University of Copenhagen

In December 1973, the Danish party system suffered a dramatic setback. Instead of five parties the election returned ten parties to parliament. The percentage share of the poll of the five 'old' parties dropped from 93 % in 1971 to 65 % in 1973. In the main this trend was confirmed at the next election in January 1975. Once again ten parties got seats, and the vote for the five old parties rose only slightly to 70 %

These events have been frequently commented upon.¹ This article concentrates on the vote for the largest of the new parties, the Progress Party (Fremskridtspartiet). The following analysis is based on material from the Danish election studies in 1973 and 1975, and the purpose is to demonstrate the impact of attitudes towards the political system by voting behaviour in Denmark. It will be shown that both political distrust and authoritarian viewpoints are clearly related to the vote for this new party in both elections. In a wider perspective this has serious implications for the evaluation of Danish political culture.

The Progress Party was founded in the autumn of 1972. It soon had a breakthrough
in the opinion polls - on one occasion rising to 25 % - and collected
15.9 % of the votes in December 1973 and 13.6 % in January 1975.

This dramatic success marked a break with the Danish tradition of slow and undramatic movement between parties. Furthermore, the politics and the style of the party were innovative. The Progress Party aimed at abolishing income tax by drastic cuts in public expenditure. Opposition to state interference in the economy was underpinned by violent verbal attacks on public servants, who were accused of being unproductive and overpaid papershufflers. Other established institutions got their share of criticism too: The old parties were referred to as being senile. In addition, the Progress Party proposed reduction of Parliament membership from 179 to 40, and suggested that royal prerogatives be increased. Finally there were frequent attacks on intellectuals in general and left-wing intellectuals in particular, indicating some amount of anti-intellectualism and lack of general tolerance.

The explosive growth as well as the character of the party's programme is reminiscent of such right-wing populism as the Poujade movement in France and the Vennamo party in Finland. The party's strength makes it more than a picturesque outsider movement, and the stability of its support among voters suggests that the Party will be a lasting phenomenon in Danish politics.

1. The Problem of Ideological Dimensions

It is obvious that one of the more important tasks facing Danish electoral research is to explain the electoral success of the Progress Party. In view of the party's character, a central question concerns the role of ideological variables. Normally, Danish politics has been structured along a typical left-right dimension, and the Progress Party's platform points to a rightist position on this dimension.² But the party's performance also indicates another ideological dimension - scepticism towards established institutions - and it is necessary to raise the question of authoritarianism. The central problem is whether these ideological dimensions are relevant when the focus is shifted from the elite to the mass level of politics.

Two comparative investigations of voter attitudes have, in fact, pointed to a rather high level of political dissatisfaction in Denmark. Before the rise of the Progress Party, in 1972, a study carried out in the Nordic Countries (except Iceland) showed that a majority of Danish voters perceived strong conflicts between politicians and the ordinary man'. The percentage was higher in Finland and Sweden, but substantially lower in Norway.³ One year later, in 1973, an analysis in all EEC countries portrayed Denmark as a country with low socio-economic dissatisfaction, but at the same time with high dissatisfaction towards the political structure. Actually, the difference between the measure of political dissatisfaction and the measure of socio-economic dissatisfaction attained its maximum value for Danish voters.⁴ These studies hint at political discontent as a likely source of political instability in Denmark. More direct evidence that some of the latent possibilities have actually been realized might be drawn from a small-space analysis of sympathy for the different parties at the 1973-election: A two-dimensional solution was obtained, and one dimension was identified as the traditional left-right axis, the other as political distrust. The Progress Party was placed to the right, but what marked it off from the other bourgeois parties was a position high on the distrust axis.₅

Small-space analysis differs from other, more multivariate techniques, and party sympathy does not necessarily imply party vote. However, when the population in the 1973 study was divided simultaneously according to a left-right index and a distrust index, it turned out that also the percentage of Progress Party vote increased with high rightist scores and with distrust scores.⁶ The core of these results will form part of the following discussion, but it should be mentioned that the same conclusions were obtained in a study in April-May 1974 of intended party vote.⁷

These results suggest a restructuring of the political scene; in view of the Progress
Party's importance, the left-right dimension must be supplemented with

dimensions concerning attitudes towards the political system. There is a high
level of discontent, and this discontent has relevance for party choice.

But there are problems, too. In 1973 the Progress Party was a new party, and in general one might assume that criticism towards (old party) politicians would be more pronounced among voters choosing something new. In that case, the connection between Progress Party vote and discontent might well recede with time. Actually, an inspection of party sympathies (not vote) in 1975 has led to the suggestion that the two-dimensionality has decreased and approached the structure of the traditional left-right one-dimensionality.⁸ Furthermore, political distrust means quite different things, depending on which other attitudes it enters into combination with. Distrust + Democratic attitudes are not the same as Distrust + Authoritarian attitudes, and in a wider perspective these two combinations might have opposite policy consequences.

In the following, it will be argued that the problem of attitudes towards the
political system is a genuine and enduring problem. The dependent variable will

be party choice. The inclusion of 1975 data will be used to test the proposition that distrust should be regarded as a transitory phenomenon. In addition, authoritarian attitudes will be used to measure the relationship between Progress Party vote and attitudes towards the political system.

The 1973 study had a rather small sample - 533 persons of which 493 actually voted - and conclusions based solely on that material might in many respects be biased. Furthermore, a concentration on traits common to the 1973 - and the 1975 - election seems advisable in order to focus on the more central ideological variables.

The dimensions considered individually

At first glance the Progress Party voters hardly differ from the electorate as a whole. Table I contains the questions that have been used in all Danish election studies to measure left-right position, political distrust, and authoritarian attitudes.⁹ Each question had codes for agreement, disagreement, and neutral answers. In the table the responses 'completely agree', and 'partly agree' have been combined.

For the total electorate the table shows a high level of political distrust, and the same could be said with regard to 'authoritarian attitudes'. For both series of questions the maximum values were attained in 1973. The Progress Party voters deviate systematically towards the right, and towards more responses indicating distrust and authoritarianism. But, apart from a few questions, the differences are quite small.

For subsequent analyses, the questions under each heading in Table I have been used to make additive indexes. The indexes are divided into three brackets, each comprising as far as possible one third of the total population.¹⁰ The Progress Party's strength in the different brackets on each criterion is displayed in Table 11.

At both elections, the Progress Party vote is related as before to rightism,
authoritarianism, and distrust. But still the correlations are moderate.

The interplay between dimensions

With this background the decisive problem is how the strength of the Progress Party is related to different combinations of scores on the three dimensions. Using a priori arguments any pattern is equally likely. It turns out, however, that each variable remains important, even when the other variables are taken into account. In Table 111 distrust and authoritarianism have been combined; the same general pattern emerges in both elections.

For each level of authoritarianism one discerns decreasing percentages from high to low distrust. At the same time, there is a decrease for each level of distrust from high authoritarianism to low authoritarianism. In combination the two dimensions produce substantial variation from the low-low corner of the matrix to the high-high corner. Nevertheless, the distinction between high and medium authoritarianism does not reveal any systematic difference in the strength of the Progress Party.

The above pattern still emerges when the left-right dimension is introduced. To avoid building percentages upon too small enumerations, 'high' and 'medium' authoritarianism have been combined. Subsequently, the total population was crosstabulated simultaneously on all three dimensions - authoritarianism, distrust, and left-right position. The strength of the Progress Party in the different cells was compared along one dimension at a time, holding the other dimensions constant. Some distinctions between brackets turned out to be of low or no relevance. However, on all dimensions the more clearcut differences in ideological scores were coupled with clear differences in Progress Party strength. The resulting matrices, omitting the less relevant distinctions, are given in Table IV.

The impact of distrust and authoritarianism is much stronger among rightwingers than among the rest of the population. This non-additivity suggests an important conclusion: An above average chance of an ideological right-winger voting for the Progress Party is heavily dependent on attitudes towards the political system. If high or medium authoritarianism is combined with high or medium distrust the likelihood is substantially above average. But at the other extreme - which is not uncommon - where low authoritarianism goes together with low distrust, the chance approaches zero, even among right-wingers. Among centrists and left-wingers the upper limit of the variation is much lower, but again the percentages are clearly dependent on attitudes towards the political system. In general, all variables remain important.

2. The Progress Party Electorate and the Electorates of Other Parties

The relevance of all three dimensions can also be illustrated by comparing Progress
Party voters with voters of other specific party groupings. For that purpose
the following categories have been used:

(1) Left-wing Parties (Socialist People's Party, Left Socialists, and Communists'
(2) Social democrats

(3) Old Bourgeois Parties (Radicals, Agrarian Liberals, and Conservatives
(4) New Bourgeois Parties minus The Progress Party

(5) The Progress Part}

The voters of each party grouping were distributed on the three brackets o:
each dimension, and the leanings of each party electorate were summarized b>

subtracting the percentage falling in the third bracket (left-wing, low authoritarianism,
and low distrust) from the percentage falling in the first bracket (rightwing,
high authoritarianism, and high distrust). The results are given in Table V.

As might be expected, the left-right dimension discriminates between the voters of left-wing parties and Social-Democrats on the one hand and the different groups of bourgeois party voters on the other hand, but there is no indication that Progress Party voters on the average are more to the right than other bourgeois voters - rather the contrary.

Attitudes towards the political system, however, make a difference. Compared with the voters of the old bourgeois parties, Progress Party voters are at both elections more prone to give responses indicating distrust and authoritarianism. Even more revealing is the comparison with the voters of the other new parties. In 1973 this latter group scored high on political distrust, but this tendency had disappeared in 1975. This change fits in very well with the common-sense expectation mentioned above that voters choosing a new party are likely to be discontent with old party politicians. However, among Progressive Party voters distrust did not decrease but increased from 1973 to 1975; thus the results suggest that the distrust of the Progress Party electorate is deep-rooted. Finally distrust is at about the same level on the left wing of the party system as it is among Progress Party voters. But the scores on authoritarianism makes a crucial difference. In sum, it is the combination of different ideological dimensions rather than any single dimension which marks the Progress Party voters off from the voters of the other parties, and hence it is necessary to take attitudes towards the political system into account.

In the final analysis, other variables have to be taken into account. It is unthinkable
that ideological variables could give a sufficient - and, least of all, a total

explanation of voting behaviour. Furthermore there is evidence that the impact

of distrust on Progress Party strength is due mainly - in 1975 only - to the distrust of the middle classes and not of the working class population.¹¹ However, from the present study a definite picture seems to emerge. The three variables - left-right position, distrust, and authoritarianism - are all related to the Progress Party vote and none of them were eliminated.

There has been a left-right dimension in Danish politics for a long time, and therefore the point most worth stressing is the two attitudes towards the political system. The Progress Party itself might have strengthened distrust and authoritarianism, but it did not create these attitudes, because the levels were about as high in 1971 as in 1975. Rather, the party was established on fertile soil.

What stands out above all is that former pictures of Danish politics have to be profoundly revised. In view of the Progress Party's strength, the left-right dimension is not sufficient. Furthermore, the incorporation of Denmark among those nations with harmonious civic cultures seems dubious. There has been a marked instability in the party system, combined with a high amount of distrust and authoritarianism, and these attitudes are related to the party most clearly associated with the upheavals in Danish politics.

---

NOTES

1. In this yearbook the following articles describe the development of the Danish party system: Erik Damgaard, 'Stability and Change in the Danish Party System over Half a Century', Scandinavian Political Studies (Oslo: Universitetsforlaget, 1974), 9/1974, Ole Borre, 'Denmarks Protest Election 1973', Scandinavian Political Studies (Oslo: Universitetsforlaget, 1974), 9/1974 and Ole Borre, 'The General Election in Denmark, January 1975: Toward a New Structure of the Party System* Scandinavian Political Studies (Oslo: Universitetsforlaget, 1975), 10/1975.

2. An analysis of the voting pattern in the parliament has also placed the Progress Party to the right. See Erik Damgaard, 'Folketingsvalget og det nye partisystem' (The General Election and the New Party System), in Per Bendix et al., Decembervalget 1973 (Copenhagen: Schultz, 1974).

3. Erik Allardt, 'About Dimensions of Welfare, An Exploratory Analysis of a Comparative Survey', Helsinki, Research Group for Comparative Sociology, University of Helsinki, 1971, p. 99.

4. J. R. Rabier and David H. Handley, 'Difference et Differenciations Interregionales dans Les Attitudes et Comportements du Public'. ECPR Joint Sessions of Workshops, London 1975 (mimeo) p. 42.

5. J. G. Rusk and Ole Borre, 'The Changing Party Space in Danish Voter Perceptions, 1971-1973', European Journal of Political Research, Vol. 2. 1974.

6. Hans Jørgen Nielsen, 'Politiske Meninger og Vælgeradfærd' (Political Attitudes and Voting Behaviour) in O. Borre et al., Vælgerskreddet 1971-1973 (Aarhus and Copenhagen, 1974).

7. Ingemar Glans, 'En liberal Reaktion - Fremskridtspartiet och dess Valjargrundlag' (A Liberal Reaction - The Progress Party and Its Voters), Aarhus, Institute for Political Science, 1975, (mimeo) p. 50.

8. Ole Borre, The General Election in Denmark, January 1975', op.cit.

9. A more detailed analysis of questions not used in all studies may be found in Hans Jørgen

---

Nielsen, 'Fremskridtspartiet - Et Højreorienteret Protestparti for hvem?' (The Progress Party - A Right-wing Protest Party for Which Groups?), Copenhagen, Institute for Political Studies, 1975 (mimeo). Because 'authoritarianism' is used in a number of ways in the literature, it should be emphasized that the two questions raised in this article concern agreement with authoritarian policies. Nothing is implied concerning the more general (authoritarian) personalities of the voters.

10. Differences between the 1973 and the 1975 distributions make this approximation difficult. It has been preferred, however, to have fixed division-points for the brackets, even at the cost of additional deviations from an equal distribution of the populations on the three brackets. In constructing the indices the responses were coded into three categories: 'agree', 'disagree' and 'neutral'. A 0-1-2 scoring was used in such a way that 0 indicates 'right-wing ideology', distrust and authoritarianism, and 2 the opposite possibilities. The bracketing was done as follows: Left-right Index Distrust Index Authoritarianism Right = 0-2 High distrust: 0-1 High Authori = 3-4 Medium distrust: 2 tarianism: 0-1 Left = 5-6 Low distrust: 5-8 Medium Authoritarianism: 2 Low Authoritarianism: 3-4

11. Hans Jørgen Nielsen, op. cit.