The State of Deterrence in International Politics Today

Abstract

In comparison with the Cold War era, deterrence in international politics has changed significantly, even though many of the basic components of that deterrence still exist and continue to have an impact. Deterrence is now less salient in national security policies and international security management, more recessed, particularly nuclear deterrence. This is primarily due to the huge changes in international politics ushered in by the end of the Cold War, particularly in great-power political relationships, and which are continuing to unfold. Important developments are underway with respect to nuclear deterrence, extended deterrence, collective actor deterrence, and other aspects of international system security. While many old topics pertaining to deterrence continue to be studied and generate continuing controversies, often along the same lines as in the past, some important investigations and theoretical analyses have also emerged on pivotal deterrence, the deterrence of cyberattacks, terrorism, and international crime. What is needed most is analysis on how to develop and apply deterrence strategy and practice in new ways as a central contribution to global and regional system security maintenance and management, a function deterrence performed during the Cold War but in ways now largely out of date. Also needed is careful attention to the possibility of the return of more traditional international politics, particularly if this were to occur in relations among the major states, and careful consideration of how best to use deterrence to hedge against such a development.

It is important to take stock of deterrence, in theory and practice, to assess where it is now and where it might be headed in security affairs. When the Cold War ended, analysts suggested almost immediately that the utility, role and conduct of deterrence would be changing. We are now better positioned to see how and where this is the case. Less obvious but soon emphasized was that deterrence theory might need adjustments as well, that it had been significantly shaped by the Cold War environment. Efforts to develop a robust theory had encouraged treatment of it as an abstract phenomenon, as if it would be basically the same everywhere and at any particular time. The impact of context had been underemphasized, even neglected, and the context had markedly shifted.

This paper grapples with a series of tasks. One is to highlight ways in which deterrence, since 1990, has been molded by surrounding conditions and circumstances. Another is to describe where deterrence now stands as a resource in security affairs. This means noting not just changes that have occurred, but the institutional, political, and intellectual inertia behind Cold War deterrence that has continued to be attractive and has been difficult to alter or eliminate. Nevertheless, international politics has changed and we have to ask how this affects or should affect deterrence. What does deterrence, in theory and practice, look like now?

As a quick overview we can say that there is:

•	insufficient appreciation of how and why Cold War conceptions of deterrence are of limited relevance now and also of the ways in which Cold War deterrence thinking remains relevant;
•	significant improvement in conceptions of how deterrence can be applied in anti-terrorism efforts;
•	spreading appreciation of the shrinking role and utility of nuclear deterrence;
•	more attention needed on political and normative contexts in assessing deterrence today;
•	a burgeoning effort to reexamine deterrence in regard to cyber-security concerns;
•	too little attention being given to the complexities of collective deterrence and to the alternative models of collective deterrence;
•	inadequate understanding of extended deterrence today;
•	insufficient use of an arms control perspective on security management in response to rising interdependence in international politics.

Deterrence is an enduring important recourse in international politics. However, its application and specific implementation must be adjusted to major shifts in the global and regional international systems, not just via individual responses of actors to their specific security threats. This is a very challenging task, as these systems are evolving amidst continuing aftershocks set off by the end of the Cold War. We are reshaping an important recourse for maintaining international order even as that order is itself being refashioned; we are altering our tools while we build on the run.

The Past

Because things have changed, assessing where we are requires comments on where we were and what used to be. Deterrence is an old practice. For instance, classic balance of power systems were based on deterrence, applied by actors not just to prevent wars but via wars. Many alternative structures, such as a hegemonic system, a great power concert or a collective security system, have involved security sought and maintained by deterrence – power accumulated by actors singly or collectively (usually in alliances) to threaten serious harmful consequences so as to ward off attacks or other noxious behaviour or, when used, to demonstrate those harmful consequences for the edification of potential opponents. Deterrence was a standard practice, like diplomacy or spying or war, readily employed within an overall security strategy.

This began to change in the first half of the 20th century, as deterrence started to become something more. Efforts were undertaken to better grasp how deterrence works or might work, how it could best be applied, and the kinds of capabilities necessary for maximizing its effectiveness. These efforts were stimulated by rising apprehension about the growing potential lethality and destructiveness of warfare. Deterrence became critical, not for preventing wars of any sort but for preventing those that could be unusually violent and destructive. Rising interest in it as a strategy was apparent even before World War II, well before nuclear weapons appeared, and sharply escalated after it.1 Additional interest was generated when the coming of the Cold War made another vast war seem possible, even likely, and peaceful resolution of the East–West conflict very unlikely.

Traditional deterrence was always a ‘sometime thing’. Often it did not work. At times it made preventing wars more difficult, provoking resentment instead of acquiescence, and instigating attacks or efforts by opponents to design around it.2 It was often difficult to make threats credible, or sufficiently daunting. It was seldom cheap or convenient. Traditional deterrence continued during the Cold War, not only among states caught up in that conflict but among others that were not involved. But with nuclear deterrence as the heart of the major nations' national security strategies, charged with preserving their existence, the need to have it work – at the highest level of warfare, and often at lower levels or with serious conflicts short of war to prevent escalation to the highest level – became overwhelming. We bet our lives, our societies, our civilization (and those of everyone else) on it. The ensuing absence of outright wars among great powers strongly suggested it was working, even though deterrence to prevent lesser conflicts or nasty provocations and challenges was much less successful.

Of significance was the appearance of extended deterrence arrangements through formal and informal alliances, patron–client relationships and other associations. Another feature of the period, present from the start, was the existence of the UN (a non-alliance multilateral actor)3 created in part to practise deterrence through the Security Council. Thus deterrence protection was extended to numerous additional states, and sometimes to actors within states. Extending deterrence became central to international politics, involving alliances, interventions, arms transfers, power projection efforts, military training programmes and non-proliferation pressures.

All this constituted the core of international and national security management for decades. Strategic (in regard to major, especially nuclear, threats) and non-strategic (in regard to lesser threats) applications of deterrence were deeply woven into the fabric of international politics. Far more attention, reliance and resources were devoted to it than any other approach to the problems of war or security that arose from military threats. The Cold War world was seemingly deterrence dependent. Clearly not always successful, deterrence remained attractive; there was always limited support for doing without it and betting on something else.

It was guided in part by wide-ranging analytical efforts, involving an array of experts who, unlike in the past, were security specialists but not military personnel. Given reliance on deterrence and the importance of its central missions, deterrence theory and related studies became a broad field, encompassing such things as the motives and perspectives that incite preparations to use attacks and war in conflicts; things that make deterrence work or fail; how best to apply it, and what to avoid; the circumstances in which it is most properly used or not; and how to reduce chances of deterrence failures, that is, to ‘stabilize’ deterrence.

Cold War deterrence settled into a distinctive pattern with the following key characteristics:

•	Superpower reliance on enormous numbers of vastly destructive weapons – far more than necessary – making it likely a deterrence failure would be catastrophic. Early efforts to avoid such huge arsenals and strategies for shattering enemy societies failed, with smaller nuclear powers adopting counter-value targeting and the superpowers seeking to be able to destroy as many of each other's weapons as possible in, and especially before, a war.4
•	Reliance by other nuclear powers on small but, from all available evidence, sufficient arsenals.
•	Many strategic nuclear weapons, eventually most, kept on constant high alert.5
•	Reliance on threats of retaliation more than defences to deter nuclear or massive conventional attacks, but maintenance of extraordinary levels of standing peacetime conventional forces for deterrence of lesser attacks.
•	Despite profound insecurities and enmities, the emergence of considerable security management via deterrence, using unilateral and cooperative efforts to stabilize it.

Since the Cold War

After 1990 the dominant political relationships in the system became quite different in character. Political relations among leading states have remained relatively moderate and significantly cooperative, remarkably free of profound security concerns. The most potent states have acute conflicts with only a few other actors, if any, and none of them great powers.

As a result deterrence is now far less central and salient. This is most apparent for nuclear weapons and nuclear deterrence. Formerly considered the ultimate recourse for preventing major wars, they have been relegated by most nuclear powers to residual functions, primarily hedging against the possible return of serious conflicts. Deployed nuclear weapons have dropped sharply, as have the total weapons in nuclear arsenals. Significant numbers remain on high alert but with few directly targeted on a clearly designated opponent. Thus the nuclear deterrence aspect of what remains of Cold War security management has frayed considerably, is fragmenting and lacks credibility. This would be fine if there was a clear and broadly satisfactory alternative, but there is not.

Great power conventional forces have also declined considerably, with few units now on high alert. Most are no longer configured as if war is a constant possibility and many lack well-defined missions for clearly anticipated contingencies because of a dearth of severe political conflicts. Deterrence at the conventional level is therefore less salient as well. In general, deterrence is much more recessed. Deterrence capacities certainly exist but there is less call for them, they are less on display, there are fewer deterrence threats issued, and so on. To the great powers, and most other states, deterrence is not an everyday necessity standing between them and disastrous attacks.

Among great powers the glaring exception is the United States, the only actor with a true global military capability, particularly for large operations anywhere on short notice,6 and the only state actively practising extended deterrence on a major scale. Its continuing preoccupation with deterrence is reflected in its wars, alliances, huge military spending and forces, very extensive intelligence resources and operations, and commitment to military superiority. The reasons for all this are unclear. Some see it as a manifestation of American ‘empire’. Others cite the American national security elite as finding ways to retain its power, status and employment.7 A third explanation is that the world remains dangerous due to ongoing nuclear proliferation, terrorist threats, the rise of China, and so on. The most potent explanation is that the United States is highly committed to the idea that international security management is important and feasible, with deterrence as the cornerstone of it, something no one else is ready, willing and able to undertake and which, whatever they may say, many governments count on the United States to provide. The United States is not unhappy being the most powerful state, but it constantly seeks to enlarge its hegemonic coalition and spread some of the burdens.

American military capabilities are the basis for that deterrence. The replacement for Cold War security management is now a Western collective hegemony loosely managed by the United States, with its forces supplying the core deterrence capabilities and other major military resources largely along Cold War lines. The primary reason for considerable continuity here is almost certainly that trying to rapidly rearrange global and regional security management, such as developing new deterrence arrangements, would have escalated interstate suspicion and conflict while inciting political controversies over national security policy in various states. With little inclination to adopt spectacular departures from the status quo, the response was settling for gradual adjustments. The United States kept most of its Cold War military trappings, without the serious threats that had shaped them, for lack of readily available alternatives in which states had great confidence. For example, eastern European states simply parked themselves within the continuing Western security orbit rather than seriously explore alternatives. All this is partly hedging – the bad old days could return, and the Cold War ended so smoothly that the existing Western arrangements could survive basically intact. Astonishing is how, except for China, the great powers have avoided unilateral efforts to sharply enhance their relative military power, and how no clear balance-of-power competition has emerged.

Consequently, the system still displays some Cold War patterns:

•	competitive great power intelligence efforts, along the same lines;
•	the great powers' core nuclear weapons capabilities, despite the absence of political conflicts that make them vital or even useful;
•	the nuclear forces basically configured as they were but on a smaller scale, with new weapons still emerging and existing ones being improved;
•	nuclear forces maintained with the same former opponents' forces in mind, despite it being even more difficult to imagine using them;
•	the main global political fault line remaining as Moscow and Beijing versus the West on many issues, which still inhibits Security Council effectiveness;
•	American alliances remain, with more members, and the network of American military bases is intact, with US forces deployed in many of the same places plus new ones;
•	American conventional forces still substantial, some other great powers maintaining rather large forces, the arms trade still vigorous.
•	American extended deterrence still widespread, based in part on the same power projection resources – bases, communications, maintenance and related facilities.

These holdovers, on a smaller scale, in such a sharply different international system politically, invite assertions that they are inappropriate for security today when major actors are far more open, interactive and interdependent, have greater information flows and much more manageable conflicts. The nature of deterrence and many of its elements, it is said, should be markedly different or even cease to exist. The main threats now are weak states, rogue states or non-states operated in ways or promoted by types of leaders that make them harder to reliably deter. The types of attacks and warfare to deal with are also changing, making a Cold War approach irrelevant. Deterrence in practice is also shifting because of expanded normative constraints on using force. Thus deterrence is outmoded and inadequate for contemporary security challenges.8

This is a somewhat flawed perspective. What particularly makes deterrence more complicated and difficult is its more normal role in today's international politics. It is more of a tactical resource, again, than a security strategy. As such it displays the basic limitations of deterrence evident before the Cold War and often displayed below the nuclear level during it. The threats employed are less frightful and it is harder to make them credible. What is often seen as a greater weakness of deterrence now, such as diverging interests among states,9 is not new and was familiar in the Cold War. One result is that the inability of the international community to make sanctions or threats of force fully effective often makes deterrence appear weak. It is also difficult because opponents more readily design around today's deterrence, so threats must often be enforced to get results, even repeatedly. This is difficult to sustain and can produce iterated low-level conflicts, even enduring rivalries, that become relationships of mutual hurting – military steps used for political posturing, communication and deterrence, making for ‘serial deterrence’ encounters.10 These facets of deterrence were well known in the past, as in the confrontations in Arab–Israeli relations or the American deterrence relationship with North Vietnam.

Where deterrence is often said to be very effective now is in its use, or potential use, by ‘rogue’ states in regard to the United States and others, particularly such a state with nuclear weapons. But even here deterrence has familiar limitations. North Korea may feel it has greater protection now from an attack but its opponents showed little desire to attack the North even before its nuclear tests. And states like North Korea, Iran, Syria, or Russia and China find deterrence of little use in coping with Western pressures (demands, sanctions, threats) that jeopardize their legitimacy and survival by condemning their behaviour on human rights, democracy, proliferation, and so on.

Thus with respect to actually using deterrence, much more interesting and innovative is how the western bloc is no longer status quo-oriented, fending off challenges to the established order. Instead, the West vigorously promotes democracy, open societies and economies, market systems and respect for human rights, in keeping with democratic peace theory and related liberal notions in which spreading those values and practices is the fundamental basis for a peaceful and secure international politics. The United States is particularly insistent, often acting as a revolutionary state of sorts by promoting societal and regime change.11 This poses an existential threat to numerous states and societies, helping incite military build-ups or nuclear weapons programmes in, for instance, Iran, North Korea and Venezuela to deter extending such ‘attacks’ into direct military ones. Deterrence now is often being sought or practised against the West.

Contemporary deterrence is also shaped by other significant contextual shifts. One is continued technological change in everything from weapons to daily living. This includes increased precision in long-range weapons. Using intercontinental ballistic missiles (ICBMs) to deliver highly accurate strategic attacks with conventional warheads is now included in US defence planning. New biological weapons are emerging out of the genetics revolution – some already developed, others on the way. In its implications the biological warfare threat parallels that posed by cyberattacks, in that cyberattack capabilities are also already widespread among states, groups and individuals, they have the potential to deliver serious harm, and they leave theorists hard pressed to show how deterring them can be effective.

An important additional shift is normative. Dominant states mostly subscribe now to norms opposing indiscriminate use of force and inflicting even modest collateral damage (by past standards), norms that blossomed in impact when the Gulf War demonstrated the rising precision in targeting and weapons. An appropriate response to an attack is supposed to be equivalent to it or less in harm inflicted and nature of the forces used (that is, responding to limited cyberattacks in limited cyber fashion).

Also receiving priority are new vulnerabilities, like those exposed in the 9/11 attacks. They complicate deterrence efforts, especially attacks that ignore constraints on using force that Western states are expected to honour. One result is less emphasis on retaliatory threats in favour of pursuing deterrence more by enhanced defences.

Finally, deterrence is frequently undertaken under global or regional multilateral security efforts, such as in the greatly expanded use of peace operations in deterring or suppressing violent internal conflicts, leading to numerous examples of ‘collective-actor deterrence’. The forces involved are multinational, with the decisions emerging from a kaleidoscope of members' political considerations. The targets face complications in trying to deter these threats since they are legitimized, in part, by their multilateral character and may be backed by almost unlimited resources.

Recent Developments in Theory

Deterrence theory was shaped by the intensity of the early Cold War and the grave danger from overwhelmingly destructive capabilities. Now those conditions apply almost entirely to concerns about terrorism being comingled with WMD proliferation, or the possible return of intense interstate conflicts in the Middle East or South Asia. Otherwise, such conflicts are in remission, maybe permanently. How relevant, therefore, is deterrence theory now?

In fact, there are numerous ways in which it can still be helpful. Much of the recent literature is immersed in perennial deterrence topics and problems. There is still debate about whether prior behaviour is important in determining the credibility of actor commitments.12 There are numerous overviews of deterrence during the Cold War, tracing or explaining the theoretical efforts in that period.13 Vigorous criticism is reiterated of assuming rationality in theory and practice, often stimulated now by concern about deterring terrorism, and there are defences of that paradigm.14 There is renewed interest in whether the United States is seeking a first-strike capability.15 The debate on nuclear proliferation and what to do about it has escalated but the issues and main positions are largely unchanged.16 The disputes over ballistic missile defence (BMD) and its potential effects on deterrence continue.

A particularly important aspect of Cold War deterrence was the way it morphed over time into providing broad security management for the international system, including various regional systems, despite deterrence often being depicted as a threat to security management by locking states into deeply competitive positions and military rivalries that inhibited serious diplomatic/political efforts to ease or resolve conflicts and reduce threats. The preoccupation with security management emerged primarily when deterrence theory focused on keeping deterrence stable and embraced arms control as the main solution.17 Arms control involved unilateral actions and cooperative arrangements to, among other things, make deterrence as stable as possible, steps pertaining directly to weapons and forces or to situations and conditions that might put undue stress on deterrence and make it fail. Given the recent shifts in the nature of conflicts, the uses of force and warfare, new theoretical efforts on arms control are needed.

For one, the preoccupation with deterrence stability has shifted. In the Cold War nuclear deterrence was supposedly critical for stability due to perceptions that if not deterred and an opportunity arose, the opponent would attack. It was also critical because only with sufficient security under stable deterrence would states moderate their conflicts, generating political stability. Without conflict reduction, under the stability-instability paradox military stability at the nuclear level could have destabilizing effects at lower levels from immoderate political clashes. Nuclear deterrence was seen as both a substitute and a stimulus for undertaking the proper political management of conflicts.

It does not play that role now, other than in a case like India and Pakistan, as interstate political stability is being more successfully pursued by other means. Where stability is poor and war is a possibility, like the conflicts that swirled around Saddam Hussein's Iraq or today's conflicts with terrorists, nuclear deterrence does not apply (except in the India–Pakistan relationship). Deterrence is needed only at the conventional level or below, and is clearly uneven in impact. What is its role now?18

Here arms control needs more attention. Many arms control techniques and practices are relevant for providing stability during significant weapons cutbacks. On the other hand, overemphasizing arms control can impede progress in promoting efforts like the zero option campaign on nuclear weapons, because of its tendency to stabilize, not erase, existing conflicts and military postures. Arms control theory says little about disarmament. Therefore we need sophisticated arms control analysis on conventional weapons and forces, on efforts to reduce weapons of mass destruction, or on subjects like cooperative approaches to BMD development.

Extended Deterrence

Today, this topic emerges mostly in debates about whether extended deterrence (ED) arrangements that reflect Cold War era concerns are still viable, useful or necessary, and how to deal with the political complications involved in altering them. Broader theoretical and empirical analysis of extended deterrence is necessary today because it remains a key element in regional and global security management. For instance, it plays an important role in non-proliferation efforts. It is what collective actor deterrence, such as by the Security Council, is about. It is at the heart of democratic peace theory as now applied. And in contemporary international politics more states practice ED than people realize.

As noted earlier, extended deterrence was woven into the fabric of Cold War security arrangements, contributing in various ways to the stability they were designed to maintain and offering protection to many states in the formal and informal American and Soviet alliances. The alliances were typically discussed and understood within a realist perspective, particularly Snyder's neorealist theory of alliances,19 leading to predictions after 1990 that they would soon disappear. This turned out to be true for most Soviet alliances and the Chinese-American association/alliance against Moscow, but the other US alliances persisted. Partly with this in mind there have been efforts to craft non-realist conceptions of alliances,20 but they do not focus on extended deterrence or other deterrence relationships. We have various kinds of ED arrangements in and through the alliances, but have done little to adjust theory accordingly.

Most significant alliances today do not involve a major power extending vital protection to a weaker in exchange for the latter's cooperation in military and other matters. While some American allies/associates really need protection, most do not face threats they cannot handle on their own. There is some hedging against a recurrence of serious threats but the alliances primarily serve other purposes, such as the United States:

•	controlling and/or compensating allies on not developing WMD or destabilizing conventional forces;
•	helping certain allies avoid disruptive internal political struggles on security issues;
•	helping to keep allies from being security threats to each other; and
•	helping to structure, with allied assistance, the international environment to enhance security.

The point is still deterrence but often now it is extended deterrence for a collective system security management. This is the extended deterrence being used to try to forestall further WMD proliferation, to deal with the Taliban or to prevent/halt internal conflicts that breed refugee crises, huge human rights violations or failed states. Using alliances this way involves variants of collective actor deterrence and aspects of ‘pivotal deterrence’.21 It rests on cooperation for security, with some governments defining the national interest entirely in realist terms while most do not.22

Thus a different conception of extended deterrence is involved. ED is normally envisioned as one actor projecting deterrence to protect another, but it can also involve projecting deterrence to keep threats geographically far away. The Soviets' Cold War security glacis in Eastern Europe was meant to help deter emergence of a new German threat and, if one arose, to keep as much ground between the Soviet Union and a recovered Germany as possible. Americans made the same argument about communism in Asia in entering the Vietnam War – if not stopped there, the spreading threat would just come closer. Now ED is often put in terms of systemic security: an unstable Middle East would put many states at risk from energy disruptions, terrorism or Islamic fundamentalism. Or nuclear proliferation in one place breeds it in others by force of example, new proliferators cooperating with would-be nuclear states, erosion of the non-proliferation regime, and so on. When ED is pursued in this fashion or to prevent genocide or promote democracy, the concept of the ‘threat’ or ‘attack’ to be deterred – and the concept of deterrence itself – is much altered. Building coalitions for such efforts involves many more states in practising it.

Finally, ED is relevant when democratic peace theory shapes and justifies foreign policies. The theory invites spreading democracy as the best way to enhance security: the international environment becomes less dangerous as stable democracies proliferate. By extension, this promotes the idea, and perhaps mandates, that democracies help defend any well-established democracy under attack, and thus extend deterrence to try to prevent such attacks in the first place. If so, there is an alliance among democracies, whether explicit or not, involving a semi-automatic extended deterrence. Numerous adjustments in thinking about security are required to encompass the complications this entails.

Collective Actor Deterrence

‘Collective actor’ designates a set of states that form a group aimed at promoting the general welfare, as opposed to one serving only the members' needs and interests: international organizations, or ad hoc groups like the one that has negotiated with Iran on its nuclear weapons ambitions. Deterrence theory developed by assuming both a unitary actor (a state) and a rational one (a state, or an individual or small group running a state) making decisions. There was no provision for a collection of states as the decision-maker. Yet this is now common in deterrence situations. We have a fair amount of evidence on how collective actors typically conduct deterrence,23 and deterrence theory should be adjusted to accommodate at least the following.

To start with, collective actor deterrence often reflects, and therefore is shaped by, a strong effort to alter important elements of the international context. Historically, deterrence was pursued in an arena largely beyond law and order and the socialization provided by organized society, particularly in its moral dimensions. It was one of an actor's responses to the ‘anarchy’ characterizing international politics. Collective actor deterrence these days is an extension of, and a tool in upholding, an emerging version of international law and order and the continuing development of norms – transnational, international and domestic – of proper behaviour. This deterrence has increasingly taken on the characteristics, burdens, limitations and legitimacy of domestic deterrence, of deterrence within an established state. The objective is to uphold those norms and in so doing reinforce socialization into behaving in accordance with them, as Lawrence Freedman has described.24 This is more clearly evident in collective actor deterrence than in any other use of deterrence today. And this is why, as noted earlier, deterrence used by many actors now is supposed to operate within greater constraints, something particularly evident in collective actor deterrence.

Next, a collective actor is typically very slow to mount a true deterrence threat. It will often strongly suggest that something being done by the target is highly unacceptable and could justify a strong response, but is loath to make that threat more explicit. One reason is that the challenger is normally not directly attacking or threatening this kind of deterrer. The target is another party, so the collective actor is engaging in extended deterrence, like trying to prevent genocidal activity or keep one state from attacking another. (Exceptions include cases where the collective actor has issued deterrence threats, seen them ignored and takes this as potentially undermining its reputation and credibility. A reoriented NATO was eventually moved to threaten, then use, force in Bosnia and later in Kosovo partly from concern that otherwise it would lose credibility with its members and future targets.)

The delay is also typically due to members disagreeing on whether to pursue serious conflict resolution, on grounds that threats will inflame the situation and make resolving matters more difficult, or get ready to take forceful action if necessary and therefore clearly threaten it in advance. Another factor is that modern collective actors with responsibilities in political-military affairs have been established with emphasis on conflict resolution and the principle of non-intervention, so their inclination is to treat using force, and thus issuing threats, as a last resort. Building a sufficient coalition for doing so is therefore hard; many members will be wary, often due to a strong commitment to sovereignty and a fear of setting precedents on breaching it. Or members line up for and against the parties to the conflict and thus disagree on threats and their enforcement. A strong disagreement about using force will raise concerns that the collective actor will itself be badly damaged by doing so.

Even when members coalesce on issuing threats, there are serious credibility problems. Credibility is at least as important for a collective actor as for a state in a conflict, because members strongly want deterrence to work – they really do prefer to not use force. They often disagree over specific coercive measures, with opposition rising as decisions on this get nearer. Typically, the more harm the proposed action may inflict, the greater the disagreement. Thus the messages sent, including threats, may well be collectively unclear, or when received they may be discounted because of members' disagreements. Mixed messages are often conveyed to encompass member disagreements or via members' private communications with the target. All this invites the target to suspect that the deterrer is not or will not be strongly determined to carry out its threats, particularly if opposed vigorously from the outset.

Finally, the collective actor is likely to implement its deterrence threat unevenly. Not all members will contribute significantly to the action taken. Support for using force or sanctions drops over time as members' domestic political backing for the costs involved declines. Some may continue opposing the coercion, covertly or overtly. Thus collective actors often cannot readily, credibly, threaten to use overwhelming force, a real handicap in trying to deter.

These behaviour patterns violate best practices on how to conduct deterrence. However, there are offsetting factors. One is that a collective actor can be good at combining threats with reassurances if the target obeys, and with the use of incentives. Both will have significant support among the members, and are known to enhance deterrence effectiveness and the probability of success. And a collective actor can more readily claim legitimacy for its threats than a state, especially in going to considerable lengths to avoid using force.

Adjusting theory to include deterrence by collective actors can build on numerous case studies that present a fairly clear picture of its strengths and weaknesses. In addition, adjusting theory might benefit from Crawford's ‘pivotal deterrence’, a theoretical perspective developed for other purposes. Based on case studies, it offers a theory of when and how certain deterrence efforts are likely to be successful.25 In pivotal deterrence a third party seeks to keep two (or more) associates – friends, allies or others considered valuable – from fighting. The theory is that to be successful the pivotal deterrer must have the capabilities to intervene in a war between the parties and determine the outcome, the parties must lack suitable alternative sources of military support to offset this and the deterrer must avoid serious commitment in advance to either party so neither can confidently initiate hostilities.

This is about situations – states headed towards direct combat – which have become rare and hopefully will remain so. However, it seems relevant, and may be crucial, for collective actors dealing with conflicts. Suitably adjusted, pivotal deterrence theory could be of considerable assistance to peace enforcement/imposition efforts in such conflicts, and also invigorate ‘muscular preventive diplomacy’, an endeavour collective actors have been urged to pursue more vigorously.26 The difficulty in this will continue to be getting collective actors to use deterrence like state actors. Whereas collective actors typically eschew serious threats until a conflict verges on fighting or even wait until fighting forces them to act, pivotal deterrence is about preventing fighting. Analysis of how this might be done more effectively could make preventive diplomacy backed by deterrence more workable and appealing. While various states are sometimes involved, only the US seems to have prepared for it fairly consistently (while not always following through), and some outsiders claim that this is Washington in pursuit of its traditional national, even imperial, objectives. However it is undertaken, the more timely use of collective actor deterrence would be very valuable. Literature on peace enforcement and peace imposition is mainly about suppressing violent conflicts, not deterring them; more efforts to change this are needed.

Nuclear Weapons in the Future

While nuclear weapons did not create heavy reliance on deterrence – analysts and military leaders had already been nudged in that direction – they considerably accelerated it. Nuclear weapons came to be considered the ultimate basis for the great powers' survival and the international system's stability, and extended deterrence then made stable security available for numerous other states. Other states explored playing the Cold War blocs against one another to achieve a similar result. All this helped compensate them for losing a possible national route to stable security when the great powers decided nuclear proliferation was destabilizing and sought to bar other states from going nuclear.

However, guiding security matters via, and by exploiting, competing blocs and nuclear deterrence has been unravelling since the Cold War and, not surprisingly, some governments believe they must compensate by obtaining nuclear weapons. Perhaps more will join in this way of thinking. Or the Cold War arrangements may further unravel due to efforts to eliminate nuclear weapons, something most NPT members have long endorsed and which is now reinforced by fears of proliferation to ‘rogue states’ or terrorist groups. Certainly the leading nuclear powers have little inclination and few real plans to ever use nuclear weapons, reflected in how most nuclear arsenals have shrunk. As a result, extended nuclear deterrence is much more tenuous than before, or at least it seems so.

What about a replacement? Will the current relaxed relationships among the great powers continue, so elaborate deterrence-based security management will not be required? If so, and if stringent anti-nuclear proliferation efforts are therefore reinforced, nuclear deterrence will largely die out in the continuing evolution of the international political climate. On the other hand, the return of a seriously conflictual bipolarity or traditional multipolarity would make reliance on deterrence almost ubiquitous and presumably generate considerable nuclear proliferation; many states now have more of the necessary resources. We have not had a serious analysis of what deterrence under true multipolarity would look like since Morton Kaplan's unit veto approach many years ago or Kenneth Waltz's suggestion that nuclear proliferation would be good for eliminating a multipolar system's propensity for spawning warfare.

If the current liberalist version of a proper international system continues to flourish, efforts to give nuclear weapons and nuclear deterrence short shrift will continue. Collective actors would have important roles in efforts to get close to or actually eliminate nuclear weapons, particularly on verification and monitoring, providing security guarantees against cheating, and suppressing or containing interstate conflicts that could provoke nuclear arming or rearming. Theoretical guidance would be needed on how best to conduct deterrence and other efforts against the return of nuclear deterrence, even if actors feel the stakes are high. We have only a marginal grip on this now. Nuclear weapons are not about to disappear, but more progress on that is plausible.

Deterrence in Practice

Nuclear Weapons and Deterrence Now

As noted earlier, the post-Cold War period has displayed a smaller, largely recessed but still potent deterrence structure in security management that reflects its Cold War origins. This elaborate arrangement is threatened in at least two directions. One is the disturbing pressures behind proliferation efforts. Extended deterrence with conventional forces, however provided, is not a fully satisfactory alternative for many states, even some rather powerful ones (South Korea, for example). States can feel that being attacked, then protected, with conventional forces is much too damaging; that is a large part of what they want to deter. And there will always be governments and rulers who want nuclear weapons for reasons of national stature.

Theoretically, nuclear proliferation is bad, but we have no fully satisfactory practical response. There is insufficient agreement among states on how serious it is and on what to do about it. Some do not consider nuclear proliferation a threat if they will not be directly targeted. Some like the additional constraints on great powers that nuclear proliferation could provide. Many studies have therefore offered advice on how to get a suitable consensus for strengthening the non-proliferation regime, which has some deterrence capabilities but hardly makes using them a high priority. Less attention has been given to how to apply deterrence more effectively, particularly by using force. This glaring gap is evident in connection with Iran and North Korea. There is concern that military strikes on either state would not succeed in destroying the nuclear weapons programme facilities. This treats the facilities as the only possible and appropriate targets even though, since proliferation is considered exceedingly dangerous, many other targets might be suitable. In fact, deterrence has never been about just threatening and inflicting harm nicely proportionate to the attacker's actions. It is about threatening enough harm to prevent an attack and then being ready to inflict it. We need more work on ways, military and non-military, to harm productively rather than proportionately and in the most precise, most suitable fashion, particularly for proliferation threats involving terrorists or unusually difficult governments.

The other unravelling element is the drive to eliminate nuclear weapons. Most NPT members have long endorsed this, one argument being that the NPT cannot long survive as a have-and-have-not regime. Another claim is that nuclear weapons are virtually unuseable anyway. Still another is fear of further nuclear proliferation. Regardless of how attractive elimination may be, it will be an incredibly complex matter and has only recently begun to receive intense study from deterrence analysts.27 A serious problem is how to make deterrence of cheating on the elimination of nuclear weapons sufficiently present and credible to reassure governments about abandoning nuclear arsenals or plans for building them, particularly if the potential cheater could be a great power. Part of the problem is maintaining the transparency vital for a comforting deterrence regime along the road to zero when some nuclear powers rely on opacity to hide their deterrence capabilities from attack. Another part is achieving the elaborate verification needed at every stage and extending well beyond zero to cover elements of any ‘virtual arsenal’, to be sure it is not suitable and primed for a quick breakout.28 There is the question of what to do about how nuclear weapons are used by several states (and by others via extended deterrence) to offset disparities in conventional capabilities.29

Do we need a substitute? If so, must it take the form of or at least include deterrence? A related question: what will be the warning signs of an impending demise of the pluralistic security community that now dominates great power relations, of the nuclear taboo or tradition of non-use, and of the current scarcity of other interstate wars? If we do away with nuclear deterrence, how do we know when that is about to become a mistake? Some analysts suggest that fear of being snookered in a zero nuclear deterrence situation could incite a very harsh version of the stability problem that initially shaped the theory of deterrence and arms control: states worried about rearming too late and thus so poised to rearm they enhance others' suspicions and hair-trigger rearming postures too.30

The American Extended Deterrence Gambit

Of more than passing interest is a major US effort to reorganize extended deterrence. After the Vietnam War the US eventually moved into the Revolution in Military Affairs (RMA) and its continuing implementation in the ‘Military Transformation’. In merging revolutions in firepower (in accuracy at great distance), information processing and surveillance (at almost any distance), coupled with greatly improved mobility, the goal was to allow much lighter, very mobile forces to win with few casualties by inflicting minimal casualties and damage while crippling the enemy's entire military effort by hitting his key organizational and communication nodes.

A less advertised goal was to liberate American forces from having to work closely with allied or associated military forces ill prepared to participate in such a war. Over time, however, the US shifted to welcoming its allies' participation for several reasons:

•	to keep the alliances viable and useful through new or newly amplified missions in security management;
•	to enhance the legitimacy of American military operations; and
•	to generate a strong Western expeditionary military capability.

Washington now expects that, under US leadership, this will generate a potent deterrence and bring direct forms of military coercion to bear when and where necessary – deterrence by a collective hegemonic actor.

As a result the United States has been working extensively with allies and associates to customize their forces. The basic function of the alliances and related security guarantees used to be to protect the allies – their forces would primarily mount a potent territorial defence which American forces would reinforce, being stationed for that purpose in or near the allies and backed up by huge American power projection capabilities. Since most of those partners today face no serious threats to their national territories, the United States wants the alliances to provide more military assistance to itself by becoming more suitable for joining American forces in conflicts and interventions around the world. This includes offering advice on military modernization, providing training, joint exercises, selling appropriate military arms and equipment, and so on. The United States discourages cuts in allied military spending and forces, urging increases instead. It opposes allies' efforts, mainly through the EU, to develop separate joint expeditionary units, suggesting these are duplicative and stretch allies' resources too far.

This is an audacious endeavour. The primary objective is a better, more reliable and more useable capacity for collective actor deterrence than the United Nations and most regional security management institutions can mount. It resembles the original plan to have (UN) standing forces for security management rather than cobbling forces together in an ad hoc fashion when needed. While the Western capability serves the UN or other IOs when necessary, it is to be ready when a relevant IO is incapable of suitable action. It is a way around great power vetoes in the Security Council, often used by Beijing and Moscow who are very uncomfortable with interventions in principle, particularly Western-dominated ones. It is also a way to offset regional organizations dragging their feet.

Along with political-military leadership, the United States envisions supplying the key forces for most operations, the better part of the necessary logistics and communications and probably the military command. However, as NATO operations in the Libya revolution in 2011 demonstrated, the United States is ready to insist others take the lead when appropriate. The backbone of all this is an unusual sort of extended deterrence. The payoff for allies and associates includes American assistance in upgrading their forces plus retaining a close association with the United States on security matters, which constitutes indirect additional reaffirmation that American extended deterrence is there for them when necessary.

This is one solution to the limitations of collective actor deterrence, but it is not guaranteed to work. A powerful collective military capability, relatively flexible and mobile, is available, which can take advantage of the regional command structure the United States combatant commands (COCOMs) constitute. What is missing is a suitably ambitious commitment by others to accepting such serious responsibilities and burdens and, in the wake of fighting in Iraq and Afghanistan, a vigorous American willingness to use force extensively, at least for the time being. Allied political systems remain too sceptical, too cognizant of the potential costs and burdens, and too narrowly focused in defining their security to embrace all this vigorously. The current test case, the struggle in Afghanistan against the Taliban and al Qaeda, is instructive. The allies are uncertain about long-term prospects for the operation, and the war has tended to reinforce the mistrust in the American armed forces about working closely with allies in combat. Even so, this is now the world's closest thing to a viable collective-actor standing force for security management, and thus for the collective use of deterrence to uphold the liberalist international order.

Deterrence and Asymmetrical Problems

During the Cold War the focus was on preventing major military attacks. Other possibilities such as spying or sabotage or threats from subversive groups were not tackled under the heading of deterrence. The major military threat – outright attacks by communist governments – was obvious, but the lesser threats had many of the characteristics that today are called ‘new’ and ‘asymmetric’ in:

•	being typically mounted by small groups, rogue states or numerous individuals, greatly multiplying the range of attacks and threats (that is, left wing guerrilla operations with Soviet, Chinese, etc. help; Nasser's operations against American supported governments);
•	involving attackers difficult to identify and locate, making after-attack attribution uncertain;
•	using participants who seemed irrational or unusually motivated (Mao was often so depicted, as were Castro, Mossadegh in Iran, the North Vietnamese in not being daunted by American bombing, and so on);
•	threatening limited harm compared to what deterrence was primarily used for preventing in the Cold War or in classic international politics; and
•	mounting attacks difficult to spot in advance, with resources difficult to detect and destroy.

This once fit guerrilla or other insurgency warfare, particularly in post-World War II anti-colonial uprisings and conflicts, as well as threats from subversive groups (that is, the anarchist movement and terrorist attacks in late 19th to early 20th century Europe.) Even before the Vietnam War analysts described guerrilla warfare as a very non-traditional challenge to regular forces. During the Cold war much was made on both sides of ‘insidious’ threats from subversive forces, and corrupting ideas and propaganda. Today, the comparable concern is about cyberattacks, biological warfare, international crime, terrorism, Islamic fundamentalist insurgencies and their international appeal.31 We didn't see how to readily deter unconventional attacks before and we don't now. This is said to leave us weak and on the defensive, with our main military capabilities, even deterrence itself, outmoded. A more accurate assessment would be that we currently face threats more annoying than deadly, much harder to detect and much more complicated to deter. We give them much attention as the threats of the day and because they might become far more than just annoying.

As a stand-in for all of them we can briefly consider cyberattacks, saying something about the others along the way. Why cyberattacks? They are particularly easy to mount and thus very common; they have huge non-military components in how harm is inflicted, who is involved and who is targeted; they readily move between something like warfare into something more criminal; their impact is often heavily psychological; the range of possible harm is vast, even from individual acts; and they readily exploit modern technology, often in unanticipated ways. The same is often said of all the others, particularly in explaining why they should be considered very important.

The cyber threat is glaring. Attacks are constant, come in vast numbers, and often go undetected. Total harm done is unknown, but enough is apparent to indicate that they are quite costly financially, damaging as espionage and potentially quite harmful as sabotage. One reason they seem so problematic, like various Cold War era threats, is the open nature of Western societies. Various analyses have been mounted on how deterrence might apply.32

Compared to other asymmetric threats, cyberattacks most resemble international crime and the two overlap a good deal. They involve constant, widespread attacks by many kinds of people – dozens of states have some cyberattack capabilities and most attacks are mounted by individuals and small groups. They are particularly difficult to detect in advance, and almost all attackers seem unmoved by penalties for getting caught – the attacks are too appealing, pay too well, lead too rarely to being punished, and so on. Often attackers act at the behest of others even more difficult to uncover, apprehend and punish. Attribution of ultimate responsibility is difficult, so precisely targeting retaliation is usually impossible.

Thus retaliation is less valuable than deterrence by defence, and the complaint is that existing defences are porous. Not that defences don't work; the great majority of attacks are fended off. But far too many attacks work, particularly the sophisticated ones which often do the most harm. Attackers constantly outrun defenders, adapting more rapidly to emerging technology or changing circumstances, learning more quickly from mistakes (just like terrorists or international criminals). Thus there is fear (as in the Cold War) that the enemy will make dangerous technological breakthroughs and that a first (cyber) strike to do great damage and erase most retaliatory capabilities will soon be possible. It would be extremely difficult to detect in advance – there would be few traces (unless the attack required prior probes or test attacks). That would leave little time to mount a deterrence campaign and issue threats, and the attack could be a shocking surprise. This could readily generate severe reciprocal-fear-of-surprise-attack problems, with opponents guessing about each other's capabilities and whether an attack is coming, each fearing the other is on the verge of gaining a crucial technological edge.

One important response has been healthy interest in seeking cooperation to contain the problem, like agreements on states' responsibilities for curbing or halting attacks originating in their territories, or joint efforts to repair current weaknesses in cyberspace that facilitate attacks. Cooperation could involve collective responses to particularly nasty attacks, sharing information and expertise, imitating what is done when new and very dangerous diseases appear. However, it will be hard to obtain and may come at a hefty price. Numerous states will seize the opportunity to impose tighter control over access to their portion of cyberspace, not just to fend off attacks but to censor it, legitimizing what they are already trying to do for political reasons.

Cyberattacks need careful attention by deterrence specialists. The difficulties are familiar, the complications are mostly those discussed earlier as part of today's more normal deterrence. Deterrence analysts know a good deal about what can be done to make deterrence more effective and when deterrence is likely to be particularly difficult. They should weigh in on two particular aspects. One is the attribution problem. A relevant analogy is how states regularly have private entities violating international trade rules, or violate the rules themselves, and can then be sued for permitting this or simply suffer retaliation (a ruling makes retaliatory measures legitimate). Thus the US might follow up its complaints about Chinese thefts of intellectual property by imposing economic sanctions. This approach could be used on cyberattacks that steal intellectual property, especially since the technology for tracing attacks back to computers of origin, or at least their general location, is improving.

The second aspect is concern that retaliation not be disproportionate. The argument is that disproportionate retaliation can readily provoke escalation, especially since cyberattacks often spread much further than the initial targets (the Stuxnet worm was aimed at Iran's violating the non-proliferation regime but spread to several other countries). Using different means to retaliate (such as destroying buildings at a school for hackers) is seen as likely to be just an escalation, inviting a further response. Again, this is not a new problem. Unwillingness to retaliate with nuclear weapons to a non-nuclear attack appeared early in the nuclear age. Objections to using nuclear weapons in the Korean War were intense, and President Truman told various people that they simply could not be used like normal weapons. Later, this helped shape the tradition of non-use, or the ‘nuclear taboo’, based in part on fear of setting a precedent that makes using nuclear weapons much easier.33

However, there is no reason to extend such limitations to responses for every attack at the conventional level. Retaliation need not be at an equivalent level nor confined to similar weapons and targets. Deterrence theory was not crafted along those lines; in fact it could not be. Deterrence is about threatening unacceptable harm, but there is no reason that the attacker's conception of unacceptable is on a par with (or less than) the level of harm he inflicted. Since he chose to attack, it is probable that his willingness to suffer harm is greater than the defender's, since an attack can lead to potentially serious costs for the attacker and presumably there was some consideration given, as part of the decision to attack, to whether the costs could be borne if necessary. In the Cold War, concern was often expressed about how communist governments seemed more willing to suffer casualties than the West. The response was not to shift to retaliating only in proportion to the attack and limiting that to roughly the same means. Today, economic sanctions are commonly used even when they are blunt instruments, harming many innocents while leaving the actual perpetrators of the unacceptable behaviour unscathed, and doing more indiscriminate harm than the target initially delivered.

Tailoring Deterrence

Recently, the idea of linking retaliation and threats of retaliation closely to attacks became prominent in another way, namely in the effort to apply what American forces refer to as ‘tailored deterrence’. The basic idea is familiar – to have deterrence efforts fit the nature of the opponent (not the nature of the attack per se). A core debate on the US nuclear posture at one point in the Cold War was about whether US deterrence could fail because Soviet leaders' conceptions of ‘unacceptable’ losses from American retaliation were much higher than Americans believed. Mao's government was considered difficult to deter because of Mao's high risk propensity and willingness to pay a heavy price for success, which worried officials in both Washington and Moscow.

When deterrence at all levels was added to STRATCOM's mission after the Cold War, its response was to begin trying to characterize governments or other actors and their decision-makers along numerous dimensions believed to bear on their vulnerability to deterrence, and to assess on that basis which deterrence steps – in combination with other means of influence – would most likely work. This is meant to refine deterrence strategy: determining which threats and other means to use, in what proportion, when and how.34 It went well beyond the initial concept of tailored deterrence, which suggested the means used in deterrence be matched to the opponent's decision calculus and the particular conflict situation, that the deterrence capabilities used reflect the opponent's nature and situation and that the objectives in using deterrence reflect appropriate moral limits.35

This is an understandable but flawed inclination. Getting away from ‘one size fits all’ is fine, but not via profoundly complicated analyses of opponents and situations that require too much information of the sort rarely available, or reducing threats to a seeming precision that is actually superficial, derived from a complex matrix built on dubious assumptions about what will effectively pressure people in another culture with an alien perspective.36 However, within realistic limits tailored deterrence could be important in appealing to US allies through its focus on limits to harmful responses, be more attractive to outside observers and fit better with today's more ‘complex’ deterrence.

Conclusion

Much of the discussion, indeed worry, about deterrence now has to do with familiar concerns: credibility, target rationality, fitting deterrence carefully to the situation (tailoring), possible/actual first-strike capabilities, and so on. More analysis is needed stressing what became, over time, the most important dimension of Cold War deterrence – system security management. This is management for a world now trying to reorder international politics and security to make deterrence steadily less relevant, with that management resting much less on nuclear deterrence in particular, and with ordinary deterrence more prominent and often meant to help sharply improve international politics.

We must think intensively about deterrence at a time when, for many states, insecurity is domestic in origin and often amounts to fear arising from lacking or losing legitimacy. Western pressure often exacerbates that fear, driving states to resist Western norms and desires via steps which, in turn, drive Western deterrence efforts. This is a major deterrence conundrum. How do we design deterrence for a hegemonic coalition when the coalition's overarching intent is at least as threatening to others as its capacity to do harm militarily? How to use deterrence when the concept of the ‘attacker’ in this situation is problematic (is Iran's nuclear programme in itself an ‘attack’ on the West?), when the deterrence is seen by others as an aggressive threat, and when their response is to seek capabilities the West sees as its most dangerous threats today?

Finally, concern that deterrence could come to the fore again in relations among major states is unavoidable. The United States is elaborately hedging against a possible sharp conflict with China, and recent Chinese behaviour has made many governments uneasy. American steps include refashioning and enhancing its alliances in East Asia, shifting more forces there, building links to associates like India, making overtures to countries like Vietnam, and the like. This goes beyond ‘recessed deterrence’ but it may be unavoidable. Having failed to see the collapse of the Cold War coming, we must be more alert in spotting something like it if and when it is coming back.

Notes

George Quester and Patrick Morgan, ‘How History and the Geopolitical Context Shape Deterrence’, in George P. Shultz, Sidney D. Drell and James E. Goodby (eds), Deterrence: Its Past and Future (Stanford, CA: Hoover Institution Press, 2011), pp. 1–45.

Alexander George and Richard Smoke, Deterrence in American Foreign Policy: Theory and Practice (New York: Columbia University Press, 1974). On the nature of deterrence failures see, for example, Stephen G. Walker and Akan Malici, U.S. Presidents and Foreign Policy Mistakes (Stanford, CA: Stanford University Press, 2011), pp. 53–77.

Alliances were typical collective actors pursuing deterrence; today the UN, the OAS, the African Union, etc. and ad hoc coalitions (like the one created prior to and to conduct the Gulf War) regularly pursue deterrence, often in internal conflicts. Hence the term ‘non-alliance collective actors’.

An unusual source on the superpower preoccupation with pre-emptive strikes is Valery E. Yarynich, C3: Nuclear Command, Control Cooperation (Washington, DC: Center for Defense Information, 2003).

Bruce Blair, The Logic of Accidental Nuclear War (Washington, DC: Brookings Institution, 1993), pp. 111–65.

Other states where deterrence seems a daily necessity include Iran, the two Koreas, Pakistan, India, Syria and Israel. China might be included, but it remains unclear how concerned China is about being attacked as opposed to expanding its forces for other purposes.

For example, Ted Galen Carpenter, A Search for Enemies: America's Alliances after the Cold War (Washington, DC: Cato Institute, 1992).

Such assertions are reviewed and critiqued in Elbridge Colby, ‘Restoring Deterrence’, Orbis, Vol. 51, No. 3 (Summer 2007), pp. 413–28.

See, for example, Michael Mazarr and James Goodby, ‘Redefining the Role of Deterrence’, in Shultz, Drell and Goodby (eds), Deterrence: Its Past and Future (note 1), pp. 99–129.

Described in Patrick M. Morgan, Deterrence Now (Cambridge: Cambridge University Press, 2003), pp. 266–7.

Classically, a revolutionary state in international relations seeks to transform the system structure to its advantage; the US seeks state domestic transformations to make a favourable system more peaceful and manageable. However, John M. Owen claims that major powers have typically attempted to spread their domestic values and arrangements, so the US is not unique. See Owen, The Clash of Ideas in World Politics: Transnational Networks, States, and Regime Changes, 1510–2010 (Princeton, NJ: Princeton University Press, 2010).

For examples see: Daryl Press, Calculating Credibility: How Leaders Assess Military Threats (Ithaca, NY: Cornell University Press, 2005); Ann Sartori, Deterrence by Diplomacy (Princeton, NJ: Princeton University Press, 2005).

For instance, Jonathan Stevenson, Learning From the Cold War: Rebuilding America's Strategic Vision in the 21st Century (New York: Penguin Books, 2008); several chapters in Jorn Gjelstad and Olav Njolstad (eds), Nuclear Rivalry and International Order (Thousand Oaks, CA: Sage, 1996); David E. Hoffman, The Dead Hand: The Untold Story of the Cold War Arms Race and Its Dangerous Legacy (New York: Anchor Books, 2009); James E. Goodby, At the Borderline of Armageddon: How American Presidents Managed the Atom Bomb (Lanham, MD: Rowman and Littlefield, 2006); James S. Pasley, ‘Chicken Pax Atomica: The Cold War Stability of Nuclear Deterrence’, Journal of International and Area Studies, Vol. 15, No. 2 ( 2008), pp. 21–39; several chapters in Morgan, Deterrence Now (note 10); Edward Ifft, ‘Deterrence, Blackmail, Friendly Persuasion’, Defense and Security Analysis, Vol. 23, No. 3 (September 2007), pp. 237–56; Keith B. Payne, The Fallacies of Cold War Deterrence and a New Direction (Louisville, KY: The University Press of Kentucky, 2001).

For examples see: Chapter 2 in Morgan, Deterrence Now (note 10); Janice Stein, ‘Rational Deterrence against “Irrational” Adversaries?’, in T. V. Paul, Patrick M. Morgan and James J. Wirtz (eds), Complex Deterrence: Strategy in the Global Age (Chicago, IL: University of Chicago Press, 2009), pp. 58–82; Bradley A. Thayer, ‘Thinking about Nuclear Deterrence Theory: Why Evolutionary Psychology Undermines Its Rational Actor Assumption’, Comparative Strategy, Vol. 26, No. 4 (2007), pp. 311–23; Frank P. Harvey and Patrick James, ‘Deterrence and Compellence in Iraq, 1991–2003: Lessons for a Complex Paradigm’, in Paul et al. (eds), Complex Deterrence (note 14), pp. 222–56.

Daryl Press and Keir Lieber, ‘The Rise of US Nuclear Primacy’, Foreign Affairs, Vol. 85, No. 2 (March/April 2006), pp. 42–54; Daryl Press and Keir Lieber, ‘The Nukes We Need’, Foreign Affairs, Vol. 88, No. 6 (November/December 2009), pp. 39–51; Jan Lodal, James M. Acton, Hans M. Kristensen, Mathew McKinzie and Ivan Oelrich, responses to Press and Lieber in ‘Second Strike: Is the U.S. Arsenal Outmoded?’, Foreign Affairs, Vol. 89, No. 2 (March/April 2010), pp. 145–52.

For examples see: Michael Horowitz, ‘The Spread of Nuclear Weapons and International Conflict: Does Experience Matter?’, Journal of Conflict Resolution, Vol. 53, No. 2 (April 2009), pp. 234–57; Christoph Bertram, ‘Rethinking Iran: From Confrontation to Cooperation’, Chaillot Paper No. 110 (Paris: EU Institute for Security Studies, 2008). For examples of the problem see: Hans Blix, Disarming Iraq (New York: Pantheon, 2004); Mike Chinoy, Meltdown: The Inside Story of the North Korean Nuclear Crisis (New York: St. Martin's Press, 2008).

There was always concern about the need to supplement deterrence with negotiations, détente, building interactions across conflicts and so on, because of its limitations. Eliciting security management from deterrence itself was a major breakthrough.

The most original approach to this is Lawrence Freedman, Deterrence (Cambridge: Polity Press, 2004).

Glenn H. Snyder, Alliance Politics (Ithaca, NY: Cornell University Press, 1997).

Such as Robert B. McCalla, ‘NATO's Persistence after the Cold War’, International Organization, Vol. 50, No. 3 (Summer 1996), pp. 445–75.

See Timothy W. Crawford, Pivotal Deterrence: Third-Party Statecraft and the Pursuit of Peace (Ithaca, NY: Cornell University Press, 2003).

Cooperation in deterrence for security management is quite possible among governments along realist lines, but the nature of the parties' association is then very different.

See, for example, Peter Viggo Jakobsen, Western Use of Coercive Diplomacy after the Cold War: A Challenge for Theory and Practice (New York: St. Martin's Press, 1998); Steven L. Burg, ‘Coercive Diplomacy in the Balkans: The U.S. Use of Force in Bosnia and Kosovo’, in Robert J. Art and Patrick M. Cronin (eds), The United States and Coercive Diplomacy (Washington, DC: US Institute of Peace, 2003), pp. 57–118.

Freedman, Deterrence (note 18).

Crawford, Pivotal Deterrence: (note 21).

See, for example, I. William Zartman, Cowardly Lions: Missed Opportunities to Prevent Deadly Conflict and State Collapse (Boulder, CO: Lynne Rienner, 2005); Michael S. Lund, Preventing Violent Conflicts: A Strategy for Preventive Diplomacy (Washington, DC: US Institute of Peace, 1996).

For examples of studies and debates on the future of nuclear weapons, particularly with regard to deterrence, see: Shultz, Drell and Goodby (eds), Deterrence: Its Past and Future (note 1); European Union Institute for Security Studies, Nuclear Weapons after the 2010 NPT Review Conference - Chaillot Papers No. 120. (Paris: EU Institute for Security Studies, 2010); European Union Institute for Security Studies, Nuclear Weapons: A New Great Debate - Chaillot Papers No. 48 (Paris: EU Institute for Security Studies, 2001).

In the Cold War small nuclear arsenals, in themselves or as remnants left after an opponent's attempted first strike, were the ultimate guarantee against being victimized by cheating or a breakout and they would be unavailable after true zero was reached (for complying states). On virtual arsenals see Michael Mazarr, ‘Virtual Nuclear Arsenals’, Survival, Vol. 37, No. 3 (2006), pp. 7–26. On other aspects of this see David Holloway, ‘Deterrence and Enforcement in a World Free of Nuclear Weapons’, in Shultz, Drell and Goodby (eds), Deterrence: Its Past and Future (note 1), pp. 335–71; and James M. Acton, Edward Ifft and John McLaughlin, ‘Arms Control and Deterrence’, in Shultz, Drell and Goodby (eds), Deterrence: Its Past and Future (note 1), pp. 279–324.

On moving towards eliminating nuclear weapons and the complications involved see: David Cortright and Raimo Vayrynen, Towards Nuclear Zero (International Institute for Strategic Studies – London) (London: Routledge, 2010); George P. Shultz, William Perry, Henry Kissinger, and Sam Nunn, ‘A World Free of Nuclear Weapons’, Wall Street Journal, 4 January 2007; George P. Shultz, Stephen P. Andreason, Sidney D. Drell and James Goodby (eds), Reykjavik Revisited: Steps Toward a World Free of Nuclear Weapons (Stanford, CA: Stanford University Press, 2008); Oliver Meier and Paul Ingram, ‘A Nuclear Posture Review for NATO’, Arms Control Today, Vol. 40, No. 8 (October 2010), pp. 8–15; Barry M. Blechman and Alexander K. Bollfrass (eds), Elements of a Nuclear Disarmament Treaty (Washington, DC: Stimson Center, 2010); George Perkovich and James Acton, ‘Abolishing Nuclear Weapons’, Adelphi Paper No. 396 (London: International Institute for Strategic Studies, 2008); Harald Muller, ‘Enforcing Zero. Forget Deterrence!’, in Shultz, Drell and Goodby (eds), Deterrence: Its Past and Future (note 1), pp. 373–82. For a more optimistic view of eliminating nuclear weapons see: Barry M. Blechman and Alexander K. Bollfrass, ‘Zero Nuclear Weapons: The Pragmatic Path to Security’, Nonproliferation Review, Vol. 17, No. 3 (November 2010), pp. 569–75.

See Christopher A. Ford, ‘Playing for Time on the Edge of the Apocalypse: Maximizing Decision Time for Nuclear Leaders’, in Shultz, Drell and Goodby (eds), Deterrence: Its Past and Future (note 1), pp. 217–77.

The literature on deterrence and terrorism is extensive. Examples of influential contributions include: Paul K. Davis and Brian Michael Jenkins, Deterrence and Influence in Counterterrorism: A Component in the War on Al Qaeda (A RAND Report) (Santa Monica, CA: RAND, 2002); James H. Lebovic, Deterring International Terrorism: US National Security Policy after 9/11 (London: Routledge, 2007); Paul Kapur, Dangerous Deterrent: Nuclear Weapons Proliferation and Conflicts in South Asia (Stanford, CA: Stanford University Press, 2007); Paul Kapur, ‘Deterring Nuclear Terrorists’, in Paul, Morgan and Wirtz (eds), Complex Deterrence (note 14), pp. 109–30. For a nice overview see Jeffrey W. Knopf, ‘The Fourth Wave in Deterrence Research’, Contemporary Security Policy, Vol. 31, No. 1 (April 2010), pp. 1–33.

For examples see: Richard A. Clarke and Robert K. Knake, Cyber War: The Next Threat to National Security and What To Do About It (New York: HarperCollins, 2010); Martin Libicki, Cyberdeterrence and Cyberwar (Santa Monica, CA: RAND, 2009); Patrick M. Morgan, ‘Applicability of Traditional Deterrence Concepts and Theory to the Cyber Realm’, in Deterring Cyberattacks: Informing Strategies and Developing Options for U.S. Policy (Washington, DC: The National Academies Press, 2010), pp. 55–76; Richard Kugler, ‘Deterrence of Cyber Attacks’, Franklin D. Kramer, Stuart H. Starr, and Larry K. Wentz (eds), Cyberpower and National Security (Washington, DC: National Defense University Press and Potomic Books, 2009), pp. 309–40.

See Nina Tannenwald, The Nuclear Taboo: The United States and the Non-use of Nuclear Weapons since 1945 (New York: Cambridge University Press, 2008); T.V. Paul, The Tradition of Nonuse of Nuclear Weapons (Stanford, CA: Stanford University Press, 2009).

See ‘Deterrence Operations Joint Operating Concept Version 20’ (Omaha, NE: STRATCOM, December 2006).

Elaine Bunn, ‘Can Deterrence Be Tailored?’, Strategic Forum, No. 225 (January 2007), Institute for National Strategic Studies, National Defense University.

On tailored deterrence see: Karl-Heinz Kamp and David S. Yost (eds), NATO and 21st Century Deterrence (Rome: NATO Defense College, 2009); Jeffrey S. Lantis, ‘Strategic Culture and Tailored Deterrence: Bridging the Gap Between Theory and Practice’, Contemporary Security Policy, Vol. 30, No. 3 (December 2009), pp. 467–85; Knopf, ‘The Fourth Wave in Deterrence Research’ (note 31), pp. 1–33; Patrick M. Morgan, ‘Evaluating Tailored Deterrence’, in Kamp and Yost (eds), ibid. pp. 32–49.