Election campaigning in Denmark has been a rather common phenomenon in the past two decades. From 1968 to 1987 there have been nine general elections, six local elections, three elections for the European Parliament and four referendums, making a total of 22 campaigns. One might reasonably therefore, that campaign studies of various kinds have flourished Danish political scientists. But this is far from the case. The main focus of research has been on various aspects of the voters' behaviour, leaving as a major lacuna the analysis of Danish political parties' election campaigning as such. This article is a modest effort to redress this situation. The article presents an exploratory study of the 1988 campaigns of the Social Democratic Party (Socialdemokratiet), the Conservative Party (Det konservative Folkeparti) and the Christian People's Party (Kristeligt Folkeparti)

These parties have been selected because: (1) the Conservative Party was the largest bourgeois party in the cabinet as well as in the Folketing; (2) the Christian People's Party was the smallest party in the cabinet, constantly fighting against the two percent electoral threshold; and (3) the Social Democratic Party was the largest socialist party and the leader of the opposition. This selection is not fully representative, of course, but it provides enough information to enable some preliminary conclusions to be

Resumé

In this article the 1988 national election campaigns of the Conservative Party, the Christian People's Party and the Social Democratic Party are analysed focusing upon: the impact of environmental factors- i.e. the constitutional setting, the party system and the media system; and the campaigns themselves - i.e. preparation and planning, control, strategy, financing and style. Findings indicate that the single most important factor for understanding national election campaigns in Denmark is that parties never know when a general election will be called. As a result, preparation and planning are virtually absent, TV is assessed as the most important channel of communication, the campaign is fairly centralized and controlled by the party leadership, the parties are dependent upon voluntary manpower and financial support, the use of 'outside' expertise is minimal, and the complex pattern of cooperation and conflict in Danish politics has a modifying impact on the amount of aggressiveness displayed in the campaign. All in all, similarities among the parties' campaigns are more pronounced than differences. It appears to make no difference whether the party is incumbent or in opposition, established or new, left-wing or right-wing, large or small.

drawn regarding some aspects of election campaigning by Danish political
parties.

The literature presents no tradition or guidelines for analysing political party campaigns which are commonly agreed upon. Campaign studies are mostly ethnocentric and do not have methodological approaches in common (Farrell & Bowler 1990). One broad avenue to follow is to distinguish between environmental factors which influence campaigning on the one hand and the campaign itself on the other. Environmental factors cover the contextual circumstances in which the campaign takes place, i.e. the type of political system, electoral laws and conventions, the media system and the type of party system. The campaign itself, by comparison, may be analysed according to categories such as preparations and strategic planning, campaign organization and control, and implementation and campaign tactics (Farrell & Bowler 1990).

Such a framework enables some specific - but admittedly broad and incoherent - questions to be raised under each of the headings mentioned above: How and to what degree did Danish electoral law and conventions, the Danish system of mass media and the Danish party system influence the campaigns of the three selected political parties? When did preparations for the campaign start and what was the strategy to be followed? Who was in control of the campaign - the central party leadership or the local branches? What role, if any, did professionals (consultants or marketing and survey agencies) play compared to voluntary manpower? How was the campaign financed? How was the campaign implemented? What channels of communication were used? Was the campaign style largely aggressive or defensive in character, orientated towards party image and/or the image of the party leader? And can any major differences be detected among the three parties with regard to these questions?¹

Environmental factors and campaigning will be considered in turn.

Environmental Factors

According to the Danish constitution, the Folketing is elected for a period of four years. Two constitutional provisions, however, provide for the possibility of calling a general election prior to the end of the four-year term. In fact, this has been the rule more than the exception: from 1945 to 1987 only seven of 19 general elections have been held after a full election period.

Various regulations, deadlines and other formalities contained in the electoral law mean that the minimum duration of any election campaign is three weeks from the day an announcement is made in the Folketing. In fact, this very short period of time has actually been the normal length of the campaigns in the 1970s and 1980s. Denmark has become accustomed

to short, sudden and mostly hectic election campaigns. This, it must be presumed, has an important effect on the campaigning of political parties insofar as they have less opportunity to prepare and plan their campaigns and must instead rely on a relatively standard set of procedures.

The electoral law does not contain any rules concerning the election campaign itself. In principle there are no constraints either on the candidates on the parties as to how they finance and conduct their campaign. The absence of regulations gives the political parties, as well as the candidates, wide degree of freedom and allows them to uphold secrecy as regards their internal affairs if they so desire.

Parties not already represented in the Folketing have to collect signatures equivalent to only 1/I7sth of the total votes cast at the previous election (roughly 20000 signatures) to gain the right to participate in the election. This means that the Danish campaigns are usually crowded with parties. In 1988, 13 parties participated, and this was three less than in 1987!

The electoral system used in national elections is a complex PR system (Johansen 1979; Tonsgaard 1986). Without going into technical details, it may be noted that this election system makes it extremely difficult - if not impossible - for a candidate to gain a seat if s/he is not nominated by a party. In practice, the political parties hold a monopoly on the seats in Parliament.

Denmark indeed qualifies as a multiparty system. The characteristics of this system need not be recapitulated here (cf. Bille 1990; Pedersen 1987). It is sufficient to stress that the absence of a single majority party or even a single predominant party has made cooperation among the parties necessary in order to establish a legislative majority. A long and deeprooted of cooperation therefore exists in the Danish party system.

From a campaign perspective this situation implies that the parties constantly have to make tactical considerations. On the one hand they have to temper their campaign messages appropriately in order not to prevent some sort of coalition with one or several parties after the election. On the other hand, the competition among the parties is keen, not only among parties from different blocs but also - and perhaps most of all - among parties inside the same bloc. This makes for a rather complex and ambiguous situation: parties attack each other at the same time as they tell the voters that they nevertheless are willing to cooperate with one another.

In some respects the structure of the Danish mass media is also unique. The number of newspapers has been steadily diminishing. From a situation in which 123 newspapers existed at the end of the Second World War, most of them with a clear and manifest affiliation to one of the 'four old parties', a process of attrition and amalgamation has taken place to the point where only 44 newspapers survive at present (Thomsen 1987). The majority of

the remaining newspapers, furthermore, no longer have a direct connection to any single political party. Direct party-owned newspapers are of no real relevance in the Danish context. Yet the bulk of the newspapers are clearly bourgeois in their general political orientation. This becomes more manifest during an election campaign. Being commercial enterprises, however, all the newspapers carry advertisements from the entire political spectrum. There are, moreover, no rules regarding political balance for the newspapers;the money a party has available, the more it can decide to spend on advertisements.

Circumstances are quite different when it comes to campaigning on the radio and television. Until the beginning of the 1980s, only 'Danmarks Radio' was allowed by law to broadcast radio and television in Denmark. Since that time, however, local radio and television stations have been allowed to take up broadcasting and in 1988-89 a second nationwide TV channel was established. The present broadcasting law stipulates that political propaganda on television is forbidden and that the programme policy of the nationwide TV networks shall be characterized by comprehensiveness fairness (Siune 1982).

When it comes to election campaigning, this demand has been observed very strictly by 'Danmarks Radio'. In particular, a special set of rules is adopted by the management of the 'Danmarks Radio' at the outset of every new general election. The main principle of these rules is to secure an even balance between the various political standpoints and to give each party exactly equal time. As soon as a general election is called, a 'political curfew' is enacted. Any programme with even the slightest affiliation to politics is cancelled and no politician can participate in any programme other than those directly connected to the election campaign. Each party participating in the election - new or old, small or big - gets exactly the same amount of time to present its platform.

Each party produces its own presentation programme with a maximum of 10 minutes duration. These programmes are subsidized by 'Danmarks Radio' and the same amount of money is given to all parties. In 1988 the sum was DKK 73 000 (US$ 11 200), but the individual party is free to spend more money on the production should it so choose. The presentation programme is shown on prime time and is immediately followed by a 30minute in which three of the party's leading politicians of its own choice are 'cross-examined' by two journalists from 'Danmarks Radio'.

Two days before the election, the campaign in 'Danmarks Radio' ends with a debate among the leaders of all the contesting parties. Each of them has exactly the same amount of time at their 'disposal' for asking and answering questions of one another. A neutral person uses a stopwatch to time the speakers and anybody surpassing the time limit immediately has their microphone cut off!

With 11 to 16 party leaders participating in the debate as has been the case in the last two decades, it is not hard to imagine how difficult it is to concentrate the debate on one issue at a time and for a single party to give a coherent presentation of some of the essential items of its platform. On the other hand, this form of debate presents a great opportunity to make the 'match-winning' - or 'match-losing' - statement. Previous research suggests that the last minute impression is important to approximately 10-15 of the electorate in terms of deciding for which party they will cast their vote (Siune 1984; Tonsgaard 1989).

This kind of debate requires special deliberations regarding the tactics to be pursued. The crux of the matter may be formulated as follows: is it a more rewarding tactic to use most - or all - of the time allotted at the beginning of the broadcast, thereby making sure that you get the message out at a time when the viewer rating is high, but at the same time running the risk that some other party may make an accusation or ask questions which you are not able to refute or answer because your time has run out? Or is it wiser to save your time and use it at the end of the debate when you are sure that nobody is in a position to contradict you, but when you are equally sure that the viewer rating is low?

This type of final debate gives a relatively good picture of the pattern of cooperation and conflict existing among the parties as well as of the style, the aggressiveness and the priority of campaign issues of the individual parties. In the final debate at the 1988 election, for example, none of the four parties in office (the Conservative Party, the Liberal Party, the Centre Democrats and the Christian Peoples Party) asked each other any questions. They concentrated on defending the politics of the government, but even more so on attacking the Social Democratic Party rather fiercely, which in its turn was then forced to answer questions and comment on contributions from four parties. The four parties in the cabinet were relatively forthcoming and friendly towards the Social-Liberal Party, on the other hand, a party which before the election had been - and most likely after the election also would be - a necessary supporter to ensure the survival of the government. The actual parliamentary situation. in other words, has a decisive impact on the parties' campaign strategy as a whole. But due to the unique structuring of the 'grand debate", it has a particularly important impact on party performance in this final phase of the campaign.

The effect of the election campaign on the voters as well as the impact of the different types of media has been studied extensively (Borre et al. 1976; Elklit & Tonsgaard 1984. 1989; Siune 1989; Worre 1987). For our purpose the relevant findings are that as many as approximately 25 percent of the voters are undecided as to how they will vote when a general election is called. This of course makes the campaign tremendously important to the parties. A well-planned and well-conducted campaign which reaches

its peak in the last few days prior to an election is imperative to the parties. But to be able to do this requires knowledge of from what sources the voters get their information on the parties' platforms during the campaign. In that respect the situation in Denmark corresponds to the åndings in other European countries: television is by far the most important channel of information, followed by newspapers and radio. It is to be expected, therefore, that campaign strategists will give high priority to the TV media.

Campaigning

Data on the 1988 campaigns of the Christian People's Party, the Conservative and the Social Democratic Party were gathered by interviewing secretaries in January 1990. These data have been supplemented with secondary information appearing in the press. The data are presented here according to the three headings mentioned above - planning, organization and control, and implementation and tactics - even though this is an artificial distinction not reflecting the real situation confronted by the parties.

Planning

The artificiality of trying to separate distinct phases of the campaign becomes clear when we approach the question of when in fact the party begins its planning and when it ends the process. Planning is an ongoing process and one might rightly state that planning finally ends on the very day of the election. As used here, however, 'planning' is taken to mean all preparations the party undertakes concerning a coming campaign before the announcement of the general election is made.

The 1988 election was completely unexpected (Bille 1988). Only eight months had passed since the previous election, so even by Danish standards it came as a total surprise to everybody. Even the Conservative Party, whose chairman was Prime Minister, was surprised by the announcement. Under the circumstances, therefore, no specific plans existed for the election By default, so to speak, concepts and routines from previous campaigns formed the initial basis for campaigning. In reality this meant that some standard posters and propaganda material left over from the 1987 campaign, or that had been printed to hold in stock, was all that was available. In fact, the real 'planning' only began after the campaign had started!

The Christian People's Party, however, presents a slightly different
case. A standing election committee is regularly appointed by the party's
executive committee at the annual party conference. As soon as possible

after a general election, the standing election committee prepares an 'election contingency plan' to be implemented whenever a new general election is called. Besides this contingency plan, the committee had - also in advance - selected eight major issues from the party manifesto after the 1987 election to serve as the party's central themes as soon as a new campaign had to be launched.

Organization and Control

Whenever a general election is called, it is common practice for all Danish political parties to concentrate all of its organizational resources on the campaign. All the employees at the headquarters throw themselves into the campaign, and the rank and file of the party members in the regional and local branches are also mobilized.

The central headquarters of the Conservative Party was staffed with 17 to 18 full-time employees in 1988. The party had no regional or local employees. Besides the employees at the party head office, the party's press agency and the party's employees in the political-economic department of the Conservative parliamentary faction - a total of 10 persons in 1988 - also actively participated in the campaign. In addition, some secretarial assistance was hired on a temporary basis. All in all, the party commanded approximately 27 to 30 paid party officials to run the campaign. This, by Danish standards, relatively high number of paid party officials did not alter the fundamental dependence of the party on the voluntary manpower of its 42 000 members. By tradition, however, only a small fraction of Conservative party members actively participate in the campaign.

As for technical facilities, the Conservative headquarters had to be content with the 'old-fashioned' telephone, copy machines and postal services. No fax or on-line computer links connecting the central office with regional and local branches were at the disposal of the party in the 1988 campaign.

The party headquarters of the Christian People's Party had six paid employees in 1988 and the secretariat of the parliamentary group had seven persons employed, making a total of 13 paid party officials to run the campaign. The party had no regional or local employees. This very modest amount of 'professional' manpower emphasizes the party's dependence on the voluntary activities of its 9000 members. In contrast to several other Danish political parties, the members of the Christian People's Party are registered at headquarters, and so labels can be produced and ail the members can be reached very quickly by direct mail. A relatively high proportion of these members actively participated in the campaign, and together with the local candidates, these members constituted the real force of the campaign.

Technical facilities at the disposal of the Christian People's Party were the traditional ones. No computer links or printing press were available, but a fax machine did exist at Christiansborg. Campaign material (posters, billboards and so on) were bought 'in town'.

The Social Democratic Party has the largest organization of all Danish political parties. Approximately 35-40 persons are employed full-time at the party's head office. During the campaign this number was temporarily increased by an additional 10-20 persons recruited mainly from former party officers and activists. In addition, the party's press agency as well as the political-economic department and the secretariat of the parliamentary faction, approximately 20 persons in all, participated in the campaign. The party also had three regional offices, Aalborg, Copenhagen and Odense in which altogether five persons were employed.

It is important to note also that the labour movement in Denmark has always been dominated by the Social Democrats. Through decades of collaboration, a comprehensive network has been formed between the Social Democratic Party and the trade unions. This means that the party can draw on the organizational and financial resources of the national federation of trade unions and all of the other organizations of the labour movement. This intertwining makes it extremely difficult to identify precisely professional manpower engaged by the party during the campaign, but an estimate is 100 persons.

Within the Social Democratic Party itself a tradition of participating in the campaign has been stronger among its 100 000 rank and file members than in several of the other Danish parties. This meant that the party was better off with respect to voluntary manpower in the 1988 campaign than the rest of the parties. In addition, the technical facilities of the party included fax and computer links to the trade unions, but not to the local branches of the party.

The organizational strength of the party is clearly an advantage to the Social Democrats, but it takes time to get an organization of this size to function optimally. Considering the very short period of campaigning, this fact was to the disadvantage of the party in the 1988 campaign.

As to the control of the campaign - i.e. the degree of centralization, coordination and standardization of the campaign - the Conservative Party constitutes a case with contrasting features. On the one hand the campaign was extremely centralized. A narrow, informal circle consisting of the chairman of the party, the party's general secretary, the chairman of the parliamentary group, one close and confidential adviser of the party chairman, and sometimes the leader of the press agency is normally formed to run the campaign. This group had overall responsibility for conducting the campaign in 1988, with the party chairman serving as the key figure. The group met at least once a day during the campaign. They decided

on the party's main campaign themes, on nationwide advertising in the newspapers (every single advertisement was controlled and sanctioned by the party chairman himself!), and on the general strategy to be pursued. They sanctioned the standard material to be produced by headquarters as background information for the candidates, including material containing information on the party's standpoints and 'results'. They also produced a daily comment on the progress of the campaign containing information and arguments for or against statements from the other parties. They decided, furthermore, on who was to represent the party at the daily press conference at Christiansborg, and in the radio and television broadcasts. This group, in short, constituted the core group with respect to all nationwide aspects of the campaign.

At the same time the Conservative Party campaign was quite decentralized character. The individual candidate was free to choose the type or style of campaign that s/he deemed best suited to local conditions. The candidates were free to use the material produced by the central party organization, and they were not obliged to report to the Conservative head office on their personal campaign. Within the general and rather broad framework of the party's centrally determined campaign strategy and themes, in other words, the degree of local self-determination and variation. was high.

The campaign of the Christian People's Party was also relatively centralized. Normal procedure is that the party's executive committee together with the candidates are called together the first weekend after an election has been called to confirm the campaign strategy and the plan developed by the party's standing election committee. This election committee also decisive with respect to the party's presentation programme for radio and TV. The weekend meeting is a rather formal assembly, however. The core group in terms of reaching day-to-day decisions during the campaign consisted of the party's executive committee (nine persons) with the party chairman and the party secretary as the central figures.

Christian People's Party candidates were nonetheless on the whole autonomous their local campaigns, so long as they stuck to the broad outlines of the party manifesto and strategy. They were not obliged to use the material produced by the party headquarters and neither were they obliged to report to central party headquarters. As is the case in all Danish parties, however, the regional party branches were important in carrying out the campaign of individual candidates.

It is also standard procedure of the Social Democratic Party₁ to summon a meeting to be held in the first weekend after the announcement of a new general election. In this case the participants are the parliamentary faction. the candidates, the party's executive committee, the chairmen of the constituency organizations, the executive committee of The Federation of

Danish Tråde Unions (LO) and the Social Democratic chairmen of the various federations of tråde unions. This large assembly is indeed a very formal assembly and the purpose of the meeting is only to confirm plans for the campaign prepared by the party leadership and to generate enthusiasmfor launching of the campaign.

The party leadership was the central controlling body during the campaign. It consisted of the party chairman and the two deputy chairmen, the chairman, deputy chairman and secretary of the parliamentary group and the party secretary. This group exercised overall responsibility for conducting the campaign politically as well as organizationally. They decided campaign strategy, themes and propaganda and on who was to represent the party at the daily press conference and on TV and radio. The real core taking care of daily details and decisions during the campaign was a triumvirate consisting of the party chairman, chairman of the parliamentary group and the party secretary. Daily coordinating meetings were arranged morning and evening.

Individual candidates were free to choose among the material sent from headquarters and could conduct their own campaign as they felt best fitted local conditions so long as they respected the centrally determined strategy and themes. All in all, however, the degree of centralization and coordination the election campaign was fairly high in the Social Democratic Party.

Implementation and Tactics

On the whole, the communication strategy of the three parties in 1988 followed a traditional path. The most important campaign channel as assessed by the party strategists was the 10-minute presentation programme followed by the 30-minute 'cross examination programme' and the final party leader debate on TV. Following this in importance was advertising in the national and local newspapers.

Only the Conservative Party engaged an advertising agency to assist in designing campaign propaganda, while the Christian People's Party to a very modest degree used some assistance from hired consultants. The Social Democratic Party used a network of sympathizers of skilled lithographic artists, lay-out people and communication experts, all on a freelance and voluntary basis. The party's press agency also contributed to the design of the campaign.

In the 1987 election the Social Democratic Party called an early morning press conference every day. In the 1988 election the other major parties adopted this idea and the parties had to enter an agreement regarding a timetable. The major parties' press conferences were well attended by journalists and the usefulness of this tactic as a suitable forum to present

the theme of the day, to comment on the other parties' campaign, documentationand forth, was assessed to be good by the party strategists of the major parties. This new arrangement clearly promoted a centralist trend in the campaigning. The Christian People's Party, however, held only four morning press conferences during the 1988 campaign. Any effort 'to conquer the agenda' or 'the theme of the day' was in practice abandoned in advance due to the minimal interest journalists had shown in the party's press conferences in the 1987 election. These earlier experiences indicate the fate of a small party and underline the importance of the principle of 'equal time' on national television.

None of the parties ordered or initiated any voter surveys during the campaign. They were viewed as being too expensive compared to their usefulness. The use of outside professionals or expertise in this form was for all practical purposes absent.

Since the election was called on a disagreement in the Folketing over security and defence policies, the Conservative Party's main 1988 campaign theme - along with the three other parties in office - was a clear stand in favour of NATO and the obligations connected with full membership. This platform was launched rather aggressively by the party, first of all in an attack against the Social Democratic Party. The Conservative Party, however, was unable to maintain this issue as the only main theme in the campaign. The usual economic and social issues gradually moved up on the agenda along with the always crucial question of which party or group of parties would form a government after the election.

No leaders' tour was planned. Only a couple of so-called 'Duels' between the Conservative Prime Minister, Poul Schliiter, and the Social Democratic opposition leader, Svend Auken, were arranged. Even so, the image of the party leader was emphasized in the campaign.

The largest single item on the Conservative campaign budget was the production of the presentation programme, a production costing approximately 200 000 (US$ 31 000). Assorted advertising in the newspapers amounted to the highest combined expenditure however. The total campaign of the central party headquarters was approximately DKK 5 million (US$77O 000).

The Christian People's Party had selected in advance eight major issues from the party programme to serve as the party's central themes during the campaign. Two more dealing with defence and security policies were added during the campaign. The timing was planned such that current defence and security issues should only be stressed at the beginning of the campaign, while the party's 'own' major issues should dominate the rest of the campaign.

The campaign focus for the Christian People's Party, therefore, was
clearly on party image and manifesto proposals. Due to a leader conflict

in the party's formative years, the party traditionally abstains from focusing on the image of the party leader and from arranging party leader's tours. Hence, no photographs of the party leader were shown on the party's posters. The party otherwise concentrated slightly more in 1988 than previously on advertising in the national and local newspapers, giving perhaps somewhat more priority to the local press.

The largest single item on the campaign budget was the production of the television programme. The costs were DKK 97 000 (US$ 15 000). DKK 104 000 (US$ 16 000) was spent on producing tapes and a video for local radio and television stations. Advertising in the press amounted to DKK 216,000 (US$ 33 000). The total campaign budget balanced at DKK 422 000 (US$65 000).

The Social Democratic Party leadership decided to run a campaign concentrated on three main themes. The first of these concerned security and defence issues. On these issues the party chose a rather defensive strategy, trying for the most part to turn the focus of the campaign away from these themes. The second main theme was to present the Social Democratic model of society as an alternative - and of course a better one - to the governing bourgeois model. The essence of this theme was that the bourgeois four-party government was no longer able to run the country in a proper, responsible and decent way. It was now time for a new government, a Social Democratic government. This theme was presented in an aggressive manner.

The third main campaign theme concentrated on building the image of the newly elected party chairman, Svend Auken. Auken had been elected after some dispute in the party. It was therefore essential to the party to present an image of a strong party united behind the new chairman. Moreover, in the debate in the Folketing which eventually led to the call of the general election, a question mark had been raised as to Auken's reliability as well as his ability as a party leader and statesman. This question mark had to be erased during the campaign. This third theme was pursued quite vigorously and the tradition of arranging a party leader's tour was intentionally emphasized in the 1988 campaign.

Approximately DKK 200,000 (US$ 31 000) was spent on producing the presentation programme on national television. The party's campaign budget amounted to DKK 6-7 million (US$ 920 000-1 080 000). The largest total amount was used on advertising in the printed press. The party changed strategies concerning this type of media in the 1988 campaign. Previously large sums were spent on whole-page advertisements in the national newspapers. In 1988 the party chose to emphasize more strongly selective advertisements in a smaller format in local newspapers, ladies' magazines, hobby magazines, trade union magazines, trade journals and the like. The section of the party's platform printed in the advertisement

was specifically selected to appeal to the magazine's readers. The 1988 campaign was in this respect more target-group orientated than previous campaigns. The party's marketing was otherwise intended to be of a fresher, less traditional character. A new style, a new party image and a new party leadership were introduced.

Conclusion

On the basis of this review, we may conclude that the single most important factor in understanding election campaigning in Denmark is that the political never know when a general election is to be called. This - and the short period of campaigning - gives few opportunities to prepare and plan the campaign. Hence all parties have a standard to fall back upon as needed, a standard campaign they can adjust to the actual issues and the general political situation which caused the election.

In 1988 the dominant features of the election campaign may otherwise
be summarized as follows:

(1) All parties viewed television as the most important channel of communication the voters. The unique rules of 'equal time' allocated to all parties on television were of invaluable importance to the small parties, to some degree balancing off differences in the various parties' organizational and financial strength.²

(2) The degree of centralization in the 1988 campaign was fairly high. The parties' headquarters controlled the main framework of the campaign, the strategy to be pursued, the themes used in nationwide advertising and so forth, leaving the candidates to conduct their campaigns within this framework with adjustments to local conditions as necessary.

(3) The use of outside expertise was limited and commissioned surveys were absent during the campaign. Having a general election almost every second year, the party leadership and the central party staff considered themselves as the real experts on campaigning, eliminating the need for buying assistance from consultants.

(4) The complex pattern of cooperation and conflict in Danish politics modified the aggressiveness in the campaign, simply because the parties never knew which party or parties was needed as a coalition partner or supporting party after the election.

(5) The similarities among the parties' campaigns were by and large more pronounced than the differences. It does not appear to make a difference whether the party is incumbent or in opposition, established or new. left-wing or right-wing, large or small.

Due to the frequency of general elections in Denmark, and the short

campaign period, campaigning has to be seen as a direct but intensive prolongation of the daily political 'battle'. No political agreement is ever reached in Danish politics without the political parties assessing its instantaneousimpact the voters, simply because experience tells the parties that a general election might be called in the very near future. In that way the Danish political parties conduct an 'election campaign' every day, year in and year out.

---

NOTES

1. Due to the fact that no systematic study of Danish political party campaigning is available, the question of whether any changes have occurred over time can only be answered in a very impressionistic way.

2. The Danish parties have small financial resources. The introduction in 1987 of a state subsidy to the parties did not remove their strong dependence on voluntary manpower and voluntary financial assistance.

REFERENCES

Bille, L. 1988. -Politisk Kronik 1. halvår 1988', Økonomi og Politik nr. 3, 58-69.

Bille. L. 1990. 'Denmark: The Oscillating Party System', in Mair, P. & Smith, G., eds.,
Understanding Party System Change in Western Europe. London: Frank Cass.

Borre, O. et al. 1976. Vælgere i 1970'erne. København: Akademisk Forlag.

Farrell, D. M. & Bowler, S. 1990. 'Comparative Election Campaigning: Towards a Framework
of Analysis', paper prepared tor the ECPR Workshop on Electoral Campaigning in Western
Europe, Bochum, Germany, 2-8 April 1990.

Johansen, L. N. 1979. 'Denmark', in Hånd, G., Georgel, J. & Sasse, C, eds., European
Electoral Systems Handbook. London: Butterworths.

Pedersen, M. N. 1987. 'The Danish "Working Multiparty System": Breakdown or Adaption',
in Daalder, H., ed.. Party Systems in Denmark, Austria, Switzerland, The Netherlands and
Belgium. London: Frances Pinter.

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

Siune. K. 1984. 'Bestemmer TV valgresultatet?', in Elklit, J. & Tonsgaard, 0., eds.. Valg og
vælgeradfærd. Århus: Politica.

Siune, K. 1989. 'Valgkampene og vælgerne', in Elklit, J. & Tonsgaard, 0., eds.. To Folketingsvalg.
Politica.

Thomsen, N. 1987. 'Politisk kommunikation, massemedier og opinion', in Dansk politik under
forandring 1945-1985. Et forskningsinitiativ under Statens humanistiske Forskningsråd.
København: Forskningssekretariatet.

Tonsgaard, O. 1986. 'Valglovgivningen - en kritisk gennemgang", in Elklit. J. & Tonsgaard,
0., eds., Valg og vælgeradfærd. 2. ed. Århus: Politica.

Tonsgaard, O. 1989. 'Vælgervandringer og vælgerusikkerhed', in Elklit. J. & Tonsgaard, 0.,
eds.. To Folketingsvalg. Århus: Politica.

Worre, T. 1987. Dansk vælgeradfærd. København: Akademisk Forlag.