Level III: International Structure as Non-Material (Norms & Ideas)
This week’s main claim is that international structure is not just material but social: identities and interests are not exogenously given but produced and transformed through interaction, shared knowledge, and norms. Wendt makes the foundational move, “anarchy is what states make of it.” Sjoberg shows that structure is also gendered. The empirical readings show norms doing real work by raising the burden of proof against nuclear use and territorial conquest.
Reading list
- Wendt. 1992. “Anarchy is What States Make of It: The Social Construction of Power Politics.” International Organization.
- Sjoberg. 2012. “Gender, Structure, and War: What Waltz Couldn’t See.” International Theory.
- Finnemore and Sikkink. 1998. “International Norm Dynamics and Political Change.” International Organization.
- Tannenwald. 1999. “The Nuclear Taboo: The United States and the Normative Basis of Nuclear Non-Use.” International Organization.
- Zacher. 2001. “The Territorial Integrity Norm.” International Organization.
- Application: Schwartz. 2024. “The Nuclear Taboo is a Myth.” Lawfare. · Fazal. 2022. “The Return of Conquest?” Foreign Affairs.
Wendt (1992): Anarchy Is What States Make of It
International Organization · the social construction of power politics
Wendt aims to build a bridge between rationalist theories (realist and liberal) and constructivism, arguing that the structure of international politics, self-help and power politics, does not follow from anarchy itself but is socially constructed through state interaction. His key move is to make identities and interests endogenous: they are not exogenously given but are formed and transformed through processes of interaction, institutions, and shared knowledge. Rationalists (neorealists like Waltz plus neoliberals like Axelrod, Keohane, and Nye) take identities and interests as given, and treat process and institutions mainly as behavioral constraints that change actions, not identities, a behavioral, individualist model focused on how behavior produces outcomes. Constructivists argue that identities and interests are socially constructed through interaction and intersubjective meaning; they focus on how practices and institutions shape what states are, not just what they do. This is the section that delivers the famous line: “Anarchy is what states make of it.” Realists reason that under anarchy there is no central authority and no collective security, so states must rely on themselves to survive; this yields a self-help system, competition, security dilemmas, and power politics. Wendt denies the causal link: there is no intrinsic “logic” of anarchy apart from the practices that constitute identities and interests. Self-help and power politics are institutions, not essential features of anarchy. So if we find ourselves in a self-help world, it is because processes, historical interactions, practices, and mutual interpretations have produced and reproduced competitive identities. It is not because the structure dictated it. “If states find themselves in a self-help system, this is because their practices made it that way.” The neoliberal institutionalists, and much of Milner’s rationalist work, generally accept the rationalist starting point (identities and interests exogenous) and then show how rules reduce transaction costs and enable cooperation within a self-help world; Wendt calls these “weak liberals” because they concede anarchy’s causal powers. Wendt’s move is deeper: he endogenizes identities and interests and shows institutions can transform them. He invites “strong liberals”, those who care about complex learning and changing conceptions of self and interest, to join constructivists in a process-oriented theory. In fairness, he notes Axelrod hints at norm internalization, but says such insights require moving beyond exogenous identities.
Two actors, “ego” and “alter,” meet for the first time; their threats are not natural but socially constructed through gestures, interpretations, and responses, and that first interaction sets the tone for cooperation or competition. It is not a historical narrative but a theoretical tool, counterfactual reasoning aimed squarely at neorealists like Waltz. Waltz, for parsimony, strips away unit-level (first- and second-image) attributes and deduces self-help directly from anarchy; Wendt turns the same abstraction against him: even two entities with no historical baggage do not get self-help automatically, it emerges through interaction. The experiment shows that threat and enmity are products of interaction, not inherent properties of anarchy. It also shows that an important intermediate variable sits between anarchy and power politics: the formation of intersubjective meaning. Its force is logical plausibility, not historical truth. It exposes the neorealist conclusion as resting on an unstated, pessimistic assumption about actors’ initial intentions. History and culture do not break the looking-glass self; they make it more powerful and easier to observe. Real actors are never blank slates: they read each other’s gestures through inherited cultural concepts, historical narratives, and ideological frames (“we are the Middle Kingdom, they are barbarians”). That very “misreading” is the mechanism at work. Continued interaction yields new information that forces revision; Gorbachev’s “new thinking” tried to change the Western mirror by changing Soviet gestures, and so shifted the USSR’s identity from enemy to partner. My reservation is one of scope: applying a micro-level cognitive theory to first encounters between states raises real questions. In historically real first contacts, especially between linguistically distinct groups with prior social complexity, resource competition and hostility often appear quickly, and where cultural legacies are thin, evidence for a “culture leads away from self-help” pathway is scarce. Anthropology suggests social differentiation is a precondition (not the cause) of state formation, that early differentiation was unstable, and that enclosure mechanisms (control of arable land) trapped populations and induced submission; where fertile land and migration options were limited, ecological constraint amplified competition and made mutual vigilance likely. So the anecdote is elegant for theory-building, but ecological and anthropological dynamics better explain why self-help became widely accepted in practice, even though, as Wendt rightly insists, it is not logically entailed by anarchy. The “Hobbesian world” is a baseline in which security rests only on material power, with no social recognition of rights. Sovereignty transforms this: it is the mutual recognition of each state’s right to exclusive political authority within territorial limits, a social institution (a form of “social closure”) that constitutes territorial property rights and stabilizes interaction. It is so strongly institutionalized because it reduces uncertainty and organizes expectations: by “closing” who has rightful authority where, it disempowers non-state challengers and lowers the baseline risk of Hobbesian predation, and repeated practice turns it into a taken-for-granted rule, a selection advantage for the norm. The EU shows the contrast between paradigms. A rationalist reads its persistence as absolute gains, lower transaction costs, and managed uncertainty, with identities exogenous; a constructivist reads decades of cooperation as having reconstructed member identities into a shared “European” identity, so interests become increasingly collective and norms internalized. For persistence I find the constructivist story stronger: it explains why cooperation endures even when the material price is right to defect, because who “we” are has changed, and identities and security interests are continuously in process.
Sjoberg (2012): Gender, Structure, and War
International Theory · what Waltz couldn’t see
Sjoberg distinguishes sex (perceived biological classes) from gender (socially learned expectations and attributes attached to “masculinity” and “femininity”), and treats gender as a system of symbolic meaning that creates social hierarchy by valuing masculine-coded traits over feminine-coded ones. Feminist IR studies how social expectations about masculinity and femininity shape global politics, including the causes and conduct of war; gender analysis is “thinking about the way that social expectations about masculinities and femininities influence the constitution, processes, and structures of global politics.” Her aim is to theorize Waltz’s third image through a feminist lens: the international system is anarchic and gender-hierarchical, and that gendered structure helps explain the recurrence of war. If gender hierarchy is part of structure, it would order the system, specify unit functions, and distribute capabilities. She presents this as preliminary, a research agenda inquiring into gender hierarchy as structural, bridging feminist IR with mainstream structural realism. Traditional third-image scholarship (Waltz) treats anarchy as the defining structural feature, driving states to compete for power and security. A feminist third image argues the system is also structured by gender hierarchy, which shapes state identities, capabilities, and interactions: “gender hierarchy is a key part of the structure of the international system, though unseen by most.” Where Waltz sees structure as material and anarchic, the feminist account sees it as socially constructed and gendered, shaping even the concepts of power, honor, and conflict. Importantly, this is a third-image claim, not a second- or first-image claim. It is not a claim about gendered states or gender subordination as human nature. Sjoberg uses Waltz’s own criteria for structure (ordering principle, functions of units, distribution of capabilities) to ask what each looks like once the structure is read as gender-hierarchical. Her answer is yes: gender hierarchy is such a property. Like anarchy in Waltz, it permits war; but because gendered ideals differ across time and place, it can also help explain variation across wars. The structure (gender as an ordering principle) is constant, while “the genders and their orders are fungible.” Structural gender hierarchy can therefore account for different levels of competition and bellicosity. It speaks to wars in the plural, not just war in general. I find the move analytically persuasive on two counts: it keeps Waltz’s virtues (parsimony, level-of-analysis clarity) by using his own structure test, and it improves on him by supplying within-structure variation through shifting masculinities and femininities. Whether one fully agrees turns on causal priors, on whether you accept that capabilities, status, and even “threat” are partly constituted by gendered meanings.
Running gender hierarchy through Waltz’s three tests: as an ordering principle, it consistently organizes social and political value by valorizing masculinities and devaluing femininities, incentivizing relative displays of masculinity and thus competition. For unit functions, gendered national identities help decide whether a state acts as protector, aggressor, or peacekeeper, and why some fuse “tough” and “tender” roles. For the distribution of capabilities, states rank one another by how they measure up to idealized masculinities (a “muscular/militarized masculinism”), and gendered scripts shape what even counts as capability. Her examples: Belgium 1914, where masculine honor (refusal to yield) was tied to the purity of feminized territory and the invasion remembered as the “Rape of Belgium”; India–Pakistan as a “clash of masculinities” valorizing martial non-compromise and feminizing the other; the United States’ “tough-but-tender” masculinity in the First Gulf War (overwhelming attacker and chivalrous protector at once); and the US–Cuba story, where conquest, loss, and recovery are read through gendered desire and emasculation. My own additions: Putin’s hypermasculine rhetoric, post-9/11 “cowboy” talk, ISIS propaganda against “impure” others, China’s “wolf-warrior” masculinity versus “responsible stakeholder” calls, and NATO’s chivalric-protector self-presentation in humanitarian interventions. Sjoberg notes that race, religion, culture, and ethnicity also pervade international life and are often performed in gendered terms (others feminized, masculinity prized), Waltz’s own work even implies the possibility of alternative structures. Foregrounding gender helps explain why states pursue symbolic goals like honor and prestige beyond material interest, how militarization and war are culturally sustained through gendered narratives, why some violence (sexual violence in war) is systematic rather than incidental, and how institutions and norms (Responsibility to Protect) are gendered in conception and implementation. The deeper point is fourfold. First, gender does not deny material power; it explains how material power is valued, nuclear weapons are coded as the “ultimate masculine power” while food security and care labor rarely count as core strength, not because the former is more material but because the system has long linked masculinity to “real power,” so a gendered structure tells us which things register as capabilities. Second, gender shapes how states use and display power: the Cold-War US stressed hard power and containment because “weakness” read as feminine and meant lost prestige, and the USSR kept up the arms race to show “manliness”; legitimacy and moral politics also run through gender, as wars and sanctions get justified to “save women and children” while target societies are feminized. Third, gender explains how capability differences become hierarchical order, European powers cast colonies as “feminized and immature” to legitimize domination, and post-Cold-War America cast “weak states” as women or children needing a masculine protector. Fourth, gendered narratives shape identity: “men fight, women are protected,” women’s bodies stand for the nation, and wartime rape becomes a tolerated mechanism of domination, Belgium’s honor lay in protecting feminized territory, while Soviet soldiers’ rapes in Germany framed violation of enemy women as restoring masculine national honor. Both show gendered symbols are indispensable variables for explaining state identity and wartime behavior.
Finnemore & Sikkink (1998): International Norm Dynamics
International Organization · the norm life cycle
A norm is a shared standard of appropriate behavior for actors with a given identity. Norms carry “oughtness” and a shared moral assessment: people judge behavior against them, praising conformity and stigmatizing violation. Because of that oughtness, a norm’s presence shows up indirectly, actors feel compelled to justify departures, leaving a “trail of communication” researchers can study (their example is US explanations for land-mine use in Korea as evidence of a growing anti-landmine norm). They deliberately separate measuring a norm’s existence and strength from measuring behavior, to avoid circularity when studying the norm’s effects. Stage 1, norm emergence: norm entrepreneurs with organizational platforms persuade by framing issues to make new standards legible and resonant, often from altruistic or ideational motives, sometimes through provocative tactics (women’s suffrage activists). The tipping point comes when enough, especially “critical”, states endorse the norm, at which point identity-based reasons for conformity kick in. Stage 2, norm cascade: states, IOs, and networks drive it through active socialization (praise, censure, incentives, sanctions, emulation), not passive contagion, motivated by legitimacy, reputation, and esteem (the ICRC monitoring after the first Geneva Convention). Stage 3, internalization: law, professions, and bureaucracies make the norm taken-for-granted, enforced by habit, routine, and professional standards. My take: as a mid-range process theory it is powerful because it links who matters to how influence works at each stage and yields observable implications, framing and justification trails in Stage 1, socialization signals and rapid treaty uptake in Stage 2, low-salience rule-following in Stage 3. Its limits (nonlinearity, reversals, hybrid logics) are real, but the authors explicitly say different logics of action dominate different stages rather than claiming a strictly linear process. Entrepreneurs are agents, often in advocacy networks, NGOs, or government, who name, dramatize, and frame problems so they resonate with existing beliefs. They are “meaning managers” competing in a crowded normative space, and they sometimes must perform deliberately “inappropriate” acts (civil disobedience) to shift the logic of appropriateness. The authors’ classic examples are Henri Dunant and colleagues (ICRC and Geneva law) and suffrage leaders (Stanton, Anthony, Fawcett, Pankhurst). My modern illustrations: Jody Williams and the ICBL on landmines, ICAN on the nuclear ban, and youth climate leaders linking climate to human rights, each fits the framing-plus-networks-plus-platform profile.
Norms stall below the tipping point for several reasons: issue characteristics and fit (norms tied to visceral harm to innocents, like land mines, travel faster); adjacency and path dependence (claims that fit existing legal-normative frameworks spread more easily); and framing challenges (activists often must re-label practices to hook into stronger accepted norms, as in shifting “female circumcision” to “female genital mutilation” under “violence against women”). Where fit is weak, frames do not resonate, or the platform is thin, the norm stalls. Emerging today (my assessment): AI accountability and autonomous-weapons constraints, climate and deforestation due diligence in trade, corporate ESG due diligence in supply chains, and data privacy as a human right. Once-contested but now largely internalized: diplomatic immunity in most contexts, the child-labor taboo, racial equality, the anti-apartheid norm, and anti-colonial self-determination, each now routinized in law and professional practice. The general test still applies: look for oughtness, justification trails, and socialization into law and professions. The authors reject a sharp rationalist/constructivist divide. Rationalists emphasize instrumental choice under constraints; constructivists emphasize identities, appropriateness, and meaning. But many important processes are strategic social construction: actors rationally deploy frames and networks to reconfigure others’ preferences and identities within a normative context, so rationality and norms are intertwined rather than opposed, we should study how strategic persuasion and socialization operate within social structures. A Waltzian would treat norms as epiphenomenal to power and the distribution of capabilities, with “socialization” largely the emulation of successful great powers under anarchy. Finnemore and Sikkink reply that socialization is active (praise, censure, incentives), not passive imitation, citing Waltz’s own discussion of emulation, praise, and ridicule, and that after the tipping point, state identity (membership in international society or a subset like liberal states) motivates rapid adherence, with IO and NGO networks pressuring laggards. That is not reducible to capabilities. For explaining variation in which rules get adopted when, I find their account more convincing because it supplies mechanisms and testable signals a purely structural account tends to miss.
Tannenwald (1999): The Nuclear Taboo
International Organization · the normative basis of nuclear non-use
Non-use cannot be explained by material factors alone. An evolving normative prohibition, the “nuclear taboo”, helped delegitimize nuclear weapons, constrain leaders, and embed deterrence practices within a wider web of norms that stabilized self-help behavior. Her mechanism is probabilistic: norms do not uniquely determine outcomes but shift probabilities and shape the realm of the possible for decision-makers, decreasing the likelihood of use. A norm is a shared expectation or standard of right and wrong that prescribes or proscribes behavior. A taboo is a particularly forceful normative prohibition tied to a “sociology of danger”, things that are “not done, said, or touched,” often carrying expectations of awful consequences if violated. Tannenwald uses “taboo” partly because leaders themselves used that language, and because over time the non-use norm acquired more taboo-like qualities (absoluteness, taken-for-grantedness), even though it was not fully a taboo in the earliest years. Regulative effects (emphasized by rationalists): norms constrain pre-existing activities, as an injunction against first use constrains an otherwise available option. Constitutive effects (emphasized by constructivists): norms create and define practices, roles, and identities, the taboo helps define the category “weapons of mass destruction” and marks what it means to be a “civilized” state, one that participates in regulating warfare. Permissive effects (a subset of constitutive): shadow effects whereby focus on the nuclear taboo diverts attention and shields other destructive practices, in the 1991 Gulf War, leaders framed advanced conventional strikes as “as destructive as nuclear weapons,” normalizing high-destruction conventional operations while avoiding nuclear stigma.
There are two materialist stories. Rational deterrence: fear of retaliation explains non-use among the superpowers, but it does not cover non-use against non-nuclear adversaries. Non-deterrence material factors: fear of long-term consequences, public-opinion constraints, organizational unreadiness, shortages of bombs or delivery vehicles, and so on. Her critique is that even where material factors matter, they are often indeterminate and socially interpreted. Key “material” variables such as public opinion or “readiness” may themselves be shaped by the taboo. The line between material and normative causes is therefore not always clean, so a strict competitive test is limited. My take: she is right that deterrence alone cannot explain US restraint against non-nuclear opponents (Vietnam, the Gulf War). The paper’s strength is that it shows how leaders actually talked: the burden of proof, the “civilized” identity, and the WMD category. This gives the taboo force beyond costs and benefits. The most telling evidence is the shifting “burden of proof”: over time it moved from justifying non-use (1945) to justifying even thinking about nuclear options (Vietnam), indicating deeper internalization. Her case contrast over time tracks this pattern. In 1945 Japan, the bomb seemed “as legitimate as any other” deadly weapon, and the burden lay on opponents of use. In Korea, Dulles recognized a taboo-like constraint and failed to “normalize” tactical nukes, while Eisenhower acknowledged strong public opposition. In Vietnam, US leaders “did not come close” to use despite costs, targets, and no fear of nuclear retaliation. In the 1991 Gulf War, they did not use nuclear weapons because they did not need to and did not want to; the taboo operated both instrumentally and as a taken-for-granted rule. The permissive shadow recurs: precisely because nuclear use was stigmatized, the coalition embraced extreme conventional destruction framed as “as destructive as nuclear weapons.” On case-study quality by our seminar criteria, the design is strong: four cases vary the dependent variable (use vs. non-use) across before, during, and after the Cold War; she sets up a competitive test of taboo vs. materialism and reflects on its limits (material variables may be norm-infused); and the evidence base is extensive, FRUS, archival notes, leader memoirs, polling, and author interviews (State Department Policy Planning). The one reasonable critique, which she anticipates, is that disentangling material from normative is intrinsically hard, but that difficulty is precisely her theoretical point, and she argues the taboo “captures more of the evidence” even where the two stories overlap.
Zacher (2001): The Territorial Integrity Norm
International Organization · the decline of conquest
Zacher rehearses the standard story: medieval Europe lacked clearly fixed state borders, and the “modern” shift was the rise of delimited territorial states, often dated to Westphalia (1648) though it emerged gradually. Early on, territory was treated as dynastic property, changing hands by inheritance, marriage, war, compensation, and purchase; conquest was legally legitimate and most wars concerned the acquisition of territory. This accords with mainstream IR history, Westphalia as a waypoint rather than a clean break, territoriality rising over time, and the early-modern legality of conquest. I would add the colonial qualifier: from the nineteenth century, European international law was projected as global but entry was conditional on “standards of civilization.” “Semi-civilized” states got only partial recognition and extraterritorial jurisdiction; “uncivilized” lands were cast as terra nullius open to annexation, a hierarchy institutionalized in Article 22 of the 1919 League Covenant, with its mandates as a “sacred trust of civilization.” Zacher crystallizes the interwar debate into three linked questions. First, may victors take territory from the defeated? Early in WWI major powers still backed this right (secret treaties), but Versailles largely limited gains in Europe, even as colonial redistribution continued via mandates. Second, should states commit to uphold one another’s territorial integrity? Wilson’s Fourteenth Point and League Covenant Article 10 were the first formal endorsement of a territorial-integrity obligation. Third, should national self-determination trump existing borders? Democratic governments were split, but self-determination for ethnic nations generally lost to boundary-stability at Versailles, it was not even in the Covenant. My read of the norms today: very strong prohibition on victors taking territory (a legal and political norm since 1945, regionally reinforced); strong mutual respect for territorial integrity in law and diplomacy, though enforcement varies with power and institutions; and mixed and contested self-determination, strong for decolonization-era “peoples” inside colonial units, but weak and conditional for ethnic self-determination or secession within existing states, where it is usually subordinated to territorial integrity. Territorial integrity is a duty not to use force to change recognized interstate boundaries (League Article 10; UN Charter Article 2(4)). National self-determination is a people’s claim to its own political community, though in practice, after 1945 the UN applied it to colonies (the 1960 and 1970 declarations) rather than to ethnic groups inside existing states. The tension is structural: self-determination, especially for ethnic nations that span borders, invites boundary change, while territorial integrity resists forcible change and privileges stability. Zacher notes democratic leaders themselves were in tension over whether self-determination should outrank existing borders, and that self-determination generally lost.
Interwar weakness: despite Kellogg–Briand (1928) and the Stimson Doctrine (1931), politics cut against enforcement, the 1919 settlements were seen as inconsistent and inequitable (Italy’s gains where “few Italians lived”; millions of Germans left outside Germany), and the great powers were divided and unwilling to use force to uphold borders. Post-1945 emergence and consolidation: Zacher applies Finnemore and Sikkink’s life cycle, a bloody emergence (learning from the 1930s), an acceptance stage beginning with UN Charter Article 2(4) and taking hold in the 1960s–70s (the 1960 UN Declaration, OAU 1963/64 principles, the 1975 Helsinki Final Act), and an institutionalization stage since 1976 via repeated responses (Ethiopia–Somalia, Iraq–Kuwait, Yugoslavia) and juridical support (the ICJ’s uti possidetis), reinforced by EU and NATO accession conditionality. I broadly agree: the interwar norm lacked legitimacy and enforcement, and after 1945 the legal shell was progressively filled by regional practice, organizational policing, and repeated case outcomes. Zacher argues the drivers were both instrumental and ideational, with instrumental dominant, fear of major war, alliance politics, and the costs of conquest, which fits the historical timing. I would add a deeper normative-shift story: the brutality of two world wars produced a sense of civilizational crisis and a need for a final reckoning (like the postwar trials) that symbolized the end of an old order and enabled a shared reorientation toward territorial integrity. Using Holsti’s 1648–1945 war list and his own post-1946 compilation, Zacher documents a sharp decline in successful redistributions: from roughly 80% of territorial wars before 1945 to 30% after, with the per-country-year rate of redistribution falling dramatically. He is transparent about limits, post-1946 cases include smaller aggressions (at least a week, with clear intent to change borders), which reduces comparability, and he disclaims statistical significance. The pattern, fewer successful conquests, stronger IO pushback, and legalization through uti possidetis, is suggestive of a robust norm. On entrepreneurs, where Zacher is thin, I see the UN (Charter Article 2(4), the 1960 and 1970 declarations, peacekeeping and monitoring), the OAU/AU (the 1964 resolution to respect borders existing at independence), the EU and NATO in the 1990s (membership conditionality requiring settled borders or ICJ referral), and Western great powers, especially the US and European states, who pressed for acceptance of the internal administrative borders of Yugoslav and Soviet successor states, often through the OSCE. A realist/materialist account, contrasting with Tannenwald’s emphasis on internalized moral opprobrium, would say states refrain from conquest because it is costly and risky in an era of nuclear deterrence, dense alliances, trade interdependence, and IO sanctions, not because of moral transformation. Zacher’s own balance is both, with interests dominant: fear of great-power war (the association of territorial revisionism with major wars was central), alliance structures, and the costs of conquest; Helsinki itself aimed to avert major war. My view is that the evolution mixes both, but the sequencing is telling, legal codification plus repeated enforcement created the material and reputational cost structures that later helped internalization, so interests came first and ideas later, very close to Zacher’s own balance. I would add a nation-building mechanism: as printing and modern communication spread national cultures, territory came to be seen as belonging to the nation rather than the dynasty, making cession intolerable; governments promoted territorial nationalism instrumentally, but once citizens internalized it, it became a genuine constraint, so interests and ideas are deeply intertwined (resource nationalism is another such alternative). Why are some skeptics? Three reasons, mostly addressable from the text: exceptions and ambiguities (Turkey’s 1974 invasion of Cyprus succeeded, though it was not widely recognized and seen as reversible, Zacher’s own point); alternative explanations (deterrence, US hegemony, alliance politics, and costs rather than norms, which Zacher himself foregrounds); and measurement concerns (datasets mix definitions across eras and include small post-1946 cases, which he notes while disclaiming statistical significance).
Application · Theory in current events
Schwartz. 2024. “The Nuclear Taboo is a Myth.” Lawfare. · Fazal. 2022. “The Return of Conquest?” Foreign Affairs.
Schwartz tests the nuclear taboo with survey experiments in the United States and India, in which respondents read scenarios where a country considers using nuclear weapons (a war with Iran; destroying a deeply buried terrorist nuclear lab), randomizing which country pushes the button, the respondent’s own government, an ally or partner, or a non-ally, while holding other features constant. His main finding: support for nuclear use is high (roughly 37–54%) and is not weaker when an ally or partner uses the bomb than when one’s own government does, though it is lower for non-allies; support also drops when respondents get vivid information about human costs or illegality. He concludes the evidence is inconsistent with an absolute, unthinking taboo, at best there is a conditional non-use norm that varies by identity and framing.
How valuable are such surveys for assessing a taboo? Informative but partial. They are valuable for mapping the limits of public aversion, how identity cues and scenario design move support, and how human-cost or illegality framing weakens it, but incomplete for a taboo in Tannenwald’s sense. For her, a taboo is a de facto prohibition that changes what counts as thinkable and shifts the burden of proof away from use; it lives not just in mass attitudes but in elite “taboo talk,” legal-institutional structures (treaties, nuclear-weapon-free zones, legal opinions), and observed practice during crises. Surveys cannot directly observe elite deliberation, so they should be triangulated with elite records, institutional constraints, and the long record of non-use. Schwartz’s policy implications follow from weak guardrails: recommit to arms control and resist a broad nuclear buildup; reprioritize nonproliferation (states cannot be tempted to use what they do not possess); and constrain first-use authority, in the US, requiring concurrence from the Vice President and Attorney General before a first strike.
Has the taboo eroded since Tannenwald wrote? Two truths coexist. Risky nuclear behaviors and rhetoric have intensified, Russia’s threats during the Ukraine war, the collapse of the INF Treaty, the suspension of New START, Trump’s “fire and fury,” Iran’s enrichment, and a rapid Chinese buildup, creating a permissive environment for thinking and talking about nuclear use. But Tannenwald’s taboo was about use, not threats, and it rested on elite taboo talk, identity, and institutionalization that raised the burden of proof for crossing the threshold; by the 1991 Gulf War, leaders called nuclear use “appalling,” “barbarian,” and outside “civilization,” evidence of both regulative and constitutive effects, which is stronger than transient public moods. My read: the discursive climate has clearly worsened, and Schwartz shows publics can be mobilized to accept use under some conditions, yet there has still been no use since 1945, and elite burden-of-proof dynamics and institutional constraints continue to matter. If anything, his results make Tannenwald’s policy implication more urgent, thicken the constitutive and regulative layers (legal regimes, alliance expectations, elite identity talk) that keep use “unthinkable,” while watching her permissive “shadow effect” that shifts destruction into advanced conventional means.
Fazal turns the same lens on the territorial-integrity norm. Russia’s full-scale invasion is the most serious test of the post-1945 norm against conquest: if the world lets Russia subsume Ukraine, it could normalize using force to revise borders, revive empires, and push some states toward extinction, and the effect depends both on how far Putin goes (puppet regime vs. annexations vs. absorbing the whole state) and on how the world responds, since norms are “nourished by enforcement.” Her prescriptions: sustain sanctions and legal pressure, not fully lifting sanctions unless Russia recognizes Ukraine’s pre-2014 borders; use international courts seriously, watching the crime-of-aggression track; pursue UN reform so the P5 cannot veto ICC referrals for aggression; but balance resolve against escalation, because upholding the norm is not worth a third world war. The consequences of failure: the highest extinction risk falls on neighbors of great powers; annexations bring more indiscriminate targeting and brutality against civilians; a return of “violent state death” and more interstate wars over borders; and even if Ukraine “resurrects,” a failure to punish would teach would-be aggressors that conquest is possible. And she insists the war “is about much more than Russia and Ukraine,” because eroding the anti-conquest norm would take the lid off dozens of disputes, Armenia–Azerbaijan over Nagorno-Karabakh, Sudan’s borders, the South and East China Seas, Taiwan.
Where I push back: I am not convinced the invasion’s effects travel so far. First, Russia is struggling, Putin is already militarily overstretched, and without a decisive victory that lets him secure and hold Ukrainian territory, the case is poorly suited to set a precedent that encourages future conquests. Second, even if some states were tempted, other powerful norms and practical limits act as barriers: in today’s order governments are expected to derive legitimacy from protecting their people’s welfare, and conquest usually imposes heavy costs, high casualties, sanctions, and resistance from populations who do not share the occupier’s identity, making it unattractive except where the weaker side has almost no capacity to resist. The most likely such candidate today is the Israeli–Palestinian conflict, yet even there Israel’s actions are already seen as deeply violating humanitarian principles, which shows that even if the territorial-integrity norm weakens, the humanitarian norms against indiscriminate violence and occupation can still serve as strong constraints.
← Back to the genealogy tree · ← Prev: Week 4, International Structure as Material · Next: Week 6, Domestic Politics & Cooperation →