By Thomas Giacoletto
There are different approaches to historical analysis, and they all bring different aspects of the studied phenomenon to light. This post will illustrate two of them, taking as an example the emergence of niche parties in Europe.
Steinmo (2008) and other historical institutionalists would, for example, ask: ‘how did electoral institutions, and the party systems that they created, influence the formation of niche parties in France after the second World War?’ This approach focuses on institutions to explain behavioural change over time, and would, for this example, help us understand how the electoral institutions shaped political actors’ behaviour. The conclusion of this study would illustrate the incentives electoral institutions create, and how these have changed over time; previously not giving incentives for the creation of niche parties, and doing it now. The main issue with this approach is that it would only give indications about the exact context of the study. This means that there is no generalisation possible, and that another country (for example) would have to be studied in such detail in order to draw conclusions.
Bates ea. (1998), on the other hand, would use their analytic narrative method in order to both use the context, and come up with “explicit and formal lines of reasoning” through the use of combined historical method and game theory. They would ask: ‘are electoral systems that give centrifugal incentives, more likely to generate niche parties?’ Their study would clearly explore the historical context of the emergence of niche parties (using documents…) and design a game that would seek to explain the actions of actors as responses to incentives given by the electoral system. With the outcome of their study, they would have modelled individual’s reactions to particular incentives given by the institutional structure.
The analytic narrative method might lead to less information to be collected about the case and the context, but it seems more useful if you want to be able to explain behavioural patterns. Furthermore, it seems to help us understand more about how individuals behave than the other method.