Mogens N. Pedersen

On Measuring Party System Change A Methodological Critique and a Suggestion

When studying party system change over time, it is tempting to use as indicators one or several of the existing measures of fractionalization and concentration. If these measures are used to study change, they will often be misleading because of their inability to discriminate among even very different diachronic patterns. Seven indicators are tested on a hypothetical set of data, and all of them fail to reflect typical patterns of change. Thus if students of party systems want to study the general phenomenon of party system stability or change, they should preferably use other types of measures. An indicator of aggregate volatility is shown to satisfy the requirements of a good indicator of party system change.

Parliamentary Roles in Modern Legislatures

When John Wahlke and his collaborators in 1962 published the monumental The Legislative System they literally started an avalanche of studies. Their approach defined and dominated the emergent field of legislative studies. Optimism was great, as was the self-assurance. ‘Political behavior is always conduct in the performance of a political role’, thus Heinz Eulau argued for the centrality of the new conceptualisation in his influential pamphlet from 1963, The Behavioral Persuasion in Politics (p. 40). A couple of decades later the studies of legislative roles gradually disappeared from the agenda. Many explanations can be offered for this. Role analysis and the functionalist approach just became unfashionable, for some even politically incorrect; the marginal utility of empirical studies tended to decrease; the representational roles of delegates and trustees were not as useful as expected, since most politicians were neither – but instead politicos; the weakness of the theoretical framework due to its American roots, that is, the low priority given to the phenomenon of political party, was also part of the explanation. The study of legislative roles never totally disappeared. Important theoretical work was still done during the 1990s by, for example, Donald Searing. Kaare Strøm tried to bring role analysis in contact with the emergent neo-institutional framework, called rational choice theory. Empirical studies were still carried out in Europe. But the heyday was over. Magnus Blomgren and Olivier Rozenberg try to revive the tradition in this edited volume. They do it by bringing together a group of writers, who are able to demonstrate that there still is a lot of empirical mileage in the old concept. When comparative and longitudinal perspectives are added to an otherwise static approach, many new questions pop up. The downside of the considerable variety is some conceptual stretching that sometimes leaves the reader with a headache. But the editors’ own contribution is also noteworthy. The introductory and concluding chapters are very well executed. The best of all chapters in this fine book is probably their own erudite discussion of the long history of legislative role research, which has ‘indeed been somewhat bumpy’, as they say. They identify and discuss three major problems for role analysis: the conceptual confusion and the problematic relation between the theoretical framework and daily politics; the problematic use of roles as independent variables; and, thirdly, the equally problematic use of roles as dependent variables.