UDK 81’246.2–053.2:159.923 159.923.2–053.2:81’246.2 Izvorni znanstveni ~lanak Prihva}eno za tisak 23. rujna 2010. Galina Chirsheva Cherepovets State University Russian Federation gnchirsheva@yahoo.com Self–interpreting by bilingual children This paper aims at describing self–interpreting as a specific skill of simultaneous bilingual children to interpret their own utterances from one language into the other. There are very few works on childhood bilingualism whose authors pay attention to the phenomenon (see Dodson 1984; Saunders 1988; Malakoff – Hakuta 1991/2000; Gergely 1997). The data for the present paper are taken from tape–recorded and video–recorded speech of three boys, who acquired Russian and English from their parents according to the principle “one parent – one language”. Self–interpreting positively influences the development of the child’s bilingualism, while the study of this phenomenon helps to reveal many aspects of childhood bilingualism: the difference in the vocabulary and semantic development in two languages; the difference in the acquisition of two grammars; specific features of hesitations; development of pragmatic bilingual competence. 1. Introduction The main objective of the paper is to show the importance of the specific skill of a bilingual child to interpret his/her own utterances from one language into the other. Self–interpreting influences the development of the child’s bilingualism, while the study of this phenomenon helps to reveal many aspects of childhood bilingualism. The research is done in three Russian families where children (boys) acquired two languages (Russian and English) simultaneously. By the age of two most bilingual children are already aware of their bilinguality and of the fact that they should choose different languages speaking to different interlocutors. The paper will deal with the situations in which bilingual children are especially eager to produce self–interpreting, the most representative of such situations being interaction in the triad “mother – child – father”. Self–interpreting in bilingual children acquiring two languages simultaneously is based on their ability to produce utterances with the same proposition in either language. This kind of activity can precede but usually follows the 173 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) stage of bilingual development when the child has acquired a substantial bilingual vocabulary. Bilingual utterances can demonstrate the volume of the child’s bilingual vocabulary, reveal what grammatical categories and syntactic structures have been acquired by the time the child begins interpreting himself/herself. Besides, informative and expressive value of the utterances having the same proposition but produced in two different languages is also different. Thus, the speech containing self–interpreted sentences reflects how balanced the child’s bilingualism is. Self–interpreting is often accompanied by hesitation; its specific features depend on the direction of interpreting and the dominance of one of the languages in the child’s bilingualism. Simultaneous bilingual children interpret their own words, phrases or utterances both spontaneously and intentionally. Intentional self–interpreting is often represented by cases of self–correction when the child realizes that s/ he has chosen some words from the “wrong” language. Self–interpreting also reflects a bilingual child’s sensitivity to the interlocutor’s language – if the child realizes that s/he has not been understood in one language s/he says the same in another language. In such situations, the process reminds that of dubbing, so such utterances may be also referred to as dubbed ones. Utterances including self–interpreted sentences may be viewed upon as a kind of intrasentential code–switches, so they can be referred to as code–switches as well. The abilities of bilingual children to code–switch and to translate/ interpret have much in common, though it is pointed out that “while translation takes advantage of similarities across two languages, code–switching takes advantage of the differences” (Malakoff – Hakuta [1991] 2000: 146). Self–interpreting characterizes bilingual development of most children, yet it is rarely paid attention to. There are very few works on childhood bilingualism whose authors describe and study this phenomenon. This paper aims at describing the most important specific features of the phenomenon in the development of simultaneous childhood bilingualism. In Section 2 theoretical background for the research is given, in section 3 the data and participants are introduced and the type of their simultaneous bilingualism is described. Sections 4 – 8 are devoted to the description of specific features of self–interpreting in bilingual children’s development. It will be shown that self–interpreting (a) has its stages of development (section 4), (b) reveals the difference in the vocabulary and semantic development of the child in two languages (section 5), (c) signals the difference in the acquisition of two grammars (section 6), (d) reflects specific features of hesitations in the speech of bilingual children (section 7), (e) plays a certain role in the development of pragmatic bilingual competence (section 8). 174 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) 2. Theoretical background When young children hear two languages spoken to them from birth, they acquire both languages as first ones and such bilingual development is referred to as Bilingual First Language Acquisition – BFLA (De Houwer 1990; 1995; 1998; 2006; 2007; 2009; Meisel 1989; 2001; 2008; Swain 1976). Researchers point out that in BFLA “there is no chronological difference between the two languages in terms of when the children started to hear them”, that is why in referring to these languages it is best not to use a notation that imply a notion of L1 and L2 (De Houwer 2009: 2). However, BFLA does not always imply that the children have the same communicative competence in either language; moreover, according to A. De Houwer, some BFLA children understand two languages but speak only one (De Houwer 2009: 2). Those children who start out hearing just one language but very soon, in the first year of life, are confronted with a second language, have both similarities and differences with BFLA and MFLA (Monolingual First Language Acquisition) or ESLA (Early Second Language Acquisition) in their language development (De Houwer 2009: 6). There is still no terminology to refer to such early language learning contexts, but it is better to refer to the child’s languages as L1 and L2. Children can become bilingual in all these language learning contexts, that is they can comprehend and speak two languages at early stages of their development. When researchers study early bilingual development, they always pay attention to bilingual input. The primary socializing agents for the bilingual development of children are their parents who may be bilingual or monolingual in different languages. They may even be native speakers of the same language but choose to speak L2 addressing their children and, in some cases, each other. In the present paper the latter type of childhood bilingualism is determined as mono–ethnic (Chirsheva 2000: 20–21) basing on the ethnolinguistic characteristics of the family where the child was born and brought up. Only one ethnic group is represented within the family, yet parents speak two different languages with their child. Such language development situations when one parent, a native speaker of L1, addresses his/her child in L2 and the other parent speaks with the child in his/her native language (L1), are also referred to as bilingual by most researchers of childhood bilingualism (De Houwer 1990; 1995; 1998; 2006; 2007; 2009; Döpke 1992; Harding and Riley 1997; Lyon 1996 among others). The first cases of mono–ethnic childhood bilingualism were described in 1950s (Stephens 1952) and 1960s (Dimitrijeviç 1965). Since that time their number has been increasing as well as the study of such cases in different countries (see: Aidman 1994, 1998, 1999; Brennan 1987; Chirsheva 1996, 1997, 1998a, 1998b, 1999, 2000; Corsetti 1996; Facey 1986; Korovushkin – Chirsheva 1992; Nataljin – Nataljina 1989; Past 1976; Past – Past 1978; Saunders 1982, 1988; Stefanik 1995; 1997; Totmjanina 1998, 1999, 2000). It is evident that the mono–ethnic bilingual child’s L2 verbal behaviour shows some his/her ethnic related (cultural, social, psychological, pragmatic) 175 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) characteristics, which can be different from those of bi–ethnic bilingual children. It is supposed that the differences between bi–ethnic and mono–ethnic simultaneous childhood bilingualism may characterize the children’s socio– and psycho–linguistic competence in their non–native language. However, most essential features of their bilingual development, as described in numerous case studies, do not differ from those in bi–ethnic childhood bilingualism. Infants are not aware of which of the two languages in their input is native or non–native before they are aware of their bilingualism (Chirsheva 2000: 107–110). Specific features of mono–ethnic childhood bilingualism may be found in self–interpreting by children at different stages of verbal development, but this phenomenon has not been studied yet. However, some researchers indicate similar patterns in the speech of children exposed to two native languages within bi–ethnic families (Dodson 1984: 3; Malakoff – Hakuta [1991] 2000: 146; Ronjat 1913; Vila 1984). Some examples from the speech of a Hungarian/English bilingual boy at the age of two years and eight months are given by Zs. Gergely in the Appendix to her paper (Gergely 1997: 151–152). The author refers to them as lexical transference without studying how they differ from other cases of code–switching and transference/interference. Like examples of self–correction, self–interpreted utterances are discussed in Saunders 1988 to show the development of metalinguistic awareness in the bilingual development of his children. C. Dodson recommends that parents should stimulate dubbing of words produced in one language with their equivalents in the other (Dodson 1984: 3). At early stages of the child’s bilingual development parents can themselves supply a word pronounced by the child in one language with its equivalent in the other language. The process and the outcomes of self–interpreting are similar to the contexts that A. De Houwer refers to as “examples of translation equivalents used side by side in a single utterance, as if to show translation” (De Houwer 2009:235). 3. Data and participants Several examples are extracted from two works on simultaneous childhood bilingualism, bi–ethnic (De Houwer 2009; Gergely 1997; Ronjat 1913; Vila 1984) and mono–ethnic (Saunders 1988) ones. However, the main data are taken from tape–recorded and video–recorded speech of three boys, as well as from the diaries of their mothers who fixed the utterances their sons said from the age of 12 months up to 36 months. Some examples to illustrate further instances of self–interpreting are extracted from tape–recordings and video–recordings at later periods of the children’s bilingual development. These boys (Peter, Andy, and Serge) are from three different families living in two Russian cities (North of Russia). Both of each boy’s parents are 176 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) Russian native speakers, but the decision of the families was that fathers would speak only English to their children according to the principle “one parent – one language”. Thus, these children’s bilingualism is mono–ethnic. In one family father began speaking English to his son (Serge) from his birth, in two other families children began to be exposed to English when they were six months old. It means that Serge acquired two languages in BFLA context, while two other boys’ language learning contexts can be characterized as mixed in–between BFLA and MFLA. One of the problems in mono–ethnic families bringing their children bilingually is the quality of L2 input. The fathers, non–native speakers of English, who spoke English with their sons, have a near–native competence in this language: Peter’s father and Andy’s father teach English at a University; Serge’s father works at a joint Russian–American company. All of them have had to speak, read and write in English regularly at work and have had a lot of opportunities to communicate with native English speakers in formal and informal contexts, both in Russia and in other countries. Those American colleagues who visited these families when the boys were 24 to 30 months old were surprised to hear that such young Russian children could speak English very much like American children of their age though with a specific Russian accent. It is quite natural that if parents intentionally choose bilingual development for their child both of them pay very much attention to his/her verbal behaviour; each parent spends a lot of time in various kinds of verbal and non–verbal activities together with the child. Therefore, bilingual children in such families are exposed to double portion of books, games, tales, etc., since their parents want them speak both languages well and are ready to do everything to achieve this aim. That is why mono–ethnic bilingualism is sometimes referred to as ’intentional’ (Stefanik 1997; 5, 7). 4. Development of self–interpreting The stages of bilingual development here are differentiated very roughly and can be observed at different age. They depend on the following processes and are related to them: separation of vocabularies, naming explosion, metalinguistic bilingual awareness, relatively balanced period of bilingual development, language attrition, language dominance. First self–interpreted utterances occur when a child begins differentiating between two vocabularies and has accumulated some interlinguistic equivalents in his bilingual vocabulary. The first of them are similar to the dubbing of separate words, for instance, see (1). (1) Ma{inka. Car! Car! (A., 1;111). ’Car. Car. Car.’ 1 Examples of a child’s speech are followed by an initial of his name and the age presented by the number of years and months, e.g. 2;1 means that the child is two years and one month old. 177 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) Similar examples from the speech of bilingual bi–ethnic children are given in Ronjat (1913: 8), see (2) and (3), Vila (1984: 42), see (4), Gergely (1997: 151–152), see (5) and (6). (2) Schiff. Bateau (L.,1;6) ’Ship. Boat.’ (3) Oeil. Auge (L.,1;6) ’Eye. Eye.’ The boy used German and French renditions of ’boat’ (2) and French and German renditions of ’eye’ (3) within the same utterance (4) Pato. Ànec (M., 2;0) ’Duck. Duck.’ The girl used Spanish and Catalan renditions of ’duck’ speaking to her parents. (5) Ott. Up. (J., 2;8) ’There. There.’ (6) Elment. All gone. (J., 2;8) ’Gone away. Gone away.’ The boy first addressed Hungarian words Ott (5) and Elment (6) to his mother and then used the English equivalents of them Up (5) and All gone (6) speaking to his father. The author writes that these are “examples of a child’s natural wish to make himself understood when, to be on the safe side, he voices his request or observation in both languages” (Gergely 1997: 150). As is seen above, such dubbing of separate words or phrases can be observed both before and after the age of two in verbal activities of not only monoethnic, but also bi–ethnic bilingual children. A child dubs his/her own words especially actively at the period when his/ her bilingual vocabulary is rapidly increasing. This stage is usually referred to as the period of naming explosion observed at the age of 1;6–2;0 (see: Markman 1994). At the period of naming explosion equivalents in two languages appear in a bilingual child’s output one after another, with a very short interval. The child uses words from both languages since these lexemes have been present in his input for quite a time and he is eager to produce them speaking to both parents almost at once. For instance, the child has seen nuts in his father’s hand and asks for them, see (7a). (7) a. Nuts. Give me. Give nuts! b. Ore{k–i papa. (A., 2;0). Nut–PL daddy. ’These are nuts. Give me nuts. Daddy has given me nuts.’ The first part of his speech, see (7a), the boy addressed to his father, while the second one, see (7b), he said to his mother after his father gave him the 178 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) nuts. In the English and Russian parts, the boy used semantically equivalent nouns though the utterances themselves cannot be considered equivalent since the English one is a request and the Russian one is a statement. Thus, the child has dubbed not the utterances but only the most important parts of them (from his point of view). At later stages, the child can dub whole utterance/s. Example (8a) presents the speech Andy pronounced addressing his father in English. (8) a. Broken car. Wheels are coming off. ’The car is broken. The wheels have come off.’ b. Sloman–a ma{inka. Koljosik–i upal–i. Netu. (A., 2;1). Broken–SG–FEM car. Wheel–PL fell–PL–PAST down. None. ’The car is broken. The wheels fell down. There are none.’ The boy addressed the second part, see (8b), in Russian to his mother turning to her and trying to express the same information. He had acquired enough translation equivalents by the age of 2;1 though some of them functioned as equivalent only for him. At the next stage of bilingual development when children are aware of their bilingualism and of its advantages (from three till five or six years old) intentional self–interpreting is observed as well as self–interpreting stimulated by adults. (9) Ja vsjo sdela–l. I have done everything. (S., 5;0) I everything do–SG–PAST–MASC I have done everything. ’I have done everything. I have done everything’. Most self–interpreted utterances at this stage like the example above, see (9), can be characterized as equivalent and well–formed. At the period when a child’s bilinguality was decreasing, i.e. he experienced L2 attrition, self–interpreting demonstrated that his L2 competence had not been lost completely. The direction of self–interpreting was based on dubbing L1 words with L2 ones. For instance, Peter began saying an English utterance, code–switched to a Russian verb but dubbed it right away by an English one because he was aware that in that situation it was more “useful” to speak English, see (10). (10) But, dad, why always spat’, sleep? (P., 10;9). But dad why always sleep sleep? ’Dad, why should I always sleep?’ Self–interpreting at the stage of L2 attrition rarely covers whole utterances, which can be explained by two main reasons: 1) the child is aware that there is no necessity in doing that in most situations, 2) his L2 competence is not sufficient to do that. 179 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) 5. Vocabulary and semantics of self–interpreted utterances The first self–interpreted units are similar to the use of baby–talk words together with “adult” words by monolingual children. Bilingual children also use baby–talk words after “adult” equivalent words from two languages to denote the same object. They combine these three equivalents in different ways; for instance, Russian baby–talk words can be used alongside with English “adult” equivalent words. The first word in the combination is an English word and the second one is its Russian baby–talk equivalent, see (11). (11) Car bibi tam. (A., 1;9). Car bip–bip (onomatopoeic for “car”) there. ’Car is there.’ For the child these elements functioned as translation equivalents for some period; later the Russian baby–talk word was replaced with the Russian “adult” word. The most frequent combinations with baby–talk equivalents in the speech of the three boys began with L2 “adult” words followed by L1 baby talk. English input did not contain English baby talk because Russian fathers did not use it in the child–directed speech. This fact can be explained by two main reasons: 1) men usually do not tend to use baby talk thinking it is bad for a child’s speech development; 2) Russian fathers do not know English baby talk, and if they do, they are not aware of its proper use since they have not heard it themselves. Children were not observed to use combinations of three or four equivalents (adult L1 word + baby–talk L1 word + adult L2 word + baby–talk L2 word, adult L1 word + adult L2 word + baby–talk L2 word, etc.) in their self–interpreted speech. In the combinations of L2 “adult” word plus L1 baby–talk equivalent the second part could serve as a kind of predication, for instance, see (12). (12) Dirty kaka! (A., 1;9). Dirty dirty ’Dirty is dirty.’ Here the Russian element (kaka) denotes not only the quality of the object, but also prohibition to touch it (dirty means that it must not be touched or taken). Thus, in such first two–word utterances switching to the Russian baby–talk equivalent means more than just dubbing the information contained in L2 element since L1 baby–talk equivalent plays the main semantic and pragmatic role in them. Sometimes, the words from two languages are not equivalents per se but one is hyponymic to the other, for instance, the Russian word is not the equivalent but denotes a kind of bird (parrot), see (13). (13) Bird there! Popugaj! Bird there… (A., 2;5). Bird there parrot bird there. ’There is a bird. It is a parrot. The bird is there.’ 180 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) Such examples as (13) can serve as the indicator of the child’s bilingual vocabulary development. When the child can use Russian verbs to express predication he sometimes uses Russian baby–talk equivalents in his English sentences that dub Russian ones. The first part of the following example, see (14a), is Russian and the second one, see (14b), is English with a Russian onomatopoeic verb denoting the action ’to fall down producing much noise’. (14) a. Soldatik–i zalez–l–i vysoko ma{ink–u. Ba–bakh! Soldier–PL got–PAST–PL high car–SG–ACC (onomatopoeic for fall down) ’Soldiers got on top of the car and fell down.’ b. Man car there! Man ba–bakh! (A., 2;3). ’The men are there. The men fell down.’ The child did not know English equivalents for Russian verbs, so he dropped down the first one (get) and repeated the Russian baby–talk verb ba–bakh in the English interpretation of his Russian utterance, see (14b). The semantic structure of English utterances usually contains fewer components than that of Russian ones. Showing a picture to his father, the child just named the object in it, see (15a), and then repeated it adding the word again while showing the second picture with the similar object, see (15b). Showing the pictures to his mother the boy was more eloquent and gave more information about the objects, see (15c). (15) a. Piggy. b. Again piggy. c. Khrju{–i taki–e. Odin, dva khrju{–i. Glazk–i {jorn–ye. (A., 2;5). Piggy–PL such–PL One two piggy–PL eye–PL black–PL ’This is a pig. This is another pig. These are pigs. One pig, two pigs. Their eyes are black.’ When the child dubs his Russian utterance, he denotes less information in his English part. For instance, while watching a cartoon the boy spoke on his observation addressing to his mother, see (16a); after that his father could hear only part of what the boy had said to his mother, see (16b): (16) a. Zaj{ik malen’k–ij v {apo~k–e. Hare little–SG–MASC in cap–SG–FEM–LOC b. Little little hare (P., 2;3). ’This is a little hare in a cap. This is a little hare.’ The boy did not express information about the cap on the hare’s head in the English interpretation of his Russian utterance, see (16b). In the following example, the information about the child’s actions is given in the Russian part, see (17a), but it is deleted in the English interpretation of it, see (17b). This fact can be explained by less developed English vocabulary of the child. 181 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) (17) a. Rybka bol’{–aja Dju{–e ku{at’. Gorja~–aja rybka. Fish big–SG–FEM Dju{–SG–DAT to eat. Hot–SG–FEM fish. Dut’. Dju{a podu–et. To blow. Andy blow–SG–3PERS. ’The big fish is for Andy to eat it. The fish is hot. I will blow to make it cool. Andy will blow and cool it.’ b. Big, big... Big fish hot! (A., 2;2). ’The fish is big and hot.’ In Russian, see (17a), the boy spoke not only about the size of the fish and its being hot, but also said that the fish was given to him to eat and as it was hot he would cool it by blowing on it. English parts of self–interpreted utterances usually contained less expressive means than Russian ones. For instance, when Andy showed how big he was, he commented on this fact addressing his father using only one English adjective, see (18a). (18) a. Andy big! ’Andy is big.’ When he repeated the same observation to his mother in Russian, he was able to express his emotional attitude to this fact by repeating the Russian adjective and adding intensifiers, see (18b): b. Dju{a bol’{–oj, o~en’ o~en’ Andy big–SG–MASC very very ’Andy is big, very, very big!’ bol’{–oj! (A., 2;3). big–SG–MASC As is seen from examples (16), (17), and (18), the boy’s Russian parts of the utterances (16a), (17a), and (18b) are more expressive and emotional than their English ones (16b), (17b), and (18a). 6. Grammar of self–interpreted sentences Sometimes grammar errors in L2 are also dubbed in the process of self– interpreting. Example (19a) is Andy’s observation he told his father in English while showing to the window. The sentence is not grammatically well–formed. (19) a. Little dark. ’It is dark a little.’ Turning to his mother, the boy interpreted his English utterance literally and produced the Russian utterance with the same word order and inappropriate morphology. See (19b) that represents self–interpretation of (19a). b. Malo temno. (A., 2;11). Little dark ’It is dark a little.’ 182 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) If the original utterance in L1 is produced without errors, the dubbed phrase in L2 can be either well–formed as well or contain the errors that demonstrate that some grammar forms have not yet been properly acquired in the weak language of the child. The following example is Andy’s Russian utterance that he produced when he broke a toy car while trying to insert a toy soldier into it; he explains to his mother what has happened in the following way, see (20a). (20) a. Djadja ma{ink–u sloma–l. Man car–SG–ACC break–PAST–MASC ’The man has broken the car.’ Then he interpreted his utterance into English for his father, see (20b). b. Man broken car! (A., 2;4). ’The man has broken the car.’ The Russian sentence, see (20a), is grammatically correct while the English one, see (20b), demonstrates the deletion of articles and an auxiliary verb has. Incomplete acquisition of L2 grammar forms is even more evident when L2 sentence is dubbed with grammatically correct L1 sentence. For illustration see (21a) and (21b). (21) a. Nuts want. ’I want nuts.’ b. Dju{a ore{k–i ku{at’ bud–et. (A., 2;5). Andy nut–PL eat will–SG–3PERS ’Andy will eat nuts.’ Andy said the first sentence, see (21a), to his father and the second one, see (21b), was addressed to his mother after his father had gone out to buy nuts. In the English part the word order is inverted in the way that is not typical for the normative English syntax, but it is normal for Russian; the deletion of the subject is also appropriate for the Russian sentence, but not for the English one. In the English parts of the dubbed combinations topics are not often verbalized, while in the Russian sentence both topic and focus are explicitly expressed. For illustration see (22) beginning with an English sentence, see (22a), and followed with its Russian interpretation, see (22b). (22) a. Truck no! Train! ’It is not a truck. It is a train.’ b. Poezd, poezd Dju{a sdela–l! (A., 2;3). Train train Andy do–PAST–MASC ’It is a train that Andy has done.’ The topic in the self–interpreted sentence can be verbalized in the language of the source part, while the focus is expressed in the target language, as 183 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) is seen in (23) taken from a paper by Gergely (1997: 152). The first part of this utterance is Hungarian, in the second one the topic is Hungarian, but the focus is expressed in English. (23) Ez a tied, Mummy. Ez a yours. (J., 2;8) ’This is yours, mummy. This is yours.’ When L1 (Russian) becomes more and more dominant in the child’s bilingual competence and Russian grammar is being acquired according to monolingual norms, L2 doublets demonstrate more clearly that L2 grammar is not being developed according to the child’s communicative age. At the period when L1 is clearly a dominant language in the child’s bilingual competence mistakes in the English (L2) sentences (word order, deletion of structurally necessary words) are not dubbed in the Russian ones while the child is self– interpreting. At the age of 2;6 Andy’s English sentences were often grammatically deficient, while Russian ones had fewer grammatical deviations. Yet, such self–interpreting also shows the ability of a bilingual child to express the same proposition in two languages. The source part of the next utterance is English, see (24a), and the second one is its Russian interpretation, see (24b). Grammar violations are tracked only in the English sentence (the deletion of articles and auxiliary verbs), while the Russian one is well–formed according to the rules of Russian grammar. (24) a. There! Helicopter! Helicopter flying! ’A helicopter is flying over there.’ b. Vertoljot let–it tam, vid–i{? Vertoljot. (A., 2;6). Helicopter fly–SG–3PERS there see–SG–2PERS helicopter A helicopter is flying over there, do you see it? It is a helicopter. The next example, see (25a), was fixed in the situation when Andy began his speech addressing his father with English sentences: (25) a. Papa, car go to Galja! To Galja go! Let’s go garage! Galja little little house there. Together go – mama, papa, Dju{a. (A., 2;6) ’Daddy, let’s go to Galja in the car. Let’s go to Galja. Let’s go to the garage. Galja has a little house there. We’ll go together – mummy, daddy and Andy.’ Then the boy goes on to self–interpret what he has said into Russian to his mother, see (25b). b. Mama, k Gal–e poed–em vmeste na ma{ink–e. Mama to Galja–SG–DAT go–PL–1PERS together in car–SG–DAT Ma{inka Oka est’ tam. Davaj odevat’sja, poed–em v gara`. (A., 2;6). Car Oka is there Let’s dress go–PL–1PERS to garage ’Mummy, let’s go together to Galja’s place in the Oka car. The Oka car is there. Let’s dress and go to the garage.’ 184 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) The boy has dubbed the focus of the utterance, which is the most important thing for the child. As for the grammar, it can be clearly seen that in L2, see (25a) the boy has demonstrated the lack of proper grammar means, while in the Russian part, see (25b), the same proposition is expressed with all appropriate grammatical and semantic elements. 7. Self–interpreting and hesitation Self–interpreting is often accompanied by pauses of hesitation that can be caused by different reasons. A new interlocutor often causes a bilingual child’s hesitation in his/her choice of a proper word. The child either changes a “mother language” word for a “father language” word or vice versa. Sometimes the child chooses a word from a “proper” language but pronounce it in an unusual way. For instance, while eating porridge Andy commented to his mother, see (26). (26) Gorja{–ij! Gorja…. Hot! Hot! (A., 1;10) Hot–SG–MASC hot hot hot ’Hot! Hot! Hot! Hot!’ The first word, see (26), is the Russian adjective meaning ’hot’. It was pronounced in a well–formed way. Then the boy began repeating it in Russian while coming up to his father, but did not finish it and left uninflected. He stopped for a while hesitating and then used the English equivalent adjective hot twice. Pauses of hesitation can be caused by too rapid change of languages the child used within one conversation. Besides, when the dubbing of the source sentence is produced in the non–dominant (weak) language of the child pauses of hesitation can occur because the child is less competent in the target language. For instance, Andy was having dinner and commented his observation in Russian, see (27a). (27) a. Rybka bol’{–aja Dju{–e ku{at’. Gorja~–aja rybka. Fish big–SG–FEM Dju{–SG–DAT to eat. Hot–SG–FEM fish. Dut’. Dju{a podu–et. To blow. Dju{a blow–SG–3PERS. ’The big fish is for Andy to eat it. The fish is hot. I will blow to make it cool. Andy will blow and cool it.’ At the moment when he was producing his Russian utterance, see (27a), his father entered the room and the boy tried to interpret in English what he had said in Russian, see (27b and 27c). b. Big, big... Hesitation followed as he tried to recollect proper English words and said only the following: 185 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) c. Big fish hot! (A., 2;2). Comparing with what he was able to express in Russian, the boy could dub only a small part of it in English. Specific features of self–interpreting and hesitation depend on which language was used for producing the source sentence. If it was said in the dominant (strong) language of the child (the direction L1 > L2), pauses of hesitation before dubbing are much longer and the sentence in the target language is less expressive then in the first one. If the dubbed utterance is the one in the dominant language (L2 > L1) pauses of hesitation occur very rarely and the second utterance is much more emotional and expressive. For instance, Andy was showing his father how big he was and commented on this observation in English, see (28a). The boy pronounced the second part in Russian when he turned to his mother, see (28b). (28) a. Andy big! ’Andy is big.’ b. Dju{a bol’{–oj, o~en’ o~en’ bol’{–oj! (A., 2;3). Andy big–SG–MASC very very big–SG–MASC ’Andy is big, very, very big!’ In his Russian sentence, he used intensifiers, repetition and more emotional intonation than in the English one that followed it. The next example also demonstrates dubbing of a source utterance in a non–dominant language, see (26a), with the utterance in the dominant language, see (29b). (29) a. Truck no! Train! Truck no train ’Not a truck, but a train.’ b. Poezd, poezd Dju{a sdela–l! (A., 2;3). Train train Andy do–PAST–MASC ’Andy has made a train!’ The boy did not hesitate to express what he wanted to say when he used first English and then Russian. He felt more confident when he had to switch to Russian and interpret his previous English utterance. 8. Self–interpreting and pragmatic bilingual competence Self–interpreted utterances are observed in bilingual children’s speech in various situations but they are especially numerous if their parents who speak to them in different languages both take part in the same conversation. Sometimes bilingual children interpret themselves in order to be better understood in the situations with new interlocutors. Since the children realize that they are to choose between their two languages, one of the reasons for their self– 186 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) interpreting can also be the attempt to repair the situation and translate their utterance/utterances into the other language, which looks like self–correction. At the age of 1;11 a bilingual child’s pragmatic competence enables him to combine in his dubbed utterances the verbs in those imperative forms that have been most frequent in his input. For instance, Andy addressed his request to his father in Russian but since his father paid no attention to his Russian words he switched to English and dubbed his request. Example below demonstrates first the Russian utterance, see (30a), and then its English interpretation, see (30b). (30) a. Pojd–jom, e{~jo, pojd–jom! Go–PL–1PERS more go–PL–1PERS ’Let’s go there again.’ b. Come, papa, come! (A., 1;11). ’Dad, let’s go.’ The direction of dubbing depends on the language of the interlocutor the child addresses. For instance, the child repeated the English word twice to attract his mother’s attention to a very interesting object, see (31a), but she did not understand him; therefore, he tried its Russian equivalent though pronounced it without the first syllable, and this time his mother understood what he wanted to say. She pronounced the word in the correct form, see (31b), and the child repeated it, very much pleased with himself, see (31c). (31) a. Child: Bus stoi–t! Bus! Toba! Bus stand–SG–3PERS bus bus ’A bus is over there.’ b. Mother: Avtobus? ’Bus.’ c. Child: Atobas. (A., 2;4) ’Bus.’ At periods of relatively balanced bilingualism self–interpreting looks like self–correction made by a bilingual child. In his book George Saunders gives an example of such self–correction by his elder son Frank, see (32). (32) Warum wollen sie das Frau shooten … schießen? (F., 3;11). ’Why do they want to shoot the woman?’ The boy addressed this question to his father, who spoke German to him, so he hurried to dub the English verb with a German one (Saunders 1988: 183). This self–correction concerned only the change of an English verb for a German one, but not the repair of incorrect grammar forms in the sentence. Zs. Gergely gives some examples of her son’s self–corrections at the age of 2;8, see (33) and (34), though she refers to them only as lexical transference (Gergely 1997: 151–152). (33) Erre megyünk. This way, Daddy. (J., 2;8) ’This way.’ 187 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) The boy addressed his Hungarian utterance to his father but realized that he had chosen the wrong language as his father spoke English to him; therefore, he repeated the same sentence but did that in English, see (33). Example 34 demonstrates the boy’s self–correction stimulated by his mother’s question. The first question included the English word frog, which made the boy’s mother ask her own question to clarify what her son wanted to know. In his reply, the boy gave the Hungarian equivalent of the word frog. (34) Hol a frog? (Mummy: Where is what?) A béka! (J., 2;8) ’Where’s the frog?’ ’The frog!’ The older a bilingual child becomes, the more clearly he realizes that the use of appropriate language produces the necessary pragmatic effect, especially in his requests. Therefore, if he uses Russian addressing the parent to whom he is to speak English he does not hesitate to dub it in English. The child knows that it is especially important when he is expected to use his weak language. The source language of the first sentence, see (35a), is Russian, but the child added its English interpretation without any delay, see (35b), because he wanted his father to fulfill his request. (35) a. Papa, nu`no kupit’ {okoladk–u. Dad necessary buy chocolate–SG–ACC ’Dad, you should buy a chocolate bar.’ b. Dad, buy a chocolate bar, please. (P., 10;4). When expressing emotions, the child dubs his L2 (English) utterances with L1 (Russian) ones, even if Russian is not expected. For instance, in his reply to his father’s request to go for a walk Peter gave his exclamation in English, see (36a). Then he dubbed the English sentences in Russian, see (36b), though it was not necessary because both his father and he were to speak English and both realized that. (36) a. I’ll go! I’ll go! b. Ja pojd–u! (P., 10;11). I go–SG–1PERS ’I’ll go.’ Thus, self–interpreting demonstrates that bilingual children know how to do things with interpreting/translation as this phenomenon is part of their pragmatic bilingual competence. Its forms, direction (L1>L2 or L2>L1), and verbal representation reflect the changes that children’s bilinguality undergoes. 9. Conclusion Self–interpreted utterances are quite natural for young children whose input contains two languages and when they have to speak with the people ad188 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) dressing them in different codes. The child wishes to express himself in both languages in order to be understood by everybody. Adults learning a second or a foreign language sometimes also do that but there are fewer situations when such self–interpreting is justified – only if they are not sure that their interlocutors have understood them properly. It is evident, that some bilingual children use self–interpreting more frequently than others, even in similar language learning contexts. For instance, of the three mono–ethnic bilingual boys who acquired Russian and English languages from the first year of their life Andy appeared to resort to combining translation equivalents most often. His non–balanced bilingualism or his personal characteristics can serve as some possible explanations of this fact. There may also be some interrelation between the frequency of code–switching and self–interpreting: Serge’s speech had very few examples of both, especially before he was five years old. However, Peter code–switched very often but rarely used self–interpreting at the earliest stages of his bilingual development. Further study of this problem, comparison of different bilingual cases and more data could help explain why self–interpreting is produced by some bilingual children more often than by others. Self–interpreting can be considered a kind of ability to combine both code– switching and natural (spontaneous) translation and as a preparatory stage for developing intentional translation. Self–interpreting can demonstrate many characteristics of linguistic competence of a bilingual child’s development: the difference in the acquisition of two grammars, expressiveness of his speech in each language, presence or absence of proper equivalents in the bilingual vocabulary, and even specific nature of his input in each language. Thus, to trace the development of childhood bilingualism, it is very important to study self–interpreting and sometimes even stimulate it. Parents can stimulate the process of their child’s self–interpreting once s/ he has begun combining translation equivalents from two languages. Adults can do that by repeating in the other language the words and sentences pronounced by the child, especially if s/he has chosen the “wrong” language in his/ her communication. This is the way parents can help their bilingual children develop their bilingual competence and fluent switching from one language to the other. The present paper is descriptive since the main objective of it is to demonstrate the significance of the phenomenon under study and its influence on the child’s bilingual abilities. The next steps, among others, are (a) to study structural characteristics of self–interpreted utterances, (b) to describe the interrelation between the acquisition of certain grammatical categories in each language and specific characteristics of self–interpreted sentences, (c) to reveal common and specific features of code–switching, self–interpreting and intentional translation, (d) to provide quantitative analysis of different aspects of self–interpreted utterances. These objectives could be achieved when more data from the speech of more simultaneous bilingual children in different language learning contexts are collected. 189 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) References Aidman, Marina 1994 Comparative analysis of written genre development by a simultaneous bilingual: A longitudinal case study of emerging biliteracy. [Unpublished MA thesis, University of Melbourne.] 1998 “Osvoenie registrov pismennoj re~i u simultanno bilingvalnogo rebjonka” [Acquisition of written genres by a simultaneous bilingual child], in: Raissa L. Smulakovskaja (ed.). Problemy detskoj re~i – 1998. [Problems of child speech – 1998]. StPetersburg – Cherepovets: Cherepovets State University Publishing House, 97–99. 1999 Biliteracy development through early and mid–primary years: A longitudinal case study of bilingual writing. [Unpublished Ph.D. dissertation, University of Melbourne]. Brennan, Anne 1987 “Teaching a trilingual child to read”, Bilingual Family Newsletter, 4/3. Chirsheva, Galina 1996 “Osobennosti formirovanija leksikona rebjonka–bilingva” [Characteristics of a bilingual child’s vocabulary development], in: Stella N. Tsejtlin (ed.). Problemy detskoj re~i – 1996. [Problems of child speech – 1996]. StPetersburg: StPetersburg State University Publishing House, 94–96. 1997 “Rol’ semanti~eskih ekvivalentov v differenciacii leksi~eskih sistem detej–bilingvov” [The role of semantic equivalents in the differentiation of vocabulary by bilingual children], in: Stella N. Tsejtlin (ed.). Problemy detskoj re~i – 1997. [Problems of child speech – 1997]. StPetersburg: StPetersburg State University Publishing House, 72–74. 1998a “O differenciacii leksi~eskih sistem det’mi–bilingvami” [On differentiation of lexical systems by bilingual children], in: Vadim B. Kasevi~ (ed.). Materialy XXVII me`vuzovskoj konferencii prepodavatelej i aspirantov. Vyp. 13. Sekcija ob{~ego jazykoznanija. [Proceedings of XXVIIth conference of professors and post–graduate students. Vol. 13. Session of general linguistics]. StPetersburg: StPetersburg State University Publishing House, 89–94. 1998b “Perevod~eskie sposobnosti detej–bilingvov” [Interpreting abilities of bilingual children], in: Raissa L. Smulakovskaja (ed.). Problemy detskoj re~i – 1998. [Problems of child speech – 1998]. StPetersburg – Cherepovets: Cherepovets State University Publishing House, 35–37. 1999 “Funkcii kodovyh pereklju~enij v re~i detej–bilingvov” [Functions of code–switching in bilingual children’s speech], in: Vadim B. Kasevi~ (ed.), Materialy XXVIII me`vuzovskoj konferencii prepodavatelej i aspirantov. Vyp. 16, ~. 2. Sekcija ob{~ego jazykoznanija. StPetersburg: SPbSU. [Proceedings of XXVIIIth conference of professors and post–graduate students. Vol. 16, part 2. Session of general linguistics]. StPetersburg: StPetersburg State University Publishing House, 71– 74. 2000 Vvedenije v ontobilingvologiju. [Introduction to childhood bilingualism]. Cherepovets: Cherepovets State University Publishing House. Corsetti, Renato 1996 “A mother tongue spoken mainly by fathers”, Language Problems and Language Planning, 20/3: 263–273. De Houwer, Annik 1990 The Acquisition of Two Languages from Birth: A Case Study. Cambridge: Cambridge University Press. 1995 “Bilingual language acquisition”, In: P. Fletcher – B. MacWhinney (eds.). Handbook of Child Language. London: Blackwell, 219–250. 1998 “By way of introduction: Methods in studies of bilingual first language acquisition”, International Journal of Bilingualism, 2 (3): 249–269. 2006 “Bilingual development in the early years”, In: K. Brown (ed.). Encyclopedia of Languages and Linguistics (2nd edition). Amsterdam: Elsevier, 781–787. 2007 “Parental language input patterns and children’s bilingual use”, Applied Psycholinguistics, 28(3): 411–424. 190 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) 2009 Bilingual First Language Acquisition. Bristol; Buffalo; Toronto: Multilingual Matters. Dimitrijeviç, N. R. 1965 “A bilingual child”, English Language Teaching, Vol. 20: 23–28. Dodson, Carl 1984 “Living with two languages”, Bilingual Family Newsletter, 1/1: 2–3. Döpke, Susanne 1992 One Parent, One Language: An Interactional Approach. Amsterdam / Philadelphia: John Benjamins. Facey, Andrea 1986 “Bilingualism with a difference”, Bilingual Family Newsletter, 3/3: 5–6. Gergely, Zsuzsanna 1997 “Code–mixing in the speech of a 2;8 year–old Hungarian–English bilingual child”, In: Zsolt Lengyel – Judith Navracsics – Orsolya Simon (eds.). Applied Linguistic Studies in Central Europe, Vol. 1. Veszprem: Faculty of Foreign Languages, University of Veszprem, 149–154. Harding, Edith – Riley, Philip 1997 The Bilingual Family: A Handbook for Parents. Cambridge: Cambridge University Press. Korovushkin, Valerij – Chirsheva, Galina 1992 “Vzaimodejstvie anglijskogo i russkogo jazykov v re~i rebjonka–bilingva v vozraste ot odnogo goda do dvuh let: iz opyta semejnogo obu~enija” [Interrelation of English and Russian in the speech of a bilingual child from the age of one year old till two years old: a family immersion case], In: Leonid M. Mikhajlov (ed.). Problemy rannego obu~enija inostrannym jazykam. [Problems of teaching foreign languages at an early stage]. Cherepovets: Cherepovets State University Publishing House, 19–21. Lyon, Jean Becoming Bilingual: Language Acquisition in a Bilingual Community. 1996 Clevedon; Philadelphia; Toronto; Adelaide; Johannesburg: Multilingual Matters. Malakoff, Marguerite – Hakuta, Kenji 1991 “Translation skill and metalinguistic awareness in bilinguals”, In: Ellen Bialystok (ed.). Language processing in bilingual children. Cambridge: Cambridge University Press, 141–166. 2000 [Reprinted Cambridge: Cambridge University Press.] Markman, E.M. 1994 “Constraints on word meaning in early language acquisition”, in: Lila Gleitman – B. Landau (eds.). The acquisition of the lexicon. Cambridge, Massachusetts: MIT Press, 199–227. Meisel, Jürgen 1989 “Early differentiation of languages in bilingual children”, In: K. Hyltenstam – L. Obler (eds.). Bilingualism across the Lifespan. Aspects of Acquisition, Maturity and Loss. Cambridge: Cambridge University Press. 2001 “The simultaneous acquisition of two first languages. Early differentiation and subsequent development of grammars”, In: J. Cenoz – F. Genesee (eds.). Trends in Bilingual Acquisition. Amsterdam/Philadelphia: John Benjamins, 11–41. 2008 “Child second language acquisition or successive first language acquisition?” In: B. Haznedar – E. Gavruseva (eds.). Current Trends in Child Second Language Acquisition: A generative Perspective. Amsterdam/Philadelphia: John Benjamins, 55–80. Nataljin, V.P. – Nataljina, S.A. 1989 “Isomorfnoje i allomorfnoje v rannem detskom dvujazy~ii” [Isomorphic and allomorphic features in early childhood bilingualism], in: De{eriev, Jurij D. (ed.). Bilingualism and diglossia, Moscow: Moscow State University Publishing House, 34–35. Past, Al – Past, Kay 1978 “Early childhood: The best time to become bilingual and biliterate”, Childhood Education, 54/3: 155–161. Past, Kay 191 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) 1976 “A case of preschool reading and speaking acquisition in two languages”, in: Robert Lado – Theodore Andersson (eds.). Georgetown University Papers on Languages and Linguistics. Washington, DC: Georgetown University Press, 58–73. Ronjat, J. 1913 Le développement du langage observé chez un enfant bilingue. Paris: Champion. Saunders, George 1982 Bilingual children: Guidance for the family. Clevedon: Multilingual Matters. 1988 Bilingual children: From birth to teens. Clevedon: Multilingual Matters. Stefanik, Josef 1995 “Grammatical category of gender in Slovak–English and English–Slovak bilinguals”, Journal of East European Studies, 4: 155–164. 1997 “Intentional bilingualism: Slovak–English bilingual family in Slovakia”, The Bilingual Family Newsletter, 14/1: 5, 7. Stephens, K. 1952 “Reader’s letter”, The Linguist, Nov.: 307. Swain, Merill 1976 “Bilingual first language acquisition”, In: W. von Raffler–Engel – Y. Lebrun (eds.). Baby Talk and Infant Speech. Amsterdam: Swets and Zeitlinger, 277–280. Totmjanina, Elena 1998 “Na~al’nyj etap intencionnogo bilingvizma” [Early stage of intentional bilingualism], in: Raissa L. Smulakovskaja (ed.). Problemy detskoj re~i – 1998. [Problems of child speech – 1998]. StPetersburg – Cherepovets: Cherepovets State University Publishing House, 84–85. 1999 “MFE kak priznak differenciacii jazykov” [Interlinguistic functional equivalents as the signal of languages separation], Stella N. Tsejtlin – Maria B. Elisseeva (eds.). Problemy detskoj re~i – 1999. [Problems of child speech – 1999]. StPetersburg: StPetersburg State University Publishing House, 175–177. 2000 “Differenciacija jazykov na rannem etape intencionnogo bilingvisma” [Differentiation of languages at an early stage of intentional bilingualism], in: Alexander V. Chernov (ed.). Criterion, Vol. IV. Cherepovets: Cherepovets State University Publishing House, 76–78. Vila, I. 1984 “Yo siempre hablo catalan y castellano: datos de una investigación en curso sobre la adquisición del lenguaje en niños bilingües”, In: M. Siguan (ed.), Adquisición precoz de una segunda lengua. Barcelona: Publicacions i edicions de la Universitat de Barcelona. Samotuma~enje kod dvojezi~ne djece Samotuma~enje je karakteristi~no u razvoju ve}ine dvojezi~ne djece, ali dosad mu se nije pridavala dovoljna pozornost. ^lanak opisuje najva`nija obilje`ja ovog fenomena u razvoju dvojezi~nosti u ranom djetinjstvu. Podaci su prikupljeni u tri ruske obitelji u kojima djeca (n=3) usvajaju dva jezika (ruski i engleski) od prve godine `ivota. Samotuma~eni iskazi prirodni su za mla|u djecu koja usvajaju dva jezika. Naj~e{}e situacije u kojima dvojezi~na djeca koriste samotuma~enje jesu razgovori unutar trijade »majka – dijete – otac«. Budu}i da dvojezi~no dijete mo`e proizvesti iskaz s istom propozicijom u oba jezika, ono to isprve ~ini spontano, a onda i namjerno. Samotuma~enje je svojevrsna sposobnost kombiniranja zamjene kodova i spontanog prevo|enja; slu`i kao pripremna faza za razvoj namjernog tuma~enja. Rabe}i samotuma~enje, dvojezi~na djeca pokazuju osjetljivost prema interlokutorovu jeziku: kada primijete da ih netko nije razumio u jednome od jezika, ka`u istu stvar na drugome jeziku. Takva aktivnost podsje}a na samoispravljanje: djeca shvate da su odabrala »krivi« jezik te se isprave. Roditelji bi to trebali poduprijeti te poticati samotuma~enje. 192 G. Chirsheva, Self–interpreting by bilingual children – SL 70, 173–193 (2010) Samotuma~enje mo`e pokazati mnoga obilje`ja jezi~ne kompetencije kod dvojezi~nog djeteta: prisutnost ili odsutnost pravih ekvivalenata u njegovu dvojezi~nom rje~niku, razliku u usvajanju dvaju leksika i dviju gramatika, izra`ajnost govora u svakom jeziku, razvoj pragmati~ke dvojezi~ne kompetencije i specifi~nu prirodu usvojenih obrazaca u svakom jeziku. Neka dvojezi~na djeca ~e{}e samotuma~e u odnosu na drugu dvojezi~nu djecu, ~ak i u sli~nim kontekstima i situacijama u~enja jezika, {to zahtijeva daljnja istra`ivanja ovog fenomena, usporedbe razli~itih slu~ajeva dvojezi~nosti, vi{e podataka i vi{e ispitanika. Key words: childhood bilingualism, self–interpreting by children, intentional bilingualism, Russian language, Croatian language Klju~ne rije~i: dvojezi~nost djece, samopoimanje djece, intencionalni bilingvizam, ruski jezik, engleski jezik 193