No abstract

Abstract.  By most definitions, the third wave of democratisation has given rise to dominant parties and dominant party systems in Africa. The effective number of parties, the most widely used method to count parties, does not adequately capture this fact. An analysis of 59 election results in 18 sub‐Saharan African countries shows that classifications of party systems on the basis of the effective number of parties are problematic and often flawed. Some of these problems are well known, but the African evidence brings them out with unusual clarity and force. It is found that Sartori's counting rules, party system typology and definition of a dominant party are still the most helpful analytical tools to arrive at an accurate classification of party systems and their dynamics in general, and of dominant party systems in particular.

Scholarly attention has started to shift from democratization and democratic consolidation to trends of democratic deconsolidation, backsliding, regression, and erosion. This article examines Hungary as a deviant and exemplary case for understanding de-democratization. The starting point is the literature on defective democracy, which provides a unified framework of analysis for the causes and the outcomes of democratization. However, as the case of Hungary shows, dedemocratization is not simply the mirror of democratization. In Hungary, both the outcome and the process of de-democratization defy expectations. The democratic defects do not conform to any of the standard types, instead resembling a "diffusely defective democracy". Moreover, existing explanations fail to account for their emergence. The case of Hungary indicates that our knowledge of democratization may be a poor guide to understanding de-democratization.

In a recent publication in this journal, Mozaffar and Scarritt claim to have found a puzzling combination of low fragmentation and high volatility in African party systems. However, if we look at national party systems rather than Africa-wide averages, include regime type as a variable and specify dominance, we find three different constellations: dominant party systems with relatively low volatility, non-dominant and pulverized party systems with high volatility and dominant authoritarian party systems with high volatility. The real surprise is that dominant parties in authoritarian regimes have higher electoral instability than dominant parties in democracies. The analyis is based on data from 78 elections in 20 African countries with at least three consecutive multiparty elections.

Consociationalism has enriched comparative politics with a whole lineage of non-majoritarian types of democracy: from consociational democracy to consensus democracy and power-sharing. This article unravels the development, interaction and succession of empirical and normative typologies in 30 years of consociational literature as embodied in the work of Lijphart. It argues that consociational theory is plagued by serious conceptual problems which remain undetected by current inquiries into proper concept formation. The problem lies both in Lijphart's empirical typology of democracies and in the presence of a complementary but incongruent normative typology. The conclusion is that in the end the two kinds of typology weaken instead of strengthen each other and lay bare fundamental weaknesses in consociational theory. It is suggested that the empirical investigation of the normative rival types of consociational and majoritarian democracy, properly defined and operationalized, should be at the heart of new research strategies.