2012/08/01 by Stewart E. Tolnay · 2 citations
Social Sciences · Psychology · Mathematics · #Social and Cultural Dynamics #Urban, Neighborhood, and Segregation Studies #Race, History, and American Society #Narrative #Data science #Psychology #Computer science #Econometrics #Mathematics #Literature #Art
paper · doi:10.1177/0081175012460855
published in Sociological Methodology 42(1), 91-93 (SAGE Publishing)
openalex publication_date 2012/08/01 · openalex created_date 2025/10/10 · openalex updated_date 2025/11/06
Franzosi, De Fazio, and Vicari (this volume, 2012:1-42) propose a strategy for using details contained in narrative accounts of social events to give life to the actors in those events and to reveal agency on the part of those actors. They use the history of lynching in the state of Georgia to demonstrate an application of the proposed meth odological tool, Quantitative Narrative Analysis (QNA). Early in this paper Franzosi and colleagues describe their primary objective as methodological rather than sub stantive. But in evaluating the contribution of this paper it is necessary to consider both methodological and substantive issues. As Franzosi and colleagues correctly observe, the quantitative study of southern lynching has relied heavily on aggregate data (spatial or temporal) to describe the strength of relationships between the intensity of lynching (usually counts of inci dents or victims) and a variety of social, economic, political, and demographic vari ables. As a result of these studies, we know that lynchings were more numerous where the African American population was relatively large, the agricultural econ omy was based predominantly on cotton, the white population was economically stressed, the Democratic Party was stronger, and multiple religious organizations competed for congregants. Savvy social scientists strive to avoid committing an eco logical fallacy by drawing inferences about individual-level motivations and beha viors from such aggregate evidence. Still, the temptation to do so is great because the loss of the actors involved in the lynchings seems a high price to pay for relying on the available, but aggregated, data. Franzosi's interest in the use of narrative for social scientific inquiry is long standing, and he and his colleagues have provided useful introductions to the QNA methodology elsewhere (see the citations in their work). The current paper offers an intriguing, and potentially very valuable, opportunity to complement the aggregate