In the late 1980s, the highly controversial issue of Norway's relationship to the European integration project re-emerged in Norwegian politics. The disruptive effect of the EU issue was a major challenge to most political parties, but the Labour Party was especially vulnerable. The 1972 referendum EC membership had caused electoral defeat and internal conflict in the Labour Party. Now a new debate was underway, and Labour's leaders feared the ghosts of 1972. Strategic choices would be critical for handling the situation properly.

To understand the party elite's choice of strategies, one must consider its objectives - as well as possible tradeoffs among its various objectives. The aim of this article is to answer the following questions: What strategies did the Labour leadership choose for reaching its objectives during the EU debate of 1988-1994? And to what extent did it learn from the 1972 debacle? I begin by sketching the development of the European issue in Norway. Then,

Resumé

The Norwegian Labour Party elite advocated European Union membership in the 1994 referendum, the party's members and voters were divided. This article examines the party leadership's conflict management strategies. The possible tradeoff between, on the one hand, seeking a given policy outcome (in this case, Norwegian membership of the EU), and, on the other, maximising votes and maintaining party cohesion, is focused upon. In the 1972 referendum on Norwegian membership of the European Community, Labour had failed to achieve any of these objectives. Electoral losses and party splits were avoided in 1994, but the main policy objective - EU membership - was still not reached. It is difficult to argue, however, that the Labour elite chose one goal at the expense of the others. This case study suggests that parties may try to reconcile seemingly incompatible objectives and to avoid choosing between goals. In addition, the results underline the importance of organisational learning and indicate that the strategies of both leadership and internal opposition should be included in research on party behaviour.

---

*Jo Saglie, Institute for Social Research, PO Box 3233, Elisenberg, N-0208 Oslo, Norway. E-mail: jo.saglie@isaf.no

in the second section, I discuss objectives, arenas, and strategies. Third, I briefly review Labour's problems in the 1972 referendum, and I explore more thoroughly the party elite's issue management strategy during the 1980s and 19905. In my concluding remarks, I discuss how the findings of this case study may contribute to the literature on party goals and strategies.

The European Issue in Norwegian Politics

Norway applied for membership of the EC for the first time in 1962. After President Charles de Gaulle of France vetoed membership for Britain, however, membership was no longer a viable prospect for Norway. The issue surfaced again after de Gaulle's resignation. In 1972, Norwegian voters rejected membership in a referendum. The EC issue brought down two governments in the early 19705, and caused considerable volatility in the parliamentary election of 1973.

Stein Rokkan and Henry Valen (1964) were the first to describe the links between the EU issue and the Norwegian cleavage structure. Thirty years later, these links were still intact. There is a strong continuity between the referendum votes of 1972 and 1994. Attitudes towards the EU were related to several cleavages (Bjørklund 1997). Both 'pro' and 'anti' arguments pointed to how the choice between membership and 'outsideship' would affect a variety of domestic policies (Oskarson & Ringdal 1998). Not all of the arguments, however, were equally persuasive. Opposition to the EU was clearly associated with egalitarian and leftist values, and with support for the interests of the periphery (Aardal 1995; Aardal et al. 1998). Viewing the matter comparatively, moreover, we find that pro-EU economic arguments were thought less persuasive in Norway than in Finland and Sweden. Oildependent made formal political integration in the EU less attractive Norwegians (Jenssen et al. 1998; Moses & Jenssen 1998; Ingebritsen 1998).

The Norwegian parties may be divided into three main groups: the left (Labour and the Socialist Left Party), the right (the Conservative Party and the Progress Party), and the centre (the Centre Party, the Christian Democratic and the Liberal Party). Confrontation along the left-right axis has usually been the predominant factor structuring party competition in Norway, but the European issue split the traditional blocs and created new political alliances. Labour, the Conservatives, and the Progress Party supported EU membership, whereas the anti-EU alliance comprised three centrist parties and the Socialist Left Party.

In the 1989 election, the anti-EU parties received only 30 percent of the
votes. However, anti-EU feelings were much stronger than was support for
anti-EU parties. In 1993, a majority of the electorate opposed Norwegian

EU membership, while less than 30 percent supported it (Valen 1994, 171). Survey data also confirm the salience of the European issue. In 1993, 65 percent of voters placed the EU among the two issues most important for their vote (Aardal & Valen 1997, 66). The 'no' parties, then, certainly saw the potential for electoral gains, whereas the 'yes' parties feared losses.

Objectives, Arenas and Strategies

Political parties are not unitary actors. Nevertheless, we can usually identify a fairly unified group of leaders. I use the terms 'party elite' and 'party leadership' to refer to such a group (which, it should be noted, is not necessarily identical with the formally elected leadership). The leadership's choice of strategy may be described as a choice between adaptation and domination (Panebianco 1988, 11-14). Party leaders may adapt their policies to demands from their environment, or they may shape and transform their environment in various ways.

The assumption here is that party elites pursue four different objectives, which may or may not conflict with each other. Strøm (1990) included three goals - policy, office, and votes - in his model. I follow Sjoblom (1968) in viewing party cohesion as a basic party goal as well. These objectives may be associated with different arenas. Most commonly, a party elite must choose strategies for the electoral, parliamentary, and internal (i.e. partyorganisational) These arenas are interlinked; strategies chosen in one arena will affect outcomes in the others (Sjoblom 1968; Tsebelis 1990).

Direct democracy has rarely been discussed in the literature on party strategies and party objectives, Miiller's (1998) study being one of the few exceptions. Accordingly, the EC/EU referendums in Norway highlight some particularly interesting party dilemmas. The introduction of direct democracy transformed the strategic game. Party strategists had to act in a double electoral arena: both a parliamentary election and a referendum would have to be confronted. Vote-maximising is usually the immediate goal in the electoral arena, and success in this arena is presumably necessary if parliamentary policy decisions are to be influenced. Holding a referendum, means transferring an important policy decision from the parliamentary to the electoral arena.

In this article, I focus on the possible tradeoff between policy-seeking and vote-maximising/cohesion-maintaining - i.e. between, on the one hand, seeking a given policy outcome (in this case, Norwegian membership of the EU), and, on the other, maximising votes and maintaining party cohesion.The issue dominated national political debate, and the handling of this issue became crucial in the struggle for votes. If vote-seeking considerationswere within the Labour elite, we would expect the

party to adapt its position to prevailing public opinion (which was EU sceptical), or at least to downplay its pro-EU position. Such a strategy might also minimise internal divisions. On the other hånd, the party elite was also deeply concerned with the outcome of the referendum. If policyseekingconsiderations i.e. the need to win the referendum - were dominant,we expect the party leadership to argue its pro-EU position openly and strongly.

The fourth goal (office-seeking) caused fewer strategic problems for the Labour leadership in the 19905. Labour minority governments were in power from 1986 to 1997, except for the period between October 1989 and November 1990, when Jan P. Syse's short-lived centre-right coalition was in office. It was disagreement over the European issue that brought down the Syse government. Labour became the only credible government alternative the default option. Furthermore, the Labour prime minister, Gro Harlem Brundtland, made it clear she would remain in office irrespective of the referendum's outcome, unlike Trygve Bratteli in 1972.

The Labour Party leaders had one advantage when they chose their strategies during the 1988-1994 EU debate: they could utilise experience gained in the previous EC referendum. As Harmel and Janda (1994) note, external shocks may explain why parties change their political strategies. Before we turn to the 1994 referendum, therefore, a brief discussion of the impact of the 1972 shock on the four party objectives will be useful.

Labour and the 1972 EC Referendum: When Party Loyalty Failed¹

With de Gaulle's fall from power, the EC issue re-emerged in Norway. Labour's 1969 national congress approved a Norwegian application for EC membership. Labour's youth organisation (AUF), on the other hand, elected an anti-EC leadership. AUF was represented in the party's national and executive committees, giving the 'no' side a voice inside Labour's chief organs.

In March 1971, the EC issue caused the demise of the centre-right coalitiongovernment by Prime Minister Per Borten, and Labour formed the new government. While Prime Minister Bratteli took on responsibility for the negotiations with the EC, he also had to keep the Labour Party united. The government's 'wait and see' strategy prevailed at the May 1971 national congress: negotiations should be completed before the party made its final decision. When the negotiations were concluded in early 1972, the minister of fisheries could not accept the result. He resigned from office, giving the 'no' side a strong case. Labour's extraordinary national congress in April 1972 voted in favour of EC membership, but former Prime Minister

Einar Gerhardsen delivered a conciliatory speech. He emphasised the danger of internal division and electoral loss: 'If the party did not have any wellknownmembers spokesmen who were against membership, they ought to get hold of some - in order to make it clear for everyone that one can belong to the Labour Party even if voting no in the referendum'.²

Gerhardsen's statement was received with great enthusiasm - and frequently - by Labour's anti-accessionists. In January 1972, they formed an organised 'no' faction: AIK (the Labour Movement Information Committee against Norwegian EEC Membership). On the other side, the Labour elite started a vigorous 'yes' campaign, under the slogan 'A Labour voter is a yes-voter'. Prime Minister Bratteli declared that the referendum would be a vote of confidence in the Labour government. That did not convince the people: 53.5 percent voted against Norwegian EC membership, Bratteli handed in the cabinet's resignation. AIK decided to continue activity after the referendum. As the 1973 parliamentary election approached, AIK joined the Socialist Electoral Alliance, which later became Socialist Left Party. Labour suffered considerable losses in the 1973 election, while the Socialist Electoral Alliance gained ground.

In short, Labour met with defeat with regard to all four objectives. The party elite failed to reach its most important policy goal - EC membership. In addition, Labour lost both office and votes, and it failed to maintain party cohesion too. These last three setbacks, however, turned out to be shortlived. regained government office in 1973, and the voters returned in 1977. And while AIK decided to join the Socialist Electoral Alliance, a large number of AIK activists chose to remain in the Labour Party. Two members of AlK's executive committee - Bjørn Tore Godal and Thorbjørn Berntsen - went on to play a key role in the EU debate 20 years later. But this time they chose the pro-EU side, as members of Brundtland's cabinet.

The Labour Party and the 1994 EU Referendum

A Dual Track towards Brussels: The EC and the EEA

Labour's foreign policy debate during the early 1980s centred mainly on nuclear weapons and NATO, but there were also signs of a return of the European issue. While the old pro-accessionists were still around, a 'New Europeanist' orientation emerged among former left-wing anti-accessionists. They spoke for a European-based security structure, which would allow ties to the US to be loosened. The 1985 party programme also indicated a reevaluation the EC issue. The previous programme had contained a reference the 1972 referendum outcome; this reference was now removed (Bull 1989; Eide 1990).

Labour regained government office in 1986 from the centre-right coalition led by Kåre Willoch, and the EC debate gathered speed. Prime Minister Brundtland's initial policy approach was an active adaptation to the EC's single market. In the short term, this policy was able to satisfy Labour's need for internal cohesion - but only as long as EC membership ruled out. When the Conservative Party expressed support for Norwegian membership in August 1988, the 'EC peace' was broken - both among the general public and within the Labour Party. The Conservative initiative put the Labour elite, which had wanted more time for debate, in an awkward position. Labour's parliamentary leader, Einar Førde, made his position clear: There will be no Norwegian EC membership unless Labour joins in. ... As a contribution to Norwegian membership, the Conservative actions are delaying and divisive. If I were burning to get Norway into the EC quickly, I would start by asking the Conservatives to stop this.'³

Labour's national congress in March 1989 adopted a new party programme, a statement that gave the leadership a great deal of latitude: 'Norway must choose the form of relationship to the EC which serves our national interests'.⁴ The party congress also voted on a resolution that concluded that 'it is neither necessary nor desirable that Norway take a position on membership now'. Anti-accessionists put forward a more critical motion, but the leadership's proposal was carried, notwithstanding the 67 votes opposed.⁵

While it treated the membership issue cautiously, the party leadership also followed another track towards Brussels. Prime Minister Brundtland and the President of the EC Commission, Jacques Delors, had been discussing cooperation between the EC and the European Free Trade Association (EFTA) since 1988. This development was reflected in Labour's 1989 programme, which stated that Norway should cooperate with EFTA when negotiating with the EC.⁶ In March 1989 - two weeks after the party congress - the EFTA summit meeting approved Delors's idea of institutionalising relations. With that, the European Economic Area (EEA) process was launched. And while the leadership sought to postpone the debate on EC membership - preferring to let it mature - it took an active and forceful approach to the EEA issue.

A resolution on European policy, which approved the EEA process, was passed unanimously by the party congress in November 1990 (Skjeie et al. 1995, 38-9). This signified the end of Labour's EEA debate; however, the EEA issue continued to divide the party's 'no' side. Some anti-accessionists thought Norwegian access to the European single market, as granted by the EEA agreement, was vital to Norwegian industry and employment. Others regarded the EEA as the worst of all possible outcomes: with integrationinto EC's single market, Norwegians were subject to a neoliberal

economic system; at the same time, however, they were without any influenceover
policy. Parliament eventually approved the EEA agreement in
October 1992. Only two of Labour's 63 representatives voted against.

Educating the Party Members

General Secretary Thorbjørn Jagland had been concerned with the role of the extraparliamentary organisation for a long time. The local organisation needed revitalising, and European policy seemed a suitable issue for study groups within local party branches.⁷ Thus, the idea of involving the Labour grassroots in a broad European debate was conceived long before the 1990 congress. Various information and debate documents on the EEA and the EC were distributed through the party organisation. Jagland himself published book, My European Dream (1990), in which he argued for European solutions to unemployment and environmental problems - which he considered be inherently transnational. However, while he thought the European should be discussed, he did not place EC membership on the agenda. Jagland wished to avoid a 'campaign discussion', and he refused to let 'minor groups decide the terms of the discussion on the most important for Norway during the 19905'.⁸ His hope seemed to be that an allparty on EC membership would emerge over the long run:

Should we not spare the time to ensure a proper foundation for our decision? This is not party tactics; rather, it deals with revitalising an important feature of our democracy: the parties' obligation to set out the premises for important decisions. Unless this happens, I do not think we will get a better EC debate than in 1972. It would be a national disaster if a new referendum in 1992-93 were to result in the narrowest possible majority for one of the alternatives. (Jagland 1990, 23-4)

Events outside Norway, however, complicated the leadership's efforts to gain time. In October 1990 (a few weeks before the party congress), the Swedish government announced that Sweden would apply for EC membership.

A committee led by MP Grete Knudsen directed the preparation of Labour's study document. Knudsen was known for her critical attitude towards the party leadership, and her appointment may be understood as a cooptation strategy: for the result to be viewed as legitimate, critical voices would have to be drawn into the European debate. And it would be difficult to convince the grassroots unless these critics eventually supported the leadership's pro-European view (Hansson & Teigene 1992, 293).

The final version of the study document - Norway in a New Europe - was presented in September 1991. This voluminous document contained a thorough description of the challenges Norway faced in regard to the environment,welfare, unemployment. Changes at the European and global levels were also discussed. The document maintained a quite neutral

tone, but EC opponents criticised its perspective and selection of facts. According to the document, 'EC membership means participation in a politicalprocess which all political forces participate. The question is: are we better served by participating in or standing outside this process?'⁹ The critics, however, did not regard the EC as an open arena. In their opinion, the EC was an institution with a neoliberal orientation and freemarketpolicies; it was detrimental to Labour's interests and values. Anti-accessionists agreed with the description of the challenges and the need for international cooperation, but they called for a critical analysis of the EC system. Their question was: is the EC a good solution to these problems?

Membership attendance at the local study circles gave legitimacy to
Labour's EC policy, as implied in Jagland's speech to the 1992 party congress:

The number of participants in our study circles was higher than these two parties' [the Socialist Left Party and the Progress Party] membership figures. And think about what this process has been. It started here at the congress. A group elected by the congress and led by Grete Knudsen drew up Norway in a New Europe. Fifteen-thousand members participated in study circles. We received piles of replies. Local associations, municipal branches, and county branches have discussed the EC issue arid passed resolutions with the greatest respect for each other.¹⁰

The official participation figures, however, were a good deal lower. Labour's educational organisation registered 1019 completed study circles with a total of 6305 participants. However, interest in the study document the number of participants. About 20,000 copies were sold.¹¹

Why did the leadership carry out this extensive debate? 'Whether this method was chosen because of a sincere wish for a democratic process or for pure convenience remains an open question', according to journalists Bjartnes and Skartveit (1995, 78). Such motives need not be incompatible. A sincere wish to activate the party grassroots in a democratic debate may be compatible with spreading information in the hope that members will adopt a pro-EC position. In any case, it was important to avoid the topdown of 1972. This time the key word would be dialogue.

From Debate to a Final 'Yes'

During the spring of 1992, the Labour Party had to draw some conclusions. The EC issue was discussed at the annual meetings of Labour's 19 county branches. The geographic split was evident, with pro-EC majorities in the south and anti-EC majorities in the four northernmost counties. At the meeting of Hordaland County, Prime Minister Brundtland announced her official support for EC membership for the first time. This was no surprise;

nobody had doubted her firm commitment to accession. Nevertheless, her refusal to express her view until April 1992 was an important signal: the local debate should be conducted without any cues from the leadership. In her speech, she attacked those who would rush the decision:

Many . . . have not been willing to accept that an historic decision demands conscious,
serious and thoroughly considered choices. Many have pushed for starting trench warfare
with slogans and demagogy, as soon as possible. ... In this party we shall open our eyes,
listen to each other - and gather new knowledge. We shall secure a debate in which as many
as possible participate - before we give our answers.12

At Labour's party congress in November 1992, Brundtland unexpectedly announced her resignation as party leader. She continued as prime minister, while Thorbjørn Jagland was elected party leader. The congress also approved application for EC membership, by a vote of 182 to 106.¹³ In addition, the congress decided that the issue would be reopened when negotiations with the EC had been completed. Thus, the 1992 'yes' was not final. On the other hand, the application certainly implied a desire for membership. The party programme stated that 'the Labour Party wants Norway to apply for membership in the EC, because we think that Norway naturally belongs to a binding cooperation between the democratic countries our own part of the world'.¹⁴ Two weeks later, Parliament passed the Norwegian application for membership. Of Labour's 63 MPs, 14 broke with party policy and voted against the application.

When negotiations between Norway and the EC started, the only antiaccessionist Brundtland's cabinet was the minister of fisheries, Jan Henry T. Olsen. The position he took was crucial. Brundtland (1998, 284-5) recounts she 'simply did not think that we could achieve a "yes" in the referendum unless the result was so good that a man with Jan Henry's attitudes background could endorse it'. If the fisheries solution was unacceptable, would not be able to recommend the negotiation result to the Norwegian people'. However, Olsen - known as 'No-fish Olsen', due to his uncompromising negotiation demands - switched to a pro-EU position. He endorsed the outcome when negotiations were concluded in March 1994. In other words, the leadership's cooptation strategy worked (Skjeie et al. 1995, 40). While the minister of fisheries in 1972 went from 'yes' to 'no', Olsen went the opposite way. Thus, the Norwegian cabinet was united in favour of membership - unlike the Swedish. Ingvar Carlsson, Sweden's Social Democratic prime minister, appointed two anti-accessionists to his new cabinet in 1994(Aylott 1997, 133).

An extraordinary party congress convened in June 1994 to make Labour's final decision. The pro-accessionist motion, which stated that Labour should recommend a 'yes' vote in the referendum, was passed by 197 votes to 93.¹⁵ A resolution on Labour's further handling of the issue was carried

unanimously. This resolution expressed a delinkage strategy - a temporary release from the bonds of party loyalty (Svåsand & Lindstrom 1996). The thinking was that demands for loyalty to congress decisions should not apply to the EU controversy, since that was to be settled by referendum. Members were free to work for either side in the referendum campaign, and the party would accept the outcome of the referendum unconditionally.¹⁶ General Secretary Dag Terje Andersen stressed that this was a special case; demands for party loyalty would not necessarily be suspended on other issues.¹⁷ In other words, both sides could find support in the congress resolution. Backed by the congress's ultimate 'yes', the party leadership could launch an extensive pro-EU campaign. On the other hand, those opposedto membership could fight against the leadership freely and without being accused of factionalism.

In addition to its domestic campaign to persuade voters, the Brundtland government apparently also took part in an international strategy of institution whereby the Nordic EU referendums were synchronised. The sequence of the three referendums was aimed at producing a domino effect (Hovi & Hellevik 1996; Ruin 1997, 67-73). The most pro-EU country

- Finland - was to hold its referendum first, followed by Sweden. Norway, where anti-EU sentiments were the strongest, would be the last country to vote. A 'yes' in Finland and in Sweden, the governments hoped, would convince sceptical Norwegians. There was no formal agreement between the governments, but there was presumably an understanding among them that such a sequence would favour the pro-EU side.

The Organised Opposition

The Labour Party leadership thought the extensive debate would remove the need for a 'no' faction like AIK in the 19705. 'The party is totally dependent an arrangement with broad support in the party, something that will remove the need for such organised factions', Thorbjørn Jagland said in 1990.¹⁸ Jagland's hopes proved unfounded. A strong 'no' faction was established within the party towards the end of the campaign. Labour's first anti-EC group, however, was not so well organised. The 'Readiness Group', a loose network led by parliamentary backbencher Inge Staldvik, was formed in August 1991. A lack of money and disagreements over the EEA created problems for Staldvik's group. The members of the group also had doubts about when - and whether - a new AIK should be formed (Staldvik 1993, 22-40). In December 1992, Staldvik gave up; he left Labour and joined the Socialist Left Party.

By that time, other anti-accessionists had made more concrete plans.
The uncompromising Staldvik was considered an unsuitable leader. The
new 'no' faction - 'Social Democrats against the EC (SME) - was intended

to be more moderate, so that more prominent anti-EC politicians could be mobilised. The timing of SME's formation was crucial. The organisers waited until after the 1993 parliamentary election. If SME had been launched before the election, it would have been blamed had Labour lost ground (Bjartnes&Skartveit 1995,167-76).

SME was formally founded in October 1993. Hallvard Bakke, an MP and former cabinet minister, became its leader. SME's executive committee included several trade union leaders, some members of Labour's executive committee, and the leader of Labour's youth organisation. Unlike Norway's main 'no' movement, 'No to the EU', SME supported the EEA agreement. Thus, it could mobilise the softer anti-accessionists, including trade unionists concerned with access to the European market. SME's founders emphasised that they did not represent a permanent opposition to the Labour leadership. According to the organisation's statutes, SME would be dissolved following the referendum. This promise was kept.¹⁹

The party elite attemped grudgingly to maintain a proper relationship with SME. Labour's executive committee - which included some SME members - stated unanimously that SME was 'not a part of the party's activities' (Skjeie et al. 1995, 42). Nevertheless, it was a difficult balance to achieve. The controversy over the sequence of Nordic referendums may illuminate this point. The anti-accessionists wanted the Finnish, Swedish, and Norwegian referendums to be held simultaneously, in order to prevent a domino effect. They put forward a parliamentary motion, calling upon the government to arrange simultaneous referendums. Such a motion would normally have been rejected. However, several pro-EC MPs were absent. Thus, the SME group in parliament helped defeat its own government. This was apparently the result of some kind of communication failure between Hallvard Bakke and Labour's party whip; and the communication failure, in turn, derived from SME's unclear status as a semi-legitimate faction.²⁰

Unlike the 'No to the EU' movement, SME never tried to become a mass-membership organisation. Nevertheless, the self-appointed executive committee enhanced its legitimacy when SME's first - and only - national congress took place in April 1994. At the same time, SME presented the document Social Democratic Alternative. Here SME outlined its view on the consequences of Norwegian EU membership, and stressed the conflict between policy and Labour's values. The party programme was compared with the principles outlined in the Maastricht Treaty: 'Our alternative to the EU is Labour's programme, which can be implemented more easily outside a union of West European countries'.²¹

Skjeie et al. (1995) underline the contrast between the party leadership's attitude towards AIK in 1972 and its attitude towards SME in 1994. The party elite had demanded loyalty in 1972; by contrast, SME was legitimised by a 'contract of disagreement'. SME was allowed to fight the party line.

on the understanding that the faction would be dissolved immediately after the referendum. Activists on both sides also pointed to a radically improved organisational culture within the party, with tolerance for divergent opinions having replaced the destructive conflict of the 19705.

Still, the contrast between 1972 and 1994 should not be exaggerated. In 1972, for instance, AIK had been accepted to some extent. For example, the party secretariat had turned to AIK to get anti-EC speakers when both sides were to be represented at a party meeting (Gleditsch & Hellevik 1977, 180). In 1994, moreover, SME's acceptance was less than complete, as a comparison with Sweden reveals. SME enjoyed a less official status than did its counterpart in Sweden. The Swedish Social Democrats established two official EC committees, one for EC membership and one against (Aylott 1997,130). SME's status was, at best, semi-official. SME may have prevented a loss of voters and members, and the faction was tolerated, but the tension between SME and the Labour elite was evident. Tove Strand Gerhardsen, SME's deputy leader, described the leadership's ambivalence in this way: 'The party leadership is not pleased with the formation of "Social Democrats against the EU", but it should be. We contribute to party unity. About one-half of Labour's voters are against the EU. A wise party leadership ignore that.'²²

The party elite had learned from the 1972 debacle - and so had the opposition. the 1972 referendum, there had been disagreement over AlK's future. A number of activists wanted to continue the struggle after the referendum victory. SME's leaders, by contrast, were determined to avoid this. They stressed that the organisation would cease to exist after the referendum, and that SME had no policy on any issue other than the EU. According to Bjartnes and Skartveit (1995, 168), anti-EU strategists had carefully studied Atle Hellevik's (1979) book on AIK, The Responsible Rebellion. The 1993-1994 rebellion, therefore, proved even more responsible than its predecessor.

Labour Youth: From Doubt to 'No'

Labour's youth movement, AUF, had played a central role in AIK in 1972. However, pro-EC currents were strong within the Labour youth movement during the European euphoria following the fall of the Berlin Wall. AUF endorsed the principle of supranational decision making at its 1990 congress. This resolution was perceived as an important step towards a pro-EC position (Bjartnes & Skartveit 1995, 105). However, the antiaccessionistssoon strength in AUF. The 'no' side won a clear victory at AUF's 1992 congress, where 215 of 350 delegates voted against EC membership. Moreover, Trond Giske - who opposed the EC as well as the

EEA - succeeded the pro-European Turid Birkeland as AUF leader. On
the other hand, the youth congress approved the EEA agreement.23

The Trade Unions: A Severe Blow to Labour's Leadership

The party leadership could tolerate the anti-EU position of the youth movement. all, the party youth occasionally took more radical stands than the main organisation. The trade unions were presumably more important to the party elite. Historically, the Labour Party and the Confederation of Trade Unions (LO) had developed as two branches of the same labour movement. LO was a weighty supporter of the Labour leadership in 1972, but it proved to be less amenable in 1994. The division between the Labour elite and LO's anti-accessionists was exposed during the May Day celebrations. several cities, local trade unions adopted slogans against both the EC and the EEA. The 1993 May Day festivities in Oslo were especially painful. Thorbjørn Jagland withdrew from the arrangements because the Oslo unions adopted anti-EEA slogans. The AUF leader, Trond Giske, replaced as the main speaker. While Jagland spoke at a minor indoor meeting, the anti-accessionists dominated the traditional celebration at Youngstorget, a downtown square. Giske's behaviour was like 'spitting in the party leader's face', according to Minister of the Environment Thorbjørn Berntsen - a former AIK activist who had converted to a pro-EC position (Bjartnes&Skartveit 1995,156-7; Staldvik 1993,137-40).

LO did not take a stand on EU membership until its September 1994 congress. At this congress, the leaders of the party and of LO itself suffered a surprising defeat. The LO leaders submitted a motion recommending that members vote 'yes' in the referendum (if, that was, Finland and Sweden had voted for membership already). But the congress adopted a resolution recommending a 'no' to the EU, by 156 votes to 149. Prime Minister Brundtland (1998, 379) was disappointed, but downplayed the importance of the decision. LO - with its pro-EU leadership - did not fight against Norwegian membership actively. Nevertheless, LO's stand played an important in the rhetoric of the 'no' side. Now LO had found its place among all the other grassroots movements on the 'no' side, while the Labour elite's only bedfellows on the 'yes' side were - according to the antiaccessionists the Conservative Party and the business elite.

The Parliamentary Election and the Referendum: The Campaigns

During the campaign for the 1989 parliamentary election, the 'no' parties claimed that Labour was keeping the EC and the EEA off the campaign agenda. The struggle for the agenda continued in the lead-up to the 1993 parliamentary election. Whereas the fiercest anti-EC parties stressed the

EC question, Labour's preferred campaign issue was unemployment. The Labour Party employed a delinkage strategy, detaching the EC issue from the parliamentary election. Since, the party reasoned, the EC issue was to be settled by referendum (i.e. outside parliament), the issue should not be central during the parliamentary election. Anti-EC voters could safely vote for pro-EC parties, according to Prime Minister Brundtland (Saglie 1998, 361).

The first step in the Labour Party's 'two-stage' strategy - parliamentary campaign first and referendum campaign later - was a success. Labour came out a winner in the 1993 election, notwithstanding the salience of the EC issue. Some pro-EU campaigners criticised Labour's low profile during the early stages of the debate. In the view of these critics, the Labour elite pursued narrow party interests (maximising votes and maintaining party cohesion) to the detriment of national interests (winning the referendum). However, the party leadership saw no contradiction between these goals. The many undecided Labour voters were important targets for the pro-EU campaign. If internal conflicts and electoral losses weakened the Labour Party, it would not be able to campaign effectively. In that case the referendum surely be lost, according to party strategists.²⁴

Whereas anti-EU parties and organisations waged a coordinated campaign,
contact between the Labour Party and other 'yes' groups was
limited. Labour wanted to keep its distance from its traditional adversaries

- the Conservative Party and the business organisations. If the image prevailed of an alliance between Labour, the Conservatives, and the business the pro-EU campaign would suffer. The saying that 'a man is known by the company he keeps' did not apply to this issue, according to General Secretary Dag Terje Andersen.²⁵

The need for party unity was evident during the last phase of the campaign well. Accordingly, Jagland formulated 'Five Commandments' for the internal debate; these emphasised tolerance and respect for divergent opinions.²⁶ Likewise, the party leadership realised the necessity for presenting nuanced message to the voters. The point of the campaign was not to rally the convinced pro-accessionists, but to persuade the doubters. Accordingly, should show respect for the doubters' views and admit that EU membership also entailed some disadvantages. The Labour Party's official 'yes' campaign started in August 1994, under the slogan 'Vote for security. Vote "yes".' In a symbolic manifestation of the delinkage strategy, the party did not use its symbol - a rose - in the EU campaign. A specially designed 'yes rose' replaced the ordinary Labour rose.²⁷

Of course, this did not prevent the party elite from promoting the party's pro-EU position vigorously. The party's ordinary electoral machine could not be employed unabatedly. A number of the volunteer activists - especially in the youth organisation - were anti-accessionists and could not be mobilised.On

lised.Onthe other hand, the prime minister's personal popularity was a great asset, and she made use of her prestige. Her speech at Labour's extraordinary party congress, which included a refutation of assertions concerning the fisheriesagreement the EU, may illustrate this point: 'Do you really believe we would have done something like that - sold out vital Norwegian interests? Do you believe that this cabinet could stand up and look people in the eye if that was the case?'²⁸

Brundtland spoke at a number of meetings around the country during
the campaign, especially in rural areas. There she attacked what she perceived
anti-EU scare tactics and propaganda. 'I have purposely sought
out environments which are [EU-]negative or [EU-]sceptical in order to
counter misinformation', she said. But her rhetoric too was charged with
emotion, as when she described the effects of a 'no': 'Then we really have
reason to fear the unknown. Norway has never, I say never, been in such a
situation. Alone - and not together with our neighbours and friends in
„ ,29
Europe.

In the heat of the battle, arguments were seldom nuanced. Minister of trade Grete Knudsen stated that 'we have to join reality',³⁰ while Brundtland (1998, 378), after a discussion with a farmer, claimed that he was brainwashed'. party members in Northern Norway wrote a letter to Jagland, complaining about Brundtland's arrogance and patronising manner ³¹ The anti-accessionists were just as outspoken. Hallvard Bakke, for example, asserted that the prime minister had consciously misled parliament ³² In short, a number of people paid little heed to the principle of respect for opponents.

Conclusion: Opinion Leadership or 'Contract of Disagreement'?

Labour performed better in 1994 than in 1972: electoral losses and party splits were avoided, and the party remained in office. But the main policy objective still was not reached: 52.2 percent of the voters rejected EU membership. this section, I present some answers to the introductory questions: What strategies did the Labour leadership choose for achieving the goals of winning the referendum, maximising votes, and maintaining party cohesion? to what extent did it learn from the 1972 debacle? Furthermore, theoretical conclusions can be drawn from the answers to these questions? How do the findings of this case study contribute to the literature on party goals and strategies?

Critics maintained that the Labour Party gave priority to maximising
votes and maintaining internal cohesion, at the expense of taking an active
role in the formation of opinion (while the anti-EU parties, by contrast.

campaigned effectively for five years). As mentioned above, Labour strategistsdisagreed. weakened by electoral defeat and internal disintegration,the Party would be unable to influence public opinion. And without a strong Labour Party as opinion leader, Norway would never join the European Union. Thus, all of the party goals were compatible. This point of view should not be dismissed as an exercise in evading conflicts between goals. The Labour Party's failure is evident only with the benefit of hindsight. The strategy was nearly successful. The Norwegian anti-EU victorywas and presumably the result of a number of factors beyond the control of party strategists. If the Norwegian 'yes' side had won a majority, Labour's leaders would certainly have been celebrated as master strategists, who had won both the referendum and the parliamentary election. Likewise, the apparently successful Swedish Social Democratic leaders would have been blamed for insufficient campaigning if the Swedes had voted 'no'. In short, it is difficult to argue that the Labour elite chose one goal at the expense of the others.

At the general level, this study underlines the complexity of party goals. To ask whether policy, office, or votes drove party behaviour in the direct democratic process - as Miiller (1998) has done - may be an oversimplification. Labour had been purely a vote-seeking party, the Labour government probably never have applied for EU membership. On the other hand, the Labour elite certainly did not ignore vote-seeking. By means of a double strategy - opinion leadership together with a 'contract of disagreement' the party elite tried to reach all of its objectives. In other words, this study indicates that parties may reconcile seemingly incompatible objectives avoid choosing between goals. Defining party strategy as a decisive choice between policy, office, and votes may be misleading.

How fruitful are the concepts of 'adaptation' to the environment and 'domination' of it? The party elite employed both strategies; however, these general concepts must be put in more concrete terms. The environment surrounding the Labour Party leadership - the party's members and its voters - was deeply divided. Thus, 'adaptation' was meant to show that there was room both for opponents and for supporters of EU membership within the party. The key words were 'delinkage' and 'legitimising'. As emphasised Skjeie et al. (1995), the party elite of the 1990s had taken Einar Gerhardsen's advice from the 1972 congress. The general secretary alluded to the unfortunate 1972 slogan ('A Labour voter is a yes-voter') in his speech at the extraordinary congress of 1994: 'Both yes- and no-people shall feel welcome in the Labour Party. A Labour voter can be either a yes- or a no-voter.'³³

As regards 'domination', the party elite could choose from a wide repertoire.Institutional
- as in the timing of the Nordic referendums -
was employed. Attempts at agenda setting were also evident: for example,

the party elite downplayed the EU issue during the early stage of the internal debate, as well as during the parliamentary and local election campaigns. Finally, the party leadership made a strong effort to persuade voters and members. Appeals to party loyalty had not worked in 1972, and they were not repeated. On the other hånd, the integrity and credibility of the prime minister and of the cabinet were prominent campaign arguments. The most distinctive feature of Labour's EU debate, however, was persuasionthrough Labour carried out an exceptionally broad and open discussion before the leadership drew any official conclusions. Still, the leadership presumably hoped that the discussion in local study circles would put an end to 'misunderstandings* about the EU, yield insight into the internationalisation of policy challenges and the shortcomings of the nation state, and thus form the basis for a more positive attitude towardsthe However, the leaders seem to have had a range of views about how the voters could be persuaded - at least their practice varied. A nuanced and balanced form of rhetoric might have been advantageous. But a number of leading Labour politicians had strong personal convictions about the EU issue. In addition, several cabinet ministers had put a great deal of time and effort into the negotiations with the EU. This situation was hardly conducive to a balanced discussion.

These findings point to another theoretical conclusion: the utility of the concepts of 'adaptation' and 'domination' may be questioned. This study indicates that these strategies may be more ambiguous in practice than in the organisational literature. Even when these concepts are put in more concrete some ambiguity remains. Party leaders may wish, for example, to persuade voters and members, but they disagree on the best way to do it. When party strategists face a divisive issue, therefore, their main choice may not be between 'adaptation' and 'domination', but rather between a clear message and a nuanced one.

The credit for Labour's successful conflict management must be shared by the leadership and the opposition. Although the anti-EU faction worked against the leaders' policy goal, it did not wish to jeopardise their voteand efforts. SME played the role of a strictly limited opposition, solely on the EU issue. It promised, moreover, to cease its activities after the referendum. Thus, the elite and SME could enter into an informal 'contract of disagreement' with a time limit. During the course of the EU controversy, relations between the Labour leadership and SME were strained. Nevertheless, the party's internal life became surprisingly idyllic after the referendum - in sharp contrast to the aftermath of the 1972 split.

Hence, this study also points to the importance of opening up the party
black box. As Strøm (1990, 569) notes, party organisations impose constraintson
behaviour of their leaders. However, we should not limit our

research to the behaviour of the party leadership. When an organised oppositionemerges a party, it should be regarded as a separate actor with its own strategies and goals. The interaction between the party elite and the internal opposition may determine the behaviour of the party as a whole.

The answer to the second introductory question should be evident from the preceding discussion: organisational learning has certainly taken place. The leadership's strategies and actions were pervaded by the 1972 experience. literature on the Labour Party and the EU issue emphasises the changes between 1972 and 1994 (see, e.g. Heidar 1994; Skjeie et al. 1995). The debate of the 1990s was conducted in a more open atmosphere, and the opposition was given a more legitimate role. This was due not only to organisational learning but presumably also to personal experience. Several leading pro-EU Labour politicians - Jagland, Foreign Minister Godal, and a number of other cabinet ministers - belonged to the anti-EC faction in 1972. Some critics have claimed that the changes were superficial. For instance, the defector Inge Staldvik (1993, 86) asserted that 'the modernised Stalinism may be softer, but it is smarter and more difficult to discern'. Nobody denies, though, that some kind of transformation has taken place.

Comparing 1972 and 1994 serves to highlight the changes, however, and the continuity may be easily overlooked. A comparison with Sweden might thus modify the picture, and highlight the limits to the 'contract of disagreement'. and Lindstrom (1996, 215) characterise the Swedish party's strategy as 'abdication' (from opinion leadership), in contrast to the hard work of the Norwegian and Finnish sister parties to persuade their supporters. This is overstated. But the Swedish Social Democrats' appointment separate 'yes' and 'no' committees, along with Carlsson's appointment two anti-EU cabinet ministers, does point to a greater emphasis on delinkage - at the cost of opinion leadership. One explanation may lie in the timing of parliamentary elections and referendums. The parliamentary election and the referendum in Sweden were held within a period of eight weeks, while the Norwegian referendum took place more than a year after the parliamentary election. Thus, the Norwegian Labour elite could more easily adopt a two-stage strategy, dealing with the election first and the referendum later.

Still, the Swedish comparison does not alter the general impression: namely, that the case of the Norwegian Labour Party underlines the importance organisational learning. It may be useful to note, however, that observers and politicians tend to draw different conclusions from history. When the European issue reappeared in the late 1980s, some observers predicted history would repeat itself: the Labour Party would suffer defeat, as it did in 1972. But Labour's strategists had learned from the 1972 experience: developed a new strategy, and they invalidated these predictions. In short, rational actors can refute historical analogies.

---

NOTES

1. This section draws upon Gleditsch & Hellevik (1977). Hellevik (1979). and Skjeie et al. (1995).

2. QuotedinHeidar(l994, 109).

3. Stortingstidende 1988-1989, 2371. Throughout this article. I am responsible for the translation from Norwegian.

4. Quoted in Aktuelt Perspektiv 1990/14. 5.

5. Aktuelt Perspektiv 1989/11, 14.

6. Aktuelt Perspektiv 1990/14, 5.

7. Aktuelt Perspektiv 1989/39, 5.

8. Aktuelt Perspektiv 1990/20, 2.

9. 'Norge i et nytt Europa'. Labour Party. 1991. 101.

10. T. Jagland's speechat Labour's party congress, Oslo. 5-8 November 1992.

11. Telephone interview with Labour's information director. P. Nordenbora, 28 November 1995.

12. 'Norske utfordringer i en ny intemasjonal virkelighet". G. H. Brundtland's speech at the annual meeting of Hordaland County Labour Party. Ullensvana. 4 April 1992.

13. Aktuelt Perspektiv 1992/41,2-3.

14. 'Mer fellesskap - sosialdemokratisk program for 90-åra". Labour Party. 1992. 52.

15. The pro-accessionist resolution. 'Et solidaritetsalternativ for Norge', is published as A-info 22/94. The anti-accessionist motion. 'Solidaritet for alle' (A-info 23/94). wasrejected.

16. A-info 22/94, 19-20. The pledge to "accept the outcome" meant that Labour would not accept any kind of minority veto in the referendum.

17. 'Arbeidet i partiet framover', D. T. Andersen's speech at Labour's extraordinary party congress. 18-19 June 1994. A-info 21/94. 5.

18. Aktuelt Perspektiv 1990/20. 2. '

19. Aktuelt Perspektiv 1993/40, 6-7. For a study of SME's members. see Geyer and Swank (1997).

20. Aftenposten. 15 December 1993. The vote was inconsequential. since the Finnish and Swedish governments refused to hold the referendums simultaneously.

21. 'Sosialdemokratisk alternativ". Sosialdemokrater mot EU. 1994. 35.

22. Aftenposten, 19 March 1994.

23. Aftenposten, 2-5 October 1992.

24. Interview with former General Secretary D. T. Andersen. 3 December 1997.

25. 'Arbeidet i partiet framover'. A-info 21 /94. 6.

26. Aktuelt Perspektiv 1993/44, 9.

27. Interview with former General Secretary D. T. Andersen. 3 December 1997: Aftenposten, September 1994.

28. innledning om forhandlingsresultatet'. G. H. Brundtland's speech at Labour's extraordinary congress. 18-19 June 1994. A-info 19 94. 4.

29. Aftenposten. 26 November 1994.

30. Sloniny.siiilende 1993-1994,2950.

31. Aftenposten. 22 October 1994.

32. Aftenposten, 27 May 1994.

33. 'Arbeidet i partiet framover." A-info 21 ''94. 6.

ACKNOWLEDGMENTS

An earlier draft of this article was presented at the annual meeting of the Midwest Political Science Association, 23-25 April 1998, Chicago. In addition to those participating in the panel debate, I would like to thank Ralph M. Goldman, Knut Heidar, Christine Ingebritsen, and the anonymous reviewers for their comments.

REFERENCES

Aardal, B. 1995. 'Ideologi på tvers?,' in Jenssen, A. T. & Valen, H., eds. Brussel midt imot.
Oslo: Ad Notam Gyldendal.

Aardal, 8., Jenssen, A. T., Oscarsson, H., Sånkiaho, R. & Såynåssalo, E. 1998. 'Can Ideology
Explain the EU Vote?,' in Jenssen, A. T., Pesonen, P. & Gilljam, M., eds, To Join or Not
to Join. Oslo: Scandinavian University Press.

Aardal, B. & Valen, H. 1997. 'The Storting Elections of 1989 and 1993: Norwegian Politics
in Perspective,' in Strøm, K. & Svåsand, L., eds, Challenges to Political Parties: The Case of
Norway. Ann Arbor: University of Michigan Press.

Aylott, N. 1997. 'Between Europe and Unity: The Case of the Swedish Social Democrats,'
West European Politics 20, 119-36.

Bjartnes, A. & Skartveit, H. 1995. Brussel tur-retur. Oslo: Tiden.

Bjørklund, T. 1997. 'Old and New Patterns: The "No" Majority in the 1972 and 1994 EC/
EU Referendums in Norway,' Acta Sociologica 40, 144-59.

Brundtland, G. H. 1998. Dramatiske år. 1986-1996. Oslo: Gyldendal.

Bull, B. 1989. 'Arbeiderpartiet og den nye Europa-debatten,' in Knudsen, B. 8., ed., Den
nye Europa-debatten. Oslo: Cappelen.

Eide, E. B. 1990. 'Europa-debatten i Arbeiderpartiet: Forholdet mellom "konføderale" og
"føderale" posisjoner,' Internasjonal Politikk 48, 403-18.

Geyer, R. & Swank, D. 1997. 'Rejecting the European Union: Norwegian Social Democratic
Opposition to the EU in the 19905,' Party Politics 3, 549-62.

Gleditsch, N. P. & Hellevik, O. 1977. Kampen om EF. Oslo: Pax.

Hansson, S. & Teigene, I. H. 1992. Makt og manne fall. Historien om Gro Harlem Brundtland.
Oslo: Cappelen.

Harmel, R. & Janda, K. 1994. 'An Integrated Theory of Party Goals and Party Change,'
Journal of Theoretical Politics 6, 259-87.

Heidar, K. 1994. 'Towards Party Irrelevance? The Decline of Both Conflict and Cohesion in
the Norwegian Labour Party,' in Bell, D. & Shaw, E., eds, Conflict and Cohesion in Western
European Social Democratic Parties. London: Pinter.

Hellevik, A. 1979. Det ansvarlige oppror: AIK og EF-motstanden i Arbeiderpartiet. Oslo:
Universitetsforlaget.

Hovi, J. & Hellevik, O. 1996. 'The Nordic Referenda on Membership in the European Union:
A Different Sequence, A Different Outcome?,' in Wiberg, M., ed., Multidisciplinary Views
on Strategic Inteructions. Turku: Department of Political Science, University of Turku.

Ingebritsen, C. 1998. The Nordic States and European Unity. Ithaca, NY: Cornell University
Press.

Jagland, T. 1990. Min europeiske dram. Oslo: Tiden.

Jenssen, A. T., Pesonen, P. & Gilljam, M., eds, 1998. To Join or Not to Join: Three Nordic
Referendums on Membership in the European Union. Oslo: Scandinavian University Press.

Moses, J. & Jenssen, A. T. 1998. 'Nordic Accession: An Analysis of the EU Referendums,'
in Eichengreen, B. & Frieden, J., eds, Forging an Integrated Europe. Ann Arbor: University
of Michigan Press.

Miiller, W. 1998. 'Party Competition and Plebiscitary Politics in Austria,' Electoral Studies
17,21-43.

Oskarson, M. & Ringdal, K. 1998. 'The Arguments,' in Jenssen, A. T., Pesonen, P. & Gilljam,
M., eds, To Join or Not to Join. Oslo: Scandinavian University Press.

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

Rokkan, S. & Valen, H. 1964. 'Regional Contrasts in Norwegian Politics,' in Allardt, E. &
Littunen, Y., eds, Cleavages, Ideologies and Party Systems. Helsinki: Westermarck Society.

Ruin, O. 1997. Folkomrostningar och parlamentarism. En jåmforande analys av EU-folkomrostningama
Norden hosten 1994. SOU 1997, 56, Appendix 1. Stockholm: Fritzes. '
Saglie, J. 1998. 'A Struggle for the Agenda? Norwegian Parties and the European Issue,

1989-1995,' Party Politics 4, 347-66.

Sjdblom, G. 1968, Pcivty Strategics in ci Miultiptiyty SystcvYi. Lund,' Studcntlittcriitur,
Skjeie, H., Langengen, L. & Rieber-Mohn, L. 1995. '"Aldri mer 1972": Behandlingen av
EU-saken i Arbeiderpartiet,' Tidsskrift for samfunnsforskning 36, 31-53.

Staldvik, J. 1993. Nok er nok. Oslo: Det Norske Samlaget.

Strøm, K. 1990. 'A Behavioral Theory of Competitive Political Parties,' American Journal of
Political Science 34, 565-98.

Svåsand, L. & Lindstrom, U. 1996. 'Scandinavian Political Parties and the European Union,'
in Gaffney, J., ed., Political Parties and the European Union. London: Routledge.
Tsebelis, G. 1990. Nested Games. Berkeley: University of California Press.

Valen, H. 1994. 'Norway: A Storting Election in the Shadow of the EU,' Electoral Studies
13, 169-79.