Can the Security Dilemma Ever be Escaped?

This content was originally written for an undergraduate or Master's program. It is published as part of our mission to showcase peer-leading papers written by students during their studies. This work can be used for background reading and research, but should not be cited as an expert source or used in place of scholarly articles/books.

The concept of security has been a major interest in International Relations theory since its inception in the aftermath of World War I. After World War II, when realist theories gained traction in the discipline, the ‘security dilemma’ emerged as one theoretical response to problems of security. John Herz, the first to coin the term, defined it as a situation in which…

‘They [states] are driven to acquire more and more power in order to escape the impact of the power of others. This, in turn, renders the others more insecure and compels them to prepare for the worst. Since none can ever feel entirely secure in such a world of competing units, power competition ensues, and the vicious circle of security and power accumulations on.’ (Herz 1950: 157)

Herz (1950) based his concept on the anarchic structure of the international system, in contrast to other classical realists who focused on human nature as the main difficulty.

Following the approach of Booth and Wheeler (2008), this essay will first address the concept of the security dilemma in three parts: the dilemma of interpretation, the dilemma of response and the security paradox. Consulting the literature discloses a range of responses to the question of whether the security dilemma can be escaped, however, this essay addresses a limited number of these, namely those of (a) John Mearsheimer, who denies the existence of a dilemma claiming that negative outcomes cannot be escaped, (b) Robert Jervis and Charles Glaser, who argue that circumstances determine the intensity of the dilemma and it can thus partly be overcome; and (c) Emanuel Adler and Michael Barnett’s theory of ‘security communities’, wherein the security dilemma is transcended rather than overcome. Within the latter the essay will also consider some of the theoretical limits of the discussion around the security dilemma.

Dilemma of Interpretation, Dilemma of Response and Security Paradox

Booth and Wheeler (2008) label Herz’s concept as a ‘two-level strategic predicament’. Firstly, the dilemma of interpretation arises from the need of policy makers to interpret other states motives, intentions and capabilities under the general condition of uncertainty, evaluating, for example whether increasing military capabilities represents defensive or offensive action. Whenever a decision regarding how to interpret another state’s intentions is made decision makers encounter the next dilemma: what is the appropriate response? If the state inappropriately decides to reassure the other party of its peaceful intentions it runs the risk of exposing itself to coercion. If the state reacts by increasing its own military capabilities, it might trigger an unintended spiralling of hostilities. In this latter instance, both states become more insecure, despite both having initially acted in order to increase their security. This is known as the ‘security paradox’ (Booth & Wheeler 2008: 4-5).

Mearsheimer on the Security Dilemma

While there is no explicit reference to the security dilemma within Mearsheimer’s work (Glaser 1997: 196), it is nonetheless important to consider his de facto rejection of the concept. Mearsheimer believes that the uncertainty about other states intentions, and therefore the dilemma of interpretation, is absolute and cannot be reduced (1994: 10). Even if states were certain about others intentions, these could change any time (Mearsheimer 2001: 31). As security matters are a question of survival, states could only accept absolute certainty. As this is according to Mearsheimer impossible states have to assume that the other state is a treat. He also stresses that it is impossible to distinguish between a state’s offensive and defensive military capability (2006: 234).

According to this conceptualisation of the dilemma of interpretation there can be no dilemma of response, because states have no choice but to increase their military capabilities to deter others. Consequently, the security paradox exists constantly and cannot be prevented (2001: 35-36). States can only feel close to secure when they become more powerful than any other state, that is, when they reach hegemony (Mearsheimer 2001: 33). Hence within Mearsheimer’s framework there is no room for the security dilemma, which is resolved into one extreme – an absolute, permanent and inescapable ‘security competition, which no amount of cooperation can eliminate’ (1994: 9). As Booth and Wheeler put it ‘Mearsheimer replaced the dilemma of interpretation with a rule of fatalism, and abolished any dilemma of response by a rule of offensive potential’ (2008: 35).

Jervis and Glaser on the Security Dilemma

Glaser recognises that states which seek more than the status quo (as Mearsheimer claims they always do) precipitate competition and conflict, making the security dilemma less relevant (Glaser 1997: 190). Generally, however, he and Jervis evaluate the security dilemma as it arises between states with strictly defensive intentions, a significant departure from Mearsheimer’s premise.

Jervis and Glaser identify a number of difficulties with the dilemma of interpretation. Firstly, Jervis argues that states are psychologically biased and can wrongly assume that others know they are only seeking security (1978: 181-183). Merely replicating another state’s security measures can lead to the other state feeling threatened, as it fails to see the dilemma of interpretation of the other (Kydd 1997). Furthermore, Jervis stresses that states have different understandings of what level of capability is necessary for defence. If another state builds up arms far beyond the level they themselves consider adequate for defence, they may assume offensive intentions (Jervis 1978: 174-176). The situation can be further exacerbated because the state expects if the other was genuinely hostile it would try to make itself look like a security-seeker (Glaser 1997: 181).

According to Jervis there are other variables that can ease or intensify the dilemma, namely the offence-defence balance and the offence-defence differentiation. The offence-defence balance reflects whether territory is easier to defend or attack. When a territory is easier to defend than attack, security and stability are more likely, states are less prone to feel threatened, and the security dilemma is consequently less intense. The opposite is the case whenever it is easier to attack than to defend. The offence-defence differentiation reflects how easy it is to distinguish defensive from offensive weapons and policies. Jervis believes that if the offence-defence differentiation is high, uncertainty, and thus the security dilemma, is effectively eliminated (Jervis 1978: 187).[1]

Following liberal ideas, Glaser adds to this argument the potential role of unit-level-knowledge in challenging the general condition of uncertainty. States which are better informed as to each other’s domestic political and economic systems are more able to correctly identify each other’s intentions and thus to escape the dilemma (1997: 191-193).

As already shown in the discussion of Mearsheimer’s ideas, the dilemma of response is dependent upon the intensity of the dilemma of interpretation. According to Jervis and Glaser, the intensity of the dilemma of interpretation is dictated by circumstances. The logical conclusion of this position is that if uncertainty (an unavoidable precondition of the dilemma of interpretation) can be diminished then states are able to make more informed choices about whether to pursue policies of reassurance[2] or deterrence. The dilemma of response is correspondingly reduced, and thus, though the security dilemma may not be escaped entirely, it may still be marginalised.

Adler and Barnett on the Security Dilemma

So far the essay has evaluated the security dilemma mainly within the framework of traditional security approaches, ranging from offensive realism to liberal institutionalism. These approaches are based on very similar core assumptions (Jervis 1999: 43), such as, the importance of the anarchic structure, the centrality of the state and its unitary character, the understanding of security as state security, the understanding of threats as mainly military and positivist methodologies. Indeed, Yosef Lapid (1989) goes as far as to claim that they form a neo-neo-synthesis.

Emanuel Adler and Michael Barnett’s theory of security communities is an attempt to move away from some of these core assumptions by deploying a broadly constructivist approach. They claim, for example, that anarchy is socially constructed and thus give more space to agency, claiming the middle-ground between positivism and post-positivism and in so doing broadening the definition of security. This constructivist approach thus ‘fully opens up the sociological bottle’ (Adler & Barnett 1998a: 13).

Adler and Barnett deploy the concept of security communities, as originally developed by Karl Deutsch (et al. 1957):

‘Deutsch observed a pluralistic security community whenever states become integrated to the point that they have a sense of community, which, in turn, creates the assurance that they will settle their differences short of war’ (Adler and Barnett 1998a: 3).[3]

They thus argue that although states cannot escape the security dilemma, security communities allow them to transcend it by reversing the condition of fear (not uncertainty) upon which the dilemma of interpretation is based. This transition is possible because states interests are not seen as fixed, but as defined by flexible identities. Constructivists thus shift the focus from static structures to processes of interaction (Adler and Barnett 1998a: 11-14).

According to Adler and Barnett security communities begin when states seek closer relations having recognised that co-operation on security or economic issues furthers their interests. At this stage articulations of identity are mainly instrumental. However, even though the cooperation is issue-based, it likely creates space for the discussion and sharing of norms and concepts. Through this interaction between states their relationship changes fundamentally (1998a: 415-418): after the first ‘nascent phase’, the norm of non-intervention is transformed into mutual accountability. Within the ‘ascendant phase’, states become more relaxed about protecting their sovereignty and autonomy in areas not likely to meet domestic opposition, and the reasons for compliance with the community’s norms gradually shift from material to ideal. Closer identification and mutual trust gradually transcend into dependable expectations of peaceful change when the security community enters the ‘mature phase’ (Adler and Barnett 1998b).

As a result of this integration process, states do not see other states within the security community as a threat and the ‘psychological anxieties that are characteristic of security dilemmas’ disappear (Adler & Barnett 1998b: 433), thus overcoming the dilemma of interpretation by overcoming fear. When the security dilemma is eliminated, non-military security dynamics concerning economic and environmental issues take its place (Adler & Barnett 1998b).

The Security Dilemma: Escape or Transcend?

The essay has discussed three perspectives on the security dilemma by Mearsheimer, Jervis and Glaser, and Adler and Barnett. They vaguely correspond to what Booth and Wheeler (2008) call fatalistic, mitigating and transcending logics. Mearsheimer believes that the only thing states are certain about is uncertainty. If they want to survive they have to assume the worst and aim for hegemony to reach relative security. This position leaves no room for the security dilemma, but instead demands the taking of a course of action that leads all states into the security paradox and constant, inescapable security competition.

Taking a different approach, Jervis and Glaser claim that the security dilemma cannot be escaped as a logical problem but may be weakened when the offence-defence balance favours defence and the differentiation between offensive and defensive capabilities is very clear. Glaser adds the possibility that states can gain understanding of the domestic processes of another state, which can in turn increase certainty. However, the dilemma is intensified by states failing to see themselves as others do and therefore underestimating other states’ security dilemmas. Thus while there is hardly any prospect of escaping the security dilemma within traditional ways of thinking, it is possible to diminish or even marginalise it.

In the final section, however, it was argued that these perspectives rest upon a number of pre-assumptions, which have to be challenged. It must be recognised that no theoretical escape to the dilemma is of value in praxis if these presumptions do not hold. Departing from some core assumptions, Adler and Barnett (1996a; 1996b) argued that the security dilemma is a transcendable problem. Fear – but not uncertainty – can be eliminated. The necessary trust, they believe, rises through long-term political cooperation and social interaction and the resulting closer identification and construction of common interests within a security community.


Adler, Emanuel and Barnett, Michael. (1998a). ‘Security Communities in theoretical perspective’, in

Adler, Emanuel and Barnett, Michael (eds), Security Communities, Cambridge: Cambridge University Press. 3-28.

Adler, Emanuel and Barnett, Michael. (1998b). ‘Studying Security Communities in Theory, Comparison, and History’, in Adler, Emanuel and Barnett, Michael (eds), Security Communities, Cambridge: Cambridge University Press. 413-441.

Booth, Ken and Wheeler, Nicholas J. (2008). The Security Dilemma: Fear Cooperation and Trust in World Politics. Basingstoke: Palgrave Macmillian.

Deutsch, Karl W., Burrell, Sidney A., Kann, Robert A., Lee, Maurice, Lichterman, Martin, Lindgren, Raymond E., Loewenheim, Francis L., Van Wagenen, Richard W. (1957). Political Community and the North Atlantic Area. Princeton: Princeton University Press.

Herz, John H. (1950). ‘Idealist Internationalism and the Security Dilemma’, World Politics, 2 (2): 157-180.

Glaser, Charles L. (1997). ‘The Security Dilemma Revisited’, World Politics, 50 (1): 171- 201.

Jervis, Robert. (1976). Perception and Misperception in International Politics. Princeton: Princeton University Press, 1976

Jervis, Robert. (1978). ‘Cooperation under the Security Dilemma’, World Politics, 30 (2): 167-214.

Jervis, Robert. (1999). ‘Realism, Neo-liberalism and Cooperation: Understanding the Debate’, International Security, 24 (1): 42-63.

Kydd, Andrew. (1997). ‘Game Theory and the Spiral Model’, World Politics, 49 (3): 371-400.

Kydd, Andrew. (2000). ‘Trust Reassurance and Cooperation’, International Organization, 54 (2): 325-357.

Lapid, Yosef. (1989). ‘The Third Debate: On the Prospects of International Theory in a Post-Positivist Era’, International Studies Quarterly, 33 (3): 235-254.

Mearsheimer, John. (1994). ‘The False Promise of International Institutions’, International Security, 19 (3): 5-49.

Mearsheimer, John. (2001). The Tragedy of Great Power Politics. New York: W.W. Norton.

Mearsheimer, John. (2006). ‘Conversations in International Relations: Interview with John J.

Mearsheimer, Part II’, International Relations, 20 (2): 231-243.

[1]  For a discussion as to how these two variables influence the security paradox see Jervis 1976.

[2]  There has been intense debate regarding whether states are actually capable of signalling their true intentions. Offensive realists believe this to be impossible (Booth & Wheeler 2008: 37; Mearsheimer 2006: 232-234). Others, such as Andrew Kydd, believe that reassurance is a credible possibility. To be persuasive the signal used has to be so costly, that a dishonest state would not give it. This can be risky, but through signals and their reciprocation cooperation is possible (Kydd 2000: 326).

[3]  Deutsch considered two types of security communities, ‘amalgamation’, i.e. the formal unification of states, and ‘pluralistic’ security communities that consist of independent and sovereign governments (Deutsch et al. 1957: 6). Adler and Barnett (1998a) explicitly discuss the latter.

  Written by: Laura Schmah
Written at: Queen’s University Belfast
Written for: Security and Terrorism
Date written: March 2012

Further Reading on E-International Relations

Please Consider Donating

Before you download your free e-book, please consider donating to support open access publishing.

E-IR is an independent non-profit publisher run by an all volunteer team. Your donations allow us to invest in new open access titles and pay our bandwidth bills to ensure we keep our existing titles free to view. Any amount, in any currency, is appreciated. Many thanks!

Donations are voluntary and not required to download the e-book - your link to download is below.


Get our weekly email