The Thucydides Trap

In brief

Graham Allison, a political scientist at Harvard's Kennedy School, coined the phrase "Thucydides Trap" in a 2012 Financial Times article and made it famous in a 2015 Atlantic essay and a 2017 book, Destined for War. The name comes from the ancient historian Thucydides (c. 460 to c. 400 BCE), who wrote of the war between Athens and Sparta that "it was the rise of Athens and the fear that this instilled in Sparta that made war inevitable." Allison's claim is structural rather than moral: when a rapidly rising power threatens to displace an established ruling power, the mismatch between the two creates pressures that historically end in war more often than not. He assembled sixteen cases of such transitions over the last five hundred years and reported that twelve of them ended in war. The book's real subject is the present: can the United States, the incumbent, and China, the challenger, escape the pattern? Allison's answer is that war is not inevitable, but avoiding it will take extraordinary statecraft.

The full treatment

The problem it answers

Why do great powers, often against their own obvious interest, stumble into catastrophic wars? Statesmen rarely want the wars they get. The First World War ruined every empire that fought it, yet the powers marched in anyway. Allison's answer borrows Thucydides' diagnosis: the deep cause is not any single decision but a structural condition. When one state grows fast enough to challenge the top spot, two emotions take over. The rising power feels a new sense of entitlement and demands more say. The ruling power feels fear and a determination to hold what it has. Each reads the other's ordinary behavior as threat. A minor incident, a provocation neither side fully controls, can then ignite a conflict that neither wanted. The trap is that rational actors, each responding sensibly to the other, produce a collectively disastrous result.

How it works

The mechanism is a feedback loop between capability and psychology. As the challenger's economy, military, and confidence swell, it presses for recognition and revision of arrangements set when it was weaker. The incumbent, seeing its margin erode, grows anxious and inflexible. Allison isolates the ingredients that turn this stress into war: a rising power's grievance and ambition, a ruling power's fear of decline, allies who can drag their patrons into quarrels not of their choosing, and accidents, third-party provocations that force a response. Sparta did not want war with Athens; it was pushed by its allies, above all Corinth, and by fear of what Athenian growth would eventually mean. The structural stress does not guarantee war. It loads the dice.

The key example and the historical survey

The founding case is the Peloponnesian War (431 to 404 BCE), in which a rising Athens and a dominant Sparta fought a generation-long conflict that wrecked the Greek world. Allison's Harvard Belfer Center project then catalogued sixteen instances over the past five centuries in which a rising power challenged a ruling power. Twelve, by his count, ended in war. The classic war cases include France against the Habsburgs in the sixteenth century, the Habsburg-Valois Wars, and Germany against Britain before 1914, the transition most often invoked as a warning. The four peaceful cases matter as much: Portugal ceding maritime primacy to Spain in the late fifteenth century, the United States overtaking Britain around the turn of the twentieth century, the American-Soviet rivalry of the Cold War held short of direct war by nuclear weapons, and Germany's peaceful rise within a unifying Europe alongside Britain and France. The survey's purpose is diagnostic, not deterministic: twelve of sixteen is a warning, and the four exceptions are the material for escape.

Distinctions that matter

The Thucydides Trap is a popularization, not an original academic theory. Its scholarly parent is power-transition theory, set out by A. F. K. Organski in World Politics (1958) and developed with Jacek Kugler. Organski argued that the international system is hierarchical, that war is most likely not when one side is dominant but when a dissatisfied challenger approaches parity with the leader, and that the danger peaks around the crossover point. Allison's contribution is to dramatize this with a memorable name and a historical case file aimed at policymakers. It should also be distinguished from balance-of-power thinking, which holds that roughly equal power keeps the peace. Power-transition theory reverses that intuition: it is the approach to parity, not imbalance, that is dangerous.

Lineage

The idea's oldest root is Thucydides himself, whose account of Athens and Sparta is often called the first work of realist political analysis. Its modern academic form is Organski's power-transition theory (1958), which sits within the broader tradition of political realism, the view that international politics is a competition for power among states with no authority above them. Structural realists such as Kenneth Waltz (1924 to 2013) supplied the underlying claim that the shape of the system, not the character of leaders, drives great-power behavior. The offensive realist John Mearsheimer (born 1947) reached a parallel pessimistic conclusion about China by a different route. Allison stands downstream of all of them, translating a century of academic argument into a phrase and a story that presidents and prime ministers actually read.

The strongest case for it

The pattern is real and the record is sobering. Rapid shifts in relative power have repeatedly coincided with the largest wars in history, and the psychological mechanism, a rising power's assertiveness meeting a declining power's fear, is visible in the diplomatic record before 1914 and in many earlier crises. The framing has genuine practical value: it directs attention away from blaming individual villains and toward the structural pressures that make even prudent leaders dangerous, which is where policy can actually intervene. It takes the risk of accidental war seriously. And Allison is careful to insist the trap can be escaped, offering the four peaceful transitions as proof that structure is not destiny. As a device for making leaders think hard about a rivalry before it hardens into confrontation, it has few equals, which is why it entered the vocabulary of officials in Washington and Beijing alike.

The strongest case against it

The critics are numerous and their objections cut deep.

Historians attack the case file. Arthur Waldron, in a widely read 2017 rebuttal, argued flatly that "there is no Thucydides Trap," and that Allison misreads Thucydides: the Peloponnesian War had many contingent causes, and the famous line about inevitability may reflect Thucydides' rhetoric more than a law of history. Waldron pointed to classicists such as Donald Kagan who read the Greek text as showing that Sparta did not want war at all. Ian Buruma, reviewing the book in The New Yorker, charged that the sixteen cases are cherry-picked and inconsistently coded, that stretching "rising power versus ruling power" across five centuries lumps together situations too different to compare, and that counting wars this way manufactures a pattern.

Political scientists question the coding and the causation. Steve Chan, in a book-length study, argued that the case selection and the labels do not withstand scrutiny and that power transitions do not reliably produce war. Critics note the selection problem: focusing only on cases where a challenger arose ignores the many rivalries that never turned violent, and correlation between transition and war does not establish that the transition caused it.

Liberals and institutionalists argue the model omits what now restrains great powers. Joseph Nye (1937 to 2025) contended that treating US-China relations as a rerun of 431 BCE ignores nuclear deterrence, deep economic interdependence, and international institutions, all of which raise the cost of war far above anything Sparta and Athens faced, and that fatalistic framing risks becoming a self-fulfilling prophecy by teaching each side to expect the worst. Others add that Allison's structure leaves little room for the choices of leaders, the very statecraft his own conclusion depends on.

Where it stands now

The phrase has escaped the academy entirely. Chinese President Xi Jinping has referred to the Thucydides Trap in public remarks, urging that it be avoided, and it recurs in American strategic debate over how to manage China's rise. That very success is part of the critique: a contested academic hypothesis has become a piece of conventional wisdom, shaping how both governments interpret each other's moves. Within international relations the scholarly verdict is mixed. Power-transition theory remains a respected research program, but Allison's specific sixteen-case argument is treated with caution, prized as a provocation and a teaching device more than as proven social science. Its lasting effect may be less as a prediction than as a warning label on a rivalry that both sides now watch through its lens.

Test yourself

Allison says war is not inevitable, yet names the dynamic a "trap." Notice the tension: if leaders on both sides are studying the trap in order to avoid it, does naming it make war less likely, or does teaching each side to expect the other's aggression make the trap more real? Decide which effect you think dominates, and what that implies about how the United States and China should talk about each other.