Defensive realism: Charles Glaser and Robert Jervis

An alternative to the pessimism of offensive realism has been developed by authors such as Charles Glaser and Robert Jervis and is often termed defensive realism. This approach agrees that anarchy is the primary characteristic of the international state system and a highly significant—though, importantly, not the sole—driver of great power behaviour in that arena. For example, defensive realism gives much more analytical space than offensive realism to the role domestic politics plays in state decision-making. In addition, whilst this approach accepts that anarchy encourages conflictual behaviour, it also emphasises that it prevents states from achieving goals that are in their ‘common interest’.40 For Jervis, the difference between the two realisms is thus that, whilst offensive realists ‘see aggression and expansionism as omnipresent’ or ‘believe that security requires expansion’, defensive realists believe that

much of international politics is a Prisoners’ Dilemma or a more complex security dilemma. The desire to gain mixes with the need for protection; much of statecraft consists of structuring situations so that states can maximize their common interests.41

The key point Jervis identifies here in relation to the security dilemma is that it may pertain in some but not all situations and can vary in intensity, being more or less ‘vicious’, depending on the circumstances.42 For Glaser meanwhile, defensive realism has identified variables that cause the ‘variation in countries’ behaviour’.43 In The Security Dilemma Revisited, he draws on Jervis’s work to explain ‘how the magnitude and nature of the security

Theories of nuclear possessionldisarmament 39 dilemma’ depends on ‘two variables’, namely ‘the offense-defense balance and offense-defense differentiation’.44 Importantly, a world where it is possible to differentiate ‘between offensive and defensive systems’ allows, for Jervis, ‘a way out of the security dilemma’.45 Yet in the past, as he also notes, the difficulty of doing this has meant that arms control treaties ‘have been rare’, something that is also partly due to states ‘not always’ being ‘willing to guarantee the security of others’.46

In addition, Glaser, who prefers the label ‘contingent realist’ for his ideas, given that the availability of security for states depends on ‘empirical assessments of the offense-defense balance’, notes that information sharing is important to help distinguish between ‘greedy states’, i.e. those with ‘motives beyond security’ and ‘security seekers’. According to this analysis, ‘the magnitude of the security dilemma’ is influenced by the ‘extent of the adversary’s greed ... and of the adversary’s unit-level knowledge of the state’s motives’. Glaser develops this point by proposing that, in cases where ‘democracies are believed not to have greedy motives’, a democracy engaged in a ‘military buildup’ won’t be seen to be as threatening as an authoritarian state acting in the same way. The result for him is thus that ‘the democracy faces a less severe security dilemma; and interactions between democracies could result in a democratic peace instead of intense competition’.47

The idea of defence-offence differentiation also informs Jervis’s argument that the nuclear revolution ‘sapped’ the driving force of the security dilemma—and was thus of major historical importance—by making defence central and overcoming the supremacy of offence.48 More precisely, Jervis sees the advent of nuclear weapons as marking a shift from ‘defense to deterrence’, in the context of which, ‘offensive weapons are those that provide defense’.49 The expansionist ambitions of greedy great powers therefore became impossible given the dangers of MAD. For him, the ‘high cost of war’, which nuclear weapons help ensure, has thus today contributed to the creation of a security community ‘among the leading states’, i.e. the US, Western Europe and Japan.50 Moreover, Jervis claims that nuclear strategy cannot work as, despite the fact that ‘at best’ nuclear possession ‘will keep the nuclear peace’, it will ‘not prevent and, indeed, may even facilitate—the use of lower levels of violence’, so that ‘military victory is impossible’.51 According to this logic, MAD may therefore weaken the security dilemma at the level of nuclear weaponry, but not necessarily at the level of conventional weaponry, depending on the policies pursued.

Despite the comments outlined, Jervis has also stated that he is ‘deeply ambivalent’ about the costs and benefits of the nuclear revolution. He therefore notes that it has brought both ‘great security and insecurity’, because ‘on the one hand, mutual second-strike capability means that a major war is extremely unlikely; on the other hand, it means that if such a conflict should erupt, it is likely to destroy our civilization’.52 This raises the question of what policies will prevent such a conflict erupting? In response, Jervis outlines several scenarios, one of which ‘in the nuclear era’ is where the superpowers relyon invulnerable submarine launched ballistic missiles, anti-submarine weapon technology is ‘not up to its task’ and ‘limited nuclear options’ are ‘not taken seriously’.53 Despite such scenarios promising stability and peace, nuclear conflict remains possible, with the ‘basic question’ of whether the nuclear revolution enhances security or not remaining ‘unanswered, if not unanswerable’ so that, for Jervis, ‘it is not surprising that so many arguments rage’.54

Moving on to the question of nuclear disarmament, as with Herz writing in the 1950s, Jervis writing in the 1980s could not see the development of the political conditions supportive of nuclear disarmament in the near future. The priority for the latter during the Cold War was thus a restrained security approach to ensure results that were mutually acceptable for the great powers.55 Moreover, in terms of precisely why meaningful disarmament processes are gridlocked, Jervis makes the important observation that ‘to the extent that America’s major interest is in preserving the status quo, nuclear weapons have brought the United States a real, although nonmilitary, victory’.56 Owing to the fact that the US has security concerns beyond its national territory—which, according to Glaser’s definition, would make it a ‘greedy’ state—’American resolve and will’ come to be seen as ‘primary’. The consequence of this for arms control agreements, according to Jervis, is that they have been treated by some in the US as ‘bad ... not because they produce a less favorable military balance than would otherwise result, but because they produce psychological demobilization and disarmament’.57

These points complement Mearsheimer’s nationalistic agenda by suggesting that the nuclear status quo supports Pax Americana. This logically means both that nuclear disarmament would jeopardise such a ‘victory’ and explains why those that support US state power and believe its security requirements include regional and/or global hegemony oppose nuclear abolition. However, the offensive realist claim that the ‘peace and stability’ of the Cold War was principally ensured by MAD and bipolarity alone is challenged by defensive realism, which, in Jervis’s work, emphasises the diplomatic cooperation built between the Soviet Union and the US, alongside other factors such as ‘the increased pain of war’ given the ‘political and economic modernization’ following WW2.58

Whilst pessimistic regarding the prospects of nuclear disarmament, defensive realism is thus more optimistic than offensive realism when it comes to international cooperation, promoting the idea that uncertainty can be overcome if the incentives and motivations of states and their leaders are understood. The possibility of sharing information in this way, it is argued, allows decision-makers to move beyond escalating threats (as per deterrence) and arms races, in order to focus on compromises, reassurances and rewards. Furthermore, statespersons are encouraged to use diplomatic and political tools in order to develop transparency and empathy and signal their intentions. As discussed above, for Glaser, regime type matters here since democracies are more likely than authoritarian states to be seen as security seekers, thus weakening the security dilemma. Ultimately this all means that,

Theories of nuclear possessionldisarmament 41 with prudent leadership and compatible political institutions, states can exercise mutual restraint as a means of preventing conflict and, particularly in the current age, nuclear war.

As well as highlighting divergent views within realism, Jervis has sought to explain the differences between realists and neoliberal institutionalists. In doing so, he proposes that the two schools’ ‘disagreement over conflict is not about its extent but about whether it is unnecessary, given states’ goals’.59 Neoliberalism thus ‘does not see more cooperation than does realism; rather, neoliberalism believes that there is much more unrealized or potential cooperation than does realism’, which can ‘at least in part’ be explained by the fact that ‘they study different worlds’.60 Drawing on what he sees as the analytical strengths of neoliberalism, Jervis thus notes that,

perhaps the most important path by which institutions can change preferences is through domestic politics. Drawing on liberalism, neoliberalism holds that states are not all alike and that preferences in part arise internally. To the extent that this is correct, international arrangements can alter the power, beliefs, and goals of groups in society in ways that will affect foreign relations.61

Whilst Jervis here accepts the neoliberal focus on how international institutions affect a state’s internal politics and thus their foreign policy, he does not himself analyse the workings of domestic regimes, though he does note that other scholars have explored how ‘the shape of domestic institutions affects both the chance of international agreement and the distribution of the benefits’.62

As with the ‘different worlds’ which realism and neoliberalism study, the domestic political sphere may thus be seen as another world which defensive realism is aware of but does not study in any significant depth. The implications of this will be seen with our subsequent discussion of nuclear possession and disarmament, given that theory in this area needs to be constructed from an analysis of both domestic and international sources of state behaviour. For example, we need to better understand how a state’s relative strategic power and regime type affects its international goals—and ability to achieve them— in relation to nuclear issues.

In addition to defensive realism’s interest in the neoliberal emphasis on institutions, Jervis and Glaser are very sympathetic to democratic peace theory. Jervis explains his enthusiasm by arguing that democracies with strong institutions are ‘very likely to remain at peace with each other and to cooperate more readily than is true for autocracies or revolutionary regimes’.63 Elsewhere, the same author notes that,

if arms are positively valued because of pressures from a militaryindustrial complex, it will be especially hard for status-quo powers to cooperate. By contrast, the security dilemma will not operate as stronglywhen pressing domestic concerns increase the opportunity costs of armaments.64

This is a rare case of Jervis outlining a domestic variable, i.e. whereby a military-industrial complex or other political ‘concern’ exerts an enabling or restraining influence over a state’s defence and foreign policy, thus making it a source of international conflict or cooperation. In this case, the author identifies the state as a ‘status quo power’, which, based on his other comments, would probably make it a democracy.65 If we accept this analysis, it is logical to extend it and propose that there may be lower or higher levels of democratic practices and processes within a state and that these may influence and correspond with the type of behaviour, for example cooperative or non-cooperative, that a state engages in on an international level, depending on the type of issue involved. Moreover, it seems reasonable to suggest that identifying the domestic political actors that enable or restrain action supportive of nuclear disarmament may both reveal to us important sources of a state’s international behaviour as well as helping us understand the internal political changes that may be required in order for a state to become a greater champion of disarmament.

Turning now to the specific question of how defensive realism approaches nuclear disarmament, Glaser has done a significant amount of work imagining what this might entail, including its political and military requirements. Notably, as with other US-based realists we have encountered, his starting point for opposing the ‘flawed case for nuclear disarmament’ is whether this initiative will ‘enhance’ US security.66 The other key question for Glaser concerns in what ‘world’ nuclear war would be more likely, to which he concludes that ‘disarmament would not reduce the probability of nuclear war, so it would not provide what is commonly understood to be its key benefit.’67 Overall, whilst like Jervis Glaser is sceptical regarding the benefits of the nuclear revolution given the risks involved, his reading of the literature on cooperation under anarchy leads him to conclude that the prospects for disarmament are ‘extremely poor’ and that there is also a ‘variety of imposing domestic political barriers’ that need to be overcome, not least for the US.68 Moreover, he argues that there are significant dangers in shifting from what he sees as the current stability in the international state system—where the security dilemma is strongly mitigated by the war preventing effects of MAD and responsible arms control is in place, facilitating reductions towards small nuclear forces.

Overall, Glaser may therefore be said to represent a school of thought where national security—particularly, in his case, for the US—must come before disarmament, which corresponds with what may be termed a security first approach to eliminating nuclear weapons. According to this view, arms control measures may also, if further developed, make nuclear disarmament possible, if not very likely. Glaser thus critiques the notion that we should move to—what he claims would be—an unpredictable and potentially very

Theories of nuclear possessionldisarmament 43 volatile disarmed world that would ‘reinforce’ a spiral down in relations, where the possibility of accidental or deliberate nuclear use would persist, the potential for proliferation would become ‘far more threatening’ than today and where ‘states’ security would be very sensitive to cheating’.69

Leading arms control theorist Thomas Schelling has made several similar points, beginning with the argument that nuclear (and other) weapons cannot be disinvented. The main question he poses is, ‘why should we expect a world without nuclear weapons to be safer than one with (some) nuclear weapons?’70 For him, a NWFW cannot be one in which the great powers see ‘every crisis’ as a ‘nuclear crisis’ so that ‘any war could become a nuclear war’.71 This could occur because at the ‘outset’ of a ‘major war’, or when this seemed like an ‘imminent possibility’, those leading ‘responsible governments’ would race to build the bomb. The first states to acquire a bomb could then respond in several different ways, including using nuclear strikes to destroy their opponent’s nuclear arms facilities ‘in the interest of self-defense’.72 Elsewhere, Schelling discusses other problems associated with ‘total disarmament’, arguing that ‘some form of world government’ would be required to ‘police the world’ and ensure against ‘war and rearmament’.73 The ‘monopoly’ of military power in such a centralised world authority would then have to find a way to ‘improve’ and ‘stabilise’ deterrence so that there is a ‘balance of prudence’, whereby states see maintaining a NWFW as preferable to rearmament.74

Jervis complements Glaser’s and Schelling’s view of disarmament, asking what will replace nuclear weapons in terms of deterrence? For Jervis, the important thing about nuclear weapons is the ‘political effects’ that they produce, ‘not the physics and chemistry of the explosion’ so that analysts therefore need to ‘determine what these effects are, how they are produced, and whether modern conventional weapons would replicate them’.75 Where Glaser’s analysis differs from Schelling’s, in particular, comes in the former’s focus on the importance of domestic political barriers to nuclear disarmament. Glaser, in a way similar to Jervis and, for that matter, Herz, thus discusses the role of the military-industrial complex and other ‘powerful interest groups’ in promoting conflict, yet does this without going into much detail about these groups and how they function—an important gap in his thinking and the defensive realist project generally.76

Moreover, Glaser argues that the flaws with the case for nuclear disarmament mean that it is preferable to discuss how a ‘permanent revolution’ in international relations may be achieved, in order to ensure ‘absolute longterm safety from the use of nuclear weapons’.77 Echoing Schelling, Glaser also points out that any disarmament agreement would need to deal with the problem of rearmament given ‘the coercive potential’ of a state gaining a lead ‘in a rearmament race’ so that ‘states would have to coordinate their potential for nuclear rearmament, including their nuclear energy facilities’.78 This observation again highlights how several authors from the defensive realist tradition see global nuclear disarmament as encompassing a range of issues beyond the disposal of the weapons themselves, including: the role played byinfluential domestic political groups, for example, those pushing for the reproduction of the bomb; the potential need for world government; the recreation of deterrence in a NWFW to ensure peace and stability; and the international management of atomic power.

< Prev   CONTENTS   Source   Next >