APSA | COMPARATIVE POLITICS THE ORGANIZED SECTION IN COMPARATIVE POLITICS OF THE AMERICAN POLITICAL SCIENCE ASSOCIATION BACK TO SUMMARY APSA-CP Newsletter Vol. XXIX, Issue 2, Fall 2019 page 58 THE STUDY OF ARMED CONFLICT AND HISTORICAL DATA 1 by Stefano Costalli and Andrea Ruggeri Why use historical data to study armed confict? Quantitative studies of armed confict, and more specifcally civil wars, have grown expo- nentially in the past few years (Cederman and Vogt 2017). However, so too, has the use of his- torical data to study armed confict. The use of historical cases to study armed confict is well established (Tilly 1978; Skocpol and Theda 1979; Petersen 2001) with other areas of Comparative Politics experiencing a “historical turn” nearly ten years ago (Capoccia and Ziblatt 2010). In recent years, the systematic collection of his- torical sources to generate new datasets for the quantitative study of armed confict has ris- en to prominence as a result of the awareness that civil wars, rebellions and all forms of con- fict involving non-state actors can be fruitfully studied using spatially and temporally disag- gregated data (Cederman and Gleditsch 2009). Kalyvas (2006) inaugurated this trend using a mix-methods approach to leverage historical data from the Greek civil war. More recent books have followed this path (e.g. Balcells 2017; Finkel 2017; Kopstein and Wittenberg 2018), and his- torical data has been efectively harnessed in a range of recent articles (Costalli and Ruggeri 2015; Zhukov 2017; Kocher, Lawrence, and Monteiro 2018). Today researchers have more opportunities to create large N-datasets from historical sources than just a few years ago, thanks to a combina- tion of factors that include the opening of some large once-private, secret, or protected archives, the evolution of technical digitalization, and the availability of powerful software to manage text and its codifcation. In this brief piece, we take stock of the publications of the past decade and provide some preliminary thoughts on why using historical data to study armed confict is benefcial for political scientists. Further, we also highlight some limitations and issues that should not be underestimated by those at- tempting this kind of research, before closing with a few suggested remedies. Historical Data and Object of Study: Legacies of Conflict and Legacies on Conflict Historical data may be intrinsically necessary to study specifc phenomena, such as the long- term efects and legacies of confict (Wittenberg 2015; Costalli and Ruggeri 2018). In these cases the “legacies of confict” are represented by past patterns of violence or confict that can be core explanatory factors for a range of subsequent social, economic or political forms (Lupu and Stefano Costalli is an Associate Professor of Political Science, Università di Firenze. His email is stefano.costalli@unifi.it. Andrea Ruggeri is a Professor of Political Science and International Relations, University of Oxford. His email is andrea. ruggeri@politics.ox.ac.uk.