Civics · Law · Geopolitics: International Relations Theory

The two great traditions start from the same fact: international society is an anarchy—note, "anarchy" here does not mean "chaos" but rather "there is no higher authority above you," no world police to right your wrongs. From this identical starting point, the two schools reach opposite conclusions. Realism says: since no one can protect you, states can only rely on self-help, and the ultimate hard currency is power, especially military power; cooperation is temporary, suspicion is the norm. Liberalism says: anarchy is real, but it doesn't have to mean kill-or-be-killed—through trade interdependence, international institutions, and shared values, states can convert a "zero-sum" gamble into "positive-sum" cooperation. In a line: realism sees a jungle, liberalism sees a market where rules can be built.

• Realism reading great-power competition: regardless of ideology, great powers of comparable strength tend to guard against one another and pursue relative advantage—realists use this to explain why arms races and spheres of influence keep recurring, and why institutions and goodwill can't suppress structural suspicion.
• Liberalism reading European integration: European states that once fought repeatedly began with a coal and steel community and bound themselves layer by layer into the EU, making war among members almost unthinkable—liberals treat it as a showcase of "taming anarchy with institutions and interdependence."

Which is closer to the truth? Realism's strongest logic: countless "cooperations" in history collapsed the instant core interests clashed; resting peace on institutions and goodwill is like building on a foundation with no enforcement—one crisis and it caves in. Liberalism's strongest logic: if international politics were only ever pure power calculation, the long-run expansion of trade, international law, and transnational bodies would be inexplicable; by treating "suspicion" as iron law, realism ignores that rules can change the payoff structure of the game itself. The trade-off: do you see "anarchy" as an unsoftenable iron cage, or as raw material that can be gradually reworked?

"Realism = warmongering hawks, liberalism = naïve doves"—this mistakes an analytical lens for a policy stance. Realists are often restrained: precisely because they weight power and risk, they frequently oppose overextension and are wary of moral crusades. Liberals don't ignore power either; they simply bet more on the role of institutions and interdependence. Both are analytical frameworks, not the platform of any party or faction.

The oldest, most central mechanism in realism is the balance of power: when one state grows strong enough to potentially dominate, others instinctively band together to counter it, pulling the system back toward equilibrium. The logic is blunt—under anarchy, no one wants to be ruled by an overwhelming hegemon, so "balancing" is almost an automatic reflex. It resembles negative feedback in mechanics: when a force grows too large, the system generates an opposing force to pull it back. States can balance internally (arm themselves) or externally (form alliances). Kenneth Waltz, founder of structural realism (Theory of International Politics, 1979), stressed that balance is often not anyone's deliberate design but a spontaneous product of the anarchic structure—much as market prices aren't set by anyone but emerge from countless self-interested acts.

• The 19th-century "Concert of Europe": after the Napoleonic Wars, the great powers deliberately maintained a multipolar equilibrium in which any state's overreach would draw a coalition against it—seen as a classic period of the balance of power operating consciously.
• Cold War bipolarity: the postwar world contracted into two superpowers facing off, and Waltz and others actually argued that bipolarity is more stable than multipolarity—with only two players, who is balancing whom is unmistakable, leaving less room for miscalculation.

Is the balance of power a "law of nature" or an "after-the-fact narrative"? The law camp: balancing behavior recurring across centuries shows it captures a deep mechanism of anarchy and has real explanatory power. The skeptics: in reality states often don't balance (bandwagoning, waiting, free-riding are all common), and "power" is hard to measure objectively, so the theory risks becoming a hindsight story that can rationalize anything. A finer revision (Stephen Walt's argument that states balance against "threat" rather than raw power) tries to patch this: it isn't the strongest that gets ganged up on, but the most threatening.

"Balance = peace"—not necessarily. The balance of power describes a state in which no one can easily dominate, and such parity can bring restrained stability, but can also slide toward war precisely because forces are so close that each side thinks it has a chance of winning. Many great wars in history broke out exactly when an equilibrium was on the verge of collapse and the parties miscalculated. Balance guards against "sole domination"; it does not directly equal "no fighting."

Liberalism's trump card is international institutions—the UN, the WTO, treaties and international law of all kinds. The core puzzle: with no world police to enforce them, why does anyone obey these rules? Robert Keohane, a leading figure of neoliberal institutionalism, answers not with "morality" but with interest: institutions are useful because they lower the cost of cooperation—providing information, increasing transparency, and making "who keeps the rules and who breaks them" visible, thereby raising the cost of defection and lengthening the shadow of interaction. Dealing with each other repeatedly and fearing damage to their reputation, states become more willing to keep their word. In other words, institutions don't abolish anarchy; they act like credit records and surveillance cameras fitted onto a crowd of strangers, turning a one-off gamble into a repeatable long-run game.

The common thread: institutions excel at scenarios where everyone wants to cooperate but distrusts one another (coordination problems); once core interests clash head-on, institutions lacking enforcement are easily bypassed or hollowed out.

Do institutions have any independent effect? The institutionalist (liberal) view: institutions provide information, lock in expectations, and raise the cost of default—genuinely changing state behavior, not mere ornaments of power. The realist rebuttal (e.g., John Mearsheimer): institutions are merely a reflection of great-power will—strong states shape the rules, obey or discard them as needed, and when it truly matters, institutions can't stop the guns. A middle position: institutions matter markedly in low politics (coordination arenas like trade, technology, health) but are clearly limited in high politics (existential security issues). The crux of the debate is therefore: are institutions an independent variable, or a dependent variable of power?

"International law has no enforcement, so it's meaningless"—this underrates the mechanisms by which constraint works even without enforcement. Most states, most of the time, obey international law—not out of fear of jail, but because breaking it damages reputation, invites retaliation, and forfeits future chances to cooperate. The power of institutions often comes from "repeated games" and "reputation," not from a baton overhead. It isn't like domestic law with courts and police, yet it isn't an empty piece of paper either.

Realism and liberalism argue about "what to do under a given anarchy," both treating anarchy as a fixed physical backdrop. Constructivism flips the table and asks: is the very meaning of anarchy something we ourselves assign? Alexander Wendt's famous 1992 line—"anarchy is what states make of it"—means exactly this: the same "no authority overhead" can be understood as an every-man-for-himself jungle or as a community of friends, depending on the identities and ideas formed through states' long-run interaction. The key variable isn't material power but ideas, identity, shared meaning: the line between friend and foe isn't inborn but constructed through interaction—and can be reconstructed. In engineering terms: the first two schools optimize algorithms on given hardware, while constructivism says don't forget—even the definition of "what system this is" is written by software.

• "500 British nuclear weapons vs 5 North Korean ones": Wendt's classic example—for most Western states, the alarm triggered by the former is far less than by the latter. Threat is determined not only by weapon count (material) but by friend-or-foe identity (ideas): who holds them matters more than how many.
• Shifts in identity: former archenemies can become close allies, and vice versa. Material strength barely changes, but the mutual perception and identity change, rewriting the whole relationship; certain once-accepted behaviors (such as open colonization) also gradually came to be seen as "illegitimate," constraining states even without enforcement—all hard for a purely material view to explain.

Can ideas really override hard power? Constructivism's strongest logic: if everything were determined by material factors, one couldn't explain how identity and norms lead equal strength toward utterly different relationships; ideas aren't power's appendage but a force that shapes interests themselves. Critics (especially realists): constructivism excels at after-the-fact explanation yet struggles with prior prediction—it can't say clearly how or when ideas change; and when the shells fly, strength still has the final word. A fair placement: constructivism fills the blind spot of the first two schools (the origins of ideas and identity), but remains weaker at yielding falsifiable predictions. It's more a corrective lens than a new map that replaces the others.

"Constructivism = everything is subjective, changeable at will"—this reads it as voluntarism. Constructivism doesn't deny that material power exists, nor claim ideas can be conjured freely; its point is that once ideas sediment through long interaction into shared identity and norms, they become extraordinarily durable and hard to shake—spirals of hostility and communities of trust can both self-reinforce. It speaks of "social construction," not "a single change of heart."