapult when children or Muslim captives were in the city. In support, they cited a verse from the Koran that reads, “Had they been separated clearly, then We would have chastised the unbelievers among them with a painful chastisement.” According to this school of thought, the “separation” of permissible targets (i.e., non-Muslim men) from impermissible targets is the precondition for a general attack. Another school of thought, by contrast, permitted the use of the catapult regardless of collateral damage in order to serve the general interest of the Muslims.

No law can exist for a millennium without being broken, and there are scattered historical reports, mostly from Christian chroniclers, of Muslim forces acting outside the bounds of lawful jihad, without the authorization of the scholars. Men were always considered legitimate targets, and Muslim armies sometimes slaughtered them just as Muslims could be slaughtered by their enemies. Remarkably enough, though, the legal principles of jihad protecting women, children and fellow Muslims survived well into the modern era, when the secular regimes of the Muslim world began to fight according to secular ideas. The World War I Armenian genocide, which took place in the last, secularizing gasp of the declining Ottoman Empire, was the first really substantial systematic violation of the ban on killing women and children in recorded Islamic history. In the bloody 20th century, when mass exterminations took place in Europe, Africa and Asia, Muslim states had a relatively better record, marred of course by Saddam Hussein’s gassing of the Kurds. And there have been the genocidal killings in Darfur in this new century. Even these horrific events, however, were not dignified by the claim that they were permitted under the law of jihad.

The last two decades have seen a challenge to this Islamic tradition of warfare under law, a challenge driven mostly by the attempt to justify suicide bombing despite its evident inconsistency with Islamic tradition. On the subject of suicide, the Koran could hardly be clearer: “Do not kill yourselves; for surely God has been merciful to you.” Faced with this explicit text, the solution of the militant Islamist ideologues has been to avoid the category of suicide altogether and to treat the bomber as a martyr rather than as one who has taken his own life. This interpretation is not very convincing in historical terms: martyrdom classically meant that another person killed the Muslim warrior, not that he pushed the button himself. Nevertheless, many Muslims now seem to find the argument convincing. Even among rather secular Muslims, it has become standard to refer to suicide bombers as martyrs.

The killing of women, children and Muslim men, however, has proved harder to explain away as a permissible exercise of jihad. The reaction to 9/11, which has (so far) been the high-water mark of suicide bombing, illustrates the nature of the difficulty of reconciling suicide bombing with Islamic law. One problem concerns the offensive nature of the attack at a time when the United States was not at war with any Muslim entity. Offensive jihad requires the authorization of a legitimate Muslim leader, absent on 9/11. A more serious concern was the obvious reality that the 9/11 attacks were certain to kill — and did kill — women, children and Muslims, all in direct contravention of classical jihad principles. Since the whole point of 9/11 was to announce and embody jihad on the international stage, the attacks quickly became the centerpiece of a high-stakes debate about whether they did or did not qualify as legitimate acts of jihad.

In the immediate aftermath of 9/11, it was sometimes asserted in the West that there were no Muslim voices condemning the attacks. This was never true. Prominent Muslim scholars expressed their disapprobation in public arenas like television and the Internet. These included senior Sunni scholars like the grand mufti of Saudi Arabia and the head of Al-Azhar, in Egypt, nominally the flagship institution of Sunni higher learning — who gave a news conference. More popular figures, like Al Jazeera’s resident cleric, Sheik Yusuf al-Qaradawi, explained that Islam “considers the attack on innocent human beings a grave sin.” Shiite scholars also spoke out, including Ayatollah Ali Khamenei, the supreme leader of Iran.

The position of the Muslim scholars and observers who condemned the 9/11 attacks was simple and consistent across the Sunni-Shiite divide: this was not jihad but an unlawful use of violence. Offensive jihad was prohibited in the absence of formal authorization by a Muslim leader. But even if the attacks could somehow be construed as defensive, the perpetrators of 9/11 broke the rules with their willingness to kill women and children. In confident and insistent tones, these critics cited the classical scholars and insisted that nothing in Islamic law could justify the tactics used by Al Qaeda. Ayatollah Muhammad Husayn Fadlallah, the Lebanese cleric whose spiritual authority is recognized by Hezbollah, gave an interview to the Beirut newspaper Al Safir in which he asserted that given their impermissible choice of targets, the 9/11 bombers were not martyrs but “merely suicides.”

At the same time, it is important to note that in 2001 few prominent Muslim scholars — the Saudi grand mufti was the main exception — condemned the use of suicide bombings in all circumstances. Fadlallah approved the attack on the U.S. Marines in 1983 and, according to the United States, played a role in ordering it. Qaradawi, whose television presence gives him reason to stay within the Islamist mainstream, distinguished the 9/11 attacks from the permissible defensive jihad of the Palestinians. He was happy to praise a God who “through his infinite wisdom ... has given the weak a weapon the strong do not have, and that is their ability to turn their bodies into bombs as Palestinians do.” Qaradawi has also repeated the common view that the killing of Israeli women is justified on the grounds that all Israelis must serve in the military, and so no Israeli is a true noncombatant: “An Israeli woman is not like women in our societies, because she is a soldier.”

The equivocation by Muslim scholars with respect to the technique of suicide bombing reflected the reality that throughout the Muslim world, Palestinian suicide bombers were by 2001 identified as martyrs dying in a just cause. This, in turn, was the natural outgrowth of the decades before suicide bombing, when Palestinian terrorists were applauded for killing Israeli civilians, including women and children. Given that embracing Palestinian suicide bombing had become a widespread social norm, it would have been essentially unthinkable for an important Muslim scholar to condemn the practice without losing his standing among Muslims worldwide. In the Islamic world, as in the U.S. Supreme Court, the legal authorities cannot get too far away from their public constituency without paying a price.

What happened, in other words, is that without the scholars paying too much attention to the question, the killing of Israeli women and children had become a kind of exception to the ordinary laws of jihad. Opportunists like bin Laden then began to widen the loophole to include new victims. With respect to the unauthorized nature of his offensive jihad, bin Laden asserted that in fact the attacks were defensive, since in his mind the U.S. was occupying the sacred soil of Saudi Arabia — just as Israel was occupying the Muslim land of Palestine. Once all of Saudi Arabia was placed on a par with the holy cities of Mecca and Medina, traditionally closed to non-Muslims, the presence of American soldiers anywhere on the Arabian Peninsula (even if their presence was with the permission of the Saudi government) could be depicted as a profanation, a violation of the Prophet’s deathbed directive to “banish the pagans from the Arabian Peninsula.”

Bin Laden was embroidering on the theories of his onetime mentor Abdullah Azzam, the intellectual godfather of Al Qaeda. Azzam was a Palestinian Islamist who made his way to Afghanistan via Saudi Arabia and established the so-called Bureau of Services to channel Arab youth into the Afghan jihad. As Azzam trod his personal path from Palestinian militancy to universal pan-Islamic jihadism, he wrote an influential treatise called “Defense of Muslim Lands.” In it, Azzam argued that not a single hand span of Muslim territory anywhere could ever be ceded to the enemy “because the land belongs to Allah and to Islam.” Though Azzam would never have acknowledged it, his account of the divine ownership of Muslim lands was probably influenced — unconsciously, to be sure — by religious-Zionist claims about the holiness of the Land of Israel.

When it came to the killing of civilians, bin Laden’s thought developed more gradually. In early pronouncements, before 9/11, he spoke as if the killing of women and children was inherently an atrocity. “Nor should one forget,” he admonished an interviewer in 1996, “the deliberate, premeditated dropping of the H bombs [sic] on cities with their entire populations of children, elderly and women, as was the case with Hiroshima and Nagasaki.” After 9/11, however, the argument changed. Now bin Laden began to suggest that American civilians were fair game. He could not argue that like Israelis, all Americans were subject to mandatory military service. Instead he proposed that because “the American people are the ones who choose their government by their own free will,” and because they “have the ability and choice to refuse the policies of their government,” attacks on American civilians were justified. Voting was now playing the role for Americans that military service played in the case of Israelis: the active step transforming civilians into fair game.

Such an appeal to collective responsibility was, however, pretty weak in Islamic legal terms. It might suffice for bin Laden’s videotaped self-justifications, and it might salve the consciences of potential jihadis hoping to join the rank and file of Al Qaeda. But it would never satisfy serious students of classical Islamic law, who found the 9/11 attacks problematic from an Islamic legal perspective.

In Saudi Arabia in particular, radical Muslim scholars with much more learning than bin Laden have sought to develop legally persuasive justifications for civilian killings. Probably the most sophisticated effort from a legal standpoint is a document titled “A Treatise on the Law of the Use of Weapons of Mass Destruction Against the Unbelievers,” written in 2003 by a brilliant Saudi dissident named Sheik Nasir bin Hamad al-Fahd. (Fahd, a theorist rather than an activist, is currently back in prison, as he has been off and on for almost a decade.) The treatise begins with the assumption that the world’s Muslims are under attack. But how are today’s Muslims supposed to defend themselves, given their military inferiority? Fahd’s response is that, if they have no other choice, they may use any means necessary — including methods that would otherwise violate the laws of jihad. “If the unbelievers can be repelled . . . only by using” weapons of mass destruction, then “their use is permissible, even if you kill them without exception.”

Lest his argument prove too much, Fahd tempers it by the claim that the Muslims fighting the jihad may not inflict disproportionately more harm on the enemy than the enemy has inflicted on them. That raises the question of the extent of American guilt. “Some Brothers have added up the number of Muslims killed directly or indirectly by [American] weapons and come up with a figure of nearly ten million,” the treatise states. This total, Fahd concludes, would authorize the use of weapons of mass destruction to kill 10 million Americans: indeed, “it would be permissible with no need for further [legal] argument.” (The number is never explained or analyzed, and you might assume that it was meant to correspond very roughly to the population of New York.)

Fahd’s arguments sit uneasily with the classical Islamic discussions of the laws of jihad. The classical Islamic law never explicitly says that women and children may be intentional targets if it is the only way to win the jihad. It does not allow violations of the law just because the enemy has broken the rules or killed many Muslims. So the treatise must fall back on whatever evidence it can muster from the classical sources that seems to modify the basic rules. The catapult rears its head and is cited as precedent for nonspecific killing. The right to fight even when Muslim hostages may be killed is brought out as proof of the permissibility of collateral damage when there is no other choice.

The legal arguments in use here are stronger than bin Laden’s makeweights, but they, too, would probably not be sufficient on their own to justify the deviation from the legal traditions of jihad wrought by today’s jihadis. The notion that it’s right because it’s necessary is doing the real work, and old-fashioned legal arguments are following along. It is no accident that the argument from necessity has been so prominent in modern Western writing about modern warfare in general and the nuclear bomb in particular. If the technology of mass destruction can be exported, why not the justification that comes with it?

Within the world of radical Islam, there are those who believe that the erosion of the laws of jihad has gone too far. There are reports of difficulty recruiting foreign candidates for suicide missions directed at Iraqi civilians. The debate about how jihad may be prosecuted is not over by any means. But it is an unavoidable fact that the classic restrictions on the killing of women, children and Muslims in jihad have been deeply undermined in the last decade.

If the Islamic laws of war are under revision, or at least the subject of intense debate, what does that mean for the question of the Islamic bomb? The answer is that the expanding religious sanction for violence once thought unacceptable opens the way for new kinds of violence to be introduced and seen as legitimate in turn. First Israeli women and children became acceptable targets; then Americans; then Shiites; and now Sunnis of unstinting orthodoxy. It would seem that no one is out of bounds.

It is therefore now possible to imagine that the classical Islamic principles governing war would not be applied even by a self-consciously Islamic regime deciding when and if to detonate a nuclear device. The traditional ban on killing women, children and fellow Muslims would have gone a long way toward banning most potential uses of nuclear power by a sincerely Islamic state actor. As those prohibitions have eroded, the reassurance that might be afforded by a state’s Islamic commitments has waned.

This means that a nuclear Islamic state would be at least as willing to use its weapons as a comparable non-Islamic state. But would an Islamic state be prepared to take the jihad to the enemy even if it would result in what amounts to collective suicide through the destruction of the state and its citizens? If the leaders of Iran or some future leaders of a radicalized, nuclear Saudi Arabia shared the aspiration to martyrdom of so many young jihadis around the world, might they be prepared to attack Israel or the United States, even if the inevitable result were the martyrdom of their entire people?

The answer depends to a large degree on whether you consider Islam susceptible to the kind of apocalyptic, millennial thought that might lead whole peoples, rather than just individuals, into suicidal behavior. It is important to note that for all his talk of the war between civilizations, bin Laden has never spoken of the end of days. For him, the battle between the Muslims and the infidels is part of earthly human life, and has indeed been with us since the days of the Prophet himself. The war intensifies and lessens with time, but it is not something that occurs out of time or with the expectation that time itself will stop. Bin Laden and his sympathizers want to re-establish the caliphate and rule the Muslim world, but unlike some earlier revivalist movements within Sunni Islam, they do not declare their leader as the mahdi, or guided one, whose appearance will usher in a golden age of justice and peace to be followed by the Day of Judgment.

From this perspective, the utter destruction of civilization would be a mistake, not the fulfillment of the divine plan. Even the most radical Sunni theorists of jihad invoke a passage from the Koran according to which civilization itself — “the crops and the cattle” — must not and cannot be destroyed completely. Bin Laden might seem to have few qualms about killing millions of Americans or other Westerners. He might well use a nuclear device if he gambled that there would be no enemy for the United States to bomb in retaliation. But even he might not be prepared to unleash a global nuclear conflagration on the expectation that a better order would emerge once many millions of Muslims and infidels died. (Bin Laden has called for Muslims to acquire nuclear weapons, and in the 1990’s reportedly tried to acquire them himself — but there is little hard evidence that he has made subsequent efforts in that direction.)

With respect to Shiite eschatology, there is greater reason for concern. Iran’s Shiism is of the “Twelver” variety, so called because the 12th imam in the line of succession from the Prophet disappeared into a state of occultation — or being hidden — from which he is expected to return as the mahdi. Ayatollah Khomeini played on the messianic overtones of this belief during the Iranian revolution, in which some of his followers went so far as to hint that he might be the returning imam. Moktada al-Sadr’s Shiite militia in Iraq is called Army of the Mahdi. Recently, Iran’s president, Mahmoud Ahmadinejad, contributed to renewed focus on the mahdi, by saying publicly that the mission of the Islamic revolution in Iran is to pave the way for the mahdi’s return, and by visiting the mosque at Jamkaran, on the outskirts of Qom, where, according to one tradition, the vanished imam was last seen. Some reports suggest that youth religion in Iran in increasingly focused on veneration of the vanished imam.

Islam has a vision of the end of days, with wars between the faithful and the tribes of Gog and Magog (Yuj and Majuj in their Arabic incarnation). Twelver Shiism is, at its core, an eschatological faith, focused on the ultimate return of the imam-mahdi, who will restore the Shiites to their rightful place and redeem their generations of suffering. Since the vanished imam is by tradition a human who has never died, but remains in occultation, he is also believed to affect the course of events even from his hidden place. And Shiite tradition fills in the picture of the mahdi’s return with an elaborate account of signs that will herald the event, including advance messengers, earthquakes and bloodshed.

But belief in redemption — even accompanied by wars and death and the defeat of the infidels — need not translate into a present impulse to create a violent crisis that would precipitate the messianic situation. Like their Jewish counterparts, Shiite religious authorities have traditionally sought to resist speculation about the imminence of a messianic return. Shiite messianic thought is less focused than its messianic Christian counterpart on generating global crisis and letting God sort things out. Khomeini himself believed that the mahdi’s advent could be hastened — but by social justice, not by provoking war. This put him on the activist side of Shiite teaching about the mahdi, much as he was also an activist about the exercise of worldly power by the mullahs. A popular revolutionary slogan urged the imam’s coming but asserted that Khomeini would govern alongside him.

Other Shiite thinkers, by contrast, take a more fatalist stance, and prefer to believe that the mahdi’s coming cannot be hastened by human activity — a view that corresponds loosely to Ayatollah Ali al-Sistani’s belief, with regard to Iraq and elsewhere, that the clerics should not themselves govern. One small, semi-secret Iranian organization, the Hojjatiya Society, was banned and persecuted by Khomeini’s government in part for its quiescent view that the mahdi’s arrival could not be hastened.

Ahmadinejad is not the only or even the most important player in Iran’s nuclear game. The supreme leader, Ayatollah Khamenei, still makes the ultimate decisions on armaments and other matters, and there are numerous factions in the country with opposed interests and ideology and goals. Nevertheless, Ahmadinejad has in some respects succeeded in making the nuclear issue his own, and as a result his personal views about the end of days have been the subject of much speculation and innuendo, inside Iran and out. The Mideast scholar Bernard Lewis, in a recent Wall Street Journal column, hinted darkly and without much evidence that Ahmadinejad might be planning a nuclear attack on Israel for the Night of Power (this year it fell on Aug. 22), when the Prophet Muhammad made his mystical journey to the Furthest Mosque, associated in tradition with al-Aqsa in Jerusalem. Rumors, possibly spread by Ahmadinejad’s enemies, have tied him to the outlawed Hojjatiya — a link mistakenly interpreted outside Iran as evidence that he might want to bring back the imam by violence, rather than that he might prefer to wait piously and prepare for the imam’s eventual return on his own schedule. It is of course impossible to gauge the man’s religious sensibilities perfectly. Yet the relative absence of a contemporary Shiite trend to messianic brinkmanship suggests that Ahmadinejad’s recent emphasis on the mahdi may be interpreted more in terms of an attempt to summon Khomeini’s legacy and Iran’s revolutionary moment than as a desperate willingness to bring the nation to the edge of war. When Ahmadinejad invoked the mahdi in his now-famous letter to George Bush, he seemed to be using the doctrine in ecumenical terms, emphasizing the Islamic tradition that Jesus — revered as a prophet, though not as the Son of God — will return alongside the mahdi and govern in tandem with him.

So although a renewed Shiite messianism does create some cause for concern about the potential uses of an Iranian bomb — in particular because it suggests that Ahmadinejad may be more a utopian than a realist — it is almost certainly a mistake to anticipate that Iran would use its nuclear power in a way that would provoke large-scale retaliation and assured self-destruction. Iranian leaders have been more than ready to sacrifice their own citizens in large numbers. During the Iran-Iraq war, major efforts went into recruiting young boys to the Basij militias, which were then sent to the front lines on what were essentially suicide missions. Religion played the central part in motivating the teenage soldiers, and it is reasonable to believe that religion helped salve the consciences of those who ordered these children into battle. Yet even this discounting of the value of human life — in a war started by Saddam Hussein, not by Iran — fell short of voluntarily putting an entire nation at risk. Ahmadinejad surely understands the consequences of using a nuclear bomb, and Shiite Islam, even in its messianic incarnation, still falls short of inviting nuclear retaliation and engendering collective suicide.

These worries about an Islamic bomb raise the question of why we trust any nation with the power that a nuclear capacity confers. Why, for instance, do we trust ourselves, given that we remain the only nation actually to have used nuclear weapons? The standard answer to why we keep our nuclear bombs — a response developed during the cold war — is that we must have the capability to deter anyone who might attack us first. The promise of mutually assured destruction was its own kind of collective suicide pact, albeit one supposed to scare both sides out of pushing the button. That is why, throughout the heyday of the unilateral disarmament movement, critics of this justification pointed out that our threat was only credible if we were, in fact, prepared to kill millions of civilians in a rapid act of retaliation. If this kind of killing was morally unjustified, went their argument, then the threat to use it was also immoral.

The truth is that we hold on to our nuclear capability not only as a matter of deterrence but also to maintain our own global strategic position. If we do not want Islamic states — or anyone else for that matter — to have a nuclear capability, it is not necessarily because we consider them especially likely to bring on their own destruction by using it. It is, rather, that we do not want to cede some substantial chunk of our own global power to them. This principle — if it is a principle — lies behind the general strategy that is embedded in the international nuclear-nonproliferation treaty. Everybody involved understands that if any government got a chance to acquire nuclear power before the other treaty members had a chance to notice and impose sanctions, it would jump at the opportunity.

So the nonproliferation regime is not and could never be based on some principle of international fairness. But it does not follow that the United States and its allies should simply accept the development of nuclear technology by just anyone. It should be relevant to our deliberations that a particular candidate is our enemy. When it comes to Islamic states, there is serious reason to worry that, both now and in the immediately foreseeable future, popular anti-American sentiment is especially likely to play an important role in the shaping of foreign policy. Over the next quarter-century, it is conceivable and certainly desirable that Islamism and anti-Americanism may be unlinked. But we must be honest and acknowledge that in the short term at least, the U.S. democratization strategy has done almost nothing to reduce Islamist anti-Americanism, whether Shiite or Sunni — this despite the fact that the same strategy has benefited Islamists across the region by allowing them to run for office and enter government.

Much of the reason for this close linkage between Islamism and anti-Americanism comes from Iran. As an enemy of the United States, which has worked consistently against American interests, Iran is in a category by itself, most nearly matched by North Korea, the other still-standing member of President Bush’s axis of evil. In this, Iran’s motives have been primarily Islamic-ideological, not pragmatic.

For many years under the shah, Iran was a natural American ally — precisely because it was Shiite and non-Arab, and uncomfortably close to the Soviet Union and its fantasy of a warm-water port. Even after the 1979 revolution and the hostage crisis, it is possible that the United States would have eventually reopened relations with an avowedly Islamic Iran had the government softened its anti-Americanism. The United States has never made secularism a condition of friendship. It has been fully prepared to support Islamic states like Saudi Arabia, and even used religion to cement the anti-Communist alliance during the cold war. The Iraqi Shiite Islamists have been willing to work alongside the Americans, and the United States has in return treated them as its allies, democratically chosen by the Iraqi electorate.

Islamist anti-Americanism is the direct legacy of Ayatollah Khomeini’s success in marrying Islamic faith to anti-imperialism — making “Death to America” into a religious chant, not just a political slogan. Of course the United States was hardly blameless. It did everything it could to open itself to the imperialist charge, including, in Iran, backing the famous 1953 countercoup that removed from power Iran’s first democratically legitimate prime minister, Mohammed Mossadegh. Contemporary Islamists can also point to America’s continuing hypocritical support of regional authoritarian regimes.

Iranian-rooted Islamist anti-Americanism has worked far better than its designers might have imagined, spreading to Sunni Islamists who have little love to lose for Iran. The marriage of Islamism and anti-Americanism will probably be considered by history as the most significant consequence of the Iranian revolution. Anti-Americanism has become a staple of Islamist sermons and Web postings, an effective tool for drawing to the movement angry young people who might not naturally be drawn to religion. Bin Ladenism, in this sense, owes much to the Iranian revolution even though Al Qaeda was never Iran’s direct ally. United States support for Israel has always been an important part of the argument for Islamist anti-Americanism, but today it is by no means a necessary component. If U.S. support of Israel were to weaken, the American presence in Iraq and elsewhere in the gulf would easily substitute as a basis for hatred.

The United States therefore has strong reason to block its enemy Iran from acquiring nuclear weapons — not simply because Iran will seek to become a greater regional power, as any nation might do, but because the Islamic Republic of Iran as currently constituted is definitionally anti-American. There need not be a direct threat of Iranian first use against either the United States or Israel for this reason to weigh heavily. A nuclear Iran will be a stronger and more effective enemy in pursuing anti-American policies under the banner of Islam. That will not change until the Iranian state abandons either its Islamic identity or its association between Islam and anti-Americanism. Iran’s eagerness to acquire nuclear capacity need not be a result of a particularly Islamic motivation, but if and when Iran does have the bomb, its enhanced power and prestige will certainly be lent to policies that it conceives as promoting the Islamic interest.

Whether force, negotiation or some combination is the right path to take to keep Iran from going nuclear is of course a hugely important question. It turns on many uncertain facts, like the true progress of Iran’s nuclear program and how much it can be affected by air attack; Iran’s capacity and will to retaliate against an attack; whether there is any chance Iran would respond to negotiations; and the ability of the United States to withstand any retaliation while 150,000 U.S. troops are in Iraq. As we have recently learned in Iraq, it is not enough to think you have a good reason to go to war — you must also have a realistic understanding of the practical and moral costs of things going horribly wrong. Any choice, though, must be made against the backdrop of the reality that the Islamic government of Iran is not only unlikely to collapse soon — it is also very unlikely to become less anti-American in the near future.

The same, unfortunately, is true of the world’s Islamist movements, for whom anti-Americanism remains a rallying cry and a principle of belief. Perhaps the promotion of democracy in the region, pursued consistently by the United States over the long term, might someday allow the rise of leaders whose Islamism is tempered by the need to satisfy their constituents’ domestic needs — and who eschew anti-Americanism as wasteful and misguided. Iraq was the test case of whether this change could occur in the short term. But we failed to make the experiment work and gave Iraq’s Islamist politicians, Shiite and Sunni alike, ample grounds to continue the anti-American rhetoric that comes so easily to them. In the wake of our tragic mismanagement of Iraq, we are certainly a generation or more from any such unlinking of Islamism and anti-Americanism, if it is to occur at all. And Islamism itself shows no signs of being on the wane as a social or political force.

That means that the best we can hope for in nuclear Islamic states in the near term is a rational dictator like Pervez Musharraf of Pakistan, who sees his bread buttered on the side of an alliance with the West. Such rulers can be very strong and can bring stability, but we also know that their rule (or reign) promotes Islamist opposition, with its often violent overtones. When such rulers die or otherwise fall from power, the Islamists will be poised to use the international power conferred by nuclear weapons to pursue their own ends — ends for now overwhelmingly likely to be anti-American.

None of this is inherent in the structure of Islam itself. Islam contains a rich and multivocal set of traditions and ideas, susceptible to being used for good or ill, for restraint or destruction. This interpretive flexibility — equally characteristic of the other great world religions — does not rob Islam of its distinctiveness. An Islamic bomb would not be just the same as the nationalist bomb of a majority-Muslim state, nor would it be the same as a Christian bomb or a Jewish one. But its role in history will depend, ultimately, on the meaning Muslims give it, and the uses to which they put their faith and their capabilities. In confronting the possibility of the Islamic bomb, we — Muslims and non-Muslims alike — need to remember that Islam exists both as an ideal system of morals and values and as a force that motivates actual people living today, with all the frailties and imperfections that make us human.

Noah Feldman, a contributing writer, is a law professor at New York University and adjunct senior fellow at the Council on Foreign Relations.

Copyright 2006 The New York Times Company. Reprinted from The New York Times Magazine of Sunday, October 29, 2006.