On the Side of Angels provides a defense of political parties. As Nancy Rosenblum notes, political philosophy has been close to utterly silent about any positive case (indeed, any case—as she notes, parties are an “orphan” in philosophy) for political parties, an odd lacuna given their centrality in actual politics and in the empirical (and theoretical) literature in political science.

The book is structured so that five of nine chapters are about antipartyism, three present her positive case, and one (the ninth) investigates when—even though political parties are understood to be good things—particular parties can justifiably be opposed or banned by democrats. While Rosenblum is critiquing antipartyism, there is something wearying, nearly convincing, about considering so many antiparty claims. Of course, as Austin Ranney showed, this ambivalence (OK, mostly negative views) about parties has been a consistent theme among American politicians and public since the Founding.Footnote ¹ For Ranney, the negative part of the ambivalence is directed mostly at the party leadership and organization. That is, he focused on the kinds of concerns voiced by those like Thomas Jefferson (would subscribing to a party mean giving up one's intellectual independence?), Mark Twain (skewering the corruption of machines and the ties between concentrated economic wealth and political officeholders), and Will Rogers (whose “I belong to no organized party. I am a Democrat” reflects the complex compromises necessary for northern and southern Democrats to form a majority to govern in the New Deal and beyond). Rosenblum's analysis of these and related problems articulated by antipartyism adds much greater rigor, clarity, and depth to Ranney's, and that alone would be enough to make the book worthwhile. But even more, she creates a defense of partisan identification that is, at least to this reviewer, totally original. Her primary contributions in this book are this defense, along with her argument of the reciprocal nature of the party and of partisan virtues.

While her philosophic reach is broad, Rosenblum's understandings of the empirical and (scientific) theoretical literature on the political party are richly informed by the scholarship on American political parties, with a lighter drawing from the comparative literature. While she points to some of the considerations that multiparty political systems create, she really focuses on a two-party system. This means that the two-option version of party systems takes precedence, such that the individual party can imagine winning a majority by itself. Mancur Olson pointed out that majority-seeking political parties differ from other organizations by being more encompassing and inclusive than the interest group or more niche-oriented parties. Majority-seeking parties are thus more likely to seek something like a public good.Footnote ² This implies a two-party exceptionalism that makes a focus on any two-party nation less than general in its claims.

What are the virtues of political parties? Antipartyism argued that parties divided rather than unified the polity or were simply divisive and thus harmful. In contrast, Rosenblum's Burkean view is that parties regulate rivalry, as in the role of loyal opposition in Britain. Her Hegelian argument is that parties provide accountability to governing. Her “proto-Millian” idea is that dialogue between the two parties improves and refines political opinions. Note that at least regulated rivalry and her proto-Millian argument are more about the virtues of a system of political parties than they are about a party per se.

And what of partisanship? If it is “[t]hrough parties [that] interests and opinions are organized and brought into opposition,” and that “[p]arty rivalry is constitutive, [so that i]t stages the battle,” then it is the “[p]artisans [who] are necessary to realize the value of parties” (p. 457). Partisanship provides inclusiveness, comprehensiveness, and a disposition to compromise (though that is compromise within parties, rather than between or among them). In her view, then, partisans are necessary for political parties, just as the parties are necessary for creating partisanship. The virtues of the party align with the virtues of partisanship.

Let me conclude with three observations. First, the defense of parties is, in my opinion, a defense of a system of political parties, not a defense of parties per se. Inclusiveness, comprehension, regulated rivalry, a Millian refinement by debate among ideas—all of these require multiple parties. Accountability further requires continuity over time. These are characteristics of a system of political parties. As V. O. Key, Jr., argued in Southern Politics, Footnote ³ in a one-party South (and presumably any situation where political parties do not form a system, recurring over time), there is no accountability, compromise, comprehensiveness, and so on.

This is my second point: The virtues of parties and partisanship are the virtues of a democratic party system. The book is, in this reviewer's opinion, a story about democratic politics. A party system is necessary to make democracy effective in an extended republic. It is not clear that the virtues outlined by Rosenblum hold for parties in a nondemocratic system. The virtues of parties, that is, are the virtues of a democracy, or at least the virtues of a democratic republic.

Finally, the ambivalence we feel about parties is not ambivalence about political parties per se. It is ambivalence about democracy.Footnote ⁴ This is the ambivalence of Churchill (i.e., democracy being the worst system except for all the rest). The worst part of democracy is, in part, its party system. However, a democracy with a party system is the worst system, except for any democracy without a party system. And thus, Rosenblum has written an original and spirited defense of a necessary part of any democracy in an extended republic. No mean feat, that.