The Evaluation of Democratic (and Other) Institutions: Some Methodological Challenges

Democratic theory has devoted extensive attention to arguing for democracy as a requirement of justice or legitimacy. Yet surprisingly little work has been done concerning the task of evaluating institutions. Therefore, political theorists largely lack an adequate methodological toolset when asked to advise on questions of institutional choice and design. What is needed are value theories of institutions. I identify three methodological challenges that any such theory must address. First, **structural completeness**: A value theory of institutions requires three distinct levels—(1) an evaluation principle specifying which features make institutions valuable, (2) an analysis determining which features institutions possess and which causal processes can change them, and (3) a resulting evaluation. We distinguish between principle-based theories (centred on where value lies) and thesis-based theories (centred on how value is achieved), showing how this typology clarifies persistent confusions in democratic theory. Second, **multiple pathways to value**: Democratic theorists typically assume that the value of democratic institutions lies either in the procedures or in their direct effects, namely their outcomes. But beyond this dichotomy, value can also arise through indirect effects. In particular, democratic institutions also have an effect on attitudes, emotions and conceptual space, with effects on value downstream. For instance, well-working democratic institutions are not only intrinsically fair, or make just decisions, they also promote value by making citizens form correct beliefs about the world, instil egalitarian attitudes, increase trust, help fostering progressive social norms, and so on. This leads to further valuable effects. Third, **context-dependence**: The value of institutions depends on circumstances—both the mental profile of citizens and external conditions. We distinguish two approaches to context-dependence: (a) incorporating context into the evaluation principle itself, and (b) developing context-specific theories. Either approach challenges the universalist aspirations of much democratic theory, while offering resources for understanding why democracy may be valuable in some contexts but not others. To illustrate these challenges, I introduce **Generativism**, the claim that institutions achieve significant value through changing citizens' mental features (attitudes, concepts, emotions), which then produce downstream effects. While standard approaches treat institutions as "external tools" operating on fixed human psychology, systematically neglecting generative effects and thus misrepresenting institutional value, a generative approach shows how democracy's value lies partly in fostering mental changes, such as knowledge, civic attitudes, or trust. These effects on mental features can cause value, for instance, by helping promote human flourishing and justice. In line with Dewey's pragmatic justification of democracy, we find that we need new methods for democratic theory. But we also note that a well-working democracy is itself a method towards social progress. This paper draws on collaborative work with Franz Dietrich for a book manuscript, "Generative Democracy: New Foundations for Democratic Theory."