Linguistics and Education 52 (2019) 13–23

Contents lists available at ScienceDirect

Linguistics and Education

journal homepage: www.elsevier.com/locate/linged

“Say ‘What happened?’ in Hebrew. He does not speak Arabic!” Early

language awareness as expressed in verbal and nonverbal interactions
in the preschool bilingual classroom
Mila Schwartz a,∗ , Inas Deeb b , Sujoud Hijazy c
a
Language Department, MA Program, Advanced Studies Faculty, Oranim Academic College of Education, Kiryat Tivon, 36006, Israel
b
Center for Jewish-Arab Education in Israel “Hand in Hand”, Israel
c
Department of Learning Disabilities, University of Haifa, Israel

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this longitudinal ethnographic study was to explore how children’s verbal and nonverbal
Received 26 October 2018 behavior reflects their language awareness at a bilingual Arabic–Hebrew-speaking preschool in Israel. We
Received in revised form 30 April 2019 adopted the perspective that children’s L2 acquisition is situated within the social events and interactional
Accepted 1 May 2019
practices of the classroom community. We took a close look at six bilingual 3-year-old children—three L1
Available online 1 June 2019
Arabic-speaking children and three L1 Hebrew-speaking children. To enhance the credibility of our study,
we triangulated the collected observations of the children’s talk via teachers and parents’ testimonies.
Keywords:
The analysis revealed that children applied diverse verbal and nonverbal mediating cues to solve their
Language awareness
Pragmatic awareness
interlocutors’ communicative troubles. Their pragmatic awareness of the interlocutors’ communicative
Social interaction troubles was encouraged by a unique classroom context where the teachers stimulated children’s lan-
Preschool bilingual classroom guage mediation and peers’ backup. The study is beneficial for language teachers’ understanding of how
Hebrew they can support children’s language awareness.
Arabic © 2019 Published by Elsevier Inc.

1. Introduction the novelty of our study is in exploring how language awareness

is expressed among young bilinguals in the natural context of the
Listening to young children’s talk about language and its use bilingual classroom community. The uniqueness of this study is
provides insight into their internal thinking mechanisms regarding also in the methodological triangulation of the data sources and
languages in their environment as they engage in language learn- participants. Thus, to enhance the credibility of the ethnographic
ing. As far as we know, few studies to date have examined very classroom observations of children’s talk about language and their
young children’s language awareness as expressed in their verbal nonverbal behaviors, we conducted semi-structured interviews
and nonverbal interactions with their peers and teachers in the with the teachers and parents. These interviews enabled a more
natural context of the preschool bilingual classroom. De Houwer comprehensive picture of how the context of preschool bilin-
(2017) called for more research to explore how young bilinguals’ gual education2 affects children’s language awareness, raises their
language awareness is related to their daily need to switch between curiosity about language in their environment, in turn promoting
languages to adjust to their interlocutors’ language competence their language awareness.
and communicative troubles, namely, to investigate their prag-
matic awareness (Cheung, Mak, Luo, & Xiao, 20101 ). 2. Theoretical background
By adopting the perspective that children’s novel language
learning is situated within the social interaction with peers and Two theoretical assumptions encouraged our work. First,
teachers (Schwartz and Gorbatt, 2016; Cekaite, 2017; Duff, 2014), our longitudinal ethnographic observations on how language

∗ Corresponding author.
E-mail addresses: [email protected] (M. Schwartz),
2
[email protected] (I. Deeb), [email protected] (S. Hijazy). Bilingual education is a term used to describe an education system where
1
Note that Cheung et al. (2010) use the notion “sociolinguistic awareness” instead instruction is given in two languages, one of which is the home language of some or
of pragmatic awareness. all the children.

https://doi.org/10.1016/j.linged.2019.05.001
0898-5898/© 2019 Published by Elsevier Inc.
14 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23

awareness is expressed in daily social interactions was inspired young children developed their language awareness in the context
bysubstantial quantitative research showing bilingual children’s of two Semitic languages, Arabic and Hebrew, sharing a common
advanced level of language awareness development, including core of linguistic features (Asli-Badarneh & Leikin, 2018; Schwartz,
pragmatic awareness of their interlocutors’ communicative needs, Taha, Assad, Khamaisi, & Eviatar, 2016). The high linguistic prox-
in comparison to their monolingual peers (e.g., Barac, Bialystok, imity between the target languages is likely to support children’s
Castro, & Sanchez, 2014; Byers-Heinlein et al., 2014; Goetz, 2003). language awareness development and might be reflected in their
Second, children’s verbal and nonverbal behavior in interaction critical evaluation of their own, and others’, grammatically incor-
with their peers and teachers has been found to promote lan- rect L2 use, as will be presented and discussed later.
guage awareness (e.g., Schwartz and Gorbatt, 2016; Blum-Kulka
& Gorbatt, 2014). 2.2. Bilingual children’s verbal and nonverbal behavior reflecting
language awareness
2.1. Language awareness among young bilingual children: What
is unique? Few qualitative studies focusing on young bilingual children’s
observations, showed that language awareness could be expressed
De Houwer (2017) recently defined language awareness as “the not only by explicit talk about language per se and its under-
totality of metacognitive skills needed to allow reflecting on lan- standing, but also by nonverbal behavior (e.g., Cruz-Ferreira, 2006;
guage as an object and the monitoring of one’s own language use Lugossy, 2018). Behaviors such as body language, gestures, visual-
and that of others” (p. 83). As early as the 1960s, several stud- ization, and expressions of happiness or anger with regard to the
ies found that bilingual children demonstrated favorable outcomes children’s use of some specific language for communication or even
about their developing conceptions of language, referred to as lan- open refusal to communicate in it, may provide clues to our under-
guage awareness, when compared with monolingual children (e.g., standing of young bilingual children’s language awareness in the
Barac et al., 2014). This idea echoes the view of Russian psycholo- case of limited L2 verbal ability (Cruz-Ferreira, 2006).
gist and developmental theorist Lev Vygotsky, who wrote, in the Children’s verbal and non-verbal behavior reflecting language
early 20th century, that expressing the same thought in more awareness can also be interpreted as an expression of the language-
than one language enables children to perceive language as one based agency as a ‘socio-culturally medicated capacity to act’
system among many (Vygotsky, 1962). Indeed, data from recent (Ahearn, 2001). This agency is vividly expressed in interaction
quantitative studies evidence that early bilingualism advances chil- with peers, teachers, and parents, and could be conveyed through
dren’s understanding that different languages constitute distinct children’s verbal behavior by means of expressing voices, ideas
systems of communication. For example, in a recent experimen- about their competence in a novel language and its understanding,
tal study by Byers-Heinlein et al.(2014), using a mutual exclusivity and beliefs about languages as well as in their nonverbal behav-
task paradigm,3 bilingual (English and other language) and mono- ior (Almér, 2017; Schwartz and Gorbatt, 2016, 2018; Bergroth &
lingual English-speaking children aged 24 months were exposed Palviainen, 2017). Children as young as 2 or 3 years old are already
to a foreign language (Mandarin), which was unfamiliar to both developing attitudes toward languages around them. These atti-
monolinguals and bilinguals. The results indicated that early bilin- tudes can influence their process of novel language learning (Almér,
gual experience facilitates children’s discovery of the nature of 2017; Crump, 2014; Lugossy, 2018). For example, recently Lugossy
foreign language words. Unlike their monolingual peers, the bilin- (2018) showed how 3-year-old children, when first exposed to a
gual 2-year-olds showed a consistent understanding that English novel language (English), might feel “uncertain, helpless, and afraid
and Mandarin are distinct conventional systems. of the unknown” (p. 122). In this study, the English as a foreign
In addition to the early development of the conception of their language teacher in the Hungarian-English speaking preschool in
languages as distinct systems, bilingual children, at ages 4–6, were Hungary, described the children’s reluctance to be engaged in struc-
found to demonstrate more developed understandings of linguistic tured English lessons. This reluctance was expressed by screaming,
constructs, and thus were able, at earlier ages, to make letter–sound crying and even kicking the teacher. The awareness of lack of com-
associations, separate sound from meaning, interpret semantic petence in a novel language, as well as open unwillingness to use
attributes, note and correct morphological and syntactic struc- this language in social interaction in the bilingual classroom, will
tural errors, and comprehend syntactic structure (Bruck & Genesee, be analyzed and discussed later in this study.
1995). This advantage has been frequently attributed to the lan-
guage typology effect (e.g., Barac & Bialystok, 2012). To illustrate, 2.3. Bilingual children’s pragmatic awareness and the Theory of
in the quantitative study Barac and Bialystok (2012) compared Mind
three groups of 6-year old bilingual children (Spanish–English,
French–English, Chinese–English) and a group of English mono- De Houwer (2017) emphasized that one of the salient char-
lingual children. These children were given a test measuring acteristics of young bilinguals is their pragmatic awareness that
morphological awareness. This is a kind of language awareness their interlocutors’ language competence, intentions, and linguis-
skill, which entails the ability to reflect on and manipulate mor- tic repertoires might differ from their own (p. 89). Pragmatic
phemes, the smallest meaningful units in words (e.g., awareness awareness as a type of language awareness has been defined as
of noun plural and past tense formation) (Kuo & Anderson, 2006). highly significant in the generation and negotiation of meaning
The Chinese–English bilingual and the French–English bilinguals (Verschueren, 2000). Within the context of the home bilingual envi-
did not differ from each other (or from the monolinguals), however, ronment, case studies show that children as young as 3 year old
the Spanish–English bilinguals showed the highest performance, can behave with pragmatic awareness by spontaneously offering
significantly different from the former groups. These bilinguals’ translations, determining which language to use in which conver-
superiority has been attributed to the considerable structural over- sational situation and asking about a language spoken by a person
lap between their two languages. In the current qualitative study, in interaction with them (Cruz-Ferreira, 2006; De Houwer, 2017).
Pragmatic awareness seemingly relies on the Theory of Mind
(hereafter TOM). TOM is a key metacognitive development that
3
This task uses the mutual exclusivity phenomenon, whereby children tend to
occurs during the preschool years and refers to children’s abil-
assume that a novel word refers to a novel object rather than one that already has ity to ascribe mental states to other people and develop social
a label. insights (Flavell, 2000). This ability develops early in infancy and
 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23 15

in casual bidirectional relations with the children’s social behavior category in American conversation as well as a tendency for strong
(Astington, 2003). Although TOM as a research field, has been con- teacher-controlled discourse in the American classroom in general.
siderably investigated (see Flavell, 2004 for review), there are still In another study, Cekaite and Aronsson (2005) focused on lan-
a number of questions that current research continue to address, guage play as a way to focus on language form, and to provide peer
including the role of such factors as early bilingual development in correction and practice among nine immigrant children acquir-
children’s TOM development. Young bilinguals’ pragmatic aware- ing beginner-level L2 Swedish in a Swedish-speaking preschool in
ness might be expressed in their higher level of awareness of Sweden. It was found that children’s intentional phonological or
intentions, desires, mental states and knowledge of their part- morphological mislabeling of L2 words as joking events provoked
ners in interaction (peers, parents, teachers) than monolingual either playful or annoyed corrections from peers and children’s talk
children (e.g., Barac et al., 2014). In a comprehensive review of about correct and incorrect forms, which could facilitate correct L2
seven quantitative studies on young preschool bilinguals versus production.
their monolingual peers’ performance on the TOM measures such As noted above, in the bilingual education context, little is
as the false-belief task,4 Barac et al. (2014) presented consistent known about the descriptive characteristics and types of chil-
data on superior understanding of mental states in preschool bilin- dren’s talk about language. Recently, Morales and Rumenapp
gual children. This advanced development was found regardless of (2017) showed that Spanish–English-speaking children aged 3–5
the languages spoken by children. This advantage among young were aware of all the linguistic resources at hand, expressing
bilinguals has been attributed among other factors to their higher the value that they attributed to their languages. In addition,
pragmatic awareness, which is required for communicating with in line with the quantitative data addressed above, intriguing
speakers of different languages and with different levels of lan- evidence was reported on bilingual children’s pragmatic aware-
guage competence (Cheung et al., 2010). Taking this quantitative ness by longitudinal ethnographic observations in the bilingual
and comparative research into consideration, the current study classroom (Angelova, Gunawardena, & Volk, 2006; Schwartz and
examined the bilingual children’s behavioral verbal and non-verbal Gorbatt, 2016). Within the context of a dual Spanish–English
patterns reflecting on their language awareness in general, and language program in the United States, Angelova et al. (2006)
pragmatic awareness specifically. These patterns cannot be identi- showed how more advanced L2 learners, 6-year-old children from
fied by means of structured tests and should be explored by taking L1-English-speaking backgrounds and L1-Spanish-speaking back-
a closer look at children’s daily interactions in the natural context grounds, were able to recognize the novices’ need for support and
of the bilingual classroom through thorough ethnographic obser- spontaneously initiated mediation without explicitly being asked
vations. for help.
In sum, the data obtained from ethnographic observations of
2.4. Preschool education environment and children’s language children’ language socialization with their peers in the contexts of
awareness monolingual education with children from diverse linguistic back-
grounds show that the children commented mostly on language
Listening to young children’s talk about their languages and form. Regarding the role of talk about language in the bilingual
their use provides insight into their internal thinking mechanisms education context, the limited current research adds that this talk
regarding languages around them as they engage in language learn- reflects on the children’s pragmatic awareness of their peers’ com-
ing. According to Blum-Kulka, Huck-Taglicht, and Avni (2004), talk petence or lack of competence in L2 and their ability to manage
about language presents a high level of language awareness, which L2 use. Nonetheless, these data are insufficient to understand how
includes a focus on language form and an analysis of language. social interactions in the natural context of preschool bilingual edu-
“In this type of talk-focused talk children’s language awareness is cation are related to development of young children’s language
manifest through comments concerned with the appropriate use of awareness. The aim of this 2-year-long ethnographic study was to
language, semantic issues, and language as topic, correct usage. . .” attempt to fill this gap by focusing on how children’s verbal and
(Blum-Kulka et al., 2004, p. 317). nonverbal behavior reflects their language awareness at a bilingual
In the monolingual education context, children’s talk about Arabic–Hebrew-speaking preschool in Israel. To enhance the cred-
language was investigated by longitudinal ethnographic observa- ibility of our study, we triangulated the collected observations of
tions of children’s language socialization in interaction with peers the children’s talk via teachers and parents’ testimonies.
and teachers (Aukrust, 2001; Blum-Kulka et al., 2004; Cekaite & The present study was motivated to address the following ques-
Aronsson, 2005; Gorbatt-Brodstein, 2012). For example, Aukrust tions:
(2001) focused on examining the comparative characteristics of
the talk about language of 3–4-year-old children in preschools in 1. What could be learned from longitudinal observations of the
Norway and the United States. Three main groups of this type of talk children’s verbal and nonverbal behavior about their language
were identified following the analysis of children’s and teachers’ awareness, in a context of dual language preschool bilingual
audio-recorded preschool interactions: talk about language form; education?
talk about discourse management and reported speech. Although it 2. How do parents’ and teachers’ testimonies contribute to our
was found that, overall, talk about language was highly infrequent understanding of children’s verbal and nonverbal expression of
in the 3–4-year-old children, some culture-related differences were language awareness in the bilingual classroom and at home?
identified between Norwegian and American samples. For example,
Norwegian children initiated talk about turn-taking in commu-
nication more often than the American children, whereas in the 3. Method
American sample, the conversation turns appeared to be more
teacher-controlled than in the Norwegian sample. This differ- 3.1. Research method
ence was attributed to turn-negotiation as an important cultural
This study was part of a larger ethnographic project, which
explored children’s bilingual development in Hebrew and Arabic.
4
The standard false-belief task, a critical test for understanding theory of mind,
In this study, we used a mixed methods design, combining quali-
was proposed by Wimmer and Perner (1983) to examine children’s ability to predict tative and quantitative ethnographic approaches to data collection
the thoughts or behavior of someone holding a false belief. and analysis. The data were collected and documented by means
16 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23

of ethnographic tools such as observations, interviews, and field We applied purposive sampling to select information-rich cases
notes. Ethnographic methodologies are chosen when researchers that were representative of the target bilingual classroom popula-
wish to situate linguistic data within a social and cultural milieu tion (Palinkas et al., 2015). We performed close observations on six
(Morse & Niehaus, 2009). The quantitative analysis of frequency children. To ensure internal diversity and representativeness of the
of the children’s verbal and nonverbal behaviors reflecting on sample, the selection criteria were gender and ethnic background.
their language awareness was important to provide a more pre- Thus, the sample included three L1-Hebrew-speakers (one girl and
cise examination of the phenomenon under study. The design is two boys) and three L1-Arabic-speakers (two girls and one boy).
described in detail below. Inclusion criteria included at least six months’ experience with
L2 acquisition and having no developmental or behavioral prob-
3.2. Brief description of the study’s socio-linguistic and lems reported by their teachers. Finally, based on Almér’s (2017)
educational background recent suggestion of how to research young bilingual children,
we focused on the children whom the teachers characterized as
Israel is a multilingual and multicultural country, with both socially active and who were particularly eager to participate in
Hebrew and Arabic as official languages. Hebrew is the dominant different classroom activities. The children’s ages ranged between
language in most life domains and is presented as the dominant 3;3 (years;months) and 3;5 at the start of the study and between 5;0
language in Israeli public spheres. Arabic is a minority language and 5;2 at the end of the data collection stage. Parental consent was
and is the national language for more than one million Arab-Israeli obtained through direct communication during parent–teacher
citizens (about 20% of the population). Given the higher status of meetings.
Hebrew as a majority language, Hebrew speakers in general do Table 1 presents the children’s background information and a
not attain a level of Arabic similar to the level of Hebrew that brief description of their L2 competence at the end of the study.
Arabic-speakers attain. Most Arab-Israelis understand and speak Teachers. The study participants were two preschool teachers:
Hebrew and use it at work and in other settings. In addition, one Hebrew model teacher, Naomi, and one Arabic model teacher,
English is considered a semiofficial language that is widely used in Jasmin.5 The teachers expressed their willingness to be observed
numerous contexts (especially in academic settings, economics and during all study periods and to be interviewed. Naomi was a novice
politics). Finally, this linguistic multiplicity is enriched by the native teacher, who had joined the preschool in September 2015, with
languages of large groups of immigrants (e.g., Russian, Amharic, no prior knowledge of Arabic. She had recently obtained a bache-
Spanish, French and scores of others). lor’s degree in preschool education followed by some pedagogical
Israel has separate Hebrew-speaking and Arabic-speaking edu- teaching experience as a preschool teacher in a monolingual set-
cation systems and therefore, Arab and Jewish children are ting. Jasmin had a bachelor’s degree in preschool and first grade
educated in different schools. To overcome this socio-educational teaching. She had been teaching in the preschool since 2013, had
and cultural separation, in 1997 the Center for Bilingual Educa- five years of professional experience as a bilingual teacher, and
tion was established. The aim of this organization is to promote spoke fluent Hebrew. In addition, Jasmin had personal experience
mutual tolerance and bilingual, bicultural education that acknowl- of bilingual schooling in the elementary Arabic–Hebrew-speaking
edges both official languages of Israel. The organization currently school in central Israel. As the more experienced teacher, Jasmin
operates six bilingual preschools and schools throughout the coun- took a leading role during the first months of the academic year
try. Both Jewish and Arab children learn in the same schools, while and helped Naomi to adjust. As was observed during the study, the
the teaching and management staff represent both ethnic groups teachers shared the floor in the classroom and alternated through
equally, with each class having two homeroom teachers—one Arab turn-taking during circle time. At other times, they divided the chil-
and one Jewish. The teachers share the educational tasks, respon- dren into small groups for project-based learning in the shared
sibilities and teaching in each class. classroom space.
Our study was conducted in one of these few bilingual Parents. Six semi-structured interviews were conducted in
preschools, in which dual language input was implemented with Hebrew, either in the preschool or in the participants’ homes, with
language separation by the teachers. In other words, the teachers the five children’s mothers and one father, at their convenience.
spoke mostly in their native languages, thus playing the role of a Each interview lasted approximately 60 min. The parents gave their
model for either Hebrew or Arabic, while providing plenty of input consent to audio-record the interviews, enabling the interviewer
in both languages and using responsible code-switching (García, (the first author) to focus on the conversation without needing to
2009, p. 299). As reported by the teachers in their interviews, at take notes.
young ages, their primary goal was to encourage bilingual devel-
opment by means of dual language input in a natural and enjoyable 3.4. Instrumentation and data generation, transcription, and
way, while gradually developing L2 perception and understanding. analysis
Later on, when most children were able to understand the L2, the
teachers tried to encourage L2 production using oriented activities Using the ethnographic approach toward data collection
and morning circles. (Blommaert & Jie, 2010), we conducted a comprehensive inves-
tigation of daily activities and social interactions in the bilingual
3.3. Participants preschool. The data were collected during two academic years,
from November 2015 through August 2017. In the first stage of the
Children. Enrolled in the preschool were 14 children aged 2;5 to study, field notes were taken from November to December 2015,
3;5 at the beginning of the study, of whom six were native speakers to enhance the credibility of our data and to ensure that the par-
of Hebrew and eight were native speakers of Arabic. Notably, some ticipants (children and teachers) had become used to our presence
of these children come from a diverse multilingual background at in the classroom (a total of eight observation sessions during this
home, applying one parent-one language policy, such as Hebrew period). This early engagement in the classroom context facilitated
or Arabic as one parent’s language and English or Russian as the our close familiarity with the participants and their life histories,
other parent’s language. During our observations, we noticed that
the teachers showed their keen interest in the children’s home lan-
guages and cultures, allowing time for listening to, and even for 5
Pontier and Gort (2016) proposed the terms “Hebrew model teacher” and “Ara-
using and learning languages other than Hebrew and Arabic. bic model teacher” to define the bilingual teachers.
 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23 17

Table 1
Children’s gender, mother’s education, age at the beginning and end of the study (years; months), L1, age when started preschool, and L2 competence.

Name Gender Mother’s education L1a Age at beginning of Age at end of study Age when started Relative level of L2
(in years) study (years;months) bilingual preschool competence at the end of
(years;months) the study

Danny Male 16 Hebrew 3;4 5;1 2;1 Receptive bilingual skills,

few cases of L2 productive
use
Hannah Female 16 Hebrew 3;5 5;2 1;1 Receptive bilingual skills,
few cases of L2 productive
use
Shahar Male 18 Hebrew 3;5 5;2 1;1 Receptive bilingual skills,
few cases of L2 productive
use
Maria Female 19 Arabic 3;5 5;2 2;9 L2 productive use
Rima Female 18 Arabic 3;3 5;0 1;0 L2 productive use
Omar Male 17 Arabic 3;3 5;0 1;0 Mostly formulaic L2 use
and few cases of L2
productive use
a
The L1 was reported by the parents when they registered the child for the preschool.

thus preparing the ground for a relationship of trust, which was interest in languages and verbal and nonverbal behaviors reflecting
crucial to ensuring that both children and teachers felt able to con- their language awareness.
fide in the researchers (Barley & Bath, 2014; Shenton, 2004). In The parents were asked questions relating to the background
addition, 58 weekly observational video recording sessions were and the history of the child’s language development and acquisi-
conducted from January 2016 to August 2017, except during school tion, as well as about family languages, sequence of their acquisition
holidays (about five weeks). Each session lasted three hours, from or learning, language practice and the children’s exposure to their
8.30 am to 11.30 am. Overall, in this study, we observed 178 h L2 outside the preschool context. In addition, they were asked
of preschool classroom time and transcribed 93 h of these obser- questions regarding the children’s queries about languages around
vations. The transcribed video observations were combined with them, and verbal and nonverbal behavior reflecting their language
field notes. All video recordings and field notes were conducted awareness. The focus of this paper is on the parents’ reports of the
by the first author and the research assistant, an MA candidate in children’s queries and reflections on languages around them.
educational consulting. We received permission to perform video- The data analysis was conducted in two stages.6 In the first
recordings in the preschool and to present the collected data stage, based on the analysis of the children’s verbal and nonverbal
(the observations and the interview extracts) from the Center for behavior elaborated on by Aukrust (2001), Blum-Kulka et al. (2004),
Bilingual Education, from the teachers, and from the children’s Gorbatt-Brodstein (2012), and Cekaite and Aronsson (2005, 2014),
parents. video-recorded and field-noted classroom interactions were selec-
To explore our research questions, we applied methodological tively transcribed to capture patterns in the children’s verbal and
triangulation, which permits a comparison of the findings derived nonverbal behavior reflecting their language awareness. In most
from different data sources to interpret the phenomenon under cases, we classified the observed examples of this talk as discourse
study and to reduce observer or interviewer bias (Shenton, 2004). comments and not as fully-fledged discursive events. It is notable
Thus, in addition to the field notes and video-recorded observa- that a discursive event is defined as a fully-fledged talk whenever it
tions, we conducted semi-structured interviews with the teachers presents coherent sequences of a number of turns-at-talk which are
and parents to obtain their reflections on the children’s early talk related to the same topic; while the discourse comment is a brief
about languages. remark which does not comprise an extensive talk around some
The methodology included ethnographic field notes, video- topic (Hamo & Blum-Kulka, 2007). We selected the transcribed dis-
recordings, and semi-structured interviews with the teachers. The course comments because they illustrated these patterns. The data
teachers were informed that the purpose of the study was to exam- were transcribed in detail, in table form, allowing for the inclusion
ine how classroom interactions are related to the children’s L2 of nonverbal information from the videos and field notes. Each tran-
progress. No further information on the focus of the study was scription was made by two transcribers, a native Hebrew-speaker
offered. They were asked to allocate a suitable day for observation and a native Arabic-speaker.
that included diverse daily activities such as a meal, circle time, In the second stage, a second transcription version was made
some structured and planned teacher-led activities within small using Excel to provide quantitative (frequency) and qualitative
groups, teachers’ storytelling, as well as unplanned and unstruc- descriptive analyses of the children’s behavior. In this version, the
tured activities such as free-play in the classroom and outdoor transcriptions were coded for information about the date of the
leisure time. observed behavior, the participants’ L1 and the language/s used,
We conducted semi-structured interviews with the teachers their partners in interaction, their verbal or nonverbal behavior and
to enhance the credibility of the obtained observations. The first its classification in terms of language awareness category and sub-
author conducted four in-depth semi-structured interviews (two category, type of classroom activity, and contextual comments (see
interviews with each teacher) individually. The interviews took Table 2).
place in March 2016 and July 2017 and were conducted in Hebrew. Based on the salient characteristics of young bilinguals’ lan-
Our goal was to obtain the teachers’ self-reports regarding the guage awareness provided by De Houwer (2017) and Cruz-Ferreira
target children’s bilingual development. The teachers were also (2006), we identified 53 cases of the following categories and
asked about their background data (i.e., education, professional
experience). It is noteworthy that, in this paper, our analysis of
the teachers’ interviews draws strictly on the study aims and was
6
Note that the developmental changes in the children’s behavior were not under
therefore limited only by the teachers’ reflections on the children’s
the scope of our analysis.
18 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23

Table 2
Example of how the transcriptions were codified.

Date Child’s name/L1 Partners in Verbal or nonverbal behavior Category/subcategory Type of Comments
interaction and classroom
their L1 activity

17.03.2016 Maria/Arabic Hanan, the Maria: (Say “What Critical evaluation of L2 Free play time Shahar was crying.
research happened?” in Hebrew, competence/self- Maria heard Hanan
assistant/Arabic he does not speak monitoring and talking to Shahar in
Shahar, Hebrew Arabic!) corrective feedback for Arabic, so she told her
others’ incorrect in both Arabic and
language use Hebrew that Shahar
doesn’t speak Arabic,
and she translated
Hanan’s question into
Hebrew

subcategories of language awareness that have been expressed in and Gorbatt, 2016). In this study, we were surprised to observe
the children’s verbal and nonverbal behavior: behaviors expressing empathy, among children as young as 3–4
years, to their peers’ and teacher’s troubles in understanding L2.
(1) Pragmatic awareness of communicative needs of peers and Excerpt 1 illustrates this point:
teachers7 (overall 20 cases): negotiating meanings (a) by Excerpt 1: Observation: January 8th, 2017
translating, and (b) by using multiple verbal and nonverbal (In the English translation, italicized text = Arabic, nonitalicized
resources; text = Hebrew)
(2) Critical evaluation of L2 competence (overall 33 cases): (a) Participants: The L1-Arabic-speaking teacher, Jasmin, two L1-
self-monitoring and corrective feedback on others’ incorrect Hebrew-speaking boys, Danny and David.
language use; and (b) talk about L2 comprehension or lack of it Situation: Classroom activity during circle time. Jasmin talked to
and about unwillingness to use L2. David in Arabic, but David did not respond. Danny thought he did
not understand so he tried to explain again in Hebrew.
To answer the second research question, we selected teachers’
Turn Name Utterance
and parents’ reflections on how children expressed their aware-
ness of languages around them through their verbal and nonverbal
behavioral patterns.
1 Jasmin:
Who can think of a game to play with these? [Jasmin points
4. Results and discussion
to the Lego game]. You [addressing David] have to take Lego
pieces that match the color of the circle. [David is confused
In this section, we will analyze and discuss how the identified and looks around trying to understand what he has been
categories and subcategories of the children’s verbal and nonver- asked to do]
bal behavioral patterns reflect children’s language awareness. To 2 Danny: Take a brick that is the color
illustrate our analysis, we will present excerpts from the classroom of the circle.
observations and from the teachers’ and parents’ testimonies. In this observation, we were interested to see that Danny
showed awareness of David’s difficulty in understanding the
4.1. Pragmatic awareness of communicative needs of peers and teacher’s instructions in Arabic and was willing to help him
teachers without being explicitly asked. In addition, it appeared that
Danny, who usually showed only receptive bilingual skills in
Overall, we observed 20 cases of the children’s pragmatic Arabic (L2), felt confident enough to translate the teacher’s
awareness of communicative needs of their peers and of the instruction in Arabic into Hebrew for his less competent
L1-Hebrew-speaking teacher. These cases will be analyzed and dis- L2 peer.
cussed in the following two subsections: (1) negotiating meanings Another interesting observation was that Danny did not trans-
by translating; and (2) negotiating meanings by using multiple ver- late Jasmin’s instructions word for word but explained what to do
bal and nonverbal resources. in Hebrew by significant syntactic simplification (“Take a brick that
is the color of the circle” instead of the teacher’s “You have to take
4.1.1. Negotiating meanings by translating Lego pieces that match the color of the circle”). In several cases, we
Learning through interaction and mediation is a characteristic of observed children’s L2 mediation to simplify the teacher’s speech.
human intelligence (Vygotsky, 1987). As was found within a dual In another case, outdoors, when Naomi was explaining about
language classroom context, by ages 5 to 6, children might play insects in Hebrew, we observed how Maria (L1-Arabic-speaking
a more knowledgeable role and negotiate meanings for their less girl) negotiated the understanding of her L1-Arabic-speaking peers.
advanced L2 peers by translating (Angelova et al., 2006; Schwartz Naomi asked them: “Where does the bug obtain the food from?”
Maria scaffolded the teacher’s question by indirect translation
and paraphrasing. She replaced Naomi’s utterance “obtain food”
7
Note that in addition to the presented above categories and subcategories, we by the semantically simpler utterance “get food.” Moreover, after
observed 17 cases of the children switching between languages to accommodate noticing that her peers did not provide an answer, she added a
language choice with their interlocutor’s language competence. However, since
scaffolding question: “Where do you get your food from?” (May 14th,
this pragmatic behavior was broadly analyzed in previous studies that focused
on young bilingual children’s interactions at home)e.g., Cruz-Ferreira, 2006; De
2017). These modifications of the teachers’ utterances to more age-
Houwer, 2017), we do not include this behavioral pattern when presenting the appropriate grammatical structures during translation might be
analysis of our results.
 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23 19

evidence of the children’s pragmatic awareness of child speech as regarded as an expression of the children’s pragmatic awareness of
distinct from adult speech. their teachers’ and parents’ needs, stimulated by the general atmo-
sphere of the target classroom. As we found, the teachers frequently
acknowledged children’s L2 mediation in the classroom.
4.1.2. Negotiating meanings by using multiple verbal and
Another fascinating facet of the children’s pragmatic awareness
nonverbal resources
was their ability to convey meaning by drawing on various nonver-
During our 2-year-long observations, we found that children
bal resources such as visualization, body language, and intonation.
used diverse and creative ways to negotiate meanings in order to
This behavioral pattern was expressed in communication with the
help their less competent peers in L2 and to help the L1-Hebrew-
interlocutors with low competence in L2. Recent experimental
speaking teacher, Naomi. They imitated their teachers’ strategies
research showed that young children growing up bilingually were
such as repeating a novel word, pronouncing it slowly, syllable by
superior to their monolingual peers in relying on such pragmatic
syllable, hinting first by pronouncing the first syllable and using
cues as pointing and eye gaze in learning novel words (e.g., Brojde,
antonyms or semantic substitution. In the following testimonies,
Ahmeda, & Colunga, 2012). Higher level of attention to nonver-
Naomi described these children’s language strategies:
bal cues facilitated the bilinguals’ word learning. It has been found
During lunch when . . . they [Rima and Maria, L1-Arabic- also that language teachers’ use of spontaneous hand and arm ges-
speaking girls] said some word [in Arabic] . . . that means not tures, together with their speech, fostered L2 understanding among
healthy. They said “it isn’t healthy”. . . and then they said similar bilingual children during classroom instruction (Church, Ayman-
things that weren’t connected to the word. And I tried to make Nolley, & Mahootian, 2004). In the same way, in our study, both
the connection between all the words and this word and then I teachers drew on their use of gestures when communicating with
said that something doesn’t make sense. So, I asked Rima what children. It is also noteworthy that the pictures on the preschool
the word is. It took her a long time to answer me; she couldn’t walls presented the application of the gesture strategy. They con-
find the word, but then she said: “It’s like sick”. . ., she told me tained images or pictures of conventional and accepted gestures.
in Hebrew “like sick” so “it is like she [the girl] isn’t healthy, do In imitating the teachers’ strategy, the children’s nonverbal behav-
you understand?” (Interview with Naomi, July 21st, 2016) ior was, in fact, an ancillary communicative tool at their disposal
. . . Maria [L1-Arabic-speaking girl] either finds the correct to convey meanings in L2. In the following reflection, Naomi high-
word [by translating the Arabic word into Hebrew] or says a lights how Maria, the L1-Arabic-speaking girl, applied this strategy
word that is similar in Hebrew, or just like I will say “fork” instead of code-switching to L1:
instead of “spoon,” it is an eating utensil, it is like in the same
In communication with me, Arab children are in no hurry to use
category. . .So, I will understand more or less what she means
words in Arabic when they do not know the Hebrew word. They’d
and then I repeat, and she says: “Yes!” (Interview with Naomi,
rather look for ways to illustrate the word they do not know in
July 21st, 2016)
Hebrew. As Maria did today, instead of saying the word “braid”
It should be noted that Naomi, as an adult Arabic learner, strug- in Arabic, she said the name of the character depicted from the
gled with L2 understanding and had a communication barrier with cartoon who has a braid. And so I understood what she meant
the L1-Arabic-speaking children. She made use of the children’s and said to her: “You mean my braid, in Hebrew it’s a braid, a
bilingual competencies as a resource to negotiate understanding of braid.” [Naomi repeated the novel word twice while touching
what was spoken in Arabic as well as to learn novel words. Inter- her braid.] (Interview with Naomi, April 7th, 2016)
estingly, even though all three Hebrew-speaking children whom
In line with TOM, this example showed us how children
we observed showed mainly receptive skills in Arabic while rarely
are sensitive to their interlocutors’ mental states including their
using the language productively, they frequently volunteered to
communicative difficulties. The children can and do solve these
play the role of L2 teachers for Naomi as well as for their parents:
difficulties by imitating the teachers’ strategy to avoid their inter-
Hannah [L1-Hebrew-speaking girl] translates a lot for me when locutors’ confusion. The teachers emphasized that the children
I ask her what something is and how to say it in Arabic. . .she often far exceeded their teachers’ expectations of their capabili-
translates for her mom. . . she tests her mom at home [for knowl- ties, such as switching to Arabic whenever they did not know a
edge of Arabic], for example “what color is this?”. . . word in Hebrew. In sociocultural theory, imitation is viewed as the
Shahar’s mom came to us two or three days ago and said that unique human ability that lays the foundation for children’s devel-
Shahar has started speaking Arabic with her at home when mak- opment (Vygotsky, 1987). Lantolf and Thorne (2006) stressed that
ing simple requests: “Give me a cup of water”. . . He explains imitation is a conscious process, which can result in transforma-
what it is [“cup of water” in Arabic] to his parents if they don’t tion of the original modeling of behavior (p. 203). Previous research
understand. (Interview with Naomi, July 21st, 2016) showed children’s imitation of what they heard from their teachers
and peers in the language classroom during the process of L2 learn-
As stated in this testimony, Hannah and Shahar, the L1-Hebrew-
ing (e.g., Lantolf & Genung, 2002; Ohta, 2001). The current example
speaking children, felt more self-confident and comfortable
adds to the theory that children as young as 3 years old could imi-
speaking with their L1-Hebrew-speaking teacher, Naomi, and their
tate not only what they heard from their teachers but also their
parents, who were novel Arabic learners. The next excerpt illus-
language mediation strategies.
trates how Iris, Hannah’s mother, reflected on her daughter’s L2
mediation at home:
Researcher: You both speak Arabic at home?
Mother: Yes, we do, we are learning, so we exchange sentences
4.2. Critical evaluation of L2 competence
and homework, so I share my homework with her [Hannah]. . .
So, Hannah really helps me with sentences, like this, she corrects
The observed cases of the critical evaluation of L2 competence
me. (Interview with Iris, January 17th, 2016)
will be analyzed and discussed in the following two subsections:
As reported above, in the intimate atmosphere at home, Han- (a) self-monitoring and corrective feedback for others’ incorrect
nah felt superior to her mother in Arabic, encouraging her to talk language use, and (b) talk about L2 comprehension or lack of it and
about language and to use it productively. These behaviors can be about L2 use.
20 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23

4.2.1. Self-monitoring and corrective feedback for others’ considered as a teachers’ strategy and means providing corrective
incorrect language use feedback minus the error, while maintaining the child’s intended
In the observed classroom, young bilingual children not only meaning (Lyster, 2007). Maria then willingly repeated the correct
paid attention to their interlocutors’ communicative troubles but feminine form after her friend, Rima, who was more competent in
also monitored their own as well as their peers’, teachers’, and par- L2 Hebrew (Turn 4).
ents’ L2 use. Overall, we observed 21 cases of self-monitoring and The teachers also paid attention to the girls’ surprising interest
critical evaluations of interlocutors’ L2 competence. The following in gender marking in both languages:
excerpt illustrates how Maria, an L1-Arabic-speaking girl, com-
I noticed that Maria and Rima are at the stage of female and male
mented critically on an adult’s inappropriate L2 use with Shahar,
word forms in Hebrew; like, when one of the Arabic-speaking
her L1-Hebrew-speaking peer:
children makes a mistake or gets confused, they correct them
Excerpt 2: Field note: March 3rd, 2016
right away. If Ribal [L1-Arabic-speaking boy] says “I want”
(In the English translation, italicized text = Arabic, nonitalicized
text = Hebrew) [ in Hebrew - the use of the incorrect female flection
Participants: The L1-Arabic-speaking research assistant, Hanan, of the verb “want”], Maria will say “No, I want” [
L1-Arabic-speaking girl, Maria, and L1-Hebrew-speaking boy, Sha- she corrects it to the male gender inflection of the verb “want”].
har. (Interview with Jasmin, April 21st, 2016)
Situation: Free play time in the classroom. Shahar is crying. Maria
Additionally, Naomi reported receiving mini-lessons from these
hears Hanan talking to Shahar in Arabic and interrupts.
girls on the inflectional gender marking in Arabic:
Turn Name Utterance
Like she [Maria] explains it to me. When I ask and then she
1 Hanan: Shahar, what happened to you? (Shahar gives me a few more examples, to make sure that I understood
does not stop crying and ignores Hanan’s question in the male and female forms and a few more words to make sure I
Arabic) understood what is male and what is female. . .She tells me: for a
2 Maria: Say “What happened?” in boy [in Arabic] say ‘inta’ [you, masculine personal pronoun] and
Hebrew, he does not speak Arabic! for a girl say ‘inti”’ [you, feminine personal pronoun]. (Interview
As demonstrated, at age 3;5, Maria was not only aware of her with Naomi, March 26th, 2017)
peer’s low competence in L2 Arabic, but also showed assertiveness One possible interpretation of this focus on the gender gram-
by criticizing Hanan, who did not take Shahar’s communication dif- matical marking in both Hebrew and Arabic is the fact that, in
ficulty into account. Maria tried to solve this confusing situation both target Semitic languages, nouns, adjectives, and verbs are
by translating Hanan’s question into Hebrew (“What happened?”). inflected for gender—masculine and feminine. Moreover, there is
This behavior indicates the girl’s pragmatic sensitivity. a relatively high degree of proximity in gender marking between
Moreover, during our 2-year-long observations, we noticed sev- these two Semitic languages, e.g., zero suffix of masculine singu-
eral cases that clearly expressed the children’s language awareness lar nouns, whereas feminine singular nouns are marked by the
whenever they reflected on language per se, as a system, by talk- suffix –a. These similarities between the languages might flag
ing about incorrect L2 grammatical production. Our observations the girls’ attention to the parsing of words into separate mor-
revealed that two Arabic-speaking girls, Rima and Maria, were phemes, thus facilitating their explicit talk about this grammar
involved, several times, in explicit critical commenting on gen- category and language awareness development (see discussion
der marking and agreement in L2 Hebrew. In line with what about typological proximity and language awareness in Lauchlan
was observed in the home environment by De Houwer (1990) et al., 2013). In addition to the possible facilitating role of the typo-
and Cruz-Ferreira (2006), they corrected each other and their L1- logical similarities between the target languages, recent research
Arabic-speaking peers’ errors by returning to the same phrase: on children’s language talk and language play in L2-immersion
“That is not how to say it in Hebrew!”, “It is wrong!” or by recasting classrooms, has shown that the grammatical features of L2 are
the incorrect utterance, as illustrated in the following excerpt: made more salient as a result of joyful language play or meta-
Excerpt 3: Observation: April 24th, 2016 talk during peer interaction (Broner & Tarone, 2001; Cekaite &
(In the English translation, italicized text = Arabic, nonitalicized Aronsson, 2005).
text = Hebrew) Overall, the teachers characterized Rima and Maria as “success-
Participants: L1-Arabic-speaking girls, Rima and Maria, and an ful L2 learners” and attributed their progress toward productive
L1-Arabic-speaking boy, Omar. use of Hebrew to their social skills (being socially active, talkative,
Situation: Outdoor play in the playground. The teacher Naomi is and “not shy”) and to their willingness to communicate with their
watching the children playing. L1-Hebrew-speaking peers. These girls seemed to be motivated
Turn Name Utterance to use L2 actively “to become a part of the social world of its
speakers” (Wong-Fillmore, 1976, p. 666). In addition, it was appar-
1 Omar: I am tall! (Omar climbs up and shouts to Naomi ent that, through their curiosity in L2, the girls showed that they
in Hebrew)
were not passive recipients of teachers’ teaching, and approached
2 Maria: You are not tall, L2 learning in an active way by frequently asking: “What is
only me, I am tall, I am tall!
this in Hebrew?” The following reflection by Jasmin stressed
3 Rima: I am tall. this point:
4 Maria: Naomi, I am tall. [Maria happily repeats the
correct utterance after Rima] There are children who are more interested in language, are enthu-
siastic about speaking in L2, repeat after Naomi more often, talk
*Erroneous use of the masculine form with zero flection instead of feminine
to themselves more at home, and sing songs at home in L2. They
form in gender marking of the adjective, “tall”.
take more interest in the other language, like Rima and Maria. . .
In this illustration, the 3;5-year-old girl Rima provided cor-
(Interview with Jasmin, June 4th, 2016).
rective feedback to Maria by recasting the erroneous masculine
Moreover, Rima assumed the role of Maria’s L2 teacher. Maria
grammatical form of the adjective, meaning “tall” with the
entered this bilingual classroom only at the start of the academic
correct feminine form, (Turn 3). In general, recasting is
 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23 21

year and showed increasing progress in Hebrew. As reported by (In the English translation, italicized text = Arabic, nonitalicized
her teachers and parents, Maria’s fast L2 socialization could be text = Hebrew)
explained by Rima’s support: Participants: The L1-Arabic-speaking teacher, Jasmin, the L1-
Hebrew-speaking teacher, Naomi, and L1-Hebrew-speaking boy
Rima loves to translate, she translated for Maria from Hebrew
Danny
to Arabic when Maria asked her to translate what Naomi said in
Situation: Morning circle time. Jasmin asks Danny to begin a
Hebrew. (Interview with Jasmin, June 4th, 2016).
game they used to play during morning circles.
It is noteworthy that Rima and Maria showed interest not only
Turn Name Utterance
in Hebrew as L2, but in languages in general:
. . .Rima’s father teaches her German. Rima translates how to say
“I love you” from Hebrew to German. (Interview with Jasmin,
June 4th, 2016).
In February, when we were in the United States, when the wait- 1 Jasmin:
We want to play the game “The King Said,” “The King Said.”
ers started asking us: “What would you like to drink? Water?”
Danny will start, but we want to play in Arabic, Danny.
So she [Maria] immediately picked up the word “water” and,
whenever the waiter came, she began to say: “I want water [in 2 Danny: The King Said, it is hard (spoken
quietly).
English], right?”. . .She has a very good ear for languages. She is
3 Naomi: It is hard.
very quick on the uptake. (Interview with Maria’s mother, Baha,
4 Jasmin: For example, clap hands. (Jasmin claps her
June 12th, 2016) hands)
We also observed how both girls, Rima and Maria, sang and 5 Danny: But. . .not in this language.
played together by imitating an English accent and pretending to 6 Naomi: . . .You
chat in English using some real English words and nonwords. The choose the language in which you want to play.
girls’ teachers and parents reported their striking curiosity about 7 Danny: The King said put your
languages: hands on your head!

Since the age of 2, the whole issue of languages really interested As can be seen, the teacher, Jasmin, gently coaxed Danny to
her. When we would meet people, first she would tell us what play in Arabic (Turn 1). Danny started playing the game, but then
language they were speaking, she would recognize the people stopped, stating that it was difficult for him to speak in Arabic, “. . .
according to what they were speaking. Once, when we were not in this language” (Turn 5). Danny’s behavior could be inter-
in Nazareth, sitting waiting for the doctor, Rima said; “What? preted as demonstrating his critical evaluation of L2 competence
Is she speaking Arabic?”. . . Rima realizes that languages exist, as well as an expression of child bilingual agency, expressed in
she can recognize them; she uses the correct intonation in each his resistance to act according to the norms of bilingual context,
language. . . She can also identify English. There was a time when namely, to switch to Arabic (L2) as his interlocutor’s designated
she first heard Russian too, and asked what language it was. At language (Turn 5). The ecological perspective on language learning
first, everything for her was Hebrew, whatever wasn’t Arabic views the children’s agentic openness to a novel language as a sig-
was Hebrew for a short period, until she acquired Hebrew, and nificant factor in the learning process (van Lier, 2011). We might
then it was English. (Interview with Rima’s mother, Rana, June suggest that Danny’s unwillingness to use L2 could be related to
17th, 2016) his beliefs about this language as not critical for his communica-
tion in the classroom since both teachers could understand his L2
Although we cannot legitimately say that we know why Rima and accept his receptive L2 skills and his L1-Arabic-speaking peers
behaved as she did, we can only offer some plausible interpretations could also understand him to some extent.
for her behavior. Over and above possible cognitive and percep-
tual predisposition to language sensitivity, her mother’s testimony
highlights Rima’s early awareness of the linguistic diversity in her 5. Conclusions
environment. This awareness seemed to be accelerated by inter-
twining of such factors as her very early exposure to Hebrew as In this study, we viewed the bilingual classroom as a natu-
L2, her father’s efforts to teach her German, multiple opportunities ral arena and unique opportunity to develop language awareness
for exposure to the multilingual social encounters, as well as the through children’s daily interactions with peers and teachers hith-
girls’ personal characteristics such as outgoingness and searching erto unseen in the home context. We focused on six children in
for interactions with L1-Hebrew-speaking peers. interactions with their peers and teachers and triangulated our lon-
gitudinal observations with the teachers’ and parents’ input. This
triangulation of different data sources strengthened the credibility
4.2.2. Talk about L2 comprehension or lack of it and talk about of the multilevel analysis of our findings. Teachers’ reflections gave
unwillingness to use L2 us better understanding of the observed as well as unobserved pat-
The children’s talk about their own or their interlocutor’s L2 terns of children’s behavior. Moreover, we believed that the fact
comprehension or its difficulties was observed 12 times in diverse that teachers were tuned to both how children behave and what
classroom contexts. Our observations identified a “strange” behav- they say during their daily classroom interactions, positioned the
ior pattern by one L1-Hebrew-speaking boy, Danny. As illustrated teachers as “co-researchers” in our study. Together with the par-
in Excerpt 1, the boy was ready to help his less competent L2-Arabic ents’ reports, they added a significant body of data about children’s
peers in understanding the L1-Arabic-speaking teacher’s instruc- reflections on languages around them and examples of their prag-
tions. At the same time, Danny repeatedly refused to use Arabic matic awareness, both inside and outside the classroom. We can
with peers and with the L1-Arabic-speaking teacher, even after conclude that the teachers’ and parents’ testimonies were congru-
three years of immersion in the preschool. In the following class- ent and complimentary with what we observed in the classroom.
room interaction, Danny talked explicitly about his difficulty in L2 The study has shown how children as young as 3 to 5 years
use during circle time: old applied diverse verbal and nonverbal mediating cues to solve
Excerpt 4: Observation: March 3rd, 2017 their interlocutors’ communicative troubles. From a sociocultural
22 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23

learning theory perspective, the mediation is viewed as an active Almér, E. (2017). Children’s beliefs about bilingualism and language use as expressed
process, central to teaching and learning through interaction. We in child–adult conversations. Multilingua, 36(4), 401–425.
Angelova, M., Gunawardena, D., & Volk, D. (2006). Peer teaching and learning: Co-
linked the observed cases of our children’s L2 mediation to their constructing language in a dual language first grade. Language and Education,
early developed pragmatic awareness, reflecting on concurrently 20(3), 173–190.
developing social insights, such as TOM (Cheung et al., 2010). We Asli-Badarneh, A., & Leikin, M. (2018). Morphological ability among monolingual
and bilingual speakers in early childhood: The case of two Semitic languages.
assumed that their pragmatic awareness of the interlocutors’ com- International Journal of Bilingualism (in press).
municative troubles was encouraged by a unique classroom context Astington, N. J. (2003). Sometimes necessary, never sufficient: False-belief under-
where the teachers stimulated children’s language mediation standing and social competence. In B. Repacholi, & V. Slaughter (Eds.), Individual
differences in theory of mind. Macquarie monographs in cognitive science (pp.
and peers’ backup. In addition, Naomi, the L1-Hebrew-speaking
12–38). Hove, E. Sussex: Psychology Press.
teacher, made use of the children’s bilingual competencies as a Aukrust, V. G. (2001). Talk focused talk in preschools – Culturally formed socializa-
resource to negotiate understanding of what was spoken in Arabic. tion for talk. First Language, 21, 57–82.
Schwartz, M., & Gorbatt, N. (2016). Ẅhy do we know Hebrew and they do not
In this way, she promoted their pragmatic awareness and social
know Arabic?C̈hildren’s meta-linguistic talk in bilingual preschool. International
skills. Finally, the teachers provided opportunities for intergroup Journal of Bilingual Education and Bilingualism, 19(6), 668–688.
interactions by encouraging play activities and communication Schwartz, M., Taha, H., Assad, H., Khamaisi, F., & Eviatar, Z. (2016). The role of emer-
between the Hebrew- and Arabic-speaking children. gent bilingualism in the development of morphological awareness in Arabic and
Hebrew, Journal of Speech. Language, and Hearing Research, 59, 797–809.
Moreover, the study highlighted a link between early develop- Schwartz, M., & Gorbatt, N. (2018). The role of language experts in novices’ language
ment of language awareness in bilingual classroom environment acquisition and socialization: Insights from Arabic-Hebrew speaking preschool
and children’s individual characteristics. Previously, a few studies in Israel. In M. Schwartz (Ed.), Preschool Bilingual Education: Agency in Interac-
tions between Children, Teachers, and Parents (pp. Series Multilingual Education..
looking at Spanish–English bilingual education in the United States Dordrecht, Netherlands: Springer.
suggested that social-personality characteristics play some role Barac, R., & Bialystok, E. (2012). Multilingual effects on cognitive and linguistic
in determining individual differences in L2 progress (e.g., Strong, development: Role of language, cultural background, and education. Child Devel-
opment, 83, 413–422.
1983; Wong Fillmore, 1983). This study added to the above data by Barac, R., Bialystok, E., Castro, D. C., & Sanchez, M. (2014). The cognitive develop-
suggesting a link between personality characteristics, such as out- ment of young dual language learners: A critical review. Early Childhood Research
goingness and gregariousness, and verbal and nonverbal behavior Quarterly, 29, 699–714.
Barley, R., & Bath, C. (2014). The importance of familiarization when doing research
expressing language awareness. We propose to view this link as
with young children. Ethnography and Education, 9(2), 182–195.
nonlinear but rather mediated by children’s social goals. As pre- Bergroth, M., & Palviainen, Å. (2017). Bilingual children as policy agents: Language
sented above, two girls, Rima and Maria, showing outgoingness and policy and education policy in minority language medium Early Childhood Edu-
cation and Care. Multilingua, 36(4), 359–375.
gregariousness, actively sought to establish friendships and social
Björk-Willén, P. (2016). Peer collaboration in front of two alphabet charts. In M.
positioning within the class group, and were more involved in nego- Theobald (Ed.), Sociological studies of children and youth: Vol. 21. Friendship
tiating meanings in L2 and in reflecting on language use, than their and peer culture in multilingual settings (pp. 143–169). Bingley, UK: Emerald
peers. Group.Brojde.
Blommaert, J., & Jie, D. (2010). Ethnographic fieldwork: A beginner’s guide. Bristol, UK:
Finally, the study can be beneficial for language teachers’ Multilingual Matters.
understanding of how they can support the children’s language Blum-Kulka, S., & Gorbatt, N. (2014). “Say princess”: The challenges and affordances
awareness development through encouraging peer mediation in of young Hebrew L2 novices’ interaction with their peers. In A. Cekaite, S. Blum-
Kulka, V. Grøver, & E. Teubal (Eds.), Children’s peer talk: Learning from each other
the bilingual classroom. Nonetheless, we are aware that, since our (pp. 169–193). Cambridge, UK: Cambridge University Press.
investigation was limited to a contextually embedded case study Blum-Kulka, S., Huck-Taglicht, D., & Avni, H. (2004). The social and discursive spec-
of a single classroom, our findings might not be fully generalizable. trum of peer talk. Thematic issue of Discourse Studies: Peer Talk and Pragmatic
Development, 6(3), 307–328.
Still, it is reasonable to assume that research exploring bilingual Brojde, C., Ahmeda, S., & Colunga, E. (2012). Bilingual and monolingual children
children’s language awareness should have, as a foreground, the attend to different cues when learning new words. Frontiers in Psychology, 3,
social context as a necessary condition for its development. Social 155.
Broner, M., & Tarone, E. (2001). Is it fun? Language play in a fifth-grade Spanish
goals at early ages such as establishing friendships and social posi-
immersion classroom. The Modern Language Journal, 85(3), 364–379.
tioning are essential for children’s well-being, and language “is used Bruck, M., & Genesee, F. (1995). Phonological awareness in young second language
as one means to achieve these social goals” (Philp & Duchene, 2008; learners. Journal of Child Language, 22, 307–324.
Byers-Heinlein, K., Chen, K. H., & Xu, F. (2014). Surmounting the Tower of Babel:
p. 84). Our analysis of classroom interactions over two years sup-
Monolingual and bilingual 2-year-olds’ understanding of the nature of foreign
ports the claim that the language classroom can be viewed as a language words. Journal of Experimental Child Psychology, 119, 87–100.
“double opportunity space” permitting both acquisition of social Cekaite, A. (2017). What makes a child a good language learner? Interactional com-
skills and development of language skills (Björk-Willén, 2016; petence, identity, and immersion in a Swedish classroom. Annual Review of
Applied Linguistics, 37, 45–61.
Cekaite, 2017; Philp & Duchene, 2008). To close, we recommend Cekaite, A., & Aronsson, K. (2005). Language play, a collaborative resource in chil-
future research for insight into how language classroom interac- dren’s L2 learning. Applied Linguistics, 26, 169–191.
tions, as well as children’s social goals and individual differences, Cekaite, A., & Aronsson, K. (2014). Language play, peer group improvisations, and L2
learning. In A. Cekaite, S. Blum-Kulka, V. Grøver, & E. Teubal (Eds.), Peer talk and
can promote their language awareness skills. peer learning in first and second language (pp. 193–213). Cambridge: Cambridge
University Press.
Cheung, H., Mak, W. Y., Luo, X., & Xiao, W. (2010). Sociolinguistic awareness and
Funding false belief in young Cantonese learners of English. Journal of Experimental Child
Psychology, 107, 188–194.
Church, B., Ayman-Nolley, S., & Mahootian, S. (2004). The role of gesture in bilin-
This work was supported by: (1) the Kindermissionswerk ’Die gual education: Does gesture enhance Learning? International Journal of Bilingual
Sternsinger’ e.V. and Abrahamszelt e.V. through Hand in Hand: Cen- Education and Bilingualism, 7(4), 303–319.
Crump, A. (2014). But your face, it looks like you’re English”: LangCrit and the
ter for Jewish-Arab Education in Israel [Grant No. D 17 2321 001/1];
experiences of Japanese-Canadian children in Montreal (Doctoral dissertation).
(2) The MOFET Institute, the Israeli national inter-collegial center Montreal, Quebec, Canada: McGill University.
for the research and development of programs in teacher education Cruz-Ferreira, M. (2006). Three is a crowd? Acquiring Portuguese in a trilingual envi-
ronment. Multilingual Matters: Clevedon, Avon.
and teaching in the colleges.
De Houwer, A. (1990). Bilingual first language acquisition. Bristol: Multilingual Mat-
ters.
De Houwer, A. (2017). Early multilingualism and language awareness. In J. Cenoz, D.
References Gorter, & S. Mey (Eds.), Language awareness and multilingualism. Encyclopedia of
language and education (pp. 88–97). Cham, Switzerland: Springer.
Ahearn, L. M. (2001). Language and agency. Annual Reviews of Anthropology, 30(1), Duff, P. (2014). Case study research on language use and learning. Annual Review of
109–137. Applied Linguistics, 34, 233–255.
 M. Schwartz et al. / Linguistics and Education 52 (2019) 13–23 23

Flavell, J. H. (2000). Development of children’s knowledge about the mental world. Palinkas, L. A., Horwitz, S. M., Green, C. A., Wisdom, J. P., Duan, N., & Hoagwood,
International Journal of Behavioral Development, 24, 15–23. K. (2015). Purposeful sampling for qualitative data collection and analysis in
Flavell, J. H. (2004). Theory-of-mind development: Retrospect and prospect. Merrill- mixed method implementation research. Administration and Policy in Mental
Palmer Quarterly, 50, 274–290. Health Services Research, 42(5), 533–544.
García, O. (2009). Bilingual education in the 21st century. West Sussex: Wiley- Philp, J., & Duchene, S. (2008). When the gate opens: The interaction between social
Blackwell. and linguistic goals in child second language development. In J. Philp, R. Oliver,
Goetz, P. (2003). The effects of bilingualism on theory of mind development. Bilin- & A. Mackey (Eds.), Second language acquisition and the younger learner: Child’s
gualism: Language and Cognition, 6, 1–15. play? (pp. 83–104). Amsterdam, the Netherlands: John Benjamins.
Gorbatt-Brodstein, N. (2012). Between two languages: Language, society and culture Pontier, R., & Gort, M. (2016). Coordinated translanguaging pedagogy as distributed
in natural discourse between young immigrants and their peers in preschool (Ph.D. cognition: A case study of two dual language bilingual education preschool
dissertation). Hebrew University. co-teachers’ languaging practices during shared book readings. International
Hamo, M., & Blum-Kulka, S. (2007). Apprenticeship in conversation and culture: Multilingual Research Journal, 10(2), 89–106.
Emerging sociability in preschool peer talk. In J. Valsiner, & A. Rosa (Eds.), The Lyster, R. (2007). Learning and teaching languages through content: A counterbalanced
Cambridge handbook of social-cultural psychology (pp. 423–444). Cambridge, Eng- approach. Amsterdam/Philadelphia: John Benjamins Publishing Company.
land: Cambridge University Press. Shenton, A. K. (2004). Strategies for ensuring trustworthiness in qualitative research
Kuo, L.-J., & Anderson, R. C. (2006). Morphological awareness and learning to read: projects. Education for Information, 22, 63–75.
A cross-language perspective. Educational Psychologist, 41(3), 161–180. Strong, M. (1983). Social styles and the second language acquisition of Spanish-
Lantolf, J. P., & Genung, P. (2002). ‘I’d rather switch than fight’: An activity theoretic speaking kindergartners. TESOL Quarterly, 17(2), 241–258.
study of power, success and failure in a foreign language classroom. In C. Kram- van Lier, L. (2011). Language learning: An ecological – semiotic approach. In E. Hinkel
sch (Ed.), Language acquisition and language socialization: Ecological perspectives (Ed.), Handbook of second language teaching and learning (Second Edition) (p (pp.
(pp. 175–196). London, UK: Continuum Press. 383–394). Mahwah, NJ: Erlbaum.
Lantolf, J. P., & Thorne, S. F. (2006). Sociocultural theory and the sociogenesis of second Verschueren, J. (2000). Notes on the role of metapragmatic awareness in language
language development. New York: Oxford University Press. use. Pragmatics, 10(4), 439–456.
Lauchlan, F., Parisi, M., & Fadda, R. (2013). Bilingualism in Sardinia and Scotland: Vygotsky, L. S. (1962). Thought and language. Cambridge, MA: MIT Press.
Exploring the cognitive benefits of speaking a ’minority’ language. International Vygotsky, L. S. (1987). Thinking and speech. The collected works of L.S. Vygotsky. pp.
Journal of Bilingualism, 17(1), 43–56. 39–285. (1).
Lugossy, R. (2018). Whose challenge is it? Learners and teachers of English in Wimmer, H., & Perner, J. (1983). Beliefs about beliefs representation and constrain-
Hungarian preschool contexts. In Schwartz (Ed.), Preschool bilingual education: ing function of wrong beliefs in young children’s understanding of deception.
Agency in interactions between children, teachers, and parent. Series multilingual Cognition, 13, 103–128.
education (pp. 99–131). Dordrecht, Netherlands: Springer. Wong-Fillmore, L. (1976). The second time around: Cognitive and social strategies in
Morales, P. Z., & Rumenapp, J. C. (2016/2017). Talking about language in pre-school: second language acquisition (Ph.D. dissertation). Stanford University.
The use of video-stimulated recall with emergent bilingual children. Journal of Wong Fillmore, L. (1983). The language learner as an individual: Implications of
Multilingual Education Research, 7, 19–42. research on individual differences for the ESL teacher. In M. Clarke, & J. Hand-
Morse, J. M., & Niehaus, L. (2009). Mixed method design: Principles and procedures. scombe (Eds.), On TESOL ’82: Pacific perspectives on language learning and teaching
Walnut Creek, CA: Left Coast Press. (pp. 157–173). Washington, DC: TESOL.
Ohta, A. S. (2001). Second language acquisition processes in the classroom setting:
Learning Japanese. Mahwah, NJ: Lawrence Erlbaum.