Abstract
This article scrutinizes the assumption that friends support each other in times of war. Picking up the notion that solidarity, or ‘other-help’, is a key feature of friendship between states, the article explores how states behave when a friend is attacked by an overwhelming enemy. It directs attention to the trade-off between solidarity and self-help that governments face in such a situation and makes the novel argument that the decision about whether and how to support the friend is significantly influenced by assessments of the distribution of material capabilities and the relationship the state has with the aggressor. This proposition is supported empirically in an examination of Sweden’s response to its Nordic friends’ need for help during the Second World War – to Finland during the 1939–1940 ‘Winter War’ with the Soviet Union, and to Norway following the invasion of Germany from 1940 to 1945.
Introduction
‘Friend’ and ‘friendship’ are commonly used terms in the study of international relations (IR), and scholars increasingly recognize and examine friendship as a distinct type of relationship in world politics. This literature often describes friends as bound together by a collective identity based on shared cultural parameters and expressed by cooperative practices. One key behavioural feature of friendship is solidarity, broadly understood as standing by someone’s side and lending support in times of need. Unsurprisingly, the practice of solidarity comes to the fore most clearly during war. As Alexander Wendt (1999: 299; see also Wolfers, 1962: 29) puts it, friends ‘fight as a team if the security of any one is threatened by a third party’.
As intuitive as this sounds, it is a rather sweeping claim that has to be examined more carefully. Wendt does not do this, and neither does existing IR literature on international friendship. Some studies have looked at tensions among friends over military operations in far-away places, exploring their ‘verbal fighting’ of international order (Mattern, 2005), their ability to overcome crises (Eznack, 2011) and how persistent disagreement over the norms guiding military operations can lead to estrangement (Berenskoetter and Giegerich, 2010). Yet an empirical and conceptual gap exists over how a state behaves when the friend faces an overwhelming external aggressor. 1 The present article attempts to fill this gap. This is an important task, as it probes the fundamental assumption that friends fight as a team. More precisely, the scenario forces analysts to confront questions about the factors influencing the nature and scope of solidarity, and how extreme situations affect the ability and willingness to support the friend. As such, it requires attention to why and how practices of ‘other-help’ (Wendt, 1999), as expressions of care, may be compromised by egoistic behaviour, or ‘self-help’ emphasized by realists.
To explore these questions, the article looks at how Sweden reacted when its Nordic friends were under military attack by a powerful external aggressor during the Second World War. Specifically, we focus on the cases of Sweden and Finland during the 1939–1940 ‘Winter War’ with the Soviet Union as the attacker, and of Sweden and Norway during the invasion and occupation of the latter by Nazi Germany from 1940 to 1945. 2 In both cases, the Swedish government decided not to formally side with and officially fight alongside its Nordic friends. Exploring the factors underpinning this decision, the article shows, however, that the decision was not taken lightly and that Sweden still supported its neighbours in a number of ways, even if this support was limited. Thus, rather than reading Swedish behaviour as evidence for the non-existence of friendship, we suggest that it reveals the need to pay more attention to egoism, or practices of self-help, in international friendship and to refine our understanding of the conditions under which friends are willing and able to support each other. Specifically, it shows that solidarity is a political choice, and that to understand this choice analysts need to consider that governments face a (perceived) trade-off between practices of solidarity and of self-help. In this vein, we argue that decisions over the scope and nature of solidarity are significantly influenced by ‘rational’ assessments of the distribution of material capabilities and how the state relates to the friend’s enemy.
Thus, in addition to presenting empirical cases that, to our knowledge, have not been discussed in the IR literature, the article advances scholarship on international friendship in two ways. First, it shows the benefit of including rationalist actors and realist factors for explaining how friends perform in war, presenting a challenge to a literature largely grounded in constructivist thinking. Second, it reminds that analysts need to look beyond joint military action (‘fighting side by side’) to see different kinds of support across society, and to take into account issues of timing. As such, the article reminds that broad and perhaps overly romantic claims about the rules of friendship must be treated with caution and supplemented by a more nuanced understanding of both the variations and limits of practices of solidarity.
The discussion proceeds in five sections. The first part focuses on the issue of solidarity in friendship and outlines the contours of a framework for assessing the trade-off between solidarity and self-help among friends. The ensuing three sections explore these issues empirically in Swedish–Finnish and Swedish–Norwegian relations. The concluding part pulls the insights together and reflects on the implications for the study of friendship in international relations and suggests avenues for future research.
Friendship and solidarity
The point of departure motivating this article is the view that friendship between states is a tangible feature of interstate relations. This is by no means a common position in the field of IR, which has traditionally preferred the default assumption of enmity, or rivalry, as the natural state of affairs. Yet, over the last two decades, scholars have started a serious discussion over the ontology of friendship and how it plays out in international relations (Berenskoetter, 2007; Haugevik, 2018; Koschut and Oelsner, 2014b; Nordin and Smith, 2018b; Roshchin, 2017). The relevant argument for our purpose is that friendship is more than, and qualitatively different from, an alliance in terms of both ontology and behavioural implications.
An alliance can broadly be understood as ‘a formal agreement among independent states to cooperate militarily’ when facing a shared threat (Leeds and Savun, 2007: 1119). It is an instrumental arrangement among rational egoists, designed for one particular purpose, namely for members to collaborate to fend off that shared threat and to protect their respective, and separate, security interests. While alliances can be formed ad hoc, they tend to be manifested in a treaty that spells out the circumstances under which cooperation occurs and the form it takes. As such, it is a type of collective security arrangement based on the expectation that members will live up to their commitments because it serves their self-interests, not necessarily because they trust or care for each other (for a comprehensive discussion, see Snyder, 1997).
Friendship is different. One distinctive feature of the reading employed here is that it entails a form of collective identity, a ‘we-feeling’ that does not presuppose the existence of a shared threat. This identity bond is based on interconnectedness expressed in shared biographical narratives, values and ideas of order, possibly even languages, and close connections between political leaders, government institutions and civil society actors (Berenskoetter and Van Hoef, 2017; Feldman, 2012; Haugevik, 2018; Koschut and Oelsner, 2014a). It renders friendship an ‘affectively charged’ relationship that is valued ‘higher than simply the sum of the material and/or strategic benefits’ derived from it (Eznack and Koschut, 2014: 73). 3 The connections are constituted and maintained through practices; that is, specific behaviour friends expect from each other and that maintain the friendship. In addition to using the language of friendship or related metaphors, such as family, when referring to each other, the most tangible indicator highlighted by IR scholars is that friends solve their disagreements peacefully (Adler and Barnett, 1998; Kupchan, 2010; Oelsner, 2007; Wendt, 1999: 299) and show patterns of routine cooperation during peacetime (Haugevik, 2018). Importantly, friends are also expected to support each other during conflict. Wendt (1999: 299) calls this the ‘rule of mutual aid’, essentially an open-ended commitment to solidarity, broadly understood as standing by someone’s side and lending support in times of need. This commitment is unique to friendship. Whereas in a traditional alliance members collaborate because they each feel individually threatened, friends care and are concerned about one another’s well-being (Koschut and Oelsner, 2014a: 14–15). Thus they support each other even if the security of only one of them is threatened by a third party. Wendt goes as far as to suggest that friends will come to each other’s military aid irrespective of when the aggression takes place and who the attacker is (Wendt, 1999: 300–301; see also Wolfers, 1962: 25–34).
While the commitment to solidarity marks friendship as a unique relationship, its expression is not as straightforward as Wendt contends. Solidarity can take many forms, and the acts by which the ‘rule of mutual aid’ is carried out cannot be specified in advance. In contrast to allies negotiating particular kinds of support in a limited but clearly defined scenario, solidarity among friends is not based on a contract laying out the terms and conditions of its fulfilment. Rather, friends trust each other to give what they can in a given circumstance (Berenskoetter, 2007: 666). Friends may still develop expectations over what they think the appropriate form of support should be, however, which raises the possibility that these expectations are not met. This possibility is evident once we take into account that friendship is not total. As the specialized literature points out, the friend is not simply ‘another Self’ as even close friends maintain a degree of separateness (Berenskoetter, 2007: 668). Even Wendt (1999: 306) notes that ‘the pull of egoism’ still exists among friends, in particular when speaking of states whose very raison d’être is to favour the needs of their own citizens and provide for their own security.
Wendt does not elaborate on this point and, thus, does not consider the potential tension the ‘pull of egoism’ generates with his ‘rule of mutual aid’. Yet, once we acknowledge that there is a place for self-interest in friendship, it is easy to see that, despite the feeling of care that propels friends to help each other, solidarity is neither automatic nor blind. Rather, solidarity and the specific behavioural form it takes is a choice. Simply put, a framework that takes the ‘pull of egoism’ seriously suggests that a government, when faced with the choice about whether and how to provide support, will consider the danger it poses to its own state. Studies have shown that domestic norms can limit the extent to which a state is willing to assist a friend in a military intervention (Berenskoetter and Giegerich, 2010), and we might expect even greater constraints when coming to the friend’s aid is likely to have serious consequences for the state’s own physical security. Thus, when there is an option to remain neutral and avert occupation and destruction, a trade-off emerges between the desire to help the friend and the protection of the security and well-being of one’s own state’s territory and citizens. Put crudely, in such a scenario governments face the dilemma of having to choose between solidarity and self-help. 4
Once this trade-off is recognized, it is necessary to consider the egoistic factors that inform the government’s response to a friend in need and compromise the commitment to solidarity. To start with, there is the question of the type of assistance requested and how it is provided. If a state is called upon for help it will consider what the friend needs, what it has at its disposal and whether support can be given overtly or covertly. It is easier to provide help when the resources requested are readily available. A state will be more hesitant to part with resources that are in short supply and/or costly to mobilize. The way support is provided also matters. Whereas open and significant aid, especially in the form of direct military action are particularly risky, the covert supply of non-military material is less likely to aggravate third parties and may allow the state to stay out of harm’s way.
Taking cues from realist and constructivist reasoning, we argue that the nature and extent of aid a state is willing to provide to the friend in need is significantly influenced by two factors: the perceived balance of material power and the relationship with the third party. Once it is established that a third party is bent on aggression towards the friend, the government will, firstly, assess the distribution of military and economic capabilities in this constellation. A favourable balance on the side of the friends vis-à-vis the aggressor will make the state more confident about the chance of winning or, at least, limits the potential repercussions for helping the friend. As such, it makes states more willing to aid their friend and to do so more extensively. In contrast, a militarily overwhelming aggressor makes it more likely that the state will get severely punished for siding with the friend, putting a high price on solidarity, and hence make aid less likely and less extensive. Secondly, considerations are affected by how the state relates to the external aggressor. Generally speaking, friends can be expected to share a cognitive outlook, or at least go along with each other’s perception of threat in the sense of ‘my friend’s enemy is my enemy’. Yet as friends are ontologically separate units they may on occasion differ in their perception of the external Other, including how they evaluate the threat posed by it and how best to respond to it. 5 When their assessments of the third party differ, friends are likely to refrain from siding unequivocally with each other and may seek a different strategy of engaging the aggressor.
Taken together, in a configuration in which A and B consider each other friends and C is an aggressor posing a military threat to B, there are two basic scenarios:
The worse A’s relationship is with C and the more advantageous the distribution of military and economic capabilities is considered to be in the favour of A and B, the more likely A is to assist B militarily, openly (directly) and extensively.
The better A’s relationship is with C and the more disadvantageous the distribution of military and economic capabilities is considered to be in favour of A and B, the less likely A is to assist B militarily, openly and extensively. Instead, support will be hesitant, covert (indirect) and limited.
These two ends of a behavioural spectrum are ‘ideal types’ in which self-interested assessment interacts with, and is balanced by, the feelings of care for the friend. While the nature and scope of solidarity is a matter of political choice on how to handle the trade-off between solidarity and self-help, it is important to keep in mind that states are not homogenous actors. Given the high stakes, one can expect considerable internal debate and contestation among the political elite, in government institutions and in the wider public over whether and how the friend can and should be supported. Moreover, wars are dynamic and evolve in unpredictable ways, and so we should expect positions and practices to develop, adjusting to new circumstances as they emerge.
Empirical analysis
Three basic indicators of amity – stable peace, shared values, and interconnectedness – suggest that it is plausible to classify the two dyads under investigation as friendships prior to the Second World War. First, both relations were characterized by decades of stable peace. Sweden and Finland have not fought against each other or engaged in serious hostilities since the latter gained independence from Russia in 1917 (Sarkees and Wayman, 2010). Even the territorial Åland Islands dispute between Sweden and Finland that would typically instigate war was settled peacefully between the two countries (Archer and Joenniemi, 2017: 91–92). Sweden and Norway fought their last (brief) war in 1814 when Norway tried to resist ceding its independence to the King of Sweden. There were frictions during the last years of the Swedish–Norwegian union, yet it was peacefully dissolved in 1905. Again, this is remarkable, as these issues were typically settled by the use of force (Cooper, 1991). Moreover, Sweden and Norway have not fought wars or engaged in militarized interstate disputes since, and a number of scholars have argued that stable peace developed between the two states (Deutsch, 1957: 7, 28; Ericson, 2017; Lindgren, 2015; Sarkees and Wayman, 2010). As is expected of friends, the stable peace between Sweden and its Nordic neighbours was also marked by routine cooperation. For instance, Finland and Norway were both among Sweden’s most important trading partners in the 1930s (Veum, 2017: 79)
Parameters of collective identity were present, second, in shared democratic values and political systems. According to the widely used Polity IV dataset (Marshall and Gurr, 2014), as well as the more refined V-Dem dataset, all three countries qualified as democracies prior to the outbreak of the war.
6
According to Jacob Westberg (2012: 92), they were ideologically close and sought to collectively protect ‘Nordic democracy’ against anti-democratic developments in other parts of Europe prior to the outbreak of the Second World War. A sense of collective identity was also manifested, third, in rich cultural and political interconnections grown out of intertwined national biographies. Finland and Sweden shared a long history and formed a unified political unit for almost 700 years. At the time of the Second World War, Swedish was an official language in Finland and the sole official language of the Finnish Åland Islands with a predominantly Swedish population, with some even describing Finland’s national identity as ‘one nation, two languages’ (Anttonen, 2005; Lavery, 2006). The fourth largest political party in Finland was the Swedish People’s Party of Finland representing the interests of the Swedish-speaking minority (Lavery, 2006: 96). In Sweden, Finland was seen as ‘a sister state’ (Agius, 2006: 78). Norway and Sweden also had developed close cultural links, having been united under the same crown for about 200 years (Barton, 2003: xii, 78; Ingebritsen, 2006: 7; Tägil, 1995: 3), including a considerable degree of mutual intelligibility between their two languages.
7
The Swedish term
Although discussions were held throughout the 1930s regarding the possibility of ‘some sort of Scandinavian defensive alliance’, this idea never materialized (Jesse and Dreyer, 2016: 115; see also Lindgren, 2015: 260; Van Roon, 1989). Yet the absence of a formal alliance did not prevent all sides from harbouring expectations and expressing informal assurances of mutual support. In fact, William Trotter posits that ‘Finnish politicians were certain that a defensive alliance with Sweden could be arranged on short notice’ and that Sweden gave many honest informal assurances that they would come to their aid in case the situation worsened (2008: 15). When Germany broke the peace, the leaders of Sweden, Norway, Denmark and Finland convened in Copenhagen in September 1939 and emphasized their neutrality ‘while upholding the right to engage in commerce with one another and the warring nations’ (Sprague, 2010: 29). The hope was that this would allow these Nordic countries to stay out of the war and ensure that they would not end up on opposite sides, while maintaining important trade links and economic ties with different parties (Wylie, 2002: 9 n.4). Nordic heads of state and their respective foreign ministers met again in Stockholm in October 1939 to discuss their worries and to affirm close ties (Veum, 2017: 14–15). When the war came to their doorstep, these affirmations were put to the test, first in Finland and subsequently in Norway.
Sweden and Finland: the Winter War
Following the Soviet invasion of Finland on 30 November 1939, one of the main goals of Finnish foreign policy was to gain Swedish aid and ideally get Sweden to join the war on Finland’s side (Nenye et al., 2015: 188). The Finnish government sent three formal pleas to its Swedish counterpart, asking for military assistance to fight their common enemy, the Soviet Union (Edwards, 2006: 86–87). The pleas referred not only to the shared links between Sweden and Finland, but also to the fact that Russia had been a traditional enemy of Sweden. After all, Sweden regularly fought wars against Russia from the 13th century to the 19th century, with Russia annexing Swedish parts of Finland during the Napoleonic Wars. However, the Swedish government rejected Finland’s pleas for direct military assistance in the war. Prime Minister Per-Albin Hansson maintained the line formulated in his speech to the nation on 1 September 1939, calling on Swedes to ‘join together with calm determination around the great task of holding our nation out of war’ (Reginbogin and Vagts, 2009: 138). He emphasized that Sweden’s foreign policy objectives were purely defensive, with the aim ‘to safeguard the existence of our people’ (Günther, 1943: 24; Moe, 2011: 147). When Finnish Finance Minister Väinö Tanner wrote a letter to Hansson, a fellow Social Democrat and personal friend, to clarify the prospects of Swedish assistance, Hansson responded promptly to his ‘brother’ noting that, even though he sympathized with Finland, Sweden would not get involved (Tanner, 1950: 48).
Hansson tried to divert the blame for this decision by noting that personally he would have liked to do a great deal more, but that his hand was tied: ‘I have to deal with a complacent people which wants to be left in peace’ (Tanner, 1950: 49). Yet the matter was not as straightforward. The Swedish public was overwhelmingly in favour of Sweden intervening in the war on the side of Finland. Large demonstrations were held in support of Finland, and senior officials within the Swedish military argued that it would be better to defend Sweden on Finnish land (Agrell, 2000; Sprague, 2010: 56–57). There also were dissident voices in the Swedish government that advocated for direct military help to Finland. In fact, disagreement over how to respond to Soviet aggression, also in the face of a strong public demand for supporting Finland, led to a crisis in government (Wylie, 2002: 314), which was solved by forming a ‘National Unity’ government. Most notably, the long-serving Foreign Minister Rickard Sandler wanted to give the Finnish government some assurances and send troops to the Åland Islands, but Hansson and the majority of the Swedish parliament rejected this proposal, fearing that such an intervention would drag Sweden into the war. Sandler resigned and was replaced by Christian Günther, a career diplomat who supported Hansson’s goal of keeping Sweden out of the war (Boberg and Wählbäck, 1966: 13, 22, 31, 35, 61; Larsson and Marklund, 2012; Reginbogin and Vagts, 2009: 138).
The government’s unwillingness to provide direct military assistance to Finland cannot be ascribed to a tradition of neutrality, a self-imposed status that had never been absolute (Agius, 2006; Barton, 2003: 10). Rather, most historians seem to agree that it was motivated by the ambition to stay out of the war in the face of an overwhelming military threat. Swedish military strength was at a low point in 1939. Following the First World War, Sweden had downsized its armed forces, a process that continued well into the 1930s and left Swedish military capacity quantitatively and qualitatively far inferior to that of the Red Army (Cronenberg, 1982: 36; Edwards, 2006: 113; Eloranta, 2009: 29). Although the government had gradually increased the defence budget, on the eve of war, the Swedish military was unprepared: strategy, direction and organization as well as attitude, logistics and equipment of the three services – army, navy and air force – were all insufficient for the task ahead (Gilmour, 2011: 210). In fact, the government was motivated by the fear of a military confrontation not only with the USSR, but also with Germany, which had entered into a treaty of non-aggression with the Soviet Union shortly before the Winter War broke out (Van Dijk, 2008: 596–597; Woodworth, 2012: 228–229). Adolf Hitler reportedly warned Sweden not to take sides in the conflict (Sprague, 2010: 163). Taking all this into account, the Swedish government concluded that fighting alongside its Finnish friends against the USSR and potentially even Nazi Germany would likely be disastrous (Edwards, 2006: 150; Johansson, 1973: 368). Similarly, the Swedish King Gustav V Adolf asserted that he could not allow Sweden to get pulled into the war by providing official military assistance but maintained that it would continue to provide humanitarian and volunteer help to its Finnish friends (Sprague, 2010: 161). In short, when weighed against the prospect of a two-front war, Finnish requests for direct military assistance were not favoured in Stockholm. Realpolitik and a concern with the distribution of material capabilities had taken the upper hand (Gilmour, 2011: 95; Scott, 2002: 376).
Yet the government’s decision against direct military support did not suppress ‘the friendship and compassion that the Swedes felt for the Finns’ or their antipathy towards the aggressor (Sprague, 2010: 55). 8 As noted earlier, the Swedish public generally showed sympathy for Finland and identified with Finnish suffering. Numerous newspapers criticized Hansson’s decision, and civil society actors launched a range of initiatives in support of Finland. Throughout the war, Swedish scholars, authors and other prominent figures gave speeches and held well-attended solidarity gatherings throughout the country (Sprague, 2010: 62). Volunteers donated food and clothing, and a ‘Committee for Finland’ organized the collection of monetary donations. Ninety thousand Swedes worked on the holiday of 6 January 1940 and raised nearly 1m kr for Finland’s war effort. According to Martina Sprague (2010: 62), the total sum of donated items, as well as credits and loans given by Swedish banks, amounted to 13b kr, or US$1.6b. 9 The government and civil society organizations sent food, clothing and medicine to Finland. Sweden also accepted the majority of the 70,000–80,000 Finnish war children who were evacuated across the Nordic countries, many of whom were taken into the homes of Swedish families (Kuusisto-Arponen, 2011: 181; Saffle, 2015).
Even the government was balancing its decision to prioritize self-help with covert acts of solidarity, walking ‘a diplomatic tightrope when determining if and how it would support Finland’s cause’ (Sprague, 2010: 1). By declaring Sweden ‘non-belligerent’ rather than ‘neutral’, the government indicated that it was open to support Finland in other ways. Militarily, Sweden strengthened its borders by placing approximately 100,000 soldiers at the Swedish side of the Torne Valley to relieve Finnish forces in the area and prepare for a potential Soviet attack against its own territory (Larsson and Marklund, 2012). Sweden also sent military aid in the form of weapons and ammunition across the border. An official Finnish announcement after the war noted that Sweden supplied Finland with 90,000 rifles, 2,000,000 rounds of ammunition, 80 anti-tank guns and 250 other cannons, among these 100 anti-aircraft guns. According to this statement, Sweden’s most vital contribution to the Finnish war effort was its supply of munitions (Tanner, 1950). The government also allowed Finnish submarines to be repaired in Sweden on the condition that the crew did not wear uniforms (Sprague, 2010: 59). In addition, volunteer soldiers from Sweden began to arrive in Finland in January 1940 (Nenye et al., 2015: 192). Initially, Swedish government forbade recruitment centres to advertise in newspapers, as it did not want the practice to become public knowledge, but eventually gave permission for advertisements on public transport. Yet the volunteers were not allowed to carry their weapons with them. Over the next two months, 8260 Swedish volunteers out of 12,705 applicants stood ready to support their neighbour (Sprague, 2010: 3), their sentiment expressed in the slogan ‘Finland’s cause is our cause’ (Saffle, 2015: 222). While it is not clear how much these volunteers actually contributed to the Finnish war effort, 10 they symbolized the commitment to solidarity within Swedish civil society. Thus, despite the government’s official rejection of Finnish pleas for Swedish troops, in the end Sweden was the greatest foreign contributor to the Finnish effort during the Winter War (Jowett and Snodgrass, 2006: 21) and played a central role in mediating the peace between Finland and the Soviet Union in March 1940 (Wylie, 2002: 314).
In sum, both the balance of power and the relationship to the aggressor significantly influenced how Sweden dealt with the solidarity/self-help trade-off, as our framework suggests. The highly unfavourable distribution of military capabilities was a major reason for the government’s reluctance to officially join the war on Finland’s side, as this would have posed great danger to Sweden’s own physical security. At the same time, identification with the Finnish plight and antipathy towards the Soviet Union generated various forms of informal support across government and civil society.
Sweden and Norway: the German occupation
German forces invaded Norway (and Denmark) on 9 April 1940. While some in the Swedish press expressed ‘shock and outrage’, the government in Stockholm was not entirely surprised by the invasion (Wylie, 2002: 316). 11 The following day, the President of the Norwegian Parliament, Carl Joachim Hambro, arrived in Stockholm to ask for help and military support on Swedish radio. The Germans found out about Hambro’s plans and told the Swedish government to forbid him from doing so or face potential consequences (Veum, 2017: 37–38). Sweden complied, quickly declared itself neutral and turned down Hambro’s requests. By proclaiming itself ‘neutral’ rather than ‘non-belligerent’, the Swedish government signalled that its aid to Norway was going to be far more limited than that provided to Finland during the Winter War (Agrell, 2000). Prime Minister Hansson ruled out an intervention and informed the Norwegian government that Swedish weapons deliveries to Finland had been an exception and that Norway could not expect such help from them.
On 12 April 1940, the government prohibited Swedish military personnel from enlisting in the Norwegian military forces or participating in acts of war in Norway, nor would Sweden supply military equipment or munitions that could be used by Norwegian soldiers. Only food supplies provided directly to Norwegian civilians were allowed. The government was not willing to risk doing anything that the Germans might interpret as a breach of Swedish neutrality (Veum, 2017: 49). That evening, Hansson gave a speech over Swedish radio voicing that Sweden felt compassion for its Scandinavian
The Swedish reluctance to fight on the side of Norway was driven by the objective to maintain a ‘separate peace’ (Veum, 2017: 58), and can be understood as a combination of three factors. First, again, was an assessment of the distribution of military capabilities. The Swedish military was still weak and poorly equipped, and the German invasion occurred right after the Winter War, when Sweden had already provided a considerable amount of resources to Finland. Moreover, the border against Norway was principally unmanned, and although Sweden mobilized its forces, these attempts were rather futile. German forces moved into Norway at a much swifter pace than the Soviet forces had done in Finland earlier. Even western allies had largely been caught by surprise, and attempts by Britain and France to prevent the German occupation of Norway did not succeed. Once Germany was in control of Norway, it could have launched an attack against Sweden from every direction and beaten its forces with ease (Agrell, 2000). Swedish fears of provoking Germany were well founded as Berlin had prepared plans for invading northern Sweden, should that become necessary (Tamelander and Zetterling, 2005: 224).
Second, this overlapped with the distribution of economic capabilities. The Swedish government was aware of Germany’s need for access to Swedish resources such as iron ore, timber products, and ball bearings. Some estimates suggest that Germany imported 50% of its iron ore from Sweden, which was vital for its war effort (Leitz, 2000: 65; Sprague, 2010: 153). The German occupation of Norway was largely motivated by the need to secure this vital resource and to prevent Britain taking control of strategically important Norwegian harbours. Encircled by Nazi Germany and with no other major trading partners in sight, Sweden became heavily dependent on Germany for trade, which accounted for 70–80% of all Sweden’s trade from the summer of 1940 until the summer of 1944. Thus, in the calculation of a Swedish government determined on maintaining its physical security, complying with German demands whilst reaping economic benefits was better than entering into a confrontation and prompting the Nazi regime to extract the resources it needed by force (Stephenson and Gilmour, 2013).
Third, in contrast to the Winter War, Sweden related differently to the aggressor. Whereas Russia/the Soviet Union had the role image of an ‘arch enemy’, Sweden had long-standing cultural, political, economic and even military ties with (pre-Nazi) Germany. As Neville Wylie (2002: 319) notes, this created ‘a situation of some normalcy with Nazi Germany’. That is not to say that the Swedish Prime Minister or the Swedish population at large regarded the Nazi regime as friends. Hansson had voiced his opposition to the Nazis as early as spring 1933, and the vast majority of the Swedish population favoured the Allies in the war (Leitz, 2000: 51). Yet Germany was clearly not regarded as being as threatening as the Soviet Union; as one historian put it, ‘to fight the “kulturstat” Germany could . . . appear unnatural’ (Wylie, 2002: 316). 12
In light of these three factors the Swedish government decided to orient its policy in a way that caused ‘the least possible friction’ with Berlin (Scott, 2002: 379; see also Agius, 2006: 81). This stance significantly reduced the scope for solidarity with Norway. In fact, together the above factors overwhelmed and effectively neutralized the dilemma between solidarity and self-help in favour of the latter, at least for the time being. The government refrained from providing Norway with weaponry or support volunteers to join the Norwegian resistance campaign. Indeed, it led to the active disarmament and detainment of Norwegian troops that were based in Swedish territory, all in line with the official protocol of strict neutrality (Agrell, 2000). When King Haakon VII of Norway, Crown Prince Olav and the members of the Norwegian government requested permission to move freely in Sweden and return to Norway when circumstances allowed, the Swedish government neglected their pleas. A Norwegian request to transport ammunition through Swedish territory was also denied (Reginbogin and Vagts, 2009: 139). By contrast, Sweden came to assist the German war effort. After initially refusing the transfer of German troops and goods through Sweden, the government soon permitted German transit traffic, even though it constituted a breach of its declared neutrality and was against the wishes of the Norwegian and the British governments (Leitz, 2000: 52–56; Lorenz-Meyer, 2007: 14). For three years, munitions and a continuous stream of uniformed but unarmed German soldiers travelled by rail from southern to northern Norway via Swedish provinces (Agrell, 2000; Leitz, 2000: 63; Wangel, 1982: 498). In addition, the government eventually let Germany use Swedish airspace and conduct military transportations through its waters (with Swedish Navy escort) and allowed Germans to use Swedish telephone and telegraph lines and to censor anti-German opinion in the Swedish press (Agrell, 2000; Scott, 2002: 374 n.4).
The Norwegian reaction to Swedish lack of support and its aid to Germany was negative (Ekman, 2005). Norwegians ridiculed Swedish neutrality, and the contact between the exile government and the Swedish authorities was sparse and anything but friendly (Veum, 2017: 377). The disappointment felt at the time was exemplified in a stinging letter that Norway’s Prime Minister Johan Nygaardsvold sent from his London exile to his party colleague based in Stockholm in December 1940. In the letter he asked his colleague to convey his anger to Hansson and tell him that he wanted to see ‘the Germans get hunted out of Norway and . . . to live long enough to give him and his entire government a proper dressing down’, adding that there ‘is nothing, nothing, nothing I hate with such passion and wild abandon as Sweden – and it is his (Hansson’s) fault’ (Local Norway, 2012). It was a reaction from a friend who feels let down, if not betrayed.
Yet Sweden did show some solidarity with Norway, and support grew as the war went on. Somewhere between 100 and 300 Swedish volunteers entered Norway to fight on the Norwegian side (Agrell, 2000; Gyllenhaal and Westberg, 2008) 13 and, for many Norwegians, Sweden became a safe place. Volunteers helped Norwegian Jews to make their way into Sweden, and, throughout the war, Swedish authorities accepted roughly 50,000 Norwegian refugees and ensured their safety once they had crossed the border (Winkler, 2015: 767). As the tides of war began to shift in 1943 and Allied pressure mounted, Sweden stepped up its support. Pro-Norwegian sentiment also grew among the Swedish public with reports of oppressive German practices against Norwegian civilians, including death sentences, deportations and mass arrests (Ekman, 2005). As a result, the Norwegian government in exile was granted permission to train ‘police troops’ in Sweden, which included the Norwegian refugees who had sought protection in Sweden. These troops received weapons from the Swedish army and by 1945 over 12,000 men had been trained and equipped with grenade launchers, anti-tank weapons and other unusual police equipment (Gilmour, 2011: 103). A small fraction of these troops were allowed to operate in Finnmark, as Sweden regarded this area ‘liberated’, but the bulk of these troops were only deployed in Norway after the German invasion ended to help with the reconstruction (Mann, 2012: 183, 208–211).
Summing up this case, once again the distribution of military capabilities and, in addition, economic interests and the relationship to the aggressor shaped Sweden’s approach to the solidarity/self-help trade-off. Initially, Sweden prioritized ‘self-help’ since military and economic factors heavily favoured compliance with German demands, facilitated by closer links with Germany compared to a history of enmity with Russia/the USSR. However, as expected by our framework, with the decline of German power, Swedish solidarity and support towards its Norwegian neighbours grew.
Conclusion
This article has critically explored the assumption that friends fight together when facing an external enemy by pointing to the tension between constructivist solidarity and realist self-help, a tension largely neglected in the IR literature on friendship. At first sight, the two empirical cases seem to show that self-help considerations reign supreme. Despite the strong bond Sweden had with both countries prior to the war, the Swedish government did not provide Finland and Norway with direct official military assistance when they were attacked by external powers. Yet a closer look revealed a more nuanced picture. The article has shown that, despite facing an overwhelming enemy, care for the friend did not simply fall by the wayside. In the case of Finland, calls for solidarity existed even in parts of the Swedish government, covert military assistance was provided, and at the level of civil society there was compassionate support and a variety of initiatives under the banner ‘the Finnish cause is ours’ (Lagercrantz and Hillebrand, 1939). Although the aid to Norway was more limited, it did become more pronounced from 1943 onwards. In both cases, one might also argue that Swedish military involvement would have escalated the war in the region and deprived Finland and Norway of a safe refuge. So rather than interpreting the official Swedish position as evidence that there had been no friendship between these nations, one must take into account that Swedish leadership grappled with the solidarity/self-help trade-off and appreciate the practices of solidarity that did occur, even if they fell far short of the ideal.
As our framework suggests, concerns about the distribution of material capabilities weighed heavily in the Swedish decision on how to respond to Finland’s and Norway’s requests for help. For the government, entering the war on the side of its friends and, thus, confronting Soviet and Nazi Germany was simply too dangerous. The expected costs to Sweden’s physical security and material interests were too high. The discussion also showed that Sweden’s relationship to the attackers played an important role in determining the ability and willingness to provide support. Whereas Sweden and Finland shared a negative view of the Soviet Union and considered it a common enemy, Sweden’s attitude towards Germany was less antagonistic. Long-standing cultural ties and vital trade relations with Germany partly explain why, at least initially, the Norwegian cause did not generate the same display of solidarity as the Finnish cause and why Swedish aid to Norway was so limited.
These insights should prompt scholars of international friendship to be more cautious with broad theoretical claims about virtues of loyalty and solidarity associated with friendship, and to refine their understanding of how friends act in the context of war. As this article has shown, it cannot simply be assumed that friends fight as a team against third parties. To understand (variation in) scope and nature of practices of support, the act of solidarity must be regarded as a choice, rather than a fixed rule, and the factors influencing this choice must be explored carefully. In this regard, scholars of international friendship need to better integrate ‘the pull of egoism’, or self-help, into their framework and acknowledge that the decision of whether and how to support the friend is made not only on the basis of care, but also by assessments of the distribution of material capabilities, as well as the relationship the state has to the aggressor. The findings also remind us of the importance of context and of paying attention to behaviour at the level of government
Footnotes
Acknowledgements
The authors thank the anonymous reviewers and the participants at the Annual BISA Conference (Bath, June 2018), the War Studies Seminar at the Swedish National Defence University (Stockholm, April 2018) and the Annual ISA Convention (Baltimore, February 2017) for their insightful suggestions.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Felix Berenskötter would like to acknowledge the financial support of the Leverhulme Trust (RF-2017-520: Friendship in International Relations).
