SCHWEIZER ANGLISTISCHE ARBEITEN SWISS STUDIES IN ENGLISH

Silja Ang-Tschachtli

# **Bilingual Couples in Conversation**

Bilingual Couples in Conversation

## Schweizer Anglistische Arbeiten Swiss Studies in English

Begründet von Bernhard Fehr Herausgegeben von Andreas Fischer (Zürich), Martin Heusser (Zürich), Daniel Schreier (Zürich)

Band 149

Silja Ang-Tschachtli

## Bilingual Couples in Conversation

Umschlaggestaltung: Martin Heusser, Zürich Umschlagabbildung: Silja Ang-Tschachtli

Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.dnb.de abrufbar.

Silja Ang-Tschachtli, ORCID: 0000-0002-3015-2246 DOI: https://doi.org/10.24053/9783772057632

Publiziert mit Unterstützung des Schweizerischen Nationalfonds zur Förderung der wissenschaftlichen Forschung

© 2022 · Silja Ang-Tschachtli

Das Werk ist eine Open Access-Publikation. Es wird unter der Creative Commons Namensnennung – Weitergabe unter gleichen Bedingungen | CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/) veröffentlicht, welche die Nutzung, Vervielfältigung, Bearbeitung, Verbreitung und Wiedergabe in jeglichem Medium und Format erlaubt, solange Sie die/den ursprünglichen Autor/innen und die Quelle ordentlich nennen, einen Link zur Creative Commons-Lizenz anfügen und angeben, ob Änderungen vorgenommen wurden. Die in diesem Werk enthaltenen Bilder und sonstiges Drittmaterial unterliegen ebenfalls der genannten Creative Commons Lizenz, sofern sich aus der am Material vermerkten Legende nichts anderes ergibt. In diesen Fällen ist für die oben genannten Weiterverwendungen des Materials die Einwilligung des jeweiligen Rechteinhabers einzuholen.

Narr Francke Attempto Verlag GmbH + Co. KG Dischingerweg 5 · D-72070 Tübingen Internet: www.narr.de eMail: info@narr.de

CPI books GmbH, Leck

ISSN 0080-7214 ISBN 978-3-7720-8763-9 (Print) ISBN 978-3-7720-5763-2 (ePDF) ISBN 978-3-7720-0171-0 (ePub)

*I would like to dedicate this book to my own bilingual family: my husband Beng the inspiration for my research — and our three children, Vincent, Jonathan, and Malina. Thank you for your love and patience, and for making me laugh every day. You are amazing.*

## **Contents**







#### **Abstract**

In this sociolinguistic study, the speech and the modes of communication of ten bilingual couples were analysed, based on a corpus of in-depth interviews with both partners. My main aim was to draw a detailed picture of various areas pertaining to the communication between native speakers of English and Swiss partners who, despite being late bilinguals, have a high level of proficiency in English, their main couple language. Thus, important gaps in previous research on bilingual couples were filled, much of which is based on couples' self-reports rather than their actual language use (Breger 1998; Khatib-Chahidi, Hill and Paton 1998) or concentrates on the female partner's perspective (Heller and Lévy 1992). Studies on fluent couples have placed their focus on the discursive construction of a bilingual couple identity (Piller 2002a) or an individual bilingual identity (Gonçalves 2013). By contrast, I considered not only the couples' reports on their linguistic practices and past experiences, but also analysed their language use during the interviews. Several areas of their speech were closely examined using qualitative as well as quantitative research methods, including the couples' language choice, their language mixing, their manner of expressing emotions, their use of and reaction to swearwords, their attitudes towards their languages and cultures, and their production of humour and laughter. In addition, I investigated the influence of the bilinguals' gender and mother tongue on various aspects of their language use. The analysis demonstrated that the couples' language is predominantly English, with relatively little direct influence from the com‐ munity language, Swiss German, and limited language mixing. Moreover, the bilinguals used only their main couple language to express emotions during the interviews, and also reported such a preference, in contrast to other bilinguals (Pavlenko 2008; Dewaele 2010). Nevertheless, the couples make use of their bilingualism in some very specific areas, for instance when coining neologisms and blended expressions, when using terms of endearment, when swearing, and when being humorous. Assimilation between the partners was evident in several areas, such as their language mixing behaviour, their swearing behaviour, and their attitudes; yet there was little indication of assimilation in the context of expressing emotions, or in the frequency and duration of laughter. At the same time, the participants' gender and mother tongue were found to have a considerable influence on their expression of emotions, the number and type of language switches they used, the frequency of their swearing and the level offensiveness of their swearwords, as well as their laughter and their production and reception of humour. Thus, the analysis provides insights into the language practices of established bilingual couples, while also contributing to the fields of gender research and fluent sequential bilingualism.

## **List of transcription conventions**

### (adapted from Du Bois et al. 1993)

#### **Final pitch movement, stress and tone**


#### **Pauses and organisation**


#### **Vocal noises**


#### **Voice quality, loudness, pitch**



#### **Marginal words**


#### **Metatranscription**


## **List of terms and abbreviations**




## **Acknowledgements**

I would like to express my sincere gratitude to the many people who offered guidance in my research and writing, especially PD Dr Sarah Chevalier and Professor Dr Andreas Fischer. I am very grateful to Andreas Fischer for encour‐ aging me to pursue my research interests and for providing me with valuable direction whenever needed; and to Sarah Chevalier, for supervising my PhD thesis, for sharing my interest in the subject of my research, and for providing excellent advice in my research and very helpful commentaries on my writing. I would also like to thank Professor Dr Daniel Schreier, for kindly offering to be a co-advisor on my thesis and for his helpful feedback, and Professor Dr Martin Heusser, for designing the book cover.

Special thanks go to Jane Dewhurst and Peter Stäuber, for proof-reading critically and providing thoughtful feedback on my writing, and to Martina Gubler, for her advice in the creation of the title image. I am also thankful to Katharina Gerhardt (Narr Francke) and Professor Dr Stephan Schmid (Pho‐ netisches Laboratorium) for their help, as well as to Christine Benesch and Nicoletta Ravizza for their advice on statistics. Furthermore, I would like to thank the Swiss National Science Foundation (SNF) for financing the publication of this book.

I am also truly grateful to my family, especially to my wonderful parents, Irene und Stephan Tschachtli, for their love and support.

Last but not least, I want to express my deep appreciation to the ten bilingual couples who were the subject of my research. Thank you for devoting your time and for sharing so much about your private lives as bilingual couples and individuals. It is due to your readiness to tell your story and to engage in dialogue that this book was possible, and that my interest in the subject never faded.

## **1 Introduction**

#### **1.1 Bilingual couple communication**

After my first semester of studying English at the University of Zurich, I met an interesting young Australian while we were both travelling in Italy. We had a long-distance relationship for a few years, ultimately got married and settled in my home country, Switzerland. Being in a bilingual, bicultural relationship has enriched my life in more ways that I could have imagined, while at the same time presenting challenges in other areas. It has also opened my eyes to many particularities in the discourse and communication of bilingual, bicultural couples and sparked a great interest in this topic. As a bilingual couple, we are in contact with other bilingual couples, and observing other bilingual relationships and comparing their communication to ours has also made me aware of many shared characteristics. However, I rarely find our own situation or the bilingual relationships around us mirrored or adequately described in literature on bilingualism, which has intensified my urge to study bilingual couple communication. On the one hand, linguistic research on bilingual couples has remained relatively scarce, even though studies in the area of individual bilingualism have been plentiful, covering areas as diverse as childhood bilin‐ gualism, second language acquisition, and the communication between people from different linguistic backgrounds. This omission is particularly surprising considering how many couples are actually bilingual, and given the diverse and interesting nature of communication among bilingual couples. On the other hand, the studies that do exist on bilingual relationships tend to focus on partners who do not speak the couple language fluently, who speak languages which are not closely related, and/or who are from very different cultural backgrounds (e.g. Rosenblatt and Stewart 2004; Beraud 2016). Only in recent years have couples with similar linguistic and cultural backgrounds shifted into focus, as researchers have begun to look at aspects such as couple identity and language desire (Piller 2002a; Gonçalves 2013). However, there are still many interesting aspects about the communication of bilingual couples that merit research, especially when it comes to partners in established relationships, who are proficient in their couple language.

Since bilingual couples exemplify a wide variety of language combinations and range of individual linguistic situations, they provide fertile ground for research on bilingualism. Depending on the bilinguals' personal backgrounds, their aptitude for languages, their partner's language skills and their place of residence, their competence in their partner's native tongue can range from virtually non-existent to completely proficient. In some cases, both partners are fluent in both languages, while in others, one partner's mother tongue is the exclusive couple language, or neither of the partners speaks the other's language, but they speak a third language (lingua franca) instead. In theory, then, it is not even necessary for both partners in a bilingual relationship to become bilingual. At the same time, there can also be substantial differences in the partners' level of motivation to learn each other's language. Whereas for some, becoming bilingual is an involuntary — but perhaps a necessary consequence of their relationship, others may have chosen their partner at least in part because of his or her native tongue, which allows them to immerse themselves in another language and to become or stay (fluent) bilinguals.

Furthermore, bilingual couple talk can offer insights into fluent bilingual communication in a manner in which few other types of bilingual interaction do. Many studies on bilingual interaction focus on one-time exchanges involving one or several bilingual speakers who often have limited language skills, rather than conversations between people who are accustomed to each other's manner of speaking. Unlike the bilinguals investigated in such studies, people in long-term intimate relationships are used to interacting with one another; their modes of communication are thus more established than most other forms of bilingual interaction. As a consequence of their repeated interaction, the partners may learn how to read and interpret each other to some degree, and misunderstandings and conflicts may become less frequent over time. If the partners in a bilingual relationship have been together for an extensive period of time and are fluent in the couple language, their communication can be expected to be more advanced than that in most bilingual encounters.<sup>1</sup>

Finally, another reason why bilingual couples can be a particularly interesting focus for the study of bilingual interaction is that communication tends to take a central role for couples. Communication may be important in any kind of relationship, but especially so in couple relationships, which depend on regular communication to develop and evolve. Thus, some even believe communication to be "the sole major cause of marital happiness or marital failure in modern postindustrial societies" (Piller 2002a: 5). As Crow puts it, "[t]he couple can be viewed, in fact, as the institution in which conversation as a system reaches its

<sup>1</sup> I use the term *advanced communication* (as opposed to *basic communication*) to refer to communication which is complex and characterised by very good understanding between the interlocutors.

fruition: Couples are formed in part to assure oneself a conversational partner with whom to develop a wealth of episodes, and conversation serves as a primary means of developing and enriching a relationship" (1983: 140). Couples will necessarily develop a distinct manner of communicating over time, and adopt aspects of each other's language use. In the case of bilingual couples, this process of assimilation might be even more marked than in monolingual couples, as language learning — even at an advanced level — is closely connected to imitation (Brown 1980: 43). Bilingual partners do not just learn aspects of the language itself from one another, but also how to cooperate with one another (for instance in turn-taking and integrating adequate pauses), and to support each other conversationally (for instance with back-channel signals or other-repetition). Fluent speakers in a long-term bilingual relationship therefore fulfil many of the requirements for successful bilingual communication and examining conversations between them should yield substantial insights into how bilingual communication works on a more advanced level.

In order to explore areas that have not been researched and to give an account of the communication of fluent bilingual couples, I recorded conversations of ten heterosexual bilingual couples, each consisting of a native English speaker and a native Swiss German speaker. In the transcriptions of their interviews, I included a large variety of suprasegmental features, ranging from terminal pitch to stress and voice quality. The comprehensive corpus resulting from the couples' conversations and the elaborate transcriptions provided an ideal foundation to closely examine a range of features of their couple language. In order to reduce the number of variables influencing the couples' communication, I chose couples who share a number of characteristics. For one, all of the couples mainly communicate in English with each other, even though all but one couple, who have a long-distance relationship, reside in Switzerland. While there are considerable differences in the (Swiss) German language skills of the English-speaking partners, all of the Swiss partners are highly proficient in English. All of the couples have been together for a considerable period of time, and lived apart from each other in different countries for some of this time. This means that they have had to face and deal with issues that commonly arise in bilingual, bicultural relationships and have become well acquainted with each other's cultures. The long duration of their relationships increases the likelihood that they have attained a high level of communicative competence in their couple language, and that they have also adopted features from each other's manner of speaking. It can be expected that all of the couples have developed their own modes of communication, in addition to having a wealth of shared experiences to report, intercultural or otherwise. Consequently, the analysis of these interviews will provide insights into the communication of established bilingual couples, whilst simultaneously filling some gaps in the field of fluent sequential bilingualism.

#### **1.2 Aim and research questions**

The main aim of this book is to paint a comprehensive picture of the communi‐ cation in a bilingual, bicultural relationship between proficient, culturally aware bilinguals. In this context, a number of questions present themselves. Which language(s) do the couples choose to communicate in, and what are the reasons behind this choice? To what extent do they mix languages, and what influences their mixing behaviour? What are the partners' attitudes towards their bilingual, bicultural relationship? How do partners in a bilingual relationship express positive and negative emotions? How do fluent bilinguals swear, and what affective attachment do expletives in both of their languages have for them? What role do humour and laughter play in their conversations, and what do the couples laugh about? And, finally, do their different mother tongues and their gender have an effect on a variety of areas of their communication?

In order to answer these questions, I give a detailed account of several areas, investigating various aspects pertaining to each of these areas. On the one hand, I use qualitative research methods to give an overview of the couples' language, experiences, thoughts, and attitudes towards a variety of aspects pertaining to their bilingual, bicultural relationships. Hence, I look at which language(s) the couples choose to communicate in what situations, what the reasons for this choice are and how and why their couple language has evolved over the course of their relationship. This includes their language mixing and factors that influence their mixing behaviour. I also consider the interviewees' attitudes towards various aspects of their bilingual, bicultural relationship and the manner in which they deal with expressing positive and negative emotions in their second language. Moreover, I look at the bilinguals' reported language preference for swearing, their emotional attachment to swearwords in both languages and their reactions to their partner's use of swearwords. I also discuss the challenge of conveying and understanding humour in a second language, as well as the role of humour and playful language in the couples' communication. All of these areas are explored by means of a content analysis as well as a close linguistic analysis of the couples' speech and interaction.

On the other hand, I use quantitative research methods to investigate a number of hitherto rather neglected areas in the language of fluent bilinguals. I examine the bilinguals' language switches with regard to a variety of features and determine whether there are any differences in the participants' mixing behaviour arising from their mother tongue or their gender. I explore the manner in which they convey positive and negative emotions during the interviews, and look at the frequency with which they use emotion words, the suprasegmental features and terminal pitch accompanying emotion words, and the correlation between the use of emotion words and the bilinguals' gender and mother tongue. Moreover, I examine their use of swearwords during the interviews and the role which their mother tongue and gender play in their swearing behaviour. Finally, I also look at who and what triggers laughter in the bilinguals' speech, their use of laughing and smiling voice quality, similarities in the partners' laughing behaviour and humour styles, but also differences in their laughing behaviour based on their mother tongue and gender.

Thus, my study encompasses both the analysis of participants' actual lan‐ guage use and modes of communication as demonstrated in the interviews, and the analysis of the couples' own reports about their linguistic behaviour and their relationships. Throughout, I tend to follow a grounded theory approach, taking my own material as the basis from which I observe behaviours and set up taxonomies, albeit bearing previous findings in mind, rather than superimposing existing theoretical frameworks onto my material. In my analysis, I explore intercultural, linguistic, as well as self-reflective and psychological aspects. This combination of different research methods and the broad range of topics that are examined will ensure a comprehensive picture of the modes of communication utilized by the bilingual couples.

#### **1.3 Limitations**

Of course, the communication between any two people is so complex that it would be impossible to give a complete account of it, and it has been necessary to omit certain aspects. For instance, nonverbal communication such as facial expressions, eye-contact, and other elements of body language, are not considered in the analysis, as, on the one hand, this would have exceeded the scope of this book, and, on the other hand, I wanted the interviewees to feel comfortable and speak as naturally as possible, and a camera might have made them feel more self-conscious than an audio recording device. All aspects of conversational analysis are thus based on their verbal expression and prosodic features alone. Moreover, the issue of conversational breakdowns is not considered, and the topic of the construction of couple identity as well as cross-cultural identity in bilingual couples are not explored at length, as this has already been examined elsewhere (see Piller 2002a; Gonçalves 2010a, 2013).

Thus, the main focus of my analysis is on a selection of aspects I believe to be of particular interest about this specific combination of languages and cultures, and about couples with such a high level of proficiency in the relationship language. This narrow focus entails that no universal statements about the communication in bilingual relationships can be made based on this research. In my opinion, however, this does not represent a problem. As Okita puts it, a small-scale study may not aim to make "*empirical generalisations*", i. e. generalisations from a sample "to a wider population, based on [the] represen‐ tativeness of the sample". Rather, it should aim for "*theoretical generalisation*, an explanation of how and why things happened in specific settings, and identifying the key explanatory factors in the process, from which questions can be asked about 'lessons for other settings', or *wider resonance*" (2002: 62, emphases in original). Thus, even though the results of this study may not be representative for bilingual, bicultural couples in general, they still give us an understanding of how bilingual couples with these specific parameters communicate. The study reveals that there are countless variables that factor into each couple's language and into each instance of bilingualism. But precisely because there are myriad forms of bilingualism, it is important to examine individual cases such as these in order to further our understanding of all facets of bilingualism.

#### **1.4 Outline**

This book is structured as follows. To begin with, I present a short overview of the language situation in Switzerland, which is the place of residence of the couples in this book (**chapter 2**). The linguistic makeup of the country is rather complex, partly due to its multilingualism, but also due to the diglossic situation in the German-speaking part of the country. I also describe the role of the English language in Switzerland, both as an influence on local varieties and as an important language in education, tourism, and commerce. This contextual information is important, as the attitudes and language ideologies attached to these different varieties may be an important factor with regard to the couples' language choice.

In the subsequent chapter, I give an overview of previous work on bilingual, bicultural couples, and define some important terms with reference to bicultur‐ alism and bilingualism (**chapter 3**). For the sake of clarity, I discuss research on bicultural and on bilingual couples separately, as although the two are closely related, most research concentrates on one or the other. In each part, I also present an outline of the challenges bilingual, bicultural couples may face. In this chapter, I only discuss general aspects of research on bilingual, bicultural couples. An overview of more specific studies in the areas which are examined in this book is offered in the corresponding chapters.

**Chapter 4** describes the sample and research methods. I present an overview of the most important information relating to each interviewee, as well as a short biography of each couple. In addition, I outline my interviewing methodology and the design of the questionnaire that formed the basis of the interviews. The transcription conventions are also described, as well as the software used for transcribing the interviews and analysing the data. The chapter also includes a discussion of the difficulties that I encountered during the process of data collection and analysis.

After the description of the setting, subjects and methodology for data collection, I present the main body of my analysis, which has been divided into six chapters. I begin in **chapter 5** with what may be the most influential factor in determining the manner in which the couples communicate, namely their language choice. After outlining the factors that have been found to influence language choice in bilinguals, I discuss the couples' reported language use with each other, and the reasons behind this. In addition, their language use outside the home is discussed, as this may be connected to their language proficiency, attitudes, and/or level of integration. In the same chapter, I also document some aspects that the couples report as unusual about their way of communicating. This includes topics such as moderating their manner of speaking, and the high level of communication and mutual understanding that they claim to have achieved as a result of their different backgrounds.

**Chapter 6** examines the couples' language mixing behaviour. This chapter is a combination of the couples' reports on their language mixing on the one hand, and their mixing behaviour during the interviews on the other. In order to analyse their language mixing, I distinguish between code-switches and borrowings, between hedged and unhedged switches, and between spontaneous and metalinguistic switches. The participants' language switching behaviour is then analysed with reference to a number of variables, such as their family situation, mother tongue and gender, and compared to their reports about their mixing behaviour and their views on language mixing.

After this, I turn to more specific areas in the couples' communication and relationships. In **chapter 7**, I look more closely at the topics of attitude and attraction, as these influence the couples' linguistic behaviour considerably. I discuss parallels in what the participants found attractive about each other initially, as well as the development of their attitudes towards their partners' language and culture over the course of their relationship. Moreover, I examine their views on being a bilingual, bicultural couple, and discuss the expectations, hopes and worries the couples voice with regard to raising bilingual children. As the topic of raising bilingual children has been widely researched and documented, this aspect is discussed only briefly; nevertheless, it is included in the interests of comprehensiveness. Furthermore, children may also have an influence on the language use of a bilingual couple, and the partners' private language planning gives us clues about covert attitudes.

The topic of expressing emotions in a bilingual relationship is addressed in **chapter 8.** In this chapter, I examine which terms the couples use to express positive and negative emotions during the interviews, as well as the voice quality and terminal pitch accompanying the expression of emotions. I also analyse to what extent speakers with different genders or mother tongues differ in their expression of emotion. The couples' thoughts on expressing positive and negative emotions in a second language are also discussed, as are potential issues for their relationship that arise from their situation, and strategies for dealing with these issues.

In **chapter 9**, the bilinguals' use of swearwords is analysed. I discuss both their reported use of swearwords and their swearing behaviour during the interviews, with particular focus on their language choice, as well as the role of gender, mother tongue, nationality and family situation. Furthermore, I explore the interviewees' reactions to their partners' use of swearwords, and the taboos associated with swearwords in their first and second language.

Finally, I look at the role and function of humour and laughter in the couples' conversations (**chapter 10).** I first analyse the manner in which the couples use humour during the interviews, with particular attention to the topics that trigger laughter, as well as who laughs about whom, and with what frequency. I also discuss the participants' use of a laughing and smiling voice quality during the interviews, and see if any aspects of their laughing behaviour seem to be influenced by the factors of gender or mother tongue. In the second part of the chapter, I offer an overview of the bilinguals' reports on their individual and couple humour. In addition, challenges that are brought about by cultural and linguistic differences in humour are examined, as well as shared couple humour. In this context, I also explore an aspect that many of the couples report as typical of the manner in which they communicate, namely the use of playful language, for instance by inventing words or imitating accents, as all of this is usually done in a humorous key. In the final chapter (**chapter 11**), I then provide a short overview of the most important findings of my analysis, and discuss the implications of these results for future research.

## **2 Language situation in Switzerland**

## **2.1 Introduction**

This chapter focuses on the language situation in Switzerland with regard to the country's multilingualism as well as the diglossic situation in the German-speaking part. Both of these aspects are potentially relevant for the couples' attitudes towards their two languages and may therefore influence their language choice and use. Moreover, the role and status of English in Switzerland is examined, as well as the linguistic and ideological challenges Anglophone immigrants may face there. With regard to all of these topics, the chapter concentrates on the German-speaking regions of Switzerland, because the interviewees live in this area.

## **2.2 Multilingualism in Switzerland**

In 2016, Switzerland had a population of about 8.4 million people, 24.9% of whom were foreign nationals (BFS 2017).<sup>1</sup> Switzerland is officially a quadrilingual country, with four national languages: German, French, Italian and Romansh. However, Swiss nationals are seldom fluent in all (or even two) of these languages, and many people do not grow up in a bilingual environment. Rather, each of the four national languages is spoken within a specific territory which is effectively monolingual (with the exception of a few bilingual areas). In the 2014 survey on the population's main language(s), 63.3% of the population indicated (Swiss) German, 22.7% French, 8.1% Italian, and 0.5% Romansh as their main language. A total of 22.3% of the population named a non-local language as their main language (or one of their main languages); English was spoken as a main language by 4.6% of the population (BFS 2016a).<sup>2</sup> While not all inhabitants use other languages regularly, at least two foreign languages<sup>3</sup> are usually learnt at school. Moreover, 38.7% of the adult population indicated in the 2014 census

<sup>1</sup> All figures in this paragraph are from the Swiss Federal Office of Statistics (BFS).

<sup>2</sup> The total of these percentages exceeds 100%, because in the 2014 census, it was possible to indicate several main languages for the first time.

<sup>3</sup> In this context, the term *foreign language* is used to refer to a language that is not the individual's native tongue, whether or not it is a national language.

that they regularly use more than one language privately and/or professionally (BFS 2016a).

Even though the majority of the Swiss population may be confined to mainly using one language in their daily lives, it has been noted that being a multilingual country is part of the Swiss national self-image and cultural identity (Stevenson 1997: 22; Murray 2003a: 103). Consequently, it is often argued especially in political debates — that "the Swiss should be multilingual and able to communicate with each other in their national languages as a way to better mutual understanding" (Murray 2003a: 103). As Rash points out, the acquisition of a second national language is often "seen as central to intercommunal harmony" (2003: 123). At the same time, many Swiss recognize the importance of speaking English well in an international context, which is why second language teaching is a hotly debated topic. There is disagreement concerning which foreign language should be learnt first in each canton — a national language or English — and at what age these foreign languages should be acquired. In 2004, it was decided that all Swiss students should start learning a second language in third grade, and a third language in fifth grade. One of these languages has to be a national language, but apart from that it is up to each canton to decide which foreign languages their students will learn, and in what order. This new curriculum, the *Lehrplan 21*, has been introduced at schools in most cantons (Lehrplan 21, 2017). As a result, English is currently the first foreign language taught in 16 of the German-speaking cantons. However, some German-speaking cantons do not follow the rule that a third language has to be introduced by the fifth grade, and do not start teaching it until the ninth grade. This recently led federal council member Alain Berset to impose an ultimatum on all cantons, which was front-page news in many Swiss newspapers (e.g. Häfliger and Burri 2016). On the other hand, a popular initiative in the canton of Zurich which aimed to postpone learning one foreign language until secondary school (seventh grade) was rejected by the majority of the voters (60.8%) in May 2017. The fact that there are frequent headlines and political discussions on the topic of second language teaching in schools shows both the controversial nature of the topic and the importance attributed to language learning in Switzerland (see also section 2.4, "English in Switzerland").

#### **2.3 Diglossia and the ideology of dialect**

All of the Swiss who were interviewed originate from the German-speaking part of Switzerland. In this area, around thirty mostly Alemannic dialects of German are spoken, which are mutually intelligible (Watts 2001: 301). The linguistic makeup of this part of the country is somewhat unusual because of its diglossia. In diglossic situations, regional dialects (traditionally called the *L variety* by researchers) co-exist with a more standardized variety of the same language (*H variety*) — in this case Swiss German (L) and Standard German (H).<sup>4</sup> Referring to Switzerland among other countries, Charles Ferguson proposed the following classic definition of diglossia:

DIGLOSSIA is a relatively stable language situation in which, in addition to the primary dialects of the language [… ], there is a very divergent, highly codified (often grammatically more complex) superposed variety, the vehicle of a large and respected body of written literature, [… ] which is learned largely by formal education and is used for most written and formal spoken purposes but is not used by any sector of the community for ordinary conversation. (1959: 244–245, emphasis in original)

According to Ferguson, the varieties differ not only in the contexts in which they are used, but also in the prestige attributed to each. He states that the H variety is regarded as superior to the L variety in a number of respects, and is often viewed as "more beautiful, more expressive, more logical" (1959: 248).

However, this does not seem to be the case for the two varieties in Switzerland nowadays, and a number of researchers have remarked on the high prestige which Swiss German carries for German-speaking Swiss. Watts believes that the fact that the local dialects are valued more highly than Standard German in Switzerland these days is the result of an "ideology of dialect" which has developed over the last century (1999: 71). In such a situation, "the symbolic value of the dialects in the majority of linguistic marketplaces in which they are in competition with the standard is not only believed to be much higher than that of the standard but is also deliberately promoted as having a higher value" (Watts 1999: 69). Thus, Swiss German is attributed positive characteristics by contrast with Standard German, and many Swiss find the former "more down-to-earth, more honest, more communicative, more direct, and, in general, more Swiss than standard German" (Watts 1999: 75).

The ideology of dialect is continually promoted by the Swiss media and educational system (Watts 1999: 89), as well as by the German-speaking Swiss themselves. As an example, Gonçalves found that the native speakers of Swiss German as well as English in her study repeatedly reproduced the ideology of dialect in their discourse. Most of the native speakers of English she interviewed

<sup>4</sup> Usually called "High German" by the Swiss, the variety used in Switzerland is effectively a Swiss version of Standard German, which is very similar to but not identical with the standard variety of German spoken in Germany or Austria (cf. Rash 1998: 49).

value Swiss German more highly than Standard German, as they "position and align themselves with Swiss German native speakers and their respective language ideologies" (2013: 151). The rise of the ideology of dialect may also have contributed to a shift in the domains in which the two varieties are used. Swiss German is now used far more frequently on the radio and on television, as well as in informal writing, such as text messaging. Consequently, the situation in Switzerland can be seen as a case of "leaky diglossia", as "one variety 'leaks' into the functions formerly reserved for the other" (Fasold 1984: 41).

Not only does Swiss German have a high symbolic value in Switzerland, many Alemannic Swiss also perceive it as central to their identity. As Watts points out, "for the German-speaking Swiss, the dialect functions [… ] as a badge of Swissness, an emblem of 'belonging' to Switzerland, which is more powerful than any other emblem" (1999: 75). Hence, the local dialect "serves as one of the most powerful markers if not *the* most powerful marker, of local, rather than national identity" (Watts 1999: 69, emphasis in original). Consequently, most Alemannic Swiss consider their dialect of Swiss German — rather than Standard German — to be their mother tongue (Watts 1999: 72) and may even classify Standard German as a 'foreign language' (Watts 1988: 328). In fact, many German-speaking Swiss do not enjoy speaking Standard German and, given the choice, "would far rather communicate in English to a foreigner than in standard German" (Watts 1999: 75).<sup>5</sup>

Since dialects are so central to the Swiss identity, they are an essential part of integration in Swiss society. As Lüdi points out, Swiss German is constitutive of "the personal and group identity of Alemannic Swiss. The use of dialect is one of the strongest in-group signals. As a consequence, dialect knowledge takes on a kind of test function in the evaluation, by the host community, of the [… ] migrant's will and ability to integrate" (1996: 111). Hence, a working knowledge of Swiss German, probably even more so than Standard German, is essential for immigrants if they want to integrate into Swiss society. This is supported by Gonçalves' study on bilingual couples in Switzerland (see section 3.3.2, "Bilingual couples"), for whom the "cultural capital" is felt to be represented by the local dialect rather than by Standard German (2013: 142). Gonçalves remarks that, while the Federal Office of Migration lists knowledge of a national language as an important criterion for immigrants who intend to stay for an extended

<sup>5</sup> Of course, this can partly be attributed to the high prestige that English carries in Switzerland. In my experience, the Swiss only prefer English to Standard German if their interlocutor's mother tongue is English or the latter is more fluent in English than in German. Whenever possible, many Swiss people try to communicate in their interlocutor's mother tongue.

period of time, immigrants actually need to learn a local dialect in order to integrate:

According to the Federal Office of Migration, learning a national language is a key factor for 'successful' integration. My data, however, indicates that learning standard German does not give foreign nationals a sense of being integrated, but perpetuates the feeling of 'difference' and 'foreignness'. In fact, understanding and speaking the local dialect rather than standard German contributed to individuals' sense of belonging and integration. (2013: 163)

Gonçalves concludes that, because Swiss German dialects are used in social interaction, speaking Standard German is not enough to guarantee successful integration (2013: 196). Instead, for the Anglophone immigrants in her study, "'successful' integration meant being competent in the local Bernese dialect rather than standard German. In fact, all of the participants who attempted to learn standard German were met with frustration [… ]" (2010a: 258). Thus, the local variety of Swiss German takes precedence over Standard German in Switzerland in terms of integrative potential as well as communicative effectiveness and cultural prestige.

## **2.4 English in Switzerland**

Like many other nations, "the Swiss are subjected to an ever-swelling tide of English language, flooding over them in the form of pop music, films, satellite television and internet sites" (Murray 2003a: 98). The high prestige of the language and its omnipresence in entertainment and the media have led to a number of English loanwords being taken up by local varieties. English loanwords are commonly used in advertising, but also by many Swiss, who feel that it makes their language sound "more up-to-date or sophisticated" (Murray 2003a: 100). As Murray points out, Swiss-German Anglicisms appear in a range of registers and sociolects, for instance in youth language or the business/work register (2003a: 100). Yet the influence of English is not limited to individual loanwords in specific contexts. Rather, the language "has found its way, both lexically and syntactically, into the oral and written language usage of speakers of Swiss German, German, French and Italian in Switzerland" (Watts 2001: 304). This should be kept in mind when one analyses the language mixing behaviour of the bilinguals in this book, in order to avoid confusing established loanwords with nonce borrowings.

Of course, many Swiss have mixed feelings about the influence of the English language, as can be seen in the following quotation:

For some, embracing English means catching up with the future and assuring Switzerland a leading place in the 'knowledge society' of the twenty-first century. For others, letting in the Trojan horse of English means opening the gates to a worldwide phenomenon that will ultimately destroy part of the linguistic and cultural heritage of multilingual Switzerland. (Murray 2003a: 105)

Watts paraphrases a 1989 report on the language situation in Switzerland in which English "is presented as an ominous presence threatening the harmonious coexistence of the four national languages within Switzerland" (2001: 308). This discourse seems to persist even though linguists postulate that the use of English as a lingua franca in Switzerland is unlikely to become the norm and should not be presented as a threat (Andres and Watts 1993: 126). Given this ambivalence that many Swiss feel with regard to the language, it is not surprising that the introduction of English at primary school level is such a controversial issue (see section 2.2, "Multilingualism in Switzerland").

English is also used frequently in a private and professional context among the Swiss population, though there is great local variation in the frequency of its use. English is spoken most commonly as a home language in the major urban regions and economic centres such as Zurich, Geneva and Basle, as well as in the tourist regions (Lüdi and Werlen 2005: 65). With regard to the places of residence of the couples in this study, this means that there are considerable differences in the number of residents who speak English as a main language in these areas. A relatively large number of residents (8.2% of the population) consider English to be (one of) their main home language(s) in the Zurich area, as opposed to only 2.8% in the St Gall area. The figures for the areas where the other participants in this study reside lie somewhere between these numbers.<sup>6</sup> The prominence of English is notable in Zurich, and, in my personal experience, it is the exception rather than the norm to take a ten-minute tram ride through the centre of Zurich without hearing someone converse in English.

English is not only used to communicate with native speakers, it also serves as a lingua franca in many situations, to communicate both with Swiss from different linguistic areas and with foreign visitors. Moreover, English is used regularly in the work environment. When asked which language(s) they speak

<sup>6</sup> These figures are based on the 2012–2014 national census by the Swiss Federal Office of Statistics. Respondents could indicate up to three main languages (BFS 2016a). See also Lüdi and Werlen (2005: 18) for an overview of the distribution of native speakers of English across the country according to the 2000 census.

regularly at work, 11.5% of the population reported that they use English. In contrast, 41.8% speak Swiss German at work, 21.2% Standard German, 18.5% French, 5.5% Italian and 0.25% Romansh (BFS 2016a). This shows that English "is an important language, but one of several important languages" (Murray 2003a: 88). At the same time, these figures reveal that English is spoken more frequently at work than Italian or Romansh, even though the latter are national languages. The percentage of people who speak English at work depends greatly on the region and its economic situation, as well as on the profession of the person.

Because of its important role in a variety of areas like business, finance, edu‐ cation and tourism, English is held in high esteem by the Swiss and many people strive to speak the language well. Consequently, it is not only the symbolic value of English that is very high in the Swiss marketplace, but also the economic value of the language. This was confirmed in an economic study, which demonstrated that English skills have a high labour market value in Switzerland, and "can be associated with remarkably high and statistically robust wage premia [ranging] from 12% to 30%" (Grin 2001: 65). These returns are dependent on the language region and the sector, as well as on the individual's gender, as the premia appear to be higher for men (Grin 2001: 73–74).<sup>7</sup>

Finally, it should be noted that, since English is an important and prestigious global language, bilingual native speakers of English are often termed elite bilinguals. This is particularly the case if they are economically privileged and if their second language is a world language (Boyd 1998: 32), or at least important in the local community. These factors tend to facilitate the integration of bilingual immigrants. Elite bilinguals are less likely to be subject to discrimination or prejudice than other bilinguals; they themselves as well as the local community tend to view their bilingualism positively, and their children often become highly proficient bilingual speakers (Boyd 1998: 32). This was confirmed in Boyd's study of the language practices of four immigrant groups living in Denmark, Finland and Sweden. When these groups were compared, "there turned out to be significant differences in their overall amount of use of majority and minority language between the Americans and the Finns on the one hand and the Turks and the Vietnamese on the other. The former two groups tend to use more majority language and the latter two more minority language"

<sup>7</sup> These figures are based on data from the mid-1990s and might look different now, but to my knowledge, no current studies are available that examine the economic value of English from the perspective of the individual. A later study (Grin, Sfreddo and Vaillancourt 2010) addresses the value of English (among other languages) for private companies in the manufacturing sector.

(1998: 47). Boyd also discovered that the children of the American respondents reached a high level of bilingual fluency, and that these elite bilinguals tended to "'disappear' in the host society" (1998: 31). There have been no comparable studies on native speakers of English in Switzerland, but it can be assumed that their situation is relatively similar. In all probability, the label of elite bilinguals may be applied to all English-speaking partners in this book, which means that they and their local community are likely to have a positive attitude towards their bilingualism.

## **3 The communication of bilingual, bicultural couples**

## **3.1 Introduction**

Bilingualism is often, though by no means always, connected to biculturalism. The couples analysed in this study are bilingual as well as bicultural; how‐ ever, there are, of course, also bicultural couples who speak the same or a similar language (e.g. Francophone Swiss-Canadian, Anglophone South Af‐ rican-British) or bilingual couples who share a (similar) culture (e.g. Frenchand German-speaking Swiss). Moreover, culture cannot always be separated from language. In fact, it has been proposed that the manner in which we speak may be influenced by an underlying "cultural script", which, in turn, is based on cultural norms and practices (Wierzbicka 2004: 98). These scripts help us "capture background norms, templates, guidelines or models for ways of thinking, acting, feeling, and speaking, in a particular cultural context" (Goddard and Wierzbicka 2004: 157). Consequently, it has been suggested that we cannot but "interpret speech through the 'filter' of our native language/culture" (Gass and Varonis 1991: 130; cf. Sapir 1958 [1929]: 69).

Yet even though culture and language are interlinked, and biculturalism and bilingualism often overlap, I have decided to discuss previous work on the two areas separately, because most studies seem to focus on one or the other. Thus, I first outline research on cross-cultural couples, and then discuss a number of studies on bilingual couples. The present chapter only includes general aspects with regard to living in a bilingual, bicultural relationship. Previous research in fields which are explored more extensively in my analysis is discussed at the beginning of the corresponding chapters.

## **3.2 Bicultural couples: Background and challenges**

#### **3.2.1 A word on culture and biculturalism**

Culture is hard to grasp, yet it permeates many aspects of our life and being, even though we may not always be aware of it. Grosjean highlights the different areas of our lives that are interconnected with culture:

Culture is the way of life of a people or society, including its rules and behavior; its economic, social, and political systems; its language; its religious beliefs; its laws; and so on. Culture is acquired, socially transmitted, and communicated in large part by language. (1982: 157)

Spencer-Oatey emphasizes the complexity of culture, defining it as "a fuzzy set of attitudes, beliefs, behavioural conventions, and basic assumptions and values that are shared by a group of people, and that influence each member's behaviour and each member's interpretations of the 'meaning' of other people's behaviour" (2000a: 4). Culture may not always be apparent; rather, it "is mani‐ fested at different layers of depth, ranging from inner core basic assumptions and values, through outer core attitudes, beliefs and social conventions, to surface-level behavioural manifestations" (Spencer-Oatey 2000a: 4).

Given the complexity and elusiveness of one single culture, the situation becomes even more complicated when multiple cultures come into contact. If this contact is extended, as in the case of cross-cultural couples, this often leads to biculturalism, "the coexistence and/or combination of two distinct cultures" (Grosjean 1982: 157). I consider all couples in this book to be bicultural, independent of the extent to which they combine their two cultures as a couple. It should be noted, though, that there may be vast differences with regard to the level of acculturation of both partners to the other's culture.

#### **3.2.2 Bicultural couples**

While bicultural and/or binational couples used to be the exception rather than the norm in Western countries, they have become much more common in recent decades. In Switzerland, for example, intermarriage rates between Swiss and foreign nationals have increased substantially over the past 35 years. In 1980, only 18.0% of the Swiss population married a foreign national; this figure had risen to 42.2% by 2015 (BFS 2016d). Exogamous marriages are a little more common among Swiss men than women, as 54.8% of all binational marriages in 2015 (compared to 60% in 1980) were between a Swiss male and a foreign national. Similarly, demographic evidence from Germany and the USA provided by Piller suggests that international intimate relationships have also increased dramatically in these countries over the past 30 years (2007: 343). Piller argues that globalisation has facilitated this trend, since there is increased international mobility, data flow and cultural exchange (2007: 344). It is also possible that "advances in travel and education, military and political incentives, the introduction and effects of broader civil rights [… ] have all contributed to the increase of marriage between people of different faiths, cultures, nationalities, and races" (Perel 2000: 178).

Scholarly attention to cross-cultural couples has been moderate in the past. Before the 1990s, a number of studies were carried out on the experiences of cross-cultural, cross-national, interfaith or interracial couples, but the majority of them were small-scale studies on couples that consisted of a Western and a non-Western spouse (see Cottrell 1990: 152 for an overview of 33 studies on intercultural marriages). Some studies on bicultural couples have attempted to gain insight into their relationships by documenting the experience of the women (often the "foreign spouse") in these relationships, rather than looking at the relationship as an entity. An example of this is Khatib-Chahidi, Hill, and Paton's (1998) study on 20 women in mixed marriages, which addresses the question of motives for entering a cross-cultural relationship. The subjects were nationals of nine different countries, who had all met and married their foreign spouse in their own country. Khatib-Chahidi, Hill and Paton were interested in ascertaining whether the intermarriages of these women were a random choice, or whether they had selected their spouses, "consciously or unconsciously, ac‐ cording to certain predetermined criteria" (1998: 49). Many of their respondents reported that they married "out" because they felt marginalized within their families or in society, or because they wanted to escape the gender roles of their own culture (1998: 57–62). At the same time, many of the women emphasized that they had a lot in common with their spouse, and that they liked that their partner was different from their countrymen, whom they portrayed as "boring" (1998: 52–53). In addition, a large number of the interviewees "reported significant interaction with foreigners before meeting their husband" (1998: 57), which suggests that they were open to engaging with somebody with a different cultural background.

The study shows that marrying someone from a different culture may be seen as a means of escaping one's own environment, and that the differences are often seen as interesting, enriching or attractive. At the same time, differing cultural practices and ideologies can also be a source of conflict (see also section 3.2.3, "Cultural challenges"). Breger and Hill put it follows:

By definition, culturally mixed marriages present those involved with a wider palette of cultural practices than culturally homogenous marriages, including such issues as gender roles, child-rearing, mores, language and general lifestyle by which to shape their lives. Sometimes, there may be little awareness of difference, or indeed the differences may be minimal [… ] On the other hand, in the process of everyday life, differences can become clearer, forming a highly charged minefield of conflict, or a source of enriching diversity, or even both. (1998: 19)

Such issues and challenges for bicultural relationships have been highlighted and examined by a relatively large body of research. Older studies in particular often assumed that bicultural relationships are problematic, and that there is a "higher risk of communication difficulties, marital dissatisfaction, [and] divorce" (Killian 2009: xix). This view has only started to change in recent years, as "cross-cultural couples have increased in both numbers and social acceptance, and there is now a growing awareness of how little we really know about them" (Killian 2009: xix).

#### **3.2.3 Cultural challenges**

The specific challenges faced by intercultural, bilingual couples depend very much on the linguistic, cultural and social backgrounds that are involved. As Rosenblatt observes, "[e]ach particular combination of cultures, social class, location where the couple lives, and so on creates a set of challenges, issues, and themes that might be quite different from what couples with other combinations of characteristics experience (Cottrell, 1990). In this sense, every intercultural couple needs its own theory" (2009: 3). Therefore, research on bilingual couples should take into account the specific socio-cultural background of the partners, which it sometimes fails to do (Killian 2009: xviii). With regard to the current study, it should also be noted that English-German-speaking couples are often viewed as unproblematic due to the proximity of the two cultures, as they are thought to share a Western or European heritage (Piller 2002a: 6). There are, however, some issues which seem to surface again and again even in close intercultural relationships, which will be discussed in this section.

Since culture is so complex and difficult to separate from other factors, bicultural couples may not recognize that some of their differences are rooted in their cultures; conversely, they may attribute personal differences to their variant cultures when they are actually due to other factors. According to Perel,

conflicts often turn out to be more connected to culture than couples realize although naturally intrapsychic sources of conflict are generally simultaneously at work. Other couples, on the other hand, often err by focusing exclusively on their cultural differences as the source of all pain, thereby eschewing taking responsibility for their actions. (2000: 196)

Indeed, not all issues potentially faced by bilingual, bicultural couples can be traced back to the fact that they are from different cultures or speak different languages. There may also be communication differences based on their social background, age, family background, or gender. The role of gender appears to be particularly prominent, and it has often been suggested that men and women differ in their styles of talking as well as in the role and importance they attribute to talk in their relationship (e.g. Cameron 1997; Tannen 1986). Thus, communicative difficulties may arise because of a combination of language, culture, gender and other factors.

#### **3.2.3.1** *Socio-cultural practices, politeness and stereotypes*

Important aspects in which cross-cultural couples are likely to differ — at least to some extent — are their socio-cultural practices, rituals, and what they consider polite or appropriate behaviour. Even in Western countries, where customs and rituals may not be quite as important as in more traditional cultures, many habits are culturally rooted. Instances of such traditions may be the British pub culture, or the Swiss Sunday brunch, which is practiced by many families in Switzerland. Gonçalves (2013: ch 7) mentions a number of socio-cultural practices which the Swiss-Anglophone couples in her study consider to be Swiss, such as not wasting food, eating (high quality) bread, cooking proper meals, or using public transport. Moreover, there are a number of everyday rituals or social customs that are considered very important in Switzerland, such as waiting for everyone to sit down at the table and saying *en Guete* ('enjoy your meal') before starting to eat, or proposing a toast (repeatedly) before consuming alcohol. In contrast, Americans tend to be more accustomed to hugging each other, or celebrating Thanksgiving, while English people may find it important to queue properly. Not complying with such cultural practices may be considered rude. Thus, people from different cultures tend to have different ideas of what appropriate behaviour or polite manners are, and they may also have different ways of expressing politeness verbally.

Another area of conflict may be the stereotypical ideas partners have about each other's culture, and expectations resulting therefrom (Rosenblatt 2009: 7). Stereotypes and expectations are likely to change if there is extended contact between cultures, but such contact "does not necessarily lead to a better understanding. On the contrary, it may reinforce mistaken judgments of the other's intentions, and increase expectations that the other will behave a certain way" (Tannen 1981: 225). Of course, this can be true for monocultural couples as well, but such mistaken expectations may be more marked when there are larger differences between the partners, and when cultural stereotypes are involved. This becomes evident in Piller's work on bilingual, cross-cultural couples, where national stereotypes often created conflicts between the partners (see section 3.3.2, "Bilingual couples"). She remarks that, in her research, "[t]he majority of instances of conflict between the partners occur when one partner 'gets carried away' with their chance to produce a list of stereotypes about their partner's national identity" (2002a: 210). This may be one of the reasons why numerous participants in her study claimed to be exceptions to their traditional national stereotypes (2002a: 211). Overall, however, it can be expected that preconceived notions of national characteristics are more complex and more differentiated among closely related cultures than among more distant ones, as the former tend to be in closer contact.

#### **3.2.3.2** *Religiousness, ideologies, values and gender*

One aspect that might lead to conflicts in cross-cultural couples is that they often have different religions or denominations, or that they at least are religiously observant to differing degrees. The home countries of the subjects in this study are all traditionally Christian, and the majority of the population in all of the countries are affiliated with a Christian religion (England 59.3%, Australia 61.1%, Switzerland 69.9%, United States 78.5%, Northern Ireland 82.3).<sup>1</sup> Between the countries, however, there are considerable differences with regard to the importance of religion and individual religiosity. In a survey, only 13% of the Swiss population stated that they would "definitely" consider themselves a religious person, and a mere 10% of the population claimed that they attended a church service or similar (almost) every week.<sup>2</sup> In contrast, 53% of the American respondents stated that religion was "very important" to them, and another 27% rated it as "fairly important" in their lives (Gallup 2014). Moreover, 37% of all Americans reported that they attended church (almost) every week; however, it should be noted that it is possible that many people over-report their church attendance in the USA (Grossman 2014), and weekly attendance rates may actually be lower than 22% (Hadaway and Marler 2005: 307). Nonetheless, these high numbers indicate that church attendance is something to which many American people aspire. The reported level of religious observance is comparably high in Northern Ireland, where 30% of the population claim that they go to church weekly (Ashworth and Farthing 2007). In Australia and England, people appear to be observant to a similar degree as in Switzerland, with 14% of the English (British Social Attitudes 2014) and 8% of the Australians (McCrindle 2013) attending a church service every week.

<sup>1</sup> These figures were taken from the British Office for National Statistics (ONS 2012), the Central Intelligence Agency (CIA 2016), the Swiss Federal Bureau of Statistics (BFS 2016b) and the Northern Ireland Statistics and Research Agency (NISRA 2011).

<sup>2</sup> These figures, which are based on a 2014 survey, include all religions, and would be lower if only Christian denominations were considered (BFS 2016c).

Such differences with regard to religiousness or denomination between the partners can create tensions within their relationship. Moreover, religious views are often linked to other values and ideologies that may differ correspondingly. For instance, strongly religious cultures are inclined to be more conservative than less religious cultures, which seems to apply to the USA when compared to most of Western Europe. Such differences are also reflected in the stereotypes that both cultures have of one another, as Europeans tend to see Americans as conservative, while the latter view the former as liberal. Gonçalves' study on bilingual, bicultural couples supports this (see section 3.3.2, "Bilingual couples"), as some of her interviewees "discursively construct and position Americans as closed and strict compared to the Swiss as open and liberal" (2010: 245). A further characteristic that is linked to the American religious and historical background is their "effort optimism", which refers to a "desire for assertiveness, action, and hard work" (Perel 2000: 187). While the Swiss are also known as hard workers, in my personal opinion, they tend to be less assertive than Americans.

Another challenge commonly faced by cross-cultural couples seems to be that one partner's culture is more individualistic, while the other's is more family-oriented (Rosenblatt 2009: 15). This appears to be an issue faced particularly by couples consisting of a Western and a non-Western partner, but European cultures also differ in their orientation towards family and individualism. Hofstede's (1980) study of individualism versus collectivism in 40 cultures indicated that many English-speaking countries are very individu‐ alistic. Thus, the USA (91), Australia (90) and Great Britain (89) ranked the highest of all 40 countries in terms of individualism, while Switzerland (68) and Ireland (70) ranked lower — though still higher than the average (m = 51) of the countries (1980: 158). Fundamental differences in the partners' family orientation sometimes emerge when they are faced with "life cycle events such as pregnancy, childbirth [… ] and deaths" (Rosenblatt 2009: 17). The same goes for ideas concerning child-rearing and parenting standards, between which there can be considerable differences based on the parents' cultural background (Rosenblatt and Stewart 2004: 52).

Other ideologies which often differ are those surrounding gender and love. What is considered appropriate behaviour for men and women depends to a large extent on the culture in question. Because the partners "may expect gender patterns from the other that do not fit the other's dispositions, beliefs, and ways of acting", some believe that "[g]ender is one of the most challenging areas for intercultural couples to negotiate" (Rosenblatt 2009: 9). Pavlenko studied the role of gender in multilingualism by analysing cross-cultural autobiographies and oral narratives about second language learning. According to Pavlenko, language learning is a gendered performance, and bilinguals tend to develop strategies of assimilation or resistance to the gender roles of their L2 (2005: 141). The accounts she analysed indicate that bilinguals may perceive their new environment as gender-free, because they escaped undergoing the gender socialisation processes of childhood in their second language. Yet gender socialisation at a later age can also pose difficulties, particularly in the context of interpersonal relationships, and "[i]ntimate relationships and friendships surface time and again as one of the most difficult areas for negotiation and an authentic performance of gender" (2005: 151).

Linked to gender ideologies are also ideas of romantic love, which may diverge in different cultures. Hence, certain aspects such as courtship rituals, the expression of love, but also expectations with regard to what love is or how it is expressed, will be strongly influenced by each culture. This was demonstrated by Ting-Toomey, who questioned 781 participants from Japan, France and the USA on aspects such as their views on commitment, self-disclosure and conflict behaviour in intimate relationships. The results of her study indicate that "specific cultural norms and rules influence the perception and the articulation of intimate expressions" in these three countries (1991: 41). Similarly, Billig argues that, while "[t]here are conventional signs of 'love', [… ] these signs may differ from culture to culture; and what counts as love may change within one culture over the course of time" (1999: 189). At the same time, however, it seems that European-based cultures also tend to share certain common denominators, most prominently, a "love ideology that includes a heritage of courtly romantic love" (Sprecher et al. 1994: 350). This ideology includes a number of notions such as "love at first sight, there is only one true love, true love lasts forever, idealization of the partner and the relationship, and love can overcome all obstacles" (Sprecher et al. 1994: 352–353). These studies suggest that partners from similar cultures, such as the couples in this study, are likely to have a similar concept of romantic love, but they may nevertheless diverge in their manner of expressing love.

#### **3.2.3.3** *Expatriate situation*

The new living situation that comes with having a partner from a different country is another challenge bicultural couples have to face. In the case of the couples in this study, one of the partners elected to leave his or her home country and move to a new cultural, linguistic and social environment. These partners may be disadvantaged socially, as they are likely to have fewer friends — at least initially — or less established relationships. In addition, they do not generally have their relatives close to hand and cannot depend on them for support. Foreign partners are often at a disadvantage linguistically, as they are not able to use their mother tongue or may be forced to use their second language in encounters with locals, where lacking language competence may be an issue or drawback. Not being a native speaker of the community language may also lead to a professional disadvantage, or reduce the individual's choice of jobs, which is often already limited due to their foreign qualifications. In addition, immigrants may also be at a disadvantage culturally, and be obliged to accustom themselves to many elements of their partner's culture. Rosenblatt explains the cultural struggle of expatriates as follows:

The cultural territory a couple is in can make an enormous difference in the resources available to the partners. Living where one partner's language, preferred foods, religion, modes of dressing, holidays, types of recreation, etc., are dominant provides enormous resources to one partner and makes him the more competent person in getting around in all sorts of ways. (2009: 12)

Having to deal with these things on a daily basis may lead to psychological problems such as homesickness, insecurity, anxiety, marginalisation, or feelings of incompetence (Rosenblatt 2009: 12). This was also found by Breger (1998), who documented the experiences of 18 mixed couples as well as members of a cross-cultural women's group and a foreign student club in Germany. Many of the subjects in her study reacted to such difficulties by

withdraw[ing] from what they perceived to be a hostile society, avoiding contact with all groups. This can lead to isolation and depression, and the self-reinforcing feeling of not coping [… ]. Many people, especially the foreign *wives*, experienced such depression, which formed a central recurrent topic at the women's group. (Breger 1998: 149, emphasis in original)

Such feelings of isolation may have been reinforced by the fact that the children of these couples were rarely balanced bilinguals, but were inclined towards the community language. As Rosenblatt remarks, children's language practices often "come to fit the country in which they are living, which further margin‐ alizes and undermines the parent who is a cultural and linguistic outsider" (2009: 12). Consequently, the partner in whose home territory they are living is at an advantage in many regards and is likely to be in a stronger position within the relationship.

Living in only one partner's home environment may create difficulties not only for the partner who emigrated, but also for the local partner, and for them as a couple. In some cases, the local partner has to take on tasks that his or her partner cannot perform, such as doing the taxes or dealing with authorities, banks, builders, or teachers, or has to lend assistance when there is a problem involving communication or a lack of cultural knowledge. Thus, a considerable amount of work can fall to the native speaker. In addition, he or she might have to lend his or her partner extensive emotional support if the latter does not have a large social network. Furthermore, the bicultural couple may struggle to fit into the local community, as can be seen in Breger's study. Many of the mixed couples she examined had difficulty in being accepted into both local and expatriate communities, and therefore joined or formed societies for mixed couples (1998: 147). The expatriate situation can thus be challenging for both partners as well as their relationship, particularly if the couple is also bilingual, and if the foreign partner is not fluent in the community language.

#### **3.2.3.4** *Overcoming cultural challenges*

Even though there are a number of challenges that bicultural couples may face, research also suggests that couples can develop various means of dealing with their situation, and that biculturalism can ultimately have a positive effect on a relationship. Bustamante et al. conducted extensive ethnographic interviews with five intercultural couples in order to determine potential culture-related stressors, as well as coping mechanisms the couples had developed. In these interviews, the following six main coping mechanisms emerged: "(a) gender role flexibility, (b) humor about differences, (c) *cultural deference* or a tendency to defer to the culture-related preferences of a partner, (d) recognition of similarities, (e) *cultural reframing*<sup>3</sup> [… ] and (f) a general appreciation for other cultures" (2011: 159, emphases in original). Similarly, the intimate cross-cultural couples in Ting-Toomey's research listed a number of advantages of living in an intercultural relationship, such as personal growth, greater diversity in their lifestyles, having a more profound relationship, developing complex cultural frames of reference and value systems, as well as "raising open-minded, resourceful children" (2009: 45). Hence, many couples feel that their intercultural challenges promote their growth as individuals and as a couple.

<sup>3</sup> Cultural reframing refers to "the development of blended values and expectations that [redefine] the intercultural relationship" (Bustamante et al. 2011: 154).

## **3.3 Bilingual couples: Background and challenges**

#### **3.3.1 A word on bilingualism**

Bilingualism has been defined as "the regular use of two or more languages" (Grosjean 1982: 10), although it can also be defined based on competence in the second language (L2). For individuals, I use the term *bilingual* to refer to the ability to speak two languages, which requires a certain level of proficiency and presupposes at least occasional use. When referring to bilingual couples, as opposed to bilingual individuals, I use the term to indicate that the partners have different mother tongues, regardless of both partners' level of proficiency. Thus, bilingual couples may not communicate in two languages with each other; instead, they may often speak one main couple language. This also means that, in theory, one of the partners might not actually be bilingual, despite being in a bilingual relationship. Moreover, it should be noted that, technically, many of the participants in this study are not just bilingual, but multilingual. Since my research only focuses on the two languages in which they are most fluent their mother tongue and their partners' L1 — I will refer to them as bilinguals, and to their partners' language as their L2, even if for some, this may not be the language which they acquired second. In contexts where I refer to any language (or several languages) other than a speaker's L1 (and not specifically his or her partner's language), I use the term LX.

#### **3.3.2 Bilingual couples**

While there is an abundance of research in the field of bilingualism, bilingual couples received little attention until the 1990s. As with research on cross-cul‐ tural couples, research on bilingual couples was conducted mostly from the perspective that they pose a problem (Piller 2002a: 19), and only in recent years have researchers begun to move away from the assumption that bilingual relationships are inherently problematic. In addition, most studies do not examine the manner in which bilingual couples talk to each other, but rather concentrate on one of the partners, most commonly the female one (see also section 3.2.2, "Bicultural couples"). An example of this is Heller and Lévy's study of 28 Francophone women married to Anglophone men in Ontario, Canada. The authors examined the women's language practices and their attitudes towards them, as well as their feelings about their own identities (1992: 20). Heller and Lévy found that the women reconstructed the French-English language border linguistically in their conversations, and that they did not necessarily assimilate to their partner's culture (1992: 39). Some of them had a strong focus on their Francophone identity, while others attempted to construct and combine two linguistic identities.

Probably the most extensive study on bilingual, bicultural couples was conducted by Piller (2002a). She examined 57 German and English-speaking couples living in a number of countries, based on self-recorded conversations or monologues, questionnaires, letters and/or focus group interviews. In her analysis, she considered a number of aspects, including the couples' reported language choice and mixing behaviour. Piller was concerned with the ideologies and other factors underlying these language choices (see section 5.2, "Factors influencing the language choice of bilingual couples", for a more in-depth overview), and studied the participants' reported language skills, processes of L2 acquisition and language ideologies, as well as their perceptions of their mother tongues. She discovered that many of the participants had a strong emotional attachment to their partner's language, a feeling that she calls "language desire" (2002a: 100; see also section 7.2.1, "Attraction in bilingual, bicultural couples").

Another focus of Piller's research is the construction of bilingual, cross-cul‐ tural couplehood and identity. She posits that, because bilingual individuals often feel or think differently depending on the language they speak, bilingual couples may also view, experience, or perform their couplehood differently. Like other identities, Piller argues, couple identity is not something fixed, but a hybrid construct that may be performed differently depending on the environment the couple is in (2002a: 11). Hence, "[c]ross-cultural couplehood is not a state of being, but an act of doing" (2002a: 2). Through detailed analyses of their speech, she demonstrated how the couples jointly constructed and performed their identity linguistically (2002a: ch 7). The couples did this on a conversational level, for instance by using their own private words, as well as on a discourse level, by reiterating "'founding myths' and stories they tell about themselves to account for their being together" (2002: 225).

Such linguistic strategies serve the partners to highlight similarities between each other and enable them to construct their couple identity, but they also discredit the prevailing idea of intermarriage as a problem (Piller 2002a: 189). The couples' reports showed that they had all come across negative opinions of intermarriage, and that "they [were] aware of a widespread framing of cross-cultural marriage as problematic" (2002a: 186). In analysing the couples' discursive strategies, Piller found that they used three major strategies to frame their relationships as unproblematic. Firstly, they contrasted themselves with other cross-cultural couples by highlighting the cultural proximity between English and German-speaking countries. They thus created a discourse that presented "European or Western culture as a homogenous cultural sphere", and, in doing so, depicted their relationship as less problematic than other bilingual relationships (2002a: 194). Secondly, the couples sometimes "downplay[ed] national identity and focus[ed] on other aspects of their identity", appealing to non-national identities instead (2002a: 189). They referred to common historical roots of their cultures, to shared values or interests, to similarities between their personalities, or to their "shared linguistic identity as bilinguals" (2002a: 199). This discourse of collective identities allowed them "to construct themselves as conventionally and appropriately similar" (2002a: 197). Thirdly, some of them used a discourse of post-national identities, for instance by presenting themselves as cosmopolitans, world citizens or fellow Europeans (2002a: 202).

The couples in Piller's study produced not only a discourse of constructing similarities, but also one of deconstructing differences or framing them posi‐ tively. Hence, many described their cultural differences as attractive, exciting or exotic, or framed their being together as their destiny. Others portrayed themselves as atypical of their cultural background or as exceptions from their national stereotypical norm (2002a: 205). Moreover, the couples emphasized that they could overcome cross-cultural difference through compromise and change. In fact, Piller states that the "discourse of change and compromise is unique in [her] corpus in its frequency, as it appears in every single conversation, without exception" (2002a: 213). Thus, the couples seem to feel that "love should and must transcend [their] differences" (2002a: 2). In addition to the couples' positive framing of their relationship, Piller claims that there is "no indication whatsoever in [her] data that the participants' relationships are in any way more or less problematic for them than they are for any monolingual and mono-cultural couple" (2002a: 186).

A particularly interesting characteristic of the couples' conversational style is their use of other-repetition and other-completion. According to Piller, "[t]he most salient feature of the transcripts [… ] is the enormous amount of repetition between speakers. All the conversations [… ] are characterized by high levels of repetition" (2002a: 228). Such other-repetition is generally seen as indicative of a collaborative floor, and is typical among people in a close relationship (2002a: 228). Piller examined the different functions of repetition and the situations in which repetition tended to occur, for instance as meta-linguistic comments, corrections, or as a consequence of word searches, word offers or mishearing. Bilingual conversations potentially provide more contexts for other-completion to occur, as there may, for instance, be more word searches (2002a: 235). Piller found that cases in which the partners finished each other's sentences were also frequent in her data. The couples appeared to value other-completion, as instances thereof were "almost always endorsed by the recipient" (2002a: 237). Both other-repetition and other-completion, Piller believes, facilitate the discursive performance of a joint couple identity (2002a: 242).

Like Piller, Gonçalves (2010a, 2010b, 2013) concentrated on the topic of iden‐ tity in her work on bilingual couples. She examined conversations with seven couples consisting of a native speaker of English and Swiss German, as well as two conversations with English speakers, each of whom was in a relationship with a Swiss partner, but who participated without their spouse. All of the couples resided in central Switzerland. Gonçalves analysed their speech with regard to their language ideologies and identity performances. Her main focus was not on the communication between the partners, or "couple talk", but rather on the female Anglophone partners' personal experiences in their partnership and their host country. In particular, she focused on the discursive construction of identities and on *doing* Swiss, by looking at the linguistic and socio-cultural practices that the participants had adopted or rejected (2013: ch 7). According to Gonçalves, people in such an expatriate situation "are constantly positioning and re-positioning themselves as certain types of individuals who perform or carry out particular local and socio-cultural practices within specific contexts" (Gonçalves 2010b: 76). To demonstrate this, she analysed a variety of linguistic resources used by the participants, "such as direct reported speech, prosodically marked utterances, pronominal use, code-mixing, and overt mentions of identity labels to position themselves and each other while simultaneously indexing various types of identities" (Gonçalves 2013: 192–193). Based on these features, she explored to what extent the identities of the English-speaking partners have shifted, and showed that some of them completely reject having an identity made up of both cultures, whereas others accept or even embrace a multiple or culturally hybrid identity (2013: 203).

In addition, Gonçalves investigated her participants' language ideologies with regard to Standard German and the local dialect, as well as their perceptions about their language learning process. She perceived a correlation between the non-native speakers' perception of their own skills in the local Bernese dialect and their notion of *having* a Swiss identity (2010b: 82). Moreover, Gonçalves found that the Anglophone partners in her study "position[ed] and align[ed] themselves with Swiss German native speakers and their respective language ideologies" (2013: 151). The participants' reports indicated that their process of language learning "was always inevitably connected to their own language ideologies and the continuous re-production of the 'ideology of the dialect' within their constructed social relations" (2013: 163). Based on her interviews, Gonçalves concludes that *doing* Swiss is "an intersubjective collaboration of social, cultural, gendered and linguistic practices and performances that emerge in social interaction" (2013: 165). Thus, language practices and ideologies may serve as a window onto the complex process of intercultural identity formation.

#### **3.3.3 Linguistic challenges**

The previous section has shown that more recent studies, such as those carried out by Gonçalves (2013) and Piller (2002a), have shifted their focus away from potential language problems arising in bilingual relationships, to topics such as the linguistic construction of identity. Miscommunication is not at the centre of my analysis either; however, in the process of analysing how the couples communicate, I inevitably also consider language problems. In the following, I will therefore delineate the most important linguistic challenges bilingual couples may face. For the most part, I will concentrate on difficulties among fluent L2 speakers, such as the couples in this study.

#### **3.3.3.1** *L2 learning and use*

Living in a bilingual relationship often means that the community language is a non-native language for one of the partners, which can be problematic if he or she is not proficient in this language. As Pavlenko remarks, learning or using a second language may lead to "feelings of anxiety, shame and embarrassment" if a bilingual's competence in the community language is not as high as desired (by the speaker or his/her environment) (2005: 32). If their level of competence is low, L2 speakers may also not be able to position themselves as mature, competent adults, or may be infantilized (2005: 218). All this is potentially applicable to the English-speaking partners in my study, most of whom were thrown into a new and complex linguistic environment without any previous knowledge of the language. Their situation is aggravated by the fact that Swiss German is not easy to learn due to its many varieties and the diglossic situation in Switzerland. Since many Swiss would rather speak English than Standard German to a foreigner (Watts 1999: 75), learning the language can be difficult and frustrating. This was confirmed by many participants in Gonçalves' study, who expressed frustration with their L2 language skills and the language situation in Switzerland, at least initially (2013: 151). Moreover, since many people in Switzerland view their dialect as central to "being Swiss", being able to speak the local variety is crucial to the Anglophone partner's integration (see section 2.3, "Diglossia and the ideology of dialect").

Being in a bilingual relationship usually means that at least one of the partners has to speak a language that is not his or her mother tongue, which may be a challenge for both partners as well as for their relationship. For one, the L2 speaker might be in a weaker position in discussions and arguments, and thus have less control within the relationship. As Piller puts it, "[i]n the linguistic construction of reality, power may also accrue to a person through being an undisputed expert manipulator of a code, a native speaker" (2002a: 142). This is particularly the case if the non-native partner is not proficient in the relationship language. In addition, not speaking one's mother tongue in one's family environment may also be perceived as a loss of cultural identity. Breger and Hill argue that the language in which the partners in a bilingual relationship decide to communicate "can be symbolic of the extent to which each partner is prepared to forego her or his cultural background and incorporate new elements" (1998: 21). This can pose a problem if the relationship language is the community language, so that one partner does not speak his or her mother tongue anywhere. As Piller remarks, "[i]n intercultural relationships the partner in whose native country the couple live is clearly privileged: legally, economically, and usually socially, too. [… ] Being a foreigner and having to use a non-native code places a person in a doubly marginalized position" (2000: section 4.3, para. 1).

In the case of the couples whom I interviewed, the situation is mitigated by their choice of couple language and the non-native partners' proficiency in this language. Since the Swiss partners are fluent in the relationship language, the native speaker might only have a slight advantage (or even none at all) in discussions or negotiations. Furthermore, potential issues are balanced out to some extent by the fact that their (main) relationship language is not the community language. This means that while one partner has the (possible) advantage of being able to use his or her mother tongue at home, the other one is able to use his or hers outside of the home. However, it should also be noted that choosing a language other than the community language as the main relationship language can be a disadvantage in the long run for the immigrant partner, since insufficient proficiency in the community language can limit access to employment (Piller and Pavlenko 2007: 18). Thus, while the speaker of the non-community language may benefit from using his or her native tongue within the relationship, this might prevent him or her from becoming fluent in the community language, and thus perpetuate his or her situation in the community as a linguistic outsider (see also section 5.2.2, "Community language").

#### **3.3.3.2** *Linguistic differences and misunderstanding*

When looking at communicative difficulties experienced by fluent bilinguals, we should bear in mind that it is often not a matter of whether one under‐ stands or not; rather, there are different grades of understanding. We can distinguish between misunderstanding and incomplete understanding; incom‐ plete understanding in itself may be divided into partial understanding and non-understanding (Gass and Varonis 1991: 124). The causes for communicative breakdowns are also varied, and may be caused by the speaker (misstatement) or be due to the listener's failure to understand (misinterpretation), or both (Banks, Ge and Baker 1991: 106). In the case of the couples in this study, non-understanding can be expected to occur only rarely due to their language proficiency, while instances of partial understanding or misunderstanding may be more frequent. Nonetheless, the latter may be particularly challenging because they are not immediately obvious. As the participants in this study are very fluent in their couple language, only few misunderstandings due to insufficient knowledge of grammar and vocabulary are to be expected, and I will not discuss such issues here.

However, there seem to be a number of other language-specific characteristics and pragmatic conventions that can lead to misunderstandings or other issues, even when fluent L2 speakers are involved. For instance, misunderstandings can arise in the context of the formulaic use of language and idioms, which appear to be difficult for non-native speakers (Gumperz and Tannen 1979: 315). A further challenging aspect can result from conversational conventions, which may vary even between speakers of different varieties of the same language, or between languages that are fairly closely related. Thus, House describes a dinner conversation between an American and a German, who disagreed on aspects such as turn-allocation, or the question of which topics are appropriate in certain situations (2000: 157). This demonstrates that speakers of different mother tongues may not have the same expectations with regard to appropriate linguistic behaviour in a given situation, especially when it comes to speech acts such as apologies, congratulations, condolences, complaints, offers, requests, or expressions of gratitude. For instance, there can be differences with regard to when an apology is expected, how such an apology is framed and formulated, what function it has and what the appropriate response to it is (Spencer-Oatey 2000b: 42). Similarly, complaints may be used with a different purpose in different cultures — as a "game of sharing complaints" or a "game of one-upmanship" (Tannen 1986: 47). Moreover, abundant or frequent use of expressions of gratitude or positive emotions (as is customary among Americans) may seem over-the-top and insincere to people from a culture where this is less common. In turn, people who are not as expressive may be viewed as unenthusiastic or boring by people from a more expressive culture.

Related to this is also the preference for indirectness or directness, which is, to some extent, culturally determined. Cross-cultural studies suggest that the conversational style of Germans is often perceived to be very direct.<sup>4</sup> Thus, House claims that "German speakers tend to interact in many different situations in ways that can be described as more direct, more explicit, more self-referenced and more content-oriented" than Americans (2000: 163). Simi‐ larly, Byrnes believes the American conversational style to be more indirect than the German conversational style and argues that "in German style, there is greater emphasis on the information-conveying function of language, as compared with its social bonding function" (1986: 200–201). Such differences may be particularly salient in the different approaches to conflict within the two cultures, as Oetzel et al. conclude from a cross-cultural comparison on facework in conflict:

The German [conflict] style is direct and confrontive. It is important to discuss issues thoroughly and completely. Discussions in Germany focus on facts and sorting through facts. [… ] U.S. Americans tend to remain calm. [Their style] focuses on talking about ideas in a calm manner in order to come to a mutually acceptable resolution. U.S. Americans often view Germans as being too blunt during conflict, while Germans view U.S. Americans as unwilling to engage in serious conflict. (2001: 253)

While, in my opinion, the Swiss are not generally as direct as the Germans, and value compromise, they may still be perceived as direct by native speakers of English.

Since the main conversational aim — conveying information or creating a rapport — appears to differ between speakers of different languages, they might also disagree with regard to the importance that is attributed to small talk. There is a persistent stereotype that Germans refuse to or are unable to engage in small talk, whereas Anglophone people are considered to be experts of small talk (Bubel 2006: 245). This seems to be the case in House's study of two reported interactions between native speakers of German and English. In a retrospective interview after a joint dinner, an American participant reported that he felt that Germans cannot — or do not want to — engage in small talk; he believed that, for German speakers, conversations always have to carry some weight (2000: 155). Based on her data, House arrives at the conclusion that

<sup>4</sup> However, Tannen notes that especially Americans "tend to associate indirectness with dishonesty and directness with honesty, a quality we see as self-evidently desirable" (1986: 55).

many misunderstandings arise because the importance attached to small talk within the respective cultures differs (2000: 161). Nevertheless, Bubel found in her analysis of phone conversations between a British sales executive and his German customers that, while it was mostly the English speaker who initiated small talk, the German customers, for their part, usually participated (2006: 254). These divergent reports could be seen as an indication of a situational dependence of small talk in both cultures.

#### **3.3.3.3** *Overcoming linguistic challenges*

The overview of previous work shows that there are a number of challenges that bicultural and bilingual couples may face. These are often seen as a disadvantage for the relationship, and they are the main reason why such relationships tend to be viewed as less stable and more problematic than monolingual, monocultural ones. However, couples may not only learn to cope with their bilingualism, but they may also view it positively and benefit from it. Couples can, for instance, attempt to overcome their linguistic differences to some extent by improving their language skills, or by being mindful of linguistic and pragmatic conventions. Furthermore, the partners may have very positive attitudes towards their bilingualism, or a particular interest in their partner's language and/or culture, and therefore enjoy speaking their L2 or being in a bilingual relationship. Due to their affinity to their partner's language and/or culture, they often see raising bilingual, bicultural children as an advantage, too. In addition, many bilingual couples consider their relationship to be more interesting than a monolingual relationship (Piller 2002a: 89). Such positive attitudes are especially common in the case of elite bilinguals, such as the participants in this study. My aim in this study is not to focus primarily on the challenges bilingual couples may face, but to explore positive aspects and particularities of their relationships as well.

## **4 Corpus and data collection**

## **4.1 Introduction**

In this chapter, I present background information about the interviewees and discuss the process of data collection and analysis. I begin with an overview of the interviewees and give a short biography of each couple. Following this, I discuss some limitations that come with my sample of participants. I then comment on the process of conducting interviews, as well as on the questionnaire that formed the basis of the interviews. As I decided to make precise transcriptions of the interviews, including a number of suprasegmental features, I describe the most important aspects of the transcription conventions. Moreover, I discuss the transcription process and the difficulties I encountered. Finally, I also comment on my method of analysis before moving on to the analysis itself in the next chapter.

### **4.2 Interviewees**

In order to examine the modes of communication employed by bilingual, bicultural couples, I interviewed ten heterosexual (cisgender) couples, each consisting of a native speaker of English and a native speaker of Swiss German. As I intended to examine fluent bilingual communication, I specifically looked for couples who had a high level of proficiency in their relationship language. To ensure that this was the case, I chose couples whose relationship language is English (although one couple uses both languages) and who have been together for a substantial period of time (on average 8.3 years).<sup>1</sup> The shortest relationship was 1 year at the time of the interview, the longest 24 years. The interviewees' ages range between 20 and 51 years; the majority of them are in their late twenties or early thirties (on average 30.15 years). On average, the couples met at a fairly young age (between 17.5 and 27, on average 21.85). Seven of the ten couples are married (not surprisingly, the three couples with the youngest average ages are not), and four of them have children. Furthermore, many of

<sup>1</sup> In contrast, Piller concentrates on the "early days" of cross-cultural intimate relation‐ ships, as she believes that "the more established a cross-cultural intimate relationship becomes, the rarer cross-cultural communication will be" (2007: 342).

the interviewees (16 of 20) have tertiary education or were pursuing a tertiary degree at the time of the interview.

In order to keep other linguistic influences to a minimum, I only interviewed partners who had grown up in a largely monolingual home. It needs to be borne in mind, however, that all of the interviewees have come into contact with other languages, especially the Swiss partners, all of whom studied several languages at school (Standard German, French, English and sometimes other languages such as Spanish or Italian; see also section 2.2, "Multilingualism in Switzerland"). Technically, the Swiss partners are all at least trilingual, although their levels of fluency in their L3 and L4 vary greatly. Regardless of which language the participants in this study feel to be their dominant language at present, or which language they currently use the most, I use the term *first language* (or L1) to refer to the language that they acquired first, which (in this case) is also the language they would call their "mother tongue". I consider their most fluent non-native language, which is also their partner's language in all cases, to be their *second language* (L2), rather than the language they acquired second (which would be French for most of the Swiss participants). All of the participants view Swiss German or English, respectively, as their mother tongue and dominant language. They all learnt their second language in their teens or in adulthood and are thus late (or sequential) bilinguals (see Harding and Riley 1986: 47).

All of the Swiss partners are from the German-speaking part of Switzer‐ land, though they speak different dialects. The countries in which the Eng‐ lish-speaking partners grew up are England (n=5), the USA (n=3), Australia (n=1) and Northern Ireland (n=1). Nine of the Swiss partners once stayed in an English-speaking country for an extended period of time, most of them for a year (1.05 years on average). Even though eight of the couples met abroad (one met in Switzerland, another one in an English-speaking online forum), all couples except for one were residing in the German-speaking part of Switzerland at the time of the interview.<sup>2</sup> On average, the English speakers had been living in Switzerland for 5.8 years. Most of the couples had also had a long-distance relationship for an extended period of time before settling in Switzerland (on average almost two years).

Finding ten couples who fit my very specific criteria and were willing to participate in an extensive interview turned out to be a difficult undertaking, and this is one of the reasons why the interviews took place over a period of several years (see section 4.3, "Interviews"). First of all, many people dislike

<sup>2</sup> While one participant (Sarah) was living in Switzerland at the time of the interview, her partner (Tim) was just visiting her for an extended period of time when the interview took place, but he was not living there permanently.

being recorded (which was also the case for potential subjects in Piller's [2002a] study), particularly in a non-native language. Some couples expressed worries about their relationship being the subject of the conversation, especially if they were having relationship problems. Moreover, many potential participants I was put in contact with had grown up bilingually themselves or turned out to be non-native speakers who use English as a lingua franca (ELF), and therefore did not fit my research profile. Several attempts at finding volunteers via social media and newsletters did not lead to any interviews with suitable couples. As a consequence, I decided to also include couples who I was acquainted with (which I initially wanted to avoid as I was a little concerned that this might influence the couples' answers and conversational behaviour). In a few cases, I knew one or both of the partners quite well; other participants were acquaintances from university or people I had met at social gatherings. Some I had never met prior to the interview, as we had only been in contact via email. In the case of the latter, I was given the couples' contact details by mutual acquaintances. Although I was sceptical at first, I believe that the interviews ultimately benefitted from the fact that there was always a personal link to the interviewees, direct or indirect, as they were all very open to sharing details about their relationships and to talking about personal or even problematic topics.

In the following, I present an overview of the most important biographic facts about all of the interviewees (Table 1), and then give a brief description of each couple and of their language choice and use. The couples are introduced in the order of the length of their relationship at the time of the interview (from shortest to longest), which is also the case in all tables involving all participants henceforth. In order to protect the interviewees' privacy, all the names that are used are pseudonyms.



Table 1: Overview of participants

Key: \* still studying at time of interview.

#### *Robert (English-Irish, 20) and Stephanie (Swiss, 24)*

*Stephanie and Robert met while they were both studying in Aberdeen, Scotland. For several months, they kept running into one another at college parties and at the library, where they always had long conversations. They became a couple shortly before the end of Stephanie's exchange year. Robert's family is originally from Ireland, but he grew up in Luton (East of England); his accent is a mixture of the varieties spoken in both locations. After Stephanie's year abroad ended, Robert decided to move to Switzerland with Stephanie, and after a few months he found a job as an English teacher at a private language school in Saint Gall. Stephanie was studying English language and literature at the University of Zurich at the time of the interview and they had been a couple for a year. Stephanie and Robert speak mostly English to each other, but they try to speak German as well now as Robert intends to stay in Switzerland and therefore wants to learn the language. They also mix languages occasionally; this mixing usually involves using loanwords from Swiss German while speaking English. Robert started doing this after moving to Switzerland because there are certain words he really likes in Swiss German, and Stephanie followed his example.*

#### *Tim (Australian, 29) and Sarah (Swiss, 23)*

*Tim and Sarah met in 2005 when Sarah was travelling in Australia and attending a language school in Cairns. They met at a local bar when they were both out with friends; Tim was there with a local group of friends, and Sarah with other language students from overseas. Tim is from Brisbane, Queensland. He has an advanced rigger licence and works as a technician and telecommunica‐ tions field worker for Telstra, Australia's largest telecommunications and media company. Sarah grew up in a village in the Zurich area, and was finishing her bachelor's degree in Business Law at a Swiss University of Applied Sciences and Arts ("Fachhochschule") at the time of the interview. She is the only one of the Swiss participants who does not have a Swiss baccalaureate ("Matura"), and has thus had less formal education in English than the others. She speaks English fluently but made more mistakes than the others did during the interview. At the time of the interview, Tim and Sarah had been together for three years. Their relationship had been largely long-distance, and they were not quite sure yet where they were going to live in the long term. Sarah and Tim always speak English to each other, as Tim speaks little (Swiss) German. Occasionally, they integrate some German words that he has learnt into English.*

#### *David (English, 32) and Susanne (Swiss, 29)*

*David and Susanne met at the end of 2003 when Susanne was working as a teaching assistant in Southern England for a year. They met through friends of theirs (an Englishman and a French exchange student, who were themselves a couple), and became a couple within two weeks of their first meeting. After the end of Susanne's exchange year the following summer, they had a long distance relationship for half a year; then David decided to resign from his job and move to Switzerland. Susanne, who is originally from a small town in the east of Switzerland, has a master's degree in English language and literature from the University of Zurich and works as an English teacher at a grammar school in Saint Gall. David holds a Bachelor of Science in business information systems and works as a senior technology officer for an IT and marketing company near the Austrian border. The business language is English, as most clients are native speakers of English. Because most of his colleagues are also fluent in English, he rarely uses German at work. The couple had been together for about five years at the time of the interview and had been married for one year. David and Susanne reportedly speak mainly English to each other, but occasionally use Swiss German expressions.*

#### *Courtney (English, 24) and Martin (Swiss, 26)*

*Martin and Courtney met at a college party in Aberdeen, Scotland, in 2004, when they were both studying at the University of Aberdeen. Martin is originally from Saint Gall and spent a year in Scotland on a scholarship. Courtney grew up mainly in England, but she has Ghanaian roots and many of her family members live in Ghana. After Martin's exchange year ended, they had a long-distance relationship for four years. Courtney decided to study in Italy for a year and Martin went to Vienna on a second exchange year, before earning his master's degree in English language and literature from the University of Zurich. After finishing her bachelor's degree in Scotland, Courtney moved to Switzerland to live with Martin — who was then working as a journalist in Basel — and to pursue a master's degree in art history at the University of Basel. At the time of the interview, she had been in Switzerland for a year and they had been together for almost five and a half years. Martin and Courtney speak mainly English to each other, as Courtney's German skills are still limited. They occasionally decide to speak German to each other so that Courtney can practice, but tend to return to English after a few sentences, because Martin feels he has to explain aspects of the language to her in English. At times, they use German in humorous situations.*

#### *Richard (English, 29) and Sophia (Swiss, 31)*

*Richard and Sophia met online in an English-speaking forum for Harry Potter fans in 2005. After communicating online regularly for some time, Richard decided to meet Sophia in person and flew to Switzerland for a weekend in late 2005. They decided to give their relationship a try that weekend and maintained a long-distance relationship for the following two-and-a-half years. In 2008, Richard moved to Switzerland, where he attended German school every day for half a year, and he has since then become fairly fluent in Swiss German as well. Richard has a bachelor's degree in business administration and works for a small company owned by Sophia's father, who usually speaks to him in Swiss German. Sophia grew up in a town in the canton of Zurich. She holds a master's degree in history and teaches history in English ("Immersion") at a Swiss grammar school. She spent a year in New York at the age of eleven and has loved English ever since. At the time of the interview, the couple had been together for seven years and had been married for three years. They have a daughter who was then seven months old, whom they address in their respective mother tongues. Sophia and Richard speak mostly English to each other, though they have also started using Swiss German, especially when they are with other people, but also in humorous situations.*

#### *Claire (Northern Irish, 28) and Simon (Swiss, 34)*

*Simon and Claire met in a bar in Belfast in 2005 when they were both out for a drink with friends. At the time, Simon was studying at Queen's University in Belfast for a year. After Simon's return to Switzerland, Claire obtained her bachelor's degree in occupational therapy in Belfast. They had a long-distance relationship for three years; then Claire found a position as an occupational therapist in Simon's hometown in eastern Switzerland. Simon holds a master's degree in English and Spanish language and literature from the University of Zurich and teaches these languages at a Swiss grammar school. At the time of the interview, the couple had been together for eight years and had been married for nine months. Claire had been living in Switzerland for four and a half years and was working in a Swiss German-speaking environment; she also had no friends who were native speakers of English. At home, Claire and Simon speak English to each other most of the time, but they address each other in Swiss German when they are with other people. They also mix Swiss German and English at times, using English as their base language and integrating Swiss German expressions, especially when they are talking about Claire's work.*

#### *Dean (English, 29) and Monika (Swiss, 29)*

*Monika and Dean met through mutual friends in 1999 while Monika was staying in London for a year. She had just finished grammar school in Zurich and had moved to London, where she was working as a language teacher. When Monika moved back to Switzerland to finish her studies the year after, Dean followed her. After many unsuccessful applications, he finally found a position, but in order for him to be allowed to stay in Switzerland, the couple had to get married. Reluctantly, they decided to marry legally at the age of twenty; however, they only celebrated their wedding eight years later. Dean does not have a university degree and works in IT recruitment. Monika, who has a master's degree in English language and literature from the University of Zurich, works as an English teacher at a grammar school. She grew up in a town near Zurich. At the time of the interview, the couple had been together for nine years and had been married for eight years. Monika and Dean speak English to each other almost exclusively, although Dean is fairly fluent in German and understands almost everything. They hardly ever mix languages, with the exception of the occasional term they borrow from the other language.*

#### *Joshua (American, 30) and Deborah (Swiss, 28)*

*Joshua and Deborah met in 1998 when Joshua travelled to Europe as a missionary for the Church of Jesus Christ of Latter-day Saints. Deborah is a member of the same church, and her family invited Joshua over for dinner on a regular basis. Deborah and Joshua were not allowed to pursue a romantic relationship during Joshua's assignment, but he returned to Switzerland shortly after and they started dating. Deborah was 18 at the time and Joshua was 20. Joshua is originally from Utah, where he attended university for three semesters, but when he moved to Switzerland, he had to complete the British A-Levels to be allowed to enrol at the University of Zurich. He was finishing his master's degree in English language and literature at the University of Zurich at the time of the interview and intended to become a teacher afterwards. To support their family, he was also working part-time at a law firm. Deborah, who grew up in the Zurich area, has a Swiss Academic Baccalaureate ("Matura") and did temporary work before becoming a stay-at-home mother to their two young girls. The couple had been together for ten years, had been married for eight years, and were expecting their third child. Deborah and Joshua are the only couple among the interviewees who do not use mostly English with each other. Deborah did not feel comfortable speaking English when they met, and Joshua's* *German was already very good at that point. Initially, they spoke to each other in Standard German, then they switched to Swiss German, and later they started mixing English and Swiss German. Now that they have children, they mix less and try to speak one language at a time when with their children. They raise their children bilingually using a "one parent, one language" strategy.*

#### *Craig (American, 32) and Katia (Swiss, 28)*

*Craig and Katia met in February 1997 when Katia spent a high school exchange year in Texas. Craig was completing an internship for his university degree at the time and staying with the same family as Katia's friend, an exchange student from Denmark. Katia was 17.5 and Craig was 22 when they met. At the end of her exchange year, Katia returned to Switzerland to graduate from grammar school. After maintaining a long-distance relationship for four years, Craig, who is an electrical engineer, lost his job in the USA, found one in Switzerland, and moved there permanently in 2002. He works in an English-speaking environment but has learnt (Swiss) German in order to converse with Katia's family and friends. Katia holds a master's degree in human movement sciences and sport and works as a grammar school teacher. At the time of the interview, they had been together for 10.5 years and had been married for three years. They have a two-year-old daughter whom they are raising bilingually. Craig and Katia usually speak English to each other, unless they are in the company of native speakers of (Swiss-)German. They also mix languages sometimes, if they cannot think of a word in the language that they are speaking or if there is a phrase in either of the languages that fits the situation better. However, they are now trying to avoid code-switching in the company of their daughter, as they find it important to each use one language when communicating with her.*

#### *Karen (American, 51) and Philipp (Swiss, 47)*

*Philipp and Karen met at the University of Minnesota in 1989 when Philipp was studying there for a year. Karen was working as a campus minister, and Philipp was her best student. He went back to continue his studies at the University of Zurich, but returned to Minnesota for another exchange year in 1991. After his return to Switzerland, they had a long-distance relationship until Karen managed to find a teaching job in Switzerland; they married the same year. Philipp has a master's degree in English language and literature, and Karen in education, and they both work as English teachers. Philipp teaches English at* *two different Swiss grammar schools, and Karen teaches business English at a University of Applied Sciences and Arts ("Fachhochschule"). At the time of the interview, they had been together for 24 years. They have three teenage children, whom they have raised bilingually, and a large network of friends who are native speakers of English. The couple speak English almost exclusively to each other. Philipp and their children also mix languages at times, but Karen does so only very rarely, apart from a few terms that she uses in Swiss German. Karen does not feel very comfortable speaking or writing in German; however, she actively seeks out situations which require passive knowledge of the language.*

The couples' biographies show that, in many ways, these relationships are different from the ones frequently depicted in previous research. For instance, Khatib-Chahidi, Hill and Paton reported that many of their subjects married nationals of other countries because they were feeling marginalized (structur‐ ally, socially or personally) (1998: 57, cf. section 3.2.2, "Bicultural couples"). None of these aspects were mentioned by any of the respondents in my study, and none of the twenty interviewees depicted their partner as a "way out". Moreover, the women in Khatib-Chahidi, Hill and Paton's study reported that they had had intensive contact with other foreigners before meeting their spouse (as had their spouse, since they met in the women's home country) (1998: 57). While this is also true for most Swiss interviewees in my study, many of the English-speaking partners had very little if any contact with Swiss or other German-speaking foreigners before meeting their partner. To my knowledge, none of my interview subjects had had previous (long-term) partners from other cultures. There are some similarities, however, to the bilinguals in Piller's (2002a) sample. For instance, many of the Swiss participants expressed very positive feelings about the English language, although only one Anglophone participant (Joshua) had similarly positive feelings about Swiss German (see section 7.4.2, "Anglophone partners' attitudes").

There are a number of particularities of the composition of the sample that should be borne in mind for the analysis. The English-speaking partners are from a variety of different cultural backgrounds, which might skew trends with regard to certain linguistic behaviour (such as swearing or language mixing) or influence the use of suprasegmental features. The Anglophone partners' cultural background and variety of English may also influence the participants' attitudes, and thus, potentially, their language choice. Moreover, there is a slight imbalance with regard to the educational level of the participants, as there are more English-speaking partners with no tertiary education than there are Swiss partners. Conversely, the sample is relatively balanced with regard to gender in combination with other factors like the age and educational level of the participants: The educational level of the female partners is slightly higher on average than that of the male partners, and the gap in the average age of the male and female participants, as well as the Swiss and English-speaking participants, is minimal (less than a year in both cases).

As regards the applicability of the findings of this study to other bilingual couples, it should be noted that the fact that one partner is a native speaker of English is bound to influence the results, for instance with respect to the partners' attitudes and language choice. Since English has a very high prestige internationally, and Swiss German is valued highly within Switzerland, the couples' place of residence, it is to be expected that the couples evaluate their bilingualism positively. If they lived in another environment (e.g. in an English-speaking country), or spoke different languages, this would possibly not be the case.<sup>3</sup> Furthermore, it should always be kept in mind that the sample includes exclusively couples in which both partners are proficient in the couple language, and very familiar with their partners' culture; most of them lived in their partners' country for an extended period of time.<sup>4</sup> Due to this selection, the couples only reported a limited range of linguistic and intercultural conflicts. In part, this might also be attributed to the fact that, as has been noted in other studies on couple relationships, couples with serious relationship issues are unlikely to participate in such a study (e.g. Fitzpatrick 1988: 80; Piller 2002a: 45). However, even though the findings of this research cannot be generalized to all bilingual couples, I believe that the focus on extensive interviews with proficient bilingual couples will enable me to draw some firm conclusions on how bilingual couple communication works on an advanced level.

### **4.3 Interviews**

The couples were interviewed in their place of choice, either at their home or at mine; I avoided public places as I needed excellent recording quality

<sup>3</sup> The influence of these factors was demonstrated by Schüpbach (2009) in a study of bilingual couples in Australia (see section 7.2, "Previous work on attitudes and attrac‐ tion") and in Boyd's (1998) study on immigrants with different linguistic backgrounds in Scandinavia (see section 2.4, "English in Switzerland").

<sup>4</sup> A total of 9 out of 10 Swiss and 9 out of 10 English-speaking partners had lived in their partner's country (or another English-speaking country) for at least ten months. The only exceptions to this are Deborah SG and Tim <sup>E</sup> .

for the purpose of my study (for instance, to be able to distinguish voice quality). Furthermore, I wanted them to feel as comfortable as possible, so that their conversation would closely resemble their usual manner of speaking to each other, and so that they would feel open to sharing their experiences. Therefore, recording the couples in a laboratory was not an option either. The interviews were recorded digitally with a high-quality stereo recorder. They took place between May 2008 and February 2013 and lasted between a little less than an hour and an hour and a half. This resulted in a corpus of 11.25 hours of conversation (see Table 48 in Appendix II for a detailed overview of the individual recording times). On average, the interviewees spoke 5154 words during the interviews (see Table 49 in Appendix III for a comparison of the number of words and intonation units each participant used during the interviews). The male partners used a greater number of words than the female partners (5529.9 vs. 4777.6 words), and the native speakers of English spoke more than their Swiss partners (5414.1 vs. 4893.3 words). The questions which the couples were asked do not account for this imbalance, as both partners were addressed equally often (see section 4.4, "Questionnaire", and Appendix I, "Complete questionnaire").

The interviews were all conducted in English; the couples were encouraged to switch to Swiss German if they felt more comfortable speaking about a certain subject in this language, but they hardly ever chose to do so. Before the interview, I told the couples that I was interested in hearing them speak to each other and discussing the questions from the questionnaire with each other, rather than in having individual answers directed at me. The couples were informed that I was going to write about the manner in which bilingual couples communicate, but they were not told which areas I planned to analyse. Aside from when I asked questions or gave some backchannel signals, I tried to keep in the background during the interviews. Of course, my presence was noticeable nonetheless, and potentially heightened by the fact that I share some essential biographical elements with the couples. I am a native speaker of Swiss German, studied English at university like many of the participants, and like them, I am also in a bilingual relationship. In my opinion, having these things in common was a considerable advantage, as it seemed that many of the interviewees were happy to tell their experiences to someone who could relate to their situation.

#### **4.4 Questionnaire**

The questionnaire (see Appendix I) was designed with the intention of covering areas that were important to the interviewees' lives as bilingual, bicultural couples, and of prompting a substantial amount of speech production and interaction between the partners. It includes questions about their relationship history, their language use (inside and outside the home) and their attitudes towards bilingualism, biculturalism and their partner's language and culture. Furthermore, there were a number of questions concerning the expression of positive and negative emotions, the use of terms of endearment and their swearing behaviour. Because I also wanted to examine how they express positive and negative emotions when conversing together, I asked questions that I expected to trigger a positive reaction (e.g. "What attracted you about each other when you first met?"; "What are the advantages [… ] of living in a bilingual, bicultural relationship?") or a negative reaction (e.g. "Were there any initial conflicts?"). I also asked a number of questions that I expected to elicit certain linguistic behaviour from the interviewees, as they would force the couples to negotiate (e.g. "Who is funnier?"; "Who is more polite?"), to criticize or commend each other ("How proficient is your partner in your language?") or to talk about potentially sensitive subjects ("How do you get along with your extended family?"; "What are your plans for the future?").

The questionnaire was structured with the objective of making the couples feel relaxed and comfortable from the beginning of the interview. I therefore started with a question that was supposed to encourage them to feel positive about their relationship and hence to speak more freely about it ("How did the two of you meet?"). I then moved on to relatively general questions about their language use and their lives together, before broaching more specific questions and, finally, addressing more sensitive or personal topics. With the help of this structure, I aimed to attain a high level of openness from the participants.

An interview situation with a pre-set questionnaire has many advantages, but also some potential drawbacks. On the one hand, my presence and my questions may have had a slight impact on the naturalness of the conversations. Some of the couples also felt a bit nervous at the beginning, or were slightly anxious about certain questions, which might have had an effect on the way they talked to each other. On the other hand, the questionnaire served to give the conversations a frame and guaranteed that I received the information necessary for the content analysis from the couples. This permitted a comparison between the conversations and the couples' answers.

#### **4.5 Transcription**

The interviews were transcribed using *Express Scribe* transcription software. Since my analysis required a relatively precise transcription including a number of suprasegmental features, I adapted the comprehensive transcription conven‐ tions proposed by Du Bois et al. (1993) to suit the foci of my analysis. I decided to put each intonation unit (see below) on a separate line, and to mark the nuclear stress in each intonation unit with a caret (^). At the end of each intonation unit, its terminal pitch was indicated, such as rising (/), falling (\) or level (\_) pitch, as well as glissando pitch movements (/~ \~) and strong rises (//) and falls (\\). I also included other aspects such as pauses (…  for standard pauses and .. for very short pauses), prosodic lengthening (:), strong/contrastive stresses (^^) and elements of conversational organisation, such as overlap [ ] and latching (=). Some intonation units contained rhythmic speech (<RH>) — with a beatable rhythm — or marked "marcato" speech (<MRC>), with several marked stresses. Vocal noises like tongue clicks (<TSK>), inhalation and exhalation (<H> <Hx>) and laughter (@) were indicated, too. Different types of non-permanent voice quality were also marked in the transcriptions, for example soft (<P>), loud (<F>), low (<LO>), high (<HI>), husky (<HSK>), creaky (<CRK>), smiling (<:-)>) or nasal (<NAS>) voice. Some changes in pitch and loudness over the course of an intonation unit or turn were indicated as well, namely increasing or decreasing loudness (<CRE> and <DIM>) and pitch (<UP> and <DOW>). In addition, I marked very fast and slow speech (<A> and <L> respectively). If necessary, comments were added in double brackets and translations in single quotation marks (see "List of transcription conventions" at the beginning of the book for a complete list).

Transcribing was painstaking, due to the length of the interviews (11.25 hours in total), the level of detail of the transcriptions and the fact that there were several speakers whose speech overlapped fairly often (on average, each participant overlapped 528 times with another speaker).<sup>5</sup> Nevertheless, I deemed the prosodic detail of the transcriptions necessary, as the prosodic features might contain clues about the speakers'communicative intentions or emotional state. Only prosodic elements that I did not believe to yield any insights for the specific purpose of this study were omitted. These include, for example, secondary stresses or standard pauses in between intonation units, which were not indicated unless they were exceptionally long. Such pauses are difficult to compare; as Edwards notes, measuring pauses is only partially meaningful,

<sup>5</sup> This number includes laughter overlaps and backchannel signals; it refers to overlaps with their partner as well as with the interviewer and indicates the number of intonation units that overlapped.

because "perceived pause length depends not only on physically measurable time, but also on speech rate, location of the pause [… ] and other factors" (2001: 322). Furthermore, it is often impossible to make a judgement on who is responsible for a pause that occurs between turns (2001: 332).

I encountered a number of difficulties during the transcription process, which were partly due to the fact that the couples have very different manners of speaking (e.g. shorter or longer pauses, one or more stresses for intonation units). First of all, it was not always clear when to "break" an intonation unit. An intonation unit has been defined as "a stretch of speech uttered under a single coherent intonation contour" (Du Bois et al. 1993: 47). It is supposed to consist of a nucleus (primary accent) and other syllables, though it has been proposed that an intonation unit is often not restricted to this (Szczepek Reed 2006: 28; Du Bois et al. 1993: 56), which can present problems. Szczepek Reed, for instance, infers from her own transcriptions that "a given intonation unit is by no means restricted to one nucleus, but may carry two or more syllables with equally prominent stress" (2006: 28). Chafe (1994) suggests that intonation units can be delimited by pausing, pitch change, duration or tempo changes, intensity, voice quality or speaker changes. Other scholars, such as Gumperz, have grouped intonation units into minor tone groups, which "delimit a message treated as a component of a larger whole, and major tone groups [… ] which are more independent" (1982: 110). Yet this distinction is also often difficult to uphold. I considered all of these factors in my own transcriptions, but was still obliged to make an instinctive decision in a number of cases. Despite these challenges, I considered distinguishing intonation units to be very important, as they provide information with regard to turn-taking behaviour (see Gumperz 1982) and are necessary to determine terminal pitch.

Another difficult decision, which is linked to the question of the intonation units, was where to place the nuclear stress. As Gumperz remarks, "accent placement is for the most part grammatically conditioned" and tends to fall on content words towards the end of the intonation unit (1982: 111). This is not always the case, though, and contrastive accents or deaccentuation may also occur. While a single nuclear stress could be clearly discerned in most intonation units in my interviews, there were nonetheless instances where there were several stresses without a significant difference between them. In these situations, I chose to set several nuclear stresses. In most cases, this correlates either with a rhythmic speech (<RH>) or marked speech (<MRC>), both of which were also indicated.

Moreover, there were considerable differences with regard to stress as well as vowel lengthening between the varieties of English spoken by the participants. Generally speaking, British English seems to be more strongly stress-timed than American English,<sup>6</sup> and there is a clearer distinction between short and long vowels. This means that the transcriptions might not be entirely uniform as regards aspects such as intonation, lengthening or pausing, since it is possible that a stress or a pause of similar intensity or length might appear more salient in the context of a conversation in one variety of English than in another.

#### **4.6 Coding and retrieval**

Once transcribed, the interviews were inserted into a software application which assists in the management and analysis of large amounts of qualitative and quantitative data (MAXQDA 12). For the coding and the analysis, I proceeded by topic, as each topic is linked to a different field of previous linguistic work, and required different methods of analysis. For each topic, a separate code system was developed, and all relevant segments were coded in MAXQDA 12. In some cases, these segments took the form of longer extracts which were then analysed quantitatively; in other cases, certain features (e.g. laughter) or individual shorter expressions (e.g. code-switches, expressions of emotions, or swearwords) were coded and assigned to categories. Whenever possible, I tried to follow my data in the creation of a code system, rather than creating categories and attempting to fit my data into them (see also Hay 2000: 727). For instance, I looked at each instance of laughter and tried to determine the main trigger; if none of the categories I had already established was suitable, a new one was created. Some of the coding could be carried out automatically (e.g. backchannel signals and suprasegmental features were auto-coded for each participant), but most of it had to be completed manually. In this case, I tended to use the search function and then to assign segments to the appropriate category.

The main advantages of coding segments in MAXQDA 12 were that I could find and draw together everything belonging to one topic (for the qualitative analysis), and that I was also able to count and compare features easily (for the quantitative analysis). In addition, I was able to revisit each coding, for instance, if I wanted to view the context, or if I decided to subdivide a category. Another major benefit of the system was that I could search, code, and finally analyse features for each speaker individually, as the software is able to recognize which

<sup>6</sup> To my knowledge, this has not been the main focus of any research, though in a study of acoustic correlates of stress-timed and syllable-timed languages, the British speaker scored higher for stress-timing than the speaker of American English (Mairano and Romano 2007: 1151).

speaker a turn belongs to in a transcript involving several speakers. This was particularly helpful in the analysis of quantitative elements for each individual speaker, such as laughter or suprasegmental features.

## **4.7 Analysis**

Once the coded segments had been retrieved from MAXQDA 12, I began with the analysis. I decided to focus on six main topics that seemed to be of particular interest based on the couples' conversations and previous research. Each of these topics is discussed extensively in a separate chapter. All the chapters are structured in a similar manner, containing an overview of previous work on the subject, a section of quantitative analysis (in the chapters where this was relevant), a qualitative analysis of the participants' reports and comments on the subject, and an overview in which I summarize the most important results of the analysis, and compare the qualitative and quantitative parts of the analysis with each other and with previous work. For the chapters containing quantitative analyses, the exact methodology used is described in the respective chapters, since this depended on the subject.

The qualitative parts of the analysis revolve around the couples' thoughts and feelings about aspects of their bilingual, bicultural relationship, whilst also taking into account their reports on their linguistic behaviour. Thus, I attempt to ascertain whether there are any parallels between the couples or between the partners, and to create a basis for comparison with the interviewees' linguistic behaviour during the interviews. In these sections, I included a relatively large number of extracts from the interviews that I considered exemplary or particularly interesting. Wherever this was considered relevant, these extracts were also examined linguistically, for instance with regard to the speakers' use of emphasis or voice quality. When I quoted from the interviews, I indicated the nationality and age of the speaker(s) for longer quotations (which were also set apart from the text). In order to improve the readability of the transcripts, I used subscript for suprasegmental features (but not for elements of conversational organisation, stresses and pauses). The paragraph (i.e. turn) numbers in which the quotations appeared in the original transcript were indexed in brackets at the end of each example. However, the speakers' turns were not numbered unless a specific turn was referred to in the analysis. If anything was omitted from the original transcripts, this was indicated with three dots in double brackets ((… )). Important parts of the transcripts were highlighted in bold print; parts that were referenced in the analysis were underlined in the extracts so as to facilitate a comparison. In order to improve the legibility of integrated quotations, I omitted all suprasegmental features except for contrastive stresses and lengthening. When referring to specific participants in the text, I indicated their mother tongue after their name in subscript (<sup>E</sup> for English or SG for Swiss German) if this was not obvious from the context.

The quantitative sections of the analysis focus on aspects of the couples' linguistic behaviour during the interviews, such as their language mixing or swearing behaviour. In all of the sections dealing with quantitative research, I also considered the potential influence of gender<sup>7</sup> and mother tongue. In this context, it is important to note that the effect of these factors on the participants' linguistic behaviour is difficult to discern clearly, not only because the sample is relatively small, but also because the distribution of combinations of gender and mother tongue is not even. While there are ten speakers of each gender and mother tongue, there are only three female English speakers to seven male English speakers; the converse is true for the Swiss participants. In order to avoid mistaking the influence of one factor for another, I therefore decided to consider the four possible combinations of gender and mother tongue in addition to the overall numbers for both genders or for both mother tongues in the analysis. The advantage of looking at the four combinations of mother tongue and gender separately is that the factor of culture or mother tongue is abstracted from the factor of gender to some degree, while it may also reveal aspects about the manner in which the two variables interact with regard to the bilinguals' linguistic behaviour. However, there are fewer participants in each group (7 and 3, respectively), and the groups are more difficult to compare. The main advantage of considering the total number of members of both genders — or speakers of both mother tongues — is that the number of participants in both groups is larger (n=10) and that tendencies are therefore more evident. At the same time, however, there are more factors that can distort the results, particularly since the distribution of genders and mother tongues is not balanced. Consequently, I always considered all members of each group in the analysis, as well as the four combinations of gender and mother tongue, to ensure that there is no interference from the respective other variable in my findings. Since the duration, content, and key of the conversation can influence the participants' linguistic behaviour, I compared the two couple combinations with one another (i.e. the Swiss women with the Anglophone men, and the Anglophone women with the Swiss men), as this reduces the likelihood that

<sup>7</sup> It should be noted that gender was treated as a binary variable in the analysis, as all of the interviewees are cisgender, i. e. their gender identity matches their biological sex.

the particular conversation in which the bilinguals are participating works as a confounding variable. In addition, a comparison of the two couple combinations can give a better indication of the interplay between the variables of gender and mother tongue.

Concerning the quantitative analysis that was carried out, it needs to be borne in mind that this is largely an exploratory study. Due to the small sample of participants, only basic descriptive statistics were used, and no tests of significance were performed. Therefore, results ought to be interpreted as potential tendencies, and need further investigation in a larger sample to be corroborated. It should also be noted that a certain degree of clustering might have occurred, due to individual behaviour (as individuals may have a tendency to use certain features especially frequently) and conversational flow (as the use of linguistic features sometimes occurs in clusters in conversations; see Hay 2000: 727). The first type of clustering could have been relevant when the role of the variables of gender and mother tongue was examined, while the second type might have affected the results regarding similarities between the partners. Moreover, there are considerable differences in the number of words the individual participants spoke during the interviews, resulting in sub-corpora of different sizes. Therefore, the number of words or suprasegmental features which the participants used were not only given as absolute numbers in the analysis, but also as normalized frequencies where relevant, as these are a more reliable indicator for a comparison. Thus, the number of instances a particular feature occurred was set in relation to either the number of words the participants spoke during the entire interview, or to their total number of intonation units. For instance, the frequency of certain expressions (such as emotion words) was normalized per 10,000 words that the interviewees said, and the occurrence of terminal pitch or voice quality (which were specified once for each intonation unit) was indicated as a percentage of the intonation units they used. In a few situations, a ratio was calculated (e.g. type-token ratio, or the ratio of laughter pulses to words).

I present the chapters in order of what I perceive to be the most general topic to the most specific one, and therefore begin with the couples' language choice and use in chapter 5.

## **5 "We have a language of our own": Language choice and use**

## **5.1 Introduction**

A bilingual couple's language choice is essential to the manner in which the partners communicate with each other, since there are a multitude of conversational elements that are language-specific (for an overview, see section 3.3.3.2, "Linguistic differences and misunderstanding"). If the couples in this study spoke mainly German — rather than English — to each other, numerous aspects of their couple language might be different: their conversational style, their use of formulaic expressions, their level of directness, their manner of expressing emotions, their use of verbal humour, their prosody, and so on. Despite the strong influence of their chosen medium of communication, however, the relationship language of a bilingual couple will not be the same as that of a monolingual couple speaking only one of the languages, since, regardless of which language the bilingual partners use as their primary means of communication, features of the other language will inevitably seep into their couple language to a certain degree. The influence of the second language may be particularly evident if both partners are bilingual, but even if the partners are not fluent in each other's mother tongues, they may adopt features from their partner's language after these have been transferred, often unwittingly, into the couple language.

Consequently, a bilingual couple's language is often a complex blend of elements from both partners' languages. Each partner contributes specific features to the couple language, and as their relationship evolves, so does their couple language. While this occurs unintentionally to some extent, there are also ways in which the partners may actively shape their joint language. Thus, their bilingualism might cause the partners to consciously modify their manner of speaking, for instance by selecting a simpler register or by expressing themselves more clearly or more explicitly. The partners can also make use of their two languages or their bilingualism to create humour, to coin their own expressions, and, ultimately, to generate a distinctive manner of commu‐ nicating with each other. As a result, partners in a bilingual relationship often perceive their couple language to be unique (Piller 2002a: 224). Thus, in spite of the challenges that are sometimes associated with being a bilingual couple, bilingual partners may also derive benefit from their linguistic situation, as their communicative efforts and their shared code can create a rapport between them or strengthen their sense of togetherness.

There are a number of factors that influence which language or languages partners speak to each other, such as their country of residence, both partners' language proficiency, their habits, their attitudes, or concerns relating to their personal and couple identity (see section 5.2, "Factors influencing the language choice of bilingual couples"). Due to the multitude of factors affecting the partners' language choice, a bilingual couple's linguistic situation is often far from simple or static. Over the course of their lives, various events can occur which may change their mode of communication, and the extent to which they speak both of their languages. They might relocate to a different country, for instance, have children, or become more fluent speakers of their L2.

In the following, I first give an overview of previous work in the areas of bilingual couples' language use and discuss a number of factors that may influence a couple's language choice (5.2). I then address the interviewees' reports of which language(s) they speak to each other, potential situational factors, as well as reasons for their language choice (5.3). Because the partners' language use outside the home is likely to influence their proficiency, and hence possibly their linguistic behaviour at home, I also comment on their reported language use with other people, and explore the interrelation between the interviewees' language use outside of and within the relationship (5.4). Furthermore, I discuss two aspects that are reported by several couples to be characteristic about the manner in which they communicate, namely the modification of their speech to suit their bilingual situation, and the achievement of a high level of communication due to their different cultural and linguistic backgrounds (5.5). In the last section of the chapter (5.6), the results are summarized and compared to previous work. On the one hand, this chapter aims to provide insights with regard to the factors influencing the language choices of fluent bilinguals in this specific setting; on the other hand, it gives an overview of the couples' past and present language use and serves as a basis for the analysis of more specific areas of the couples' language use in the subsequent chapters.

## **5.2 Factors influencing the language choice of bilingual couples**

#### **5.2.1 Proficiency and habit**

There are many different factors that interact to determine the language choice of a bilingual couple, and it can be difficult to decide which factor is dominant in each situation. One factor that inevitably plays an important role, at least initially, is the partners' language proficiency. It seems highly likely that a couple will choose the relationship language in which their joint level of language proficiency is the highest. This may be the mother tongue of either of the partners, or a lingua franca. Beraud, who examined the language use of five Norwegian-Ukrainian bilingual couples, reports that many of them use English as a lingua franca, since this is "the strongest and fairest shared language" for them (2016: para. 1). When examining the language choice and use of German-English bilingual couples living in a number of different countries, Piller also found that language proficiency "is the most basic constraint on language choice" (2002a: 5). Their language choice is possibly linked to the partners' formal education in each other's mother tongues, as individuals who have studied their partner's language at school tend to be more confident about their L2 skills and have a sense of language ownership (Piller 2002a: 99).

As the partners' language skills evolve, the relationship language may also change. The partners may, for instance, start mixing languages, or using the respective other language in certain situations. However, once a language has been firmly established as a relationship language, habit renders it unlikely that the main language will change. Indeed, many couples in a study by Piller claimed that they use their relationship language out of habit, as it is the language they used when they first met (2000: section 4.2). One reason why habit is an important factor with regard to language choice is that the established language tends to be connected to the couples' sense of couple identity. Thus, couples often "stick to the language of their first meeting because they might lose the sense of knowing each other, the sense of connectedness and the rapport derived from being able to anticipate what the partner might be about to say if they switched" (Piller 2002a: 138). The language they are accustomed to using — as well as bilingualism itself may be part of their idea of themselves as a couple. Furthermore, linguistic habits are only likely to change in conjunction with a considerable change in circumstances (Piller 2002a: 140). Since there is a high probability that, for many bilingual couples involving a native speaker of English, including the couples in this study, the language of the first interaction will be English, this frequently becomes their couple language, especially if English also happens to be the community language in their place of residence.

#### **5.2.2 Community language**

A factor that is relevant with regard to the partners' L2 proficiency, and thus potentially also their language choice in the long term, is the community or majority language in the bilingual couple's country of residence. It is often assumed that the majority language is the natural choice for bilingual couples (Piller 2002a: 133), and that "[i]n mixed marriages there is usually a shift to the majority language" (Romaine 1995: 42). This was confirmed in Piller's study of 51 couples, which suggested that "the community language is the most powerful indicator of the language bilingual couples will use for marital communication" (2000: section 4.1, para. 5). In a different sample of 36 couples, though, Piller found that only 29.2% of the couples residing in Germany used German with each other, whereas 80% and 40% of the UK and US-based couples respectively used English as their common language, and almost a third of the couples claimed to use a mixed code (2002a: 134). Nonetheless, it seems that couples usually consider the community language the more obvious or preferable choice; Piller reports that the couples in her study who used the minority language or a mixed code "tend[ed] to engage in lengthy discussions, explanations and justifications in their responses", whereas couples who used the community language rarely did so (2002a: 135). However, there may also be feelings of solidarity towards the less prestigious language (Siguan 1980, in Piller 2002a: 22). Thus, couples might use the non-community language to compensate for the immigrant's potentially weaker position and attempt to even out the power imbalance to some extent (see also section 3.3.3.1, "L2 Learning and use").

#### **5.2.3 Attitude and motivation**

Attitude is an essential component in language learning and may have a considerable influence on both language competence and language choice.<sup>1</sup> Besides the attitudes of the two partners, the attitudes within their immediate

<sup>1</sup> This section only includes work on the role of attitude in language choice. Other aspects relating to the attitudes of bilingual couples are discussed extensively in chapter 7, "'This uh foreign girl with a great accent': Attitude and attraction".

environment can also play a vital role. Thus, De Klerk discovered in a case study of ten bilingual South African couples that their choice between English and Afrikaans depended greatly on their personal attitudes, and that it was affected by the negative connotations Afrikaans had acquired in recent history (2001: 199). The Afrikaans-speaking partners in her study made more of an effort to learn English than their partners — some of the English speakers had not tried to learn their partner's language at all (2001: 202–203). Moreover, nine of the ten couples conceded that, despite their intentions to raise their children bilingually, their family language had a strong inclination towards English (2001: 210). This indicates that bilingual couples' attitudes will also influence the linguistic upbringing of their children (see also section 7.6, "The couples' views on raising bilingual children"). The influence of attitude on language choice was also noted by Gal (1979), who studied German-Hungarian couples in the Austrian town of Oberwart and found that they were shifting toward the language with the higher symbolic value, which was German in their case.

In addition to the attitudes towards each language, the partners' attitudes towards bilingualism itself may also affect their language choice. In the past, negative views on bilingualism used to be common, especially as regards childhood bilingualism, as it was often assumed that "increasing one language will automatically cause a decrease in the second language" (Baker 1988: 170), which was thought to potentially result in deficient language skills in both languages. Views towards bilingualism have generally become considerably more positive, but not concerning all types of bilingualism. Thus, attitudes towards bilingualism seem to be connected to aspects such as which languages are involved, the age of L2 acquisition, possibly the bilingual's level of education, as well as the situation within the community with regard to multilingualism. For the couples in this study, this means that the prevalent attitude towards bilingualism is probably much more positive in Switzerland than in the native countries of the Anglophone partners. Whereas many Swiss embrace their country's multilingualism (see section 2.2, "Multilingualism in Switzerland"), such positive views of bilingualism are rare in English-speaking countries:

It is a cultural fact that no Anglophone nation anywhere has exhibited enthusiasm for any kind of bilingualism other than traditional. Only where English speakers are themselves a numerical, but elite, minority, as they used to be in South Africa, have they accepted bilingualism, but even there it is asymmetrical. There are many more speakers of Afrikaans who are bilingual than English speakers. (Romaine 1995: 324)

When bilingualism is met with scepticism, elements pertaining to bilingualism such as language mixing may be viewed equally critically; negative attitudes may thus inhibit the use of a mixed code (see section 6.2.1, "Factors that influence language mixing").

In addition to the attitudes towards bilingualism and towards both languages, motivation has been determined as a key factor in second language acquisition and, consequently, in bilingual language use (Gardner and Lambert 1959). Language learning may be affected by many different motivational factors, which are generally divided into integrative and instrumental motives (Baker 1988: 152). Instrumental motivation is mostly self-oriented, as there is "a desire to gain social recognition or economic advantages through knowledge of a foreign language" (Gardner and Lambert 1972: 14, in Baker 1988: 153). In contrast, integrative motivation tends to be other-oriented; it is "a concern to develop personal relationships with a group speaking another language" (Baker 1988: 153). Both instrumental and integrative motives may affect an individual's desire to learn his or her partner's language, and hence influence the language use within their relationship.

#### **5.2.4 Identity and language emotionality**

As has been mentioned earlier, being bilingual and bicultural may be important for the partners' sense of couple identity (see section 3.3.2, "Bilingual couples"), but it may also be linked to aspects of individual identity.<sup>2</sup> A number of studies indicate that fluent bilinguals talk, act, view or experience things differently in their two languages. For instance, Ervin-Tripp showed that bilinguals sometimes complete the same sentence differently (in terms of content) depending on which language they are using (1964: 96). Koven, who examined narratives of personal experiences by two Portuguese-French bilinguals, demonstrated that they perform different kinds of "selves" in their two languages (1998: 436). In a later study on a Portuguese-French woman, Koven (2004) found differences in the bilingual's performance of affect and self in her two languages. To give an example, the woman claimed to be quieter and more reserved in Portuguese than in French. Koven argues "that it is not the structural differences of the two languages that account for [her differing performances], but rather, the reper‐

<sup>2</sup> In-depth discussions on the concept of couple identity have been published elsewhere (see e. g. Gonçalves 2010a, 2013, for a summary), and I will not go into detail on the topic unless it is directly relevant for the present study. The topic of language emotionality, which is linked to identity, will be discussed in more detail in chapter 8, "'In Swiss German, *I lieb di*, that's strange': Expressing emotions".

toire of personas to which this speaker has access in each language" (2004: 471). Similar results were obtained by Marian and Kaushanskaya (2004), who com‐ pared narratives of English-Russian bilinguals, and tried to determine the effect of their languages on self-construal and emotional expression. They discovered that speaking in English brings about "a more individualistic self-construal, whereas speaking Russian [results] in a more collectivist self-construal" (2004: 197). These findings suggest that bilinguals may have a preference for using one or the other of their languages to express or perform themselves in a certain way.

As a consequence, bilinguals sometimes feel that they have multiple or fragmented selves, as can be seen in Pavlenko's analysis of the manner in which multilingual writers position themselves in autobiographical narratives. In their writing, these writers frequently present themselves as "agents whose multiple identities are dynamic and flexible" (2001a: 319). Pavlenko believes that

[the] notion of self as fluid, fragmented and multiple [… ] allows the authors to explore the links between multiple languages and selves in ways that were previously non-ex‐ istent and/or impossible: challenging the essentialist notions of self, deconstructing various ethnic, national, colonial, and gender identities, creating new discourses of hybridity and multiplicity [… ]. (2001a: 339)

Since language has a close connection to our feelings, thoughts and self-image, not speaking one's mother tongue may engender feelings of loss of identity. Pavlenko notes that "[d]iscourses of language and identity commonly present mother tongue as the language of the self, of the heart, of one's ethnic, national, and cultural identity, and argue that losing one's language is tantamount to losing one's self" (2005: 200). In the case of bilingual couples, moreover, it has been claimed that "the choice to become fluent in the partner's language is a symbol of willingness to give up one's own culture and language-based self in order to get along with one's partner" (Rosenblatt 2009: 14). A bilingual couple's language choice may hence also be influenced by concerns of identity, sacrifice and assimilation.

Bilinguals may not just construct their identity differently in their two languages, they may also feel and interpret things differently, depending on which language they speak. Therefore, Wierzbicka suggests that being bilingual "means to live with two different sets of concepts, and through them, different ways of interpreting human relations" (2005: 15). She is of the opinion that different languages are not only linked to different manners of feeling and thinking, but that "they are linked with different attitudes, different ways of relating to people, different ways of expressing one's feelings [… ]. They are linked with different 'cultural scripts', including 'emotional scripts'" (2004: 98). The level of emotionality a language carries for a bilingual may thus have an effect on his or her language choice, both in general and in specific situations. This will be discussed in more detail in chapter 8, "'In Swiss German, *I lieb di*, that's strange': Expressing emotions".

#### **5.2.5 Gender and family situation**

A number of researchers have proposed that multilingualism itself may be gendered, which could have consequences for the language choice of bilingual couples. It is noteworthy that even the selection of a cross-cultural partner is often influenced by gender roles and cultural stereotypes. As Piller points out, this is particularly evident in romance travel: "men from industrialized nations go on 'romance tours' to choose an overseas bride, while women from underdeveloped nations migrate to join their overseas husbands and take up residency with them" (2007: 346). Further aspects that are potentially gendered are the choice of the place of residence and of the relationship language. Piller reports that, in her sample of 47 couples living in the home country of one partner, the native language of both genders was used with comparable frequency. However, since more female partners had migrated to their partner's country (43 out of 47), a high number of females, namely 22, found themselves in a doubly disadvantaged position, having "given up their status as natives and their status as native speakers" (2000: section 4.3, para. 3). In contrast, only 3 out of 47 male partners in her sample were in a similarly disadvantaged position (see also section 3.3.3.1, "L2 learning and use"). In a different study of 13 German and English-speaking couples, Piller found that sometimes, "one partner may be stuck with all the deficient labels, 'female', 'migrant', and 'non-native speaker", while the other receives all the prestige labels, 'male', 'native', and 'native speaker'" (1999: 119).

Moreover, a person's gender may be of significance to his or her access to and view of multilingualism, as well as to the value that is attributed to his or her multilingualism. According to Piller and Pavlenko, there are "two ways in which gender structures multilingualism in [the] domains [of economic production and social reproduction]: gender structures access to language as symbolic capital, and 'doing multilingualism' may in itself be a gendered practice" (2007: 26). They postulate that "[l]anguage work, language learning, and bilingual childrearing have all become sites that are implicated in the reproduction of hegemonic gender ideologies" (2007: 27). Thus, access to languages is often limited for immigrant women, especially for members of a lower social class. Gender can also play a role with regard to the economic benefits of speaking a second language. This applies to English in Switzerland: The language carries a high economic value, yet apparently more so for men than for women. As the economist François Grin discovered, in Switzerland, men who are highly proficient in English have a significantly higher income than women who have a high level of proficiency in the language (2001: 73).

Finally, whether or not a couple has children will also inevitably influence their language choice(s) to some extent. It is conceivable that each parent's language use with his or her children might be influenced by his or her gender, too, since language socialisation and transmission tended to be allocated to the female partner in the past (Schüpbach 2009: 20). Consequently, the non-community language may be more likely to be transmitted if it is the mother's native tongue than the father's. Moreover, while many couples may not take conscious decisions relating to their joint language at the beginning of their relationship, this usually changes once they have a family. At this point, careful language planning often takes place (see Piller 2002a: ch 9). Consequently, the family situation may affect their language choice(s) in general as well as their language mixing behaviour (see section 7.6, "The couples' views on raising bilingual children").

#### **5.3 The partners' language use with each other**

As the overview of previous work has highlighted, research on the language choices of bilingual couples indicates that, depending on the personal back‐ ground of each partner and on their relationship history, there are a multitude of factors that interact in determining their relationship language. Consequently, my first aim is to see which factors are of prime importance for the ten bilingual couples within this specific setting, whether the prominence of these factors has shifted over the course of their relationship, and whether this has led to any changes in their language use. To this end, the couples were asked which language they usually speak to each other, as well as if, and why, they ever mix languages. Nine of the couples responded that they mostly use English, while only one couple (Deborah/Joshua) uses both languages regularly. The inclination towards English that most couples demonstrate can partly be explained by the selection of the participants for this study: The language of the interviews was English, and I specifically looked for couples whose relationship language was English or a mixture of both languages, as I wanted to examine proficient L2 speakers of English (though it may be noteworthy that I did not happen to come across any couples who spoke mostly (Swiss) German to each other in my search for participants). Nevertheless, it is somewhat surprising just how little German most of the couples claimed to speak to each other, considering that their place of residence is Switzerland, and that most of the native English speakers know at least some (Swiss) German; a number of them even speak the language fluently. This high propensity to use the non-community language can probably be attributed to the fact that the mother tongue of one of the partners is English. The non-community language does not seem to be an unusual choice for couples consisting of speakers of (Swiss) German and of English (see Piller 2002a; Gonçalves 2013), yet it appears to be used far less commonly when other nationalities or mother tongues are involved.<sup>3</sup> Even so, the reasons for the couples' language choice may be manifold, and I will attempt to shed light on them by discussing their answers while bearing in mind previous research on the subject.

According to previous studies, the most important factors in the language choice of bilingual couples are both partners' proficiency in their L2, their country of residence, their habits, as well as their attitudes towards both languages. Although the interplay between these factors can be quite complex, there are a number of similarities between the couples in this study with regard to these factors. When they met, all English-speaking partners, apart from Joshua <sup>E</sup> , were residing in an English-speaking country,<sup>4</sup> mostly their country of origin, and they had had little or no prior exposure to German. One of them, Richard <sup>E</sup> , had studied the language briefly at school, and David <sup>E</sup> had enrolled in a German course at university (but failed the course, see example 5.1 below), while the others had no knowledge of German. In contrast, all of the Swiss participants had studied English formally for several years, primarily at a grammar school: 7 of them had a Swiss Academic Baccalaureate ("Matura") in English, and 2 were in the process of obtaining their Academic Baccalaureate when they met. The only Swiss participant without an Academic Baccalaureate, Sarah SG, was taught English for three years in secondary school, and later attended a language school in Australia for a few months. Moreover, 6 of the Swiss interviewees were pursuing a university degree in English literature and

<sup>3</sup> As an example, Damigella, Licciardello, and Longo studied 50 mixed couples in Italy consisting of one Italian and one foreign partner with a number of different nationalities. Interestingly, 82% of the couples stated that "Italian was the main language spoken in the family, justifying this choice in relation to contingent and contextual factors" (2013: 1069) (which are not discussed further). The remaining 18% used both languages; no couples used the foreign partner's language exclusively.

<sup>4</sup> Because Deborah and Joshua's situation differs from the other couples' situations in various ways, it will be discussed separately at the end of this section.

linguistics when they met their partners. In addition to their extensive formal education in English, all Swiss partners had also been exposed to English in a natural environment. In fact, all of the Swiss participants except for Deborah SG and Sophia SG were spending an exchange year in an English-speaking country at the time of their first meeting (Sophia did, however, live in America for a year when she was 11). Considering the fact that the Swiss participants possibly with the exception of Deborah SG — were much more fluent in their L2 than their partners when they met, it is hardly surprising that they started out using English as their couple language. The interviewees' reports confirm that their language skills had the greatest influence on their language choice at the beginning of their relationship.

At some point in their relationship, however, the couples were confronted with a new situation, as the English-speaking partners decided to move to Switzerland permanently (with the exception of Tim <sup>E</sup> , who was only staying in Switzerland temporarily at the time of the interview). Thus, the community language changed, and, as a consequence, the German skills of the Anglophone partners improved. A number of them have even become fluent in German, although, according to their reports, they have not quite reached the level of skills that the Swiss partners have attained in their L2 (again, Joshua <sup>E</sup> is the exception to this). The difference in language skills may partially account for (but may also be a result of) the fact that most of the couples adhered to English as their primary couple language. Moreover, the Swiss partners might have felt more confident and comfortable using their L2 than the English speakers, due to the formers' formal education combined with their exposure to English in a natural environment. It has been suggested that a combination of natural and instructed learning tends to result in a comparatively strong sense of L2 ownership (Piller 2002a: 99), which may have been the case for the Swiss partners in this study.

Indeed, a number of participants name insufficient language skills in (Swiss) German as a reason for still using English as their relationship language. As an example, David <sup>E</sup> states that he and his wife speak English to each other almost exclusively because of his low proficiency in German:

```
Example 5.1: David (English, 32) and Susanne (Swiss, 29)
238 David: <PP<BR because ^my German is ^not uh BR>PP> \
               … <PP<BR #well par^ticularly BR>PP> _
               … <PP<BR a^mazing BR>PP> \
               <H:> <PP<HSK uh ^I HSK>PP> --
239 Susanne: = <P<HI your ^German's HI><P B1 P> \
240 David: <CRK ^yea:h CRK> \\~
               <P<HSK but uh [I ^don't] HSK>P> _
241 Susanne: [^that's:] _
242 David: <P<HSK<HI I didn't HI> ^^study German at HSK>P> _
               <P<HSK [at] .. ^school HSK>P> /
243 Susanne: [<e>] _
244 David: <PP<LO<CRK ^so uh CRK>LO>PP> _
               … <P<CRK ^I uh CRK>P> _
               <P<NAS I only ^started NAS>P> \
               well ^I _ 
               [<NAS I took <HI ^one se^mester HI>NAS>] \
245 Susanne: [@@@@] ((chuckling))
246 David: <NAS at university of ^German NAS> \ 
               <P<NAS<HSK ^a:nd uh HSK>NAS>P> _
               <P<NAS<:-)<LO I ^failed that course LO>:-)>NAS>P> _
247 Silja: @@@@@ [<@ o^kay @>] \/~
248 David: [<@ ^so @> @@@] /~
249 Susanne: [@@@] ((chuckling))
250 David: <@ [so I <HI ^stopped HI>] German @> _ 
251 Susanne: [<P<HI you ^didn't HI>P>] \
252 David: [<@ that was the end of my ca^reer @>] /~
253 Silja: [@@@@@] 
254 David: <:-) and [^then u:h] :-)> _
255 Susanne: [you didn't] <:-)<LO ^know me then LO>:-)> _ (238–255)
```
The example shows that David explains his (reportedly) rather poor L2 skills with his lack of formal schooling in German. Before meeting Susanne, he had even attempted to learn the language, but unsuccessfully, which may explain his low motivation to speak it at present. Susanne implies that his level of motivation was not high enough because he had not yet met her at that point in his life. David speaks haltingly throughout his turn, using a lot of hesitation markers (*uh* 7 times, *well* 2 times), pauses, lengthening ("a:nd" [246]) as well as false starts ("I" [244]) (all underlined). This indicates that the subject of his limited language skills in German makes him feel uncomfortable or frustrated, and probably also, in turn, less interested in speaking the language. Responses from other interviewees suggest that a low level of proficiency in Swiss German may lead to frustration in both partners, for instance if they are not able to

<sup>509</sup> Monika: <H> <P ^it's: P> \_

<sup>510</sup> Silja: <PP m^hm: PP> /

<P **it it's ^hard** P> /

<P<LO be^cause LO>P> \_

<sup>511</sup> Monika: and **if we changed to ^German** \

<P be^caus:e P> \_ <CRK for ^me: CRK> /\~

<P **to change that ^habit** P> \

<P **that would be ^awful** P> //

discuss complex subjects with each other, or if one of them feels that the other one gives up and reverts to English too quickly (see below).

Besides language proficiency, an important reason for the couples not having transitioned at least partially to the community language is that they were already accustomed to communicating in English when they took up residence in Switzerland. The participants' answers underline the important role of habit, as many of them believe that they speak English to each other "cause it was English at the start" (Richard <sup>E</sup> 253), and "because that's what [they] [… ] always did" (Katia SG 585–587). When the English-speaking partner migrated to Switzerland, most of the couples already regarded English as their relationship language and as an essential component of their couple identity. This also explains why many couples still find it difficult to switch to (Swiss) German even for a short period of time, although they would want to do so in order for the non-native speaker to improve his or her language skills. The following example illustrates this:

```
Example 5.2: Dean (English, 29) and Monika (Swiss, 29)
500 Dean: <CRK it's ^hard I mean I CRK> /
               <P I to be honest with you I ^wish that P> _
               we spoke a lot ^more //
               <P<LO ^but it LO>P> <%> \
               <H>
               ((…)) <P<BR but it's ^just BR>P> _
               it's too ^hard //
               [you ^know] _ 
501 Monika: [<H:>] 
502 Dean: <W <HI maybe HI> we do it for a couple of <HI ^minutes HI>W> \ 
               1[and ^then:] _ 
503 Monika: 1[<Hx::>] 
504 Dean: 2[the concen^tration's] gone _
505 Monika: 2[<P ^yea:h P>] \~
506 Dean: and we're back in ^English \\
               and it just ^you know \ 
               <H> [we're so ^used to speaking English] _
507 Monika: [<W I think e^xactly W>] \
               we sort of ^^grew up /
               <P<HSK together speaking ^E@nglish HSK>P> \ 
               <PP<CRK if you ^know what I mean CRK>PP> _
508 Silja: <P m^hm P> /
```
<H> <HSK **it's our ^language** <P **of ^intimacy** P>HSK> \

<%> … <W<MRC<F **es^^pecially** F> **^^High German** MRC>W> \_

```
509 Monika: <H> <P ^it's: P> _
              <P it it's ^hard P> /
              <P to change that ^habit P> \
              <P<LO be^cause LO>P> _ 
              <H> <HSK it's our ^language <P of ^intimacy P>HSK> \ 
510 Silja: <PP m^hm: PP> /
511 Monika: and if we changed to ^German \ 
              <%> … <W<MRC<F es^^pecially F> ^^High German MRC>W> _
              <P that would be ^awful P> //
              <P be^caus:e P> _
              <CRK for ^me: CRK> /\~
              … ^that is: \
              … <P<LO a ^language of of: LO>P> _
              <%> … <PP ^what's it PP> _
              <P it's <HI not [^intimate HI> at all] P> /
512 Dean: [<P m^hm: P>] /\~ (500–512)
```
<H> [**we're so ^used to speaking English**] \_

<PP<CRK if you ^know what I mean CRK>PP> \_

<sup>507</sup> Monika: [<W I think e^xactly W>] \

<sup>508</sup> Silja: <P m^hm P> /

**we sort of ^^grew up** /

Dean and Monika enumerate several reasons why they rarely speak German to each other, although Dean states that he would, in theory, like to do so (500). The couple repeat three times that it is "hard" for them to speak (Swiss) German to each other (500, 509), and suggest that, because they are so accustomed to communicating in English, it requires a great deal of concentration for them to speak a different language to each other. More importantly, English is the language of their shared past, their language of intimacy, while German lacks this emotional quality. Monika says emphatically that this is especially the case for Standard German, which is not at all an intimate language for her (511). This highlights that the diglossic language situation in Switzerland may keep the couples from shifting to the community language, as the non-native speakers initially learn Standard German, yet many Swiss do not view this as their mother tongue and might feel uncomfortable using Standard German as a language of intimacy (see section 2.3, "Diglossia and the ideology of dialect").

Other couples report similar difficulties in their attempts to switch to German, since both partners feel that this language does not allow them to establish an emotional or intellectual connection (Robert/Stephanie, Courtney/Martin, Sophia/Richard, Philipp/Karen). As an example, Stephanie SG and Robert <sup>E</sup> occasionally decide to speak German to each other in order for Robert to improve his language proficiency, but they often do not succeed in having an extended conversation, because they become increasingly frustrated with the lack of intimacy during their interactions:

 

```
Example 5.3: Robert (English-Irish, 20) and Stephanie (Swiss, 24) 
275 Robert: <P<HSK it's ^hard to HSK>P> \ 
               … to do for so ^long //
               <W because you just feel <HI ^distance HI> from each #other W> \ 
276 Silja: = <P m^hm P> /
277 Robert: = <P<HSK y- you .. you ^kno:w HSK>P> /
               <PP #^words PP> _
               <P<A<HSK exactly we're just saying these basic ^sentences like
               HSK>A>P> \
               … <H::> and then you ^can't _ 
               <P<HSK you ^can't really HSK>P> _
               … <P<CRK <%> ^touch each <LO other LO>CRK>P> _
               <P<HSK you can't really s:- ^say HSK>P> _
               <P<HSK how you're ^feeling HSK>P> \\
               <P<LO and ^so LO>P> \ 
               <P<HLT after a f:- after a d- f:- a week HLT>P> //~
               <P we <HI ^done it HI> for like P> _ ((sic))
               really <HI ^^long one time HI> _
               ^didn't /dɪn/ we \\ 
278 Stephanie: [<P ^yeah like s::- P>] --
279 Robert: <A [and then it was] really ^frustrating towards the end A> _ 
284 Stephanie: ((…)) <A it's ^just <LO<BR when we want to BR>LO>A> _
               <LO<BR discuss something im^portant BR>LO> //
               <H> ^and like \
               <BR about i^deas BR> /~
               we can't do this in ^High <LO German LO> _ (275–284)
```
Robert and Stephanie both struggle with the absence of an emotional connection and the inability to express feelings precisely, but also with the challenges of having an intellectual exchange in German.

The three examples above underline that the attitude and motivation of the partners also play a crucial role in their language choice. While many of the English speakers claim to be motivated to speak (Swiss) German in order to improve their language skills, they and their partners find it difficult to put their intentions into practice. Martin SG reports that he and his partner sometimes decide to speak German to each other for at least half an hour, but that it usually ends up being "five minutes, a few sentences" (171). In their case, Courtney's E rather basic German skills cause Martin to revert to English in order to explain grammar rules to her. Others report similar problems with switching to German, which they fail to do even for a short period of time, despite their best efforts. Besides habit and proficiency, this failure to follow through with their intentions may partly be due to the fact that there is often no real necessity for the Anglophone partner to become proficient in (Swiss) German, and that the local variety has little ideological or economic value outside of Switzerland (see chapter 2, "Language situation in Switzerland").

Another relevant factor that causes the couples to continue communicating in English is that many of the Swiss partners have a high motivation and a personal interest in speaking English with their partners. Philipp SG says that English seems an obvious choice as their couple language, since he and his wife got to know each other in the United States, and he "was there for … language purposes" (Philipp SG 233). Simon SG also states that he and Claire speak English to each other out of self-interest, as she prefers using her mother tongue at home, and he benefits from the frequent exposure to the language:

```
Example 5.4: Claire (Northern Irish, 28) and Simon (Swiss, 34)
Simon: <W I <HI think HI> it's <HI ^two reasons HI>W>_
               the f- <ph:> ^well \~
               … ((lengthy pause)) <HSK<HI pretty HI> ego^istic reasons HSK> _
               <P<HSK that work out ^fine HSK>P> /~
               <P<HSK ^she's HSK>P> _
               … <P<HSK she thinks that ^well HSK>P> \~
               <P<HSK ^I don't <LO know if I can: LO>HSK>P> _
               <P<HSK<LO if it's ^true LO>HSK>P> /~
               <P but you think you speak enough German during the ^day P> //~
               <P<HSK<LO ^so LO>HSK>P> _
Silja: @@@
Claire: @@ [@@@@] ((chuckling softly))
Simon: [<P<:-) you s- can ^speak :-)>P>] _
               <P you can speak English at ^home P> //~
               <P<HSK at [^least] HSK>P> /
Claire: [<P<HI m^hm HI>P>] /
Simon: <P<HSK ^and: um: HSK>P> \\
               … <P<HSK I teach ^English so HSK>P> \
               @@ ((chuckling))
Silja: you [##] _
Simon:
Claire: 
                 [<P I ^have a P>] _
                 <W [d- do you] [pre<HI ^fer] to speak HI>W> _
Simon:
Claire:
Simon:
                                 [<P I ^hav:e P>] /
               [<P<HSK<:-) a stay a^broad :-)>HSK>P>] _
               [<P ^English to me then P>] \
               <%> <PP<HSK<:-) all the ^time basically :-)>HSK>PP> _ (441–452)
```
Claire appreciates the fact that she does not have to speak German at home, as she uses the language at work all day. Meanwhile, Simon feels that he profits personally and professionally from speaking English with his wife. He later claims that some of his fellow English teachers even envy him a little, because he is able to practice English at home, while they only use the language at school (787). Indeed, many of the Swiss participants express very positive attitudes towards their partner's mother tongue, which are rarely matched by the Anglophone participants despite their generally positive views on bilingualism, biculturalism, and their host country (see chapter 7, "'This uh foreign girl with a great accent': Attitudes and attraction", for an extensive discussion). The Swiss participants appear to have a very high level of motivation to maintain English as their relationship language, partially owing to the high prestige of English and its status as a world language. Consequently, the couples are inclined to continue communicating in English, even if some of the Anglophone partners have reached a fairly high level of proficiency in (Swiss) German over the years. This underlines that, as the partners' relationship progresses and their L2 skills develop, language proficiency becomes less of a constraint, while habit, attitudes, and motivation gain in importance.

Motivation and attitude also play a crucial role in the language choice of the only couple who did not start out speaking English, Joshua <sup>E</sup> and Deborah SG. Their situation differs from the other nine couples' in many ways. For one, they met in Switzerland, and Joshua already spoke Standard German fairly fluently when they began a relationship. Moreover, he had personal and professional motivation to speak German due to his work as a missionary and his involvement in the church. In contrast, Deborah reports that, while she has always liked English, she also has some hesitations about speaking the language, as she did not get along at all with her grammar school English teacher, who was American. She therefore preferred to speak Standard German to Joshua when they first met. In the first few years of their relationship, Joshua and Deborah took a number of conscious decisions to change their mode of communicating, based on what was most suitable to their life situation at the time. First, Joshua decided that he wanted to learn the local variety, and Deborah started speaking Swiss German to him, until he became fluent enough to use it himself. Later on, Joshua wanted Deborah to overcome her inhibitions about speaking English, and they started speaking both languages to each other, and also mixing them. Occasionally, they would also speak dual-lingually, i. e. in their respective mother tongues (see Piller 2002a: 150). This meant that Deborah would speak Swiss German and Joshua would respond in English, a manner of communicating that was often considered odd by those around them. When they became parents, the couple decided that each of them should make an effort to use solely his or her mother tongue, in order to raise their children according to a "one parent, one language" strategy. Since then, they have reduced their code-switching considerably, at least when their children are present, though Joshua occasionally still finds it difficult to use English exclusively. As a matter of fact, he found both of their attempts at shifting towards English challenging, despite that fact that these were made on his own initiative, and that English is his native tongue. He attributes these difficulties to the force of habit:

```
Example 5.5: Joshua (American, 30)
Joshua: <P<W I <HI feel HI> like I've ^^tried to W>P> \
               <A exchange more and more to ^English A> // ((sic))
               but at ^first //
               it was <HI ^hard for me HI> _
               <PP<HSK cause ^I: HSK>PP> _
               <P<A<W was really used to speaking to her in <HI ^Swiss German HI>W>A>P> _
               <P and ^^no:w P> /~
               <P I'd ^say it's: P> _
               … <P<RH almost ^fifty ^fifty RH>P> \\ (145)
```
Despite these challenges, Deborah and Joshua have been effective in their language planning and have managed to implement their ideas to a great extent. Their example shows that, while habit is an important factor, couples can successfully change their mode of communicating if they are determined enough to do so. Moreover, it demonstrates that English is not necessarily the default language, even if it is spoken fluently by the non-native partner, as long as both partners are motivated to communicate in the other language and have a positive attitude towards it.

## **5.4 The partners' language use outside the home**

The language that the bilinguals use with the people around them is likely to have an influence on their language skills and the level of activation of the lan‐ guage, and hence also on the manner in which the bilinguals communicate with each other. For this reason, the couples were questioned on their language use in their social and professional environment. An overview of the participants' reported language use in a professional context is given in Table 2. It shows that 17 out of 19 participants use English regularly at work (Deborah SG was not included in the total number, as she is a stay-at-home mother).


Table 2: Reported language use at work

Key: E = English; SG = Swiss German; G = Standard German or Swiss German.

The majority of the Swiss participants speak English at work on a regular basis, albeit not exclusively. This can be explained by their choice of occupation, as six of them are students of English language and literature (Stephanie SG, Martin SG) or English teachers (Susanne SG, Monika SG, Simon SG, Philipp SG). One participant teaches history in English (Sophia SG), while another one is working at an airport and pursuing a degree in tourism at the same time (Sarah SG). Only two of the Swiss participants do not use English professionally on a regular basis; one of them teaches physical education at a grammar school (Katia SG), and the other one is a stay-at-home mother (Deborah SG).

The majority of the native speakers of English also use English in their professional environment, but, unlike the Swiss participants, they tend to use little or no German. Four of them work for international corporations whose business language is English (David <sup>E</sup> , Dean <sup>E</sup> , Craig <sup>E</sup> , Tim <sup>E</sup> ), one is an English teacher (Karen <sup>E</sup> ), and another one is attending a university course for international students (Courtney <sup>E</sup> ). Three of them speak both languages while pursuing their occupations as a language teacher (Robert <sup>E</sup> ), as a project manager in a small company (Richard <sup>E</sup> ), and as a student of English language and literature (Joshua <sup>E</sup> ). One of the native speakers uses mainly German (Claire <sup>E</sup> ), as she works as an occupational therapist with predominantly Swiss clients and colleagues. In this context it is noteworthy that using English as a working language is not uncommon in Switzerland, as roughly 11.5% of the Swiss population speak English professionally on a regular basis (BFS 2016a; see also section 2.4, "English in Switzerland").

The language(s) that the bilinguals speak professionally may be a matter of choice, a consequence of their occupational preferences, the result of their (current or initial) language skills, due to a lack of employment alternatives, or simply a coincidence. In contrast, the language they speak to their friends and family depends largely on the language skills of the interlocutors, though choice can also play an important role, as the expatriates might seek out other native speakers or Swiss friends who are fluent in English. In addition, the couples' geographic location might also influence their opportunities to speak English, as people in urban regions may be more internationally oriented, especially in the Zurich area, and thus tend to be more willing to speak English. Moreover, the number of residents who speak English as (one of) their main language(s) varies between the regions in which the couples are living (see section 2.4, "English in Switzerland"). Thus, in Zurich (Robert <sup>E</sup> , Dean <sup>E</sup> ) and Basle (Courtney <sup>E</sup> ), as many as 8.2% and 8.5% of the permanent residents are native speakers of English, as opposed to only 2.5% and 2.8% in the Claire's <sup>E</sup> and David's <sup>E</sup> places of residence. Table 3 gives an overview of the couples' reported language choice with their friends and family members in Switzerland.


Table 3: Reported language use with friends and family

Key: E = English; SG = Swiss German; G = Standard German or Swiss German.

Interestingly, the participants' reported language use with family and friends is largely congruent with their professional language use. There are only three bilinguals who report a different usage in the two areas: Simon SG, Dean <sup>E</sup> and Craig <sup>E</sup> all use more German privately than they do in a professional context. The majority of the participants (n=12) use both languages with friends and family members; none of the Swiss uses solely English, as opposed to four English speakers who do. Swiss German is used as the principal language by some Swiss interviewees (Deborah SG, Katia SG, Simon SG) and an English speaker (Claire <sup>E</sup> ). It should be noted that most of the Swiss participants only speak English in social situations in which their partner is present.

Moreover, the couples' responses reveal that (Swiss) German is used more frequently with family members than with friends, and particularly often with the Swiss partners' parents. The reason for this may be that the level of fluency in English tends to be higher among younger Swiss than among older generations, owing to the relatively recent introduction of English as part of the compulsory school curriculum (see section 2.2, "Multilingualism in Switzerland"). Dean <sup>E</sup> , for instance, states that his mother-in-law "really didn't speak ((… )) any English" (1236) when they first met. Thus, many of the Anglophone partners report that they regularly communicate in (Swiss) German with their parents-in-law (Dean, David, Richard, Joshua, Claire, Craig). It appears that some Swiss parents also speak (Swiss) German to their son- or daughter-in-law in order to offer them an opportunity to practice the language. David's parents-in-law, for instance, who are both English teachers and thus fluent in the language, try to encourage him to speak German — though often unsuccessfully:


Susanne's interpretation of the situation is worth mentioning, as she describes her parents (perhaps humorously) as coercing her husband into speaking German. David, however, seems to have no interest in conversing in German with his parents-in-law, maybe partly because they speak English fluently.

The social and linguistic context in which friends and family members are usually encountered also seems to be of relevance for the language use with them. Thus, some Swiss interviewees have Swiss friends with whom they sometimes speak English, even when no native speakers of English are present, and Joshua <sup>E</sup> and Deborah SG have an American sister-in-law in Switzerland with whom they both speak Swiss German. Piller reports a similar case in which family members use their L2 to communicate with each other (2002a: 166). In these cases, habit and the social environment appear to outrank the unwritten rule that prohibits speaking an L2 with native speakers of one's own language, although the fact that the interviewees specifically mention and justify their language use indicates that they themselves consider it a marked choice.

While the majority of the interviewees speak English at least occasionally, their conversational partners are not necessarily native speakers, but often Swiss who have a high level of proficiency in English. In fact, not all of the Anglophone participants maintain regular contact with other native speakers of English, which is somewhat surprising in view of previous findings. Piller, who studied German-English bilingual couples living a multitude of countries, reports that all the participants in her study, "without exception, have personal networks in the minority language" (2002a: 161), and that there is "not a single couple who do not participate in minority language networks to some extent" (2002a: 161). She believes that this is because it gives them a chance to speak their L1, but adds that "minority language friends may also be sought out for the shared experiences they may offer" (2002a: 172). However, when I asked the Anglophone partners among my participants if they had friends who were native speakers of English, and if they had specifically looked for them, only two English speakers reported that they have a large network of Anglophone friends and acquaintances (Karen <sup>E</sup> , Dean <sup>E</sup> ). Another two interviewees have some contact with native speakers of English: Courtney <sup>E</sup> has one friend who is American, while Craig <sup>E</sup> is friends with some English speakers from work. The remaining six Anglophone participants said that they do not have any friends in Switzerland who are native speakers of English, though they may be acquainted with some from work or have encountered some at social gatherings (Claire, David, Robert, Richard, Joshua, Tim). These reports differ greatly from the findings obtained by Piller (2002a). The reasons for these differences are unclear; one possible explanation is that the Anglophone partners have Swiss friends who speak English proficiently, thus compensating for friends who are native speakers.

When asked whether they specifically looked for other native speakers, none of the Anglophone partners answered affirmatively. Some who are acquainted with few or no native speakers emphasize that sharing the same mother tongue does not necessarily entail common interests (David <sup>E</sup> , Robert <sup>E</sup> , Richard <sup>E</sup> , Claire <sup>E</sup> ). David states that he has encountered English compatriots on three occasions since migrating to Switzerland, but found them rather irritating. Similarly, Robert declares that, while he misses the ease of talking to his friends at home, he does not particularly like the native speakers he has met in Switzerland. Richard, who does not have any friends who are native speakers, even becomes somewhat defensive when this topic is addressed (example 5.7 below). Frustrated, he inhales and exhales, saying that he has "just never got around to it really" (505). His wife reports that she used to encourage him to engage with an Englishman who lived nearby, but Richard expresses annoyance and disinterest in her suggestion, and retorts that this particular man has since moved away. Sophia insists that looking for other native speakers could be beneficial ("I ^^told you" [508], "you would benefit from it" [512], "you should" [514]), yet Richard remains non-committal ("yea:::h" [509], "yeah maybe one day" [511], "maybe I'll" [513]):

```
Example 5.7: Richard (English, 29) and Sophia (Swiss, 31)
505 Richard: <Hx:> <H::>
               <W s- #I've <LO just never got LO> a^round <LO to it LO> really W> \
508 Sophia: ((…)) <W I <HI ^^told you HI> to meet [the English man <LO down LO>] there
               W> /~
509 Richard: [<P<HI ^yea:::h HI>P>] \~ ((annoyed, disinterested))
               <P<HI ^yea::h HI>P> \~
510 Sophia: = @@@@
511 Richard: <P<DOW yea:h now he ^moved DOW>P> \
               @@@
               <H:> … <P<DOW yeah maybe ^one day DOW>P> \
512 Sophia: … <W<HI no well ^I think HI>W> \\
               … [you wou- you would <HI ^benefit HI>] from it \\
513 Richard: [maybe I'll ###] _
514 Sophia: <W [you should] .. ^go look for W> \
515 Richard: <P [m^hm:] P> /~
516 Sophia: [((sniffs))] ((…))
518 Richard: <P<HSK [go look] for some ^English people HSK>P> _
               <H:>
519 Sophia: <P<W<BR<MRC Anglo-^^Saxon ^^people MRC>BR>W>P> \
520 Richard: <P<:-) ^yes:: :-)>P> _ (505–520)
```
Despite their difference in opinion, the couple are very conciliatory towards the end of their mild dispute, as both of them laugh, and they also repeat, paraphrase, and finish each other's utterances (510–519, underlined). Claire <sup>E</sup> seems to be in a similar position to Richard <sup>E</sup> . She reports having been told repeatedly after her relocation that she should get in contact with an English-speaking expatriate in the area, an American woman who formerly worked at her clinic and resides in a nearby village. Like Richard, Claire was not interested in doing this at all, as she felt that they would not have anything in common:

```
Example 5.8: Claire (Northern Irish, 28)
527 Claire: <P<BR and ^I was like BR>P> \
              <P<BR who ^is she BR>P> \
              <P<HSK ^and HSK>P> _
              <H> <P she's like P> ^forty-o:dd \
              and jus' cause she speaks ^English /
              ^they were like \
              <H> … <P you ^should <%%> P> _
              … try ^yea:h /~
              <W<A get in ^contact <HI with HI> her and I was like A>W> \
              … <PP<:-) ^no :-)>PP> //~
              <PP ^you know PP> \
528 Silja: <P [^yeah] P> _
529 Claire: <H> <P [what do] I have in: in ^common [with like] P> \
530 Silja: <P [^yeah yeah] P> \
531 Claire: <P a forty-#year-#old who lives in Malans ((a village nearby)) P> /
539 ((…)) <:-) you'd rather ^find :-)> _
              <A<:-)<LO somebody you have somethin' in LO> ^common with :-)>A> //
              <A<P<LO rather than just LO> ^look ler- P>A> --
              <A<P<LO lookin' for somebody who speaks LO> ^English P>A> //
              <P at the end of the ^da:y P> \ (527–539)
```
To Claire, having the same mother tongue seems to be of lesser importance than being of a similar age and sharing interests with another person. Her narrative shows that she finds the suggestion of meeting this particular native speaker absurd; she expresses her feelings in a breathy voice (<P<BR who ^is she BR>P>), and outright rejects the idea with a high final rise (<PP<:-) ^no :-)>PP> //~) (527, both underlined). Other interviewees display a similar defensiveness as Claire to the notion of specifically looking for other native speakers. At the same time, the two examples above show that many people do attribute importance to being in contact with individuals who share the same mother tongue, as various people have encouraged Claire and Richard repeatedly to get in contact with other expatriates.

It is worthy of note that Claire has considerably less contact with speakers of English, be they Swiss or native, than the other Anglophone participants in this study. This is not merely due to her lack of interest in actively searching for English speakers, but probably also a consequence of the couple's place of residence. Claire <sup>E</sup> and Simon SG live in a small town in a rural area of the canton of St. Gall (north-eastern Switzerland). Simon says that most of their friends are reluctant to speak English, possibly because their level of proficiency is relatively low:

```
Example 5.9: Simon (Swiss, 34)
Simon: <W<HI maybe HI> it's just the mentality that we have ^here but W> \
              <P ^people who:: P> \~
              … <P<HSK don't speak English that ^well HSK>P> /~
              <PP<HSK seem to be a bit re^served HSK>PP> /
              … <P in ^doing it P> /
              <P in ^doing so: P> \~ (509)
```
Even acquaintances of theirs who speak English prefer not to use the language with them; Simon thinks that they may be intimidated because of his own high level of proficiency. There is one friend of Claire's who spent some time in Australia, and occasionally wants to practice her English with Claire. However, whenever Claire wants to tell her friend something important, she switches to German, in order to increase the pace of the conversation and to ensure that her friend understands her.

It is evident from the couples' reports that the environments of the immi‐ grants differ considerably with regard to their orientation towards English. While Claire <sup>E</sup> and David <sup>E</sup> hardly ever use English in their private or professional lives, Dean <sup>E</sup> and Karen <sup>E</sup> are in contact with many native speakers and can get by in their daily lives without speaking much (Swiss) German. Dean <sup>E</sup> , who lives and works in the city of Zurich, feels that his entire life has remained in English:

```
Example 5.10: Dean (English, 29)
Dean: <%> <W<MRC my en^^vironment ^^still MRC>W> /\~
               ^after: _
               <NAS more than five ^years of being in NAS> \
               in ^Zurich \\
               <LO<RH is ^still an English en^vironment RH>LO> \
               <P<LO for the ^most part LO>P> \
               <H:> <LO my LO> ^work environment _
               ^v:ironment is still English _ ((sic))
               <H> here at ^home is still English _
               <H> the times when it's ^^not //
               <W is: most of the time when I'm with Monika's <HI ^family HI>W> \\
               <P<HSK with her ^mum HSK>P> _ (1130)
```
The fact that Dean repeats the adverb *still* four times suggests that not only does he find this peculiar, but he may also believe that this will change at a future point in his life. It becomes apparent that, despite the relatively homogenous level of education of the Swiss partners, there are vast differences with regard to the international orientation of the environments of the couples in this study. David <sup>E</sup> and Susanne SG, who, like Claire, live in places where there are few opportunities to converse with fluent speakers of English, remedy this deficit by occasionally travelling to Zurich to meet their (Swiss) English-speaking friends from university at the pub.

The majority of the Anglophone partners have the option of speaking English in most areas of their lives, and many of them appreciate this. However, there is also a potential downside to this, as it limits their opportunities to improve their (Swiss) German, and might discourage them from speaking the language, as it is ultimately not a necessity for them. This is particularly obvious in Karen's <sup>E</sup> case. She describes how, when she first arrived in Switzerland twenty-five years earlier, and was learning German, she went into a department store and "tried [her] little baby German" (358), but the sales lady offered her help in English with a British accent. This was very frustrating to Karen and discouraged her from using the language in other situations:

**Example 5.11: Karen (American, 51)** Karen: <PP<CRK and it was ^like CRK>PP> \_ <WH okay: **give ^up** WH> \ (360)

In example 5.11, Karen expresses her resignation in a low creaky voice that turns into a whisper. Twenty-five years after this incident, she still does not feel comfortable using German in a variety of situations, for instance, when she has to speak in front of people, or when she has to write anything in German. The only person she writes to in German is her dog trainer, as she figures that the trainer's English must be worse than her own German. At home, Karen does not usually speak German either, as her children laugh at her when she does. At teachers' meetings, which take place in German, she sits with her fellow English teachers, and "of course [they] don't speak German to each other" (399). There are only a few situations in which she is forced to speak German, for instance with neighbours or the members of the local women's association ("*Landfrauenverein*"):

```
Example 5.12: Karen (American, 51) and Philipp (Swiss, 47)
405 Karen: <W and then our <HI ^neighbours speak HI>W> _
               <CRK ^German CRK> \
               <P<CRK and [^they:] CRK>P> /~
406 Silja: [<PP m^hm PP>] /
407 Karen: <P<HSK ^I have to speak German with them HSK>P> \
               <W and I <HI ^^do belong to the HI> Landfrauenverein W> \/~
408 Silja: @@@@@ ((loud))
409 Karen: = 1[<W and I <HI ^^do HI> go:: W>] \/~
410 Silja: 1[@@@@@] 
411 Karen: 2[^you know] \
412 Silja: 2[@@@] ((soft))
413 Karen: maybe th::ree or four times a ^yea:r \~
414 Philipp: [<Hx:> ((maybe @))] 
415 Silja: [<:-) ^yeah :-)>] _
416 Karin: [<BR and I ^do BR>] <Hx> //
               <BR<HI my HI> ^best BR> \\
               <P<BR<HI to speak HI> ^German BR>P> _
               <P<A<W<LO oh but it LO> <HI sometimes HI> <LO it's LO> ^terrible W>A>P> \
               <A because it's so ^loud A> /~
               <P<HSK you can't hear ^anything HSK>P> \ (405–416)
```
Her choice of words ("I *have to* speak German to them" [407], "it's terrible" [416]) emphasizes that Karen still feels uncomfortable speaking the community language. The extract also shows that the occasions in which she has to speak German are relatively rare. However, she says that she "set[s] [herself] up for at least some … activities" (426), such as reading the local newspaper, which she tries to do "^^every day" (434). Hence, Karen deliberately seeks out encounters with the German language, but apparently prefers receptive activities to productive ones. Her husband defends her reluctance to speak German, saying that she is an English teacher, and therefore need not focus on speaking German:

```
Example 5.13: Karen (American, 51) and Philipp (Swiss, 47)
363 Philipp: <W #but <HI she's HI> an <HI ^English teacher HI>W> _
              <DOW so obviously her [business] is to teach ^English DOW> //
364 Silja: [<PP ^m:h PP>] /~
365 Philipp: <P<LO 1[and not] to speak LO> ^German P> //
366 Silja: 1[<P ^yeah P>] _
367 Philipp: <H:> 2[<P ^um: P>] _
368 Karen: 2[<P ^yea:h P>] \~ 
              <Hx::>
369 Philipp: … <L plus we ^have eh L> \
370 Karen: … <CRK we have a ^network CRK> _
              <CRK of <HI [^English speakers] HI>CRK> _
371 Philipp: [<P we we have a .. ^u:h P>] \~
372 Karen: = <P<LO<HSK that's pretty [^s:trong] and HSK>LO>P> \
373 Philipp: [<CRK ^e::h CRK>] _
              <W very strong ^network W> \\
              <LO<HSK of ^English speakers yeah HSK>LO> /~
              … <P<LO<HSK inter^nationals HSK>LO>P> _
382 ((…)) <W and <HI that makes HI> it of course ^difficult W> \
              … <P<LO for: a ^partner to LO>P> \
383 Silja: … <PP ^yeah PP> _
384 Philipp: = <P<LO you know sort of ^make a LO>P> _
              <P<LO make a real .. ^effort at LO>P> \\ 
385 Silja: = <PP m^hm: PP> /~
386 Philipp: = <P I'd say a^ssimilating P> _
              <P<LO<HSK in in in terms of ^language HSK<LO>P> \ (363–386)
```
With his comment, Philipp highlights that it is not necessary for his wife to speak German in any domain of her life, neither professionally, nor privately. He also points out that the fact that they have such a strong network of English speakers is one of the reasons why their children are such "solid bilingual speakers" (380–382). Other Swiss participants (Martin SG, Simon SG) also emphasize that their partners have emigrated because of them, and not with the objective of learning the language, so they do not expect them to put much effort into learning or perfecting their L2. Nonetheless, many of the Anglophone partners express a desire to have more opportunities for practicing the language.

While limited skills in the community language can be a disadvantage in social situations, being a native speaker of English may, conversely, be an advantage in social as well as professional situations, as the language carries a high prestige, and many Swiss enjoy speaking it. Joshua <sup>E</sup> , who is working part-time at a lawyer's office while finishing his degree in English language and literature, remarks that his colleagues tend to be friendlier when they speak to him in English:

```
Example 5.14: Joshua (American, 30) and Deborah (Swiss, 28)
293 Joshua: <HI<HSK the people ^at work HSK>HI> /
               they <HI ^like to speak English with me HI> /
               <P<:-) they like to <HI ^practice speaking English HI>:-)>P> / 
               <W<:-)<LO and I've LO> also: ^reali:zed that :-)>W> \ 
               <H:>
294 Silja: @@
295 Joshua: = <:-) ^it's :-)> _
               … <W an advantage for ^^me W> //~
               @@@ <@ maybe I shouldn't ^say this @> /
               1[@@@ <@ o@n tape @> @@@] /~
296 Deborah: 1[@@@@@@@@] 2[@@@] 
297 Silja: 1[@@@@@@@@] 2[@@@] 
298 Joshua: <W cause when they <HI ^^speak HI> <LO English LO>W> _
               <P<:-) I have the feeling that they're a lot <HSK ^^nicer HSK>:-)>P> \\
302 ((…)) cause it's <HI ^^fun for them to speak English HI> _
303 Silja: <P m^hm: P> /
304 Joshua: = <P<CRK so <# that's u:h #>CRK>P> \
               … <P #if ^they P> /~
               <HSK are up^^set HSK> //
               … <W then they'll speak <HI ^Swiss HI><LO German LO>W> _
               cause [then] they can ex^press themselves _
305 Silja: [<P m^hm P>] _
306 Joshua: and ^get out \\
               <A whatever they wanna ^say A> //
               … <P<HSK but if ^I: HSK>P> \~
               <P<W speak <HI ^English HI> with them then just like W>P> \
               … <P<A ^I don't know A>P> \\
               … <P the ^atmph- P> --
               <UP the ^atmosphere changes UP> /
               <W<HSK and it's <HI ^fun HI><LO for them LO>HSK>W> _
307 Silja: <P m^hm P> \
308 Joshua: <P<HSK and ^it's like HSK>P> _
               … <PP they are ^nicer #then PP> _ (293–308)
```
Joshua emphasizes that people are "a lot ^^nicer" to him when he addresses them in English (298), and that the entire mood of the conversation changes with the language, which can be advantageous for him (306). Paradoxically, Joshua is the most fluent L2 speaker among the Anglophone partners, and the one who enjoys speaking the local variety the most (see section 7.4, "The partners' attitudes towards each other's culture and language"). For English speakers who are less fluent than Joshua, the potential social value of their mother tongue can be another hurdle in the process of learning (Swiss) German, as the preference for English of the local population might limit the expatriates' opportunities to practice the community language.

Ultimately, the language choice with other people can also influence the couples' mode of communication, both immediately and in the long term. On the one hand, the presence of other people can lead to a temporary language switch, as some couples change their language to accommodate people who are present, or to be part of the group (Joshua/Deborah, Claire/Simon, Katia/ Craig, Sophia/Richard). In other contexts, they may specifically decide not to communicate in the language spoken by the people present, in order for them not to understand their conversation, and may thus speak German when they are abroad (Sophia/Richard, Karen/Philipp). In the long term, on the other hand, the frequent use of the community language can improve the expatriate's L2 skills, and result in a higher level of activation of the language, which might, in turn, lead to an increased use of the language within the relationship. However, with the exception of occasional switches to German, this does not seem to be the case for the couples analysed here. While six of the nine native English speakers permanently residing in Switzerland report that they speak (Swiss) German regularly outside the relationship, and the other three use it occasionally, only one of them (Joshua <sup>E</sup> ) uses the language frequently to communicate with his partner. Rather, the English speakers who use German on a regular basis appreciate being able to use English at home. This suggests that, in the case of the couples in this study, the influence of their language use outside the relationship on the language use within the relationship is ultimately not as substantial as might be expected.

## **5.5 Developing a bilingual couple language**

In addition to general questions about their language use, the couples were asked if there is anything particular about their manner of communicating, and if they feel that they have, in a sense, developed their own language. All couples except for one answered affirmatively; only Karen <sup>E</sup> and Philipp SG — who incidentally have also been in a relationship the longest — could not think of anything that stands out in the way they communicate, although they believe that their faith defines the manner in which speak to each other to a certain extent. These findings mirror those of Piller's study on English and German-speaking bilingual couples. She discovered that some of the couples in her study had created their own "unique couple language", which they viewed as "the constitutive element of their relationship" (2002a: 224) and as essential to their joint identity as a couple. This also appears to be the case for the majority of the couples in my study, and I aim to explore some aspects they consider to be particular to their couple language.

There are three topics that recurred frequently when the couples discussed particular aspects of their manner of communicating with each other. These are (a) modifying their manner of speaking in order to improve their mutual understanding, (b) having achieved an exceptionally high level of communica‐ tion within the relationship as a consequence of their bilingualism, and (c) using their languages in a playful or humorous manner, for instance by inventing words, mixing languages, or imitating accents. The last one of these topics will be addressed in the context of couple humour (section 10.4.4, "Playful language in bilingual couple talk"); the other two will be discussed in the following two sections.

#### **5.5.1 Modifying one's manner of speaking**

An essential element of the communication of bilingual couples is that there is a mutual process of adaptation, which is usually particularly evident in the early stages of the relationship. While it is to be expected that the non-native speaker will attempt to adapt to the native speaker's manner of speaking and will strive to reach a higher level of proficiency, the interviewees' reports indicate that the native speaker may also modify his or her manner of speaking to some extent as a consequence of being with a non-native speaker. Tim <sup>E</sup> , for instance, states that he speaks more slowly when conversing with Sarah SG, which becomes evident when he addresses his friends after having spoken to Sarah on the telephone ("I put the phone down and start speaking fast again" [1092]). Sarah also thinks that he speaks differently to his friends, though not specifically with regard to his pace. In Switzerland, Tim attempts to speak "more clearly and distinctly" (1098), especially in the presence of Sarah's family. Robert <sup>E</sup> also reports that he started speaking more slowly and distinctly when he moved to Switzerland, and he even changed his accent so that it was easier for others to understand him. Although Robert did not change his manner of speaking because of his partner, but rather because of her environment, this ultimately affected his manner of communicating with her as well. Robert reports that has found a new "voice" for himself, a new accent that he cannot place geographically:

```
Example 5.15: Robert (English-Irish, 20) and Stephanie (Swiss, 24)
397 Robert: my ^English \\
               … <W has really [<LO changed LO> dra^matically] W> \\
398 Stephanie: [@ <H:>] ((softly))
399 Robert: <HSK<UP my ^accent UP>HSK> //
               <LO ^and the way LO> \
               <LO<HSK and the way I ^speak HSK>LO> /
400 Stephanie: <@ i@n a ^ba@d way @> /~
               @@ <Hx> [@@@@] 
401 Silja: [<@ oh ^yeah @>] /~
402 Robert: [@@] _
406 ((…)) ^no _
408 ((…)) <W<P I just beg- I <HI ^somehow HI>P>W> \
               <W<HI ^^had HI><HLT to slow down HLT>W> /~
409 Silja: … <PP m^hm PP> /
410 Robert: = <W and ^^had to gain a different <HI ^accent HI>W> _
               <A had to <HI ^find HI>[<P<HSK an accent for myself HSK>P>]A> _
411 Stephanie: [@@] ((softly))
412 Robert: <W and somehow <RH ^this ^voice ^now RH>P>W> /~
               <H:>
413 Silja: [@@] ((chuckling))
414 Stephanie: [@@] ((chuckling softly))
415 Robert: <W<:-) I don't know <HI ^what HI> it i:s :-)>W> \\
               [<:-) it's like ^somehow: :-)>] \
416 Stephanie: [@@@] ((softly))
417 Robert: … <P<HI <%%> ^I don't HI>P> /
               <:-)<HI w:here ^is it HI>:-)> \\
               <:-) it's ^like :-)> \ 
               <:-) <%%> to A^merica :-)> \
               <:-)<P to ^Eng- P>:-)> --
               <HI you know ^somehow HI> like \
               … <W what^ever it is W> /
               <H> t'is somehow what I ^fell <HI into HI> _
               <P<HSK ^just because this: t- HSK>P> --
               <W for her <HI ^friends HI> to understand me W> _ (397–417)
```
It is notable that Robert finds it difficult to describe and locate his new accent and manner of speaking. While he speaks freely and without hesitation markers about the change in his language, he speaks haltingly when he attempts to define his new accent (417, underlined). He wonders if the Irish accent that he inherited from his parents has become more American or English since his relocation to Switzerland. Stephanie is worried that Robert might view this change in his language negatively (400, underlined), and laughs (somewhat nervously) throughout his account. Robert denies this; nevertheless, the manner in which he describes his language contains a mixture of positive and negative images. On the one hand, the vocabulary he uses emphasizes that he felt forced to change his manner of speaking and had no control over it (repeating a stressed "had to" three times [408 and 410]; describing it as something that he "fell into" [417]); on the other hand, there is also some positive terminology ("gain a different accent" [410], "find an accent for myself" [410], "this voice" [412]) and he uses a smiling voice quality towards the end of the extract (<:-)>).

Other native speakers also mention that they have changed their accent or their manner of speaking. Claire <sup>E</sup> claims that she "definitely speak[s] different English" (755), because she is so used to communicating with non-native speakers. She believes that her English has not changed substantially in terms of the manner in which she speaks to her husband, Simon SG, but he sometimes tells her that she is "speakin' weird" (Claire <sup>E</sup> 769) after a day at work. They both agree that her accent is not as broad as it used to be, although, like Tim <sup>E</sup> , Claire reverts to her original accent when she interacts with her family and friends. According to Martin SG, Courtney <sup>E</sup> has also changed her manner of speaking since migrating to Switzerland. He does not specify in what regard, but states that she occasionally declares that she needs to "speak properly again" (295), or that "[she]'ll just start speaking English again, like ((… )) English people do that" (297–299). Similarly, Susanne SG thinks that David's <sup>E</sup> language is completely different — much more colourful and idiomatic — when he is in England. The couple are both of the opinion that David's English is boring when he is not in his native country:

#### **Example 5.16: David (English, 32) and Susanne (Swiss, 29)**


<P<BR<LO **de^generates to the level of the person you're** LO>BR>P> \_ <P<BR<LO **co^mmunicating with** LO>BR>P> \ (510–534)

<sup>518</sup> David: [@@@@]<A **#normally in** <HI **^techni** HI> **color** A> \_

… <P<NAS **your ^language** NAS>P> \

<H:> [@@@] <sup>519</sup> Susanne: [<HI ^yeah HI> @@] ((softly)) \~ <sup>534</sup> David: ((…)) <P<NAS so ^your: your: NAS>P> \_

```
518 David: [@@@@]<A #normally in <HI ^techni HI> color A> _
               <H:> [@@@] 
519 Susanne: [<HI ^yeah HI> @@] ((softly)) \~ 
534 David: ((…)) <P<NAS so ^your: your: NAS>P> _
               … <P<NAS your ^language NAS>P> \
               <P<BR<LO de^generates to the level of the person you're LO>BR>P> _
               <P<BR<LO co^mmunicating with LO>BR>P> \ (510–534)
```
<sup>514</sup> Susanne: = <@<HI **^boring** HI> [**English**] @> \_ ((identical pronunciation))

[<HI **^boring** HI>] \\

<sup>512</sup> Susanne: [<:-) you ^speak :-)>] / <sup>513</sup> David: = <LO **^English** LO> \_

<sup>517</sup> Susanne: [<@ ^ye:s @> @@@] \\~

The extract above is another example for the negative terminology that the bilinguals use to describe the manner in which the native speakers have adapted their language to their partners and/or their new environment. They describe it as "boring English" (Susanne SG 514, David <sup>E</sup> 511) and "speakin' weird" (Claire E 769); David even recapitulates that one's language "degenerates" (534) as a consequence of being in a bilingual relationship. In contrast, their original language is described as "speaking properly" (Martin SG 295), "colourful:" (Susanne SG 510) and even, jokingly, as speaking "in technicolor" (David <sup>E</sup> 518). However, it also appears that most of the native speakers do not, or only to some extent, attribute this change to their partner, but rather to the change in environment when they migrated to Switzerland, which is probably due to the Swiss partners' high level of proficiency in English. In fact, the development of the language within the relationship — and on the part of both partners — is described very positively by most interviewees, as will be demonstrated in the following section.

#### **5.5.2 Effective communication and implicit understanding**

Many of the couples identify their ability to communicate very well with each other as an element that stands out in their couple language, and describe communication as an essential component in their relationship. The following extract exemplifies how highly the interviewees rate the importance of commu‐ nication, and shows that they are willing to invest in it for the sake of their relationship:

<sup>516</sup> Silja: [@@@@]

```
Example 5.17: Craig (American, 32) and Katia (Swiss, 28)
432 Craig: we ^realize \/~
              ^that _
              ((…)) communi^cation i:s \
              ex^^tremely important \
433 Katia: [<PP m^hm PP>] /
434 Craig: [<HSK ^a:nd HSK>] _
              <HSK we::: .. ^keep HSK> //
              ^saying: _
              to each other and to our^selves that \
              it's ^^very important <LO that we're LO> _
              … <LO that we LO> ^^listen to [each other] _
435 Katia: [<P m^hm P>] _
436 Craig: <LO and that we LO> ^^talk //
437 Katia: 
              and that .. the: communi^cation \
              that there ^isn't u:h \
              … <P ^this u:h um: P> _
              … <PP<HSK ^um: HSK>PP> _
              we don't get .. ^too far: \
              … let's say a^part /
               <HLT before .. ^realizing that HLT> \
              <P<HSK we're having communication ^problems HSK>P> \
              <P<CRK<LO that we: LO>CRK>P> _
              … <H> <P<LO ^you know: LO>P> \
              <MRC dis^cuss what we're ^thinking MRC> \
              <P<LO ^and u:h LO>P> _
              ex^plain why: an:d \
              <H> ^and _
              … <MRC<L<HI ^this ^focus HI> probably comes be^cause we are: L>MRC> _
              = <PP m^hm PP> /
              <PP I think so [^too] PP> /
438 Craig: <HSK<DIM<DOW [^different] native .. speakers DIM>HSK>DOW> \
439 Silja: = <PP m^hm PP> /
440 Katia: = <P<HSK and I think it's a ^positive thing HSK>P> //
441 Craig: ((…)) ^this is u:h \
              <P<A very good for the [re^lationship] A>P> \
442 Katia: [<PP m^hm PP>] \
443 Craig: <P<A and so in that sense then it turns into a ^positive A>P> \
              ((very fast)) (432–443)
```
Katia and Craig both appear to be very committed to the communication that underpins their partnership, and Craig lists and highlights a number of conver‐ sational elements that he and his wife find crucial ("^^listen" [433], "^^talk", "discuss what we're thinking", "explain why" [435]). According to Craig, the partners often remind each other of the great importance of these elements, and they are intent on preventing communication problems before they even appear. His speech is very emphatic, with several lengthened expressions ("why:", "are:" [436]), sequences in marked speech (<MRC> [436]), and strong stresses ("^^very important", "^^listen" [433]) (all underlined). Interestingly, Craig believes that their different mother tongues are the reason why they focus so much on their communication, with which his wife agrees. They are both convinced that, ultimately, the fact that they had to work on their communication due to their bilingualism has had a positive effect on their relationship ("a positive thing" [Katia 440], "very good for the relationship" [Craig 441], "it turns into a positive" [Craig 443]). The couple's commitment to their own communication is mirrored in their commitment to their daughter's bilingualism (see example 7.34 in section 7.6, "The couples' views on raising bilingual children").

Thus, many interviewees believe that the communication between them is central to their relationship, and a key factor for its success. When asked to rank the importance of communication in their relationships, almost all of the participants ranked it at 9 or 10 out of 10.<sup>5</sup> The majority of the women (7 of 10) ranked it at 10, while their partners tended to rate it slightly lower on average. The only participant who rated the importance of communication lower than 9 was Simon SG, who rated it at 7. He explains that this is a point of argument in their relationship, as even though he believes that "communication is very very important, of course" (1889), he is also of the opinion that there are moments in which communicating is not necessary, and that there are topics that are not worth talking about, while his partner values talking for its own sake.<sup>6</sup> The importance of communication may also be dependent on the couple's living situation to some extent, as communication often plays a particularly prominent role in long-distance relationships. Thus, Sarah SG believes that, because they live in separate countries, communication is the foundation of her relationship with Tim, as they usually speak on the phone and cannot see or touch each other. Tim <sup>E</sup> agrees and states that, for them, communication is "^^hu:ge" (1299), and that "[their] relationship runs on it at the moment" (1304).

A related topic that recurs in the bilinguals' discourse is that they feel that they understand each other implicitly or even without words, and that other people do not necessarily understand their manner of communicating. Thus,

<sup>5</sup> Three participants decided not to respond to this question with a number, but rated (the importance of) communication "^^very important" (Sarah SG 1317), "^^hu:ge" (Tim <sup>E</sup> 1299), "high" and "obviously important" (Philipp SG 1079 and 1075).

<sup>6</sup> It should be noted that these differences in the participants' ratings of the importance of communication may also stem from different concepts of the term itself.

Courtney E is of the opinion that she and her partner understand each other so well that it seems to her that they have a language of their own:

```
Example 5.18: Courtney (English-Ghanaian, 24)
Courtney: <CRK I'm very aware of that we have a ^language CRK> \\
                <P<CRK of our ^own CRK>P> \~
                … ((2.0)) <HSK ^^not like <PF:> HSK> \\
                … <HSK different ^wor:ds <LO and things LO>HSK> _
                <P<L but because we under^s:ta::nd <LO each other LO>L>P> _
                <P<HSK<LO or understand what we we LO> ^mean sometimes HSK>P> /(300)
```
Similarly, Joshua <sup>E</sup> points to a tacit understanding between him and his partner, highlighting that they communicated very well with each other even before they were fluent in each other's languages:

```
Example 5.19: Joshua (American, 30) and Deborah (Swiss, 28)
88 Joshua: we .. under <HI ^stood each other well HI> /~
               <H> ^she'd s- --
               <DIM<DOW ^^even though we're speak- we: DOW>DIM> \~
               … <A you know my <HI German HI> wasn't that ^grea:t A> /~
               ((…)) <H:> <P<HSK ^but HSK>P> _
               <P<A ^not just the language but everything else we just A>P> _
               … <PP<HI ^we just HI>PP> _
               <P<DOW really co^mmunicated well DOW>P> \ 
89 Silja: <P<HI o^kay HI>P> \/~
90 Joshua: <PP ^we <Hx> PP> _
               … <P ^pulled P> _
               <P a ^fast one on our: P> _
               <P<HI<:-) ^parents once :-)>HI>P> _
91 Deborah: @@[@@@] 
92 Joshua: [@@@] ((very soft))
96 ((…)) <P<A we just kind of ^looked at each other like A>P> \
               .. <A<PP<:-) and ^then it was like :-)>PP>A> \
               <A<:-) we knew e^^xactly what the other person was <HI ^thinking HI>:-)>A> _ 
                      (88–96)
```
Joshua believes that a large part of their communication is not verbal, but "everything else" (88), so that they sometimes understand each other without words. A similar notion is evident in Katia's SG statement below. She reports that she and her husband have certain "codes" in their communication, which they both understand without an explanation, while outsiders may not know what they are talking about:

```
Example 5.20: Katia (Swiss, 28)
Katia: <H:> like .. certain things ^mean certain things _
              without anybody else ^knowing \
              ((…)) <F<W I mean <HI certain HI> things don't have to be ^said W>F> /
              because: _
              <A<P<DOW he understands ^anyway DOW>P>A> \
              <A<PP<DOW without me ^saying it DOW>PP>A> \
              … <HSK I just have to .. <HI ^look at him him HI>HSK> _
              <NAS and he ^^knows NAS> \~ (1095)
```
Katia later points out that monolingual couples may also have such codes. Nevertheless, it is striking how many of the bilingual couples comment on their exceptionally high level of communication, both verbal and nonverbal, and how much importance the couples attribute to communication in their relationships. Statements such as those cited above appear repeatedly in the interviews; this discourse potentially serves the couples to emphasize that their language differences and bilingualism are not impediments to their communication, but rather the contrary.

While some participants stress how much they are willing to invest in their communication, and that they deliberately work on it, others describe how easily or naturally communication comes to them. Monika SG and Dean <sup>E</sup> comment extensively on the ease with which they communicate, and ascribe this largely to the fact that they have grown into a joint language together. They express highly positive feelings about their manner of communicating:


Monika is very emphatic about her positive emotions, highlighting them with lengthening ("lov::e" [1500]), strong stresses ("^^easy" [1504]), as well as fairly short, measured intonation units. On the one hand, she states that they can talk to each other on many different levels, and have a similar sense of humour. On the other hand, she highlights the effortlessness with which they communicate ("so ^^easy", "not an effort" [1504], "natural" [1506]). This stands in stark contrast to the prevalent view that the communication between people in a bilingual relationship is often difficult or problematic. In fact, Dean and Monika even believe that their communication works so well precisely because she largely learnt to communicate in English from him. Dean claims that his partner has grown into the English language alongside growing into the relationship with him, and that, with the language, she has taken on a large part of his mentality and culture:

```
Example 5.22: Dean (English, 29)
Dean: <W<HI Monika HI> has grown into the English <HI ^language HI>W> \\
                <W but <HI ^also HI>W> \\
                … <A<W in a lot of sense it's from my <HI per^spective HI>W>A> \\
                <W or from my .. <HI ^side HI> of the English language W> _
                <P<LO ^and LO>P> _
                <H> … <W along with learning the <HI ^language HI>W> \\
                she's taken as well a ^lot /
                <CRK of the men^tality CRK> \\ 
                … <CRK from ^me: CRK> //\\~
                <NAS of ^how NAS> /~
                <W<CRK ^I <LO am I think LO>CRK>W> _
                ((…)) ^she's _
                <H> been like a ^sponge //
                <W<RH in ab^sorbing the ^language RH>W> \\ ((melodious))
                <RH<HSK she's also ab^sorbed a ^lot of the: HSK>RH> \~
                ((…)) <CRK a a ^lot that CRK> \
                <NAS that en^tails NAS> /\~ ((sic))
                <W<HSK with the language and the <HI ^culture HI>HSK>W> \\ 
                                                                           (1516)
```
Dean describes Monika as a "sponge", absorbing everything about him — his language, culture, personality. Monika agrees with this assessment, saying that she learnt how to communicate from her partner, not just with regard to the language itself, but also in terms of general communication skills:

```
Example 5.23: Dean (English, 29) and Monika (Swiss, 29)
1532 Monika: <HI ^I just know that HI> \
              <P<HI I've ^learnt HI>P> \~
              <P<W<HSK how to communicate through ^you HSK>W>P> //
              … <P<HI ^^you taught HI> me how to communicate P> \
1533 Dean: = <A<W but [<HI this is HI>] what I ^mean W>A> /
1534 Monika: [<H:>] 
1535 Dean: <NAS<W and I <HI ^don't HI> think that's: W>NAS> \
1536 Monika: = <W<HI I ^don't think that's ^language HI> dependent A> _
              <P<BR I ^think BR>P> //
              <%> <Hx> … <P<HI maybe [it ^i:s] HI>P> /
1537 Dean: <A [it's more] perso^nality A> _
1538 Monika: = <HI I [^th:ink] HI> //
1539 Dean: <W [or a <HI bit of] ^both HI>W> \\~ (1532–1539)
```
In this extract, Monika and Dean support each other verbally in a number of ways, for instance by confirming that they mean the same thing (Dean 1533), picking up on and finishing each other's sentence (Monika 1536, Dean 1537), or assenting to their partner's opinion (Dean 1539) (all underlined). Later in the interview, Monika mentions that, because she has learnt so much from her partner in terms of communication, it would be more difficult for her to communicate a problem in Swiss German than in English. The fact that they speak English to each other, that Monika has acquired many aspects relating to the language from Dean, rather than the other way around, appears to be central to the way they communicate as well as to their sense of couplehood. This becomes evident in Dean's statement in example 5.24:


This statement ties in with example 5.2 (above), where Monika describes herself and her partner as having grown up together speaking English (508). It highlights once again the fact that the language partners speak to each other may be essential to their identity as a couple, and that being in a bilingual relationship may engender both individual and joint linguistic and personal growth.

#### **5.6 Discussion and summary**

The analysis has shown substantial similarities between the couples, with regard both to their language use and to the most important factors in their language choice. Nine out of the ten couples in this study speak English to each other predominantly, rather than the community language, and only one couple regularly use both languages. Even though English may not be an uncommon choice for bilingual couples involving a native English speaker, the fact that most couples claimed to hardly use the community language is somewhat surprising considering that most of the native speakers of English are able to speak at least some German, and many of them are relatively fluent. Nevertheless, for all couples, their language skills were the key factor for their language choice at the beginning of their relationship. This is consistent with previous studies, which have also found language proficiency to be a major influence in the language choice of bilingual couples (Beraud 2016; Piller 2002a). Since most couples in this study met in Anglophone countries, and all of the English speakers only relocated to Switzerland at a later time, the latter initially had no or little knowledge of (Swiss) German, while the Swiss partners all spoke English to a certain level of fluency. The only couple who met in Switzerland is also the only couple to have spoken German to each other at the beginning of their relationship. For some couples, their language proficiency continued to be an important factor as their relationship progressed, as some of the English-speaking partners' L2 skills remained limited. For others, habit seems to be the main reason for adhering to English once it had become their established relationship language.

Language emotionality and concerns of identity emerged as further impor‐ tant influences on the couples' language use. Many couples view English as their language of intimacy and lament the fact that they have difficulties in establishing an emotional or intellectual connection to each other in German, due to custom, inadequate skills, but also characteristics they attribute to the language itself. This echoes findings reported by Piller, who discovered that the language the bilingual couples in her study used at the beginning of their relationship was closely connected to their sense of couple identity (2002a: 138). For the couples in this study, concerns of personal identity are also relevant to their language choice. The Swiss participants enjoy expressing or 'performing' themselves in their second language, and most of them regard it as part of their identity, similar to the fluent bilinguals described by Koven (1998) and Pavlenko (2001a). In contrast, most of the Anglophone partners view German merely as a medium of communication, and use it primarily for instrumental reasons. It seems that the lack of emotionality and personal connection are part of what prevents them from speaking Standard German, while Swiss German, which is viewed more favourably by most of them, is considered difficult to master. Thus, it appears that the diglossic situation in Switzerland constitutes an additional impediment to switching to the community language, especially since many Swiss do not regard Standard German as an intimate language either, and feel a closer emotional connection to English. This reminds of the case of Swiss migrants to Australia, who justified abandoning their Swiss German in terms of its perceived lack of usefulness and its non-standard status, while shifting the family language to Standard German would have entailed too great an effort (Schüpbach 2009: 22; see also section 7.2.2 "Attitudes towards languages, cultures, bilingualism and biculturalism").

For many couples, motivation and attitude are important reasons for adhering to English rather than transitioning to the community language, even as their relationship progresses. This contradicts Romaine's observation that "there is usually a shift to the majority language" in bilingual couples (1995: 42). It is possible that some couples in this study have not been together long enough for such a shift to occur, yet this is not the case for the majority of them. Like the couples in Piller's (2002a) study, several of the interviewees volunteered explanations for their language choice, which might suggest that they consider the community language the more obvious choice, even if a language shift is entirely out of the question for most of them. Many of the Swiss partners stated that they genuinely enjoy speaking English, and that they benefit personally from the frequent exposure to the language. Moreover, although some English-speaking partners would in principle like to speak (Swiss) German and to improve their language skills, they are not usually motivated enough to persist. The English speakers who do not use German regularly outside the home see no necessity to speak it with their partners; those who do speak German at their workplace feel that at least their home language should be English. The importance of motivation and attitude is also underlined by the fact that the only couple to use (Swiss) German frequently consists of the English speaker with the most positive attitude towards the language (Joshua), and the only Swiss participant who seems to have mixed feelings towards English (Deborah). While most of the Anglophone interviewees like Swiss German, Joshua is the only one who expresses genuine enthusiasm for it, and has always been highly motivated to speak the language. He and his wife have taken a number of conscious decisions regarding their language planning, have altered their couple language several times over the course of the years, and are currently speaking both languages. This demonstrates that, although habit and fluency are very important factors, determination and strong motivation can lead to successful changes in the relationship language.

The most important factors with regard to the language choice of these bilingual couples are thus language skills, habit, and attitudes. The community language and the partners' gender, on the other hand, appear to have a minimal influence in the case of these specific couples. Among the interviewees, the common observation that women are more likely to follow their partners to their native country (Piller 2002a) does not apply: 3 of the expatriates are female, and 6 of them male (one couple had not yet decided where they were going to live at the time of the interview). At the same time, 3 women and 6 men speak mainly their mother tongue, 6 women and 3 men speak their L2, and 1 woman and 1 man speak both. This, however, seems to be mainly determined by which of the participants' mother tongue is English, rather than the bilinguals' gender. Although Piller observes that women are more often placed in a "doubly marginalized position", because they emigrate and also use their L2 with their partner (Piller 2000: section 4.3, para. 3), this "double marginalization" is only the case for one of my interviewees — who happens to be male and enthuses about his L2 (Joshua). Other findings of previous research, which suggest that access to, as well as the value and economic benefits of, multilingualism may be gendered, do not apply to the couples in this study either, potentially due to their overall high level of education. For the participants in this study, therefore, gender does not appear to be a central factor in their language choice within the relationship.

The bilinguals' language use in their professional and social environment was also examined, as this was expected to shape their couple language to some extent. The responses revealed that, in these environments, the partners' language choice also depends on a number of factors, such as their language skills and those of their interlocutor, their profession, their geographic location, as well as their motivation and attitude. At work or university, the majority of the participants use English, either exclusively or in combination with German. Only two participants, one Swiss and one English-speaking, mostly use German at work. With friends and family, the majority of the interviewees use both languages (n=12), while 4 use mainly English, and another 4 use mainly German. (Swiss) German appears to be used most frequently with the Swiss partner's relatives, either because some family members are not fluent in English, or because they feel that the Anglophone partner should practice (Swiss) German. Apart from these occasions, the vast majority of the expatriates use English regularly in their daily lives, both professionally and socially. This might partly account for the fact that their relationship language has not shifted to the community language.

Interestingly, the Anglophone partners speak their mother tongue most frequently in interactions with proficient L2 speakers of English, rather than with other native speakers. In fact, only two couples claimed to have a large network of friends who are native speakers of English; another two have some friends who share their mother tongue, yet the majority do not have any native English-speaking friends in Switzerland. This stands in contrast with previous findings, in particular Piller (2002a: 171), who reported that all of her interviewees had networks in the non-community language. Furthermore, the interviewees all denied having specifically looked for other native speakers; some even emphasized that speaking the same mother tongue does not auto‐ matically lead to commonalities. The contact with other native speakers, as well as fluent non-native speakers, seems to depend greatly the couples' place of residence, as this determines the availability of such speakers. Yet, even though many of the Anglophone partners do not use their mother tongue with other native speakers, and some use German regularly, there is little indication of their daily language use leading to any major changes in the couple language.

Finally, there are also some features besides the couples' language choice which appear to shape the manner in which they communicate. One recurring topic during the interviews was the modification of the partners' manner of speaking; this is reminiscent of the frequent discourse of change and compro‐ mise observed among the bilingual couples in Piller's study (2002a: 213). Indeed, the interviewees' reports indicate that it is not only the non-native speakers who have adapted their speech to their partners'; the native speakers of English have also modified their manner of speaking as a consequence of being in a foreign-language environment. Several of them reported that they now speak more slowly and articulately, that they have developed a new and less distinct accent, or that they use less idiomatic language. Even though these changes may be due to their environment rather than to the non-native partner, this new manner of speaking tends to seep into their couple language. Some (but not all) bilinguals described their new manner of speaking in deficient terms as "weird", "boring", "less colourful", or not "proper".

In contrast, the couples expressed overwhelmingly positive views on their communication and believe that they have achieved a very high level of communication as a consequence of their bilingual situation. For many of them, their ability to communicate, both verbally and nonverbally, is a key component of their relationship and central to its success. On the one hand, some couples reported that they invest in and work on their communication; on the other hand, there are couples who emphasized the naturalness and ease with which they communicate, and the almost implicit understanding between them. In this manner, the couples challenge the notion that their bilingualism may be an impediment to their communication and portray it in a positive light. Their reports suggest that, by committing to the communication between them, bilingual couples can develop a great rapport and increase mutual understanding.

In conclusion, there are several elements that constitute the couples' mode of communication: their language choice, their language mixing, their individual manner of speaking, and the verbal and nonverbal communication between them. Many of the couples in this study are convinced that they have grown together in their couple language, and have developed a unique couple language, which they consider vital to their couplehood, just as Piller (2002a) observed in the bilingual couples she studied. While changes in professional and personal circumstances may lead to changes in the partners' language use outside of their relationship, the couple language seems to be much more stable. The analysis corroborates previous findings that the language a couple chooses early on in their relationship can be expected to influence their interaction permanently, as major changes to an established couple language tend to be rare. The only times when changes are more likely to occur are when one or both partners relocate to another country, or if they start a family. The moment the first child is born, the mode of communication of the couple has to be adapted to a triadic situation, and develops from a couple language into one or several family languages.<sup>7</sup> While the partners are unlikely to shift completely from one language to the other, it might mean that they start to mix languages or use them alternately in certain situations. This will be discussed in more detail in the next chapter.

<sup>7</sup> The development of such a family language will be discussed in section 7.6, "The couples' views on raising bilingual children".

## **6 "German sorta creeps into it": Language mixing**

## **6.1 Introduction**

As the previous chapter has shown, language choice is often not merely a matter of selecting one language or the other, but languages may also be combined, or used in alternation. While the overall choice of a couple language tends to be relatively permanent, bilinguals also make spontaneous language choices, and may consequently switch between their languages. In the current chapter, I aim to investigate such situational or temporary language mixing among fluent bilingual couples. I will analyse the couples' language mixing behaviour during the interviews, and will also examine their reports on their habitual language mixing and their attitudes towards it. I am interested in determining in which situations the partners switch, in the context of which topics switches occur, and whether there is a preference for shorter or longer switches. Moreover, I intend to examine whether the interviewees' language mixing behaviour correlates with other factors, namely their family situation, their mother tongue, and their gender.

To my knowledge, the only aspect of language mixing that has received considerable attention with regard to bilingual couples is their attitude towards mixing, whereas their language mixing behaviour itself has rarely been studied to date (see section 6.2.2, "Language mixing in bilingual couples"). Rather than focusing on bilingual couples, research on actual language mixing has mostly examined the mixing practices in bilingual language communities (Poplack 1980; Gardner-Chloros 1991) — usually among relatively fluent or even balanced bilinguals — as well as language mixing in the context of bilingual language acquisition (Lanza 1997). Bilingual couples tend to differ from these in that they are established communicative dyads, who have probably developed their mixing behaviour jointly over time. The bilingual couples in this study, moreover, are also characterized by a discrepancy between the partners' ability in each other's language, and in many cases, only one of them is a fluent bilingual. My analysis will demonstrate to what extent and in what manner bilinguals mix languages in these specific conditions.

Language mixing can take a variety of forms, and switches may range in length from a single word to entire sentences or extended turns. I am using the term *language mixing* as a cover term to include all types of temporary language switches. The type of switch, as well as the amount of hedging accompanying the switch, may be indicative of the speaker's level of language proficiency, but also his or her attitude towards language mixing. I therefore distinguish between shorter and longer switches in my analysis, as well as between hedged and unhedged switches. Shorter switches that only involve a single expression are referred to as *borrowings*. These are similar to *loanwords*, except that they occur spontaneously, while loanwords have usually become established in the vocabu‐ lary of the target language, and tend to have been adapted phonologically and morphologically.<sup>1</sup> In contrast, I use the term *code-switching* to refer to extended switches, i. e. entire phrases or sentences (see section 6.3.2.2, "Code-switching vs. borrowing", for a more detailed discussion of the distinction). Whenever the distinction is not relevant, I refer to both code-switches and borrowings as *switches* in this chapter.

In the following section, I first give an overview of previous work on language mixing (6.2). Since language mixing has rarely been examined in bilingual couples, the majority of the findings that are discussed are not specific to language mixing of bilingual couples, but apply to other speech communities; even so, the overview serves to explain some basic principles underlying language mixing which are necessary for the analysis. I then look at the couples' language mixing during the interviews (6.3). For this purpose, I set up a taxonomy of potential triggers and reasons for switches, which helped me to determine in which situations switches happened most commonly, in which situations these switches were hedged, and which types of switches were used most frequently. I also look at similarities between the partners, and explore the influence of gender, mother tongue and family situation on the bilinguals' language mixing. Finally, I discuss the interviewees' reports on their language mixing behaviour, as well as their attitudes towards language mixing (6.4), and compare these with their mixing behaviour during the interviews as well as with previous research (6.5).

<sup>1</sup> Speakers may borrow individual expressions from one language while speaking the other, by borrowing either the form and content of a word (*nonce borrowings*) or only the content (*loan shifts*) (Grosjean 2001: 6). I will only consider the first type.

#### **6.2 Previous work on language mixing**

#### **6.2.1 Factors that influence language mixing**

Previous research indicates that when, in which manner and how extensively a bilingual switches languages is determined by a number of factors. One of the most important factors seems to be the bilingual's interlocutor and speech community. In an exploratory study, Poplack analysed the code-switching behaviour of a fluent Spanish-English bilingual, and found that the extent to which she mixed languages depended greatly on her conversational partner. The bilingual switched four times as frequently in conversations with an in-group interlocutor than with an outsider, and used far more intra-sentential switches with the former than the latter (1981: 179–180). Extensive switching behaviour was also found by Gardner-Chloros in a case study on German-French bilinguals in Alsace, France. In subsequent interviews, many of them explained their frequent language mixing by referring to their interlocutor's switching behaviour (1991: 98–99). Gardner-Chloros concludes from this that language mixing can be a way of accommodating one's interlocutor, yet it can also be a compromise in terms of language choice (2009: 78).

Language mixing behaviour may also be influenced by the personality of the speaker, the topic of the conversation, as well as language-specific characteris‐ tics. This can be seen in Poplack's (1980) study of the code-switching behaviour of Puerto Rican bilingual speakers of Spanish and English residing in the USA, many of whom were balanced bilinguals and had a habit of code-switching frequently with each other. Poplack discovered that some bilinguals in her study preferred extra-sentential switches, while others used more intra-sentential ones, which could partly be attributed to the speakers' personality (1980: 240). Similarly, Gardner-Chloros found that, particularly in intensive code-switching, there is a strong individual personality factor (1991: 161). The topic of the conversation, moreover, may trigger language mixing at a specific point in the interaction (Schmid 2005: 140), for instance if the subject matter is linked to key cultural concepts. Thus, bilinguals sometimes shift from one language to another because they sense that "the meaning that they want to express 'belongs' to the other language" (Wierzbicka 2004: 102). Switches can also occur as a result of similarities between the languages. Hence, the triggering hypothesis brought forward by Clyne (1967; 2003) suggests that there are certain expressions, such as cognates or bilingual homophones, which potentially trigger switches. This hypothesis was tested empirically by Broersma and De Bot, who found evidence that code-switches may indeed be facilitated by trigger words in a small-scale study on the code-switching of Dutch-Moroccan Arabic bilinguals (2006: 11).

In addition, the amount and type of language mixing that appears in the speech of a bilingual can also be influenced by his or her level of fluency in both languages. In particular, it seems that the level of proficiency influences the manner in which the languages are mixed. Thus, Poplack found that bilinguals who are dominant in one language tend to use this language as a base from which they switch into their L2, whereas balanced bilinguals alternate between both languages (1980: 597). She concludes from her sample that

alternation between languages requires a high level of bilingual competence. Code-switching involves enough knowledge of two (or more) grammatical systems to allow the speaker to draw from each system only those rules which the other shares, when alternating one language with another. Surprisingly enough, this knowledge appears to be shared by even the non-fluent bilinguals in the sample. (1980: 601)

Hence, Poplack posits that, rather than being a sign of low language proficiency, code-switching behaviour may actually be an indicator of bilingual competence (1980: 616). This is only the case for code-switching, though, and not for all types of language mixing, as borrowing can occur in the speech of less fluent bilinguals, and even monolinguals (Pfaff 1979: 295).

Other important factors that influence language mixing are the attitude of the bilingual and the speech community towards both languages, and towards language mixing in general. Attitudes towards language mixing vary considerably, and mixing used to be viewed negatively in the past. Thus, there are numerous reports of language mixing (or even bilingualism) being discouraged by professionals such as speech therapists, doctors, or teachers (Romaine 1995: 237). Even among bilinguals themselves, language mixing is often viewed negatively and attributed "to lack of education, bad manners, or improper control of the two grammars" (Gumperz 1982: 62). Nevertheless, the general perception of language mixing appears to have become more positive in recent years, and many bilingual individuals and language communities do not only enjoy mixing languages, but consider it "a legitimate style of informal talk" (Gumperz 1982: 62).

The amount of mixing and the manner in which languages are mixed seem to depend greatly on the conventions and habits of the speech community in question. A comparison between the language mixing of a Puerto Rican community in the USA and French-English bilinguals in Quebec by Poplack (1988) demonstrated that the two speech communities differed considerably in their language mixing behaviour as well as their views on language mixing. While many of the Puerto Rican bilinguals had a positive attitude towards language mixing and code-switched frequently, often without hedging their switches, the French-English bilinguals appeared to view it more negatively, hedged their switches, and only mixed languages when there was an obvious reason for it.

The French-English case presented made clear another point: evaluation of the equivalence or any syntactic constraint is a fruitless pursuit in situations where 'smooth' code-switching is not a community-wide discourse mode. Here English use as well as speaker attitudes towards it are consistent with highlighting, flagging or otherwise calling attention to the switch. Indeed, in order for the switch to accomplish its purpose — be it metalinguistic commentary, finding the *mot juste*, providing an explanation and so on — it must be salient, and should not pass unnoticed. (Poplack 1988: 238)

Poplack's comparison suggests that switching does not happen randomly, but usually has a purpose, be it to reinforce one's group identity or to express oneself more precisely. This view is shared by others who believe, for instance, that "bilinguals often switch varieties in order to communicate something beyond the superficial meaning of their words" (Gardner-Chloros 2009: 4).

Language mixing can also be influenced by the social class, level of education and gender of the bilingual, though these factors are difficult to discern as they are often contingent on other variables. Thus, research on the role of gender in language mixing has been inconclusive, as it appears to depend on the specific speech community. In a study comparing Greek Cypriot and Punjabi immigrants in the UK, Cheshire and Gardner-Chloros found no significant differences in the code-switching behaviour of the sexes in either community (1998: 28), though gender differences have been identified in other communities, for instance among the Puerto Ricans studied by Poplack (1980: 608). Gardner-Chloros hypothesizes that any gender-specific use of code-switching may depend on its function, and that women possibly code-switch more frequently in order to mitigate directness, to create humour, or to bond with their interlocutors (2009: 85–86). Social class or education, on the other hand, can have an indirect influence, as they may affect attitudes towards language mixing. Thus, a study of London Greek Cypriots found that well-educated bilinguals viewed code-switching less favourably than bilinguals with a lower level of education (Gardner-Chloros, McEntee-Atalianis and Finnis 2005: 71).

Finally, an important aspect to consider when one examines spontaneous language mixing in fluent bilinguals is whether the speaker is in a monolingual or a bilingual speech mode at the time of speaking. The concept of speech modes was proposed by Grosjean, who believes that there is a continuum of different speech modes ranging from monolingual to bilingual. Speakers are likely to be in bilingual mode "when they are interacting with other bilinguals who share their two languages and with whom they feel comfortable mixing languages" (Grosjean 2001: 4). In this mode, "the bilingual speaker chooses a base language, activates the other language and calls on it from time to time in the form of code-switches and borrowings" (Grosjean 2001: 2). In monolingual speech mode, on the other hand, bilinguals deactivate one of their languages, though never completely, which results in little or no language mixing. Speakers may also take an intermediate position, "when, for example, the interlocutor knows the other language but either is not very proficient in it or does not like to mix languages" (Grosjean 2001: 4). Grosjean (1997) tested his theoretical model in a laboratory-based study, in which French-English bilinguals were asked to retell French stories that included English code-switches to three virtual conversa‐ tional partners with different linguistic backgrounds. The study demonstrated that the participants were in a more bilingual mode with imaginary interlocutors who were fluent bilinguals, as they hesitated less and code-switched more. This confirms that bilinguals adjust their language mode and, consequently, their language use, to suit their interlocutor's linguistic situation (Grosjean 1997). The bilingual language modes may be a confounding variable in research, which was potentially the case for Poplack's (1988) study, as one speech community was recorded in a more formal context than the other. The bilingual couples' language modes are therefore an important variable that should be kept in mind for the analysis of their language mixing behaviour during the interviews.

#### **6.2.2 Language mixing in bilingual couples**

In one of the few studies on language mixing in bilingual couples, Pietikäinen (2014) examined the language mixing behaviour of six couples who use English as a lingua franca (ELF). All of the couples in her study reported that they mix languages (2014: 12). Many of the switches in her interviews were unmarked and seemed to happen automatically, from which the author concludes that the couples have a relaxed attitude towards language mixing (2014: 21). In contrast, many of the couples in Piller's study expressed negative attitudes towards language mixing, and only 30% of them claimed to mix languages, though many of them actually did so during their conversations (2002a: 143–148). At the same time, several of the bilinguals who had expressed negative opinions "simultaneously evaluate[d] mixing positively", as their two languages allow them to express different thoughts and feelings (2002a: 148), which suggests that they may attach a covert prestige to language mixing. These two reports on language mixing in bilingual couples underline that there is a large variance in couples' attitudes and mixing behaviour, depending on their combination of languages and linguistic environments.

#### **6.3 The couples' language mixing during the interviews**

#### **6.3.1 Language mixing in an interview situation**

When analysing the couples' language mixing behaviour, it is important to bear in mind that the language mixing they displayed in the interview situation may not precisely reflect their mixing habits when they are on their own. On the other hand, there were situational factors that potentially inhibited language mixing. English had been set as the base language of the interviews, and it is possible that the participants made an effort to adhere to English due to the perceived formality of the situation; this is supported by the fact that, as soon as we were off the record, many of the Swiss bilinguals switched to Swiss German when they addressed me directly, unless the topic concerned their partner. On the other hand, there were several variables that potentially encouraged language mixing. For one, I was present during the interviews, and the participants were all aware that I am a native speaker of Swiss German. Moreover, I had communicated with many of the Swiss participants in (Swiss) German prior to the interviews,<sup>2</sup> and I had informed all of the couples that they could switch to (Swiss) German whenever they desired to do so during the interviews. Consequently, it is to be expected that the participants' bilingual language mode was at least partially activated (Grosjean 2001). Switches to (Swiss) German may also have been triggered by some interview questions, particularly when they concerned cultural or linguistic topics. Overall, it may be assumed that the variables promoting and inhibiting language mixing balanced each other out to some extent; this assumption is further borne out by the fact that the analysis of the couples' language mixing in these specific circumstances largely corresponds to their reports regarding their usual mixing behaviour.

<sup>2</sup> I was acquainted with some of the participants before the interviews; the language of communication had usually been Swiss German when speaking to the Swiss partner, and English when speaking to the English-speaking partner or both partners.

#### **6.3.2 Categorisation of language switches and research questions**

In this chapter, my aim is to explore the participants' language mixing behaviour by examining the frequency of switching during the interviews, the types of switches they used, and the topics which triggered switching. I also intend to ascertain whether similarities can be discerned in the language mixing behaviour of the partners, and whether there were any correlations between switching and mother tongue or gender. To this end, I categorized all switches to (Swiss) German that occurred during the interviews. I decided to also include words that are not technically German expressions, but that were uttered using German pronunciation and phonology and/or contained German elements, for instance *Miörgler* (an invented term of endearment; Courtney <sup>E</sup> 628), or loanblends like *^blade-bar:* 'blade-able' (i.e. doable with rollerblades; Joshua <sup>E</sup> 1170). In the case of cognates or homophones, the pronunciation was taken to be indicative of the language in which they were spoken (thus *Nicole* /ˈnɪgol/ [Karen <sup>E</sup> 887], for instance, was regarded as a borrowing, even though it is a personal name). Four different aspects were considered with regard to the couples' switches, namely (a) whether they occurred spontaneously or in the context of metalinguistic discourse; (b) whether they were code-switches or borrowings; (c) whether they were hedged or unhedged, and (d) what triggers appeared to have caused them. I will elaborate on the first three categories in the following sections, before proceeding to the analysis of the switches that occurred during the interviews.

#### **6.3.2.1** *Spontaneous vs. metalinguistic switches*

Due to the interview situation, I deemed it important to distinguish between switches that happened spontaneously and switches that were prompted by interview questions or served as metalinguistic examples (labelled "s" [*sponta‐ neous*] and "m" [*metalinguistic*] in the following tables). Spontaneous switches served, for instance, to fill a lexical gap or to refer to cultural concepts; such a switch can be seen in example 6.1, where Claire initially does not remember the English word for *sledging* and uses the Swiss German term instead:

```
Example 6.1: Claire (Northern Irish, 28)
Claire: we ^went em: \\
             <PP<HI ^Schlitteln HI>PP> // (('sledging'))
            … <P<HI ^sledgin' HI>P> // (300)
```
In contrast, the category of metalinguistic switches includes, for instance, responses to a question concerning the partners' nicknames for each other or the swearwords they use, as well as metalinguistic references to particular Swiss German terms, and instances in which expressions were quoted. The following example demonstrates this:

**Example 6.2: Susanne (Swiss, 29)** Susanne: <PP<HSK that's because in ^Swiss German HSK>PP> \_ <PP<BR *I ^lieb di* BR>PP> \ (('I love you')) <PP<HI that's HI> ^strange PP> /~ <P ^isn't it P> // (1054)

At least some of the metalinguistic switches may be attributed to the subject of the interviews, suggesting that there might have been fewer switches if the conversations had revolved around different topics.

#### **6.3.2.2** *Code-switching vs. borrowing*

The second aspect I took into consideration was whether the switches concerned single expressions (borrowings), or if they were more extensive (code-switches). In the first category, I included all switches that referred to a single concept. Usually they consisted of one word in German, though the English translation may involve several words ("and I ^^do belong to the *Landfrauenverein*" 'rural women's association' [Karen <sup>E</sup> 407]). Two-word expressions referring to a single concept were counted as borrowings, like *Britische Matura* 'A-levels' (Joshua <sup>E</sup> 696) or *Migros butschi* (a humorous corruption of *Migros budget*, the name of the economy range of the largest Swiss retailer [David <sup>E</sup> 819]). Words that occurred several times were counted individually unless they appeared in immediate succession in the same turn (*schnell ^schnell* 'quickly' [David <sup>E</sup> 1373]). In the case of all of these borrowings, the base language of the conversation did not shift to German, but instead individual terms (or several consecutive ones) were inserted into the English base language. As Cheshire and Gardner-Chloros note, such switches are generally less stigmatized, as they are akin to established borrowings or loanwords, which are also used by monolinguals (1998: 21). Established loanwords were not considered, though, as I was only interested in spontaneous speech borrowings.

Code-switching, in contrast, which has been defined as "the alternate use of two codes in a fully grammatical way, in the same discourse" (Poplack 1988: 44), usually involves a shift of the base language, even if this may be only brief. Occasionally, it is difficult to distinguish between code-switches and borrowings; as Poplack points out, "[t]he smaller the switched constituent […  ], the more difficult it is to resolve the question of whether we are dealing with a code-switch or a loanword [i.e. borrowing]" (1988: 47). I categorized switches as code-switches whenever they consisted of a phrase (*Ich habe eine Katze* 'I have a cat' [Claire <sup>E</sup> 100]), even if some of them were short (^*kennsch da* 'do you know that' [Susanne SG 1489] or *du ^tubel* 'you idiot' [Susanne SG 945]). Most code-switches only lasted a single turn or less. For the few code-switches that exceeded a turn, it was difficult to decide whether they should be counted as a single code-switch or several ones. This can be seen in example 6.3:

```
Example 6.3: Craig (American, 32) and Katia (Swiss, 28)
Craig: … ^or that _
              when[she came to the U-] --
Katia: <FF [ah die ^Sieche] mached ^süchtig FF> \ (('these suckers [the cookies] are addictive'))
              1[@@@] 
Silja: 1[@@@@] 
Craig: 1[@@@@] ((softly))
Katia: 2[<@ Das isch jetzt ebe <HI my native tongue HI> gsi @>] \ (('that was my native tongue now'))
Silja: 2[##### @@@] 
              3[@@@@@] 
Katia: 3[@@@@@] 
Craig: 3[@@@@@] 
              ^um: _
              4[or ^the um:] \
Katia: 4[@@@] ((chuckling))
Craig: or that she only ^knew uh [one year] --
Katia: <F [das isch wichtig] für d ^Uswärtig F> _ 
              (('that's important for the evaluation')) (912–922)
```
In example 6.3, Katia interrupts Craig with a code-switch to Swiss German, which is met with laughter, and she then continues for a second turn (which technically includes an English borrowing). Craig carries on in English, yet Katia adds another turn in Swiss German. In this case, I counted the first two of Katia's turns as one code-switch, and the last turn as a separate one, as it was preceded by an English turn. Thus, switches lasting for several turns were considered as one single switch as long as they were not interrupted by a turn in the other language.

#### **6.3.2.3** *Hedged vs. unhedged switches*

Both types of switches — borrowings and code-switches — may be *hedged* (flagged) or *unhedged* (fluent). Hedged switching, which is accompanied by features such as hesitation markers or metalinguistic comments, is potentially linked to low L2 proficiency, to a lack of habitual switching, or also to a negative view of language mixing. A hedged switch can be seen in the following example, in which Craig hesitates and pauses before switching to German:

```
Example 6.4: Craig (American, 32)
Craig: <P<HSK i- ^in uh: HSK>P> \
                <P<HSK ^um: HSK>P> _
                … <P<A<DOW ^Berner Oberland DOW>A>P> \ (('Bernese uplands')) (533)
```
Unhedged switching, in contrast, is characterized by smooth transitions be‐ tween both languages and is "unmarked by false starts, hesitations or lengthy pauses" (Poplack 1980: 601). It can — but does not have to — be indicative of a high level of bilingual proficiency, of habitual switching, or of a positive attitude towards language mixing. This is shown in example 6.5, where Karen transitions to German smoothly and without flagging her switch:

```
Example 6.5: Karen (American, 52)
Karen: <W<HI Philipp HI> doesn't believe in <LO ^Ordnung LO>W> \//~
             (('order, tidiness, regulation')) (467)
```
Since hedging can point to the bilinguals' proficiency as well as attitudes, I distinguished between hedged and unhedged switches in my analysis. Switches were counted as hedged if they were accompanied by hesitation markers such as *uh* or *um*, long pauses, or repeated words. Short pauses between turns were not considered hedges, as they were relatively common throughout the interviews.

#### **6.3.2.4** *Switches to Swiss German vs. Standard German*

One aspect that was not considered in the analysis was whether the language switches were to Swiss German or to Standard German. The reason for this is that the two varieties can be difficult to distinguish for short expressions, and that extended code-switches at times consisted of both varieties. While longer switches were usually to Swiss German in the case of the Swiss participants, speakers of English tended to use either or both varieties when code-switching (depending on their level of proficiency in Swiss German). However, it is worthy of note that many of the borrowings were taken from Standard German, rather than Swiss German, even by the Swiss participants. To give an example, Simon SG, whose partner speaks Swiss German fluently, uses the Standard German term *Ausländerausweis* 'identity card for foreign nationals' [252] instead of *Usländeruswiis*. 3 It is possible that some Standard German expressions have become established as part of the couples' vocabulary, that they are easier for the non-native speaker to understand, or that the bilinguals perceive a switch to Standard German to be more formal and therefore preferable in an interview situation.

#### **6.3.3 Overview of language switches in the couples' conversations**

All of the switches that occurred during the interviews were divided into the categories discussed above. Table 4 below gives an overview of all instances in which the participants switched to (Swiss) German, indicating the type of switch (borrowing or code-switch), the circumstances (spontaneous or metalinguistic), and hedging (hedged or unhedged).


Table 4: Overview of number of switches of each type during the interviews

Table 4 demonstrates that, overall, there were relatively few language switches during the interviews. The total number of switches is 306 (bor‐ rowings or code-switches), but many of those switches were used in a metalinguistic context (n=141), which means that there were only 165 sponta‐ neous language switches. Consequently, the total number of words in (Swiss) German that were used in non-metalinguistic situations only constitutes a small percentage of all words (roughly 0.24%).<sup>4</sup> What also becomes imme‐ diately apparent in Table 4 is that the bilinguals used considerably more borrowings than code-switches. The participants borrowed 150 expressions spontaneously and used 114 in a metalinguistic context (n=264 borrowings in total), while they code-switched 15 times spontaneously and 27 times in a

<sup>3</sup> As there is no standardized version of Swiss German, Swiss German terms are written in the manner in which a Swiss person from the linguistic region of the speaker might write them (mostly Zurich German).

<sup>4</sup> The bilinguals used 150 borrowings and 15 code-switches (93 words) spontaneously; 103,074 words were used in total.

metalinguistic context (n=42 code-switches in total). This makes an average of 7.5 spontaneous borrowings and less than one spontaneous code-switch (n=0.75) for each speaker. With regard to the code-switches, it should be noted that they were usually rather short, and only consisted of 4.2 words on average. In fact, only six of the participants code-switched in a non-metalinguistic context — three Swiss and three English-speaking participants — and even in these cases it was only for a few words or sentences. Moreover, there was no instance of what has been called *mixed discourse*, in which the two languages are mixed seemingly meaninglessly, or the codes appear to converge (for examples of mixed discourse, see Gardner-Chloros 2009: 1–3). The fact that the bilinguals rarely mixed their two languages is probably tribute to their high level of proficiency in the couple language.

The couples' hedging behaviour gives an indication of their attitudes towards language mixing. As can be seen in Table 4, there were far more unhedged than hedged switches during the interviews — there were 46 hedged versus 260 unhedged switches, which means that only 15.0% of all switches were hedged. Not surprisingly, switches that occurred spontaneously (30 of 165 [18.2%] hedged) were hedged more often than those that took place in a metalinguistic context (16 of 141 [11.3%] hedged).<sup>5</sup> The small percentage of hedging suggests that, when the speakers changed languages, they did not view this negatively, and may have felt that a switch was necessary, or even desirable. It might also be linked to the context of the switches; only very few switches happened as a result of a speaker's inability to express him- or herself in the language of the conversation, but many of the switches were motivated by the topic of the conversation and therefore brought about by a conscious decision (see also section 6.3.5, "Triggers in the couples' language mixing"). It should further be noted that, even if a switch co-occurs with hesitation markers, the latter are not necessarily due to the switch itself, but can also have other causes, such as the use of a taboo word, or the need to think of an example word or sentence.

#### **6.3.4 Similarities in the partners' language mixing**

When the couples' switching behaviour is considered individually, it becomes evident that, while there were considerable differences in the mixing behaviour of the individual speakers, the participants often displayed language mixing

<sup>5</sup> This difference might have been even greater if the metalinguistic switches had not included a discussion of swearwords (which might have been hedged for other reasons), or a prompt to think of example sentences or words.

behaviour similar to that of their partner. As can be seen in Table 5, which gives an overview of the different types of switches for each participant relative to the number of words he or she used, this is the case for both metalinguistic and spontaneous switches.


Table 5: Overview of types of switches for each participant (normalized per 10,000 words)

Key: CS = code-switch; B = borrowing; m = metalinguistic context; s = spontaneous use.

Table 5 demonstrates that there are some couples who mixed very little, like Robert/Stephanie and Monika/Dean, who each only used between 2.0 and 3.3 spontaneous switches per 10,000 words. In contrast, Susanne SG and David <sup>E</sup> both used spontaneous switches comparatively often, namely 43.0 and 35.4 times per 10,000 words. The largest difference in spontaneous switching was between Claire <sup>E</sup> and Simon SG (29.4 and 8.8 switches per 10,000 words) and Joshua <sup>E</sup> and Deborah SG (28.4 and 67.1 switches per 10,000 words). The couples' use of spontaneous switches is visualized in Figure 1 below.

is visualized in Figure 1 below.

Table 5 demonstrates that there are some couples who mixed very little, like Robert/Stephanie and Monika/Dean, who each only used between 2.0 and 3.3 spontaneous switches per 10,000 words. In contrast, Susanne SG and David <sup>E</sup> both used spontaneous switches comparatively often, namely 43.0 and 35.4 times per 10,000 words. The largest

**Figure 1.** Number of spontaneous switches for both partners (normalized per 10,000 words), including length of their relationship Figure 1: Number of spontaneous switches for both partners (normalized per 10,000 words), including length of their relationship

The similarities in the partners' mixing behaviour might have been influenced by the conversational flow, but they

probably also indicate that the partners have assimilated their mixing behaviour to each other. 6.3.5Triggers in the couples' language mixing The similarities in the partners' mixing behaviour might have been influenced by the conversational flow, but they probably also indicate that the partners have assimilated their mixing behaviour to each other.

but also the type of switch that is used. For my analysis, I tried to identify the main or most likely trigger or reason for each switch, and categorized the switches accordingly. This categorisation renders it possible to determine

#### Switches can be initiated by a variety of triggers, which may influence not only the bilinguals' hedging behaviour, **6.3.5 Triggers in the couples' language mixing**

#### whether there are any differences in the bilinguals' specific switching behaviour based on their mother tongue or **6.3.5.1** *Overview of triggers*

*6.3.5.1 Overview of triggers*

70 gender; in addition, the triggers may provide an explanation for specific hedging behaviour. Thus, speakers might consider certain borrowings less acceptable than others and therefore flag them: it could be expected, for instance, that hedging occurs more commonly in the context of searches for the appropriate word than before borrowings of cultural concepts or proper nouns. In line with a grounded theory approach, I decided to set up my own classification, as I did not want to superimpose fixed categories on my material, even though others have also developed theoretical classification systems for code-switches (albeit, to my knowledge, less detailed ones, e.g. Gumperz 1982). Therefore, the taxonomy presented here does not include all possible triggers for language switches; instead, it focuses on the ones that caused switches in these specific interviews. It should also be noted that the categories Switches can be initiated by a variety of triggers, which may influence not only the bilinguals' hedging behaviour, but also the type of switch that is used. For my analysis, I tried to identify the main or most likely trigger or reason for each switch, and categorized the switches accordingly. This categorisation renders it possible to determine whether there are any differences in the bilinguals' specific switching behaviour based on their mother tongue or gender; in addition, the triggers may provide an explanation for specific hedging behaviour. Thus, speakers might consider certain borrowings less acceptable than others and therefore flag them: it could be expected, for instance, that hedging occurs more commonly in the context of searches for the appropriate word than before borrowings of cultural concepts or proper nouns. In line with a grounded theory approach, I decided to set up my own classification, as I did not want to superimpose fixed categories on my material, even though others have also developed theoretical classification systems for code-switches (albeit, to my knowledge, less detailed ones, e. g. Gumperz 1982). Therefore, the taxonomy

presented here does not include all possible triggers for language switches; instead, it focuses on the ones that caused switches in these specific interviews. It should also be noted that the categories are not clear-cut, as some switches may have several triggers. In each case, I attributed them to the trigger that I felt to be dominant. In theory, all of the categories could trigger both code-switches and borrowings, as well as both hedged and unhedged switches, but not all of these combinations appeared in the interviews, and only the ones that did were included in Table 6 below. I will first present an overview of the triggers for all the language switches, before taking a closer look at each trigger.

Table 6 gives an overview of all of the instances in which the participants switched to (Swiss) German, indicating the type of switch, the circumstances, hedging, and the most likely trigger for each switch. The table demonstrates that language switches were caused by a variety of different triggers, and that there were great differences among the bilinguals with regard to what triggered their language switches. Most frequently, spontaneous switches were triggered by names for places (n=41) and people (n=27), as well as cultural concepts (n=29). Together, these three groups make up 58.8% of all spontaneous switches. In contrast, only 8.5% of all spontaneous language switches were triggered by a gap in the speaker's lexicon (2+4+8). Table 6 also shows that the vast majority of the bilinguals' language switches can be attributed to a clear trigger — only for 4.6% of all switches (4 code-switches and 10 borrowings), it was not entirely clear what triggered the switch, though it was usually assumed to be the content of the conversation.

When the hedging context of the triggers is examined more closely, it emerges that the only switches that were usually hedged were those that occurred when the speaker could not think of a word and used a translation instead (7 of 8 hedged [87.5%]). Instances in which the speakers asked for a word (1 of 2 code-switches and 1 of 4 borrowings [in total 33%]), as well as switches that were triggered by the topic of the conversation (1 of 4 code-switches and 4 of 10 borrowings [35.7%]), were hedged the second most frequently, with about a third of the switches being hedged. They are followed by cultural concepts (6 of 29 [20.7%]), names of institutions (2 of 11 [18.2%]), general switches in a metalinguistic context (13 of 90 [14.4%]) and place names (5 of 41 [12.2%]). Several types of switches were hedged only rarely — switches triggered by previous switches (0 of 3 code-switches and 1 of 10 borrowings [total 7.7%]), people's names (2 of 27 [7.4%]), or were never hedged at all (corrections, translations, quotations, spontaneous or emotional reactions, direct address). It is likely that these types of switches are viewed more positively or perceived as less disruptive or more natural than others.


Table 6: Overview of triggers for language mixing for each participant

Key: h = hedged; nh = unhedged; CS = code-switch; B = borrowing; m = metalinguistic; s = spontaneous; pp = per person; bold = in text.

#### **6.3.5.2** *Triggers for spontaneous switches*

For switches that occurred spontaneously, I distinguished eleven different categories (see Table 6), which could all comprise borrowings as well as code-switches (although only categories [c] to [g] did). In the following, I will discuss the categorisation of these triggers in more detail, give examples of each trigger, and address their role in the bilinguals' language mixing behaviour.


to a master's degree] [Courtney <sup>E</sup> 1051]), local foods (*Luxemburgerli* 'macaroons' Claire <sup>E</sup> 11]), or were linked to the interviewees' lives as a bicultural couple (*Migrationsamt* 'migration office' [Katia SG 1365], *Ausländerausweis* 'identity card for foreign nationals' [Simon SG 252]). These switches were used with similar frequency by speakers of both mother tongues.

c. **Triggered by topic:** Other spontaneous switches were triggered by the topic of the conversation (n=14 [4 cs + 10 b]),<sup>6</sup> as in the example below:

```
Example 6.6: Claire (Northern Irish, 28) 
Claire: <W but I <HI think HI> my thing is as ^well W> /~
               <%%> like if I speak ^German //
               co^rrect /
               Hochdeutsch // (('High German')) 
               and langsam / (('slowly'))
               ((…)) kann ich auch besser ^sprechen // (('I can speak better, too'))
                                                                     (618–620)
```
 In this case, the subject of speaking German appears to have triggered a code-switch to the language, and switching to Standard German which occurs although Claire is fluent in Swiss German — appears to serve her to exemplify her statement on the topic. In other situations, the reason for the language switch was not as clear as in the example given (e.g. *Steuerklärung* (sic) 'tax declaration' [David <sup>E</sup> 666] or *Ausdrücke* 'expressions' [Craig <sup>E</sup> 970]), but if there was no definite reason apart from the topic of the conversation, such switches were thought to be internally motivated and assigned to this category. It is conceivable that these switches occurred with German terms that have been used so often by the couple that they have become "habitualized" (Poplack 1980: 598). Another reason for some of these switches might be that the English speakers attempted to demonstrate their L2 proficiency or their familiarity with the language in this manner, which is supported by the fact that this type of switch was used considerably more often by English-speaking participants (n=12) than by the Swiss (n=2). It should be noted that, although the reference to names and cultural concepts can technically be regarded as a switch triggered by the topic, these switches were so specific and numerous that they were categorized separately.

<sup>6</sup> In the following, the first number in brackets refers to code-switches (CS), the second one to borrowings (B).

d. **Triggered by previous switch:** Some switches, as David's <sup>E</sup> switch in example 6.7, were triggered by another speaker's switch (n=13 [3 cs + 10 b]).

```
Example 6.7: David (English, 32) and Susanne (Swiss, 29)
Susanne: with <HI ^other HI> people _
                <P<LO<HSK he's more direct than ^I am HSK>LO>P> _
                … ((lengthy pause))
                ^gell //~ (('right?'))
                da ^schtimmt _ (('that's true'; addressing David))
David: <H> <P<LO ^jo: LO>P> _ (('yes'))
```
 Such switches were used similarly often by speakers of both mother tongues.


```
Example 6.8: Claire (Northern Irish, 28)
Claire: = <P<LO<BR and he wants to discuss it all ^through BR>LO>P> /~
               <P<HSK<LO ^and LO>HSK>P> _
               <PP<A ^talk about it A>PP> _
               <PP<LO ^and LO>PP> _
               <H> … <PP wieder^ho:l- PP> / (('repeat'))
               <PP rep- re- <HI ^peat it HI>PP> _ (1641)
```
 According to Lüdi (2003: 178), this is a well-known communicative strategy used by L2 learners, which he calls *translinguistic wording.* However, since the bilinguals in this study are highly proficient, such switches were relatively infrequent. While the Swiss participants might be expected to use this strategy more often than the native speakers of English due to gaps in their L2 vocabulary, it was in fact used by participants of both mother tongues, and only slightly more often by the Swiss (n=8) than by English speakers (n=6). The majority of these switches, 9 out of 14 instances, were hedged, which could indicate that such switches are not perceived as desirable, but could also be due to the fact that the speakers were searching for words.

g. **Direct address:** Some switches marked a change in addressee (n=5 [3 cs + 2 b]); occasionally, these switches simultaneously indicated that a comment was off the record. This happened, for instance, when someone said *danke* 'thanks' to someone for passing something. Such "addressee specification" (Gumperz 1982: 77) enables the speaker "to continue the conversation smoothly, without undue flagging of who they mean to address" (Gardner-Chloros 2009: 79). Interestingly, switches to German were used to address both native speakers of Swiss German (n=2) and non-native speakers (n=3). In example 6.7 above, for instance, Susanne addresses her Anglophone partner in Swiss German.

#### **6.3.5.3** *Triggers for metalinguistic switches*

With regard to switches in a metalinguistic context, there were not as many different scenarios as for the spontaneous switches, and I distinguished five subcategories. The majority of the metalinguistic switches were attributed to one category ("general metalinguistic comments"), a category that mainly com‐ prises switches that were due to the interview situation. In addition, there were also a number of borrowings at the intersection between metalinguistic and spontaneous triggers: quoting someone, translating for others' benefit, helping or correcting. Although there were examples of these five subcategories among the metalinguistic borrowings, I did not distinguish subcategories for metalin‐ guistic code-switches, as they are often hard to tell apart for code-switches, as the following quote demonstrates:

```
Example 6.9: David (English, 32)
David: = <W<P<HSK I've never <HI ^heard HI> you say HSK>P>W> \
              <W<PP I liebe ^dich to me PP>W> // (('I love you')) (909)
```
Such switches were, therefore, not assigned to separate categories, resulting in a single category of (general) metalinguistic code-switches.

h. **General metalinguistic comments**: This type of switch was by far the most frequent, with 90 instances of borrowings and 27 code-switches. Switches were assigned to this category when a specific word or sentence was discussed, or when an example of a word or sentence was given ("I say ^*Schatzi*" 'treasure, darling' [Joshua E 1054]). English speakers used such switches a little more often than the Swiss (63 [14 cs + 49 b] vs. 54 [13 cs + 41 b] switches).



l. **Translations**: Finally, a (Swiss) German translation was sometimes provided in addition to the synonymous English expression (n=3), usually for the interviewer's benefit ("I ((… )) had a co-op, *Praktikum*" 'internship' [Craig <sup>E</sup> 11]).

This overview has shown that bilinguals switch languages due to a large variety of different triggers, and that there appears to be a correlation between the speakers' mother tongues and the prevalence of certain triggers in their language mixing. This variable and the variable of gender will be examined more closely in the following sections.

#### **6.3.6 Language mixing and gender**

Previous research indicates that the element of gender potentially plays a role with regard to code-switching or borrowing. Since it has been observed that women use fewer non-standard forms than men, it seems likely that they will also mix languages less extensively than men do (see Cheshire and Gardner-Chloros 1998), although this may depend on the setting and the cultural backgrounds involved. To test this hypothesis, Table 7 below provides an overview of the switching and hedging behaviour of the two genders, comparing the groups that comprise all members of each gender and the groups that simultaneously include the variable of their mother tongue (see section 4.7, "Analysis", for a discussion of these groups).


Table 7: Code-switching and borrowing for each gender and each combination of language and gender (normalized per 10,000 words)

Key: CS = code-switch; B = borrowing; m = metalinguistic context; s = spontaneous use; h = hedged; nh = not hedged; bold = in text.

As can be seen in Table 7, the female participants used more spontaneous bor‐ rowings, as well as more spontaneous code-switches, than the male participants (18.63 vs. 11.03 borrowings and 2.30 vs. 0.72 code-switches per 10,000 words). This is the case for the groups consisting of all members of each gender, as well as for the groups that include the variable of the speakers' mother tongue, and for both types of switches. Hence, the Swiss German and English-speaking females used more spontaneous switches than their respective partners (17.93 vs. 11.30, and 27.93 vs. 12.93 per 10,000 words). Overall, the female participants were responsible for 73.3% of all spontaneous code-switches and 59.6% of all borrowings, which demonstrates that there is a clear gender trend with regard to language mixing for speakers of both mother tongues.

In addition to the frequency of switching, there are also differences in the hedging behaviour of the genders. The female bilinguals hedged fewer of their spontaneous switches than the male participants did — only 14.0% (2.93 of 20.93) of the women's switches were hedged, as opposed to 24.6% (2.89 of 11.75) of the men's. When the mother tongues of the speakers are taken into account, it is evident that the Swiss women hedged their spontaneous switches less frequently than their partners (10.0% [1.80 of 17.93] vs. 26.6% [3.01 of 11.30]), while there was little difference in the other couple combination (20.0% [5.59 of 27.93] vs. 20.0% [2.58 of 12.93]). Nevertheless, the female participants hedged fewer switches than the male participants with the same mother tongue. The fact that the female participants not only switched more frequently than the male participants, but also tended to hedge fewer of their spontaneous switches than the male participants did, suggests that women may have a more positive attitude towards language mixing and may be more inclined to practice it themselves. These findings are contrary to the expectation that women will not code-switch as much as men because they might prefer a more standardized mode of communication (see section 6.2.1, "Factors that influence language mixing").

Unlike spontaneous switches, language switches that were prompted by the interview situation are probably only minimally indicative of the partici‐ pants' usual language mixing behaviour, and may also be perceived to be less marked, or more necessary, by the bilinguals. It is therefore not surprising that there appear to be no gender-specific trends with regard to switches in a metalinguistic context. At first sight, it seems that the male participants used slightly more metalinguistic switches than the female participants did (13.92 vs. 13.40 per 10,000 words); when the bilinguals' mother tongue is taken into consideration, however, there is no uniform pattern (as English-speaking males were responsible for the most instances, but Swiss males for the fewest). Moreover, while the number of hedged metalinguistic switches is similar for both genders (11.0% [1.47 of 13.40] for female and 11.7% [1.63 of 13.92] for male participants), there was considerable variation between the four groups. Hence, specific gender differences in the participants' language mixing behaviour were only found in the context of spontaneous switches, which highlights the importance of the context in which switches occur for the analysis of language mixing.

#### **6.3.7 Language mixing and mother tongue**

In addition to gender, the mother tongue of the bilinguals is likely to influence their language mixing. The base language of the conversation may be their stronger or weaker language, and it is possible that the bilinguals for whom it is an L2 — the Swiss participants in this case — will be more inclined to switch than the native speakers, even though the latter may, of course, also switch to their L2. Table 8 provides an overview of the switching and hedging behaviour of the speakers of both mother tongues, comparing the groups that comprise all native speakers of both languages, and the groups that also include the variable of their gender.


Table 8: Code-switching and borrowing for each mother tongue and each combination of language and gender (normalized per 10,000 words)

Key: CS = code-switch; B = borrowing; m = metalinguistic context; s = spontaneous use; h = hedged; nh = not hedged; bold = in text.

Somewhat surprisingly, the results indicate that the Swiss speakers did not switch to (Swiss) German more often, or more extensively, overall than the native speakers of English did (on average 28.61 vs. 30.66 switches per 10,000 words). The English speakers switched more often in a metalinguistic context than their partners (14.96 vs. 12.26 instances per 10,000 words), while the Swiss participants switched slightly more often spontaneously (16.35 vs. 15.70 instances per 10,000 words). On average, the Swiss used more spontaneous borrowings (15.12 vs. 14.04 per 10,000 words), whereas the Anglophone partners used more spontaneous code-switches (1.66 vs. 1.23 per 10,000 words). When the variable of gender is also taken into account, it appears that there is no clear trend with regard to spontaneous switching among the speakers of both mother tongues: The largest number of spontaneous switches occurred among Eng‐ lish-speaking women (27.93 switches per 10,000 words), yet English-speaking men switched the least (11.30 switches). With regard to metalinguistic switches, however, the tendency for the Anglophone participants to switch more persists, as the male and female Anglophone participants switched more than their respective partners (16.07 vs. 14.04, and 11.88 vs. 8.39).

These findings indicate that the bilinguals for whom the base language of the interaction is an L2 will not automatically — as might be expected — mix languages more frequently than those for whom it is their native tongue. In the case of the interviewees in this study, this is probably due to the Swiss partners' high level of proficiency in the language of the interaction. In addition, it is also imaginable that, as non-native speakers, the Swiss partners had a higher motivation to refrain from language mixing than their partners so as to demonstrate L2 proficiency. This is supported by the fact that, on the few occasions in which the Swiss participants struggled to express themselves in the exact manner they wanted, their partners sometimes invited them to switch to German, but the Swiss never did. In contrast, the Anglophone participants might have wanted to demonstrate their L2 skills by switching to (Swiss) German. The fact that not a great deal of language mixing occurred during the interviews confirms that the fluent bilinguals in this study are willing and able to adhere to one of their languages if they feel that the situation calls for it, even if they theoretically have the option of using the other language.

While speakers of both mother tongues used a similar number of switches, there was a marked difference in the frequency with which spontaneous as well as metalinguistic switches were hedged (Table 8 above). Thus, the native speakers of English hedged their spontaneous switches considerably more frequently than the Swiss interviewees (23.5% [3.69 of 15.70] vs. 12.5% [2.04 of 16.35] hedged). In the two couple combinations, this trend is only distinct in the couples consisting of an English-speaking male and a Swiss female (26.6% [3.01 of 11.30] vs. 10.0% [1.80 of 17.93]), as there is little difference between the partners in the other couple combination (20.0% [5.59 of 27.93] vs. 20.0% [2.58 of 12.93]). This also means, though, that the two Anglophone groups hedged more than the respective Swiss group of the same gender. A similar trend was found with regard to the hedging of metalinguistic switches; in this case, too, the English-speaking participants hedged a higher percentage of switches than the Swiss bilinguals (14.8% [2.22 of 14.96] vs. 6.7% [0.82 of 12.26]). This difference also remains evident when the variable of gender is taken into account, as the Anglophone men hedged more metalinguistic switches than their partners (12.4% [2.00 of 16.07] vs. 6.4% [0.90 of 14.04]), and the Anglophone women hedged more than theirs (23.5% [2.79 of 11.88] vs. 7.6% [0.64 of 8.39]). The fact that the English-speaking participants hedged a substantially higher percentage of their switches than the Swiss participants in both contexts could be due to a number of factors. It might indicate that the latter view language mixing more positively than the former, or that they are more accustomed to it because of Switzerland's multilingualism (see also chapter 7, "'This uh foreign girl with a great accent': Attitudes and attraction"). Moreover, the English speakers could be hesitant about switching to German not because they have a negative attitude towards language mixing, but because they feel self-conscious about speaking their L2. In addition, it is also conceivable that the Swiss used fewer hedges because they switched in more contexts where they felt that switching is acceptable.

However, a closer look at the contexts in which both groups mixed languages indicates that the total number of switches in each category, as well as the distribution of hedged versus unhedged switches, are remarkably similar in both groups (see Table 50 in Appendix IV). In terms of the frequency of mixing across the categories of triggers, the main differences that are noticeable are that spontaneous emotional reactions in Swiss German were more common among the Swiss participants (n=5) than the speakers of English (n=1), and that the English-speaking participants produced more switches that were triggered by the topic (n=12) than the Swiss did (n=2). The most obvious difference regarding hedging is that the Swiss speakers hedged fewer of their switches that concerned proper names (1 of 40 hedged) than the English speakers (8 of 41 hedged). Interestingly, there is no difference with regard to switches that were triggered by gaps in the bilinguals' vocabulary. Not only were there few such switches, but they also occurred with equal frequency in the discourse of speakers of both mother tongues (n=6 for both). This, again, emphasizes that the high level of proficiency of the Swiss participants in their partner's language in all probability had a major influence on their switching behaviour.

#### **6.3.8 Language mixing and family situation**

One more factor that potentially influences the language mixing behaviour of bilingual couples is their family situation, in particular, how long they have been a couple, and whether or not they have children. In order to analyse this, the couples were assigned to three groups. The couples in the first group had been together for less than four years at the time of the interview, were not living together and had no children (Tim/Sarah, Robert/Stephanie). The native English speakers in this group were also the least fluent in German, as they had been exposed to the language for the shortest period of time. The second group consisted of couples who had been together for a longer period of time (more than 4 years), were living together, and had no children (David/Susanne, Courtney/Martin, Claire/Simon, Dean/Monika). The couples in the third group had also been in a relationship for a longer period of time, and they also had children together (Richard/Sophia, Joshua/Deborah, Craig/Katia, Karen/Philipp). Table 9 shows how often the speakers of each of the three groups switched spontaneously on average, and how many of these switches were hedged in each group:


Table 9: Correlation of spontaneous switches with relationship length and family situation Key: h = hedged; nh = unhedged; bold = in text.

Table 9 indicates that the couples in the first group switched the least (on average only 4.18 per 10,000 words). The couples in the second group — who had been together for a longer period of time but did not have children — switched the most, almost 4 times as much as the speakers in the first group (19.47 per 10,000 words). The participants who had children switched less than the second group in absolute numbers, but when their number of switches is set in relation to the number of words they used, it emerges that they mixed the most in relative terms (22.99 per 10,000 words). While there is no clear indication that having children influences the couples' mixing behaviour, these figures suggest that the frequency of language mixing increases with the duration of the relationship, since the couples with children have been together the longest. This trend can also be observed in Figure 1 above (in section 6.3.4, "Similarities in the partners' language mixing"); there are only two couples who stand out, namely Susanne and David, who switched particularly often considering the length of their relationship (5 years), and Monika and Dean, who switched very little despite having been in a relationship for 9 years.

It is also possible that there is a connection between relationship length and the couples' hedging behaviour, as the percentage of hedged switches was relatively similar among the two groups who had been in a relationship for a longer period of time (19.3% and 18.2%, respectively), while the two couples who had been together for a shorter period of time hedged fewer of their switches (9.1%). Due to the small number of switches among them (n=11), however, this percentage may not be representative of a larger sample.

#### **6.4 The couples' reports on their language mixing**

Since the couples' language mixing during the interviews only depicts their behaviour in one specific setting, I also questioned them on their habitual switching behaviour. The couples were asked if they ever mix languages, which eight of the ten couples answered affirmatively, though many of them immediately qualified that they only mix languages in very specific situations, and that it does not happen very often. The interviewees were not specifically asked which type of switches (code-switching or borrowing) they use more frequently, but they all described their switching behaviour of their own accord. Their responses are summarized in Table 10, which gives an overview of the participants' reported language mixing behaviour, both with regard to type of switch and frequency of switching.


Table 10: Reported language mixing behaviour in couple conversation for each participant Key: B = borrowing; CS = code-switching; E = English; SG = Swiss German.

Since most of the couples usually speak English to each other when they are on their own, Table 10 relates to situations in which the couples use English as a base language from which they switch to (Swiss) German (with the exception of Joshua/Deborah, who use both languages as a base language). The table shows that there are five participants who sometimes code-switch, and who use some or a fair number of borrowings (Richard/Sophia, Joshua/ Deborah, Philipp). Eleven participants said that they sometimes borrow (Swiss) German expressions while speaking English, but that they rarely switch to German for entire sentences or for an extended period of time. The remaining four participants, Tim/Sarah and Courtney/Martin, hardly ever mix languages, though they may use a small number of Swiss expressions with which the English speaker is familiar as part of their couple language. No-one professed to code-switch frequently at the time of the interview (though Joshua and Deborah used to code-switch regularly before they had children).

It is notable that in all but two cases, the participants claimed that their language mixing behaviour is similar to that of their partner. It is possible that this is because the more fluent bilingual speaker adapts to the language level of the less fluent partner; if the latter is not a skilled code-switcher, the former will not practice code-switching either. Only two couples reported that they differ slightly in their language mixing. In one case, the Swiss partner claimed to mix more: Philipp code-switches primarily with his children, and occasionally with his wife, Karen, but Karen does not code-switch, as their children tease her when she speaks German. In the other case, the English speaker, Claire, borrows more terms from German than her husband does, though neither of them codes-switches extensively. The reason for this may be that, professionally, Claire speaks German exclusively, so that the language is constantly activated to some extent, while Simon, who is an English teacher, might place more value on keeping the languages separate. At any rate, the difference appears to be minor even in the case of these two couples.

According to their reports, native speakers of both languages switch equally often, which might be due to the assimilation of the partners to each other's mixing behaviour (or the tendency to at least report a similar behaviour); for the same reason, there appears to be no difference between the genders. The only variable that clearly correlates with the participants' language mixing behaviour (apart from their partner's behaviour) is the level of proficiency of the less fluent partner in his or her L2. Thus, the four couples who have been together the shortest period of time (and of whom the native speaker of English has conse‐ quently had the least exposure to German) rarely code-switch; some of them report borrowing (Swiss) German expressions occasionally (Robert/Stephanie, David/Susanne), while others hardly ever do so (Tim/Sarah, Courtney/Martin). Thus, many of the speakers who are less fluent or less comfortable speaking German only use the occasional (Swiss) German loanword, which is also true for their partners. In contrast, the couples who have been together for a longer period of time do not necessarily code-switch, but they all use borrowings at least occasionally, and some of them even do so frequently. Finally, there are also two couples who reportedly practice skilled code-switching sometimes (Richard/Sophia and Joshua/Deborah), both of whom include an English speaker who is fairly fluent in his L2. The couple who reportedly code-switch most frequently, though not as much as they used to, are Joshua <sup>E</sup> and Deborah SG (see below). Of the native speakers of English, Joshua is the most fluent in Swiss German, and the couple are both comfortable and fluent enough to use both languages. In contrast, many other native speakers of English may not be proficient enough in their L2 to practice fluent code-switching.

The couples' reports indicate that their language mixing is also contingent upon the topic as well as the circumstances of the conversation. Some couples use (Swiss) German when they speak about situations or aspects that are associated with the language, such as their work, or local foods (Claire/Simon, Karen/Philipp, Sophia/Richard). They feel that it is convenient, for instance when they forget a word, or if there is an expression in either of the languages that they deem to be more suitable (Claire/Simon, Monika/Dean, Karen/Philipp, Katia/Craig). Other couples use German or Swiss German when they are speaking in a playful key, for example Richard and Sophia:


The extract shows that Sophia and Richard do not speak Swiss German in serious situations, even though Sophia points out that there would be no problem in terms of understanding. Instead, they only use the language in a playful or humorous way. Other participants also state that they use (Swiss) German or mix the two languages for fun or to create humour (Joshua/Deborah, Courtney/Martin, Stephanie/Robert). In a sense, this goes against the prevalent perception in Anglophone countries of German as a rather serious — and not very humorous — language (see section 10.4.4, "Playful language in bilingual couple talk", for a more detailed discussion of humorous language use).

Despite this positive function, the couples' views on language mixing seem to be somewhat negative overall. While most of the participants report that they do mix languages, they usually frame this as a kind of confession, or hesitate when the topic is addressed. In the following extract, for instance, German is portrayed as sneaking its way into the bilingual's language without him being able to control it:

**Example 6.12: Richard (English, 29)** Richard: <P the the .. the ^longer P> \ … I've lived here the ^more / … <H> <CRK German: sorta creeps ^into it CRK> \ (253)

There seems to be a general consensus that language mixing is often due to laziness ("sometimes you get a bit lazy as well ((… )), whenever a word comes first you just say it, German or English" [Simon SG 668–671]), and some of the interviewees think that it renders one's language very inaccurate (Karen E ). Karen <sup>E</sup> states that she and her husband view language mixing negatively because they are English teachers and have a background in linguistics, which indicates that she believes her critical attitude to be supported by linguistic research. Her husband even says explicitly that he believes mixing to be bad ("we d- we d- I do: [mix], but I know it's not good" [Philipp SG 299]). Yet, while many of the interviewees believe that mixing languages may not be ideal, they also appear to feel that it is not an issue as long as they are in control of it or do not do it too frequently. Thus, the interviewees emphasize that they always know which language they are speaking (Craig <sup>E</sup> ), that they do not mix a lot compared to "what's going on on the street" (Katia SG 597), or that they do not mix "in a way that is ba:d", as they "do it on purpose a lot of times" (Karen <sup>E</sup> 296). Many of the interviewees' statements reflect previous research, which reports a tendency for bilinguals to display negative views on language mixing and to attribute it to laziness (Gardner-Chloros 2009: 14).

The couples' accounts indicate that, especially once they have children, they try to separate the two languages, because they believe that mixing languages would confuse the children and thus be detrimental to their linguistic development. This is the case for Joshua <sup>E</sup> and Deborah SG, who used to speak both languages on a regular basis, and also used to code-switch frequently, but took a conscious decision to stop doing this when their first child was born:


Joshua emphasizes that they are putting considerable effort into keeping their languages separate with their children, which to them means that they should abstain from mixing languages (see also section 7.6, "The couples' views on raising bilingual children", for an extensive discussion). This demonstrates that particularly the couples who have children attempt to monitor their language mixing carefully.

Interestingly, when we look at each couple individually, about half of their reports on their usual mixing behaviour were not entirely congruous with their mixing behaviour during the interviews (see section 6.3, "The couples' language mixing during the interviews"). On the one hand, the couple who switched by far the most frequently during the interview, Susanne/David, did not state that they usually switch a lot, but only reported that they sometimes use borrowings. On the other hand, a number of couples claimed to switch regularly, but hardly did so during the interviews. Thus, Robert/Stephanie and Monika/Dean reported that they sometimes use borrowings, but they did so very rarely during their conversations (each 1–2 times). Richard <sup>E</sup> , Sophia SG, Philipp SG, Deborah SG and Joshua E claimed that they code-switch, yet of the five, Joshua was the only one who code-switched during the interview. In the case of most interviewees who switched less than they reported, it appears that switching is a part of their intimate couple language, and that they are less inclined to do so when others are present. The exception to this is Philipp, for whom code-switching is a part of the manner in which he speaks to his children; since they were not present during the interview, it is no surprise that he did not code-switch. These findings offer further confirmation for the view that a bilingual's switching behaviour in a specific situation may not be an accurate reflection of his or her habits in other situations, as it depends greatly on the circumstances and the interlocutor; it also confirms that self-reports have to be viewed critically.

#### **6.5 Discussion and summary**

The preceding analysis of the couples' language switching behaviour has demonstrated that there were relatively few language switches during the interviews, with 165 instances of spontaneous switches and 141 switches in a metalinguistic context. Spontaneous borrowings occurred far more often than code-switches, namely ten times as often. Both of these observations mirror the couples' own assessments of their usual language switching behaviour, as the majority of the participants claimed to use borrowings sometimes, but only a quarter of them said that they code-switch on occasion. It is therefore not surprising that there were no instances of extended skilled code-switching in the interviews, such as the switching behaviour observed by Schmid among Italian immigrants in Switzerland (2005: 137), or by Gardner-Chloros among German-French bilinguals in the French Alsace (1991: 98–99). Even if some couples occasionally code-switch when they are on their own, the setting of the interview may have been too formal to trigger such behaviour. There was also no indication in the couples' speech of anything resembling a mixed language, or "fused lect", as described by Auer (1999), and there were hardly any lexical hybrids, which are known to occur in some bilingual speech communities. It is possible that such phenomena are more common in the discourse of larger bilingual communities than in the speech of bilingual dyads; most of the bilingual couples in this study are not part of a larger bilingual network. It is also conceivable that code-switching occurs more frequently among second-gener‐ ation bilinguals. Schmid, for instance, observed that code-switching is popular among second-generation Italians in a number of different countries, most of whom "are really fluent bilinguals and want to distinguish themselves both from their basically monolingual parents and from the local society" (2005: 145). In addition, the fact that the bilinguals in this study switched relatively little could also be due to their high levels of education.

Research suggests that the topic of the conversation can also trigger language switches, as speakers may feel that the subject matter belongs to another language (Wierzbicka 2004: 102), especially if it is connected to key cultural concepts. This was explored in more detail by categorizing the switches according to the triggers which caused each of them. Triggers that were found to be responsible for spontaneous switches most often comprised proper names, cultural concepts, and other aspects of the topic of the conversation, but also included previous switches and gaps in the bilinguals' lexicon. Some spontaneous switches were the result of an impulsive or emotional reaction, or marked a change in addressee. In contrast, most metalinguistic switches were grouped as general metalinguistic examples or comments; translations, corrections, quotations and loanblends were also considered metalinguistic switches, as these were caused directly by the interview situation. Of all of these triggers, proper nouns (n=81) and general metalinguistic comments (n=117) were by far the most frequent, the former composing almost half of all spontaneous switches, and the latter 83% of all metalinguistic ones. The couples' reports are largely in agreement with this, as many of them claimed that they use German terms when they speak about aspects that are associated with the language, if they forget a word or feel that a German term would be more suitable or convenient, but also for humour. Furthermore, the analysis has also shown that the partners' switches were often caused by similar types of triggers. This suggests that the partners are likely to assimilate their switching behaviour to each other over time.

Most switches that occurred in the interviews were unhedged, and only 15% of all switches were accompanied by pauses or hesitation markers. As was to be expected, hedging was more common in the vicinity of spontaneous switches than in a metalinguistic context (18.2% and 11.3% hedged). Furthermore, it appears that the bilinguals' hedging depended to a large extent on the reason for the switch. For instance, the majority of switches were hedged when a speaker could not think of a word, while other switches, like corrections, translations, or emotional reactions, were never hedged. Even though hedges may have a variety of causes, the bilinguals' hedging behaviour is likely to be influenced by their attitude towards switching (Pietkännen 2014: 21); it may therefore be assumed that their views on switching are, to some extent, dependent on the situation or trigger for the switch.

The relatively low percentage of hedged switches overall can be seen as indicative of a generally positive or neutral attitude towards switching among the bilinguals. However, by their own accounts, the couples' views on switching are ambivalent. Many of them expressed slightly negative views on language mixing, relating it to laziness and linguistic inaccuracy. Nonetheless, several interviewees simultaneously portrayed language mixing as convenient and enjoyable, and consider it to be unproblematic in their case, as they believe themselves able to retain control over their mixing behaviour. A similar mixture of positive and negative attitudes has been observed in other bilinguals (e.g. Romaine 1995; Poplack 1988). Among the bilingual couples examined by Piller (2002a), such mixed feelings were especially common among couples who themselves practice language mixing. Thus, she reports that, "[o]n the one hand, they [felt] somewhat guilty about it and evaluate[d] language mixing negatively. On the other hand, they gleefully report[ed] how they enjoy the fact that they are not easy to categorize and how mixing best expresses their dual selves" (Piller 2002a: 179). From this discrepancy, Piller concludes that bilingual couples might attach a covert prestige to language mixing, which is also possible for the bilingual couples in this study.

The bilinguals' accounts indicate that their attitudes toward language mixing tend to be very similar to their partners'. Moreover, the participants also reported that they mix languages to a similar extent as their partner, with only two couples reporting slight differences between them. This suggests that there may be a large degree of assimilation between the partners, in terms of both their attitudes towards language mixing and their actual language mixing behaviour. This assimilation may have attenuated potential differences with regard to gender or mother tongue. However, it should also be noted that the couples' switching behaviour may not be as similar as their reports suggest. While, in many instances, the partners switched similarly often during the interviews, this was not the case for all of them, and some of the interviewees' reports were not entirely consistent with their switching behaviour during the interviews, since a number of participants switched more or less than they reported. Such a discrepancy between bilinguals' reported and actual mixing behaviour has been observed by others (Piller 2002a: 148, see also section 7.6, "The couples' views on raising bilingual children"). Potential reasons for this divergence could be changes in the bilinguals' mixing behaviour owing to the interview situation, a desire to report similar switching behaviour to one's partner, or imprecise or inaccurate reports due to the bilinguals' attitudes or due to the short time available to them during the interview to reflect and analyse their behaviour.

Despite the potential assimilation between the partners, some gender differ‐ ences were discernible in this particular setting, as the female participants used more spontaneous borrowings as well as code-switches than the male participants. In fact, 59.6% of all borrowings and 73.3% of all code-switches in a spontaneous context could be attributed to women, and this trend could also be observed irrespective of the participants' mother tongue. These results are consistent with some previous findings, such as Poplack (1980), who found more intrasentential switches in the speech of female than of male bilinguals among Puerto Rican Americans, though, in other settings, the opposite trend was found (Poplack 1988), or there was no gender-specific trend (Cheshire and Gardner-Chloros 1998). These divergences suggest that gender-specific switching behaviour may be dependent on the language community in question. In addition to switching with different frequencies during the interviews, representatives of the two genders also displayed different hedging behaviour, with women hedging fewer spontaneous switches (14.0%) than men did (24.6%). Both their hedging behaviour and the frequency with which they switched might indicate a more positive or relaxed attitude towards switching among the female participants. In contrast, no unambiguous gender trends were found with regard to metalinguistic switches, which lends further weight to the hypothesis that the context in which switches occur is highly relevant to their analysis.

When the switching behaviour of the genders during the interviews is compared with their own reports, it emerges that the two are not entirely congruent, as interviewees from both genders claimed to mix equally often. It cannot be determined whether this discrepancy is due to a tendency for the partners to report similar frequencies of switching, despite differences in their actual behaviour, or because there was a disparity in the partners' switching behaviour during the interviews, although they switch similarly often when they are on their own. It could be hypothesized that the female participants felt more compelled to switch in the presence of an interviewer of their own gender, which would confirm Gardner-Chloros' hypothesis that women may code-switch more in order to bond with their interlocutor (2009: 86). This supposition would, however, need further exploration.

While I expected the Swiss participants to mix languages more often and more extensively than their Anglophone partners, no indication of such a tendency was found when the influence of the bilinguals' mother tongue was examined. This observation is consistent with the couples' own accounts, as native speakers of both languages reported that they switch equally often. During the interviews, the Anglophone participants switched more in a metalinguistic context, while there were slightly more spontaneous switches among the Swiss, though the second tendency was not confirmed when the variable of gender was also considered. This indicates that, provided that their level of L2 proficiency is high, bilinguals do not automatically resort to their mother tongue more often when the language of the interaction is their L2 as opposed to their L1. If the L2 speakers switched less than the native speakers, this was potentially due to a higher motivation among the Swiss to adhere to the language of the conversation; the Swiss demonstrated their L2 skills by refraining from switching, whereas the Anglophone participants did so precisely by switching. The bilinguals' hedging behaviour, however, does not lend support to this hypothesis, as the native speakers of English hedged a much larger percentage of their spontaneous switches than the Swiss participants did (23.5% vs. 12.5% hedged). This tendency was also evident in the hedging of metalinguistic switches (14.8% vs. 6.7% hedged), even when the variable of gender was accounted for. It is conceivable that the Anglophone partners used more hedges because they have a more negative attitude towards switching, or because some of them may feel self-conscious about speaking their non-native tongue.

In addition to gender and mother tongue, the relationship length of the couples was shown to play a role in their mixing behaviour. Thus, there was a tendency for the bilinguals who had been together for the shortest period of time to switch the least, while participants in a more long-term relationship switched considerably more often. This indicates that switching may increase as the partners become more fluent in each other's languages and more familiar with one another. The frequency of switching appears to be most clearly linked to the L2 proficiency of the less fluent speaker, as only two couples reported that they code-switch with one another, both including two very fluent bilinguals. At the same time, less fluent speakers tended to only use occasional borrowings, but rarely code-switched, which also applied to their partner. This is in line with Poplack's observation that fluent code-switching requires a high level of language proficiency (1980: 601). While I anticipated that the couples' switching may decrease once children are born, there was no evidence of such a tendency during the interviews, potentially because their children were not present. This stands in contrast to their reports, according to which the couples who have children endeavour to separate the two languages in order to avoid confusing their children or impeding their language acquisition, indicating a considerable amount of private language planning.

Due to the small size of the sample, the tendencies found with regard to language mixing need to be explored in a larger sample to be confirmed with certainty. There are other variables, too, which may have been influential with regard to the bilinguals' switching behaviour, such as their cultural background, or their children's age. Moreover, the couples' language mixing behaviour may be different when they are on their own, with other people, or in a different setting. Nonetheless, I believe that the combination of the couples' reports and various aspects of their switching behaviour during the interviews has provided important insights into language mixing among proficient bilingual couples.

## **7 "This uh foreign girl with a great accent": Attitudes and attraction**

## **7.1 Introduction**

In the previous two chapters, I discussed two essential elements of the manner in which bilingual couples communicate with each other: language choice and language mixing. The chapters have confirmed that situational as well as long-term language choices are influenced by a variety of elements, which all interact to shape the bilingual couples' communication. Among these, the partners' attitudes towards a multitude of aspects relating to their bilingual, bicultural relationship have proven to play a crucial role in the couples' language choice and their mixing behaviour. In the case of the bilingual couples in this study, there is an interplay between their attitudes towards both of their languages, their cultures, bilingualism in general, bilingual couplehood, and, potentially, raising bilingual children. Moreover, the bilingual partners' attitudes are by no means static, but may evolve over time with additional contact and experiences. Because attitudes are so complex, and so influential in shaping bilingual, bicultural couplehood, they will be examined more closely in this chapter.

Research on attitudes of and towards bilingual, bicultural couples has shown that these tend to depend on the cultures and languages of the respective partners. While some studies have concluded that many bilingual, bicultural relationships are inherently problematic, or are at least viewed as such, due to cultural or linguistic barriers, there have also been mentions of very positive attitudes (e.g. Piller 2002a). This highlights the importance of outlining the attitudes of bilingual couples for each specific combination of languages and cultures. It needs to be borne in mind that attitudes are difficult to measure due to their fluidity and complexity, and attitude reports are often unreliable (see section 7.2, "Previous work on attitudes and attraction"). Nonetheless, attitudes ought to be taken into account, since they may help to explain bilinguals' language choices and use.

The element of attraction among bilingual couples will also be considered in this context, as attraction is often linked to underlying attitudes towards one's partner's culture and/or language. Indeed, when attraction is triggered by linguistic or cultural characteristics of the partner, it may be viewed as the manifestation of a highly positive attitude. Conversely, partners may also have been attracted to each other initially for other reasons, and may have found each other appealing not because, but in spite of their different backgrounds. Since many of the interviewees, particularly the Anglophone partners, may not have had a differentiated view on their partner's language and culture before they met, the attraction they experienced should be taken into account, as it may give an indication of their covert or latent attitudes.

My aim in the present chapter is to give an overview of the partners' attitudes and the development of these attitudes, by considering their first attraction to each other, their initial attitudes as well as their current attitudes. I am also interested in ascertaining whether the participants' cultural or linguistic backgrounds affect their attitudes. Furthermore, I want to find out to what extent their attitudes influence their linguistic behaviour, both concerning their couple language and the language(s) they speak to their children. In this context, also the couples' attitudes towards raising bilingual children will be examined, as well as their intended and actual strategies in their children's linguistic upbringing. As the languages the couples speak to their children are likely to influence their couple language to a certain extent, this will be included in order to create a comprehensive picture of the manner in which the couples communicate.

The couples were asked a number of questions so as to determine their attitudes towards each other's culture and language, as well as towards their bilingualism and biculturalism. After an overview of previous research on attitudes and attraction (7.2), I discuss the role of cultural and linguistic aspects in their initial attraction to each other (7.3). I then proceed to examine their past and current opinions about each other's language and country (7.4), and their views on being a bilingual, bicultural couple (7.5). Moreover, I discuss the development from a couple language to a family language, which occurs when the couples start raising bilingual children, as well as the couples' attitudes towards their (potential) children's bilingualism (7.6). In the final section (7.7), I summarize my findings, compare them with previous research, and trace the role of the interviewees' attitudes in their language choice.

### **7.2 Previous work on attitudes and attraction**

#### **7.2.1 Attraction in bilingual, bicultural couples**

Ending up with a partner from a different linguistic or cultural background may, of course, be a coincidence to some extent, as there are a number of other reasons for which a partner may be selected (such as his or her personality, attractive‐ ness, or intelligence). However, a recurring theme in conversations with people in bilingual relationships seems to be a strong interest in and attraction to their partner's language and/or cultural background. For instance, Pavlenko observes that some "multilinguals acknowledge being emotionally attracted to their new languages which allow them to perform new and different emotional selves. [… ] At times, attraction to language may also lead to attraction to speakers of a particular language" (2002: 49). Piller (2009) studied the phenomenon of "language desire" in bilingual couples by analysing conversations between 36 couples consisting of a native speaker of English and a speaker of German. She discovered that there are various desires that are frequently exhibited by bilinguals, even before entering a bilingual relationship. These are:

the desire to master another language; romantic and sexual desire for a partner who is a native speaker of a particular language; a desire for access to interactional partners in the target language; a desire for one's children to become fluent bilinguals; and a desire to be legitimate members of both language communities. (2009: 66)

According to Piller, "a bilingual relationship can serve as a way into a second-language community" (2009: 54) and the "desire for a partner who speaks another language may also be related to a desire for greater interactional opportunities in that language" (2009: 54). As a consequence, several of the respondents in her study claimed that their bilingual relationship was far from coincidental. In earlier research based in part on the same data, Piller also identified many "Germanophiles" or "Anglophiles" among the bilingual respondents, particularly the female partners, many of whom "express[ed] a longstanding desire for their L2 which culminated in pursuing the language in a university degree course" (2002a: 100, see section 3.3.2, "Bilingual couples"). Most frequently, the desired language was English, which may be linked to the fact that English is usually framed as desirable in the media and in the public discourse. In English-speaking contexts, in contrast, foreign languages — or at least accents — that are often linked to desirability appear to be French and Italian (Jenkin 2014).

Attraction may not only be caused by a foreign language, but also by another culture. The attraction towards members of a foreign culture is often felt to be immediate and strong, as Visson discovered in an exploration of Russian-American marriages, as many couples "claimed that they were instantly attracted to their future partner, as though hit by lightning" (2009: 148). Such a fascination with one's partner and his or her culture seems to be especially prominent at the beginning of the relationship when partners tend to view each other as representative of their culture. In contrast, "[t]he more established the relationship is the less partners see each other as cultural representatives, and the more they see each other as individuals" (Piller 2007: 349).

There are a number of factors that potentially contribute to the appeal of a partner from another culture. There may be purely practical reasons, as was the case for some of the Russian mail-order brides described by Visson, who were reportedly looking for a more comfortable life (2009: 154).<sup>1</sup> It has also been suggested that people who do not fit in well or are unable to find a suitable partner in their own society may attempt to find "a mate who is unaware of the extent of their alienation" elsewhere (Visson 2009: 161). In addition, the novelty or otherness of the language or culture can make it seem interesting and desirable. For instance, certain features of the L2, such as idiomatic or formulaic language, are at times viewed very positively by non-native speakers. As Tannen points out, "the turn of phrases and common expressions of another language, when translated into one's own, can seem especially charming, novel, or creative, and one can therefore attribute special creative verbal ability to speakers of other languages" (1985: 210).

Another variable that can influence the desire for a specific language or culture is the manner in which these are framed discursively (Cameron and Kulick 2003: 104). Language desire may be promoted by the media, advertising, or certain social groups, as can be seen in Piller and Takahashi's (2006) study. The authors interviewed five Japanese women living in Sydney, Australia, who relocated in order to improve their language skills and, potentially, to meet an Australian partner. They found that there was a strong desire for English-speaking white men among these Japanese women, which might be attributed to propagation by the media and by language schools. For instance, Japanese language school advertisements often suggest that female students will acquire English quickly because they are attracted to their good-looking, white

<sup>1</sup> While this does not apply to the couples in this study, the financial aspect may have contributed to the couples' choice of Switzerland as their country of residence.

teachers, and Western men tend to be portrayed as well-educated, sophisticated, kind and attractive (2006: 66).

These findings indicate that the desire for a particular language or culture can also be gender-specific, which is supported by previous research. In her interviews with bilingual couples, Piller discovered that there seems to be "a romantic desire for a type of masculinity (or femininity) that is stereotypically associated with another language" (2009: 57). Similarly, Visson (2009) found that some of the American men in her study seemed to look for '"femininity" in Russian brides, and portrayed American women, by contrast, as terrible partners. This demonstrates that the perceived attractiveness of the partner's language or culture is sometimes contrasted with a more negative image of the bilingual's own language and culture. It also underlines that the initial attraction between partners of different linguistic or cultural backgrounds is often based on stereotypical or idealized notions of the other language or culture. While attraction may thus be taken as an indicator of latent attitudes, these attitudes are often vague at the beginning of a relationship, and more concrete or complex attitudes usually only develop with increased contact.

#### **7.2.2 Attitudes towards languages, cultures, bilingualism and biculturalism**

The role that a bilingual couples' attitudes towards their languages, cultures and bilingualism play in terms of their language choice has been discussed in an earlier section (see section 5.2.3, "Attitude and motivation"). However, their attitudes may have an influence on several other areas, too, such as their partner selection, their choice of country of residence, their level of integration, as well as language learning and proficiency. The specific impact of attitudes on these areas can be hard to determine, as many other variables come into play simultaneously. Moreover, attitudes themselves can be difficult to measure, as they cannot be observed directly, and self-reports are not always reliable. Therefore, emotional responses to linguistic varieties — in matched guise tests, questionnaires or interviews — are often taken as manifestations of underlying language attitudes and used for the analysis of such attitudes (Chevalier 2014: 198). Attitudes are also highly complex and shaped by socio-cultural influences and personal experiences. This can be observed, for instance, in the results of a survey by Vildomec (1963) described by Pavlenko (2005: 38) of the individual language preferences of 40 European multilinguals. The participants named various (subjective) reasons for preferring one of their languages over others, such as "language dominance, beauty, expressivity, and family associations", whereas "[r]easons for disliking particular languages included ugliness, diffi‐ culty, and personal and political associations" (Pavlenko 2005: 38). Finally, "[a]ttitudes toward foreign languages, and the cultures they represent, are very often ambivalent. They are often dependent on the political climate and can shift quite rapidly" (Dewaele 2010: 136). All of these aspects ought to be borne in mind when one analyses language attitudes.

Whether a bilingual has a positive or a negative attitude towards his or her L2 and the culture associated with it, may have various long-term consequences. Positive attitudes, on the one hand, may lead to attraction or "language desire" (Piller 2002a; see previous section) and, ultimately, to a high level of fluency in the L2. Thus, Piller identified a high linguistic competence in many partners in bilingual relationships; out of 73 bilinguals of English and German who were interviewed, 27 claimed "that they had achieved high-level proficiency in their L2 and that they were passing for native speakers in some contexts" (2002b: 179). In fact, the author notes that the topic of passing as a native speaker was addressed in almost half of the recorded conversations (17 out of 38). In addition to resulting in a high level of language proficiency, very positive attitudes towards bilingualism and biculturalism may be connected to a sense of double cultural belonging. This was suggested by Licciardello and Damigella, who examined the attitudes of 50 mixed couples towards bilingualism and bicultur‐ alism, all including one Italian partner and one partner from various countries. Most of these couples reportedly transmit both languages and cultures to their children, and do not believe that this has any adverse effects on their children's linguistic development, but rather that it provides "an enrichment which is not only linguistic, but also relational, cultural and cognitive" (2013: 750). Positive attitudes may thus lead to multiple cultural and linguistic identities, for bilingual couples as well as their children. In contrast, negative attitudes can inhibit the development of such multiple identities and may pose a variety of problems. The non-native partner may lack the motivation to engage with the local population or culture, or to achieve a high level of fluency in his or her L2. This can result in a sense of isolation, and also present difficulties for the relationship. Such issues were reported by some of the Francophone women married to Anglophone men in Heller and Lévy's study (1992: 41; see section 3.3.2, "Bilingual couples").

Attitudes towards both languages and both cultures, as well as bilingualism, may determine language choices within the relationship as well as language transmission to children. Schüpbach (2009) analysed the life stories of 14 mi‐ grants to Australia from German-speaking Switzerland, examining the factors influencing their language transmission to their children. Eleven of the Swiss partners whom the author interviewed were in exogamous relationships with an Australian partner, though only two of them were transmitting their mother tongue to their children, and none of them spoke Swiss German to his or her partner. The author discovered that, besides family type and linguistic environment, the immigrants' linguistic decisions had been strongly influenced by their attitudes and past life experiences (2009: 25). Furthermore, she posits that attitudes and beliefs also help the bilinguals to "make sense of their experiences and to construct a coherent narrative of their decisions and of the outcomes they report" (2009: 28). Thus, some migrants justified abandoning their mother tongues on the grounds that it is a non-standard variety and, in their opinion, of no immediate use, or explained their choice by the rather negative attitudes towards languages other than English in the socio-cultural climate at the time of their migration. Thus, the author concludes that among migrants and their partners, Swiss German is "likely to be viewed in a different light to a standardised, codified and more widely used variety such as Standard German" (2009: 28).

At times, attitudes and linguistic practices are also connected to concerns of cultural and linguistic identity. This can be seen in earlier research by Schüpbach, which focuses on the language practices, attitudes and identity construction among the same Swiss immigrants. Her analysis revealed a very complex and diverse pattern of identity construction among the bilinguals, who "present[ed] themselves as exceptional Swiss while retaining the notion of Switzerland as a *Sonderfall* ['special case'] and as exceptional immigrants in Australia" (2008: 35). At the same time, there were various degrees of multilingualism and assimilation among the migrants, ranging from largely maintaining Swiss German within the home environment to complete immer‐ sion in the community language both privately and publicly. The bilinguals' level of linguistic integration tended to be connected to the value they attached to bilingualism and their mother tongue, and their level of identification with their adopted country.

In the case of bilingual couples, not only may their attitudes towards both of their languages be of relevance, but also their attitudes towards specific varieties of these languages. For one, there could be substantial differences in the partners' attitudes towards Standard German as opposed to specific Swiss dialects, which may influence their linguistic choices. Moreover, the Swiss partner may favour a specific variety of English, due to personal experiences or his or her linguistic education. Such a preference was found in research carried out by Henderson et al. (2012), who surveyed English teachers in several European countries in order to determine if they favoured a particular variety of English in class. Among these teachers, the Swiss respondents displayed some preference for RP (Received Pronunciation) over GA (General American), though "a type of International English" (2012: 20) was frequently mentioned, too. Interestingly, the Swiss teachers believed their students' preferences to be largely in line with their own, with only a slightly smaller preference for RP over GA among the students than the teachers.

Attitudes towards different varieties of English in Switzerland were investi‐ gated more extensively by Chevalier (2014), who surveyed 98 Swiss university students. She discovered that, although British English is the national variety that has traditionally been taught most often at Swiss schools, the students favoured a large range of national or regional varieties. These preferences were frequently linked to emotional attachments that the students had formed with a particular region, as in many cases, there was a correlation between the variety the students preferred and the location where they spent time abroad (2014: 210). According to Chevalier, this could indicate that a "rise of the regional", which has been observed among native speakers (Mugglestone 2003: 273), is also occurring among non-native speakers. Moreover, the responses revealed that the students had different associations with different varieties, possibly due, to some degree, to their amount and type of exposure to each variety. Thus, British English was considered elegant, sophisticated, and arrogant, while American English was perceived to be more direct and natural, though there was no clear preference between the two varieties (Chevalier 2014: 208).

Besides attitudes towards inner-circle varieties of English, attitudes towards non-native varieties may also be relevant. Murray analysed the attitudes of native and non-native English teachers in Switzerland towards Euro-English (a hypothetical European variety of English including non-native features). She found that non-native speakers tended to have a more conservative view than native speakers, since they were less likely to accept non-native structures and displayed less enthusiasm for Euro-English, while the latter were more tolerant of errors (2003b: 159). For partners in a bilingual relationship, this might mean that the non-native speaker is more critical of his or her linguistic shortcomings, or that he or she has a high motivation to achieve native-like competence (which may not be equally important to his or her partner).

Lastly, attitudes towards languages and/or cultures can also be gendered, and gender may influence both language learning and the attraction to the other sex. In an investigation into the motivation to learn foreign languages among English secondary school students, Williams, Burden and Lanvers (2002) discovered that, while both genders were more motivated to learn German than French, this preference was especially prominent among boys, who explained their preference in terms of the former being a more masculine language than the latter. The fact that French is deemed a very feminine language may thus deter male students from learning it — even though the language may sound attractive to Anglophone men (when spoken by a French woman, for instance). A similar case was discussed by Pavlenko, who described the case of a male author who failed in multiple attempts to master French, which he himself attributed to his own perception of the language as effeminate (2001b: 147). This underlines that gender ideologies and language attitudes may either inhibit or promote language learning.

#### **7.2.3 Attitudes towards raising bilingual children**

Bilingual couples' attitudes towards their languages and cultures are likely to influence their attitudes towards (childhood) bilingualism, as well as their strategies in raising bilingual children. In turn, the language choices bilingual couples make in their children's upbringing can reveal hidden facets to their attitudes. As has been mentioned earlier, childhood bilingualism is often viewed extremely positively among elite bilinguals (Boyd 1998; see section 2.4, "English in Switzerland"), as has been demonstrated in several studies. For instance, all of the couples in Piller's research answered the question of whether they raised their children bilingually and biculturally affirmatively. In fact, many couples expressed a strong commitment to their children's bilingualism and framed it as an investment and as effortless learning (2002a: 252–253). Moreover, the participants shared the notion that their children would obtain the best of both worlds by being bilingual, and believed that they would thus become world citizens. Many bilingual couples also believed that their children's bilingualism may be useful later in life, and that it helped them maintain a connection to their extended family and both cultures (2002a: ch 9). Comparably positive attitudes are reported by Grearson and Smith, who relate their own stories as partners in intercultural relationships; they think that their children are lucky, "not because they are American, but because they have two cultures to draw from, a growing knowledge of and commitment to a wider world" (2009: 86).

The desire to raise children bilingually is often linked to the hope that they will become native (and thus legitimate) speakers of both languages. Such aspirations could be observed in many non-native participants in Piller's study, which the author explains with prevailing ideologies surrounding nativeness:

Even as highly advanced and proficient second-language users, [the non-native speakers] still grapple with their legitimacy as speakers of that language. Conse‐ quently, their dreams are for their children to be fully legitimate speakers of both. A pervasive ideology of native speakership engenders the belief that only speakers who learned a language from birth are legitimate speakers of that language. (Piller 2009: 63)

The aspect of legitimacy is also related to the concept of the mother tongue. Bilinguals may wish to speak their mother tongue with their children so that they acquire this language as a mother tongue themselves; another reason is that many bilinguals find it easier to establish an emotional connection in their mother tongue, as it is a more emotional language for them (see chapter 8, "'In Swiss German, *I lieb di*, that's strange': Expressing emotions"). Moreover, some parents feel that speaking in their L1 to their children is more natural or genuine, which has been ascribed to a "romantic ideology of first language primacy" (Pavlenko 2004: 192).

It may come as no surprise that such overwhelmingly positive expectations of one's children's bilingualism are often not fulfilled. This can be seen in Piller's research, which revealed that many elite bilinguals were disappointed with their children's linguistic development, as they expected native-like proficiency in return for their commitment to their children's bilingualism. Especially the parents of older children frequently "exhibit[ed] a palpable sense of linguistic failure" (2009: 63). The same applies in situations where raising a bilingual child is met with difficulties due to external factors. Heller and Lévy, who studied 28 Francophone women who married Anglophone men in Ontario, Canada, found that these women "[tried] to bring up their children to be bilingual, but for many it [was] an uphill battle, because English [was] so overwhelming a force in their children's lives" (1992: 12). As a consequence, some of them expressed feelings of shame, or blamed themselves and their environment for not encouraging their children to speak more French (1992: 36–37). This demonstrates that bilinguals' positive attitudes can be met with disappointment if they are not shared by their children and/or their environment.

#### **7.3 Initial attraction between the partners**

The overview of previous research on attraction and attitudes has underlined the importance of these elements with regard to bilingual couples' language choices, but also concerning other aspects of their relationships. In order to explore these topics further, I therefore aim to identify what linguistic or cultural aspects the interviewees found attractive in their partner, and to ascertain whether their attraction was connected to their partner's nationality, mother tongue, or gender. The couples were asked what they found attractive about each other when they first met, and, as a follow-up question, whether they found the fact that they are from different linguistic and cultural backgrounds attractive. An overview of their answers is given in Table 11.


Table 11: Perceived attractiveness of otherness Key: E = English, SG = Swiss German.

Table 11 indicates that most of the interviewees found the fact that their partner had a different linguistic and/or cultural background attractive, or even very attractive, when they met for the first time. In fact, only three of the interviewees — the Anglophone women — deemed these characteristics not particularly attractive, while one participant said that he had no reaction to his partner's Swiss background. Since the answers of the Swiss and the Anglophone participants concerning their initial attraction differ to a great extent, they will be discussed in two separate sections.

## **7.3.1 Swiss partners' attraction**

As Table 11 shows, all of the Swiss partners initially found their partner's language and/or culture attractive, which implies that they had a positive attitude towards these elements even before they met. A number of Swiss interviewees explain that they considered their partner's variety of English or manner of speaking appealing (Stephanie, Susanne, Martin, Sophia, Simon, Monika). In several cases, the attraction to their partner's variety of English ties in with their general interest in the country or the language:

```
Example 7.1: Susanne (Swiss, 29)
Susanne: <P<BR I .. <LO ^did LO> (find it attractive) because BR>P> / ((very soft))
                <PP I studied <HI ^English HI>PP> \
                <P ^so: P> _
                <P I was interested in the ^country: P> /
                <P ^and P> _ ((enumerative))
                <P<BR<:-) the ^language :-)>BR>P> /
                <P ^and P> _
                <P<:-) and he had <HI such HI> a beautiful ^English :-)>P> / (1287)
```
Interestingly, only the interviewees whose partners are from the British Isles mention that they specifically liked their partner's accent or manner of speaking, unlike any of the interviewees whose partners are from the USA and Australia. Moreover, not only accents which are close to RP are considered attractive by the Swiss partners; Simon SG and Stephanie SG, for instance, emphasize that they particularly like their partners' Irish accents ("he's ^^Irish, like I think ^^that's very attractive" [Stephanie 237]). Monika SG reports that, even though it took her some time to understand Dean's Essex accent, she found it extremely appealing:

```
Example 7.2: Monika (Swiss, 29)
447 Monika: <W<MRC it <HI ^^did HI> con^^tribute to that whole MRC>W> _
               <P<@ bad boy ^image @ P>@> \
               @@ <P<@ of the the English ^lad P>@> /
               <H> <P ^and e:h P> _
               … <A it was just very A> <P<HSK a^ttractive HSK>P> \\
448 Silja: <PP m^hm PP> //
449 Monika: = <P<HSK ^very HSK>P> \
               <PP<HSK very a^ttractive to me HSK>PP> _ (447–449)
```
In this segment, Monika stresses just how attractive she found Dean's language and the "bad boy image" she associated with his way of speaking. She speaks very softly and with a husky voice towards the end, which underlines her emotional involvement, and she repeats herself for additional emphasis ("very attractive", "very very attractive" [447, 449]).

The attraction to one's partner's language is often linked to cultural aspects, and the Swiss partners mention a number of culture-specific features which they found attractive about their partner. Some Swiss women mention that their partner is more progressive in his thinking than a typical Swiss man (Sophia SG), or that he is a gentleman (Monika SG, Stephanie SG):

```
Example 7.3: Monika (Swiss, 29)
Monika: <P ^he used to P> \
               <P<CRK and still ^i:s CRK>P> \\
               <MRC<P<CRK a a a ^big CRK> <LO ^charmer LO>P>MRC> \
               <CRE he's just the most ^charming CRE> _
               … <P<:-) ^guy :-)>P> \~
               <P ^I've: P> _
               <P<A<CRK ^probably met CRK>A>P> \
               <P he's the ^gentleman P> _
               <PP the ^English gentleman PP> _ (367)
```
Others report that meeting their partner was connected to their stay abroad and the wonderful time they were having (Sarah, Simon, Philipp, Katia). Thus, Sarah and Simon felt that their relationships gave them a chance to interact with the locals and have access to their culture. Simon had been living in Northern Ireland for almost seven months and loved being there, and being with Claire gave him "a bit of a … of a connection to the country" (960). Philipp explains that he felt "ex^^tremely comfortable that year in the States" (563), as, despite being an introvert, he found it so easy to interact with the people he met, which was something that added to the attraction. Katia, on the other hand, reportedly felt so American by the time she met her partner during her high school exchange year that she hardly thought there to be cultural differences.

Many of the Swiss participants explained that when they met their partner, they already had a long-term interest in their partner's language or culture, for instance because of a passion for literature (Susanne, Philipp, Sophia) or history (Simon, Sophia). In Philipp's case, this can be traced back to a childhood interest in Native Americans, which contributed to his decision to study English at university. He eventually spent two semesters in Minnesota, where his fascination was rekindled:

584 ((…)) <W **I** <HI **probably** HI> **read about** <MRC **every docum:::entary ^^book** MRC>W> /

<P<W **that has ever walked .. the ^earth** and W>P> \

… <PP<HSK<MRC **^^this is the ^^country** MRC>HSK>PP> \\ ((emphatic))

<H> **on** <MRC **^any ^major ^^chief** MRC> //

**that I've always ^^read about**

and I ended up in Minne^sota \

```
Example 7.4: Karen (American, 51) and Philipp (Swiss, 47)
578 Philipp: <:-) when I was a ^kid :-)> \~
               <P I ^was::: P> _
               … <PP<BR I was totally ^hooked on: BR>PP> \
               <P<HSK on Native A^mericans HSK>P> \
               <P and ^sto:ries and P> _
               <H:> [<P I ^read P>] /
579 Karen: [<P Karl May P>] /~
580 Philipp: = I I ^read /
               … <H> Karl May is just the ^prelude \\
```
**that there ^i:s** / <LO on on LO> ^^any: \_

<sup>586</sup> ((…)) <P then I ended up ^there P> //~

**and I ^realized** \

 

 

#### <sup>579</sup> Karen: [<P Karl May P>] /~ 182 7 "This uh foreign girl with a great accent": Attitudes and attraction

**I was totally ^hooked on:** \

<P<HSK **on Native A^mericans** HSK>P> \

… <H> Karl May is just the ^prelude \\

<P and ^sto:ries and P> \_ <H:> [<P I ^read P>] /

<sup>580</sup> Philipp: = I I ^read /


Philipp's speech is extremely emphatic, with several of strong stresses, length‐ ening, as well as marked sequences (e.g. <MRC every docum:::entary ^^book MRC> [584]). All of these features are used more often in this sequence than he used them overall.<sup>2</sup> Thus, Philipp underlines the depth of his fascination with Native Americans, but also his strong emotional reaction to finally being in the place about which he had read so much. This emotional involvement is further accentuated by his halting speech, repetitions (e.g. "it was:" [588]) and a very soft, creaky, husky or breathy voice quality.<sup>3</sup> Philipp highlights that, because of

<sup>2</sup> Relative to the number of intonation units, strong stresses (^^) were used in 228% more often in this extract than overall; lengthening (:) occurred 7% more frequently and marked speech (<MRC>) was used even 579% more often here than overall.

<sup>3</sup> While halting speech and repetitions cannot be compared to Philipp's speaking behaviour overall, he used more very soft speech, husky, breathy and creaky voice quality here than overall (on average 25% more, relative to the number of IUs).

his long-term interest in the country, this was not a superficial attraction to a beautiful English-speaking woman, but rather "a ^^lot more profound" (588). The extract is exemplary of the strong attraction most of the Swiss participants initially felt for their partner's linguistic and cultural background.

#### **7.3.2 Anglophone partners' attraction**

While all of the Swiss participants believe that their partner's background constitutes part of their initial appeal, this only applies to six of the English speakers. The Anglophone partners' responses also differed considerably from the Swiss interviewees' in terms of what exactly they found attractive about their partner's background. Thus, rather than being attracted to the language or the specific culture of their partner, some of the English speakers generally found the fact that the Swiss were foreign and or spoke with an accent attractive. A few of the English speakers report that they found their partner or their differences "exotic", as the following two examples show:

```
Example 7.5: David (English, 32)
David: <HI there was HI> something e^^xotic <P<BR<LO about her LO>BR>P> _
               … <H::> … <WH about this this WH> <:-) ^^foreign girl :-)> _
               <:-) coming into my <WH ^life WH>:-)> /~ (500)
Example 7.6: Dean (English, 29) and Monika (Swiss, 29) 
Dean:
Monika:
Dean:
               <W for <HI sure HI> I mean there <HI ^is HI> some kind of W> _
               <CRK you ^know em:: CRK> \\~
               <H::> <Hx:> <PP<HI<BR what's it what's the #expr- #so- BR>HI>PP> _
               <HSK something e^xotic [about being] HSK> _
                                      [<P m^hm P>] \
               <W with someone <LO from another LO> ^country: W> _
               <HI ^you know HI> \
               there's there's a lot of ^mystery involved _ (1482–1484)
```
In these extracts, Dean and David portray their partners as unusual and mysterious, and underline their fascination by alternating pitch (e.g. wide intonation contour <W>), loudness (e.g. whispering <WH>) and voice quality (e.g. breathy or husky <BR> <HSK>). In the eyes of some (male) Anglophone partners, their partner's foreignness was accentuated by the manner in which she dressed back then. For instance, David <sup>E</sup> found it interesting that Susanne SG often wore a skirt and stockings, and Robert <sup>E</sup> liked the fact that Stephanie SG frequently wore scarves. To his mind, this gave her a French look, which he considered very appealing:

```
Example 7.7: Robert (English-Irish, 20) and Stephanie (Swiss, 24) 
Robert: <P<HI ^I was HI> always P> _
               <P<HI ^I was interested HI> at the beginning P> _
               <P because ^somehow P> _
               … <%%> <P<:-)<HI I thought she looked like .. ^French HI>:-)>P> //
               <P<:-) you ^know :-)>P> /
               <P<:-) and she reminded me of this like ^French girl :-)>P> \
               [<:-) and I .. I was <HI ^really HI>:-)>] \
Silja: [@@@@@]@@
Stephanie: [@@@] 
Robert: <@ I was <HI really a^ttra@cted HI> to@ tha@t @> @@@@@ _ (204–207)
```
Even though most English-speaking partners were aware that their partner was Swiss, a number of them declare that they found their partner attractive because her appearance reminded them of another nationality they associate with attractiveness, namely French (Robert) and Swedish (Tim, Craig, David).

While none of the Anglophone interviewees reports that he or she considered it attractive that his or her partner spoke (Swiss) German, a number of them reveal that they liked the fact that she had an accent when they met (Tim, Robert, Dean, Craig). Dean, for instance, remembers that he and his friends really liked the mixture between Monika's Swiss accent and Essex English, and thought that it was sexy, so that they repeatedly asked her to say certain phrases. The appeal of non-native accents can be seen in the following examples:

```
Example 7.8: Dean (English, 29)
Dean: <H> <P<HI ^I thought HI> Monika's P> _
          <PP<CRK accent was ^sexy CRK>PP> \ (400)
Example 7.9: Craig (American, 32) and Katia (Swiss, 28) 
Craig: <:-) this uh ^foreign :-)> _
 girl with [a ..] great ^accent and \
Katia: [@@] 
 [<WH shut: ^up WH>] _
Craig: [@@@@] (231–233)
```
**Example 7.10: Tim (Australian, 29) and Sarah (Swiss, 23)** 

Sarah: = <P<LO uh **<Hx> and I wanna** <:-) **^lose** [**it**] :-)>LO>P> /

Tim: [^and] \_ <DOW I [I'd I'd .. made a mis^take DOW>] \~

Tim: <P<DOW **I thought she was** <LO **^Swedish** but LO>DOW>P> \ (377–382)

<HI **to to ^start off** HI> \ **it was the ^accent** \

Tim: <HI for ^me: HI> //

Silja: … <P<:-) o^kay :-)>P> \/~

Sarah: [@@@ <:-) ^yeah :-)>] \~

Craig: <:-) **this uh ^foreign** :-)> \_

```
Example 7.10: Tim (Australian, 29) and Sarah (Swiss, 23) 
Tim: <HI for ^me: HI> //
             <HI to to ^start off HI> \
             it was the ^accent \
Silja: … <P<:-) o^kay :-)>P> \/~
Sarah: = <P<LO uh <Hx> and I wanna <:-) ^lose [it] :-)>LO>P> /
Tim: [^and] _
             <DOW I [I'd I'd .. made a mis^take DOW>] \~
Sarah:
Tim:
              [@@@ <:-) ^yeah :-)>] \~
             <P<DOW I thought she was <LO ^Swedish but LO>DOW>P> \ (377–382)
```
Craig: [@@@@] (231–233)

**girl with** [a ..] **great ^accent** and \

As the extracts demonstrate, the Swiss interviewees are often somewhat embar‐ rassed at their partner's mention of their accent, even if it is phrased positively, as they strive to lose their accent. Most of them have accomplished this to a remarkable degree, and many of them have adopted a manner of speaking that is close to their partner's variety. As expected, those Swiss participants who have been with their partner for the shortest period of time, Stephanie (1 year) and Sarah (3 years), have adjusted their accent the least to their partner's. Both of their partners view this positively; Tim states that "the accent still does it for [him]" (429), and Robert reports he really likes Stephanie's accent and would hate for her to lose it.

```
Example 7.11: Robert (English-Irish, 20)
Robert: <H:> and I ^like /
               <P<HSK I like ^Stephanie's like accent HSK>P> \ 
               <P sinc:e P> _
               <P I find it really <HI ^charming HI>P> _
               <P<HSK I I would HSK><HI ^hate HI> that somehow P> _
               <P<HSK sh:e ^she HSK>P> \
               <P ^tried to speak P> _
               … <P<HSK E- ^English or so HSK>P> /
               <P<HSK like with this English ^accent HSK>P> //
               … <PP I th- think it's very ^charming PP> \
               … <PP<HSK<HI I'm very ^happy yeah HI>HSK>PP> \~ (661)
```
Not surprisingly, the positive attitude towards a foreign language appears to coincide with a positive attitude towards the corresponding culture, as the English speakers who initially considered the fact that their partner is from a different country attractive also seem to be the ones who found their accent appealing. At the same time, these positive views of non-native accents seem to stand in contrast with those of some Swiss partners, who do not find it attractive when their partner speaks (Swiss) German with an accent.

There are, however, also four native speakers of English who report that they did not find their partner's foreign background particularly attractive, or even somewhat unattractive. According to Richard, his wife's background evoked no reaction in him. Courtney states that she was not "so: impressed that he was Swiss or anything" (866–868), and that she did not even realize that English was not Martin's mother tongue for the first half an hour of their first conversation. Karen reports that she did not find her partner's background attractive, since she was not at all interested in going to the German-speaking part of Switzerland (though she mentions that people sometimes assume that this was a motivating factor). Claire even found it somewhat off-putting that Simon was not a local:

```
Example 7.12: Claire (Northern Irish, 28)
Claire: <P at the start I ^didn't (find it attractive) P> /
              ((…)) <P at the start I though like ^what P> //
              <HI<P ^where is he from P>HI> _
              <P a ^foreigner kinda thing P> _
              <P<@ I r@emember thinkin' at the very ^start @>P> _ (964)
```
While many interviewees see their partner's foreignness in a positive light, it is portrayed as unattractive by Claire. It is conceivable that the element of gender plays a role in this regard, as all three female English speakers did not find their partner's foreignness or non-native accent appealing, unlike the Anglophone men.

## **7.4 The partners' attitudes towards each other's culture and language**

The element of attraction already gives an impression of the partners' initial attitudes, though it is only one component of their attitudes, and may neither reflect the interviewees' current attitudes towards their partner's language and culture, nor the development of their attitudes. In order to explore these aspects further, the interviewees were asked how they felt about each other's language and culture before they met, and whether there were any aspects about their partner, their partner's country, their culture or their language that they currently really liked or disliked. The answers to these questions are summarized in Table 12 (first question) and Table 13 (second and third question) below.


Table 12: Initial attitude towards partner's language and culture Key: E = English, SG = Swiss German.


Table 13: Current attitude towards partner's language and culture

Key: – = tendency towards negative attitude, –/+ = relatively balanced or neutral attitude, + = tendency towards positive attitude.

Tables 12 and 13 are to be interpreted as a rough indication of the bilinguals' attitudes based on their accounts during the interviews. However, it is worth noting that measuring attitudes based on conversations is difficult and poten‐ tially imprecise, for instance because some speakers are more straightforward with their opinions than others, or they may be reluctant to criticize their partner's language or culture. One participant, to give an example, seemed to avoid criticizing anything relating to his wife's language or culture, stating that there is nothing about her or her culture that he would "*openly* dislike" (Richard E 962, emphasis mine), which may imply that there are nonetheless aspects that he dislikes. Meanwhile, other participants immediately seized the opportunity to produce a catalogue of aspects they like and dislike. Despite these differences in candour, the participants' extensive comments on the subject can be expected to serve as a relatively reliable indicator of their attitudes. The bilinguals' answers will again be discussed in two separate sections, depending on their mother tongue, as many of their answers are culture- or language-specific.

#### **7.4.1 Swiss partners' attitudes**

Table 12 shows that, on first meeting each other, the Swiss partners had a positive or very positive attitude towards their partner's language and/or culture. This was to be expected, considering that the majority of them met in an English-speaking country, and that their partner's language and culture were important elements of their attraction (see section 7.3, "Initial attraction between the partners"). A comparison between Tables 12 and 13 also indicates that, as their relationships progressed, the Swiss participants' positive attitudes towards the English language have remained, while their attitudes towards their partner's culture (as well as cultural traits they perceive in their partner) have become more ambivalent. Thus, all interviewees mention aspects that they presently like as well as aspects that they do not like about their partner's culture.

The most critical opinions that were expressed during the interviews were the Swiss participants' current views on their partner's culture or home country, and many of them list a variety of characteristics that they view negatively. These are often tied in with common stereotypes about their partner's country, such as the English drinking culture (Stephanie) or whinging (Sophia), the American superficiality (Deborah) and the perception of the US as a very legalistic country (Philipp). Other negative views are based on personal experiences, for example that some Anglophone people are very non-committal and say things that they do not really mean (Susanne, Sarah). While most of the Swiss participants do not appear to be bothered greatly by the negative aspects they perceive, some feel quite strongly about certain aspects of their partner's country or culture. This is most obviously the case for Sophia and Simon, as can be seen in the two extracts below.

```
Example 7.13: Richard (British, 29) and Sophia (Swiss, 31). 
810 Sophia: … <NAS<LO and what I LO> ^^absolutely ^^dislike is that NAS> \
               … <L they ^still: L> /
               … <CRK seem to .. <HI ^think HI>CRK> /
               <CRK that they're this [great <HI ^empire HI>]CRK> _
811 Richard: [@@@@] ((softly))
812 Sophia: [<BR and that they're better than the ^rest BR>] //
813 Richard: [@@@@@@] 
               … ((clears his throat))
             [<H:> <Hx>] 
814 Sophia: <W [for example when they] <HI always HI> talk about the <HI
               ^^continent HI>W> \\
               <P<CRK<LO and them^selves LO>CRK>P> \~
               <P<CRK so <HI they're ^not HI> European CRK>P> \
               <P<CRK<LO they're LO> ^British CRK>P> \
               … <P<CRK<BR<MRC I ^^absolutely [^^hate that] MRC>BR>CRK>P> \
815 Richard: [@@@@@] ((chuckling))
816 Sophia: this .. this <HI ^patriotism HI> _
               this <HI ^nationalism HI> //
               … <DIM is something <CRK I ^don't understand CRK>DIM> \
817 Richard: … <H><Hx:>
818 Silja: = <PP m^hm PP> /
819 Richard: … <Hx>
820 Sophia: <P<CRK<DOW that is ^horrible DOW>CRK>P> _
821 Silja: … @@ ((softly))
822 Sophia: <CRK because ^sometimes I think CRK> _ ((shaky voice throughout turn))
               <P<CRK there's <HI ^^not HI><DOW really that much reason to be DOW>CRK>P> _
               <P<CRK<HI ^^so HI> proud of things CRK>P> _
               <P<CRK any ^more: CRK>P> \\ (810–822)
```
Sophia expresses her opinion in a forthright manner, and does not hesitate to openly criticize the English (unlike her husband, see previous section). She does not include her husband explicitly, referring to the English people as "they" throughout, rather than "you" (810, 812, 814). Nonetheless, Richard is noticeably uncomfortable here, which can be seen in the fact that he is chuckling, clearing his throat, inhaling and exhaling repeatedly, even though he does not object to any of Sophia's statements.

Similarly, Simon criticizes the ignorance of many people in Northern Ireland, both with regard to the country's own issues and with regard to world events. In his opinion, many Irish people are neither interested in these topics nor informed well enough:

```
Example 7.14: Claire (Northern Irish, 28) and Simon (Swiss, 34)
1050 Simon: <W I I <HI still HI> think people are quite ^ignorant W> \
               <P<CRK<LO when it ^comes to LO>CRK>P> \
               … <P<HSK<LO your^selves LO>HSK>P> /~
               … <P<HSK the whole .. <LO ^p:roblem LO>HSK>P> /
               <PP<LO<HSK about ^politics [and all] that HSK>LO>PP> \
1051 Claire: [<P<HSK ^yeah HSK>P>] \~
1052 Simon: <P<HSK<W and .. I <HI ^think HI> that's W>HSK>P> \
               <H:>
1053 Claire: … and it a^nnoys you _
1096 Simon: ((…)) <W<A<HI youse HI> don't have a clue what's going on in the ^world A>W> //~
               <LO<HSK ^basically HSK>LO> \
               … <H> <W you <HI buy HI> a ^newspaper W> _
                <HSK<W<A<MRC and <HI all HI> you read is about ^bloody <HI ^Jordan HI>MRC>A>W>HSK> \
                    ((English television personality Katie Price))
               <HSK<W and .. and and her <HI ^husbands HI> W>HSK> \\ 
1105 ((…)) like your ^soaps //~
                <W and your:: ce^lebrities W> \
                <W are <HI ^^much HI> more important W> \
                <A than what's going on in the ^world A> /~ (1050–1105)
```
While Sophia does not include her partner in her outburst against the English, Simon uses the second person throughout ("yourselves", "youse", "your"). These extracts demonstrate that, despite the fact that culture appears to have played an important role in the initial attraction to their partner, the Swiss participants can also be very critical of aspects of their partner's culture.

At same time, the Swiss also make reference to a number of characteristics of their partner's culture that they appreciate and admire. For instance, several participants mention that people from Great Britain are relaxed (Simon, Sophia, Stephanie), can truly party (Sophia), or that they are very friendly:

```
Example 7.15: Sophia (Swiss, 31)
Sophia: <HI<CRK ^^I li:ke CRK>HI> \
              … <HLT<CRK ^that .. the CRK>HLT> _
              <P<BR<HI British HI> are ^^so friendly BR>P> \\
              … <CRK<P they are P><HI ^^s:o HI> <DOW incredibly friendly DOW>CRK> \ 
              <P<CRK ^everywhere: CRK>P> \ (762)
```
This extract shows that, while Sophia is very critical of certain characteristics of her husband's compatriots, she thinks very highly of others. Thus, she reports that she loves English history, literature, and the British sense of humour, and she remembers being called an "Anglophile" by a university professor once.

In addition to certain cultural traits the Swiss partners appreciate, several of them remark that they have positive feelings towards their partner's culture because of people whom they met that represent this culture (Deborah, Katia, Sarah, Philipp, Monika). The following example gives an indication of the high level of enthusiasm and affection with which the participants remember these encounters:

```
Example 7.16: Craig (American, 32) and Katia (Swiss, 28)
360 Katia: <H> the <HI ^people HI> I met \
               were just .. ^awesome <LO people LO> \
               <P ^and P> _
               <P<HSK ^^so ex^tremely: HSK>P> _
               … <H::> ^um: <Hx> _
               I don't ^know \/~
               <Hx> ^just <Hx> _
               in [^love] \
361 Craig: [<P ^open P>] _
362 Katia: in ^love <P with these people P> \
366 ((…)) <P I met the ^right people at the right time P> \
370 ((…)) <PP for ^me it's just PP> _
               <CRK I think I <HI ^realized HI> that CRK> _
               <H> <HLT it ^doesn't .. really: HLT> _
               <BR I mean ^yeah there: BR> \
               <HI there's ^cultural differences and whatever HI> \
               but I ^think /
               <H:> <MRC no matter <HI ^^where you HI> ^^are it ^^matters MRC> _ 
               <P<MRC ^^what people you're a^^round MRC>P> /~ (360–370)
```
Katia's husband, Craig, recalls that he thought that Katia was "in love with the US", but she insists that this was solely due to the people she had met. Several of the interviewees report that they enjoyed themselves immensely during their stay abroad, or that their partner's country feels like home to them (Sarah, Simon, Katia, Philipp). Overall, the Swiss partners appear to have developed a rather balanced attitude, seeing both negative and positive aspects in their partner's culture.

In contrast, the Swiss interviewees' attitudes towards their partner's mother tongue are far less ambiguous; without an exception, they express positive feelings for the English language. This does not only apply to the participants who were studying English language and literature at university when they met (Stephanie, Susanne, Monika, Martin, Simon, Philipp); also the other Swiss participants have highly positive associations with the language. For many of the Swiss, English is not merely their second language, but has become central to their identity. Monika, for instance, states that she feels at home in the language, and that it has become a second mother tongue to her:

```
Example 7.17: Monika (Swiss, 29)
Monika: <P<W it's be^com:e W>P> //\~
               … <PP<HSK my second ^language HSK>PP> /
               <P it's be^come like a m:- m% P> --
               <P<A<W my mother <HI ^second HI> mother tongue W>A>P> /\\~
               … ((long pause))
               ((…)) <P<BR I feel ^^very comfortable BR>P> \
               <PP<HSK expressing myself in ^English HSK>PP> //
               <P and I feel at ^home P> /\~
               <PP<HSK in ^English HSK>PP> // (471–477)
```
The same is true for Philipp, who even professes that he feels more comfortable speaking his L2 than his L1. He reports that he is a very introverted person, especially when he has to speak (Swiss) German. In English, on the other hand, he is more extroverted and feels that he can connect with people in a manner in which he cannot in German. He explains his situation as follows:

```
Example 7.18: Karen (American, 51) and Philipp (Swiss, 47)
539 Philipp: <W I <HI ^feel HI> that I'm: W> \ 
                <W<LO I'm: LO> .. ^^two different people W> / 
                <LO when I ^us:e LO> _
                … <P<LO whichever ^language LO>P> _
                … and <HI ^English HI> <P<DOW allow:s for me: to be: DOW>P> \ ((rhythmical)) 
                … <PP<HI<HSK ^um: HSK>HI>PP> _
540 Karen: … <P<W<CRK more out^going CRK>W>P> _
541 Philipp: = <P ^mor:e P> _
                <P<HSK<DOW ^ja: I mean DOW>HSK>P> \
                <P it comes <LO ^down to LO>P> _
                <P<LO to more out^going LO>P> _
                <P<DOW and that's something <HSK I feel a^ttracted HSK>DOW>P> _
                <P<LO attracted ^to LO>P> /
543 ((…)) <P there's ^just P> //
                <P<W there's <HI so many HI> <DOW ^things about the German language DOW>W>P> _
                <P<W that are <HI so HI> ^^awkward W>P> \\
569 ((…)) <P I ^I: P> _
                <W<HI come HI> across daily situations ^^here W> \\~
                <LO ^where LO> _
                … I ^knew: //~
```
<W<RH **what I could ^do and ^say:** RH>W> /\~ <P<HSK **if I could speak ^English** HSK>P> \\

<PP<W<HSK in that [situ^ation] HSK>W>PP> \

@@@ <P<:-) **cause of the ^vehicle** :-)>P> \\ <H> <:-)<PP **I have to ^use** PP>:-)> \ ^and so

<P<MRC **it's it's ^^not ^^working** <LO **because** LO>MRC>P> \_

<P whoever is ^ther:e P> //\\~

<P ^to: P> \_

<sup>570</sup> Silja: [<PP ^yeah PP>] \_ <sup>571</sup> Philipp: <PP<HSK but it ^just HSK>PP> /

<LO **^where** LO> \_ … **I ^knew**: //~

<sup>569</sup> ((…)) <P I **^I:** P> \_

<P<W **that are** <HI **so** HI> **^^awkward** W>P> \\

 

```
<W<RH what I could ^do and ^say: RH>W> /\~
               <P<HSK if I could speak ^English HSK>P> \\
               <P ^to: P> _
               <P whoever is ^ther:e P> //\\~
               <PP<W<HSK in that [situ^ation] HSK>W>PP> \
570 Silja: [<PP ^yeah PP>] _
571 Philipp: <PP<HSK but it ^just HSK>PP> /
               <P<MRC it's it's ^^not ^^working <LO because LO>MRC>P> _
               @@@ <P<:-) cause of the ^vehicle :-)>P> \\
               <H> <:-)<PP I have to ^use PP>:-)> \
               <H:> <PP<HSK ^and so HSK>PP> \
               <F<HI I s- I s'^^ppose there's HI>F> _
               <W<F<HI I don't know HI>F> what that ^i:s W> \\~
               … <TSK> ^u:m _
               but there is <HI ^something HI> _
               … that ^^resonates with me: _
               … <P<HSK when it comes to the English ^language HSK>P> \
               <P<CRK ^that u:h CRK>P> _
               … <PP<HSK ^that HSK>PP> _
               <P I just ^haven't P> _
               <P<HSK known be^fore HSK>P> /\~ (539–571)
```
<P<W **there's** <HI **so many** HI> <DOW **^things about the German language** DOW>W>P> \_

Philipp expresses great frustration over the fact that he has to use (Swiss) German in his daily life, and views both languages very differently, which becomes evident in his choice of words in the extract above. While he describes the German language as "so awkward" (543), and sees it as a mere "vehicle" he is forced to use (571), the English language "allows" him to be the way he wants (539), and it "resonates" with him (571). For Philipp, the two languages differ in their level of emotionality and their social function, which is underlined by the fact that he corresponded with me in English before and after the interview — even though my first email to him was written in German — and that he advised me to switch to German during the interview if he was speaking too much. Philipp's case highlights that for some bilinguals, their L2 can become central to their personal identity or perform an important social function.

#### **7.4.2 Anglophone partners' attitudes**

Whereas the Swiss participants all had a fixed notion of their partner's language and culture when they met, this was not the case for the native speakers of English. Four of them said that they simply had no concept of Switzerland or Swiss German, and hence no associations — positive or negative — with the

culture or language (Robert, Richard, Dean, Tim). In the following example, Robert stresses just how little he knew about Switzerland back then:

```
Example 7.19: Robert (English-Irish, 20)
246 Robert: 
249
253
254 Silja:
255 Robert:
                <P <%%> uh ^really un:- P> --
                <P un^known P> //
                <P<HSK I didn't know <HI ^anything HI> really HSK>P> _
                ((…)) <W nothing o^ccurred really W> _
                ((…)) <PP<HSK when she said ^Switzerland there was really HSK>PP>\
                … <PP<HSK<HI ^nothing HI>HSK>PP> \\
                <P<HSK you know [no re^action] HSK>P> \
                                [<PP yeah .. ^yeah PP>] _
                <PP<BR I was like what's ^that BR>PP> // (246–255)
```
This reaction may be explained by Switzerland's small size, its distance to their home country (Tim), or the interviewees' relatively young age at the time (Robert, Richard, Dean). Another five of the Anglophone partners said that they had some notion of Switzerland, but that they knew fairly little about it, and nothing that would have resulted in a strong opinion. They had heard some common stereotypes, for instance, that it was a very clean country (Claire), or that it was small, and famous for chocolate and gold:

```
Example 7.20: David (English, 32)
David: <HI ^Switzerland HI> was just _
                … <%> eh this little: .. ^place /~
                <:-) the size of ^Wa:les :-)> /\~
                <@ in the mi@ddle of ^Europe @> \
                <P<@ ^which was @>P> \
                <H:> <P<:-) which: had a ^^bad reputation regarding ^^gold :-)>P> /\~ (1334)
```
Only one Anglophone participant, Joshua, stated that he had a generally positive attitude, mainly because he was excited to be in Switzerland and to learn (Swiss) German, as he considered this a rather difficult language and enjoyed the challenge.

As the Anglophone partners came into contact with their partner's language and culture, their attitudes changed and became more differentiated. Like the Swiss interviewees, the English speakers mention both aspects they like and aspects they dislike at present about the local culture and the Swiss. For instance, they appreciate that the Swiss are friendly and reliable (Tim, Claire), that it is a safe and clean country (Tim, Robert), or that there is less commercialism than in America (Karen). On the other hand, some criticize the level of formality (Claire, Robert), that the Swiss are self-conscious and not very easy-going (Claire, Robert, Dean), or that they cannot queue properly (Richard). An interesting account of the manner in which an attitude can evolve is given by Robert. He claims that he did not have any associations with the country before relocating; when he first arrived in Switzerland, he did not like how seemingly perfect it was, and that everyone looked beautiful. It all appeared fake to him, and he states that he wanted to "rebel" by "drop[ping] litter and stuff like this" (1017). Robert started looking for negative aspects to Switzerland and idealizing his home country. Yet after a visit to England, he began to change his mind about his host country:

```
Example 7.21: Robert (English-Irish, 20)
Robert: <W I came back to <HI ^Switzerland HI><CRK and I thought CRK>W> _
                <P<HSK I'm really <HI ^lucky HI>HSK>P> \
                <P<HSK you know ^this is HSK>P> \
                … <P<HSK I'm really ^lucky so HSK>P> \
                … I'm ^really \
                … <P<HSK<HI ^^happy HI> to be here HSK>P> _
                <P I'm ^really uh P> _
                … <W I <HI really ^like HI> Switzerland W> _
                <DIM<HSK I'm really a su^^pporter of Switzerland now HSK>DIM> /~ (1048)
```
Thus, Robert's attitude towards Switzerland evolved from being neutral to being the most negative of all the participants, and finally turned into one of the most expressly positive ones. The complete change in attitude that Robert underwent underlines that attitudes can be far from stable, and have the potential to evolve greatly with added exposure to the culture or language in question.

Thus, the attitudes towards Swiss culture among the Anglophone participants appear to have developed from rather vague ideas into much more specific views, which are overall positive for all of the interviewees, even though there are also some points of criticism. In contrast, the participants' attitudes towards (Swiss) German are very heterogeneous. Two of the participants, Karen and David, say that, while they do not really dislike the language anymore, they had a negative attitude towards it at the beginning. Karen remarks that she "used to not like German" (925), because she believes that it is much more difficult than French, which, in turn, she finds a "^^lovely language" (930). She also mentions that she does not "have anything that [she] prefer[s] to talk about in German" (1091). David, on the other hand, laments that it is difficult to learn German in Switzerland, and appears to like the local dialect less than Standard German (for instance, he jokingly asks his wife: "Aren't there disadvantages ((… )) of Swiss German compared to German?" [500–502], knowing that this will trigger a reaction). David used to dislike Standard German, but over time, he has become accustomed to it, and it has lost some of the negative connotations it used to carry for him:

```
Example 7.22: David (English, 32)
1371 David:
1372 Silja:
1373 David:
                <DOW<HI ^German's okay HI>DOW> \~
                <HI the ^more HI> //
                <P<W the <HI more HI> I ^hear it W>P> \
                <P the ^more P> /
                <P the ^more I P> \
                … <P you ^know I P> _
                <P I under^stand it # P> _
                … <P<HSK ^and HSK>P> _
                <P<HSK the more <LO fa^miliar it becomes LO>HSK>P> _
                <HSK and it is <MRC ^not ^just MRC> a HSK> _
                <H> <P<MRC a ^Second ^World ^War ^language MRC>P> \
                <P which is shouted ^out to P> _
                English ^prisoners \
                <P ^^climbing over the concentration P> _
                <P m^hm P> /
                <HSK camp ^^fence HSK> /
                <P<HSK or what^ever HSK>P> /
                … <PP ^and eh PP> _
                <P there's a bit ^more to it than P> _
                <P schnell ^schnell P> / (('fast, fast', i.e. 'hurry up'))
                … [@@@@@] ((chuckling))
1374 Susanne: [@@@@@] ((chuckling))
1375 Silja:
1376 David:
                @@@@@ 
                <PP<HSK ^^recktam HSK>PP> _ ((rechts um, halt! = 'right turn, stop!'))
                <PP<HI<HSK ^^halt .. ^^halt HSK>HI>PP> /~ ((shouting, aggressive voice quality, but soft))
                <P<:-) et^cetera :-)>P> \ (1371–1376)
```
David's attitude towards German is difficult to ascertain, partly because he is obviously joking to some extent, but also because, even though he asserts that his attitude has become less negative, his wording is only minimally positive: he states that he finds the language "okay" (1371), that it is not "just" a WWII language (1371) anymore, but that there is "a bit" more to it now for him (1373). At the same time, he portrays it negatively as a language which German Nazis shouted out "to *English* prisoners climbing over the concentration camp fence" (1371, emphasis mine).

In contrast to these rather negative attitudes, five of the Anglophone participants seem to have a relatively neutral or balanced attitude. Richard, Craig and Claire do not have a problem with the language, but simply view it as a practical necessity for their life in Switzerland. Claire, for instance, does not mind speaking the language, and speaks it regularly and naturally, yet for her it is mainly a "means to an end" (Simon SG 602). Richard does not say anything about Swiss German except that "language-wise, ((… )) there is no issue there at all" (854). This tends to be his stance throughout the interview; rather than mentioning positive or negative aspects, he emphasizes that things are unproblematic, stating that his wife's different background was "never ((… )) an obstacle" (752), for instance, or that "culture-wise, ((… )) [he's] just happy" (854–856). It is noticeable that, while Richard, Craig and Claire are fluent enough in Swiss German not to consider using it problematic, they do not express positive feelings towards the language itself either. In contrast, there are a few English speakers who appear to have a rather positive attitude towards the language itself, but at the same time struggle with their own language learning situation. Hence, Tim feels like an outsider because he does not speak or understand enough Swiss German to follow the conversation in social interactions. For Dean, who started learning German after settling in Switzerland, the country's diglossia is a source of frustration:

```
Example 7.23: Dean (English, 29)
```

```
Dean: <H::> <P and for ^me: P> //\~ 
                <ph> <Hx> <P it's it's ^hard I mean P> _
                <H:> <P<BR it's kind of a love ^hate BR>P> /~
                <P relationship with Swiss ^German P> \
                <P<HSK becaus:e you know HSK>P> \~
                <P<CRK I <%%> I learn <HI ^High German here HI>CRK>P> _
                <P<LO ^and: LO>P> _
                … <W<RH you know you <HI ^^still HI> feel like an outsider RH>W> \
                <W<RH when you <HI ^learn HI> High ^German RH>W> \
                <P<LO<HSK ^u::m HSK>LO>P> _
                … <P<LO<CRK ^an::d CRK>LO>P> _
                … ^you know I I _
                … I I <HI much HI> prefer the ^sound /~
                <H> <P<HSK and the ^tones <LO of e:h LO>HSK>P> \
                <P of <HI ^Swiss German HI>P> _ (478)
```
While Dean theoretically has a positive attitude towards Swiss German, he also emphasizes the difficulties that arise from the local language situation; for professional reasons, he has to acquire Standard German, which he likes less than Swiss German, and which does not suffice to become fully integrated into the local community.

Finally, there are also some participants who feel very positively about Swiss German (Joshua, Courtney, Robert). Joshua reports that he feels very comfortable speaking in Swiss German, and says that it "just rolls off [his] tongue" (246). The manner in which he speaks about the language echoes the feelings many Swiss participants articulate towards their L2:

```
Example 7.24: Joshua (American, 30) 
1156 Joshua: <PP<:-) <HI Swiss HI> German's ^awesome :-)>PP> /
1187 ((…)) <P<HSK<HI I think HI> it's ^coo:l HSK>P> /~
              <P<:-) I think it ^sounds cool: :-)>P> \
1195 ((…)) <P<BR I think it's ^so c- BR>P> --
              <P<BR I think it's ^cool BR>P> //
              it just <HI ^^flow:s HI> and it's _
              <H:> … <DOW<:-) and when I speak ^High German :-)>DOW> _
              <P<:-) it's ^like it's eh :-)>P> _
              <P<LO<A I don't ^know A>LO>P> _
              more ^formal: an:d \ (1156–1195)
```
Throughout the interview, Joshua repeats four times that he thinks that Swiss German is "cool", and often brings up the topic unsolicited. While part of the reason why Joshua enjoys the language may be his proficiency, there are also some less fluent speakers who like the language. Courtney would love to speak the language fluently, as she thinks that this would develop her identity, whereas Robert enjoys using German because it is "a new language" (615), and he finds it "quite cute" (615). Neither of them is proficient in the language (yet), but they do not perceive this as a problem, probably because they had only been living in Switzerland for approximately a year at the time of the interview. They are motivated to learn the language, and still view it as a novelty, which seems to be a key reason for their positive attitudes. In sum, the English-speaking partners have rather heterogeneous views on (Swiss) German, as some do not particularly like the language or struggle with learning it, whereas others express positive feelings towards it.

## **7.5 The couples' views on being in a bilingual, bicultural relationship**

To acquire a more profound understanding of the couples' attitudes, also their views on bilingualism and biculturalism in general should be explored. In order to accomplish this, the couples were asked what the advantages and disadvantages of living in a bilingual, bicultural relationship are. Overall, their responses were very positive, and almost all couples immediately started listing advantages, or aspects that can be seen as either advantages or disadvantages. Some participants even claimed that they could only think of advantages for themselves (Karen <sup>E</sup> , Deborah


X X X 1 2 3

X X 1 1 2

SG), or that there may be disadvantages, but they are only minor (Simon SG). Table 14 gives an overview of the advantages that were mentioned by the interviewees.<sup>4</sup>

Table 14: Perceived advantages of being in a bilingual, bicultural relationship Key: E = English; SG = Swiss German.

the best of both worlds

opportunity to speak another language

having bilingual

valuing one's

children X X 2 2

own culture more X 1 1 total 2 2 1 3 3 3 1 2 3 3 2 3 2 2 1 0 2 2 2 3 18 24 42

<sup>4</sup> It is worthy of note that these are simply the first aspects that came to the respondents' minds. If they did not mention a particular aspect, this does not mean that they do not also consider it an advantage. Which partner mentioned a particular point was often just a matter of who spoke first, and usually the interviewees agreed with their partner's statement.

The interviewees mentioned several aspects that they perceive to be advanta‐ geous for their relationship. Thus, many participants believe that being in a bilingual, bicultural relationship is more interesting, more special or more fun than being in a monolingual, monocultural one (n=9). The following example demonstrates this:


Robert is extremely positive about their relationship, repeating how interesting it is (355 and 357) and that he never gets tired of it (361 and 363). He adds emphasis to his statement by using the terms "always" and "never" several times, as well as marked speech ("it ^never ^seems to ^stop" [355]). Others are equally enthusiastic about their relationship. In example 7.26 below, Monika stresses how much she loves being in a bicultural relationship, and how special this relationship is to her.

```
Example 7.26: Monika (Swiss, 29)
Monika: <P<W<:-)<HI I ^^love HI> being in a bicultural relationship :-)>W>P> _
               <P and es^^pecially love being with P> _
               <P with an ^English: P> \
               … <P<HSK ^man HSK>P> /~
               ((…)) <P<DOW although it has become ^^normal for me DOW>P> \
               <P<LO ^an:d LO>P> _
               <P<W<LO is nothing LO> ^^special anymore W>P> _
               <P<DOW<CRK<MRC it ^still .. ^i:s .. ^special MRC>CRK>DOW>P> \\
               ((…)) <P it's ^great P> \~
               <P I ^love <LO being LO>P> _
               <P<LO<HSK in that re^lationship HSK>LO>P> _
               … <P be^^cause [<LO of its LO>]P> _
Silja: [<H> <P ^yeah P>] \~
Monika: <PP<LO<HSK di^versity HSK>LO>PP> \ (1477–1479)
```
Monika reports that many friends of hers say that they are jealous of her relationship with Dean, and that they find his English very beautiful. The fact that he is an Englishman and speaks with a British accent plays an important role for Monika's positive attitude towards her relationship. In addition to the interest in each other and their relationship, an advantage mentioned by a number of participants is that their communication has improved due to their situation (n=5) (see section 5.5.2, "Effective communication and implicit understanding").

Besides the positive effects on one's relationship, personal benefits may be derived from being with someone with a different cultural and linguistic background. Thus, some participants believe that it has expanded their mind and opened their eyes to the world, or that there are positive mental effects of being bilingual (n=8). Others say that the constant exposure to their L2 has benefitted them professionally, in their work as language teachers or during their final exams at university (n=4). There are also participants who think that they would not have had the chance to travel as extensively as they have, or to move countries, if it had not been for their partner (n=5). In addition to this, being part of a bilingual, bicultural couple affords them the opportunity to interact with people from various backgrounds and offers them access to international communities (n=3). Some interviewees mention the fact that they can raise their children bilingually as an advantage of their relationship (n=2) (see also section 7.6, "The couples' views on raising bilingual children"), or that they have started to appreciate their own culture more (n=1). Finally, the enjoyment of speaking another language is also perceived as an advantage (n=2).

While the majority of the participants believe that there are mainly advan‐ tages to being in a bilingual, bicultural relationship, the bilinguals also named some disadvantages. An overview of their responses is given in Table 15:


Table 15: Perceived disadvantages of being in a bilingual, bicultural relationship Key: E = English; SG = Swiss German.

Most of the disadvantages the couples see with regard to their bilingual, bicultural relationship are language-related. The disadvantage mentioned most frequently is that there can be misunderstandings, which at times lead to arguments (n=5). All of the participants who mention this also emphasize that there is some, but not much miscommunication between them, or that they have fewer conflicts now than they did initially. They highlight that their cultures are not that dissimilar, or that they are fluent L2 speakers, and that more misunderstandings would arise if this were not the case. Courtney <sup>E</sup> , for instance, points out that Martin SG and herself are both urbanites, and that their situation might be different if they were not:

```
Example 7.27: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
Courtney: <L I mean if I was ^^s:traight L> /~
              <L<MRC from a ^village in ^Ghana MRC>L> \
Martin: <W would be ^different W> //
Courtney: <P<LO it would be ^wa:y different LO>P> \
Martin: <P<CRK ^yeah CRK>P> \~ …
Courtney: … <P<LO and if you were straight from a village in LO> ^Appenzell P> \\ 
                                                              (1009–1013)
```
Some interviewees also declare that they experience no difficulties in under‐ standing each other, but are acquainted with bilingual couples who do, mainly because these couples are not as proficient in each other's native language (Susanne/David, Joshua/Deborah). Other language-related drawbacks that the bilinguals mention are that one's L1 can become less varied or less colourful, because there is a tendency to simplify one's speech when one is conversing with an L2 speaker (n=2), or that expressing oneself in one's L2 can occasionally pose difficulties (n=1). Moreover, a low level of proficiency in the community language can be challenging in social situations (n=1). There are also negative consequences of the distance between the partners' home countries, as it entails that one partner's extended family is far away (n=1), and that there are high travelling expenses (n=1).

Two interviewees voice concerns that are partly linguistic, though they are, at the same time, connected to aspects of social or personal identity. Robert feels that he cannot see all sides of his partner if he is not able to speak her mother tongue. He senses that she is a different person when she speaks Swiss German, and laments that other people can see this part of her, but he is not (yet) able to do so:

```
Example 7.28: Robert (English-Irish, 20) and Stephanie (Swiss, 24)
Robert: <P<HSK I ^had this problem like HSK>P> \\
               <P<HSK at the be^ginning s- a bit HSK>P> \\
               <P you ^kno:w when I P> \
               <P<W<HSK when I ^used to <LO hear her speak in Swiss German LO>HSK>W>P>_
               <LO<PP<CRK and it was ^like CRK>PP>LO> _
               <HSK this other ^person HSK> /
               ((…)) ^I was always like \
               <A<:-) is that two t- you ^know like :-)>A> \
               <P<:-)<HI ^yeah HI>:-)>P> /~
               <A there's lots of people who see ^^this: A> / 
               … <P<HSK ^this u:h HSK>P> \
               <HSK<P<HI ^person HI>P>HSK> _
               <HSK you ^kno:w and HSK> \
               <:-) somehow ^I never get :-)> /
               <P<:-) ^get #there :-)>P> // (374–376)
```
The desire to see Stephanie in her entirety contributes to Robert's motivation to learn her mother tongue; he reports that he would like to speak Swiss German better in order to understand this hidden side of her, and to experience what her friends are seeing. In contrast, Philipp SG struggles with his own dual identity, as he believes that, to some degree, establishing a new cultural or linguistic identity comes at the expense of one's existing identity. Partly due to his immersion in the Anglophone culture, he often does not really feel Swiss, or part of the local community:


Philipp's statement indicates that a struggle to connect with one community does not necessarily have to do with language skills. It is his L1 community, rather than speakers of his L2, that Philipp finds difficult to engage with, despite his conscious efforts to participate in the local community, for instance by joining the local men's choir. The difficulties he experiences in connecting with others are thus not due to a low level of language proficiency, but rather appear to be caused by an emotional barrier.

Finally, there are also a number of interviewees who do not mention any disadvantages of being in a bilingual, bicultural relationship (n=6); they either evade the question, or claim to be unable to answer it. Karen <sup>E</sup> , for example, says that she would not be in Switzerland if it was not for her partner, "so it's hard for [her] to think of disadvantages" (776). Overall, the couples' views on their bilingual and bicultural relationships appear to be very positive, as they mention a greater number of advantages than disadvantages (42 vs. 15), and several of them remark that the former clearly outweigh the latter. In most cases, I had to ask the couples a second time if they perceived any disadvantages. Moreover, several of them qualified immediately that this is either a minor issue, or one that does not arise often. In sum, it can be said that the couples all have very positive attitudes towards their bilingualism and biculturalism, even if they also see minor disadvantages in being in such a relationship.

#### **7.6 The couples' views on raising bilingual children**

In order to complete the overview of the couples' attitudes towards various aspects of their bilingual couplehood, I will also consider the participants' attitudes, expectations, and strategies with regard to raising bilingual children. As only four of the ten couples had children at the time of the interviews, and the children of three of these couples were still very young, not much can be said about the success of their chosen strategies. In spite of the children's young age, though, the parents among the interviewees have made new language choices, not just with regard to which language(s) they speak to their child, but also with regard to their couple language(s) and their mixing behaviour. Indeed, having children creates a dilemma for most bilinguals in this study, as they have agreed upon and gotten accustomed to a relationship language, but both partners may want to use their respective mother tongue with their children. As a consequence, several languages may be spoken in conversations that involve the bilinguals' children.

The four couples who have children are all raising them bilingually, and report that they have never even considered not doing so, which can be seen as indicative of their positive attitudes towards childhood bilingualism. They all agree that the best strategy to use is the *one parent, one language* (OPOL) strategy. In this strategy, parents speak to the child in their own native tongue, which results in a simultaneous bilingual language acquisition (see Harding and Riley 1986: 47). All of the couples with children explicitly mentioned this strategy during the interviews, which demonstrates that they have obtained information about potential methods of raising children bilingually. In order to raise their children in keeping with the OPOL strategy, Joshua <sup>E</sup> and Deborah SG have reduced their code-switching drastically in front of their children, and attempt to speak to them in their respective mother tongues (see example 6.13 in section 6.4, "The couples' reports on their language mixing"). Katia SG and Craig <sup>E</sup> have also been attempting to adhere to their native tongues since their daughter's birth. Sophia SG and Richard <sup>E</sup> have adopted an OPOL strategy as well, but they do not implement it consistently. Finally, Karen <sup>E</sup> and Philipp SG, whose children are already teenagers, originally intended to pursue the same strategy, but have had to become more flexible over time, as they view English as their family language, and Philipp found it hard to speak Swiss German to their children initially (see example 7.36 below). He speaks both languages to them equally often, and also code-switches on occasion. In contrast, Karen E speaks almost no German with the children "because they make fun of it" (252). Despite certain difficulties in implementing an OPOL strategy, however, all parents among the participants were initially determined to pursue such a strategy.

Although the couples who did not have children at the time of the interview may not have put as much thought into the linguistic upbringing of their potential future children, they may still hold an opinion on the subject. Thus, they were asked if they plan to raise their children bilingually, which they all answered in the affirmative. In fact, their responses revealed that, while all in‐ terviewees displayed very positive attitudes towards raising bilingual children, these were particularly pronounced among the couples without children. The following extract is exemplary of this:


Monika and Dean use very emphatic expressions, as do other interviewees ("of course" [Susanne SG 438], "definitely" [Robert <sup>E</sup> 494]). The idea of not raising one's children bilingually almost seems absurd to many of the bilinguals, as can be construed from Monika's use of husky voice (<HSK>), strong stresses (^^), and the high final rise (//) in her turn. The same applies to the other interviewees, for all of whom not raising their children bilingually is out of the question. Furthermore, the couples' answers reveal that many of them have already talked about the topic previously, even if they may not be planning to start a family any time soon.

The remarkably positive views on childhood bilingualism become evident in the manner in which the couples talk about their (potential) children's bilingual upbringing. The expressions used by the interviewees revolve around the notion that being raised bilingually generates a considerable personal advantage for the child. The following three examples demonstrate this:

```
Example 7.31: Sophia (Swiss, 31)
Sophia: <W<HI I think HI> it's <HI such HI> a ^gift W> / (466)
Example 7.32: Craig (American, 32) and Katia (Swiss, 28)
Craig: <PP<HSK it ^would be: uh um: HSK>PP> /
              … <DIM [let's say a big mistake not] ^not [to give] her this DIM> \
                                                            ((their daughter))
Katia: [<H:> <P that's no ^question P>] /
                                                   [<P ^that's: P>] _
Craig: [<PP<BR oppor^tunity or eh BR>PP>] \
Katia: [<PP ^yeah PP>] \~ (801–804)
Example 7.33: Courtney (English, 24) and Martin (Swiss, 26)
Martin:
Courtney:
Martin: 
              yea:h ^yeah /
              I ^would //
              <W you ^have to W> \\
              be^cause it's um:: \~
              … ^such a \
              @[@@@] ((Courtney expresses nonverbally that she is not enthusiastic about the topic of future children))
               [@@] 
              <W<:-) such an ad^vantage :-)>W> \\ (412–414)
```
Being raised bilingually is seen as an "advantage", an "opportunity" and a "gift" by the interviewees. The participants add emphasis to their positive assessments with the intensifier *such a/an* (Sophia SG, Martin SG) or a wide intonation contour (<W>) (Sophia SG). They portray raising their children bilingually as an obligation that they have as parents, something "you have to" do (Martin SG); not following through with it would be "a big mistake" (Craig <sup>E</sup> ). Usually the advantages that the couples expect their children to gain from their bilingualism are not specified, except that some believe that it will facilitate learning other languages (Stephanie SG, Katia SG). These positive attitudes are not surprising, considering that in Switzerland, the community language as well as English have a high prestige (the situation might, however, be different if they were living in an English-speaking country). Moreover, these reports mirror research on bilingual couples in similar situations, in particular the German-English bilingual couples studied by Piller (2002a), who also expressed very positive attitudes towards their children's bilingualism.

Due to their positive attitudes towards bilingualism, the couples want their children to be fluent, balanced bilinguals, and are worried about not succeeding, or about making mistakes in their children's upbringing. Such concerns are expressed most emphatically by the three couples who have young children (Joshua/Deborah, Richard/Sophia, Craig/Katia). Moreover, three couples volun‐ teer stories of children they know, or of whom they have heard, who grew up in a bilingual family, but did not end up being fluent, 'successful' bilinguals (Courtney/Martin, Claire/Simon, Craig/Katia). These accounts seem to serve as cautionary tales of what should be avoided in their children's linguistic upbringing, such as depriving them of communication with their extended family, making them feel disconnected from their cultural heritage, or creating a disadvantage for them at school (e.g. by only speaking the non-community language at home). A worry that is voiced repeatedly is that their child might get confused, which they fear will happen if do not consistently adhere to one language, or if they mix languages. Katia SG and Craig <sup>E</sup> discuss this topic extensively. When asked what their plan is with regard to their daughter's linguistic upbringing, Craig answers that "the plan is to not confuse her" (664–666), which they try to achieve by reading parenting books and by taking advice from other bilingual parents. Their biggest fear seems to be that their daughter will start confounding her languages. To prevent this, they try to keep their two languages strictly apart, and observe her closely:

#### **Example 7.34: Craig (American, 32) and Katia (Swiss, 28)**

724 Craig: 3[**^careful** about uh:] \_ 725 Katia: 3[<P **^definitely aware** P>] \

728 Craig: <P<BR **we're ^trying to** be uh: BR>P> \

726 Craig: 4[but even ^now] \_ 727 Katia: 4[<PP ^yeah PP>] \_

729 Katia: [<PP m^hm PP>] \_


730 Craig: [<PP<HSK trying to] **keep them ^separate** HSK>PP> \ (695–730)

**starts to mix ^up** / then I ^think we will \

722 Craig: 1[**we will**] **be a bit** 2[^**more** uh:] \ 723 Katia: 2[<P<LO ^**we're:** LO>P>] \_

```
724 Craig: 3[^careful about uh:] _
725 Katia: 3[<P ^definitely aware P>] \ 
726 Craig: 4[but even ^now] _
727 Katia: 4[<PP ^yeah PP>] _
728 Craig: <P<BR we're ^trying to be uh: BR>P> \
729 Katia: [<PP m^hm PP>] _
730 Craig: [<PP<HSK trying to] keep them ^separate HSK>PP> \ (695–730)
```
**since we're ^watching for this** \_

In the manner in which they phrase their statement, we can see that Katia and Craig have discussed the subject previously, and that they are very dedicated to their daughter's bilingual upbringing. The vocabulary they use revolves around paying close attention to her development ("as soon as we notice" [720], "we're watching for this" [720], "we'll be a bit more careful" [722–724], "we're definitely aware" [723–725]), so that they can intervene and modify their strategy if necessary. These efforts, Craig hopes, will lead to a successful bilingual upbringing (note also the repetition of "hopefully" [695–697]). Craig has the floor for most of example 7.34, but Katia supports him with several backchannel signals and by paraphrasing his statements, thus highlighting that this is a family issue, to which they are both committed. The case of Katia and Craig exemplifies the large amount of private language planning that is done by the couples who have children, even if these are still too young to speak.

For most of the participants, using their mother tongue to speak to their children is an obvious choice, and they cite a number of reasons for doing so. Thus, some believe that speaking a language other than their native tongue to their children would feel uncomfortable or weird to them (Katia SG 784 and 788), or that it would be unnatural (Sophia SG 466). In addition to the aspect of transmitting their own language, an important concern is that they may not be able to pass on their cultural identity to their children or establish a deep emotional connection if they are not able to speak their mother tongue. Moreover, a reservation that is shared by many of the bilinguals is that they do not speak their L2 competently enough to raise a child in this language, despite their high level of fluency (though there are exceptions to this, which will be discussed below). As an example, Katia SG says that she would not speak English to their daughter because she will always have an accent and will always make mistakes (even though her accent is barely perceptible and there were very few mistakes during the interview). She thinks that, because English is their family language, it would, in theory, be possible for her to speak English to their daughter, but she does not want to do this, as it could have a detrimental effect on her child:

\_

```
Example 7.35: Katia (Swiss, 28)
Katia: <H> I ^could (speak English) /
               <W but .. I don't really ^^wa:nt to W> //
               <P<LO because I ^know that you can: LO>P> _
               <P<LO screw ^up a kid LO>P> /
               <P<BR I mean it's ^not my mother tongue BR>P> _
               ((…)) <P<HSK and so I ^think HSK>P> /
               <H:::> <WH<@ I don't wanna screw ^up my kid @>WH> / (755–759)
```
Katia is convinced that speaking her mother tongue is the best for her daughter, and thinks that she would confuse her by using her L2 with her, even though she is proficient in the language. In the example above, she speaks very emphati‐ cally, using a wide intonation contour (<W>), stress and lengthening (^^wa:nt), repetition ("screw up [a/my] kid") and whispering (<WH>). Such worries are especially prominent among the bilinguals who already have children, while the other interviewees tend to have a more carefree view on their future children's bilingual upbringing.

Even though all of the participants agree that they ought to speak their respective mother tongues to their (potential) children, translating their inten‐ tions into reality sometimes proves difficult. This is the case for two of the participants with children, Joshua <sup>E</sup> and Philipp SG, who both sometimes use their L2 with their children. Joshua reports that he has to make a conscious effort to speak English to his children, and, as a consequence, sometimes mixes languages. Philipp, on the other hand, recounts that he initially spoke English to his first-born daughter, contrary to his original plans and despite his wife's protest:

```
Example 7.36: Karen (American, 51) and Philipp (Swiss, 47)
265 Karen: <F<W but <HI you s- HI> you ^used to W>F> \
                <F<W when <HI Erin was a HI> ^baby W>F> \
                <DOW you wanted to speak ^English to her DOW> \
                … <H:> and it ^took me: \~
                a f::- <HI ^several HI> months \
                <HSK<DOW to get you to speak DOW> .. ^German HSK> \
                … <P<CRK<W to her but then you got ^used to it W>CRK>P> \
                <PP<W<HSK<L and <HI ^you HI> mostly speak L>HSK>W>PP> \~
                <PP<W<HSK we:ll <LO it's about half LO> ^half now HSK>W>PP> /~
                <PP I don't [know ##] PP> /
```
<sup>266</sup> Philipp: <W [okay we're we're] both ^linguists W> \\

… <P as the educated ^linguists P> \

<P **the best of all** P> **^possible** \_

<P<LO ^which [is the <H:>] LO>P> \_

<P<HSK **we thought we'd we ^choose the** HSK>P> \

… ((chewing)) <TSK> <LO ^so LO> \_

<P **^strategie:s** P> \

<sup>267</sup> Silja: [<P ### P>] <sup>268</sup> Philipp: one ^person one \ <sup>269</sup> Silja: = <HI<NAS ah ^okay NAS>HI> \/~

<sup>270</sup> Philipp: = <TSK>

a f::- <HI **^several** HI> **months** \

<PP I don't [know ##] PP> /

<HSK<DOW **to get you to speak** DOW> .. **^German** HSK> \ … <P<CRK<W **to her but then you got ^used to it** W>CRK>P> \

<PP<W<HSK we:ll <LO it's about half LO> ^half now HSK>W>PP> /~

```
266 Philipp: <W [okay we're we're] both ^linguists W> \\
               … ((chewing)) <TSK> <LO ^so LO> _
               … <P as the educated ^linguists P> \
               <P<HSK we thought we'd we ^choose the HSK>P> \
               <P the best of all P> ^possible _
               <P ^strategie:s P> \
               <P<LO ^which [is the <H:>] LO>P> _
267 Silja: [<P ### P>] 
268 Philipp: one ^person one \
269 Silja: = <HI<NAS ah ^okay NAS>HI> \/~
270 Philipp: = <TSK>
271 Karen: = one [^language] _
272 Philipp:
273 Silja:
                 <DOW [one] ^language strategy and uh DOW> \
               = m^hm \\//~
274 Philipp: I was all ^up for \
               … <P<HSK speaking German to my ^children HSK>P> _
              <P<HSK when they come a^long and HSK>P> \
               … <TSK> <P<HSK ^but u:h HSK>P> _
 <P it ^proved P> _
275 Karen: … <P ^difficult P> //
276 Philipp: <P diffic- to be very ^difficult P> //
               <P<CRK and: and close to im^possible CRK>P> \
               … <H:> <TSK> <P re^memberin:g P> _
               <TSK> ^having my \
               … ((swallows)) <DOW ^first-born in my: DOW> \
               … <DOW<HI ^arms HI> the first time DOW> \
               ^it was just _
               there was ^^nothing \\
               <LO<P ^German P>LO> \
               … <LO<P coming ^out P>LO> _ (265–276)
```
The example above indicates that, before their daughter's birth, Philipp and Karen were both convinced that Philipp should only speak in his mother tongue to her. Due to their background in linguistics, the couple were well-informed about possible methods of raising a child bilingually, and considered the OPOL strategy "the best of all possible strategie:s" (266). In spite of his best intentions, however, Philipp did not manage to speak his mother tongue to their daughter. While Karen says that he "wanted to speak English" to Erin (265), Philipp describes his language choice as something that was out of his control, rather than a conscious decision ("it proved ((… )) very difficult"; "close to impossible"; "there was ^^nothing German … coming out" [274–276 in example 7.36]; "I couldn't bring myself to speak in German" [284 in example 7.37 below]). His turns are halting, with a large number of pauses, and his speech is accompanied by a number of nonverbal vocalisations (Trouvain 2014: 598), namely five tongue clicks (<TSK>), two lengthened inhalations (<H:>), as well as swallowing (all underlined in example 7.36). Such vocalisations often co-occur with word searches or hesitating, but also when the speaker is taking a stance (Ogden 2013: 308–309). In the present case, they convey Philipp's own sense of helplessness in the matter. His wife Karen appears to have devoted a considerable amount of time and effort to convincing him to speak German ("it took me several months ((… )) to convince you" [265]); she evidently saw it as her responsibility to make her husband speak German to their daughter, or deemed it rather important. Nonetheless, Philipp spoke English to his daughter for a few months, until she was exposed to Swiss German more regularly outside the family home, and then decided to switch to Swiss German at least part of the time. Now that she is a teenager, he speaks both languages to her, as he does to their two sons.

An important factor in using one's L2 to speak to one's children appears to be the perceived level of emotionality of this language. For some of the bilinguals, their couple language has acquired an emotional component and become part of their identity; using this language with their children therefore feels natural them. When asked why he initially found it difficult to speak Swiss German to his daughter, Philipp replies as follows:

```
Example 7.37: Karen (American, 51) and Philipp (Swiss, 47)
282 Philipp: we- ^I always thought \\
               <A<W<LO you know partner language is LO> <HI ^one HI> thing W>A> \\
               <A<P<LO and the language you speak with [your children] is a^nother but LO>P>A> _
283 Silja: [m^hm] / 
284 Philipp: <H> <P emotionally <LO<HSK they're so much co^nnected HSK>LO>P> \
               <PP<LO<CRK ^tha:t CRK>LO>PP> _
               <PP<LO<CRK<PP like .. we- CRK>LO>PP> _
               ^yeah \~
               <A<W in <HI some ways HI> I was <LO ^not surprised LO>W>A> \
               <PP<HSK that ^I HSK>PP> _
               … <P<HSK couldn't ^bring myself HSK>P> _
               … [<P to speak in ^German P>] _
285 Silja: [<P<HI m^hm HI>P>] /
286 Karen: [((drinking))]
287 Philipp: <PP<LO ^to her LO>PP> _ (282–287)
```
Because their couple language, his L2, carries more emotional weight for him than his mother tongue, Philipp could not make himself speak his mother tongue to his child. This underlines the crucial role of the level of emotionality attached to both languages (see also chapter 8, "'In Swiss German, *I lieb di*, that's strange': Expressing emotions"). Ultimately, the bilinguals' strategies and their language choice with their children are influenced by the level of emotionality they attribute to their languages, their personal circumstances, considerations of practicality, as well as their ideas of what might be advantageous for the child. These ideas, in turn, tend to be shaped by guides on bilingual parenting, linguistic research, as well as the opinions of their environment and the experiences of other people.

Ultimately, the couples' language use with their children is likely to have an influence on the manner in which the partners communicate with one another. Thus, some of the non-native speakers express worries about changing or even losing their couple language as a consequence of developing a family language. Several interviewees mention that it is important to them to maintain their present couple language after they have children, as can be seen in the follow example:

```
Example 7.38: Simon (Swiss, 34)
812 Simon: <W<A I <HI think HI> it would be really really im^portant A>W> \
              <HI<HSK ^^definitely HSK>HI> \\
              to speak ^English \\ 
816 ((…)) <P<HSK<HI I would HI> probably make an ^effort at the start HSK>P> \
              <P to speak Swiss ^German (to the child) P> // 
              … <P<HSK but make ^sure that HSK>P> \
              … <P ^our P> _
              <P language ^here <LO<HSK would stay HSK>LO>P> _
817 Silja: … <PP ^yeah PP> \~ 
818 Simon: <P ^English P> \ (812–818)
```
Simon puts a lot of emphasis on the importance of speaking English to each other and the child; in contrast, speaking Swiss German is something that he would "probably make an effort" to do at the beginning (816), which does not convey a great desire or urgency to do so. The interplay between their family language and their couple language may be part of the reason why only one of the four couples who have children (Katia/Craig) appears to have remained consistent in this respect so far, even though before their children's birth, all of the bilinguals envisioned that they would only speak their mother tongue to them. Furthermore, it highlights again that the decisions bilingual couples take before their children's birth about their linguistic upbringing are not always implemented later.

#### **7.7 Discussion and summary**

The analysis has revealed that there are considerable differences between the attitudes of the Swiss and the Anglophone interviewees towards their partners' language and culture, which are also noticeable in their initial attraction to each other. The Swiss partners exhibited highly positive attitudes towards most aspects of their bilingual relationships, which was reflected in the fact that they deemed their partners' linguistic and cultural backgrounds very attractive when they first met. Thus, many of the Swiss interviewees liked the fact that their relationship gave them access to native speakers and a closer connection to the country. Long before meeting their partner, several of them had an avid interest in their partner's culture or language, which had been kindled by a passion for literature or history. Hence, one partner in most couples expressed a type of "language desire" as described by Piller (2002a) — with the exception of one English speaker, Joshua <sup>E</sup> , this was usually the Swiss partner. Especially the Swiss participants who had partners from the British Isles showed great enthusiasm for their partners' specific varieties. This concurs with the results of Chevalier's (2014) survey on the attitudes of Swiss university students, as her respondents also showed a large range of preferences for regional or national varieties of English, often depending on the location where they had spent time abroad. It is also conceivable that this very positive view of regional varieties among the Swiss participants is linked to the ideology of dialect that is prevalent in Switzerland (see section 2.3, "Diglossia and the ideology of dialect").

The Swiss partners' reports of their initial attitudes are in line with their accounts of their first attraction to their partners, since their views of English and of many cultural aspects were highly positive when they first met. Over the course of their relationship, the Swiss participants' attitudes towards their partners' culture have become less one-sided, and they have begun to see both negative and positive aspects. While many of them still feel very positive about their partner's culture in general and the people whom they met, all of the Swiss also mentioned aspects they disliked about their partner's culture, and there were openly critical statements. In contrast, their highly positive attitudes towards the English language have remained, and have even become more pronounced for some, as the language has come to be a second mother tongue to them, and an integral component of their identity.

Unlike the Swiss, all of whom believe that their partner's language and/or culture constituted part of their initial attraction, only some of the Anglophone interviewees found their partner's background appealing at first. Their attrac‐ tion was also considerably less language- and culture-specific than their part‐ ners'; some simply found the fact that their partner was foreign and spoke with an accent attractive. Interestingly, gender may play a role in this regard, as all three Anglophone women among the participants neither felt attracted to their partner's language nor to their cultural background, and, to various degrees, they even considered their partner's foreignness unattractive. In contrast, all but one of the Anglophone men reported finding these characteristics attractive, and several of them portrayed their partners as exotic or mysterious, as they had a foreign appearance or an accent when they met. Since their first encounter, most Swiss partners have adapted their accent to their partner's to a great extent, though the partners of the two women who have retained their accent the most — Stephanie and Sarah — expressed very positive feelings about this. To some extent, these gender differences may be explained by Piller's proposition that there is "a romantic desire for a type of masculinity (or femininity) that is stereotypically associated with another language" (2009: 57), although in the case of the Anglophone males in this study, this desire does not seem to be specific to a language, but triggered by the notion of foreignness.

Similar to their feelings of attraction, the initial attitudes of the Anglophone partners were not particularly specific or differentiated, which was largely due to a lack of familiarity with their partner's language and culture. Four of the English speakers claimed to have had no associations with either, while another five had a rough idea based on popular stereotypes. Only one participant, Joshua E , who is the only one who met his partner in her home country, claimed that he had a generally positive attitude, because he was excited to be in Switzerland and to learn Swiss German. Nonetheless, the Anglophone partners were willing to relocate to Switzerland, which could be taken to indicate an absence of very negative views. With their exposure to the local language and culture, their attitudes became more differentiated, and they started to discern both positive and negative characteristics. Like the Swiss participants, they expressed overall favourable views on the local culture, but also voiced some criticism. With regard to the local language, in contrast, their attitudes have remained rather heterogeneous. Two of the English speakers declared that they initially did not like (Swiss) German, but have become accustomed to it over time. Five of them took a balanced or neutral attitude, stating that it was simply a practical necessity for them, and did not constitute a problem, or that they liked the language, but struggled with learning it or with the diglossic language situation. Three Anglophone interviewees expressed primarily positive feelings towards Swiss German, and reported that they liked the sound or novelty of the language, or enjoyed learning a new language.

It becomes apparent that, whereas the attitudes of the Swiss participants towards their partner's language and culture were initially more positive than the Anglophone partners', their attitudes towards each other's cultures largely converged over the course of their relationships, so that all participants had adopted a favourable yet somewhat critical attitude towards their partner's culture at the time of the interview. With regard to language, in contrast, differences have persisted. All of the Swiss interviewees still hold a highly positive view of English, yet many English speakers are struggling with their partner's language. With the exception of Joshua, even the Anglophone partners who speak it rather fluently did not express exceedingly positive feelings towards the language. This may be an important reason why — aside from occasional language mixing — nine of the couples have not shifted their relationship language to (Swiss) German even partially (see chapter 6, "'German sorta creeps into it': Language mixing").

Interestingly, even the partners who do not have a very positive attitude towards their L2, nonetheless have a high opinion of bilingualism, and all participants showed immense enthusiasm about their bilingual, bicultural relationships. Such positive opinions towards bilingualism seem to be common for elite bilinguals (Boyd 1998). The couples believe that their bilingualism has made their relationship more interesting, that their communication has improved as a result of their linguistic situation, and that they have benefitted personally and professionally from being in such a relationship, as it has given them access to international communities, the opportunity to travel, to move countries, or to raise bilingual children. This is reminiscent of the highly positive attitudes of the intercultural couples studied by Khatib-Chahidi, Hill and Paton (1998), as well as the bilingual couples studied by Piller (2002a), who used a variety of strategies to present their relationship as unproblematic, and to frame differences positively. The couples I interviewed also named some disadvantages, most of which were language-related, such as the potential for misunderstandings, or difficulties expressing oneself in one's L1 or L2 as a result of using the respective other language more frequently. Overall, though, almost three times as many positive as negative aspects were mentioned, and potential disadvantages were framed as minor issues. It is also worthy of note that, despite their positive views on bilingualism and biculturalism, almost all of the couples are still communicating in English for the most part — rather than bilingually — unless they are simultaneously addressing their children.

The couples also conveyed highly positive views on raising bilingual children, particularly the participants who did not have children (yet). The couples think that their children will gain a personal advantage from their bilingualism, and view it as their parental obligation to raise them bilingually. Similarly positive attitudes were expressed by the bilingual couples studied by Licciardello and Damigella, who believe that being bilingual and bicultural enriches their children not only linguistically and culturally, but also cognitively and relation‐ ally (2013: 750). Consequently, the couples with children in this study are all raising their children bilingually and are highly committed to their children's bilingual upbringing, like the bilingual couples examined by Piller (2002a: 252). Given their endeavours and high expectations, it is not surprising that several participants expressed worries about not succeeding in raising fluent bilinguals, which, in their opinion, might deprive them of a connection to their extended family or their cultural heritage. As most children were still very young at the time of the interviews, no definitive statements can be made about the success of their efforts yet. Their children's young age also accounts for the fact that none of the couples in this study expressed disappointment over their children's language proficiency, unlike some bilingual parents in studies by Piller (2002a) and Heller and Lévy (1992), whose expectations for their children's bilingualism were not met.

The four couples with children had originally decided to use their respective mother tongues with them, thus implementing an OPOL strategy. The fact that the parents demonstrated a clear preference for the OPOL strategy was to be ex‐ pected, as previous research indicates that this strategy is particularly common among elite bilinguals, and that many parents deliberately keep their languages separate in order to manage their children's bilingualism (Gardner-Chloros 2009: 144). Thus, several studies describe how parents employ a variety of strategies to encourage their child to address them in their language, and to refrain from language mixing (Juan-Garau and Pérez-Vidal 2001: 60; Chevalier 2013: 29). However, following through with their intentions of keeping their languages apart has proved to be difficult for the couples in this study, and only one of the four couples — whose child is still very young — has remained consistent in using the OPOL strategy. This is probably not uncommon, since, as Gartner-Chloros posits, even "parents who claim to be strict followers of OPOL often turn out, when their speech is systematically observed, to be subconsciously code-switching in spite of their best intentions" (2009: 144).

While the bilingual parents all share the opinion that OPOL is the best strategy to follow, they cited reasons for, as well as against, implementing this strategy. On the one hand, the bilinguals believe that speaking one's mother tongue with one's children is the obvious, natural, and preferable choice, and if they did not do this consistently, they offered explanations for this. In addition, several bilinguals claimed that they did not speak their L2 perfectly enough to raise a child in that language, despite their very high level of proficiency. The underlying notion of first language primacy, which seems to be particularly common in association with European languages (Pavlenko 2004: 192), was thus also noticeable among the couples. On the other hand, some bilinguals expressed worries about losing their couple language if they started speaking their mother tongue to their children. Others find it challenging to speak (only) their native languages to their children, as they view their L2 as their more accessible or more emotional language. Indeed, the language choice with one's children seems to correlate largely with the perceived emotionality of the two languages. While for the majority of bilinguals, their mother tongue is the most emotional language, for some of them, this is their L2, which is especially likely if their L2 is also their couple language. This concurs with Pavlenko's (2004) observation that, although bilingual parents are more likely to use their L1 with their children, when other languages are felt to be emotional, "many bi- and multilingual parents use more than one language to create an emotional connection to their children" (Pavlenko 2004: 196). Hence, the language choice with one's children is influenced by diverse factors such as language emotionality, ideologies, language proficiency, and aspects of practicality.

The analysis has confirmed that, while there are participants with a favour‐ able attitude who are not fluent in their L2 (yet), highly positive attitudes towards the participants' L2 correlated with a high level of proficiency, while more negative attitudes tended to correlate with language skills that are geared towards a practical use of the language. Even though highly positive attitudes were more common among the Swiss participants, they were also expressed by one Anglophone participant, Joshua, who has achieved a similarly high level of fluency in his L2 as the Swiss bilinguals. A comparable concurrence of positive attitudes and high linguistic competence in bilingual couples was identified by Piller (2002b). However, the overall positive attitudes of the interviewees towards various aspects of their bilingualism and biculturalism, which led to their desire to transmit both languages and cultures to their children, are not necessarily representative for bilingual couples in different settings, particularly for couples residing in English-speaking countries. Thus, the Swiss-Australian couples residing in Australia examined by Schüpbach (2009) displayed far more negative attitudes towards Swiss German and bilingualism. In fact, only two of the eleven Swiss bilinguals in her study were transmitting their mother tongue to their children, and none of the couples spoke the non-community language with each other. Even though the couples in this study consist of speakers of the same languages, there are considerable differences in their attitudes and their language choices. To some extent, this variance can be attributed to the attitudes of their surroundings towards non-community languages and bilingualism in general, as well as the status of Swiss German in Australia as opposed to English in Switzerland. This highlights the strong influence of a bilingual couples' environment on their attitudes, and, ultimately, their language use.

It should also be considered that the extremely positive views that the couples presented in the interviews may not be entirely accurate, as it is possible that the couples stress positive aspects of living in a bilingual, bicultural relationship so as to compensate for negative attitudes that they believe to prevail towards such relationships (Piller 2002a). Since the couples feel that the advantages of being in such a relationship outweigh the disadvantages, cultural or linguistic problems might often be downplayed, or portrayed as minor issues. Moreover, the sample may be biased with regard to the bilinguals' attitudes, since couples who have a problematic relationship (whether this is due to their personalities, cultural differences or language problems) are less likely to participate in sociolinguistic studies. Likewise, parents who are committed to their children's bilingual education, as Piller points out, "are simply more likely to be accessible to linguistic research and to participate in linguistic research than parents who do not share such a commitment" (2002a: 251).

This chapter has given an overview of the interplay and development of various attitudes surrounding the bilingual, bicultural relationships of speakers of Swiss German and English. The analysis has brought to mind, once again, that attitudes are often highly complex. On the one hand, they are shaped by a multitude of factors, such as an individual's environment and past experiences. On the other hand, there are a number of different attitudes that can be relevant in the case of bilingual couples: each partner's attitude towards both languages, towards both cultures, as well as towards bilingualism and biculturalism in general. The couples' reports have shown that their attitudes have an influence on their language choices among themselves and with their children. Thus, the fact that the couples hold overall more favourable views on English than Swiss German seems to be part of the reason why none of them have shifted their couple language towards the community language, which underlines the important role of attitudes in the communication of bilingual couples.

## **8 "In Swiss German,** *I lieb di***, that's strange": Expressing emotions**

## **8.1 Introduction**

The previous chapter has demonstrated that the attitudes and the emotionality that are attached to a language can be key to a bilingual's linguistic behaviour, for example when it comes to choosing a couple or family language. In this chapter, I intend to explore the expression of emotions, both positive and negative, in more detail, as well as potential difficulties that arise from having to express emotions in a second language. In a sense, bilingual couples provide the perfect context for studying the topic of expressing emotions in a bilingual setting. For one, the expression of emotions is an integral part of a couple's communication. Even if emotions do not necessarily have to be vocalized to be experienced, "the natural condition of emotion is that they are interpersonally expressed" (Guerrero, Andersen and Trost 1998: 9). Not only are emotions known to flourish in social situations, they are also more likely to be stirred by people who are close to each other (Planalp 1999: 18), such as partners in a romantic relationship. Notwithstanding these findings, it appears that emotion research in linguistics has often focused on the experiences and communication of emotions by individuals, rather than the negotiation of such experiences in dyads (Andersen and Guerrero 1998: 83); only in recent years, the linguistic expression and negotiation of emotion in interaction has received some attention (Sorjonen and Peräkylä 2012). For these reasons, the study of the manner in which bilingual couples communicate emotions may provide valuable insights.

At the same time, the expression of emotions constitutes an interesting research area within the field of bilingualism, since emotions are a highly complex matter, and a number of difficulties may surface when they are expressed in a non-native language. For one, there is the issue of language competence, as both positive and negative emotions can be difficult to articulate in an L2, and not finding the *mots justes* can lead to conflicts among the partners. Hence, "LX learners and users may not have the linguistic and pragmatic means to express the full range of their emotions in a way that would satisfy their communicative needs and be considered appropriate by their interlocutors" (Dewaele 2010: 6). Secondly, there may be inherent differences in the emotion vocabularies of the two languages, or in their use of emotion concepts (Pavlenko 2008: 147). Thirdly, the aspect of language emotionality may be relevant for bilinguals' language preferences, as different languages may carry different levels of emotional attachment for them (Pavlenko 2008: 147). In addition to this, the element of power can play a role when it comes to expressing negative emotions or resolving conflicts among bilingual couples, as the native speaker may be in a stronger position than the non-native speaker. Finally, the suprasegmental features accompanying the expression of emotions, such as voice quality, loudness or intonation contour, may differ between cultures and/or languages (see section 8.2, "Previous work on expressing emotions", for a discussion). Thus, there are many different factors that are involved in shaping the expression of emotions among bilingual couples.

Since emotions are often elusive and highly complex, comparing the manner in which bilinguals express them can be challenging. Indeed, the instances in which the bilinguals in this study became emotional during the interviews did not lend themselves to a direct comparison, as they differed in length and intensity, and were triggered by different topics. I decided to focus my analysis on the lexical level instead, and searched for expressions in the interviewees' speech which I deemed to be emotionally charged. Such *emotion words*, as Pavlenko terms them, tend to describe emotional states or processes, and potentially vary in their frequency of use in different languages (2008: 147). On the one hand, I want to determine which words the bilinguals used to express positive and negative emotions, and the frequency with which they used the individual terms, as well as to compare the usage of emotion terms by native versus non-native speakers, and male versus female partners. On the other hand, I am interested in suprasegmental features accompanying these emotion words, such as pitch, loudness, or intonation contour.

In the following, I first provide an overview of previous work on the expression of emotions in a bilingual setting (8.2), before discussing the couples' use of emotion words during the interviews (8.3). In addition to the frequency of use, I examine the bilinguals' use of voice quality and terminal pitch in the context of these emotion words, as well as differences in the use of emotion words based on the speakers' mother tongue and gender. Subsequently, the couples' reports on various aspects relating to the expression of emotions in a bilingual relationship are addressed (8.4). This section covers the expression of both negative and positive emotions, as well as the partners' use of terms of endearment. In the final section (8.5), I summarize my results and discuss them in light of previous research.

#### **8.2 Previous work on expressing emotions**

#### **8.2.1 On the study of emotions**

Emotions may be identified in various ways: speakers may either report their own experiences directly, or others can observe their nonverbal communication or their physiological responses (Berscheid 1983, in Feeney, Noller and Roberts 1998: 476). In addition, suprasegmental features can give us an indication of a speaker's level of emotionality (see section 8.2.5, "The suprasegmental expression of emotions across languages and cultures"). However, it is worthy of note that the emotions that are communicated may not necessarily match what is actually felt. Thus, emotions that are expressed verbally may not be based on underlying feelings, and not all existing emotions are conveyed, resulting in simulation or inhibition (Andersen and Guerrero 1998: 55). Whether emotions are expressed or suppressed tends to depend on whether this is situationally appropriate. For instance, people often simulate emotions so as to conform to standards of politeness, or avoid expressing them if they do not match what is expected in a certain situation (Andersen and Guerrero 1998: 55). In the present study, the only aspect that can be analysed is the manner in which the partners express emotions verbally and paraverbally, but not the accuracy or depth of these emotions.

Besides situational variables, the extent to which people convey their emo‐ tions can also depend on their culture and/or mother tongue. Hence, it has been argued that cultures vary with respect to their "display rules" for the expression of emotions, which tend to dictate that members of a culture either simulate, intensify, neutralize, deintensify or mask their emotions (Ekman and Friesen 1975, in Porter and Samovar 1998: 456). In addition to the level of intensity with which emotions are expressed, there appear to be certain emotions that are expressed in a different manner by members of different cultures or by speakers of different mother tongues. While there are some primary emotions which tend to be expressed similarly across cultures (anger, fear, surprise, sadness, disgust, and happiness), secondary emotions like pride, shame or guilt, by contrast, "arise culturally through participation in the sociocultural environment and tend to vary based on age, gender, and culture" (Porter and Samovar 1998: 452, based on Ekman and Friesen 1986). Such language-immanent differences in the expression of emotions may also influence bilinguals' language choices, which will be discussed in the following section.

#### **8.2.2 Expressing emotions in a second language**

There are a number of variables that may complicate the expression of emotions in a second language and/or among bilingual couples. Even if, as some believe, emotions themselves may be universal and independent of language, the manner in which emotions are expressed is to some extent determined by the language in question. This is partly due to characteristics inherent in each language, as each language appears to have a set of language-specific emotion words, and the exact meaning or connotations of emotion words may not match in both language (Pavlenko 2002, 2008; Wierzbicka 2004). In order to investigate how languages differ in the manner in which emotions are expressed, Wierzbicka examined the emotion lexicons of different languages, and explored differences in the emotional vocabulary of these languages through semantic analysis (Wierzbicka 1992, 1999), or by comparing and analysing key words in several languages (Wierzbicka 1997). She concludes from her research that emotions are categorized and valued differently in the languages she studied, and that these differences in the language-specific emotion lexicons may, in turn, shape the emotions of the individual. In Wierzbicka's words, "[w]e learn to make sense of our raw feelings through the categories imposed on them by our language, and this categorisation enters into the fabric of our feelings, and gives them shape and direction" (2004: 103).

The emotion vocabulary of a language is often influenced by the underlying culture. Thus, Wierzbicka inferred from her semantic analysis that, since "each language imposes its own classification upon human emotional experience, English words such as *anger* and *sadness* are cultural artefacts of the English language, not culture-free analytical tools" (1995: 249, italics in original). Consequently, the extent of the difference between the emotion vocabularies of two languages tends to be connected to the level of similarity of the cultures associated with them. In Wierzbicka's opinion, "each language has its own set of ready-made emotion words, designating those emotions that members of a given culture recognise as particularly salient. Presumably, these language-specific sets overlap and, presumably, the closer two cultures are, the greater the overlap between their respective sets of emotions" (1992: 124). Yet even among closely related languages, cultural attitudes can lead to differences in the usage of emotion words (Wierzbicka and Harkins 2001: 3). Such differences in the emotion vocabulary between English and German were shown in a corpus-based analysis of terms referring to anger in the two languages, which demonstrated that, despite the genetic relationship between the languages, there is no exact match for these terms (Durst 2001: 118). Since the participants in this study are all citizens of Western countries and their mother tongues are closely related, it may be expected that their emotion vocabularies are relatively similar, but not identical.

In addition to differences in the terminology referring to emotions, languages may also differ with regard to the size of their emotion lexicon. Pavlenko refers to a number of studies which have shown that German contains about 230–250 emotion words, while English is said to have more than 2,000 of them (2005: 88). However, it is doubtful that the differences between these two closely related languages are that great, and, as Pavlenko points out, the results may partly be due to a different selection or definition of emotion words in these studies. Their results are difficult to assess, as there are no cross-cultural studies actually comparing the two languages in detail. Besides the size of the emotion lexicon, moreover, it seems that certain languages have particular characteristics that make them attractive for being used as an emotional language. For instance, many people find it easy to say *I love you* to others in English (which is reserved for couples in many other languages).

Such variance in the emotion vocabulary of different languages may have a number of consequences. Firstly, it means that emotions can be very difficult if not impossible — to translate, which, in turn, can engender specific linguistic behaviour in a bilingual couple's communication, such as code-switching. Thus, bilingual speakers may shift from one language to another because they feel that they cannot express certain emotions in one of their languages (Wierzbicka 2004: 102; see section 6.2.1, "Factors that influence language mixing"). Moreover, different manners of expressing emotions may also lead to instances of miscom‐ munication in cross-cultural conversations (Pavlenko 2002: 50).

In view of these differences in the emotion lexicons of languages, it is not surprising that bilinguals often report that they express emotions differently in their two languages. This can be seen in research conducted by Besemeres, who examined the writing of three bilinguals with regard to cultural differences in their emotional expression. The three writers reported differences relating to the amount and the type of emotions (positive or negative) that are expressed in their two languages (2004: 140). At the same time, it appears that fluent bilinguals are often able to switch between their two languages and produce adequate emotion discourses in either of them. This was demonstrated in Pavlenko's (2002) comparison of the English and Russian emotion discourse of 31 late bilinguals. The bilinguals in her study used and interpreted emotion states adequately in both languages, but they preferred more emotionally charged words when speaking Russian than in English. Pavlenko concludes from her research that "in the process of second language socialization some adults may transform their verbal repertoires and conceptualizations of emotions, or at least internalize new emotion concepts and scripts" (2002: 71). Besides the level of second language socialisation, there are a variety of factors that influence a bilingual's use of emotion vocabulary in his or her L2, such as his or her extroversion, gender and language proficiency, as well as the topic of the conversation (Dewaele and Pavlenko 2002: 298, see next section).

In addition to the manner in which emotions are expressed, also the manner in which they are experienced tends to be determined by the language in which an event occurred. Hence, bilingual speakers may exhibit differences in their emotional reaction depending on the language and cultural context. This was shown by Panayiotou in a study of the emotional responses of 10 Greek-English bilinguals to two similar stories with an American and a Cypriot protagonist. According to the author, the participants' emotional reactions to the two stories differed and were culturally appropriate (2004: 132). In addition, emotions may be assessed and remembered differently depending on the culture concerned, as was demonstrated in a cross-cultural study by Scollon et al. (2004). The authors assessed reports of pleasant and unpleasant emotions in 416 participants from a variety of cultures and found the specific feelings which were reported to be consistent with the value of these feelings in the participants' respective cultures (2004: 320). Moreover, many bilinguals feel different depending on which of their languages they use, which can be seen in Pavlenko's (2006) research based on an extensive online questionnaire on bilingualism and emotions (BEQ) among 1039 multilinguals with a variety of mother tongues. Almost two thirds of the respondents claimed that they feel different when they use a different language, whereas a quarter reported feeling the same, and ten percent of the multilinguals gave ambiguous responses.

Besides the language-specific differences in emotion vocabularies, also a bilingual's personal history with both languages, as well as the context of their acquisition, will determine its emotionality for him or her. Thus, according to Pavlenko, there is evidence "that languages of bi- and multilingual individuals may differ in terms of their perceived embodiment, and more specifically in physiological reactions, levels of arousal and anxiety, autobiographic memo‐ ries, sensory associations, and perceptions of emotionality elicited by their emotion-related words" (2005: 184). Factors that are believed to determine the emotionality of a language for a bilingual are the age and circumstances of acquisition, personal history, language proficiency and dominance (Pavlenko 2004: 185). The most important factor among these seems to be the age of language acquisition, as languages that are learnt later in life are likely to carry less emotional weight than our first language (Dewaele 2004a: 87; Pavlenko 2004: 182). Therefore, as Pavlenko points out, it is often assumed that the L1 is the language of emotions, while the L2 is a language of distance (2004: 200).

A critical element with regard to the emotionality of a language seems to be whether it was acquired before and after puberty. Thus, various studies indicate that "when a second language is learned after puberty the two languages may differ in their emotional impact, with the first being the language of personal involvement and the second the language of distance and detachment" (Pavlenko 2002: 47). This may partly be due to the fact that "the first language is acquired in a natural environment, through perceptual and affective channels, and [therefore] becomes integrated into the limbic system" (Pavlenko 2004: 182; see also Paradis 1994). When we are socialized in our first language, the words in this language are linked to connotations and emotionally charged memories, both positive and negative (Pavlenko 2004: 182). When languages are acquired in adulthood, by contrast, the speaker is not exposed to the same level of socialisation, even if these languages are acquired in a natural environment. For this reason, being fluent in and accustomed to speaking an L2 does not automatically entail that there is an emotional attachment to the language.

Language emotionality also influences the choice of language(s) in which a bilingual expresses emotions. Pavlenko (2008) gathers from self-reports of multilinguals to the BEQ that many speakers prefer their L1 to express emotions, even if their L2 happens to be their main language of communication. On the other hand, some bilinguals may have a preference for their L2 precisely because it does not make them experience "the undue emotionality elicited by the L1" (2008: 160). Similar results were obtained by Dewaele (2010), who investigated factors which influence bilinguals' choice of language for expressing emotions and their perceived language emotionality, also based on the BEQ. His research indicates that extensive LX socialisation, frequent LX use, a wide network of interlocutors, language acquisition in a naturalistic environment, and an early age of acquisition all lead to increased LX use to express emotions (Dewaele 2010: 103). All of these factors tend to increase the level of emotionality of the language for the bilingual as well. Dewaele's study also shows that, even when a non-native speaker is capable of expressing emotions in his or her LX, it often takes a considerable amount of time until emotions expressed in the LX are felt to be equally strong as those in his or her mother tongue (2010: 218). This may explain why overall, only a minority of the respondents reported a preference for communicating emotions in their LX (2010: 215). Nonetheless, there were also bilinguals for whom their LX had become their most emotional language; such a shift in language preference was commonly "linked to new partners or simply to the fact of having moved to a different country and, subsequently, having acculturated to the new language and culture" (2010: 217). This indicates that a second language can carry a considerable emotional weight, even if it is acquired later in life, provided that it is spoken fluently and used frequently in an intimate environment.

#### **8.2.3 Bilingual couples and emotions**

The expression of emotions among bilingual couples has been the subject of a few studies. Pavlenko (2004) examined the language choice and emotionality of bi- and multilingual families, as well as the parents' perception of the emotionality of their languages; her research is based on the BEQ, but only considers respondents with children. She found that, in their case, language emotionality did not necessarily coincide with language competence or order of acquisition. Thus, while language dominance was a much stronger factor than language emotionality in the parents' overall language choice, in situations that involve praising and disciplining, their language choice tended to correlate with their perceived language emotionality (2004: 188). Moreover, the parents' responses suggested that an L2 can become emotionally charged if it is used in an intimate environment, as it "is not necessarily perceived as the language of detachment by respondents who are married to LX speakers and/or are bringing up children in that language" (2004: 193). Similar results were obtained by Dewaele and Salomidou, who investigated the emotional communication among cross-cultural couples based on a large-scale online questionnaire and written interviews. They found that longer relationships resulted in affective socialisation in the bilinguals' non-native language, and that female participants initially found communicating emotions in their LX more challenging than male participants (2017: 128).

If an L2 carries positive connotations, expressions of emotions as well as terms of endearment in this language can become very meaningful to a non-native speaker. Such cases are reported by Piller and Takahashi (2006), who studied the L2 desire of Japanese women living in Australia. A number of these women expressed the wish to be addressed with English terms of endearment by Western men:

English terms of endearment are seen as a particular expression of Western hetero‐ sexual relationships. Many participants confessed their desire to be called 'darling', 'honey', or 'sweetheart' […  ]. Conversely, attempts by English-speaking boyfriends to use Japanese terms of endearment or expressions of love are considered a real turn-off. (Piller and Takahashi 2006: 71–72)

A similar preference for terms of endearment in their L2 was reported by a number of partners in Piller's (2002a) study on bilingual couples. According to Pavlenko, especially native speakers of English often display a preference for pet names in their L2, as they feel that other languages have a larger array of such terms, and these seem more novel and less "worn out" to them than terms of endearment in their mother tongue (2004: 198). For some multilinguals, moreover, their L2 can provide "new emotion terms and affective repertoires that may best capture feelings 'unnamed' in the native language" (Pavlenko 2005: 139–140).

Of course, bilingual couples may also want to express negative emotions such as criticism, or they may argue with each other. Piller observes that, while "[m]ost couples share a preference for a particular conflict language", it is difficult to accommodate both partners if this is not the case (2002a: 155). Moreover, the fact that one partner is forced to use his or her L2 in these situations potentially presents a problem. As Piller notes, partners in bilingual relationships may lack "rowing proficiency" in their L2 (2002a: 156), which could mean that they are in a weaker position than their partner during arguments and disputes, or they may not find the right words and thus hurt their partner more (or less) than intended. It is also possible that certain rowing conventions have to be learnt upon entering into a bilingual relationship. Due to these potential difficulties, it is not surprising that for most partners in Piller's study, their L1 tended to be "preferred in an event involving strong emotions, such as a marital conflict" (2002a: 154). This was also the case for the respondents to the BEQ examined by Pavlenko, who "often talk[ed] about reverting spontaneously to the L1 in fights and arguments with their partner and spouses" (2005: 133), even if their partner did not understand this language. This demonstrates that arguments may present an exception to the usual language choice of many couples (2005: 134). At the same time, there are also bilinguals who prefer using their L2 to express negative emotions. As Pavlenko explains, "[u]sing the LX in an argument offers a double advantage: Not only do speakers place themselves at a greater emotional distance, they also wield a language that has more power to hurt the interlocutor, for whom the language is native" (2005: 135).

In addition to the partners' mother tongues, there are several other variables that potentially have an influence on the manner in which bilingual couples communicate emotions. One aspect that has proved to be relevant with regard to the frequency with which emotions are expressed, for instance, is the state or duration of the relationship. Thus, Aune, Buller and Aune (1996) found that romantic partners felt that at an earlier stage of the relationship, expressing both positive and negative emotions was less appropriate; accordingly, the partners paid particular attention to managing their expression of negative emotions. Another important factor is the level of emotional security within the relationship. In a study by Feeney (1995), couples with an insecure emotional attachment reported more control of negative emotions than secure couples. Finally, an interesting aspect to consider when one studies the expression of emotions in dyads is that partners may mirror each other's emotional state in their communication. Such "emotional matching" has been observed repeatedly among couples (Anderson and Guerrero 1998: 84). The expression of emotions among bilingual couples is thus shaped by a variety of factors, such as their proficiency in their L2, their past experiences and subjective language emotionality, as well as their relationship status.

#### **8.2.4 Gender and emotions**

Furthermore, research on the communication of emotions among couples tends to suggest that the element of gender plays a crucial role when it comes to expressing emotions. Notarius and Johnson state that in North American culture, there is a gender stereotype that "females are more overtly expressive than males" (1982: 483). To corroborate this, they videotaped six married couples during a discussion of relationship issues, and discovered that the wives' speech contained more negative expressions of emotions than the husbands' speech. Moreover, the wives "reciprocated their husbands' positive and negative speech while husbands did not reciprocate their wives' speech" (1982: 483). A gender difference in expressiveness is also visible in Piasecka's study, even though it should be noted that in this case, the emotional situations were not actually experienced by the participants. The author instructed graduate students to write down what they would feel and say in a number of hypothetical emotional situations and analysed the answers of ten (randomly selected) female and ten male participants for ten situations. The female participants produced consistently more words to express their emotions than the male participants, and there were gender differences both in the identification and naming of emotions (2013: 161).

Gender differences can also be observed in dyadic situations. In interviews with 13 German- and English-speaking couples, Piller (1999) discovered differ‐ ences in the genders' expression of criticism. Thus, the female participants in her study tended to produce negative national stereotypes about their own country of origin, while the male participants were more likely to express negative stereotypical views about their partner's country. Furthermore, men and women reacted differently to their partners' criticism; while women reacted with positive minimal responses or laughter, men often matched negative comments with negatives statements about their partner's country of origin (1999: 125).

The element of gender also emerges as relevant in the bilingual expression of emotions, as similar results were obtained by Dewaele and Pavlenko, who examined the impact of gender (among other variables) on the use of emotion vocabulary of Dutch-French and Russian-English bilinguals. One of their find‐ ings was that the female L2 speakers used a wider range and a larger number of emotion words than the male participants (2002: 285). Moreover, gender may influence a bilingual's language choice for the expression of emotions, as it seems that female bilinguals are more likely to express their feelings in their L2 than male bilinguals (Dewaele 2010: 104). Gender may also influence the pitch and voice quality with which emotions are expressed in different cultures. Thus, Pavlenko concludes from autobiographic reports of bilinguals that "L2 users may consciously or unconsciously attempt to adjust their pitch and voice quality in order to come across as more feminine or masculine in their L2 community" (2001b: 166). Gender may thus influence a speaker's pitch and voice quality when expressing emotions, the level of expressiveness, the extent of the emotion vocabulary, as well as the language choice of bilinguals.

#### **8.2.5 The suprasegmental expression of emotions across languages and cultures**

Finally, languages and/or cultures may also differ with regard to the vocal cues (or suprasegmental features) that accompany the verbal expression of emotions. Vocal cues "are commonly discussed in terms of two types of *suprasegmental* features, that is, aspects of speech that involve more than single segments: (a) *paralinguistic* features, such as voice quality and vocal gestures, and (b) *prosodic* features, such as pitch, loudness, or rate of articulation" (Pavlenko 2005: 49, emphases in original). These cues assist us in conveying emotions to others, but also in interpreting the emotions of others (Pavlenko 2005: 45). However, relatively little is known about the manner in which individual cues work together, since this depends on a multitude of factors and is therefore difficult to determine (Planalp 1999: 50). Furthermore, vocal cues are far from clear indicators of emotions or the speaker's level of emotionality. On the one hand, cues may be inconsistent — for example if they are reflecting mixed feelings and speakers can differ in their level of expressiveness (Planalp 1999: 52). On the other hand, a vocal cue may be indicative of a variety of different emotions, and a particular emotion can trigger a myriad of different vocal cues:

For instance, higher pitch, increased loudness, and fast rate of speech are commonly associated with such diverse emotions as anger, happiness, joy and fear. On the other hand, the same emotions may be conveyed by a variety of different cues, depending on the […  ] speaker, the situation they are in, and their communicative intentions. (Pavlenko 2005: 45–48)

Planalp examined various cues that are used to communicate emotions, in‐ cluding verbal, facial and body cues, as well as vocal cues such as increased loudness or pitch. After monitoring a person that they knew well, the vast majority of the respondents in her study reported detecting several cues (1 to 13 cues), with the largest number of respondents reporting four cues. Among these, vocal cues were reported most frequently, in particularly loudness, speed and amount of talking (1998: 38). These findings confirm that there is often a combination of cues involved in the expression of emotions (1998: 37).

While a number of researchers have pointed to the differences in vocal cues between different languages and cultures, little research has been conducted on how emotions are expressed vocally across cultures. One thing that is known is that voice setting — and thus voice quality such as breathiness — differs across languages, and bicultural bilinguals have been found to shift voice quality when switching languages (Esling 1994, in Pavlenko 2005: 50). Consequently, native speakers tend to be at an advantage in identifying emotions of speakers of the same language, even though L2 speakers can become very accurate at reading emotions over time. Pavlenko states that

in the absence of visual and contextual cues, native speakers of a particular language are typically more accurate than non-native speakers in interpreting emotional content of the speech of other native speakers. At the same time, the native-speaker advantage is not an across-the-board phenomenon. Research shows that L2 users improve their performance over time, and in some cases may be as good or even more accurate than native speakers in emotion identification, at least in experimental conditions […  ]. (Pavlenko 2005: 74)

However, the fact that a bilingual is able to interpret emotional cues in his or her L2 correctly does not automatically mean that he or she can produce appropriate cues when speaking his or her L2. For instance, it is possible that bilinguals do not imitate prosodic patterns of their L2 because they perceive them as unnatural or exaggerated (Pavlenko 2005: 75). This might be the case for the Swiss German speakers in this study, as, in my personal opinion, emotions tend to be communicated with stronger verbal cues in English than in Swiss German (e.g. with a wider intonation contour, stronger stresses), though to my knowledge, no comparative research has been carried out on the subject.

#### **8.3 The couples' expression of emotions during the interviews**

#### **8.3.1 Methodology and research questions**

Previous research indicates that there may be differences between speakers with different mother tongues and genders, both in the manner in which emotions are expressed and the frequency with which they are expressed. The vast majority of research on the influence of these factors, but also on the expression of emotions among bilinguals, has been of a qualitative nature, and there have not been many quantitative studies on the subject to date. Moreover, many studies examine the emotional expression in hypothetical situations (e.g. Piasecka 2013: 158), asking participants how they imagine that they would feel and express themselves in certain situations. By contrast, my aim is to examine the spontaneous expression of positive and negative emotions among the bilingual couples whom I interviewed. I am interested in determining (a) which expressions they used to express positive and negative emotions, and with what frequency these terms were used; (b) which suprasegmental features accompanied their expression of emotions; and (c) what effect the expression of emotions had on their intonation contours (terminal pitch). For all of these topics, I also want to ascertain if (d) the participants' gender or mother tongue affected their manner of communicating emotions.

In order to achieve this, I decided to search for expressions that indicate positive and negative emotions in the transcriptions, so as to be able to analyse the frequency of use for each speaker, as well as the suprasegmental features accompanying each expression of emotions. In my analysis, I draw on Pavlen‐ ko's concept of *emotion words*, which refers to words that describe or express emotional states (e.g. "being happy"), or processes (e.g. "worrying") (2008: 148). Pavlenko further distinguishes between *emotion-related words*, which refer to behaviours caused by particular emotions (e.g. "yelling"), and *emotion-laden words*, which may describe or elicit emotions without referring to them directly (e.g. "loser") (2008: 148). In addition to terms that explicitly refer to affective states or processes, emotion-related words and emotion-laden words were also included in the analysis, provided that they were deemed to be indicative of a clear affective stance.

It needs to be noted that, of course, the utterance of a positive or negative expression does not guarantee that there is a corresponding underlying emotion (see section 8.2.1, "On the study of emotions"). Speakers may be reporting past emotions without experiencing them to the same extent at the moment of speaking; they may report someone else's emotions, or make a positive or negative statement about something without feeling strongly — if anything at all — about it. Conversely, emotions may be present when no positive or negative expressions are uttered, for instance when one speaker disagrees with the other using non-emotion words, or remains silent. As there is still a higher probability of the speaker experiencing emotions in these contexts, though, I do not believe this to pose a problem for the analysis. If only situations in which the speaker appears to be very emotional were considered, the resulting effects on voice quality or intonation contour might be more noticeable. However, these situations would probably be selected based on precisely those vocal cues whose presence I want to examine in this study. Considering all situations in which positive or negative expressions appeared is therefore likely to lead to a more objective result.

Not all potentially positive or negative terms that were used during the interviews were coded, but rather the ones that were deemed to be clearly positive or negative in the context in which they appeared, and were likely to serve the expression of a positive or negative emotion. If, for instance, a word was used ironically (e.g. when Joshua [<sup>E</sup> ] refers to an early misunderstanding as "an interesting experience" [403]), this was not included. There were also several words that were clearly positive or negative in certain contexts, but not in others (e.g. well, like, open, fine; strange, mean, mind, crazy). In these cases, only the instances where there was a definite positive or negative meaning were counted.

Expressions with the same stem were grouped together, and all related forms of each positive or negative expression were coded in the same group. Hence, for the word *amazing*, for instance, I also searched for *amaze*, *amazed*, *amazement*, *amazingly*. These word groups are listed in Table 16 below (one term representing each word group). The word groups are listed in order of frequency, including the number of instances that each word group appeared during the interviews. Since emotions are often composite and difficult to abstract from other emotions, I decided not to distinguish between different types of positive or negative emotions (e.g. anger, sadness, frustration), although this may, of course, have had an impact on the prosodic features accompanying the participants' speech.

The entire intonation unit in which each expression of emotion appeared was coded in each instance, so as to be able to analyse different suprasegmental features that co-occur with these expressions, as well as the terminal pitch following it (to give an indication of the intonation contour of the intonation unit in which the emotion was expressed). The frequency of the paralinguistic and prosodic features in the intonation units in which the emotion words occurred was then compared to their incidence during the entire interviews. In doing so, I was able to determine whether there were any suprasegmental features that occurred especially frequently when either positive or negative emotions were expressed. Since intonation units containing emotion words were coded separately for each speaker, I could also examine whether speakers with different mother tongues and/or genders differed with regard to the frequency with which they expressed positive or negative emotions, the types of emotion words they used, and their use of paralinguistic and prosodic features.

In total, I coded 928 positive and 1,000 negative expressions (Table 16), which amounts to an average of almost 96.4 emotion expressions per inter‐ viewee. This number seems to be relatively large, which may be due to the fact that several of the interview topics revolved around positive or negative experiences, but also demonstrates that the couples were very forthcoming in expressing their emotions during the interviews. There were fewer groups with the same stem referring to positive emotion words (68 groups and 928 instances, thus 13.6 per group, TTR [type-token ratio] = 7.3%) than negative emotion words (106 groups and 1,000 instances, thus 9.4 per group, TTR = 10.6%). This could indicate that the speakers were more careful or precise in their choice of words when voicing criticism or negative emotions than when offering praise or communicating positive feelings. In the following, I will take a closer look at the bilinguals' use of emotion words, as well as the suprasegmental features accompanying them.


<sup>1</sup> These expressions were grouped together although they do not technically share a stem. If this group were subdivided, there would, of course, be more groups in the negative category; if this group were disregarded, the TTR for negative expressions would be 11.5%. I decided to include these expressions in the analysis in order to have a larger number of instances in which emotions were expressed on which I could base by findings with regard to suprasegmental features.


Table 16: Positive and negative emotion words used during the interviews

#### **8.3.2 Frequency and types of emotion expressions**

Emotion words often occurred in clusters, as can be seen in the following example:

```
Example 8.1: Katia (Swiss, 28)
Katia: <PP<HSK there's a ^lot of things that fascinated me HSK>PP> /
             … <P ^just because you were so P> _
             <P you were so ^^ni:ce P> //
             <P and <HI ^^friendly HI>P> _
             <P and <HI ^^open HI>P> _ (240)
```
To allow for a better comparison of the participants' use of positive and negative expressions, the number of expressions was put in relation to the number of words each bilingual spoke during the interview (frequency normalized per 10,000 words). In addition, the number of different types of expressions each participant used was specified (expressions with the same root were counted as one type), and the type-token ratio (TTR) was calculated to measure differences in the level of variety. An overview of each speaker's use of emotion words is given in Table 17.


Table 17: Use of emotion words for each participant (absolute numbers and normalized per 10,000 words)

Key: pos = positive; neg = negative; types = number of words with different stems; TTR = type-token ratio; bold = in text.

Table 17 indicates that, while the interviewees used similar numbers of positive and negative expressions overall (46.4 positive and 50.0 negative expressions on average), there was considerable variance among the individual speakers with regard to how many expressions each one used. On average, 90.0 per 10,000 words were positive expressions (with a standard deviation of σ = 36.9) and 97.0 were negative expressions (σ = 31.7). The greater standard deviation for positive expressions shows that there was a greater difference in the number of positive expressions each interviewee uttered than the number of negative expressions. Furthermore, it is evident that among all but two couples (Courtney/Martin and Claire/Simon), the female partner used a higher proportion of emotion words than the male partner (total normalized); and for only one couple, it was the speaker of English who used a higher percentage of expressions of emotion (Karen/Philipp).<sup>2</sup> On average, each participant used 18.3 positive word types (which corresponds to 35.5 per 10,000 words, σ = 19.8), and 22.1 negative word types (42.9 per 10,000 words, σ = 12.4). The type-token ratio for each speaker individually is 39.4 in the positive and 44.2 in the negative context on average.

#### **8.3.3 Emotions and suprasegmental features**

When one examines situations in which emotion words are used, it becomes apparent that these are usually accompanied by a number of suprasegmental features. The following extract illustrates the abundance of vocal cues that co-occur with emotional speech:


Monika is very emotional when she talks about the difficulties she faced at the beginning of her relationship with Dean, as she felt discriminated against by his brother for not being Jewish. Besides a number of negative emotion words ("discriminated", "given up", "nasty"), the extract contains a variety of suprasegmental features which are often found in the context of the expression of emotions. It demonstrates that emotional speakers may lend emphasis to their

<sup>2</sup> The elements of gender and mother tongue will be discussed in more detail in sections 8.3.5, "Emotions and gender", and 8.3.6, "Emotions and mother tongue".

speech by using contrastive stresses (^^) or lengthening (:), or they may show that they are uncomfortable by inhaling and exhaling (<H> <Hx>). In addition, their emotional state is reflected in the voice quality of their speech, as there appears to be a large amount of husky, creaky, or breathy voice quality (<HSK>, <CRK>, <BR>), as well as a soft voice (<P>), and a lowered, heightened or widened pitch level (<LO>, <HI>, <W>).

In order to determine whether these suprasegmental features do indeed occur more frequently when emotions are expressed than they do otherwise, it is necessary to compare the frequency of their use in the vicinity of emotions to the overall frequency with which they appear during the interviews. Table 18 below gives an overview of the use of each type of voice quality, loudness and intonation in the context of positive and negative emotions, as well as their occurrence during the entire interviews (on average per person).<sup>3</sup> I also indicated the percentage of intonation units in which each suprasegmental feature appeared ("% of IUs"), so as to be able to compare the numbers with each other.


<sup>3</sup> Note that emotion words are also included in the overall category, and that emotions may also have been expressed in intonation units where no emotion expressions were coded. If this were not the case, the differences could be expected to be even more marked (cf. section 8.3.1, "Methodology and research questions").


Table 18: Voice quality, loudness and intonation in the context of emotion words (average per person)

Key: IUs = intonation units; bold = tendencies described in text. See "List of transcription conventions" at the beginning of the book for an overview of all abbreviations of suprasegmental features.

**Figure 2.** Suprasegmental features in the context of emotion words (in % of intonation units) Table 18 confirms that a variety of different suprasegmental features occurred more frequently in those instances Figure 2: Suprasegmental features in the context of emotion words (in % of intonation units)

where emotions were expressed than they did overall, which is visualized in Figure 2. In terms of pitch (high <HI> or low <LO> voice) and pitch movement (wide intonation contour <W>, upwards <UP> or downwards <DOW>), it is remarkable that all types occurred more frequently in both positive and negative contexts than overall. Not surprisingly, a low voice and downward pitch (<LO>, <DOW>) was most frequent in negative contexts, while a high voice and an upward pitch (<HI>, <UP>) were particularly frequent in positive contexts, though they were also more frequent in negative contexts than overall. In addition, there was slightly more loudness variance (crescendo <CRE> and diminuendo <DIM>) when emotions were expressed. This demonstrates that loudness and intonation are important means of lending emphasis to one's speech when one expresses emotions. The same applies to rhythmic (<RH>) and marked Table 18 confirms that a variety of different suprasegmental features occurred more frequently when emotions were expressed than they did overall, which is visualized in Figure 2. In terms of pitch (high <HI> or low <LO> voice) and pitch movement (wide intonation contour <W>, upwards <UP> or downwards <DOW>), it is remarkable that all types occurred more frequently in both positive and negative contexts than overall. Not surprisingly, a low voice and downward pitch (<LO>,

The difference between the emotional contexts and the entirety of the interviews is particularly marked with regard to breathy, creaky and husky voice quality (<BR>, <CRK>, <HSK>). All of these were more frequent in the context of negative emotions, particularly creaky voice, than overall. When positive emotions were expressed, a breathy and creaky voice quality occurred more frequently than overall, but a husky voice was used similarly often as overall. While the occurrence of these types of voice quality was anticipated in negative contexts, they could be considered somewhat unexpected in positive contexts. The reverse was the case for smiling and laughing voice quality; even

speech (<MRC>), which were both more common when emotions were expressed than overall.

<DOW>) was most frequent in negative contexts, while a high voice and an upward pitch (<HI>, <UP>) were particularly frequent in positive contexts, though they were also more frequent in negative contexts than overall. In addition, there was slightly more loudness variance (crescendo <CRE> and diminuendo <DIM>) when emotions were expressed. This demonstrates that loudness and intonation are important means of lending emphasis to one's speech when one expresses emotions. The same applies to rhythmic (<RH>) and marked speech (<MRC>), which were both more common when emotions were expressed than overall.

The difference between the emotional contexts and the entirety of the interviews is particularly marked with regard to breathy, creaky and husky voice quality (<BR>, <CRK>, <HSK>). All of these were more frequent in the context of negative emotions, particularly creaky voice, than overall. When positive emotions were expressed, a breathy and creaky voice quality occurred more frequently than overall, but a husky voice was used similarly often as overall. While the occurrence of these types of voice quality was anticipated in negative contexts, they could be considered somewhat unexpected in positive contexts. The reverse was the case for smiling and laughing voice quality; even though smiling voice quality was particularly present in the context of positive emo‐ tions, it was also more frequent in negative contexts than in general. A reason for this might be that speakers mitigate the force of negative statements or criticism by adding a smiling voice quality. The same does not apply to the use of laughing voice quality, though: mostly, it accompanied positive emotions, yet it occurred slightly less than average when negative emotions were expressed.

The results with regard to the rate of speaking are as expected. A slow speech rate (<L>) was most common in negative contexts and less common than average in positive ones. Conversely, a fast speech rate (<A>) was used most frequently in positive contexts, and least frequently in negative ones. Meanwhile, halting speech occurred with similar frequency in all three contexts; my expectation that it would be most frequent in negative contexts — was not confirmed. This is also true for short pauses, which were similarly frequent in negative contexts as they were overall, though there was a higher number of short pauses in positive contexts (see Table 19 below).

Besides voice quality and intonation, speakers may add emphasis to their emotional speech by using contrastive (strong) stresses or lengthening. Table 19 below indicates the average number of nuclear (^) and contrastive (^^) stresses in all three situations, as well as lengthening (:).


Table 19: Stresses, lengthening and pausing in the context of emotion words (average per person)

Key: IUs = intonation units; ^ = nuclear stress; ^^ = contrastive (strong) stress; : = lengthening; .. = short pause; bold = in text.

As can be seen in Table 19, strong stresses were used very frequently when positive and negative emotions were expressed — more than twice as often as overall (12.8% and 12.1% of IUs, vs. 5.2% overall). In both emotional contexts, there were far more units containing several stresses than in the overall context (5.5% and 5.6% of IUs, vs. 0.8% overall). However, lengthening appears to be less common in the context of emotions than overall. This might be due to a relatively high percentage of lengthening occurring with hesitation markers or similar (e.g. *um:*), which were usually placed in a separate intonation unit in the transcription even if they occurred in the vicinity of emotion expressions. Another explanation might be that emphasis is often created with strong stresses and pitch variation in emotional contexts, reducing the need for lengthening as a stress marker.

Overall, the majority of the trends that are discernible with regard to the use of suprasegmental features in an emotional context are in line with expectations. Nevertheless, the large difference between the number of suprasegmental features used in emotional contexts and overall is remarkable. This difference is noticeable in a great variety of suprasegmental features and is particularly dis‐ tinct in the case of exceptionally high or low pitch and strong pitch movements.

#### **8.3.4 Emotions and terminal pitch**

The previous section has demonstrated that the interviewees were more likely to use a wide intonation contour (<W>) as well as a greater number of rising and falling pitch movements (<UP> and <DOW>) when speaking about emotions. To learn more about the intonation contour in the context of expressing emotions, I also examined the terminal pitch of the intonation units in which emotion expressions appeared. Table 20 and Figure 3 give an overview of the average use of each terminal pitch, in the context of positive emotions, negative emotions, and in the entire transcription. In order to be able to compare the figures better, I also calculated the percentage of intonation units which contained each type of terminal pitch.


Table 20: Terminal pitch in the context of emotion words (average per person) Key: IUs = intonation units; -- = truncated contour (interruption); ~ = glissando; bold = in text. See "List of transcription conventions" at the beginning of the book for an overview of all symbols used for terminal pitch.

**Table 20.** Terminal pitch in the context of emotion words (average per person)

all up or down without ~ (/ // )

scription conventions" at the beginning of the book for an overview of all symbols used for terminal pitch.

**Figure 3.** Terminal pitch in the context of emotion words (in % of intonation units) Figure 3: Terminal pitch in the context of emotion words (in % of intonation units)

Table 20 and Figure 3 indicate that the terminal pitch movements of intonation units in which emotions are expressed are similar for negative and positive emotions, yet differ from the terminal pitch movements in the entire interviews. For one, there were fewer level pitches in the context of emotion expressions: overall, more than a third (35.0%) of the intonation units had a level terminal pitch (\_), as opposed to 21.8% and 23.9% of the intonation units in the vicinity of positive and negative emotion expressions, respectively. This might be due to the fact that this is the least dynamic (and hence the least emphatic) terminal pitch. A considerable difference is noticeable in the number of strong pitch movements (// //~ \\ \\~), which occurred in 17.5% of all intonation units, but in 27.4% and 24.7% of the intonation units containing positive and negative emotion terms, respectively. The difference with regard to regular pitch movements (/ /~ \ \~) is not as marked, as they occurred in 46.1% of the intonation units overall, versus 50.2% and 50.7% of the intonation units in the positive and negative contexts. Both rising and falling terminal pitch movements were more common in the context of emotions than overall, though interestingly, the pitch movement in both emotional contexts did not concur with the use I had predicted. Since rising pitch movements tend to be associated with positivity, I expected there to be more upwards pitch movements in the positive context than in the negative one, but there was only a slight difference (34.3% of the intonation units in the positive context and 32.9% in the negative one, vs. 25.5% overall). Furthermore, I anticipated that falling pitch would be most common in negative contexts, but in fact there were even slightly fewer downward pitch movements in negative than in positive contexts (43.4% of the intonation units in the positive context and 42.4% in the negative one, vs. 38.0% overall). Moreover, when emotions were expressed, there were considerably fewer intonation units with a glissando terminal pitch movement (gliding up or down, indicated with a tilde) than without. In the context of positive and negat emotions, 11.4% and 12.6% of the intonation units featured glissando, versus 15.9% overall. It is possible that this is partly attributable to the fact that the emotional intonation units did not include many backchannel signals short answers, which often include glissando; these were usually transcribed in a separate intonation unit. Finally, it is also evident that there were fewer interruptions when emotions were being expressed than overall. While 1.5% of Table 20 and Figure 3 indicate that the terminal pitch movements of intonation units in which emotions are expressed are similar for negative and positive emotions, yet differ from the terminal pitch movements in the entire interviews. For one, there were fewer level pitches in the context of emotion expressions: overall, more than a third (35.0%) of the intonation units had a level terminal pitch (\_), as opposed to 21.8% and 23.9% of the intonation units in the vicinity of positive and negative emotion expressions, respectively. This might be due to the fact that this is the least dynamic (and hence the least emphatic) terminal pitch. A considerable difference is noticeable in the number of strong pitch movements (// //~ \\ \\~), which occurred in 17.5% of all intonation units, but in 27.4% and 24.7% of the intonation units containing positive and negative emotion terms, respectively. The difference with regard to regular pitch movements (/ /~ \ \~) is not as marked, as they occurred in 46.1% of the intonation units overall, versus 50.2% and 50.7% of the intonation units in the positive and negative contexts. Both rising and falling terminal pitch movements were more common in the context of emotions than overall, though interestingly, the pitch movement in both emotional contexts did not concur with the use I had predicted. Since rising pitch movements tend to be associated with positivity, I expected there to be more upwards pitch movements in the positive context than in the negative one, but there was only a slight difference (34.3% of the intonation units in the positive context and 32.9% in the negative one, vs. 25.5% overall). Furthermore, I anticipated that falling pitch would be most common in negative contexts, but in fact there were even slightly fewer downward pitch movements in negative

important, and therefore interrupt each other less than they do otherwise.

130

all intonation units overall ended in a truncated intonation contour, only 0.5% of the intonation units in positive and 0.7% in negative emotional contexts were truncated. It might be that the couples perceive emotional topics to be than in positive contexts (43.4% of the intonation units in the positive context and 42.4% in the negative one, vs. 38.0% overall).

Moreover, when emotions were expressed, there were considerably fewer intonation units with a glissando terminal pitch movement (gliding up or down, indicated with a tilde) than without. In the context of positive and negative emotions, 11.4% and 12.6% of the intonation units featured glissando, versus 15.9% overall. It is possible that this is partly attributable to the fact that the emotional intonation units did not include many backchannel signals or short answers, which often include glissando; these were usually transcribed in a separate intonation unit. Finally, it is also evident that there were fewer interruptions when emotions were being expressed than overall. While 1.5% of all intonation units overall ended in a truncated intonation contour, only 0.5% of the intonation units in positive and 0.7% in negative emotional contexts were truncated. It might be that the couples perceive emotional topics to be important, and therefore interrupt each other less than they do otherwise.

In sum, there is a distinct difference between the terminal pitch in emotional contexts and overall. In both emotional contexts, there were fewer level terminal pitches, more regular as well as marked rising and falling pitch movements, and fewer glissando pitch movements than overall. All of this is in accord with expectations and demonstrates that the interviewees were more emphatic in those intonation units in which they used emotion expres‐ sions than they were otherwise. Contrary to expectation, however, there is relatively little difference between the terminal pitch in positive and negative emotional contexts.

#### **8.3.5 Emotions and gender**

Since it is often assumed that women are more expressive of their emotions than men, it is conceivable that there are differences in the male and the female participants' usage of emotion terms during the interviews, as well as the suprasegmental features accompanying their emotional speech; I therefore decided to explore the role of gender in the expression of emotions. The first aspect I examined was the frequency with which men and women used positive and negative expressions of emotions. Table 21 below shows the average number of expressions of emotions for speakers of each combination of gender and mother tongue, as well as the average for all members of each gender. In addition, it shows the number of different types of expressions they used on average.


Table 21: Use of emotion words for each gender and each combination of language and gender (absolute numbers and normalized per 10,000 words) Key: SG = Swiss German; E = English; bold = in text.

If only the absolute numbers are considered, it seems that the hypothesis that women use more expressions of emotions than men is not confirmed, since there is no difference in their use of positive expressions (on average 46.4 expressions), and women only used a slightly greater number of negative expressions (50.8 vs. 49.2 expressions). However, when these numbers are put in relation to the total number of words members of each group uttered on average, it is evident that expressions of emotion constitute a greater part of the women's speech than the men's (97.1 vs. 83.9 per 10,000 words for positive terms and 106.3 vs. 89.0 for negative ones). In fact, both female groups used a higher proportion of positive expressions than their respective partners (98.6 vs. 90.4 and 93.6 vs. 67.2 per 10,000 words). As regards negative emotions, the tendency is the same for the Swiss women and their partners (110.0 vs. 84.6), but the Swiss men used slightly more negative expressions than their partners (100.2 vs. 97.8 per 10,000 words).

Differences were also evident in the number of different types of emotion expressions the men and women used. Relative to the overall number of words they spoke during the interviews, the female participants used a greater number of different types of words than the male participants (38.3 vs. 33.1 per 10,000 words for positive types, and 45.4 vs. 40.7 for negative types). This also applies to the individual groups that take into account the factor of the genders' mother tongues. In contrast, the type-token ratio is identical for male and female participants for positive types (TTR = 39.4%) and a little lower amongst female participants for negative types (TTR 42.7% vs. 45.7%), with no clear trend in the two couple combinations.

In addition to the frequency of usage and the variety of emotion expressions, I was also interested in determining if the genders used different suprasegmental features to accompany their expressions of emotion. An overview of the average use of suprasegmental features by each combination of gender and mother tongue is given in Table 22 below. The figures refer to the percentage of intonation units that were accompanied by various suprasegmental features in each one of the three settings. The absolute numbers are not indicated in the table for the sake of clarity, as a complete table would be very extensive and rather painstaking to interpret. For the same reason, I decided to group similar types of suprasegmental features together, and to omit some types of suprasegmental features that occurred so rarely that they did not yield any definitive results. In addition, the results are displayed in graphic form in Figure 4.


Table 22: Percentage of intonation units with different suprasegmental features for each gender and each combination of language and gender

Key: pos = positive expressions; neg = negative expressions; SG = Swiss German; E = English; bold = in text. See "List of transcription conventions" at the beginning of the book for an overview of all abbreviations for suprasegmental features.

Key: pos = positive expressions; neg = negative expressions; SG = Swiss German; E = English; bold = in text.

tion of language and gender

**Figure 4.** Suprasegmental features in the context of emotion words for each gender (in % of intonation units) Figure 4: Suprasegmental features in the context of emotion words for each gender (in % of intonation units)

132 In many cases, the frequency with which the men and women used a partic‐ ular suprasegmental feature in positive or negative contexts is more or less proportional to their use of this feature overall. The female participants used a soft voice (<P/PP>) slightly more often than the men in all three contexts; both genders used it most in the positive context (in 49.6% [female] and 45.4% [male] of all intonation units) and least frequently in the overall context (46.4% female and 42.3% male), which also applies to the combinations of gender and mother tongue. Similarly, both genders used a wide intonation contour (<W>) far more often when expressing emotions than overall (18.2% and 20.8% in positive contexts; 17.6% and 24.8% in negative contexts; 8.5% and 14.3% overall); in all three contexts, such a contour was used more often by the male speakers, also in the two couple combinations. In contrast, there is no clear correlation of gender with the use of low voice, nor with husky, breathy and creaky voice.

Interestingly, the frequency with which the male and the female participants used a high voice (<HI>) is not proportional in the three contexts. While overall the women used a high voice slightly more often than their partners (22.7% vs. 21.9%), they did so far less frequently than the men when expressing positive emotions (31.8% vs. 36.2%) as well as negative emotions (27.9% vs. 34.9%); this trend is also evident in both couple combinations. In contrast, the women added more emphasis to their emotional speech than the men by using more variation in pitch (<UP/DOW> ) and rhythm (<RH/MRC>), though in the couple combinations, the difference in rhythm was only reflected for negative emotions. While these suprasegmental features were all used more frequently by women than men in the emotional context, they were used similarly often by both genders over the course of the entire interviews.

Since the female bilinguals used more rhythmic and marcato speech when expressing emotions, it is not surprising that they also used more strong stresses than the male partners in these contexts; this can be seen in Table 23 below.


Table 23: Percentage of intonation units with strong and nuclear stresses for each gender and each combination of language and gender

Key: pos = positive expressions; neg = negative expressions; SG = Swiss German; E = English; ^ = nuclear stress; ^^ = strong stress; multi ^ = IUs with multiple stresses; bold = in text.

Table 23 indicates that all four groups used strong stress more frequently in the two emotional contexts than they did overall, and the same applies to intonation units with more than one stress. In all three contexts, there was a higher percentage of intonation units with several stresses in the female participants' speech, yet this tendency was not confirmed in the two couple combinations, which suggests that the participants' mother tongue may have a stronger influence than their gender. There is, however, a gender pattern with regard to the use of strong stresses. The female participants used more strong stresses than the male participants overall (in 5.9% vs. 4.6% of their intonation units), and this difference was even more marked when they expressed positive (13.9% vs. 11.7%) and negative (13.7% vs. 10.5%) emotions. The individual groups taking into account the variable of the bilinguals' mother tongue show the same tendency. It is noteworthy that, in all three contexts, the Swiss women used most strong stresses, and the Swiss men used the least. It is possible that, in different cultures, the level of expressiveness for emotions is gendered differently, although the sample is too small to be able to draw any firm conclusions.

Lastly, I also intended to ascertain if there were any differences in the terminal pitch used by the two genders when expressing emotions. Table 24 below gives an overview of the most important elements of terminal pitch for each combination of gender and mother tongue in all three contexts.


Table 24: Percentage of intonation units with each terminal pitch for each gender and each combination of language and gender

Key: pos = positive expressions; neg = negative expressions; E = English; SG = Swiss German; \_ level pitch; -- = truncated intonation contour; ~ = glissando; bold = in text. See Table 20 above for details on each category.

On average, both genders used fewer level pitches (\_) in the context of emotions than overall, and a greater number of upward as well as downward pitch movements. Furthermore, both genders used fewer glissando rises and falls in emotional contexts than overall. However, while the tendency for the women to use slightly more rising terminal pitches than their partners did overall (in 27.0% vs. 24.3% of their intonation units) was present in the positive context (35.9% vs. 32.7%), it was reversed in the negative one (31.7% vs. 34.3%), and this was also the case in the two couple combinations. Moreover, it appears that the female participants used fewer strong rises and falls than the male partners when expressing emotions (26.7% vs 28.1% in the positive context, and 22.4% vs. 27.1% in the negative one), although their numbers in the interview overall were very close (17.3% vs. 17.7%); however, this trend was not entirely mirrored in the four groups considering the variable of mother tongue. Finally, it is worthy of note that, in the context of negative emotions, there were more instances of truncated speech (--) among the female participants than the male participants (1.1% vs. 0.2%), also in the two couple combinations. This means that, when negative emotions were expressed, the male participants interrupted their partner more frequently than the reverse, even though overall, the female participants interrupted their partner slightly more frequently (1.3% vs. 1.6%). Thus, there are some gender differences with regard to terminal pitch in emotional contexts, but only for some types of terminal pitch.

#### **8.3.6 Emotions and mother tongue**

The final aspect I chose to focus on was whether the participants' mother tongues had an influence on their use of emotion words as well as on the voice quality in the vicinity of these terms. In order to examine this, I again considered each combination of gender and mother tongue, as well as the average numbers of the participants with the same mother tongue. Table 25 shows the frequency with which these groups used positive and negative expressions of emotions, both in absolute terms and relative to the number of words they used overall (normalized per 10,000 words). In addition, the average number of different types of expressions the participants used is indicated.


Table 25: Use of emotion words for each mother tongue and each combination of language and gender (absolute numbers and normalized per 10,000 words) Key: SG = Swiss German; E = English, bold = in text.

Table 25 demonstrates that the English speakers used a similar number of positive and negative expressions, whereas the Swiss participants used more negative than positive ones. Relative to the number of words they spoke during the interviews, the native speakers of English used positive expressions at a slightly higher frequency than the Swiss (91.2 vs. 88.7 instances per 10,000 words) and negative ones at a lower frequency (88.1 vs. 106.9 per 10,000 words). In the case of negative emotion expressions, this is in accordance with the numbers for the groups including the participants' gender. However, in the case of positive expressions, the tendency for all speakers of both mother tongues does not entirely concur with the numbers for the four groups; while the Swiss men used by far the fewest positive expressions (67.2 per 10,000 words), the Swiss women used the most (98.6 per 10,000 words).

In terms of the variety of expressions the participants used, I anticipated that the English speakers would use a greater number of different types of expressions, as they had the advantage of speaking in their mother tongue during the interviews. However, relative to the number of words they used during the interviews, the Swiss used a larger variety of positive and negative expressions (36.4 and 43.5) than the English speakers (34.7 and 42.3), though this tendency was only found for negative emotion words in the two couple combinations. Moreover, the type-token ratio indicates that, relative to the number of emotion expressions they used, the English-speaking partners used a greater number of negative types (TTR = 48.0% vs. 40.7%), but the Swiss partners used a greater number of positive types (TTR = 41.0% vs. 38.1%), which was also the case in the four groups including the variable of gender. My expectations were thus not confirmed, which is probably testimony to the Swiss participants' high level of fluency in their partner's language.

In addition to the frequency of usage of emotion expressions, I was interested in the suprasegmental features accompanying these expressions. Table 26 below gives an overview of the percentage of intonation units with different suprasegmental features in the context of positive and negative emotion words, as well as overall, for speakers of both mother tongues and for each combination of gender and mother tongue. In addition, Figure 5 provides a visualisation of the most important trends regarding the use of suprasegmental features in the three contexts.


Table 26: Percentage of intonation units with different suprasegmental features for each mother tongue and each combination of language and gender

Key: pos = positive expressions; neg = negative expressions; SG = Swiss German; E = English; bold = in text. See "List of transcription conventions" at the beginning of the book for an overview of all abbreviations for suprasegmental features. Key: pos = positive expressions; neg = negative expressions; SG = Swiss German; E = English; bold = in text See "List of transcription conventions" at the beginning of the book for an overview of all abbreviations for suprasegmental features.

**Figure 5.** Suprasegmental features in the context of emotion words for each mother tongue (in % of intonation units) Figure 5: Suprasegmental features in the context of emotion words for each mother tongue (in % of intonation units)

Table 26 and Figure 5 demonstrate that the use of suprasegmental features is relatively similar for both native languages. With regard to loudness (<P/PP>, <F/FF>), the speakers of Swiss German and English differed little from each

used slightly more often by English speakers, though this was the case in all three settings and thus not specific to the expression of emotions. Other types of pitch (<HI> <UP/DOWN>) were used with similar frequency by speakers of both mother tongues; all groups used them more often in the emotional contexts than overall. The largest differences can be seen with regard to a wide intonation contour (<W>); while this featured considerably less in the speech

other, and the frequency of use in the emotional contexts was similar to their overall usage. A low voice (

Table 26 and Figure 5 demonstrate that the use of suprasegmental features is relatively similar for native speakers of both languages. With regard to loudness (<P/PP>, <F/FF>), the speakers of Swiss German and English differed little from each other, and the frequency of use in the emotional contexts was similar to their overall usage. A low voice (<LO>) was used slightly more often by English speakers, though this was the case in all three settings and thus not specific to the expression of emotions. Other types of pitch (<HI> <UP/DOWN>) were used with similar frequency by speakers of both mother tongues; all groups used them more often in the emotional contexts than overall. The largest differences can be seen with regard to a wide intonation contour (<W>); while this featured considerably less in the speech of the Swiss participants than the English speakers overall (in 8.5% vs. 14.4% of their intonation units), also in the two couple combinations, the Swiss used it with a similar frequency as the English speakers in the context of positive emotions (19.6% vs. 19.4%) and negative emotions (19.8% vs. 22.6%). Finally, the Swiss also accompanied their emotion speech more frequently with husky, breathy or creaky voice (<HSK/BR/CRK>) than the English speakers, in positive (29.2% vs. 25.3%) and negative contexts (37.4% vs. 30.0%), which was also visible in the two couple combinations, even though there was little difference in their usage overall (24.1% vs. 23.5%).

Since the use of stress is language-specific to some extent, it is possible that speakers of different mother tongues use it more or less frequently to add emphasis when they are expressing emotions. Table 27 gives an overview of the percentage of intonation units that were accompanied by nuclear and strong stresses for each combination of gender and mother tongue.



Key: pos = positive expressions; neg = negative expressions; SG = Swiss German; E = English; ^ = nuclear stress; ^^ = strong stress; multi ^ = IUs with multiple stresses.

Table 27 indicates that, while both groups used multiple stresses more frequently in the emotional context than overall, the Swiss participants used multiple stresses more frequently in the context of positive emotion words than their partners (7.3% vs. 3.7%), whereas the Anglophone participants used them more often in the negative context (6.4% vs. 5.0%). This tendency persists when the groups including the participants' gender are considered. Furthermore, the averages of the native speakers of both languages suggest that the Swiss participants used more strong stresses than the speakers of English overall (5.7% vs. 4.7% of IUs), while both groups used them similarly often when expressing emotions (12.8% vs. 12.7%, and 12.3% vs. 11.9%). However, it is also evident that, in all three contexts, the Swiss women used strong stresses most often, while the Swiss men used them the least. Thus, a clear trend can be observed with regard to the influence of the speakers' mother tongue on their use of multiple stresses, but not on their use of strong stresses in emotional speech.

The final aspect I decided to examine was whether the speakers of both mother tongues used different types of terminal pitch to accompany their expressions of emotion. The most important elements of terminal pitch are summarized in Table 28 below.


Table 28: Percentage of intonation units with each terminal pitch for each mother tongue and each combination of language and gender

Key: pos = positive expressions; neg = negative expressions; SG = Swiss German; E = English; \_ = level pitch; -- = truncated intonation contour; ~ = glissando. See Table 20 above for details on each category.

Table 28 reveals that the participants' use of terminal pitch in all three contexts is relatively similar in some respects. Hence, all four groups used more rising and falling pitch overall ("up" and "down") when expressing emotions than they did overall and used more regular as well as strong pitch movements. Consequently, all groups used fewer level pitches to express emotions than they did overall. However, the Swiss participants used more falling pitch contours than the Anglophone partners in all three contexts, with a slightly more marked difference in the positive context (45.2 vs. 41.8), and this trend was reflected in the individual groups including gender. Meanwhile, the English speakers used more level pitches than the Swiss partners in the positive and negative contexts (24.3 vs. 18.9, and 25.5 vs. 22.5), which was also evident in the two couple combinations. These findings concur with the greater use of a wide intonation contour (<W>) in the emotional context among the Swiss participants that was described earlier in this section.

## **8.4 The couples' reports on expressing emotions as a bilingual couple**

The previous sections have demonstrated that there are a number of subtle differences in the manner in which emotions are expressed by speakers of different genders and/or mother tongues, ranging from the type and number of expressions to the intonation contour and voice quality that accompany the expression of positive and negative emotions. In order to develop a deeper understanding of the manner in which bilingual, bicultural couples express emotions, we will now turn to the interviewees' reports on the subject. The couples were asked which language(s) they use to express positive or negative expressions, why they chose this particular language (or these languages), and whether the non-native speaker ever considers this an issue. They were also questioned about their use of terms of endearment and swearwords (the latter will be discussed separately in chapter 9, "'You're not gonna say this word!': Swearing"). Moreover, the couples were asked whether they have different ways of arguing or of categorizing arguments. My aim is to determine potential difficulties, to see if and how the couples have overcome these, and to find out whether the partners also see advantages in expressing emotions in a second language. Furthermore, I am interested in establishing whether the couples' language choice for the expression of emotions diverges from their general language use, and whether the couple language, as it is regularly used in an emotional context, tends to become the non-native speaker's preferred language to express emotions over time. In addition, I want to see whether it makes any difference to the bilinguals if the emotions that they express are positive or negative, in terms of their language choice as well as potential difficulties. In the following, I will first give an overview of their experiences with arguing and expressing negative emotions and then discuss their thoughts on expressing positive emotions, as well as their use of terms of endearment.

#### **8.4.1 Expressing negative emotions and arguing**

Almost all of the couples use English to express negative emotions or to argue (9/10). Several of them stated that this an obvious choice, as English is their stronger language, and they deem it important to really understand each other in these situations (David <sup>E</sup> ). The only couple who habitually express negative emotions in both languages are Joshua and Deborah, who often use their respective mother tongues for this purpose. Thus, all of the couples avail themselves of their usual couple language(s) to express emotions as well. Interestingly, when this subject was addressed, the majority of the couples also mentioned that they argue very rarely, even though they may have heated discussions (Monika/Dean, Sophia/Richard, Susanne/David, Courtney/Martin, Joshua/Deborah, Stephanie/Robert). One couple reported that they argue quite often (Claire/Simon), while the remaining three couples did not comment on how much they argue.

The couples' answers to the question if they have culturally different manners of arguing or categorizing argument were heterogeneous. While three of the couples do not think that there are cultural differences in the manner in which they argue, or are uncertain about it (Sophia/Richard, Susanne/David, Deborah/Joshua), the remaining seven answered in the affirmative (Courtney/ Martin, Katia/Craig, Karen/Philipp, Monika/Dean, Stephanie/Robert, Sarah/ Tim, Claire/Simon). For example, Katia SG and Craig <sup>E</sup> believe that there are differences in the way they react, explain things, show emotions, or argue a point (Craig 1142–1144). In contrast, Sarah SG and Courtney <sup>E</sup> report that they want to discuss issues, no matter what the time, and dig for the root of the problem, while their partners consider this less urgent or prefer to focus on the problem at hand. However, while many partners believe that they differ in the manner in which they think a problem should be addressed or in their style of arguing, they are often undecided whether these differences are down to their cultural backgrounds. Many interviewees think that such differences could also be gender-related (Sarah SG, Courtney <sup>E</sup> ), due to their family backgrounds (Courtney <sup>E</sup> , Monika SG) or their personalities (Katia SG, Monika SG, Philipp SG). Some interviewees also emphasize how much they have adapted their style of arguing or resolving disagreements to each other. This is particularly evident in the case of Monika SG, who initially struggled with the fact that she and her partner handled conflicts differently. Monika says that there is a "culture of argumentation" in Dean's <sup>E</sup> family, and there are lots of loud arguments (637), while she did not grow up in a family environment where arguments were cultivated. In the past, this caused her to give in very quickly when they had arguments — in order to re-establish harmony between them — which, in turn, really annoyed and frustrated Dean:

```
Example 8.3: Dean (English, 29) and Monika (Swiss, 29) 
637 Monika: 
638 Dean:
639 Monika:
640 Dean:
642
643 Monika:
644 Dean:
645 Monika:
                <P<HSK I always a^ssociated an argument with: HSK>P> _
                <PP<HSK a lack of ^love or: HSK>PP> \
                <PP<HSK ^em: HSK>PP> _
                a ^^loss: /
                <PP<LO of ^love LO>PP> _
                <H> <P so I'd ^always: P> _
                … <P ad^mit P> /
                <P that ^I was wrong P> \
                <P<DOW and say ^sorry even though I was: DOW>P> _
                <P<HSK ^in the right and HSK>P> \
                <H> <PP<L<HSK ^he would HSK>L>PP> _
                <P<L<RH ^force me to ^argue with ^him RH>L>P> \
                <P<L<HSK just to ^learn it a bit HSK>L>P> \
                <PP ^and PP> _
                1[<H>]
                1[<H> <TSK>] <P<HI ^yea:h HI>P> \\~
                2[<HI<L I ^did: L>HI>] \\ ~
                2[<A it p- it p- it p- A>] --
                <W<:-) it pissed me <HI ^off in the beginning HI> :-)>W> _
                ((…)) <HSK you know if we got an <HI ^argument HI>HSK> \\
                <P<LO<HSK about ^something HSK>LO>P> _
                <P<LO<HSK ^e::m HSK>LO>P> _
                <H> <P and ^I would: P> _
                <Hx> … @ <@ I would win it hands ^down @> /
                <@ wi^thin a: @> _
                <@<W within a ^minute or two W>@> \\
                = <P ^m:hm P> \~
                ^em: _
                <A<W and <HI she HI> would a^pologize to me W>A> \
                <A<W and I there was <HI ^many times HI>W>A> \
                <P<A<HI<HSK where I ^turned around and said HSK>HI>A>P> _
                <P<A<W<HI what the hell HI> are you a^pologizing for W>A>P> \\
                <W<:-)<CRK<HI I'm ^^wrong HI> here CRK>:-)>W> \\
                = <P m^hm: P> /\~
```
646 Dean: = <H> <A<HSK ^you know HSK>A> \ ^and \_

<sup>650</sup> ((…)) **she was a ^push-over** \ ^and \_

<W<HSK **it was frus^trating** HSK>W> \\

<%> <A<HSK y- **I ^don't want someone** HSK>A> \_ <A<HSK that admits they're ^wrong HSK>A> \_

… <A **because they think it's the right thing to ^do** A> //

… <W **it frus^trated me** W> \\

<sup>652</sup> ((…)) or someone **that giv:es ^in** /

<A<HSK when they're ^not HSK>A> //

```
646 Dean: = <H> <A<HSK ^you know HSK>A> \
              ^and _
              <W<HSK it was frus^trating HSK>W> \\
650 ((…)) she was a ^push-over \
              ^and _
              … <W it frus^trated me W> \\
              <%> <A<HSK y- I ^don't want someone HSK>A> _
              <A<HSK that admits they're ^wrong HSK>A> _
              <A<HSK when they're ^not HSK>A> //
652 ((…)) or someone that giv:es ^in /
              … <A because they think it's the right thing to ^do A> //
              or ^they're [<H>] _
653 Monika: [<PP<HI m^hm: HI>PP>] \~
654 Dean: they d- ^don't want to: \
              <W to <HI ^argue HI>W> \\ (637–654)
```
<A<W **and** <HI **she** HI> **would a^pologize to me** W>A> \ <A<W and I **there was** <HI **^many times** HI>W>A> \

<W<:-)<CRK<HI **I'm ^^wrong** HI> **here** CRK>:-)>W> \\

<P<A<HI<HSK **where I ^turned around and said** HSK>HI>A>P> \_

The extract shows that arguing carries very different connotations for Monika and Dean. While Monika associates arguing with emotional distance, Dean does not view it negatively per se, and thinks that it can also be stimulating. He does not want his interlocutor to give in because he or she thinks it is the right thing to do (652) or because he or she does not want to argue (654), and equates the attempt to avoid conflict with weakness (646).<sup>4</sup> Therefore, Dean would sometimes force Monika to take a stance, and thus, in her opinion, taught her how to argue (637–639). As a result, the partners seem to have developed a method of dealing with disagreements which lies somewhere between the two approaches with which they grew up.

In addition to having different manners of arguing, some couples also feel that they disagree on the point at which a disagreement becomes an argument. Karen <sup>E</sup> thinks that she has a lower threshold in this regard than her husband does, with which her husband agrees. For Claire <sup>E</sup> and Simon SG, this appears to be a major source of conflict, as many arguments start with Claire getting upset with Simon for voicing what he perceives to be mild criticism:

<sup>645</sup> Monika: = <P m^hm: P> /\~

<sup>4</sup> One is tempted to attribute the differences in the interviewees' apologizing behaviour to gender; but the only other time this is mentioned in the interviews, the male partner apologizes substantially more often (Simon SG, see next paragraph).

```
Example 8.4: Claire (Northern Irish, 28) and Simon (Swiss, 34)
1606 Simon: <P<A<W I have the <HI feeling HI> that sometimes I <HI ^say something: HI>W>A>P> _
               <P<A ^maybe A>P> \
               <H> <CRK ^^slightly CRK> \
               … <P<CRK in a criticizing ^way CRK>P> /~
               <P<HSK and ^she: HSK>P> /
               <P and [and ^you s-] P> _
1607 Claire: [<P take a ^huff P>] \
1608 Simon: <P and ^you tell me P> _
               <P ^oh P> _
               stop ^^shoutin' //
               <P and .. [I'm ^not] P> /
1609 Claire: [<P o:h yea:h P>] \~
1610 Simon: I'm <HI ^^not shoutin' HI> \\
               … I'm just .. ^sayin:' \
               … <%> … it's not ^great //
1629 Claire: ((…)) <P I <HI ^huff HI> very [<LO quickly then LO>] P> _
1630 Simon: <P<HSK [you ^huff very] quickly HSK>P> _
1631 Claire: = <P I [^do yeah] P> /~
1632 Simon:
1633 Claire:
1634 Simon:
1635 Claire:
1636 Simon:
1637 Claire:
1638 Simon:
1639 Claire:
1642 Simon:
1643 Claire:
                [<P ^and P>] _
               = <P<LO and and you tell me I'm ^shoutin' LO>P> //
               <P<LO whenever I'm .. I'm LO> ^not P> //
               <PP ^yea:h PP> /~
               … ((long pause))
               <PP<LO an' I'm far [more LO> for^giving <LO than you but LO>PP>] _
                                 [<PP yeah PP> <H> and <HI ^you HI> want] \
               <W y- you want to dis^cuss things W> //
               <P<LO and I just ^want [to go] LO>P> _
                                  [<PP ^yeah PP>] _
               <PP<W<BR oh <HI piss HI> ^o:ff BR>W>PP> \\
               <P [and] .. ^just P> _
                  [<PP<HSK ^yeah HSK>PP>] _
               <P no:t to talk to #him ^huff with him P> /
               ((…)) <P and you say <HI ^sorry a HI> lot P> \
               <P<HSK but you never a^pologize HSK>P> \
               … ((long pause)) [@@@@] ((chuckling))
                             [<P ^em:: P>] _
               <P<W<HI I usually HI> think I'm ^right yes W>P> /
               [<%> ^eh <%m:>] 
1644 Simon: [@@@@] ((chuckling softly))
1645 Claire: <P<A ^ninety percent of the time A>P> \
               <P<:-) ^Simon is: .. apologizin' :-)>P> _ (1606–1645)
```
Claire and Simon seem to differ not only in the manner in which they perceive or categorize arguments, but also in the way they argue. Simon wants to discuss the point of conflict, whereas Claire wants to stop talking and sulk in silence (1635–1637). Moreover, Simon believes that he is far more forgiving than his partner (1634) and, as Claire notes, he also apologizes more readily and frequently (1645). Simon later expresses annoyance that Claire says sorry "all the time" without being truly sorry (1653). To his mind, the expression carries no weight for her, which he attributes to her cultural background. In Claire and Simon's case, their different manner of arguing and categorizing arguments thus appears to add fuel to their quarrels.

The issue of being misunderstood by one's partner, which ultimately tends to result in an argument, was raised by various interviewees, and is one of the reasons why expressing emotions in a non-native tongue can be difficult. When asked whether they ever see it as a problem when they have to express negative emotions or have to argue in their L2, five of the nine participants to whom this applies answered affirmatively. In the following example, Sophia describes the frustration she sometimes experiences when she cannot express precisely what she intends to during a disagreement or an argument:


According to Sophia and Richard, not finding the right words in these contexts has several consequences. It can be exasperating to the non-native speaker (1209), it often aggravates the argument (1210), and it can also turn the conversation into a discussion about semantics rather than the issue at hand (1218, 1220). Moreover, being in a weaker position linguistically during a debate or argument can cause the non-native speaker to feel at a disadvantage (1209), or even evoke feelings of inferiority. This is also the case for Stephanie SG, as the extract below shows.

```
Example 8.6: Robert (English-Irish, 20) and Stephanie (Swiss, 24)
606 Stephanie: ((gulp)) because when ^I /
               <P get e^motional P> //
               <P<CRK like and I have to talk <HI ^English HI>CRK>P> _
               <P and I have to make it ^clear P> //
               <H> <P then <HI ^sometimes HI>P> _
               <P<MRC I <HI really HI> lack ^words MRC>P> /\~
               <P or so that for ^me P> //
               <P would be ^easier P> \
               <P<LO to maybe say it in LO> ^German P> /
               <H::> … <@ and so I ge@t even [^more emotional] @> /
607 Robert: [@@@] ((chuckling))
608 Stephanie: [@@@] 
609 Robert: [@@] ((softly))
611 Stephanie: ((…)) <DOW<MRC<HI he:'s HI> always su^perior MRC>DOW> \\
               <P be^cause P> _
               <:-)<MRC it's ^his ^language MRC>:-)> \\
               <P it's his ^mother tongue P> \
               <H:> so he ^always has \
               better ^words /
               <UP and can ex^press himself better UP> /
               ((…)) I <HI really HI> lack ^words then \
612 Richard: [<P m^hm P>] \~
613 Stephanie: [<P and ^I P>] _
               … <W I <HI ^feel HI> like inferior W> \\
               <W<BR because I <MRC ^can't really ex^press MRC>BR>W> /
               <P<LO what I LO> ^mean sometimes P> _ (606–613)
```
Stephanie describes that she becomes even more emotional when she senses that she lacks the right words to express herself in emotional situations, which renders the conversation even more difficult, and makes her feel even more inferior to her partner. In the extract above, she repeats twice that she lacks words (606, 611) and that her boyfriend is able to express himself better than she is (611, 611). Similarly, she not only states that her partner is "always superior" (611), but also that she herself feels "inferior" to him (613), which underlines the gravity of the issue for her.

Interestingly, various interviewees seem to have ambivalent feelings about expressing negative emotions in their L2. Martin, for instance, laments that he sometimes does not find the right words when he is angry, which he thinks diminishes the force of expression (see example 9.3 in the next chapter), but at the same time, he cannot imagine using his native tongue to express emotions to a partner (see example 8.9 below). Similarly, even though four of the nine participants who habitually express negative emotions in their L2 initially denied that they ever feel disadvantaged because of this, two of them contradicted their statement to some extent at another point in the interview (Katia SG, Simon SG). When evaluating her own language skills, Katia SG mentions that she occasionally feels that she cannot say certain things in her L2, especially when they are speaking about emotional or personal matters. She finds it frustrating that this happens even though she has been speaking the language "for years and years and years" (839). Similarly, Simon SG states that he has difficulties expressing exactly what he wants on occasion, which can lead to misunderstandings. Although he professes that this is not a major problem, he and his wife agree that many of their arguments start with one of them saying something that is taken the wrong way by the other one.

```
Example 8.7: Claire (Northern Irish, 28) and Simon (Swiss, 34)
702 Simon: <Hx::> … ((long pause))
                <P<W<HSK I <HI think HI> ^^sometimes HSK>W>P> \
                <P<HSK<A there's ^still a bit of a A>HSK>P> \
                … <P<HSK a bit of a misunder^standing HSK>P> \
                … <P<HSK at ^times HSK>P> \\
                <P<HSK ^that HSK>P> _
                <P<HSK <HI ^I HI> have the feeling HSK>P> _
                <P<HSK I <HI can't HI> really e^xpress myself HSK>P> \
                <P<HSK e^^xactly the way I would HSK>P> _
                … <PP<LO<HSK I would ^want to HSK>LO>PP> \
704 ((…)) <:-)<W we <HI ^argue HI> qui@te a lot but W>:-)> \
705 Claire: @@ ((chuckling))
706 Simon: = <HSK<W<HI [^some] HI> arguments are probably: W>HSK> \~
707 Silja: [@@] ((softly))
708 Simon: … <HSK ^a:lso due to that HSK> /
                … ((lengthy pause)) <P I ^just P> /
                <HSK I'm ^not a native <LO speaker LO>HSK> _
                <HSK<LO ^and LO>HSK> _
                <HSK ^yes: HSK> _
                … <P<W<HSK I can still <LO express myself LO> ^much <LO better in
                my: LO>HSK>W>P> _
                … <P<HSK<LO in my native tongue than I can in LO> ^English HSK>P> _ 
                                                                       (702–708)
```
Simon attributes his inability to express himself precisely to the fact that he is not a native speaker (708); at the same time, he repeatedly asserted during the interview that communicating in English is no issue at all for him. Interestingly, the topic of being at a loss for words surfaced in the answers of most bilinguals, but only ever in the context of expressing negative emotions. While many of the interviewees theoretically do not mind expressing negative emotions in their L2, they dislike the fact that their failure to find the right words can lead to misunderstandings or arguments, and that it can put them in a weaker position in situations where they have to assert themselves.

The fact that one partner has to express negative emotions in his or her non-native tongue can also be difficult for the native speaker. On the one hand, the latter is sometimes confronted with imprecisely worded criticism and may take this personally or might also misunderstand his or her partner; on the other hand, the native speaker may feel bad that his or her partner is put in this position. Courtney <sup>E</sup> , for instance, sometimes feels guilty that she does not speak German well enough, and that her partner has to speak English to her in challenging situations:


In this example, Courtney speaks in an emotional manner, using lengthening, pausing as well as a creaky, husky, and breathy voice quality. Moreover, she switches from referring to Martin as "him" to addressing him directly at the end of the extract, thus adding emphasis to her statement. The example highlights that this is not only an issue that the non-native speaker is faced with, but often one that concerns both partners.

Nonetheless, there are also various non-native speakers who claim not to feel disadvantaged by their situation, or who even perceive advantages in using their L2 to express negative emotions. Katia SG, for instance, states laughingly that she does not mind because "[she]'ll talk about [her] emotions in German with other people" (1025). Her husband adds that she has a very strong personality, so he is never worried that "she doesn't get her point ac@ross" (Craig <sup>E</sup> 1052). Moreover, many of the participants are much more accustomed to expressing emotions in their L2 than their L1, as they have been with their partner for most of their adult life. This is the case for Martin SG, who had never had a long-term relationship before he met Courtney <sup>E</sup> . In the extract below, he describes the bewilderment he would expect to feel if he had to express his emotions in German instead of English:


Martin repeats three times how "strange" or "weird" it would be for him to use German in these situations, and uses intensifiers ("rea::lly", "very") as well as vocal cues such as lengthening and a wide intonation contour (<W>) for emphasis. This highlights that, while having to express emotions in an L2 might initially be detrimental, potential disadvantages tend to diminish or even disappear with added cultural experience, longer duration of the relationship, and increased language skills.

#### **8.4.2 Expressing positive emotions and using terms of endearment**

The previous section has shown that almost all of the couples express negative emotions exclusively in English. The only couple who use both languages, Deborah SG and Joshua <sup>E</sup> , also use both languages for positive emotions. The other couples use mainly English to communicate positive emotions, though there are three participants — Stephanie SG, Robert <sup>E</sup> and David <sup>E</sup> — who occasionally use (Swiss) German for this purpose. Stephanie SG reports that if she wants to express certain positive emotions in German, she does so and explains to Robert <sup>E</sup> what the expression in question means, so that he "recognizes [it] when it comes again" (651). Robert thinks that it is nice to use German to speak about love, because it is a new language for him and he feels that it is easy to say *I love you* in German. David E sometimes tells Susanne SG that he loves her in Swiss German, even though she does not do this ("I've never heard you say *I liebe dich* to me in German" [David, 909]). It is notable that none of the Swiss participants voiced a need or preference for expressing positive emotions in their mother tongue, which stands in contrast to their reports on expressing negative emotions.

Upon being asked whether they mind expressing positive emotions in their L2, and whether they ever miss using their mother tongue, all of the interviewees who regularly do so denied this, some of them even vehemently. In fact, the majority of them (8/10) even professed that it is often easier for them to express positive emotions in their L2 than in their L1. They named a variety of reasons for this, yet most of them attribute it to the respective characteristics of both languages. Deborah SG, for example, thinks that it is sometimes easier to say *I love you* in English than in Swiss German. Susanne SG feels the same way, because she is of the opinion that the phrase *I love you* sounds strange in Swiss German (910).<sup>5</sup> Other interviewees regard English as a particularly creative language, or find that Swiss German is sometimes unable to convey the emotion they intend to express:

<sup>5</sup> The reasons for this are twofold. First, *I love you* is used much more frequently and to address more people than the corresponding (Swiss) German expression. While the English phrase can be said to friends or family members, this is not the case for the German translation, which is reserved for romantic love. Second, the phrase is used in Standard German and can theoretically be translated literally to Swiss German (*Ich liebe dich* 'I love you'), but instead, it is much more common to say *Ich ha di gärn* (literally 'I like / care about you'). The verb *liebe* 'to love' does exist, but it sounds odd to many Swiss German ears.

```
Example 8.10: Stephanie (Swiss, 24)
Stephanie: <P<LO ^no: LO>P> _ ((answering the question if she minds expressing positive emotions in her L2))
               <P<LO not at ^all LO>P> \
               <P<:-) be^cause :-)>P> _
               <P<BR ^I think BR>P> _
               the English ^language \
               <UP is is ^so free UP> /~
               <UP I think it's ^so creative UP> // (634)
Example 8.11: Monika (Swiss, 29)
Monika: <P it ((English)) ^^mea:ns a lot to me P> \
               <H:> ^sometimes \
               <HSK ^I feel HSK> \\~
               <P<BR that I have to express myself in ^^English: BR>P> \
               <P because Swiss German would ^^not P> /
               <P<HSK be able to capture the e^motion HSK>P> \
Silja: <PP m^hm PP> \
Monika: <PP<HSK that I would ^want to HSK>PP> \
               <PP<BR ex^press BR>PP> / (1050–1052)
```
Other Swiss partners claim that they are more accustomed to or experienced in speaking about their positive emotions in their L2 than their L1. This may be because they have been together for a long time (Philipp SG, Monika SG), or because they have not had any long-term relationships other than their current one (Martin SG; see example 8.9 above). Philipp attributes his preference to the fact that English has become a language with a high emotional attachment, so that he does not even consider speaking German to his wife a viable alternative (1099), and even finds it difficult to express emotions in German to his children (see section 7.6, "The couples' views on raising bilingual children"). Similarly, Sophia SG, who enjoys expressing her feelings in English very much, wishes that she could sometimes use her L2 instead of her mother tongue to address her daughter, which points to the high level of emotionality her non-native language carries for her:

```
Example 8.12: Sophia (Swiss, 31)
Sophia: <HSK<MRC I ^^like ex^^pressing my ^^feelings MRC>HSK> \
               <P in ^English P> _
               … to ^him \\
               <CRK<W<RH and <HI ^some HI> times I re^gret RH>W>CRK> /
               <P<HSK<RH that I can't ^do it with our <HI ^daughter HI>RH>HSK>P> \
               … <P<HSK that I ^have to use this: this:: HSK>P> _
               <P<HSK ^harder language HSK>P> \ 
               … <P<HSK with ^her HSK>P> _ (1166)
```
Unlike Philipp SG, however, Sophia does not consider using English to articulate feelings towards her daughter an option at all, even though she conveys great enthusiasm about using the language with her husband, and the manner in which she refers to Swiss German indicates some alienation from her mother tongue ("this: this:: harder language" [1166]). While in the case of Philipp and Sophia, the high level of emotionality of their L2 leads to a desire to use this language with their children, other bilinguals prefer using their L2 in certain situations precisely because it is *less* emotionally charged for them. An example of this is Claire <sup>E</sup> , who switched to German in her wedding speech to prevent herself from crying, as she was becoming too emotional when speaking her mother tongue. Thus, the subjective language emotionality may influence which language the bilinguals prefer to express emotions in a given situation.

Within the context of expressing positive emotions, the use of terms of endearment seems to occupy a special position, as these evolve and establish themselves over time. Interestingly, the interviewees use Swiss German much more frequently to nickname each other than to express (positive or negative) emotions in general. Thus, only three couples reported that they employ mainly English terms of endearment (Sarah/Tim, Katia/Craig, Karen/Philipp); Martin and Courtney, for instance, use the term *baby*, which they initially used to mock other couples, and it just "stuck" (Courtney <sup>E</sup> 621). One couple, Stephanie and Robert, use both languages to nickname each other, while remaining six couples use mainly Swiss German terms of endearment (e.g. *Schatzi* 'treasure' [Deborah/Joshua], *Bibii* 'chick' [Sophia/Richard]) as well as made-up and blended words, which usually have a Swiss German phonology (e.g. *Knüffel* [Courtney/Martin], *Schminzi* [Monika/Dean], *shnugglesy* [Claire/Simon]; for more examples, see section 10.4.4, "Playful language in bilingual couple talk"). Various interviewees emphasized that they greatly enjoy using or hearing terms of endearment in their L2 (Dean <sup>E</sup> , Monika SG, Robert <sup>E</sup> , David <sup>E</sup> ). Dean, for instance, uses more Swiss German terms of endearment, and only nicknames his partner in English on special occasions, particularly when he intends to make her feel good. The following extract shows that the couple both like nicknames in each other's native languages; Monika considers English terms of endearment to be more special, whereas Dean finds Swiss German nicknames cute.

```
Example 8.13: Dean (English, 29) and Monika (Swiss, 29)
Monika:
Dean:
Monika:
Dean:
               <P<DOW ^I like it as m- <%%> DOW>P> --
               <P<HSK<HI ^very HI> much when you HSK>P> _
               <P<HSK you use ^English words HSK>P> \\
               … <H>
               <P to: P> _
               <P<HSK ^nick <LO name me LO>HSK>P> _
               … <P<W instead of <HI ^Swiss HI> German words W>P> \
               <P<W be^^cause <LO it's so special LO>W>P> /
               <H:> and the ^worst thing is like \
               … <BR because ^sometimes BR> _
               <W those .. terms of en^dearment W> \ ((in Swiss German))
               … <P<HSK when you ^say them out HSK>P> _
               <PP<HSK<:-) in ^public: :-)>HSK>PP> \
               <PP<:-) and it's just like ^mh :-)>PP> \\~ ((indicating embarrassment))
               <H::> [^so] _
                [<LO ^what LO>] _
               <W<RH the ^Swiss <HI ones or [the HI> ^English] RH>W> \
Monika: [^m:h] \~
               <P ^Swiss [ones] P> \\
Dean: [<H:>] but ^then: /
               <A <%> I don't ^know A> //
               there's ^something \
               about a lot of the ^Swiss words \
               <W<DIM<HSK that sounds very <HI ^cute HI><LO to me: LO>HSK>DIM>W> _
Monika: <P ^mh P> _
               <HI yeah of ^course HI> //
               for ^^you //\\~
               <HSK but it's ^lik:e HSK> _
               … and for ^us: /
               <P<HLT<HSK but then for ^other people it's: HSK>HLT>P> \
               <P<LO oh my ^God LO>P> / (972–978)
```
The extract also indicates that people are often more critical of or more sensitive about terms of endearment in their native tongue than in other languages. Thus, Monika insinuates that she is highly embarrassed when they use Swiss German terms of endearment in public (974, 978), anticipating that other people will find these terms awkward ("for other people it's oh my God" [978]). A reaction similar to Monika's was reported by Craig <sup>E</sup> , who called his wife Katia SG *Schatzi* 'treasure' once, which she did not appreciate at all (and the couple jokingly indicated that this resulted in a slap in the face).

Just like in the context of other expressions of emotion, different connotations attached to particular terms of endearment can lead to misunderstandings. Such a misunderstanding can be witnessed in the extract below. A few days prior to the interview, Claire <sup>E</sup> called Simon SG *Liebling* 'darling' in a text message, a term which is not commonly used in Switzerland, and is perceived to be German and potentially antiquated. To tease her for what he thought to be an odd choice of a nickname, Simon called her *Mausi* 'little mouse' in response. The joke was lost on Claire, though, and she liked being nicknamed *Mausi*.

```
Example 8.14: Claire (Northern Irish, 28) and Simon (Swiss, 34)
Claire: <W you called me <LO ^Mausi LO><HI the other HI> day W> \ (('little mouse'))
             <P<W and eh <%i> [ma- I ^like that] W>P> /
Simon: [<P<HI just HI> to ^be: P>] \
             … <P<:-)<CRK<HI a^nnoying HI>CRK>:-)>P> \
Claire: <W<HI oh I HI> thought that was ^cute W> /~ 
Simon: = [<@ ^oh @> @@@@@ ((softly))] \~ 
Silja: [@@@@@@@] 
Claire: <W cause I [<LO ^wrote to LO>]W> _
Simon: [<HI ^okay: HI>] \
Claire: <W li- [I wrote <HI hallo HI> ^Lieblin:g] W> _ (('darling')
Simon: [<HI<@ I ^meant it like that @>HI>] _
Claire: [<W<:-) ^didn't I write hallo Liebling :-)>W>] //
Simon: [<:-) # ^yes because you wrote Liebling :-)>] \
             <:-)<HSK and I wrote back [ha-] HSK>:-)> _
Claire: [^Liebling] /
Simon: <P<:-)<HSK hallo ^^Mausi HSK>:-)>P> \\ ((savouring, teasing))
Claire: … [<P ^Mausi P>] \
Simon: [@@@] ((chuckling))
Claire: <P<:-) ^yea:h :-)>P> /~ (1515–1529)
```
Simon, who only comes to realize the misunderstanding during this interaction, is laughing a lot and speaking with a smiling voice throughout the extract. To him, the fact that his wife called him *Liebling* 'darling' is a source of amusement, as is his mocking response with what he deemed to be a similarly weird nickname. As Claire was unaware of the connotations of either of the terms, she could neither predict her partner's reaction nor interpret his response correctly. This example highlights that particular terms have different associations for native and non-native speakers, which potentially results in misunderstandings.

## **8.5 Discussion and summary**

In this chapter, the expression of emotions among bilingual couples was explored based on the couples' reports and the analysis of their use of *emotion words* as well as *emotion-laden* and *emotion-related words* (Pavlenko 2008: 148) during the interviews. In their conversations, the participants used a slightly greater number of terms expressing negative emotions than positive emotions, and also a greater variety of different negative expressions than positive ones, potentially because the expression of negative emotions or criticism requires a more careful wording. Furthermore, there were a remarkable number of suprasegmental features that occurred particularly often in the context of emotional speech. Most obviously, all types of prominent pitch (high, low voice) or pitch movement (wide intonation contour, upwards or downwards pitch) were present more frequently in the emotional contexts than overall. Thus, the most prominent vocal cues in the context of emotions in this study were not the same as the vocal cues that the participants in Planalp's (1998) study most frequently associated with emotionality, namely loudness and speed. As might be expected, low and downward pitches were most common in negative contexts, while high and upward pitches occurred most frequently in positive contexts; breathy and creaky voice was also very common in both emotional contexts. Interestingly, smiling voice quality was not only frequent in the context of positive emotions, but also in the negative context, which indicates that it can serve speakers to mitigate the force of negative comments. This is not the case for laughing voice quality, however, which was much less common in the vicinity of negative emotion expressions than positive ones (see also section 10.3.5, "Use of laughing and smiling voice quality", for a discussion of the participants' use of smiling and laughing voice quality overall). Overall, the findings with regard to suprasegmental features confirm Pavlenko's (2005: 45) observation that emotions can be conveyed by a range of different cues, and that particular vocal cues may be associated with various emotions.

As regards the rate of speaking, the findings are in line with expectations, with a slow speech rate being most common in the negative context and a fast speech rate most common in the positive one. Unexpectedly, halting speech was not more frequent in either of the emotional contexts than overall. There were fewer truncated intonation contours in both emotional contexts than overall, which might indicate that the partners deemed these topics too important to interrupt each other. Moreover, strong stresses were used very frequently when emotions were expressed, and many intonation units had several stresses (but less lengthening than overall). Differences were also noted with regard to terminal pitch. Thus, there were fewer level pitches in contexts where emotions were expressed, a greater number of strong upward and downward pitch movements, and fewer glissando pitch movements. Contrary to expectations, there was only a minor difference between upwards and downwards pitch movements in the two emotional contexts.

Clear gender differences were noticeable in the expression of emotions during the interviews. Relative to the number of words they used during the interviews, both negative and positive emotion terms constituted a slightly greater part of the speech of the female than the male participants, which is in line with the gender stereotype of women as more expressive of their emotions than men, and with findings of previous research (Notarius and Johnson 1982; Piasecka 2013). It is worth mentioning, though, that in absolute numbers, the males (who spoke more words during the interviews than their partners) did not use fewer emotion terms than the female participants. It is possible that the male participants in this study were more inclined to speak about emotions than an average male, as only men who are willing to do so might participate in an extended interview about their relationship. Moreover, the female partners used a greater number of different types of expressions than the male partners relative to the number of words they spoke during the interviews, which is in accord with Dewaele and Pavlenko's (2002) findings. Meanwhile, the majority of the suprasegmental features were used proportionally to their use overall by both genders, with the exception of emphatic speech. Thus, female participants were more inclined to add emphasis to their emotional speech by using pitch variation and a strong rhythm, including a greater number of strong stresses, while male participants used a high voice more often than their partners in the emotional context. Moreover, the female bilinguals used more rising terminal pitch contours overall and in the context of positive emotion words than the male participants, yet this tendency was reversed in the negative context.

The participants' mother tongue appears to play a slightly lesser role in the expression of emotions than their gender. In relation to the number of words they used, the Anglophone participants used slightly more positive emotion expressions and fewer negative ones than the Swiss partners, which might be due to differences in the emotion vocabularies of the cultures, or could indicate that the Swiss partners perceive their relationship (or their position as non-native speakers) as more problematic. However, no other differences were obvious with regard to the participants' word choice, which could be because the two cultures are similar, and their emotion vocabularies therefore overlap to a certain extent (see Wierzbicka 1992: 124). Moreover, the expectation that the Swiss participants, for whom the language of the conversation was an L2, would use fewer different types of emotion words was not met; the type-token ratio indicates that the English-speaking partners used a greater variety of negative emotion terms, but the Swiss participants used a greater variety of positive emotion terms, which can probably be ascribed to the Swiss partners' high proficiency in their L2. With respect to the use of suprasegmental features, previous research suggests that bicultural bilinguals may shift voice quality when they speak in their L2 (Esling 1994, in Pavlenko 2005), but also that they do not always imitate prosodic patterns in their L2 if they do not deem them natural (Pavlenko 2005). In the present study, the speakers of both mother tongues behaved similarly in this regard, though there was a stronger increase in husky/breathy/creaky voice quality and wide pitch in the emotional context among the Swiss partners compared to their overall usage. The Swiss bilinguals also used multiple stresses more often in the context of positive emotions, while the English speakers used them more often in the context of negative emotions. In addition, the Swiss participants used more falling terminal pitch movements than the Anglophone participants in all three contexts, with a more marked difference in the positive context, while the Anglophone participants used more level pitches in the emotional contexts.

According to their own reports, all of the couples use their couple language to express negative emotions, which is exclusively English in all cases but one. The majority of the bilinguals believe that there are differences in their manner of arguing, possibly due to their cultural backgrounds, but maybe also because of their gender, upbringing or different personalities. Furthermore, there were several couples who reported that they disagree on the point at which a disagreement becomes an argument. Some of the interviewees stated that they do not mind expressing negative emotions in their non-native language, and some even expressed a preference for using their L2, as they are more accustomed to doing so. At the same time, several of the bilinguals displayed ambivalence about expressing negative emotions in their L2, and the majority of the interviewees who regularly do so even find it occasionally problematic, stating that they sometimes feel disadvantaged, or that they are frustrated because they do not find the right words to express themselves precisely, which can aggravate an argument. This reflects the reports of the bilingual partners examined by Piller (2002a), many of whom also felt they lacked "rowing competence". At times, the situation is also challenging for the native speakers in this study, who are not only faced with misunderstandings, but some of whom also feel guilty about their partner's struggle.

Even if the bilinguals sometimes find expressing emotions in their L2 difficult, though, they do not switch to their mother tongue. In this regard, the couples in this study seem to differ from other bilinguals, for whom the L1 has often been found to be the preferred language for expressing emotions (Dewaele 2006: 146). Thus, several studies have suggested that

in heated arguments partners and spouses with different L1s often revert spontane‐ ously to their respective L1s because this language choice feels more satisfying and 'natural' (Piller, 2002; Pavlenko, 2005; Dewaele, 2006). This may happen even if the partner has a weak knowledge of the speaker's L1, or no understanding of it at all. (Pavlenko 2008: 159)

In contrast to these bilinguals, for the interviewees in this study it is self-evident that they express negative emotions in the relationship language, even if their partner understands their L1, and none of them consider doing so unnatural.

In order to express positive emotions, the couples also use their couple languages for the most part, but there are some bilinguals who occasionally use their respective other language for this purpose; this is more commonly the non-native speaker of the couple language. Unlike in the case of negative emotions, none of the Swiss participants voiced a preference or need for expressing positive emotions in their L1, and several of them even stated that doing this in their L2 is easier, because they view English as a very creative or free language, or because they are used to it. Some bilinguals consider using their L2 advantageous in certain situations, as it is a less emotional language for them, while for others, their L2 has become a language with a high emotional attachment, and some even expressed a desire to use this language with their children (even though they believe that they should not do so). In fact, while the level of L2 emotionality varies among the interviewees, it is emotionally charged for all bilinguals for whom it is a relationship language. This reflects previous research, which has found that bilinguals who use their L2 in a romantic relationship do not necessarily view their L2 as a language of detachment, but may develop a high level of emotionality even if they are late L2 learners (Pavlenko 2004; Piller 2002a). As a consequence, it seems that partners in a bilingual relationship usually do not mind expressing positive emotions in their L2 (Piller 2002a; Dewaele 2010), while other multilinguals tend to prefer their L1 to do so, even if their L2 is their principal means of communication (Pavlenko 2008).

With regard to the bilinguals' language choice, the use of terms of endearment stands out, as the bilinguals use (Swiss) German much more frequently to nickname each other than they do otherwise. Several of the couples reported that they use made-up or blended expressions, usually with a Swiss German phonology. Various interviewees emphasized how much they enjoy terms of endearment in their L2, and admitted that they are much more critical of (and more easily embarrassed by) nicknames in their own mother tongues; similar feelings were reported by the bilingual partners in studies by Piller (2002a) and Piller and Takahashi (2006). In addition, the interviews demonstrated that the different connotations a term of endearment carries for L1 and L2 speakers can lead to misinterpretations.

In conclusion, this chapter has shown that expressing emotions, both positive and negative, can pose difficulties for bilingual couples, but that benefits may also be derived from expressing emotions in a non-native language. Especially when it comes to expressing positive emotions, the perceived advantages seem to far outweigh the disadvantages for the bilinguals in this study, which is probably due to their fluency in their L2. This is supported by the fact that there was little difference in the emotional expression of L1 and L2 speakers during the interview, in terms of the number and the variety of expressions they used. With regard to suprasegmental features accompanying the expression of emotions, further research is necessary in order to confirm the observations made in this study, as definitive statements are difficult to make due to the multitude of factors influencing the use of suprasegmental features, as well as the size of the sample. Nevertheless, in analysing the expression of emotions in a context in which they actually occur, and in a natural rather than in a laboratory setting, this study contributes further to the area of emotion research.

## **9 "You're not gonna say this word!": Swearing**

## **9.1 Introduction**

For many people, swearing is an important way of expressing themselves, and an integral part of their manner of communicating with others. Swearing does not only serve to convey negative emotions, but may have a variety of causes and social functions. As Beers Fägersten and Stapleton point out,

[t]he act of swearing to release emotion does not only apply to the experience of negative feelings such as anger, fear, pain, or even frustration, sadness etc. Swearwords can also be used in conjunction with decidedly positive emotional experiences, such as happiness, surprise, intimacy, or humour. Swearing can also occur neutrally, neither deliberately nor strategically, and in situations that are neither saliently positive or negative. (2017: 4)

Moreover, there are numerous variables that influence a person's swearing behaviour, including his or her gender, age, and social background. In the case of bilingual speakers, even more factors come into play, such as their proficiency in their L2, the onset of second language learning, or the level of socialisation in their L2. For bilingual or non-native speakers, swearing is often challenging, since it "requires a strong sense of what is considered appropriate within a particular speech community" (Dewaele 2017: 330), and "their knowledge of conventions surrounding the use of swearing might be incomplete or inaccurate" (2017: 331) unless they have undergone extensive socialisation in their L2.

Consequently, the use of swearwords can be an area of conflict for bilingual couples, and partners in a bilingual relationship may differ greatly in their use of swearwords as well as their attitudes towards swearing and specific swearwords. The aim of this chapter is to provide an in-depth overview of the swearing behaviour of the bilingual couples in this study. In order to do this, I will explore differences in the swearing behaviour of the partners with regard to their language choice, their selection of swearwords and the level of offensiveness of these swearwords. Moreover, I will examine the extent to which the partners have overcome their differences and assimilated their swearing behaviour to each other over time. In addition, the bilinguals' reasons for their preferences and their reactions to their partner's use of swearwords are investigated.

There are many advantages to analysing the swearing behaviour of bilingual couples, even if they may not seem to be an obvious choice for the study of swearing at first sight. For one, swearing is more commonly examined in specific social groups or age groups, yet it has rarely been studied among couples. In the case of the couples in this study, moreover, a large amount of information is available on the speakers, their relationship and the setting. This is important because the context in which a swearword is uttered is relevant, and swearing is usually an interpersonal act, even if the interlocutor is not being addressed directly. In addition, reactions to swearing can be examined in the couples' conversations, which may give an indication of covert attitudes towards swearing. Furthermore, since positive as well as negative emotions are likely to be expressed quite frequently in intimate relationships (see chapter 8, "'In Swiss German, *I lieb di*, that's strange': Expressing emotions"), the majority of the partners have probably experienced each other swearing on a number of occasions, and have already formed an opinion on the other person's swearing behaviour. Finally, in-depth interviews with two partners also present the opportunity of combining the actual use of swearwords and reactions to them with the reported use and attitudes towards swearing.

In the following, I first provide an overview of previous work on the subject of swearing, considering aspects such as linguistic taboos, swearing in an L2, as well as the influence of social and cultural background, and gender (9.2). After this, I examine the participants' use of swearwords during the interviews, considering how many swearwords they used, in which language they chose to swear, and the estimated level of offensiveness of these terms (9.3). As the interviews present a very specific setting, the bilinguals' behaviour during the interviews is compared with their reports about their swearing behaviour in their every-day lives (9.4.1). After this, the interviewees' reported language choice and their reasons for this preference are discussed (9.4.2). In the last and most extensive part of the analysis, I then consider the interviewees' reactions to their partners' use of swearwords, particularly to swearwords in their own mother tongue (9.5). Finally, the results of the analysis are summarized and compared to previous research (9.6).

#### **9.2 Previous work on swearing**

#### **9.2.1 Swearing and linguistic taboos**

Swearing has been studied within a number of academic fields, including linguistics, psychology, neurology and anthropology (Ljung 2011). While lin‐ guistic research on swearing was relatively scarce before 1960, it has become an increasingly popular area of research since then (Ljung 2011: 3). Most commonly, linguistic research on swearing focuses on the historic development of swearwords (e.g. Hughes 1991; McEnery 2006; Chirico 2014), cross-linguistic comparisons (e.g. Montagu 1967; Ljung 2011), the swearing behaviour of individuals or social groups (e.g. Rayson, Leech and Hodges 1997), attitudes towards swearwords, as well as the pragmatic functions of swearing (e.g. Jay and Janschewitz 2008).

A *swearword* is defined as "a profane word" (OED), "an offensive word" (Merriam-Webster) or a "word regarded as obscene or blasphemous" (The Chambers Dictionary). In spite of this, swearwords are rarely used with the purpose of offending. In fact, it seems that, while swearing can be an aggressive act in theory, this is seldom the case, and "most instances of swearing are con‐ versational; they are not highly emotional, confrontational, rude or aggressive" (Jay and Janschewitz 2008: 268). Instead, there are a variety of reasons why individuals swear, and swearing can fulfil numerous functions. For instance, it has been hypothesized that it serves to release tension, to express the speaker's emotional state to the listener, to convey humour, to aid story-telling and to help getting one's message across (Stapleton 2003: 28). When the reasons for using swearwords were examined by Stapleton among a group of 30 undergraduate drinking friends, humour/story-telling emerged as the most common motive for using expletives, followed by verbal emphasis and expressing anger or releasing tension (2003: 28; see also section 9.2.4, "Swearing and gender").

In addition, there are a number of situational factors that have an effect on the frequency and the level of offensiveness of the swearwords that are used:

Swearing is influenced by pragmatic (contextual) variables such as the conversational topic, the speaker-listener relationship, including gender, occupation, and status, and the social-physical setting of the communication with respect to whether the swearing takes place in a public or private location, one's jurisdiction over the location, and the level of formality of the occasion. (Jay and Janschewitz 2008: 272)

The situation is a key factor, as swearwords may be considered acceptable or even viewed positively in some situations (for instance amongst members of certain social or cultural groups),<sup>1</sup> while the same expressions may be socially taboo in others.

Taboos are an important aspect when it comes to swearing, and influence people's swearing behaviour. Words or expressions are called *taboo* when they are "considered offensive and hence avoided or prohibited by social custom" (OED). Children learn early on which words are taboo, because in the process of first-language acquisition, learning taboos often goes hand in hand with learning the language itself. Billig explains this as follows:

The parent invariably informs the child about rudeness, as an unforeseen consequence of instructing the child in the ways of politeness. Thus, the child learns what must not be spoken. Temptation is created, for, as Freud recognized, prohibitions create their own desires. In this way, the child is provided with a forbidden world of discursive pleasures […  ]. (1999: 97)

The linguistic taboos that are engrained in the child usually lead to lasting emotional connotations, in particular if they are associated with punishment (Jay, King and Duncan 2006: 124). Jay and Janschewitz propose that "one's personal experience of being punished for swearing, for example, having soap put in one's mouth, classically conditions the emotional reaction to swear words" (2008: 272), even if they found in their research that being punished for swearing as a child does not keep individuals from swearing later on in life.

#### **9.2.2 Swearing in multiple languages**

Even more factors come into play with regard to the swearing behaviour of bilinguals, and may influence their choice of expression as well as their language preference. The most obvious determinant for a bilingual's language choice for swearing is said to be the level of skills he or she has in his or her L2. This was confirmed in research by Dewaele, who studied the effects of various factors on multilinguals' language choice for swearing based on an extensive online questionnaire, and discovered "that language dominance is clearly linked to frequency of use of the L1 for swearing" (2004a: 96). Potentially connected to this is the age of the onset of language learning, which Dewaele found to be "negatively correlated with self-reported language choice for swearing for the L2" (2004a: 100) — the later bilinguals acquire a language, the less likely they are

<sup>1</sup> An example of this are the ritual insults practiced by African-American adolescents described by Labov (1972), which are rarely meant to be offensive, but are rather verbal duels that reinforce in-group solidarity.

to swear in this language. Moreover, bilinguals who learnt their L2 in a purely instructed environment are less likely to use this language for swearing than speakers who acquired it in a naturalistic context (Dewaele 2004a: 102).

The bilingual's degree of socialisation in his or her second language is also important, as "[s]trong socialisation in the L2 is linked to a more frequent choice of that language for the expression of anger" (Dewaele 2010: 118). The emotional attachment to swearwords in an L2 often takes years to fully develop, and Dewaele's research on the swearing behaviour of multilinguals "confirmed that it is often only decades of living in the LX environment that they dared to use some of the swearwords in that language, and even then only the mild ones" (2010: 218). In fact, many native and non-native speakers appear to feel that bilinguals should not use swearwords in their second language. As an example, the bilingual woman studied by Koven (2004) reported that she did not feel entitled to use swearwords in French, her second language (even though she did swear in French; see next paragraph). It seems that the privilege to swear in a language is tied to fluency in this language, and, potentially, to being a native speaker. In order to swear in a language, many speakers require a sense of legitimacy or ownership of this language, which non-native speakers may (only) gain by habitually using a language and becoming fluent, as a "proficient and frequent user of a language not only possesses the correct perception of emotional force but may also feel he/she is close enough to the in-group to dare use these powerful words" (Dewaele 2004b: 220). However, there is also evidence that, while most multilinguals prefer their L1 to voice anger, "the LX can become the preferred language for anger expression, once emotion scripts have been acquired in the process of socialisation" (Dewaele 2010: 121). Thus, Dewaele found that especially bilinguals who speak their L2 to their partner, and have thus reached a high level of L2 socialisation, often prefer their L2 for swearing (2010: 131).

Taboos also play an important role when it comes to bilinguals' swearing behaviour. As Pavlenko notes, taboos become emotionally charged when they are acquired in the process of first language socialisation. In a language learnt later in life, on the other hand, "the speaker may escape strict socialisation common for L1 acquisition and thus experience less arousal in response to L2 words and expressions" (2004: 183). As a consequence, multilinguals often feel that swearwords in their L1 have a greater emotional force (Dewaele 2004a, 2004b, 2010: 156). Because some words are felt to be very strong in one's mother tongue, they are avoided by many native speakers. At the same time, "[t]he smaller emotional impact of L2 ST-words [swearwords or taboo words] on the speaker means that they can be used more freely" (Dewaele 2004a: 95). As Pavlenko puts it, "[t]he language learned later in life […  ] allows speakers to use taboo and swearwords, avoiding the feelings of guilt and discomfort internalized in childhood with regard to these expressions" (2005: 136). Because the L2 is associated with fewer taboos, it may become the bilingual's preferred language for swearing. This can be seen in the example of a bilingual woman studied by Koven, who reported that she does not use taboo words in Portuguese (her home language and the language that she acquired first), but uses them more frequently and liberally in her second language, French (2004: 492).

Of course, this does not mean that bilinguals do not experience any taboos in their L2, or that they are not at least conscious of the fact that native speakers sense such taboos in connection to certain swearwords. Fluent bilinguals are usually aware that swearing is both culture- and language-specific, particularly if they acquired their L2 in a natural or mixed environment (Dewaele 2004a: 98). This may explain why Jay and Janschewitz did not find any difference between the offensiveness ratings of swearwords of native and proficient non-native speakers (2008: 280). In a study of the perceived offensiveness of swearwords among multilinguals, Dewaele (2016) found that LX speakers tend to overestimate the offensiveness of swearwords, with the exception of the most offensive terms. Nonetheless, it should be noted that proficient speakers such as the Swiss partners in this study may well be aware of the offensiveness of swearwords in theory, but it is possible that they are still inclined to use the terms more readily, because they themselves do not feel the full emotional force of the taboos.

#### **9.2.3 Swearing in different social and cultural groups**

Even people who share the same mother tongue may show considerable differences in their swearing behaviour, as the use of and attitudes towards swearwords can differ between cultural and social groups. It has been observed repeatedly that members of lower socio-economic classes tend to swear more than members of higher classes (Rayson, Leech and Hodges 1997; McEnery 2006: 45), and that young people swear more than older ones (Rayson, Leech and Hodges 1997; McEnery and Xiao 2004; McEnery 2006: 40; Thelwall 2008). For a long time, it was almost seen as self-evident that "the distribution of taboo vocabulary is highly significant along all three dimensions of gender, age and social group, demonstrating (for those who needed it) that the archetypal user of swearwords is to be found among male speakers in the social range C2/D/E [working class] under the age of 35" (Rayson, Leech and Hodges 1997: 144). Growing older, but also having children of one's own, is likely to reduce swearing, as parents may attempt to set a good example for their children (Holmes 1992: 183; McEnery and Xiao 2004: 241). Moreover, swearing behaviour can be influenced by an individual's character type. In an analysis of self-reported swearing behaviour by 2347 L1 and LX users of English, Dewaele found that participants who had high scores in the areas of extroversion, neuroticism, as well as psychotism, reported more swearing than those with lower scores (2017: 330).

There also appear to be considerable differences in the swearing behaviour of residents of different Anglophone countries. The USA seems to be the English-speaking country where people swear the least, which has been attributed to the historical development of the United States (Hughes 2006: 13). British people are generally thought to swear more and use more offensive swearwords than Americans, which has been commented on in a number of online blogs and newspaper articles (e.g. Moore 2015). However, to my knowledge, there has been no research confirming such a trend, and in a recent study comparing the self-reported use of swearwords of American and British respondents, no such tendency was found (Dewaele 2015). Some take the view that, while swearing used to be less common in the United States, it has become more accepted in recent years. Chirico examined the use of swearwords in areas such as literature, the media, and music in the USA, and concluded that the country is presently "experiencing a linguistic, and therefore cultural, shift that is passively opening up to an amplified inclusion of profanity" (2014: 30). In contrast, swearing is traditionally more common in Australia, and of all varieties of English, "the Australian is most noted for the liberal use of swearing and profane language" (Hughes 2006: 13). According to Hughes, this is "a reflection of the nature of the founding population, which was made up principally of 160,000 convicts, very unlike the Pilgrim Fathers of the United States" (2006: 13). His observation is supported by Wierzbicka, who postulates that swearing is more accepted in Australia than in other English-speaking nations, as for instance the four B-words — *bullshit, bastard, bugger,* and *bloody* — are part of everyday language. In Wierzbicka's opinion, these swearwords "are felt to be an important means of self-expression, self-identification, and effective communication with others" (1997: 217). Her view is shared by others (e.g. Green 2012), and it has been suggested that swearing is an essential part of the colloquial speech of Australian men as well as women (Power and Walker 2012). According to popular opinion, the Irish seem to be a close second if not on a par with the Australians — when it comes to swearing (Rennicks 2013). However, as far as I am aware, there are no empirical studies on the frequency of swearing in all of these Anglophone countries, which could either confirm or refute these stereotypes.

Of course, the difference might be even greater among people who do not share the same mother tongue or whose cultures are very different. While there has been no comparative research on the subject, in my personal opinion, many Swiss have a habit of swearing rather freely and frequently, but the majority of them use relatively mild swearwords, or expressions that have lost their original offensiveness. Aman, a German philologist and researcher of swearwords, takes the view that the Swiss are "average" when it comes to swearing (Kern 1996). It is also relatively common in the German-speaking part of Switzerland, especially among younger people, to use English swear‐ words such as *shit* or *fuck* (Hafner 2017; Müller 2013). My own observation is that there is a tendency for English-speaking parents to try to ingrain linguistic taboos more strongly in their children than Swiss parents do; for example, the practice of washing children's mouths out with soap (or even just the threat of doing this) does not, to my knowledge, exist in Switzerland, while 20% of the respondents to Jay, King and Duncan's (2006) questionnaire on punishment for swearing stated that they were punished in this manner when they were children.

#### **9.2.4 Swearing and gender**

An individual's swearing behaviour can also be influenced by his or her gender. It has been noted repeatedly, especially in older studies, that women swear less frequently than men and that they use weaker swearwords (Lakoff 1975; Holmes 1992; Coates 2004; Rayson, Leech and Hodges 1997), although some believe that the difference may not be as great as it is traditionally thought to be (de Klerk 1990, 1991). More recent studies seem to have produced somewhat conflicting results, depending on which culture, age group, and social background was analysed in connection with gender. In a study based on the British National Corpus, McEnery (2006: 47) found that women in some age groups used a greater number of mild swearwords (*bloody, shit*) than men, whereas the latter used strong swearwords more frequently (especially the term *fuck*). Similar results were obtained by Gauthier and Guille, who studied the genders' use of swearwords in tweets in the UK. They found that, regardless of the age of the users, strong swearwords (*cunt* and *fuck*) were used far more frequently by men, while women used the term *bitch* significantly more often than men (2017: 154). Examining the use of swearwords among university students in New Zealand, in contrast, Bayard and Krishnayya (2001) discovered that the male students in their study swore slightly more than the female students, but that there was little difference with regard to the strength of the swearwords that were used. Moreover, the male participants used fewer very mild swearwords, and reduced their use of swearwords to a greater extent than the female participants did in more structured (or formal) contexts. Based on an analysis of the use of swearwords in narratives about punishments for swearing, Jay, King and Duncan reported that female American college students produced as many swearwords as their male counterparts in their accounts (2006: 127), though it may be relevant that these words were elicited in a metalinguistic context. In a study of taboo words based on the Bergen Corpus of London Teenager Language, Stenström (1995) identified no gender differences among British teenagers, but a tendency for adult women to use weaker taboo words than men (in Dewaele 2004b: 206), which highlights the importance of the age factor when it comes to gender-specific swearing behaviour.

It appears that a person's cultural background can play a particularly impor‐ tant role in correlation with gender, as the conventions and expectations for the use of swearwords of men and women tend to depend on the specific culture. Analysing the swearing behaviour of members of the social networking site *MySpace* in the US and the UK, Thelwall (2008) detected no significant gender difference in the use of strong swearwords in the UK. However, he found that in the US, male participants used a significantly higher number of strong swear‐ words than female users (2008: 101). It is possible that the *ladette* movement<sup>2</sup> in the UK has led to increased swearing by females in recent years (Jackson 2006: 353; Thelwall 2008: 102). Swearing is often linked to ideas of femininity prevalent in a culture or subculture. This was demonstrated by Stapleton, who examined the practice of swearing within a group of 30 Irish undergraduates who regularly met in a pub. On the one hand, she found that women considered a greater number of terms offensive or obscene, and refrained from using them, and that the majority of the students deemed certain expressions "more acceptable for men than for women" (2003: 24). On the other hand, the author observed that, "[c]ontrary to common sociocultural conceptions, both female and male participants in this community report[ed] habitually deploying strong language […  ]. In doing so, they constitute[d] themselves as members of the particular 'drinking' culture in which they [were] engaged […  ]" (2003: 32). Thus, it appears that, "through the strategic use of 'obscenity', the women in [Stapleton's] study [were] able to forge community-specific versions of femininity, in which they

<sup>2</sup> A ladette is a "young woman characterized by her enjoyment of social drinking, sport, or other activities typically considered to be male-oriented, and often by attitudes or behaviour regarded as irresponsible or brash" (OED, s.v. Ladette, n.), such as swearing.

constitute[d] themselves […  ] as 'drinking women'" (2003: 32). All of these findings suggest that gender is not an independent determinant of swearing behaviour, but is interrelated with other factors such as the specific culture, age, social class, and possibly the medium.

The role of gender in swearing by non-native speakers has not been re‐ searched extensively. Register (1996) studied the reported use of English taboo words by non-native learners of English at university level in the USA and found that male learners used more swearwords than female learners. However, it should be noted that it is likely that this depends to a great extent on the country of origin of the non-native speaker, and its cultural conventions around swearing for both genders. Jay and Janschewitz, who questioned native and non-native speakers on their perception of the offensiveness of swearwords, found that female native speakers provided higher offensiveness ratings than male native speakers, but detected no significant gender differences among the non-native speakers (2008: 285).

Finally, there seem to be cultural differences in the perception of the appro‐ priateness of swearing in the presence of women. I tentatively hypothesize that swearing by women as well as in the company of women is perceived more negatively in most English-speaking cultures than in Switzerland, at least among the older generations. As Hughes posits, it is "[t]he generally accepted traditional norm in English-speaking societies […  ] that it is highly impolite to swear in the presence of a woman and unacceptable or taboo for a woman to swear at all" (2006: 502). In contrast, I personally believe that Swiss men would not usually attempt to avoid swearing because "there are ladies present" (although they might swear more if they are among themselves). In my opinion, swearing is not generally perceived to be unladylike in German-speaking Switzerland, and swearing is not deemed something from which a woman's ears would have to be shielded.

#### **9.2.5 Swearing in bilingual couples**

Few studies on swearing have focused on bilingual couples, even though the differences in cultural and linguistic background in combination with gender make it a complex and potentially challenging area to negotiate in such relationships. For instance, differences in the perception of and emotional reaction to swearwords may lead to misunderstandings or conflicts among bilingual couples. Since the non-native partner does not have the same taboos attached to swearwords as the native speaker does, the former may use these terms more freely, which may be offensive to his or her partner. This can be observed in Piller and Takahashi's (2006) study, in which Japanese women reported being offended by their English-speaking partners' use of Japanese dirty talk. In addition, the inappropriate or incorrect use of swearwords is also a potential source of embarrassment for the non-native partner (Dewaele 2004b: 205).

To summarize, the main factors that determine how much bilinguals swear in each of their languages seem to be (a) their level of proficiency in each language, (b) the onset of language learning, (c) the environment and attitudes with regard to swearing in both cultures, (d) their level of socialisation in the L2, (e) childhood taboos and the perceived emotional force of swearwords, (f) their gender, (g) their age and family situation, (h) their social background, as well as (i) the sense of ownership of the languages. Furthermore, in the case of bilingual couples, (j) their partners' swearing behaviour can be expected to influence the bilinguals' own behaviour, too. These factors interact and might enhance or inhibit each other; for instance, bilinguals are more likely to swear in the language in which they are more proficient, yet it is likely that this is also the language in which they sense taboos most strongly.

#### **9.3 The couples' swearing during the interviews**

#### **9.3.1 Swearing in an interview situation**

Before turning to the analysis of the couples' swearing behaviour during the interviews, some comments should be made on the setting. The interviews were not specifically designed to examine how much couples swear among themselves in a natural environment. Despite the intentionally familiar loca‐ tion of the interviews, the setting can be considered semi-formal, which did not particularly encourage swearing; neither did the majority of the topics of the interviews. Moreover, a third party was present, which may have had an in‐ fluence on how readily the partners used swearwords. However, the partners' swearing behaviour during the interviews, in particular the language in which they cursed and the level of offensiveness of the swearwords they uttered, can still give some indication of their use of and attitude towards swearwords. Due to differences in age as well as social and cultural backgrounds of the bilinguals, I expected their use of swearwords and their attitudes towards swearing to vary considerably. While the number of subjects in this study is too small to allow for any general statements on the swearing behaviour of bilingual couples, certain tendencies may be revealed nonetheless, for instance regarding the extent to which the partners have assimilated their swearing behaviour to each other.

#### **9.3.2 Categorisation of swearwords and overview**

The overview of previous work has indicated that swearing is often examined in particular social groups, and most commonly in informal contexts, as it tends to be especially frequent in these situations. By contrast, the present study examines the swearing behaviour of dyads in a semi-formal situation, which is likely to yield different results. My aim is to explore the couples' swearing behaviour in this specific setting with regard to (a) their frequency of swearing, (b) their choice of swearwords and the level of offensiveness of their expressions, (c) their language choice, (d) the degree of assimilation in the partners' swearing behaviour, as well as (e) the influence of factors such as their gender and mother tongue on their swearing behaviour. In addition, I intend to compare my observations with the couples' own reports (see section 9.4, "The couples' reports on their swearing behaviour").

In order to investigate these areas, the swearwords that the participants used were divided into four categories in each language, according to their level of offensiveness: (a) "very mild" (the only expressions used in the interviews were the mild oaths *oh (my) God* and *good God,* which many people might not even consider swearwords; no comparable Swiss German expressions were used); (b) "quite mild" (expletive interjections such as *bloody*, *shit*, *damn*, *piss,* as well as expressions including these words like *pissed off;* or similar Swiss German expressions like *Scheisse* 'shit', *dami* 'damn (me)' or *huere*, a common intensifier originating from the term *Huere* 'whore', which it is now largely devoid of meaning [Lüscher 2008; Landolt 2016]); (c) "quite offensive" (potentially "directive abuse" [Millwood-Hargrave 2000: 8] such as *bitch, whore, dick,* or *Arschloch* 'asshole'); and (d) "very offensive" (the obscene words *fuck* and *cunt*, 3 and the corresponding Swiss German terms

<sup>3</sup> These expressions were grouped together even though the term *cunt* is generally regarded as more offensive than *fuck.* This appears to be the case in America in particular, as the term *cunt*, albeit still considered vulgar, seems to have lost its shock value in the UK in recent years, and is used more frequently in the UK (Dewaele 2015: 330). This development has been commented on in a number of articles and blogs (e.g. Fogg 2014a, 2014b). It should also be noted that the term *fuck* is perceived as offensive in English-speaking countries, but not to the same extent in Switzerland, where the English term is used by younger people. In my opinion, the term *fuck* is considerably less offensive to a Swiss speaker than the German equivalent *ficken* 'fuck'.

*figge* and *Fotze*). It should be noted that none of these expressions was used as a personal insult to another person present, which would have added another level of offensiveness to the categorisation. Euphemisms (substitutions like *f off, the c-word,* as well as minced oaths<sup>4</sup> like *oh goodness*) were included in the tables below, but not counted in the total number of swearwords. A table of all the swearwords and euphemistic expressions that appeared in the interviews, as well as their categorisation, can be found in Appendix V (Table 51, "Categorisation of swearwords").

These four categories are based on a study on the offensiveness of swear‐ words in English (Beers Fägersten 2007) and two public opinion research reports undertaken by broadcasting corporations (Millwood-Hargrave 2000; Broadcasting Standards Authority 2010). Since there are no similar studies for Swiss German, the Swiss German swearwords were divided into categories in a similar manner as the English ones, while also taking into account my own perception of their level of offensiveness. However, it should be noted that the aforementioned surveys indicate that there are considerable differences between English-speaking countries in the perceived level of offensiveness of some swearwords, which was also confirmed by Dewaele's (2015) comparison of the offensiveness of some swearwords in the USA and the UK. Such differences can be seen in an anecdote related by Horan, who reports finding a colleague's use of *oh shit* "mildly shocking", while she finds the term *fuck* far less offensive; she attributes this to her Irish background (2013: 294).

Moreover, when categorizing swearwords, one should bear in mind that the context in which a word is uttered can be crucial, as well as who or what it is said to or about. As Beers Fägersten points out, "offensiveness ratings of isolated words are unreliable, since it is impossible to know how a rating task participant interprets the individual words" in context (2007:19). To give an example, an expression carries a different weight if it is used as a general expletive, or as an insult directed at someone. Apparently, swearwords are seen as less offensive if they are felt to be "*contextually appropriate*" (Stapleton 2010: 291, emphasis in original). In addition, some swearwords have a range of potential functions<sup>5</sup> and it could be argued that certain variants of the same swearword (like the

<sup>4</sup> A minced oath is "a clipped or euphemistically altered oath" (OED, s.v. Minced oath, n.); Montagu calls this *euphemistic swearing*, "a form of swearing in which mild, vague or corrupted expressions are substituted for the original strong ones" (1967: 105).

<sup>5</sup> See e. g. McEnery (2006) for a detailed classification of fifteen linguistically distinct manners of swearing. Due to the relatively small number of offensive swearwords, many of which were used metalinguistically, I decided not to classify the terms according to their function in this study.

cursing expletive *fuck you)* are far more offensive than others (an idiomatic set phrase such as *fuck all* or an adverbial booster like in *fucking awesome*). To accommodate the element of context to some extent, I divided each category into swearwords that were used spontaneously (e.g. "and all you read is about bloody Jordan" [Simon SG 1103]), and swearwords that were used metalinguistically, i. e. when someone talked about swearing (e.g. "[some words] have a different meaning in ^^English, like for example *bitch*" [Sarah SG 1190]).

Table 29 below shows how many swearwords each speaker used, in which language these terms were uttered, their level of offensiveness, and whether they were used spontaneously ("s") or mentioned in a metalinguistic context ("m"). In addition, Table 30 gives an overview of the participants' spontaneous use of swearwords relative to the number of words they used during the interviews (normalized per 10,000 words). Table 29 shows that 120 swearwords were uttered during the interviews; half of them were used spontaneously, the other half in a metalinguistic context. Overall, the participants did not swear a lot; on average, each person uttered 3.0 swearwords spontaneously [1/20 of 60], which equates to about 5.82 swearwords per 10,000 words. There was only one instance in which a Swiss German swearword was used spontaneously compared to 59 expressions in English — which is not surprising since the base language of the interviews was English. While 21 euphemistic expressions were uttered in English in both contexts, no euphemisms were used in Swiss German in either context, which may be due to the fact that, although minced oaths exist in Swiss German, for example *gopf* or *gopfridstutz* instead of *gopferdami* (originally *Gott verdamme mich* 'God damn me' [Landolt 2015]), there are no corresponding terms to expressions such as *the f-word* or *a four-letter word* in Swiss German. If a speaker wants to refer to a specific swearword in Swiss German in a metalinguistic context, he or she has no choice but to explicitly utter this word.


Table 29: Use of swearwords for each participant, including language, level of offensive‐ ness, and context

Key: E = English term; G = Standard or Swiss German term; m = metalinguistic context; s = spontaneous use. \* Euphemisms were not included in the total number of swearwords and are therefore in italics.


Table 30: Use of spontaneous swearwords for each participant (normalized per 10,000 words)

Key: bold = in text. \* Euphemisms were not included in the total number of swearwords and are therefore in italics.

Furthermore, Tables 29 and 30 demonstrate that there are considerable differ‐ ences in the interviewees' swearing behaviour. While some of the participants did not use any swearwords (Deborah SG, Philipp SG, Sophia SG, Craig <sup>E</sup> , Karen <sup>E</sup> ), used only mild ones or used them only in a metalinguistic context (Stephanie SG, Susanne SG, Joshua <sup>E</sup> ), there were some who used swearwords relatively liberally (Monika SG, Simon SG, Claire <sup>E</sup> , Robert <sup>E</sup> , Dean <sup>E</sup> , David <sup>E</sup> ). Yet another group of interviewees showed moderate swearing behaviour (Sarah SG, Martin SG, Katia SG, Courtney <sup>E</sup> , Richard <sup>E</sup> , Tim <sup>E</sup> ). Some very offensive swearwords were uttered during the interviews, but mostly in a metalinguistic context (20 out of 23 [E] and 5 out of 5 [G] were used metalinguistically).

It is remarkable that many of the couples displayed fairly homogenous swearing behaviour, with both partners using a similar number and type of swearwords. Four of the couples did not swear at all or swore only very little (Deborah/Joshua, Sophia/Richard, Katia/Craig, Karen/Philipp); two of them swore occasionally (Courtney/Martin, Sarah/Tim), and two of them swore rather frequently (Monika/Dean, Claire/Simon). Only for two couples (Ste‐ phanie/Robert, Susanne/Richard) was there a considerable gap in the number of swearwords used: Both Stephanie SG and Susanne SG did not use any swearwords spontaneously, whereas their partners swore on a number of occasions. It may be relevant that these two were among the couples who had been in a relationship for the shortest period of time. This suggests that over time, the swearing behaviour of couples converges, at least while they are in each other's company.

#### **9.3.3 Swearing and gender**

On the basis of previous research, the female partners were expected to use fewer and less offensive swearwords than the male partners. In order to test this supposition, the genders' use of swearwords was examined with regard to the context, language choice, and level of offensiveness. An overview is provided in Table 31 below, which indicates the number of swearwords in relation to the average number of words the speakers used during the interview (normalized per 10,000 words). In addition, the use of swearwords of both genders is visualized in Figure 6.



very offensive E – s 0.00 1.29 0.00 0.25 **0.00** very offensive E – m 2.09 1.94 2.79 1.51 2.30 very offensive G – s 0.00 0.00 0.00 0.00 0.00

total mild – s \* 2.10 3.88 12.57 5.02 5.23 total mild – m \* 1.20 1.29 3.50 3.01 1.89

Table 31: Use of swearwords for each gender and each combination of language and gender (normalized per 10,000 words) 10,000 words)

Key: E = English; SG = Swiss German; G = Standard or Swiss German; m = metalinguistic context; s = spontaneous use; bold = in text. \* Euphemisms were not included in the total number of swearwords and are therefore in italics. Key: E = English; SG = Swiss German; G = Standard or Swiss German; m = metalinguistic context; s = spontaneous use; bold = in text. \* Euphemisms were not included in the total number of swearwords and are therefore in italics.

**Figure 6.** Use of swearwords for each gender (normalized per 10,000 words) Figure 6: Use of swearwords for each gender (normalized per 10,000 words)

159 Table 31 and Figure 6 demonstrate that the male participants used slightly more swearwords spontaneously than the female participants did (5.97 [5.97 + 0.00] vs. 5.65 [5.44 + 0.21] per 10,000 words), and the expressions they used spontaneously were considered more offensive (0.72 vs. 0.42 "quite offensive", and 0.54 vs. 0.00 "very offensive"). The tendency for the male bilinguals to use more offensive spontaneous swearwords is also visible in the individual groups including the variable of the participants' mother tongue (3.23 vs. 0.79, and 0.50 vs. 0.30). In contrast, the female participants in both couple combinations used far more very mild swearwords spontaneously than their partners (0.90 vs. 0.25, and 3.49 vs. 0.65) and more euphemisms (1.79 vs. 1.00, and 4.89 vs. 1.94). Interestingly, the female bilinguals also used more offensive swearwords in the metalinguistic context than their respective partners (4.48 vs. 2.51, and 6.29 vs. 1.94). While none of the women used very offensive swearwords in spontaneous speech, many of them thus did not seem to mind using potentially offensive expressions when speaking about swearing and swearwords, neither in their L1 nor their L2. Conversely, some male participants also used swearwords that are considered very offensive (variants of *fuck*) in spontaneous speech.

#### **9.3.4 Swearing and mother tongue**

In addition to gender, it was anticipated that the mother tongue of the partici‐ pants might have an effect on their use of swearwords. Table 32 demonstrates the correlation between the speakers' mother tongues and their choice of swearwords, with regard to the level of offensiveness, context and language (normalized per 10,000 words). To facilitate the interpretation of these results, they are visualized in Figure 7.



Table 32: Use of swearwords for each mother tongue and each combination of language and gender (normalized per 10,000 words)

Key: E = English; SG = Swiss German; G = Standard or Swiss German; m = metalinguistic context; s = spontaneous use; bold = in text. \* Euphemisms were not included in the total number of swearwords and are therefore in italics. Key: E = English; SG = Swiss German; G = Standard or Swiss German; m = metalinguistic context; s = spontaneous use; bold = in text. \* Euphemisms were not included in the total number of swearwords and are therefore in italics.

**Figure 7.** Use of swearwords for speakers of each mother tongue (normalized per 10,000 words) Figure 7: Use of swearwords for speakers of each mother tongue (normalized per 10,000 words)

As can be seen in Table 32 and Figure 7, the native speakers of English used more swearwords overall, and especially more spontaneous ones, than the Swiss participants. In particular, the English speakers used a greater number of spontaneous swearwords that were categorized as mild (on average 7.02 vs. 2.65 instances per 10,000 words), and this trend also persists in the individual groups including the variable of gender (12.57 vs. 3.88, and 5.02 vs. 2.10). In addition, the Anglophone participants referred to more mild swearwords in a metalinguistic context (3.14 vs. 1.22), also in both couple combinations (3.50 vs. 1.29, and 3.01 vs. 1.20), and used some euphemisms spontaneously, which none of the Swiss did (1.85 instances, all in English). Conversely, the Swiss As can be seen in Table 32 and Figure 7, the native speakers of English used more swearwords overall, and especially more spontaneous ones, than the Swiss participants. In particular, the English speakers used a greater number of spontaneous swearwords that were categorized as mild (on average 7.02 vs. 2.65 instances per 10,000 words), and this trend also persists in the individual groups including the variable of gender (12.57 vs. 3.88, and 5.02 vs. 2.10). In addition, the Anglophone participants referred to more mild swearwords in a metalinguistic context (3.14 vs. 1.22), also in both couple combinations (3.50 vs. 1.29, and 3.01 vs. 1.20), and used some euphemisms spontaneously, which none of the Swiss did (1.85 instances, all in English). Conversely, the Swiss participants spontaneously used a greater number of quite offensive terms overall than the Anglophone participants (0.82 vs. 0.37 instances), as well as very offensive terms (0.41 vs. 0.18; all in English), though this tendency is not confirmed in the two couple combinations. While there was only one instance in which a Swiss German swearword was used spontaneously, a number of Swiss German swearwords (including some offensive ones) were mentioned in the metalinguistic context. Interestingly, the English-speaking partners made

reference to more than twice as many Swiss German swearwords than the Swiss interviewees in the metalinguistic context (3.14 vs. 1.43), and to more offensive ones, which is also the case for the two couple combinations. Thus, it is noteworthy that, in both languages, fewer offensive swearwords tended to be used by native than non-native speakers of the respective language (in both contexts), which is probably due to stronger taboos attached to such swearwords in the case of native speakers.

#### **9.3.5 Other factors influencing swearing**

Besides gender and native language, the cultural background of the bilinguals might also have an influence on their use of swearwords. An overview of the use of swearwords by participants with each cultural background is given in Table 33 and Figure 8 below (normalized per 10,000 words). It should be noted that some of these groups comprise only few members — there were only three Americans and one Australian among the interviewees — so that their figures only give some indication of potential tendencies.


Table 33: Use of swearwords for each cultural background (normalized per 10,000 words) Key: E = English; SG = Swiss German; G = Standard or Swiss German; m = metalinguistic context; s = spontaneous use; bold = in text. \* Euphemisms were not included in the total number of swearwords and are therefore in italics.

**Table 33.** Use of swearwords for each cultural background (normalized per 10,000 words)

Key: E = English; SG = Swiss German; G = Standard or Swiss German; m = metalinguistic context; s = spontane-

**Figure 8**. Use of swearwords based on cultural background (normalized per 10,000 words) Figure 8: Use of swearwords based on cultural background (normalized per 10,000 words)

162 On average, the participants from the United Kingdom deployed by far the greatest number of spontaneous words (11.63 per 10,000 words), while the Australian and the Swiss interviewees used less than a third as many swearwords (3.13 [3.13 + 0.00] and 3.88 [3.68 + 0.20]). None of the Americans used any swearwords spontaneously during the interviews, which may be linked to the fact that they all happen to have a religious background of them are practicing Christians. Indeed, religious groups have been found to avoid using conventional swearwords, and to use milder or euphemistic forms instead (Dutton 2007), which also seems to apply to these three participants. In the case of the interviewees in this study, it cannot be determined whether the main reason for the absence of swearing is their religiousness or their cultural background. On average, the participants from the United Kingdom deployed by far the greatest number of spontaneous swearwords (11.63 per 10,000 words), while the Australian and the Swiss interviewees used less than a third as many swearwords (3.13 [3.13 + 0.00] and 3.88 [3.68 + 0.20]). None of the Americans used any swearwords spontaneously during the interviews, which may be linked to the fact that they all happen to have a religious background; two of them are practicing Christians. Indeed, religious groups have been found to avoid using conventional swearwords, and to use milder or euphemistic forms instead (Dutton 2007), which also seems to apply to these three participants. In the case of the interviewees in this study, it cannot be determined whether the main reason for the absence of swearing is their religiousness or their cultural background.

The speakers' age and family situation also appear to play a role. As can be seen in Table 34, participants aged 30 and over swore less frequently than those younger than 30 (on average 3.79 vs. 6.83 spontaneous swearwords per 10,000 words).


Table 34: Use of swearwords for both age groups and family situations (normalized per 10,000 words)

Key: <30 = younger than 30 years, 30+ = 30 and older; m = metalinguistic context; s = spontaneous use.

Having children of one's own appears to have an even stronger effect: The twelve participants who had no children at the time of the interview used more than 12 times as many spontaneous swearwords than those who did (8.11 vs. 0.63 swearwords per 10,000 words). In fact, the interviewees with children uttered only 2 swearwords out of the 60 swearwords that were used spontaneously during the interviews. It is possible that bilinguals who have a family are more used to adjusting their swearing behaviour to a specific situation. It also needs to be pointed out that 3 of the 4 couples with children in this study include an American. The elements of the religious and cultural background as well as having children coincide in their case, and the most influential factor cannot be determined with certainty. It is remarkable, however, that the Swiss partners of the Americans — one of whom does not have any religious affiliation — used hardly any swearwords either, and neither did the fourth couple who have a child.

## **9.4 The couples' reports on their swearing behaviour**

#### **9.4.1 Frequency of swearing**

As the interviews involved a very specific setting, and the language of the interview was largely pre-set, I asked the bilinguals about their swearing habits in their everyday lives in order to complement the analysis of their swearing behaviour during the interviews. Their reports confirm the observation that their swearing behaviour is very diverse, as 6 of the interviewees stated that they never or rarely swear; 6 reported that they swear sometimes, and 8 indicated that they swear (quite) often (see Table 35).

Table 35: Reported swearing behaviour of each speaker

The participants' reports are largely congruent with their swearing behaviour during the interviews. The three Americans claimed to swear rarely or not at all, while most British participants stated that they swear (quite) often, or at least sometimes, which is in accord with the observations made during the interviews. Three men and three women claimed that they swear very little, but there are more men (n=6) who reportedly swear (quite) frequently than women (n=2). This corresponds in most cases with the participants' swearing behaviour during the interviews, with the notable exception of Tim <sup>E</sup> and Richard <sup>E</sup> , who swore rather little during the interviews, but indicated that they swear more in everyday speech. They may have used fewer swearwords than they would in other situations due to the semi-formal setting, the company of their partner, or my presence during the interviews. It is also conceivable that they professed to swear more than they actually do because they attribute some covert prestige to swearing. Thus, Tim said that he does not usually swear in the company of his girlfriend, but if he were with his friends, "being Australian", he would "swear [his] heart out" (1150–1152). The fact that Tim insists on only swearing a lot when he is in the presence of other Australians may be linked to Wierzbicka's (1997: 217) observation that swearing is a part of the Australian self-image and suggests that it could be a cultural performance.

#### **9.4.2 Language choice**

#### *Overview*

The couples were also questioned regarding their language preference for swearing as well as the reasons for their preference. In terms of their language choice, the accounts of the bilinguals were rather heterogeneous (Table 36). According to their self-reports, 6 of the interviewees mainly swear in their L1, 5 swear in their L2 sometimes, 6 swear in both languages equally, while 1 swears more in her L2, and 2 swear in neither of their languages. No-one claimed to have a strong preference for swearing in his or her L2.


Table 36: Reported language choice for swearing of each speaker Key: E = English; SG = Swiss German.

The swearing behaviour of the participants during the interviews (9.3, "The couples' swearing during the interviews") suggests that the language of the conversation is a very important factor in determining the language of swear‐ words, since there was only one instance in which a swearword in Swiss German was uttered, and this occurred during a short exchange spoken in Swiss German. Thus, all 60 swearwords that were used spontaneously were spoken in the language of the on-going conversation. When it comes to the intervie‐ wees' use of swearwords in general, on the other hand, the most important factor determining language choice appears to be their proficiency and/or the frequency with which both languages are spoken. Hence, the majority of the Swiss participants habitually swear in both languages — to various degrees while 4 of the Anglophone partners swear mainly in English, 4 occasionally use German for swearing, and only 1 swears in both languages. Overall, then, the Swiss speakers, all of whom use their L2 regularly and have a high level of fluency, also swear more often in their L2 than the English speakers. Among the English speakers, there are considerable differences with regard to their L2 skills, which range from weak to proficient, and, not surprisingly, the least fluent among them hardly ever swear in their L2. It is also notable that the two couples who have been in a relationship for the shortest period of time stated that they swear almost exclusively in their respective mother tongues, even though the Swiss partners are fluent in English, while fluent L2 speakers in a more long-term relationship readily swear in their L2. This confirms that the socialisation in a bilingual's L2 is indeed an important variable in his or her language choice for swearing.

In the case of the bilinguals in this study, it is difficult to determine the role of other factors that have been found to be important with regard to language preference, such as age and environment of the L2 acquisition. The Swiss speakers all started studying English at school in early adolescence, whereas most English speakers began learning (Swiss) German in their twenties. The Swiss partners hence started learning English in a formal environment, but also acquired a large part of their language skills naturally (during stays abroad and from their partner). As English is their relationship language, they have been socialized extensively in their L2, and it comes as no surprise that they have started using terms their partner uses. On the other hand, some of the English speakers attended a German language school upon taking up residence in Switzerland, but they all learnt their L2 mainly in a natural environment; yet they use it less for swearing. Finally, it should also be considered that certain English swearwords are commonly used in German-speaking Switzerland, even if the language of the conversation is Swiss German, which could have contributed to the more frequent use of L2 swearwords by the Swiss partners.

#### *Reasons for L1 preference*

The interviewees who avoid swearing in their L2 volunteered a number of reasons for their L1 preference. Some explained that they choose not to utilize their L2 because they do not speak the language well enough, or because they have been taught not to swear by their partner (Robert <sup>E</sup> , Tim <sup>E</sup> ). For one of the interviewees, the main reason is that he thinks that swearing in his L1 is "nicer", that it is "more colourful" (David <sup>E</sup> 75), and that there is a limited choice of swearwords in German ("what have you got in German; *Scheisse* ['ʃaɪze], and that's about it really, and a few other things" [David E 761]). Some bilinguals do not swear in their L2, or at least avoid saying certain words, because their partner or other native speakers told them that it bothers them personally or that it is not socially acceptable (Robert <sup>E</sup> , Claire E , Katia SG, Monika SG, Dean <sup>E</sup> ; see also section 9.5, "Reaction to partner's use of swearwords").

```
Example 9.1: Claire (Northern Irish, 28) and Simon (Swiss, 34)
Claire: <BR at the <HI ^start I HI> remember sayin' W>BR> _
              <H:> … <PP<:-) ^like this: :-)>PP> /
              <PP<@ like ^figge@n @>PP> _ (('fuck'))
              <A<PP<@<HI like figg ^this HI> figg it @>PP>A> _ 
              <A<:-) like ^kind of: too :-)>A> / 
              <LO<A and ^everybody's like A>LO> _ 
              <W<@<HI you don't HI> ^^say <HI that HI> [in German] @>W> _
Simon: [@@@] ((softly))
Claire: <@<HI you ^can't HI>@> / 
              <:-)<HI or what^ever HI>:-)> _ 
              <P<LO so then I ^didn't LO>P> / (1605–1607)
```
In Claire's case, part of the reason why people told her not to use the word *figge* 'fuck' is because she translated some expressions literally ('fuck this / it'), creating forms which do not exist in German. In addition to this linguistic issue, it is evident that Claire lacks the emotional attachment to swearwords in her L2 that a native speaker might have, which is also named by Karen <sup>E</sup> as a reason not to swear in her L2. Karen explains that she used to spend time with international students, who used swearwords without sensing an emotional attachment, which caused her to avoid swearing in her L2:

```
Example 9.2: Karen (American, 51) and Philipp (Swiss, 47)
Karen: <W when <HI I HI> was a ^young person W> \ 
               … I I hung ^out with: \ 
               <P a lot of inter^national <LO students LO>P> _ 
               <P<LO<A when I was in ^college and A>LO>P> \ 
               <H> ^one of the things I <LO noticed LO> // 
               <P<LO was ^the LO>P> _
               <H:> <P<W<MRC their <HI ^use HI> <LO of ^swearwords LO>MRC>W>P> \
               … <H> <P<CRK and the ^fact that CRK>P> \
               … e^motionally _
               <W<BR they didn't have an a^ttachment BR>W> //
               <P<HSK so ^it HSK>P> _
               <P<W<MRC didn't ^mea:n the same ^thing MRC>W>P> /~
               <P as it would <LO for ^me: LO>P> //~
Silja: = <P m^hm P> //
Karen: = <H:> and so I ^think // 
               I a^voided \\ 
               … using ^swearwords \\ 
               in another ^language _
               cause I ^knew / 
               <H:> <UP it just didn't ^^work UP> / 
               <:-)<BR<MRC somehow it ^didn't ^^work MRC>BR>:-)> // (1200–1202)
```
In addition, there are interviewees who hardly ever swear in their mother tongue because of religious or social taboos (e.g. gender norms) or because, as they put it, it is "not in [their] nature" (Joshua <sup>E</sup> 531). Consequently, they do not swear (much) in their L2 either (Stephanie SG, Courtney <sup>E</sup> , Joshua <sup>E</sup> , Deborah SG, Karen E , Philipp SG).

For some of the bilinguals, the fact that they do not feel entitled to use certain expressions (Simon SG), or that they are unable to swear competently, may keep them from swearing in their L2, or make them feel inferior when they do. Thus, some claim that, even though they are otherwise proficient L2 speakers, they lack swearing competence (Stephanie SG, Sarah SG, Martin SG). Martin SG, for instance, dislikes the fact that he cannot express anger in English as naturally or eloquently as a native speaker:

```
Example 9.3: Martin (Swiss, 26)
Martin: da- ^when I'm _
              when I'm ^angry \
              I: and ^I: _
              <P ^I: P> _
              <HLT want to .. ^s:wear HLT> /\~
              or I want to ^say something _
              ^p:roperly \\
              <H> and it's ^um: _
              … <HSK<P I ^say it in like a P>HSK> _
              <W a very .. clumsy ^way W> /~
              <HSK ^then I HSK> _
              <P<HSK I think <HI ^why should HI> I: HSK>P> \
              <P<HSK you ^know HSK>P> /~
              <P<HSK I'm ^angry I have to HSK>P> _
              … say in this ^way /\~ (644)
```
It is also worth noting that this appears to be an emotional matter for Martin; he hesitates and speaks haltingly when he explains his situation, and his voice quality changes to soft and husky. According to Martin, not being very fluent at swearing also takes away "the force … of your expression" (648), which he finds deeply frustrating. Nevertheless, he does not seem to consider using his L1 a viable alternative ("I *have to* … say in this way" [644]), which may be due to his partner's limited German skills.

#### *Reasons for L2 preference*

A number of the interviewees use both of their languages equally (Martin SG, Richard <sup>E</sup> , Simon SG, Monika SG, Katia SG, Philipp SG). In fact, many do not seem have a clear language preference, and simply use the language of the on-going conversation. Only one interviewee, Susanne SG, generally prefers to swear in her non-native language, because it does not have the same emotional force for her as her L1 has:

```
Example 9.4: David (English, 32) and Susanne (Swiss, 29)
Susanne: <HSK<HI I don't HI> really ^swear in German HSK> _
               <P<HSK because <HI ^I HI> find swearwords really HSK>P> \
               … ^^strong _
David: = @@@ [@@] ((chuckling))
Susanne: [<P<CRK ^and um CRK>P>] _
               … <PP<HSK<HI when I HI> use English ^swearwords HSK>PP> \\
               <PP<W<:-) I think they're quite ^swee:t :-)>W>PP> /\~ (728–730)
```
Whereas Susanne pauses when talking about swearwords in her mother tongue, she starts giggling every time she refers to English swearwords during the interview. Part of the reason why Susanne thinks that it is acceptable for her to utilize English swearwords is precisely because it is not her native language, and therefore words do not carry a much weight when she utters them (see also section 9.5.5, "Personal offence"). Monika SG explains her liberal use of English swearwords in a similar way:

```
Example 9.5: Dean (English, 29) and Monika (Swiss, 29)
858 Monika: <HI<A it's because it doesn't ^mean anything to me A>HI> \\
               <W<MRC [it's an <HI ^empty] HI> ^word MRC>W> \~
859 Dean: [<HI e^xactly HI>] \
860 Monika: <P<HSK<LO for ^me LO>HSK>P> _
               [<DIM it doesn't have any c:onno^tation DIM>] _
861 Dean: [<L<HSK be:^cause i- <%> ^you: HSK>L>] _
               <L<HSK because it's ^not HSK>L> _
               <W<HSK <%%> [necessarily in:] in your own <HI ^language HI>HSK>W> \\
862 Monika: [<HI<A I didn't grow ^up with it A>HI>] \ (858–862)
```
To Monika, a swearword in her non-native language is an "empty word" without any connotations, which the couple explain with the fact that she did not grow up with it, and that it is not "[her] own language" (Dean 861). Even though Monika's partner is aware of this, he dislikes it when she uses certain expressions in his native language, which illustrates that not only the utterance, but also the reception of swearwords can pose a challenge in bilingual relationships. Thus, the bilinguals describe many different reactions to swearwords which are spoken in their mother tongue by their non-native partner; these will be discussed in the following section.

#### **9.5 Reactions to partner's use of swearwords**

The reaction of the native speakers to their partner's use of swearwords is determined by a number of factors. It depends on the specific swearword that is used, the language in which this word is spoken, the situation in which it is used, the social setting, and the manner in which it is framed (for instance, as a quotation, an expletive interjection, or a personal insult). It has to be borne in mind that the participants might have reacted less strongly to certain swearwords if they had not been in an interview setting. In the following, I will give an overview of the reactions the interviewees displayed and reported with regard to their partner's use of swearwords during the interview or in the past, and will attempt to categorize them. Of course, specific swearwords may trigger different reactions, so that it is possible for the participants to show several or mixed reactions.

#### **9.5.1 Indifference**

While the majority of the interviewees exhibit or report some reaction to their partner's swearing, there are some who claim that they generally do not care if their partner swears in their native tongue (Martin SG, Richard <sup>E</sup> , Katia SG). These three participants do not find anything odd or special about their partner using swearwords in their mother tongue, and do not mind it. It may be relevant that all of their partners do not swear very frequently or tend to use rather mild swearwords. At the same time, many of the interviewees are indifferent to swearing in their non-native language. They show no reaction to the use of swearwords in their L2 during the interviews, nor do they comment on this topic. The following extract is an example of this:


It is notable that Stephanie does not react at all to Robert's accumulation of swearwords, but simply supports his statement with a soft backchannel signal.

#### **9.5.2 Amusement**

A common reaction to the non-native speaker's utterance of swearwords in his or her L2 is for the native speaker to chuckle or laugh (Stephanie SG, Tim <sup>E</sup> , David E , Susanne SG, Dean <sup>E</sup> , Monika SG, Katia SG, Karen <sup>E</sup> ). Tim <sup>E</sup> , for instance, finds it amusing when Sarah SG uses an English swearword on occasion, as it sounds strange and unexpected to him:

```
Example 9.7: Tim (Australian, 29)
Tim: <W it <HI sounds HI> so ^funny to my ears W> /~
              <P it's ^it's like P> \\
              … <W<:-) did you really <HI ^say HI> that :-)>W> \ (1154)
```
The fact that hearing a non-native speaker swear can sound weird or comical to a native speaker is also the reason why L2 speakers are sometimes encouraged to say particular swearwords by native speakers. For instance, Stephanie's SG friends tried to make Robert <sup>E</sup> utter the terms *Fressi* ('gob', 'trap' as in *shut your trap*) and *Arschloch* ('asshole') — much to her chagrin — or Susanne SG taught David <sup>E</sup> to say *Siech* (a colloquial term for 'bloke', 'dude'; also 'damn'), finding this entertaining.

While some of the native speakers simply appear to be amused by their part‐ ner's swearing, others might instead be masking a slight sense of embarrassment with laughter (see also section 9.5.4, "Embarrassment"). It is often difficult to tell amusement and embarrassment apart, as the following example demonstrates:


Dean seems to consider Monika's statement entertaining and responds with soft laughter, yet at a later point in the conversation he voices criticism over her repeated use of the c-word (see example 9.13 below).

Reacting with amusement seems to be particularly common when the non-native speaker is not proficient in his or her L2, or speaks with an accent. As Monika SG points out, the fact that Swiss German is not her partner's native language and that he does not pronounce swearwords properly makes his swearing sound peculiar:

```
Example 9.9: Dean (English, 29) and Monika (Swiss, 29)
Monika: = <P<W<HI yeah but it's also you <CRK ^saying it and CRK>HI>W>P> \
              <P and you ^don't speak P> \
              ^p:roper \\ ((using fingers to indicate inverted commas))
              <P<CRK Swiss ^German CRK>P> \
              … <P then it P> makes it ^^even \
              … <P<HSK ^weirder HSK>P> \\ (888)
```
Because of his accent, Monika dislikes Dean using swearwords in Swiss German. Similarly, Katia SG thinks that it sounds "kind of funny" when Craig <sup>E</sup> uses swearwords in her mother tongue, but she views this positively, as it might break the tension (1214). In Karen's case <sup>E</sup> , it is her partner's choice of swearwords that she finds amusing, because the terms are very weak and sometimes old-fashioned:

```
Example 9.10: Karen (American, 51) and Philipp (Swiss, 47)
Karen: <F<HI that's really ^funny HI>F> //
             <F<HI and it de^pends on HI>F> \
             <F<HI what you consider a ^swearword HI>F> \\
             <W<:-) because <HI even HI> [his ^swearwords] :-)>W> \
Philipp: [well ^that's the f-] [that's the fr-] --
Karen: [<W are pretty ^weak: W>] /
Philipp: that's the ^first /
             that's the [^first point] \\
Karen: [<:-) you ^know :-)>] \\~
             <F<:-) they're not ^really :-)>F> \\ 
             <HSK<:-) ^swearwords :-)>HSK> \\
             they're: sort of what ^^grandpa \\
             <LO<HSK might consider a ^swearword HSK>LO> _ (1186–1190)
```
These examples demonstrate that finding a suitable swearword can be difficult for bilinguals even if they are highly proficient L2 speakers — while they may use swearwords that are too offensive, they may also choose a swearword that is perceived as too weak, too old-fashioned, too feminine, or too masculine. This is a possible source of amusement for native speakers and can subject non-native speakers to ridicule.

#### **9.5.3 Disappointment**

Some interviewees display a more negative reaction to their partner's use of swearwords. One of them, Robert <sup>E</sup> , reports that he experiences a sense of disenchantment or disappointment when his partner swears in his mother tongue. He thinks that it is "a shame" when she swears, because it destroys the image of the lovely Swiss girl with the charming accent for him:

```
Example 9.11: Robert (English-Irish, 20)
659 Robert:
660 Silja:
661 Robert:
                <:-)<HI I .. I ^I HI> always feel like somehow :-)> \
                <:-) it's .. it's a ^^shame :-)> //
                <P<LO you ^know LO>P> /
                <P<:-) I ^always think it's like :-)>P> \
                <A<:-) not because like <HI ^English HI> people say this but :-)>A> \
                <A<DOW<HI I always HI> think it doesn't ^sound DOW>A> \~
                … <P<A doesn't ^sound <LO right you LO> know A>P> /
                <PP ^cause PP> _
                <P I ^think when P> \\
                … <P<W<LO<A sh- she speaks LO> ^English it sounds W>P>A> /~ ((slurred))
                <P<W<HI it sounds really ^cute HI> you know W>P> /
                <P it ^sounds P> _
                <A with this ^accent A> \\
                <A<HI it sounds really ^beautiful HI><LO I think LO>A> _
                … <HI and then to say uh some a ^swearword HI> \\ ((snapping fingers))
                <LO you ^know like LO> \
                ^shit _
                <LO<HSK or so you ^know HSK>LO> /
                <WH I think it ^just WH> /
                <P<HSK<HI it's eh almost ^wrecks HI> this HSK>P> \\
                 <P<HSK and it des^troys this HSK>P> \
                <W and it <HI doesn't ^sound HI>W> \~
                … <P<:-) as good to ^me :-)>P> //
                <P<:-) when I ^hear this :-)>P> \
                <PP<# you ^know #>PP> /
                … <CRE is .. it doesn't ^^suit CRE> /~
                = <P m^hm P> \
                <HSK ^^her HSK> \~
                her <HI ^accent HI> \\
                <P<LO ^almost LO>P> \
                <P you know this this ^ha- P> --
                <HI ^these HI> harsh words /
                … well maybe it would suit ^my accent <LO you know LO> /
                <P ^cause P> _
                cause I'm <HI ^English HI> \\
                <P<HSK<LO I can LO> ^say like HSK>P> _
                <P<HSK these ^words HSK>P> / (659–661)
```
Robert's impression that swearwords do not "suit her" (659) highlights the fact that the dislike of his partner's swearing is not so much due to a taboo attached to specific swearwords, but rather because it ruins the image that he has of his partner (as innocent, lovely, girlish, and maybe sophisticated). The element of gender may play a role in this regard, as Robert characterises Stephanie's manner of speaking with the (more feminine) attributes "cute" and "beautiful" (659), as opposed to "harsh" (661) words that suit his own speech.

It should also be noted that Stephanie swears considerably less frequently than Robert, and did not use any swearwords during the interview, whereas Robert does not seem to have any reservations about using swearwords himself (as demonstrated in example 9.6 above). Nevertheless, the vocabulary with which he describes the effect of Stephanie's use of swearwords ("shame", "doesn't sound right", "wrecks", "destroys", "doesn't sound as good", "doesn't suit […  ] her") and the urgency of his speech (using strong stresses and repetition) highlight that he feels strongly about this. Moreover, it is remarkable that Robert thinks that swearwords do not suit his girlfriend's accent, but that they do suit his, although in many ways, hers is not that dissimilar to his. For instance, they both tend to slur their words (and comment on this during the interview), and they claim that they have adapted their manner of speaking to each other to some degree (see section 5.5.1, "Modifying one's manner of speaking"). Roberts attributes the fact that swearwords suit him better to his British-Irish origin, thinking that his cultural identity and heritage entitle him to use these words. Thus, his emotional reaction and his attitude are also influenced by a sense of ownership of the language.

#### **9.5.4 Embarrassment**

Some of the native speakers feel that it is not socially acceptable for their partner to use swearwords in their mother tongue — at least not offensive ones — and react with embarrassment and protest when their partner swears (Courtney <sup>E</sup> , David <sup>E</sup> , Monika SG, Dean <sup>E</sup> ). Interestingly, it occurred repeatedly during the interviews that one partner claimed that he or she did not mind his or her partner's swearing, to which the latter objected, reporting that he or she does remember being criticized for using certain swearwords. For instance, Courtney <sup>E</sup> initially states there are no words that Martin SG uses about which she has any negative feelings, but when he disagrees and provides an example, she reacts strongly:

```
Example 9.12: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
769 Martin: <HSK but s- ^some HSK> // 
               some you ^do (mind) /
770 Courtney: … <CRK like ^what CRK> \
771 Martin: but I don't really ^use it \
               ^the: _ 
               <UP<HSK the ^c-word HSK>UP> /
772 Courtney: … <HI<CRK ^yeah CRK>HI> \~
773 Martin: … ^you u:h _
               <A if [if I say if I <HI ^say HI>] that like as a: A> \
774 Courtney: [<L ^obviously L>] \ 
775 Martin: as a ^joke /
               <P<LO you: LO> .. ^cunt P> /
776 Courtney: <HI<L<CRK [you can't ^say] that CRK>L>HI> \\ ((very high))
777 Martin: [<F<W not to ^her W>F>] /
               <F<HI not HI> to ^her F> /
778 Courtney: <HI you can't ^say that HI> \\ ((very high))
779 Martin: ((…)) <P and then ^Court- (says) P> --
               <P<HSK<HI ^^Martin HI>HSK>P> \ ((very high)) (769–779)
```
Courtney's reaction — repeating "you can't say that" in a very high voice (776, 778) — demonstrates that she is embarrassed about her partner's use of the term *cunt*. The extract also shows that Martin is acutely aware of the taboo that Courtney associates with this particular expression, as he anticipates and imitates her reaction. Such a difference in perception of the offensiveness of the term is mentioned several times; interestingly, this is the only swearword (out of 30) in Dewaele's study that was rated less offensive by L2 than by L1 speakers.

Many of the bilinguals insist that offensive terms are particularly unaccept‐ able when the non-native partner uses them in the company of others (as in the interview situation). Dean <sup>E</sup> , for instance, states that he does not mind Monika SG saying the c-word as long as she is alone with him, but worries about how this (or she) is perceived by others:

```
Example 9.13: Dean (English, 29) and Monika (Swiss, 29) 
849 Monika:
850 Dean:
851 Monika:
852 Dean:
                <HI<W but you ^did say to me once W>HI> \
                <P<HSK that you don't ^like it HSK>P> _
                <HSK<P when c- I say the word P> <PP ^cunt PP>HSK> \
                … ((lengthy pause)) <H> <Hx>
                … <W<:-)<A<DIM cause it's <HI ^obviously HI> <LO not very lady-like
                LO>DIM>A>:-)>W> _
                <H> <HI it's no- ^no HI> /\\~
                <W<HSK<A I mean I <HI don't HI> mind when you say it with ^me
                A>HSK>W> /\\~
                … I mean ^anyon:e \
                that would listen to this <HSK eh <HI ^interview HI>HSK> \\
                <P<W<HSK<LO will think LO> you <HI ^are one HI>HSK>W>P> \\
                <RH<:-) because you ca- ^don't stop saying the ^word :-)>RH> //
                                                                        (849–852)
```
Dean avoids saying the word *cunt* itself in this situation, and while Monika utters it five times during the interview, he never uses it. Yet he cleverly manages to use her own frequent mention of the word to criticize (and potentially insult) her, saying that others would think that she is "one" if they heard her using the word (852). When Dean turns the tables and intends to demonstrate that Monika would be bothered by his mention of the same word in her native tongue, she reacts indeed much more strongly than he does in the reverse situation (and he manages to make his point):

```
Example 9.14: Dean (English, 29) and Monika (Swiss, 29)
864 Dean: <H:> <A but if I turn a^round and A> _
                ^e::m \~
865 Monika: = <P<NAS ^no NAS>P> \~ (("oh no" voice quality))
                [<P<A<:-) don't ^say it :-)>A>P>] _
866 Dean: [^said] _
867 Monika: @@ ((chuckling very softly; possibly inhaling and exhaling))
868 Dean: <:-)<W<HI see you don't HI> even want me [to <HI ^say HI> it] W>:-)> \\
869 Monika: [<HI ^^no HI>] \~
                <HI that's [the ^thing] HI> //\~
870 Dean: <:-)<HI [if .. if I] [^said] HI>:-)> \
871 Silja:
872 Dean:
873 Monika:
874 Dean:
875 Monika:
876 Dean:
877 Monika:
                                       [^what] _
                <DOW<:-)<HI if I ^said it HI> in your: :-)>DOW> \~
                ^e:m _
                … <P<HI ^futz HI>P> _ (('cunt'))
                1[@@@@@@@@@@@@] ((hearty laughter throughout Dean's turn))
                1[<:-) if I if I said if I called you a futz in your <HI ^language HI>:-)>] \\
                2[@@@@@@] 
                2[<W<HI you'd HI> be ^^really pissed off W>] //
                ^wouldn't you \\
                = <P<:-) I ^would :-)>P> \
                <PP<HI<HSK<:-) ^yeah :-)>HSK>HI>PP> \~
                <P I ^would P> \
                [<PP<HI<HSK ^yeah HSK>HI>PP>] \~
878 Dean: [if I] ^said _
                <PP du bisch [u ^huere] PP> -- (('you are bloody …'))
879 Monika: [<HI<F ^^nei F>HI>] _ (('no'))
                <HI<F ^etzt F>HI> _ (('now', meaning 'come on now'))
880 Dean: = <:-)<HI<F yeah e^^xactly F>HI>:-)> \\ (864–880)
```
Monika's reaction goes from begging her partner softly not to say the swearword ("no, don't say it" [865]) and expressing frustration (chuckling <@> and inhaling <H> [867]) to laughing loudly when he does say it (873), to finally objecting loudly and vehemently and even switching to Swiss German when he threatens to say it again ("^^nei, etzt" 'no, come on now' [879]). Interestingly, this is the only instance in which Monika speaks Swiss German during the interview. Her switch may have been triggered by Dean's use of a Swiss German word, but it could also be a rhetorical device. Words which are uttered in a language other than the language of the ongoing conversation stand out and may therefore carry more weight; code-switching can thus be a convenient means to lend substance to one's words. David <sup>E</sup> uses this strategy as well when he fears that his wife might say the c-word, as can be seen in example 9.15 (755):

```
Example 9.15: David (English, 32) and Susanne (Swiss, 29) 
733 Susanne: = like the c-word \
               <HI I don't ^mind HI> \~
              1[<W saying the ^c-word W>] \
734 Silja: 1[@@@] ((softly))
735 David: 2[you ^can't] //
736 Susanne: 2[but he] [^blushes] \
737 David: [<F<@ you're ^^not @>F>] //
              3[<@ gonna say this ^word @>] \~
738 Susanne: 3[<@ when I ^sa@y it @>] \
739 Silja: 3[@@@] 
740 Susanne: 4[<HI<F no I'm ^^not F>HI>] // 
741 Silja: 4[<@ you ^can't @>] /
742 Susanne 5[<@ going to ^say <LO it on here LO>@>] _
743 Silja: 5[@@@] /
744 Susanne: <:-) I know I'm not <CRK a^llowed to CRK> :-)> \\
745 David: 6[<DOW<@ not when we've got ^company @>DOW>] \\
746 Susanne: 6[<H> ^but .. ^but] _
              7[<P<:-)<HI I find it HI>] a very sweet ^wo:rd :-)>P> /\~
747 Silja: 7[@@@] ((softly))
748 Susanne: <P<CRK<:-) ^so :-)>CRK>P> _
               [<F ^just because F>] _
749 David: <FF<W [it's <HI ^^not HI> a@] sweet [^^wor:d] W>FF> /\\~
750 Susanne: [@@@]@@ ((chuckling softly))
751 David: 8[<PP<HI<HSK ^good ^^God HSK>HI>PP>] /\~
752 Susanne: 8[@@@@] ((chuckling))
753 Silja: 9[@@@@@] 
754 Susanne: 9[@@ <:-)<HI is #was the next HI> argument in ^German :-)>] _
              @@ <H:>
755 David: <P<:-) das ist <HI un:^^glaublich HI>:-)>P> \ (('that's incredibly […]'))
              [<P<HI das ist ein ^ganz schleckt ^Wort HI>P>] \ (('that is a very bad word')) 
756 Susanne: [@@@@@@] ((chuckling)) (733–756)
```
David reacts even more strongly to the possibility of Susanne saying the c-word than Monika does to the Swiss German equivalent, and uses a variety of supra‐ segmental features to lend weight to his statements, such as increased loudness (<FF>), contrastive stresses ("good ^^God" [751]), lengthening ("un:^^glaublich" [755]), a high pitch (<HI>) and a wide intonation contour (<W>) (all underlined in example 9.15). Not only does he tell his partner repeatedly that it is not a sweet — but rather an incredibly bad — word, he expressly forbids her to use the term, to which she complies ("you can't […  ] you're ^^not gonna say this word" [David 735, 737]; "I know I'm not allowed to" [Susanne 744]). It is worthy of note that this is something that never happened when a bilingual swore in his or her own mother tongue during the interview. Even though some may not like it when their partner swears in his or her own native language — and some explicitly say this (Sophia SG, Susanne SG) — they neither correct their partner nor prohibit the use of such language. However, when a partner swears in his or her L2, the native speaker tends to assume the role of an educator or parent, and acts as a kind of gatekeeper, who decides which expressions can and cannot be used. It is also remarkable that, during the interviews, the L2 speakers never questioned the native speakers' authority on the matter.

Interestingly, David <sup>E</sup> , after asserting that Susanne SG "breaks the ^^major taboo" (1079) by saying the c-word, and telling her repeatedly not to say it (see example 9.15 above), then proceeds to dare her to utter the word:

```
Example 9.16: David (English, 32) and Susanne (Swiss, 29) 
1082 David: <P<HI<HSK y:ou ^cannot HSK>HI>P> \ 
               <PP<HI s- ^look HI>PP> _
               <P<HI ^can you just HI>P> \
               1[<:-) put it on ^record :-)>] \\
1083 Susanne: 1[@@@] ((chuckling))
               2[@@@] ((chuckling))
1084 David: 2[^^say the word] \
               3[<HSK ^^say it HSK>] \
1085 Susanne: 3[@ <:-) ^no: :-)>] _
               <@ I'm ^not saying it @> \ (1082–1085)
```
Of course, Susanne refuses ("no:, I'm not saying it" [1085]), claiming that if she did, she would hear about it for the rest of her life, which underlines her deep awareness of the taboo attached to the word. Furthermore, she reacts strongly when David later actually utters the word, interrupting him mid-turn in a high and loud voice (1087):

```
Example 9.17: David (English, 32) and Susanne (Swiss, 29) 
1086 David: <P<BR there is <HI only ^one HI> swearword in English BR>P> \
             <P ^which: .. which is P> _
             … <P<HSK<HI ^^really HI> the ultimate ta^^boo <LO nowadays LO>HSK>P> _
             <P<HSK<LO and that that's ^cunt LO>HSK>P> \\
             <HSK<W and you're <HI ^^not HI> allowed [<P<LO to say this LO>P>W>HSK>] \
1087 Susanne: [<F<HI you ^said i:t HI>F>] \
              (1086–1087)
```
It is evident that Susanne is not offended by her partner's use of the c-word, but rather genuinely surprised and somewhat perplexed by it. In addition, the extract demonstrates that David feels that different rules apply to him — as a native speaker or, potentially, as a man — than to his wife: Immediately after uttering the word himself, he states that "you" (it is unclear whether this is a generic *you*, or refers to Susanne) are not allowed to do the same.

A similar double standard becomes apparent in the conversations between other couples. Courtney <sup>E</sup> , for instance, insists that her partner is not allowed to say certain words, even though she has used them herself in the past. In the following extract, Martin SG cites what Courtney said in an earlier situation when he describes a friend's ex-girlfriend as "a bitch" (1528); yet Courtney declares that "he can't say that" (1531), before remembering that she herself said that "she was a bitch" (1534):

```
Example 9.18: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
1528 Martin: <P<CRK<DOW because she was a ^bitch DOW>CRK>P> \\
1529 Silja: … @@@[@@@] 
1530 Martin: [because she ^was] \
1531 Courtney: <W<PP you see he <HI can't HI> <CRK say ^that CRK>PP>W> /
              <PP<CRK for ^instance CRK>PP> _
1532 Silja: [<P he ^can't P>] /
1533 Martin: [<PP<HI #^why HI>PP>] _
1534 Courtney: … <PP<HI ^no:: HI>PP> \\~
              <PP<HSK well <HI ^I HI> said that didn't I HSK>PP> _
              <PP 1[she] was a ^bitch 2[#to #me] PP> \
1535 Martin: 1[^yeah] \~
               <P<:-) 2[well of] <HI ^course HI> you #did :-)>P> _
1536 Courtney: @@@@ (1528–1536)
```
Overall, about half of the partners of speakers who occasionally or regularly swear in their L2 (5/10) appear to be embarrassed by this behaviour in public, even though many of them do not seem to mind, or at least not as much, when they are on their own.

#### **9.5.5 Personal offence**

In contrast, some native speakers do not only consider using certain swearwords to be inappropriate in a social context, but they are also personally bothered or even offended by their partner's use of coarse language (Sophia SG, Claire <sup>E</sup> , Simon SG, Craig <sup>E</sup> ). To give an example, Craig <sup>E</sup> comes from a rather conservative and religious background, where swearwords were used very rarely, while his wife Katia SG, who was a teenage exchange student when they met, used swearwords rather liberally back then. Katia remembers that he was "extremely offended" by her using the f-word (1234), while she was hardly aware of using it. Craig explains that for him, there are some words that have a "^^rea:lly … heavy meaning" (1255). He stresses that it makes a difference whether or not someone swears in his or her native tongue, and that the word *fuck* means something else to him than it does to a non-native speaker:

```
Example 9.19: Craig (American, 32) and Katia (Swiss, 28) 
1259 Craig: really #you can't# say o^ffen- <HI not off- HI>] --
              = ^yeah it's [it's] --
1260 Katia: <F<DOW [yeah you] ^^were offended DOW>F> \
1261 Craig: [I I wou- ^I would say I wasn't] --
1262 Katia: [<DOW I mean you can ^say that DOW>] \\
1263 Craig: … <HI yea:h I ^guess HI> /
              I <HI ^wasn't HI> _
              let's ^say \
              … <A<W<HI personally HI> offended as in personally a^ttacked W>A> /
              [by it .. ^but] _
1264 Katia: [<HI ^^no ^^no but HI>] \
1265 Craig: but it <HI ^really HI> _
              <W but it ^bothered me for sure W> /
1266 Katia: <CRK ^right CRK> _
              <PP<HI m^hm HI>PP> //
1267 Craig: = just because of what my associ^ation and u:h _
              … the <HI ^meaning HI> this word has \
              … <LO in my: .. native ^tongue LO> \ (1259–1267)
```
Craig finds it hard to define his precise reaction to the f-word, and his speech is halting and full of truncated words and false starts in this sequence (underlined). He initially contradicts Katia's assessment that he was offended by her use of the f-word, though he later agrees reluctantly ("yea:h I guess" [1263]) and acknowledges that it bothered him "for sure" (1265). The extract highlights how much her use of the word upset him back then, even though it was not directed at him, but simply because of his associations with the term itself.

Likewise, Sophia SG is occasionally bothered by her husband's use of swear‐ words in her L1, mainly because she thinks that he sometimes "doesn't have a clue what he's saying" (1258). She believes that the reason for this is that Richard E learnt most swearwords from her father, whose first language is Spanish. In her eyes, her father himself does not swear appropriately in Swiss German, because his own parents never taught him "that there are various forms of swearwords, and different levels" (1267). Her husband simply adopts the swearwords her father uses, which she finds problematic:

```
Example 9.20: Richard (English, 29) and Sophia (Swiss, 31) 
1267 Sophia: and he ((her father)) just throws them all <HI to^gether HI> _
1268 Richard: @@@ ((softly))
1269 Sophia: … <HSK and then it's <HI ^^really HI> ^^bad HSK> //~
               … <CRK and <HI he ((Richard)) just HI> takes it ^over CRK> \
1270 Richard: @@@@ ((chuckling))
1271 Sophia: = <P and has <HI<MRC ^no ^^clue MRC>HI>P> \\~ (1267–1271)
```
While Sophia is criticising her husband's swearing, Richard is chuckling softy, which suggests that this either amuses him or makes him slightly uncomfortable.

The couples' reports indicate that when one partner uses a swearword in the other's L1, this can be a source of disagreement and cause frustration on both parts. Sophia SG and Craig <sup>E</sup> in the examples above are bothered by their partner's use of swearwords, but not necessarily personally offended; Claire <sup>E</sup> , in contrast, reacts very strongly when Simon SG uses the f-word. Although she occasionally makes use of the term herself, she takes offence when he does and goes "ballistic" (Simon 1565). Neither of them can explain why Claire reacts so sensitively to Simon using this word. For Simon, it is frustrating "that she:'s allowed to use the f-word" (1552) and he is not, and he feels that it puts him in a weaker position. The fact that she forbids him to use the term shows again that being a native speaker is connected to legitimacy and a sense of ownership of a language, and with the ability to decide who is allowed to say what (see section 9.5.4, "Embarrassment", above).

The bilinguals describe the strongest reactions to swearwords in their mother tongue, yet there are a number of interviewees who report having felt personally offended by swearwords in their L2 (Sarah SG, Susanne SG, David <sup>E</sup> , Sophia SG, Monika SG). In fact, their accounts suggest that, although taboos, especially social ones, are usually felt most intensely in one's native tongue, it is also possible for a non-native speaker to feel strongly about swearwords in his or her L2. As an example, Susanne SG expresses a deep dislike for the expression *sod off*, which she attributes to her awareness of its original meaning. Other swearwords are felt to be offensive because of their present denotation. In this study, it was mainly the female partners who reported negative feelings about swearing or specific swearwords in their L2, particularly with regard to derogative terms for women. Thus, Sarah SG initially found it hard to get accustomed to certain words that are used rather frequently in English, such as *bitch*. The term is difficult to translate into German, and is often (mis-)translated as *Schlampe* ('slut'), which is perceived as very offensive. Similarly, Monika SG finds it highly offensive when a woman is referred to as *bitch* or a *whore* (though, interestingly, she minds the c-word much less; see example 9.13 above). Besides the original or current meaning of swearwords, associations with similar swearwords in the bilinguals' L1 may also explain their reaction to specific expressions to some extent. Since swearwords in English and (Swiss) German are often similar in sound and etymology (e.g. *whore* — *Huere*; *shit* — *Scheisse*, *damn(ed)* — *verdammt, dami*), they might trigger a similar reaction in someone's L2 as they would in his or her L1. The equivalents of the terms the Swiss bilinguals find offensive in English tend to be less commonly used in Swiss German, or to carry more weight than they do in English.

A number of couples disagree considerably on the level of offensiveness of certain swearwords (Claire/Simon, David/Susanne, Dean/Monika), which can lead to misunderstandings or conflicts. Simon SG reports that there are some words that he and Claire call each other occasionally which one of them perceives as "really offending", while the other one thinks "they're nothing" (Simon 727). For instance, he dislikes it when Claire <sup>E</sup> calls him a *dick*, while she finds the term "quite endearing, like come on silly" (Claire 737), because she and her three sisters use this expression rather often to address each other. Example 9.21 demonstrates the extent to which they disagree on this particular word:

```
Example 9.21: Claire (Northern-Irish, 28)
Claire: <P and ^he's like P> \
             <P<W[<HI don't HI>] ^call <HI me HI> that W>P> \
Silja: [<PP m^hm PP>] _
Claire: <P<HSK [that's] not ^nice HSK>P> //~
Silja: [<PP m^hm PP>] _
Claire <P<HI ^I'm like HI>P> _
             <H:> … <P ts- think it <@ ^I think it's funny @>P> //
             [<@ I think it's no@t .. o@ffensive at ^^all @>] //
Silja: [@@@ ##] ((softly))
Claire: <@ and he doesn't ^like <PP it PP>@> // (737–743)
```
Part of the reason for these differences in perception may be that among particular groups of English speakers, it is not unusual to use personal insults (such as *bitch* or *slut*) to achieve a humorous effect, or as terms of affection,<sup>6</sup> whereas this is rarely practiced in Switzerland. Therefore, bilingual couples can have very different ideas about which terms are offensive and which ones are not, and bilinguals sometimes have to adapt their swearing behaviour to their partner in order not to give offence.

<sup>6</sup> In this case, swearwords may be used as markers of in-group solidarity, similar to the ritual insults among speakers of AAVE described by Labov (1972). As calling each other *bitch* or *slut* appears to be a relatively new phenomenon among women, I did not find any scholarly articles on the subject, but a number of people have commented on this online (e.g. Hetzel 2016).

Some couples have different notions about swearing in general, partly due to differences in their upbringing or background (Robert/Stephanie, David/ Susanne, Richard/Sophia, Craig/Katia). Thus, Sophia SG takes the view that Richard <sup>E</sup> should limit his overall use of swearwords, especially since she does not want their baby daughter to learn "these words" (1244). She states that she does not like it when Richard uses the f-word "for ^every ^single ^thing" (1282) and says that, even though she is not religious, she particularly has a problem with him (or anyone) using swearwords that insult God. Susanne SG also dislikes it when David E swears in his mother tongue, and used to feel very uncomfortable when he was cursing at his computer, for instance. She explains that she did not grow up in a family where strong language was used, and tends to take it personally, even if the swearing is evidently not aimed at her. When David argues that this is a contradiction to her perception of the c-word as a "sweet word" which she does not mind using, Susanne insists that there is a crucial difference, because the force and the meaning are not the same when one swears in one's L1 as opposed to one's L2:

```
Example 9.22: David (English, 32) and Susanne (Swiss, 29)
```


According to Susanne, English swearwords are far stronger when her partner utters them, since they are in his language. When she swears in her L2, on the other hand, she does not mean to offend, and therefore feels at liberty to say whatever she wants (see also section 9.4.2.3, "Reasons for L2 preference", above). In Susanne's opinion, the level of offensiveness is thus not inherent in the word itself, but rather depends on who the speaker is.

Another reason for a bilingual to feel offended by a swearword in his or her L2 might be that he or she finds it difficult to judge the level of offensiveness of the word in question. This can leave the non-native speaker with the impression that their interlocutor is taking advantage of the fact that he or she is not proficient in the language. David relates an anecdote of a situation in which he felt offended by an expression in Swiss German. He had helped his father-in-law install a lamp, and because of an unintentional error on David's part, the fuse burned out, and his father-in-law used the word *Tubel* (a mild swearword meaning 'idiot'). While Susanne insists that her father was not referring to David but the lamp, David does not believe this ("that's what she says" [938]), and states that he would prefer it "if you just called me you stupid idiot *in English*, rather than that" (954, my emphasis). What upset him the most back then appears to be that he felt that his father-in-law called him a name thinking that he would not understand this term. Despite all protestations from Susanne's family that this was not the case, David was so offended that he just disappeared for a while, and he still does not seem to fully believe them. This underlines the fact that the assertion that an expression is not supposed to be offensive does not help if that word is felt to be offensive.

In sum, the couples' reports suggest that bilinguals can take just as much personal offence to swearwords in their L2 as their L1. In spite of this, however, the non-native speakers in this study never protested against their partner's use of words they consider offensive, or forbade their partner to use them, whereas the native speakers sometimes intervened when their partner used a word which they disliked (see section 9.5.4, "Embarrassment"). Moreover, it seems to be relatively rare for the partners to be personally offended by each other's swearing, especially since many of them have been in a relationship for a substantial period of time and both parties know what triggers such a reaction in their partner. Thus, although various interviewees shared memories of situations in which they felt offended by a particular word in the past, no-one showed signs of being personally offended by a word their partner used during the interview.

#### **9.6 Discussion and summary**

The participants displayed and reported very diverse swearing behaviour, ranging from never swearing to swearing very frequently. Overall, they did not use a great number of swearwords during the interviews — on average 6 swear‐ words per person — and half of the swearwords were used in a metalinguistic context. The fact that they did not swear very much may partly be attributed to the semi-formal setting, as the setting and the level of formality have been found to influence swearing behaviour (Jay and Janschewitz 2008: 272). The participants' swearing behaviour during the interviews was largely congruent with their reports on their frequency of swearing. The only exception to this were two English-speaking male participants, who did not swear very much during the interviews, but claimed to swear rather frequently in their everyday life. As for the couples, 8 out of 10 displayed relatively homogenous swearing behaviour, with both partners using a similar number of swearwords with comparable levels of offensiveness. While this may be linked to the specific 'key' of each conversation to some extent, it might also be interpreted as an indication of a long-term assimilation of the partners' swearing behaviour to each other, especially since the two couples who showed the most divergent swearing behaviour were among the three couples with the shortest relationships.

As was expected based on previous research, the bilinguals' swearing behav‐ iour seems to be influenced by their gender. During the interviews, the female participants swore on average slightly less than the male participants, and they used milder swearwords when swearing spontaneously. This is in line with previous research, most of which suggests that men swear more than women (e.g. de Klerk 1990, 1991; Rayson, Leech and Hodges 1997; Bayard and Krishnayya 2001). However, it is noteworthy that, in the metalinguistic context, the female participants used more swearwords than the male participants, especially more offensive terms. It appears that, while there is a tendency for women to swear less frequently in spontaneous speech than men, women have no problem with vocalising swearwords per se. This provides an explanation for the results of Jay, King and Duncan's study of the use of swearwords in narratives about punishments for swearing, in which no gender differences were found (2006: 127). It could be hypothesized that this gendered pattern in the spontaneous versus the metalinguistic context indicates that the main reason why women use fewer and weaker swearwords in spontaneous speech is social custom, rather than strong taboos associated with uttering swearwords, as such taboos would probably also be felt in a metalinguistic context.

Besides gender, the cultural and linguistic background of the bilinguals has proved to have an influence on their swearing. In both contexts, the Eng‐ lish-speaking participants used more swearwords than the Swiss participants, especially more swearwords categorized as mild. The Anglophone partners' more frequent use of spontaneous swearwords might be attributed to the fact that the language of the interviews was English, and there appears to be a preference for swearing in the language of the conversation, but also for swearing in one's mother tongue. Interestingly, offensive swearwords in both languages were uttered more frequently by the respective non-native than the native speakers, which might be because the latter deem these terms more socially inappropriate. During the interviews, moreover, the six British participants swore by far the most of all participants, while none of the three Americans used any swearwords spontaneously during the interviews. This might not come entirely as a surprise, as British people are generally thought to swear more and use stronger swearwords than Americans (Hughes 2006: 13), even though this tendency was not confirmed in self-reports on the use of swearwords of British and American respondents (Dewaele 2015). In the case of the participants in this study, the British participants both reported using more swearwords than the Americans and the Swiss, and actually produced more of them during the interviews.

Other important factors influencing the participants' swearing behaviour are their age and family situation. Younger participants swore more frequently than older participants, even though there were no interviewees in their teens, which is the age group that is generally believed to swear the most (McEnery and Xiao 2004; Thelwall 2008). An even stronger factor than age appears to be whether or not the couples have children of their own. On average, the couples who did not have children used more than twelve times as many swearwords as those who did. This reflects earlier studies, according to which parents attempt to reduce swearing in order to set a good example for their children (Holmes 1992; McEnery and Xiao 2004). The present study indicates that such resolutions may influence the swearing behaviour of parents even when their children are not present.

In terms of language choice for swearing, the most important factors were confirmed as the language of the ongoing conversation, the speaker's profi‐ ciency, and his or her socialisation in the language. All of the 60 swearwords that were used spontaneously during the interviews were produced in the language of the ongoing conversation (59 in English, 1 in Swiss German). With regard to their usual swearing behaviour, 6 of the participants reported that they swear mainly in their L1, while 12 participants use both languages to swear (to various extents); none of the interviewees reported that they swear mostly in their L2. Especially the speakers who are less proficient or who have an accent are less inclined to swear in their L2. This is in line with research carried out by Dewaele, who found language dominance to influence bilinguals' frequency of L1 and L2 use for swearing (2004a: 94). The differences in language proficiency, together with a high level of socialization, are likely to be the reason why the Swiss participants reported swearing more frequently in their L2 than the Anglophone participants. Their proficiency and their socialization in their L2, in turn, are linked to the couples' choice of English as their relationship language.

The bilinguals listed a number of reasons for their language preference when swearing. Some participants prefer swearing in their L1 as they consider it a more colourful language, or generally better suited to swearing. Others swear in their L1 because they feel that they lack swearing proficiency in their L2, or because they are concerned about causing offense to their partner or other native speakers. In contrast, some bilinguals swear in their L2 because they consider it easier to swear in a non-native language as swearwords carry less weight than they do in their mother tongue, which is probably linked to the fact that linguistic taboos are acquired at a young age (Billig 1999; Jay, King and Duncan 2006; Jay and Janschewitz 2008). These results also underline that the choice of L1 or L2 for swearing is very much dependent on the individual, as for some, the reduced emotional impact of their L2 is perceived as an advantage (see Koven 2004, 2006; Dewaele 2004a: 95; Pavlenko 2005: 136), while others are reluctant about using their L2 for swearing because they do not feel entitled to do so (see Dewaele 2004b: 220, 2010: 218). Yet despite some hesitations, particularly with regard to using strong swearwords in one's L2, the majority of the interviewees (12 of 20) swear in both of their languages.

The participants displayed and reported various reactions to their partner's use of swearwords, depending on the situation, the language, and the swearword in question. Most of the bilinguals do not mind swearwords being used in their L2, though a few of them reported that there are specific words in their L2 which they dislike. In contrast, their reactions varied greatly with regard to swearing in their mother tongue. While some participants claimed that they are usually indifferent to their partner's swearing, especially if their partner has a habit of using relatively mild swearwords, others find their partner's swearing amusing or entertaining, which seems to be particularly common if their partner has an accent, or chooses odd or unsuitable swearwords. This is even seen as an advantage by some, as it may break the tension during an argument, but it also emphasizes that swearing in their L2 potentially exposes bilinguals to ridicule. One interviewee reported being disappointed when his Swiss partner swears in her L2, as it ruins her beautiful manner of speaking. Other interviewees feel embarrassed when their native tongue is used by their partner, particularly if their partner uses relatively offensive swearwords in public. Finally, there are also interviewees who are personally offended by their partner's use of particular swearwords. This appears to be relatively rare, as even though many of the interviewees recalled situations in which they were offended by a specific swearword in the past, no-one appeared to be personally offended by a swearword that was used during the interview. Overall, the interviewees displayed and reported much stronger reactions to swearwords in their L1 than their L2, which suggests that L1 swearwords have a greater emotional force for them, as has been shown to be the case for other multilinguals (Dewaele 2004a, 2004b, 2010). At the same time, the utterance of a swearword by a native speaker is viewed as more forceful, which suggests that the offensiveness of a word can depend as much on the speaker and the situation as on the word itself.

Language ownership emerged as a recurrent topic in the bilingual couples' conversations. As Dewaele noted, only proficient users of a language may feel close enough to the in-group to use powerful swearwords (2004b: 220). Yet despite the very high level of L2 proficiency of some of the participants, they (and their partners) often feel that, as L2 speakers, they are not entitled to utter strong swearwords. Indeed, many native speakers believe that, because of their cultural and linguistic background, only they have the right to use swearwords and to decide which words their non-native partner can or cannot use. Several native speakers protested when their non-native partner threatened to use an offensive swearword during the interview, asking them emphatically not to utter the expression or even forbidding its use. Interestingly, some native speakers seem to think that the rules which they have established for their partner do not apply to them, and they themselves use words which they do not allow their partner to use (this double-standard was also personally experienced by Dewaele 2010: 7). In contrast, none of the non-native speakers explicitly forbid their partner to swear, even if they dislike a particular expression. It becomes evident that ownership of a language comes with responsibilities: words can carry more meaning when spoken by a native speaker, who may also feel responsible for his or her non-native partner's manner of speaking.

## **10 "Then we're the only two people laughing in the room": Laughter and humour**

## **10.1 Introduction**

The final chapter of this book is dedicated to the aspects of humour and laughter in bilingual, bicultural relationships. Laughter and humour are important elements in human interactions, especially if the speakers have a close relationship. Shared laughter and humour can help establish a positive atmosphere, create a sense of togetherness or intimacy, and diffuse difficult situations. Conversely, laughter can also engender negative feelings, or establish an imbalance between the speakers, as it can "serve to mock, deride, and belittle others" (Glenn 2003: 1). The kind of humour that is utilized tends to be connected to the relationship between the speakers; therefore, humour style and laughing behaviour can be indicative of aspects such as the conversationalists' level of familiarity, or the distribution of power within the relationship. Moreover, humour appears to have "an active role in both interpersonal attraction and mate selection" (Hahn and Campbell 2016: 407), and also plays a central part in intimate relationships (Ziv 1988; Priest and Thein 2003; Barelds and Barelds-Dijkstra 2010).

Humour and laughter have been studied in a variety of social interactions, by psychologists, anthropologists, and linguists. Nevertheless, the manner in which humour is expressed among bilingual couples, and the issues that potentially arise from bilingual partners' different cultural and linguistic backgrounds, have received little attention to date. I intend to fill some of these gaps by considering the topic from several different angles. As in previous chapters, I combine an analysis of the couples' behaviour during the interviews with their reports on the subject. In the quantitative analysis, my aim is to determine how much the bilinguals laughed during the interviews, as well as what topics and which speaker triggered their laughter. Furthermore, I am interested in their use of laughing and smiling voice quality. Since self-reports have indicated that married couples often share similar humour styles (Hahn and Campbell 2016: 414), I aim to determine whether such similarities in humour styles and laughing behaviour can also be observed among bilingual, bicultural couples, in spite of the fact that they speak different mother tongues. The bilinguals' cultural and linguistic backgrounds are relevant because, on the one hand, humour has often been found to be culturally determined (e.g. Cheng 2003; Yue et al. 2016), and, on the other hand, the native speaker in the relationship may have an advantage over the non-native speaker in producing and understanding humour. Besides culture and language, a speaker's gender can also influence his or her laughing behaviour (e.g. Lampert and Ervin-Tripp 2006; Holmes 2006). Therefore, I also attempt to determine the role of the participants' gender in their laughing behaviour and their production of humorous comments.

Since humour is often subjective, I am also interested in the couples' personal opinions on various topics relating to humour in their relationships, and their accounts of the development of their joint bilingual couple humour. I examine how their backgrounds influence their perception of each other's funniness, as well as challenges that they have encountered as a result of their different linguistic and cultural backgrounds. Moreover, I aim to explore features which the couples believe to be specific to their couple humour, particularly their use of playful language, such as invented words, mixed and blended language, or imitated accents. The interviewees' reports indicate that they are able to turn their bilingual, bicultural situation into something humorous; this also becomes obvious in the topics that trigger laughter during the interviews. The couples' accounts, in combination with actual data on their laughing behaviour, will aid in furthering our understanding of the subject of laughter and humour in a bicultural, bilingual context.

In what follows, I first provide an overview of previous work on laughter and humour (10.2). After this, I examine the couples' laughing behaviour during the interviews, and attempt to find out which topics trigger laughter, who triggers laughter, who laughs more frequently, and whether the bilin‐ guals' laughing behaviour is linked to the factors of gender or mother tongue. I also discuss the bilinguals' use of laughing and smiling voice quality during the interviews (10.3). Subsequently, I look at the topic from a qualitative perspective, and examine the couples' reports on what they usually laugh about, their perceptions of who is funnier, and challenges in expressing and understanding humour as a bilingual couple, as well as their use of playful language (10.4). In my conclusion, the results of the analysis are synopsized and compared to previous research (10.5).

## **10.2 Previous work on laughter and humour**

#### **10.2.1 Why we laugh: Triggers and social function**

There are many different reasons why people laugh. For instance, people may laugh because they find themselves at an advantage, or in a position superior to other people, or because there is an incongruity resulting from what was expected and what actually happens or what is said (Glenn 2003: 19). Alternatively, people may laugh when experiencing relief from psychological tension or a threat (2003: 22). As laughter can fulfil a variety of functions, ranging from protecting the recipient's face to expressing surprise or embarrassment (see Brock 2008: 545 for an overview), laughter may even occur when difficult subjects are discussed. In a study of conversations in which troubles were addressed, Jefferson found the recurrent phenomenon of speakers describing their troubles and then laughing, while the recipients often (though not always) reacted with a serious response. She explains this as follows:

It appears that in troubles-talk, a laughing troubles-teller is doing a recognizable sort of job. He is exhibiting that, although there is this trouble, it is not getting the better of him; he is managing; he is in good spirits and in a position to take the trouble lightly. He is exhibiting what we might call 'troubles-resistance.' But this does not mean that […  ] a recipient is invited to join in the merriment, to also find the thing laughable […  ]. (Jefferson 1984: 351)

The same may apply to uncomfortable subjects, such as taboo topics. For instance, obscene or explicit utterances are sometimes distorted with laughter, which is probably not coincidental (Jefferson 1985: 33).

It appears that social elements play a particularly important role in making people laugh. As Glenn points out, "social variables such as whether others are present, whether others are laughing, who they are, what is going on between participants, what the laughable<sup>1</sup> is about, and so forth" (2003: 26) can have a considerable effect on laughter. Hence, a number of studies have observed that people's laughing behaviour is affected by the presence of others. In natural group interactions, Mehu and Dunbar observed that people tended to smile and laugh more when more people were involved in the interaction

<sup>1</sup> Glenn uses this term to designate any referent or stimulus for laughter, or anything for which can be argued that it is intended to draw laughter (2003: 48). He takes laughter as a starting point in his research and then looks for potential triggers (laughables) in its vicinity; I take the same approach in my analysis.

(2008: 1766). Moreover, the participants in Osborne and Chapman's (1977) study, who were exposed to identical humour stimuli, laughed the most when their partners responded with laughter, less often by themselves, and the least in the presence of partners who did not respond to the stimuli with laughter (in Glenn 2003: 26). In addition, laughing behaviour depends on the relationship between the conversational partners, since "facilitation is more likely to occur when the co-participant is a friend, classmate or familiar other; when the other is unfamiliar, inhibition more likely occurs" (Glenn 2003: 27). Thus, laughter often serves to express affiliation with others.

In light of this, it does not come a surprise that various researchers have observed that few utterances which produce laughter are truly humorous (Provine 1996: 42). As a matter of fact, as Glenn points out, "[l]aughter is so inconsistently associated with humor that experimental psychologists have abandoned using it as a reliable indicator that the subject perceives something as funny" (2003: 24). The observation that laughter is often triggered by non-humorous statements was confirmed in Adelswärd's study of laughter in a variety of interactions in institutional settings in Sweden, which showed that speakers "often laugh alone and not always at things considered particularly funny" (1989: 107). Similarly, Clift's examination of video-taped interactions and audio recordings of interviews suggests that "*any* action may be a potential target of laughter" (2016: 86, emphasis in original). These reports highlight that the element of humour may be less central to laughter than its social function.

#### **10.2.2 The organisation of laughter in interaction**

Besides the triggers and functions of laughter, linguists have also studied the organisation of laughter in interaction. Jefferson (1979) analysed the moment in which speakers start laughing and thus aim to provoke laughter in others. According to Jefferson, recipients may start laughing at a variety of recognition points: immediately after utterance completion, after a break, or in response to the prior speaker's laughter. A common technique for inviting laughter is that "[t]he speaker constructs a laugh-specific recognition point by inserting particles of *within-speech laughter* into his utterance" (1979: 82, emphasis in original). Such laughter particles may provide the recipient with clues regarding the humorous nature of the utterance. The recipient then has two possibilities: he or she may either accept the invitation to laugh, or decline it, and instead speak seriously about a topic or remain silent (1979: 84). A third alternative was proposed by Lavin and Maynard, who analysed 13 conversations between survey interviewers and respondents. The authors found that the interviewers often used a smile voice or "quasi-laugh" to react to respondent-initiated laughter, which allowed them to acknowledge and respond positively to their interlocutor's laughter, without interrupting the interview or diverging from the interview script. Such pseudo laughter can serve as a "minimal reciprocation of respondent-initiated laughter" (Lavin and Maynard 2001: 469), and as a compromise between accepting and declining an invitation to laugh.

Laughter can be initiated by the producer of a laughable, or it can be volunteered by someone other than the current speaker. In his research on the subject, Glenn (2003) found that an important factor in determining who laughs first is the number of parties involved in the interaction. He compared two-party interactions (often on the telephone) with conversations among three or more people (mainly face-to-face) and found that, in the case of the former, the current speaker would usually laugh first (87% of the time), while this rarely happened in the multi-party interactions (17%). This may partly be attributed to the type of interaction (face-to-face or on the telephone), yet there are also other potential reasons. Glenn believes that, in multi-party interactions, there is greater flexibility with regard to the manner in which laughter may begin (2003: 89). He postulates that, since speakers usually own their laughables, "there may be a participant bias against current speaker providing the first laugh. Laughing first at one's own designedly laughable materials may be heard as engaging in self-praise" (2003: 91). This can be avoided if, for instance, a participant who is familiar with the laughable initiates shared laughter and thus provides a laughter cue for the other participants (2003: 92). In contrast, speakers are more likely to laugh first in two-party interactions in order to mark the utterance as "not-serious" (2003: 102).

Since the recipient(s) as well as the producer of a laughable may laugh at the same time, a considerable number of laughs tend to overlap and thus create shared laughter. As Jefferson, Sacks and Schegloff put it, "[l]aughter can be an achieved product of methodic, co-ordinated processes, with occasion of laughing together oriented to, produced, extended, as an event in itself" (1987: 159). Shared laughter happens most frequently "at locations where the speaker with the turn invites the partner to join in the common laugh in order to take the turn" (Trouvain 2014: 599). Conversationalists appear to aspire to shared laughter, as it can help them to demonstrate affiliation with one another, and to show like-mindedness (Glenn 2003: 29). Due to this affiliative function, shared laughter can play an important role in intimate relationships.

There are a variety of suprasegmental features that assist speakers in framing their turns as humorous. The existence of such features was first proposed by Bateson (1953), who discovered that speakers tend to mark their actions either as 'play' or as 'serious'. In the case of the former, speakers may create a *play frame* by means of hints and clues, which allows the interlocutor to recognize the humorous nature of the utterance, and potentially provokes laughter. Contextual clues for laughter are, for instance, the "speaker's tone of voice, sudden changes in pitch or rhythm, and paralinguistic clues such as the use of a laughing or smiling voice" (Holmes and Hay 1997: 132). According to Kotthoff, suprasegmental features such as laughter in speech may not necessarily "signal something funny, but [create] a special form of reading (a special frame or keying) for what is said, in the sense of evoking a funny perspective on it. What is offered can be taken lightly" (2006: 274). Coates analysed such humorous talk, or "talk as play", in informal conversations among friends, and demonstrated that laughter served them to indicate playful speech (2007: 45). Through close collaboration, the conversational partners in her study achieved a joint humorous frame (2007: 29). These reports highlight again the social function of joint laughter.

#### **10.2.3 Humour in bilingual and bicultural settings**

People who have different linguistic or cultural backgrounds are often found to differ in what they consider humorous (Ruch and Forabosco 1996; Cheng 2003). Moreover, there may be cultural differences in the manner in which and the extent to which they frame humorous utterances. There can also be differences relating to the tolerance of humour and the frequency of its use (Yue 2011). Cultural differences in humour were, for instance, detected by Yue et al. (2016), who interviewed Canadian and Chinese undergraduate students about their perceptions of humour. They found that the former not only rated humour as more important than the latter, but that they also believed themselves to be more humorous. In contrast, Cheng (2003) found in her study of humorous conversations between Hong Kong Chinese and native speakers of English that the overall key in their intercultural interactions was one of playfulness, and that they jointly created a play frame.

In bilingual interactions, non-native speakers may struggle to understand or produce humour, as a high level of fluency is necessary for certain types of humour. According to Ritchie (2010: 33–34), verbal humour is particularly difficult for L2 speakers to understand, as it relies on the idiosyncratic features of a language, while referential humour is largely based on meaning and may therefore not be as challenging for non-native speakers. Adelswärd and Öberg (1998) explored potential challenges of humour for non-native speakers in authentic international negotiation activities between native and non-native speakers of English. The authors intended to determine if the non-native speakers were at risk of becoming targets of the laughter of the native speakers. They discovered, however, that the aspect of hierarchy was a much stronger factor than whether or not the conversationalists were native speakers. Hence, most joint laughter was initiated by team leaders, while unilateral laughter was more frequent among speakers who were at a lower position in the hierarchy. They conclude from their study that being "a native speaker does not per se present a person with an upper hand" (1998: 425).

In situations where there is no pre-existing hierarchy, however, being a non-native speaker may nevertheless put bilinguals at a relative disadvantage when it comes to producing humour, or create difficulties among conversational partners. This can be seen in Chiaro's (2009) study, in which she examined humorous talk in 59 bilingual, cross-cultural couples based on their responses to a web-questionnaire, and compared the results with qualitative data collected in face-to-face interviews with six couples. It emerged from her data that many couples believe that there are differences in their sense of humour, and that "language itself is an undeniable barrier in the transmission of humour" (2009: 222). The differences between the partners were especially marked among couples including a native English speaker, as they tended to rate their own humour as superior to their partner's. Among these couples, typical features of English humour, such as sarcasm, irony, or understatement, were presented as more sophisticated than other types of humour, and English humour was "frequently taken as a (positive) benchmark of what humor should be" (2009: 223). At the same time, the non-native English speakers in these couples reported that their partner's humour confused them at first, and that they misconstrued it as rudeness at times (2009: 223).

Despite the potential challenges of being humorous as a non-native speaker, using humour in one's L2 can also offer new opportunities. Norrick (2007) proposes that bilinguals may use their two discourse systems to mix them for special effects and to produce humour (when code-switching, for in‐ stance). He also notes that they can overcome potential misunderstandings by "mak[ing] adjustments, slowing down, repeating, defining, and explaining" (2007: 390). Similar adjustments were found by Bell (2007), who examined humour in interactions between native and non-native speakers of English and interviewed them on the topic. She discovered that the conversationalists used several strategies to prevent conflicts, such as avoiding taboo topics, choosing their words carefully, and approaching humour with an attitude of leniency. Despite some difficulties with L2 humour, she found that joking did not seem to harm the relationship between the native and non-native speakers in her study (2007: 42). This underlines that there are strategies available to deal with challenges surrounding the use of humour in bilingual, bicultural interactions, as well as the possibility to utilize the uniqueness of this situation to create humour.

#### **10.2.4 Humour in intimate relationships**

Various studies have noted the importance of humour in intimate relationships. Ziv questioned 51 married couples on the role of humour in their relationships and found that more than 90 percent of them considered humour important in their marriages (1988: 228). The social function of humour, especially through humorous comments understood only by the partners, was the feature men‐ tioned with the highest frequency, which might be indicative of the couples' use of humour to establish intimacy (1988: 228).

Moreover, several studies have suggested that couples tend to share a similar sense of humour. Hahn and Campbell (2016) analysed the degree of similarity in the humour styles of 116 married couples, and found that — according to their self-reports — the couples had similar styles. Barelds and Barelds-Dijkstra also discovered parallels in couples' sense of humour when they questioned 114 couples in long-term relationships and examined the connection between humour and relationship quality. According to the couples' self-reports, however, the similarities in their humour were not related to the quality of their relationship (2010: 447). In their analysis of the humour appreciation among 100 married couples, Priest and Thein found that the appreciation for the humour categories they had defined was similar among the couples, but they did not find a significant relation between the couples' similarity in humour appreciation and their marital disaffection, or emotional detachment (2003: 73). There was also no correlation between the level of similarity of the couples' humour and the duration of their relationship, which, in the opinion of the authors, may indicate that similarities in humour appreciation do not evolve over time, but that instead, partners tend to be selected partly based on a similar taste in humour (2003: 71). It is possible that, due to such a selection, the differences in the humour of the couples in this study are less pronounced than they would be between bilinguals with the same backgrounds who are not in a romantic relationship.

#### **10.2.5 Humour and gender**

Gender has often been found to influence people's laughing behaviour, as well as their production and reception of humour. Many of the assumptions relating to such gender differences have been empirically researched and critically examined. One common observation is that men produce a greater number of humorous statements than women, and, consequently, are perceived as funnier. Robinson and Smith-Lovin (2001) studied the production of humour in discus‐ sion groups with different gender compositions and found that men produced more humorous statements in mixed-gender groups than women did. However, all-female groups produced the most humour of all group compositions (2001: 137). Though they did not study spontaneous speech, Greengross and Miller's (2011) survey supports the hypothesis that men produce more humour. The authors asked 400 male and female undergraduate students to produce captions for cartoons. Not only were the men's humour captions rated funnier than the women's, but the former also produced a significantly larger number of captions than the latter. It is conceivable that there is a tendency for men to produce more humour than women do as "appearing witty seems more central to a male personal identity than to a female identity" (Hay 2000: 33), even though these studies highlight that the genders' use of humour is also connected to the composition of the group.

Another common observation is that females laugh more than males during interactions. Adams and Kirkevold, who observed conversations in fast-food restaurants, discovered that women smiled and laughed more frequently than men (1978: 118). Similarly, women laughed more often than men during conflict negotiations among couples examined by Winterheld, Simpson and Oriña (2013: 501). It appears that women laugh especially frequently in mixed-gender groups, as Mehu and Dunbar inferred from observing smiling and laughter in natural group interactions. They found that the men's laughter was not influenced by the group composition, yet women laughed more in mixed-gender groups than they did in all-female groups (2008: 1757). In addition, young women laughed more than more mature women in mixed groups, while no age-related differences were observed for men (2008: 1758). This suggests that factors such as age and group composition can have a different effect on laughing behaviour according to gender.

The third common observation is that women laugh more at men's humorous statements than the reverse. Adelswärd observed such a tendency in various interactions in an institutional context. She discovered that female applicants laughed along with the interviewer more frequently during job interviews, and also laughed alone more often than male interviewees. The female speakers laughed more often overall, as they partook in 59% of all laughing events, while men only featured in 41% of them (1989: 122). When observing young adults in public, Provine (1993) discovered that male speakers triggered laughter more frequently in both male and female interlocutors than female speakers did. The same pattern was found in a later study, in which Provine examined 1,200 natural conversations between men and women (1996: 42). This may be linked to gender differences in the receptiveness to humour, as Jefferson (2004) found when examining gender patterns in the laughter of one woman and different male interlocutors in telephone conversations. Her data indicates that the woman tended to join in with men's laughter, thus exhibiting "laugh-re‐ ceptiveness", whereas the men were less likely to laugh along, and hence showed "laugh-resistance" (2004: 125). However, there was an exception to this, namely when laughter accompanied trouble-talk; in these situations, the men were more likely to laugh along than the woman (2004: 125). In contrast to these findings, Glenn, Hoffman, and Hopper (1996) discovered no statistically significant difference in speakers' likeliness to laugh along with each other in twelve male-female interactions. Their data suggested, however, that whether the participants laughed along with each other or not depended on the nature of their relationship. In courtship-relevant interactions, female participants were more likely than male participants to respond with laughter to their interactional partner, yet no such tendency was noticeable in interactions which were not courtship-relevant (in Glenn 2003: 156). This ties in with Martin's observation that women find men who make them laugh especially attractive, while men feel attracted to women who respond to their jokes with laughter (2014: 145).

The fourth hypothesis is that men and women have different styles of humour and appreciate different types of humour. It is generally assumed that women have a more supportive or cohesive humour style, while men use more compet‐ itive or differentiating humour; this hypothesis has been confirmed by a number of studies. Robinson and Smith-Lovin (see above) found that, in mixed-sex groups, men and women used cohesive humour with a similar frequency, yet men used differentiating humour more often. In same-sex groups, women used more cohesive and less differentiating humour than men (2001: 144). Holmes' (2006) analysis of humour and gender in everyday work interactions in New Zealand also revealed gender differences in humour type. Like Robinson and Smith-Lovin, she found that the type of humour used by both genders partly depended on the gender composition of the participants of the meeting. Hence, supportive conjoint humour "tended to be more frequent in groups that included women (whether the groups were single or mixed-gender). Contestive conjoint humour, on the other hand, tended to occur more frequently in meetings in which men were involved (again, whether single or mixed-gender)" (2006: 40). There were also differences with regard to collaboration, as groups involving only or mainly men used a minimally collaborative type of floor most often, while a maximally shared floor was most frequent in situations where both genders were present (2006: 41).

These differences in humour styles are probably linked to the different functions humour is said to fulfil for the two genders. Based on observations of humour in all-female and all-male groups in a variety of social contexts, Coates demonstrated that the female speakers in these groups used humour as a form of self-protection, while the male speakers used it to exert dominance. At the same time, both genders collaborated to create humorous talk, and, thus, solidarity (2014: 164), which can be seen to contradict the results of earlier studies. Similarly, Hay analysed the function of humour in conversations among 18 friendship groups, including all-female, all-male and mixed-gender groups, and discovered that the women in these groups used humour to create or maintain solidarity with a far greater frequency than the men did (2000: 734).

In addition, Hay also discovered differences in the teasing behaviour of the sexes. While the men and the women in her study both appeared to enjoy teasing, teasing tended to occur far more frequently in single-sex groups than in mixed-sex ones (2000: 736). Hay's results are not entirely congruent with those presented by Lampert and Ervin-Tripp (2006), who examined conversational humour in naturally occurring conversations of mixed- and same-sex groups of friends, and found that men teased less frequently in mixed groups than in same-sex groups, while women teased more in mixed groups, but only infrequently in same-sex groups.<sup>2</sup> Based on this, the authors hypothesize that

[m]en may avoid teasing a female friend, even when teased, because of the social prohibitions on aggressive behavior by men towards women in friendship groups, sensitivity towards maintaining symmetrical power relationships within the group, and recognition that women view being teased more negatively than men. (2006: 66)

Furthermore, Lampert and Ervin-Tripp believe that women may adopt a male conversational style with their male friends. The men in their mixed groups often responded to women's teases "with a supportive self-directed remark,

<sup>2</sup> It is possible that the differences between these findings are partly due to cultural norms, as Hay (2000) examined young New Zealanders of European descent, while Lampert and Ervin-Tripp (2006)'s research was on European Americans.

which the women could easily interpret as acceptance, a sign of bonding, and encouragement to tease even more" (2006: 67). These findings suggest that the gender of the interlocutor may be just as important as the gender of the speaker when it comes to laughing and humour.

#### **10.3 Humour and laughter during the interviews**

#### **10.3.1 Frequency of laughter during the interviews**

I will now turn to the analysis of the couples' use of humour and laughter during the interviews. My first aim is to examine the frequency with which the bilinguals laughed during the interviews in order to ascertain whether the partners displayed similar behaviour patterns, and whether their gender and mother tongue influenced their laughing behaviour (see sections 10.3.6, "Laughter and gender", and 10.3.7, "Laughter and mother tongue"). The duration of each instance of laughter was indicated by marking each pulse of laughter with an "at" symbol (@) in the transcripts, rather than a time measurement in seconds. In this manner, the number of laughter pulses each participant produced was established, as well as the number of instances in which laughter occurred, or laughter episodes (which consist of one or several pulses of laughter). It should be kept in mind that there may have been instances where a speaker smiled or laughed inaudibly, which would not appear in the transcript.

Table 37 below indicates the number of laughter pulses (@), the number of laughter episodes, as well as the average laughter length (number of laughter pulses per laughter episode) for each bilingual. In addition, I calculated the average number of laughter pulses and laughter episodes for each minute of interview time, to give some indication of the relative frequency. This was deemed preferable to putting the laughter pulses in relation to the number of words or intonation units for each speaker, since laughter can, to some extent, be seen as dissociated from speech.


Table 37: Number of laughter pulses, laughter sequences and average laughter length for each participant

Key: @ = laughter pulse; duration (min) = interview duration in minutes; bold = in text.

Table 37 indicates that there are considerable differences in the frequency and duration of laughter among the interviewees. The bilinguals averaged at 503.1 laughing pulses — or 7.45 per minute — with a large standard deviation (σ = 302.3). The number of laughing pulses ranged from 154 (2.2 per minute, Robert <sup>E</sup> ) to 1079 (18.3 per minute, Richard E) and 1254 (17.0 per minute, Susanne SG). It is notable that one couple, Susanne and David, laughed considerably more than the other couples — together they produced 30.4 [13.4 + 17.0] laughter pulses per minute (2244 [990 + 1254] in total), which is twice the average per couple (14.9 [2 \* 7.45] laughter pulses per minute, 1006.2 [2 \* 503.1] in total) and 4.7 times as much as the couple who laughed the least, Claire and Simon (6.5 [3.1 + 3.4] laughter pulses per minute, 476 [225 + 251] in total). It is also evident that most partners differed considerably from each other in the number of laughter pulses, and thus the duration of laughter during the interviews — on average, one of the partners used 253.8 pulses (83%) more than the other one, and only two couples differed less than 20% (Claire/Simon, Karen/Philipp).

The variance between the speakers regarding the number of laughter epi‐ sodes is almost as great as it is for the number of laughter pulses. The speakers ranged between 56 instances (Karen <sup>E</sup> ) and 222 instances (Susanne SG), with an average of 108 instances, and a standard deviation of σ = 52.9. On average, one partner in each couple laughed 40.1 instances (42.4%) more than the other one. In relative terms, the participants laughed 1.60 times per minute, ranging between 0.77 (Tim <sup>E</sup> ) and 3.71 (Richard <sup>E</sup> ) laughter episodes per minute (σ = 0.83). Regarding the average length of each laughter sequence (laughter pulses per laughter episode), the numbers were somewhat more homogenous, with an average of 4.66, and a standard deviation of σ = 0.88. The average difference between the partners in each couple was 1.1 laughter pulses per laughter sequence. While the duration of laughter was similar for half of the couples, the other half displayed considerable differences in their average duration of laughter per episode (with the one speaker laughing between 25% and 82% longer than the other).

In sum, there are vast differences with regard to the frequency and duration of laughter among the bilinguals, in spite of the fact that (largely) the same topics were discussed in the interviews. Moreover, the interviewees show considerable differences from their partners in the overall number of laughter pulses, the number of laughter episodes, as well as the duration of laughter, even though they were participating in the same conversation. These results will be discussed more extensively with reference to gender and mother tongue at a later point in this chapter (see sections 10.3.6, "Laughter and gender" and 10.3.7, "Laughter and mother tongue").

#### **10.3.2 Categorisation and overview of laughter triggers**

#### **10.3.2.1** *Overview of laughter triggers*

Besides the frequency with which the interviewees laughed, I am interested in the reasons for their laughter. On the one hand, I intend to determine who caused whom to laugh; on the other hand, I want to identify what type of utterances (humorous, positive or negative) were followed by laughter the most frequently. For this purpose, I examined all instances in which a speaker produced laughter, tried to establish the utterance or event that triggered this laughter, and categorized each laughter episode based on the predominant trigger. In this section, I will give a broad overview of the frequency with which these triggers provoked laughter in the bilinguals, before looking at each trigger type individually in the following sections. The number of laughter episodes that were provoked by each trigger are indicated in Table 38 below. In addition, Figure 9 provides a visualization of the distribution of laughter triggers overall. It should be noted that the numbers refer to sequences of laughter of various durations, ranging from one pulse of laughter to prolonged laughter. Moreover, during longer sequences with several humorous comments, multiple triggers sometimes immediately followed each other. When prolonged laughter could be attributed to several triggers, it was counted as several instances of laughter (1 for each trigger).


Table 38: Laughter triggers for each participant

Key: part = partner; int = interview situation or interviewer; pos = positive; neg = negative; s / p = self and/or partner; bold = in text.

As can be seen in Table 38, the first aspect that I examined was whose utterance triggered each laughter episode. This was either the speaker him- or herself ("self"), his or her partner ("partner"), the interviewer or interview situation ("interview"), or both participants jointly ("both"). In the last category, I also included laughter that may have been triggered by either of the participants but could not be assigned clearly to either. Table 38 and Figure 9 indicate that, overall, laughter was triggered most frequently by the seaker him- or herself (1001 of 2159 instances, 46.4%). Instances which were triggered by the laugher's partner were somewhat less frequent (800 instances, 37.1%). This is not surprising, considering humorous utterances are often framed by laughter in order to make clear that a statement is to be interpreted in a humorous key (Kotthoff 2006: 274). In addition to these two main groups, approximately a seventh of the laughter was triggered by the interviewer or the interview situation (318 instances, 14.7%). Finally, a comparatively small percentage of laughter was attributed to a joint or unclear trigger (40 in-

**Figure 9.** Visualization of laughter trigger distribution Figure 9: Visualization of laughter trigger distribution

As can be seen in Table 38, the first aspect that I examined was whose utterance triggered each laughter episode. This was either the speaker him- or herself ("self"), his or her partner ("partner"), the interviewer or interview situation ("interview"), or both participants jointly ("both"). In the last category, I also included laughter that may have been triggered by either of the participants but could not be assigned clearly to either. Table 38 and Figure 9 indicate that, overall, laughter was triggered most frequently by the speaker him- or herself (1001 of 2159 instances, 46.4%). Instances which were triggered by the laugher's partner were somewhat less frequent (800 instances, 37.1%). This is not surprising, considering humorous utterances are often framed by laughter in order to make clear that a statement is to be interpreted in a humorous key (Kotthoff 2006: 274). In addition to these two main groups, approximately a seventh of the laughter was triggered by the interviewer or the interview situation (318 instances, 14.7%). Finally, a comparatively small percentage of laughter was attributed to a joint or unclear trigger (40 instances, 1.9%).

The second aspect that I included in the categorisation was whether laughter was caused by a humorous comment, a positive utterance, or a negative one. Figure 9 above shows that the majority of all laughter sequences (62.3%) occurred in a humorous context (triggered by oneself [31.9%], one's partner [20.8%] or the interviewer [9.6%]). This was followed by laughter which accompanied a negative utterance (14.8%), either by the laugher him- or herself (7.5%), or his or her partner (7.3%). Moreover, 8.6% of the laughter was triggered by a positive utterance (3.9% self and 4.7% partner), and 5.1% of the laughter sequences were due to an uncomfortable topic, or the interview situation in general. Finally, 6.8% [3.0% + 3.8%] of all laughter could not be assigned to a trigger with certainty and 0.5% were considered involuntary laughter triggers.

I believe that two aspects of these findings are particularly interesting. First, the results underline that laughter is, with surprising frequency, not actually associated with a humorous statement — in 37.7% of the laughter sequences during the interviews, the trigger was not deemed to be humorous. While this percentage might be different in another type of conversation, it is nonetheless remarkably high. Secondly, the results suggest that, in a non-humorous context, it is more common for laughter to accompany negative than positive statements. Laughter was triggered more than twice as often by a negative statement or an uncomfortable topic (14.8% + 5.1%, thus 19.9% in total) than by a positive comment (8.6%). In the following, I will discuss the individual triggers in more detail, and provide examples of each trigger type.

#### **10.3.2.2** *Humorous laughter triggers*

As the previous section showed, the type of utterances which provoked laughter the most frequently were humorous statements or laughables ("humorous"). In these cases, laughter preceded, accompanied, or followed a laughable. Humorous statements occurred in all categories (self, partner, interviewer), but they were not distinguished in the fourth category ("both"), since the trigger was usually not entirely clear in this category. Example 10.1 shows a typical humorous statement (underlined), which resulted in laughter from both partners:

```
Example 10.1: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
Martin: <HSK so I I ^wouldn't know how I HSK> _ 
             <P<LO ^um:: LO>P> _
             … how it would have been <HI ^different HI> with uh: \
             <P<HSK with a <HI ^Swiss HI> girlfriend HSK>P> _
             <H:>
Courtney: … <P<HSK I'd say [it'd be ..] ^dull HSK>P> /
Martin: [<PP couldn't ^say PP>] //
Courtney: <P<LO ^no LO>P> \~
             <P<LO ^no LO>P> \~
             [@@@]@@@ ((claps hands))
Martin: [@@@] (939–943)
```
Interestingly, far more instances of laughter were provoked by the laughers' own humorous comments (689 instances, 31.9%) than by their partners' (448 instances, 20.8%). In addition, there were various situations in which a humorous comment by the interviewer triggered laughter (208 instances, 9.6%). It is noteworthy that few of the humorous laughables would generally be classified as truly hilarious. Rather, the majority of these triggers were somewhat funny remarks which were presented in a humorous key.

#### **10.3.2.3** *Positive and negative laughter triggers*

Although humorous comments constituted the majority of all laughter triggers, it is notable that more than a third of all laughter episodes were provoked by non-humorous statements. Notably, speakers often framed positive statements, for example about their partner or their relationship, with laughter. In example 10.2, Sophia accompanies a positive comment about her husband with laughing voice quality; towards the end of her statement, her speech is interspersed with laughing particles (underlined):

```
Sophia: <W when you <HI ^meet HI> him you know it W> \
           … <W from the very be^ginning W> \\
           <@<HSK that he's [such a ni@@ce ^gu:@y] HSK>@> /~
Richard: [@@@@] (123–124)
```
There were 85 instances in which a speaker framed a positive statement with laughter (3.9% of all instances), and 101 instances in which laughter was triggered by a partner's positive statement (4.7%), while no such instances occurred in the other settings.

Even more frequently, laughter was used to frame or mitigate (non-hu‐ morous) negative statements such as criticism or disagreement, or accompanied a reference to a conflict or negative past event.<sup>3</sup> In example 10.3, Simon criticizes his partner for never apologizing to him, and then starts chuckling:


The statement above is preceded by the mention of several other points of conflict, such as that Claire "huffs" very quickly (Simon 1630), that he is much more forgiving than her (Simon 1634), or that she does not want to talk things out but wants to argue instead (Claire 1635). Simon's chuckling at the end of this list appears to be an invitation to start depicting their conflict less negatively. His partner takes him up on this, as she switches to a smiley voice (<:-)>) and frames the issue in a humorous key, which results in shared laughter:

```
negative events, which were assigned to the first category ("humorous"). Example 10.2: Richard (English, 29) and Sophia (Swiss, 31)
Example 10.4: Claire (Northern Irish, 28) and Simon (Swiss, 34)
Simon:
Claire:
Simon:
                  @@@@ ((chuckling softly))
                  <P<A ^ninety percent of the time A>P> \
                  <P<:-) ^Simon is: .. apologizin' :-)>P> _
                  <P<A sayin' ^sorry an' I'm like A>P> _
                  @@ [@@@@] ((softly))
Silja: [@@@@@] 
Claire: [<H::> <P<A<:-) ^thank you :-)>A>P>] //
```
Claire: [**@@@** ((softly)) <P<:-) ^yeah but <%%> <m> <%%> :-)>P>] \_

Silja: [**@@@@@**]

Simon: [**@@@**] ((softly)) (1644–1651)

<sup>3</sup> Note that this category does not include humorous comments about conflicts or

Claire: [<H::> <P<A<:-) **^thank you** :-)>A>P>] //

Silja: [**@@@@@**]

 [**@@@@**] ((chuckling softly)) Claire: <P<A **^ninety percent of the time** A>P> \

```
Silja: [@@@@@] 
Claire: [@@@ ((softly)) <P<:-) ^yeah but <%%> <m> <%%> :-)>P>] _
Simon: [@@@] ((softly)) (1644–1651)
```
<P<:-) **^Simon is: .. apologizin'** :-)>P> \_ <P<A sayin' ^sorry **an' I'm like** A>P> \_

Such laughter also occurred frequently when one participant disagreed with the other during the interview. In the following example, Robert and Stephanie are asked to decide which one of them is a more positive person, and are in disagreement on the matter:

```
Example 10.5: Robert (English-Irish, 20) and Stephanie (Swiss, 24)
1335 Silja: an:d who's the more ^positive person HI> \
1335 Robert: <W<CRE<HI I'd #have HI> say ^I am CRE>W> \\
               ^no //~
1337 Stephanie: = @@@ ((loud laughter)) 
               <@<HI I think ^I@ a@@m HI>@> \ 
               @@@[<H::>] 
1338 Robert: [<PP <%m> PP>] --
1339 Stephanie: <@<HI ^^positive HI> person @>_
1340 Robert: = <:-)<HI yeah HI> I think ^I'm like :-)> \
               <:-) I've got the most positive <HI ^outlook HI>:-)> \\
               <:-) ^no :-)> //~
               .. @[@@] 
1341 Stephanie: [@@]@@
1342 Robert: @ <HI<@ #al- ^alright# I- @>HI> _
               <:-)<W<HI I'll argue HI> ^^my <P case here P>W>:-)> //~
1343 Stephanie: <:-)<P ^okay P>:-)> \/~ (1335–1343)
```
Example 10.5 shows that Stephanie and Robert both consider themselves to be more positive than their partner. Stephanie bursts into loud laughter when Robert states that he is more positive than she is. She openly disagrees with him, framing her disagreement with laughing voice quality and laughing particles ("I think I@ a@@m", 1337). Robert insists in a smiling voice quality that he has "the most positive outlook" (1340), yet seeks confirmation about this from his partner (using the question tag "no"); when she does not give an affirmative response, he starts laughing, as does she. In the instances of laughter in this example, there are no intentionally humorous comments, but laughter results from the disagreement between the partners, and, potentially, their desire to frame this more positively. As was the case for positive statements, laughter following negative statements or disagreements was only differentiated for selfand partner-triggered laughter. There were 163 instances (7.5% of all) in which a speaker framed his or her own negative statement with laughter and 158 instances (7.3%) in which laughter followed a partner's negative statement.

#### **10.3.2.4** *Other laughter triggers*

Some laughter could be assigned to an utterance by one of the partners, but the exact reason for laughing was not entirely clear. These laughter triggers were thus grouped in another subcategory ("neutral / unclear"). An instance of such a trigger is given in example 10.6:


In this extract, it is not clear why Philipp chuckles; it may be because of the longer pauses, or something he recalls (but does not elaborate). It is equally possible that he starts laughing because of a visual stimulus, such as his wife's smile. In 64 instances (3.0% of all instances), the reason for self-triggered laughter could not be determined for sure, and in another 82 instances (3.8%), the reason for partner-triggered laughter was not entirely clear.

In some cases, an utterance seemed to provoke (humorous) laughter invol‐ untarily, or there was unintentional situational humour (11 instances, 0.5%). This was usually due to a sudden language switch, a mispronounced word, a slip of the tongue, a misunderstanding, or an unexpected question:

```
Example 10.7: David (English, 32) and Susanne (Swiss, 29)
David: <PP<A ^anyway A>PP> \
           <P we di^gress P> /
Susanne: <HI what was the HI> ^question _
David: @@@@ (764–766)
```
In these situations, it was usually the partner rather than the speaker who started laughing first; sometimes the producer of the involuntary trigger joined in or switched to a smiling or laughing voice quality.

In addition, there was some laughter which was caused by an uncomfortable topic or question, or the interview situation in general (110 instances, 5.1%). Such laughter has also been observed by Lavin and Maynard, who discovered that respondents often initiated laughter after an interview question, which they termed "question-oriented respondent laughter" (2001: 460). In my study, this was usually nervous laughter, which occurred particularly frequently at the beginning of the interview, or right after I had asked a question, as in example 10.8:

```
Example 10.8: Robert (English-Irish, 20) and Stephanie (Swiss, 24)
Silja: how did the two of you ^meet /
Stephanie: … <:-) ^you @ :-)> //~
              @[@@@@@] 
Robert: [@@@@] 
              <P<:-) no go a^head :-)>P> \\
Stephanie: = <P<HI<:-) no ^you :-)>HI>P> /\~ 
Robert: <P<:-) but you're ^better <LO at it LO>:-)>P> _ (1–5)
```
The extract above shows the very first question of the interview; the negotiation between the partners about who should start speaking indicates that they were still feeling somewhat uncomfortable or reserved.

Finally, there were also instances in which laughter was triggered jointly by the couple, or in which it was not clear whose utterance provoked laughter. A joint trigger can be seen in example 10.9, in which Susanne is telling the story of how she met her parents-in-law. Due to his clothes, she mistook her future father-in-law for a handyman, and thought that a bank manager who was present was David's father:

```
Example 10.9: David (English, 32) and Susanne (Swiss, 29)
```

```
1234 Susanne: <PP (David's father was wearing) a ^blue: PP> _
               <PP ^overall so I thought PP> _
               <PP maybe he was ^just PP> _
               <P in the <LO<CRK ^kitchen: an:d CRK>LO>P> _
               … 1[the <LO<@ ^plumber maybe @>LO> @@@] //
1235 David: 1[<HI but ^that'd be my ^da:d HI> @@@] \\~
1236 Susanne: 2[<@ so <HI that was his HI> ^da:d @>] \\ 
1237 David: 2[@@@@] 
1238 Silja: 2[oh <HI ^really HI>] \
1239 Susanne: 3[<:-) and the <HI other one was the HI> ^bank :-)>] \\
1240 David: 3[@@@@] 
1241 Susanne: 4[<@ ^bank guy @> @@@] \
1242 David: 4[<@ yeah the other one was the ^bank manager @>] _ (1234–1242)
```
Susanne introduces the story and builds up towards the punchline (i.e. mistaking the father for a plumber), yet David interjects and delivers the main punchline for her, which she then repeats after him. Subsequently, Susanne relates another part of the anecdote (i.e. mistaking the bank manager for his father), which her husband repeats after her. Due to the partners' joint story-telling, most of their laughter cannot be attributed to either of the partners with certainty. Laughter provoked by joint or ambiguous triggers was relatively rare, and only 40 of all instances of laughter (1.9%) were assigned to this category. Of these, 31 were triggered jointly or by either of the speakers, and another 9 were due to misunderstandings or conversational breakdowns (e.g. interruptions). While joint triggers were infrequent, there were relatively frequent occurrences of several laughables being produced successively by the partners.

#### **10.3.3 Distribution of trigger types among the couples**

In the previous section, I examined the average distribution of different triggers for laughter among all interviewees. In the following, I will turn my attention to the distribution of the main laughter triggers for each participant, so as to compare the distribution among each couple. Table 39 below summarises the percentage of laughter episodes provoked by each main trigger type for each speaker.


Table 39: Distribution of laughter triggers for each participant (% of all laughter episodes of each participant) Key: bold = in text.

As can be seen in Table 39, there are considerable similarities in the distribu‐ tion of laughter triggers among the interviewees. For all of them, humorous comments were the most frequent trigger for laughter, and for all but two, Joshua <sup>E</sup> and Deborah SG, laughter in a negative context was more common than in a positive context. Moreover, it appears that the partners resemble each other in their relative distribution of laughter among the categories. To give some examples: Robert <sup>E</sup> and Stephanie SG both have a relatively low percentage of humorous instances of laughter (44.1% and 42.9%, compared to 62.3% on average), but high percentages in the positive (11.9% and 16.1%, compared to 8.6% on average) and negative context (30.5% and 25.9%, compared to 14.9% on average). In contrast, Craig <sup>E</sup> and Katia SG both have a very high percentage of humorous laughter (79.8% and 73.8%), but low percentages both for positive (2.1% and 1.4%) and negative statements (7.4% and 7.6%). What is apparent in the case of Joshua <sup>E</sup> and Deborah SG is that both of them used laughter to frame positive statements very frequently (14.3% and 14.6%, compared to 8.6% on average). Only one couple — Richard <sup>E</sup> and Sophia SG — differed considerably in their distribution of laughter triggers. Richard used laughter to frame positive and negative statements much more frequently than his partner did (19.6% and 23.7%, compared to 7.1% and 11.8%). As Richard laughed far more often than his partner during the interview, the difference is even more marked when absolute numbers are considered (43 [19 + 24] and 52 [17 + 35] instances, vs. 6 [3 + 3] and 10 [6 + 4] instances; see Table 38 in section 10.3.2.1, "Overview of laughter triggers").

The similarities between the partners with regard to laughter triggers could be due to a number of reasons. On the one hand, the fact that the partners were participating in the same conversation may be relevant, as it is possible that, for instance, more humorous statements were made during some of the interviews, while negative topics were discussed more extensively during others. On the other hand, it is conceivable that partners tend to laugh in response to similar triggers, either because they have a similar sense of humour, or because they have assimilated their laughing behaviour to each other over time.

#### **10.3.4 Use of laughing and smiling voice quality**

In addition to the triggering of laughter and the frequency of laughter, I am also interested in the couples' use of laughing and smiling voice quality. Since the interviews were not recorded visually, smiling itself was disregarded, but smiling voice quality was nevertheless considered. Table 40 below gives an overview of the interviewees' use of laughing voice quality (<@>) and smiling voice quality (<:-)>) during the interviews. Voice quality was indicated for each intonation unit (rather than for each turn or word) in the transcripts. Thus, I was able to calculate the percentage of intonation units which were spoken with a laughing or smiling voice.


Table 40: Use of laughing and smiling voice quality for each participant Key: IUs = intonation units; <@> laughing voice quality; <:-)> smiling voice quality; bold = in text.

Table 40 demonstrates that the interviewees differed greatly in their use of smiling and laughing voice quality. On average, they spoke with a laughing voice quality in 47.2 (or 3.1%) of their intonation units, with a standard deviation of 40.22. To some extent, this high standard deviation can be attributed to one couple, David <sup>E</sup> and Susanne SG, who used laughing voice quality particularly frequently (148 and 172 times, which correlates to 9.2% and 8.8% of their intonations units). In contrast, only 1.0% to 4.6% of the other participants' speech was accompanied by laughing voice quality. While there was great variance overall, most interviewees tended to use laughing voice quality to a similar extent as their partner; on average, the difference between the partners in the amount of speech with laughing voice quality is only 1.13 percentage points (σ = 2.20). Another aspect that becomes apparent is that the person who laughed the least during the interviews (Robert <sup>E</sup> ) also used the least laughing voice quality, while the couple who laughed the most (David/Susanne) used it the most. This could suggest that laughing voice quality tends to function more often as a gateway to actual laughter (as in example 10.9 above), rather than a substitute for laughter.

Smiling voice quality was more frequent than laughing voice quality; it was used almost three times as often. The variation among the participants' use was rather high as well, with an average of 138.0 instances per person, and a standard deviation of σ = 82.33. The difference in the partners' use of smiling voice quality was greater than in their use of laughing voice quality, and there appears to be little assimilation in this regard. On average, 8.8% of the participants' speech was accompanied by smiling voice quality, and the partners differed 4.09 percentage points from one another (σ = 3.79). Moreover, there seems to be no positive correlation between the interviewees' use of smiling voice quality and how much they laughed, or used laughing voice quality. On the contrary: The speakers with the highest percentages of smiling voice quality (Robert <sup>E</sup> , Richard E, Joshua <sup>E</sup> ) did not use much laughing voice quality. Robert, who laughed and used laughing voice quality the least, accompanied the highest percentage of intonation units with smiling voice quality. These results suggest that smiling voice quality functions very differently from laughing voice quality. The latter appears to be connected to laughter, as people who laugh a lot also use more laughing voice quality. In contrast, smiling voice quality does not seem to be directly linked to laughter, but rather seems to be used independently of it, or even to replace laughter. Moreover, it appears to be less "infectious" than laughter or laughing voice quality, resulting in less assimilation between the partners.

#### **10.3.5 Topics provoking laughter**

In addition to the broad triggers discussed so far — humorous, positive and negative utterances — I want to examine what topics caused laughter among the couples. An overview of the topics which triggered laughter, as well as the number of instances of laughter each topic triggered, is presented in Table 40 below. Because the interviews were semi-structured, and many topics were pre-set, the topics which the couples laughed about during the interviews may not reflect precisely what they laugh about when they are on their own. Nonetheless, it gives some indication of the kinds of topics they find funny. Moreover, the analysis can show us differences in what speakers of different genders and mother tongues consider amusing.

In the humorous context, I distinguished 19 different topics that made the interviewees laugh. Most instances of laughter in this category were produced accompanying or following humorous comments about cultural or linguistic differences and stereotypes (147 [91 self-triggered and 56 partner-triggered] of 1137 instances of humorous laughter, 12.9%). Almost as much laughter was triggered by humorous remarks about conflicts or difficulties (131 [75 self and 56 partner], 11.5%). Humorous statements about their relationship and their past were also frequent (100 [64 self and 36 partner], 8.8%), as well as the topic of their relationship language, wordplays, and invented or misused words (95 [56 self and 39 partner], 8.4%; see also section 10.4.4, "Playful language in bilingual couple talk"). Similarly frequent was humorous laughter about oneself (98 [57 self and 41 partner], 8.6%) and one's partner (73 [51 self and 22 partner], 6.4%). Thus, more than half of the humorous laughter in the interviews (56.7%) was triggered by these six topics. These findings indicate that, for the couples, their bilingual, bicultural situation is a source of humour, even when it comes to conflicts resulting from it. Moreover, it is worth noting that, while there were fewer instances of laughter triggered by humorous statements made by their partners than by the laughers themselves (448 [38.4%] vs. 689 [60.6%]), the distribution among the topics is very similar in both cases.



Table 41: Overview of laughter triggers according to content and speaker Key: bold = in text.

In terms of positive statements that were accompanied or followed by laughter, the most frequent ones concerned the bilinguals' relationship (68 [36 self and 32 partner] of 186 instances, 36.6%) and the laughers' partner, family, or culture (57 [28 self and 29 partner], 30.6%). In these situations, the interviewees might have been accentuating their positive feelings by laughing, or might have laughed because they perceived the topic to be awkward or very personal. Interestingly, positive statements about the speakers' own language or culture triggered laughter in their partners far more often (30 of 101 [partner-triggered] instances, 29.7%) than in the speakers themselves (3 of 85 [self-triggered] instances, 3.5%). Possible explanations for this are that the listeners did not agree with the positive assessment (but chose not to mention this explicitly), that they viewed it as indirect criticism of their own language or culture, or that the listeners simply wanted to display positive feelings or agreement with their partner's statement.

As I mentioned earlier, laughter accompanying negative statements was overall more frequent than laughter accompanying positive statements. Most commonly, the interviewees laughed when either of them referred to conflicts or negative events in the past (82 [44 self and 38 partner] of 321 instances, 25.5%). This was followed by general negative statements and indirect or mitigated criticism (50 [18 self and 32 partner], 15.6%) and laughter about disagreement, objections or commands which occurred during the interviews (40 [27 self and 13 partner], 12.5%). In contrast to most of the topics discussed so far, there were substantial differences in the interviewees' responses to their own negative statements and to those made by their partner. On the one hand, the interviewees were more likely to frame their own disagreement or objections with laughter than to laugh in response to their partner's (27 vs. 13 instances); in these situations, the use of laughter may have mitigated the force of a potentially face-threatening act. On the other hand, the interviewees were more likely to laugh when criticism was directed at them than when they voiced criticism about their partner. This was the case for all types of criticism, i. e. direct criticism (24 vs. 11 instances), indirect/mitigated criticism (32 vs. 18) and anticipated criticism (14 vs. 5). In these situations, the bilinguals may have laughed because they were feeling (consciously or subconsciously) uncomfortable or embarrassed in some way; potentially, such emotions are felt more strongly by the recipients of criticism, which might explain the fact that they laughed more frequently.

#### **10.3.6 Laughter and gender**

Because laughter and its triggers have often been deemed to be gendered, I want to investigate if there are any gender patterns in the interviewees' laughing behaviour. Previous research tends to suggest that women produce fewer laughables than men, that they laugh more than men in mixed-gender interactions, that they laugh more at men's laughables than the reverse, and that men and women may have different humour styles (see section 10.2, "Previous work on laughter and humour"). In the following, I will examine the genders' laughing behaviour during the interviews, bearing in mind these observations. In addition, I will also discuss the genders' use of smiling and laughing voice quality, as well as their use of laughter to frame humorous and non-humorous statements. I will look at the overall numbers for both genders, and compare them with the numbers for each combination of gender and mother tongue.<sup>4</sup>

#### **10.3.6.1** *Frequency of laughter among the genders*

Table 42 below lists the gender distribution with regard to laughter pulses, ep‐ isodes of laughter, and length of laughter episodes. Additionally, the bilinguals' laughter was put in relation to the average duration of the interviews.


Table 42: Number of laughter pulses and laughter sequences for each gender and each combination of language and gender (average per person)

Key: SG = Swiss German; E = English; @ = laughter pulse; bold = in text.

<sup>4</sup> The interaction between the factors of gender and mother tongue in the bilinguals' laughing behaviour has been examined more closely elsewhere, and compared with gender patterns that are commonly observed in laughter (Ang-Tschachtli, 2021).

Table 42 indicates that the female participants produced more laughter pulses than the male participants (568.0 vs. 438.2 on average), and the tendency for the female speakers to laugh more than the male speakers is also visible when the factor of their mother tongues is considered (as Swiss German and English-speaking females laughed more than their respective partners, producing 674.1 vs. 496.9 and 320.3 vs. 301.3 laughter pulses). This means that the female participants laughed 29.6% longer than the male participants. These findings provide support for the observation that women tend to laugh more than men in mixed-gender interactions. The difference between the genders is even more marked when the ratio of laughter to words is considered, as the male interviewees spoke more overall, but laughed less than the female interviewees.

In contrast, the difference in the number of laughter episodes produced by both genders was not as marked, as women only laughed 7.4% more frequently than men (111.8 vs. 104.1 episodes of laughter), and this tendency was not visible in the four groups considering their mother tongue. While women may only have laughed on slightly more occasions, though, they laughed longer than men on each occasion. On average, each episode of laughter lasted 5.08 pulses for female participants, compared to 4.21 for male participants. This tendency remains the same in the two couple combinations, as the Swiss and the Anglophone women's laughter episodes were longer than their partners' (5.17 vs. 4.23, and 4.67 vs. 4.15). As a matter of fact, the women's laughter episodes were longer than their partners' for every single couple (see Table 37 in section 10.3.1, "Frequency of laughter during the interviews").

Another aspect that becomes apparent in Table 42 is that the laughing behaviour of the two couple combinations differs considerably. On average, the Swiss German-speaking females and the English-speaking males laughed far more frequently than the other two groups (130.3 and 117.6 vs. 68.7 and 72.7 laughter episodes). Interestingly, this pattern is not just visible in the total number of laughter episodes, but also in the three main subcategories (humorous, positive, and negative triggers; see Table 44 in section 10.3.6.3, "Humorous laughter triggers"). Indeed, across all subcategories, there is a trend for the Swiss German-speaking females and the English-speaking males to rank higher than the other speakers. It seems that, on average, both partners in these relationships laughed more readily than the other couples, about both humorous and non-humorous topics, as well as about interviewer laughables. This demonstrates that the factors of gender and mother tongue may be interlinked in determining laughing behaviour, and it also highlights that the mood of the conversation and the relationship between the speakers have a great influence on their laughing behaviour.

**10.3.6.2** *The genders' use of laughing and smiling voice quality* Table 43 below lists the gender distribution with regard to laughing and smiling voice quality, both as average numbers, and as a percentage of all intonation units for each group.


Table 43: Use of laughing and smiling voice quality for each gender and each combination of language and gender (average per person and % of intonation units)

Key: SG = Swiss German; E = English; <@> laughing; <:-)> smiling; bold = in text.

As can be seen in Table 43, the interviewees' use of smiling and laughing voice quality is distributed across the genders in a similar manner as their laughing behaviour. On average, the female participants used more laughing voice quality (3.6% vs. 2.5% of their intonation units) as well as smiling voice quality (9.5% vs. 8.2%) than the male participants. In both couple combinations, moreover, the female speakers framed a larger percentage of their intonation units with laughing voice quality (4.1% vs. 2.7%, and 2.1% vs. 2.0%) and smiling voice quality (9.9% vs. 9.1%, and 8.4% vs 5.8%) than their partners did. This indicates that gender may play an important role in the use of smiling and laughing voice quality, and that it outweighs the influence of a bilingual's mother tongue. It also becomes evident that the couples consisting of a male Swiss and a female English speaker used less smiling and laughing voice quality than the other couples, which is congruent with their laughing behaviour during the interviews, and suggests that, like their laughing behaviour, their use of voice quality may also be shaped by the flow or 'key' of the conversation and the relationship between the speakers.

#### **10.3.6.3** *Humorous laughter triggers*

A closer look at which speaker and what type of trigger provoked laughter can provide additional insights into the genders' laughing behaviour. In the following, I will therefore consider the number of humorous laughables ("self humorous" and "partner humorous") that triggered laughter for both genders, before discussing non-humorous laughter triggers. Table 44 gives an overview of the distribution of laughter triggers among both genders, as well as among the groups including the factor of their mother tongues. In addition, the numbers are indicated as percentages of all laughter episodes of each group.


Table 44: Laughter triggers for each gender and each combination of language and gender (absolute numbers and % distribution)

Key: pos = positive; neg = negative; s / p = self and/or partner; f = female; m = male; E = English; SG = Swiss German; bold = in text.

Table 44 indicates that the distribution of the laughter triggers is similar among the genders. This may not come as a surprise, as the individual couples tend to show a relatively similar distribution of laughter triggers (see section 10.3.3, "Distribution of trigger types among the couples"). For both genders, and the four groups including the variable of the bilinguals' mother tongue, humorous triggers comprised the largest percentage of laughter triggers in all three categories (self, partner, and interviewer), while negative statements ranked second and positive statements third.

Self-triggered humorous statements were the most frequent stimulus for laughter for both genders and for all groups including the variable of the participants' mother tongue, triggering almost a third of the women's (31.1%) and men's (32.8%) laughter episodes. The male participants framed a slightly higher percentage of their own humorous laughables with laughter, yet in absolute numbers, the female interviewees laughed slightly more often about their own humorous laughables than the male interviewees did (34.8 vs. 34.1 instances on average). Moreover, no clear gender trend is visible with regard to humorous self-triggers in the individual groups including the variable of the bilinguals' mother tongue. In line with their overall laughing behaviour (see sec‐ tion 10.3.6.1, "Frequency of laughter among the genders"), the English-speaking women and Swiss men laughed far less frequently about their own humorous utterances (28.7 and 23.3 instances) than the Swiss women and English-speaking men (37.4 and 38.7 instances).

While gender does not appear to have a great influence on the extent to which the participants framed their own humorous statements with laughter, there could still be differences in the genders' response to humorous comments by their partners. Indeed, the overall numbers (Table 44) suggest that men's humorous laughables were more successful in provoking laughter in their partner than the reverse (24.1 vs. 20.7 instances on average), which seems to lend some support to the common observation that men laugh less at women's jokes than reverse (Jefferson 2004). However, if the individual groups are considered, it also becomes apparent that other factors such as the key of the conversation or the participants' mother tongue (see section 10.3.7, "Laughter and mother tongue") may be more influential than gender in this particular case, as the Swiss women laughed more frequently at their partners' humorous statements (29.4 vs. 21.0 instances), while Anglophone women laughed considerably less about their partners than the reverse (11.7 vs. 20.0 instances). The individual groups hence do not follow the expected gender patterns.

If the bilinguals' laughing behaviour is used to try to determine which gender produced more (successful) humorous statements, it should be borne in mind that neither laughter about one's own laughables, nor about one's partner's, is a clear indicator of the production of laughables; laughter may not coincide exactly with humorous statements, nor does it reveal a great extent about the level of funniness of a laughable. Thus, a person does not necessarily laugh about everything that he or she finds funny, and laughter may be dependent not only on the funniness of a laughable, but also on the individual's inclination to laugh. In this setting, it seems that women were generally more responsive to humour, as — in absolute numbers — they laughed more overall (111.8 vs. 104.1 instances on average), and they also laughed more frequently about their own, their partner's, and the interviewer's (11.9 vs. 8.9) humorous statements. Consequently, it cannot be determined which gender produced a greater number of laughables.

#### **10.3.6.4** *Positive and negative laughter triggers*

There are a few aspects that become apparent with respect to the genders' use of laughter in a non-humorous content (see Table 44 in the previous section). First, the male participants laughed more frequently about positive statements than the female speakers did (on average 10.0 [5.0 self + 5.0 partner] vs. 8.6 [3.5 self + 5.1 partner] instances). This difference can be attributed to a tendency for male speakers to frame their own positive statements with laughter more frequently than for female speakers (on average 5.0 vs. 3.5 instances). Secondly, female speakers accompanied a greater number of their own negative statements with laughter than the male speaker did (9.3 vs. 7.0 instances), and male speakers laughed more frequently about negative statements by their partner than female speakers did (8.9 vs. 6.9 instances); of course, the second tendency may be a consequence of the former. Thirdly, it is notable that the couples consisting of a Swiss female and an English-speaking male, who laughed more in a humorous context (see section 10.3.6.1, "Frequency of laughter among the genders"), also laughed more in a non-humorous context than the other couples. Indeed, they laughed more frequently in all non-humorous categories for which a clear trigger could be determined. In the case of the participants in this study, their laughter caused by non-humorous triggers thus correlated with the occurrence of humorous laughter. These findings underline the importance of taking the context into consideration when one studies the genders' laughing behaviour.

#### **10.3.7 Laughter and mother tongue**

The last aspect I intend to examine in the participants' laughing behaviour is the role of their respective mother tongues. This factor can be expected to influence their laughing behaviour for two reasons. First, their laughing behaviour may be determined by their culture to some extent, as it is, for instance, possible that it is more common among the speakers of one of the languages to laugh about themselves, or to laugh about non-humorous triggers. Second, the fact that one of the partners is a native speaker of the couple language, while the other is not, might influence the number of laughables they produce, and how many of these succeed in triggering laughter. It is to be expected that the native speaker produces more laughables, and that the non-native speaker laughs more at the native speaker's laughables than the reverse.

## **10.3.7.1** *Frequency of laughter and use of laughing and smiling voice quality*

Table 45 below displays the number of laughter pulses, episodes of laughter, as well as laughing and smiling voice quality used by the speakers of both mother tongues. In addition, laughter is indicated per minute of interview time, and voice quality is indicated as a percentage of all intonation units.


Table 45: Use of laughter, laughing and smiling voice quality for each mother tongue and each combination of language and gender (average per person and % of intonation units) Key: SG = Swiss German; E = English; IUs = intonation units; @ = laughter pulse; bold = in text.

The table demonstrates that the factor of the participants' mother tongues has an impact on the frequency of their laughter as well as on their use of laughing and smiling voice quality. Thus, the Anglophone participants laughed considerably less extensively than the Swiss participants, both in terms of the number of pulses of laughter (443.9 vs. 562.3, or 6.48 vs. 8.33 per minute), and their number of laughter episodes (102.9 vs. 113.0 instances, or 1.52 vs. 1.67). A comparison of the individual groups indicates that the Swiss participants in each couple combination produced more laughter episodes (130.3 vs. 117.6, and 72.7 vs. 68.7), but gender appears to have a greater impact than linguistic background when it comes to the number of laughter pulses (674.1 vs. 496.9, and 301.3 vs. 320.3). Overall, the native English speakers also used laughing voice quality less frequently than the Swiss (on average in 2.6% vs. 3.5% of their intonation units). However, they used smiling voice quality similarly frequently (in 8.9% vs. 8.7% of their intonation units). The reason for these differences is difficult to determine, given the variety of triggers and contexts for laughter that have been observed. The observation previously made — that laughing voice quality correlates with laughter, while smiling voice quality does not — still holds.

#### **10.3.7.2** *Laughter triggers*

The participants' mother tongues did not only have an influence on the frequency of their laughter, but also on what triggered their laughter. Table 46 below gives an overview of the average number of laughter triggers for speakers of both mother tongues, as well as for all combinations of mother tongue and gender. In addition, the numbers are indicated as percentages of all laughter episodes of each group.



Table 46: Laughter triggers for each mother tongue and each combination of language and gender (absolute numbers and % distribution)

Key: pos = positive; neg = negative; s / p = self and/or partner; f = female; m = male; E = English; SG = Swiss German; bold = in text.

Table 46 reveals that the overall distribution of the laughter triggers is relatively similar between speakers of both mother tongues and that the two groups primarily differ in their response to their partners' humorous comments. The Swiss German speakers laughed far more often about humorous comments by their partner than the reverse (26.6 vs. 18.2 instances on average). This is also the case in both couple combinations, as the Anglophone partners laughed considerably less frequently about their partners' humorous comments than their partners did about theirs (11.7 vs. 20.0, and 21.0 vs 29.4 instances).

When it comes to laughing about their own humorous comments, the difference between the speakers of both mother tongues is not quite as marked, but the trend is similar. Thus, the native English speakers laughed about their own humorous comments slightly more frequently than the speakers of Swiss German did (35.7 vs. 33.2 instances on average), which was also the case for the individual groups (28.7 vs. 23.3, and 38.7 vs. 37.4 instances on average). In both couple combinations, it was hence the native speaker of English regardless of gender — who framed slightly more of their own laughables with laughter, and triggered more laughter in his or her partner (both in absolute terms and relative to all of their laughter episodes). This suggests that the native speakers produced more laughables, or at the very least more successful ones, and that the bilinguals' mother tongues, the couple relationship and the key of the conversation may play a more important role than gender in the production of and response to laughables.

One reason for this might be that the English speakers, having more linguistic resources at their disposal, were able to produce laughables in their mother tongue that are funnier than those of non-native speakers, or that the former are simply funnier than the latter. Another reason may be that the Swiss participants are more accustomed to Anglophone humour and therefore understand their partner better than their partner understands them, yet they may not be able (or may not want to) reproduce this type of humour themselves. It is also conceivable that the Swiss German speakers were more inclined to respond to someone else's humorous comments with laughter than the English speakers. The fact that the Swiss participants laughed more about the (Swiss) interviewer's laughables than the English speakers (11.5 vs. 9.3 instances on average) would support either of the last two hypotheses.

## **10.4 The couples' reports on humour in bicultural, bilingual relationships**

In this final section of the analysis, I will discuss the couples' thoughts on humour and laughter in their relationships. The discussion includes their assess‐ ments of who is funnier, challenges with regard to expressing and understanding humour in a bilingual, bicultural relationship, as well as particularities of their couple humour.

#### **10.4.1 Perception of each other's sense of humour**

To explore the topic of couple humour, as well as the partners' perceptions of each other's funniness, the couples were asked which of them is funnier, and what they laugh about as a couple. The first question aimed to engender a discussion between the partners, and to ascertain whether they considered the native speaker to be funnier than the non-native speaker, though I was aware that the element of gender might influence their judgement as well. An overview of their answers is provided in Table 47.


Table 47: Perception of each other's funniness Key: E = English; SG = Swiss German.

As can be seen in Table 47, the couples did not automatically name the native speaker of the couple language as the funnier partner; only three native speakers were thought to be funnier by both partners (Robert <sup>E</sup> , David <sup>E</sup> , Dean <sup>E</sup> ), versus one non-native speaker (Philipp SG). Five of the couples agreed that they are both equally funny (Courtney/Martin, Sophia/Richard, Deborah/Joshua, Katia/Craig, Sarah/Tim). One couple could not agree on who is funnier (Claire/Simon). Surprisingly, the couples' perceptions of who is funnier often do not correlate with the person who provoked more laughter during the interview. The couples' responses also suggest that the influence of gender on perceived funniness may be at least as strong as the influence of native tongue or cultural background: In the case of four couples, the female partner named her partner as the funnier one (Stephanie SG, Susanne SG, Monika SG, Karen <sup>E</sup> ); the partners usually did not object to this assessment. Moreover, five of the male interviewees consider themselves funnier (Robert <sup>E</sup> , David <sup>E</sup> , Simon SG, Dean <sup>E</sup> , Philipp SG). Only one male participant, Tim <sup>E</sup> , first claimed that his partner was funnier than him; she in turn was surprised by this and contradicted him ("you can't say [that]" [Sarah SG 2145]), and they ultimately agreed that they are both equally funny.

In a number of instances, the couples not only chose the male partner as the funnier, they even raised the question of whether the female partner is funny at all. One partner, Richard <sup>E</sup> , comments on this jokingly, saying that his wife is funny to him, yet that he does not know if other people share his opinion:


Richard's statement is difficult to interpret; his voice quality suggests that he is just teasing his wife. He could also be insinuating that he finds her funny because he loves her, for instance; yet if this is the case, it also means that he is limiting her funniness to his subjective view, rather than attributing a general funniness to her.

The topic of the female partner's inferior or lacking sense of humour also emerges in other conversations, and is propagated by both male and female participants. When asked who of them is funnier, Karen <sup>E</sup> , for instance, claims that her husband "is probably funnier" than her (987), that she is "not very funny" (989) and that she "has to work on" being funnier (991). Her husband does not object to any of her self-deprecating statements and simply chuckles in response. At a later point in the conversation, the couple agree that Karen's allegedly inferior sense of humour is due to her upbringing, as she did not grow up being teased, for instance, and still does not appreciate it at all (unlike her husband; see section 10.4.2, "Challenges with regard to bilingual couple humour"). For Karen and Philipp, but also for Monika SG and Dean <sup>E</sup> , the superiority of the male partner's humour appears to be a fixed part of their couple discourse. Thus, Monika reports that her husband has told her that she is not funny in the past ("he actually says I'm not funny" [1320]), to which Dean responds with laughter, without explaining or mitigating his previous statement. Similarly, when Monika states that he is funnier than her, Dean simply laughs and then even assents to her view ("I won't disagree" [1318]).

The funniness of the female partner, or lack thereof, also causes disagreement between Simon SG and Claire <sup>E</sup> . Claire thinks that she is funny, and enjoys laughing at her own jokes, yet her husband frequently does not respond with laughter:

#### **Example 10.11:Claire (Northern Irish, 28) and Simon (Swiss, 34)**

```
1824 Claire: = <@<W [<HI I HI> laugh at my own] j@okes [a@ll the ti@me] W>@> /~
1825 Simon: [<:-) o^ka:y :-)>] \
              [@@ <@ I ^know @>] \~
1826 Claire: [@@ <@ he ^do@@esn't @> @@] _
1827 Silja: [@@@@@@] 
1828 Claire: <W<@<HI I HI> think I'm hi^larious @>W> //
1829 Simon: [<@<HI this is an HI>@>] \
1830 Claire: [<DIM<W but probably #^he's #more #funny W>DIM>] _ ((sic))
1832 Simon: ((…)) <:-)<P you <LO think your jokes are [hi^larious] LO>P>:-)> \
1833 Claire: [@@@@@] 
1834 Simon: <:-)<P<LO and most of the time [I'm like ^yes] LO>P>:-)> _
1835 Claire: <:-)<P [and they're ^stupid] [like but] P>:-)> \
1836 Simon: <:-)<P<LO [there's it's] o^kay 
             but LO>P>:-)> \\
1837 Silja: [@@@@@@@] 
1838 Simon: … <:-)<PP<LO it's not as LO> ^funny as you think PP>:-)> \\ (1824–1838)
```
In this example, Simon is openly critical of his wife's jokes, yet Claire does not contradict his assessment. While her first statements are lively, accompanied by laughing voice quality and a wide intonation contour (1824, 1828), her voice quality becomes softer over the course of the extract, as she concedes that her partner might be funnier (1830, <DIM> diminuendo) and that her jokes are "stupid" (1835, <P> soft). Simon joins in with a similar tone, in a low soft voice (1834), which becomes even softer (<PP>) as he claims that Claire's jokes are "not as funny as [she] think[s]" (1838). In addition, both partners are framing their statements with smiling voice quality, which might assist in reducing the potential for conflict of the subject. Interestingly, the couple only debate whether Claire is funny or not, while Simon's sense of humour is never challenged. As a matter of fact, the funniness of the male partner was not questioned by any of the female bilinguals. It is also noteworthy that two of the four examples just mentioned include a female native speaker of English, which suggests that being a native speaker of the couple language may play a lesser role than gender in the participants' subjective perception of funniness.

#### **10.4.2 Challenges with regard to bilingual couple humour**

During the interviews, the couples also commented on a number of aspects that they find challenging about expressing humour in a bilingual, bicultural relationship. One reason for this is that humour does not always translate well from one language to another. Thus, various couples lament that jokes are often hard to translate or retell in the other language (Claire/Simon, Sarah/Tim, Courtney/ Martin). Courtney <sup>E</sup> and Martin SG explain that most misunderstandings between them happen in the context of retelling a joke or a humorous situation, as the punchline gets lost and the retelling ends up not being funny:

```
Example 10.12: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
246 Courtney: <W<A when you want to A><HI ex^plai:n HI><LO something or: LO>W> _
               <HSK y- you come ^up with a: HSK> \
               <HSK colloquial ^term or something HSK> _
247 Martin: ^yeah _
248 Courtney:
249 Martin:
               … <HSK<DOW and then all the joke is ^missed DOW>HSK> \
               <HSK<P<LO or ^something LO>P>HSK> _
               = yeah e^xactly \\
               ^jokes /~
               ((…)) you ^want to explain a ^joke //~
               … uh <su> <su:> you <HI ^heard something HI> _
               <W<A [and] somebody told you something ^really funny A>W> //
250 Courtney: [^yeah] \~
251 Martin: <W in uh in Swiss ^German W> //
               <A and then if you want to .. translate it into ^English A> _
252 Courtney: = ^yea:h /~
253 Martin: = <W<A and it <HI doesn't HI> really ^work A>W> /
               <LO because ^um:: LO> \
               … <P<LO yeah it's just the LO> ^different P> \
               … <P [different kind] of uh ^words P> //
```
[**in English** <F<HI ^yeah HI> <# and I get #>F>] /

<W<LO **when you're** LO> **telling the ^joke** W> \~

<sup>258</sup> Courtney: [or .. bu- but <F<HI **you can even**] **^feel it** HI>F> \\

<sup>254</sup> Courtney: [<P #lacking P>] \ <sup>255</sup> Martin: <P that you [would ^use] P> // <sup>256</sup> Courtney: [<P #### P>] \_

<sup>257</sup> Martin: <P **and uh it's not ^funny** <LO in e:h LO>P> \_

<LO ^when you're LO> \_

… <P<LO yeah it's just the LO> ^different P> \ … <P [different kind] of uh ^words P> //

<sup>253</sup> Martin: = <W<A **and it** <HI **doesn't** HI> **really ^work** A>W> /

<A and **then if you want to** .. **translate it into ^English** A> \_

<sup>252</sup> Courtney: = ^yea:h /~

<sup>251</sup> Martin: <W in uh **in Swiss ^German** W> //

```
254 Courtney: [<P #lacking P>] \ 
255 Martin: <P that you [would ^use] P> //
256 Courtney: [<P #### P>] _
257 Martin: <P and uh it's not ^funny <LO in e:h LO>P> _
             [in English <F<HI ^yeah HI> <# and I get #>F>] /
258 Courtney: [or .. bu- but <F<HI you can even] ^feel it HI>F> \\
             <LO ^when you're LO> _
             <W<LO when you're LO> telling the ^joke W> \~
259 Martin: = ^yeah [it's <%>] \
260 Courtney: [you can] feel that [it's just ^^not] <LO<A<:-) gonna be funny
             :-)>A>LO> _
261 Martin: [<A it jus:t ^feels like A>] \
             = it doesn't ^sound right \\
             <P<W<HSK<LO it's: LO> not really ^funny W>HSK>P> \
262 Courtney: <PP ^yeah PP> /~ (246–262)
```
Martin and Courtney are both very engaged in this sequence, which suggests that this is a relevant issue for them. Their involvement can be seen in the manner in which they interact, as there are a number supportive backchannels (*yeah*, 6 times), as well as simultaneous speech ([ ], 6 times) and latches (=, 4 times). Moreover, there is a great amount of repetition between them (*explain*, *joke(s)*, *it's not funny*, *feel*, and *just* are all used 2–4 times [all underlined]). This indicates that the partners are making an effort to explain the issue jointly, and that they are in tune with each other. An extensive use of other-repetition was also observed in the bilingual couples studied by Piller, and can serve partners to perform a joint couple identity linguistically (2002a: 242).

The language barrier can be a problem not just within the relationship, but also in social situations, as limited L2 skills may exclude non-native speakers from humorous social interactions. Thus, Tim <sup>E</sup> says that he often feels like the odd one out when Sarah SG has to stop and explain why everyone is laughing, and she, in turn, finds it challenging to translate humorous remarks for him. Sarah also feels left out sometimes, as she does not always see what is funny when Tim's Australian friends are joking around, nor does she always appreciate Tim's dry sense of humour. Other non-native speakers report that it is difficult to convey humour in social situations, or to present themselves as humorous individuals. Thus, three interviewees (Courtney <sup>E</sup> , Stephanie SG, Monika SG) observe that their partner is not as funny in his L2 as he is in his native tongue. Monika reports that this is an issue that bothers her partner greatly, as he cannot show his witty personality in German and people may perceive him as a different person than he actually is:

```
Example 10.13: Dean (English, 29) and Monika (Swiss, 29)
553 Monika: <F<HI the ^one thing that you say HI>F> \~
               <L is is ^terrible <LO for you LO>L> _
               … <P<CRK when you speak Swiss German or ^German CRK>P> _
               <P<CRK is ^that CRK>P> _
               <H> <P<BR you're not your^self BR>P> \\~ 
554 Dean: … <PP m^hm PP> //
555 Monika: <P you ^^cannot P> \ 
               <P express your r- real perso^nality P> _
563 ((…)) <MRC so a <HI ^whole .. ^part HI>MRC> \~ 
               <P of your personali- ^nality gets lost P> \
566 Dean: ((…)) <%> <P hundred per^cent P> \ 
               <P<CRK I think that I ^can't CRK>P> _
               <H> I can't be ^witty or: _ 
               <NAS take the kind of [English ^sarcasm] NAS> _
567 Monika: [<PP m^hm: PP>] \~
568 Dean: <P<HI into ^German HI>P> \
               ((…)) <P<DOW that's something that is very frus^trating for me DOW>P> _ 
               the one thing I ^^hate // ((Monika drinks and puts her glass back down))
               ^e::m \~
               <H> in a social situ^ation \
               ((…)) <H> is I ^end up _ 
               <NAS being ^quite NAS> \~ 
               ^quiet \\
               <H>
569 Monika: [<P<NAS m^hm: NAS>P>] \~
570 Dean: [^a:nd] _
               <A I'm coming a^cross as perhaps being ^shy A> //
572 ((…)) <DOW that ^that's one thing that DOW> _
               ^really: \
               … <HI that I ^don't HI> like \
               I ^feel that I'm: \~ 
               <W I'm <HI ^^not HI> myself W> \ (553–572)
```
Dean believes that the image that he projects of himself when he is speaking Swiss German is not an accurate reflection of his true character. Even though he can have a conversation and express himself adequately, he cannot interject witty remarks. In the example above, Dean is emotional and speaks emphatically about this issue, stating that it is "very frustrating" to him (568), that he really does not like it (572) or even hates it (568). His level of emotionality and frustration is underscored by multiple inhalations (<H>), lengthening, and various false starts throughout his turns (underlined).

In addition to the difficulty of expressing and understanding humour in a non-native tongue, there is also the element of culture that may pose a problem. An example of culture-specific humour is slapstick humour, which is traditionally not as popular in Switzerland as it is in many English-speaking countries. Katia SG and Sarah SG both state that they do not like slapstick at all, while their partners find it entertaining. Katia SG and Craig <sup>E</sup> disagree strongly on whether this is because of their different cultural backgrounds or due to differences in their personalities; Katia believes that it is probably just her personality, while Craig is convinced that this is cultural, as such humour was a part of his younger life. He is aware of her dislike of this type of humour, and states that he would therefore not use such humour with her:

```
Example 10.14:Craig (American, 32)
Craig: <A<F<W yeah <HI ^no: HI> but what I what I'm saying is W>F>A> \
              <A<F the- the <HI these thi- ^this type of humour HI>F> I would not A> \
              … <P<DOW try to ^use <LO or to: LO>P> \
              … <@ ex^plain @@ or: @@ @> _ (1663)
```
Other Swiss participants report that Anglo-Saxon humour is at times too bizarre, too dry, or too mean for their liking (Stephanie SG, Sarah SG, Susanne SG, Sophia SG). To give an example, Susanne remembers that she was a little taken aback when the conversation in the staff room at an English school started with a humorous remark such as "oh, you're wearing your ugly orange jumper" (2353) — even if this description was accurate — and could not see the humour in this statement. Similarly, two of the Anglophone participants occasionally take issue with the Swiss manner of teasing each other (Robert <sup>E</sup> , Karen <sup>E</sup> ).5 Philipp SG explains that his wife Karen <sup>E</sup> did not grow up being teased; it was not even allowed in her family. For him, on the other hand, teasing was part of his family's humorous discourse:

<sup>5</sup> Although no gender trend was visible among the interviewees in this study, gender potentially plays a role as well, since men have been found to be "more inclined than women to view being teased as a positive experience that reinforces a sense of affiliation and liking (Keltner et al., 2001)" (Lampert and Ervin-Tripp 2006: 56).

```
Example 10.15: Karen (American, 51) and Philipp (Swiss, 47)
Philipp: <PP<LO it was LO> ^family culture <LO<HSK it was: HSK>LO>PP> _
               … <PP ^it was::: PP> \
                [<P<LO the way we grew ^up LO>P>] \
Karen: <PP [at s::]- <LO ^school too LO> probably PP> /
Philipp: = <P<W ^and for <HI ^her HI> it's a: W>P> \ 
               <PP<LO it's an o^ffenc:e [almost] LO>PP> \ 
Karen: [<PP ^yeah PP>] \~
               … <PP ^yeah PP> /~
Philipp: … <P<LO and for ^me it's LO>P> \ 
               <P<LO part of who I ^am LO>P> \ 
               <P<LO so ^that makes: LO>P> \ 
               <H> … <H> <PP<BR<LO for some .. interesting ^tensions sometimes LO>BR>PP> _ 
                                                              (1021–1025)
```
This different reaction to teasing is part of the reason why the partners think that Philipp is funnier than Karen (see section 10.4.1, "Perception of each other's sense of humour"). Not only do the two partners react differently to teasing, but, as the example above shows, this is also a potential source of conflict. The same is true for Stephanie SG and Robert <sup>E</sup> . While Stephanie and her friends frequently mock or tease each other, Robert cannot see the humour in this. He feels that it is patronizing and reacts very sensitively when Stephanie teases him:

```
Example 10.16: Robert (English-Irish, 20) and Stephanie (Swiss, 24)
1452 Robert: <W<HI this is HI> a really ^culture <LO thing LO> yeah W> //~ ((sic))
                ((…)) <HSK she ^goes like HSK> \ 
                ((smacks lips)) <:-) ^aw:: :-)> \~ 
                <HI<LEG<NAS<:-) ^o::h ^you:: :-)>NAS>LEG>HI> \~ 
                <P<A<:-) you know like ^this :-)>A>P> /
                <P<HI<:-) ^well done :-)>HI>P> \ ((teasing or belittling voice quality))
                <P<HI<:-) you tried to ^do that :-)>HI>P> /
1454 ((…)) <P<W<A if you do this in <HI ^England HI>A>W>P> _
1455 Silja: … @@ ((softly))
1456 Robert: <W it's really ^patro<LO nizing LO>W> _
1457 Stephanie: e^xactly yeah so \
1458 Robert: = <A you know [^someone's going] A> _
1459 Stephanie: <:-)[<HI we HI>] ^talked about that yea:h :-)> /\~
1470 Robert: ((…)) <@ in <HI ^^England HI>@> _
                <W<A<:-) 1[you don't] <HI ^do HI> this you know :-)>A>W> //
1471 Stephanie: 1[<:-) ^yea:h :-)>] /~
```
3[<HI **I didn't ^^realize** HI>] **that** \

<P<CRK [**feel**] .. **feel ^bad** CRK>P> //

**I think it's: really ^negative** \ <sup>1486</sup> ((…)) <H> <W<A **but in** <HI **^^England** HI> **you just say** A>W> \ … <A<HSK it's more di^rect and go HSK>A> \_

<sup>1474</sup> Robert: 3[<W **this is ^^patro**<LO **nizing** LO>W>] \_ <sup>1482</sup> ((…)) <W<HSK **I auto^matically** HSK>W> \_

<sup>1472</sup> Robert: 2[^this is] \ <sup>1473</sup> Stephanie: 2[**but ^see**] /

<sup>1483</sup> Stephanie: [^yea:h] \~ <sup>1484</sup> Robert: <P<HSK you ^know HSK>P> /~

<W<A<:-) 1[**you don't**] <HI **^do** HI> **this you know** :-)>A>W> //

<sup>1459</sup> Stephanie: <:-)[<HI **we** HI>] **^talked about that yea:h** :-)> /\~

**it's really ^patro nizing** LO>W> \_

<sup>1458</sup> Robert: = <A you know [^someone's going] A> \_

<sup>1457</sup> Stephanie: e^xactly yeah so \

<sup>1471</sup> Stephanie: 1[<:-) ^yea:h :-)>] /~

```
1472 Robert: 2[^this is] \
1473 Stephanie: 2[but ^see] / 
              3[<HI I didn't ^^realize HI>] that \
1474 Robert: 3[<W this is ^^patro<LO nizing LO>W>] _
1482 ((…)) <W<HSK I auto^matically HSK>W> _
              <P<CRK [feel] .. feel ^bad CRK>P> //
1483 Stephanie: [^yea:h] \~
1484 Robert: <P<HSK you ^know HSK>P> /~
              I think it's: really ^negative \
1486 ((…)) <H> <W<A but in <HI ^^England HI> you just say A>W> \
              … <A<HSK it's more di^rect and go HSK>A> _
              <PP<HSK uh that's: that's fucking ^shit HSK>PP> //
1496 ((…)) <HI<:-) because ^then you go :-)>HI> \ 
              <:-)<DOW<MRC ^no it's ^not MRC>DOW>:-)> \\ 
              <:-)<HI ^come on HI>:-)> \~ 
              <DOW<:-) you know it's more like ^this sort of dialogue :-)>DOW> _
   (1452-1496)
```
Robert and Stephanie are aware of their differing manners of expressing humour; this has led to conflicts in the past, and, as Stephanie points out, they have discussed this issue previously (1459). However, they have not yet managed to align their humour styles with each other. Despite being aware of Stephanie's intentions, and of the cultural differences in their humour styles, Robert cannot help but feel that the manner in which Stephanie teases him is "patronizing" (1456 and 1474) and "really negative" (1484). In the example above, he presents his reaction as beyond his control, saying that her teasing "automatically" makes him feel bad (1482). Stephanie, on the other hand, does not understand how the English manner of teasing described by Robert could be perceived more positively. These reports demonstrate that, when the couples are aware of differences in their humour styles, they may avoid particular types of humour (like Craig <sup>E</sup> or Tim <sup>E</sup> ), but adapting one's humour style (or one's reaction) to one's partner's is nonetheless challenging.

#### **10.4.3 Developing joint couple humour**

When asked what they laugh about as a couple, all of the interviewees immediately listed specific topics or situations that make them laugh. While there are some parallels between the couples, their answers were overall relatively diverse. They reported that they laugh about sarcasm (Courtney/ Martin), absurd things (Stephanie/Robert), word plays (Karen/Philipp), dif‐ ferent accents (Sarah/Tim), or simply anything when they are tired (Joshua/ Deborah). Several couples mentioned that they often laugh at other people, or also about animals (Sarah/Tim, Sophia/Richard, Claire/Simon); Karen/Philipp laugh most frequently about their children. Such laughter directed at others may be common in couple relationships due to its potentially unifying function, since it has been found that "[l]aughing at people external to the group can strengthen boundaries, solidifying members in their group identity against outsiders" (Glenn 2003: 30).

At the same time, the couples also laugh about each other, especially about each other's language mistakes. Thus, five couples reported that they often laugh about each other's linguistic errors or each other's manner of speaking (Claire/Simon, Katia/Craig, Monika/Dean, Susanne/David, Stephanie/Robert). Usually, though not always, this involves the native speaker laughing about the non-native speaker. In some cases, one partner's mistake becomes a running gag after he or she has stopped making it, which is particularly the case for words that are mispronounced (see section 10.4.4, "Playful language in bilingual couple talk"). Craig <sup>E</sup> , for instance, says that there were many situations at the beginning of their relationship where Katia SG said something that "[came] out funny" (1691) that they still remember and laugh about, and there are also stories about him learning German that resurface every now and again (1717). Craig asks his wife not to take what he is about to say personally before he mentions these situations, which demonstrates that laughing about language mistakes can potentially be hurtful to non-native speakers, as it can be construed as criticism of their language skills. However, none of the non-native speakers expressed any negative feelings about this during the interviews, but rather, they appear to enjoy laughing jointly about past mistakes and view it as an indispensable part of their bilingual relationship.

The couples' answers suggest that, despite their different cultural and lin‐ guistic backgrounds, they do not struggle to find commonalities when it comes to humour. In fact, many of the couples consider laughing together and sharing a common sense of humour integral components of their relationship. Thus, Courtney <sup>E</sup> believes that humour is something that defines her relationship with Martin SG, with which her partner agrees. They have many private jokes between them, and, unlike many other people they know, both of them appreciate and understand sarcasm:

```
Example 10.17: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
Martin: ^^sarcasm _
               <HSK is very im^portant HSK> \\
               … <A<W and sometimes <HI people HI> don't get ^sarcasm W>A> \
               <A and it's <HI ^^really a^^nnoying HI>A> \\
               ((…)) and ^we: _
               <P<MRC<HI usually HI> ^both ^get MRC>P> / 
               <P<HSK the ^sarcasm but HSK>P> \
Courtney: <P<:-)<A so <LO then we're the only two people LO> ^laughing :-)>P> \
               <PP<:-)<LO in the ^room LO>:-)>PP> /
               [@@@@] ((high))
Martin: [<:-) ^yeah :-)> @@] /~ (1481–1485)
```
The fact that they feel that they are the only people laughing about sarcasm sets them apart from others and thus potentially strengthens their sense of togetherness. In fact, joint laughter and sharing a similar sense of humour seems to serve many of the couples as in-group markers of solidarity. Thus, several couples emphasized that they laugh a lot together (Sophia/Richard, Susanne/David, Martin/Courtney), or that they have a similar sense of humour to their partner (Deborah/Joshua, Sarah/Tim, Philipp/Karen). This can even be the case if the partners do not believe that they are equally funny: Philipp SG, for instance, states that, even though he is funnier than Karen <sup>E</sup> , they still "pretty much appreciate the same type of humour" (1068). The couples' accounts demonstrate that, despite some differences in their humour (see section 10.4.2, "Challenges with regard to bilingual couple humour"), they manage to find common ground in their humour and to utilize humour to reinforce the bond between them.

#### **10.4.4 Playful language in bilingual couple talk**

When asked what is special about the way that they talk to each other (see section 5.5, "Developing a bilingual couple language"), many of the couples stated that the manner in which they use language playfully to create humour is an essential part of their couple language. Thus, a number of couples frequently invent and reuse words (10.4.4.1), mix and blend their language (10.4.4.2) or imitate different accents (10.4.4.3). The high frequency with which they claim to do this suggests that the use of playful language is a feature that is typical of the speech of bilingual couples, and a vital component of their shared sense of humour.

#### **10.4.4.1** *Inventing and reusing words*

Several of the couples often play with their bilingualism by inventing words, usually in a humorous key. Richard <sup>E</sup> and Sophia SG feel that they have their own language, which is defined by "made up words really" (Richard <sup>E</sup> 281) such as *tschurulu* /tʃʊɾʊlʊ:/ 'crazy' (Sophia SG 284). According to Sophia, this is some‐ thing that other members of her family — who are also bilingual — do as well. Among other things, Sophia and Richard have invented various terms of endear‐ ment for each other. Indeed, terms of endearment seem to be one area in which couples are often creative, as six of the couples report that they regularly invent nicknames or terms of endearment for each other (Sophia/Richard, Monika/ Dean, Susanne/David, Courtney/Martin, Deborah/Joshua, Claire/Simon; see also section 8.4.2, "Expressing positive emotions and using terms of endear‐ ment"). Monika SG and Dean <sup>E</sup> , for instance, report that they have made up many terms of endearment over the past years, which they start using on their cats once they are "done using them for each other" (Monika SG 1058). An example of such invented terms are their cats' current nicknames, *Schminzi* and *Schminkel McFinkel* (1079). Courtney <sup>E</sup> and Martin SG particularly enjoy inventing nicknames and terms of endearment for each other, as the following extract demonstrates:

```
Example 10.18: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
593 Martin: 1[I <HI usually HI> just make up ^names] /~
594 Courtney: 1[@@@@@] ((chuckling softly))
595 Martin: 2[^like] _
596 Courtney: 2[@@@]@@ ((chuckling softly))
597 Martin: <P ^Schnuffel P> \\
598 Silja: @@@@
599 Martin: = <W<P<HI o::r HI> ^Schnaffel P>W> //
600 Silja: @@@@@
601 Martin: or or <HI ^really HI> uh there's a: _
               there's [about <HI two HI> thousand] ^words /~
602 Courtney: [<:-) quite a ^few :-)>] /~
               @@ ((softly))
603 Martin: <TSK> ^and but uh: _
               they're not ^really: \~
               [<P<HSK ^u:h HSK>P>] _
604 Courtney: [<HI ^we HI>] _
               well ^I /
```
… <PP<A spoken by ^other people A>PP> \

<W<A **I I** <HI **wouldn't** HI><LO **call you** LO> **^^honey** A>W> \

<sup>605</sup> Martin: = <P<CRK ^words that are CRK>P> \

<sup>623</sup> ((…)) **and I** <HI **^call** HI> **him** \ <P<LO *^Knaufler* LO>P> \\

<sup>606</sup> Courtney: **^^no** \\~

<sup>625</sup> Courtney: = **^***Knüffel* \ <sup>626</sup> Martin: <P<:-) m^hm :-)>P> / <sup>627</sup> Silja: <HI<:-) ^*^Knüffel* :-)>HI> \

<sup>628</sup> Courtney: = **^***Miörgler* \\

<sup>624</sup> Silja: … @@@

 

 

<sup>604</sup> Courtney: [<HI ^we HI>] \_

@@ ((softly)) <sup>603</sup> Martin: <TSK> ^and but uh: \_

well ^I /

they're not ^really: \~

```
605 Martin: = <P<CRK ^words that are CRK>P> \
               … <PP<A spoken by ^other people A>PP> \
               <W<A I I <HI wouldn't HI><LO call you LO> ^^honey A>W> \
606 Courtney: ^^no \\~ 
623 ((…)) and I <HI ^call HI> him \
               <P<LO ^Knaufler LO>P> \\ 
624 Silja: … @@@
625 Courtney: = ^Knüffel \
626 Martin: <P<:-) m^hm :-)>P> /
627 Silja: <HI<:-) ^^Knüffel :-)>HI> \ 
628 Courtney: = ^Miörgler \\
629 Martin: 1[@@@@@@@] 
630 Silja: 1[@@@ <@ o^kay @> @@] _
631 Martin: 2[<PP<@ ^Miö@rgler @>PP> @@] _
632 Courtney: 2[@@@@] ((softly))
633 Martin: 3[@@@@] ((softly))
634 Courtney: 3[<P<:-) ^Mi:örgler :-)>P>] _ 
638 Martin: ((…)) <PP<:-) that's a ^good one :-)>PP> \~ (623–638)
```
Martin and Courtney emphasize that they have coined a multitude of nicknames for one another; Martin hyperbolically states that "really ((…  )) there's about two thousand words" (601), and Courtney assents to this ("quite a few" [602]). They also assert that they would not use common terms of endearment for each other another; when Martin states that he would not call his girlfriend *honey*, she responds with an emphatic "no" (606). It is notable that the partners both find generating such neologisms very entertaining, as they are both giggling throughout the extract, and Martin savours one of Courtney's word creations ("Miö@rgler @@" [631]), and compliments her on it ("that's a good one" [638]). Martin and Courtney volunteer a number of nicknames they might use for their partner (*Schnuffel, Schnaffel, Knaufler, Knüffel, Miörgler*). Although none of the words means anything, they all have a Swiss German phonology and are pronounced with a Swiss German accent. It is remarkable that, despite the fact that most couples in this study use English as their main couple language, they mostly invent terms of endearment that either sound (Swiss) German, or are a mix of both languages (see section 8.4.2, "Expressing positive emotions and using terms of endearment").

In addition to coining completely new words, some couples also have a habit of changing the meaning of existing words in their language use. For instance, Martin and Courtney replace words with random other expressions if they do not feel like saying the word in question (Martin SG 306). This new term may then come to replace yet other words. According to Martin, they sometimes even get "stuck in that language", as it is "hard to get out [of it] again" (312 and 316). Another example of a new use for a word are the terms of endearment that Susanne and David use for each other. Thus, Susanne occasionally nicknames David *gnogi* (/gnɔgi/ instead of Italian *gnocchi* /ɲɔkːi/ 'dumplings'), because this is how he once pronounced the word when they went out for dinner to an Italian restaurant in Brighton, and she found this very sweet. Conversely, he sometimes calls her *Huckehäschtli,* which is derived from *Chuchichäschtli* 'kitchen cabinet' (a word the Swiss often ask foreigners to say because it is difficult to pronounce, and because they perceive it as typically Swiss). These examples demonstrate that the couples invent new terms, but also appropriate and reuse existing expressions, and thereby enhance their joint couple language.

#### **10.4.4.2** *Mixing and blending languages*

In addition to inventing and reusing words, combining the languages — whether in the form of language mixing or by creating blended terms — was also mentioned by some of the interviewees as an important part of their couple language. For instance, Richard <sup>E</sup> and Sophia SG believe that a special feature of their common language is that they frequently mix languages when they are being silly, although they hardly ever mix in serious situations. While Joshua <sup>E</sup> and Deborah SG do not generally switch with a humorous purpose, they also think that the frequency with which they switch is something that defines their couple language. Moreover, the partners sometimes use different languages in the same conversation, as both speak in their respective mother tongues, which people tend to find peculiar (see also section 5.3, "The partners' language use with each other"). Despite this feedback, Joshua and Deborah both agree that they never felt that this was an unusual thing to do; instead, they regard it as sensible, since they understand each other, and can both express themselves in their mother tongue. Besides mixing their languages by code-switching or speaking dual-lingually, the couple occasionally combine their two languages to create loanblends, such as the expressions *Schatzi-love* 'treasure, darling' or *blade*-*bar* 'blade-able' (doable with rollerblades). The term *shnugglesy*, which Claire and Simon use for each other, might also be a mix between the two languages (namely *snuggle / snugly* and *Schnüggel(i)* 'little cutie'). While only a few such terms were mentioned by the interviewees, many couples do occasionally blend languages by using words from one language with the pronunciation of the other, which will be discussed in the next section.

#### **10.4.4.3** *Imitating accents*

The couples' accounts indicate that many of them enjoy imitating different varieties of English, as well as different foreign accents. Various participants mentioned this, and eight participants spontaneously imitated one or several accent(s) during the interviews (without being provoked to do so). For in‐ stance, they imitated the Swiss train conductors saying *Grüezi mitena:nd* 'good morning/afternoon/evening everyone' (Robert <sup>E</sup> 470), an Indian friend (Susanne SG) or a Scottish boy they met on holiday (Sarah SG). Often, the accents the interviewees imitated are linked to the reported speakers' cultural identity, as in example 10.19. Courtney describes an American friend of theirs who is in a relationship with a Swiss woman and likes to comment on the cultural differences between them:

#### **Example 10.19: Courtney (English-Ghanaian, 24)** Courtney: <P<:-) he's ^always saying :-)>P> \_ <Hx:> <HI ^o:h I'd HI> \ ((imitating drawling American accent throughout, rhotarized /r/)) <W I I would have ^lo:ved to have a ^s:teak yesterday: W> \_ <P<BR but you know <LO my ^girlfriend LO>BR>P> \_ <P<BR she had a little <LO ^sa:la:d LO>BR>P> \_ (558)

In cases such as this, the imitation of an accent or a particular manner of speaking could be considered part of the reported speech, or constructed dialogue (Tannen 1986: 311).

When particular expressions or modes of speaking are imitated repeatedly, they sometimes become a permanent part of the couple language. For instance, Robert <sup>E</sup> enjoys imitating the pronunciation of Swiss words that he finds typical, and the couple have integrated a number of expressions in their English language use, such as *genau* 'exactly' (Robert <sup>E</sup> 297), *sägemal* 'tell me' (Robert E 308), and *gäll* 'right' (Stephanie SG 305). There are also some features which they have adopted from other languages, for instance a *no* question tag at the end of a sentence (analogous to *right*), which was used by the Spanish exchange students in Aberdeen, where they met. Stephanie reports that she borrowed the term because she "kinda liked it", and Robert was quick to adopt "this curious no question word" (Robert <sup>E</sup> 442) from her (see example 10.5 in section 10.3.2.3, "Positive and negative laughter triggers", for an example of their use of *no*). Likewise, Susanne SG and David E have incorporated some foreign expressions in their English base language, such as versions of *gnocchi* and *Chuchichäschtli* (see section 10.4.4.1, "Inventing and reusing words"), which they pronounce humorously, with an English accent. Another term in their joint vocabulary is *Migros butschi* /mɪgrəʊ bʊdʒi/, which is an Anglophone distortion of the Swiss pronunciation *Migros budget* /migʀo bytʃe/ [the economy line of the largest Swiss retailer]). These examples show that Susanne and David enjoy imitating the pronunciation of non-native speakers to establish a humorous discourse.

Similarly, Courtney <sup>E</sup> and Martin SG often play with different languages and accents to create humour. In the following extract, Courtney describes how Martin sometimes pretends to translate their conversations into Italian for fun, something which she evidently finds as entertaining as he does:

```
Example 10.20: Courtney (English-Ghanaian, 24) and Martin (Swiss, 26)
317 Courtney: ^sometimes: eh \
               <W ^Martin feels that ^everything W> \
               … <CRK that <%> e- ^um: CRK> \
               <CRK ^well CRK> \~
               his conversation in ^English //
               <CRK ^needs to be: CRK> \ 
               <CRK translated into I^talian CRK> \\
               but ^not /
               … <CRK ^rea:l <LO Italian LO>CRK> _
               <CRK but just ^everything CRK> _
               <H> <CRK<DOW at the end of every word there needs to be an ^o or
               DOW>CRK> \
318 Silja: … <P o^kay P> /~
319 Courtney: = <P<HSK ^something like HSK>P> \
               <@ he@ he'd ^ask me @ @> //
                <:-)<HI ^how HI> do you say: :-)> \
               … <P<HSK ^how do you say: HSK>P> \
               … <PP<HSK ^u:::h HSK>PP> //~
               … ((thump, probably slap on leg)) <P ^what's his name P> \
               <P<W<HSK<:-) Clark Gable Gable Clark ^Gable :-)>HSK>W>P> \
320 Martin: <P<:-) yeah Clark ^Gable :-)>P> \
321 Courtney: = DOW<P<HSK<:-) how do you say Clark ^Gable :-)>HSK>P>DOW> \
322 Martin: … <:-) what's his <HI ^real HI> name :-)> \\ 
               <A and his ^real name is A> _
               <DOW /ˈglʌrke ˈga:ble/ DOW> \ ((imitates Italian pronunciation, savouring))
323 Silja: @@@[@@] 
324 Courtney: [<:-) ^a:nd :-)>] _
               <:-)<DOW<A it's ^often like this A>DOW>:-)> \
325 Silja: @@ [<@ o^kay @>] \
326 Martin: <W<:-) [and <HI ^just HI>] [that's just] in: in: :-)>W> \
327 Courtney: [and ^then] _
328 Martin: <W in boring ^situations W> \
               when ^things have to be \
329 Courtney: … <P<HSK lightened ^up HSK>P> \
330 Martin: [<P<:-) ^yeah :-)>P>] \~
331 Silja: [so you] lighten [them ^up] _ 
332 Martin: [<P just a ^little bit P>] /
333 Silja: by making things <LO I^talian LO> _
```
[<P<MRC<:-) **it's ^really ^funny** :-)>MRC>P>] \

<sup>336</sup> Courtney: <P<DOW [**it's**] **^re:::ally funny** DOW>P> \_ (317–336)

<sup>334</sup> Martin: <P<:-) ^yeah :-)>P> \~

[o^kay] \/~

<sup>335</sup> Silja: [<P ### P>] \_

```
334 Martin: <P<:-) ^yeah :-)>P> \~
               [<P<MRC<:-) it's ^really ^funny :-)>MRC>P>] \
335 Silja: [<P ### P>] _
                    [o^kay] \/~
336 Courtney: <P<DOW [it's] ^re:::ally funny DOW>P> _ (317–336)
```
**when ^things have to be** \

<sup>332</sup> Martin: [<P **just a ^little bit** P>] /

<sup>329</sup> Courtney: … <P<HSK **lightened ^up** HSK>P> \ <sup>330</sup> Martin: [<P<:-) ^yeah :-)>P>] \~

<sup>333</sup> Silja: by making things <LO I^talian LO> \_

According to Courtney, this may go on for "a good forty minutes" (341). It is obvious that they both greatly enjoy this style of playing with language, as they describe it very positively, and believe that it is hilarious and lightens the mood. The extract is also a very good example of a jointly constructed conversation in a humorous key: the partners' contributions are perfectly synchronized, and they repeat each other (319–320; 334–336), rephrase each other's utterances (321–322), and finish each other's sentences (328–332). This example epitomises the general finding that the bilingual couples in this study have achieved a very high level of communication in their relationships and manage to make use of their multilingualism to create humorous discourse.

#### **10.5 Discussion and summary**

In this chapter, I examined the manner in which bilingual, bicultural couples express humour, as well as their laughing behaviour in conversation. When the bilinguals' laughing behaviour was analysed, the first aspect that was apparent was that the frequency and duration of their laughter during the interviews varied considerably. Moreover, the partners displayed considerable differences from one another in the number of laughter pulses, the number of laughter episodes, as well as the average duration of their laughter episodes, despite participating in the same conversation. This suggests that little assimilation of the partners' laughing behaviour to each other has occurred, or that other factors such as their gender, mother tongue and culture are more influential.

The analysis of the context of the participants' laughter revealed that even though the majority of laughter episodes occurred in a humorous context (62.3%), a high proportion of laughter episodes were triggered by non-humorous topics (37.7%). While others have observed that laughter often follows topics or statements that do not seem humorous (Provine 1993: 291; Adelswärd 1989: 107), and that any action or topic may potentially trigger laughter (Clift 2016: 86), the number of non-humorous statements provoking laughter was nevertheless surprisingly high. Interestingly, laughter was caused by negative statements (14.9%) far more frequently than by positive ones (8.6%), which provides empirical evidence for Jefferson's (1984, 1985) observation that people

often accompany trouble-telling, as well as uncomfortable subjects or obscene topics, with laughter.

The distribution among these three contexts (humorous, negative, positive) followed this pattern for almost all of the interviewees, with the exception of one couple, who accompanied more positive than negative utterances with laughter. Furthermore, the partners tended to resemble each other in their relative distribution of laughter among the categories, which could be indicative of a similar humour style. In self-reports, a positive association between partners' humour styles has been found among monolingual couples (Priest and Thein 2003; Barelds and Barelds-Dijkstra 2010; Hahn and Campbell 2016), and these results suggest that this may also be the case for bilingual couples. However, the similarities may also be attributed to the specific topics discussed during each interview to some extent.

A closer look at who provoked most laughter revealed that laughter was triggered most frequently by an utterance made by the laugher him- or herself (46.4%), second most frequently by his or her partner (37.1%), and occasion‐ ally by the interviewer or interview situation (14.7%). Far less frequent were instances where laughter was provoked by both speakers jointly or an unclear trigger (1.9%). The percentage of self-triggered laughter differs from the results of Glenn's (2003) study, in which conversationalists initiated laughter very frequently in two-party interactions (87%), and far less often in multi-party interactions (17%). It is possible that the frequency with which laughter was triggered by the laugher's own comments was influenced by the setup of the interviews and the number of participants, and that the findings would have been very different if more or fewer people had been present. When the trigger type (humorous, positive or negative) was considered in combination with the speaker who triggered laughter, it became apparent that far more laughter was provoked by the laughers' own humorous comments than by those of their partners. This supports the observation that speakers frame their own humorous utterances with laughter in order to indicate a humorous key and to invite laughter by their interlocutors (Jefferson 1979: 82; Kotthoff 2006: 274). In contrast, more laughter was triggered by positive statements by the laughers' partner than by the laughers themselves, while negative statements triggered a similar number of laughter episodes in both parties. These findings indicate that the speaker and the context are interlinked when it comes to triggering laughter.

Besides the general context in which laughter occurred, I also looked at the specific topics which provoked the participants to laugh. In the humorous context, comments about cultural or linguistic differences, conflicts, as well as about each of the partners and their relationship, were most common. The positive statements which were accompanied by laughter most frequently concerned the interviewees' relationship, their partner, their family, or their culture. In the negative context, the bilinguals laughed most often when past conflicts were discussed, with criticism or general negative statements, or when disagreement occurred during the interviews. Again, whether the participants laughed more about their own non-humorous statements or their partner's depended on the specific topic.

In addition, I examined the participants' use of smiling and laughing voice quality during the interviews. These features are of interest because they can indicate a humorous frame or serve as contextual clues for laughter (Holmes and Hay 1997: 132). On average, laughing voice quality was used in 3.1% of all intonation units, with great variance among the participants, although, in relative terms, most interviewees used laughing voice quality with a similar frequency to their partner. Smiling voice quality was almost three times more frequent than laughing voice quality — accompanying 8.8% of the participants' intonation units — and there was less similarity between the partners' use of smiling voice quality. Interestingly, the use of laughing voice quality tended to correlate with the frequency of laughter; this was not the case for smiling voice quality, which seems to be used independently of laughter, or even to replace it.

Since humour and laughter have often been found to be gendered, I explored potential gender patterns in the bilinguals' laughing behaviour. Previous studies have repeatedly suggested that women laugh more than men in interactions (Adams and Kirkevold 1978; Mehu and Dunbar 2008; Winterheld, Simpson and Oriña 2013), which was also the case for the interviewees in this study. Thus, the female participants produced almost 30% more laughter pulses during the interviews than the male participants. Interestingly, when the number of laughter episodes were considered instead of the individual laughter pulses, the difference was far less marked (7%). This shows that, while women laughed on only a few more occasions, their laughter episodes lasted longer than their partners'. Moreover, the female interviewees used more laughing voice quality and more smiling voice quality than the male participants.

In addition, I examined whether there were any gender trends with regard to which speaker (self or partner) and what type of trigger (humorous, positive, or negative) provoked laughter. The relative distribution of the trigger types among these categories was similar for both genders. While both genders laughed similarly often about their own humorous comments, men's humorous laughables triggered more laughter in their partners than the reverse. However, a look at the two couple combinations indicates that other factors may be more influential than gender in this case, since the native speakers of English in both groups laughed less about their partner and more about themselves. Thus, no evidence was found for the assumption that women produce fewer laughables than men, nor that women laugh more about men's laughables than the reverse (Provine 1993, 1996; Jefferson 2004). In spite of this, the couples' own evaluations of their perceived funniness suggest that their subjective perceptions of each other's funniness are gendered. Thus, the couples either claimed to be equally funny, or tended to name the male partner as the funnier in the relationship, and several male participants even put their partner's funniness somewhat into question.

A person's cultural background and mother tongue may not only influence his or her laughing behaviour, but also what he or she finds humorous (Ruch and Forabosco 1996; Cheng 2003). In addition, the factor of mother tongue is relevant because the native speaker may be at an advantage in producing successful laughables when conversing in that language. Indeed, the analysis demonstrated that the participants' mother tongue had an impact on the frequency of their laughter as well as their use of laughing and smiling voice quality. Thus, the English speakers laughed considerably less often than the Swiss participants, and they also used a laughing voice quality less frequently. They did, however, use smiling voice quality similarly often as the Swiss bilinguals. It is possible that this is the result of culturally learnt behaviour, but it may also be linked to the participants' production of laughables. Thus, it was evident that humorous laughables produced by native speakers of English were far more successful in provoking laughter than those by the Swiss partners, and the English speakers also marked a slightly higher number of their own humorous statements with laughter. It cannot be determined whether this is because the former are funnier than the latter, because the English speakers are at an advantage as they are conversing in their mother tongue, or because the Swiss are simply more ready to laugh. Potentially, it is connected to the belief expressed by some Anglophone participants that their humour is better than their partner's. Such a tendency was also observed by Chiaro in bilingual couples, among whom the native speakers of English tended to view their sense of humour as superior to their partner's and presented it as a positive benchmark for humour (2009: 223). Interestingly, such views were only put forward by participants from Great Britain — and no other English-speaking countries — in the present study.

The couples also discussed various challenges posed by expressing humour in a bilingual, bicultural relationship. They lamented that humour can be difficult to translate, which can be an impediment to the retelling of funny remarks and stories. It was also reported that their different sense of humour or limited language skills sometimes exclude the non-native speakers from participating in humorous social interactions, or do not allow them to present their witty side. The bilinguals' preferences for certain types of humour, such as slapstick, dry humour, or sarcasm, are also influenced by their cultural background, and not every humour style is appreciated by both partners. In particular, their different manners of teasing tend to create tension among some of the partners. Moreover, even when the partners are aware of differences in their humour styles, adapting to each other's style still appears to be challenging.

The bilinguals named a variety of topics and situations that make them laugh, such as sarcasm, absurdity, word plays, accents, other people, animals, or their children. The partners stated that they also laugh at each other, in particular about each other's language mistakes and words that are mispronounced. Interestingly, while verbal humour is more language-specific than referential humour (Ritchie 2010: 34), and therefore more difficult for L2 speakers, much of the couples' humour is verbal, and many of the couples view the use of playful language as an essential part of their couple language. A number of couples enjoy inventing new words, or reusing existing words in a different context, which they do especially with terms of endearment. Moreover, several of them enjoy combining the languages, either by language mixing or coining blended terms. Finally, their accounts and their behaviour during the interviews also indicate that many of them enjoy imitating different varieties of English as well as foreign accents, which is in line with Vaid's (2006) observation that bilinguals enjoy accent humour and bilingual puns. In fact, the interviewees mentioned several expressions that they like to imitate, which have become a permanent part of their couple language.

The couples' answers, as well as their humorous talk during the interviews, show that they have all succeeded in finding common ground when it comes to humour, and that many of them consider laughing together and having a similar sense of humour essential components of their relationships. This stands in opposition to the bilingual couples examined by Chiaro, who believed that they had a different sense of humour, and viewed language as a major obstacle in the communication of humour (2009: 222). None of the bilinguals in this study framed their language differences as an impediment to producing or understanding humour, but rather, they utilize their bilingual, bicultural situation to create humour, which may, in turn, strengthen their shared bilingual couple identity.

## **11 Conclusion**

## **11.1 Introduction**

My primary goal in this book was to present a detailed picture of the com‐ munication among bilingual, bicultural couples with a high level of fluency in their main couple language. In order to achieve this, I interviewed ten bilingual couples with similar backgrounds, and explored a number of areas pertaining to their couple language: their language choice(s), their language mixing behaviour, their attitudes towards their languages and cultures, their expression of emotions, their swearing behaviour, as well as their humour and laughter. In each of these areas, I analysed various aspects, and, wherever possible, I combined the couples' reports about their linguistic practices with their actual language use during the interviews.

In the following, I synthesize the main findings with reference to my research questions. Since the findings have already been discussed and related to previous work at the end of each chapter, the focus here is on consolidating and drawing together the most important findings relating to the bilingual couples' language in all areas (11.2). I then proceed to a discussion of the role of the variables of gender (11.3) and mother tongue (11.4), as well as of other factors which might influence the couples' communication (11.5). Subsequently, I consider potential implications of my findings (11.6), offer recommendations for future research (11.7) and discuss the strengths and limitations of my research (11.8), before making closing remarks (11.9).

## **11.2 Developing a bilingual couple language**

My initial research questions related to the language(s) the couples choose for communication, the reasons behind this choice, and how their language use has evolved over the course of their relationship. According to their own reports, all except for one couple speak primarily English to each other and hardly ever use the community language in their relationship. Moreover, their couple language is characterized by a high level of stability, as only one couple gave account of major changes in their language choice and in the extent of their language mixing over the course of their relationship. The couples' answers indicated that the key factor determining their language choice at the beginning of their relationship was language proficiency, and that, for most of the couples, the main reason for adhering to English as their relationship progressed was habit; this was also the case for many of the bilingual couples studied by Piller (2000, 2002a). Other influences on the couples' language use were considerations of individual and couple identity, as well as language emotionality, but also per‐ sonal motivation and attitudes. In addition, the diglossic situation in Switzerland seems to be a considerable impediment to a shift to the community language. The majority of the interviewees also use English outside the home on a regular basis, but primarily with L2 speakers, as only two couples have established extensive networks of native English speakers. These findings stand in contrast with those of other studies, particularly Piller (2002a).

While most of the couples use English as their primary language, this does not mean that the community language does not feature in their interactions. To assess the extent of the presence of (Swiss) German, I examined how frequently the partners mix languages, and what factors influence their language mixing behaviour. Most of the couples reported that they do not mix languages very often, and that, while they occasionally use borrowings, they rarely code-switch. These reports were corroborated by their mixing behaviour during the inter‐ views, since there were relatively few language switches (on average only 8.25 spontaneous switches per person), and borrowings occurred ten times more often than code-switches. These results confirm that fluent bilinguals can adhere to the language of the conversation without falling back on their mother tongue if they desire to do so — even if their conversational partners understand their L1. Most of the switches (85.0%) during the interviews were unhedged, which could be indicative of a positive attitude towards language switching; spontaneous switches were hedged more frequently than metalinguistic switches (18.2% vs. 11.3%). Despite the high percentage of fluent switches, the couples tended to voice mixed attitudes towards language mixing, similar to the bilingual couples examined by Piller (2002a). Furthermore, the analysis of the bilinguals' switching behaviour also demonstrated that, while there were considerable differences in the participants' language mixing, the couples usually switched similarly often, and used similar types of switches.

In addition, I investigated the interviewees' attitudes towards various aspects of their bilingual, bicultural relationship, and the interplay between these attitudes and their language use and attraction to each other*.* Their reports revealed that the Swiss participants' attitudes towards their partner's language and culture were highly positive when they met, which contributed to their initial attraction to their partner, whereas most of the Anglophone participants had relatively neutral and rather stereotypical views of the Swiss language and culture at the beginning of their relationship. The partners' attitudes towards both cultures have converged over time; both partners expressed some critical views, even though their attitudes towards their partner's culture were positive on the whole. Meanwhile, the partners' attitudes towards the two languages have remained divergent: the Swiss participants still have a very positive attitude towards English — many even expressed a type of "language desire" as observed by Piller (2002a) — while most of the Anglophone participants articulated mixed feelings towards their L2. This may account for the fact that their relationship language has not even partially shifted to the community language. At the same time, all participants expressed very positive attitudes towards bilingualism in general, and the couples with children displayed an immense commitment to their children's bilingualism. Their attitudes seem to reflect the positive views of their environment of English and bilingualism in general, and contrast strongly with those of bilingual couples in other settings (e.g. Schüpbach 2009). As was expected, highly positive attitudes among the bilinguals towards their partner's language tended to correlate with high L2 proficiency, but also with a high language emotionality and with a preference for using their L2 with their partner, and at times even with their children.

Other questions that I aimed to answer concerned the language the partners choose to express emotions with each other, and the manner in which they deal with having to express positive and negative emotions in their L2. According to the couples' responses, all of them primarily use their couple language to express positive and negative emotions. Even though expressing negative emotions in an L2 can sometimes be challenging or lead to misunderstandings, the Swiss participants do not view using their L1 as a viable option, which contrasts with reports on couples in other studies (Piller 2002a; Pavlenko 2005). However, the participants occasionally use the non-couple language for the purpose of expressing positive emotions and they frequently give each other Swiss German nicknames, or names consisting of invented or blended expressions with a Swiss German phonology. Many interviewees emphasized that they enjoy using and hearing nicknames in their L2, and all of the participants displayed a high emotional attachment to their relationship language, which is in line with findings by Pavlenko (2004) and Dewaele and Salomidou (2017). During the interviews, positive and negative emotions were expressed exclusively in English, and there was great variance in the frequency of the interviewees' use of emotion words, and little evidence of assimilation between the partners. Overall, the participants used slightly fewer positive than negative emotion words, and a greater variety of negative than positive emotion words. A number of suprasegmental features appeared particularly often in the vicinity of these emotion words, most notably prominent pitch and pitch movement, strong and multiple stresses, as well as rising and falling terminal pitch. Interestingly, smiling voice was very frequent in both emotional contexts, while laughing voice quality was much less common in the context of negative than positive emotion expression.

In addition to the couples' expression of emotions in general, I was interested in the bilinguals' swearing behaviour in both languages, and the affective attachment of expletives in both of their languages. Most of the participants stated that they do not swear very frequently, which was borne out by the fact that they did not use a great number of swearwords during the interviews. On average, each person only used three expletives spontaneously, all of which were uttered in the language of the ongoing conversation. In most of the couples, both partners used a similar number of swearwords with comparable levels of offensiveness, which suggests that couples tend to assimilate their swearing behaviour to each other. With regard to the language choice for swearing, the most influential factors appear to be the language of the ongoing conversation, the speaker's L2 proficiency, and his or her socialisation in the language. Thus, most of the bilinguals reported that they use both languages for swearing; some swear mainly in their L1, yet no-one primarily uses his or her L2. The interviewees listed a number of reasons for their language preference, such as the view that one of the languages is more colourful or more suitable for swearing, that they lack swearing proficiency in their L2, or that they do not feel entitled to swear in their L2. At the same time, some reported that swearing in one's L2 is easier, less of a taboo, and potentially less offensive, because swearwords carry less weight in one's L2 than in one's native language. The interviewees displayed and referred to various reactions to swearwords in their mother tongue which were uttered by their partner, ranging from indifference and amusement to disappointment, embarrassment, and personal offence, which demonstrates that swearing can be a challenging area to negotiate for bilingual couples.

The final questions I decided to investigate related to what the couples laugh about, whether there are similarities in the partners' laughing behaviour and humour styles, and to what extent they utilize their bilingual situation to create humour. The analysis of their laughing behaviour during the interviews indicated that the participants' laughter was very heterogeneous in terms of its frequency as well as the duration of each laughter episode. Moreover, the assimilation between the partners with regard to the frequency of laughter appeared to be minimal, as they differed considerably in the number of laughter pulses they produced as well as the number of laughter episodes. However, the participants' laughing behaviour was more homogenous concerning the context which triggered their laughter, since, for all of them, laughter occurred most frequently in a humorous context, and, for all but one couple, this was followed by negative and finally by positive utterances. The partners resembled each other in the relative distribution of laughter among these categories, which could be taken to indicate a similar humour style. Such similarities between partners' humour styles have been found among monolingual couples (Priest and Thein 2003; Barelds and Barelds-Dijkstra 2010; Hahn and Campbell 2016). With regard to the topics which usually make the couples laugh, it can be said that many of them relate to their situation as bilingual, bicultural couples, and a large part of their humour is of a linguistic nature. The bilinguals view the use of playful language as an essential component of their couple language, and enjoy inventing neologisms, reusing existing words in a different context, and combining their languages by mixing and coining blended terms, but they also like imitating different native and non-native accents. The interviews demonstrated that these bilingual couples manage to find common ground in their humour, and that many of them view laughing together and sharing a similar sense of humour as constitutive elements of their relationship.

In sum, this study marks a further contribution to the variety of different findings on the attitudes and modes of communication of bilingual couples. Although there are a number of areas in the couples' interaction that are potentially challenging, they have achieved a high level of communication and a considerable degree of linguistic assimilation, and exhibit highly positive attitudes towards their bilingual, bicultural relationships. This contradicts the assumption that such relationships are problematic and have a higher risk of communication difficulties (Killian 2009), and stands in contrast with Heller and Lévy's (1992) study of English- and French-speaking couples in Canada, who expressed rather negative views concerning their bilingualism. It is, however, in line with some reports of couples who find their relationship and their partner's different linguistic and cultural background highly interesting (Khatib-Chahidi, Hill and Paton 1998; Piller 2002a), and who do not believe that their relationship or their communication are more problematic than those of a monolingual couple (Piller 2002a). This underlines once more that, while the situation and communication of each bilingual couple is unique, there are also broader patterns, which are influenced by aspects such as language proficiency, environment, prevailing attitudes, education, and the similarity of the partners' cultures.

#### **11.3 The gender variable**

Gender has often been found to be an influential factor in shaping the manner in which people speak. It may also influence the manner in which couples interact, and account for differences in the partners' conversational styles. On the one hand, I decided to pay attention to the role of gender as it can be a confounding variable when the influence of other variables (such as their mother tongue) on the couples' language is analysed; on the other hand, by its nature, my data lent itself to the study of gender, as all of the conversations took place between dyads who were balanced in terms of gender.

Some questions that I aimed to answer were whether gender plays a role in the participants' preference of couple language, and whether it influences the manner and the extent of their language mixing. The couples' reports suggested that the influence of gender on their language choice is minimal. Even though previous research has repeatedly indicated that multilingualism is gendered in terms of access to it, the value attached to it, and the economic benefits derived from it (Grin 2001; Piller and Pavlenko 2007), there was no indication of any such trend among the participants in this study. Similarly, the observation that women in bilingual relationships are frequently placed in a doubly marginalized position (Piller 2000), since they are immigrants as well as non-native speakers of the relationship language, does not pertain to the participants in this study, among whom only one participant, who happens to be male, is sometimes faced with such a situation (Joshua <sup>E</sup> ).

While gender does not seem to affect their choice of couple language, the men and women in this study differed considerably in their language mixing behaviour during the interviews, as the female participants produced more spontaneous borrowings and code-switches than their partners. These findings add to the varying research results on the factor of gender in language mixing, as gender differences have been found in some groups but not in others (Poplack 1980, 1988; Cheshire and Gardner-Chloros 1998). This highlights that the context and group composition, as well as the practices of the specific speech communities, may affect the genders' mixing behaviour in different ways. Furthermore, women hedged fewer of their switches than men, which might be symptomatic of a more relaxed or more positive view of language mixing among the female partners.

The couples' accounts did not point to gender differences in their attitudes towards each other's language or culture, or towards bilingualism in general. However, the factor of gender seems to have played a role in their initial attraction to each other, at least among the speakers of English. While many Anglophone men reported being attracted to their partner's non-native accent and foreign appearance, the women stated that they did not consider their partner's foreignness particularly attractive or even deemed it unattractive. No such gender differences were found among the Swiss participants, who all found their partner's cultural and linguistic background attractive.

Since women are generally believed to be more expressive of their emotions than men (Notarius and Johnson 1982; Piasecka 2013), I expected the female participants to use a greater number of emotion words than their partners. The analysis demonstrated that, relative to the number of words they used during the interviews, women used more negative as well as positive emotion words than men, and they also used a greater number of different types of emotion words (yet in absolute terms, not much difference was found between the genders). Moreover, the analysis of suprasegmental features demonstrated that, while many of these features tended to be used proportionally to their overall use in the context of emotions by both genders, women displayed a tendency to lend emphasis to their emotional speech with variations in pitch, as well as with strong stresses and rhythmic speech, while men frequently used a high voice for this purpose. Thus, there were clear gender differences in the area of expressing emotions, in the frequency and diversity of emotional expression, and as regards suprasegmental features accompanying emotion words.

The factor of gender was also analysed in the context of swearing, since many studies have come to the conclusion that women swear less frequently than men, and that they use weaker swearwords (Coates 2004; Rayson, Leech and Hodges 1997). Indeed, gender appeared to play a role with regard to partic‐ ipants' swearing behaviour. When using swearwords spontaneously, the male participants used more offensive swearwords, while the female participants used more euphemisms and very mild swearwords. Moreover, the women used more offensive swearwords than their partners in the metalinguistic context. This could be taken to suggest that, for women, the main reasons for avoiding the use of swearwords are social custom or gender expectations, rather than a sense of strong taboos attached to these swearwords.

Lastly, gender patterns were also examined in the bilinguals' laughing behav‐ iour and humour. It has been suggested repeatedly that women laugh more than men in conversations (Mehu and Dunbar 2008; Winterheld, Simpson and Oriña 2013), which was corroborated in these interviews. The female participants produced 7% more laughter episodes, and almost 30% more laughter pulses than the male participants, and, consequently, laughed considerably longer than their partners on each occasion. Men and women also differed in their use of voice quality, as women used more laughing and smiling voice quality than men. The observations that women laugh more about men's laughables than the reverse (Provine 1993), and that they produce fewer laughables than men (Robinson and Smith-Lovin 2001), were not confirmed. Thus, both genders laughed similarly frequently about their own humorous comments, and while the overall numbers indicate that the female participants laughed more often about their partner's humorous comments than the reverse, a look at the two couple combinations suggests that the variable of the participants' mother tongue is more influential than their gender. Nevertheless, when asked to decide which partner is funnier, the couples selected the male participant more frequently and some men even presented their partner as humourless. Thus, gender emerged as an important influence on various areas of the partners' language use, including their laughing behaviour.

#### **11.4 The influence of mother tongue and culture**

The participants' cultural and linguistic backgrounds are relevant to the study of their communication for two reasons. On the one hand, both partners' speech and manner of interacting may be influenced by conventions specific to their culture or mother tongue, which could, for instance, shape their use of emotion terms, suprasegmental features, or the production of humour. On the other hand, the fact that one of the partners is a non-native speaker of the couple language may, for example, influence his or her language mixing and swearing behaviour, and potentially place him or her at a disadvantage. Furthermore, as was the case for the analysis of gender, the fact that the interviews were conducted with dyads that were balanced in terms of their mother tongues provided a great opportunity to study the influence of their linguistic backgrounds on several language features.

As I anticipated, the speakers of the two mother tongues differed in terms of the reasons for their language preference and their attitudes towards their second language. The Swiss participants, who are all fluent in English, stated that they enjoy expressing themselves in their L2, and tend to consider it a part of their identity and an emotional language. Thus, their attitudes towards their partner's language are highly positive. In contrast, most of the Anglophone participants indicated that they view their L2 primarily as a medium of commu‐ nication and showed a mixed attitude and little emotional attachment to the language. The English speakers are also considerably less fluent in their L2 than their partners are; in addition to having had less formal education in their L2, their efforts to improve their language skills are obstructed by the diglossic situation in Switzerland. In the light of the partners' attitudes and language skills, the choice of English as their relationship language comes as no surprise.

I was also interested in determining whether there were any differences in the language mixing behaviour and in the expression of emotions between the speakers of both languages. According to the partners' reports, they mix languages similarly often, despite the differences in their L2 proficiency. This was paralleled by their mixing behaviour during the interviews, since the Swiss participants did not switch to (Swiss) German more frequently than the Anglophone interviewees did, despite conversing in their L2. However, the native speakers of English hedged a larger percentage of their spontaneous as well as metalinguistic switches than the Swiss participants, which could indicate that the former have a more negative attitude towards language mixing, or that they felt self-conscious about speaking in their L2.

In addition, the analysis of emotion words demonstrated that the Swiss used slightly more negative and fewer positive emotion expressions than the Anglo‐ phone participants and that, contrary to expectation, the Swiss participants used a slightly greater variety of positive emotion terms, while their partners used a greater variety of negative emotion terms. The Swiss participants used more intonation units with multiple stresses in the context of positive emotion words, whereas the Anglophone participants used more of them with negative emotion words. Moreover, the Swiss used more husky/breathy voice quality and falling terminal pitch when expressing emotions than the speakers of English did, while the latter used more level pitches in the emotional contexts.

With regard to swearing, I anticipated that the native speakers would swear more frequently than their non-native partners. This was indeed the case: the Anglophone participants used more swearwords, in both languages, and in the spontaneous as well as the metalinguistic context, than the Swiss participants did. Particularly, they used a far greater number of mild swearwords than their partners in both contexts. In addition, culture emerged as an influential variable, as the British participants swore the most among the interviewees, and the Americans the least, which matched their own reports. The range of reactions that the interviewees reported and demonstrated to swearing in their first language emphasizes that swearwords have a greater emotional force in a bilingual's L1 (Dewaele 2004a, 2004b, 2010), and that the offensiveness of a word is also connected to the speaker, not just the word itself. Language ownership was thus a recurrent topic in the interviews, and several native speakers forbade their partners to use specific terms.

Finally, I aimed to determine the role of the participants' mother tongue in their laughing behaviour, and in their production of humour. The native speakers were expected to have a certain advantage in producing successful laughables, which proved to be true, as the Anglophone partners laughed less about their partners' humorous comments, and slightly more frequently about their own, than the Swiss participants did. It cannot be determined whether this is due to the Anglophone participants' advantage as native speakers, or due to cultural aspects of their sense of humour. In the couples' self-evaluations, there was a tendency to rate the English speakers as funnier than the Swiss, and their responses also indicated that several participants from Great Britain believe their (cultural) sense of humour to be superior to their partner's; such a tendency was also observed by Chiaro (2009) among Anglophone partners in bilingual relationships. In addition, the interviewees referred to a variety of cultural differences in their humour preferences and their senses of humour. Some participants stated that humour can be difficult to translate, or that non-native speakers are sometimes excluded from participating in humorous social interactions due to their limited language skills or their different sense of humour, and may not be able to present their humorous side in their L2. This indicates that humour can also pose challenges for bilingual couples.

#### **11.5 Other factors**

In addition to the partners' gender and mother tongue, there are several other factors which potentially influence their speech. Thus, relationship length turned out to have an impact on the mixing behaviour of the couples, as the bilinguals who had been in a relationship for the shortest period of time switched the least, while the couples who had been together the longest switched the most. Given that relationship length is also linked to the bilinguals' level of skills in their L2, this comes as no surprise.

Moreover, the family situation has emerged as an influential variable in a variety of areas of the partners' speech. When children are born, the couple language evolves into a family language, and language use and mixing practices may change. Often, the non-native partner of the relationship language starts speaking his or her mother tongue to address their children, which increases the presence of this language in their relationship. As a result, the couples' language mixing may increase, or it may decrease because the parents attempt to keep their languages separate for the sake of their children; while the couples in this study reportedly reduce their language mixing for their children, no such tendency was observed during the interviews. Meanwhile, the influence of children on the couples' swearing behaviour was evident; during the interviews, the couples without children swore more than twelve times as much as the participants who have children, even though no children were present.

Another factor which possibly affects the couples' language use is their place of residence. While all of the couples live in Switzerland, they live in different geographic locations, which has a considerable influence on their language use outside the home, since the availability of other native speakers, but also of fluent non-native speakers, is largely determined by the couples' place of residence. The proximity of other native speakers is likely to have an effect on the immigrants' L2 skills, and, potentially, on the partners' language use with each other, though the couples' reports provided little evidence for the latter.

#### **11.6 Implications**

The findings of this study carry a number of implications for bilingual couples, particularly with regard to expressing negative emotions, swearing, and cre‐ ating humour. The couples' reports indicate that the expression of negative emotions can be very challenging even for highly proficient L2 speakers, and that this is — as is probably the case for monolingual couples — one of the most difficult areas to navigate in their relationships. Thus, while the non-native interviewees in this study believe that they have acquired considerable arguing skills, they still feel in a weaker position on occasion, or cause offense to their partner unintentionally. Differences may be as subtle as the suprasegmental features that are used to accompany expressions of negative emotion. Such expressions were framed particularly often by husky, creaky or breathy voice quality and a wide intonation contour among the Swiss participants, and there was also a greater occurrence of rising and falling terminal pitch among them. Native partners should be mindful of the difficulties that their partners face and show some leniency where possible. This also applies to the production of humour, where non-native speakers are sometimes placed at a disadvantage, as their humour may not be appreciated, or they do not manage to find the right tone or words to produce successful laughables. As a consequence, they may not be able to present their humorous or witty side.

Despite these challenges, the analysis of the partners' interactions also demonstrated that they communicate very successfully, and that they have adapted their language practices to each other to a considerable extent. Many of the couples believe the effective communication between them to be a consequence of their bilingualism, and of their long-lasting efforts to engage with each other and to advance their mutual understanding. This indicates that we need to move away from the assumption that bilingual relationships are inherently more problematic than monolingual ones. Instead, bilingual couples who are at an earlier stage in their relationship, or who have a lower proficiency in the couple language, should be heartened by the fact that it is possible to attain successful and satisfying couple communication with dedication and time.

The results of this study also have some practical implications for foreign language teaching, in particular with regard to the use of swearwords and the production of humour in a non-native language. The partners' different views of — and emotional reactions to — swearwords in their L1 and L2 suggest that this is a potential area for misunderstandings between native and non-native speakers. It is therefore important that foreign language students are made aware of the affective meaning of swearwords, especially very offensive ones, as well as the taboos and the notion of ownership associated with swearing in one's mother tongue. With regard to the use of humour, the interviews demonstrated that this is a challenging area for language learners, who may not only be at a linguistic disadvantage, but whose sense of humour may also not be fully appreciated by native speakers. Consequently, foreign language students should be taught about cultural and linguistic idiosyncrasies of humour in their target language.

#### **11.7 Future research**

The analysis and the results presented in this book are by no means complete, and there are many areas which merit further research. On the one hand, the tendencies which were noted in terms of the assimilation between the partners, but also with regard to the influence of gender and mother tongue on their language use, need to be corroborated in a larger sample. On the other hand, there are many aspects that pertain to the communication within couples that could not be examined, as this would have gone beyond the scope of this study. One aspect that stood out in the couples' interactions, which was also evident among the bilingual couples studied by Piller (2002a), was the large amount of conversational collaboration between the partners. Even though Piller examined examples of such cooperative elements, a quantitative analysis of bilingual couples' use of strategies such as joint story-telling, other-repetition, paraphrase, backchannel, simultaneous speech, and shared laughter, would be of interest. Another aspect in the communication of bilingual couples that would benefit from broad-based empirical research is their management of conflict, including negotiation, disagreement, and repair.

In this study, the partners' use of suprasegmental features was mainly researched in the context of laughter and the expression of emotions, besides being taken into account for the interpretation of the interviewees' statements. A more extensive analysis of bilingual partners' use of voice quality, loudness, and pitch, including terminal pitch movements, could provide insights with regard to the assimilation between the partners, and point to more differences relating to gender and mother tongue. In this context, it would be interesting to examine the connection between laughing and smiling voice quality more closely, and the use of laughter, as the analysis suggested that the former correlates with laughter, which was not the case for the latter.

Finally, the two main variables that have been found to affect the partners' speech and modes of interaction — namely gender and mother tongues — are by no means the only influential variables. Variables that were not examined in detail here are the age, level of education, religiousness and social class of the bilinguals, as well as their family situation and relationship length. The impact of these variables on aspects such as the partners' language mixing, swearing, laughter or attitudes should be researched more extensively in the future.

#### **11.8 Strengths and limitations**

The primary appeal of this study is that it provides a detailed linguistic analysis of the speech and modes of communication of bilingual couples who are highly proficient in the relationship language, despite being late bilinguals. Using innovative and diverse research methods, it fills some gaps in previous research on bilingual couples, much of which focused on the experiences of the female partner (e.g. Heller and Lévy 1992; Gonçalves 2013), or on couples' self-reports rather than their actual language use (e.g. Breger 1998; Khatib-Chahidi, Hill and Paton 1998). Studies on couples with similar linguistic and cultural backgrounds as the couples I interviewed concentrated on the discursive construction of their bilingual identity and couplehood (Gonçalves 2013; Piller 2002a), as well as on the conversational cooperation between the partners (Piller 2002a). Through the focus on the partners' speech and interaction, my research generates new knowledge about the communication of bilingual couples, as well as fluent late bilingualism, a field which has received less scholarly attention than childhood bilingualism or less proficient bilingualism in the past. Meanwhile, the focus of my research also differs from most other research in the field of bilingualism in that bilingual couples form an established communicative dyad, which entails that their language practices have evolved and assimilated over time.

In addition, there were many advantages to the design of the research, as it enabled me to analyse each topic from several perspectives, and to consider a number of variables. By combining the couples' reports with the analysis of their actual linguistic behaviour, and by using quantitative as well as qualitative research methods, I avoided giving only a snapshot of their current situation and managed to provide an extensive account of their communication and language use. Furthermore, the fact that the interviews were of similar lengths, that the groups were balanced in terms of gender and mother tongue, and that the same general topics were discussed, contributed to enabling comparisons with reference to certain features (such as language mixing, laughter, and the use of swearwords) and to the factors of gender and mother tongue. Since group size and group composition during a conversation can be confounding variables with regard to certain linguistic modes of behaviour, the fact that these were identical in all interviews was another advantage of the research design. For instance, this was relevant for the comparison of language mixing and laughter, as group size has been found to affect the frequency of laughter (Mehu and Dunbar 2008), and group composition can influence the genders' laughing behaviour, their choice of humour type, and their mixing behaviour (Robinson and Smith-Lovin 2001; Holmes 2006).

Nevertheless, there are also some limitations to this study. First, the con‐ versations that were analysed did not occur completely naturally, and even though the interviews were designed to engender as much spontaneous speech between the partners as possible, the interviews were still semi-structured, and my presence might have affected their language use and their manner of communicating. As a consequence, the observations made with regard to the couples' language mixing, laughing, and swearing behaviour during the interviews are only completely valid for this setting. Secondly, self-reports can provide valuable insights, but they can also be unreliable; the interviewees may not have remembered their past attitudes and linguistic practices clearly or accurately, or they may have misreported certain aspects. This issue was partly remedied by comparing the couples' accounts with their actual language practices. To assess the development of their attitudes and their language use more precisely, however, a longitudinal study would be necessary.

Thirdly, it should be called to mind again that many of the findings only apply to couples in these specific circumstances, rather than to bilingual couples in general, and some aspects are strongly influenced by the couples' environ‐ ment. For instance, the couples' extremely positive attitudes towards their bilingualism, and their strong inclination towards English, are most likely linked to the fact that English is viewed very positively in Switzerland, which shapes not only the Swiss partners' attitudes, but also determines the Anglophone partners' experience as migrants in Switzerland. On the one hand, the latter are looked upon very favourably, often accommodated linguistically, and welcomed as elite immigrants; on the other hand, their mother tongue may make it more difficult to acquire the community language, and might thus keep them from fully integrating. The strong influence of the environment becomes evident when the attitudes of the couples in this study are compared with those of the bilingual couples in Australia examined by Schüpbach (2009), who expressed far less positive attitudes towards bilingualism, even though the couples in both studies speak the same languages. Nonetheless, the fact that the sample was restricted to couples with similar backgrounds and in similar situations provided a comparable basis for the analysis.

#### **11.9 Closing remarks**

When I started out with my research, I hypothesized that the high level of proficiency of the bilingual couples I interviewed entails that they have not only very positive attitudes towards their bilingualism, but that they have also learnt to communicate on a very advanced level, including the successful expression of positive and negative emotions, as well as of humour. In the analysis of the couples' interactions and reports, all of these expectations were met. While it might be assumed that the partners' different languages, cultures, genders, and family backgrounds present considerable challenges in their communication, the analysis demonstrated that there are also advantages, which, in the case of these fluent bilinguals, are felt to far outweigh potential disadvantages. My research also showed that, even though the community language does not feature very prominently in the couples' discourse, it has nevertheless found its way into their shared language, for instance through the use of playful language, neologisms and blended terms, all of which enrich their private couple language. Like the bilingual couples studied by Piller, who "describe their private language as a central element of their relationship, as a glue that binds them together" (2002a: 222), and use it to perform and construct their bilingual couple identity, the couples in this study view their shared language as an essential component and defining feature of their partnership. Despite the differences between them, they have all managed to find a mode of communication that incorporates elements pertaining to their individual backgrounds, while simultaneously representing their bilingual couple identity.

## **References**


Chafe, Wallace L. 1994. *Discourse, Consciousness, and Time: The Flow and Displacement of Conscious Experience in Speaking and Writing.* Chicago: University of Chicago Press.

Cheng, Winnie. 2003. "Humor in Intercultural Conversations". *Semiotica* 146: 287–306.


Feeney, Judith A. 1995. "Adult Attachment and Emotional Control". *Personal Relationships* 2: 143–159.


## **Appendices**

## **Appendix I: Complete questionnaire**

## **Relationship history**


## **Language use**


## **Language and identity**


## **Attitude (towards languages, cultures, bilingualism)**


## **Language and emotions**


36. How do you feel about using swearwords or taboo words in your second language? How do you feel when your partner uses swearwords in your language?

## **Characteristics**


## **Relationship and future**


\* *Questions in square brackets were only asked if the couple did not volunteer enough information, and/or if there was spare time. Moreover, it occasionally happened that the couple gave an answer to a question in response to another question, in which case it was not asked again.*


#### **Appendix II: Overview of length of recordings**


Table 48: Overview of length of recordings and time transcribed

## **Appendix III: Table of number of words and intonation units for each participant**


Table 49: Number of words, intonation units and words per minute for each participant / couple

Key: IUs = intonation units.

#### English Swiss German total hedged total unhedged group total % all total hedged total unhedged group total % all CS meta 3 11 14 51.9 0 13 13 48.1 CS used emotional reaction 0 2 9 60.0 0 1 6 40.0 triggered by previous switch 0 3 0 0 triggered by topic / unclear 1 2 0 1 asking for a word 0 0 1 1 addressing sb; "off-record" 0 1 0 2 B meta general 9 40 67 58.8 4 37 47 41.2 translating 0 2 0 1 correcting & helping 0 0 0 3 loanblends 0 8 0 0 quoting / imitating 0 8 0 2 B used cultural concepts 3 11 76 50.7 3 12 74 49.3 names people 1 12 1 13 places 5 17 0 19 institutions 2 4 0 5 titles 0 0 0 2 triggered by previous switch 0 4 1 5 triggered by topic / unclear 3 6 1 0 spontaneous reaction, filler 0 1 0 5 asking for a word in E / G 1 1 0 2 can't think of a word 4 0 3 1 off-record 0 1 0 1 total meta 12 69 81 57.4 4 56 60 42.6 total spontaneous 20 65 85 51.5 10 70 80 48.5 total 32 134 166 54.2 14 126 140 45.8

## **Appendix IV: Detailed table of language switches**

Table 50: Language switches by type and mother tongue

Key: B = borrowing; CS = code-switching.


## **Appendix V: Categorisation of swearwords**

Table 51: Categorisation of swearwords used during the interviews Key: E = English; G = (Swiss) German.

None of the swearwords was personally directed at anyone present.

No euphemisms or very mild swearwords were used in (Swiss) German.

## **Index**

accent modifying 113, 126 native 180, 201, 372, 377, 384 non-native 171, 183f., 209, 215, 309, 324, 377, 391 advanced bilingualism 27, 73, 177, 399 advanced communication 26 arguing 118, 202, 258, 274 assimilation 27, 89, 141, 158, 164, 175, 323, 349f., 380, 387 attitudes 31, 86, 99, 124, 169, 173, 390, 398 covert 278 towards L2 186, 188, 193, 214, 218, 386, 392 towards language mixing 130, 134, 139, 152, 154, 163, 386 towards partner's culture 186, 188, 193, 214, 386 towards raising bilingual children 177, 205 towards swearing 307, 317 attraction 169, 178f., 183, 190, 214, 386, 390 backchannel signals 209, 246, 368 base language 134f., 137, 152, 290 bicultural couples 44, 171, 198, 216, 351, 367, 383, 386 individuals 88, 274, 332 biculturalism 43, 52, 99, 173, 205, 216 bilingual children 32 couples 25, 53, 73, 112, 157, 178, 198, 257, 300, 339, 364, 385 elite 41 individuals 53, 88, 100, 140, 144, 238,

300, 338 bilingualism 43, 53 sequential 64 bilingual language mode 133, 135 body language *see* nonverbal communica‐ tion borrowing 130, 136f., 140, 143, 149, 152, 157 challenges 63258 cultural 25, 43, 46, 239, 258 linguistic 35, 53, 96, 100, 197, 203, 221, 274 methodological 64, 72, 77 personal 204 with regard to humour 367, 383 children 51 language choice 394 language mixing 155, 158, 160, 166 raising bilingual 61, 87, 99, 110, 174, 177, 201, 205, 387 cisgender 63, 80 code-switching 130, 137, 140, 144, 149, 152, 157, 206, 314 cognate 131, 136, 146 community language 51, 57, 86, 93, 110, 123, 207, 386 couple 112 covert attitudes 135, 164, 278, 301 covert prestige *see* covert attitudes cultural background 46, 72, 258, 285, 297, 332, 365, 370, 383 cultural concept 131, 146, 163 cultural reframing 52 cultural script 43, 90 culture 43

dialect 36 ideology of 36, 57, 214 regional 36, 64, 98, 111, 175, 195 difficulties *see* challenges diglossia 36, 96, 124, 197, 215, 386, 392 dual-lingual 99, 377 education level of 72, 87, 125, 133, 163 elite bilingual 41, 61, 177, 216, 399 emotion expression 237 emotion-laden word 233 emotion-related word 233 emotion word 222, 233, 237, 243, 246, 252, 393 environment attitudes of 87, 178, 213, 219, 387, 398 different 45, 50, 57, 73, 116, 218 family 259 linguistic 35, 107, 175 natural 93 professional 40, 100, 110, 115, 125 sociocultural 104, 125, 223 euphemism 289, 294, 297, 299, 391 exhalation 105, 189, 240 fall *see* terminal pitch false start 318, 369 family situation 90, 155, 299, 394 final rise 106 first language 53, 64, 80, 104, 146, 152, 203, 209, 226, 302f., 307, 318, 393 first language primacy 178, 218 gender 49, 80, 90, 125, 133, 150, 165, 176, 186, 230, 246, 273, 284, 293, 311, 323, 335, 355, 365, 382, 390 habit 85, 95, 100, 123, 148, 300, 325, 376 halting speech 182 hedging 130, 138, 140, 144, 151, 154, 156 hesitation marker 94, 138, 163 H variety 37

identity couple 25, 54, 84f., 95, 113, 123, 368, 384, 399 cultural 36, 175, 209, 311, 378 discursive construction of 56, 397 dual 164, 204 gender 335 individual 88, 123, 192, 198, 203, 212, 214, 392 loss of 58, 89 multiple 56, 89, 174 national 38, 48, 55f. performance of 56 ideology of dialect 214 of love 50 of native speakership 177 imitation 27, 150, 232, 377, 384 inhalation 105, 189, 212, 240, 314, 369 interruption 138, 246, 252, 272, 316, 348 interview setting 73, 162, 287, 301, 322 intonation contour truncated 246, 251, 272 wide 183, 210, 241, 249, 255, 272, 315, 395 intonation unit 76, 81, 235, 240, 248, 349, 357 joint story-telling 347, 368, 380 ladette 285 language acquisition bilingual 205, 226 first 280 formal 85, 93, 303, 392 natural 93, 227, 303 second 49, 57, 87, 226, 303 language choice at work 100, 125 for swearing 280, 288, 301, 388 influence of gender on 90, 125, 293, 390

with children 91, 205, 218, 268, 394 with each other 65, 83, 91, 123, 135, 140, 275, 385 with friends 102, 107 language desire 25, 54, 171, 214, 387 language ideologies 38, 56, 177 language mixing 129, 135f., 140f., 143, 152, 157, 385, 394 language mode *see* bilingual language mode language skills *see* proficiency language teaching 36, 396 latch 368 laughable 329, 331, 343, 357, 363, 382, 392, 394 laughter episode 338, 340, 348, 355, 361, 380, 388 laughter pulse 338, 355, 361, 382 laughter trigger 340, 348, 351, 362 humorous 343, 357 non-humorous 343, 346, 360 Lehrplan 21 36 lengthening 77, 182, 240, 242 lingua franca 26, 40, 65, 85 loanblend 136, 150, 377 loan shift 130 loanword 39, 130, 137, 158 loudness 240, 255 crescendo 242 diminuendo 242, 366 loud voice 232, 272, 314ff. variation 183, 242 L variety 37 LX 53 marginalization 45, 51, 58, 72, 125 marked speech 76 MAXQDA 78 minced oath 289f. miscommunication 57, 202

misunderstandings 59, 202, 234, 264, 270, 320, 346, 367, 396 mixed discourse 141 mother tongue 252, 267, 295, 307, 360, 392, *see* first language motivation 26, 86, 94, 97, 124, 174, 204 instrumental 88 integrative 88 multilingual 53, 64, 280 multilingualism 35, 87, 90, 125, 390 native tongue *see* mother tongue neologism 375f., 389 network 104, 110, 126, 162 nickname 136, 269, 275, 375 nonce borrowing 39 non-community language 58, 86, 91f., 126, 218 non-native tongue *see* second language nonverbal communication 29, 120, 127, 211, 223 normal rise *see* terminal pitch offensiveness 277, 282, 288, 290, 312, 320 one parent, one language strategy 71, 99, 205, 211, 217 other-completion 55 otherness 172 other-repetition 55, 105, 368 overlap 76, 331, 368 paralinguistic features 231, 235, 332, *see* suprasegmental features pauses 76, 139, 211, 242 performing couplehood 54, 116, 368, 373, 399 culture 56, 301 gender 50 self 56, 88, 123, 171, 192 pitch 272 downward 241, 243, 245, 250, 255, 272 falling 245

high 242, 255, 272, 315 low 241, 272 rising 243 upward 242f., 245, 250, 255, 272 pitch movement *see* terminal pitch prestige 37, 39, 73, 99, 110, 208 proficiency 27, 61, 85, 92, 100, 123, 158, 264, 302, 386 high 58, 63, 139, 141, 153, 218 low 57, 138, 203, 368, 384 prosodic features 231, 235, *see* supraseg‐ mental features pulse of laughter *see* laughter pulse questionnaire 75, 419 recording 29, 73, 422 relationship 364 long-distance 64, 67, 118 long-term 26, 72, 166, 303 relationship length 63, 65, 155, 166, 394 religiousness 48, 299, 305, 317 repetition 182, 209f., 311, 368 rhythmic speech 76 rise *see* terminal pitch second language 53, 57, 64, 166, 224, 265, 267, 302, 305, 370 sense of humour 364, 373, 383 sequential bilingualism 28, 64 setting 165, 387, 398 simultaneous bilingualism 205 simultaneous speech *see* overlap slapstick 370 small talk 60, 192 socialisation first language 281 gender 50 second language 226ff., 281, 303 speech halting 182, 272 marked 76f., 118, 182, 200, 242

rhythmic 76, 242, 391 speech rate fast 113, 232, 242, 272 slow 113, 126, 242, 272 stem 234 stereotypes cultural 47, 49, 60, 188, 194, 230, 351 gender 230, 273 stresses multiple 250, 393 nuclear 76f., 242, 255 strong (contrastive) 182, 206, 242, 250, 255, 315 strong rise *see* terminal pitch struggle *see* challenges subscript 79 suprasegmental features 29, 76, 231, 234, 239, 248, 253, 388 swearing metalinguistic 290, 295, 297, 323 spontaneous 290, 294, 297, 299, 323 swearword 279, 288, 302, 307, 426 switch 130 switching 130 metalinguistic 136, 140, 149, 152f., 163 spontaneous 136, 140, 142, 144, 146, 153, 163 taboo 279, 281, 285, 305, 312, 316, 323 terminal pitch 15, 76, 233, 244, 251, 256, 272, 388 falling 245, 251, 256, 393 glissando 246, 251 level 245, 251, 257, 272, 393 rising 106, 245, 251, 256 strong fall 245, 251, 256 strong rise 245, 251, 256 terms of endearment 228, 266, 269, 275, 375 tongue click 76, 212

Index 431

transcription 15, 27, 76, 422 translinguistic wording 148 trilingual 64 truncated intonation contour 246 word 318 TTR *see* type-token ratio type-token ratio 237, 248, 253, 273 variety of English 72, 78, 175, 179 of Swiss German 98 vocal cues *see* suprasegmental features vocal noises 16, 76

voice quality 16, 76, 231, 240, 255, 366 breathy 106, 182f., 240, 242, 249, 255, 265, 272 creaky 108, 182, 240, 242, 249, 255, 265, 272 husky 180, 183, 206, 240, 249, 255, 265, 305, 393 laughing 242, 272, 345, 349, 357, 361, 382, 388 smiling 242, 272, 349, 357, 361, 382, 388 whispered 108, 183, 210 work environment 40

## **List of figures**


## **List of tables**




## Schweizer Anglistische Arbeiten (SAA)

Begründet von Bernhard Fehr

Herausgegeben von Andreas Fischer (Zürich), Martin Heusser (Zürich) und Daniel Schreier (Zürich)

## Bisher sind erschienen:

Frühere Bände finden Sie unter: http://www.narr-shop.de/reihen/s/schweizeranglistische-arbeiten-saa.html

#### 116

Neff-Mayson Redemption 1990 €[D] 24,– ISBN 978-3-7720-1676-9

#### 117

Lotti Arter-Lamprecht Deutsch-englischer Sprachkontakt 1992, 152 Seiten €[D] 27,– ISBN 978-3-7720-1930-2

#### 118

Karin Schmidli Models and Modifications Early African-American Women Writers From the Slave Narrative to the Novel 1995, XIV, 204 Seiten €[D] 29,– ISBN 978-3-7720-2430-6

#### 119

Marianne Dada-Büchel Katherine Mansfield's Dual Vision Concepts of Duality and Unity in Her Fictional Work 1995, XII, 286 Seiten €[D] 39,–

ISBN 978-3-7720-2431-3

#### 120

Thomas Honegger From Phoenix to Chauntecleer Medieval English Animal Poetry 1996, X, 288 Seiten €[D] 43,– ISBN 978-3-7720-2432-0

#### 121 Esther Loehndorf The Master's Voices Robert Browning, the Dramatic Monologue, and Modern Poetry 1997, VIII, 210 Seiten €[D] 39,– ISBN 978-3-7720-2433-7

#### 122

Michael C. Prusse "Tomorrow is Another Day" The Fictions of James Gordon Farrell 1997, XII, 237 Seiten €[D] 39,– ISBN 978-3-7720-2434-4

#### 123

Thomas Hermann ››Quite a Little About Painters‹‹ Art and Artists in Hemingway's Life and Work 1997, XII, 240 Seiten €[D] 39,– ISBN 978-3-7720-2435-1

#### 124

Verena Bühler Roth Wilderness and the Natural Environment Margaret Atwood's Recycling of a Canadian Theme 1998, VIII, 206 Seiten €[D] 34,– ISBN 978-3-7720-2436-8

#### 125

Ruth Frehner The Colonizers' Daughters Gender in the Anglo-Irish Big House Novel 1999, XIV, 256 Seiten €[D] 39,– ISBN 978-3-7720-2437-5

#### 126

David Stamm A Pathway to Reality Visual and Aural Concepts in Dorothy Richardson's "Pilgrimage" 1999, VIII, 267 Seiten

€[D] 39,– ISBN 978-3-7720-2438-2

#### 127

## Katharina Gutmann

Celebrating the Senses An Analysis of the Sensual in Toni Morrison's Fiction 2000, VIII, 232 Seiten €[D] 39,– ISBN 978-3-7720-2439-9

#### 128

## Yvonne Studer

Ideas, Obsessions, Intertexts A Nonlinear Approach to Russell Hoban's Fiction 2000, XII, 539 Seiten €[D] 48,– ISBN 978-3-7720-2440-5

#### 129

#### Sandra Engler "A Career's Wonderful, but Love Is More Wonderful Still" Femininity and Masculinity in the Fiction of Mills & Boon 2004, XIV, 204 Seiten €[D] 44,– ISBN 978-3-7720-8065-4

#### 130

Roger Meyenberg Capel Lofft and the English Sonnet Tradition 1770–1815 2005, XVIII, 334 Seiten €[D] 49,– ISBN 978-3-7720-8104-0

#### 131

Sarah Chevalier Ava to Zac A Sociolinguistic Study of Given Names and Nicknames in Australia 2006, XVI, 219 seiten €[D] 39,– ISBN 978-3-7720-8139-2

#### 132

Cordula Hawes-Bilger War Zone Language Language and the Conflict in Northern Ireland 2007, 399 Seiten €[D] 74,– ISBN 978-3-7720-8200-9

#### 133

## Nicole Frey Büchel

Perpetual Performance Selfhood and Representation in Byron's Writing 2007, X, 224 Seiten €[D] 52,– ISBN 978-3-7720-8201-6

#### 134

#### Sabina J. Müller Through the Mythographer's Eye Myth and Legend in the Work of Seamus Heaney and Eavan Boland 2007, XII, 291 Seiten €[D] 58,– ISBN 978-3-7720-8215-3

#### 135

Lukas Bleichenbacher Multilingualism in the Movies Hollywood Characters and Their Language Choices 2008, XIV, 236 Seiten €[D] 54,– ISBN 978-3-7720-8270-2

#### 136

Stefan Daniel Keller The Development of Shakespeare's Rhetoric A Study of Nine Plays 2009, 310 Seiten €[D] 58,– ISBN 978-3-7720-8324-2

#### 137

Lukas Rosenberger The Swiss English Hypothesis 2009, XVI, 254 Seiten €[D] 58,– ISBN 978-3-7720-8333-4

#### 138

Simone Heller-Andrist The Friction of the Frame

Derrida's *Parergon* in Literature 2012, 277 Seiten €[D] 58,– ISBN 978-3-7720-8426-3

#### 139

Philipp Gwerder Gothic Identities

War, Atrocities and Doubles in Philip Caputo's Fiction 2013, 294 Seiten €[D] 60,– ISBN 978-3-7720-8460-7

#### 140

Gaby Frey Private Goes Public: Self-Narrativisation in Brian Friel's Plays €[D] 58,– 2015, IX, 261 Seiten ISBN 978-3-7720-8534-5

#### 141

#### Nicole Studer-Joho Diffusion and Change in Early Middle English

Methodological and Theoretical Implications from the LAEME Corpus of Tagged Texts 2014, XVIII, 231 Seiten €[D] 58,– ISBN 978-3-7720-8539-0

#### 142

Dieter Studer-Joho A Catalogue of Manuscripts Known to Contain Old English Dry-Point Glosses 2017, 306 Seiten €[D] 58,– ISBN 978-3-7720-8617-5

#### 143

Martin Mühlheim Fictions of Home Narratives of Alienation and Belonging, 1850-2000 2018, 384 Seiten €[D] 78,– ISBN 978-3-7720-8637-3

#### 144

Alexander G.Z. Myers Pastoral, Identity, and Memory in the Works of John Banville 2018, 223 Seiten €[D] 58,– ISBN 978-3-7720-8647-2

#### 145

Lyutsiya Staub Revisiting Renoir, Manet and Degas Impressionist Figure Paintings in Contemporary Anglophone Art Fiction 2019, 233 Seiten €[D] 58,– ISBN 978-3-7720-8700-4

#### 146

Rahel Rivera Godoy-Benesch

#### The Production of Lateness

Old Age and Creativity in Contemporary Narrative 2020, 205 Seiten €[D] 58,– ISBN 978-3-7720-8698-4

#### 147

Simone Eva Höhn One Great Family: Domestic Relationships in Samuel Richardson's Novels 2021, 296 Seiten €[D] 58,– ISBN 978-3-7720-8731-8

#### 148

Stefanie Strebel Between Dream Houses and "God's Own Junkyard": Architecture and the Built Environment in American Suburban Fiction 2021, 228 Seiten €[D] 58,– ISBN 978-3-7720-8751-6

#### 149

Silja Ang-Tschachtli Bilingual Couples in Conversation 2022, ca. 400 Seiten €[D] 58,– ISBN 978-3-7720-8763-9

This book provides a detailed linguistic analysis of the communication between highly proficient bilingual couples, each consisting of a native speaker of English and of Swiss German. Combining the accounts of ten couples on their language use with an analysis of their actual linguistic behaviour, several areas of the partners' speech and interaction were closely examined. These include their language choice and language mixing, attitudes, expression of emotions, swearing, as well as their humour and laughter. In addition, the influence of the bilinguals' mother tongue and gender on their language use was explored. Thus, the study provides valuable insights into the language practices of established bilingual couples, while also contributing to the fields of fluent late bilingualism and gender research.

SCHWEIZER ANGLISTISCHE ARBEITEN

**Bilingual Couples**

**in Conversation**

SWISS STUDIES IN ENGLISH

Silja Ang-Tschachtli

SAA 149

Ang-Tschachtli

**Bilingual Couples in Conversation**