Intra-sentential code switching in the speech of bilingual children

Table of Content

1 Introduction

2 Principles and constraints of code switching
2.1 Poplack’s Equivalence Constraint
2.2 Sankoff and Poplack’s Free Morpheme Constraint
2.3 Myers-Scotton’s Matrix Frame Language Model

3 Analysis of data
3.1 Data 1, Hannah (2;4)
3.2 Data 2, Jeremy (2;5.05)
3.3 Data 3, Jeremy (2;5.12)
3.4 Data 4, Olivier (2;8)
3.5 Data 5, Annick (3;1)
3.6 Data 6, Olivier (3;7)
3.7 Data 7, Paula (4;3)
3.8 Data 8, Paula (4;5)

4 Conclusion

5 References

6 Appendix

1 Introduction

In this paper, I analyse the patterns of intra-sentential code switching in the speech of bilingual children aged between of two and four years.

The first section will deal with some of the theoretical approaches to code switching which try to establish certain grammatical rules for this linguistic phenomenon.

Afterwards, I will analyse the code switches in the speech acts of different children in order to find out more about the nature of these switches. The aim is to find out whether code switches, within the sentence boundaries, always occur in the same position or if they can be found anywhere in the sentence. Furthermore, I will analyse what types of words are being switched and if the switches violate the word order of the sentences. All this will be looked at with regard to the principles mentioned in the first section and whether the code switches adhere to these principles or violate them.

The findings and results presented here should be considered with reservation since the brevity of this paper does not allow the analysis of a representative amount of data.

The data is taken from the CHILDES Corpus as well as from Gawlitzek-Maiwald and Tracy’s article Bilingual

Bootstrapping and from Albrecht’s publication I can speak German- und Deutsch: The development and use of code-switching among simultaneous and successive English-German bilingual children.

2 Principles and constraints of code switching

First of all, I will give a short overview over some of the principles and constraints of code switching which have been elaborated by different linguists.

2.1 Poplack’s Equivalence Constraint

According to Poplack’s constraint, a code switch will only appear at that point of the sentence where the structure of the two languages are identical and a switch from L1 to L2 does not violate the syntactic structure of either language (MacSwan 2006: 286).

2.2 Sankoff and Poplack’s Free Morpheme Constraint

This constraint is concerned with the morphology of the bilingual’s languages. It implies that “a switch may not occur between a bound morpheme and a lexical item unless the latter has been phonologically integrated into the language of the bound morpheme” (MacSwan 2006: 286). According to this constraint, the switch in the sentence

“* estoy eat- iendo” (MacSwan 2006: 286) would not be allowed since the lexical item ‘eat’ has not been integrated into the Spanish language.

2.3 Myers-Scotton’s Matrix Frame Language Model

Myers-Scotton categorises one of the two languages as the matrix language (ML) and the other one as the embedded language (EL). The ML is the language dominating the sentence and which “defines the surface structure positions for content words and functional elements” (MacSwan 2006: 291). This model adheres to a “frequency-based criterion” with the ML being the language contributing the majority of morphemes to the discourse of the bilingual speaker (MasSwan 2006: 292). However, the ML is nothing static but can change from one language to the other over a period of time or even within a conversation (MacSwan 2006: 292).

3 Analysis of data

In this section, I will give an analysis of the data in order to find out if the switches adhere to one of the constraints above or not and demonstrate my findings.

3.1 Data 1, Hannah (2;4)

In Hannah’s case the code switch is situated in the middle of the sentence in line 1. The girl starts her sentence with a German subject and a German auxiliary verb. Then, the code switch takes place within the past participle. She starts with the German morpheme ‘ge’ which indicates that it is a past participle and follows with the English past participle ‘climbed’, which is a lexical item. Hannah does not know the correct German past participle ‘geklettert’ and therefore, creates her own variety by blending together the two components which are known to her. The preposition of the sentence is given in English as well. This code switch violates the Free Morpheme Constraint since the child created a word by putting together a bound morpheme of the German language and an English lexical item which has not been incorporated into German.

When the other person asks “what” (l. 2), she takes it as a request for clarification. Therefore, she repeats the whole sentence in English since all words are known to her in this language.

The ML changes throughout the conversation: while the ML is German in line 1, it changes to English in line 3. The word order follows the syntactic rules of English throughout the transcript, even in line 1 where the ML is German. Therefore, this code switch also violates the Equivalence Constraint.

3.2 Data 2, Jeremy (2;5.05)

In line 2, the child is responding to the father’s English question by starting his sentence with the French word “pas”, which is only the last part of the French negative adverb “ne … pas”. This negative adverb is in the same position as the negated English imperative ‘don’t’ would be. Although both English and French follow the same word order, namely SVO, the French word order for this utterance would differ from the English. Therefore, it is obvious that this sentence follows the English word order.

The next code switch is to be found in line 7. Once again, Jeremy switches the word which is in the initial position of the sentence. This time, he switches the conjunction and the personal pronoun in the sentence. Interestingly, he does not use the correct French pronoun. While there is no different pronoun in English, the French language differentiates between the objective pronoun “moi” being used by Jeremy and the subjective pronoun “je” which would have been the correct form to use.

The ML for this transcript is English. Moreover, this transcript confirms Polack’s Equivalence Constraint since all switches are made at positions in the sentence where the sentence structures of both languages are identical.

3.3 Data 3, Jeremy (2;5.12)

In line 3 of this conversation, Jeremy switches the adjective which is at the end of the sentence. The child uses the feminine form of the adjective. This shows that he transferred this word directly from the French language rather than translating it since the French word for ‘stone’ is ‘une pierre’. Therefore, the use of the feminine form seems perfectly justified to him. Moreover, this switch has not taken place due to pronunciation reasons as the English word ‘small’ is being used by the child a little later in the conversation.

An initial switch can be found in the lines 20 and 23 when Jeremy uses, as in Data 1, the French objective pronoun rather than the French subjective pronoun. Another case of the pronoun being switched is to be found in line 82. Here, he uses the incorrect French pronoun again, but this time at the beginning of the sentence.

In line 32, Jeremy switches the adverbial “out” at the end of his sentence while the verb and the negative adverb are in French. One could argue that the subject and the first part of the negative adverb are missing in his utterance. However, the first part ‘ne’ is not used in negations in spoken French.

The next intra-sentential code switch is to be found in line 43. Here it is at the beginning of the sentence. Jeremy uses a French subject and a French verb. Once again, he starts off by using “moi” instead of ‘je’. This time however, he realises his mistake and corrects himself by using the correct pronoun. The child uses a French verb after this but then continues in English.

[...]