Empirical political scientists (especially Americanists) differ sharply with political theorists regarding how they measure democracy’s quality or proper functioning (Sabl Reference Bawn, Cohen, Karol, Masket, Noel and Zaller2015). This book exemplifies both the size of this gap and some of its costs. Christopher Achen and Larry Bartels aim to refute a prevalent model of democracy under which “what the majority wants becomes government policy” (p. 1). They call this theory “the familiar textbook portrait of democracy” (viii), “a kind of ‘folk theory’” (1; compare 9, 21, 67 and passim), “current thinking about democracy” (327), or simply “democratic theory” (whether “contemporary” [2], “conventional” [14, 306, 310], “standard” [20], “romanticized” [298], or unmodified [8, 37, 114, 316, 321]). However, real democratic theorists, i.e., political theorists who write on democracy—the authors oddly call them “academic theorists” and portray them as little concerned with mass publics and electoral institutions (2, n.2; 301; 306)—have rarely imagined that democracy normally translates public preferences into policy, or judged that democracy’s health depends on whether it does (though I have argued for their taking the responsiveness criterion more seriously than they do).
On the other hand, actual democratic theorists have long regarded as fairly obvious some of the theses that Achen and Bartels regard as new (to their fellows; as the authors rightly note, they date back at least to Publius). That politics is driven largely by “group and partisan loyalties,” not well-informed policy preferences translated into votes (18; cf. 236, 264, 313, and passim); that politicians’ claims to embody the people’s desires are often fatuous or dangerous (327); and that good government requires not simply responsiveness to public wishes but also “political experience and judgment” (86) will strike many political theorists, and in fact many scholars in diverse disciplines, as constituting a brave new world of “realist” insights that is new to academic empiricists and them alone.
Consider the “retrospective” model of democratic control, whose extensive rebuttal (persuasive, in my non-expert view) lies at the book’s heart. Under this model, voters control politicians by throwing them out if a few blunt social indicators (GDP or income growth, war casualties, crime) fail to please. Achen and Bartels argue that this fails: vote decisions only track these indicators “myopically,” for the period right before the election (enabling easy gaming), and the aggregate vote is slightly, but often decisively, influenced by events outside politicians’ control (weather, shark attacks).
But who exactly will mourn the retrospective model’s passing? As the authors admit, its partisans include neither “political philosophers” nor “ordinary citizens” but mostly Americanists who study public opinion and elections (91; compare 306). Here as in other respects, the “fairy tales” or “romantic theories” of democratic control at which the authors take aim (pp. 7, 20) seem to be tales and romances to which only academic empiricists ever thrilled.
If academic empiricists have sought democratic control where they are unlikely to find it, they also tend to neglect it where it might be present. Consider shark attacks (and other nonpolitical events) more closely. This book never claims that such events cause huge swings in the vote, only that small ones can quash hopes for reliable responsiveness given that most elections are close. But focusing on larger structures rather than individual elections might prompt the question of why they are close.
The answer, I submit, is that both parties are constrained to play the game within quite narrow bounds defined by policy. Casting a Jaws vote seems harmless because, and only because, people know that the government functions which are most fundamental, and to which they have long become accustomed—e.g., social insurance against unemployment, disability and old age; publicly-funded schools and universities; basic civil-rights legislation; a large military and the de facto welfare state that goes with it; roads and rails; clean air and water regulation—are so popular that no set of politicians they might select is likely to fundamentally change them. (Achen and Bartels’ examples of what seem arbitrary voter decisions over policy all concern new policy proposals whose import was unpredictable and abstract. Moreover, they all took place many decades ago, when government did much less and most of its workings little implicated daily habits and life plans.) The authors’ much-cited conference paper, which served as a précis for this book, contained a crucial proviso to this effect, as does recent work stressing the “electoral blind spot” with regard to lower-profile issues (Achen and Bartels Reference Achen and Bartels2004, 38–9; Bawn et al. Reference Bawn, Cohen, Karol, Masket, Noel and Zaller2012. 577). The current book, for some reason, lacks such a proviso (though 312n6, mentioning the case in which “alternatives to the incumbent government are … uncertain or not credible,” comes closest).
The effect is to exaggerate the arbitrariness and variability of political decisions. By stressing instances in which democratic control is undermined when voting decisions are affected by random events, Achen and Bartels obscure the pervasive degree to which control might be enabled by non-decisions that are immune to such events (but tied to very clear preferences—that large swathes of policy change fairly little).
This book’s contribution is still great. Among academic empiricists in the American subfield, this book will be, and deserves to be, a bombshell. It fearlessly questions the deep, apostolic (and, arguably, subfield-defining; Seidelman Reference Seidelman and Harpham2015) assumption that democracy “must” mean a system for converting public preferences into policy. Its arguments are thorough and relentless; its use of evidence, careful and exhaustive; its rhetoric, appropriately bracing. The authors’ stature supplies their argument with both weight and salutary shock value.
Democratic theorists will find in this book few normative or conceptual innovations, but mostly reminders that we must grapple more fully with findings that empiricists have long taken for granted (e.g. the degree to which all of us derive our beliefs from our partisan allegiance). But if Democracy for Realists does not represent the original contribution to “democratic theory” that its authors imagine, it does constitute a powerful license to reject parochial superstitions: to dispense with durable professional myths in favor of old and precious truths, both about what drives political change and about what might productively channel and contain it.
