By Brajabandhu Mahanta•November 17, 2026•19 min read
Structural realism explains some patterns of South Asian security well, such as balancing, arms races and the failure of regional cooperation (see the previous topic). But it struggles to explain the sources, nature and content of most security threats in the region. This topic sets out its main limits, which together form the heart of Unit II.
The core argument in one paragraph
Structural realism was built to explain the behaviour of great powers in a system of strong, cohesive, sovereign states whose main threats come from other states. South Asian states are mostly post-colonial, internally divided and weakly consolidated. Their most frequent and deadly threats come from within (insurgencies, ethnic and religious conflict, regime crises) or from non-state and non-military sources (terrorism, migration, water, climate). Their foreign policies are shaped by identity, history and domestic politics, not only by the distribution of power. Therefore, the realist model of unitary states balancing each other in anarchy gives an incomplete and sometimes misleading picture of South Asian security.
Realist assumptions to be tested
Realist assumption
Question for South Asia
The main threats are external and military
Are most threats actually internal or non-military?
States are unitary, rational actors
Do armies, parties, provinces and leaders pursue their own interests?
State interests come from structure, not identity
Are rivalries driven by religion, nationhood and history?
States are the main actors
What about terrorist groups, insurgents and diasporas?
States are “like units” with full sovereignty
Are South Asian states weak, with porous borders and contested legitimacy?
Cooperation is rare and fragile
Why have some treaties and friendships lasted?
The theory is universal
Was it built mainly on European and Cold War experience?
Structural realism assumes that security threats come mainly from other states across the border. In South Asia, the most frequent, long-lasting and deadly threats have been internal.
Insurgencies and separatism: Kashmir and the north-east (India), Balochistan and the tribal areas (Pakistan), the Tamil north and east (Sri Lanka, 1983–2009), the Chittagong Hill Tracts (Bangladesh, till 1997), the Maoist People’s War (Nepal, 1996–2006).
Breakup of a state from within: Pakistan lost half its territory and population in 1971 not because of a structural shift in the balance of power, but because of an internal crisis of identity and legitimacy (Bengali nationalism), which India then exploited.
Regime crises and uprisings: Sri Lanka’s Aragalaya (2022), Bangladesh’s 2024 uprising, Nepal’s 2025 Gen Z protests, and repeated coups and political crises in Pakistan show that governments often fall due to internal pressure, not foreign attack.
Over the decades, internal conflicts in South Asia have killed far more people than the inter-state wars since 1971.
Concepts that explain this better
Mohammed Ayoob’s “subaltern realism” (The Third World Security Predicament, 1995): Third World states are “late state-makers”. Because they are still building their states, their main security problem is internal: establishing control, legitimacy and national integration. External threats often matter mainly because they exploit internal weaknesses.
Barry Buzan’s “weak states” (People, States and Fear): states with weak socio-political cohesion feel threatened mainly by internal forces; in such states, national security often means regime security.
Brian Job’s “insecurity dilemma” (The Insecurity Dilemma: National Security of Third World States, 1992): in weak states, the regime, rival groups and society all feel insecure from each other. Unlike the classical security dilemma between states, the main dilemma lies inside the state.
K.J. Holsti (The State, War, and the State of War, 1996): since 1945, most wars have been wars within states, not between them.
Examples of regime security driving “national” security
Pakistani governments and the army often present domestic opposition (Bengali, Baloch, Pashtun movements) as foreign-backed threats to national security.
Sri Lankan governments treated Tamil demands for autonomy as threats to the unitary state.
Bangladesh’s Awami League government (2009–2024) used security laws against political opponents.
Why this matters
If the main threats are internal, then balance of power and military alliances cannot solve them. They require state-building, inclusive governance, federalism, development and political accommodation, which are outside structural realism’s scope.
Limit 2: The State Is Not a Unitary Actor
The problem
Waltz treats states as unitary “like units” and says his theory does not explain foreign policy. But in South Asia, security policy is often driven by domestic institutions, leaders, parties, provinces and public opinion, which can push states in directions that structure alone would not predict.
Civil-military relations in Pakistan
The army controls India, Afghanistan and nuclear policy, even under civilian governments. Its institutional interests (budget, prestige, economic empire, “Military Inc.” in Ayesha Siddiqa’s term) and identity are tied to the India threat.
Kargil (1999): the operation was planned by the army under General Pervez Musharraf, apparently without the full knowledge or approval of Prime Minister Nawaz Sharif, just months after the Lahore Declaration (February 1999). This shows how internal divisions, not systemic logic, produced a war.
Civilian attempts at peace with India (Nawaz Sharif in 1999 and 2014–15; the 2004–07 “composite dialogue”) were often undermined by the army or militant attacks.
Domestic politics in India
Coalition and regional parties shape foreign policy: Tamil Nadu parties influenced India’s Sri Lanka policy (the IPKF, votes on UN resolutions); West Bengal’s opposition blocked the Teesta water-sharing agreement with Bangladesh in 2011.
Public opinion and elections: responses to terror attacks (surgical strikes 2016, Balakot 2019, Operation Sindoor 2025) are shaped by domestic political pressure and nationalism, not only by strategic calculation.
Leadership: Nehru’s non-alignment, Indira Gandhi’s assertiveness, Gujral’s generosity, Vajpayee’s Lahore bus journey (1999) and Modi’s “Neighbourhood First” show how much individual leaders matter.
Other examples
Sri Lanka: Sinhala nationalist parties and the Buddhist clergy prevented compromises with Tamils that might have ended the conflict early.
Nepal and Maldives: foreign policy swings (pro-India or pro-China) often follow changes of government, not changes in the balance of power.
Bangladesh: relations with India changed sharply after the fall of Sheikh Hasina (2024), again a domestic, not structural, cause.
Better explanations
Neoclassical realism (domestic “intervening variables”), liberal theories (domestic preferences, regime type), bureaucratic politics models (Graham Allison) and leadership studies explain these cases better than Waltz’s structural model.
Limit 3: Identity, Ideology and History
The problem
Structural realism says state interests come from material structure (anarchy and the distribution of power), not from identity or ideas. But in South Asia, many conflicts are driven by who states think they are and how they see the “other”.
The India–Pakistan rivalry as an identity conflict
The rivalry began with competing ideas of nationhood: India as a secular, civic, plural nation; Pakistan as a homeland for the Muslims of the subcontinent (the two-nation theory).
Kashmir is not just territory or strategic real estate. For Pakistan, a Muslim-majority Kashmir in India challenges the logic of the two-nation theory; for India, giving up Kashmir on religious grounds would challenge its secular identity. This makes compromise very hard, even when material costs are high.
Memory of Partition and 1971: trauma, humiliation and mutual stereotypes shape perceptions on both sides. Pakistan’s identity is partly built in opposition to India.
Religious nationalism on both sides (political Islam in Pakistan; Hindu nationalism in India) intensifies hostility.
A purely realist view cannot explain why the rivalry has persisted even when power asymmetry was overwhelming (after 1971) or when cooperation would clearly bring large economic gains.
Other examples
Sri Lanka: Sinhala-Buddhist nationalism, not the balance of power, drove policy towards Tamils and shaped relations with India.
Bangladesh: relations with India and Pakistan depend on how the nation’s identity is defined (Bengali/secular vs Bangladeshi/Islamic).
India–Bhutan amity: shared cultural ties and a long history of trust explain a friendship that realism, which expects suspicion of a powerful neighbour, cannot fully explain.
India and Nepal: the “roti-beti” relationship (open borders, intermarriage, shared religion) coexists with Nepali nationalism against Indian dominance.
Better explanations
Constructivism (Alexander Wendt): “anarchy is what states make of it”. India–Pakistan relations reflect a Hobbesian culture of enmity, India–Bhutan a more Kantian culture of friendship, and India–China a Lockean culture of rivalry, all under the same anarchy.
Strategic culture approaches (George Tanham, Indian Strategic Thought, 1992; Kanti Bajpai on Indian grand strategy; Christine Fair, Fighting to the End: The Pakistan Army’s Way of War, 2014): states’ historical experiences and beliefs shape how they use force. Fair argues that the Pakistan army is an ideological institution that will not accept the status quo even when it loses.
Ontological security (security of identity, not only of territory) helps explain why states pursue costly policies to protect their sense of self.
Limit 4: Non-state Actors, Terrorism and the Limits of Deterrence
The problem
Structural realism is state-centric: only states matter as actors. But in South Asia, many of the most serious security events have been caused by non-state actors.
Terrorist and militant groups: Lashkar-e-Taiba (Mumbai 2008, Pahalgam 2025, according to India), Jaish-e-Mohammed (Parliament attack 2001, Pulwama 2019), the TTP (attacks inside Pakistan, the 2014 Peshawar school massacre), ISIS-K, and earlier the LTTE (which assassinated Rajiv Gandhi in 1991).
Insurgent groups with cross-border bases (Naga and Manipuri groups in Myanmar; Baloch groups; the TTP in Afghanistan).
Crime and smuggling networks, and diasporas that fund or lobby for causes (Tamil, Sikh, Kashmiri, Baloch).
Why non-state actors challenge realism
They have no territory or population to hold hostage, so classical deterrence does not work directly on them.
State sponsorship and deniability: when states support militant groups as proxies, it becomes unclear who is responsible. Realism has no clear framework for this “grey zone”.
They can trigger inter-state crises: Kargil involved irregulars; the 2001–02 standoff, the 2016 and 2019 strikes and the 2025 conflict were all triggered by terrorist attacks. The decision to escalate was effectively influenced by non-state actors.
They often turn against their sponsors: Pakistan has suffered heavily from the TTP, a product of the same jihadi networks it once used.
The stability-instability paradox
Glenn Snyder (1965): when nuclear weapons make full-scale war too dangerous (strategic stability), states may feel freer to engage in lower-level conflict (instability).
S. Paul Kapur (Dangerous Deterrent: Nuclear Weapons Proliferation and Conflict in South Asia, 2007) argued that nuclear weapons increased instability in South Asia: Pakistan felt protected by its nuclear shield to support militancy and limited operations (Kargil), which in turn provoked Indian responses.
The sequence of crises (1999, 2001–02, 2008, 2016, 2019, 2025) suggests that nuclear deterrence has prevented full-scale war but not prevented recurring violence. The 2025 conflict even involved missile and drone strikes between two nuclear powers.
This challenges Waltz’s optimism that “more may be better” and supports Scott Sagan’s argument that organisational problems, civil-military issues and militant groups make proliferation dangerous.
Better explanations
Studies of terrorism, insurgency and proxy war; critical terrorism studies; Sagan’s organisational theory; and Unit III‘s analysis of deterrence in South Asia.
Limit 5: Non-traditional and Human Security
The problem
Structural realism defines security narrowly as military protection of the state. In South Asia, the threats that affect the most people are often non-military and transnational, and the referent object of security is frequently the individual or community, not the state.
Major non-traditional threats
Threat
South Asian examples
Why realism misses it
Poverty and hunger
Large shares of the world’s poor and malnourished live in the region
Not a military threat; needs development
Climate change and disasters
Floods in Pakistan (2022); cyclones in Bangladesh and Odisha; sea-level rise in Maldives; glacial melt in the Himalayas
Sri Lanka’s default (2022); Pakistan’s repeated IMF bailouts
Internal economic management
Human security
The concept of human security was popularised by the UNDP Human Development Report (1994), strongly shaped by the Pakistani economist Mahbub ul Haq (with Amartya Sen’s ideas on capability and development).
It shifts the focus from the security of the state to “freedom from fear” and “freedom from want” for people.
In South Asia, states have spent heavily on defence while large parts of their populations lacked basic security of food, health and livelihood; Mahbub ul Haq famously criticised this “guns versus butter” trade-off.
Better explanations
Broadened security studies (Buzan’s five sectors), human security, the Copenhagen School (securitisation: how issues like migration become “security” issues through political speech), and environmental security studies.
Limit 6: Post-colonial States and Contested Sovereignty
The problem
Waltz assumes that states are “like units” with full Westphalian sovereignty: clear borders, control over their territory and the same basic functions. South Asian states do not fit this assumption well.
How South Asian states differ
Colonial borders: lines such as the Durand, McMahon and Radcliffe Lines were drawn by outsiders and remain disputed. Afghanistan has never accepted the Durand Line; China rejects the McMahon Line; Kashmir remains divided.
Porous borders: people, goods, weapons and militants move easily across borders in Punjab, Bengal, the Terai and the Pashtun belt. The India–Nepal border is open by treaty.
Cross-border ethnic communities: Pashtuns, Baloch, Punjabis, Bengalis, Tamils, Madhesis and Nagas live on both sides of borders. Domestic ethnic policies therefore immediately become inter-state issues (for example, the Tamil issue between India and Sri Lanka, the Madhesi issue between India and Nepal).
Uneven state presence: in areas such as Pakistan’s former tribal areas, parts of north-east India or much of Afghanistan, the state’s control has been limited.
Unequal sovereignty in practice: Bhutan accepted Indian “guidance” in foreign affairs until 2007; Nepal’s 1950 treaty and economic dependence limit its freedom; small states are heavily shaped by bigger neighbours.
Stephen Krasner (Sovereignty: Organized Hypocrisy, 1999) showed that sovereignty is often violated in practice, which is especially visible in South Asia (interventions in 1971, 1987, 1988; proxy wars).
Why this matters
If borders are porous and identities cross them, the clear line between “internal” and “external” security that realism assumes does not exist.
Weak states are less able to mobilise power for balancing and more vulnerable to outside interference.
Security problems such as migration, smuggling, militancy and ethnic conflict arise from the nature of the post-colonial state itself, a factor outside structural realism.
Better explanations
Post-colonial approaches, Ayoob’s subaltern realism, and Buzan and Wæver’s Regional Security Complex Theory, which pays attention to geography, history and cross-border links within a region.
Limit 7: Cooperation, Amity and Change That Realism Cannot Explain
The problem
Structural realism expects cooperation to be rare, fragile and limited by relative-gains concerns. Yet South Asia also shows durable cooperation, friendships and peaceful settlement of disputes, as well as sharp changes in policy that structure does not predict.
Durable cooperation
Indus Waters Treaty (1960): survived the wars of 1965, 1971 and 1999 and many crises for 65 years, until India placed it “in abeyance” in 2025. Its long survival shows that institutions and shared interests can moderate rivalry, as neoliberal institutionalists argue.
India–Bangladesh: the Ganga Water Treaty (1996), the Land Boundary Agreement (2015), which settled the enclave issue peacefully, and the maritime boundary settlements (with Myanmar in 2012 and India in 2014, through international arbitration, which Bangladesh won substantially and India accepted).
India–Pakistan nuclear CBMs: the annual exchange of lists of nuclear installations since 1992, continuing even in crisis years.
India–Nepal open border and India–Bhutan hydropower cooperation.
Disaster cooperation: India’s assistance to Sri Lanka and Maldives in the 2004 tsunami; regional COVID-19 cooperation; India’s emergency aid to Sri Lanka in 2022.
Amity and the “security community” idea
India–Bhutan relations resemble a security community (Karl Deutsch): a relationship in which war has become unthinkable. Realism, which expects a small state to fear a large neighbour, cannot explain this; constructivism (shared identity and trust) can.
Change not explained by structure
India’s non-alignment (1950s): a choice based on ideas and leadership, not on balancing.
The Gujral Doctrine (1996–97): India’s non-reciprocal generosity towards smaller neighbours contradicts relative-gains logic.
Peace processes: the Lahore Declaration (1999) and the 2004–07 composite dialogue between India and Pakistan, which came close to solutions on Kashmir (the “four-point formula” discussed by Manmohan Singh and Musharraf), followed by reversals; these came from domestic and leadership factors.
Rapid foreign policy swings in Bangladesh, Maldives, Nepal and Sri Lanka after changes of government.
Better explanations
Neoliberal institutionalism (Keohane and Nye): institutions and interdependence enable cooperation.
Constructivism: shared identities, norms and trust create friendship.
Liberal theories of domestic politics and leadership explain changes in policy.
Limit 8: Eurocentric and Great-power Bias
The problem
Structural realism was developed mainly from the experience of European great powers (from the 17th to the 20th century) and the US–Soviet Cold War. Waltz himself said his theory is about great powers, because they shape the system. This creates several problems when the theory is applied to South Asia.
Main points of criticism
Ignores the colonial experience: the theory has nothing to say about how colonial rule created South Asian states, borders and conflicts, which is the root of most regional insecurity (Unit I).
Ignores small and weak states: most South Asian states (Nepal, Bhutan, Sri Lanka, Maldives, Bangladesh) are not great powers; their main concerns are survival as states and societies, development and autonomy from bigger neighbours.
Assumes the Western state as the norm: a strong, cohesive nation-state with clear borders, which South Asian states are still trying to become.
Treats the region as a passive arena: as if regional states only react to great-power competition. In fact, regional dynamics (India–Pakistan rivalry, internal conflicts) often drive the involvement of outside powers, not the other way round.
Universal claims: realism presents itself as a timeless, universal theory, but its concepts and examples come from a specific Western history.
Third World and non-Western critiques
Mohammed Ayoob: “subaltern realism”, focused on the security predicament of late state-makers.
Stephanie Neuman (ed.), International Relations Theory and the Third World (1998): mainstream IR theories do not fit the Third World well.
Amitav Acharya and Barry Buzan (eds.), Non-Western International Relations Theory (2010), and Acharya’s idea of “Global IR”: IR theory must include non-Western histories and ideas.
Navnita Chadha Behera and other South Asian scholars: call for “re-imagining IR” from South Asian experiences.
Post-colonial and feminist scholars (for example, Sankaran Krishna, Postcolonial Insecurities, 1999, on India and Sri Lanka): security discourses are tied to national identity, territory and anxieties produced by colonial history.
Why this matters
A theory built for great-power politics among strong states can capture the India–Pakistan–China military balance, but not the everyday insecurity of most South Asian states and peoples. This is why many scholars argue for alternative or complementary approaches, discussed in the next topic.
Summary: Realist Assumptions vs South Asian Reality
Realist assumption
South Asian reality
Better explanation
Threats are mainly external and military
Most violence is internal: insurgencies, ethnic conflict, regime crises
Subaltern realism (Ayoob); weak states (Buzan); insecurity dilemma (Job)
States are unitary rational actors
Army, parties, provinces and leaders shape policy (Kargil, Teesta, Sri Lanka)
Built on European and Cold War great-power experience
Global IR; Third World and post-colonial critiques
Conclusion
Structural realism is not wrong about South Asia, but it is incomplete. It captures the inter-state military competition among India, Pakistan and China, but it misses the internal, identity-based, non-state and non-traditional dimensions that dominate everyday insecurity in the region. It treats states as cohesive, sovereign actors, when South Asian states are post-colonial, plural and still building themselves. It cannot explain why rivalries persist despite their costs, why some partnerships are deeply friendly, or why policies change sharply with governments and leaders.
A realistic understanding of South Asian security therefore needs to combine realism’s insights about power with approaches that take seriously the nature of the post-colonial state, identity, domestic politics, non-state actors and human security. These approaches (subaltern realism, Regional Security Complex Theory, constructivism, securitisation, human security and feminist perspectives) are the subject of the final topic of this unit.
Quick Revision: Thinkers and Concepts
Concept
Thinker
Key work
Subaltern realism; late state-makers
Mohammed Ayoob
The Third World Security Predicament (1995)
Weak states; regime security
Barry Buzan
People, States and Fear (1983, 1991)
Insecurity dilemma
Brian Job (ed.)
The Insecurity Dilemma (1992)
Wars within states
K.J. Holsti
The State, War, and the State of War (1996)
Stability-instability paradox
Glenn Snyder
“The Balance of Power and the Balance of Terror” (1965)
Dangerous deterrent
S. Paul Kapur
Dangerous Deterrent (2007)
Organisational dangers of proliferation
Scott Sagan
The Spread of Nuclear Weapons: A Debate (with Waltz, 1995)
The concept of human security was popularised by the UNDP report shaped by:
(a) Amartya Sen (b) Mahbub ul Haq (c) Jagdish Bhagwati (d) Muhammad Yunus
Answer: (b)
Match the following:
A. Sovereignty: Organized Hypocrisy – 1. C. Christine Fair
B. Fighting to the End – 2. Stephen Krasner
C. Postcolonial Insecurities – 3. K.J. Holsti
D. The State, War, and the State of War – 4. Sankaran Krishna
Answer: A-2, B-1, C-4, D-3
Assertion (A): Nuclear deterrence has not prevented repeated crises between India and Pakistan. Reason (R): Nuclear weapons may encourage limited conflict and proxy war under the nuclear umbrella.
Answer: Both A and R are true, and R is the correct explanation of A.
Which relationship is often described as resembling a “security community”?