LINGUISTIC DIVERSITY OR LINGUISTIC IMPERIALISM?

(1)

LINGUISTIC DIVERSITY OR

LINGUISTIC IMPERIALISM?

Language policy and the Eurovision Song Contest

ANNE-MIEKE EVELINE VAN HELDEN, 1988808 Prof. Dr. G.T. (Goffe) Jensma

Dr. H. (Hanneke) Loerts April 9, 2018 Word Count: 18737

MA Thesis

Departments of Applied Linguistics and Frisian Language and Culture Faculty of Arts

(2)

1

0. Abstract

The use of English has increased over the years and its effects on multilingualism and linguistic diversity are under debate. In this study, the Eurovision Song Contest was used as a platform to look at the European relationship with language and particularly English. Through analysis of the contest’s language policies and history, as well as by looking into language choice practices and the effect of language on scores, the research answers the question whether the ESC has been an expression of linguistic diversity policy or of linguistic imperialism. The EBU’s diversity policy was created through the national language rule, which obligated all countries to sing in a national language. However, this rule was seen as unfair and created a hierarchy between languages, as well as an increased wish from the participating countries to use English. The rule was cancelled in 1999 and since then, the English language has become dominant in the ESC. The participants attitudes towards- and beliefs surrounding English changed the language policy and the linguistic landscape of the ESC and one can question whether the position of English in the contest today is neutral, or that linguistic imperialism is at play.

(4)

3

1. Introduction

The European continent is a diverse one, home to many different languages and cultures. Since the Second World War, European countries have been striving for more unity, for instance in the political arena and in trade. Arguably the most visible of these efforts for unity is the creation of the European Union. Over the years, the EU has grown: More countries have become member states, making Europe come together in a political and economic union. Additionally, thanks to technological developments and the rise of the internet, it is easier to connect across borders and therefore Europeans have many new ways and reasons to communicate with others on the continent. In this communication, the role of English draws attention. The use of this language has surged over the years. In international communication, English often functions as a lingua franca.

The position of the English language and the effects this has on multilingualism, are topics of debate. Some scholars argue that the role of English is too great and that its widespread use has a negative effect on multilingualism and linguistic diversity. Phillipson (2003; 2008a) fears that the use of English globally and in Europe is a threat to multilingualism. He also describes a phenomenon he calls ‘linguistic imperialism’, where a hierarchy is created between languages and between the speakers of different languages (Phillipson 1997, 238ff).

The dominance of English is also visible in communication amongst the member states of the European Union and other European organisations. This creates an apparent contrast between the organisations’ own language choices and the language policies they create, considering the latter are heavily focused on linguistic diversity. These organisations created various initiatives for the

promotion of linguistic and cultural diversity. The European Union itself aims to protect

multilingualism and linguistic diversity (“Linguistic Diversity,” 2017). The Council of Europe promotes linguistic diversity as well (“Education and Languages,” 2014) and the European commission aims to promote cultural diversity (“European Capitals of Culture,” 2017).

One initiative in the cultural area that unites many countries across the European continent every year, is the Eurovision Song Contest. The contest was created to help unite Europe after the War and its first edition took place in 1956 (Motschenbacher 2016, 13). The ESC comes from similar ideological origins as the European Union and has become a true phenomenon over the past sixty years, with the 2017 edition drawing a total of 182 million viewers worldwide (“Eurovision Song Contest 2017,” 2017).

The Eurovision Song Contest is thus a large-scale event, which brings many European

countries and speakers of many different languages together. It might therefore be relevant to take a look at the relationship between language and the Eurovision Song Contest. The ESC is a good

(5)

4

platform for research, considering it is a large arena for European communication. The contest’s organisation has a language policy as well and also has to deal with the predominant position of English. To the average TV-viewer, the recent editions give the impression that English is dominant and that there is little diversity.

By looking at the ESC, one could thus look at the diversity in language choice in Europe, at linguistic diversity policy and the role of English. This leads to the research question for this research: Can the Eurovision Song Contest through the years be interpreted as an expression of European diversity policy or as an expression of linguistic imperialism?

This question will be answered in two ways: The ESC’s language policy will be analysed first and secondly the event’s practical situation will be investigated. Through a literature review a theoretical framework will be created, describing what linguistic diversity policy and linguistic imperialism are, looking at both terms with a critical eye. Then the contest’s language policy and ESC’s relationship with language over the years will be analysed in a qualitative manner, through literature, by comparing it to the European Union and measuring it according to the definitions of linguistic diversity policy and linguistic imperialism. The consequences of this policy will be discussed, as well as the reasons behind it.

For further information on the effect of the language policy on the practical situation, the research also looks into the data that is available on the songs that have participated in the contest. Looking at the language choices countries made in the contest and the role of language in the points songs receive, the research can provide insight into the languages attitudes in the ESC, particularly towards English. This gives information about the possible tension between the presence of English and linguistic diversity policies in the event. In the discussion the results for both parts will then be compared and analysed and conclusions will be drawn in the final chapter.

(6)

5

2. Background

In order to be able to analyse the role of language in the Eurovision Song Contest, as well as the organisation’s language policy, definitions and additional information on linguistic diversity policy and linguistic imperialism are necessary. Findings of previous research on the role of language in the voting process of the Eurovision Song Contest are also included in this section.

2.1 Linguistic diversity policy in Europe

In a linguistically diverse environment, many different languages are present. However, linguistic diversity has been under pressure for years and many languages have disappeared over time. Today, languages are still in danger of disappearing (Brenzinger 2007, ix). Several languages in Europe, too, are close to disappearing. This endangerment is due to the fact that, until the most recent decades, language policy in Europe was aimed at homogeneity and not at the promotion of minority

languages. The objective was to marginalise minority languages (Salminen 2007, 205ff).

There are two sides to linguistic diversity: On the one hand, diversity is seen as something positive, to be promoted and even celebrated. On the other hand, many things that go wrong in society are easily attributed to linguistic and cultural diversity, particularly when immigrants are involved (Piller 2016, 2f). While Europe is becoming increasingly politically unified, the people within this union have to deal with increased cultural and linguistic diversification. This causes a fear of losing one’s language and even one’s identity. This anxiety is part of the public debate and because of this, there is increased attention for language policy (Studer & Werlen 2012, 1).

To deal with matters of languages and their disappearance, policies have to be made. For Wright (1995) language policy is not only something that is done by governments; different groups in society are influential in deciding which (variety of a) language people speak. Language policy is related to all areas of politics and society (Wright 1995, 148). Languages should be maintained, because they are a way to keep contact with history and with cultural heritage, for instance. Language is also an instrumental part of one’s identity and therefore worthy of protection (Wright 2016, 273). According to Wright (1995), language policy thus affects everyone and therefore there is a joint responsibility for language policy. This policy should be aimed at diversity. This change in the attitude towards language policy and in language policy itself is only of the last few decades.

Changing situations and changing opinions towards endangered languages ask for different language policy than has been in place for decades. It appears to be a struggle how to best deal with the changes and new expectations of language policy (Studer & Werlen 2012, 1f).

The European Union has to deal with these new demands of language policy as well. The European Union aims to be ‘united in diversity’ (“Linguistic Diversity” 2017). Article 22 of the

(7)

6

European Charter of Fundamental Rights and Article 3 of the Treaty on the European Union warrant linguistic diversity. To promote linguistic diversity, the EU focuses mostly on the learning of new languages by funding and promoting various ways of learning languages. The European Day of Languages is a day of events co-organised by the EU to promote language learning and linguistic diversity (“Linguistic Diversity” 2017).

Promoting linguistic diversity is part of the EU’s policy on multilingualism. Studer & Werlen (2012) claim this is difficult for the EU, because the diversity of languages in the Union is based on the diversity of nation states. Those nation states have all striven -and many still do- for linguistic homogeneity within their own countries. All of their national languages are official working

languages of the European Union. In practice, however, this led to English becoming the non-official but real working language of European communication (Studer & Werlen 2012, 1f). According to House (2003, 561) the multilingualism policy of the European Union contradicts itself and cannot work. Motschenbacher (2013) agrees and adds that the multilingualism policies by the EU are top-down, while the want for English and its increased dominance is bottom-up. Top-down

multilingualism will therefore never be successful (Motschenbacher 2013, 7). Other criticism of the EU’s diversity policy comes from De Witte (2008), who discusses the legal protection of linguistic diversity, as offered by the enshrinement in Article 22. The Article does protect minorities, but in a very basic and minimal form. It is mostly an Article that withholds the EU from interfering with minorities, but it does nothing to actively help them.

In summary, a changing world has created different attitudes towards linguistic diversity and language policy. In Europe, several organisations, like the European Union, strive to protect and promote linguistic diversity. These organisations are held back in their endeavours by the fact that their policies need to be accepted and agreed to by all of their member states. This dilutes the organisations’ efforts and weakens the actual policies. On the other hand, it is to be complimented that the EU, among others, makes an effort to promote linguistic diversity.

2.2 Linguistic imperialism

Robert Phillipson is one of the academics who has written the most on the topic of linguistic imperialism. Particular focus in research on linguistic imperialism is on the role of the English language in international communication. In his research, Phillipson uses the term linguistic imperialism to describe how some languages have a higher status and are used more than others. The term describes how a hierarchy can be created between languages and the consequences this has. Various underlying structures and organisations are responsible for establishing and maintaining such a hierarchy (Phillipson 1997, 238f). About this hierarchisation, Motschenbacher (2013), while writing on the status of English as a lingua franca, states that multilingualism in Europe is organised

(8)

7

after national attitudes. These attitudes decide the hierarchisation, with national languages on top and other languages having a lower status. As the main working-languages of the European Union, English and French have the highest status in Europe (Motschenbacher 2013, 5). According to Reagan (2009, 76), some scholars link the widespread use of English to the imperialism agenda of the British and more recently the Americans. They do not see it as a natural, neutral occurrence.

Phillipson offers a definition of linguistic imperialism for English, which is as follows: “The dominance of English is asserted and maintained by the establishment and continuous reconstitution of structural and cultural inequalities between English and other languages” (Phillipson 1992, 47). According to Phillipson (2008), stating that English is used as a lingua franca is not correct. Saying it is a lingua franca would suggest English is merely used as a neutral instrument for communication between speakers of different languages. In Phillipson’s view, this fails to show the other uses and implications of the English language. To him, it is a colonial language; the language of dominating forces and/or the elite, for instance. The fear is, that this increased use of English in different functions leads to domain loss for native languages (Phillipson 2008, 250ff).

In his view, global English is a project, set up by the establishment, for many different reasons, including economic profit (Phillipson 2008, 260). In the case of the Eurovision Song Contest, this means that there would be more to the use of English than merely communication. It could also mean English is used for profit: Being successful in the ESC and having commercial success outside of the contest.

For Phillipson, linguistic imperialism is a subtype of linguicism (Phillipson 1997, 238f). This term he attributes to Tove Skutnabb-Kangas, who describes it as “ideologies and structures which are used to legitimate, effectuate and reproduce an unequal division of power and resources (both material and non-material) between groups which are defined on the basis of language (on the basis of their mother tongues)” (Skutnabb-Kangas 1988, 13). When linguicism is at play, some languages are stigmatised as less valuable, while others are considered a “language of progress and modernity”, like English (Phillipson 1997, 239). The fact that these beliefs continue to exist and are even being promoted, be it publicly or in secret, consciously or subconsciously, is part of linguicism (Phillipson 1997, 240). Linguistic imperialism is thus not necessarily a clear policy or action, but it can also be a subconscious and subtle effort. This makes analysing policies more difficult, for beliefs are not always visible on the surface. It is an underlying pattern that needs to be discovered.

To Motschenbacher (2013), Phillipson and Skutnabb-Kangas link the role of English to human rights, particularly the right to speak your mother tongue. However, in inter-European

communication it is much more difficult to apply linguistic human rights, considering this can seriously hamper the communication. In addition to this, only wanting to speak one’s own language could come across as an expression of nationalism (Motschenbacher 2013, 6).

(9)

8

Phillipson and Skutnabb-Kangas’ views, particularly of English being a threat to multilingualism, are thus disputed by Motschenbacher (2013) and also by House (2003), who distinguishes between “languages for identification” and “languages for communication” (House 2003, 556). A language for communication is merely a tool for a speaker to use to facilitate

communicating in a multilingual situation (House 2003, 559). In her view, native languages are much more influential in creating one’s identity than a lingua franca is. English is a true lingua franca, so English and native languages are completely different. There is no “collective cultural capital” that comes with English, so it cannot be a part of forming an identity (House 2003, 560). The imperialism Phillipson attributes to English as a lingua franca no longer rings true, according to House. In former colonies in Africa English is now purely a tool for communication (House 2003, 560). Motschenbacher goes on to add that people choose to learn English because it gives them access to better jobs or other international success and not because of imperialism from English-speaking countries (Motschenbacher 2013, 7). He refers to Yano (2009, 253), who claims that speakers of English as a lingua franca in new, hybrid and multilingual societies do not feel they speak English because of English imperialism, but for other reasons.

However, not all speakers of English as a lingua franca live in the type of societies Yano (2009) describes. Imperialism is not necessarily like colonial imperialism anymore, with a visibly occupying foreign power. As Phillipson (1997) writes, it is represented by the values that are

attached to a language. In the examples given by Motschenbacher (2013) and House (2013), English has a higher status than other languages and is associated with positivity, modernity and democracy. It is also used for profit. These things fit in with Phillipson’s explanations of linguistic imperialism and linguicism.

2.3 Voting and language in the Eurovision Song Contest

Research has been done on voting and language in the Eurovision Song Contest. Spierdijk & Vellekoop (2009) looked into voting bias in the Eurovision Song Contest and various factors that could possibly influence voting. They discovered that, on average, songs in a related language are preferred in voting. However, this is not the case for many other countries. Every situation is different. For most Nordic countries, but not for Iceland, language does play a part in the voting (Spierdijk en Vellekoop 2009, 423f).

Ginsburgh & Noury (2008) investigated whether there is political voting in the ESC and found that the quality and the cultural and linguistic proximity of the song are larger factors for success in the contest than political reasons.

Kirkegaard (2013) looked at the position of the Nordic countries in the ESC and states that there is indeed a certain level of unity between the Nordic countries and they do appear to vote for each

(10)

9

other often. There are, however, also many situations where this was not the case and many different reason are possibly of influence in the Nordic voting. She sees increased cooperation between the Nordic countries on political levels as a reason for buddy voting.

Charron (2013) found that friend-voting is clearly visible in the Eurovision Song Contest, but also argues that states with weaker democratic systems and less political impartiality are more likely to vote for friends. Nations with politically impartial institutions vote for a greater variety of countries and are thus less predictable in their voting behaviour.

2.4 Statement of purpose

For the research that follows, the aforementioned literature creates a framework with which to analyse the role of language in the Eurovision Song Contest. How to best investigate the possible presence of linguistic imperialism, is formulated by Phillipson:

“For example if an aid project provides funds for language X, and not for language Y, when both X and Y are central to the linguistic ecology of a given country, there may be linguistic imperialism at play, especially if X is associated with the donor country, is the former colonial language, and is being used as a medium of education (for instance French in Senegal or English in Nigeria). Empirical validation of a hypothesis that linguistic

imperialism is in operation requires study of the nature of the local linguistic ecology and linguistic hierarchies, the purposes which language X serves, whose interests are promoted by the specific language policy, and the likely outcomes of any proposed activities.”

(Phillipson 1997, 239f)

This is what will be done in the qualitative part of this research, by looking at the linguistic ecology and linguistic hierarchies in the songs for the Eurovision Song Contest. The linguistic ecology will further be analysed by comparing it to situations in the EU.

It is expected that the language policy for the Eurovision Song Contest has been focused on linguistic diversity for many years, but that it suffers from the same troubles that the European Union has to deal with. It is also expected that the role of English has increased and that a form of linguistic imperialism is present.

For part two of the research, this study will see if the numbers reflect any existing linguistic hierarchies or hegemonic beliefs, represented in language choice for a song and a possible effect of language factors like language choice on the points a song is awarded. In previous research, language has been investigated as one of many factors in voting in the ESC. In this research, the role of

language in voting, will be looked into further, by looking at the presence of English first and secondly comparing the mutual voting behaviour for two language groups. For this data, the hypothesis is that

(11)

10

the use of English for all countries has increased significantly over the years, in the rule-free periods. For the two language groups that will be studied further, Scandinavian and Romance, the

expectation is that the use of English has increased significantly as well. Estimated is that the

countries in the smaller, more closely related language group award each other more points than the larger, more distant languages do, considering previous research has shown the relationship

between languages can play a part in voting. Language choice for a song is expected to have a significant effect on the awarded points and English songs are thought to score better.

(12)

11

3. Language, language policy and linguistic diversity in the

Eurovision Song Contest

The first edition of the Eurovision Song Contest took place in Lugano, Switzerland, on the 24th_{of May}

1956. Seven countries participated, all of whom performed two songs. Switzerland not only hosted this first edition, but won as well, with Lys Assia’s Refrain chosen as favourite by the juries from the other countries. The first event came two years after the European Broadcasting Union, the

organisation of European public broadcasters, was given the assignment to develop new

programmes, organised in cooperation and for viewers of all the national broadcasters. Modelled after the Italian San Remo Festival, the Eurovision Song Contest was founded. The original aim of the contest was to stimulate the production and quality of original, popular music by creating

international competition between composers and lyricists (O’Connor 2005, 8). The contest was also developed to unite Europe after the War, through music and the then new medium of television (Motschenbacher 2016, 13). This desire to unite the continent was also a reason for the founding of the European Union, which came into being around the same time. Among the seven participants of the first edition were all the founders of the European Coal and Steel Community (Raykoff 2007, 1).

During this first contest all participants performed two songs. Belgium, France and Luxembourg performed both of their songs in French, while Switzerland chose German for one of their songs and French for the other. Countries with French as one of their national languages chose to sing in French often in the years that followed. It was a dominant language in the first years of the Contest and was, from the start, one of the two official languages of the presentation of the Contest. (Motschenbacher 2016, 118). With the years, French has become less dominant and French today no longer has the position it had in the early years of the ESC. According to Motschenbacher (2016, 118), the position of French in the ESC has followed the same path as the position of French as a working language in the European Union: English has eclipsed and replaced it.

There are various possible reasons for the fact that many countries chose to sing in French. With a large part of the participants having French as one of their national languages, it makes it easier to communicate a message by singing in a language that the juries in the other countries understand as well. French also had a stronger position as a language in the 1950’s and 1960’s than it does today. In addition to this, the music that was presented in the contest could be a part of this: There was a relationship between the genre of music in the contest and the language they were performed in. The songs that were sung in these first years were often ballads and chanson-like songs. The chanson is a French form of music. French chansons and German schlagers were popular during the first years of the contest (Bohlman 2013, 46).

(13)

12

There were only a few rules for the first contest. All songs had to -as they still have to- be original songs and be sung live, but there were no specific rules for language during this first contest and thus no real language policy; all participants were allowed to sing in whatever language they chose (O’Connor 2005, 9). In the years that followed, rules were created when the situation asked for it and particularly when EBU-members complained. For example, the minimum age of participants was set at sixteen in 1990, following complaints that contestants aged twelve had participated the year before (O’Connor 2005, 117ff). The dominance of French in song choice in the early contests did not lead to the creation of a diversity policy. It was not until Sweden sent in a song in English in 1965, being the first non-English speaking country to use this language, that rules about language choice were changed and that a language policy was created. Sweden was the first and only country to use the lack of language rules in this first decade of the ESC to sing in English. O’Connor (2005, 28) states that the national language rule was implemented in 1966 because of Sweden’s choice for English the year before. This led to complaints from the other participants and therefore the EBU decided to create the national language rule (Motschenbacher 2016, 127).

The EBU, as organising company, is in charge of the rules. This is a complicated task for the EBU, because it is not a political organisation. It is an organisation made up of national broadcasters, not of national governments, but was, on the other hand, powered by ideological principles (Raykoff 2007, 3). Songs with obvious political or politically offensive messages are not allowed, but the execution of this policy is rather ambiguous, as not all songs with visible political messages are excluded from the competition (Motschenbacher 2016, 14f). Political messages in songs have caused a big debate during the entire history of the Eurovision Song Contest. Anti-Russian songs have often been part of this debate. An Anti-Eastern European sentiment is present in the perception of the contest (Georgiou 2008, 151). One extreme example is the 2016 winner from Ukraine. Jamala’s song

1944 told of the deportation of the Crimean Tatar minority by Stalin and was subsequently

complained about by Russia. The song was allowed to compete, however, and even took home the victory. The Ukrainian entry in 2005, a rap-version of the battle song of the Orange Revolution, is another controversial entry that was allowed, albeit after mild revision (Raykoff 2007, 4f).

Other political messages have been present throughout history, too. In 1998, Israel won the contest with Dana International’s song Diva. Dana International was a transgender singer (O’Connor 2005, 153). This was a clear example of the support for the LGBTQ+-community that has since then been very present in the Eurovision Song Contest (Tobin 2007, 25); the win of the ‘bearded lady’ Conchita Wurst for Austria in 2014 being a more recent example.

The EBU is thus, perhaps unwantedly, involved in political matters, which happens because the ESC’s participants are all from different nations. The EBU’s inconsistency in the execution of its policy often leads to discussion. Another issue with the organisation’s policy, is that the Broadcasting

(14)

13

Union is not very transparent about the reasons behind its language rules. A history of the rules, or explanation of the language rules that are currently in function is not provided; the website of the Eurovision Song Contest only states the rules that are valid for the current contest. Decisions on rule changes on other subjects can occasionally be found, but official reasons for changes of rules cannot. O’Connor (2005) and others offer their interpretation of- and assumptions about the changes and the reasons behind them, and often complaints from members appear to lead to change, but no official clarity is provided.

1966-1972

The first rules about language came into action during the contest of 1966 and were valid through the contest of 1972. During this time in the contest, all participants’ songs had to be performed in one of their nation’s official languages (O’Connor 2005, 28). A minority language with the status of national language was thus also allowed.

This was the beginning of an official language policy by the EBU. The newly initiated language policy did not appear to create a visible effect immediately. Practice had shown in the years before that all countries already sang in one of their national languages, so very little changed. French was still very prevalent as language of choice for countries that have French as one of their national languages.

One could argue that this new language rule and effectively the creation of a language policy was an effort to protect the linguistic diversity of the contest, even to prevent English from taking over. However, with only one entry having ever been performed in English, English taking over does not seem to have been a likely threat at this time. The change in rules appears to show a preference of the EBU for French over English in these early years. This may have to do with the popular music at the time, the status of French, and the fact that French simply was a national language for many participating countries.

The fact that language choice was free before 1966, created the impression to the EBU members that national boundaries were not that important and that the event was not very nation-centred (Kirkegaard 2013, 81). The EBU’s national language rule was aimed at the maintaining of these national structures (Motschenbacher 2016, 127). Although the official reasons behind the change in language policy are unclear, scholars say that the obligation to sing in a national language was created to protect national identity (Kirkegaard 2013, 81).

It would appear that the EBU has a more linguistically inclusive policy than the European Union, because minority languages or other languages that do not have an official status in the European Union, have been used in songs in the ESC (Motschenbacher 2016, 121). It is more difficult for these languages to participate, however. This has to do with the construction of the Eurovision

(15)

14

Song Contest. The EBU members, the national broadcasting companies, send a representative for a whole country and they ultimately decide what criteria an entry has to meet. For the United

Kingdom, for instance, the BBC is the representing member of the EBU, so they alone are allowed to send an artist. The broadcasters in Wales and Scotland are members of the EBU, but because the BBC has the first right to represent the United Kingdom, including Scotland and Wales, the Scottish cannot participate in the contest as a separate country as of yet (“Scotland” 2008).

The national language rules have always been subject to debate and frustration, particularly for smaller languages that claimed that ‘bigger’ languages had a great advantage. These larger languages had more success, in the eyes of speakers of the marginal languages, because it was easier for people in other countries to understand their lyrics. The larger languages, like English and French, were also languages used in popular music and fashion, so songs in these languages would come across as more modern. They were seen as commercially more viable. To deal with this, it was usual for songs during the first decades of the contest to be translated into various other languages after the contest was over (Kirkegaard 2013, 81). Lenny Kuhr, one of the Dutch winners, did so with her song ‘De Troubadour’ from 1969. Seven different versions of the song were recorded (O’Connor 2005, 39).

The great results achieved by the English-language songs of the United Kingdom and Ireland may have contributed to the sentiment that larger languages did well in this period. The United Kingdom won the Contest in 1967 and was one of the four winners in the edition of 1969. The UK came second in 1968, 1969, 1970 and 1972 and Ireland won in 1970 and came second in 1967. In addition to this, France, Luxembourg and Monaco all won as well, while performing in French.

There were several ways for smaller languages to deal with this perceived inequality. Code-switching, for instance: While the lyrics to a song officially had to be in a national language, several entries broke this rule by singing partially in English or French, alongside their national language. The EBU allowed these songs to participate, apparently agreeing to songs as long as their lyrics partially contained the national language (Motschenbacher 2016, 122f). Spain won the contest in 1968, singing a chorus which only included the words ‘la la la’ – which was also the title of the song. The trend for making up words or using ‘la la la’ has continued throughout the history of the ESC as a way of avoiding language barriers (Raykoff 2007, 2). In a way, a new lingua franca was created through these nonsensical syllabic lyrics, a musical lingua franca as it were. The use of this musical lingua franca and the translations of songs after the contest were a way for countries with marginal languages to feel like they could do better in the contest and do better commercially after the contest.

In a similar fashion to the European Union, all national languages were official working languages in this era of the contest. The EU suffers from the difficulty of being a union consisting of

(16)

15

nation states and this is the same for the ESC. While it is not intended as a political event, the songs represent their country. Georgiou (2008, 143) notes that the ESC does not revolve around the singers, but around the countries these singers represent. Countries participate in the ESC and the attention goes to them as nations: Their flags are shown and points are awarded per country. This frames the contest as a nation-centred event. It also gives way to stereotypes in the performances and in the reaction of viewers to the various songs. Georgiou adds, that “for some, the song contest becomes almost synonymous to European Union politics” (Georgiou 2008, 145). The added difficulty for the ESC, is that, although the event appears to be nation-centred, the nations are not the ones making the decisions; the broadcasting companies in those nations are. Painting it as an event of nations, creates a risk of politicising the contest.

The language policy of this period appears to be aimed at the protection of the diversity of the linguistic ecology and the national structures of the ESC, but it had as a consequence that a hierarchy was created between the different languages. At least, the speakers from the marginal languages felt this was the case. If we, as mentioned before, look at the success of French and English songs, this feeling is not difficult to understand.

1973-1977

Seven years after implementing the national language rule, the EBU decided to change its language policy again; countries were now allowed to sing in whatever language they chose.

During the first language rule-free contest in 1973, only Finland and Sweden decided to sing in English and Norway chose to sing in various different languages. In the editions that followed, the option to sing in English became more popular (O’Connor 2005, 53). All the winning songs in this period were either sung in French or in English. No explanation was provided for this rule change of 1973, either. It would appear, however, that the EBU again gave in to complaints, now coming from countries with minor languages, about the perceived advantages the larger languages had. The rules changed regularly, which come across as though the EBU was looking for a balance between a language policy aimed at diversity and the maintaining of national structures on the one hand and the want for commercial success expressed by the participating countries, on the other.

The use of English became more popular throughout this second language rule-free period, than it had been during the first years of the contest. Particularly the Nordic countries were prone to using English instead of their national language: Finland and Sweden only sang in English during this period. For code-switching, English was now used more often than French (Motschenbacher 2016, 127). Examples of the musical lingua franca are also visible in this period: The Netherlands won with a song called ‘Ding-a-Dong’ and many lyrics contained verses of ‘la la la’ (O’Connor 2005).

(17)

16

The winner of the 1974 edition of the contest is the most famous Eurovision-winner of all time: ABBA, for Sweden. Their song was performed in Swedish during the Swedish national selection, but it was translated to English for the international contest (O’Connor 2005, 58). This also happened with the Dutch entry ‘Ding-a-Dong’, which won the year after. This option has been very popular in many countries over the years. In more recent years, for instance Albania often selected its song in Albanian first, but then translated it before the international event. Since the rule change in 1999, all Danish songs have been performed in English. According to the official DR (the Danish broadcaster) rules, however, the language of the winning song at its ESC-performance is optional and is to be decided together with the broadcaster (Plank 2017).

These translations once more illustrate the importance of the national broadcasting

company. In the end, the power to decide lies with them. If the broadcaster does not agree with the language a song is performed in, it can decide to do something about this. It is also up to the

broadcaster to choose the manner in which this representative is selected. There are many options for this; some countries have a national song contest of their own and allow the public at home to choose, other broadcasters select an artist and/or song internally (Charron 2013, 485). In Sweden, the popular annual selection Melodifestivalen is an enormous event, spread out over several weeks. The Netherlands has had this type of selection in the past, but during the most recent years,

broadcaster AVROTROS has picked an artist internally. The artist has then been allowed to choose a song without input from the Dutch public. This happens in several countries.

The rules of the Contest were changed again in 1977, but not in time for the selection of the German and Belgian songs. Both countries were allowed to choose their song according to the previous rules of free choice and chose to sing in English (O’Connor 2005, 69).

1978-1998

A language policy change in 1977 meant that, once again, all songs at the Eurovision Song Contest had to be performed in a national language from the contest of 1978 on. In addition to the

arguments provided for previous language rule-changes, one could give as a positive argument for this change, that the contest had grown in size and a bigger field of participants, with more variation in languages present, would perhaps create a more even playing field. This could, in theory, create more equal chances of winning. But, whatever the EBU’s reasons may have been, a policy that appeared to be created to prevent the use of English, was in fact a pro-English policy. The effects of it were still beneficial for English – or were at least perceived as such. This can be seen in the

continuous hegemony of English-speaking nations. The United Kingdom and Ireland in particular, but Malta as well, achieved great results in the contest during these national language-rule periods and particularly in this period. Their successes decreased when the use of English was allowed to all

(18)

17

countries. For instance, Ireland won in 1992, 1993, 1994 and 1996 and came second in 1997. The United Kingdom took home the victory in 1997 and came second in 1992, 1993 and 1998, but neither won or placed in the top tree after the language rules changed in 1999 (O’Connor 2005;

Motschenbacher 2016, 145f).

The language rules by the EBU are top-down policies, but they are created after bottom-up wishes. The EBU is the policy making authority for the competition, but in most cases, it changes its policies based on the demands the members make. The broadcasting companies in the various countries are therefore the ones actually making the decisions in the Eurovision Song Contest. The EBU is, after all, made up of these broadcasting companies. There is a board that governs the organisation, but it appears the members decide. Motschenbacher (2013, 7) wrote about the multilingualism policies in the EU, stating that they do not work because they are top-down, while the want for- and increased use of English is bottom-up. This is arguably also the case for the Eurovision Song Contest. The EBU obligated the participating countries to sing in their national language, but the feeling among the nations was that English was more successful and for that reason, they wanted to use it.

This increased longing for the use of English is further visible in the use of code-switching and thus the number of participants that did not entirely obey the rules about the use of national

languages. This happened more often during this rule period than in previous times

(Motschenbacher 2016, 123). An example of this is the Italian winner of 1990, with the singer Toto Cutugno exclaiming ‘unite, unite Europe’ throughout his song. The policies on the use of national language that the EBU had in place, were thus starting to show serious cracks during this period.

Minority languages were scarcely represented in the ESC. France used minority languages in a few of its entries during the 1990’s, for example by singing in Breton in 1996 and in Creole in 1992. The country used minority languages in the ESC, because it needed to improve its public image, after receiving criticism in the 1990’s for its purist language policy and for its negative policies towards minority languages (Motschenbacher 2016, 124f).

With the decades and the changing language rules, the styles of music in the contest changed as well. For instance, after ABBA’s win, the Swedish opinion towards the domestic popular music performed at the contest, shifted. This “national” music was now seen as ethnic and strange. Music shifted from national popular music to an international musical culture, maybe even specifically European, that had very few national markers (Björnberg 2007, 19f). There was a typically

Eurovision-style of music in the contest, at least up until the 1990’s (Björnberg 2007, 13). By some, this more commercial type of music was seen as a marker of cultural imperialism by the United States. The effect of this imperialism would, in this view, be the disappearance of traditional music (Badenoch 2013, 845).

(19)

18

The evolvement of a typical Eurovision sound could also partially be an effect of increased European cooperation on other levels that was going on at the same time, for instance in the EU’s predecessors, and the increasing idea of European unity, working towards the Maastricht Treaty of 1992. At the end of this rule period, from the early 1990’s on, this trend of stripping away national traits in the music, was countered by a trend of folkloristic music in the Contest (Björnberg 2007, 21).

1999-2017

If the Eurovision Song Contest were to be split in two, 1999 would be the breaking point. The language policy changed, but that was far from the only change the contest went through. The contests after 1999 took place in a different Europe, with many new participating countries and in a globalized world, with the European Union and without the Iron Curtain. Technology was becoming more advanced and the world connected through new media. From 1998 on, the vast majority of countries used televoting, where the viewers at home vote by phone call, instead of juries. The jury system had been in place from the very first edition of the contest onwards (Charron 2013, 485). The voting system was changed by the EBU after complaints from its members, which deemed the juries’ decisions of the previous years to be unfair. The hope was that televotes would change this

(O’Connor 2005, 147f). This hope did not seem to be realised, there still appeared to be block-voting, and during the recent editions, votes of the public (which can also be cast in the Eurovision app) make up half of the score. The other half of the score is once again provided by national juries. In the most recent years, the contest has also been broadcast live on the internet. Since 1999, the contest has thus made an effort to modernise itself and to stay relevant in a changing world.

The rules about language use were changed after the 1998 contest, because of the

continuous complaints from EBU-members that the national language rule gave songs in English too much of an advantage (O’Connor 2005, 69). Several other rules changed, too, which happened for economic reasons, as it allowed the contest to become more of a commercial event and made it easier for the songs to be used in a commercial way as well (Wolther 2012, 168). The rule change led the way for English to properly take over in the contest. While in the previous rule-free period English had been used often, but not in the majority of songs, this shifted after the 1999 rule change (O’Connor 2005, 157).

The idea behind the choice for an English-language song, was that English songs achieve better scores. Previous research has not been able to confirm this as true and it appears to mostly be the attitude people have towards English that makes them believe it is true. It is suggested that English is what the audiences prefer (Wolther 2012, 170). Because the EBU-members think English is the road to success - success being the winning of the contest and commercial success - singing English appears to have a more positive connotation than singing in another language.

(20)

19

English is strongly associated with globalisation. The ESC has also been influenced by globalisation, which is visible for instance in the globalising context of the event. Australia has broadcast the ESC for more than three decades and has participated in the contest since 2015. The country’s broadcaster SBS is an EBU-member (“Australia to compete” 2015). Nowadays, the show is also broadcast live in China (“Wake Up” 2016) and the United States (“Eurovision returns” 2017), for instance. One could wonder whether this would also be the case if English had not been the

dominant language in the contest since 1999. Nowadays, the whole event is in English, including many of the songs and this makes the contest more appealing to a world-wide audience. Reaching a wider audience in a greater variety of countries, is positive for the EBU, as it creates a bigger market for the organisation.

Since 1999, many new countries have participated in the ESC, but this new diversity is hardly seen in the entries’ language choices. Many of the former Soviet states have sung in English since their participation. Azerbaijan, for instance, has been very successful in the contest, but has never used its national language. This may have to do with the attitude towards English in these Eastern European states, where English is promoted as a language of democracy and human rights, as well as commercial success (Phillipson & Skutnabb-Kangas 1996, 431). Particularly for non-native speakers of English, using it “possesses a certain prestige in the ESC” (Motschenbacher 2016, 143). These

attitudes towards English suggest that English in the ESC is not a neutral lingua franca, because there are beliefs and perceptions about the function of English. Using English in the Contest is not merely done to communicate, but for status and for commercial gain.

The changing language rules coincided with the creation of the Charter for Regional or Minority Languages, by the European Council. Though the use of English has visibly increased, more countries have used minority languages after the 1999 rule change than before. Examples of this are the use of Udmurt in the 2012 Russian entry, which came second, Finland’s Swedish-language entry in the same year and Ukraine’s winning song in 2016, which was partially in Tatar. Estonia, France and Greece are some other countries that have been represented by a song in a minority language since 1999 (“Rules” 2017).

Minority languages are thus more present in the ESC since 1999. Songs with nonsensical words still can be found in the contest since 1999 and, because performers can choose every

language they want, several entrants have even sung in a made-up language. For example, the Dutch entry in 2006 (Treble’s Amambanda) was sung in an imaginary language, as well as the Belgian entries of 2003 (Urban Trad’s Sanomi) and 2008 (O Julissi by Ishtar). The success of made-up

languages differs; Belgium came second in 2003 (O’Connor 2005, 175), but neither Treble, nor Ishtar made it past the semi-finals.

(21)

20

The biggest musical trend since 1999 has been the ethnic trend Björnberg (2007) discusses. The trend is also visible in clothing. This can be seen as a countermovement to the Europop and typical Eurovision-songs of the time, but it has become a solid part of the competition in its own right. Not only does it appear to be a musical countermovement to the typical Eurovision music, it also seems to be a countermovement to the decreasing linguistic diversity in the ESC. There are many successful examples of the ethnic trend. It is visible most clearly in songs outside of Western Europe (Björnberg 2007, 21). Turkey’s 2003 winner Every Way That I Can was a successful example, as well as the 2004 Ukrainian winner Ruslana, with her Wild Dances and even Greece’s win in 2005 (Helena Paparizou, with My Number One) (Björnberg 2007, 21f). Along with these winners, the trend has taken flight during the past decade. All editions of the contest between 2005 and 2010 were filled with songs with tribal drums. The trend is not strictly limited to Eastern Europe, however. The Dutch entry in 2006 was played on drums. Norway won the ESC in 2009 with Alexander Rybak’s

Fairytale; a folk song, centred around the violin and Denmark won in 2013 with tribal drums and

flute. While the linguistic ecology of the contest is becoming less diverse, the music at least

celebrates cultural diversity (Björnberg 2007, 23). So far, the last part of the contest has been lacking linguistic diversity, with some exceptions.

(22)

21

4. Linguistic Imperialism? - An analysis of language choice and

scores

Phillipson (1997, 239f) informs us that, in order to find out whether we can speak of linguistic imperialism, one needs to map out the linguistic environment of the situation that is under

investigation. In addition to this, one needs to see whether one language is dominant in this situation and which attitudes and beliefs surround the languages present. These attitudes decide which languages are used. English, in his opinion, is not a neutral lingua franca, it is used for more than merely communication (Phillipson 2008, 250). In chapter three, we saw that English appears to have become dominant in the ESC since 1999, partially because of the belief that the language is the most successful one in the competition. The positive attitude towards English and the idea that national languages were not as successful, created a hierarchy between the different nations. English was seen as commercially successful, so the use of it was not purely for communication.

This, however, is the theory. There is also a practical side to the Eurovision Song Contest. This needs to be studied to see the practical effects and consequences of the EBU’s language policy. The songs are the most essential part of the contest and they also provide us with quantitative

information. For instance, many people immediately think of the famous ‘douze points’ when thinking about the ESC. These points make the competition and ought to be studied as well. In order to properly map out the linguistic ecology of the ESC, some quantitative data and statistical analysis is useful.

The question then is, how the term ‘linguistic imperialism’ can be operationalised for quantitative research. The first thing that was looked at in this following chapter was the presence – and possible dominance – of English, in the language choices made for songs. Literature showed a preference for English, but it is good to test this and see it for ourselves. Test were run to see

whether the use of English has increased significantly during the history of the ESC. Furthermore, the language choice patterns for several countries were mapped out in detail.

The second aspect to studying linguistic imperialism, the beliefs and attitudes surrounding language, is not straightforward. How does one know whether English is used for other reasons than communication? For this research, language choice was also seen as an expression of language attitude. Choosing a language for a song, says something about the attitude towards this language. For this part of the study, eight countries were studied as a sample. Four of them have a Romance language as their national majority language and four of them have a Scandinavian language as national majority language. The two groups have different characteristics: There is one group with large languages – languages with many speakers – and one with small languages. The languages in both groups are closely related to the other languages in their group. The Scandinavian languages are

(23)

22

particularly well-known for their mutual intelligibility, the Romance languages perhaps less so, meaning there is a difference between the two groups in this regard as well.

A cross-tabulation was done to see whether certain types of language groups are more likely to sing in a certain language. For instance, to see whether smaller languages choose to sing in English more than larger languages do. This can provide insight into whether the perceived hierarchy, caused by the EBU’s national language rule, indeed led to the increased use of English by smaller languages. It also creates a detailed picture of language choice patterns in the ESC.

The points that these eight countries scored, were also analysed. The points voters award to a song, are an expression of the attitude towards this song. There are of course many different factors that decide why a song gets these points, not merely language, and this complicates the investigation into the presence of linguistic imperialism. This research aims to figure out the possible effect of various language-related variables on these scores. These variables include the different language policies, the level of linguistic distance between the nations’ languages, the ‘size’ of the languages and the song languages. More details on why these variables were chosen, can be found in chapter 4.1. The idea is that the combination of these factors will show whether English-language songs are significantly more successful than those in another language. This would show that linguistic imperialism is at play, because the use of English would then create a commercial advantage. Should English-language songs be more successful, especially among closely related languages, for whom English is not necessary for communication, this would be remarkable. Should this be the case, could we then still say English is a neutral facilitator of communication? Or are there attitudes at play that decide that English is a success? Again, the influence of language is difficult to filter out of all the different factors that play a part in voting, but the combination of variables in this study may shed some light on the role of language.

4.1 Method

This chapter thus analyses the linguistic environment of the ESC and the awarding of points. This analysis is divided into several steps, each investigating a different aspect. The research is therefore split into five sub-questions. The answer to each sub-question is a part of the map of the linguistic ecology and hierarchies of the ESC. The questions are as follows:

1) Has the use of English for songs for all countries increased significantly over the years? 2) Has the use of English for songs increased significantly for the chosen language groups? 3) Is there an association between language group and the language choice for a song?

4) Are points awarded by the members of the other group affected by song language, language policy and/or language group?

(24)

23

5) Are the points awarded by the members of the own group affected by song language, language policy and/or language group?

In order to answer these questions, data was gathered through the official Eurovision.tv website, through the Wikipedia pages of all editions of the contest and through O’Connor’s work (2005). This data has been subject to several statistical tests, to be able to see whether the theory of the

relationship between ESC and languages, matches the practical situation.

First, the presence and possible dominance of English in the ESC was investigated, to see if English the preferred language to sing in when countries are given the choice. Therefore, this part of the study investigated the presence of English in the contests held while the national language rule did not apply. The three periods in which participants could choose their language freely were

compared, to see whether the use of English indeed increased, as the theory stated. The number of songs in English per year was added up and expressed as a percentage of the total number of songs in that year. This was done in percentages, because the number of participants varied most years. Songs were counted as an occurrence of English when they were monolingually or bilingually

performed in English (thus including songs with code-switching). The presence of English in the songs by the two language groups was investigated in the same manner. In this part of the research, the dependent variable was the percentage of songs in English. The independent variable was the language policy period, with three levels, because there were three periods of time during the ESC during which participants were free in their language choice.

Secondly, the language choices of the two language groups were analysed, during the same three periods during which language choice was free. This, as explained in the introduction, creates a detailed map of the ESC’s linguistic ecology: It extensively shows the language choices of different types of participants and thus their attitudes towards these languages. Additionally, knowledge was gained on whether the language choice patterns differ per type of language group. In contrast to the research into the presence of English in the previous questions, this part placed songs that were bilingual (English and another language) in a separate category to the ‘songs in English’ category. This was done, to create a more in-depth map of the countries’ language choice practices and to look at code-switching, too, for example.

The final parts of this statistical analysis focused on the points received by the two language groups. For questions four and five, the dependent variables were respectively the percentages of points awarded by the members of the other group and of the own language group. The independent

(25)

24

variables in this part were: Song language (nominal, with six levels), language group (nominal, two levels) and language policy type (nominal, with two levels). More about the levels is stated in chapter 4.1.3. These independent variables were chosen, because they all represent a role of language in the contest. The combination of them analysed the points according to language factors. This is to show differences between the different types of languages in the competition, that were mentioned in chapter three. One of the variables is the type of language policy in place; whether the song was performed under the national language rule or not. This is investigated because minor languages, like the Scandinavian languages, have complained in the past that major languages like English had an advantage during the national language-rule. This might mean that, within the groups, more points may have been awarded under the national language rule, as previous research found that often songs in a related language are preferred (Spierdijk & Vellekoop 2009).

Language group was a variable to see whether different types of languages have different voting patterns. Language choice was used as a variable to see whether it is true that songs in English score better in the contest or that songs in a related language are indeed preferred.

4.1.1 Participants

The participating songs in the Eurovision Song Contest are also the participants in this study. For the first test all songs were investigated, for the latter parts the songs of eight countries, which form a sample. They are divided into two language groups. The national languages used in the Eurovision Song Contest belong to various language families and to various sub-groups of these families. There are, for example, various Slavic languages in the contest, as well as Romance languages and Uralic languages. For this research, a sample of the countries was necessary. It would be too time-consuming to investigate the points of all entries in the ESC. Additionally, chapter three showed a hierarchy between different types of languages. Therefore, this study needed a way to arrange the participating countries by language and therefore nations were chosen with related languages. Language groups were also chosen to be able to shed more light on the neutrality of the role of English. With related languages that do not need English for communication, it would be noteworthy if English songs do better. We could subsequently investigate why they do so and see whether English is used neutrally or that linguistic imperialism is at play.

There were various criteria for the language groups in this study: Both groups needed to contain languages that were part of the same branch of a language family tree, one perhaps more closely related than the other, because a higher score for English songs by closely related languages could mean English is more than just a facilitator of communication. Additionally, one group needed to consist of larger languages (in number of speakers) and one of smaller languages. This was done

(26)

25

because of the perceived hierarchy between these types of languages, which is explained in chapter three. An equal number of countries was picked per language group (four).

The first group is the Scandinavian language group, with Sweden, Norway, Iceland and Denmark. The close relationship between these languages is well-documented. Figure 1 gives an image of the presence of these Scandinavian countries in the Eurovision Song Contest; it shows how long they have been participating and how well they have done. Denmark was the first of these four countries to participate in the Eurovision Song Contest, in 1957. Iceland was the last country of the four to start taking part, in 1986. The table shows that for instance Sweden and Denmark have performed especially well in the contest since the national language rule was cancelled in 1999. The Scandinavian languages are also ‘small’ languages; languages with a relatively small number of speakers. The difference in attitudes towards English in Eurovision between speakers of smaller and of larger languages has been shown in chapter 3.

Figure 1: The history of the Scandinavian countries in the ESC

First

participation

Missed editions since 1st_entry

Wins Last places (in the final) Sweden 1958 1964, 1970 1974, 1984, 1991, 1999, 2012, 2015 1963, 1977 Norway 1960 1970, 2002 1985, 1995, 2009 1963, 1969, 1974, 1976, 1978, 1981, 1990, 1997, 2001, 2004, 2012 Denmark 1957 1967-1977, 1994, 1998, 2003 1963, 2000, 2013 2002 Iceland 1986 1998, 2002 - 1989, 2001

To contrast with this, another language group had to be compared. Four countries with Romance languages were used for this comparison. The Romance languages are globally spoken by more people than the Scandinavian languages. Four countries were randomly picked from this language group: France, Italy, Portugal and Spain. Three of the countries with Romance languages in this research are part of the Big Five and are thus always allowed to take part in the final of the contest, namely Spain, France and Italy. All four countries have been ESC-participants since early on:

(27)

26

France and Italy even took part in the first edition of the ESC. The last Romance country of these four to join was Portugal, which first took part in 1964. Figure 2 presents information on the history of these Romance language-countries in the Eurovision Song Contest. The table shows that, with the exception of Portugal’s win last year, none of these Romance languages have won since the rule change in 1999. France has not won the contest since the 1970’s.

Figure 2: The history of the four Romance language-nations in the Eurovision Song Contest

First

participation

Missed editions since 1st_entry

Wins Last places (in the final) Italy 1956 1981, 1982, 1986, 1994-1996, 1998-2010 1964, 1990 1966 France 1956 1982 1958, 1960, 1962, 1969, 1977 2014 Portugal 1964 1970, 2000, 2002, 2013, 2016 2017 1964, 1974, 1997 Spain 1961 - 1968, 1969 1962, 1965, 1983, 1999, 2017

4.1.2 Data

The first question maps out the presence of English in the history of the ESC. The data for this part of study was gathered through Eurovision.tv and Wikipedia. The latter has more direct information on song language than the ESC website, because it shows a table with a column dedicated to song language, whereas the ESC website only provides lyrics to check language. Wikipedia is not generally considered a fail-safe way place for the gathering of information. To test this information’s reliability, several of the songs’ lyrics of various editions were checked to see whether Wikipedia was correct. This was the case, so the information was used.

Information was collected for the years in which the national language rule did not apply, meaning the editions between 1956-1965, 1973-1977 and 1999-2017. The data for each year was given a number 1, 2 or 3; 1 meaning it was an edition between 1956 and 1965, 2 for a year between 1973 and 1977 and so forth. The number of English songs was counted and expressed in a

percentage of the total number of songs participating in that year of the contest. In this data, a song counted as an occurrence of English when it included sentences in English.

(28)

27

To find out whether the use of English increased significantly for the two language groups, data was gathered in the same way. The difference is that the percentages there express the number of English occurrences by language group participants, out of the total number of participants for both language groups in that year. As an example, in 1973 Sweden sung in English and Norway in a mixture of English and various other languages. Six of the language group members participated in this edition of the contest: Spain, Portugal, Italy, France, Sweden and Norway. The percentage of English occurrences used as a score is thus 33.33.

In order to investigate the detailed language choices and awarding of points in the language groups, data was collected for all past editions of the contest, so from 1956 to 2017. The information was found through the Eurovision.tv website, through Wikipedia and through O’Connor (2005). For the language groups the data was coded per country. The data that was collected per country was as follows:

1) Year

2) Type of language policy (obligatory national language rule or free choice) 3) Country

4) Language of the song 5) Placing in the Final 6) Total Points

7) Total points awarded by others in the own language group

8) Number 7, expressed in a percentage of the total points that in theory could have been awarded by the own language group

9) Total points awarded by the members of the other language group

10) Number 9, expressed in a percentage of the total points that in theory could have been awarded by the members of the other language group

As mentioned before, the points used in the statistical analysis are percentages. This was done because the scoring system has varied greatly over the years. There are two percentages in this data: the points received by a country from the others in its own language group and the points received by a country from the countries in the other language group. For instance, in 2011 Denmark received 29 points from the other Scandinavian countries. Iceland, Sweden and Norway all voted in this contest, so in theory Denmark could have received a total of 36 points from its group members. The number used for analysis here is thus the percentage of 80.56.

LINGUISTIC DIVERSITY OR LINGUISTIC IMPERIALISM?