Backsliding vs. Building: How Changes in Regime Type Affect Attitudes Toward War

• In Event: People Get Ready: Public Opinion and Conflict

Abstract

Countries that become democratic do not necessarily stay democratic. Democratic backsliding can be blatant—there is no mistaking a coup d’état or flagrant election rigging—but its contemporary forms are often insidious and incremental, using the very institutions of democracy to hollow it out from the inside. Will the current wave of backsliding make international politics more conflictual by undermining the democratic peace? Historically, democracies have been more peaceful in their relations with one another than with autocracies, and recent work suggests that public opinion plays an important role in this phenomenon. Democratic publics are substantially less likely to support attacking another democracy, giving leaders an added incentive to resolve those conflicts peacefully, rather than risk the political costs of an unpopular war. Threat and morality appear to be the most important mechanisms driving this democratic peace effect in public opinion. All else equal, democratic citizens perceive other democracies as less threatening than autocracies, and they are less likely to see the use of force against a democracy as morally justified.

However, the existing literature does not speak directly to the question of whether and how incremental changes in a country’s regime type influence support for the use of force. Studies that show a democratic peace effect in public opinion typically rely on experimental designs in which the adversary is either a full democracy or a full autocracy. This approach shows the cross-sectional effect of regime type, and at first glance it may seem to imply that backsliding—moving away from democracy toward autocracy—should increase support for the use of force, while democratic building—moving away from autocracy toward democracy—should have the opposite effect of decreasing support for the use of force. However, there is good reason to be cautious about extrapolating conclusions about incremental change from cross-sectional effects. The evaluation of changes in behavior is governed by its own psychology and set of biases.

Drawing on literature in social psychology, we argue that as the existing cross-sectional results suggest, democratic backsliding will increase support for the use of force and democratic building will decrease support for the use of force, but that the magnitude of these effects will differ. People tend to treat negative behavior as more diagnostic of threat than positive behavior. Thus, we expect that backsliding will have a greater effect than building, and that it will consequently be easier for countries to lose the benefit of the democratic peace effect by backsliding than it will be for them to regain it through building. In addition, we expect the effects of backsliding to be conditional on ideology. Specifically, backsliding should have a greater impact on liberals compared to conservatives due to their different moral foundations. The impact of negative behavior is magnified when it is moralized and democratic backsliding, which often takes the form of restrictions on political inclusivity, sits squarely in the liberal moral domain, which is primarily concerned with reducing harm and ensuring fairness. In contrast, conservatives tend to moralize harm and fairness less than liberals do, while placing greater moral weight on concerns about loyalty, authority, than purity. Consequently, while conservatives may see backsliding as a negative, they are less likely than liberals to view it as a moral issue.

We test these expectations using a novel online survey experiment conducted in the United States. The results will have potentially far-reaching implications for how we think about the role of morality in public opinion and international relations more broadly, as well as for larger debates about democracy and war and the possibility of great power conflict in the twenty-first century.

Authors

• Rachel M. Stein, George Washington University

• Miles Evers, George Washington University