Rashomon Goes to Rwanda
19
coding full articles and not just leads) tends to give humans a distinct
advantage.
Differences also exist with regard to how individuals code from sources.
Some pay close attention to events (discrete activities demarcated within
space and time): identifying their frequency, magnitude, location, and so forth
(e.g., Sorokin 1937; Tilly et al. 1975; Taylor and Jodice 1983; Francisco 2000).
These individuals believe that such information is intricately connected with
understanding what takes place out in the world (Tarrow 1998). From this
perspective, the most important things that humans engage in are events and
to understand what individuals as well as groups do therefore one must focus on
these ‘events-based’ aspects of existence. Others believe that such efforts are
ultimately misleading for they maintain that sources are incapable of providing
an accurate characterization of events. Instead, they suggest that researchers
should use sources to construct rough pictures of what took place by using
categories (e.g., civil war/no civil war; violent protest/non-violent protest/no
protest).
Following the literature, we employ a research design where we take
advantage of the detail allowed within a case and combine this with the
precision allowed by a focus on events. Deviating from much of the literature
but drawing upon the work of numerous individuals (e.g., Scott 1985; Lustick
1996; Davenport and Ball 2002), we also explore the diversity that exists within
Convention