The debate on the preemptive doctrine has dominated international law since the Bush Administration's National Security Strategy was published in autumn 2002. So far the discussion has been unhelpfully polarized, between those who defend the legal status-quo at all costs and those who contend that the law is outdated since the challenges of our era make preemption a necessity. The time is now ripe for a reassessment of these arguments.

Preemption is, in itself, neither irrational nor immoral. As already observed by Henry Kissinger in 1959, the strategic incentive to preempt the enemy has been intensified by technological change that has made a surprise attack more feasible and infinitely more destructive than in the past. At the same time, of course, the preemptive incentive is a potential danger for two main reasons. Firstly, the consequences of a mistake would be disastrous. Secondly, it can be easily exploited as a pretext for aggression. Ever since nuclear weapons were invented, national security strategists have had to deal with these conflicting considerations: the rationality of the incentive to act preemptively, and the dangers inherent to preemption.

In the days of the Cold War, the preemptive incentive was offset by a synergy of political doctrines and legal instruments. The most powerful device proved to be what Philip Bobbit calls an ‘intellectual creation': the doctrine of deterrence and its corollary, the notion of mutually assured destruction. Deterrence was the pivotal element in the system, but legal principles and instruments buttressed it. Firstly, central to the legal regulation of the use of force is its prohibition in article 2 (4) of the UN Charter, which places the surprise attacker squarely outside the boundaries of legality. Secondly, various bilateral agreements, most importantly strategic arms reduction treaties, were concluded between the Soviet Union and the US, in order to reduce the risk of nuclear war as a result of ‘accident, miscalculation or surprise attack'. Thirdly, the legal regime on non-proliferation played an important role. Since the 1970s, its main instruments have been the Nuclear Non-Proliferation Treaty (NPT) and the Biological Weapons Convention (BWC), with the addition of the Chemical Weapons Convention (CWC) in the 1990s. Moreover, various portions of territory - including the South Pacific, Latin America, Antarctica, outer space and the moon - have been declared nuclear weapons-free in other treaties.

The starting point for a critical analysis of the preemptive doctrine should be the crisis of this system based on deterrence. Deterrence is, to some extent, a victim of its own success. During the Cold War decades, the system worked well to avoid a nuclear conflict between the two blocs. Nowadays, although it would be far-fetched to say that deterrence is obsolete, its role in the avoidance of conflict between the two major nuclear powers has changed - one of the considerations that probably caused the Bush administration to withdraw from the Anti-Ballistic Missile Treaty.

But the system of deterrence shows its limits, especially vis-à-vis terrorist groups, in respect of which legal and political means of coercion and persuasion are not effective. Terrorists are not subject to inspection and arms control, and their aim is precisely to launch as devastating a surprise attack against civilians as possible. Moreover, the terrorists' attempt to obtain weapons of mass destruction (WMD) has given their non-deterrability an altogether new dimension. The preemptive doctrine is therefore a response to this new phenomenon in a context in which proliferation of WMD technology has gained momentum, with some 35 countries capable at present of developing nuclear weapons according to the International Atomic Energy Agency (IAEA). The central argument in the National Security Strategy was based precisely on the need to adapt the law to a changed reality.

The real challenge is therefore to identify ways of dealing with this threat that are effective from a national security point of view and that can counterbalance the strategic need for preemption. As mentioned, there is a close link between, on the one hand, proliferation of WMDs and the consequent increase in the risk of terrorist groups obtaining them, and, on the other, the resurgence of preemption in the national security strategy of those states that are the most likely targets of the terrorists. The non-proliferation regime is at present beleaguered by weak enforcement mechanisms. Compliance of state parties to the NPT, for example, is based on the implementation of the safeguards agreements, which state parties have to conclude with the IAEA. The limits of this system have been shown by the recent crises with North Korea and Iran. In the case of the CWC, monitoring occurs mainly through inspections, although the Conference of State Parties ‘may recommend collective measures… in conformity with international law'. The Biological Weapons Convention does not yet have even a proper verification system in place.

The current treaty regime on proliferation is also limited by the possibility of withdrawal. The three main agreements contain almost identical provisions that affirm that a state party ‘shall in exercising its natural sovereignty have the right to withdraw from the Convention if it decides that extraordinary events… have jeopardized the supreme interests of its country.' This is the provision that North Korea invoked when it withdrew from the NPT on January 10, 2003. Finally, the non-proliferation regime ultimately depends on the state becoming a party to these agreements. Although the vast majority of states have done so, there might still be states to which these obligations and the scant monitoring mechanisms available are not even applicable. For example, Syria, which featured in the US administration's list of countries suspected of pursuing a biological weapons programme, is not yet a party to the BWC.

Outside the non-proliferation regime, there is the possibility, at least in principle, for the Security Council to act under Chapter VII, which accords it the power to authorise collective enforcement if it determines that a threat to the peace and security exists. However, the Council's record on non-proliferation is all but encouraging. In addition, as the main target of a non-conventional terrorist attack, it would simply be unreasonable to expect US administrations, with all the military might they possess, to delegate the most pressing issue on their national security agenda to the Security Council, with its history of indecisiveness, division and ineptitude. As the UN Secretary General himself admitted, the international community cannot become ‘a suicide pact': if rules and mechanisms do not exist, or if they do not operate effectively, ‘states will resort to other means to reduce or eliminate threats to their way of life - or to their very existence'.

The limits of the non-proliferation regime have already prompted the US and ten allies, including the UK, France and Germany, to adopt the Proliferation Security Initiative (PSI), which aims to prevent the transport and delivery of weapons of mass destruction.

A step further is needed. International law should move towards allowing the use of force, as a last resort, to enforce compliance with non-proliferation agreements. Systematically non-compliant states should face the prospect of the ultimate sanction available under international law, particularly if they cooperate with terrorist groups. Forcible counter-proliferation would not solve all the security problems of our era, but it would constitute another important instrument in the global security toolkit. While not excluding it, forcible counter-proliferation would have some important advantages if compared with pre-emption. Firstly, counter-proliferation by force does stand a chance of becoming part of general international law. The pre-emptive doctrine, on the other hand, has been rejected by the vast majority of states, and is viewed with concern even by some of the US' closest allies, for example the UK. It also proved to be of no use legally in the case of Iraq, when the UK and the US chose to rely instead on a more orthodox argument based on the enforcement of Security Council resolutions.

Secondly, forcible counter-proliferation would be part of a multilateral system - the non-proliferation regime - and, as such, would be subject to further checks and balances. This should not deter the US, because, in case of paralysis in the multilateral system, it would give states prepared to intervene unilaterally for compelling national security reasons a (or another) legal basis for doing so.

Furthermore, the express provision of the ultimate sanction - the use of military force - against systematically non-compliant states would constitute an incentive for the multilateral system to work effectively. Indeed, while the recent breakthrough on WMD in Libya shows that important results can still be obtained through diplomacy, it is clear that this was made possible, or certainly accelerated, by the US and UK intervention in Iraq. The emergence of a system of counter-proliferation could catalyze similar reactions and thus help diplomatic efforts: the more serious are the legal and political consequences of non-compliance, the more likely diplomatic efforts are to succeed.

Forcible counter-proliferation could become part of international law in three ways: firstly, through an amendment to the UN Charter; secondly, through a resolution of the Security Council, adopted under Chapter VII; thirdly, through an amendment to the proliferation treaties. Each of these routes has its advantages and disadvantages. The process for amending the UN Charter would be lengthy, but it would signify a constitutional moment for the international community, with states solemnly accepting these obligations as a fundamental component of their membership of the community of nations.