Sofia Näsström is right: an exclusive focus on the critical infrastructure of democracy—political parties, and professional news organizations—is far too constricted or, as she also puts it, conservative. An endnote buried in Democracy Rules suggests that plenty of other institutions—trade unions, for instance—also matter a great deal, but I failed to make that point clear in the book’s main part. So, I agree that revitalizing democracy (and remaining faithful to its spirit, as so clearly articulated in Näsström’s book) does not happen in a vacuum.

I would maintain, however, that parties and professional news organizations, although not sufficient for democracy, remain absolutely necessary. What also remains necessary is a widely shared recognition that conflict, rather than always posing a danger to democracy as those singing paeans to “overcoming our divisions” claim, can ultimately create cohesion. But not just any conflict has this effect. Conflict has to remain within two borders— which, I hasten to add, is not to suggest that political theorists should be thought of as a kind of normative border police; there are plenty of other things that theory can and should do, including serving as arguments for transgressions in other contexts: just think of democratic disobedience.

So which borders? First, one must not deny the standing of others engaged in a conflict; if one does, surely one cannot truly recognize them as adversarial partners to the conflict. For those who claim uniquely to speak in the name of a “real people,” for instance, the other side should not really be here in the first place or is, at best, a group of second-rate citizens. Second, some minimal respect for facts is required. Of course, the notion of facts is hardly uncontested. But Hannah Arendt’s basic point about facts and opinions remains crucial. Again, if we have absolutely no shared understanding of facts of the matter, one cannot recognize the other side as any kind of partner: there is just nothing to talk about.

This framing also leads me to reorient debates on democracy away from an image that Näsström employs in her generous comments (but with which my book does not operate): the center versus extremism. The notion of a center is not always meaningless or a cover for opportunism, but the equation of “being democratic” with “being centrist” is a mistake. Centrism has no lasting substantive content, and one surely can be “extremist” in policy without in any sense being anti-democratic. Centrism might either be positional (placing oneself between real or presumed extremes) or procedural (an imperative to work with the other side in systems characterized by intricate checks and balances, and hence in need of some cooperation “across the aisle”). However, such notions lose all plausibility when, let’s say, one party in a two-party system to a significant degree turns against democracy itself; for instance, by not recognizing election losses. In that case, positional or even procedural centrism might become complicit in the destruction of democracy itself (which is why invocations of centrism in the United States today ring hollow). So, although I share the concern articulated with Yeats’s and Didion’s and Näsström’s various riffs on the center maybe not holding, I do not think the notion of the center is helpful for understanding our moment and for acting in it.