Preemptive Action: When, How, and to What Effect?

Posted in United States , Other | 22-Oct-03 | Author: M. Bunn| Source: National Defense University

Key Points

The Bush administration's concept of preemptive action has become a lightning rod in domestic and international politics. Proponents see it as a prudent response to terrorism and rogue states, which may not be deterred from threatening or using weapons of mass destruction (WMD). Critics contend that the concept will exacerbate distrust of the United States on the part of allies and potential partners and may make rogue states more risk-prone.

There is a good deal of misunderstanding and confusion about this concept. Preemption usually is associated with military strikes, but financial, diplomatic, and law enforcement measures also can be used in preemptive ways to enhance security.

Preemption is not a new option. U.S. officials have contemplated preemptive military actions against WMD several times, usually without taking action. What is new is open discussion of preemption.

Successful preemptive strikes require precise, prompt actions with decisive effects. Thus, the United States must be prepared to invest in intelligence, surveillance, and reconnaissance capabilities, planning assets, and, potentially, new military forces.

The National Security Strategy of the United States does not establish clear criteria for preemptive military action. Apprehension about the current policy might be assuaged with an elaboration of the general conditions and circumstances, or the factors to be weighed, for the preemptive use of force.

What role should preemptive action play in U.S. national strategy? In the wake of the first public statements by President George W. Bush in June 2002, and in the buildup to military action against Iraq, the issue quickly became a lightning rod for controversy. While some commentators hailed preemption as a valuable concept whose time had come, others condemned it as a dangerous precedent that could damage American interests, strain our relations overseas, and make the United States a feared unilateralist in the international system. All the hue and cry has done little to clarify the issues and choices that policymakers face in weighing the utility and limits of the concept.

The National Security Strategy of the United States of America (September 2002) states that "the United States will, if necessary, act preemptively" to prevent rogue states or terrorists from threatening or using weapons of mass destruction (WMD) against the United States or its friends and allies (see page 2). Yes there is much misunderstanding and confusion about the administration's concept of preemption, which has led to a great deal of apprehension. Some of the confusion is self-inflicted, some is circumstantial, and some results from willful misreading.

While the administration was laying out a general concept, it did so against the backdrop of Iraq. It is not surprising, then, that some have failed to distinguish between the two. Iraq may be the first case study in the new policy--although some argue that action against Iraq was not preemption but a preventive war, while others argue it was a continuation of action from the 1991 Gulf War. In any event, Iraq is not the sum total of the policy.

In the popular mind, preemption is synonymous with the use of force, and specifically with military strikes. But the concept has a broader meaning and application, as implied by the administration's careful emphasis on preemptive action. To preempt is defined in Merriam-Webster's Collegiate Dictionary as "to prevent from happening or taking place: forestall; preclude." Many of the preemptive actions that the United States is likely to undertake will be nonmilitary. And the nonmilitary methods of preemptive action are likely to be less controversial than military preemption.

Historical Antecedents

There are a number of historical examples of the United States contemplating preemptive military actions including against WMD-armed adversaries (at least partially motivated by preemption or prevention, sometimes in addition to other motives), but only a small number of examples in which preemptive military action was actually taken.

U.S. Government deliberations in the late 1940s and early 1950s considered attacking Soviet nuclear capabilities while they were a fledgling force, and discussions in the early 1960s debated whether to take out Chinese nuclear capabilities.1 In neither case did the United States go forward with preemptive strikes.

The Cuban Missile Crisis of October 1962 has been cited repeatedly by Bush administration officials as an example of preemptive action (as opposed to a preemptive strike); the quarantine or blockade of Cuba was intended to prevent any further buildup of offensive arms (medium-range ballistic missiles and nuclear warheads were already on the island, although the presence of nuclear warheads was not clear at the time).2

In 1989-1990, America threatened to take military action to shut down a Libyan chemical facility at Rabta, which the United States suspected was intended to produce mustard or nerve gas. (This was against the backdrop of earlier tensions with Libyan leader Muammar Qadhafi and the 1986 U.S. bombing of Libyan leadership sites [Operation El Dorado Canyon] following the bombing of a Berlin night club that killed two Americans.) Qadhafi subsequently shut down the Rabta facility, claiming that a fire had destroyed it. U.S. officials in 1996 also threatened military strikes against a suspected chemical weapons plant that Qadhafi was building under a mountain near Tarhunah.3

In 1994, when North Korea threatened to remove fuel rods from the Yongbyon nuclear reactor, U.S. officials considered a preemptive strike on the reactor with conventional precision weapons. Former officials have recently testified that they were confident that such a strike would have eliminated the facilities at Yongbyon without causing any radioactive plume to be emitted downwind, but they also recognized that the result might well have been a very destructive North Korean conventional attack on South Korea. (Such a war was averted by the negotiation of the 1994 Agreed Framework.4)

The United States is not the only country that has considered or carried out preemptive actions. The classic case is the 1981 Israeli attack on Iraq's Osirak nuclear facility because of concern that Iraq would use the reactor to produce highly enriched uranium for a weapons program. (Iran had previously bombed Osirak, in the opening days of the Iran-Iraq war in 1980, lightly damaging the facility.5) After its 1981 attack, Israel claimed it was exercising its inherent right of self-defense, consistent with Article 51 of the United Nations (UN) charter. The UN Security Council censured Israel, and the U.S. Ambassador to the UN spoke against Israel for its action.6

In these historical cases--none of which involved nonstate actors--the risk/benefit analysis of preemptive action was controversial. These examples point to the difficult choices associated with preemption and to the likelihood that, in the future, there will probably be relatively few situations where the risks of preemptive military action are worth the costs.

Though the option for preemption is not new, talking about it publicly and raising it (arguably) to the level of doctrine is new, which may be both helpful and harmful. Moreover, while the National Security Strategy sensibly refrains from setting down rigid criteria for when preemptive military action should be seriously considered or used, its reticence to elaborate the general conditions and circumstances, or the factors to be weighed in deciding on the preemptive use of force, has failed to resolve uncertainty regarding administration decisionmaking when faced with difficult future situations.

To better understand the issues at stake in the concept of preemptive action, consider the following questions: why the new focus on preemption; against whom and what; when, how, and to what effect preemptive action might be considered?

The New Focus

The motivations underlying the Bush administration's emphasis on preemption spring from three strongly held assessments of the changing nature of international security and the heightened vulnerability of Americans to new-era threats.

The first, and foremost, perception is a growing nexus between transnational terrorism and WMD proliferation. It is hard to overestimate the impact of September 11, 2001, on administration thinking. In the months after the attacks, Congress and the public were asking what administration officials knew beforehand and whether they had done everything possible to prevent it. As the weight of responsibility for protecting the United States from an even worse occurrence--a potential September 11 with nuclear, biological, or chemical weapons--settled on the shoulders of administration officials, the need to act swiftly and firmly before threats become attacks is perhaps the clearest lesson they have drawn from that experience.

Second, there is a growing pessimism about deterrence. In the case of rogue state leaders, deterrence--while still applicable--may not always work. The perceived high-risk strategies of the former Iraqi and current North Korean leadership have added fuel to such arguments. In addition, some analysts have come to a working assumption that "deterring terrorists" is an oxymoron and that in the case of terrorists and WMD, possession equals use. While some believe that seeking to deny terrorists their objectives may have some (modest) long-term deterrent effect, many believe that terrorist organizations--which lack populations to protect or territory to safeguard and whose operatives may be willing to die for their cause--are essentially undeterrable, or at least very difficult to deter given international standards and political norms (for example, the unacceptability of threatening reprisals against innocent family members).

Third, there is a realization that if deterrence fails, defenses will never be perfect. Even if America had defenses of all types--ballistic missile defense, cruise missile and other air defenses, civil defense, detection, vaccines, port/border checks, and so forth--those defenses would not be 100 percent effective against WMD threats 100 percent of the time. As Secretary Donald Rumsfeld has stressed, "It is not possible to defend against every conceivable kind of attack in every conceivable location at every minute of the day or night. . . . The best, and in some cases, the only defense, is a good offense."7

Statements pointing to this administration's lack of confidence in traditional deterrence precede its statements on preemption. The President's May 2001 speech at the National Defense University addressed the need to broaden the concept of deterrence and put in new, more capable offensive and defensive elements to bolster this concept.8 His campaign speech at The Citadel in September 1999 also sounded the theme that traditional deterrence is not sufficient in the new security situation.9 However, the first explicit mention of preemption during the Bush administration was in the statement of the Chairman of the Joint Chiefs of Staff in the Quadrennial Defense Review Report issued September 30, 2001.10 In recognizing that defense of the U.S. homeland is the highest priority for the Armed Forces, the report states that the United States "must deter, preempt, and defend against aggression" targeted at the United States.

What to Preempt?

Administration officials have consistently stated that the concept of preemption is focused on terrorists and rogue states--thereby presumably easing the minds of leaders in Russia and China, while potentially heightening concerns in Iraq (prior to the fall of Saddam), Iran, North Korea, and possibly Libya and Syria.

Rogue states. Though the President seems to have avoided the term "axis of evil" since he used it in his January 2002 State of the Union address, the National Security Strategy's characterization of rogue states appears to define the common traits of countries in the axis: those who brutalize their own people, display no regard for international law, threaten their neighbors, are determined to acquire weapons of mass destruction, sponsor terrorism around the globe, and reject basic human values. (Interestingly, the National Security Strategy applies this characterization and the phrase rogue state only to Iraq and North Korea--omitting Iran, the third country named in the axis of evil speech.)

Terrorists. The National Security Strategy defines terrorism as "premeditated, politically motivated violence perpetrated against innocents" and focuses as a priority on disrupting and destroying "terrorist organizations of global reach."

While the National Security Strategy lumps together rogue states and terrorists when discussing preemption, it is useful to consider them separately. There are important differences that may yield different preemptive options. In addition, the threshold for obtaining international support for preemptive action against a recognized sovereign state (even a rogue state) is likely to be significantly higher than against terrorist groups. In fact, many argue that because al Qaeda has attacked the United States (even before September 11), it is not "preemption" for America to seek to destroy al Qaeda infrastructure and leadership.11

Targets. The U.S. Government might want the option of taking preemptive action against a range of threats by rogue states and terrorists, including aggression of any kind, and particularly conventional attacks with mass effects, such as the September 11 attacks. However, American officials have focused their discussion almost exclusively on WMD--that is, chemical, biological, or nuclear weapons.

Even within the restricted subset of WMD, one might consider preempting the development of weapons-useable material, acquisition of weapons, or their use. Specific WMD targets might include:
  • pre-capability components of weapons or precursor agents
  • production sites, storage, deployment sites, launchers or other delivery systems
  • military forces (of rogue states) or operatives (of terrorist networks) that might employ WMD
  • command and control systems (which may be very different for terrorists than for rogue states), including leadership or a regime.

When to Preempt?

There has been considerable discussion of the distinction between preemptive action (when adversary use of force is imminent) and preventive action (for example, depriving an adversary of a capability that it might someday have or averting an action that it might someday contemplate). Some observers maintain that the issue being raised by the administration is not preemption but instead preventive military operations or preventive war--that is, taking preventive action before the need is certain and balancing the risks of acting against those of not acting. The threshold of a decision to take preventive action will be much higher than for preemptive action, for which there is actual, presentable intelligence of an imminent threat. However, from a practical standpoint, since the President and the National Security Strategy first used the phrases preemptive action and preemption, they have taken hold in public discussion. The debate really revolves, then, around the definition of imminent.

There are two ways to think about a preemption timeline: either the traditional peacetime-to-crisis-to-conflict-to-postconflict timeline, in which preemptive action is taken in peacetime or a building crisis; or, in the case of WMD, where an adversary is on the acquisition-to-use timeline. The points on the latter continuum range from just prior to the time at which an adversary has acquired a useable capability (for example, disrupting/interdicting delivery of the last item needed for a completed WMD capability, or in the terrorist case, interdicting the team before it infiltrates the WMD site, or before the handover of materials takes place), to the point where an adversary has the capability and generally hostile intent, to the point where there is a specific hostile intent and use is imminent.

At what point is danger imminent? Condoleezza Rice has noted, "new technology requires new thinking about when a threat actually becomes ‘imminent'...We must adapt the concept of imminent threat to the capabilities and objectives of today's adversaries."12 We can imagine a scenario in which preemption would be justified based on imminent danger. For instance, suppose that the United States had firm intelligence that a hostile rogue government was about to launch missiles with nuclear warheads that could kill millions of Americans (in the case of no or imperfect missile defenses). A President would at least have to consider the possibility of preemption in such a situation. If one is absolutely sure that a devastating attack is coming, preemption is an easier choice. That scenario is at one end of a WMD time continuum. At the other end is a nation with general hostile intentions and for which there is ambiguous intelligence regarding efforts to acquire WMD. Where along that spectrum is preemptive action justifiable? Presidents will likely hesitate to preempt when today looks quite a bit like yesterday, and they anticipate that tomorrow is going to look quite a bit like today. This raises the question of what U.S. thresholds will be and how to describe the action points.

Michael Waltzer, in his 1977 book Just and Unjust Wars, drew the line between legitimate and illegitimate first strikes not at the point of imminent attack but at the point of sufficient threat--a phrase he recognized as "necessarily vague." He meant it to cover three things: "a manifest intent to injure, a degree of active preparation that makes that intent a positive danger, and a general situation in which waiting, or doing anything other than fighting, greatly magnifies the risk."13

It is ironic that the closer to the "imminent use" end of the spectrum a situation falls, the easier preemption is to justify politically, but the harder it may be to be operationally decisive, because the adversary will likely have protected the intended targets of preemption through deception, hardening, burial, dispersal, or predelegation of release. Conversely, the farther from the imminent use end of the continuum a situation is, the less acceptable it is likely to be to world opinion, though preemptive strikes at that stage are more likely to be effective in eliminating or postponing the threat.

How to Preempt?

Though most tend to think of preemption in terms of strike--kinetic weapons on target--the ways in which preemptive action can be carried out are much broader. Preemptive action also could employ a variety of nonmilitary, as well as military and intelligence, capabilities. The National Security Strategy states that the "United States will not use force in all cases to preempt emerging threats." Rice, in interviews shortly after the President's June 2002 West Point speech, made clear that preemptive action did not necessarily imply force. She stated, "It really means early action of some kind. It means forestalling certain destructive acts against you by an adversary."14

The National Strategy for Homeland Security, released in July 2002, has been little noticed in the preemption debate, although it addresses the issue extensively--in terms of "preventing and preempting future attacks."15 It notes that:

with advance warning, we have various federal, state, and local response assets that can intercede and prevent terrorists from carrying out attacks. These include law enforcement, emergency response, and military teams. In the most dangerous of incidents, particularly when terrorists have chemical, biological, radiological, or nuclear weapons in their possession, it is crucial that the individuals who preempt the terrorists do so flawlessly, no matter if they are part of the local SWAT team or the FBI's Hostage Rescue Team.16

Where preemptive action occurs will have implications for how it is done. Places might include international territory (such as interdicting capabilities, precursors, or terrorists at sea); the territory of a rogue state; the territory of a falling or failing state; or the territory of a friend or ally. Preemptive action in the latter case--that is, in nonmilitary ways with the cooperation of an ally--may be very different from that against rogue or failing states.

At times, military action will be required for preemption to be effective. Even in those cases, there are a variety of methods for applying military forces preemptively. While most people tend to focus on kinetic strike capabilities, there are also nonkinetic offensive means such as information operations (for example, disrupting command, control, and communications); special operations forces; or Coast Guard/Navy boarding ships, imposing blockades, or turning back/sabotaging shipments. The December 2002 interception of a North Korean ship at sea carrying 15 Scud missiles is an example of what that type of preemptive action--interdiction at sea as a kind of aggressive nonproliferation--might look like.17

When preemptive military action is required, there will likely be a need for decisive effect and the ability to act promptly for such action to be successful--not only to take advantage of potentially fleeting opportunities but also to allow the maximum amount of time for consultation, consideration, and deliberation of whether the United States will take action. America must be prepared to make investments in improving intelligence, surveillance, and reconnaissance capabilities; planning, including analysis of the effects of executing a plan (especially attacking WMD sites while trying to minimize collateral damage); and the military forces themselves (see box on page 5).

Complicating military options is the fact that adversaries, or the targets of preemption, are taking actions to constrain the U.S. ability to preempt. These enemy actions are designed to defeat the military effects of preemption, essentially eliminating the opportunity for surprise or raising costs to an unacceptable level. For example, adversaries may embed a target within commercial activities, cloak it in denial-deception operations, put it underground, conceal it, make it mobile, or embed it in populated areas that make it difficult to target without substantial collateral damage. Aided by the globalization of technology, adversaries are working to frustrate American precision-guided munitions. To be effective in such circumstances requires an extraordinary degree of precision, and precision strike requires intelligence collection on a scale that is extraordinarily difficult and very costly in terms of time and resources.18

Preemption sometimes is taken to mean nuclear preemption. This confusion is partly a holdover from Cold War usage and partly a result of recent developments. The 2002 Nuclear Posture Review (NPR)--which addressed not only nuclear issues but also nonnuclear strike as well as active and passive defenses--first came into public view through selective leaks of the classified NPR Report to Congress by critics who focused on its nuclear aspects. More recently, the discussion of preemption against Iraqi WMD coincided with the discussion of a new (or repackaged) nuclear weapon to target hard and deeply buried targets and chemical and biological weapons.

It is an understatement to say that the threshold of any Presidential decision for nuclear preemption would be higher than for conventional strike. It is difficult to imagine that a President would preempt with nuclear weapons in other than the most dire circumstances, with no other option to prevent massive American or allied casualties. The use of nuclear weapons would be among the most difficult decisions a President could make--given their destructiveness, the breaking of the 50-plus-year barrier to nuclear use, as well as the international opprobrium likely to follow. It would be hard enough for any President to make the decision to use nuclear weapons in response to the use of WMD; deciding to preempt with nuclear weapons would be far more difficult.

So why, some might ask, is the administration studying whether lower-yield, lower-collateral-damage, earth-penetrating nuclear weapons--either new or revamped--should be pursued? Administration officials argue that their purpose is (still) largely for deterrence--so that in the mind of a rogue state leader, it is credible that the United States might use nuclear weapons and thus the risks are not worth the gains.

To What Effect?

In any assessment of the effect preemptive action will have, both operational/military effect and political effect are important considerations.

On the operational/military side, the U.S. Government needs to think through the effectiveness of any preemptive operation in achieving its objectives and how to measure that effectiveness. If the overriding objective of preemption is regime change, the measure is fairly straightforward, though not necessarily easy to achieve. On the other hand, if the objective is to disarm, rather than to remove the leadership of the adversary, the measure of effectiveness is more temporal. Specifically, for what period of time would a particular preemptive action prevent or delay acquisition of a WMD capability? For example, the Israeli attack against the Osirak reactor may have set the Iraqi nuclear program back--it is unclear whether by several months or several years--but as the inspections after Operation Desert Storm in 1991 found, the attack did not end the Iraqi program. If anything, preemption may have led Iraq to take a much more expanded effort, pursuing several paths to nuclear capability in parallel.19

What level of confidence is required (in both the intelligence information and the U.S. ability to effectively carry out the mission) to make a decision to preempt? If, for example, the objective were to destroy WMD, how much must be destroyed for the preemptive action to be considered effective--all, most, or some? There is unlikely to be absolute confidence in the intelligence capability to locate all of an adversary's WMD, or in the capability to destroy it. Will preemption prompt the adversary to use the remainder? And is the United States better or worse off after such a preemptive action?

The answer depends fundamentally upon the assumption we make about the likelihood of use. Would a rogue leader use WMD in any event (either early, as a last chance to save the regime, or as the regime falls, in an effort to inflict as much pain as possible)? If he were going to use WMD anyway, and the United States destroys even some of it, preemption makes sense; but if it prompts adversary use of WMD capability that otherwise would not have been used, preemption would turn out to be a bad choice. The difficulty is in knowing the answer in advance. Are terrorists a different case? Does the assumption "possession equals use" apply? If so, the "better-or-worse-off" presumption in terrorist cases would be in favor of preemption.

Since preemptive action is unlikely to be 100 percent effective, the role defense plays in preemption must be considered. Should defenses of all types--ballistic missile defense, cruise missile and other air defense, defenses of borders and ports--be put on a higher state of alert or readiness prior to preemptive strikes? What effect would that have on tactical surprise? And is the U.S. Government more likely to consider preemption if it believes its defenses can be effective against any WMD capabilities that it fails to destroy? Conversely, do limited defenses reduce the pressure to preempt if U.S. leaders believe that the first few adversary WMD used could be neutralized?

With regard to the political effect, worldwide reaction to preemptive action by the United States would depend on a number of factors:
  • What was the reason for preemptive action, its effectiveness, and the capability used?
  • Was the level of force, if required, commensurate with the task, or was it perceived as excessive?
  • How was the action viewed under international law?
  • How effective was U.S. consultation and intelligence sharing as the crisis ramped up?
  • What was the level of unintended damage (including the contaminating effects of partially destroyed WMD materials) to civilians and the territory of others?
  • How effective was consequence management after the fact, and what was the American role in it?
  • How effective was U.S. public diplomacy on the issues just mentioned (what proof was offered; how convincing was the case that preemption was the best of a number of bad options and that it was done in self-defense)?
  • How effective was the public diplomacy by adversaries and others?

The reactions of allies and friends, conditioned as they would surely be by some or all of these factors, would have consequences for coalition building and cohesion as well as basing rights, access, and overflight.

The United States would also have to judge the policy effect of the message to the next potential adversary, as well as whether others--such as China versus Taiwan, or India and Pakistan--would see U.S. preemption policy as a green light for their own preemptive actions--or at least as rhetorical cover. The administration was obviously cognizant of the "precedent for other nations" criticism. National Security Advisor Rice has stated, "But this approach must be treated with great caution. . . . It does not give a green light--to the United States or any other nation--to act first without exhausting other means, including diplomacy. . . . The threat must be very grave. And the risks of waiting must far outweigh the risks of action."20 This statement implies that the bar is set high, but administration statements are vague about the conditions under which preemption would be undertaken and what the decision factors should be.

Why Talk about It?

Although preemption as an option is not new, the level of public discussion of it by American officials is. Previously, preemption had been discussed internally in defense or policy circles and very little externally. If other administrations have had the quiet option for preemption, why has this administration talked about it so publicly, and even written it explicitly into its National Security Strategy?

What effect does public discussion of preemption by U.S. officials--either generally or specifically, in peacetime and in crisis--have on friends as well as on potential adversaries? There are both advantages and disadvantages of U.S. officials talking publicly about preemption.

Discussion might signal a new seriousness in dealing with WMD and terrorism that will strengthen deterrence. Could discussion, for instance, have a deterrent effect on states that might let WMD fall into the hands of terrorists--especially if the weapons might be traced to them and they were held accountable? Might it make rogue states hesitate before an aggressive action that might be used to justify preemption? Might it send a message to rogue states that their continued existence depends on whatever WMD capabilities they may have not falling into the hands of terrorist groups, whether voluntarily or involuntarily?

While it may dissuade or deter rogue state brandishing of WMD, U.S. talk of preemption could, on the other hand, make rogue states more determined to acquire or maintain WMD, in order to make Americans think long and hard about the costs of preemption or regime change against them, as well as to work even harder to hide or disperse those capabilities or leaders against whom they think preemptive action might be taken. This concern appears to be animating North Korean determination to acquire nuclear weapons despite international opprobrium.

Public discussion of preemption may also be seen by the administration as opening a critical debate necessary to prepare the public for use of force in ways not previously thought to be consonant with the American sense of fairness. There is a public presumption (and even a military presumption) that "good guys don't preempt," which some believe, in the new security environment, stands as an inappropriate restraint on U.S. action. So the President or his advisors may have decided that it is necessary to engage in this public debate in order to prepare the intellectual groundwork (both internationally and domestically) that would otherwise make the use of preemptive options, if not impossible, then exceedingly difficult.

Secretary Rumsfeld has observed:

It is difficult for all of us who have grown up in this country and believed in the principle that unless attacked, one does not attack . . . for the most part, our country has had a view that that was the way we do things. It was other countries that have attacked us for the most part and initiated conflicts. The question, though, is in the 21st century, with biological weapons, for example, that could kill hundreds of thousands of people, what does one do? Does one wait until they're attacked? . . . There is no doubt in my mind, but that the overwhelming majority of the American people would prefer that their government take the kinds of steps necessary to prevent that type of attack.21

Officials in other nations have also discussed the issue of preemptive action. Australian Prime Minister John Howard observed, "It stands to reason that if you believe that somebody was going to launch an attack on your country, either of a conventional kind or a terrorist kind, and you had a capacity to stop it and there was no alternative other than to use that capacity, then of course you would have to use it."22 He also reportedly said that he would like to see the United Nations Charter changed to allow preemptive action against terrorists.

The French government document issued in January for its 2003-2008 military program addressed preemption as well: "We must . . . be prepared to identify and forestall threats as soon as possible. In this context, the possibility of preemptive action might be considered, from the time that an explicit and confirmed threatening situation is identified"23

In Japan, which has been reluctant since World War II even to contemplate taking actions except in self-defense, Japan Defense Agency Director General Shigeru Ishiba said in January:

If North Korea expresses the intention of turning Tokyo into a sea of fire and if it begins preparations [to attack], for instance by fueling [its missiles], we will consider [North Korea] is initiating [a military attack]. . . . Once North Korea declares it will demolish Tokyo and begins preparing for a missile launch, we will consider it the start of a military attack against Japan.24

Ishiba later stressed that, even with Japan's Peace Constitution, "Just to be on the receiving end of the attack is not what our constitution had in mind. . . . Just to wait for another country's attack and lose thousands and tens of thousands of people, that is not what the constitution assumes."25

Public discussion of American preemption may also alienate friends as well as warn enemies. Some have expressed concern that the public debate has a strategic consequence of complicating relationships with some allies and coalition partners. Talk of preemption may have exacerbated what some countries see as a U.S. tendency toward unilateralism and acting above the law and international norms. A German editorialist called the doctrine "immature, even dangerous . . . it is the law of the strongest."26 Others have expressed concern that by stating the threat to take preemptive action, it sounds as if America is threatening to execute a "series of Pearl Harbors"--and that, if what the United States is trying to do is to create a consensus as to what is right and what is wrong, the result could be the opposite.

Much of the debate over preemption confuses threats of preemption with preemptive operations--that is, talking about it versus actually doing it. A preemptive threat may actually complicate a preemptive operation by undercutting the element of surprise. With regard to terrorist and rogue state adversaries, public talk of preemption may encourage them to redouble efforts to hide, disperse, or quickly use WMD capabilities that they anticipate may be the targets of preemptive action. Indeed, there may be a tradeoff between the operational pluses of surprise versus the political downsides of not having built the case and influenced public opinion before preemptive action (keeping in mind the distinction between tactical and strategic surprise). When the Defense Secretary was asked, "Is the United States contemplating preemptive strikes against other nations' WMD holdings?" Rumsfeld replied, "On the record. Why would anyone answer that question if they were contemplating it?"27

Option or Doctrine?

An important issue for the national security community is whether preemption, which was treated as merely an option in the past, has been raised to the level of a doctrine by the Bush administration. The President himself said in remarks (perhaps impromptu) on June 14, 2002:

I was at West Point the other day and I was honored to give a graduation speech where I laid out a new doctrine [emphasis added] called preemption. . . . In the past, we used to have a doctrine called containment and deterrence. You can't contain a shadowy terrorist network. You can't deter somebody who doesn't have a country. And you're not going to be able--future Presidents won't be able to deter or contain one of these nations which harbors weapons of mass destruction, nations who hate America.28

This statement sounded like the death knell for previous doctrines. However, Secretary of State Colin Powell, responding to Congressional questions, observed that it is not "as if all other strategies and doctrines have gone away and suddenly preemption is the only strategy doctrine. That's just not the case."29 Equally, Rice has made clear that "The National Security Strategy does not overturn five decades of doctrine and jettison either containment or deterrence. These strategic concepts can and will continue to be employed where appropriate."30

The question of "option or doctrine" is not purely a semantic distinction. To call preemption a doctrine implies that it is a central organizing principle for marshalling the instruments of national power in support of national objectives and that in relevant cases, action will be taken in accordance with established governing principles. Seen in this light, the recent use of force against Iraq may be the first application of a new doctrine of preemptive action.31 On the other hand, preemption may have been an option employed in that specific situation, without its rising to the level of doctrine. With respect to North Korea, the administration has downplayed the idea that Pyongyang's recently disclosed nuclear program constitutes a crisis or that military force might be required, and it has insisted that a diplomatic solution is possible. This inconsistency could lead to the conclusion that since preemption has not been chosen in all cases where it conceivably applies, it must not be a doctrine. But the administration also reportedly moved B-52s and B-1s to Guam so that they would be available for military options in North Korea--one of which might be preemptive action. Probably only in hindsight, with historical perspective, will we be able to judge the administration's application of the concept of preemption.


The National Security Strategy is notably silent on the issue of what, if any, decision criteria our national leadership would apply in considering the possibility of preemptive military action to counter WMD. Perhaps public elaboration of such criteria was too much to ask--given the sensitivity of the issue--but silence begs the question: Is there a roadmap that can be seriously considered or used?

Over a decade ago, Michéle Flournoy and Philip Zelikow, as part of the early debate on counterproliferation, developed a lengthy set of questions, which include the following points that still serve as a useful guide to decisions about preemption:
  1. Do we have less than high confidence that we can deter a given country from using [nuclear, biological, or chemical] weapons against U.S., our allies, and our interests?
  2. Do we believe that the country might transfer the capabilities to others who might use them?
  3. Will the mere acquisition by this country of WMD significantly increase the danger of war or coercion or attack that would threaten our interests? If the answer to any of these three is yes, then the next question is:
  4. Are we confident that these dangers can be significantly dealt with short of preemption? If not, then we go down the list.
  5. Do the actions and behavior of the country ethically and legally justify offensive military action by the United States?
  6. What kind of public reaction--domestic and foreign--can we anticipate? And if highly negative, would it be prohibitive, or is it something we'd be willing to weather criticism for?
  7. Can military action contain or eliminate the danger without risking either retaliation or unintended consequences that would pose an even greater risk to our interests? Is preemption worth the risk of the response?
  8. Do we know enough about the adversary's capabilities, their defenses, and their operations, to actually launch a strike that will be effective with high confidence of success and with acceptable levels of collateral damage?
  9. Can we do it? Do we have the military and intelligence capabilities to pull it off?
  10. Looking strategically at the long-term implications, what precedents does this action set? Do the anticipated benefits outweigh the costs and conversely, are we willing to live with not preempting in a given case? Are we willing to live with a given country having a capability that could be used in the future?32

These questions are as relevant today as they were 10 years ago. While the first several questions focus on countries--nonstate terrorists were not as great a concern then as now--they can be adapted to apply to terrorist groups as well as rogue states. The key question will be how great is the danger, but it is also important to ask whether the dangers can be dealt with in other ways, whether the actions ethically and legally justify U.S. military action, and whether the United States has the intelligence and military capabilities to be effective with high confidence of success and acceptable levels of unintended damage.

The 10 questions above are a starting point for decisionmakers and may also help the public think through preemption. Answers from today's perspective may differ from the day-after perspective; that is, if in fact WMD had already been used against the United States or American forces or allies with many thousands of casualties, would leaders and the public have a different perspective on the advisability and the desirability of preemption?

These points are geared to thinking about when military preemption might be appropriate, although they could be adapted to nonmilitary preemptive action as well. However, military preemption will be more difficult to justify than nonmilitary options. Adopting some version of these considerations as criteria for preemption might reduce the apprehension and misperceptions about administration policy, because when these factors are seriously weighed, there are likely to be few cases where the preemptive use of military strikes will be chosen as the best option.

Looking beyond Iraq, it is difficult to predict whether preemptive military action will be the doctrine of the future or a seldom-used option. The challenge here is somewhat like divining the future of case law, in which established practice is created step by step and can only be understood by looking backward, not forward.

Clearly, the United States is feeling its way carefully as it goes along. The recent U.S. focus on preemption was born of a defensive reaction in the face of emerging threats. If used too frequently, without good justification, and with ill effect, America is more likely to be increasingly viewed as an arrogant nation carrying out the "law of the strongest." Used judiciously, with good reason, and effectively--only in the toughest cases--preemption is likely to be accepted, if somewhat grudgingly, by the international community.


1 William Burr and Jeffrey T. Richelson, "Whether to 'Strangle the Baby in the Cradle': The United States and the Chinese Nuclear Program, 1960-64," International Security 25, no. 3 (Winter 2000/2001), 54-99.

2 Anatoli I. Gribkov and William Y. Smith, Operation ANADYR: U.S. and Soviet Generals Recount the Cuban Missile Crisis (Chicago: Edition q, 1994).

3 Richard J. Newman, "The Qadhafi Question," U.S. News and World Report, April 15, 1996, 15.

4 Ashton B. Carter, "Three Crises with North Korea," prepared testimony before the Senate Foreign Relations Committee, February 4, 2003.

5 Rebecca Grant, "Osirak and Beyond," Air Force Magazine, August 2002, 74. Grant cites a statement issued by the official Iraqi news agency following the 1980 Iranian bombing of Osirak: "The Iranian people should not fear the Iraqi nuclear reactor, which is not intended to be used against Iran, but against the Zionist entity."

6 Anthony Clark Arend, "International Law and the Preemptive Use of Military Force," The Washington Quarterly 26, no. 2 (Spring 2003), 89.

7 Donald H. Rumsfeld, "21st Century Transformation of U.S. Armed Forces," presentation at the National Defense University, Washington, DC, January 31, 2002.

8 George W. Bush, remarks at the National Defense University, Washington, DC, May 1, 2001.

9 George W. Bush, remarks at The Citadel, September 23, 1999.

10 Department of Defense, Quadrennial Defense Review Report (Washington, DC: Department of Defense, September 30, 2001), particularly Chapter VIII: Statement of the Chairman of the Joint Chiefs of Staff.

11 See Walter B. Slocombe, "Force, Pre-emption and Legitimacy," Survival 45, no. 1 (Spring 2003), 123.

12 Condoleezza Rice, remarks on the President's National Security Strategy, Waldorf-Astoria Hotel, New York, NY, October 1, 2002.

13 Michael Walzer, Just and Unjust Wars (New York: Basic Books, 1977), 81.

14 David E. Sanger, "Bush to Formalize a Defense Policy of Hitting First," The New York Times, June 17, 2002.

15 The Office of Homeland Security, The National Strategy for Homeland Security, July 2002, accessed at, 16.

16 The National Strategy for Homeland Security, 43.

17 In that case, it was determined the vessel was bound for Yemen. The Yemeni government asserted that the missiles were legally purchased, and the United States decided that there was no clear authority to hold the ship or prevent it from delivering the missiles.

18 See Slocombe, "Force, Preemption and Legitimacy," 126: "The problem with pre-emption, unfortunately, is not lack of legal legitimacy, but operational practicality . . . Precision weapons require precision intelligence, and pre-emption requires that intelligence be comprehensive as well as precise."

19 See Marc Dean Millot, "Facing the Emerging Reality of Regional Nuclear Adversaries," The Washington Quarterly 17, no. 3 (Summer 1994), 41. See also David A. Kay, "Denial and Deception Practices of WMD Proliferators: Iraq and Beyond," The Washington Quarterly 18, no. 1 (Winter 1995), 83.

20 Rice, remarks on the President's National Security Strategy, October 1, 2002.

21 Donald H. Rumsfeld, Department of Defense Regular News Briefing, January 29, 2003, accessed at

22 John Shaw, "Startling His Neighbors, Australian Leader Favors First Strikes," The New York Times, December 2, 2002, A11.

23 LOI n° 2003-73 du 27 janvier 2003 relative à la programmation militaire pour les années 2003 à 2008 (1), section 2.3.1, accessed at

24 "Ishiba: Japan to 'Counterattack' If N. Korea Prepares to Attack," The Yomiuri Shimbun/Daily Yomiuri, January 25, 2003.

25 James Brooke, "Japanese Official Wants Defense Against Missiles Expanded," The New York Times, April 17, 2003, 13.

26 Wolfgang Koydl, "The Law of the Strongest," Sueddeutsche Zeitung, June 3, 2002 (as translated by the Department of State "Foreign Media Reaction Early Report," June 13, 2002).

27 Donald H. Rumsfeld, meeting with the editorial board of The Washington Post, June 3, 2002, accessed at

28 George W. Bush, remarks at Texans for Rick Perry Reception, Houston, TX, June 14, 2002, accessed at

29 Hearing before the Senate Foreign Relations Committee on Treaty on Strategic Offensive Reductions, Washington, DC, July 9, 2002.

30 Rice, remarks on the President's National Security Strategy, October 1, 2002.

31 On the other hand, some argue that it was not preemption, but a continuation of the 1991 Gulf War that ended with a ceasefire and conditions that Iraq failed to meet.

32 Drawn from chapters by Michéle A. Flournoy and Philip Zelikow in New Nuclear Nations: Consequences for U.S. Policy, ed. Robert D. Blackwill and Albert Carnesale (New York: Council on Foreign Relations Press, 1993), as well as more recent statements by Flournoy.