Code-switching in bilingual communication

Contents

I Introduction

II Bilingualism and Code-switching
1 Introduction to Bilingualism
1.1 Definition
1.2 Who is Bilingual?
1.3 Research Perspectives
2 Code-switching
2.1 Introduction
2.2 Definition
2.3 Problems in Defining Code-switching
2.3.1 Interference
2.3.2 Borrowing
2.4 Short Research Overview
2.5 Types of Code-switching
2.6 Sociolinguistic and Pragmatic Aspects of Code-switching
2.6.1 Language Choice, Reasons and Functions
2.6.2 Attitudes and Feelings

III Theories of Code-switching
3 Introduction
4 Poplack’s Two-Constraint Model
4.1 The Free Morpheme Constraint
4.2 The Equivalence Constraint
4.3 Discussion and Hypotheses
5 Myers-Scotton’s Matrix Language Frame Model
5.1 Defining the Matrix Language and Background
5.2 The Hypotheses of the Matrix Language Frame Model
5.3 Discussion and Hypotheses

IV Case Study on Russian-German Code-switching
6 Introduction
7 Historical Background
8 Research Method
8.1 Data Collection
8.2 Participants
9 Relevant Characteristics of Participating Languages
9.1 Russian
9.2 German
10 Analysis of Bilingual Speech Data
10.1 Introduction
10.2 Intra-sentential Code-switching
10.3 Tag-switching
10.4 Inter-sentential Code-switching
10.5 Summary
11 Analysis of Language Situation, Use and Attitudes
11.1 Introduction
11.2 Questionnaire Results
11.3 Summary

V Conclusion
Indexes
12 Bibliography
13 List of Figures
14 List of Tables
15 List of Abbreviations
Appendix
16 Recorded Conversation Samples
17 Questionnaire
Eigenständigkeitserklärung

I Introduction

This thesis is about the use of two languages in everyday life. Bilingualism is a facet of nearly every country in the world and code-switching is a widespread characteristic of bilingual speech. Today, bilingualism and code-switching research constitute a great area in linguistics. An obvious and at the same time interesting aspect is that bilinguals will, of course, stay within one language when talking to monolinguals. However, when talking to other bilinguals, they will probably use both languages. Thus, in bilingual conversations, they often switch from one language to another and frequently even within an utterance. The following example shows how a code-switched sentence looks like.

(1) Es ist schon interessant как мы так говорим. Я даже не замечаю, dass wir so reden.

It’s really interesting how we’re talking like that. I don’t even notice that we’re talking like that.

Such kinds of switches call for a special competence of the two languages involved. But how well the bilinguals have to know each of the languages is a justifiable question. These switches are not arbitrary since they may depend on the situation of the conversation, the topic of the conversation, the emotional aspects involved, the language preference of the speaker and the need to express the own identity. Bilingualism and especially code-switching were long considered as “a sign of linguistic decay” (Appel & Muysken 1987: 117). It was claimed that bilinguals are not able to acquire two languages properly and language mixing was often considered ungrammatical. However, both negative and positive views on the consequences of bilingualism have been studied and many studies showed positive evidence on bilingualism and code-switching. In recent years, many researchers agreed that code-switching even results from complex bilingual skills and is a natural communicative strategy.

The goal of this thesis is to look in detail at code-switching in bilingual communication and to contribute to the research on code-switching with the help of the present study on Russian-German bilinguals. I would like to say some words to my personal background and my motivation for having chosen code-switching as the topic for my thesis. My interest in bilingualism originated as a personal concern. In 1991, after the immigration from Kazakhstan to Germany, I was faced with bilingualism. Russian is the language which I first acquired in my first seven years in Kazakhstan whereas my knowledge of German grew out of my life in Germany – that means of my experience with new German friends, schoolmates and teachers. I think that I was conscious of the existence of both languages and that it was not difficult to keep them apart. The reason for telling my personal story is to illustrate that growing up with two languages can also mean integrating two cultures and maybe also two identities.

In informal conversations between people who are familiar with each other and have a shared background, code-switching can occur quite often. When I’m talking with my parents, for instance, we are code-switching all the time. In a formal speech situation and among people who have little in common, however, code-switching may be avoided. I recognized this aspect at work. I have a colleague who has a similar background to me. But switching occurs only very rarely in our conversations – even if we are alone without other monolingual persons participating in the conversation. According to me, this is related to factors such as prestige or formality which influence our language behaviour. Moreover, I noticed that, if it does occur, and I am the one who starts talking Russian, he follows my language choice and inserts some Russian expressions into the conversation. Moreover, my impression is that switching signals, in a sense, a personal relationship with the other bilingual. To conclude, I would say that for me, as a bilingual, bilingualism is a natural fact of life.

The thesis is divided into three main parts. Apart from introduction and conclusion, section II and III provide the theoretical part while section IV provides the practical part. Section II deals with general information on bilingualism and code-switching. The first chapter presents the introduction to bilingualism. After the definition and the question of who is bilingual, the research perspectives on bilingualism are presented in subchapter 1.3. Chapter 2 presents the code-switching research. After the introduction, the definition and the associated problems in defining code-switching are discussed in subchapter 2.2 and 2.3. In subchapter 2.4, a short research overview is presented. Subchapter 2.5 considers the various types of code-switching. Moreover, subchapter 2.6 presents the sociolinguistic and pragmatic aspects of code-switching. This subchapter looks at language choice, reasons and functions of code-switching as well as on attitudes and feelings about code-switching behaviour.

Section III presents two central theories of code-switching which are introduced in chapter 3. Clearly, as many studies have also shown, code-switching is not the random mixing of two languages. Switching is systematic and rule-governed, in other words, follows some core principles. Accordingly, bilinguals have the capacity to differentiate between well-formed and ungrammatical patterns of code-switching. First, in chapter 4, the focus is on Shana Poplack’s grammatical constraints. The Free Morpheme Constraint is described in subchapter 4.1 followed by the Equivalence Constraint in subchapter 4.2. Discussion and formulation of hypotheses are the subjects of subchapter 4.3. In addition, Carol Myers-Scotton’s Matrix Language Frame (MLF) model is presented in chapter 5. After the definition of the Matrix Language in subchapter 5.1, the hypotheses of the MLF model are the subjects of subchapter 5.2. As in chapter 4, discussion and formulation of hypotheses are the subjects of the last subchapter 5.3.

After the presentation of the theories, section IV deals with the present case study on Russian-German code-switching which is introduced in chapter 6. In chapter 7, the historical immigration background of the participants is provided followed by the research method of the case study in chapter 8. Data collection methods are described in subchapter 8.1 and the participants are introduced in subchapter 8.2. Chapter 9 demonstrates relevant characteristics of the participating languages: subchapter 9.1 introduces the Russian and subchapter 9.2 the German language. The next two chapters, 10 and 11, deal with the analysis of data. Chapter 10 is focused on the grammatical analysis and presents the first part of the actual case study. What happens during a language change in the middle of an utterance? How can this be explained grammatically? This chapter discusses different types of code-switching and provides an accurate analysis of real bilingual speech data with respect to the presented theories in chapter 4 and 5: after the introductory part in subchapter 10.1, subchapter 10.2 discusses intra-sentential code-switching, subchapter 10.3 tag-switching, subchapter 10.4 inter-sentential code-switching and subchapter 10.5 summarises the main findings. Chapter 11 represents the second part of the case study which was carried out with the help of a questionnaire. The focus is on the analysis of language situation, use and attitudes in a Russian-German bilingual speech community. Thus, the data will be analysed from a sociolinguistic perspective. This part of the case study is introduced in subchapter 11.1. In subchapter 11.2, the questionnaire results are presented. Afterwards, a summary of the results is given in subchapter 11.3.

In section V, the conclusion will comprise the main points, some suggestions for further discussion and directions for future research.

II Bilingualism and Code-switching

1 Introduction to Bilingualism

Bilingualism is a world-wide phenomenon today. Because of increased information and communication technologies as well as several travel possibilities, it is nowadays easier for people from all over the world speaking different languages and having distinct cultures to get in contact with each other. One consequence of such contacts is that in some way the speech behaviour and therefore also the languages of the people involved are influenced. This leads to several language contact phenomena and consequently, bilingualism arises as a result of contact. Due to migrations all over the world people nowadays cannot elude the phenomenon of bilingualism and nearly every individual is getting confronted with it.

In contrast to multilingualism, which is seen as a societal phenomenon, bilingualism is regarded as an individual phenomenon. In literature bilingualism is normally used for individuals and communities in which two languages are present, whereas the term multilingualism refers to societies where more than two languages are found (cf. Hoffmann 1991: 10). Some linguists make a clear distinction between bilingualism and multilingualism, whereas others, e.g Romaine (1995: 12), use the term bilingualism for speaking more than two languages as well.

According to Wei (2008: 3) there are 193 countries in the world with over 6,000 different languages. These numbers show that the world is a multilingual society – a far more multilingual society than some centuries ago. Several centuries of political, cultural and economic interaction have made the world a world with widespread bilingualism and multilingualism. Multilingualism exists as a result of historical events and migration. All over the world we can find many bilingual communities of migrant origin. Moreover, new multilingual countries have emerged and many linguistic minorities have become bilingual “not only in the language of their own social group and the national language, but often additionally in one of these international languages” such as English, French or Spanish (Milroy & Muysken 1995: 1).

There are nations which are officially multilingual, such as Belgium or Switzerland (cf. Wei 2008:4). It is not true that people of multilingual countries are necessarily multilingual themselves. Belgium or Switzerland may have many monolinguals in their population, while monolingual countries, e.g. Germany or France, are considered as countries with a great number of bilingual people. As a consequence, linguists agree that there is a fundamental distinction between individual and societal bilingualism (cf. Baetens Beardsmore 1982: 36).

With respect to societal bilingualism, language choice and the different domains of language use have to be mentioned here. One language is often used in a certain domain, e.g. family, whereas the other language is used in other domains, e.g. employment or education (cf. Romaine 1995: 30). This can be compared to a kind of diglossia which is the term Ferguson (2000: 65) used for “two or more varieties of the same language” which are used “in many speech communities [...] by some speakers under different conditions.” He distinguished between informal Low (L) and formal High (H) varieties. Ferguson’s term diglossia can also be extended to languages. Fishman refines Ferguson’s phenomenon of diglossia and tries to connect it with bilingualism. He relates “these two research traditions to each other by tracing the interaction between [the] two major constructs: bilingualism [...] and diglossia” (Fishman 2000a: 81). When bilingualism and diglossia interact, the degree of influence is much higher and has considerable effects on the two languages. Societal bilingualism analyses social structures where two or more languages or varieties are spoken.

The focus of this thesis will be on languages, not on varieties, mainly on individual bilingualism and likewise rather on bilingual than on multilingual speakers. The following subchapter gives some definitions of bilingualism. The factors which play a role in defining who counts as a bilingual speaker will be presented afterwards.

1.1 Definition

Till this day there is no standard definition of bilingualism. There does exist a variety of definitions, interpretations and descriptions of bilingualism which were formulated by many linguists. Bilingualism can generally be considered as the use or the presence of two languages. Myers-Scotton (2006: 44) defines bilingualism as “the ability to use two or more languages sufficiently to carry on a limited casual conversation.”

According to Weinreich (1968: 1) “[t]he practice of alternately using two languages will be called BILINGUALISM, and the persons involved, BILINGUAL.” Bloomfield’s interpretation of bilingualism is found in one of the early books on modern linguistics, Language, which was first published in 1933. It is a definition which was also often quoted by other linguists (cf. Hoffmann 1991: 15).

In the extreme case of foreign-language learning the speaker becomes so proficient as to be indistinguishable from the native speakers round him. This happens occasionally in adult shifts of language and frequently in the childhood shift just described. In the cases where this perfect foreign-language learning is not accompanied by loss of the native language, it results in bilingualism, native-like control of two languages. [...] Of course, one cannot define a degree of perfection at which a good foreign speaker becomes a bilingual: the distinction is relative. (Bloomfield 1970: 55-56)

Like Bloomfield, Mackey (2000: 27) amongst others defines bilingualism in connection to the degree of proficiency: “Since bilingualism is a relative concept, it involves the question of DEGREE. [...] Bilingualism is a behavioural pattern of mutually modifying linguistic practices varying in degree, function, alternation, and interference.” Moreover, he states that “[b]ilingualism is not a phenomenon of language; it is a characteristic of its use” (Mackey 2000: 26). Furthermore, Haugen (1987: 14) rephrased Bloomfield’s definition as “native competence in more than one language” stating that this definition is an ideal model and only few people actually achieve this. So, he states that the definitions of bilingualism can be classified in narrow and wide definitions.

1.2 Who is Bilingual?

Bilingual people normally use their languages, either separately or together, because of different reasons, in different domains of life and with different people. It is not quite clear how well the languages need to be known for counting as bilingual and there exist several types of bilinguals. There are many factors which have to be taken into account when describing bilingualism – these are factors such as age, context, relationship between sign and meaning, order and consequence of bilingual language acquisition, competence or level of proficiency, function or use and attitudes (cf. Hoffmann 1991: 18f).

Multilingual and bilingual speakers can communicate in more than one language. There are distinct groups of people speaking more than one language. Wei (2008: 4) distinguishes between early and late bilinguals. People have become bilingual through different experiences. Early bilinguals have acquired two or more first languages since birth while late bilinguals have acquired and maintained one language during childhood and learned other languages later in life. In bilingualism research, the distinction between early and late bilingualism is a crucial one. The most common way to become bilingual is being born into a bilingual family. There are, however, many other reasons like personal, social or cultural reasons why individuals become bilingual. It has to be distinguished between bilinguals who started their speaking years as monolinguals and bilinguals who acquired two languages simultaneously as young children. When the languages are acquired one after the other, the researchers speak of a first language L1, the language that is acquired first as a child, and a second language L2 which is acquired later, either as a child or adult (cf. Myers-Scotton 2006: 2). One central aspect is the critical age hypothesis which

presents the notion that all children easily acquire any language to which they are exposed up until about the age of puberty; after that, acquiring a language becomes more arduous and more of a conscious procedure. (Myers-Scotton 2006: 36)

This hypothesis is attributed to Lenneberg (1967) and was discussed by many researchers. During the critical period the ability to acquire language reaches its peak, thereafter, the ability declines. There is a lot of disagreement over the critical age hypothesis. Researchers, however, do agree that “late learners are much less successful in language learning than young children” (Myers-Scotton 2006: 350). Language learning after a certain age is effortful and foreign accents are usual. Late acquisition generally means less proficiency. The study of second language acquisition (SLA) is separately an entire field of research, but also an aspect of the broad concept of bilingualism.

The two languages may have been learned either in a fused context or in separate contexts, i.e. it has to be differentiated between both parents using both languages to the child or the one-parent-one-language principle. The distinction can also be made between the two languages spoken in the same locality or one language is learnt in one country and the other in a different one (cf. Hoffmann 1991: 18f).

Another important aspect is the order of bilingual language acquisition and the consequences which may arise.

Qualitative, as well as quantitative, judgement is expressed by the terms ‘additive bilingualism’ and ‘subtractive bilingualism’ [...]. The former implies that the addition of a second language to a person’s first can result in enriched, or at least complementary, social, cognitive and linguistic abilities, whereas the latter suggests that the L2 is learnt at the expense of the L1. [...] This means that, although they are becoming more proficient in the L2, they are losing skills in the L1; therefore, as the latter is not being maintained, it is actually being ‘subtracted’ from their bilingual proficiency. (Hoffmann 1991: 21)

The processes of language maintenance and language shift are possible consequences of becoming bilingual. An interesting aspect is what happens to the L1 of speakers when they become bilingual. One possibility is that the L1 is maintained and the L2 is learned as an additional language. Language shift means that bilingual speakers shift to the new L2 as their main language.

But in most cases, it takes an entirely different generation for the shift to the L2 to go to any sort of completion. In fact, shift to an L2 generally doesn’t happen until the third generation, with the second generation bilingual in both languages. (Myers-Scotton 2006: 68)

As for the purpose of this thesis, the focus is mainly on immigrant communities. There are many factors which influence maintenance or shift of L1.^[1] Myers-Scotton classifies maintenance and shift within a bilingual community by means of a continuum: those bilinguals “who use only the L1 at one end to those who use only the L2 at the other end” (2006: 89). Regarding language shift, an interesting aspect is the younger generation. “[E]specially in immigrant communities, shift by the third generation is almost a foregone conclusion” (2006: 100). Further research on that aspect will be presented in chapter 11 of section IV, Case Study on Russian-German Code-switching.

The question of how proficient a bilingual person has to be in both languages for counting as a bilingual is connected to the notion of relativism and the degree of bilingual competence. In literature there is made a distinction between the two poles of the continuum, the maximalist and minimalist view of bilingual competence. With respect to the maximalist view of bilingual competence, linguists speak of perfect bilinguals. The minimalist view of bilingual competence, however, considers a person who has only a minimal competence in one of the four language skills, listening, reading, speaking and writing in a language different from his or her mother tongue, as bilingual as well (cf. Hamers 2005: 34). Somewhere in the middle of the continuum there can be situated the notion of balanced bilingualism. According to Hoffmann (1991: 22) most bilinguals tend to be more fluent or proficient in one language which can be considered as the dominant one, i.e. they have a stronger or preferred language and a weaker one.

Bilinguals talking to each other have to decide which language or languages they want to use. Most bilinguals, however, seem to know which language is the most adequate in a certain situation. Bilinguals can communicate with each other either using the one or the other language. Many bilingual speakers, however, tend to switch between the languages when they are talking to other bilinguals which are also able to speak the same languages. This linguistic phenomenon, which arises because of the fact that the structures of two languages are in contact, is known as code-switching. Code-switching is one of the most studied topics in language contact phenomena.

1.3 Research Perspectives

Research on bilingualism has a long history but especially from the 1970s bilingualism has become a major focus of research. Bilingualism can be studied from different perspectives – from a linguistic, a psycholinguistic and a sociolinguistic point of view (cf. Wei 2008: 4f). It is, however, not always easy to draw a clear line between the various research perspectives; there exist complex psychological, linguistic and social interrelationships in bilingualism research. The main aspects of all three research areas will be presented in this chapter.

With respect to the linguistic perspective, there are three important questions which play a significant role in linguistics as well as in research on bilingualism. The first question deals with the nature of language or grammar in the mind of a bilingual person and with the interaction of different systems of language knowledge. Wei (2008: 5) points out that one substantial feature is the bilingual speakers’ ability to move between different languages, i.e. to code-switch. This can happen between sentences, in the same sentence, clause or even in one word. Code-switching takes place at specific points in a conversation and seems to agree with the grammatical constraints of the languages involved. There are several grammatical constraints on code-switching. The theories significant for this thesis will be presented in section III, Theories of Code-switching. The second question concerns the bilingual acquisition of more than one language, i.e. of more than one grammatical system. Much of the language acquisition research focuses on infants. But the aspect of SLA has to be considered as well and “[c]learly one of the key objectives of second language acquisition is to become bilingual” (Wei 2008: 8). Additionally, Wei states that SLA is generally considered to be different from the research on bilingualism, but “second language learners and other later-acquired language users are regarded as an important and distinctive group of bilinguals” (2008: 8). The third question is about the use of language knowledge – how is the knowledge of two or even more languages used by the same speaker in bilingual interaction. Language choice is an important aspect here – bilinguals choose the language for example according to the topic of the conversation, the participants involved or the setting.

While studying bilingualism from the psycholinguistic perspective, psycholinguists are more concentrated on investigating bilingual behaviour, language choices and the cognitive processes involved than on the structure of bilingual speech. But they are interested in basically the same three key points from the linguistic point of view – namely in bilingual knowledge, acquisition and use (cf. Wei 2008: 9).

Psycholinguistic research on bilingualism was much influenced by the work of Weinreich, Languages in Contact. Weinreich (1968: 1) describes different language contact phenomena and puts emphasis on the fact that each individual has to be analyzed: “[...] two or more languages will be said to be IN CONTACT if they are used alternately by the same persons. The language-using individuals are thus the locus of the contact.” He classified three different types of bilinguals – “co-ordinate (systems kept apart), compound (systems drawing on common mental representations), and subordinate (one system is represented in terms of the other)” bilinguals (Milroy & Muysken 1995: 5). For psycholinguists it is important to investigate what is going on in a bilingual person’s mind and to develop models of bilingual mental lexicon. Many psycholinguistic factors may influence or affect the linguistic behaviour of bilingual people. One major aspect is the question as to how a bilingual person accesses or selects words from different languages. Bilinguals are able to produce sentences with elements from more than one language. Psycholinguists try to find out what the rules that the bilinguals seem to follow are and how language is processed in the brain of a bilingual person.

From the sociolinguistic perspective, bilingualism is seen as a socially constructed phenomenon and the bilingual person as a social actor. Wei (2008: 13) states that “the issue of language use that linguists and psycholinguists are concerned with becomes an issue of identity and identification for the sociolinguist”, i.e. every time a bilingual person says something in one language when he or she could also have said it using the other language, the bilingual person reconnects with situations or people from past interactions. Wei (2008: 13) continues arguing that through language choice, bilinguals “maintain and change ethnic group boundaries and personal relationships.”

Self-identification, i.e. whether a bilingual person feels himself or herself totally identified with his or her languages, and identification by others, i.e. whether the bilingual is accepted by others as belonging to both linguistic communities, are the two features which Hoffmann (1991: 26) uses when talking about attitudes. Attitudes are a significant factor when describing bilingualism. Consequently, this leads to the fact that a bilingual person should be conscious of his or her bilingualism. On the one hand, the focus of sociolinguistic research is on the bilingual individual and on the other hand on multilingual communities.

2 Code-switching

2.1 Introduction

As has been indicated in previous chapters, one of the most studied and maybe most important phenomenon in bilingualism research and language contact is code-switching. Up to now many studies on code-switching have been carried out and often different terminologies as well as spellings were used. As a consequence, various definitions and interpretations of code-switching can be found throughout the history of research on code-switching. Moreover, there are different spellings of code-switching in literature. Some linguists write ‘codeswitching’, ‘code switching’ or ‘Code Switching’, whereas others use the hyphenated form ‘Code-switching’ or ‘code-switching’. The latter spelling will be used in this thesis.

In many situations bilingual speakers make use of the grammar and lexicon of two languages.

(2) У неё Stress на работе… с работы она пошла zum Arzt и он её на всю неделю krankgeschrieben и сказал, чтобы она am Freitag ещё раз пришла… sie nimmt die Arbeit echt ernst.

She is under a great deal of stress at work... from work she went to a doctor and he signed her off sick for the whole week and said that she should come again on Friday... she takes work seriously.

This example is a normal everyday occurrence of bilingual language use. Utterances of this sort probably occur in most bilingual speakers’ conversations regardless of which languages. Weinreich is of the opinion that the

ideal bilingual switches from one language to the other according to appropriate changes in the speech situation (interlocutors, topics, etc.), but not in an unchanged speech situation, and certainly not within a single sentence. (Weinreich 1968: 73)

In literature there are many contradictions to the statement made by Weinreich (cf. Romaine 1995: 124). Firstly, there are many doubts whether the ideal bilingual exists or what is exactly meant by an ideal bilingual. Moreover, as the example above shows, switches within a single sentence are not uncommon among bilingual speakers. Romaine (1995: 124) also shows that switching with no apparent change in topic or interlocutor is possible.

One important question is what is meant by the term code-switching and how linguists describe this phenomenon. What are other language contact phenomena? Where does code-switching seem to be possible syntactically? Are there different types of code-switching? In what kind of situations is code-switching possible or probable? In which activities do bilingual people tend to switch from one language into the other? Do bilingual speakers behave in another way in different social situations? Another significant question which arises is why bilingual speakers are doing this kind of switches. Thus, what motivates bilingual people to mix two languages? This is related to the bilinguals’ awareness of switching between the languages in a conversation. What do bilinguals think about their language mixing? What are their attitudes about bilingualism and code-switching? How do they perceive bilingualism and what are their feelings about bilingualism? Do bilinguals have two identities or personalities or just one? These are the most important questions which will be discussed in this thesis, particularly in this chapter and in section III, Theories of Code-switching. In section IV, I will apply the theoretical background on my case study and try to answer the above listed questions with respect to my case study.

First, some definitions of the term code-switching will be discussed in this chapter. As many books have been written about this topic, there are many definitions of various linguists. Since it is not always easy to find a clear-cut definition for code-switching, it has to be distinguished between code-switching and other similar language contact phenomena. Subchapter 2.3 will deal with this kind of definition problems. The phenomenon of code-switching has its origin in bilingualism research and code-switching can also be examined from different perspectives. So, after a short research overview with respect to grammatical, psycholinguistic, sociolinguistic and pragmatic approaches to code-switching, the various types of code-switching will be described. The last subchapter 2.6 considers the sociolinguistic and pragmatic aspects of code-switching.

2.2 Definition

Code-switching is the term for different languages coming into contact with one another in a conversation. This could be a general definition. Most of the linguists doing research on code-switching refer to it simply as the alternating use of several languages or the alternation of two or more languages. Thus, general terms for the code-switching phenomenon are also code alternation and language mixing. Bilingual or multilingual speakers, who speak two or even more languages, normally tend to code-switch when speaking to another bilingual person; that means that they often change from one language to the other and use words and phrases from distinct languages, even in the same sentence. Hamers & Blanc (1989: 148) even say that code-switching is “one of the most common and original strategies used by bilingual speakers among themselves.”

Grosjean (1982: 145) defines code-switching as “the alternate use of two or more languages in the same utterance or conversation.” According to Wei (2000: 16) code-switching occurs “when a bilingual talks to another bilingual with the same linguistic background and changes from one language to another in the course of conversation.” Milroy & Muysken (1995: 7) describe the phenomenon of code-switching as “the alternative use by bilinguals of two or more languages in the same conversation.” They point out that switching sometimes “occurs between the turns of different speakers in the conversation, sometimes between utterances within a single turn, and sometimes even within a single utterance.” Poplack (2000: 224) defines code-switching in a similar way as “the alternation of two languages within a single discourse, sentence or constituent.” Similar to Poplack, Grosjean shows that code-switching can involve several sentences, only one sentence, a phrase or a word (cf. Grosjean 1982: 146). For Hamers & Blanc (1989: 148), code-switching occurs when “two languages are present in the discourse, chunks from one language alternating with chunks from the other.” Moreover, they say that a “chunk can vary in length from a morpheme to an utterance.” These observations lead to the classification of different code-switching types which will be presented in subchapter 2.5.

2.3 Problems in Defining Code-switching

In the study of language contact there has been little agreement on the definitions of the numerous language contact phenomena. Some linguists use, for instance, the term language mixing or code-mixing in a more general way for explaining several language contact effects while others describe code-mixing as switching within a sentence. This subchapter will deal with these terminology problems and differentiate between code-switching, interference and borrowing. Haugen distinguishes between switching, interference and integration. Switching is “the alternate use of two languages”, interference “the overlapping of two languages” and integration “the regular use of material from one language in another” (1956: 40). Besides code-switching, interference and borrowing are two of the most important phenomena of language contact which will be described below. In Haugen’s terminology integration is the phenomenon of borrowing.

2.3.1 Interference

“[T]he influence of the bilingual’s language systems upon each other” – this is how Hoffmann (1991: 19) defines the phenomenon of interference. Already Weinreich (1968: 1) has studied the language contact phenomenon of interference.

The term interference implies the rearrangement of patterns that result from the introduction of foreign elements into the more highly structured domains of language, such as the bulk of the phonemic system, a large part of the morphology and syntax, and some areas of the vocabulary.

Appel & Muysken (1987: 84) subscribe to Weinreich’s interpretation and define interference as “the use of elements, structures and rules from the source language in the production of the target language.” For Mackey (2000: 40) interference is “the use of features belonging to one language while speaking or writing another.” Haugen (1956: 50) describes interference as a “linguistic overlap, when two systems are simultaneously applied to a linguistic item.” In earlier studies there was the tendency to group all language contact phenomena under the general term interference.

Romaine uses the term cross-linguistic influence. She states that cross-linguistic influence can occur on different levels of the language and that influence upon one level can have consequences for other levels of language (cf. 1995: 53). Many other researchers, such as Weinreich or Mackey, were of the same opinion and investigated the interference phenomenon taking place at various levels of language. There is interference at the phonological, grammatical or syntactic, and lexical level. The syntactic level is mainly affected with respect to word order divergence, e.g. the placing of adjectives before or after the noun (cf. Romaine 1995: 55). “The determination of the nature and occurrence of interference is clearly due to the structure of the languages in contact which affects the resistance and encouragement of transfer” (Baetens Beardsmore 1982: 63). Lexical interference is also interpreted as lexical borrowing. Furthermore, Mackey (2000: 44f) examines cultural and semantic interference. We can conclude that interference is basically the influence of one language on the other with respect to various levels of language, such as phonology or syntax.

While investigating the phenomenon of interference, as maintained by Weinreich (1968: 3), different extra-linguistic factors are to be analyzed as well. These are factors such as the speaker’s ability to keep two languages apart, proficiency in each language, manner of learning each language and attitudes toward each language (cf. Weinreich 1968: 3). He does not only take bilinguals as individuals into account but also groups of bilinguals.

Interference, “a phenomenon which is also often called ‘negative transfer’” (Appel & Muysken 1987: 84). This is another aspect of the expression interference; it has, to some extent, a negative connotation. For that reason, some linguists also prefer the term transference instead of interference (cf. Romaine 1995: 52).

Negative interference may be used to describe the avoidance of native forms because of their diamorphic equivalence to foreign ones. In bilingual situations this is most conspicuous in the case of lexemes which are socially unacceptable in one, but not the other language. (Haugen 1956: 54)

2.3.2 Borrowing

The phenomenon of interference has to be distinguished from borrowing. Haugen (1956) made one of the earliest attempts to classify types of borrowing based on integration.^[2] Interference is seen as “individual and contingent”, while borrowing is regarded as “collective and systematic” (Mackey 2000: 40). The phenomenon of borrowing has many similarities with code-switching and it is not always easy to make a clear division between the two terms. Moreover, the distinction between intra-sentential code-switching and borrowing is a far more difficult distinction. Intra-sentential code-switching often involves only single-word switches. Borrowing normally involves lexical items which generally also consist of one word. Myers-Scotton (2006: 209) defines lexical borrowing in the following way: “Words from one language appearing in another are lexical borrowings.” Moreover, she states that nouns are the category which is borrowed more frequently. There is a general agreement among linguists that lexical material is the most easily borrowed and that nouns are freer of syntactic restrictions than other word-classes.

Halmari (1997: 17) refers, in some way, to Poplack’s opinion that “a lexical item is not a codeswitch, if it is phonologically, morphologically and syntactically integrated into the host language.” This means also that borrowing is used “to refer to the embedded language items that have been integrated to the matrix language” (Halmari 1997: 17). Borrowed items are incorporated into the target language; they become a part of the other language. The degree of integration of the embedded language items into the matrix language plays a big role with reference to borrowing. This is regarded as a characteristic feature of borrowings. Consequently, a word that has been integrated according to the phonology and morphology of the target language is called borrowing. Thus, borrowed elements often look quite different in the target language. Many linguists distinguish between these two terms defining that borrowing – because of the integration of two languages – belongs to the level of langue of bilingual people while code-switching is a part of the bilingual’s parole^[3] (cf. Appel & Muysken 1987: 121).

[...]

^[1] Just to name some, the following factors promote the L1: ability to visit the home country easily; radio, TV and literature available in L1; types of social networks, i.e. strong ties with in-group memberships; personal emphasis on family ties and psychological attachment to L1 for self-identity (cf. 2006: 90). Their opposites result in language shift.

^[2] Haugen (1956: 59f) distinguishes between two main types of borrowings – between loanwords and loanshifts. Loanwords are phonologically and morphologically integrated into the target language. They are further divided into loanwords and loanblends. Loanblends are a kind of hybrid category – one part is borrowed and the other part belongs to the base language (cf. Grosjean 1982: 313f). In loanshifts, however, the meaning of a word in the base language is extended by an additional meaning – in this case Haugen speaks of extensions. As opposed to loanwords, there is a semantic and not a phonetic influence from the other language. Creations constitute the second type of loanshifts involving “rearranging words in the base language along a pattern provided by the other and thus [creating] a new meaning” (Romaine 1995: 57). Bilinguals often borrow whole idioms from one language into another by translating them literally. These types of borrowings are also called calques or loan translations (cf. Grosjean 1982: 318).

^[3] Langue is the abstract system of signs and rules whereas parole is the realisation of langue in a concrete situation.