The ideological convergence of political parties in the Western world is a perennial issue in political science as well as in general political debate. It seems to be generally accepted that parties increasingly resemble each other in terms of ideological content and mass appeal (Budge & Farlie 1985, 304-305; Helenius 1969, 27-136; Panebianco 1988, 262-274). Moreover, causes of this development are usually depicted in a rather uniform manner. Increased standards of living, universal access to higher education and a general structural transformation of society have reduced, and in many instances even obliterated those cleavages on which the ideological distinctiveness of political parties was based. Political parties have increasingly developed into catch-all electoral machines (Kirchheimer 1966, 190; Dittrich 1985, 258-259).

All the same, warnings have been issued against premature and exaggerated about the demise of ideological differences among Western political parties. The radical wave of the late 1960s clearly entailed a certain re-ideologization of party politics. Analysts like Klaus von Beyme. while not denying the long-term tendency towards convergence, have found it necessary to underline that it is 'too early yet to proclaim the end of parties' (1985, 364, 371). The de-ideologization of political parties, in short, is not a simple linear process.

Nor can the process of de-ideologization be a simple one within individual

Resumé

This study addresses the hypothesis concerning the ideological convergence of political parties in light of election campaigns in Finland. The basic expectation is that the parties have become more alike in terms of their orientation vis-å-vis the electorate and in their use of propaganda techniques. The empirical data consists of editorials in the leading newspapers of the four largest Finnish parties in connection with the parliamentary election campaigns of 1954, 1966. 1975 and 1987. The main hypothesis is clearly corroborated, as traditional elements such as class orientation have been replaced by references to the nation at large and to the political elite. A re-ideologization could be noted in connection with the 1975 campaign, followed by a strong de-ideologization in 1987. Despite the virtual disappearance of offensive propaganda techniques, recent campaigns were not found to be decisively less 'propagandists" than the earlier ones.

parties. The rationale behind a less ideologically distinct posture is to attract new groups of voters. But such shifts entail the danger of alienating parts of the previous hard-core constituency of the party; for these people, deideologizationmay an unacceptable deviation from those very principles that led them to rally behind the party in the first place. Hence, the general convergence of the major parties may make room for new, more ideologically pronounced political alternatives (cf. Przeworski 1987, 26-29). The most probable strategy on the part of the parties is to try to have their cake and eat it too. In other words, they can be expected to continue to appeal to their hard-core constituencies while at the same time adding elements intended to attract new strata of voters (Siune 1982, 264-265).How such a mixture is in the eyes of new as well as old voters is, of course, open to test.

Despite the fundamental nature of these questions, the literature offers
relatively little to go on in terms of systematic empirical analysis. According
to a recent assessment

Beyond Kirchheimer, precious little research has appeared delineating trends in program and ideology, despite the burgeoning literature on transformation of social cleavages and values. Possibly, the greater difficulties involved in gathering data and in measuring the inherently soft concepts in this area have contributed to this lacuna. Serious questions suggest themselves. Foremost is the extent to which new or traditional parties have presented programs for or based their appeal on post-bourgeois concerns (Mayer 1989, 160).

This article focuses on the Finnish party system since the mid-19505. Its objective is to inquire whether and to what extent the hypothesis concerning the ideological convergence of political parties is manifested in the electoral campaigns of the four major parties. The Finnish party system - as indeed party systems throughout Scandinavia - is particularly interesting due to the traditionally strong electoral links between class and party. At the same time, the structural transformation of Finnish society has been extremely rapid in the postwar era, thus thoroughly affecting the very foundation of party alignments. Consequently, one may expect to find considerable tension in Finnish parties between the demands of ideological orthodoxy on the one hand and the need to amend their mass appeal on the other.

Structural Change in Finland, 1940-1985

The prolonged predominance of the primary sector sets Finland apart from the rest of Western Europe and indeed even from many East European countries. Up until the end of the 19505, agriculture was the most important means of livelihood in Finland. From that point on, in a matter of a couple

of decades, Finland leaped to the post-industrial structure characteristic of
modern West Europe (Table 1).

The process of modernization in Finland, particularly dramatic in the 1960s and early 19705, never resulted in a predominantly industrial society. Instead, Finland passed directly from a mainly agricultural into a "service economy. Indeed, as of the early 19605, the share of the tertiary sector exceeded those of the primary and secondary sectors, and this difference has grown rapidly since then (Table 2). Thus, those who left the country for the cities were not to form a homogeneous industrial working class.

Finland, in other words, has joined the bandwagon of societies dominated the 'new middle strata'. The rapidity of the structural transformation Finnish society, however, is a feature that largely separates Finland from the rest of Western Europe.

Parties, Classes and the Press

Given the many dramatic features of modern Finnish history (cf. Kirby

1979), the stability of the party system is rather remarkable. In the parliamentaryelections 1922, the four major parties - the Conservatives, the Center Party (until 1965 the Agrarian Party), the Social Democrats, and the Communists - polled 78.3 percent of the total vote. Some six decades later, in 1983, their share was 79.9 percent. True, there has been some variation in their relative strength. One can note, for instance, the decline of the Communists and the increased support of the Conservatives during the last two decades. Still, the striking feature is the stability rather than the change.

All in all, the link between class and party in Finland has declined slightly over time, but it is still reasonably clear, as is apparent from Table 3. As this table suggests, the Conservatives have become increasingly the party of the urban middle and upper strata; they have lost much of their former support of the well-to-do farmers of Southern and Western Finland. The Center Party, on the other hand, has managed to make considerable inroads into both the working class and the middle strata, thus compensating for much of the decline in the share of the farmers in the electorate at large. In fact, this party has the most diversified social basis of all four parties today. The Social Democrats still depend on the workers for approximately two-thirds of their total support. Nevertheless, the party has managed to gain considerable middle-class support during the past two decades. The Communists, by contrast, still seem to be as proletarian as they were during the immediate postwar period. Given the facts at hand, one may say that the radicalization of the late sixties and early seventies did not result in the kind of 'white collar radicalism' which has been noticeable elsewhere, especially in Denmark (cf. Borre 1987, 350).

Concerning political parties and the press, two things should be noted in an international comparison. Both are characteristic of Scandinavia as a whole rather than of Finland alone. First, Scandinavians are newspaper readers par excellence; literacy has long been higher than in any other

world region, and the number and circulation of newspapers uniquely high (Lindblad et al. 1984, 118-120). In other words, the potential political influence of the press is tremendous in Scandinavia. Second, party-affiliated newspapers have had a remarkably strong position. Here, however, a considerable change has occurred in the postwar era. Thus, while approximatelytwo-thirds the daily newspapers circulated in Finland in 1946 were party organs, the figure was around 40 percent at the beginning of the 1980s (Nousiainen 1985, 128). The general process of concentration typical of the press in the West has hit the party press particularly hard in Finland. The socialist parties today control about 10 percent of the total daily circulation, whereas the shares of the conservative and center parties are slightly higher. Despite this decline over time, the party press continues to be one of the chief vehicles for electoral campaigning.

In sum, the basis of Finnish party politics has evolved in a direction quite similar to that found elsewhere in Western Europe. The socio-economic transformation of society has produced large middle strata which are difficult to place in a traditional class analytic framework. Parallel to this, the class dependence of the major parties has declined somewhat, but it continues to be higher than one might expect on the basis of the change in the general class structure. Similarly, the Finnish party press has declined in importance as compared to politically non-affiliated newspapers. This too, however, is far from a quantité negligeable in Finnish politics.

Hypotheses

The overall development of Finnish electoral campaigns is expected to comply with the de-ideologization thesis. In other words, the parties are expected to increasingly resemble each other concerning campaign style, and the content of their campaign messages is expected to show evidence of losing its ideological edge.

In operational terms, this study has two foci. On the one hand, the convergence thesis would predict changes concerning whom the parties wish to address through their campaigns. On the other hand, how they do this is likely to reflect the same forces of change. As to the first dimension, a modified version of Olavi Borg's conceptualization of levels of group orientation (1964, 162-184) is applied in this study. The framework used distinguishes between eight levels of orientation:

(1) The individual level: "the individual', 'the citizen". "the voter":
(2) Classes: 'workers', 'farmers", "functionaries", 'the middle class';

(3) Quasi-classes: 'high. medium and low income earners". 'wage and salary
earners", 'employees";

(4) Non-classes: "families with children", 'pensioners", 'savers" etc;

(5) Demographic groups: 'young people', 'senior citizens', 'women';

(6) Geographic groups: 'urban people', 'country people', inhabitants in
given regions;

(7) The national level: 'the nation', 'the Finnish people', 'the Finns', 'the
electorate';

(8) Other: groups not belonging to any of the above, the Government or
its institutions, governments or groups abroad, individual politicians
etc.

For analysis of the 'how'-dimension, a study from the very beginning of modern propaganda analysis was revisited. In a classical study of radio propaganda, A. M. Lee and E. B. Lee (1939) presented a number of propaganda techniques, which they call the propagandist's 'tricks of the trade'. These are instruments of linguistic manipulation intended to persuade audience without its being aware of the fact of manipulation. After initial applications, the technique called 'Transfer' was omitted from the present study. Consequently, the way in which Finnish parties seek to persuade the electorate through their press is studied in terms of the following 'tricks of the trade':

(1) Name calling: 'giving an idea a bad label - is used to make us reject
and condemn an idea without examining the evidence';

(2) Testimonial: 'consists in having some respected or hated person say that
a given idea or program or product or person is good or bad';

(3) Card stacking: 'involves the selection and use of facts or falsehoods, illustrations or distractions, and logical or illogical statements in order to give the best or the worst possible case for an idea, program, person, or product';

(4) Bandwagon: 'has as its theme, everybody - at least all of us - is doing it; with it, the propagandist attempts to convince us that all members of a group to which we belong are accepting his program and that we must therefore follow our crowd and jump on the bandwagon';

(5) Glittering generality: 'associating something with a "virtue word" - is
used to make us accept and approve the thing without examining the
evidence';

(6) Plain folks: 'is the method by which a speaker attempts to convince his
audience that he and his ideas are good because they are "of the
people", the "plain folks'" (ibid., 23-24).

In addition to an increasing similarity between the parties with respect
to their campaign profiles, the following trends are expected in terms of
the group orientation and propaganda techniques employed:
-

- references to social classes proper have declined in favor of quasi-classes,
non-classes and demographic groups;

- the territorial dimension, including the urban-rural distinction, has
declined in importance;

- references to the individual level have increased;

- it has become increasingly common to appeal to the nation as a whole;
- references to 'other' categories have increased, including references to
individual politicians and party leaders;

- the use of propaganda techniques has generally declined somewhat;
- 'name calling' and 'card stacking' are used less frequently than earlier:
- 'glittering generality', 'plain folks', 'bandwagon' as well as 'testimonial'
techniques have a more prominent position today than earlier, at least
in relative terms.

In sum, the expectation is that class orientation and corresponding 'divisive' propaganda techniques have increasingly been replaced by a stronger emphasis on the individual, both in terms of the individual citizen and voter and in terms of individual politicians as persons. At the same time, 'commonness' is emphasized at the collective level with stronger appeals to the 'nation' and the 'people' as a whole, accompanied by the increased use of propaganda techniques appealing to that which unites rather than separates the citizens. These expectations regarding changes over time are summarized in Table 4.

The above model is, of course, based on ideal types used for the purpose of clarity. The changes we expect to find concern degree more than kind; none of the 'orientations' or propaganda techniques listed above can be expected to vanish from political campaigns totally. Moreover, if it is true that the late 1960s entailed a re-ideologization of West European politics, the differences between the 1950s and the middle of the period studied can be expected to be smaller than those between 1975 and 1987. In fact, the

late 1960s and early 1970s may indeed turn out to be the very heyday of
ideological campaigning.

Finally, the four parties have different 'starting points' concerning several of our categories. For instance, conservatism is, unlike leftist party programs, traditionally associated with an emphasis on the nation. Hence, the Conservatives' 'need' to increase this campaign element may not be as great as that of the other parties. Similarly, for instance, the necessity to make inroads into the urban electorate has been much more exacting for the Center Party than for the other parties. Consequently, one may expect to find a much clearer change in centrist propaganda concerning the 'geographical level'. A list of these kinds of special hypotheses could probably be made much longer. This study contents itself, however, with merely pointing at such possible deviations from the overall trend of convergence predicted by our general hypothesis.

Empirical Analysis

Data

The empirical material of this study consists of the editorials in the leading organs of the four parties during a two-week period preceding each election. The papers included are, for the Communists, Kansan Uutiset (The People's News', formerly Vapaa Sana, 'Free Speech'), for the Social Democrats, Sosialidemokraatti ('The Finnish Social Democrat'), for the Center Party, Suomenmaa ('The Country of Finland', formerly Maakansa, 'Country People') and for the Conservatives, Aamulehti ('Morning News'). Below, party labels will be substituted for the names of the papers.

The total number of editorials studied was 501. Both the number and size of the editorials has varied to some extent over time. Especially notable is the increased use of short comments jointly with longer editorials. This trend has mainly contributed to the increased number of articles in the Conservative paper. All in all, the four newspapers carry a sufficient number of editorials before each election to permit a comparison (Table 5).

Method

The research problem calls for a content-analytical approach. While resting on qualitative judgments regarding both definitions of the variables and operative classification of the empirical material, the method used aims at presenting the results in a quantified form. That is to say that the relative occurrence of the various levels of orientation and propaganda techniques is to be presented numerically. The unit of analysis is the individual party

in each election. The aim is, consequently, to ascribe variable values to each party at four different points in time: for instance, for the Communists, the share of 'class orientation' at time 1 is x and the relative frequency of 'name calling' is y.

As for the recording unit, the two empirical foci of the study called for different solutions. After initial applications it was found that it was possible and therefore rational to use entire editorials as recording units with respect to the level of orientation. In other words, each editorial was ascribed to one category of orientation, and one category only. As to the use of propaganda techniques, the recording unit had to be smaller since various techniques could be used in one and the same editorial. This meant that the occurrence of these techniques had to be recorded separately for each sentence.

The value of any content analysis largely depends on the reliability of the coding procedures. In this study, the coding rested entirely on qualitative judgments, i.e. the coder's interpretation of the categories used and of the substance of the editorial texts. Consequently, it was necessary to check the reliability of the classifications by recoding a considerable portion of the material. An intersubjective reliability test was carried out so that another person recoded a random sample of 10 percent (n = 51) of the editorials. Concerning level of orientation, 78 percent of the codings were identical at the two instances. The figure suggests that the categories used are by no means self-explanatory. Still, the level of reliability can be considered sufficiently high (Pietilä 1973, 242-244). For the six propaganda techniques studied, separate reliability tests were conducted. Reliability ranged from 87 to 94 percent, which can be characterized as entirely satisfactory. For examples of individual codings, see Appendix 1.

Findings

Level of orientation. In accordance with the convergence hypothesis, the

Finnish parties have become remarkably alike over the four decades studied. In the 19505, the Communists above all defined their message in terms of social class. To some extent, they also argued in terms of the nation and a few other categories. Nevertheless, the predominance of the class element was quite apparent. The two non-socialist parties, by contrast, displayed an equally strong emphasis on the nation. This is especially true of the Center Party, which in faet stands out as the most uni-oriented party in connection with the 1954 election. The Conservatives also appealed to 'class' to a moderate degree. This finding, however, is not similar to the Communists' message, since they naturally referred to quite different class interests. The Social Democrats had the most diversified pattern of orientation. They largely campaigned in terms of references to other political actors, including the Government and individual politicians. At the same time, they also appealed to the nation as well as to class and group interests.

In 1987, in contrast, campaigning in terms of social class has all but vanished. If the parties appeal to group interests at all, it is to quasi- and non-classes or demographic groups such as 'employees', 'women' or 'senior citizens'. Moreover, it is the nation at large rather than any clearly defined segment of the electorate or population the parties address when they wish to speak to the people. Equally remarkable, however, is the fact that campaigning is less and less orientated towards a dialog with the people; an increasing share of the editorials deal with the political elite (included in the 'other' category) rather than with any segment or level of the electorate. This is especially true of the Conservatives, whose editorials contain little in terms of a group orientation today.

Table 6 depicts the share of various orientation levels of the total volume of the editorials (in terms of centimeters of column) by party and by election year. The data indicate that as the class element proper has decreased in importance, references to quasi-classes and non-classes have become more frequent. There has, in other words, been an increasing tendency to address groups like 'employees', 'families with children', 'students' and 'pensioners' whenever the parties have wished to single out any specific segment of the electorate. This finding is basically in accordance with our expectations. It should, however, be borne in mind that it is less and less common to address specific groups. The intra-elite debate and references to the nation at large are the predominant feature of party propaganda today.

As to the geographic dimension, our hypothesis is not supported by the data. This level is mainly of importance to the Center Party, which continues to underline the importance of a 'living countryside' and to advocate the particular interests of Northern and Eastern Finland. Interestingly, the importance of the geographic dimension temporarily drops in connection with the 1966 campaign. The year before, the party had abandoned its old

name (Agrarian Union) and commenced its drive to gain a foothold in cities and towns. Apparently, this reorientation was regarded as too drastic; by 1975, the geographic dimension had nearly regained its original position in centrist propaganda.

By the same token, the hypothesis about the increased importance of the individual level is clearly falsified. In fact, references to the individual level have decreased over time. This level had a certain position in Conservative propaganda at the beginning of the period, but by 1987 it had become peripheral. Besides pointing to the conclusion that the possible 'individualization' of political campaigns has not meant a stronger emphasis on the individual voter, 'the citizen' or 'the individual', our data indicate that the individual level has never been of major importance in Finnish party propaganda.

By contrast, the national level has become increasingly prominent. Taken together, it is now the most important level of orientation for Finnish party propaganda. Interestingly, the Social Democrats and the non-socialist parties have changed places concerning the intensity of references to the national level. The Social Democrats, the Staatstragende party par excellence, are now the chief advocates of 'the common good" or 'the good of the nation', that which used to be the leading bourgeois trademark. Defense of the welfare state against the neo-liberal challenge apparently must now be conducted in terms of the nation rather than specific group interests:

This election points the way for the entire nation. The SDP has embarked on the campaign
by stressing this fact (16 March 1987).

The residual category ('other') shows an increase quite in accordance with our expectations. Little of this material contains references to other population groups; instead, the intra-elite perspective prevails. Political parties, the government and individual politicians are usually addressed in these editorials. It is here that the individualization of Finnish politics is noticeable: it is the individual politician rather than the voter as an individual has come to the fore. The practice of conducting extensive personal interviews with leading politicians (none in 1954, five in 1966, three in 1975, 20 in 1987) in close connection with the editorials has accentuated the impression that politics is to an increasing extent a question of the individual qualities of party leaders (Esaiasson 1985, 61-67, 181— 182).

Finally, the re-ideologization hypothesis seems to have some validity in the Finnish case. Between 1954 and 1966 the parties become much more alike as class references in particular drop rather considerably. At the same time, the national level becomes more important. By 1975, however, the class dimension has regained much of its earlier importance in Communist Party campaigns. For the Social Democrats, 'quasi-classes' have attained a similar importance. Thus, both the 'end of ideology' trend of the first half of the 1960s and the re-ideologization from the late 1960s on can be found in the Finnish campaign data. All in all, the similarities between the parties are more overwhelming today than in 1966. By the same token, the differences were sharper in the mid-1950s than 20 years later.

In sum, the level of orientation of Finnish parties has evolved in accordance our main hypothesis. There is considerable convergence between the parties both in 1966 and particularly in 1987. Moreover, the re-ideologization in the middle of the period was found in the data for 1975. References to class and group interests have largely been replaced by an orientation towards the nation at large and by an increased intraelite As to the individualization of political campaigns, it was found to concern the parties rather than the electorate. The individual voter has never been a major focus of Finnish electoral campaigning. Finally, contrary to our expectation, the geographic dimension has not become obsolete as a level of orientation, although it is not of major importance for the entire party field.

Propaganda techniques. 'Propaganda' and 'ideology' are often intuitively regarded as related, if not coincident concepts. If 'propaganda' is perceived as the opposite of 'fact', then 'ideology' is regarded as the opposite of 'matter-of-factness' in politics. Ideology means that not only individual issues, but basic differences in Weltanschauung separate political actors.

To further an ideology not only requires facts in favor of stands on
specific issues; propaganda is necessary to convince the public of the more
fundamental lines of demarcation between parties and movements.

If this is the case, then politics will be the more 'propagandists' the more 'ideological' it becomes. Conversely, the more 'de-ideologized' politics becomes, the lesser will the extent of propaganda be in political debate. And if 'propaganda techniques' are a way of introducing propagandistic elements into the debate, 'the end of ideology' should entail a reduced use of these techniques.

Figure 1 indicates that the overall development does not entirely support this general hypothesis. The total frequency with which the parties have resorted to propaganda techniques has not declined in a linear fashion. Propaganda techniques were somewhat more common at the beginning of

the period than at the end. However, they were least frequent in 1966 and clearly most extensively used in 1975. Moreover, the general trend has been rather similar for all four parties, particularly with regard to the sharp rise in the use of propaganda techniques in 1975 and the subsequent decline in 1987. All in all, the use of propaganda techniques seems to have been a normal element in the political debate throughout the period.

Nevertheless, interesting changes can be noted both with respect to the various techniques as well as between the parties. Quite in accordance with our expectations, 'name calling' has all but disappeared from the verbal arsenal of the Finnish parties: it was the third most frequent technique in 1954, but was used less than any other technique in 1987. The bulk of this change is accounted for by the fact that the Communists more or less dropped this technique after 1975. Thus, while an editorial in 1954 spoke of. . . this Government of Reaction and the bigwig parties behind it . . .' (24 February 1954), the same paper in 1987 simply stated that '. . .the SKDL represents the only realistic alternative to present Government policy . . .' (10 March 1987). There is a clear decrease in the use of 'name calling' for the other parties as well (cf. Appendix 2).

By contrast, 'card stacking', which we expected to decrease over time, has retained its position as one of the main 'tricks of the trade'. All four parties 'stack cards' to about the same extent throughout the period; there is no clear trend over time or a difference between the various parties. The use of 'testimonial' was expected to increase with the growing emphasis on personalities and party leaders. This has been the case, although the trend is by no means dramatic. Contrary to our hypothesis, 'bandwagon' and 'plain folks' display a decline rather than a growth. The basic impression, however, is that these two techniques have never been particularly important the Finnish political debate. This type of populism remains surprisingly in Finland. Finally, we expected a growth in the use of 'glittering generality' in the wake of blurred class distinctions in the Finnish society. Again, the basic trend confirms this hypothesis, but the extent of change is not particularly great.

Comparing the parties, a few features may be noted. The general tendency greater similarity over time was mentioned above. What is most striking is that the Communists, up until 1975 the propagandists par excellence, now use propaganda techniques less than the other parties. It is the Conservatives who in purely quantitative terms come closest to them. Conservative campaigning started from a low level of propaganda and displayed an increase up to 1975. In contrast to the other parties, the level of Conservative Party propaganda rose also between 1954 and 1966. Since then, however, the basic trend has been identical for all four parties. This indicates a convergence also in terms of techniques used by them; their profiles are now roughly identical (cf. Appendix 2).

What is perhaps most striking in our data is the increased use of propaganda in 1975. The trend towards re-ideologization noted earlier is indeed visible here; in fact it appears clearer than does level of orientation. By the same token, the falling trends in both 1966 and 1987 confirm our previous impression. The general roller-coaster pattern is compatible with the early 'end of ideology' thesis, the hypothesis about re-ideologization, and with the idea of ideological convergence in the 1980s.

Summary and Conclusions

To sum up, the empirical data came out in support of our hypotheses in
the following respects:

(1) The four parties have become more similar both in terms of their level
of orientation and in their use of propaganda techniques;

(2) The 1975 campaign gives evidence of a 're-ideologization' of the political
debate;

(3) Class orientation has declined;

(4) References to the national level have increased;

(5) The 'intra-elite' perspective has become more dominant, including a
stronger focus on individual politicians;

(6) 'Name calling' as a propaganda technique has nearly disappeared from
the campaigns; and

(7) The use of 'testimonial' and 'glittering generality' have displayed an
increase over time.

The following findings ran contrary to our expectations:

(1) The individual voter has not become more important as a focus of
political campaigns;

(2) The geographic dimension does not display a major decline over time;
(3) There is no clear decrease in the total use of propaganda techniques;
(4) The use of 'card stacking' has not diminished; and

(5) 'Bandwagon' and 'plain folks' have not become more important as
propaganda techniques.

The Finnish parties have clearly not replaced the traditional focus on social classes by a dialog with the individual voter. Rather, it seems that the parties are no longer primarily addressing the electorate. Instead, the campaign rhetoric presents politics as a game among the main political actors. This reflects the professionalization of party politics as well as the function of politics as a sort of pastime rather than a channel of interest articulation for the citizenry. On the other hand, the relatively strong position of the geographic dimension in party propaganda represents a traditional element in political campaigns.

The use of propaganda techniques as such does not seem to be clearly related to traditional ideological campaigning. Rather, it is the most offensive these techniques, 'name calling', that seems to covary with such traditional elements as class orientation. As far as the other 'tricks of the trade' are concerned, they seem to remain part and parcel of the campaign arsenal. In this sense, no definite Versachlichung of political campaigns has resulted from the process of de-ideologization.

Nevertheless, the bulk of our empirical findings do corroborate the basic hypothesis: the campaign style of Finnish parties is less ideological than a few decades ago, and it is increasingly difficult to tell the main parties apart. By the same token, the idea of a re-ideologization in the wake of the radical wave of the late 1960s and early 1970s receives support in our data. To a large extent, Finnish campaigns appear to follow the mainstream of the political current in the Western World as this is depicted in the literature.

Concluding Remarks

As elsewhere in Europe, the rise to prominence of the 'new middle classes' has created a large electoral no-man's-land between the political parties in Finland. These new groups form a considerable reservoir of electoral support. Any party that manages to exploit this potential to a decisive extent is likely to change the power relations in Finnish party politics.

So far no party has emerged as the middle-class party par excellence. With the partial exception of the Center Party, all major parties in Finland continue largely to depend on their traditional core strata. As for the Center Party, its support outside the agricultural community still does not originate primarily from the middle class. The middle-class potential, in short, is there for the taking. At the same time, no party can really afford the risk of losing major parts of their traditional support.

Against this background, the de-ideologization of political campaigns in Finland does not have to be interpreted solely in terms of the homogenizationof structures and values. The electoral dilemma mentioned above may offer an equally important source of explanation. To effectively reach the new middle strata would require a clear shift in class orientation, a focus on clearly defined problems common to these groups. So far, such a shift has not occurred. Instead, all four parties studied have chosen to blur class and group distinctions by switching the focus to the nation and to the level of the political elite. The parties apparently count on the fidelity of their core strata as long as they do not expressedly woo other groups. At the same time, the heaviest verbal artillery has more or less vanished from party propaganda. The parties direct their messages to everybody,

but nobody in particular, and they do this in a more inoffensive way than
earlier. Consensus and convergence can well be an effect of everybody's
playing it safe.

What about the response of the electorate? The question certainly lies beyond our empirical focus, but some thoughts may nevertheless be in order. In a way, the new middle classes have responded 'rationally' in a situation where the fundamental differences between the parties have all but vanished - i.e. they do not form a cohesive electoral block but spread their vote relatively evenly across the political spectrum (cf. Rantala 1982, 172-175). The Conservatives have profited by the rise of the new middle strata somewhat more than other parties, but the difference is not dramatic, and it is possibly diminishing at present. What is perhaps even more 'rational' in this perspective is not to vote at all as long as the partisan choices offered do not represent real alternatives. The figures on voting turnout support this perspective, but may give cause for some concern: turnout fell from 81 percent in 1983 to 75.9 percent in 1987, about 10 percentage points below the all-time high reached in 1962.

REFERENCES

Berglund, S. 1988. 'The 1987 Eduskunta Election in Finland', Scandinavian Political Studies
11, 69-76.

Borg, O. 1964. Suomen puolueideologiat. Periaateohjelmien sisäitöanalyytlinen vertailu sekä
katsaus niiden historialliseen syntyprosessiin. Helsinki: Valtiotieteellinen yhdistys.

Borre, O. 1987. 'Some Results from the 1987 Danish Election', Scandinavian Political Studies
10, 345-355.

Budge, I. & Farlie, D. 1985. 'Party Competition-SelectiveEmphasis or Direct Confrontation?
An Alternative View with Data", in Daalder, H. & Mair, P., eds.. Western European Party
Systems. Continuity and Change. Beverly Hills, California: Sage Publications.

Dittrich, K. 1985. 'Testing the Catch-all Thesis: Some Difficulties and Possibilities', in Daalder.
H. & Mair, P. eds.. Western European Party Systems. Continuity and Change. Beverly
Hills, California: Sage Publications.

Esaiasson, P. 1985. 'Partiledarna infor valjarna. Partiledarnas popularitet och betydelse
for valresultatet', Gothenburg: Statsvetenskapliga institutionen, Goteborgs Universitet,
Forskningsrapport, 4.

Helenius, R. 1969. The Profile of Party ldeologies. Helsinki: Scandinavian University Books.

Kirchheimer, O. 1966. 'The Transformation of the Western Party Systems', in LaPalombara.
J. & Weiner, M., eds., Political Parties and Political Development. Princeton: Princeton
University Press.

Lee, A. & Lee, E.. eds., 1939. The Fine Art of Propaganda. Published by the Institute of
Propaganda Analysis. New York: Harcourt. Brace and Company.

Lindblad, 1., Stålvant. C-E., Wahlback, K. & Wiklund. C. 1984. Politik i Norden. En
jåmforande oversikt. Stockholm: Liber Forlag.

Mayer, L. 1989. Redefining Comparative Politics. Promise Versus Performance. London: Sage
Publications.

Nousiainen, J. 1985. Suomen poliittinen jarjestelmd. Juva: WSOY.

Panebianco, A. 1988. Political Parties: Organization and Power. Cambridge: Cambridge
University Press.

Pietilå. V. 1973. Sisållon erittely. Helsinki: Gaudeamus.

Przeworski, A. 1987. Capitalism and Social Democracy. Cambridge: Cambridge University
Press.

Rantala, O. 1982. Suomen puolueiden muuttuminen. Helsinki: Gaudeamus.

Siune, K. 1982. Valgkampe i TV og radio. Politiske partiers anvendelse af ti- og radio, herunder
samspillet med journalisterne. Århus: Forlaget Politica.

Suhonen, P. 1984. 'Luokka-asema ja puoluekanta', Politiikka XXVI, 176-189.

Toivonen, T. 1988. Rakennemuutos 1930-1985. Aineistoa toimiala-, luokka- ja kerrostumarakenteen Publications of the Turku School of Economics, Series A-l:1988 (Summary: Structural Change 1930-1985: Data and Material on Transformation of Industries, Social Classes, and Strata),

von Beyme, K. 1985. Political Parties in Western Democracies. Aldershot: Gower Publishing
Company.