Jørgen Elklit & Ole Tonsgaard, eds., Valg og vælgeradfærd: Studier i dansk politik Aarhus: Forlaget Politica, 1984, 416 pp.

The Department of Political Science at the University of Aarhus, which was founded in 1959, celebrated its quarter centennial in 1984 by publishing a series of books. These publications illustrate both the breadth and the vitality of the scholarship produced at this young institute. The volume under review is devoted to electoral research. The fourteen contributors to this volume discuss a number of central topics dealing with the interpretation of elections such as the relationship between social class and party choice, the dimensions of the party system, protest behavior and the role of mass media.

The time span covered in some of the chapters is much longer than the history of the Aarhus institute. Jørgen Elklit takes a new look at the problem of stability of the party system. Despite recent earthquake elections, the 1973 election being the most famous example, the basic configuration of Danish political parties remains unchanged. Elklit goes back to the formative years in the late 19th century and the early 1900s. His chapter includes corrected data on votes and seats in the Folketing elections between 1881 and 1920. These tables will no doubt be a standard reference for those interested in early Danish electoral history. Elklit traces the development of the initial two-party system, via the three-party system (following the split within Venstre in the early 1890s), into the four-party system a decade later. He concludes that the subsequent stability of the party system can be attributed to the fact that Denmark already by the turn of the century had a set of parties which corresponded to the basic groupings in the economic system.

Søren Risbjerg Thomsen also puts recent developments in a long historic perspective when analyzing all the elections held under the P.R. formula (1920-1984). His use of territorial data is interesting both in a methodological and a substantial way. Thomsen criticizes much of the previous work basedon ecological analysis for using statistical techniques that overestimate the importance of the socio-economic dimension. His own alternative is basedon certain assumptions about the relationship between ecological and individual

levels and he estimates individual level relations by magnifying intra-regional variation. Not surprisingly, Thomsen finds a long-term decrease in class voting. The interesting thing, however, is that this trend is not uniform. Those periods in which there is a particularly marked drop in class voting tend to coincide with the years the Social Democratic Party held office. Thomsen concludes that the Social Democratic Party, by pursuing a policy of compromise, has lost support within the the working class and has contributed to a decreasing polarization of class society.

In the same chapter, Thomsen also relates election results to government structure. Since no Danish party — primarily due to the proportional election system — has ever been able to secure an absolute majority in the Folketing, all governments have been minority governments or coalition governments (or a combination of the two). The formal election period is 4 years, but elections have actually been held every 2 years and 8 months. Only once (during the German occupation in 1943) has an election been held because of the fact that the 'normal' election term ran out. Quite frequently, lack of parliamentary support has caused a government to call a new election. What then is the connection between the reasons behind the calling of an election and the results of that same election? Obviously, it is not advantageous to wield political power. Parties in government lose on the average 0.5 percentage points. It is even more of a liability to support the government without formally being represented in the government. The average election performance for these 'support parties' and government parties together was minus 1.7 percentage points. A certain 'prime minister effect' is, however, visible, but only for non-socialist parties. Within the nonsocialist camp, it has proved beneficial, in terms of votes gained, to be the leading party in the government.

A book on recent developments in Danish electoral history would not be complete without a discussion of the Glistrup phenomenon. Ingemar Glans presents a few of the more common theories on the rise and decline of the Progress Party. He rejects the marxist-inspired theory of the petty-bourgeois, anti-monopolistic, populist revolt. Social groups classified as petty-bourgeois constitute only a minority of the Glistrup party's electoral support. As a matter of fact, the Progress Party, in its heyday, drew votes from almost all social strata (with the exception of salaried employees in the public sector). Glans finds more evidence in favour of the proposition that the Progress Party is but one, albeit a peculiar, non-socialist party. Second preferences and voter attitudes indicate that the Progress Party can be safely located on the right wing of the sociopolitical spectrum. The primary reason why Glistrup succeeded in gaining 15.9 percent of the votes in 1973 was the inability of the regular non-socialist parties to present a viable alternative to the then predominant policy of high taxation and extensive public spending. But by the time the non-socialist parties, notably the Conservatives, regained momentum, Glistrup's popularity decreased and the Progress Party, by the 1984 election, had shrunk to 3.6 percent.

In the mid-19705, the Glistrup party was particularly successful among the young voters. Ten years later the situation had changed. The Progress Party was as unpopular among the youth as among the adult voters. But, in other respects, young voters in Denmark exhibit a few special characteristics, as is shown in a chapter by Palle Svensson. As in the rest of Scandinavia, young Danish voters in the 1970s leaned to the left. What is somewhat surprising is that this trend does not seem to be affected by the current right-wing tendencies in the electorate. On the contrary, the proportion among the youth voting for any of the extreme left parties reached an all-time high of 37 percent in 1984. Although support for the Conservatives increased in all population groups, they were as underrepresented among the young voters as they had been during the preceding decade.

The election system in all five Nordic countries is based on the principle of proportional representation. However, the actual practice differs significantly from one country to another. Denmark has without doubt the most complicated system of the five countries. As Ole Tonsgaard points out in an admirably daring quest into the intricacies of Danish electoral legislation, the conversion of votes into seats takes a multistage process in which no less than five different distribution rules are used (the largest remainder, d'Hondt's rule, Sainte-Lagué's rule, a modified Sainte-Lagué and a homespun technique called the '1-4-7' method). A bewildering set of circumscriptions (hovedområder, amts- og storkredse and opstillingskredse), plus the 20 or so possible combinations of list types that are available to the parties, would be confusing to any political science student — not to mention the voters in general. In the next chapter Tonsgaard and Elklit deliberately move from descriptive to normative efforts and suggest a revised election system, the greatest advantage of which would be intelligibility and transparency.

Other contributions to this book include a discussion on contending theories of electoral behaviour, a reassessment of data on class voting, an interpretation of women as an electoral group, data on the relationship between national and local elections, a comparison of the 14 national referenda held so far in Denmark, data on Danish attitudes toward the Common Market and electoral behaviour in the first two European elections held in 1979 and 1984. This review has only offered some brief samples from a rich and very readable volume.