A New Kind of Social Science: Moving Ahead with Reverse Wolfram Models Applied to Event Data.

Existing formal models of political behavior have followed the lead of the natural sciences and generally focused on methods that use continuous-variable mathematics. Stephen Wolfram has recently produced an extended critique of that approach in the natural sciences, and suggested that a great deal of natural behavior can be accounted for using rules that involve discrete patterns. This paper reports some initial findings from a new NSF-funded project designed to apply this pattern-based method to political event data. The core of this project is a new, publicly-accessible web-based tool designed for the analysis of event data patterns. Using data on the Israel-Palestine conflict for the period 1979-2004, we first consider some differences between the activities of various sub-state actors. While most prior event data analysis has simply aggregated all activities, we demonstrate that some sub-state actors produce streams of activities that are statistically independent. We continue the analysis by showing there are distinctive--and very plausible-- differences in the patterns found during the tenures of various Israeli prime ministers, and the "interesting" patterns go well beyond the simple variations on tit-for-tat that we found in our earlier work. [ABSTRACT FROM AUTHOR]