Among the European democracies Denmark seems to be one of the most dull countries to deal with for an empirically oriented student of voting behavior. Apparently the Danish political system lacks most of those characteristics that form the point of departure for many modern research workers, i. e. conflicts, cleavages, and instabilities.

Homogeneity characterizes the Danish electorate. No religious, ethnic, regional, or other types of significant subcultures exist, which might threaten the maintenance of the political system or at least produce conflicts and tensions among the voters. Apart maybe from a few rudiments in the Labor movement Danish society can in no significant way be characterized as a "segmented" society. The only cleavages seem to be those of the traditional socio-economic character, but although most of the parties have their strongholds in different social categories, there is a pronounced overlapping; every party tries to make appeals to a broad section of the electorate in order to expand its share. Traditional differences between urban and rural Denmark are diminishing, as industry is decentralized, and as the development of the communications makes small differences still smaller. Particularly the appearance of television has revolutionized the climate of Danish politics; the electoral campaign is now dominantly a national contest, and regional problems are practically banished.

The unique Danish electoral system gives the parties a fair share of the seats in the Folketing, and as new parties do not meet insuperable obstacles in getting represented, it is possible for the system to adjust itself and to channel new political movements into the chore of the system. Several of the smaller parties have lost their seats and have been replaced by other small parties in the post-war period, without any resulting significant changes in the composition and method of work of the political, authorities.

The minority status of most Danish governments has made collaboration among several parties a necessity from the very first day in the life of a government, and collaboration has made it difficult to establish platforms for the electoral campaigns. The Danish parties are not without reason often described as becoming more and more similar in their outlook.

The Social Democratic Party on the one hand and the Liberals and the Conservatives on the other hand have been considered the only possible holders of government during the postwar period. Their respective shares of the electorate have constantly been about 30—35 %. This stable balance has seemingly been cemented as the distributions of the votes in the various regions of the country gradually have become more and more alike.

None of the two constellations have been able to rise above these magic numbers; they have
had to form minority governments or majority coalitions, which have not been enduring, because
the smaller parties in the coalitions have been worn out.1

No wonder that this balance in the Danish electorate has attracted the students of voting behavior. An explanation would provide political scientists with a clue to many problems. And as politicans often ask the question how long time this deadlock will continue, these studies will be given much attention.

Recently three studies have emerged, all of them dealing with the problem of the electoral
balance and the problem of stability in the electorate.

One of them, Aksel Lassen's Vælgere på Vandring, is written by a leading Danish political
journalist. His aim has been to investigate not the Danish case alone, but also the partly
similar Norwegian and Swedish cases.

A fundamental assumption in Aksel Lassen's analysis is his belief in the impossibility of changing partypreference between the two main constellations of parties. Once a voter has been put in one of the categories he will stay there, or eventually change for a while to one of the minor parties or refrain from voting. The stability is thus the given point of departure for Aksel Lassen, and his real problem has been to explain the small fluctuations occurring at every single election; variations in turn-out and movements to and from the small parties function as his clues.

The next problem in the analysis is the explanation of the tendencies to homogeneity
instead of regional variations, visible in every single Danish party. Aksel Lassen introduces

a single explanatory variable, the domestic movements of the Danish population, and as since, contrary to the findings of international electoral research, he assumes that moving citizens do not generally change party preference, he becomes able to account for the growing homogeneity by a reference to the movements between rural and urban Denmark, and between the Central and Greater Copenhagen.

The analysis is exclusively based upon the net results of Danish elections, and as most conclusions are about the individual voters or categories of voters, many of the arguments are patently false examples of what Johan Galtung has recently termed "the fallacy of the wrong level".²

In their Partistyrke og Social Struktur i 1960 Ole Borre & Jan Stehouwer have tried to lay the foundation for an analysis of the problem of the stable balance in the Danish electorate, following up and extending a previous study by Erik Høgh. They have divided the country into 9 regions, each of them consisting of a number of municipalities, ranging from 73 to 250. Dissimilarities in social structure and party strength in these regions are described; in contrast to Aksel Lassen the authors have tried to isolate, one by one, the effects of several ecological factors on relative party strength in order to estimate the extent to which it is possible to relate "political regionalism" to the specific social structure of the geographically determined regions. An examination of the distribution of the votes in the municipalities will show significant variations from the national average. In some small, well defined regions some of the parties have had strongholds for a long time. Borre & Stehouwer have desisted from an analysis of these specific regional variations: they are considered outcomes of historical traditions, and as such their place in the analysis is defined as a non-explained residuum, handed over to the historians or to other political scientists.

Examining the vote in the 9 regions the authors find some variations in party strength. The Social Democratic Party, the Radical Liberals, the Conservatives, the Socialist Peoples' Party, and the Communists have their greatest share in the Eastern parts of Denmark, while the Liberals, the Independents, and the Singletaxers have their strongholds in the Western parts. Copenhagen and Western Jutland are political as well as geographical poles, but a check on urbanization in the municipalities reveals that intraparty regional variations can to_a significant extent be attributed to a different degree of urbanization in the regions.

These findings are elaborated in a detailed study of the impact on party strength of several
other ecological variables, such as occupational structure, average taxable income, and population
growth in the 19505.

What emerges from this analysis is a map of party strength in Denmark, a verbal description that, although it only touches the central questions about stability and change, provides a starting point for this kind of investigation. As the authors are well aware of the limits of their method, they do not jump into the trap of the ecological fallacy, but as a result of their caution they have to pass over many questions to further research. The only common quality of the books by Lassen and Borre & Stehouwer thus seems to be that they furnish future Danish survey research with many questions and hints of working hypotheses.

The second volume in the series of ecological analysis of Danish voting behavior is a longitudinal study of elections in the 1960s (Stehouwer & Borre, 1969).³ Ecological and political data from the municipalities in 1960 and 1964 are combined with data from 1966 and 1968 from the constituencies, and the aim of the analysis has been nearly the same as that of Aksel Lassen: to describe the apparent stability in the jiet-results, expjain the_relatiyely small fluctuations from election to election, and account for the tendencies to intra-party homogenization visible in the post-war period.

The 1950s have been described as a period of ambivalence in Danish Politics:⁴ relatively low electoral participation was a characteristic of this period, as were modest fluctuations in the shares of the three great parties. The Communist vote declined steadily, while the relatively small gains and losses of the two other small parties, the Radical Liberals and the Singletaxers, had decisive influence on the formation of governments.

This ambivalence came to an end in the 1960 election. The turnout rose and reached a

peak in 1966 (88,6 %);^B two of the older small parties, the Communists and the Singletaxers, were replaced by two new parties, the Socialist People's Party and the Independents, in 1960, and again in 1966 there was a new replacement of the Independents by a new Liberal Center Party, formed around two former members of the Liberals.⁶

These apparently turbulent fluctuations in the 1960s are scrutinized by Jan Stehouwer & Ole Borre, and it appears that a great many of them can be seen as continuations of tendencies already existing in the previous decade. The stable balance between the three .great parties has continued, although a close examination of the results from the last elections shows that the fates of.Ae. parties have been very differentJn various types of municipalities: The Social Democratic share has decreased in the urban parts but increased in rural Denmark, while the Liberals, traditionally regarded as an agrarian party, have increased their share of the votes in the cities, at least until the appearance of the Liberal Center Party. The Conservatives improved their position in all types of municipalities in the 1960 and 1964 elections. These trends, which are visible back in the 19505, apparently were partly broken in the 1966 and the 1968 elections with their extraordinary high electoral participation, but nevertheless these trends can in all probability be interpreted as, signalling profound changes in the Danish electorate.

The authors have also tried to account for the recent fluctuations in the share of the parties on the right and the left wings, primarily by an analysis of the co-variation of turnout and the strength of these parties. The ecological method cannot give any kind of proof, and the authors have cautiously desisted from causal interpretations. Their conclusion is that the primary advantage of the ecological method in longitudinal analysis is that it makes it possible to discern otherwise invisible long term tendencies in the electorate, and separate these tendencies from the_shoj± Jean fluctuations caused by, the specific political situation.

Apart from Aksel Lassen, who in addition to his analysis of the Danish elections has tried to give parallel descriptions of the post-war elections in Norway and Sweden, the three abovementioned statistical-ecological studies have in common that the researchers have focused theii whole attention upon the Danish problems, particularly the problem of stability, and the problem of converging regionalism.

From an international point of view it might be a disadvantage that this strong national focus generates studies which, although two of them are technically very much up to date, are only to a minor degree theoretically and comparative oriented. On the other hand, at least the two studies from the data project at the University of Aarhus undoubtedly give a very clear picture of the Danish electorate. When these studies are extended backwards in the century as it is planned (Stehouwer, 1967), a firm basis for further electoral research will exist in Denmark. When they become elements in cross-national comparisons their theoretical value will increase tremendously.

Mogens
University of Aarhus

---

NOTES

1 It is too early to comment on the majority coalition, formed after the 1968-election. The massive victory of the Radical Liberals gave them a strong and entirely new position in Danish post-war politics, and the lead in the three-party government.

2 1967. A detailedcan be found in. 595-614.

3 This study also contains some comments to the 1968-election.

4 Stehouwer, 1967, pp. 94-116. See especially p. 100.

5 In the 1968 election the participation was still higher (89.3 %). Only in the 1943 election during the German occupation has Denmark experienced a turnout that can be compared with the turnouts of the latest election (1943: 89.5 %).

6 See further articles by Curt Sørensen, Ingemar Glans, and Mogens N. Pedersen in Scandinavian Political Studies, 2 (1967). In the 1968 election the Liberal Center Party was replaced as the smallest Danish party by the new Left Socialist Party, which, by a very small margin, obtained 4 seats.

REFERENCES

Borre, Ole & Stehouwer, Jan. Partistyrke og social struktur i 1960 (Party strength and
social structure in 1960), with English summary, Aarhus, 1968.

Galtung, Johan, Theory and Methods of Social Research, Oslo, Universitetsforlaget. 1967.

Høgh, Erik. "En økologisk analyse af folketingsvalget 1950" (An ecological analysis of the
national elections in 1950), with English summary, Sociologiske Meddelelser, 4. serie, 1958.

Lassen, Aksel. Vælgere på vandring. Nordiske valg 1947-66 (Floating voters. Movements
between political parties and constituencies in the Northern countries 1947—66), with English
summary, Aarhus, Skrifter udgivet af Jysk Selskab for Historie, 1967.

Pedersen Mogens N. Valgforklaringer, Historie, Ny Række VII, 4, (1967).

Stehouwer, Jan. Long term ecological analysis of electoral statistics in Denmark, Scandinavian
Political Studies, 2, (1967).

Stehouwer, Jan & Borre, Ole. Fire Folketingsvalg 1960, 1964, 1966 og 1968 (Four National
elections 1960, 1964, 1966, and 1968), forthcoming, 1369. )IVO.