What are the sources of democratic legitimacy? Pierre Rosanvallon raises this question in the introductory section of Democratic Legitimacy (first published in French in 2008). From the late nineteenth century until the 1980s, he notes, democratic legitimacy was connected to two key institutions. The first was the ballot box, which conferred legitimacy by means of elections—elections eventually characterized by universal suffrage. The second was a professional bureaucracy or civil service—positions that were open to all on the basis of competitive examination—which grounded legitimacy in identification with social generality. Since the 1980s, however, this dual legitimacy has been increasingly questioned, for two principal reasons. The first is that elections have been “desacralized,” owing in part to a changing understanding of “majority” and in part to an increasing emphasis on minorities, resulting in changes in understanding of who the “people” are. The second is the damaged credibility of bureaucracy (5). Rosanvallon explores the development and the decline of these two institutionalized expressions of legitimacy in part 1.

Their decline provides an opportunity to consider new varieties of democratic legitimacy. Though we cannot simply set aside concerns about procedural legitimacy (particularly as expressed through the mechanism of elections), Rosanvallon focuses his attention on social generality as he explores these new forms. He sees three ways of establishing social generality other than through universal suffrage or traditional forms of bureaucracy. There is a “negative generality,” which he associates with what he calls the “legitimacy of impartiality”; a “generality of multiplication,” which he associates with the “legitimacy of reflexivity”; and a “generality of attention to particularity,” which he associates with the “legitimacy of proximity” (6–7). As the work unfolds, Rosanvallon takes up each of these forms of legitimacy in turn: the legitimacy of impartiality in part 2, reflexive legitimacy in part 3, and the legitimacy of proximity in part 4. His intent is “to develop a conceptual framework for evaluating the democratic potential” of institutions (such as independent oversight and regulatory bodies and constitutional courts) and practices associated with these newer forms of legitimacy (11).

The first of these forms is what Rosanvallon calls the “legitimacy of impartiality.” It is most clearly expressed in the development of independent oversight or regulatory agencies. Such agencies, designed to protect the public interest, supply what neither legislative bodies nor bureaucracies can: a high level of expertise, a flexible system of regulation, and the possibility of arbitration (77–78). To carry out their functions, they need to be able to function independently of both elected authorities and the bureaucracy (though elected officials might play a role in their constitution). Such agencies have at times been held suspect, and their constitutional status questioned. Nonetheless, such agencies are genuinely representative (both in their composition and in their accessibility and openness to input from the general public), and therefore have democratic legitimacy. They simply implement a different type of generality than elections do. Power need not be elective to be representative.

Social generality can also be expressed through what Rosanvallon calls “reflexive legitimacy.” This second form is the subject of part 3, and Rosanvallon connects it closely (though not exclusively) with constitutional courts and judicial review. Democracy requires reflexivity as part of an effort to maintain an appropriate understanding of who and what “the people” is. “The people” is not a monolithic entity, nor can it simply be equated with the majority. Furthermore, “the people” exists across time. Constitutional courts, though they do so in a less immediate way than elections, do represent the people, fostering the collective memory, improving our political deliberation by keeping multiple points of view (including those of previous generations) in the discussion, and making clear the distinction between the sovereign people and a mere majority. It is true that reflexive institutions such as constitutional courts are countermajoritarian. However, they are useful for meeting two key requirements of democracy: “it has to arrange for periodic choice among sufficiently different individuals and programs, and it must establish institutions that rise above those differences to promote the general interest” (163).

In part 4, Rosanvallon introduces readers to what he calls the “legitimacy of proximity.” With its emphasis on particularity, this form of democratic legitimacy may seem an odd means of expressing social generality. How can generality result from attention to particularity? Each person wants to count for something, and this desire to be valued requires that attention be paid “to the infinite variety of singularities that exist in the real world” (186). But this does not imply a focus on singularity for its own sake; the intent is that “greater attention be paid to social diversity so that no one is sacrificed on the altar of abstract principle” (185, emphasis added). The key is for as many singularities as possible (ideally, all singularities) to be taken into account, so that generality ends up being expressed precisely through active engagement with the particular. Citizens concerned—consciously or otherwise—with this form of generality will seek representatives who are present to them, who will engage with them, and who understand them, whether or not they resemble them demographically.

These three forms of legitimacy underlie the vision of what Rosanvallon calls the “democracy of appropriation,” which is characterized by three main features: (1) a certain distrust, which leads to such “counterdemocratic” (or, more accurately, “countermajoritarian”) practices as oversight, impeachment, and judgment, providing a means to keep government in check; (2) agencies of indirect democracy that express social generality in ways the ballot box cannot, such as oversight bodies, regulatory agencies, and constitutional courts; and (3) an insistence that leaders “conduct themselves democratically”—that they be present to those they represent—which places constraints on their actions, regardless of how they are chosen (220–21).

This “democracy of appropriation” is, in Rosanvallon's view, more realistic than what he terms the “democracy of identification”; it takes account of the ways elected officials behave and the distance between them and the governed, while remaining positive in that it “points the way toward an effective social reappropriation of power,” bringing power closer to those over whom it is exercised (221). It also gives us a new way to think about the separation of powers. Rosanvallon notes that we can no longer argue that there is a real separation between the executive and legislative branches of government. Nevertheless, there is a genuine separation of powers about which we can speak meaningfully; there is what Rosanvallon has elsewhere (in Counter-Democracy: Politics in an Age of Distrust [Cambridge University Press, 2008], 318–20) called the “mixed regime of the moderns.” This division of powers is seen “in the existence of counterdemocratic and indirect democratic institutions in tension with the sphere of majoritarian powers” (221–22).

As he builds his case for a “democracy of appropriation,” Rosanvallon carefully traces the historical development of the newer expressions of social generality that underlie it (taking his examples primarily from France and the United States). His argument is well grounded in both the historical and the theoretical literature, and demonstrates the value of critical reflection on historical experience, showing clearly how new understandings of generality arise from the actual practice of politics. Rosanvallon has provided us with a carefully written, accessible work that takes full account of the complexities of contemporary democracy. It will be of interest to anyone concerned with issues of democratic legitimacy.