Cncerns about democratic decline, backsliding, and creeping authoritarianism in the United States have led to a good deal of ink being spilled in public discourse. Although such concerns have long been present, they were accelerated and amplified by the presidency of Donald Trump. Nicholas T. Davis, Keith Gåddie, and Kirby Goidel’s new book argues that, to understand whether the American public is backsliding, we must start by asking how the public conceives of democracy in the first place. In an impressive combination, the authors apply a synthesis of existing democratic theory to new survey research to gauge public opinion on the meaning of democracy. They show that Americans express agreement about procedural definitions of democracy but express divergent opinion about egalitarian definitions of democracy.

This smart book not only makes its reader think about democratic backsliding among the public but also offers a classification approach that might be applied to the conceptual definition of backsliding and democracy more generally. The authors use latent class analysis to group respondents to two opinion surveys into four classes, each with a different conception of democracy. A proceduralist group conceives of democracy as a system in which a familiar set of rules and norms are effectively followed, such as competitive elections, freedom of association and speech, peaceful transition of power, and minority rights. The authors find proceduralists compose around 20% of the public. A maximalist or social group (about 40% of the public) adds more expectations to the endorsement of proceduralist rules and norms. Maximalists believe a democratic state should act to enforce some level of social and economic equality. Importantly, for this group, states that do not attain sufficient levels of social and economic equality do not meet a full definition of democracy. The authors call members of a third group moderates (about 30%), whose views on social and economic equality fall somewhere between the maximalist and the proceduralist groups. A residual group (indifferent, about 10%) does not appear to have coherent views about the meaning of democracy.

The implications of the theoretical approach and findings are important. Do Americans conceive of democracy as a set of civic procedures and rights, or are specific outcomes necessary to achieve “democracy”? The authors write, “It is possible that an individual’s specific conception of democracy constitutes more than naïve or abstract support for a set of process-based institutional rules, but, instead, an evaluated framework of social, political, and economic preferences” (p. 5). Their data and analysis suggest some do and some do not hold this conception.

What does the evidence say, then, about the prevalence of democratic decline in American public opinion? As the authors note, of course, it would be preferable to have a long time-series analysis of these questions to see whether and how opinions changed over time. But even absent the time-series, there is some good news. There is nearly universal recognition that textbook democratic norms—equal access to the franchise, equal treatment before the law, free and competitive elections, free speech, and freedom of religion (Figure 3.2, p. 47)—are essential to definitions of democracy. There is strong support for free speech even if the speech is offensive. Endorsement of procedural criteria holds across the three main types in the authors’ taxonomy (Figure 4.2, p. 65), which suggests public commitment to these democratic ideals.

Concern arises on two fronts, however. First, individuals in the social group define democracy more broadly than do proceduralists. They endorse the provision of basic necessities, economic prosperity, and, to a lesser degree, combating income inequality as essential traits of a democratic system. To the extent that members of this group evaluate a state as failing to sufficiently deliver these outcomes, such individuals may sour on “democracy” more broadly. Indeed, one of the authors’ conclusions is, “The current crisis, then, is rooted not in declining commitments to democracy as an ideal but in the realities of democratic governance” (p. 167). When some voters hold a definition of democracy that includes features subject to political contestation rather than near universally accepted norms, their evaluation of democracy in any setting is necessarily more precarious than the evaluation of a proceduralist holding more minimal criteria.

A second front of concern is that the opinion survey asks respondents to rate each characteristic of democracy on its own, rather than in relationship with other goals for the respondent. Voters might say that free and fair elections or freedom of association is essential to democracy in the abstract, but when application of those values leads to political results contrary to other values they hold, they must compromise on one of the two. Connecting to my work under review in this Critical Dialogue, if voters care more intensely about the political result than about the democratic norm, their action might follow politics rather than norms.

Indeed, one might even define democratic backsliding as a change in relative intensity for political outcomes versus democratic norms. Americans might continue to endorse free speech and fair elections as before, but if their perception is that the policy consequences of elections are of greater salience—as might be the case with increased polarization between the party coalitions— voter willingness to swallow political defeat in deference to democracy might decline. Despite the rhetoric of “Stop the Steal” around election fraud, my suspicion is that many who entered the Capitol on January 6, 2021 did so more to prevent what they saw as an unacceptable Biden presidency than to prevent certification of a stolen election.

More broadly, if readers adopted the perspective on public opinion presented in John Zaller’s 1992 book, The Nature and Origins of Mass Opinion, they might be concerned about the empirical enterprise of Democracy’s Meanings. Many Americans have not thought carefully about what features they deem essential to a democratic system. Their opinion survey responses might simply reflect “what they’ve heard” from the elite political rhetoric in their information milieu. Proceduralists might give responses reflecting the rhetoric they hear about the rule of law and fair elections, and indeed the authors find proceduralists more likely identify as conservative and Republican. Maximalists might give responses reflecting the rhetoric they hear about inequality and disenfranchisement, and indeed the authors find that maximalists more likely identify as liberal and Democratic.

Under this Zaller-type story, the relevant influence on meanings of democracy would be elite rhetoric, rather than individual opinion. Democratic backsliding would follow, then, from a change in the elite rhetoric surrounding norms of democracy. My sense is that there is ample evidence of this phenomenon taking place. Research documenting and quantifying this trend so we can better understand its causes and evaluate its effect on individual citizens strikes me as a natural and important part of the project started in this book.

If elite rhetoric drives public opinion on the meaning of democracy, it does imply a potential problem of accountability, as the authors suggest. Political elites who defy norms of democracy might use rhetoric to influence the public’s definition, upend the evaluative criteria that might have been held against them, and proceed with their action without risk of voter retribution. It is crucial to understand whether voters hold ethical standards for democratic conduct external to elite rhetoric.

The authors, on my read, accept the premise that the United States is experiencing democratic backsliding and argue that public opinion is part of the story: “We are struck by the democratic deficit that faces the United States. Americans are socially divided, and yet, they share a set of expectations for good governance that are woefully unfulfilled” (p. xiii). I am not certain why they make this claim. Although it is true that the authors classify 40% of American opinion as maximalist, 50% is either proceduralist or moderate. Therefore, we should not expect that the maximalist position should gain full representation in public policy. We might instead expect some kind of weighted average, which I would suggest is roughly what we have. The American state enacts massive redistribution that counteracts some, though not all, of the recent increase in income inequality. The Congressional Budget Office, for example, estimates that federal taxes and means-tested transfers increase income for households in the lowest quintile by 64% and decrease incomes in the highest quintile by 24% (“The Distribution of Household Income, 2019,” Washington, DC, Exhibit S-1). This is not to say that the extent of federal efforts toward economic equality matches the perceptions of many Americans about what the American democracy should be doing, only to push back on claims that the effort is demonstrably inadequate.

Davis, Gåddie, and Goidel’s book pushes scholarly inquiry of democratic decline into the public mind and highlights that how individuals (scholars not excepted) define democracy directly influences any evaluation of its functioning, vibrancy, and backsliding.