Publications

Prevailing theories about the nature and development of the democratic party system fail to account for the important case of the United States. Using a novel dataset on legislators and roll call votes in the New York State Assembly after the ratification of the U.S. Constitution, we show that, contrary to existing accounts, legislative parties had already formed at this early stage. Yet these parties did not arise from the translation of social cleavages such as economic or social class into political oppositions, as sociologists might expect, nor were they merely networks of powerful elites disconnected from the polity, as political scientists and historians have suggested. Instead, these parties coalesced around formal issues—structural questions like the procedures for election and appointment, questions whose answer would determine the rules of the game for future contests. Parties emerged, we argue, not because of an inherent need to adjudicate conflicts between sectors of the polity, but because of the organizational affordances of the modern democratic state. Our findings suggest the formation of party systems is an integral part of the formation of the modern state.

Social scientists have long been interested in elite cohesion in American society, recognizing its potential implications for democracy and governance. While empirical research has focused on corporate elites and, in particular, on cohesion derived from shared board memberships, cohesion among those in highest positions in the American state and historical change in that cohesion have been little studied. Drawing on a novel dataset of the career histories of 2,221 people in these elite positions who were appointed to them between 1898 and 1998, I examine whether administrative elites, prior to their elite appointment, attended the same educational institutions or worked in the same agencies of the federal government at the same time. I find evidence of increasing elite cohesion during the twentieth century. Educational cohesion increases significantly in the three decades following the Second World War and then declines slightly toward the end of the century. This increase goes hand in hand with a change from college to graduate education as the primary site generating educational cohesion. Federal government workplace cohesion increases markedly in the 1930s and 1940s and then remains high. As people are appointed to different organizations within the American state, their educational and workplace connections create inter-agency networks that, it is expected, facilitate mutual understanding and coordination and thus help integrate the American administrative state.

While much of economic sociology has been interested in how culture shapes economic networks, there has been less attention to how economic networks shape culture. In this chapter, we focus on the case of political culture. Surveying literature in political and economic sociology and history, we argue that there are general patterns that link formal characteristics of economic networks (whether actors have direct ties or occupy structurally equivalent positions, whether relations are voluntary or involuntary) to aspects of political culture (whether it is particularistic or universalistic, whether it is conflictual or pacific). We discuss how these patterns vary for elite and non-elite political actors, and conclude with some preliminary generalizations about the relationship between economic and political relations that can guide future exploration.

It is common for social scientists to use formal quantitative methods to compare ecological units such as towns, schools, or nations. In many cases, the size of these units in terms of the number of individuals subsumed in each differs substantially. When the variables in question are counts, there is generally some attempt to neutralize differences in size by turning variables into ratios or by controlling for size. But methods that are appropriate in many demographic and epidemiological contexts have been used in settings where they may not be justified and may well introduce spurious relations between variables. We suggest local regressions as a simple diagnostic and generalized additive models as a superior modeling strategy, with double-residualized regressions as a backup for certain cases.

The Florentine *Consulte e Pratiche* is the oldest recorded series of speech-by-speech policy discussion by political elites in European history, over one hundred and fifty years in length. This article is the first of an extended two-article sequence on political discussion in the *Consulte e Pratiche*, during the 1376–1378 period of the War of Eight Saints, which led up to the famous Ciompi Revolt. Our interest is in discovering both the semantic- network (article 1) and the factional-network (article 2) mechanics of this unexpected spillover from foreign-policy conflict into domestic revolt. Our central finding at the semantic level, in this first article, is that the spillover from war to revolution was mediated through the ceremonial and political-economy sides of religion. The methodology in this first article is to uncover the evolving narrative-network structures exhibited in Florentine political discussion – namely, changing inter-correlations among keywords about topics, through chapters and subplots. "Narrative-network analysis" for us means (a) uncovering changing topological portraits of how subplots interlink through time, and (b) discovering interlocking linguistic "hinges" through which new historical trajectories of subplot combinations become defined. In our case, the linguistic hinges between foreign policy and domestic revolt were rooted in religion. How the evolving issues and topics discussed in this article express themselves in domestic (and eventually violent) political conflict between the anti-war Parte Guelfa faction and the pro-war Civic ‘faction’ will be the subject of the second of this complementary pair of articles.

We analyze public-policy speeches in the Florentine *Consulte e Pratiche*, immediately prior to the Ciompi Revolt, for signs of elite factional conflict, in the context of self-proclaimed unity. We employ three statistical analyses of these speeches in Latin: namely, scatterplots of word frequencies, Wordfish scaling, and regressions on speech-similarities. Plus we employ two qualitative analyses: a case study of the speeches of Lapo da Castiglionchio, leader of the Parte Guelfa faction, and a close examination of the rhetoric of unity in three important sets of meetings. Our main finding is this: The runup to the Ciompi Revolt was crystalization of "unity of citizens" in the room of the *Consulte e Pratiche* and, among the same actors, crystallization of "unity of Guelfs" in the room of the Parte Guelfa, with a lack of recognition in the multivocal speeches in the former of the obvious contradiction with actions in the latter. In our opinion, the tragedy of "the valiant failure of republicanism" in Florence was that intense wishful yearning for unity in speech induced, under background conditions of deep social-class contestation about "Who is Florence?," an intensification in action of the very revolutionary forces that it most desperately wanted to suppress.