How to increase cosmopolitanism on an individual level : the effect of interstate war on the development of a cosmopolitan identity

(1)

How to increase cosmopolitanism on an individual level

the effect of interstate war on the development of a

cosmopolitan identity

Sjoerd Jelmer Dreteler

Master thesis Political Science

International Relations

June 2017

Student no. 11009551

Supervisor: A. F. Voicu

(2)

Abstract

For a cosmopolitan system to succeed, there must be legitimacy in the form of support from individuals. However, it has only been recent that scholars started to explore cosmopolitanism on an individual level (the cosmopolitan identity). Up until now, only travel, use of the internet, place of living, occupational status, income and freedom of the press has been found to be related to cosmopolitanism. This thesis explores if there is also a relationship between these indicators and the cosmopolitan identity. Next, this thesis will also examine what the effect of war on the development of a cosmopolitan identity is. The thesis is divided by three sub-questions. First, does the national identity form a barrier to the development of a cosmopolitan identity? Second, how can the relationship between the previously found indicators of cosmopolitanism and cosmopolitanism itself be explained from a psychological perspective? Third, what is the effect of interstate war on the development of a cosmopolitan identity during and after the interstate war? All three sub-questions contribute to the main research question: how can cosmopolitanism on the individual level be increased?

To answer the sub-questions and, eventually, the research question, the thesis consists of several chapters. In the first theoretical chapter, I explored if the national identity could form a barrier to the cosmopolitan identity. Based on the theory it seemed indeed that the national identity could form a barrier to the cosmopolitan identity to develop, especially during interstate war. This is the first contribution to the scientific literature. In chapter three, I used psychological theories to provide a theoretical framework for the previously found indicators of cosmopolitanism. According to this theoretical framework, the indicators can be clustered into three categories, indicators related to the amount of contact, indicators related to perceived threat, and indicators related to a barrier. This theoretical framework is also a contribution to the scientific literature. In chapter 4 I explained more about the possible role of interstate war on the cosmopolitan identity. Besides the previously found indicators of the cosmopolitan identity, I added ‘fear of war’ and ‘the time after the last interstate war’ as a potential indicator. The latter one was also related to a barrier. In chapter 5 and 6 I discuss my research and the results. I conducted 4 multiple linear regression analyses and was able to confirm 7 out of my 8 hypotheses. Only the relationship between travel, as an indicator, and the cosmopolitan identity appeared negative, while it was hypothesised as a positive relationship. The results also showed that freedom of the press and the time after the last interstate war have relatively strong predictive values. Based on the results in chapter 6 I answered my third sub-question and concluded that war has a negative influence on the cosmopolitan identity in two ways. First, war strengthens the national identity which in turn forms a barrier to the development of a cosmopolitan identity. Second, fear of war has a negative effect on the cosmopolitan identity.

(3)

So, in the absence of and without fear of war, the cosmopolitan identity will develop. These findings were also a contribution to the scientific literature. Based om the research I was able to provide a final answer to my research question in chapter 7. One of the most important findings of this thesis is that barriers play a big role in preventing the cosmopolitan identity to develop. Taking away barriers, for example by increasing freedom of press or ending interstate war, leads to increased cosmopolitanism. The findings in this thesis are relevant because it showed how political identities could be influenced top-down and it contributes to existing knowledge and theories, such as the liberal/democratic peace theory.

(4)

Chapter 1. Introduction

1. 1.

Political identities

Studying social identity started decades ago, and over the years our understanding of identity has grown. Identity addresses the “structure and function of the socially constructed self”, in which the socially constructed self can be seen as “a dynamic construct that mediates the relationship between social structure or society and individual social behaviour” (Hogg, Terry & White, 1995, p. 262). Put differently, social identity is defined as the part of the self that is created by placing the self in a social context. The first studies about identity focused on the different roles people have. A person can be a son or daughter but in the meantime also a grandson or granddaughter (Brewer, 2001; Brown, 2000). According to Brown (2000), identity is always related to others. This means identities can coexist and serve different functions in different situations. Even identifying oneself as, for example, a sports lover is unintentionally related to others because it creates awareness of the fact that one’s love for sports distinguishes them from others (Tajfel & Turner, 2004). Therefore, the way in which one identifies oneself is socially related, and identities can coexist.

There are more things known about the social identity. First, the context also determines which part of the identity is salient (Ashforth & Mael, 1989). When someone asks a person that is on holiday to a foreign country to describe him- or herself, he or she is likely to identify in relation to the country of origin. The context makes, in this case, the ethnic identity salient. Second, being made aware of one’s identity can also have an influence on identity. Bruter (2003) showed that media play a role in shaping and creating identities and Scheffler (1999) argued that the individual identity is not fixed. Third, people’s identities change, are “being modified, updated, altered, supplemented, recast, and reconceived” (Scheffler, 1999. p. 257). This means identities can also develop. So, identities are not fixed, being made aware of one’s identity influences one’s identity and identities can develop.

Part of the identity is the political identity, which is also a social identity. For the purpose of this thesis, I describe political identities as social identities related to one’s political view on the world. Since the creation of modern states, people have a regional identity and a national identity (Beck & Sznaider, 2010) and more recently scholars acknowledge more superordinate identities such as the European identity and the cosmopolitan identity (Beck, 2002). All of these identities can be seen as political identities. The regional identity can be defined as the extent to which one identifies with the region in which one lives and the extent to which one perceives others and behaves towards others as being part or as not being part of this exclusive local group. In line with the definition of regional identity, national identity can be defined according to the “nation within a state territory, on the basis of citizenship. In this definition, it is understood that all the citizens of one state, regardless of their ethnic, cultural, language, historical, religious or other origin make a nation” (Čorkalo &

(7)

Kamenov, 2003, p. 86). Moreover, the national identity can be seen as a “multidimensional concept, with a sense of belonging, positive feelings and loyalty as being essential parts” (Čorkalo & Kamenov, 2003, p. 87). This means that people within state territory with a national identity consider themselves as belonging to the state, have positive feelings towards the state or its members and are loyal to the state. Everybody in state territory is included in this definition and they can be seen as a unitary group.

Research about political identities is divided about the interaction between political identities. Some scholars argue that strong political identities cannot coexist (Carey, 2002; Hoffmann, 1966; Hooghe & Marks, 2008; Medrano & Gutiérrez, 2001), meaning, for example, that a strong regional identity cannot go together with a strong national identity. While others argue for the possible existence of a positive relationship between regional/national identity and a supranational identity (Cinnirella, 1997; Medrano & Gutiérrez, 2001; Stevenson, 2005), meaning, for example, that a strong regional identity coexists and even stimulates a strong national identity. This indicates that the relationship between different political identities is complex and is in need of further exploration. Previous research already focused extensively on the relationship between the regional identity and the national identity (Faulkner, 2016; Hoffmann, 1966; Hooghe & Marks, 2008; Medrano & Gutiérrez, 2001; Stevenson, 2005) but not on the relationship between the national identity and the cosmopolitan identity. Therefore, part of this thesis will focus on the relationship between the national identity and the cosmopolitan identity.

The central idea in cosmopolitanism is “that all humans, can and/or should, consider themselves as belonging to a universal group that includes every person, and should treat all persons with equal moral concerns” (Faulkner, 2016. p. 1-2). Cosmopolitans perceive themselves as citizens of the world (Pichler, 2012; Robbins, 1992; Scheffler, 1999; Schueth & O’loughlin, 2008; Szerszynski & Urry, 2002) and have an openness towards other cultures and other people (Cwerner, 2000; Matthews & Sidhu, 2005; Scheffler, 1999; Szerszinski & Urry, 2002). The cosmopolitan identity is the level of cosmopolitanism in the individual (Beardsworth, 2008). For the purposes of this thesis, I define cosmopolitanism as the extent to which individuals identify themselves with the world, are open to other individuals and trust other individuals. The more one identifies with the world, is open to others, and trust others, the stronger the cosmopolitan identity. In relation to the regional identity and the national identity, it seems that there is a hierarchical order in which the regional identity is the lowest level of identity and cosmopolitanism the highest level of identity. The national identity stands between the regional identity and the cosmopolitan identity and is therefore theoretically the closest political identity to the regional identity and to the cosmopolitan identity.

Because identities are well studied in psychology and some scholars suggest for more interdisciplinary research between psychology and political science (Brock & Atkinson, 2008) “in which concepts from different disciplines and theoretical traditions can be brought together in an

(8)

integrative framework” (Brewer, 2001, p. 123), this thesis will use psychological theories and applies them to a political topic area. The first contribution of this thesis to the scientific literature is, therefore, to explore the relationship between the national identity and the cosmopolitan identity, from a psychological perspective. By going through psychological identity theories I will slowly build a psychological framework to understand the relationship between the national identity and the cosmopolitan identity. To do so, my first sub-question will be: does the national identity form a barrier to the development of a cosmopolitan identity? This is important because if the national identity indeed is a barrier to the development of a cosmopolitan identity there might be other barriers for the development of the cosmopolitan identity as well. It is also important because if the national identity is indeed a barrier to the development of the cosmopolitan identity, this would suggest that taking away this barrier could lead to increased cosmopolitan identity.

1. 2.

Previously found indicators of cosmopolitanism

It has only been recent that theorists started to explore “how individuals might be encouraged to act as cosmopolitans in practice” (Faulkner, 2016. p. 2). Thus far, several indicators have been found to influence cosmopolitanism or might be related to cosmopolitanism on an individual level. First, higher educated people tend to be more cosmopolitan compared to less educated people (Pichler, 2012; Schueth & O’loughlin, 2008; Skrbis, Kendall & Woodward, 2004; Teney, Lacewell & de Wilde, 2014). Second, people that are cosmopolitan have been found to travel more and use the internet more than non-cosmopolitan people (Phillips & Smith, 2008). Therefore, it seems plausible that travel and internet usage are determinants of cosmopolitanism. Third, the place of living was also found to be related to cosmopolitanism. People in urban areas are found to be more cosmopolitan than people in rural areas (Pichler, 2012). The main argumentation behind the latter three is the assumption that people who travel, use the internet and live in urban areas are more in contact with foreign cultures. However, a more theoretical explanation is still lacking and the relationships between the indicators and cosmopolitanism are still unclear. Fourth, income and occupation were found to be positively related to cosmopolitanism (Pichler, 2012). The reasoning behind this was, the same as for education, that people with higher education, higher income and an occupation, would be less at risk in a competitive globalising world (Teney, Lacewell & de Wilde, 2014). Therefore, scholars suggested that cosmopolitanism was only possible for wealthy countries and the more fortunate people in the world (Matthews & Sidhu, 2005; Roudometof, 2005) and was also called elite-cosmopolitanism. However, the effect of the ‘elite cosmopolitan variables’ appeared less straightforward than it had seemed. Furia (2005) explored if cosmopolitanism disproportionately appeals to privileged individuals and privileged societies. In his study, Furia (2005) compared different societies on a global scale and concluded that cosmopolitanism was not a purely elitist

(9)

phenomenon. Therefore, also a clear theoretical explanation for these ‘elite cosmopolitan variables’ is missing.

On the state level, some research about indicators of cosmopolitanism has been conducted. Norris and Inglehart (as cited in Pichler, 2012) conducted research in which they used a combined scale as an indicator. This scale consisted of economic globalisation, press freedom, and economic development. However, because their scale was combining different indicators, it was hard to draw separate conclusions about economic globalisation, press freedom, and economic development. Pichler (2012) also conducted research with this scale and suggested that these indicators could be “strong predictors of the ethical dimension of cosmopolitan orientations” (p. 34). Because Furia (2005) showed that economic prosperity on a state level was not to be related to cosmopolitanism, the findings of Pichler (2012) only suggests a relationship between freedom of press and cosmopolitanism. But again, a theoretical explanation is missing.

Concluding, several indicators have been found to relate to cosmopolitanism. However, a clear theoretical explanation is missing at the moment, although it seems clear that a good theoretical explanation is a necessity. Subsequently, the second contribution of this thesis to the scientific literature is to provide a theoretical framework that can explain the relationship between the previously found indicators of cosmopolitanism and cosmopolitanism. Therefore, my second sub-question will be: how can the relationship between the previously found indicators of cosmopolitanism and cosmopolitanism itself be explained from a psychological perspective?

1. 3.

Previous research and current research

Despite the fact that political identities are part of the self, several scholars found that it can be externally influenced (Bruter, 2003; Čorkalo & Kamenov, 2003; Cram, Patrikios & Mitchell, 2011; Medrano & Gutiérrez, 2001). First, Čorkalo and Kamenov (2003) found that “war certainly shaped…national values and feelings about their national group, making this aspect of group identity especially salient and important” (p. 91). Second, Reading (2011) argued that: “historic examples show that during war… regimes ban and burn books, demolish memorials and museums, build roads through cemeteries and erect shopping centres on historic sites” (p. 4) with the purpose of enforcing the national identity or creating a national identity. And third, according to Medrano and Gutiérrez (2001) rulers in the twentieth century have been using media as an instrument to instil a strong sense of national belonging in the population. So during a war, regimes/governments are actively engaging in actions that stimulate the national identity.

However, the kind of wars in the nineteenth and twentieth century differ (Kaldor, 2007). Therefore it is needed to mention explicitly how wars will be treated in this thesis. This thesis will focus on national group identity and cosmopolitan group identity, so the lowest group identification

(10)

level is on nation-state level. Subsequently, only interstate wars will be used in this thesis. In interstate war, individual nations are at war with each other which emphasises the national identity.

Thus far I argued that the national identity might form a barrier to the development of a cosmopolitan identity and the national identity will be actively stimulated during an interstate war. This means that if my first sub-question is correct, there would not be a cosmopolitan identity or at least no strong cosmopolitan identity during an interstate war. However, after an interstate war, the cosmopolitan identity can develop. This means that interstate war might have consequences for the development of the cosmopolitan identity during and after the interstate war. Therefore, the third contribution of this thesis to the scientific literature is to explore the effect of interstate war on the cosmopolitan identity and therefore third sub-question is: what is the effect of interstate war on the development of a cosmopolitan identity during and after the interstate war? This is important to explore because it would suggest the cosmopolitan identity can be influenced top down. Put differently, if there are certain barriers such as war that prevent the cosmopolitan identity to develop it would imply that governments can actively shape policies to deliberately stimulate or oppress the cosmopolitan identity.

According to Beck (2002), “after communism and neoliberalism, the next big idea is needed – this could be cosmopolitanism” (p. 20). This could be a perfect system because the world is getting more and more interconnected as a consequence of globalisation. This globalisation leads to a pluralization of borders (Beck, 2002) which is defined as “the pluralization of nation-state borders or the implosion of the dualism between the national and the international” (Beck, 2002, p. 19). Put differently, state borders are becoming less relevant because of globalisation and, therefore, the relationship between states and the international system changes. The position of the nation state weakens and states lose power legitimacy (Beardsworth, 2008; Bromley, 2009). One of the consequences of this process is the shift of power to a higher level, such as a European level or an international level (Bromley, 2009). However, for an integrated system on such a higher level to succeed, wielders of power must be seen to be legitimate not only in the eyes of other wielders of power but also in the eyes of individuals (Beardsworth, 2008). It is, therefore, important to explore how this legitimacy can be increased for the individual. So, how can the individual become more cosmopolitan, in order to create strengthened power legitimacy for a cosmopolitan world system? Subsequently, the 3 sub-questions contribute to the main research question: how can cosmopolitanism on the individual level be increased?

1. 4.

Thesis overview

To explore this research question, chapter 2 will focus on the relationship between the national identity and the cosmopolitan identity. In this chapter, the social identity theory (SIT) (Tajfel & Turner, 2004) and the self-categorization theory (SCT) (Turner & Reynolds, 2012) will be used to

(11)

explain how group identities are formed and why the national identity is emphasised during an interstate war. After that, I will explain how this strengthened national identity can function as a barrier to the cosmopolitan identity to develop during a war. This will also answer the first sub-question.

In the third chapter, the relationship between group identity and the cosmopolitan identity will be explored more thoroughly. Using the contact hypothesis (CH) (Allport, as cited in Brown, 1993) and the mutual intergroup differentiation model (MIDM) (Hornsey & Hogg, 2000) I will provide a theoretical framework to explain the relationship between the previously found indicators of cosmopolitanism and the cosmopolitan identity itself. This will answer the second sub-question. Chapter three will also include the hypotheses of the relationship of previously found indicators of cosmopolitanism and the cosmopolitan identity. ‘Fear of war’ will be added as a possible new indicator, which will, therefore, be included in the analyses and will help to answer the third sub-question.

Chapter four will focus on the development of identities after an interstate war. I will argue how the cosmopolitan identity can develop when a war ends and why time after the interstate war, therefore, can be a possible indicator of the cosmopolitan identity. The role of collective memory (Beck, 2002) will help to understand how intergroup boundaries slowly decrease over time and how the cosmopolitan identity slowly increases over time.1

In chapter five and six I will explain and conduct my analyses. By conducting multiple linear regression analyses, the hypotheses were tested. At the end of chapter six I will provide a complete answer on the third sub-question: what is the effect of interstate war on the cosmopolitan identity? Based on the results, time after an interstate war is positively related to the cosmopolitan identity. This shows that war indeed forms a barrier to the cosmopolitan identity to develop. As soon as war ends, the cosmopolitan identity will slowly increase. The findings also show that fear of war has a negative influence on the cosmopolitan identity, which means that increased fear of war leads to decreased cosmopolitan identity. Together, this shows that war has a negative influence on the cosmopolitan identity in two ways. First, when war is present, the cosmopolitan identity cannot develop, second when fear of war increases, the cosmopolitan identity decreases. In chapter seven, I will use all my results to answer the research question and mention the implications. Finally, I will mention limitations, give suggestions for further research and conclude.

1

Due to use of psychological theories in a political science topic area, I deliberately decided to divide the theoretical chapters in three separate chapters, each with a different focus and building forward on each other.

(12)

Chapter 2. The national identity as a barrier to the development of the

cosmopolitan identity

After the second world war, psychologists were struggling to find an explanation for the increased nationalism in Germany during the second world war (Hornsey, 2008). At first, explanations were sought in Freudian theories, focusing on frustration and unresolved inner conflicts. However, these explanations were not sufficient to explain group behaviour and as a response, two new theories were created; the SIT and the SCT (Hornsey, 2008). In this chapter, I will combine the SIT and the SCT to explain how the national identity is strengthened during an interstate war and how this influences the cosmopolitan identity.

2. 1.

The social identity theory

The first theory that was developed to explain group behaviour and group identity was the social identity theory (SIT). In the early 1970’s Tajfel conducted a set of experiments to explore human group behaviour (Hornsey, 2008). In these experiments, participants were first divided into groups and were subsequently asked to perform a task. Dividing into groups was in every experiment based on meaningless or arbitrary criteria. However, other conditions differed, in some experiments this meaningless dividing of participants was made explicit while in other experiments this was not explicit. Also, in some experiments, there was no interaction between groups while in other experiments the groups were able to interact. Lastly, in some experiments, group membership was completely anonymous while in other experiments it was clear which participant belonged to which group. Despite all the differences in the experimental conditions, the outcomes were the same in every experiment. After being allocated to a group, every participant showed favouritism towards their own group and discrimination or a form of hostility towards the other groups in the task later in the experiment (Ashforth & Mael, 1989; Hornsey, 2008; Tajfel & Turner, 2004). These experiments showed that even when allocating in groups was completely based on arbitrary or meaningless criteria and there was no sufficient reason, i.e. there were no gains, people behaved differently towards members of the own group compared to members of another group. So, there was intergroup discrimination. This phenomenon is therefore called the ‘minimal group paradigm’ (Tajfel & Turner, 2004). These experiments were the foundation for the SIT.

In the SIT Tajfel and Turner (2004) introduced the term ‘in-group bias’. The ‘in-group bias’ is “the tendency to favour the in-group over the out-group in evaluations and behaviour” (Tajfel & Turner, 2004, p. 13) and is part of the ‘minimal group paradigm’. This in-group bias is so strong that people sometimes even prefer to reduce their own gain if that increases the relative gains towards other groups (Hornsey, 2008). The more aware people are of the existence of an out-group, the stronger the in-group bias will be. Put differently, the more aware people are of the existence of an

(13)

out-group, the more intergroup competition will appear and the more discriminatory people will react towards the out-group (Tajfel & Turner, 2004).

The SIT contributes to the understanding of how and when the national identity is formed. Based on the ‘minimal group paradigm’ even arbitrary or meaningless criteria were enough to create in-group favouritism and out-group discrimination. In relation to the national identity, state borders are in my opinion comparable arbitrary criteria. Although everybody in a state is different the fact that they are located in an area surrounded by a border causes an automatic in-group favouritism and an automatic out-group discrimination. When people are aware of a group distinction, they already show in-group favouritism and out-group discrimination. The more aware people are of the existence of an in-group and an out-group, the more intergroup competition will appear and the more discriminatory people will react towards the out-group. This helps to understand why the national identity is strong during war. As mentioned in Chapter 1, during war regimes actively use media to emphasise national belonging in people. So, the existence of borders leads to the national identity and during war people are strongly aware of their in-group and out-group. Subsequently, the national identity is strong.

2. 2.

Self-categorization theory

The second theory that was developed to explain group behaviour and group identity was the self-categorization theory (SCT). According to the SCT, “social categorizations define[d] people’s place in society and through being internalised into the self, together with their emotional and value significance, provide[d] people with social identities” (Turner & Reynolds, 2011, p. 3). The SCT differentiates social identities on various levels. The first level is the interpersonal level, second the intergroup level and third the superordinate level (Hornsey, 2008; Turner & Reynolds, 2011). Each identity, on each level, is constructed through contrasting. On the interpersonal level the self is defined as an individual relative to other individuals, on an intergroup level the self is defined as member of a group relative to another group, and on a superordinate level the self is defined as being a human species relative to other lifeforms such as apes (Turner & Reynolds, 2011). This is important to point out because it acknowledges the existence of multiple identities in the individual and therefore the theory reaches further than only experimentally constructed intergroup identity distinction.

The SCT consists of several points. First, as a function of the SCT, comparison leads to salience. This means for example that when people identify on an intergroup level, the group differences will be pointed out. During comparisons, people are also more aware of their own group values and group norms. Because people within a group want to be perceived as a unity, the differences compared to other groups will be exaggerated (Ashforth & Mael, 1989). So, the differences between the different groups will be considered as more substantial than the differences between members of the same group. People will interpret their own group as ‘all the same’ and the

(14)

other group as ‘totally different from us’. As a consequence people will create stereotypes towards the outgroup (Turner & Reynolds, 2011). This means that members of groups will perceive members of other groups according to their group membership and not according to their individual characteristics (Tajfel & Turner, 2004). At the same time, individual group norms and values are being influenced by group norms and values, which slowly become internalised. Moreover, when group members initially have a different opinion than the most favoured opinion by the group, group members are likely to shift their opinion in the direction of the most favoured group opinion or make their opinion compatible with the most favoured group opinion (Hornsey, 2008), a process described as group polarization and depersonalization (Ashforth & Mael, 1989; Turner & Reynolds, 2011).

Second, Hornsey (2008) pointed out that stereotypes depend on the comparative context. So, different contexts will make different out-groups salient which in turn makes different in-groups salient. This contextual salience can be on different levels of the self-concept (Hornsey, 2008). Put differently, the context can make one aware of one’s identity on the interpersonal level, the intergroup level or the superordinate level. For example, if a mother visits the zoo with her husband and her child, she will at some moments be aware of the fact that she is a human compared to the animals, which is a superordinate comparison, that her husband her child and she, herself, belong to one family compared to other families visiting the zoo, which is an intergroup comparison, and at other moments aware of the fact that she is a mother compared to her child, which is an individual comparison.

Third, identification according to a group, and thereby also group-polarization, becomes more likely when this categorization is more available. Frequently categorization leads to more availability and as a consequence, some categories become chronical accessible even without a context that makes the in-group and out-group salient (Hornsey, 2008). If this occurs, the frequent categorization becomes internalised or imprinted into the self and forms a more permanent social identity.

Lastly, according to self-categorization scholars, the different levels of the self-concept are incompatible. This means that when a categorization is salient on the interpersonal level, there cannot be a categorization salient on the superordinate level and vice versa (Hornsey, 2008). This might not be relevant for understanding how the national identity originates, but it is important for understanding how the national identity and the cosmopolitan identity interact. I will come back to this in chapter 2.4.2..

The SCT contributes in two ways to the understanding of the national identity, especially during interstate war. First, during an interstate war, people will compare on a state level, therefore differences between states will be pointed out. This means that during interstate war that there will be a strong sense of ‘we’, our state, versus ‘them’, the other state(s), in which ‘we’ will be perceived as all the same, unity, and ‘them’ will be perceived as totally different from the way ‘we’ are. Second, during an interstate war, the intergroup distinction will frequently be accessed, becomes chronically

(15)

accessible and becomes part of the permanent identity. As said above, through awareness of intergroup distinction there will be an increased ‘we’ versus ‘them’ feeling. This increases cohesion within a state and strengthens the distinctions towards the other state(s) which makes an appeal on the national identity.

2. 3.

The effect of war on the relationship between the national identity and the

cosmopolitan identity during interstate war

Thus far, I explained how the national identity originate. In this part of the chapter, I will indicate how the SIT and the SCT separately help to understand the relationship between the national identity and the cosmopolitan identity. Through a combination of the SIT and the SCT, I will answer my second sub-question: does the national identity form a barrier to the development of a cosmopolitan identity?

2.3.1. The relationship between the national identity and the cosmopolitan identity

according to the SIT

According to the SIT, people have group identities, which can be created or of which people can become aware only by being allocated as a group, this is called the ‘minimal group paradigm’. As a consequence of the ‘minimal group paradigm’, people show in-group favouritism and out-group discrimination even if they are aware of the fact that their group is based on arbitrary or meaningless criteria. Theoretically, people know in which country they live and that they belong to the group human. Therefore, the social identity as a citizen of a certain state or being a human exists in everyone but the context determines if people categorise themselves as a citizen of a certain state or as a human.

The SIT also state that in-group favouritism and out-group discrimination occurred only because of awareness of group membership. During the interstate war, people will be made aware of their national identity, but to a lesser extent on their cosmopolitan identity. So, it seems logical that during an interstate war the national identity is strong and the cosmopolitan identity is weaker. To clarify this process, I made it visual in figure 1.

(16)

Based on the SIT, I cannot say anything about the relationship between the national identity and the cosmopolitan identity. The SIT does not focus on relationships between different identities and as a consequence, it does not contribute much to the knowledge about the interaction between different identities. However, as also visible in figure 1. the SIT explains that the presence of an identity depends on the situation. During an interstate war, people are made aware of their national identity so this identity is more present during a war.

2.3.2. The relationship between the national identity and the cosmopolitan identity

according to the SCT

Based on the SCT, I will explain on which level the political identities can be differentiated. As mentioned in chapter 2.3, social identities can be differentiated on different levels. The SCT differentiates on the interpersonal level, the intergroup level and on the superordinate level. In chapter 1 of this thesis, I explained that the regional identity, the national identity and the cosmopolitan identity are political identities. The regional identity and the national identity are social identities on a group level. However, the cosmopolitan identity includes all human and can, therefore, be seen as a social identity on the superordinate level. The SCT also says that different levels of the self-concept are incompatible which has implications for political identities as well. Based on the SCT a regional identity and a cosmopolitan identity cannot be present together and a national identity and a cosmopolitan identity can also not be present together because the cosmopolitan identity is on another level than the regional and national identity. As mentioned in the SIT, in a normal situation the national identity and the cosmopolitan identity are equally present. So combining these theories would mean that in a normal situation people are not aware of neither the regional/national identity nor the cosmopolitan identity. To clarify this, I made it visual in figure 2.

The last interesting point of the SCT is the assumption that, when a distinction is frequently made salient and accessible it becomes part of the permanent identity, which has implications for identity during an interstate war. As I already mentioned, during the interstate war people are frequently made aware of their national identity. So, according to the SCT, the national identity will become part of the identity in a way that it is constantly present, which I therefore call the permanent identity. This is also visible in figure 2.

(17)

2.3.3. Combining the SIT and the SCT

If I combine the theory to the situation during an interstate war, the national identity can indeed form a barrier to the development of the cosmopolitan identity. War stimulates the national identity, which is an identity on the intergroup level. But, war does not emphasise the cosmopolitan identity, which is an identity on the superordinate level. According to the SCT, if an identity on the intergroup level is present there cannot be an identity on the superordinate level present. So during an interstate war, the cosmopolitan identity cannot be present because the national identity is made present. This would provide an answer to my first sub-question, does the national identity form a barrier to the development of a cosmopolitan identity: yes, in situations in which the national identity is made present, the national identity can form a barrier to the development of a cosmopolitan identity.

Although the theory provides a plausible answer to my first sub-question, I also have some critics that I want to point out before drawing a conclusion. First of all, the SCT does not provide an in-depth reason why social identities on the different levels cannot be present at the same time. Second, research by Medrano and Gutiérrez (2001) showed that in a real-life situation a group identity and a superordinate identity can be present. In their research, Medrano and Gutiérrez (2001) explored the relationship between the regional identity, national identity and European identity among Spanish people. Their results showed a positive relationship between regional or national identity and European identity, which is also often perceived as a social identity on the superordinate level (Medrano & Gutiérrez, 2001). This research showed that the national identity and the cosmopolitan identity can be present together, which suggests that the use of the SIT and the SCT as presented thus far is incorrect.

(18)

As an alternative, I propose a more nuanced view which is still generally in line with the SIT and the SCT. Based on the SIT, social identities are present independent from each other so the national identity and the cosmopolitan identity could theoretically be present at the same time. Contrary to this view, the SCT states that identities that are on different levels cannot be present at the same time, so the national identity and the cosmopolitan identity could theoretically not be present at the same time. However, I argue, in line with the research of Medrano and Gutiérrez (2001), that the national identity and cosmopolitan identity can be present at the same time, as long as one of the two is not strongly present. This would mean that in a normal situation, both political identities are present and depending on the context one of the identities is strengthened. When one of the identities is frequently made salient and becomes part of the permanent identity, which happens to the national identity during an interstate war, then identities on another level, such as the cosmopolitan identity, cannot become strong. To clarify this, I made this visible in figure 3.

Based on my view, the answer on the first sub-question, does the national identity form a barrier to the development of a cosmopolitan identity, would be yes. In some situations, for example, during an interstate war, it seems plausible that the national identity can form a barrier to developing a strong cosmopolitan identity. However, even during an interstate war, the cosmopolitan identity can be present although not strong.

(19)

2. 4.

Conclusion

In this chapter, I explained how the SIT and the SCT are both useful to understand how the national identity is stimulated during an interstate war. Second, I explained the relationship between the national identity and the cosmopolitan identity based on the SIT and the SCT. Lastly, I combined the SIT and the SCT to form my own perspective on the relationship between the national identity and the cosmopolitan identity and concluded that a strong national identity could form a barrier to the development of a strong cosmopolitan identity.

I explained that barriers could prevent the cosmopolitan identity from becoming part of the permanent identity. However, there are questions unanswered, such as: what happens to identities after a war and how can the cosmopolitan identity be enforced? Also, according to the SIT and the SCT, the relationship between the national identity and the cosmopolitan identity is very strict. In practice, the relationship between the national identity and the cosmopolitan identity will be more fluid and there will be more of an interaction between the national identity and the cosmopolitan identity. In the next part, I will use the CH and the MIDM to give a psychological explanation for the previously found indicators of cosmopolitanism, which will answer my second sub-question. At the same time, using the CH and the MIDM, I will continue to explore the interaction between the national identity and the cosmopolitan and present a more fluid relationship between the national identity and the cosmopolitan identity.

(20)

Chapter 3. Explaining previously found indicators of the cosmopolitan identity

from a psychological perspective

In this chapter I will explain more about the interaction between the national identity and the cosmopolitan identity and I will come up with a theoretical framework that can explain the previously found indicators of cosmopolitanism. First, the CH and the MIDM help understand how intergroup boundaries increase and the cosmopolitan identity decreases. This will help to understand the interaction between the national identity and the cosmopolitan identity even more. Second, in chapter 1.2 I already indicated that education, travel, internet use, place of living, income and occupational status were positively related to cosmopolitanism although, theoretical explanation seemed incorrect or insufficient. Therefore, in this chapter, I will use the CH and the MIDM to explain why the previously found indicators are related to cosmopolitanism.

3.1.1. The contact hypothesis

The contact hypothesis (CH) was on the one hand created to explain why groups were hostile towards each other and on the other hand to explore how groups can work together and can be united into one group (as cited in Brown, 1993). According to the CH, group hostility and intergroup discrimination are caused by prejudices about members of other groups (Dovidio, Gaertner & Kawakami, 2003). In the original CH posed by Allport in 1954, he argued that: “Prejudice (unless deeply rooted in the character structure of the individual) may be reduced by equal status contact between majority and minority groups in the pursuit of common goals” (as quoted in Brown, 1993, p. 613). This means that if groups work together in a situation in which they are equal to each other, this will reduce prejudice. Subsequently, relationships become less hostile and groups can eventually unite. This suggests that if prejudices reduce, also the awareness of group identity reduces. In relation to political identities, this suggests that if nations work together as equals, prejudice and hostility would be reduced between people from different nations and so would the national identity which eventually could lead to a more superordinate identity, such as the cosmopolitan identity. The barrier of the national identity can slowly disappear and the cosmopolitan identity can strengthen.

To test the CH, Sherif, Harvey, White, Hood and Sherif (1961) started with an experiment in 1954, the Robbers Cave Experiment. In this experiment, a group of 22 schoolchildren were divided into two comparative groups on a summer camp. The schoolchildren did not know each other beforehand and initially, the groups did not have contact. The real experiment consisted of three stages. In the first stage, the two groups had to perform tasks. In this phase, there was only competitive contact, which meant that during every trial the groups were competing against each other and only one group could win. This successfully created feelings of anger, frustration, competitiveness, in-group solidarity and led to stereotypes towards the other group. In the next stage,

(21)

a situation was created in which the two groups made contact during non-competitive activities such as eating dinner. Despite the expectation, this setting did not lead to reduced friction between the two groups. In the last stage of the experiment, the two groups had to perform a task with a “superordinate goal which had compelling appeal value for both groups but which could not be achieved by the efforts and resources of one group alone” (Sherif et al., 1961, p. 132). In this phase, the groups showed “reciprocally cooperative and helpful intergroup actions, in friendly exchange of tools, in developing standard procedures for alternating responsibilities and in meeting problems [together]” (Sherif et al., 1961, p. 132). The last phase of the experiment showed a “decrease of unfavourable stereotypes towards the out-group. Favourable conceptions of the out-group developed, so that ratings of the in-group and out-group were no longer a set of contrasted polarities” (Sherif et al., 1961, p. 132).

This study was important because it confirmed the CH and because it proved that only simple contact between groups is not enough to reduce hostility. However, there are severe limitations of the Robbers Cave Experiment. First, the experiment only included two groups and therefore it is hard to determine if there are generalizable effects on other out-groups because of the effects of the creation of more favourable conceptions towards one out-group. Second, the participants in the experiment were schoolchildren which makes it hard to determine to what extent this phenomenon, reducing hostility because of contact with a mutual goal, would occur with adults in a real-life situation. Despite these limitations, the CH has been used as a basis and complementary for other theories, such as the ‘extended contact theory’ (Boomgaarden & Vliegenthart, 2009), and is still seen as an important and valuable theory.

3.1.2. Theoretical argument for the relationship between travel, internet use and the place

of living and cosmopolitanism

Now that I have described the CH, I will explain how the CH provides an explanation for the relationship between travel, internet usage and place of living and cosmopolitanism. First, I will explain the relationship between travel and cosmopolitanism. Second, I will explain the relationship between internet use and cosmopolitanism. Third, I will explain the relationship between place of living and cosmopolitanism. Lastly, I will hypothesise, based on previous studies, the direction of the relationship between these indicators and the cosmopolitan identity. However, before starting with explaining the relationship between travel and cosmopolitanism I do want to point out that I only provide a possible explanation. There might be more reasons or better reasons why there is a relationship between these indicators and cosmopolitanism. Further research could go more into the individual indicators.

According to the CH, it seems plausible that the relationship between travel and cosmopolitanism might be explained because travel to a foreign nation brings people in contact with

(22)

people from other cultures and nationalities. The CH states that simple contact is not enough but there has to be a common goal. This common goal could be found in the desire to learn to know the other culture. Citizens in the foreign nation interacting with the travelling individual might also have the goal of learning to know another culture. Another goal could be found in daily activities during travel. A simple activity such as asking the direction to the nearest tourist office already serves a mutual goal of helping to find the way. This active approach to exploring other cultures provides a theoretical explanation for the positive relationship between travel and the cosmopolitan identity. This contact with a mutual goal reduces intergroup boundaries as a result of reduced prejudices. At the same time, the contact with a mutual goal increases the cosmopolitan identity. To make this process more visual, I illustrated this in figure 4.

Following the theory, it sounds plausible that the relationship between internet use and cosmopolitanism might be explained because through the internet, one can interact with people from all over the world while being at home. Through the use of different languages and proverbs, one is made aware of the fact that other people, one is interacting with, do not belong to the same nation. Especially online gaming through the internet would fit the theory because in online gaming there often is a mutual goals, such as defeating a common enemy. However, interacting online with people from other nations, through for example online gaming, is not always the purpose of the use of internet. Therefore, more research could be done on the way internet is used and the relationship to cosmopolitanism. For now, I argue that the relationship between internet use and cosmopolitanism can, in some circumstances, be explained by the CH because, in some circumstances, online internet use can lead to decrease in prejudice which leads to decreased intergroup boundaries. At the same time, internet use can lead to a more positive attitude to the out-group, which increases cosmopolitanism. See figure 4. for a visual representation of this process.

Thirdly, the relationship between the place of living and cosmopolitanism might be explained by the fact that “people living in urban areas are generally more exposed to global others (e.g. foreigners, tourists) than people residing in the countryside” (Pichler, 2012, p. 25). The underlying reasoning could be the same as for the relationship between travel and cosmopolitanism, people in cities are more frequently asked to work together in urban areas compared to people in rural areas. This does not only have to be cooperation with tourists but can, for example, also be the cooperation in a work situation with expats. This reduces prejudices which decrease intergroup boundaries, and increases the cosmopolitan identity. For a visualisation of this process see figure 4. However, as mentioned before, there might be other explanations or theories that could provide a better explanation. For place of living this could, for example, be the ‘mere exposure effect’, which states that people create a positive attitude towards someone or something if they are exposed to it several times (Harmon-Jones & Allen, 2001).

(23)

As also mentioned in chapter 1 cosmopolitanism has been found to positively related to travel and use the internet (Phillips & Smith, 2008) and people living in urban areas appeared more cosmopolitan than people in rural areas (Pichler, 2012). Subsequently, my expectation is that there is a positive relationship between the indicators of cosmopolitanism, that can be explained by the CH, and the cosmopolitan identity. Therefore my first hypotheses are:

H1a: there is a positive relationship between the amount of travel and the cosmopolitan identity.

H1b: there is a positive relationship between the amount of internet usage and the cosmopolitan identity

H1c: there is a positive relationship between the extent to which one lives in an urban area and the cosmopolitan identity

The CH could provide an explanation for the relationship between travel, internet use and place of living and cosmopolitanism. However, occupational status, income and freedom of press, as being indicators of cosmopolitanism, are still not theoretically explained. Therefore, in the next part, I will first explain the mutual ingroup differentiation model. Next, I will explain how the mutual ingroup differentiation model provides a theoretical explanation for occupational status and income. Based on this theory, I will also add fear of war as a possible indicator of the cosmopolitan identity. After this part, I will focus on freedom of press, as an indicator of cosmopolitanism.

(24)

3. 2.

The mutual intergroup differentiation model

The mutual intergroup differentiation model (MIDM) could help to explain how occupational status, income and fear of war are related to cosmopolitanism. The MIDM, just like the CH, helps to explore how groups can work together and different groups can be united into one group. The model argues “that intergroup contact should be structured in such a way that group identities are not threatened” (Hornsey & Hogg, 2000, p. 243) which can be done in two ways. First, qualities of different groups should be pointed out instead of weaknesses. Groups should not focus on single qualities and compare each group on that quality, instead, each group has to gain credit for the qualities they possess most. Second, in line with the CH, group category, as well as superordinate category, should be salient during the contact to experience minimal in-group bias (Hornsey & Hogg, 2000). However, this model has hardly been tested. Also, the relationship between the individual and out-groups is unexplored. This leaves room for discussion because it is possible that a threat to the individual or a threat to identity on interpersonal as well as on intergroup level could lead to increased bias and a stronger group identity, which in turn could be a barrier to the cosmopolitan identity. I argue, in line with evolutionary psychology, that a threat on a personal level as well as a threat on group level results in self-protection, in order to secure the survival (Neuberg, Kenrick & Schaller, 2011). Following this reasoning, every threat to the individual could result in a protectionist response in which the individual will first secure its own individual- or its own group survival over the survival of other individuals or other groups as long as the individual perceives this threat as a threat. This protectionist response means that intergroup boundaries are strengthened. However, this effect only occurs when the individual compares itself on interpersonal or intergroup level as a result of the perceived threat. If one compares with individuals or groups that experience the same threat, there is no need to strengthen the intergroup boundary as a protective mechanism towards other groups. Instead they will look for opportunities to work together for survival. This reasoning suggests that there has to be some insurance for survival in order for intergroup boundaries to decrease and to create a superordinate identity.

3.2.1. Theoretical argument for the relationship between occupational status, income and

fear of war and cosmopolitanism

Now that I have described the MIDM, I will explain how this theory can provide an explanation for the relationship between occupational status and income and cosmopolitanism. Based on previous research and my own argument, I will hypothesise the direction of the relationship between occupational status and income and the cosmopolitan identity. After that, I will use the same argument to theorise how there could also be a relationship between fear of war and the cosmopolitan identity. Lastly, I will hypothesise based on previous studies and my own argument, the direction of the relationship between fear of war and the cosmopolitan identity.

(25)

As mentioned in the previous part, I want to point out that I only provide a possible explanation. There might be more reasons or better reasons why there is a relationship between these indicators and cosmopolitanism. Further research could go more into the individual indicators.

The relationship between occupational status and income and cosmopolitanism can be explained because not having an occupational or earning less than others can be perceived as a threat to the individual. Being unemployed or earning less give people disadvantage for survival compared to other people. Because of this comparison, people will react protectionist in order to survive. Subsequently, intergroup boundaries will be strengthened and the cosmopolitan identity will weaken. To make this more visual I illustrated this in figure 5. Income has also been found to be related to cosmopolitanism. Pichler (2012) found that earning less compared to others was negatively related to cosmopolitanism. This is in line with Teney, Lacewell and de Wilde (2014) who stated that economic scarcity is perceived as a threat to the identity and suggested that there has to be some kind of economic insurance for survival to create a cosmopolitan identity. Economic scarcity in the form of income or a bad economic position because of unemployment are perceived as threats which lead to strengthened intergroup boundaries.

Based on my own argumentation and research by Pichler (2012) I argue that there is a negative relationship between perceived threat to the individual and the amount of cosmopolitan identity, if the perceived threat makes people compare themselves towards other individuals or members of other groups. Therefore, I hypothesise that:

H2a: there is a negative relationship between unemployment status and the cosmopolitan identity. Being unemployed leads to less cosmopolitanism than being employed.

H2b: there is a negative relationship between economic scarcity in the form of income and the cosmopolitan identity. Earning less than others leads to less cosmopolitanism.

Lastly, there could also be a relationship between fear of war and the cosmopolitan identity. The MIDM argues that there would be a protectionist reaction after group comparison. Occupational status and income are indicators on the individual level, however, in this thesis, the main focus is war. Therefore, fear of war could be an indicator on group level related to the cosmopolitan identity. Following the reasoning presented in this chapter, interstate war can be perceived as a threat to the group and fear of war subsequently activates protectionist mechanisms to secure survival. As a consequence, intergroup boundaries are strengthened in order to work together to survive. When talking about fear of war, the strengthened intergroup boundaries would logically be related to the national identity. As already explained in chapter 2.1., if the national identity is strengthened, this could form a barrier to the cosmopolitan identity. This process is illustrated in figure 5. So this would suggest that increased fear of war would lead to decreased cosmopolitan identity. Therefore, I hypothesise that:

(26)

H2c: there is a negative relationship between fear of war and the cosmopolitan identity.

So, the MIDM could provide an explanation for the relationship between occupational status and cosmopolitanism and income and cosmopolitanism. Based on the MIDM I also argue that there would be a relationship between fear of war and the cosmopolitan identity. However, freedom of press, as an indicator for cosmopolitanism, is still not theorised. Therefore, in the next part, I will explain how theoretically freedom of press can be related to cosmopolitanism.

3.2.2. Theoretical argument for the relationship between freedom of press and

cosmopolitanism

The relationship between freedom of press and cosmopolitanism is not extensively explored but I argue that in countries with restricted freedom of press, national group boundaries are maintained which function, just like war, as a barrier to the development of a cosmopolitan identity. This can be explained by the way in which governments use the media. Medrano and Gutiérrez (2001) and Skrbis, Kendall and Woodward (2004) argued that the way media representation can influence the way in which other nations are perceived. In their articles, they argued that broadcasting positive messages about the European Union lead to increased European identity. Therefore, portraying the own country as superior and positive lead to a strengthened national identity.

This seems also supported by the research of Pichler (2012). In his research, Pichler (2012) found more cosmopolitanism in ‘failed states’ such as Mali, compared to successful states, such as European states. Based on these results Pichler (2012) argued that ‘failed states’ have more cosmopolitanism because the state identity is less strong than in ‘non-failed states’. So, this could

How to increase cosmopolitanism on an individual level : the effect of interstate war on the development of a cosmopolitan identity