Early code-switches in young Russian bilingual siblings

Cover Page

Cite item


The number of bilinguals and multilinguals of all age groups, including the youngest ones, constantly grows in many countries of the world, which explains the topicality of the study. While researchers explore various aspects of bilingual development, linguists concentrate their efforts on the analysis of bilingual speech, including code-switches. The author deals with Russian-English code-switches in the utterances of two siblings at the earliest stages of their bilingual development - before they were 36 months old. The children had acquired two languages since their first month in a monoethnic Russian family, their first language being Russian and their second (non-native) language being English. The aim of the study is to reveal specific structural, semantic, and pragmatic characteristics of mixed utterances observed in early code-switches of the siblings. The author argues that children’s code-switches at the earliest stages within one family reflect specific features of the communicative situations where children have to cope with the choice between the two languages spoken by their adult interlocutors. It is shown that the application of the Matrix Language Frame Model to the analysis of early childhood bilingual speech is possible, but has several restrictions and limitations related to incomplete and imperfect acquisition of both grammars. The originality of the research is not only in the use of specific data (the earliest stage of bilingual speech observed in simultaneous Russian-English dual-language development), but also in the employment of the Matrix Language Frame Model to study them. This work can contribute to the research of typical features of emerging code-switches in developmental perspective.

Full Text


The topicality of this paper is determined by the fact that the number of bilinguals and multilinguals of all age groups, including the youngest ones, constantly grows in many countries of the world, including Russia.

From the earliest stages of their development, young bilinguals often find themselves in the so-called bilingual mode (Grosjean, 2001: 5–20) when they activate both of their languages to a different extent, which stimulates them to use mixed utterances. In the families where caregivers interact with children in two languages, such situations are very frequent.

Since the 1990s, researchers have studied early child code-switches in a variety of language combinations: German-French (Meisel, 1994); English-Spanish (Deuchar, Quay, 1998); English-German (Bauer et al., 2002); Norwegian-English (Lanza, 2004); French-English (Paradis et al., 2000; Comeau et al., 2003; Kuzyk et al., 2020); Finnish-Swedish (Rontu, 2007); Estonian-English (Vihman, 2016, 2018); English-Mandarin (Yow et al., 2018); German-English (Lanvers, 2001; Quick, Hartman, 2021); Cantonese-English (Lam, Matthews, 2020); Spanish-English and French-English (Smolak et al., 2020); Spanish-English (Gross et al., 2022); Dutch-English (Sczepurek et al., 2022); German-French and Turkish-Italian (Schächinger Tenés et al., 2023).

Researchers from many countries, who study early child code-switches, have already dealt with various problems, including the following ones: unequal acquisition of two grammars by children, which causes violations of some constraints of bilingual speech patterns (Paradis et al., 2000); the pragmatic role of code-switching in play contexts (Bauer et al., 2002); children’s code-mixing as their sensitiveness to the interlocutors’ code choices (Comeau et al., 2003); reasons and functions of early child code-switches and their developmental patterns (Ervin-Tripp, Reyes, 2005); code-switching in triadic conversational situations between two bilingual siblings and their mother (Rontu, 2007); code-switching as a marker of linguistic competence in children (Yow et al., 2018); the interrelation of inter-sententional and intra-sententional code-switching with language dominance in bilingual, trilingual, and multilingual children (Poeste et al., 2019); the development of code-switching longitudinally in two distinct groups of children from 31 to 39 months of age (Smolak et al., 2020); inter-sententional code-switching as a sign of language dominance status (Lam, Matthews, 2020); code-switching behavior in children as the sign of their cognitive benefits revealed in executive tasks (Kuzyk et al., 2020); the interplay of two developing grammars in bilingual children and their individual differences (Quick, Hartman, 2021); input-output effects in bilingual language processing (Gaskins et al., 2022); code-switching reflecting high linguistic competence in preschool children (Schächinger Tenés et al., 2023).

Code-switches at the earliest stages in simultaneous childhood bilingual siblings have not been studied to the extent they deserve; therefore, this paper can contribute to understanding how two languages interact in a very close contact.

The aim of the study is to describe structural, semantic, and pragmatic aspects of early code-switches in two siblings who acquired Russian and English simultaneously in Russian monoethnic settings.

The research is based on two interrelated hypotheses. First, we suppose that the more balanced bilingualism children develop, the more varied their code-switches are both structurally and pragmatically. Second, we can analyze early code-switches within the Matrix Language Frame Model, though in a specific restricted variant.

Materials and methods

This study uses the mixed-method approach – it combines observation and description of everyday informal bilingual interaction within an extended Russian family.

Data collection included video-taping, audio-recording, and written records by parents and grandparents. Every two weeks, adults video-taped or audio-recorded the children’s speech in home settings. The overall time of video-records for five years is approximately 40 hours. Besides, immediately after parents and grandparents observed the siblings’ utterances that contained code-switching, they wrote them down in their diaries. The total amount of mixed utterances observed in the siblings’ speech at the age stages before each of them was 36 months old comprises 450 (240 in Mile and 210 in Alex).

The author later analyzed structural, semantic, and pragmatic characteristics of children’s code-switches in the extracts from video-, audioand written records. The analysis employed the MLF Model and 4-M Model in combination with the developmental study of early bilingual development and the comparison of each type of code-switches in the speech of two siblings.

The participants were two brothers, the elder boy (Mike) being 28 months older than his younger brother (Alex). The boys produced the observed code-switches when they were younger than 36 months old.

Bilingual upbringing in the family developed according to the “one parent – one language” strategy: the mother and her relatives were speaking their native language (Russian) to the children, while the father and his parents were using their non-native language (English) in their interactions with the children. The children’s father had learned Russian and English in the same way: his father spoke English and his mother spoke Russian to him. Russian is the only native language for everybody in the family. The development of Russian-English bilingualism in both families was the free choice of the parents. Being University professors of linguistics and researchers of linguistics and bilingualism, Mike and Alex’s grandparents regularly used English in their professional activities.

The children’s bilingualism developed naturally because they learned both languages in everyday communication, without special instruction, as a native language. However, this kind of bilingualism, which we refer to as monoethnic, did not combine with natural biculturalism because the children had only Russian cultural background and were socializing only in a Russian community. Before they were 36 months old, neither of them had interacted with native speakers of English and had never been abroad. They learned some elements of British or American culture only via films, cartoons, books (in English) and other occasional artefacts brought from other countries (toys, clothes, etc.).

Structural characteristics of code-switches in this paper are analyzed in terms and principles of C. Myers-Scotton’s Matrix Language Frame Model (MLFM), elaborated on in the mid-1990s and ever since developed in her further works, including those co-authored with her colleagues (Myers-Scotton, 1997; Myers-Scotton, Jake, 1995; 2017).

Structurally, code-switches can be inter-sentential or intra-sentential. The latter are further classified as insertions, Embedded Language islands (EL islands), clause-switches, parenthetical switches, and tag-switches. All insertions follow the rules of the Matrix Language. Some of them may appear as bare forms, i.e. without system morphemes required by the Matrix Language morphosyntactic frame, for instance:

Ona vyshla pogulyat so svoim dog (= She went out to walk with her dog).

In order to be well-formed, the word “dog” in this position (prepositional object) within the Russian morphosyntactic frame must have the ending -om (the Instrumental Case, Singular), like the preceding possessive adjective (svoim) that agrees with the noun in this NP. However, the English word (content morpheme) remains a bare form, i.e. lacks proper system morphemes that are required by Russian grammar.

In EL islands, EL grammar is present when one of the fundamental principles (the Morpheme Order Principle or the System Morpheme Principle) or both of them are at work. Thus, an EL island consists either of one EL content morpheme and at least one EL system morpheme (article, plural inflexion, etc.) or several EL content morphemes combined in agreement with the EL order.

We suppose that the Matrix Language Frame Model (MLF model) can predict not only the order of system and content morphemes production in bilingual speech, but also the order in which children learn them in their bilingual speech development.

Early code switches can reveal which of the languages a bilingual child learns earlier and which of them becomes more active as the Matrix Language. Its system morphemes are also more abundant in the child’s bilingual speech at all stages. Yet, at the first stage of child bilingual development it is highly problematic to differentiate between the Matrix language (ML) and the Embedded language (EL) since the earliest utterances are too short (two-morpheme or three-morpheme ones), and young children do not use grammatically relevant system morphemes (Chirsheva, Korovushkin, 2015).

I argue that in early bilingual speech, it is also possible to track how two fundamental principles of the MLF model – the Morpheme Order Principle and the System Morpheme Principle – are used to build the morphosyntactic frame of children’s utterances. The way and order a child activates these principles provides some evidence of his/her bilingual awareness and attitude toward each language.

C. Myers-Scotton’s 4-M Model explains even more details of bilingual speech because it differentiates between four groups of morphemes: content and three groups of system ones – early, late bridge and late outsiders. The authors of the model have proved that early system morphemes are the first in speech production, late bridges act at phrase level and late outsiders become active at the sentence level (Myers-Scotton, 2002; Myers-Scotton, Jake, 1995, 2017).

Researchers of childhood bilingualism argue that this order is also relevant for bilingual child speech development: only some early system morphemes appear in first mixed utterances (Chirsheva, Korovushkin, 2017: 89).

The analysis of stages of simultaneous childhood bilingualism requires the consideration of the following aspects: the development of the children’s vocabulary in both languages, the amount and age characteristics of functional bilingual equivalents in their speech, interference and code-switches, spontaneous and intentional interpreting by children, their metalinguistic awareness and language attitudes. These aspects in their combination can serve as criteria for differentiating stages of children’s bilingualism development.


Before the age of 36 months both Mike and Alex had passed the first stage (Mike’s lasted between 18 and 20 months, Alex’s lasted 8 months longer – between 13 and 23 months), when they used only receptive and reproductive functional inter-linguistic equivalents (FBEs) and started using productive FBEs that stimulated their self-interpreting and spontaneous interpreting in bilingual situations.

At the first stage, both boys used only content morphemes in their bilingual utterances. Some early system morphemes (definite and indefinite articles and plural endings) were only emerging at the second stage, but they were not in systematic use. It means that they were not proper system morphemes since the boys have not perceived the corresponding grammatical categories (definiteness and plurality) by that time. Therefore, grammatically relevant morphemes (grammemes) were sometimes present and sometimes absent in similar structures.

Structural aspects of mixed utterances at the first stage of bilingual development were the following:

1) they were not variable and limited to simple insertions of content morphemes represented by short words (Russian one + English one);

2) only Mike used mixed structures;

3) the Matrix and the Embedded languages were not differentiated;

4) neither principle of the MLF model could be tracked.

The second stage of bilingual development lasted till 36 months of their age in both Mike and Alex. Utterances with code-switches at this period preserved several problems that were characteristic for the first stage: it was highly problematic to distinguish between ML and EL because the children did not acquire either grammar systematically. However, some cases showed clear signs of the emerging differentiation between the two languages.

The dominant language of the children was Russian, which caused more active acquisition of its grammatical properties by both Mike and Alex. Therefore, it was Russian that usually acted as ML, because the children used its system morphemes more correctly. Neither of the siblings utterly refused to speak English; however, both of them, especially Mike, could ignore requests to repeat in English those words or phrases that they had said in Russian to their father or grandparents.

Semantically, early code-switches from Russian to English in both children referred to the objects they used in their everyday life and the ones they associated with English-speaking adults (father, grandfather, and grandmother).

Pragmatics of early code-switches reflected the most important intentions in the children’s interactions: addressee-oriented and message-oriented.


Among the first signs of bilingual speech development are the so called “translation equivalents”, i.e. the words that function in a specific child’s communication as equivalents. They are not always true equivalents because in “adult vocabulary” they may have different semantics. Therefore, they should be more precisely called functional inter-linguistic equivalents (FIE).

First FIEs may appear very early, when a child’s bilingual vocabulary includes only a few words. The way they come to the vocabulary and the intervals between their emergences are quite symptomatic since they help to track both the development of bilingualism and the first structures of code-switches.

The first signals of Mike’s perceiving functional equivalence of lexemes from two languages appeared at 18 months, while Alex showed them when he was only 13 months. These were identical non-verbal reactions to the questions to show a nose, a mouth, eyes, ears, etc. (e.g. Gde tvoj nosik? (Russian) and Where is your nose?) and requests to say goodbye (Skazhi do svidaniya (Russian) or  Say goodbye) their parents and grandparents expressed in Russian or English.

Later (Mike at 18 months and Alex at 17 months), the boys started reacting in such situations verbally, but used only Russian in response to both Russian and English stimuli. For instance, when father asked Mike: Where is your granny? the boy pointed at his grandmother and said in Russian: Baba! (= granny). When father asked him: Say goodbye, Mike waved his hand and pronounced in Russian: Poka!

Alex reacted in the same way when his grandmother warned him at the table that a cup of tea was hot: It’s hot. Then Alex pointed at an oven and pronounced his Russian variant of the word goryacho (= hot): Gy!, which showed that English hot and Russian goryacho were equivalent. Other examples of passive acquisition of equivalency at this age occurred in thanking situations: father gave Alex grapes and asked him: Say: Thank yo to which the boy replied in Russian: Si! (his variant of spaSIbo). Such were his answers when somebody asked him to say thank you in Russian.

Such reactions revealed comprehension, but not production of equivalents. Therefore, we refer to them as to receptive functional bilingual equivalents (receptive FBEs).

The reasons for receptive FBEs in children’s utterances were to fill in lexical gaps in productive vocabularies or overcome difficult pronunciation of some sounds in English or Russian words. Thus, FBEs inserted into utterances in the “wrong” language resulted in the children’s first bilingual structures.

One more reason for mixed utterances that was observed not only at the first stage, but later on as well, was excessive bilingual mode produced by adults who spoke to the children and to each other in different languages. This was especially salient in Alex’s speech at the second stage. For instance, looking at both his mother and father, he pointed to the corridor where he was afraid to go and explained: Dark, (tem)no (his variant of saying dark in Russian (25 months). Mike used such bilingual dubbings only at the end of his second bilingual stage – by 36 months.

Mixed utterances in Mike’s speech appeared much earlier (at 18 months) than those in Alex’s. Among the first of Mike’s bilingual constituents, were those that referred to cars:

(1) Papa car. Baba car. Dedya car.

Such utterances were abundant: the boy pointed at cars every time he saw them. With the help of them, he referred either to the fact that the car belonged to daddy, granny or granddad, like in examples (2), (3), and (4), or that somebody was sitting in the car.

(2) Papa car (= Dad’s car).

(3) Baba car (= Granny’s car).

(4) Dedya car (= Granddad’s car).

A month later (at 19 months) the boy started distinguishing between these semantically differing structures; thus, he added to his locatives specific sounds resembling a preposition and a definite article:

(5) Papa i-i car (= Dad is in the car).

(6) Baba i-i car (= Granny is in the car).

(7) Mama i-i car (= Mom is in the car).

Alex’s first mixed utterance appeared only when he was 25 months. It occurred in the situation when he asked Mike to give him a puck:

(8) Daj puck (= Give me the puck).

On the same day, he spoke about a favorite Russian cartoon which he entitled Makha i bear. This was his invented title for a popular Russian cartoon Masha i medved (= Masha and the bear).

Like all babies of his age, Mike often used Russian onomatopoeic reduplicates, which he sometimes combined with English proper words. For instance, when once he heard loud bumping sounds from a warehouse, he commented:

(9) Big tuk-tuk! (= Loud bump-bump) (30 months).

He used no system morphemes here, which was habitual for his speech at this period.

Occasionally, Mike used system morphemes in his mixed utterances, though in most situations, they looked like the imitation of adults’ speech. For instance, while doing a puzzle, the boy (27 months) commented on the pictures that he got:

(10) Pi-pi in the car (= Squeek-squeek in the car ‒ a mouse is in the car).

(11) Av-av in the car (= Bow-wow in the car ‒ a dog is in the car).

Here the boy used Russian onomatopoeic reduplicates for a mouse (10) and for a dog (11).

Alex did not use any system morphemes at the second stage, which made his mixed utterances less variable than his brother’s at the same age.

Even when Mike knew conventional words, he combined them with semantically equivalent onomatopoeic ones. For instance, once he pointed at a dog and said (31 months):

(12) A big dog! О! А big av-av! (= A big dog! Oh! A big bow-wow!).

These minor sentences, the English only and the mixed English-Russian ones, were grammatically different from those at the first stage, because the boy added indefinite articles to both an English conventional word (dog) and a Russian onomatopoeic word (av-av). It means that English definitely acted as the Matrix Language in the bilingual constituent with the English morphosyntactic frame: the boy used an English early system morpheme (the indefinite article).

Another English early system morpheme that Mike used at the second stage of his bilingual development was plural ending in those English nouns that he learned at the first stage and that were very frequent in his speech. When he was 30 months of age, such instances appeared in his short mixed utterances:

(13) Tam cars (= There are cars).

(14) Eto keys (= These are keys).

As for Alex’s mixed utterances, at the second stage, he never built English morphosyntactic frames because he did not use any English system morphemes consciously – he only reproduced them in the sentences that he repeated after adults.

The most salient components that signaled of the ML frame are verbs. In Mike’s and Alex’s utterances even at the second stage they were predominantly Russian, so at the age between 26 and 29 months, they built Russian morphosyntactic frames with the verbs in the imperative mood (see (15), (16), and (17)):

(15) Baba, daj key (= Granny, give me the key).

(16) Papa, daj key (= Dad, give me the key).

(17) Papa, daj key car! (= Dad, give me the key to the car).

Some of Alex’s similar requests (at 26 months) had no verbs at all:

(18) Papa, key! (= Dad, give me the key).

Both boys omitted verbs, even in their mixed replies to English stimuli in the adults’ speech addressed to them. For instance, Alex (30 months) gave a towel to his granny and commented:

(19) Baba, towel (= Granny, take the towel).

When his grandmother said: It’s cold. You should put on your socks, Alex (26 months) put his socks on and showed his feet to her:

(20) Ya socks (= I have put on my socks).

In the situation when granny was leaving, Alex (26 months) pointed at her shoes and asked whether she was going to put them on:

(21) Baba, boot? (= Granny, will you put on your boots?).

Mike constructed Russian morphosyntactic frames in his utterances denoting the nomination of objects. They included Russian demonstrative pronouns and English nouns and no linking verbs, which reflected Russian syntactic rules. Therefore, the ML in such utterances was Russian (see examples (22), (23), and (24)).

(22) Eto big car (= This is a big car) (27 months).

(23) Eto spoon (= This is a spoon) (30 months).

(24) Eto big bear (= This is a big bear) (30 months).

In his replies to adults, Mike sometimes repeated their English utterances but supplied them with the Russian system morphemes, which resulted in mixed constituents with Russian ML. For instance, when his grandmother showed him a picture of a mouse and said: It’s a mouse, Mike changed it into a mixed sentence (24). Alex changed English utterances similarly (25).

(24) Eto mouse (= This is a mouse) (30 months).

(25) Eto you (= This is you) (36 months).

Following the same pattern, Alex changed other types of speech acts. For instance, when Mike asked their father: And where is mummy?, Alex (28 months) also asked this question, but his interrogative utterance became a mixed one with Russian ML (see (26)).

(26) A gde mummy? (= And where is mummy?).

In all these cases, English components represented the focus (rheme) of the mixed utterance, while morphosyntactic frame that was Russian contained topic (theme) elements.

English words in such reconstructions could be adjectives, like in the following sentences in Mike’s speech (see (27)). When he tried to climb an exercise bike, his grandmother warned him: Be careful, it’s too big for you. Mike repeated a part of the English utterance: It’s big, but then rebuilt it into a Russian morphosyntactic frame with only adjective left from the English one:

(27) Eto big (= This is big) (30 months).

Alex also produced such reconstructions. For instance (see (28)), to his grandmother’s request: Say: I am big, he said (26 months):

(28) Ya big (= I am big).

Such mixed utterances (27 and 28) do not violate the MLF model principles because they represent correct Russian syntactic structures.

Only Mike produced the Embedded Language islands at the second stage of his bilingual development, whereas Alex began constructing them later. Mike’s EL islands (at 27 months) often included onomatopoeic Russian words combined with English phrases (see examples (10) and (11)).

These English EL islands did not violate of Russian grammar rules.

Though rarely, Mike evidently built English morphosyntactic frames when he used English verbs in his mixed utterances. For instance, he peeped into the bedroom and informed his mother (30 months):

(29) Mama, tam Sasha sleep! (= Mummy, Alex is sleeping there).

We suppose that the ML here was English, though the boy violated a principle of the MLF model: having chosen the wrong aspect form, he did not use the system morphemes required by the English frame. Only if the English verb is a bare form in the Russian morphosyntactic frame, there is no violation of the MLF model.

Alex did not use English bare forms in his Russian utterances before he was three years old.

Bare forms do not violate the principles of the MLF model, so they are quite frequent in the speech of adult bilinguals when they choose not to add system morphemes for pragmatic reasons. As for bilingual children, bare forms first appear in their mixed speech when they are too young for proper knowledge of grammars. They seem to remove all grammatically relevant morphemes, which shows that they do not imitate adult utterances where such morphemes are present. Therefore, the usage of bare forms can serve as a signal that a child constructs his/her own bilingual grammar.

Those code-switches that the MLF model views as “classical” ones, since they follow both fundamental principles (the System Morpheme Principle and the Morpheme Order Principle), appeared in Mike’s speech in the middle of the second stage but they were quite few. Alex did not use them at the second stage at all. Such switches follow all the rules of the ML morphosyntax: EL content morphemes are supplied with the ML system morphemes and are arranged in the order proper for the ML. Classical switches in Mike’s mixed speech (30 months) manifested themselves only with Russian as the ML.

(30) No phon-ya (= There is no phone).

In (30), the English content morpheme is supplied with the Russian Genitive-case ending in full accordance with the Russian morphosyntactic frame.

Another signal of proper bilingual sentence construction is the use of English adverbial modifiers within Russian constituents (clauses). They follow the EL Implicational Hierarchy Hypothesis: the EL components that are peripheral for the ML syntactic structure are most frequent in bilingual constituents (Myers-Scotton, 1997: 7). Mike used adverbial modifiers of location expressed by pro-adverbials here and there, which do not interact with the Russian morphosyntactic frame and can locate in any part of the sentence. For instance, when his grandfather asked: Where is your cup?, Mike (31 months) pointed at the cup and replied:

(31) Von there upal! (= It fell down over there).

Alex also used such switches at the second stage (29 months), for instance:

(32) Kto there? (= Who is there?).

The same rule applied to the mixed speech of both children when they used English interrogative words to construct Russian questions. The most frequent was the interrogative pro-adverb where. For instance, Mike (31 months) asked such questions (33) and (34):

(33) Dedya, а where papa? (= Granddad, and where is daddy?).

(34) A Sasha? Where Sasha? (= And Alex? Where is Alex?).

Alex (27 months) asked questions with similar structures (see (35)).

(35) Where mama? (= Where is mom?).

A rule for appropriate bilingual constructions requires that subjects expressed by personal pronouns should combine with a predicate expressed by the verb from the same language, because personal pronouns are system morphemes and EL system morphemes cannot appear in the ML frame by themselves, without EL content morphemes. Children who are not aware of the grammatical status of words in their early vocabulary often violate this rule. Alex (28 months) especially actively used such mixed constructions. For instance, among his variations of the same question And me? were monolingual Russian (A ya?, A mne?) and monolingual English (And me?), as well as mixed ones (see (36) and (37)):

(36) A me? (= And me?).

(37) A ya – some more? A ya – more? (= And some for me? And I also want some more!).

Mike could say in English: I’m here and its equivalent in Russian: Ya zdes; however, he also used mixed structures to express the same idea (33 months):

(38) I zdes (= I am here).

(39) Ya here (= I am here).

The following utterance by Mike (33 months) presents the violation of the same rule:

(40) Ya sleep (= I am sleeping).

Before they were three years old, neither boy invented mixed compounds. Only one example can show this kind of bilingual creativity in Mike (29 months). Once he brought a broken toy car and said to his father in Russian,

(41) Slomal (= I have broken it).

His father asked him in English: The car is broken?, to which Mike replied,

(42) Na-bobo-kan (= It is broken).

In his hybrid word (42) he combined the Russian baby-talk word bo-bo, which children say when they feel some pain, with the English word broken. The bilingual derivative could appear because Mike used to say bo-bo while sympathetically taking care of broken toy cars.

We can view mixed utterances in young bilinguals as dubbing, self-interpreting or partial translation: a child interprets into Russian only structural components of an English sentence while leaving an English content word (morpheme) intact. For instance, Mike did it in the following way (see (43) and (44)):

It is big. → (43) Eto big (= It is big).

It is a dog. → (44) Eto dog (= It is a dog).

Mike evidently translated separate words or word combinations in the situations when he wanted to clear up something and asked echo-questions. For instance, grandmother took keys and said: They are granddad’s, to which Mike (32 months) asked in reply:

(45) Dedini? (= Are they granddad’s?).

Alex used to produce such a “translation/interpreting” in order to agree with his interlocutor. For instance, his father asked him: Say: granny Galya, to which Alex (28 months) replied in Russian:

(46) Baba Galya (= Granny Galya).

When Alex was looking at photos and saw a baby in one of them, he asked his grandmother in Russian: Kto – ya? (= Who is that – me?). She said: No, it’s Mike. Alex nodded and confirmed, using his brother’s Russian variant of the name: Misya (29 months).

At a later age, Mike and Alex spontaneously translated the utterances that adults told them. For instance, father asked Mike: Ask mummy what we are going to have for dinner. Mike asked his mother this question in Russian, then “translated” her answer to his father into English.

Such situations also occurred when father came to take Alex from the kindergarten. Father asked him: Ask your teacher permission to go home. Alex asked his teacher in Russian (31 months):

(47) Mozhno idti domoj? (= May I go home?).

Mike sometimes helped his younger brother by translating some words from English into Russian. For instance, grandmother showed a toy to Alex and said: It’s a bee, and Mike (33 months), who heard that, “explained” or interpreted it into Russian for Alex:

(48) Eto pchyolka (= This is a bee).

One specific type of switch is dubbing, or a self-interpreting switch. Children often find themselves in bilingual communicative situations when they have to tell something to both Russian and English interlocutors. Therefore, they repeat the same ideas in two languages. Alex began doing that at an earlier age than Mike, but he dubbed only separate words. In all cases, English words followed Russian ones. For instance:

(49) Brru, car (26 months);

Sobaka, dog (26 months);

Luna, moon (26 months);

Temno, dark (27 months);

Deda, granddad (28 months);

Da, yes (29 months).


We can analyze bilingual speech at the very early stages of simultaneous bilingual development with the Matrix Language Frame Model, though certain developmental characteristics limit some premises and add specific features.

Code-switches in the two siblings’ speech at the first two stages of bilingual development had the following characteristics:

1) it was difficult to distinguish between the Matrix and the Embedded languages;

2) the children often violated the Principle of System Morphemes because of their imperfect acquisition of both grammars;

3) tracking violations of the Morpheme Order Principle in Russian-English early mixed constituents was a hard task for two reasons: a) the utterances were very short, b) Russian and English word order in statements has much in common;

4) there were more violations of the MLF model when the non-dominant language acted as the Matrix Language;

5) at the first stage of their bilingual development, the children used only EL insertions expressed by short words; at the second stage, EL insertions remained the most frequent structural types of code-switches, but occasional new ones (EL islands and bare forms) also appeared;

6) intra-sentential code-switches were predominantly nouns;

7) the relief strategy can explain the earliest code-switches: children chose words from either vocabulary if they learned them earlier or these words were easier to pronounce;

8) many code-switches in children occurred as their reaction to the bilingual mode of the communicative situations they find themselves in;

9) some mixed utterances looked like a partial translation of the stimuli addressed to the children by adults.

Thus, early code-switches are not chaotic mixed speech at stages of emerging bilingualism. It means that we can analyze young children’s speech when they use units of both languages within one utterance according to the principles that are implemented on adults’ code-switches. It is also important that the analysis of early switches can add to the study of childhood bilingualism, since structural characteristics of two languages in such close and developing contact can predict some ways of their acquisition or attrition.

More case studies (data and participants), as well as statistics, should be involved and researched in order to come to proper assumptions about the stages of bilingual development in the process of simultaneous acquisition of two languages, about two or more bilingual siblings, about the possibility to analyze early mixed utterances with the help of the MLF model. Further research can show how mixed constituents change in the process of bilingual development and of L1 or L2 attrition.


About the authors

Galina N. Chirsheva

Cherepovets State University

Author for correspondence.
Email: chirsheva@mail.ru
ORCID iD: 0000-0001-6627-693X

Doctor of Philology, Professor, Professor of the Department of Germanic Philology and Intercultural Communication

5 Lunacharskii Prospekt, Cherepovets, 162600, Russian Federation


  1. Bauer, E.B., Hall, J.K., & Kruth, K. (2002). The pragmatic role of codeswitching in play contexts. International Journal of Bilingualism, 6(1), 53-74. https://doi.org/10.1177/13670069020060010401
  2. Chirsheva, G.N., & Korovushkin, P.V. (2015). Periods of a bilingual child’s development. In G.N. Chirsheva (Ed.), Contacts of Languages and Cultures: Proceedings of the IV International Conference (pp. 143-148). Cherepovets: Cherepovets State University. (In Russ.)
  3. Чиршева Г.Н., Коровушкин П.В. Периоды билингвального развития ребенка // Взаимодействие языков и культур : исследования выпускников и потенциальных участников программ Фулбрайта. Череповец: Череповецкий государственный университет, 2015. С. 143-148.
  4. Chirsheva, G.N., & Korovushkin, P.V. (2017). Mixed speech of bilingual children in a Russian family. Tomsk State University Journal of Philology, (48), 84-97. (In Russ.) https://doi.org/10.17223/19986645/48/6
  5. Чиршева Г.Н., Коровушкин П.В. Смешанные высказывания билингвальных детей в русскоязычной семье // Вестник Томского государственного университета. 2017. № 48. С. 84-97. https://doi.org/10.17223/19986645/48/6
  6. Comeau, L., Genesee, F., & Lapaquette, L. (2003). The modeling hypothesis and child bilingual codemixing. International Journal of Bilingualism, 7(2), 113-126. https://doi.org/10.1177/13670069030070020101
  7. Deuchar, M., & Quay, S. (1998). One vs. two systems in early bilingual syntax: Two versions of the question. Bilingualism: Language and Cognition, 1(3), 231-243. https://doi.org/10.1017/S1366728998000376
  8. Ervin-Tripp, S., & Reyes, I. (2005). Child codeswitching and adult content contrasts. International Journal of Bilingualism, 9(1), 85-102. https://doi.org/10.1177/13670069050090010601
  9. Gaskins, D., Quick, A.E., Verschik, A., & Backus A. (2022). Usage-based approaches to child code-switching: State of the art and ways forward. Cognitive Development, 64, 1-12. https://doi.org/10.1016/j.cogdev.2022.101269
  10. Grosjean, F. (2001). The bilingual’s language modes. In J.L. Nicol (Ed.), One Mind, Two Languages: Bilingual Language Processing (pp. 1-22). Oxford: Blackwell.
  11. Gross, M.C., López González, A.C., Girardin, M.G., & Almeida, A.M. (2022). Code-switching by Spanish-English bilingual children in a code-switching conversation sample: Roles of language proficiency, interlocutor behavior, and parent-reported code-switching experience. Languages, 7(4), 1-34. https://doi.org/10.3390/languages7040246
  12. Kuzyk, O., Friend, M., Severdija, V., Zesiger, P., & Poulin-Dubois, D. (2020). Are there cognitive benefits of code-switching in bilingual children? A longitudinal study. Bilingualism: Language and Cognition, 23(3), 542-553. https://doi.org/10.1017/S1366728918001207
  13. Lam, C.F., & Matthews, S. (2020). Inter-sentential code-switching and language dominance in Cantonese-English bilingual children. Journal of Monolingual and Bilingual Speech, 2(1), 73-105. https://doi.org/10.1558/jmbs.13308
  14. Lanvers, U. (2001). Language alternation in infant bilinguals: A developmental approach to codeswitching. International Journal of Bilingualism, 5(4), 437-464. https://doi.org/10.1177/13670069010050040301
  15. Lanza, E. (2004). Language mixing in infant bilingualism: A sociolinguistic perspective. Oxford: Clarendon Press.
  16. Meisel, J.M. (1994). Code-switching in bilingual children: The acquisition of grammatical constraints. Studies in Second Language Acquisition, 16(4), 413--439. https://doi.org/10.1017/S0272263100013449
  17. Myers-Scotton, C. (1997). Dueling languages: Grammatical structure in code-switching. 2nd ed. Oxford: Clarendon Press.
  18. Myers-Scotton, C. (2002). Contact linguistics: Bilingual encounters and grammatical outcomes. Oxford: Oxford University Press.
  19. Myers-Scotton, C., & Jake, J.C. (1995). Matching lemmas in a bilingual language competence and production model: Evidence from intrasentential codeswitching. Linguistics, 33(5), 981-1024. https://doi.org/10.1515/ling.1995.33.5.981
  20. Myers-Scotton, C., & Jake, J.C. (2017). Revisiting the 4-M model: Code-switching and morpheme election at the abstract level. International Journal of Bilingualism, 21(3), 340-366. https://doi.org/10.1177/1367006915626588
  21. Paradis, J., Nicoladis, E., & Genesee, F. (2000). Early emergence of structural constraints on code-mixing: Evidence from French-English bilingual children. Bilingualism: Language and Cognition, 3(3), 245-261.
  22. Poeste, M., Müller, N., & Gil, L.A. (2019). Code-mixing and language dominance: Bilingual, trilingual and multilingual children compared. International Journal of Multilingualism, 16(4), 459-491. https://doi.org/10.1080/14790718.2019.1569017
  23. Quick, A.E., & Hartmann, S. (2021). The building blocks of child bilingual code-mixing: A cross-corpus traceback approach. Frontiers in Psychology, 12, 1-9. https://doi.org/10.3389/fpsyg.2021.682838
  24. Rontu, H. (2007). Codeswitching in triadic conversational situations in early bilingualism. International Journal of Bilingualism, 11(4), 337-358. https://doi.org/10.1177/13670069070110040101
  25. Schächinger Tenés, L.T., Weiner-Bühler, J.C., Volpin, L., Grob, A., Skoruppa, K., & Segerer, R.K. (2023). Language proficiency predictors of code-switching behavior in dual-language-learning children. Bilingualism: Language and Cognition, 1-17. https://doi.org/10.1017/S1366728923000081
  26. Sczepurek, N.-S., Aalberse, S.P., & Verhagen, J. (2022). Multilingual children’s motivations to code-switch: A qualitative analysis of code-switching in Dutch-English bilingual daycares. Languages, 7, 1-28. https://doi.org/10.3390/languages7040274
  27. Smolak, E., de Anda, S., Enriquez, B., Poulin-Dubois, D., & Friend, M. (2020). Code-switching in young bilingual toddlers: A longitudinal, cross-language investigation. Bilingualism: Language and Cognition, 23(3), 500-518. https://doi.org/10.1017/S1366728919000257
  28. Vihman, V.-A. (2016). Code-switching in emergent grammars: Verb marking in bilingual children’s speech. Phililogia Estonica Tallinensis, 1, 175-198. https://doi.org/10.22601/PET.2016.01.10
  29. Vihman, V.-A. (2018). Language interaction in emergent grammars: Morphology and word order in bilingual children's code-switching. Languages, 3(4), 1-23. https://doi.org/10.3390/languages3040040
  30. Yow, W.Q., Tan, J.S.H., & Flynn, S. (2018). Code-switching as a marker of linguistic competence in bilingual children. Bilingualism: Language and Cognition, 21(5), 1075-1090. https://doi.org/10.1017/S1366728917000335

Copyright (c) 2023 Chirsheva G.N.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies