Between 1850 and 1868, the United States struggled through a long and bloody Civil War, attempted to reconfigure southern class and economic relations, settled much of the western prairie, and began the transition from an agrarian to an industrial society. During these two decades, Americans went to the polls, whether located in hamlets swarming with Union soldiers, rural churches set in the midst of the southern cotton belt, wooden cabins so isolated that even the nearest neighbors had difficulty finding them, or saloons in the most densely populated sections of great cities. Their votes have come down to us as election returns reporting tens of millions of officially sanctioned exercises in democratic participation. Neatly collated and arrayed in columns by office, candidate, and party, these returns are routinely interpreted as more or less accurate reflections of the popular preferences of the individuals within the communities in which they were made out. Seen this way, the returns themselves appear to constitute unambiguous and overwhelming evidence of the existence of a robust popular democratic system and ethos during the mid-nineteenth century. The purpose of this article is to suggest some important caveats that must attend this conclusion.

“The American Ballot Box” is a most interesting exploration into the deep past of American electoral politics. Well-researched and argued, it provides examples of sometimes very funny interview sessions with nineteenth-century American voters. The basic data method here rests upon using congressional-election contest hearings as sources of information. The author does well, however, to remind us that by definition, election contests may have reflected “the dark side of the force,” so to speak, but also necessarily raised claims of fraud and other abuse by the losers. Naturally, the overwhelming majority of U.S. House elections in the nineteenth century were not contested. The strength of the method lies in the fact of contest itself, which permitted relatively searching inquiry into the conduct of elections, and thus revealed much about their organizational pattern that would be taken for granted (and hence not discussed or brought to light otherwise). But bias may be implicit in the method.

The precipitous decline in collective bargaining between employees and employers during the last twenty-five years of the twentieth century in the United States should tell us something useful about American political development. The decline of collective bargaining is about more than waning union influence or competitive conditions in the labor market because collective bargaining was the primary expression of the regime established by the New Deal Democratic Party electoral coalition. For a generation, labor-management bargaining epitomized empowered liberal democracy which, acting through the legislative branch, established a new legal regime of industrial order. The National Labor Relations Act of 1935 (NLRA) shaped a pluralist industrial order by creating new rules that curbed the authority of managers and compelled business firms to participate with unions in a process of rational bargaining based on the assumption of competing employee and employer interests.I do not mean to suggest that the place of collective bargaining was uncontested. Scholars, however, launched major efforts in the 1950s and 1960s to understand how “industrial relations” fit into the emerging conception of pluralism. This was an important effort because union-management relations were central to the New Deal Democracy. For example, John Kenneth Galbraith, American Capitalism (New York: Houghton Mifflin, 1956); Robert Dahl and Charles Lindblom, Politics, Economics and Welfare: Planning and Politico-Economic Systems Resolved into Basic Social Processes (New York: Harper and Row, 1953); Clark Kerr, Labor Markets and Wage Determination: The Balkanization of Labor Markets and Others Essays (Berkeley: University of California Press, 1977); Clark Kerr, John T. Dunlop, Frederick Harbison, and Charles A. Myers, Industrialism and Industrial Man: The Problems of Labor and Management in Economic Growth (Cambridge, MA: Harvard University Press, 1960); Grant McConnell, Private Power and American Democracy (New York: Knopf, 1966); J. David Greenstone, Labor in American Politics (New York: Vintage, 1969). By the mid-1960s, industrial relations was a rare topic for political science investigation; aside from Greenstone, political science conceived “labor” and “work” to be “private interests” and not a matter of political significance. Yet collective bargaining is no longer the focal point for industrial relations, and industrial relations no longer occupies the center of governing attention. The various reasons for this trend have attracted significant scholarly attention; however, the issue is confounded by arguments that union decline is a wrongly conceived topic to investigate.

While attracting ardent interest from social scientists on a number of fronts, the Progressive Era is insufficiently examined within presidential studies. Scholars have, instead, portrayed Theodore Roosevelt and Woodrow Wilson as prefiguring a later, modern presidency. For example, Sidney Milkis writes that the emergence of the modern presidency began in the Progressive Era, “especially as shaped by the statesmanship of Theodore Roosevelt and Woodrow Wilson.”Sidney Milkis, The President and the Parties: The Transformation of the American Party System Since the New Deal (New York: Oxford University Press, 1993), 8. William Leuchtenburg observes that Franklin Roosevelt contributed to the modern presidency by giving the office “an importance which went well beyond what even Theodore Roosevelt and Woodrow Wilson had done.”William E. Leuchtenburg, Franklin D. Roosevelt and the New Deal (New York: Harper Torchbooks, 1963), 327. However, scholarship has not explained fully why these two Progressive Era presidents were precociously “modern.”

Although in existence for only a few decades, the field of American political development (APD) has been an agent of many changes within the discipline of political science. It has been a leading force in the rediscovery of institutions; it has bolstered a comparative approach to American politics; it has helped to draw attention to previously neglected issues such as race, class, and gender; and it deserves primary responsibility for bringing the temporal dimension of politics into consideration.

John Gerring's methodological critique of American political development can be viewed from at least three perspectives. First and most generally, Gerring proposes procedural and analytical standards that should apply to all the social and natural sciences. As a methodological virtue, for instance, conceptual clarity strengthens the underpinnings of knowledge, regardless of the substantive focus. Rigorous specification of causal relations is equally universal. From a second perspective, Gerring invites his readers to at least entertain the possibility that methodological rigor might vary over the lifespan of a research field. From this angle, seminal work often involves more relaxed methodological styles because initial exploration of a topic or area, almost by definition, must be unguided by research conventions and norms. Such conventions and norms only emerge if and when that exploration attracts a self-sustaining community of scholars.

In his methodological critique of APD research, John Gerring steers clear of the issues that have bogged down debates about methodology in recent years. He does not argue that historically-oriented students of American politics should adopt some allegedly superior method current in other sectors of political science, nor does he demand clarification of an alternative method that might be more pertinent given some allegedly distinctive qualities of the APD project. Instead, he scores the project for its failure to measure up on universal standards of good research design—clear concepts, clear propositions, clear logic. The resulting critique is, on its face, singularly disarming. What self-respecting scholar wants to be unclear?

John Gerring's call for greater methodological self-consciousness and rigor in APD scholarship is valuable in several ways. Although some APD scholars deploy formal models and statistical analyses of large n data sets, many do not. Today, those choices inevitably raise questions of the rigor of such work. Gerring's essay usefully pushes us to provide answers. I also generally agree with his sense of what appropriate methods involve. The fact that his recommendations are fairly standard and noncontroversial then poses the further issue Steve Skowronek stresses: whether there is any distinctive method or substance that defines APD scholarship as a field.