1. Introduction

"Extended deterrence," or "active deterrence," as it is sometimes called, threatens a nuclear-strategic response in case of a nuclear attack on the territory or troops of one's allies. This paper aims to explore the possibilities of extended deterrence in the Middle East in light of an Iranian nuclear military capability. Two preliminary remarks are necessary in order to frame the line of reasoning on the issue.

First, discussion of the possibilities and pitfalls of extended deterrence in the Middle East does not intend to insinuate that diplomatic efforts to stop the Iranian regime from constructing a nuclear device have failed or that a nuclear Iran is already a given. Exploring the possibilities of extended deterrence in the Middle East is, rather, an attempt to be intellectually honest and anticipate that all the efforts underway for almost a decade will fail because the Iranian regime is determined to produce nuclear warheads or reach the breakout point, in which it will become a “virtual nuclear power.” Either trajectory will have a decisive impact on the nuclear realm, but even more so, on the political balance of power in the region; each has the potential to reshuffle relations not only between Iran and Israel but also between Iran and the Arab states in the Middle East. If either development is perceived as detrimental to the already fragile security situation in the Middle East, academics and practitioners must start thinking about a “Plan B”.

The second preliminary remark that must precede any analysis of extended deterrence and its applicability to the Middle East concerns the nature of the subject to be explored. Although for about six decades there has been a profusion of literature on the mechanisms of deterrence and extended deterrence (in conjunction with the same number of critical studies on why deterrence and extended deterrence might not work),^[1] Lawrence Freedman, Deterrence (Cambridge: Polity Press, 2004). we still don’t know much about these two concepts. This paradox can be explained by the simple fact that so far we have not experienced the failure of a deterrence relationship, i.e., resulting in a nuclear war between two powers. Proponents and opponents of deterrence believe – in the theological sense of the word – that deterrence either works or doesn’t, respectively, but neither camp knows for certain. The consequences of this highly unsatisfying state of the art is that neither the “more may be better” nor the “dead end of deterrence”^[2] Kenneth N. Waltz, “The Spread of Nuclear Weapons: More May Be Better,” Adelphi Papers 171 (London: International Institute for Strategic Studies, 1981), p. 15. approach provides any form of guidance for policymakers. If academics want to speak truth to power they need to be aware of, first, the limitations of their theories, and second, that the real world can’t be grasped with parsimonious concepts.

With these words of caution the essay proceeds as follows: It first approaches the topic by defining extended deterrence, which in the twenty-first century is much broader in its instruments than the old East-West conflict concept. The essay then argues that for extended deterrence to work it must be able to resolve three conceptual problems related to the credibility of a threat. After this conceptual clarification, the essay introduces the two extended deterrence models familiar from past and present, namely, the European and the Asian models. They differ slightly but decisively. The purpose of presenting these two models is to determine if they are applicable to the Middle East, and it will be shown that for different reasons neither the European nor the Asian model seems to be a viable approach.^[3] Bruno Tertrais, "In Defense of Deterrence: The Relevance, Morality and Cost-Effectiveness of Nuclear Weapons," (Paris: Institut français des relations internationales, 2011). The last section of this paper looks at different ways deterrence can be extended to the Middle East. It argues that for the time being only unilateral US guarantees can pave the way for something that comes close to extended deterrence in this highly volatile region.

2.The Definition and Difficulties of Extended Deterrence

During the Cold War extended deterrence was a public good provided by the US and the USSR to some of their allies. Usually extended deterrence materialized in a system of formal alliance relationships among states, with either the US or the USSR as a guarantor. At the time, extended deterrence was mainly nuclear. Stretching a nuclear umbrella over allies served two purposes: first, preventing allies from going nuclear themselves, and second, preventing an adversary from attacking an ally (either in a conventional or a nuclear strike). It might seem surprising that extended deterrence is also mentioned here as a tool against conventional aggression, but in the early days of the Cold War, NATO’s strategy of massive retaliation threatened the USSR and its allies with a nuclear attack in the case of conventional aggression. Extended nuclear deterrence as an instrument against conventional aggression is essential if the opponent is perceived as a predatory, revisionist state that wants to shift existing balances of power to its own advantage by all available means.

The main purpose of extending nuclear guarantees, however, was to prevent the proliferation of weapons of mass destruction. With the demise of the Soviet Union, the concept of extended deterrence occupied a smaller academic focus, and especially with the rise of violent non-state actors, the question has arisen whether deterrence, and thereby also extended deterrence, plays any role in the security politics of the twenty-first century.^[4]Tertrais, “In Defense of Deterrence”.  Interestingly, this academic debate is out of sync with political reality. In light of existing or emerging nuclear powers, states in Asia as well as in the Middle East are exploring the possibilities of sheltering under a renewed or new nuclear umbrella to gain more security vis-à-vis a potential nuclear threat.

Today, extended deterrence is only partly nuclear; it also entails missile defense and, to a certain extent, means such as prompt global strike (PGS) capabilities.^[5] Prompt global strike is a US effort to develop a system that can deliver a precision conventional weapon strike anywhere in the world within one hour, in a similar manner to a nuclear Intercontinental Ballistic Missile ( ICBM). If provided to prevent a nuclear attack on an ally, extended deterrence rests on a mix of instruments that make it at least theoretically possible to tailor it more precisely to regional needs or to the needs of the guarantor and the guarantee.

3.Conceptual Problems of Extended Deterrence

If we turn to the question of essential prerequisites for viable extended deterrence, it must be kept in mind that three problems must be solved before the concept can be considered valid.

a.In extended deterrence, there must be a credible threat to an adversary on behalf of or in collaboration with a third party:

Virtually all efforts at extended deterrence face credibility problems, which is probably a major source of collective actor difficulties on this score, rather than the nature of a specific actor being an issue. Striking a credible balance between the identical imperatives of pursuing a vision of a world free of nuclear weapons and that of maintaining credible extended deterrence in the eyes of US allies poses political challenges to third parties. Since credibility is in the eye of the beholder, the credibility of the balance being struck will depend on the audience and a changing international security environment. Allies will be closely watching the US’s efforts to balance the imperatives of pursuing global nuclear disarmament and maintaining effective deterrence as long as nuclear weapons exist. Moreover, as long as this balance is managed effectively, these seemingly contradictory objectives are actually mutually reinforcing, because credible and effective extended deterrence commitments will provide stability at sharply reduced levels of nuclear weapons, which is a necessary waypoint to the elimination of all nuclear weapons.

The credibility of extended deterrence currently depends on stationing American forces in Europe, which must be balanced by forces from the European states. If the European states were to fail in their part of the bargain  they could end up without the protection of American extended deterrence. The alternative would be providing a purely European deterrent, although there is little likelihood of this at presentsince most arab states do not want Sub strategic weapons stationed on their soil. To strike this balance, the US must consistently educate the public about the importance of extended deterrence to maintain adequate support for the capabilities, actions, and statements of intent that are necessary for credibility.^[6] George Fink, ed. Stress of War, Conflict and Disaster (Amsterdam: Elsevier, 2010), 76.

b. The elite of the guarantor and the guarantee must be convinced on a bipartisan basis that extended deterrence is credible:

As a state moves along a continuum from outright aggression to terrorism, deterrence becomes harder for a collective actor. In essence, consensus for deterrence is weaker when the challenge is less clear-cut and the guilty party is harder to ascertain.  Most of the important alliance dynamics pose true dilemmas, with factors pulling in opposite directions. Depending on the political context, any single adjustment in an alliance can undercut either deterrent threats or assurances. Keeping smaller allies weak and dependent does more than increase the burden of the alliance on the alliance leader;  this strategy can undermine deterrence for the alliance as a whole if it makes it seem weak or irresolute, particularly in the regions where the weaker allies reside.^[7] Fink, Stress, 76.  Efforts to get allies to do more, however, can trigger negative reactions among allies and adversaries alike.For example, a more robust military posture by a local ally as part of burden sharing can send provocative signals to regional adversaries about long-term dangers posed by the ally in question and the alliance as a whole. Such signals might undercut coercive diplomacy if an adversary’s perception of an increase in the ally’s power over time undermines the alliance’s efforts at reassurance.^[8] Brian H. Reid, The Science of War: Back to First Principles (Burlington: Routledge Taylor & Francis Group, 1993), 34. At other times, the dilemma might cut in the other direction. If the leader appears increasingly reliant on newly mobilized allies, this apparent need for military help can undercut deterrence by making the leader seem too weak or irresolute to get involved directly in a local conflict. To the degree that it looks like the leader needs its weaker allies to fight, its own ability to mobilize effectively for war might appear to be reduced, thus undermining deterrence for the overall alliance. But even should each of the actors be deterrable by the costs of war, they must also have sufficient respect for their opponent’s resolve in order to make their retaliatory threats credible. Credibility in part reflects interests. Sometimes, the US’s interests are so high that its credibility is not in doubt. At other times, its interests are so low that its reputation cannot enhance its credibility regardless of its capabilities. In between lurks the extended deterrence problem, where uncertainty about the US’s interests and resolve can determine the level of credibility of the threats.

c.The domestic audiences of the guarantor and the guarantee must believe that extended deterrence is necessary and practicable:

On a broader scale, the long-term viability of the nonproliferation treaty regime may hinge on the credibility of the existing official nuclear powers.^[9] Frank Zagare and Marc D. Kigour, Perfect Deterrence (New York: Cambridge University Press, 2000), 112. The main challenge with extended deterrence is to make retaliatory threats credible against a nuclear-armed opponent, as extended deterrence is in direct conflict with the requirements for stable central deterrence. Two main problems facing US strategists during the Cold War were how to deter a nuclear attack on the homeland and how to deter a conventional or nuclear attack on US allies. To illumine further, extended deterrence was extended in two senses: extended to less-vital US interests like protecting allies, and extended to cover non-nuclear attacks against these interests. Developing plausible strategies for each type of deterrence and resolving tensions between the requirements for each preoccupied American strategic thinkers throughout the Cold War.

Western Europe was of two minds about flexible response. They welcomed US nuclear guarantees for extended deterrence but were fearful that nuclear escalation would leave their homeland a smoking irradiated ruin, while the US and the Soviet Union stopped short of attacks against each other’s homelands. To address this fear, western Europeans argued in favor of deploying US ballistic missiles and ground launched cruise missiles in Europe in the late 1970s.

These intermediate range nuclear forces had the capability to strike the Soviet homeland from European soil. If the USSR was attacked by US intermediate range forces stationed in Europe, the Soviet Union would retaliate against the US homeland, or so the argument went. These forces coupled the vulnerability of Europe to the vulnerability of the superpower homelands, thereby ensuring that nuclear war would not be confined to Europe. By deploying intermediate-range nuclear forces in Europe, US policy tried to strike a balance between deterrence and assurance.

4. Extended Deterrence in the Middle East: Difficult but Possible?

With the nature of and the conceptual problems that accompany extended deterrence as a background, the paper now examines the applicability of extended deterrence to the Middle East given an Iranian nuclear capability. Broadly speaking, there are two familiar models of extended deterrence in the twenty-first century: the European and the Asian models. Both models rest on a significant number of conventional ground, air, and naval forces stationed in the respective regions. They differ with regard to the forward deployment of non-strategic nuclear weapons (NSNW). While nuclear deterrence to US allies in Asia is provided through capabilities stationed in the US, the European model rests on the forward deployment of NSNW as well as a form of nuclear participation within NATO.^[10] Clark A. Murdock and Jessica M. Yeats, Exploring the Nuclear Posture Implications of Extended Deterrence and Assurance: Workshop Proceedings and Key Takeaways (Washington, DC: CSIS Press, 2010).

At first glance, both models are not applicable to the Middle East. It is neither thinkable that Arab countries will accept the deployment of US forces on their soil (especially given the anti-US sentiments among large parts of the population), nor is it likely that the US will deploy NSNW in the region (given the volatility of existing regimes). Although large numbers of US ground, air, and naval forces are already stationed in the Persian Gulf, extended deterrence rests on a country-based strategy, meaning that in every country that enjoys a nuclear umbrella, a tactical link such as US installations or US troops must be present. In today’s political climate, it is hard to envision US forces stationed in Egypt or Jordan. Indeed, the already existing US ground presence in some countries of the Arabian Peninsula is a constant source of tension between the leaderships of those countries and their populations. As long as the population is not convinced that such a presence is needed to guarantee national sovereignty and survival, the credibility of extended deterrence is weakened.

Thus if both models are not applicable to the region, how can extended deterrence be tailored to the Middle East?

The answer depends partly on solving four “known unknowns.” First, how will a nuclear Iran behave? Will it be a defensive status quo or an offensive revisionist power? Second, how can extended deterrence be provided to the region given the Arab-Israeli divide? Third, given their security cultures, will the Arab states and/or Israel trust external guarantees? And last, if Tehran develops long-range delivery systems, how can Iran be made to believe that the US will follow through on its commitments?

Based on these unknowns, four models on how extended deterrence can be guaranteed for the region are plausible: a multilateral agreement, a regional security system, the Holocaust declaration, and unilateral US guarantees. The paper discusses each model  with regard to its applicability.

One possibility of providing the region with a kind of extended deterrence entails the great nuclear P5 powers ( china, Russia, the US, France , and Great Britain) declaring their willingness and readiness to defend Israel and the Arab states, by nuclear means if necessary, if Iran attacks. Together with a declared willingness to use PGS capabilities and the Israeli Arrow system, this form of guarantee could either be provided by a joint P5 declaration or a Russian-US statement on the Middle East and nuclear weapons. At first glance this option looks appealing, since the top three nuclear powers would pool their capabilities and send a clear and strong signal into the region. Even if Russia and China currently object to stronger sanctions than currently exist (not to speak of military action) against Iran, they both share a strategic interest in no nuclear escalation in the Middle East. From a mid-term perspective, it is possible that these three countries, together with the two European nuclear powers, would be willing to extend their deterring capabilities to the Middle East.

Such an option, however, would face an enormous credibility gap, which makes it unlikely to materialize. The likelihood that Israel would consider such a guarantee as credible must be considered extremely low. The option of multilateral guarantees might be appealing to some or all Arab states in the Middle East, but given Israel’s historical record with Russia and France and the current behavior of China and Russia vis-à-vis the Iranian file, it is hard to imagine that the Israeli elite as well as public opinion would perceive such guarantees as credible. Multilateral agreements would also give Iran an opportunity to try to drive a wedge among those countries that would provide extended deterrence to the Middle East. The conclusion, therefore, is that multilateral agreements provided by the P5 or by a Russian-US consortium could not be implemented due to a lack of credibility.

A veteran idea that is frequently aired when it comes to Middle Eastern security is that of a regional security system. With regard to the purpose of extending deterrence, such a system would include the Arab states and Israel as well as external powers such as the US, and possibly Russia. Participants in such a system would commit themselves to defend any member of the system attacked by an outsider through all available means (nuclear, PGS, and missile defense). Such an arrangement would look very much like a formal alliance. A regional security system could be designed as single-purpose (exclusively against the external threat posed by a nuclear Iran) or multi-purpose (trying to create interdependencies among signatory states in the field of security). Although the theoretical literature on building alliances suggests that given an external threat, alliance building is possible even among states that have enmities, it seems unlikely that Arab states would be willing to form an institutionalized regional security system to oppose the Iranian threat. Furthermore, if issues between Israel and the Arab states were not be settled beforehand, such a system would always have a high degree of instability, and intra-system balancing would impede its credibility in the eyes of the Iranian regime.

Charles Krauthammer has proposed the so-called “Holocaust declaration”^[11] Charles Krauthammer, "The Holocaust Declaration," Washington Post, April 11, 2008, p. 8. as one form of extending deterrence to parts of the Middle East. Within this framework, the US would state unilaterally that it would not allow a second Holocaust to take place, meaning that the US would be willing to use nuclear weapons to prevent Iran from exterminating the Jewish state. This kind of unilateral extended deterrence just for Israel would face two major obstacles. First, it would single out Israel as the only state in the Middle East of concern to the US and thereby potentially have a detrimental effect on US-Arab relations, and second, the Israeli elite might feel limited in its freedom to maneuver vis-à-vis Iran and beyond.

These three models on how to extend deterrence to the Middle East suffer from logical flaws given the political reality in the region. Currently, the major obstacles for establishing an overall (meaning including Israel and the Arab states) system of extended deterrence are the lack of trust among Arab states and Israel, and Arab security cultures, which make it hard to believe that Arab leaders and the Arab street could be convinced that the US would defend them in case of an Iranian assault.

Realistically speaking, the creation of a comprehensive and credible system of extended deterrence must start from unilateral US statements to Israel and the Arab states that the US will not allow any other country to blackmail or threaten its allies in the region. This means of extending unilateral deterrence guarantees is far from perfect. It is weak in the sense that there will be no link between the strategic nuclear capabilities of the US and the security of its allies in the region (as in the case of Europe or Asian countries) because of the opposition to US forces in those states. It will suffer from the basic credibility problem of extended deterrence,^[12] Vesna Danilovic, "The Sources of Threat Credibility in Extended Deterrence," Journal of Conflict Resolution 45, No. 3 (2001): 341-69. which Charles de Gaulle captured so precisely in the 1960s when he asked Konrad Adenauer if the German chancellor really believed that the US would risk the destruction of New York for the liberation of Hamburg. The credibility problem nowadays has become even worse since the current US administration has shifted its attention to the Pacific and does not seem too determined to stop Iran “by all means necessary” from going nuclear. Added to this, US credibility and its commitment to get tough on Iran if the mullah regime, once nuclear, threatens US allies, might suffer from the fact that the US has lost two conventional wars in the broader region (Iraq and Afghanistan) and public opinion does not support getting bogged down again in the Middle Eastern quagmire.

But given the aforementioned obstacles facing other forms of extended deterrence in the Middle East, unilateral guarantees might currently be the only form of extending deterrence to the region. Those who point to the fact that Israel has sufficient deterrence capabilities of its own and does not need any kind of extended deterrence^[13]Shen Dingli, “Extended Nuclear Deterrence: Fading Fast," Interpreter, February 3, 2011, http://www.lowyinterpreter.org/post/2011/02/03/Extended-deterrence-fading-fast.aspx. are right from a purely military perspective, but utterly wrong given the political signal sent to Iran if the US extended its deterrence only to Arab states. This signal could be interpreted by the political and religious leadership in Tehran as a crack in US-Israeli relations and as an isolation of Israel in the Middle East. In turn, such a policy could cause Iran to step up its aggressive provocations (via its proxies in the region) to below the threshold of a direct attack against the Jewish state. For political reasons, it would thus be necessary for the US to also extend its deterrence to Israel.

5.Analysis of the Four Models and their Applicability in the Middle East: A Critical Overview

To critically evaluate and augment the four models discussed above, it is fundamental to pose the following questions: To what extent should or must US deterrence strategy depend on multilateral or collective security? Can US military supremacy be brought to bear without coalition support? Will the ability of the US to act decisively unilaterally drive states to oppose it in order to retain some level of strategic independence? Indeed, the question of the role of nuclear weapons in extended deterrence and the level of multilateralism required in US strategy have stirred up great controversy in the policy analysis community.^[14]Air Force Research Institute. Deterrence in the Twenty-first Century: Proceedings (London: Creative Commons CCO, 2010), 138. Conversely, there is a more general agreement on the necessary criteria for US forces to underwrite a US deterrence strategy. In whatever form they may take, efforts to bolster the US’s extended deterrence commitments against Iran are likely to persist for two reasons beyond the obvious one. First, doing so offers a plausible alternative to the certain costs, and almost certain failure, of the preventive war option. Second, it is hoped that US security commitments can dissuade Middle East allies from pursuing their own nuclear options. On a broader scale, the longer-term viability of the nonproliferation treaty regime may hinge on the credibility of the nuclear powers’s positive assurances to non-nuclear weapons states that they will be protected from nuclear aggression and coercion. For example, Israel’s confidence in the US’s statements and actions regarding preventing Iran from acquiring nuclear weapons and coping with a nuclear Iran will make it less likely for Israel to feel compelled to pre-emptively attack Iran’s nuclear facilities. Recent developments, particularly in missile defense cooperation, appear to be helping in this regard. In a stark change of tone, Israel seems to exhibit some strength based on its nuclear capabilities, the assumption that the US would stand behind it if it came under attack, and the calculation that enough of the country’s air bases and military facilities would survive a first strike to retaliate effectively. Moreover, it is publicly acknowledged that Israel’s nuclearresponse would make it politically difficult for Arab states to remain non-nuclear, thus unleashing a possible number of defensive forms: a bilateral defense treaty, a joint congressional resolution, an executive agreement, or a presidential declaration. The broader, more public, and more formalized the security guarantee, the greater its deterrent value, but also the greater the obstacles to the two sides’ abilities to reach an agreement.

Extended deterrence thinking has also penetrated into the growth areas of international security, especially under the aegis of collective-actor deterrence. For instance, in the heyday of the 1990s, the new peacekeeping UN Security Council passed numerous resolutions concerning safe areas in the former Yugoslavia and elsewhere that explicitly tied deployment of peacekeepers to the goal of deterring attacks on civilian enclaves including, Srebrenica.^[15] Antony H. Cordesman  and Adam C. Seitz,  Iranian Weapons of Mass Destruction: The Birth of a Regional Nuclear Arms Race? (California: ABC-CLIO, 2009), 315. The frequently heard claims that the Rwandan genocide could have been averted by a timely deployment of just a few thousand robust UN peacekeepers rests on an implicit and heroic assumption about the effectiveness of extended deterrence. In recent years, from Eastern Europe to the Middle East, Africa, Central Asia, and Southeast Asia, the US has recruited new formal and informal allies, sinking money into these partners in the form of concrete based infrastructure, military-to-military training and exercises, and preferential arms sales, investing in prestige and political capital.^[16]Air Force Research Institute, Deterrence, 139. The conventional wisdom is that these efforts to shape the international environment through alliances and a forward presence will tend to discourage regional aggression; in other words, it will tend to encourage general deterrence.

If the only areas of conflict between Iran and the US were threats posed to each other’s territory, deterrence alone might be sufficient. Iran certainly might commit money and effort to build nuclear weapons for status quo purposes, as the UK, China, and France have done. However, Iran is more likely to try to exploit the political value of nuclear weapons to jockey for advantage in other areas. The Iranian perception of itself as a natural leader in the Gulf, a cultural hegemon in the Middle East, and a challenger to the US presence and role as protector suggest that Iran would want to use the clout of its nuclear force to further its aims beyond the defense of its homeland.^[17] Murdock and Yeats, Exploring the Nuclear Posture Implications, 35. While there is global uncertainty surrounding how Iran may evolve politically in the near to medium term, Iran’s overall national security interests are broadly supported by its political elite and a large section of its population. These interests involve ensuring the survival of the current regime by deterring a US invasion of Iran, protecting the homeland against all external threats, and maintaining and expanding Iran’s influence and power in the Middle East and beyond.

Most participants in Murdock and Yeats’ study (commissioned by the Center for Strategic and International Studies (CSIS)) supported continued efforts to prevent Iran from going nuclear, though there was little support for preventive US or Israel military action. Moreover, study participants rejected the view that continued pursuit of a diplomatic solution is inconsistent with discussing options for how the US and its allies could cope with a nuclear Iran. To the contrary, such discussions could bolster the international community’s negotiating leverage by shaping Iran’s cost-benefit calculations. The Gulf Cooperation Council, in particular Saudi Arabia, would like to benefit from US extended deterrence and assurance.^[18] T.V. Paul,  Patrick M. Morgan and James J. Wirtz, eds., Complex Deterrence Strategy in the Global Age (Chicago: University of Chicago Press, 2009), 55. General stability in the Middle East is important, but Saudi Arabia, because of its oil reserves, is the only country whose independence and survival is of vital interest to the US. Assurances to Egypt are often discussed (in part because Egypt has the most advanced civilian nuclear program among the potential Arab recipients) as well as arrangements with Jordan, Iraq, Afghanistan, and Syria, but because the political and credibility-related challenges would be amplified in each of those cases, Saudi Arabia is the litmus test for the spread of nuclear weapons if Iran will acquire a military capability.

Lack of the public support for deterrence also cautions against overly suggestive, high-level statements on extended nuclear deterrence in the Middle East in the media. Whereas the formalization or explicit nature of an assurance commitment has been identified elsewhere as a factor of assurance strengthening, unintended fallout may cause those reassurance mechanisms to backfire. Were such comments to provoke a prolonged domestic debate, the American public’s lack of resolve would be projected internationally and the damage to US security commitments globally would exceed the value of the public reassurance.

Existing and potential extended deterrence and assurance relationships fall along a spectrum, encompassing varying degrees of formality, transparency, clarity, and relevance to US nuclear weapons. The election of President Ahmadinejad in Iran in 2005 and the progress of Iran’s nuclear program since then have made military options more plausible and prominent.^[19]Paul, Morgan and Wirtz, eds., Complex Deterrence, 56. At the same time, however, growing military difficulties in Iraq make the idea of another major American military effort in the Middle East less credible, and the unilateralist and militaristic image of the US that developed after the 2003 invasion of Iraq further constrains Washington.

6. Conclusion

Extended deterrence consists of merits and demerits based on a variety of factors inherent in both parties. I summarize the advantages and challenges of extended deterrence, as shown in Figure 1.

Figure 1: Pros and cons of providing new or strengthened extended nuclear deterrence guarantees

Unilateral declarations by the US to be willing to extend its deterrence to the Middle East is the weakest form of extended deterrence, but currently the only option that appears at all realistic. In the mid-term (assuming that Iran goes nuclear) a more credible and stable system of extended deterrence for the region will be needed. Such a system might be composed of unilateral Israeli capabilities, multilateral security agreements between Israel and the Arab states, and US nuclear guarantees for all members. There is still some time, however, before such a system must be in place.