The deployment of missile defences has become one of the first steps to be taken when a NATO Ally is threatened during a crisis. The Alliance, for example, sent Patriot missile defences to Turkey during the 1990-1991 and 2003 conflicts involving Iraq. Moreover, NATO has a history of successful cooperation in missile defence for the protection of its forces in the field, and it is improving this capability. In March 2005 the North Atlantic Council launched the Active Layered Theatre Ballistic Missile Defence (ALTBMD) programme. ALTBMD will integrate various theatre missile defence systems into a coherent network for the protection of deployed forces, with an initial operational capability in 2010.

Now NATO may be on the verge of moving beyond force protection in the Alliance's missile defence capability. When NATO's defence ministers met in June 2006, they took note of the completion of the missile defence feasibility study that the Alliance had launched at the 2002 Prague Summit to examine options for the protection of Alliance territory, forces and population centres.

The question of the technical feasibility of such defences against ballistic missiles has become less salient than the political-military issues, which are still under discussion. Within this debate it is increasingly accepted that "full-spectrum" missile defences - that is, against all missile ranges - could serve Alliance security interests. Despite efforts to discourage and contain missile proliferation, the numbers of ballistic missiles outside NATO (and their range and sophistication) are increasing, and these missiles could be equipped with weapons of mass destruction (WMD). Moreover, terrorists and offensive missiles may constitute overlapping challenges, as Hezbollah's connections with Syria and Iran have demonstrated with short-range systems.

Threat assessment - forecasting the requirement for missile defence - is obviously not an exact science. How serious (and how imminent) are long-range ballistic missile threats in relation to the costs of procuring missile defences and pursuing other defence priorities? The general threat assessment among Allied governments is not fully in agreement with that of the US government, notably with regard to the time-scale on which foreign ballistic missile capabilities may mature. However, it is widely acknowledged that the availability of assets could change rapidly with transfers from a foreign supplier such as North Korea and that intentions could shift overnight with a change of regime. Moreover, even with poor accuracy and uncertain technical reliability, WMD-armed long-range ballistic missiles could pose significant threats to NATO countries.

Missile defences protecting NATO cities and territory could in principle reinforce deterrence, enable the Alliance to navigate crises with greater steadiness and solidarity, and (perhaps in some cases) even dissuade adversaries from acquiring ballistic missiles that could be used to threaten Allied homelands. Missile defences could bolster deterrence because an adversary would face the risk of operational defeat of his attack as well as the prospect of NATO's retaliation. Moreover, with defences against such ballistic missiles, the Allies would be freer to consider options other than preventive or pre-emptive action against adversaries. Such defences could help the Allies reach consensus on strategy in a crisis and hold together during a conflict. Missile defence capabilities could also offer signalling possibilities in a crisis. NATO governments could exercise the option of announcing an increase in the missile defence alert level to send a message of cohesion, determination, and preparedness. The employment of Patriot to protect Israel from Iraqi missile attacks in 1991 demonstrated that missile defences can be employed for purposes of de-escalation, conflict limitation, and crisis management.

The potential advantages have, however, often been lost from sight as experts and officials have debated the many unresolved questions associated with missile defence. These questions include command and control, the prioritisation of scarce missile defence assets, the prospect of continuing (and even deepened) Allied dependence on US capabilities, liability for debris from successful interceptions of WMD-armed enemy missiles, enemy response options, technology transfer, cost, threat assessment, missile defence architecture priorities, and possible reactions by Russia.

Command and control for defences against ballistic missile attack cannot be effectively improvised in the midst of a crisis. The arrangements must be thought through and agreed upon in deliberations far in advance, with clear rules of engagement spelled out for the military commander to whom authority to act in the event of ballistic missile attack would be delegated. In practice, this commander would almost certainly be an American, perhaps the Supreme Allied Commander, Europe (SACEUR). The rules of engagement could be determined through Alliance consultations and the operation of the missile defences could include Allied contributions.

Such command and consultations arrangements could address questions of prioritisation and debris. For instance, how would the Allies determine which assets should be defended with priority in particular contingencies? Would national capitals or the largest population centres be defended first? Would precedence be given to nations in a coalition of NATO allies undertaking a specific operation? What about defending neutral and non-NATO countries? Would some interceptors be dedicated exclusively to the defence of Europe or North America? NATO has recognised these issues, but has not yet reached conclusions. The need to determine priorities for limited defensive assets stands out as an important question for Alliance consultations.

Another command and control issue that requires more analysis is debris, in view of the risk that fragments of an intercepted warhead might fall on the territory of a non-NATO country or a NATO country that was itself not the object of the attack and that had no direct role in the engagement decision. Potential debris damage should be compared with the consequences of a ballistic missile attack conducted as intended. Not one piece of debris from the Columbia shuttle hit a human being, and the shuttle was much bigger than a warhead, which would be (unlike the Columbia shuttle) pulverised into bits by a non-nuclear kinetic interception. In the case of an exoatmospheric intercept, therefore, debris would be likely to burn up entirely during re-entry. Even if the enemy employed a nuclear warhead designed to explode on interception and thereby cause an electro-magnetic pulse (EMP), which is not a simple design task, the damaging effects of a high-altitude EMP burst would (although significant) almost certainly be much less than those of a nuclear attack against a city. Beyond the vast immediate destruction, a successful nuclear attack would produce fallout and impose catastrophic long-term medical, social, and economic costs. Nick Witney, then a British official, may have had these consequences in mind when he wrote in 2003, "No European state, I hope, would refuse to run the risk of a debris shower from a successful interception falling on its territory, if this was the price to be paid for protecting a friend or ally, near or far, from ballistic missile strike."

The Russian offensive deterrent would not be threatened by US or NATO missile defences in Europe, because these defences would involve only a small number of interceptors. Despite criticism from Russian commentators, consultations with Moscow and transparency measures might clear up possible misunderstandings about the purpose and potential functions of these interceptors. In fact, since 2002 NATO and Russia have been engaged in theatre missile defence cooperation and dialogue.

The issue of the response options open to adversaries is related to threat assessment. Ballistic missiles are seen by those pursuing them as useful for various purposes - deterrence, coercion, and prestige, as well as strike options. Ballistic missile defences might nonetheless drive some adversaries to consider other attack options, such as manned aircraft, cruise missiles, unmanned aerial vehicles (UAVs), or bombs hidden in ships in harbours. In the last case, there would be no defence, if maritime surveillance failed, other than the threat of identifying who planted the bomb and (as with declared deterrence policy) punishing the guilty party. Raising the issue of an adversary's circumvention options is a paradoxical tribute to missile defences, in that it implies that these defences might succeed in dissuading the adversary from seeking ballistic missiles. Adversaries determined to rely on ballistic missiles might shop for Russian- and Chinese-developed penetration aids and other countermeasures, but could not be sure of defeating technologies designed to identify and overcome them.

It is precisely because other means of attack are possible that NATO is examining improved defences against manned aircraft, cruise missiles, and UAVs. Except for short-range interceptors, ballistic missile defence is practically useless against cruise missiles. With the proliferation of access to technologies, actual and hypothetical threats are expanding. Allied governments must make choices and keep in mind the risk that investing heavily in defence against one form of attack might leave NATO exposed in another domain.

The most serious cost issue may not be the fact that adversaries have multiple attack options but that the defence spending of most Allied governments is at low levels and these governments have other urgent military priorities. In recent years they have emphasised current operations, particularly in the Balkans and Afghanistan, and transformational issues related to the NATO Response Force. Many Allies would probably be reluctant to commit to common funding of a major new procurement programme, and see advantages in postponing such a decision while Alliance experts grapple with the various unresolved questions, particularly since answers to some of these questions may depend on technological advances. For example, sophisticated discrimination, tracking, and trajectory-projection capabilities may make it possible to design command and control arrangements that would optimise intercept opportunities and minimise the risk of debris.

It is not clear whether at the November 2006 Riga Summit the Allies will go beyond the call for "further work on political-military considerations" made by NATO defence ministers in June 2006. However, the Summit may provide guidance on analytical and practical steps forward, perhaps including the establishment of a NATO Headquarters Missile Defence Centre.

Moreover, it is possible that the United States and a few other NATO Allies will go forward with "full-spectrum" missile defences without waiting for the Alliance as a whole to make a procurement decision. The main precedents for this in NATO theatre missile defence include the SAMP-T programme involving France and Italy; the Patriot programmes involving Germany, the Netherlands, and the United States; and the Medium Extended Air Defence System (MEADS) programme involving Germany, Italy, and the United States. SAMP-T, Patriot and MEADS all defend against shorter-range missiles. To counter long-range missile threats to Europe and North America, the United States may work with some NATO Allies to deploy a missile defence site in Europe. (This would be the third site after the US sites with ground-based interceptor missiles in Alaska and California. It might be more accurately termed "the first European site.") The site with interceptor missiles would be supported by other sites in Europe for sensors and communications, and it could offer protection to much of NATO Europe and North America against long-range ballistic missile threats originating in the Middle East.

If the United States and some other NATO Allies go forward with "full-spectrum" missile defences with a set of bilateral programmes of cooperation, this may accelerate the Alliance's collective decision-making about missile defence. As in some other areas of Alliance policy, all the Allies may well be involved in consultations, even if only a few participate directly in the practical arrangements. NATO communications could link together national, multinational, and Alliance missile defence efforts. Coordination of national programmes and those involving groups of allies with NATO-wide efforts is essential to preserve Alliance unity.

The US is already sharing missile launch information with NATO and other allies. A missile defence arrangement pioneered by the United States and other NATO Allies could give all the Allies insight into capabilities and operational principles, including changing readiness conditions. A US-led missile defence system in cooperation with selected allies could consist of ground-based interceptor missiles networked with other capabilities (such as THAAD, Patriot, and Aegis) to offer a measure of protection to the Alliance as a whole. The system might be paid for mainly by the United States, with the contributions in Europe consisting of funding, real estate, national sensors and missile defence systems, and/or support staff, including for force protection. Full-spectrum protection for NATO European cities and territory might well include significant ALTBMD components, because networking among sensors and interceptors is essential for a comprehensive architecture to defend against short- as well as long-range missiles.

An expression of NATO support for "full-spectrum" missile defence would reassure potential host nations that a US missile defence site in Europe would not be divisive or a source of disputes with other nations. Indeed, such NATO support, along with consultations with Russia, could help to create a positive framework for future missile defence cooperation among the Allies and could be seen by potential adversaries as a sign of Alliance resolve.