Sungmoon Kim’s justification of pragmatic Confucian democracy hinges on its effective response to what he describes as the circumstances of modern politics in East Asia. These circumstances are, notably, the loss of monarchy, the associated impossibility of any appeal to the mandate of heaven, and growing pluralism and so reasonable moral disagreement. The kind of liberal modernity and representative democracy already imported into several Confucian societies require accommodation. Kim is to be applauded for taking contemporary political reality seriously; political theorists still do this too rarely, and can sometimes stumble when they try.Pragmatic Confucian democracy, as its name suggests, synthesizes pragmatism, Confucianism, and democracy. But pragmatism is not assimilated in a full and demanding Deweyan “great community” sense, because power is delegated to winners of elections for broadly Schumpeterian reasons (with a Confucian riff, for elections are necessary to help establish bonds between rulers and ruled). Confucianism is not assimilated in any full sense, because social change has diminished the depth of commitment to Confucian values in East Asia, while political change has made the overarching imposition of those values implausible. And democracy is assimilated not in Barber’s (1984) sense of “strong democracy,” but rather in terms set by the fact that existing democracies in East Asia developed mainly as rejections of authoritarianism, importing liberal electoral institutions. In short, pragmatic Confucian democracy is thinly pragmatic, thinly Confucian, and thinly democratic. These three thins help locate Kim’s theory; they do not imply any deficiencies in that theory.Thin Confucianism means that virtue does still need to be cultivated in both citizens and leaders—Kim’s theory is “inevitably perfectionist.”1 Though whether moral growth is simply a welcome “by-product”2 or central to pragmatic Confucian democracy is not clear. In traditional Confucian thought (and Bell’s 2015 Confucian meritocracy), it is the job of rulers to inculcate moral virtue in the people. Pragmatic Confucian democracy must endogenize both the definition and the cultivation of virtue in the people themselves, while respecting some democratic constraints, as well as some Confucian constraints. If this kind of cultivation remains, I don’t quite see why the book is titled Democracy After Virtue, for virtue remains central—unless “after” means “pursuing” rather than “subsequent to.”Here I want to suggest two things. First, that especially given pragmatic Confucian democracy is in large measure a response to empirical circumstances, it would do well to be informed by some more systematic empirical inquiry. And second, that there are further aspects of contemporary political circumstances in East Asia (and elsewhere) beyond those specified by Kim that merit a response. I will do my best to ascertain the degree to which pragmatic Confucian democracy can respond to these further aspects.How then should a pragmatic Confucian democratic theorist deploy empirical inquiry? Normative theories cannot be tested empirically, but they can (and should) be informed by empirical inquiry in several different ways. Possibilities (following Dryzek 2007) include: No facts, just assumptions. This kind of political theory is evidence-free, and so has universalist pretensions as an ideal. Rawls’s reasoning about the political arrangements a rational individual would support behind a veil of ignorance is a well-known example. The essential cultural dimension of Confucian theory presumably rules out this kind of approach to political theory.Stylized facts. Stylized facts are ubiquitous in political theory. Rawls’s (1993) “fact of reasonable pluralism” is perhaps the most famous. Stylized facts are not necessarily true. For example, “those with more formal education are better able to take advantage of deliberative processes” is widely accepted, but may not be true (and it certainly does not have to be true, given the possibilities of process design).Bad facts. Bad facts are worse than stylized facts. They are not necessarily false, rather it is their erroneous deployment that is so bad. So the findings from monological social science (survey research) often get deployed to criticize the prospects for dialogical politics—and that kind of politics would include pragmatism in general, and pragmatic Confucian democracy in particular (though a thin version of the latter might survive). But as Chambers (2018) points out in criticizing Achen and Bartels (2016), the fact that individuals are poor solitary reasoners has no bearing on their much better capacities to act together as group reasoners. Indeed it is this group capacity that provides a strong epistemic case for the kind of approach to democracy that pragmatism takes.Better facts. Better facts result from the deployment of social science guided by political theory. As I already pointed out, the idea is not to test theories, because normative theories cannot be falsified by empirical test. Instead, the idea is for political theory to guide research questions, such that the results of the empirical inquiry can then inform the normative project. For example, responding to a charge made by Sunstein (2000) that deliberation will normally lead to polarization among its participants, democratic theorists started to think how about how polarization could be countered in institutional design, and what kinds of conditions would ameliorate the problem. Grönlund, Herne, and Setälä (2015) then carried out some empirical comparisons of citizen forums on immigration composed only of people hostile to immigration, which should (according to Sunstein) lead them to polarize further. The empirical finding was that unstructured dialogue did indeed increase polarization, but that dialogue under deliberative conditions (with a facilitator, norms of respect for others present, attentive listening, etc.) actually produced moderation in the group, not polarization, thus providing support for the normative project of deliberative institutional design. In terms of these categories, the circumstances of modern politics in East Asia as set out by Kim consists of three undeniable facts (the passing of monarchy, the implausibility of the mandate of heaven, and the presence of liberal modernity accompanied by representative democracy), and one stylized fact (pluralism). The good news is that there are no bad facts in the book, as no dubious or inappropriate social science is called upon. The bad news is that there are no better facts. For there is no social science at all, either good or bad, that gets deployed in the book—despite the claim on that “pragmatic Confucian democracy aims to combine the insight from the political science of democracy . . . with that from traditional democratic theory.”3 This turns out to be little more than a gesture toward using social science.So what kind of social science inquiry would be helpful to the project? Rather than go through everything in the book that social science might illuminate, let me just suggest two possibilities. I will focus on the (ambiguous) place that the Deweyan dialogical aspect has in the project rather than the electoral aspect that Kim also identifies as integral to pragmatic Confucian democracy. If nothing else, such inquiry might clarify the difference between the prospects for pragmatic Confucian democracy and a stronger Deweyan-Confucian democracy.The endogenization of virtue in the citizenry means that what is required of citizens in Kim’s theory is very different from traditional Confucian perfectionism, much more demanding than Schumpeterian minimalism, but much less demanding than strong Deweyan pragmatism.What are the prospects for citizen participation of this sort? Here some empirical inquiry ought to be able to help. I have already suggested that conventional opinion surveys are not much good when it comes to ascertaining the interactive competence of individuals, and I think they would be too blunt an instrument to illuminate the very subtle differences that Kim sets out between what is required of citizens in different varieties of Confucian democratic theory. We could learn more from what happens when we actually invite citizens to participate in politics. Neblo, Esterling, and Lazer’s work (2018) on deliberative participation is exemplary. They find that given the chance to participate in meaningful deliberation with their members of the U.S. Congress, most Americans wish to do so; and the very individuals most turned off by normal partisan politics are the most interested in deliberative participation. Neblo, Esterling, and Lazer are very precise on the nature of the participatory opportunities they offer people. What would a comparable research design look like for a pragmatic Confucian process? Should we imagine such processes on the small scale? Or is pragmatic Confucian democracy an aspiration only at the scale of the polity as a whole? If so, is it not after all amenable to being informed by experiments of the kind I have described?Second, would pragmatic Confucian democracy successfully both define and then promote modern Confucian-democratic virtues in practice? Local experimentation might provide some useful evidence, especially if it were preceded by the participants asked to think about the terms of their subsequent interaction. Would participants conclude that in their subsequent deliberations they should be expected to show deference to those in the group with greater age, expertise, or education, reflecting lingering Confucian cultural traces? Or would they think that their interactions should be more egalitarian? If both kinds of principles were present, could participants reconcile these two sorts of procedural commitments? We could then take a look at whether and how individuals become transformed in their participation. Do they exhibit moral growth in the terms that (modern) Confucian ethics would presumably applaud? Does participation indeed facilitate moral growth? The ancient Confucian values or virtues can no longer be taken for granted and so provide incomplete guidance, and modern Confucian-democratic virtues are more appropriate in Kim’s account, but presumably democracy means the relevant virtues here themselves require definition and validation in processes of democratic interaction: the endogenization of virtue I spoke of earlier. Appropriate measures here might include commitment to the community (as opposed to self-interest), concern with the wellbeing of the people as a whole, long-term thinking, and a capacity to reflect upon deeper ethical questions, concerning for example equality, merit, and justice.Let me turn now to some circumstances of contemporary politics—in East Asia and elsewhere—that Kim does not address, but might be worthy of some attention.One reason for the recent turn to systems thinking in (deliberative) democratic theory is the question of large scale (Parkinson and Mansbridge 2012). A deliberative system does not require all the deliberative virtues to be manifested at the same time in the same place. So for example the virtue of good justification might be sought in an elected legislature. The virtue of good reflection and independent judgment are much more likely to be found in forums composed of lay citizens with no history of partisanship on an issue (Dryzek 2017; Benson 2018). The virtue of questioning power may be found in a more unruly public sphere, separate from the state.A strongly pragmatic democracy, with a Deweyan conceptualization of a great democratic community, does not easily allow for such differentiation. As Kim points out,4 for Dewey (and Barber) there is properly no separation between state and civil society. And institutions are a decidedly secondary aspect of a strong Deweyan democracy. In pragmatic Confucian democracy, in contrast, there is a clear difference between (elected) leaders and ordinary citizens, so perhaps greater scope for the differentiation of sites, actors, and associated virtues that the idea of a deliberative system requires. At the same time it is not entirely clear how what we expect from leaders when it comes to (modest) perfectionism would be different from what we would expect from citizens—if it is in the interactions of leaders and citizens that (endogenized) virtue is to be generated. Thus it is not obvious that pragmatic Confucian democracy overcomes the problem of imposing the burden of too many demands on a single political site.For better or for worse, there is a something like a deliberative system (though its degree of deliberativeness might be questionable) implicit in Bell’s Confucian democratic meritocracy (2015): key decision making and the operationalization of virtue are in the hands of the most able, while inclusion and so presumably democratic legitimation should be sought in the subordinate democratic parts of the political system. Pragmatic Confucian democracy might involve a complex combination of representative and participatory aspects, but that is not the same as a deliberative system with a clear differentiation of sites, virtues, and tasks.I note in passing that one reason any Confucian deliberative system must differ from a Western liberal deliberative system is in the axiom accepted by theorists of the latter that non-deliberative acts can have deliberative consequences at the level of the system as a whole. The standard example is unruly activism that gets suppressed issues onto the political agenda for subsequent deliberative resolution. Presumably Confucians could not allow such departures from intrinsically virtuous action. The corollary is that all parts of a Confucian deliberative system should themselves be deliberative. This may be an advantage compared to Western liberal deliberative systems, which have come in for substantial criticism for allowing non-deliberative practices to enter deliberative systems, thus undermining the whole idea of deliberative democracy (see, for example, Owen and Smith 2015).Social movements exist in East Asian democracies, no less than elsewhere. Viewed in a differentiated deliberative system perspective, it is easy to see the positive roles social movements can perform in both the operation of democracy and even a deepening of the democratic vitality of the system as a whole. But what is the role for social movements in a pragmatic Confucian democracy? The claims of social movements are often partial rather than comprehensive: that is, they are selective in the causes and values they advance, not all of which refer to the common good. However, one of the causes of such movements may be democracy itself (think of the Umbrella Movement in Hong Kong), and so clearly a conception of the common good. If social movements enter as civil participants in democratic development then presumably pragmatic Confucian democrats would welcome them. In earlier work, Kim (2006) showed that in Korea, protests can be seen as positive aspects of ethical civil society. There, Kim linked civic protests to the intersubjective, quasi-familial concept of jeong, which entails taking on collective responsibility for individual harms that could have been prevented if the community had been stronger (one of his examples concerns protests in the wake of two teenage girls killed by a U.S. military vehicle). But what if protests are more unruly, and see themselves as challenging the political economy from the outside—for example, Occupy, which has its own radical conception of the common good?East Asian societies no less than others have recently featured a vast increase in the volume of, and possibilities for, political expression accompanying the rise of social media (Keane 2013). (The degree to which individuals in China dare take advantage of the opportunity is perhaps a different question.) In many parts of the world, this means an overload of expression: everyone is talking but nobody is listening, and especially not listening to anyone who does not share their views. This sort of social media environment does not seem to be especially hospitable to the advance of Confucian-democratic values, and may indeed undermine those values. What is the response of pragmatic Confucian democracy? What place (if any) should social media have in a pragmatic Confucian democracy? Do Confucian-democratic virtues themselves need to be rethought for an online age? Should listening be highlighted as a virtue? Is it possible to think about interventions in, even reconfiguration of, the social media landscape and its algorithms guided by pragmatic Confucian democracy? Presumably any such interventions should themselves be carried out through inclusive democratic means.The global epidemic of authoritarian populism (often associated with right-wing authoritarianism) has not yet reached East Asia. But it is important to be vigilant. While it is hard to generalize about the causes of the epidemic, one factor is resentment against modernizing and globalizing elites on the part of those who do not share in the fruits of economic growth, or whose material position may indeed have been worsened by marketization and globalization. Another is resentment of actual or potential immigrants. To the degree it prizes individual integration into the political community, it seems to me that pragmatic Confucian democracy could help constitute a good defense against populism—or at least a better defense than minimalist electoral democracies, which provide few opportunities for meaningful participation in politics, but plenty of opportunities for rage to be expressed in the vote.East Asia, like the rest of the world, is vulnerable to catastrophic climate change (Vietnam is usually ranked among the most vulnerable countries in the world). The government of China at least has recognized this, and the commitment of its leadership to building an “ecological civilization” is truly remarkable—at least at the level of rhetoric. Whether performance on the ground matches up is a different question. There remains considerable debate among environmental governance scholars as to whether democratic states are up to the task of effective response to climate change, and adversarial democracies such as Australia and the United States show just how badly it is possible to perform when the issue is processed through the lens of short-term partisan advantage. On the other hand, no authoritarian state has ever turned in an adequate environmental performance, and there may be limits to which complex social-ecological systems are amenable to centralized authoritarian steering. Pragmatic Confucian democracy could make a pretty good case for offering an approach that could overcome both the pathologies of adversarial democratic states and the epistemic limitations of authoritarian systems. This could be done through its cultivation of common-good and long-term virtues. At any rate, I think this is an argument worth exploring in the political theory of ecological pragmatic Confucian democracy.Kim stresses pluralism and associated moral disagreement as key aspects of the circumstances of modern politics in East Asia. But disagreement does not necessarily stop at moral questions. Much more serious are deep differences of identity that underlie many ethnic, racial, religious, and national conflicts in today’s world. Deep difference means that one identity can be validated only by the rejection of the identity of the other side—especially when it comes to a share in sovereign authority. Such differences are not of course uniquely modern, and extend back over the centuries in different parts of the world. They are not unknown in contemporary East Asia (for example, Uighurs and Tibet in China).Such conflicts can provide some of the most intractable challenges for any political system. The literature on politics in deeply divided societies contains a number of possible remedies. The most popular is perhaps consociational democracy, which involves the leadership of different sides constructing a power-sharing arrangement—while limiting interaction between the different sides below the elite level. This limitation means it is hard to see how consociationalism could coexist with any pragmatist democracy, Confucian or otherwise, because it would mean dividing the democratic conversation into non-interacting segments defined by ethnicity, religion, or whatever. (On the other hand, it is easy to see how meritocracy could combine with the elitism of consociationalism.) Deliberative approaches to deep division would in contrast seek dialogue transcending the divide at all levels. But the terms of that dialogue have to receive careful attention. For example, it may be preferable to concentrate the dialogue at least to begin with on shared practical problems (such as the environment, policing, or education) rather than big questions of who should exercise sovereign power, or share in sovereignty. This dialogue may profitably begin at some remove from peak state institutions (Dryzek 2005). At first sight, this might seem closer than consociationalism to the disposition of pragmatic Confucian democracy. However, pragmatic Confucian democracy may I think find such an approach challenging, given its presupposition is the existence of a community whose existence or even construction in the light of deep division cannot be assumed. The idea of a single community may indeed have to give way to lesser ambition about how different sides can learn to live together in relative harmony.Virtuous individuals can jointly produce bad outcomes if they must act and interact within bad social, economic, and political structures. I live in Canberra, home to an environmentally conscious and politically progressive population. Structurally, Canberra is designed for cars. So even people like me who hate the idea of driving end up driving a lot—with associated environmentally unsound collective outcomes. The general point is that ethical perfectionism has its limits. For example, if virtuous individuals are located within a capitalist political economy, that will force them to do bad things.Democracy After Virtue opens up a rich agenda linking questions of normative political theory, practical application, and empirical study. The normative core of the project is well established in Democracy After Virtue. Less well developed is what exactly the project would mean in practice, how exactly it would negotiate the multidimensional contemporary political challenges in real contexts, and to what effect.