Peter Gundelach/Karen Siune (eds.): From Voters to Participants. Political Institute of Political Science, University of Aarhus, 1992, 262 pp.

As is often the case with a Festschrift (the book is subtitled Essays in Honour of Ole Borre), this book covers a number of different topics, not always closely related to each other. It follows from this that a short review will not do justice to all the contributions and contributors. Nevertheless, the fifteen chapters (plus a bibliography Ole Borre's work) may tentatively be assembled under the following headings: (1) elections and voters, (2) -new politics" and (3) the mass media.

As befitting a book in honour of Denmark's "Mr. election study", the grand seigneur of not only American, but also international, electoral research. Professor Warren E. Miller has written the first chapter: "Changing Generations: Some Consequences of Compositional Change in the American Electorate". Party identification not only been one of the most important explanatory variables in American electoral research, but it has also been one of the most controversial concepts, especially in a European setting. In the US the decline of party identification been seen as a confirmation of an on-going process of dealignment from traditional parties and politics. Miller explains, more than defends, the original concept of party identification by pointing to three very different patterns of change in the American electorate: one affecting Blacks, one affecting white Southerners and one affecting non-Southern, non-Black citizens. Miller emphasizes the fact that the post-New Deal generation of non-Southern. non-Black voters was "almost single-handedly responsible for the increased number of citizens with no party identification in the 1970s . At the same time there has been an increase in the proportion of party identifiers in this group since 1976. Miller therefore argues that the pessimistic projection of a fundamental change in the American party system is out of tune with more recent developments. Although confined in space. Miller's chapter gives an informative picture of one of the most hotly debated topics in American electoral research.

Soren Holmberg's "Undermining of a Stable Party System" represents a followup Miller's chapter, with its emphasis on the decrease in party identification in Sweden. Typical of the way party identification has been treated in European studies, Holmberg concentrates on the strength of identification rather than the directional aspect that has been integral in the American studies. In a Nordic perspective, Holmberg shows that while Swedish voters had the highest level of party identification in the mid-19605, the situation is exactly the opposite in 1991. Fewer Swedes identify with a party than do Norwegians and Danes. This development with an increase in political distrust in the same period. Although Holmberg mentions '"sociological macro explanations", his main explanatory variables found within the political realm. More specifically. Holmberg emphasizes an increasing distance between voters and parties on the left-right divide on one hand, and decreasing agreement between voters and parties on political issues on the

other, as "one of the explanatory factors" behind the decline of party identification in Sweden. This conclusion is definitely in accordance with Holmberg's earlier thesis on the primacy of politics and the impotence of sociology. A critical note from this reviewer is that Holmberg runs the risk of explaining a subjective political phenomenon (party identification) with another subjective political phenomenon (viz. ideological affinity and policy evaluations). In other words, explaining politics with politics. This, however, does not prevent Holmberg's chapter from giving valuable insight into important trends with respect to the voter-party bond in modern societies.

Hans Jørgen Nilsen's "Basic Questions" and "Utopian Goal" leave the theme of party identification and focus on political trust, or rather distrust. Reporting from a comparative study including both mass surveys and in-depth interviews of the same respondents, he confronts what he coins the most basic criticism of Danish election studies: not having asked voters directly whether they trust politicians or not. The in-depth interviews revealed that the answers gathered by telephone did not present the respondents' real opinions accurately, although the overall picture was much the same. The substantial content of the study showed that the level of trust, as well as of distrust, was not high (below 50 percent). People indicating trust in politicians and the political system, did not shy away from voicing critical remarks, much in the same way as the more distrustful did not distrust all aspects of politics. Nielsen's conclusions are hardly surprising to the political science community: political trust is a very complex phenomenon that needs careful operationalization and analysis. Criticism of politicians does not necessarily mean rejection of the political system as such. But the interesting thing with Nielsen's analysis is that this "wisdom" also applies to a system that has seen major political turbulence in the last 20 years: emergence of an "anti-system" party (Glistrup), numerous changes of government and increasing individual distrust. Although Nilsen refers to some international studies of political trust, I would like to have seen more systematic references to the huge number of studies that have dealt with this subject in other countries. Not the least because this book addresses an international audience.

Søren Risbjerg Thomsen changes the subject once again and turns to local elections. He argues that local politics has been left too much in the shadow of national politics, and deserves more attention. Because local and national elections in Denmark are not held at the same time. Risbjerg Thomsen "invents" an expected national election by linear interpolation of the national elections before and after the local elections. By examining election results and turnout rates in different types of elections, the empirical patterns seen to be inconclusive. It is hardly surprising that "national politics has a considerable impact on local politics, but also that 'local politics matters'".

Max Kaase's chapter on direct political participation in the EC countries in the Late Eighties changes the topic to "New Politics". The analysis and conclusions in this chapter represent a continuation of the famous Political Action volume, of which Kaase was a principal contributor. New data affirm that "legal forms of direct-action politics have been firmly established as an element in the political repertoire of citizens, both in attitudinal acceptance and in real action". This statement is not very controversial in most countries today. But Kaase goes on to analyse the structural and ideological "determinants' of various operationalizations of civil disobedience. Despite a multitude of multivariate regression tables, I am surprised to find no reference whatsoever to statistical significance. Even comments that "the signs are in the expected direction" are of little interest if the coefficients are not significant.

Jørgen Goul Andersen's chapter gives an interesting overview of the main

questions in the Decline of Class debate, in which Goul Andersen himself has been an active participant. Although one can see major changes in old patterns between occupational position (class) and voting preferences, Goul Andersen's argument is that there has been a change in class voting, more than that class voting as such has disappeared. The public-private divide is one candidate for a new class cleavage, but whether it constitutes an enduring cleavage is open to question, according to Goul Andersen. Although he admits that part of his empirical analysis is based on few observations and quite often is not statistically significant. Goul Andersen's chapter presents an intellectually stimulating perspective. He does, however, indicate "culture" and "structure" are competing explanations with respect to the political impact of occupational class, a position to which Stein Rokkan may have raised an eyebrow in disapproval.

Palle Svensson and Lise Togeby pursue the theme of new social and political classes in post-industrial society. They approach the question of a "new class" by taking into account three criteria: education, occupation and sector. By combining these variables the authors distinguish between "new middle class" (well-educated professionals in the public sector), "yuppies" (well-educated professionals in the private sector) and "plebs" (low education, no attachment to the labour market). Analysing a sample of persons between 19 and 37 years of age, Svensson and Togeby interpret the data as supporting the distinction between the aforementioned three groups. Even though they find clear interaction effects between the explanatory they claim that interaction is not "vital in this respect. Taking into account the complex nature of all three explanatory variables, this reviewer feels that "disentangling" the effects may indeed be vital to the thrust of the article, namely the argument that post-industrial development really has resulted in new social classes. One important question left unanswered by Svensson and Togeby is to what extent the new "class has developed any kind of sub|ective class identify (consciousness).

Ole Tonsgaard concludes the section on voters and elections, although stretching elections to include referendums. He develops a theoretical model of referendum behaviour, building upon Fishbein's well-known attitude model. In this model it is important to distinguish between attitudes to a specific object and evaluations of relevant consequences. This may create problems, since it is often difficult, if not out-right impossible, to obtain measures of evaluations that are independent of the respondents attitude to the object. Even though Tonsgaard's attempt to apply this model on Danish referendums on the EC is interesting, his analysis raises doubts as to the appropriateness of treating evaluations as "causes" of attitudes toward the

Karen Siune's chapter on the agenda-setting function of the mass-media does perhaps not present breathtaking news to most readers, but it gives a thorough discussion of modern research in this area. Concentration of coverage is still pivotal: the more concentrated the media coverage, the more increased salience of a given issue for a greater number of people.

Marit Bakke addresses the political implications of changes in media structure. Her point of departure is the "demonopolization" of Norwegian radio and television in the 1980s. Despite the increasing number of TV-channels, the "Dallasitication" of TV programmes does not guarantee a real pluralism and choice for the individual consumer.

Peter Gundelach has compared attitudes to abortion in Denmark and Norway. His findings indicate that the effect of the Women's Movement has been smaller in Denmark than in Norway, perhaps because the right to legal abortion from the beginning was less controversial in Denmark than in Norway. Diffusion theory does

not present a good explanation of changing attitudes to abortion in either country.
It is noteworthy that there are indicators of increasing critical views on abortion,
and increasing emphasis on family values, especially in Denmark.

Nils Mortensen's chapter on "Future Norms" also finds ""the revival of conventional
in the family" intriguing. This implies that individuals can change
values and norms after their formative years.

Chitra K. Tiwari and Sushil Ray Pandey present an interesting glimpse of election
research in the Third World, dealing with the first democratic elections in Nepal,
a country where Danish social scientists have played a major role.

The last chapter, but definitely not the least, is Jørgen Elklit's comparison of the Danish electoral system between 1915 and 1920 with as adverse systems as the German system of 1949, the Bulgarian and the Hungarian systems of 1990. Elklit shows how decisive the institutional framework is for the outcome of elections. This reminder is particularly appropriate after a number of behaviour-oriented studies presented earlier in the book. In this perspective Elklit's chapter is a worthy Finale.

Although From Voters to Participants contains a number of disparate contributions, scope of the volume is perhaps its major virtue. In today's world of ever increasing specialization, even within social science, a volume of this kind deserves many readers.