This article is part of the Foreign Affairs Iraq Retrospective.


With war in the Middle East imminent, it is clear that the United States has painted itself -- as well as Iraq -- into a corner. The Bush administration's success in engineering international support for a preventive war in the Persian Gulf is impressive, both politically and diplomatically. But Washington's case rests on two crucial errors. It understates the very real risk that an assault on Iraq will trigger a counterattack on American civilians. And even when that risk is admitted, the pro-war camp conflates it with the threat of unprovoked attack by Iraq in the future.

Many Americans still take for granted that a war to topple Saddam Hussein can be fought as it was in 1991: on American terms. Even when they recognize that the blood price may prove greater than the optimists hope, most still assume it will be paid by the U.S. military or by people in the region. Until very late in the game, few Americans focused on the chance that the battlefield could extend back to their own homeland. Yet if a U.S. invasion succeeds, Saddam will have no reason to withhold his best parting shot -- which could be the use of weapons of mass destruction (WMD) inside the United States. Such an Iraqi attack on U.S. civilians could make the death toll from September 11 look small. But Washington has done little to prepare the country for this possibility and seems to have forgotten Bismarck's characterization of preventive war as "suicide from fear of death."

America's political leaders have not just lost faith in deterrence as a means to contain Iraq, they have also lost sight of the fact that, when it comes to a showdown between two countries that both possess WMD, deterrence can work both ways. The United States is about to poke a snake out of fear that the snake might strike sometime in the future, while virtually ignoring the danger that it may strike back when America pokes it. True, not everyone demanding an American attack ignores the immediate threat such an attack might raise -- but even this camp misreads that threat, thinking it reinforces the urgency of preventive war. The consequences, they argue, will only get worse if Washington waits. This argument may seem like common sense at first. But it dangerously confuses two sets of odds: the chance that Iraq will eventually challenge America even without being provoked, and the risk that Baghdad will retaliate against Washington if struck first.

The probability that Iraq could bring off a WMD attack on American soil may not be high, but even a modest probability warrants concern. By mistakenly conflating the immediate and long-term risks of Iraqi attack and by exaggerating the dangers in alternatives to war, the advocates of a preventive war against Saddam have miscast a modest probability of catastrophe as an acceptable risk.


An invasion to get rid of Saddam would represent an American attempt to do what no government has ever done before: destroy a regime that possesses WMD. Countries with WMD have fought each other twice before, but these events (when China and the Soviet Union came to blows on the Ussuri River in 1969, and when India and Pakistan fought over Kargil in 2000) were mere skirmishes. In both of those limited clashes, neither side's leadership was truly threatened. The opposite is true this time, and yet the difference has not been digested by pro-war strategists.

During Congress' debate over whether to authorize the war, for example, the danger that a preventive assault might provoke Iraqi retaliation against the American heartland went almost unmentioned. In an October letter, Director of Central Intelligence George Tenet stated that Saddam would be more likely to attempt a WMD attack against the United States as "his last chance to exact vengeance" if he believed he could no longer deter an American onslaught -- but this comment received scant notice. Attention focused instead on less immediate, less likely, and less dangerous threats. Hawks argued that Iraq will get nuclear weapons in the future. But the fact is that the biological weapons Iraq already has are dangerous enough to do tremendous damage -- even if the worst estimates of U.S. vulnerability are excessive.

A 1993 study by the Office of Technology Assessment concluded that one plane, delivering anthrax by aerosol under good weather conditions over the Washington, D.C., area, could kill between one million and three million people. That figure is probably far too pessimistic even for an efficiently executed attack, since among other things, the medical response would be quicker and more effective today than it would have been a decade ago. So discount this estimate by, say, 90 percent. Even then, fatalities could still exceed 100,000. This reduced figure may still be excessive, since clandestine Iraqi operations to infect U.S. cities might be crude and inefficient. Yet if you reduce the death toll by another 90 percent, fatalities would still be more than triple those of September 11. Multiple attacks, even clumsy ones, could yield tens of thousands of casualties. Worst of all, Iraq may have bioengineered new pathogens for which no defense is available. Chemical weapons, although less destructive than biological ones, could also exact a dramatic toll.

But is an Iraqi counterattack on U.S. soil really plausible? Hawks argue that Saddam must be eliminated because he may decide to use WMD in the future or give them to terrorists -- even if the United States threatens him with devastating retaliation. This argument assumes that Saddam would be prepared to cut his own throat without provocation. If that is true, it certainly follows that he will lash out with anything he has if Washington goes for his jugular and puts his back against the wall.

Yet Washington now seems determined to push him to that wall. Few are proposing that Saddam be retired to a villa on the Riviera next to "Baby Doc" Duvalier's. The option of a golden parachute should be considered, but it is unlikely to be accepted. Saddam would demand protection from extradition so that he could avoid joining Milosevic in court. And even Saddam knows he has too many bitter enemies to survive for long outside Iraq. Regime change in Baghdad, therefore, probably means an end to Saddam Hussein. And he will not go gently if he has nothing left to lose. If a military assault to overthrow the Iraqi regime looks likely to succeed, there is no reason to doubt Saddam will try to use biological weapons where they would hurt Americans the most.

Instead of considering the chances of a strike on the American heartland, however, war planners have tended to focus on the vulnerability of U.S. invasion forces, or on local supporters such as Israel, Saudi Arabia, and Kuwait -- as if they are the only likely targets of an Iraqi WMD attack. Awful as attacks on these targets would be, the consequences would be nowhere near as large from the American perspective as those of a strike on the United States itself. The only remaining question, then, is whether Saddam would have the capability to carry out such an attack.

Maybe he won't. Saddam may not be crafty enough to figure out how to strike the American homeland. Iraqi intelligence may be too incompetent to smuggle biological weapons into the United States and set them off. Or Saddam's underlings might disobey orders to do so. The terrorists to whom Iraq subcontracts the job might bungle it. Or perhaps American forces could find and neutralize all of Iraq's WMD before they could be detonated. But it would be reckless to bank on maybes. Washington has given Saddam more than enough time to concoct retaliation, since he has had months of notice that the Americans are coming. The Bush administration has made this war the most telegraphed punch in military history.

Is it alarmist to emphasize the danger of an Iraqi counterattack on American soil? The odds may be low -- perhaps as low as the odds were on September 10, 2001, that 19 Arab civilians would level the World Trade Center and tear a chunk out of the Pentagon. Even if the odds are as high as one out of six, however, that makes the risks inherent in overthrowing Saddam look like Russian roulette. It would be one thing for Americans to hope that they can wage war without triggering effective retaliation. But it would be altogether different to blithely assume that outcome; such unwarranted optimism represents the kind of "best case" planning that should shame any self-respecting hawk.

Taking the threat of retaliation seriously means two big things: preparing to cope with it, and reconsidering the need to start the war that could bring it on. If war on Iraq is deemed necessary despite the risk of mass destruction, Washington is dangerously far behind in preparing the home front. The United States must not wait until the war begins to put homeland defense into high gear. Studies and plans to prepare for future biological or chemical attack should be implemented in advance, not left on the drawing board until American tanks start rolling into Baghdad. The American people deserve immediate, loud, clear, and detailed instructions about how to know, what to do, where to go, and how to cope if they encounter anthrax, ricin, smallpox, vx, or other pathogens or chemicals Iraq might use against them. It is already too late now to do what should have been done much earlier -- to cut through the production problems and other complications in making anthrax vaccines available to civilians (much of the military has already been vaccinated). At least there should be a crash program to test and put in place mechanisms for detecting anthrax attacks promptly and dispensing antibiotics on a massive scale; these are the minimum steps the Bush administration should take before it pokes the snake. Smallpox is a less likely threat, and much planning has been done for mass vaccination in an emergency. But at a minimum, health-care workers should be immunized in advance. Until the U.S. government is ready to do all these things, it will not be ready to start a war.


Although it is already terribly late in the day, the risk of Iraqi retaliation also underlines the need to reconsider the alternative to provoking it. Why are containment and deterrence -- the strategies that worked for the four decades of the Cold War -- suddenly considered more dangerous than poking the snake? Proponents of war against Iraq have provided an answer -- but they are wrong.

Deterrence rests on the assumption that a rational actor will not take a step if the consequences of that action are guaranteed to be devastating to him. The United States can therefore deter Iraqi aggression unless or until Saddam deliberately chooses to bring on his own demise, when he could otherwise continue to survive, scheme, and hope for an opportunity to improve his hand. Of course, Saddam's record is so filled with rash mistakes that many now consider him undeterrable. But there is no good evidence to prove that is the case. Reckless as he has been, he has never yet done something Washington told him would be suicidal.

It is true that Saddam has a bad record of miscalculation and risk-taking. But he made his worst mistake precisely because Bush the Elder did not try to deter him. In fact, Washington effectively gave Baghdad a green light prior to its 1990 invasion of Kuwait. Ambassador April Glaspie was never instructed to warn Saddam that the United States would go to war if he grabbed Kuwait. During the ensuing war, in contrast, American leaders did issue a deterrent threat, warning Saddam against using biological or chemical weapons. And that deterrent worked. (The threat in that case was only elliptical; to make future deterrence less uncertain, threats should be much more explicit.) Despite humiliating defeat, Saddam held back his high cards in 1991 because he was never forced to the wall or confronted with his own demise. That war, unlike the one now contemplated, was limited.

Bush the Younger has quite aptly compared Saddam to Stalin but has drawn the wrong lesson from that parallel. Like Saddam, Stalin miscalculated in approving the invasion of South Korea in 1950, because President Truman (like the elder Bush in 1990) had not tried to deter him in advance. In fact, Secretary of State Dean Acheson had indicated publicly that South Korea was outside the U.S. defense perimeter. On the other hand, Stalin never invaded Western Europe, where the NATO deterrent was clear. For his part, Saddam's record shows that he is foolishly self-destructive when the consequences of his gambles are unclear, but not when they are unmistakable.

Should Saddam be compared to terrorists instead of to Stalin? If the Iraqi regime is viewed as similar to al Qaeda (a conflation of threats that official rhetoric has encouraged), deterrence would indeed be impractical. But Saddam and his Baath Party supporters are not religious fanatics bent on martyrdom. They are secularist thugs focused on their fortunes in this world. Nor can they hide from the United States, as al Qaeda members can. The crucial difference between a rogue state and a terrorist group is that the state has a return address.

None of this is meant to imply that containment and deterrence are risk-free strategies. They are simply less risky than would be starting a war that could precipitate the very danger it aims to prevent. Besides, what makes hawks so sure that long-term deterrence is more dangerous than immediate provocation? Saddam could be a greater threat in five years than he is today. But he could also be dead. He is now 65, and although he has so far been adept at foiling coups and assassination attempts, his continued success is hardly guaranteed. His stocks of WMD will grow more potent over time, but why should Saddam suddenly decide in the future that they afford him options he now lacks? And at what point in the growth of his arsenal would he plausibly choose to bring down a decisive American assault on himself and all his works?

It is also worth remembering that briefs made for preventive war in the past have proved terribly wrong. Truman, for example, did not buy arguments for attacking the Soviet Union -- despite the fact that, as the historian Paul Schroeder wrote recently in The American Conservative, "Stalin had nuclear weapons, was a worse sociopath than Hussein ... and his record of atrocities against his own people was far worse than Hussein's." Moreover, within a few years of Navy Secretary Francis Matthews' and others' having recommended preventive war against him, Stalin was dead. In 1968, similarly, Robert Lawrence and William Van Cleave (who served a dozen years later as head of Reagan's Pentagon transition team) published a detailed rationale in National Review for attacks on China's nascent nuclear facilities. It is easy today to forget that at that time, Mao was considered as fanatically aggressive and crazy as Saddam is today. But within a few years of Lawrence and Van Cleave's article, Washington and Beijing had become tacit allies. How history could have turned out had either of these preventive wars actually been fought is a sobering thought, and one that the White House should now consider.


Relying on deterrence indefinitely is not foolproof. Unfortunately, international politics is full of cases where the only policy choices are between risky options and even riskier ones. In the current era of U.S. primacy, Americans often forget this fact, mistakenly assuming that the only problems they cannot solve satisfactorily are those about which they are inattentive or irresolute. Overconfident in U.S. capacity to eliminate Saddam without disastrous side effects, leaders in Washington have also become curiously pessimistic about deterrence and containment, which sustained U.S. strategy through 40 years of Cold War against a far more formidable adversary. Why has Washington lost its faith?

One explanation is psychological and moral. Many people think of deterrence as something the good guys do to the bad, not the reverse. To use the current danger of Iraqi retaliation as a reason not to attack seems dishonorable, like taking counsel from fear, a wimpy submission to blackmail. Moreover, it strikes Americans as presumptuous for a country such as Iraq to aspire to paralyze U.S. power. And it is a matter of American honor not to be deterred from suppressing evil. The cold logic of deterrence, however, has nothing to do with which side is good or evil. Deterrence depends only on the hard facts of capability, which should constrain the good as well as the bad.

Some Americans also become indignant when it is suggested that an Iraqi counterattack could be considered the fault of American initiative. This stance, they argue, is like blaming the victim. But this argument again confuses moral and material interests. If the snake strikes back when you poke it, you may blame the snake rather than yourself for being bitten. But you will still wish that you had not poked it.

Of course, Iraq has undermined its own deterrent potential by not making it explicit. Because he always denies that he possesses prohibited WMD, Saddam cannot declare a deterrent capability or doctrine. Iraq's bugs in the basement should work like Israel's bomb in the basement -- as an undeclared deterrent, known about by those who need to know. But Iraq's WMD have not worked like Israel's, because, despite their potentially comparable killing power, biological weapons just do not instill the same fear as their nuclear equivalents.

At this late date, it would be awkward for Washington to step back from war -- an embarrassing retreat, unless it was cushioned by apparent success in imposing inspections. (Administration leaders are correct in believing that genuinely successful inspections are nearly impossible. To work, they would have to prove a negative -- that Saddam has not stashed WMD somewhere in his vast country that inspectors have not been clued in to search.) The only thing worse than such embarrassment, however, would be to go ahead with a mistaken strategy that risks retaliation against American civilians, extraordinarily bloody urban combat, and damage to the war on terrorism. No good alternatives to war exist at this point, but there are several that are less bad.

The first such option is to squeeze the box in which Saddam is currently being contained. This means selectively tightening sanctions -- not those that allegedly harm civilians, but the prohibitions on imports of materials for military use and the illicit export of oil. More monitors could be deployed, and the inspection of cargoes could be increased. The squeeze would continue at least until absolutely unimpeded disarmament inspections -- anytime, anywhere, undelayed, and institutionalized until the regime changes -- had been under way for a long period. There would be no international enthusiasm for more serious sanctions, but reluctant allies would embrace such a course if it were offered as the alternative to war. The crumbling of sanctions was one of the motives for the Bush administration's move toward war; stepping back from the war will reinvigorate containment and disabuse Saddam of the hope that he can wriggle away from it.

Second, Washington should continue to foment internal overthrow of Iraq's regime. Saddam seems immune to covert action, but even long-shot possibilities sometimes pan out.

Third, the Bush administration could consider quasi war. U.S. forces might occupy the Kurdish area of northern Iraq (where Saddam has not exercised control for years) and build up the wherewithal to move quickly against him at some unspecified future date -- to enforce inspections, to protect Iraqi garrisons that revolt against his rule, or, ultimately, to invade Baghdad.

As the noose tightens, Washington or its allies should offer Saddam safe haven if he and his henchmen step down. Of course, he is not likely to accept, and if he does, it would lead to an international chorus of clucking tongues as a heinous criminal escaped justice. But it would not hurt to leave open a bad alternative that remains better than unlimited war.

In pondering Bismarck's line about preventive war, it helps to recall the consequences of the Prussian's passing. He was soon replaced by leaders who saw more logic and necessity in the course Bismarck had derided. In 1914, such European leaders thought they had no alternative but to confront current threats with decisive preventive war, and they believed the war would be a short one. As often happens in war, however, their expectations were rudely confounded, and instead of resolving the threat, they produced four years of catastrophic carnage.

Applying Bismarck's definition of preventive war to the current case is a bit hyperbolic. Iraqi retaliation would not destroy the United States -- it might not even occur. But running even a modest risk of tens of thousands of American civilian casualties is unacceptable when compared to the exaggerated risk that Iraq will court its own suicide by using or helping others use WMD without provocation, and will do so before Saddam's regime is overthrown from within.

If war is to be, the United States must win it as quickly and decisively as possible. If no catastrophic Iraqi counterattack occurs, these warnings will be seen as needless alarmism. But before deciding on waging a war, President Bush should consider that if that war results in consequences even a fraction of those of 1914, those results will thoroughly discredit his decision to start it.

This article is part of the Foreign Affairs Iraq Retrospective.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Richard K. Betts is Director of the Institute of War and Peace Studies at Columbia University and Adjunct Senior Fellow at the Council on Foreign Relations, and was a member of the National Commission on Terrorism. Among his books are Surprise Attack, Nuclear Blackmail and Nuclear Balance, Military Readiness, and Conflict After the Cold War.
  • More By Richard K. Betts