John Thrasher and I have published an article in the American Journal of Political Science, “Political Stability in the Open Society,” that BHL readers may find of interest. If you’re interested in how to have a diverse and free but stable social order, take a look. I’m cross-posting the blog post linked here.

In “Political Stability in the Open Society,” we argue that John Rawls’s model of a well-ordered society—as an account of a realistic utopia—is defective for two reasons. First, the well-ordered society model wrongly excludes the deep disagreement and diversity that we find in contemporary political life from figuring into a model of liberal order. Second, when deep disagreement and diversity are integrated into the model, discovery becomes an important part of modeling a stable liberal order. A liberal society is one where people are free to experiment with different approaches to the good life and justice given that we know much less than we might about how to live together.

If we are committed to recognizing deep diversity and the need for social discovery in modeling a stable liberal order, we must modify the idea of a well-ordered society and the ideas most closely associated with it in a liberal theory of justice. In particular, a more dynamic notion of stability for the right reasons is required for a new model that we call an open society. An open society is a liberal society that allows for deep disagreement about the good and justice and which sustains institutions that can adapt to new discoveries about what justice requires.

Our goal is to explain the idea of stability appropriate for an open society. The challenge is that, given the importance of respecting diversity and openness to social change, stability for the right reasons now seems to have a cost; stable rules are hard to replace with better rules. On the other hand, some rules need to remain stable to support productive social change and experimentation.

Given these challenges, we distinguish two different kinds of stability that apply at different levels of social organization. The first kind of stability applies to constitutional rules that set out the general legal rules within which our lower-level institutional rules operate. These constitutional rules must remain in equilibrium despite challenges and threats in order to preserve the social conditions that foster experimentation. But we reject a similar form of stability for lower-level legal and institutional rules. Experimentation at that level can be productive in ways that constitutional experimentation is not. Instead, lower level legal and institutional rules need to be robust in the sense that, when challenged, old rules can be replaced by stable new rules without undermining the system of rules as a whole.

One important implication of our analysis is that, in the open society, a shared conception of justice is less important than a stable constitutional framework where many aspects of the open society, including justice, are open for debate. Rather than focusing on the particular principles of justice that are most reasonable for a well-ordered society, theorists should focus on the properties of different constitutional orders that encourage productive social evolution and experimentation. A second implication of our analysis is that open societies may turn out to be substantially different from one another. There will likely be no single type of social order that suits any given open society. This is all to the good because these diverse orders can learn from each other’s experiments.