After 9/11, U.S. President George W. Bush announced his determination to do whatever was necessary to prevent future terrorist attacks against the United States. Following the lead of several countries that had recently come to similar conclusions after their own bitter experiences -- including India, Israel, Japan, Russia, Spain, and the United Kingdom -- the United States tightened its immigration laws; increased the protection of its borders, ports, and infrastructure; criminalized providing "material support" for terrorist groups; and tore down the wall between the intelligence agencies and law enforcement agencies, which had crippled counterterrorist efforts for decades. Washington did not authorize preventive detention, as other countries had, but it used other measures to hold persons against whom criminal charges could not be brought -- thereby preventing terrorist attacks. The U.S. government also led or joined various international efforts aimed at warding off new dangers, such as the Proliferation Security Initiative, through which over 70 states cooperate to interdict the movement of nuclear materials across international borders.

But the Bush administration's call for preventive action went further: it endorsed using force against states that supported terrorism or failed to prevent it. This was a particularly controversial position, since using (or threatening to use) preventive force across international borders is generally considered to be a violation of international law: the International Court of Justice (ICJ) and most international legal authorities currently construe the United Nations Charter as prohibiting any use of force not sanctioned by the UN Security Council, with the exception of actions taken in self-defense against an actual or imminent state-sponsored "armed attack."

Now that the Bush administration is no longer in power, some argue that its approach to this subject should be shelved. But the objective of preventing terrorist threats before they are realized -- rather than primarily treating terrorism as a crime warranting punishment after the fact -- is now established as an essential element of U.S. national security. Indeed, in 2008, Barack Obama, then a presidential candidate, told the American Society of International Law that "the preventive use of force may be necessary, but rarely." And Vice President Joe Biden announced in February 2009, "We'll strive to act preventively, not preemptively, to avoid whenever possible the choice of last resort between the risks of war and the dangers of inaction." Other senior officials have made similar statements, both before joining the Obama administration and since.


The case for considering preventive force stems largely from the threat posed by terrorists, especially their potential use of weapons of mass destruction (WMD). But it is justified by other threats as well, including the proliferation of WMD to irresponsible or fanatical regimes; the spread of criminal activities, such as piracy and drug and human trafficking; and genocide or other massive violations of human rights.

The UN, NATO, other regional organizations, and many individual countries agree that these threats exist and need to be countered. The UN High-Level Panel on Threats, Challenges and Change, appointed by Secretary-General Kofi Annan, concluded in December 2004 that "in the world of the twenty-first century, the international community does have to be concerned about nightmare scenarios combining terrorists, weapons of mass destruction and irresponsible States, and much more besides, which may conceivably justify the use of force, not just reactively but preventively and before a latent threat becomes imminent."

Preventive action seeks to counter threats before they are imminent. It is thus distinguished from preemptive action, which, in Daniel Webster's classic formulation, is taken when a government has "a necessity of self-defence, instant, overwhelming, leaving no choice of means, and no moment for deliberation." In contrast, preventive action seeks to head off dangers that are further in the future and therefore less tangible, less likely to occur, and possibly more avoidable through diplomacy. Prevention is not considered an act of self-defense under the UN Charter. But although these distinctions are valid and significant, they can be overstated.

Many contemporary threats do not involve conventional forces that can be observed as they prepare to attack; rather, they involve unconventional uses of force that remain invisible until their sudden deployment. Such unconventional uses of force can seldom be preempted at their moment of imminence. Moreover, many contemporary threats do not qualify as "armed attacks" under current international law, since they come from nonstate actors, are aimed at a state's citizens or interests outside of its territory, or are considered insufficiently substantial to entitle a state to resort to the full range of actions allowed in cases of recognized self-defense. Some threats stem from criminal activities, not "armed attacks," and some involve governments targeting their own populations via massive deprivations of human rights. All these unconventional threats can be very serious and may result in great potential harm. Although international uses of force to counter such threats would be considered illegal without Security Council approval, such actions could be as vital to maintaining international peace and security as the actions of domestic law enforcement are to maintaining domestic security.

Preventive actions pose serious risks. Rather than deterring a state from attacking, the prospect of being targeted by preventive action may provoke it to strike first. Moreover, since preventive actions are based on predictions of future conduct, they are subject to error. The use of force always causes human suffering, intended and unintended, but the costs are more difficult to justify if they result from an action later revealed to have been unnecessary. And preventive action can do more harm than good, opening attackers to condemnation and alienating the public in the states that are attacked. But a decision must be made one way or the other, and either option may prove in retrospect to have been wrong. Although being wrong is an unavoidable risk, it can be limited through the disciplined and effective collection and analysis of intelligence.


Despite the illegality and the risks involved, states have used unauthorized force for preventive purposes in well over 100 instances since the UN Charter was signed in 1945. This has happened in part because during the Cold War stalemate, the Security Council failed to perform its duty to maintain international peace and security. Instead of cooperating on the council, the two superpowers competed to support proxies, which, in turn, sought power and territory across the world. In some cases, parties used unauthorized force to overthrow colonial regimes. These coups (which had no preventive purpose) demonstrated the unwillingness of states to be bound by narrow legal limits on the use of force, including the requirement of Security Council authorization. Other unauthorized uses of force were preventive, including hostage-rescue operations, abductions of criminals and terrorists, targeted killings, attacks on terrorists and their infrastructure, actions to prevent the subversion of established governments, the disruption of WMD development, actions to prevent gross violations of human rights, preventive cyberattacks, and preventive war.

Although virtually all these were technically illegal, they elicited different reactions from the international community depending on their purpose, duration, and consequences. Some unauthorized actions have been condemned, many have been accepted without comment, and some have even been widely praised or formally supported.

Uses of force whose purpose has seemed inconsistent with the UN Charter have generally been opposed. For example, whereas countries' efforts to protect their citizens have generally been tolerated, they have been opposed when colonial powers have exploited them in order to continue exercising control in former colonies (such as in the British and French seizure of the Suez Canal in 1956). Sometimes, the same type of action has been tolerated in one set of circumstances and condemned in another. Israel's abduction from Argentina of the Nazi war criminal Adolf Eichmann in 1960, for example, was formally criticized but effectively condoned, whereas the United States' 1990 abduction of a doctor from Mexico for cooperating with drug dealers was widely viewed as an improper extension of national law enforcement.

Although interventions have often earned criticism when they have seemed to advance national interests rather than objectives based on the UN Charter, the motive has often mattered less than the result. Consider two events from the 1970s. In late 1978, Vietnam invaded Cambodia to remove the Khmer Rouge from power and to limit China's control in Southeast Asia. And in 1979, the Tanzanian army deposed Idi Amin in Uganda in order to prevent him from attempting to capture part of Tanzania. Ultimately, Vietnam was criticized more for occupying Cambodia too long than for having overthrown the murderous Pol Pot regime. And although Amin rightly complained that he was attacked before any of his forces had even entered Tanzania, the UN Security Council and then UN Secretary-General Kurt Waldheim simply ignored his complaints, no doubt recognizing the benefits of his removal from power.

Likewise, international responses to unauthorized interventions (preventive and otherwise) have depended on the credibility of the intervening states' justifications. For example, Israel's 1976 operation to rescue Israeli and Jewish passengers from a hijacked plane at Uganda's Entebbe airport was deemed acceptable, since Amin had embraced the cause of the kidnappers and the involvement of Israel was tailored to the narrow objective it had claimed. The United States' 1983 invasion of Grenada, on the other hand -- which was also a hostage-rescue operation, in that over 500 U.S. medical students were being barred from leaving the island -- was widely opposed because in addition to rescuing the students, the United States used the opportunity to oust the country's new leaders and to eliminate all Cuban political influence.

International attitudes toward some categories of actions have changed over time, which further complicates the task of evaluating the acceptability of these actions. For example, the UN Security Council at first condemned Israel's 1981 bombing of Iraq's Osirak nuclear reactor. But a decade later, after Iraq repeatedly attacked other countries and used chemical weapons at home and abroad, Israel's strike received widespread praise. And in 2007, by which time the world had grown increasingly concerned about the proliferation of WMD, Israel's attack on what seemed to be the foundations of a nuclear reactor in Syria was barely criticized.

The use of force to prevent humanitarian disasters (or halt their escalation) also elicits more support today than it would have when the UN Charter was adopted. The movement to establish a "responsibility to protect" reflects a growing acceptance of the need to prevent gross violations of human rights, even those taking place within another country's borders. The 1999 U.S.-led intervention in Kosovo was technically illegal because the UN Security Council did not approve it, but it was sanctioned by NATO and widely supported. Similarly, regional organizations in Africa -- including the Economic Community of West African States, its cease-fire monitoring group, and the Southern African Development Community -- have authorized the removal of coup leaders on the continent to prevent the suffering that these groups have concluded regularly follows the overthrow of democratically elected regimes.

Preventive force, in other words, has been used widely even though it is generally regarded as illegal. This discrepancy poses a challenge for international law, whose strength and credibility depend partly on consistency and objectivity. It would be worthwhile, therefore, to develop criteria for identifying and approving those uses of force that enhance the values enshrined in the UN Charter while devising ways to constrain more effectively those that do not.


The existence of a UN Security Council with the will and the means to prevent threats to international peace and security would help reduce the need for unauthorized preventive action. But although the council has sometimes undertaken specific peacekeeping operations, responded to aggression, and adopted important rules regarding terrorism, the proliferation of WMD, and human rights, it has often failed to enforce its own resolutions to prevent threats from being realized.

One way to remedy such inadequacies is to make sure that the existing rules serve the UN Charter's purposes. Some aspects of the law governing the use of force -- including the concepts of necessity and proportionality -- remain universally accepted. But other aspects are seldom taken seriously and deserve review and amendment. For example, states could read Article 2, paragraph 4, of the charter more literally -- as prohibiting the threat or use of force when it is employed to undermine the territorial integrity of states "or in any other manner inconsistent with the Purposes of the United Nations." They could also recognize an exception to Article 2, paragraph 4, permitting compelling humanitarian interventions or the use of force for objectives approved as proper by the Security Council. Also, the right of self-defense under Article 51 could more often be treated as "inherent" (as is written in the article) and thereby consistent with the historic right to use force for that purpose. And, as suggested by the British Foreign Office, states could understand the meaning of "imminent" to include situations in which a known terrorist group had both the intent to carry out attacks and the ability to do so without being detected. This would expand the concept's meaning to fit a world in which grave threats can be realized at a time and a place impossible to anticipate in advance.

Instead of accepting such changes, which increasingly reflect actual state practice, the ICJ has continued to support and craft rules that effectively protect terrorists, proliferators, and irresponsible states. In 1986, for example, the ICJ rejected El Salvador's claim to self-defense against Nicaragua, which was aiding rebels seeking to undermine the elected Salvadoran government. By doing so, the ICJ effectively denied El Salvador the right to cooperate with the United States in an exercise of collective self-defense. In 2004, the ICJ found that Israel's right of self-defense did not apply to protecting itself against attacks on its population (including suicide bombings) by nonstate actors. This meant that Israel could not legally build a fence or take other self-defense measures on any territory beyond its internationally accepted borders. Similarly, the next year, the court ruled that Uganda had no right to self-defense against rebels attacking it from the Democratic Republic of the Congo because the attackers had not been sent by the Congolese government, even though that government had failed to stop them, as international law requires. These decisions -- along with the ICJ's refusal to treat as precedents those uses of force that the international community has widely accepted as just, such as NATO's intervention in Kosovo -- have undermined, rather than enhanced, the objectives of the UN Charter.

One possible reaction to international law's failure to deal effectively with current threats would be to treat it as irrelevant and regard national self-interest as the only reliable guide for when to use force, preventive or otherwise. But the United States cannot and should not respond this way. It is obliged to abide by the UN Charter's rules, and it has a strong interest in having other states do so as well. Disregarding international law would free all states to act as they please when the goal should be to encourage them to act in ways that advance universally accepted objectives.

The 2004 report of the Secretary General's High-Level Panel on Threats, Challenges and Change suggested a far better approach to enhancing international security. The panel was unwilling to abandon existing international legal norms, but -- recognizing both that the UN Security Council does not always act when it should and that the legality of a use of force does not assure its wisdom or utility -- it proposed that the Security Council adopt "a set of agreed guidelines, going directly not to whether force can legally be used but whether, as a matter of good conscience and good sense, it should be." The panel, in other words, rightly acknowledged that the legitimacy of an action can differ from its legality.

Legality is relevant in determining an action's legitimacy, but so are other values and norms, including the propriety of dealing effectively with substantial threats to charter-based values. The UN Security Council's decision to deny weapons to victims of ethnic and religious abuse in Yugoslavia in the early 1990s, for example, was legal but arguably illegitimate, whereas NATO's unauthorized use of force to prevent abuses in Kosovo was illegal but arguably legitimate.

Using legitimacy as a guide in determining whether to use preventive force would allow states to take into account a broader range of considerations than current international law typically dictates. Such states would weigh not only their own views regarding possible uses of preventive force but also the views of other states, nongovernmental organizations, and knowledgeable or interested groups and individuals (including UN agencies, regional organizations, religious leaders, and others). Legitimacy is not a yes-or-no proposition but a matter of degree; it does not demand the definitive conclusions required of legal opinions.

The concept of legitimacy, moreover, can be made more concrete through the establishment of a process to judge state conduct according to various important standards, such as the seriousness of a perceived threat, the necessity of using force to counter the threat, the proportionality of the force used, the extent of international support for the action in question, the action's consistency with the values of the UN Charter, the strength of the evidence supporting the intervention, and whether the action meets the chief criterion of "just war," that is, whether its expected benefits outweigh its potential costs.

In addition to subjecting potential preventive actions to these standards, states should take procedural steps to help establish the actions' legitimacy. Although preventive actions must sometimes be secret in order to be effective, their causes are often well known and should be discussed within the UN Security Council in advance. In all instances, a state that has taken preventive action should report the action to the council after the fact in order to justify its conduct. And states that use force preventively should accept accountability for errors; they should conduct thorough inquiries of their actions and seriously consider paying compensation to the victims of their mistakes.


Some threats to international peace and security are so potentially damaging that preventing them in advance may be preferable to remedying their effects. Prevention can often be achieved by means short of force (including diplomacy, sanctions, and deterrence), and the unauthorized preventive use of force should be considered only as a last resort, when all alternatives to force have been exhausted and UN Security Council authorization for the use of force cannot be secured. When such a case arises, states should evaluate possible courses of preventive action based on standards of international legitimacy.

Some argue against using the concept of legitimacy to evaluate the use of force on the grounds that its criteria are broad, subjective, and too permissive and so its application would undermine current international law. But relying on legitimacy is unlikely to result in less adherence to international law, since the current use-of-force rules are already routinely disregarded as impractical or unsound. Other critics argue that the notion of legitimacy is no more likely to govern the use of force effectively than are current legal standards. But the utility of legitimacy lies, partly, in the modesty of its claims. Unlike traditional legal arguments, which purport to rely on established rules to vindicate or condemn state behavior, arguments based on legitimacy claim only to guide complicated decision-making by subjecting that process to a survey of the full range of relevant international opinion.

States have much to gain and little, if anything, to lose by subjecting their decisions to use preventive (or other) force to systematic legitimacy tests. Encouraging such disciplined examination should enhance the prospect that states will use preventive force in ways consistent with the goals of the UN Charter -- and that their actions will, thanks to the international support they receive, stand a greater chance of succeeding. A state that disregards this process, on the other hand, is more likely to fail, or to pay higher costs in achieving its objectives.

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
  • ABRAHAM D. SOFAER, George P. Shultz Senior Fellow in Foreign Policy and National Security Affairs at the Hoover Institution at Stanford University, served as Legal Adviser to the U.S. State Department from 1985 to 1990. This essay draws on a report, titled "The Best Defense? Legitimacy and Preventive Force," prepared for the Stanford University Task Force on Preventive Force, which will be published by Hoover Press on February 1, 2010.
  • More By Abraham D. Sofaer