Deliberative democracy is unrealistic, but so are rational-choice models of democracy. The elements of reality that rationalistic theories of democracy leave out are the very elements that deliberative democrats would need to subtract if their theory were to be applied to reality. The key problem is not, however, the altruistic orientation that deliberative democrats require; opinion researchers know that voters are already sociotropic, not self-interested. Rather, as Schumpeter saw, the problems lie in understanding politics, government, and economics under modern—and postmodern—conditions. (...) Modernity entails levels of complexity that stymie voters' attempts to be well informed about which political promises will actually solve social problems. Postmodernity adds to complexity an overabundance of fragmented information experienced impatiently and fleetingly by citizens who confuse entertainment with knowledge, and whose political demands are as unrealistic as their personal desire for a unique identity. These factors can be added, one by one, to the minimalist rational-choice model of politics, with each step leading us farther away from deliberative democracy. (shrink)

Advocates of ?deliberative democracy? want citizens to actively participate in serious dialogue over political issues, not merely go to the polls every few years. Unfortunately, these ideals don't take into account widespread political ignorance and irrationality. Most voters neither attain the level of knowledge needed to make deliberative democracy work, nor do they rationally evaluate the political information they do possess. The vast size and complexity of modern government make it unlikely that most citizens can ever reach the levels of (...) knowledge and rationality required by deliberative democracy, even if they were better informed than they are at present. (shrink)

Abstract Scholars find that political elites are badly polarized over a large range of policy issues, but they tend to agree that the mass public is much more ambivalent. The abortion war in particular is regarded as one in which millions of ambivalent citizens are caught in the crossfire of polarized activists. Yet even abortion activists struggle to escape the very ambivalent sentiments that plague ordinary Americans. These common sentiments even exert a moderating influence on both movements in ways that (...) are consistent with the preferences of the American public. They also suggest that liberalism may be mired in permanent conflict and ambivalence over the scope of basic human rights. (shrink)

Over recent years support for deliberative democracy as a regulative ideal against which political and economic institutions should be judged has become the dominant tradition within political theory. Deliberative democrats such as Amy Guttman and Dennis Thompson argue that deliberative public decision making would bring with it important epistemological and ethical gains. Closer inspection of these claims, however, suggests that deliberative democratic arrangements are not only impractical but are fundamentally at odds with the epistemic and ethical goals that their supporters (...) seek to advance. Arrangements based on ?exit? rather than ?voice? may be better placed to meet these challenges in a world characterized by complexity and conflicting moral values. (shrink)

In his remarkably meticulous and even-handed 2005 book, Expert Political Judgment, Philip E. Tetlock establishes that the only thing we can count on in the political experts' predictions is that they will underperform-in some cases significantly-the predictions made by mechanical statistical procedures, including random chance. Experts have many uses and Tetlock does not claim that they have no value. However, Tetlock zeroes in on experts' important political role-as prognosticators. Tetlock does not attempt the impossible by trying to judge experts on (...) their ability to prognosticate the effects of public policies; this would require that Tetlock himself could know what those effects were, and even in retrospect, it isn't that simple. Experts' failure to predict much simpler matters, however, such as GDP growth within a broad range, strongly suggests that technocracy is worse than putting dart-throwing chimps in charge of public policy. (shrink)

Do people tend to disagree over political issues because of conflicting values? Or do they disagree about which policies will most effectively promote shared values? In a previous article, I argued that the issues most people think are most important tend to fall into the latter category. On the issues of greatest importance to the mass public, most citizens agree about the ends that are desirable, but disagree about which policy means would best effectuate those ends. Consequently, disputes about facts—disputes (...) about the actual effects of proposed public policies—lie at the heart of the most important divisions in contemporary American public opinion. However, people do not necessarily interpret their political disagreements this way. If they fail to recognize that facts that they see as self-evident are disputed by their opponents, they may see their opponents as having different values, since there could be few other explanations for their opponents' disagreement with them. However, evidence that disagreements about facts are really driving public opinion can be found by using conditional surveys, which ask respondents if they would support a given policy if they believed that it would cause specific negative consequences. (shrink)

Any workable ideal of deliberative democracy that includes elections will need modern democracy's ever-present ally, parties. Since the primary function of parties is to win office rather than to reflect on public questions, parties are potential problems for the deliberative enterprise. They are more at home in aggregative models of democracy than in deliberative models. While deliberative democracy will need its moments of aggregation—and therefore, must have parties—partisans as they actually arise in the political world possess traits that undermine the (...) deliberative ideal. If partisans of partisanship are to be unembarrassed by (or are to correct) these defects, even workable ideals of deliberative democracy need to stand at some distance from the partisan imperative. (shrink)

In the ongoing debate about the impact that studies of public ignorance should have on the study of culture, Mark Fenster and Bret Chandler assume that wider political participation must be our goal, because, to them, political ignorance is a culturally imposed, and therefore removable, obstacle—as if, without the baleful influence of culture, political participants would be well informed. Culture is indeed a primary influence on people's political opinions, so political scientists should indeed study the role it plays in the (...) formation of political beliefs. But such studies should begin with ignorance as people's default position, and should refrain from assuming that the source of ignorance is cultural domination or manipulation. (shrink)

A conditional survey establishes a preliminary case for believing that policy differences are to some extent driven by fundamental beliefs about empirical aspects of society and economics. The survey shows willingness in about a third of all respondents to shift their expressed policy preferences when asked a hypothetical question positing negative consequences of their initial preferences. This suggests that assumptions about the consequences of public policies may play as important a role in policy preferences, or a more important role, than (...) do values, personality traits, or motivation. However, roughly a third of those with the strongest initial attitudes expressed even stronger attitudes—but in the same direction—when confronted with counterattitudinal hypotheticals. This may reflect attitude polarization of the sort encountered by previous researchers, such as Lord et al. (1979) and Taber and Lodge (2006). Or it may indicate a tacit objection to the very idea of basing policy preferences on policy consequences. (shrink)

American democracy is not unconstrained or autonomous, but instead achieves what could be termed value representation. Rather than affording representation on policy issues, elections transmit priorities among competing normative ends, while elite politics address the more complex matching of ends and means within the value boundaries established by voters. This results in neither policy representation nor state autonomy, but instead in a specific and limited form of democratic representation.

On August 30, 2013, the American Political Science Association sponsored a roundtable on political epistemology as part of its annual meetings. Co-chairing the roundtable were Jeffrey Friedman, Department of Government, University of Texas at Austin; and Hélène Landemore, Department of Political Science, Yale University. The other participants were Scott Althaus, Department of Political Science, University of Illinois at Urbana-Champaign; Mark Bevir, Department of Political Science, University of California at Berkeley; Rogers Smith, Department of Political Science, University of Pennsylvania; and Susan (...) Stokes, Department of Political Science, Yale University. We thank the participants for permission to republish their remarks, which they subsequently edited for clarity. (shrink)