Resumé

Due to their foreign policy opposition, the left socialist parties in Denmark. Norway, and Sweden remain among the few parties in Western democracies that lack governmental experience. When political parties confront political issues, they can either choose a competitive or a cooperative strategy. The Norwegian and Swedish left socialists chose competition when the EU issue appeared on the scene in the early 19905. The Danish Socialist People's Party, on the other hand, opted for a cooperative strategy and accepted EU membership and the 1993 Edinburgh Agreement. Drawing on coalition theory, this article asks why.

Introduction

This article studies foreign policy constraints on coalition formation. The three major left-socialist parties¹ in Denmark, Norway and Sweden have traditionally been defined as non-legitimate coalition partners (Strom & Leiphart 1992; Särlvik 1983). The Swedish Left Party's (Vänsterpartier) background as a traditional communist party, the Danish Socialist People's Party (Socialistisk Folkeparti) and the Norwegian Left Socialist Party's (Sosialistisk Venstreparti) opposition to NATO and EC/EU membership have made the parties non-ministerial. Opposition to NATO and EC/EU would cause problems with these institutions if those parties were to enter the government. However, the Danish Socialist People's Party's acceptance of EU membership and the Edinburgh Agreement in May 1993 at least officially placed the party among the pro-EU membership parties (Svensson 1993). Both the Swedish and Norwegian left wing parties advised their followers to reject membership in the referenda held in the autumn of 1994. They thereby

---

* Dag Arne Christensen, Department of Comparative Politics. University of Bereen, Christiesgt. 15. N-5007 Bergen. Norway.

held on to previous party positions (Christensen 1996). We raise two questions: Under which circumstances do parties redefine their self-imposed foreign policy constraints? What are their motives when they formulate their foreign policies?

Drawing on coalition theory, we formulate a framework for the study of constraints on coalition bargaining. We map key foreign policy dimensions within the three party systems and study foreign policy distance at the parliamentary level. Finally we ask how foreign policy distance affects internal party constraints in the bargaining process.

Constraints on Coalition Bargaining

Two factors are important when we study coalition building in Scandinavian politics. First, the traditional left-right dimension is not the only one to affect the process of coalition bargaining. Second, parties are not necessarily "unitary actors" when they confront potential coalition partners (Laver & Budge 1992; Narud 1996). Parties may unite on some dimensions of conflict and at the same time evidently confront each other on others. Recent contributions in coalition theory have modified traditional perspectives along these lines. The one-dimensional analysis of coalition formation does not adequately account for why certain coalitions form and why they fall. The nucleus of parliamentary work in Scandinavian multi-party systems is to build parliamentary coalitions in different policy areas, coalitions that may eventually turn out to be executive coalitions. Obviously other issues than those identified within the traditional left-right dimension are important when parties decide whether to join coalitions or not.²

Policy distance or cleavage mobilization is the essence of Schattschneider's (1960/1975) book The Semisovereign People. Schattschneider claims that among the possible cleavages in a political system, few become salient in electoral competition while others remain irrelevant. Parties try to define political conflicts in a way that increases their chances of building coalitions and getting access to political power. The salience of different issues and changes in the weight different dimensions have in a polity over time define the parties' space of action when it comes to coalition bargaining. Enemies may become friends as the patterns of conflict change, as issues that previously divided parties are forgotten or swept under the rug.

We also treat parties as constrained actors in the process of government formation. As pointed out by Strøm, Budge & Laver (1994, 307), "the real world of coalition politics is one of constraints, in which it is quite definitely not the case that everything is possible." Not everything is possible because parties are institutions that apply rules and procedures for their own conduct. Parties put internal constraints on their own behavior when they confront

other parties. It is not accidental that the left socialist parties studied here remain among the few parties in Western democracies that lack governmental experience. However, differences in policy objectives may be reduced over time and hence lead parties to soften their demands in the bargaining process. The most straightforward constraints parties put on themselves are made through formal party decisions. Party congresses usually choose among potential coalition partners. These choices impose restrictions on the party leadership in political negotiations. The development within the Danish Socialist People's Party can, as we shall see, be understood along this line of reasoning.

The crucial questions are: What are the motives when parties assess foreign policy dimensions? And what bargaining options does the party leadership have? If no options exist, we would not expect changes in party policy. First we turn to the foreign policy conflicts embedded in the Scandinavian party systems.

Party Systems and Foreign Policy Dimensions

No such concept as a common "socialist foreign policy" exists (Sassoon 1996). Note also that the collapse of communism and the intensified integration in the EU has made it difficult to draw a clear line between foreign and domestic politics.³ Socialist parties have primarily been oriented towards national politics, and in foreign affairs they have embraced the idea of national interests. However, two foreign policy dimensions divide leftwingers in Scandinavia. Both the security dimension and the European economic dimension have played a significant, and at times, decisive role in dividing social democrats and left-wing opposition.

Table 1 shows three distinct positions that are useful when we study the security dimension in Scandinavian politics. Two foreign policy alternatives existed from 1949 to 1989, i.e., nationalist neutralism and supra-national Atlanticism. In both Denmark and Norway, the issue of NATO membership divided the left wing into two distinct blocs. The left socialists in both countries originated as splinter parties in the late 19505. The Danish Socialist People's Party was established in 1959. when several prominent members of the pro-Soviet Communist Party either left the party voluntarily or were excluded. The split was a result of a severe conflict within the Communist Party after the Soviet invasion of Hungary in 1956.

In Norway the issue of NATO's nuclear policy broke the Labor Party from within when the Socialist People's Party (Sosialistisk Folkeparti) was established in 1961. The fight against EEC membership in 1971 triggered the consolidation of the Socialist People's Party and other left-wingers, which led to a broad electoral coalition in 1973 (Kvarn 1973: Stenersen 1977). In

contrast, the Swedish Left Party has its background in the conflicts within the Social Democratic Party (Socialdemokraterna) immediately after the Russian revolution in 1917. The Swedish Communist Party was established in 1921 and confronted the Social Democrats especially on economic policy. In Denmark and Norway, NATO membership was the most important cleavage between the leftists and the Social Democrats, while in Sweden domestic economic policy was the central cleavage line. Note also that neutralism in Sweden was not a distinctly socialist position, but was accepted by all major political parties. In both Norway and Denmark, the socialist left has replaced the old pro-Soviet communist left which vanished as sizable parties. In Denmark, the crowdedness on the left is complicated by smaller parties. The Left Socialists (Venstresocialisterne) and the Communists (Danmarks Kommunistiske Parti) have joined forces in general elections in the 1990s on the so-called Unity List (Enhedslisten). In Norway, both the Communist Party (Norges Kommunistiske Parti) and the Workers' Communist Party (Marxist-Leninist) (Arbeidernes Kommunistparti Marxist-Leninisterne) have been electorally insignificant.

As part of the "New Left's" bridge building neutralism, the left socialists represented a break with what until then had been labeled left politics. The "New Left" placed itself between the social democrats and the old communist parties and gave priority to issues beyond the traditional class cleavage. Anti-militarism, anti-hierarchy, solidarity with the Third World, and opposition to the development in the socialist states of Eastern Europe formed the ideological backbone of the movement. In the 19705, this ideological platform made adherence to the EEC impossible. Valen (1985) also stresses that left-wingers have perceived NATO within the framework of the left-right dimension as an economic and political power, first and foremost identified with American capitalism.

The abnormal and only non-socialist party in the Scandinavian context
with an explicit background in foreign policy opposition is the Danish
Radical Liberals (Det Radikale Venstre). The party was established on the

basis of anti-militaristic attitudes within the Liberal Party (Venstre) in 1905, a background that even led it to vote against Danish NATO membership in 1949 (Larsen 1994). The party has a history of close cooperation with the Social Democrats (Socialdemokratiet) and did not accept the question of NATO membership before it entered into a coalition with them in 1957. The Radical Liberals have functioned as a pivotal party in economic policy, while simultaneously breaking with the traditional left-right cleavage in foreign policy, unlike its sister parties in Norway and Sweden (Schou & Hearl 1992).

Left wing positions on the security dimension were heavily altered in the aftermath of the 1989 Eastern European revolution. Pro-Sovietism was wiped off the political agenda at the same time as the background for both pro and "bridge building neutralism" lost their meaning. The onedimensional communist/anti-communist cleavage pattern was replaced by a more complex security scenario. The left was confronted with new policy dilemmas - security now means more than nuclear threats. The East-West axis, however, remains an important dividing line. Social Democrats aimed at enlarging the European Union, a solution where EU could fuse the security and economic dimensions in an integrated Europe. Still, however, the Nordic Social Democratic parties remain committed to NATO and are at the same time constrained by an EU-skeptical national electorate. The left socialists, on the other hand, point towards lighter solutions such as the OSCE, the European Council, and the UN. As true sixties parties, Norwegian and Danish left socialists had no ideological baggage to shed when communism collapsed in 1989. The Swedish Left Party, however, was heavily affected. As a part of the "old left," its history has been a never ending fight between "reformers" and "dogmatist."⁴ A pro-Moscow group from the districts of Norbotten and Göteborg left the party in 1977 and established the Communist Workers" Party (Arbetarpartiet Kommunisterna). In the 19605, the Swedish Left Party tried to orient itself towards the position of the "New Left." The party has even changed its name twice as a symbol of a redefined party identity. For Danish and Norwegian left socialists it was unthinkable in the 1980s to send economic delegations to the Soviet Union and party members to theoretical conferences in Yugoslavia, as the Swedish Left Party actually did until the bitter end. However, pro-Sovietism was obviously not an electoral platform that could make voters rally around the party, and the Swedish Left Party, as many other pro-Soviet parties, tried to hide its "true self" behind the concept of neutrality.'¹ In 1967, the party took the name the Communist Left Party (Vänsterpartiet Kommunisterna) and finally tried to wipe out the old "dogmatists" in 1990 when the party assumed its current name. After the end of the Cold War the security dimension has become more complex. multidimensional and less ridden by the black/white ideological demonology created in the heyday of the Cold War. In the post-1989 period, security has

become a matter of degree among left-wingers, not a question of life or death
within ideologically divided camps.

The EU dimension appeared on the agenda in both Denmark and Norway in the late 19605. The issue attacked the relatively peaceful situation that prevailed among and within parties (illustrated by Table 2). In the Danish and Norwegian cases, the divide can be traced all the way back to the 1972 referendum (Siune, Svensson & Tonsgaard 1994). Note that a majority of the Socialist People's Party's voters still veto membership in the European Union.

In the Swedish case, the 1994 EU referendum introduced foreign policy as a permanent dimension of conflict in the party system. In both Denmark and Norway, foreign policy has been a major cleavage between Social Democrats and left socialists. How have the different foreign policy settings expressed themselves at the parliamentary level?

Foreign Policy Distance: Consensus or Conflict?

Policy-based coalition theories assume that parties are policy motivated. The possibility of affecting public policy outputs is crucial when parties decide whether or not to enter coalitions. However, a party with a particularly distinct policy in one area, a policy that appears to be a cornerstone of its electoral appeal, may refrain from entering a coalition if that policy is not accepted by its partners (Ware 1996, 337). Thus, one would expect parties to enter coalitions that minimize the ideological distance between coalition

partners. One should also expect that when a party aims at office, it indicates that it sees the possibility of moving public policy in its preferred direction. The fundamental question is which parties cooperate on foreign policy issues within the three countries? Let us look at the two dimensions previously outlined, i.e., security and European integration.

In the Danish case, the period 1982 to 1988 was marked by foreign policy convergence between the opposition parties. Damgaard (1992, 204) argues that the Danish parliamentary system in this period developed a new form of "minority rule" in foreign policy making. The Conservative-led "four-leaf clover" government (Conservatives (Det konservative Folkeparti), Liberals (Venstre), Center Democrats (Centrumdemokraterne) and the Christian People's Party (Kristeligt Folkeparti)) faced a united opposition on foreign policy issues in this period. It survived for eight years despite a weak parliamentary basis. It governed on the basis of agreements with the Radical Liberals on economic policy (Damgaard & Svensson 1989, 738). In foreign policy, the government faced the so-called "alternative majority." The Social Democrats, Radicals Liberals and the Socialist People's Party converged on several foreign policy issues in this period, especially nuclear arms policy. The "alternative majority" pushed through several issues: Denmark vetoed the stationing of nuclear missiles in Europe, advocated a nuclear arms "freeze," and the opposition ordered the government to work for the creation of a Nordic nuclear free zone. The government had to implement decisions it vigorously opposed. Obviously, the defeats in security policy could be too heavy to bear. A formalized governmental cooperation based on the "alternative majority" was, according to the Socialist People's Party, a credible alternative, but the question of European integration ruled out such a coalition.

The debate on Denmark's relationship with the EU falls into three phases (based on Petersen 1995): 1) The period 1973-86 was marked by a passive political strategy, a period that certainly does not place Denmark among the most "Euro-centric" countries; 2) from 1986-92 the country followed a more active policy combined with focus on integration along specific policy areas: and 3) the rejection of the Maastricht Treaty in 1992 put Denmark in a more reactive position, the Maastricht "opt-ouls" set the frame for the country's current EU policy.

How was the "alternative majority to perform along the EC EU dimension? Petersen (1994, 220) underlines thai the referendum on the Single European Act (SEA) in 1986 removed the question of membership from Ihe political agenda and placed the development in the EU at the center of the political debate. The conflict over Europe developed from what Sehattschneider (1975) calls a "high priority" conflict with a permanent and stable conflict pattern to a fight about setting the agenda for future Danish EU

policy. The traditional membership cleavage was played down and replaced
with that of degree of integration on different policy areas (Worre 1992, 122).

The "alternative majority" found a common platform in their attitudes towards European integration. Among other things, they turned against the rebirth of the Western European Union (WEU) as a forum for EC members to discuss security policy (Petersen 1994, 200). While the government had a positive view of the WEU, the Social Democrats thought Denmark should keep out of the organization. The "alternative majority" also confronted the government in the debate over the Single European Act in 1986. A parliamentary majority opposed the Act, which led the government to propose a referendum on the issue. A comfortable majority of 56.2 percent of the voters supported the Act.

However, policy disagreements still existed among the three opposition parties. The Social Democrats and the Radical Liberals did not share the Socialist People's Party's view on EU membership. They both saw EU mainly as a question of economic cooperation, but they dismissed the tendency towards a political union (ibid., 216). The Social Democrats reevaluated their attitude in the aftermath of the East European revolution. The party chairman at the time, Svend Auken, declared in a speech in 1990 that the party had given the conservative forces in Europe too much space, and it was time for the party to strengthen the EC in their own image (Haahr 1993).⁷ In the 19905, the Socialist People's Party followed, first convinced and later half-heartedly.

The EU issue made the Socialist People's Party leadership believe in the possibility of reshaping the "alternative majority" into a coalition government. The Danish "no" to Maastricht on June 2, 1992 was followed by intense negotiations between the political parties. Again the three parties united in the so-called "national compromise" which became the solution to the political crisis created by the Danish veto on Maastricht. The "national compromise" was initiated by the Socialist People's Party in May 1992 before the Maastricht referendum (Christensen 1996). By December 1993, the party leadership finally accepted the question of membership when it approved the Edinburgh Agreement. Although only 20 percent of its voters went along with the leadership in the 1993 referendum, the Socialist People's Party was for the first time in a position to determine the direction for Danish EU policy.⁸ An extraordinary EU party congress decided to acknowledge the leadership's decision in 1993. The party stated that a precondition for its EU membership approval was "that the fundament for Danish policy towards Europe is the national compromise."⁹

The Swedish case contradicts the Danish one in several ways. First of all, we have seen how foreign policy has been at the center in Danish politics. In Sweden, foreign policy issues have centered on a debate of a strongly normative character (foreign aid, South Africa, human rights in other

countries, and peace and disarmament) (Goldman, Berglund & Sjostedt 1986, 109-10). Swedish neutrality has taken key foreign policy issues off the political agenda and fostered consensus among the political parties. Kite (1996,117) finds that in Sweden, in contrast to Denmark and Norway, the EU membership question has been very much an international security issue. Before 1990, all parties agreed that membership was impossible.¹⁰ In faet, the Left Party has traditionally seen Sweden's foreign policy as a model for other countries. However, this has been combined with lots of support for Marxist oriented liberation armies in every jungle of the world and Marxist-Leninist dictators in the Third World. Consequently, the Left Party's support for Swedish neutrality does not mean that the political distance between the party and the Social Democrats has not existed. In the pre-1989 communist/anticommunist environment, the Left Party traditionally identified itself with Soviet communism and thus made itself unacceptable as a coalition partner during the Cold War. The collapse of communism led the party into a severe ideological crisis which threatened to dissolve the party, and it decided to "freeze its international contacts."¹¹ However, the EU issue appeared on the scene in Swedish politics and made it possible for the Left Party to avoid addressing its own internal confusion. The party leadership decided to throw itself heavily into the anti-EU campaign instead of facing the discussion of how to deal with its communist past (Hammar 1992, 25). By chance, the "New Left" parties in Denmark and Norway became identification models at a time of internal crisis.¹" At the May 1990 congress, the Left Party disposed of the communist label in its name, but the party membership was strongly divided. An internal membership poll showed that 41 percent wanted to keep the old name, while 41 percent wanted to change it (Vånsterpress 1990b).¹³ It was a party marked by internal divisions and defections that entered the 1991 general elections. Several prominent party members saw the renewal process as incomplete, claimed that they wished to be part of a political party not a "church," and left (Vånsterpress 1990a).¹⁴

In Norway, foreign policy gave birth to left wing opposition among Labor Party grassroots. This made foreign policy issues a question of identity for the socialist left. In Norway, NATO membership has received strong support across the political spectrum, with opposition only during periods of Soviet disagreements (Valen 1985, 29).^1:> The Norwegian Labor Party (Arbeiderpartiet) grassroots have been much more pro-NATO than their Danish counterparts. Among Norwegian Labor voters, the support for NATO membership increased from 70 percent in 1965 to 90 percent in 1981 (ibid.). Danish Social Democratic supporters have been more reluctant towards NATO, and support for membership stabilized around 65 percent in the 1980s.¹⁶ In Norway, the socialist left has been alone in its struggle against NATO membership. The issue has not created severe parliamentary confrontations as described in the Danish case. However, the Norwegian Labor

Party, when it was in opposition in the beginning of the 1980s, promoted the same security policy as its Danish counterpart. The question of EU membership has, however, affected the relationships between the Norwegian parties. Opposition to Norwegian EU membership cut across the left-right cleavage and made non-socialist cooperation a fictional alternative. In the 19905, the parliamentary opposition to membership was broader in Norway than in Sweden.¹⁷ In Sweden, 88 percent of representatives supported EU negotiations, but only 65 percent of the Norwegian Storting members favored membership negotiations (Kite 1996, 135). The Center Party (Senterpartiet), previously keen on non-socialist cooperation, decided to leave the government in 1989 and pointed to the Labor Party to handle the European Economic Area negotiations with the EU.^lIS From 1990 to 1994, Norwegian politics was dominated by the fight over membership, a competitive environment with no room for coalition building.

In the Danish case, the Radical Liberals, the Socialist People's Party and the Social Democrats united on several foreign policy issues in the period 1982-1993, first along the security dimension and later they rejoined to handle the problems created by the rejection of the Maastricht Treaty in 1992. In the Swedish case, foreign policy has been a conflict between Social Democrats and left-wingers after 1989. The Swedish Left Party's opposition to EU membership has made the party an even more unwanted coalition partner in the eyes of the Social Democrats. In Norway, the security dimension did not create any severe conflicts at the parliamentary level. However, the EU dispute ruined non-socialist cooperation and brought potential new coalition patterns on the agenda, at least within Socialist Left Party ranks.

How do the foreign policy positions in the three systems correlate with
official party positions on different coalition alternatives?

Internal Constraints on Coalition Formation

Table 3 shows left wing parliamentary strength and fragmentation. Note the
short period between elections in Denmark, and an electorally weak and
fragmented left wing in the Danish Folketing.

In Denmark the socialist parties have not been in a majority position since the 1966 election.¹⁹ Consequently, it has been necessary for both the Social Democrats and the Socialist People's Party to include one of the bourgeois parties as a coalition partner. The inclusion of the Radical Liberals is naturally based on its previous cooperation with the Social Democrats and the foreign policy convergence described here.

Note also that the left wing in Norway and Sweden has formed a majority
in considerable parts of the period. The Social Democratic parties of Norway

and Sweden have both been able, when in government, to lean on shifting alliances in parliament, joined by the left wing opposition on some issues and one or several of the bourgeois parties on others (Strøm 1990. 225-26). Thus, both Swedish and Norwegian Social Democrats have confronted a less complex parliamentary arena than the Danish Social Democrats.

The period of strong foreign policy agreements within the Danish "alternative majority" corresponds directly with weak bargaining constraints on the Socialist People's Party leadership (Table 4). The SEA referendum in 1986 resulted in immediate changes in the party's coalition strategy. The National Congress in 1986 accepted the leadership's view that withdrawal from the EC would no longer be an ultimatum regarding potential governmental cooperation with the Social Democrats."¹

Before the June 1997 Amsterdam meeting, the "national compromise" determined the direction for a Socialist People's Party would-be government's EU policy. In 1998, Denmark will face its fifth EC/EU referendum, this time on the Amsterdam Treaty. The Socialist People's Party will enter the campaign marked by a strong internal conflict. Before the party's extraordinary EU congress in September 1997, the debate between its different EU fractions was intense. "Euro-optimists" and "Euro-realists" came out with the smallest majority possible in the parliamentary group (seven against six), while the "Euro-opponents ' dominated where it really mattered, in the National Executive and at the local party level.²' Evidently, the party leadership miscalculated when they believed that the "national compromise" could be used as a springboard to a common governmenlal declaration with

the Radical Liberals and the Social Democrats. The Radical Liberals had no intentions of forming a coalition government that included the Socialist People's Party, and in 1993 the Social Democrats took the initiative to form a majority government with the parties at the center of the political system (the Christian People's Party and the Center Democrats). While the Social Democrats²² in 1987 had opened for a possible governmental cooperation with the Socialist People's Party, the door was closed in 1990. After more than ten years in opposition, the Social Democrats headed a "broad majority coalition" (Bille 1991, 40).

Why did the Socialist People's Party fail in its governmental ambitions? The Danish case underlines the huge difference between decisions to enter governmental coalitions and decisions to form parliamentary coalitions. Parliamentary coalitions usually rally around specific policy arenas, while government coalitions have to deal with a wide range of issues. The Radical Liberals' key argument against any cooperation with the Socialist People's Party has always followed the traditional left-right dimension. This is hardly surprising since the Socialist People's Party usually turns against all budget proposals in parliament. Except for the 1996 Social Democratic-led minority government's budget proposal, it has opposed all budget proposals since 1981. In addition, a coalition between two parties divided on European integration (Social Democrats and Radical Liberals) and the Socialist People's Party united in its opposition to the United States of Europe could indicate problems in EU negotiations. More so when the Socialist People's Party vigorously declared itself "watchdog" when it came to the Danish optouts and the Edinburgh Agreement. Euro-skeptics among Radical Liberal and Social Democratic grassroots could also lean towards the Socialist People's Party in order to ensure that the "national compromise" became governmental EU policy.

In the 19705, the Swedish Left Party claimed that the "Social Democrats
mainly restricted themselves to the conduct of capitalism" (VPK-Informationennforma-

1976). Although the party periodically has been in a key parliamentary position, it still more or less passively supported Social Democratic governments up until the mid-1980s (Sjolin 1993, 36). However, their support became less unconditional as the competitiveness among the Swedish parties grew stronger (Sannerstedt & Sjolin 1992, 106). From 1976 to 1982. the Social Democrats were expelled from office by a non-socialist majority. When the Social Democrats returned to office after the 1982 general elections, they had to rely on a more reluctant Left Party (Sjolin 1993, 45^8). The traditional backing of Social Democratic policy became a permanent internal conflict in the Left Party. At the 1985 party convention. many motions concerned the relationship with the Social Democrats. The party leadership was seen as being too gentle with the Social Democrats and the unions (VPK-Informationen 1985). In addition, the Green Party's entrance into the Riksdag in 1988 forced the Left Party to promote a more active parliamentary strategy. It now confronted the Social Democrats and demanded negotiations in different policy arenas. For the first time in the history of Swedish politics, the Left Party helped overthrow the Social Democratic government in 1990, when it refused to accept the cabinet's proposal for a ban on strikes and wage increases. However, the Left Party supported the formation of a new Social Democratic government. Its opposition to EU membership was backed by a united party organization. and currently it opposes Swedish entrance into the European Monetary Union (EMU) and fights the cabinet's convergence policy."" The Social Democratic government's adaptation to EU policy is the main dividing line between the two parties. A less reliable Left Party has also made the Social Democratic Party reformulate its parliamentary strategy and lean towards the Center Party for support. The old coalition partners from 1951-57 rejoined after the 1994 general elections on several issues in the Riksdag. The introduction of the EU dimension in Swedish politics has created a new arena²⁴ for left wing opposition and helped the Left Party change the party's image as a ghost from the Cold War. The EU dimension united the party, gave it a "new start" and increased its electoral support to 6.2 percent in the 1994 general elections and to a record-high 12.9 percent in the 1995 European elections (Widfeldt 1996). The voters apparently believe that the Left Party has finally renounced its communist past.

In Norway, the Left Socialist Party has primarily aimed at pushing a governing Labor Party to the left. Former party leader Finn Gustafsen underlines that the Labor Party, up to 1963. treated the Left Socialist Party as a fraction within its own ranks that eventually would fall into place. When the Socialist People's Par!y^2> contributed to the termination of the 1963 Labor government on the so-called Kings Bay issue.²'¹ Labor could no longer take its support for granted. Nevertheless, the party has never been in a situation where it has been necessary to take concrete initiatives to formalize the

cooperation between the two parties. The Socialist Left, on the other hånd, has been more inclined to enter into a governmental cooperation with the Labor Party over time (Sørensen 1988). The EU dispute resulted in a reevaluation of the Socialist Left's parliamentary strategy at the 1993 Party Congress. Encouraged by the opinion polis, the leadership aimed at governmental cooperation with Labor in the 1993 election.²⁷ However, party leader Erik Solheim's invitation to the Labor Party triggered strong internal opposition. The 1993 Party Congress sought to include the hard-core anti-EU Center Party in its governmental ambitions together with their previous invitations to the Labor Party. The congress made it clear to the party leadership that it opposed governmental cooperation with Labor without demanding withdrawal from the European Economic Area agreement (EEA). A congress compromise constrained the leadership when it decided that a potential coalition government with Labor should not stop Norway from withdrawing from the EEA if it wished to do so (Seierstad 1996,104). In 1996, the party newspaper, Ny Tid, presented a poll in which 15 of the Socialist Left's 17 members of the National Council supported a government that included the anti-EU parties. The combination of a salient EU dimension and a parliamentary strategy aimed at governmental power was dynamite among Socialist Left members. It touched on the very identity structures of the party. Since the 1993 general elections, the party has been in a more or less permanent internal conflict,²⁸ which is one of the reasons why party leader Erik Solheim decided to step down at the 1997 Party Congress.

Conclusion

We started with two questions: under which circumstances do parties redefine their self-imposed foreign policy constraints? What are the motives when parties reformulate their foreign policies? In a review article on coalition theory, Grofman & Van Roozendaal (1997) agree with Narud (1996) that political parties change from a cooperative to a competitive strategy when "heartland" issues become salient. Strøm et al. (1994) also point out that parties are internally constrained when they enter bargaining processes, policy distance and ideology matter to party members and they, through party congresses, usually take the final decision. Internal constraints are important, precisely because policy positions are closely related to the very identity structures of political parties. Strøm et al. add external constraints to the list, the most important being the numerical strength of potential coalition parties (ibid.). We have highlighted three variables: the content of the EU issue, foreign policy distance and the structure of government.

How do we explain why the Danish Socialist People's Party moved from
competition to cooperation in 1993, and why the two other parties aimed at
competition? How do the three parties score on our three variables?

The content of the EU issue: the Socialist People's Party aimed at cooperation in 1993 although the EU issue was fundamental to party identity. This can be used against the "heartland" theory. Coalition theory often fails to account for the content of politics. Content matters, and the contents of issues vary both over time and among countries. Political parties cannot choose freely among existing options when they formulate their parliamentary strategies. They are constrained by how issues are defined. The dimensions of conflict determine the strategy of politics. "High priority"' conflicts, with unified majorities and minorities, are followed by stability in politics. The Danish leftists confronted a conflict over Europe where the question to answer was the degree of integration, a question that opened for negotiations among the parties. In Sweden and Norway, the EU issue was "high politics," the question to answer was "yes" or "no," a situation that definitely ruled out any bargaining processes. This explains why the Socialist People's Party followed a cooperative strategy and became interested in involving itself in the policy process. In our case, self-imposed, left socialist foreign policy positions put strong internal constraints on the party leadership's bargaining options. For the Danish Socialist People's Party and the Norwegian Socialist Left, opposition to NATO and EU has been fundamental for party identity. The collapse of communism did not affect the identities of these parties. The Swedish Left Party's identity, on the other hand, is rooted in the classical onedimensional cleavage between communists and anti-communists. 1989 therefore meant a time of confusion for the party. According to party identity, we would expect changes in EU policy in neither case. For the Norwegian Socialist Left and the Swedish Left Party, this is evidently so - none of them had incentive to change their policy. In Norway, members of the Socialist Left have EU opposition in their "blood." Although it is hard to find a more united anti-EU party in Europe, the party leadership even faced strong internal criticism for being too soft in its opposition to EU membership. For the Swedish Left Party, the EU issue arrived on the scene at a time when the party was struggling to survive. By holding on to Swedish neutrality as its prime policy objective, the party was capable of uniting at the same time as it helped wash the Cold War stains off its clothes.

Foreign policy distance: Foreign policy issues have not only been salient on the agenda in Danish politics, they have also led to a convergence among the opposition parties. Convergence along the security dimension among the Socialist Democrats, the Socialist People's Party and the Radical Liberals made a governmental coalition possible, at least in the minds of the leadership of the Socialist People's Party. The three parties even joined in the "national compromise" in order to renegotiate the Maastricht Treaty. The Radical

Liberals' foreign policy was acceptable to the Socialist People's Party. In the Swedish case, foreign policy conflicts have not dominated the political agenda and there has been no legislative coalitions. In the Norwegian case, the Labor Party could rely on bourgeois parties to push through foreign policy issues on the legislative arena.

The structure of government: The structure of government is the most obvious external constraint in coalition politics. Counting parliamentary seats is the essence of coalition bargaining. Unstable and weak minority governments led the Danish Socialist People's Party to reformulate its parliamentary strategy. It aimed at office and convinced the party membership that it had to soften its foreign policy demands if it was to head for political power. The complex situation that faced the Danish Social Democrats also made it include the Socialist People's Party in its parliamentary strategy in the mid-1980s. In the Norwegian and Swedish cases, the Social Democrats have been left socialist "office blockers." Both the Norwegian Labor Party and the Swedish Social Democrats have been strong enough to effectively block left socialist governmental ambitions.

The Danish Socialist People's Party scores high on all three variables, while the Swedish Left Party and the Norwegian Socialist Left score low. This explains why it was possible for Denmark's Socialist People's Party to reformulate its policy. Norwegian and Swedish left socialists did not face the same external pressure.

---

NOTES

1. Left socialist parties differ from traditional socialist parties. Kitschelt (1990, 180) defines left socialist parties as "left libertarians," they "accept important issues on the socialist agenda (mistrust of the marketplace), but reject traditional socialism's paternalist-bureaucratic solutions." Left socialists represent the "new politics": feminism, environmentalism and pacifism (Sassoon 1996). The so-called "old left" has its roots in the material oriented labor movement, while the "new left" shows sympathy for the new social movements.

2. The decision of the Norwegian Christian People's Party not to enter the non-socialist majority coalition after the 1981 election is a good example. Abolition of free abortion was a key issue for the party and it would not enter a coalition government that committed itself to a change in the abortion law (Sarlvik 1983, 123).

3. Goldman, Berglund & Sjöstedt (1986, 25) argue that the "chief problem of definition concerns the foreignness of the activity called foreign policy." They further stress that international policy may be a better word and that it is hard to find non-international issues if, "international is defined in terms of external dependence." Obviously the dividing line between international and domestic politics is blurred and a number of issues today involve bargaining with other countries.

---

4. For a discussion of ihe Left Party's history, see Hermansson 1984; Sparring 1973; Kokk 1974; Olsson 1976 and Lindkvist 1989.

5. VPK-Informationen 1983 (membership paper).

6. As pointed out by Sassoon (1996, 209) "this does not mean that all or even most communists were not sincerely committed to peace, or that pacifists were communist stooges or political imbeciles." For a discussion on communist damage control. see Bull & Heywood 1994.

7. The Social Democratic Party wanted to strengthen the European Parliament's powers vis-å-vis the Commission, close the so-called democratic deficit, and make it easier for new member states to join (ibid., 211-14).

8. The decision in the National Executive was taken with a majority of 25 against 3. The minority agreed to negotiate in case of a "no" in the June 1992 referendum, but maintained that the Edinburgh Agreement lacked clear-cut changes in the Maastricht Treaty (Socialist People's Party. Statement from the National Executive minority to the extraordinary national congress, March 13-14. 1993).

9. The Socialist People's Party recommended a "ves" to the Edinburgh Agreement. Socialist People's Party. Decision taken by the National Executive on December 13. 1992.

10. In December 1990, changes in the international system made all parties, except the Left Party and the Greens (Miljopartiet de grona). agree that EU membership was compatibie with neutrality (Kite 1996).

11. The Left Party: Report to the 1993 Party Congress. p. 9.

12. The contacts between the Danish and Norwegian Left Socialists have tradhionally been close. A regular cooperation that included the Swedish Left Party was not established until after 1989. The Left Party's report to the 1993 Party Congress. p. 20.

13. Vansterpress is the Left Party's membership paper. However. 68 percent approved to replace the old party program. 5,600 out of 13.000 party members answered the questionnaire.

14. Several key politicians felt that the renewal process was half-hearted. Before the 1991 general elections, Margo Ingvardsson. Ylva Johansson and Bo Hammar vctoed renomination and left the party. In addition, Annika Åhnberg and John Andersson left the party's parliamentary group (The Left Party: 1993 and 1996 Party Congress reports).

15. The support for NATO membership has increased. 44 percent supported membership in 1965, 56 percent in 1969, 67 percent in 1973, and 72 percent in 1981 (ibid.).

16. The exact figures are 66 percent support in 1986. 58 percentjn 1988. 60 percent in 1989, and 66 percent in 1990. See Dansk Udenrigspolitisk Årbog (Danish Foreien Policy Yearbook) 1986, 1988, 1989 and 1990.

17. The opposition in Norwav was even stronger in 1992 than in 1961. In 1961. 25 percent of the Storting opposed the government's membership position (Kite 1996).

18. See Narud 1996 for a good analysis of the termination. See Christensen 1997a for a discussion of the Norwegian Center Party's parliamentary strategy.

19. The arrival of the Progress Party (Fremskridtspartiet) in 1973 has made non-socialist cooperation difficult in Denmark. The party has been seen as non-ministerial by the bourgeois parties (Sarlvik 1983. 120).

20. However, the party stressed that "exit from the EC could only be possible if the EC blocked the politics of a majority in the Folketing" (Christensen 1996. 532).

21. The confusion at the leadership level is complete. The party chairman. Holger K. Nielsen, opposes the Amsterdam Treaty. the vice-chairman. Christine Antorini. supports it and is joined by the parliamentary leader. Steen Gade. At the local party level, "Euro-opposition" prevails. A clear majority (28 against 10) in the National Executive decided to recommend the extraordinary September EU congress to reject the treaty (Dagbladet Information 1997a). In a newspaper survey of local party chairmen, two out of three opposed (Morgenavisen Jyllands-Posten 1997). The survey was basedon interviews with 137 out of 201 local party chairmen. At the extraordinary congress, 67 percent of the delegates rejected the Treaty, but the congress statement opened for a "ves" in case of improvements in a possible renegotiated treaty (Dagblader Information 1997b). The congress decision made parliamentary leader Steen Gade step down in order to avoid intensified conflicts within the party. Prior to the congress, the division was so intense that key party members proposed that the party should refrain from taking a decision (Politiken 1997). The idea was inspired by the Finnish Left Party which, before the 1994 EU membership referendum, could not agree on the membership issue and thus decided to have no official party position.

---

22. Even in 1987, skepticism prevailed at the Social Democratic leadership level towards including the Socialist People's Party in the government. Many thought it would be difficult to formulate a coherent NATO policy with the Socialist People's Party as a government party (Boel 1988, 228).

23. See biannual report to the 1996 Party Congress, p. 14. The Social Democrats decided in June 1997 that Sweden should not enter the EMU from the start in January 1999. Prime Minister Göran Persson explicitly referred to a critical and uncertain Swedish EMU opinion as one of the most important reasons for the decision (Svenska Dagbladet 1997). This also means that the EMU issue does not have the same potential to become a key issue in the upcoming 1998 general elections.

24. Christensen's 1996 study of the level of activity within the Swedish parliament's EU Committee shows that the Left Party, together with the Greens, is responsible for 33 percent of the activity, even though the two parties only command two out of seventeen representatives in the Committee.

25. The Socialist People's Party prefigured the Socialist Left in 1961. The fight against EEC membership in 1972 consolidated the Socialist People's Party with other leftwingers into a broad election coalition in 1973. The process culminated when the Socialist Election Alliance (Sosialistisk Valgallian.se) was renamed the Left Socialist Partv in 1975.

26. 21 people lost their lives in an accident in the state-run Kings Bay mine at Svalbard in 1962. The motion of no confidence against the government was basedon the fact that the government had not followed up on the parliament's decision to improve the

27. The opinion polls before the election showed a record-high level of support for the Socialist Left, between 15 and 20 percent. (Seirstad 1996).

28. For an overview of the internal opposition's main arguments, see a special issue of the left-wing journal Vardøger. "SV på Sotteseng" (1996).

ACKNOWLEDGEMENTS

I thank Erik Damgaard, Ulf Lindström, Lars Svåsand and two anonymous reviewers for valuable comments. This article is part of the research program "ARENA" (Advanced Research on the Europeanization of the Nation-State) funded by the Norwegian Research Council. NorFa (Nordic Academy for Advanced Study) has given me financial support to work in a very stimulating environment at the Department of Political Science, University of Aarhus.

REFERENCES

Bille, L. 1991. -Politisk kronik 2. halvår 1990," Økonomi oi; politik. DJØF-Forlaget.

Bull. M. J. & Heywood, P. 1994. West European Commumst Parties after the Revolution of
1989. London: St. Martin's Press.

Christensen. D. A. 1996. "The Left-Wing Opposition in Denmark, Norway and Sweden: Cases
of Euro-phobia?," West European Polities 19, 3, 525-46.

Christensen, D. A. 1997a. "Adaptation of agrarian parties in Norway and Sweden," Partv
Polities 3, 3, 391-406.

Christensen, D. A. 1997b. "Europautvala i Danmark, Sverige og Norge: Sandpåstrøingsorgan
eller politiske muldvarpar/. Nordisk Ådmmistrali\'t Tidsskrift 78, 2, 143—62.

Dagbladet Information. 1997a. August 18.

Damgaard, E. 1992. Parliamentary Change in the Nordie Countries. Oslo: Scandinavian
University Press.

Damgaard, L. & Svensson, P. 1989. "Who Governs? Parties and Policies in Denmark,"
European Journal of Political Research 17, 731-45.

Dciusk Udctifi^spodtisk At'ho^ (Oiinish FFHorci^n Policy YCtirhook) 1 vHfi, I^SH, I^HV) tv I^'JO.

Goldman, K., Berglund, S. & Sjostedt, G. 1986. Democracy and Foreign Policx. The Case of
Sweden. Gower: Aldershot.

Grofman, B. & Van Roozendaal, P. 1997. "Review Article: Modelling Cabinet Durability and
Termination," British Journal of Political Science 27, 419-51.

Haahr, J.-H. 1993. Looking to Europe: The EC Policies of the British Labor Party and the
Danish Social Democrats. Aarhus: Aarhus University Press.

Hammar, B. 1992. Ett langt fan-ål till kommunismen. Stockholm: Brombergs.

Heidar, K. & Svåsand, L. 1994. Partiene i en brymingstid. Bergen: Alma Mater.

Kite, C. 1996. Scandinavia Faces EU, debates and decisions on membership 1961-1994.
Umeå University: Department of Political Science.

Kitschelt, H. 1990. "New Social Movements and the Decline of Party Organization." in
Dalton, R. J. & Kuechler, M., eds.. Challenging the Political Order. New Social and
Political Movements in Western Democracies. Cambridge: Polity Press.

Kokk, E. 1974. VPK och SKP. De gralande n-illingpartierna. Stockholm: Tidens Forlag.

Larsen, H. 1994. "Efter P. Munch. De Radikale og dansk udenrigspolitik 1945-1992."
Vandkunsten 9/10, 285-314.

Laver. M. L & Budge, 1., eds. 1992. Party Policy ami Government Coalitions. Basingstoke: St.
Martin's Press.

Lindkvist, K. 1989. "Vad skulle VPK heta'?." ZENIT 4. 73-84.
Morgenavisen Jyllandsposten. 1997. July 12.

Narud, H. M. 1996. Voters, Parties and Covcrnmcnts. Elcaoral competuum, Policy Distance
and Government Formation in Multi-Party Systems. Oslo: Institute tor Social Research.
Report 96:7.

Ny Tid. 1996. No. 5, Fcbuary 02.

Olsson, S. E. 1976. Från SKP till VPK - en antologi. Lund: Bo Cavefors Bokforlag, idem
1978. Utanfor systemet. Yanstern i Sverige 1968-78. Stockholm: Råben & Sjogren.

Petersen, N. 1994. ■■Vejen til den europæiske Union 1980-93." in Swienty. T. cd.."Danmark i
Europa 1945-93. København: Munksgaard. 195-271.

Petersen. N. 1995. "Denmark and the European Community 1985-93," in Due-Nilsen. C. &
Petersen. N„ eds.. Adaption & Activism. The Foreign Folier of Denmark 1967-1993.
København: DJØF-Forlaget. 189-224.

Pierre, J. & Widfeldt. A. 1992. "Sweden." in Kat/. R. & Mair, P.. eds.. Party organizations: A
Data Handbook. London: Sage.

Politiken. 1997. July 26.

Sannerstedt. A. and Sjolin. M. 1992. "Sweden: changing party relations in a more active
parliament." in Damgaard. F... ed.. Parliamentary Clninge in the Sordic Countries. Oslo:
Scandinavian University Press.

Sassoon, D. 1996. One Hundred Ycars ofSocialism. The West European Lefl in the Twentieth
Century. London: 18. Tauris Publishers.

Schattschneider, E. E. 1960/1975. The Semisovereign People. A Realist's view of Democraey
in America, llinsdale: The Dryden Press.

Schou, T.-L. & Hearl, D. J. 1992. "Party and Coalition Policy in Denmark." in Laver. M. J. &
Budge, L, eds.. Party Policv and (iovernmeni (\>iilitions. Basinustokc: St. Martins Press.

Seierstad. D. 1996. "Med SF og SV mot ED." Vanhgcr 2396. 92-119.

Sjolin, M. 1993. Coalition Politics and Parliamcmarv Power. Lund: Lund University Press.

Sparring, Å. 1973. "The Communist Party of Sweden," in Upton, A. F., ed., The Communist
Parties of Scandinavia and Finland. London: Weidenfeld.

Stenersen, 0. 1977. "Venstrekreftene i Norsk politisk historie 1945-1965," in Bergh, T. &
Pharo, H. 0., eds., Vekst og Velstand. Norsk politisk historie 1945-1965. Oslo:
Universitetsforlaget.

Strøm, K. 1990. Minority Government and Majority Rule. Cambridge: Cambridge University
Press.

Strøm, K. & Leiphart, J. 1992. "Norway: Policy Pursuit and Coalition Avoidance," in Laver,
M. J. and Budge, 1., eds., Party Policy and Government Coalitions. Basingstoke: St.
Martin's Press.

Strøm, K., Budge, I. & Laver, M. J. 1994. "Constraints on Cabinet Formation in Parliamentary
Democracies," American Journal of Political Science 38, 2, 303-35.

Svenska Dagbladet. 1997. June 4.

Svensson, P. 1993. "The Danish Yes to Maastricht and Edinburgh. The EC Referendum of
May 1993," Scandinavian Political Studies 17, 1, 69-82.

Svåsand, L. & Lindström, U. 1996. "Scandinavian political parties and the European Union,"
in Gaffney, J., ed., Political Parties and the European Union. London: Routledge.

Särlvik, B. 1983. "Coalition Politics and Policy Output in Scandinavia: Sweden, Denmark and
Norway," in Bogdanor, V., Coalition Government in Western Europe. London: Heineman
Educational Books.

Sørensen, K. H. 1988. "Sosialdemokrati som læringsprosess." Vardøger 18/88, 221-27.

Valen, H. 1985. "Cleavages in the Norwegian Electorate as a Constraint on Foreign Policy
in Holst, J. J., ed.. Norwegian Foreign Policy in the 1980s. Oslo:
Universitetsforlaget, 26-54.

Vardøger. 1996. No. 23.

VPK-Informationen. 1976. No. 6.

Ware, A. 1996. Political Parties and Party Systems. Oxford: Oxford University Press.

Widfeldt, A. 1996. "Sweden and the European Union. Implications for the Swedish party
system," in Miles, L., ed., The European Union and the Nordic Countries. London:
Routledge, 101-116.

Worre, T. 1992. 'Tolkeafstemningen om Maastricht-Traktaten 2. juni 1992," Dansk
Udenrigspolitisk Årbog. København.

Wörlund, I. 1995. "The Swedish Parliamentary Election of September 1994," Scandinavian
Political Studies 18, 4, 285-91.