The field of institutional design in plural societies has evolved around the debate as to which ideal type of institutional design is better suited to cope with the difficulties of plural societies. A number of authors expressed doubts as to whether these ideal types of institutional design can be fully implemented in any real empirical set-up. Additionally concerns have been raised as to the empirical validity of ideal models and the arising uncertainty of with regard to real world configurations of political institutions. To answer these concerns this paper undertakes a configurational approach to the study of empirical configurations of political institutions in plural societies. The aim of the paper is to find patterns of configuration of five crucial political institutions (executive type, type of electoral system, type of party system, type of territorial division of the state and coalition size) in 18 plural societies in the years 1960-2008. Using the exploratory capacities of the Qualitative Comparative Analysis, I first test the frequency of occurrence of ideal models which have dominated the field of institutional design in divided societies. Later on I explore alternative, real world patterns of configuration of institutions which explain institutional variation across cases. The research sheds doubt on the empirical relevance of the ideal models, showing that ideal models are rarely implemented in the empirical reality, while some of them do not have their empirical references at all. Through an exploratory analysis of the available data I identify seven patterns of configurations of institutions in plural societies which cluster to each other following the common logic of institutional design and regularities of bi-variate relationships between institutions. The research also finds that the commonly accepted regularities of relationships among institutions are weak in plural societies, where crucial intervening variables weaken these relationships. The configurational perspective applied in the research also gives a new perspective to assessment of centrifugal and centripetal characteristics of separate institutions and their configurations.
