Offensive vs defensive realism & the security dilemma in depth

A shared foundation, a contested conclusion

Structural realism (neorealism), codified by Kenneth Waltz in Theory of International Politics (1979), holds that the anarchic ordering principle of the international system—the absence of a world government—compels states to prioritise survival. From this single premise two rival schools emerge that agree on the diagnosis but split on the prescription: how much power should a rational state seek?

Defensive realism

Defensive realism, the position Waltz himself adopted, argues that the system rewards states that seek an appropriate, not maximal, amount of power. Because balancing coalitions form against any state that grows too strong—Waltz's 'balance-of-power tendency'—excessive expansion is self-defeating. The optimal grand strategy is to preserve the existing balance. Robert Jervis ("Cooperation Under the Security Dilemma," World Politics, 1978) and Stephen Van Evera supplied the supporting logic: under most conditions, conquest is hard, defence is easier than offence, and security is plentiful. Aggression therefore reflects domestic pathology or misperception, not structural necessity. The historical exhibit is Wilhelmine and Nazi Germany, whose bids for hegemony provoked the very encircling coalitions that destroyed them in 1918 and 1945.

Offensive realism

John Mearsheimer, in The Tragedy of Great Power Politics (2001), rejects the idea that the system supplies enough security. Because a state can never be certain of another's present or future intentions, and because military power is the ultimate arbiter, the only rational guarantee of survival is to maximise relative power—ideally to become a regional hegemon, as the United States did in the Western Hemisphere by roughly 1900. Great powers are revisionist by default, restrained only when the costs of expansion exceed the gains. Mearsheimer's five bedrock assumptions—anarchy, offensive capability, uncertainty of intentions, survival as the primary goal, and rational calculation—generate relentless competition. His doctrine of offshore balancing and his prediction that a rising China cannot rise peacefully follow directly from this premise.

Where the schools diverge

The practical fork is over the offence–defence balance. Defensive realists say technology and geography usually favour the defender, so restraint pays; offensive realists say states cannot reliably read that balance and must assume the worst. Both reject the liberal claim that institutions or commerce can override anarchy, distinguishing them sharply from neoliberal institutionalism (Keohane). For exam purposes, fix the names to the texts: Waltz/Jervis/Van Evera = defensive; Mearsheimer = offensive; both are structural and therefore distinct from classical realism's Morgenthau, who located the drive for power in flawed human nature (animus dominandi) rather than in system structure.

The security dilemma in depth

The security dilemma, named by John Herz (Idealist Internationalism and the Security Dilemma, 1950) and elaborated by Herbert Butterfield and Robert Jervis (1978), is the engine common to both realisms. Its logic: under anarchy, measures a state takes to make itself more secure—building arms, forming alliances, acquiring buffer territory—necessarily reduce the security of others, who respond in kind. The outcome is a spiral in which all parties end up less secure despite none harbouring aggressive intent. It is a tragedy because it arises even among purely defensive actors.

Jervis's two variables

Jervis showed the dilemma's intensity is governed by two variables. First, whether offensive and defensive weapons are distinguishable: if they are not, every acquisition looks threatening. Second, whether offence or defence has the advantage: when offence dominates and weapons are indistinguishable (his 'doubly dangerous' world), the spiral is most violent. When defence dominates and the two postures are distinguishable, security-seekers can signal benign intent and cooperation is feasible. The pre-1914 'cult of the offensive' and rigid mobilisation timetables are the textbook case of offence-dominance turning a crisis into general war by August 1914.

Spiral model versus deterrence model

Jervis paired the dilemma with two competing prescriptions. The spiral model holds that an adversary is a frightened security-seeker; conciliation defuses the spiral, while toughness inflames it—appeasement's intellectual cousin, though Jervis warned it can be misapplied. The deterrence model holds that an adversary is a genuine aggressor; firmness deters, concession invites further demands—the 'Munich 1938' lesson. The central policy problem is diagnosis: treating a Hitler as a security-seeker (spiral logic) invites disaster, while treating a security-seeker as a Hitler (deterrence logic) manufactures an avoidable enemy. The Cold War arms race, NATO–Russia friction after 1999 enlargement, and Sino-American naval competition in the South China Sea are all live applications.

Mitigating the dilemma

The dilemma is not destiny. Charles Glaser's Rational Theory of International Politics (2010) shows that arms-control agreements, transparency, and costly signals (unilateral restraint that an aggressor would not undertake) can let security-seekers identify one another. Confidence- and security-building measures (CSBMs) institutionalised in the 1975 Helsinki Final Act, the 1972 ABM Treaty, and the 1990 Treaty on Conventional Armed Forces in Europe were deliberate engineering against the dilemma—shifting weapons toward defensive, distinguishable, verifiable postures. Their erosion (US withdrawal from the ABM Treaty in 2002 and the INF Treaty in 2019) sharpens the dilemma anew.

Why this matters for the exam

This lesson is high-yield across every paper in the shared layer. For UPSC GS-2, the offensive/defensive split and the security dilemma are the analytical scaffolding for questions on India–China relations, the Quad, and 'India's strategic autonomy'—examiners reward candidates who name the theory and then apply it, rather than narrating events. For the FSOT, the Job Knowledge Test directly samples IR theory and Cold War strategic concepts; Mearsheimer, Waltz, Jervis and the deterrence-versus-spiral distinction are canonical. CSS International Relations Paper-I explicitly lists neorealism and the security dilemma in its syllabus, and BCS international-affairs questions test the same theorists.

How it is tested

Three formats recur. First, direct conceptual questions—'Distinguish offensive from defensive realism' or 'Explain the security dilemma.' Here you must anchor each label to its author and text and avoid conflating structural realism with Morgenthau's classical realism. Second, applied analytical questions that hand you a contemporary case (NATO enlargement, the South China Sea, Indo-Pacific balancing) and expect you to deploy the theory as a lens. The discriminator is Jervis's two variables: a top answer asks whether weapons are distinguishable and whether offence or defence dominates, then diagnoses spiral versus deterrence. Third, evaluative questions that ask you to critique realism—deploy the liberal-institutionalist and constructivist rebuttals (Wendt's 'anarchy is what states make of it,' 1992) to show range.

High-yield facts to retain

Lock down the attributions: Waltz 1979 (structural realism, defensive); Mearsheimer 2001 (offensive realism, The Tragedy of Great Power Politics); Herz 1950 (coined 'security dilemma'); Jervis 1978 (offence–defence balance, spiral vs deterrence). Remember the PYQ-relevant theme that India's 'strategic autonomy' and hedging behaviour can be read as a defensive-realist balancing response to a security dilemma with China. Pair each abstract point with one dated instance—1914 mobilisation, Munich 1938, NATO's 1999/2004 enlargements, the 2002 ABM withdrawal, the 2019 INF collapse, Galwan 2020. The examiner's pen rewards the candidate who fuses a named theorist, a precise mechanism, and a dated case in a single paragraph.