Liberalism in IR is not the same as "liberal" in everyday party politics. It is a family of theories that share a hopeful but not naive claim: the way states are governed inside, the economic ties among them and the institutions they build can all make cooperation more likely and war less likely. Its best defenders do not deny anarchy; they deny that anarchy settles everything.

Kant and the idea of a peaceful federation

The tradition's founding text is Immanuel Kant's essay Perpetual Peace (1795), which set out three "definitive articles". The first said that every state's constitution should be republican, meaning a government of law with representation and a separation of powers. Kant reasoned that citizens who would bear the costs of war themselves would hesitate to begin one, while a monarch could declare war as casually as he might order a hunting party. The second held, in M. Campbell Smith's 1903 translation, that "the law of nations shall be founded on a federation of free states." The third was a limited cosmopolitan right of hospitality: strangers who arrive peacefully should not be treated as enemies. Kant's three pillars were republican government, a federation of free states, and a right of hospitality to strangers.

The democratic peace

In 1983 the political scientist Michael Doyle revived Kant's argument with evidence. The democratic peace is the finding that established democracies have rarely, if ever, fought wars against one another, even though they fight non-democracies about as often as other states do. Scholars offer two main explanations. The institutional explanation points to voters, a free press and legislatures that slow decisions and make bluffing hard. The normative explanation holds that democracies expect each other to settle disputes by compromise, as they do at home. Some scholars have called the democratic peace the closest thing to an empirical law in IR.

Institutions and cooperation

A second strand, often called neoliberal institutionalism, accepts much of the realist starting point and still reaches more hopeful conclusions. Robert Keohane's After Hegemony (1984) argued that international institutions help states cooperate by sharing information, lowering the costs of making deals and making cheating easier to detect.

The political scientist Robert Axelrod showed why repetition matters. In a one-time prisoner's dilemma, each player gains by betraying the other, so both betray and both lose. Axelrod held computer tournaments in which strategies played the game against each other again and again, published in The Evolution of Cooperation (1984). A simple strategy won. Tit for tat cooperates on the first move and then copies whatever the other player did on the previous move. It is nice, since it never betrays first; retaliatory, since it punishes betrayal at once; and forgiving, since it returns to cooperation as soon as the other side does. Institutions that bring states back to the table again and again lengthen what Axelrod called the shadow of the future, giving everyone a stake in keeping a good reputation.

A third strand, commercial liberalism, stresses the economy. Commercial liberals argue that states that grow rich by trade have less reason to grow rich by conquest. Level 8 returns to this idea and its limits.

Criticisms

Realists answer that institutions reflect the power of those who create them and are ignored when great interests are at stake. Critics of the democratic peace raise problems of definition and timing. One objection is that for most of history there were too few democracies for their peace with each other to prove much. Another is that the result depends on how "democracy" and "war" are defined: was Britain in 1812, when only a small minority of men could vote, a democracy when it fought the United States? Others suggest that the peace among Western democracies after 1945 owed more to a shared Soviet enemy and American protection than to democracy itself.

There is also an uncomfortable record. Democracies have helped overthrow elected governments abroad through covert operations, especially during the Cold War. Defenders reply that the core finding has survived decades of testing with many definitions, and that such operations were kept secret precisely because democratic publics would not have accepted open war against fellow democracies.

Worked example: Fashoda, 1898

In 1898 a small French expedition under Captain Jean-Baptiste Marchand reached Fashoda on the upper Nile, hoping to secure a French belt of territory across Africa. A much larger British and Egyptian force under Herbert Kitchener arrived soon after. For several weeks Britain and France, both parliamentary states with a free press, seemed close to war. In November France ordered Marchand to withdraw.

A liberal reads the crisis as the democratic peace at work: public debate, parliamentary politics and shared habits of compromise gave both sides time and ways to step back, and within six years the two countries signed the Entente Cordiale of 1904. A realist reads it differently. In a 1994 article the political scientist Christopher Layne argued that France backed down because the British navy was far stronger, and that French public opinion was furious rather than conciliatory. The case shows how one event can be claimed by rival theories, and why scholars now test the democratic peace on hundreds of cases rather than one. It is also worth noticing that the dispute was over African land whose inhabitants were never consulted.