Eufracio 1
Nancy Eufracio
Dr. Pate
13 November 2014
Bilingual teachers' language strategies: The case of an Arabic–Hebrew kindergarten in Israel
The goal of this study was to investigate the language-teaching strategies used in a
bilingual Arabic–Hebrew kindergarten in Israel. We used an ethnographic approach by applying
a mixed methods design. The results demonstrate that the language-teaching strategy most
frequently used by teachers was flexible bilingualism, through translanguaging that involved
code-switching. This is in contrast to traditional instruction using language separation. In the
teachers' opinion, translanguaging enables bilingual children to learn their second language
efficiently, especially since Arabic is a socially weaker language in Israel, and it encourages
children's interactive involvement in the kindergarten.
Keywords
Bilingual Arabic–Hebrew kindergarten;
Bilingual teaching;
Language strategies;
Flexible bilingualism;
Translanguaging
Eufracio 2
The issue of bilingual education has occupied researchers, policymakers, and educators
for some decades, and a variety of bilingual teaching strategies have been examined. Recent
research calls for rethinking and reevaluation of constructions of traditional pedagogy, e.g.,
usage of translation, and language separation and negation of flexible transfer between languages
(Creese and Blackledge, 2010 and Cummins, 2005). This study aimed to investigate languageteaching strategies which are applied in a bilingual Arabic–Hebrew kindergarten in Israel and
challenge the term “parallel monolingualism” (Heller, 1999) by stressing flexibility and no
language separation as main principles of their bilingual pedagogy. To reach this aim, we used
linguistic ethnography as a methodological framework including field-notes, video-recorded
observations, the documents pertaining to kindergarten and semi-structured interviews with
teachers.
Bilingual education around the world is based on several different models. Most of these
indicate that languages should be strictly segregated in learning. Thus, parallel monolingualism
was traditionally considered as a positive means of acquiring a new linguistic system. This
approach is in keeping with the rationale of the two-way system used in the USA, in which half
of the students are L1 speakers of English and half are L1 speakers of another language, such as
Spanish (Freeman, 2007). In this system, all the children receive instruction in both languages,
but separately, using the “language-time” strategy. The aim of this strategy is to produce highlevel bilingualism in children from English-speaking families and children who speak a minority
language at home. In the two-way system, a mixture of languages or code-switching in the
classroom is rarely endorsed. It is stressed that languages should be kept discrete and separate; in
this case, bilingualism is accepted as double monolingualism. This pedagogy of language
Eufracio 3
separation has been criticized in current ethnographic studies on the application of language-time
strategy in bilingual preschools and kindergartens in the USA (see Lee, 2007).
Recently introduced in the research literature was an interactional approach to the
teaching of two languages (Arthur & Martin, 2006). This approach allows code-switching,
claiming that flexible transfer between languages is necessary for effective learning. In an
educational context, code-switching is defined as the practice of switching between a first and a
second language or discourse (Coffey, 2009). In this context, it was claimed that for bilinguals
and multilinguals, languages are not discrete, but form an integrated language system (García,
2009a). The following section presents the main language strategies for teaching bilingual
children.
2.2. Strategies of language learning in bilingual and monolingual education of bilingual
children
Based on the research analysis performed for the purposes of this study, we distinguished
four main groups of language-teaching strategies for young bilingual children that were
implemented in the framework of bilingual and monolingual education: (1) bilingual resource
strategies; (2) metalinguistic strategies; (3) nonlinguistic strategies, i.e., gestures and (4)
translanguaging (e.g., Creese & Backledge, 2010; Cummins, 2010; Kenner et al., 2008 and
Nicoladis, 2002).
2.2.1. Bilingual resources
Eufracio 4
At the same time, there is growing evidence that translation, as a dominant strategy, is
inefficient in bilingual education (Cummins, 2005 and Montague, 1997). Montague (1997)
claimed that translation as a main strategy leads to passive waiting for translation instead of
active involvement in L2 learning. Thus, there is an increasing call to critically address an
overreliance on translation as a bilingual resource.
Providing bilingual resources such as parallel versions of storybooks and poems in L1
and L2 and creating dual language multimedia books and projects was also found to be an
effective language-teaching strategy ( Cummins, 2010 and Kenner et al., 2008). This strategy
facilitates children's understanding of the content, and enables them to compare meanings that
were expressed in the different languages. In addition, Ernst-Slavit and Mulhern (2003) found
that the presence of books in the classroom in children's L1 conveys a clear message to the
children about the value ascribed to minority languages, scripts and cultures.
2.2.2. Metalinguistic strategies
The second group includes metalinguistic strategies. The goal of metalinguistic strategies
is to develop metalinguistic awareness, which is defined by psycholinguists as a person's explicit
knowledge about language, knowledge that can be brought into awareness, verbally reported and
declaratively presented (Bialystok, 2001 and Bruck and Genesee, 1995). Concerning bilingual
education, a growing number of studies have shown that this education can provide a basis for
progress in young bilingual children's cognitive and linguistic development (Cummins, 2000).
Eufracio 5
Several studies suggest that, at least at an early age, bilingual children appear to be more
sensitive to the structure of language, in both L1 and L2, than monolingual children (Bialystok,
2001). For example, Kenner et al. (2008) found that the children within a community class
strengthened their explicit knowledge of how language works. They discussed differences
between language structures in Bangla and English in regard to the presence and absence of the
definite article in these languages.
In the present study, the focus was on the metalinguistic strategies which involve issues
such as awareness of cognates and association between words. Cognates are words in different
languages that share a common historical origin (Whitely, 2002). Cognates are defined as
vocabulary items in two different languages that are similar both “orthographically and
semantically” (August, Carlo, Dressler, & Snow, 2005). For example, the English word advance,
the Spanish word avance, and the French word avancer. Research studies provide evidence of a
positive cross-linguistic transfer between L1 and L2 through the use of cognates. Nagy, García,
Durgunoglu, and Hancin-Bhatt (1993) found that bilingual English–Spanish students made use
of cognate relationships in their English reading. L1 Spanish-speaking students recognized many
cognates with English and had an advantage in English vocabulary recognition, but they often
required explicit instructions to optimize transfer for comprehension. The findings led the
researchers to conclude that students had an emergent concept of cognates, and used cognate
strategies in their reading. However, García (1991) noted that cognate strategies have constraints.
For example, it appeared that some aspects of word knowledge are understood only through
experience. Furthermore, it was stressed that a positive transfer of vocabulary knowledge can
occur if languages involve similar origins. In our case, Hebrew and Arabic share a large number
Eufracio 6
of cognates which are unambiguous in their meaning since they represent concrete objects, e.g.,
animals, vegetables. This is a reason to believe that cognate awareness facilitates vocabulary
learning for Arabic and Hebrew speakers.
2.2.3. Nonlinguistic strategies
The third group of strategies addresses nonlinguistic approaches. It was suggested that
teachers use nonlinguistic strategies such as spontaneous hand and arm gestures together with
their speech to foster L2 understanding among bilingual children during classroom instruction
(Zukow-Goldring, Romo, & Duncan, 1994). Researchers have found that gestures may improve
speech recall and enhance the learning process (Church, Ayman-Nolley, & Mahootian, 2004). In
addition, the use of signs and gestures appears to facilitate speech articulation (Alibali & Nathan,
2007). Finally, concerning bilingual education, Hadar, Teitelman, and Dar (2001) claimed that
L1 speakers tend to use more gestures in their L2 speech.
Teachers face the challenge of helping students to identify connections between ideas and
objects. In this context, research shows that children attain conceptual and essential information
through gestures, which can help them to understand obscure sentences (Kelly & Church, 1998).
It was also found that teachers who used gestures in their instruction, and the information they
expressed by means of gestures, had a positive influence on learning, by helping children to
understand the speech more quickly (Church et al., 2004).
Eufracio 7
There are three basic types of gestures (Nicoladis, 2002). The first type is pointing
gestures, which direct the child's attention to the referent. This type of gesture is defined as the
simultaneous use of the arm and index finger pointing toward an object. The second type of
gesture is conventional gestures, which are those body movements or limited actions that have a
firmly established and agreed-upon meaning. The third type is iconic gestures, which use the
shape or the motion of the hands and arms to mime a key aspect of the referent and to produce its
dynamic visual representation; for example, using the hands to indicate the size of the referent
( Bernardis & Caramelli, 2007).
2.2.4. Translanguaging
Bilingual education programs all over the world traditionally suggested implementation
of separate teaching of languages. However, some researches calls for rethinking this separation
by appealing that keeping the languages separate (“two solitudes”, Cummins, 2008) is argued to
be the continuing prevalent monolingual learning strategy, as a way to ensure students'
understanding of each individual language (Cummins, 2005 and Cummins, 2008). Instead of
“language-time” strategy, in which each language is instructed in turn, a new, alternative and
innovative approach was suggested to challenge the separated bilingual resources. According to
this approach, L1 and L2 may be used concurrently as realization of truly bilingual pedagogy
(Arthur and Martin, 2006, Creese and Blackledge, 2010 and Cummins, 2005). This approach is
called translanguaging, involves code-switching and is expressed by teaching two or more
languages in parallel (Creese & Blackledge, 2010).
Eufracio 8
New pedagogies and approaches have consolidated academic interest around the term
translanguaging (Greese & Backledge, 2010). Williams (1996) used it to describe a bilingual
pedagogy that alternates language modes such as reading in one language but writing in another
during the bilingual classes. Moreover, García (2009a) expanded this term to all bilingual
practices including literacy and transfer between diverse languages and contexts. Studies such as
those by García, 2009a and Creese and Blackledge, 2010 have adopted this term to describe the
normal practice in a bilingual environment of 'bilingualism without diglossic functional
separation' or flexible bilingualism that views language as fluid and changing with its flexible
movement across and between ‘languages’. However, proactive usage of translanguaging raises
various theoretical and practical questions that have received inadequate attention to date, such
as the pedagogical implications and benefits of this strategy (Canagarajah, 2011). The purpose of
this study, therefore, was to take a closer look at translanguaging as an innovative bilingual
strategy as well as to examine other bilingual strategies that are applied within the framework of
an Arabic–Hebrew kindergarten and have been found to be effective by its bilingual teachers.
3. Background of the present study
3.1. A brief description of the Israeli socio-linguistic context
Israel is a multilingual and multicultural country, with both Hebrew and Arabic as official
languages. Arabic was recognized as the second official language in Israel by force of legislation
in 1948, but is not a competing partner in a dyadic bilingual state (Lambert, 1999). Hebrew is the
Eufracio 9
dominant language in most life domains and is presented as the dominant language in Israeli
public spheres. Arabic is a minority language and is the national language for more than one
million Arab–Israeli citizens, who comprise one fifth of the population. Hebrew is studied as a
second language in Arab schools and is part of their curriculum from third to twelfth grade.
Israel has separate Hebrew-speaking and Arabic-speaking education systems and therefore, Arab
and Jewish children are educated in different schools (Mor-Sommerfeld, Azaiza, & HertzLazarowitz, 2007).
Most Arab–Israelis understand and speak Hebrew, and use it at work and in other
settings. According to Amara (2002b), in Israel, Arab citizens choose the strategy of linguistic
integration over assimilation. They aim to acquire high socio-linguistic ability in Hebrew, which
facilitates their functioning in the social network of the majority. Alongside Hebrew acquisition,
Arab citizens still preserve Arabic as their mother tongue to maintain their identity. At the same
time, Arabic is scarcely presented in the landscape of cities with a Jewish population and the
level of competence of the Jewish children in Arabic is relatively low, even though the study of
Arabic is obligatory in the Jewish secular-school curriculum from 5th grade (Amara, 2002a).
3.2. General information about Arabic–Hebrew bilingual education in Israel
In 1997, the Center for Bilingual Education initiated two bilingual schools, the first in
Jerusalem and the second in the Western Galilee. Later in 2004, the Center for Bilingual
Education established a third school in the Arab city of Kfar Kara'a, where the kindergarten was
an integral part. The Center for Bilingual Education believes that both language groups should be
Eufracio 10
represented equally in schools on all levels. From the outset, they have been consistent in
choosing a teaching and management staff that represents both groups equally, with each class
having two homeroom teachers—one Arab and one Jewish. The teachers share the educational
tasks, responsibilities and teaching in each class. Teaching in these educational settings is the
individual choice of the teachers. It is also important to mention that the Ministry of Education
supports these bilingual schools.
Concerning biculturalism as an integral target of this educational system, no bicultural
approach exists that is separate from the socio-political context. Therefore, each context has its
unique challenges. A potential problem facing the Center for Bilingual Education is determining
cultural and identity issues that should be dealt with at the school, and the exposure of both
groups to historical interpretations of the conflict.
Regarding language policy, these settings aimed to challenge the segregated monolingual
and monocultural education system in Israel and adopted the model of cultural bilingualism
(Mor-Sommerfeld, 2005). Bilingual education challenges teachers to provide practical strategies.
It advises the teaching of vocabulary and encouraging students' language-learning strategies. Up
to now, no study has investigated the language-teaching strategies which are applied in
kindergartens. This study begins to fill this void. We examined a bilingual Arabic–Hebrew
kindergarten in Kfar Kara'a that applied translanguaging as a main principle of its bilingual
pedagogy. We also addressed how this principle works, as well as certain challenges in its
realization. In addition, this study focused on other language-learning strategies (e.g., concept
transfer, cognates, translation and gestures) and their rationales. The findings were compared to
Eufracio 11
strategies mentioned in the research literature dealing with bilingual or monolingual elementary
education for bilingual children.
3.3. Research questions
The study was designed to examine the following questions:
(1)
What are the bilingual language-learning strategies which are applied in the
kindergarten?
(2)
What is the place of translanguaging in the teachers' bilingual pedagogy?
(3)
How do the teachers reflect on their bilingual pedagogical experience, expectations,
perceptions and language policies relevant to their bilingual strategies?
4. Method
4.1. Setting
Education has been acknowledged as a main field for linguistic ethnographic research
(Maybin & Tusting, 2011). In this context, we used a linguistic ethnography as a methodological
framework of the study given that it allows “making linkages between language, culture, society
Eufracio 12
and cognition in complex ways” (Creese, 2008, p. 232). As will be presented in the following
sections, a combination of ethnographic (observations, field-notes) and linguistic (analysis of
language use) methodologies permitted us to address language teaching not as a separate
discipline but as viewed from the social contexts in which it exists (Creese, 2008; Rampton et al.,
2004). This combination helped us understand the teachers' perspectives on their language
strategies while also thoroughly analyzing their behavior (Maybin & Tusting, 2011).
Our focus was on a bilingual kindergarten which was attended by Arab and Jewish
children, and aimed to promote L2 (Hebrew/Arabic) acquisition and L1 (Arabic/Hebrew)
development. Beyond bilingualism as an objective, this kindergarten as well as a bilingual school
in Kfar Kara'a which included the preschool setting as its integral part, aimed to help the Arab
and Jewish children develop a high level of mutual tolerance, respect and acknowledgment from
early childhood. Accordingly, the challenge in realizing this aim is to take into consideration that
these children belong to two groups who have a longstanding history of intolerance.
In addition, the kindergarten curriculum provides the children with a meeting of the two
cultures. During teacher–children circle meetings and communications, the children become
familiar with customs and cultural traditions of a second ethnic group. It is notable in their
interviews that the teachers stressed that the elaboration and implementation of curriculum which
equally emphasizes the narratives of both cultures is a real challenge. Beyond the coordination of
the day-by-day instruction, the implementation of such a curriculum demands from the teachers'
significant preparation, reciprocal confidence and the ability to work together and to cooperate.
Eufracio 13
4.2. Participants
The study participants were three kindergarten teachers: two native Hebrew-speaking
teachers, Aviva and Orit, at different time periods and one native Arabic-speaking teacher,
Sokaina. All teachers expressed their willingness to participate in the study. The teachers started
working in the kindergartens of their own initiative and for ideological reasons. From the
kindergarten establishment in 2004, Aviva, the Hebrew-speaking teacher, and Sokaina, the
Arabic-speaking teacher, began their pedagogical bilingual process using the language separation
approach, in which each teacher spoke her native language. After the first year, they realized that
this approach to bilingual instruction was not working. They changed the method of instruction,
and both teachers began to speak in both languages, without setting aside time for each
individual language. This led to a combination between the languages, i.e., the application of
translanguaging—a strategy that they believed “works.” Note that this modification resulted not
from their professional supervision and training but from the teachers' daily experience, without
previous theoretical knowledge.
Together with Sokaina, Orit, the Hebrew-speaking teacher who replaced Aviva, has
continued the guidelines that were developed before she came. We will address the teachers'
reflections on their bilingual teaching in the Results and Discussion section. The following
passage will relate to each teacher and her role in the building of the linguistic policies of the
kindergarten.
Eufracio 14
4.2.1. Aviva
Our first five video-recorded observations were conducted with the first Hebrewspeaking teacher, Aviva, from April 2011 to June 2012, who then left the position at the end of
the academic year. Aviva had more than 16 years of professional experience. She had been
working in the kindergarten since its establishment, for six years. For Aviva it was critical to
speak Arabic as well as to show Jewish children that she is a model of someone who learned and
continues to learn Arabic every day. In this continuous process of learning Arabic, Aviva
emphasized the role of peer learning and often turned to the Arabic-speaking children for help
and to be her teachers. She put a lot of effort into raising the motivation of the Jewish children to
learn Arabic. Aviva also was very decisive about the need to create changes in the linguistic
policies of the kindergarten. She was the driving force of the discussion about the effectiveness
of the separation between the languages which existed as the foundation of the teaching method
since the establishment of the kindergarten.
4.2.2. Orit
With Orit, the Hebrew-speaking teacher who replaced Aviva, the next five videorecorded observations were performed from September 2012 to November 2012. Orit had more
than 10 years of professional experience; similar to Aviva, Orit came to the kindergarten of her
own initiative after gaining experience as a special needs teacher in a different school and
working in a bilingual school as a homeroom teacher in first and second grade. Together with
Eufracio 15
Sokaina, Orit has continued the guidelines that were developed before she came. Orit lives in a
Jewish settlement next to Kfar Kara'a.
4.2.3. Sokaina
Sokaina is an Arabic-speaking teacher living nearby the kindergarten. She had more than
10 years of professional experience and spoke fluent Hebrew. Together with Aviva, Sokaina was
the first teacher in the kindergarten since its establishment in 2004. Sokaina specialized in the
didactic development of acquiring languages such as using songs and gestures. She supported the
approach of “the earlier the better” in acquiring a second language. After working in an
elementary school, it was very important to Sokaina to work with preschoolers. In her opinion,
the framework of preschool and the atmosphere in kindergarten allows for more speaking
between Jewish and Arabic children and is therefore very positive in acquiring languages. After
changing Hebrew-speaking teachers, it seemed liked Sokaina as the more experienced teacher,
took a more dominate role in the kindergarten and helped Orit acclimatize to the kindergarten.
4.3. Procedure
Field-notes were taken during each visit to the kindergarten throughout the entire
research period. Ten planned video-recorded observation sessions were conducted in the
kindergarten twice a month from April 2011 to November 2011 (five observations up to the end
of the academic year in the end of June, and five observations from the beginning of the new
academic year, from September to November 2011).
Eufracio 16
All of the teacher observations were performed during teacher–child circle time. The
activities and communication were typically dedicated to a specific topic (e.g., religious festivals,
seasons and weather, family) and the children actively interacted during the presentation and
discussion. We decided to conduct only two video recordings per month to avoid disturbance of
the daily routine in the kindergarten. Still, it was important to obtain data from a large number of
observations at a regular time interval in order to investigate a relative degree of regularity of the
observed teachers' language practices. Note also, that to reduce the “observer's paradox” (Labov,
1972) and to collect data as unobtrusively as possible, we started our video recording only after
three months after our first visit to the kindergarten in January 2011. Thus, this initial period of
familiarity with the setting and teachers, was aimed to make the teachers and children more
accustomed to our presence. Each observation lasted approximately one hour and was digitally
videoed during teacher–children circle meetings and communications. The teachers were the
focus of the observations, which supplied information about instructional strategies and informal
linguistic and social interaction between teachers and children. The interviews with the teachers
were conducted individually with each teacher in the kindergartens at a convenient time for the
teachers. Each interview lasted approximately 60 min and was tape-recorded with the
interviewees' consent.
4.4. Instrumentation
The methodology included field-notes, video-recorded observations and semi-structured
interviews with the teachers. We received permission for video recordings in the kindergarten
Eufracio 17
from the Israeli Ministry of Education. In addition, the documents pertaining to kindergarten
(e.g., teachers' journals and other materials) were collected.
In addition to the qualitative analysis of audio and video recordings of the lessons, we
analyzed the frequency of the observed strategies. Frequency analysis was important to provide a
panoramic view of the situation under study (Marsland, Wilson, Abeyasekera, & Kleih, 1999).
As was noted by Mackey and Gass (2005), the quantification in the qualitative research, permits
more precise examination of phenomenon occurrence and then one may use these data to draw
inferences. In addition, the frequency analysis was necessary to pinpoint the teachers' regularly
used versus rarely used strategies, and as a result, we obtained a deeper understanding of the
rationale behind their choice. In addition, given the fact that observation and interview data are
mainly categorical, the frequency analysis was used as an appropriate way to divide the types of
variables, as this prevented the loss of information, which might have resulted from the division
into categories (Mackey & Gass, 2005). The results of this analysis were then interpreted and
discussed in light of the teachers' reflections on their teaching and with regard to the broader
socio-linguistic context of the study.
4.4.1. Field-notes and video-recorded observations
During field-notes we wrote detailed impressions of the teachers' linguistic behavior
during their instructions as well as spontaneous communication with the children. Observation
protocols were used while videoing the observations for the initial analysis of the observed data,
which included the following categories: (1) time of each activity (in seconds), (2) type of
Eufracio 18
activity, (3) target audience (who is involved in each activity) and (4) resources or tools of
instruction. The video-recoding of the observations permitted us a detailed analysis of the
language-teaching strategies.
4.4.2. Semi-structured interviews with the teachers
The authors conducted in-depth semi-structured interviews with three teachers. Semistructured interviews can improve the quality of generated data due to their flexible and open
nature, which is more suited to the particular local environment. The primary reason for using
the interviews was to obtain the teachers' reflections on their bilingual pedagogical experience,
expectations, perceptions and language policies relevant to their bilingual strategies. As the study
focused on bilingual strategies used by the teachers, we asked about the rationale behind the
observed strategies, about how their classroom practices are linked to their language beliefs and
about the linguistic development of the Arab and Jewish children. The teachers were also asked
about their background, and their interaction with the parents of children in the kindergarten.
4.5. Data analysis
The methodological framework for the data analysis comes from linguistic ethnography.
This orientation offers descriptive and analytic procedures, which investigate social processes,
institutions and communities of practice and holds that language and social life are mutually
formative. Linguistic ethnography uses case study methodology to engage with everyday activity
issues. The interviews with the teachers were transcribed, coded and divided into different
Eufracio 19
thematic categories relevant to our analysis. They were analyzed using the standard procedures
for analysis of qualitative data (placing labels on themes and concepts that emerged from the
data, open-coding, building connections between categories to form larger, core categories, and
axial-coding) (Bogdan & Biklen 1992).
The transcribed videos of the observations were analyzed in two stages. First, all teachers'
verbal data and nonverbal material (gestures) were divided into units of meaning. Second, the
data were coded under categories of strategies on the basis of similarity between categories that
were repeated in the observations and were then discussed during the interviews. The frequency
of each observed behavior was calculated in the following way. The number of times each
strategy and gesture was observed during the entire length of the study were counted by
observation and then summarized to present a frequency of their usage, which is labeled
“frequency” in Table 1. The percentage was calculated by dividing each observed strategy's
frequency by the total number of all of the frequencies.
Table 1.
Frequency of the observed strategies and percentage of usage.
Eufracio 20
Observed Strategy
Frequency
Percentage of usage
Translanguaging: code-switching intra-sentence 198 33%
Translanguaging: code-switching inter-sentence 85 14.19%
Overall translanguaging 283 47.19%
Gestures 242 40.4%
Translation 35 5.84%
Cognates 25 4.17%
Double books 14 2.33%
Table options
Eufracio 21
5. Results and discussion
The study's results included two types of data analysis: qualitative and quantitative. The
qualitative method was applied to analyze the semi-structured interviews with teachers, the
video-observations and field-notes on the language-teaching strategies, and the collected
linguistic resources and artifacts. The quantitative focused on the frequency of the observed
strategies. The analysis of the frequency of teachers' strategies has revealed the following order:
translanguaging, gestures, translation, cognates, double books. Table 1 provides the total
frequency of use for each strategy. The observed strategies will be presented sequentially based
on the frequency principle.
5.1. Translanguaging
Translanguaging was manifested in teachers' speech and dialogues with children as the
most frequently applied strategy. Two types of code-switching were observed and they are
analyzed in this section: inter-sentence code-switching and intra-sentence code-switching.
5.1.1. Inter-sentence code-switching
Inter-sentence code-switching was the first type, meaning switching languages between
sentences. This strategy was applied both during instruction time and during spontaneous
teacher–child interactions in the kindergarten. The following excerpt illustrates inter-sentence
code-switching during instruction time:
Eufracio 22
Sokaina (Arabic-speaking): (We hear the sounds to remember them better). (in Hebrew)
(We have a game, a memory game, right?).
Three main contexts of inter-sentence code-switching emerged from the data analysis.
The first context was related to the use of Arabic for common and frequent phrases (e.g.,
greetings, instructions and imperatives) and for semantic categories such as animals, numbers,
food, body parts and musical instruments. In a bilingual kindergarten context of English (L2)
acquisition in the US, these basic phrases were found to be acquired first (Tabors, 1997). It was
also found that adults used techniques such as definitions and semantic contexts to help children
understand the new words' meanings.
In their interviews, the teachers explained that this context of inter-sentence codeswitching was frequent and was aimed at promoting Arabic acquisition among Jewish children.
Even though both the Arab and Jewish teachers aimed to support bilingual development in both
target languages, the teachers were aware that the Arab children had more opportunities of
exposure to Hebrew in daily life. Therefore, their acquisition of Hebrew was more accelerated
and successful than their Jewish peers' progress in Arabic. In contrast, Hebrew-speaking children
are not exposed to Arabic at home, and Arabic is the language of the minority in Israel. In
Eufracio 23
addition to the unique and complex socio-political context in which Jews and Arabs live in
Israel, the minority language status of Arabic is one fact that might reduce Jewish children's
motivation to learn this language (Amara, 2002a). In the kindergarten under study, the goal of
using translanguaging as a main bilingual pedagogy strategy was to increase the Jewish
children's motivation for Arabic acquisition. This is illustrated by the following excerpt from a
Jewish teacher's interview:
Aviva (Hebrew-speaking): Arab families encourage and motivate the children to learn
more and more Hebrew, faster and faster, and using the correct pronunciation. Their homes are
full of Hebrew tapes, discs and books. Arabs are highly motivated to learn Hebrew. For them, it
is a source of pride. Arab children speak Hebrew at home; they get encouragement and
reinforcement from the environment, from siblings and parents. They (Arabs) ought to learn
good Hebrew at an early age. As they acquire better Hebrew, it is easier to fit into society and
university. Proficiency in Hebrew facilitates their life in Israel, but Jewish children do not have
to learn Arabic; it is just good if they know. It is different in Jewish homes, because the parents
have no knowledge of Arabic. They cannot enhance the children's learning, and they have no
vocabulary to enrich the children's language at home. If the child speaks to them in Arabic, they
don't understand him. There is no motivational factor in the Hebrew-speaking home, and the
adults are not motivated to learn Arabic.
Eufracio 24
In summary, it was clear to us that the teachers attempted to speak more Arabic with both
the Hebrew (L1) speaking children and the Arabic (L1) speaking children, especially commonly
used sentences, for two reasons: to advance and strengthen Arabic among the Hebrew (L1)
speaking children, and to preserve the Arabic (L1) of the Arabic-speaking children.
The second context in which inter-sentence code-switching was used was the switching
from Arabic to Hebrew to facilitate the understanding of (L1) Hebrew-speaking children and to
keep them attentive and active (“…and not to lose them”). Note that the teachers avoided direct
translation of the Arabic words into Hebrew. This is because they believed that the translating
strategy discourages children from learning the L2 (see the Translating strategy section).
Thus, instead of direct translation, the teachers clarified what was said in Arabic. The
following example illustrates the clarification strategy:
Sokaina (Arabic-speaking): (Let's get to know and remember the new children's names
and wish them a good morning using the greeting that we learned).
The third context of the inter-sentence code-switching was related to language games and
cognate usage. Teachers switched to L2 for direct language instruction, often accompanied by
songs and melodies.
Eufracio 25
5.1.2. Intra-sentence code-switching
The second type of code-switching was intra-sentence, which means switching between
languages within the sentence. The following example illustrates this strategy:
Aviva (Hebrew-speaking): (Because we were on vacation for a long time) (in Arabic) (a
long, long time) (in Hebrew) (who remembers, who remembers) (in Arabic) (How do we play
this game?)
Sokaina (Arabic-speaking):
(in Arabic and then in Hebrew) (You know and remember how to play this game)
This example illustrates flexible bilingualism. Both these cases of code-switching are
constructions, which characterize the bilingual context and are examples of translanguaging. The
teachers shared the instruction by following each other and not separating the languages or
providing translations.
In summary, the observations showed that translanguaging practice is a natural and
spontaneous phenomenon, which was used by the teachers as a language-switching tool. Note
Eufracio 26
also that this instructional strategy did not encourage translation. We can see that teachers
adopted this strategy from their personal experience of teaching; an experience that was related
to and corresponded with their pedagogical orientation. This strategy is also congruent with the
theoretical rationale of teaching for cross-linguistic transfer that is called translanguaging, and
challenges the “two solitudes” teaching approach (Cummins, 2008).
As reported above, traditionally, bilingual educators insisted on separating the languages
in teaching, as applied within the two-way bilingual model in the USA (Cummins, 2008). This
traditional approach views translanguaging as a threatening factor in language learning.
However, it is important to note that ethnographic research on diverse language communities
presents translanguaging as a flexible use of the entire linguistic and cultural resources in family
and community communication (e.g., Goodz, 1989). As for the educational context, the impact
of translanguaging on language learning as an innovative language-teaching strategy has scarcely
been examined to date (Creese & Blackledge, 2010). Consequently, the current data might
extend our understanding of this innovation. We found that the teachers in our target bilingual
kindergarten applied the language separation approach from the outset. The interviews clarified
that, from the beginning, the teachers were instructed to make a separation between the
languages and to maintain the one-teacher-one-language strategy. However, after six months of
keeping the languages separate (“two solitudes” assumption, Cummins, 2008), the teachers
realized that the Jewish children had made very minor progress in Arabic. Thus, the language
separation and application of the one-teacher-one-language instructional model resulted, in their
opinion, in the continuation of the prevalent monolingual learning strategy and a lack of
motivation to use Arabic among the Jewish children. During the interviewers, the teachers,
Eufracio 27
Aviva and Sokaina, described the inefficiency of this model and the necessity to rethink and
modify the initial one-teacher-one-language strategy existing approach. In sum, this modification
highlighted the teachers' flexibility and openness to a transformation of the traditional approach
of language separation in light of the challenges posed by the Arabic–Hebrew bilingual context.
5.2. Gestures
In this study, the analysis of the observations showed that using gestures was one of the
most dominant strategies (40.5% of the observed strategies). Teachers used three types of
gestures, pointing, conventional and iconic ones, to facilitate children's bilingual development. It
is noteworthy also that the pictures on the kindergarten walls also presented the application of
the gesture strategy. They contained images or pictures of conventional and accepted gestures as
a tool for language instruction and for teacher–child communication (see Appendix).
5.2.1. Pointing gestures
As reported above, the pointing gestures were used to direct the child's attention to the
referent. In the following extract, teachers used pointing and speaking to enhance the children's
vocabulary and to develop their understanding.
Aviva and Sokaina (Hebrew and Arabic-speaking) spoke simultaneously: (Good
morning to all the nice boys and girls, good, take a deep breath) [breathing gesture].
Eufracio 28
… (Good morning to the whole world) [they make a gesture of pointing to the world].
5.2.2. Conventional gestures
As presented previously, the conventional gestures are body movements that have a
firmly established and agreed-upon meaning. This example illustrates the way Orit used
conventional finger gestures that represent numbers to arrange the items she mentioned. Both
Arab and Jewish children easily understand the use of fingers for counting.
Orit (Hebrew-speaking): (The first thing) [she raises her finger] (Number one) (We
bring margarine), (in Arabic) (and in Arabic?).
5.2.3. Iconic gestures
Iconic gestures, as mentioned above, means using the shape or the motion of the hands
and arms to represent the referent. This strategy was the most frequently applied. For example,
Eufracio 29
during a lesson about planting grains, when the teachers mentioned the new word—bean—they
indicated its physical size by using the hands to show how big or small it is. In addition, the
iconic gestures were for discipline and to set limits for the children. Teachers also used iconic
gestures to represent abstract nouns and verbs, such as feelings. For example, Sokaina painted a
heart to show the children that she loves them, and taught them the abstract noun ‘love’ and the
verb ‘to love.’ In another example, Orit illustrated physical balance to teach children about the
similarity between Arabic and Hebrew cognates: (in Arabic, arnab, rabbit) and (in Hebrew,
arnav, rabbit) [she showed a weighing scale]. These examples demonstrate how the teachers' use
of the iconic gesture is closely related to speech. By illustrating what is being said by using hand
movements, the teachers used the iconic gestures differently from other gestures. Their aim was
to make tangible the physical characteristics of concrete items and to add details to the mental
images that the teachers were trying to convey.
It should also be noted that teachers used gestures mainly to translate words from Arabic
to Hebrew, or vice versa, and to teach new words and enrich the children's vocabulary. The
rationale behind the use of gestures and pointing is that this nonlinguistic strategy plays a
significant role in language development. In their interviews, the teachers evidenced the
importance of this strategy in bilingual instruction:
Sokaina (Arabic-speaking): We depend on signs and gestures as our main means of
communication, particularly movements, gestures, visual schedules and concrete objects of
reference to enhance learning. We encourage body language and use gestures to enhance the
Eufracio 30
children's understanding. Children learn words when they are paying attention to the object we
are naming for them. There are children who need gestures as a transitional stage in speech
development. I think that children need signs to help their understanding and speech perception.
To summarize, the interviews with teachers revealed that they believed that gestures
facilitate bilingual teaching. The teachers perceived this strategy as facilitating the early
bilinguals' rapid progress in the acquisition of Hebrew or Arabic vocabulary. It also accelerates
conceptual transfer from one language to another. Moreover, the teachers believed that visual
schedules and concrete objects of reference helped children to know what they were supposed to
be doing, to understand and recall words. These teachers' beliefs are in accordance with research
findings, pointing out the role of gestures and signing in the development of speech (Capone and
McGregor, 2004 and Goldin-Meadow and Morford, 1990).
The application of the gesture strategy brings additional evidence supporting the “speechgesture system” approach as mature ground for language acquisition. This is in accordance with
McNeill's (1992) claim that language and gestures must be examined together to unveil the
operations of the mind. Finally, we would like to note that the application of the multi-sensory
approach as a bilingual teaching strategy has not been addressed sufficiently in the current
research, and therefore, requires deeper examination in further research.
5.3. Translating strategy
Eufracio 31
How do teachers relate to the translating strategy? We observed that the teachers'
orientation was “no translation.” They pointed out that this strategy did not inspire children to
learn a second language, but encouraged them to become passive learners, as discussed above:
Aviva (Hebrew-speaking): …In a conversation, for example, Sokaina, the Arabic (L1)
speaking teacher, begins speaking in Arabic. I don't translate what she says, but in my answer to
her, I include the main idea. I believe that children understand from contexts, and learn to make
the connections. In this way, they learn better by listening than when the words are translated for
them.
Indeed, in the observed lessons, the translating strategy was used minimally (in 5.8% of
observed strategies). It should be noted that the teachers translated words and not sentences. The
strategy was used in only one context, when teaching new vocabulary in the two target
languages. Thus, in contrast to their initial orientation, the teachers avoided frequent application
of the translating strategy. This minimal application was not in line with Manyak's (2004)
approach, which suggested that translation is a powerful way of facilitating children's literacy
learning. At the same time, the avoidance of direct translation usage is in line with Montague's
(1997) statement that translation as a main strategy is inefficient in bilingual education and leads
to waiting patiently for translation instead of active involvement in L2 learning. This assumption
corresponded to the explanation by Cummins (2005), who argued that translation between first
and second languages has no place in teaching. Cummins also described this strategy as a
Eufracio 32
reversion to a discredited method. To sum up, in line with the theoretical assumption presented
above, our teachers were convinced in inefficiency of translation as the main strategy of
language instruction based on their reflective thinking about children's passive behavior when
this strategy has been applied. In our opinion, the efficiency of the translation should be
evaluated in light of specific social contexts in which it exists as we illustrate by the current
study.
5.4. Cognate usage
The teachers used Arabic and Hebrew cognates to stimulate the children's ability for
word-transfer between languages. In this context, the teachers focused on the metalinguistic
strategy that is based on highlighting similar roots of words in L1 and L2, and therefore, the
shared origin of the Hebrew and Arabic languages. Thereby, they tended to enhance L1–L2
association to accelerate acquisition of the target languages. Aviva illustrated the application of
this strategy:
Aviva (Hebrew-speaking): We have cards or pictures of words that sound similar in
Arabic and Hebrew: (in Arabic and Hebrew, home) (in Arabic and Hebrew, mother) (in Arabic
and Hebrew, father) (in Arabic and Hebrew, boy) (in Arabic and Hebrew, girl) (in Arabic and
Hebrew, dog) (in Arabic and Hebrew, giraffe). Also, we have cards containing words that sound
similar in Arabic and Hebrew but that have different semantic meanings, for example: the word
(soos) in Arabic is a chick, but (soos) in Hebrew, is a horse. There are language games of similar
Eufracio 33
and different-sounding words that charm the children. We also use parallel songs and finger
games that highlight the languages' similarity.
The cognate usage develops children's ability to remember these homophonic words
based on meaning ambiguity. Indeed, these strategies enable children to develop sensitivity to
the target languages, to compare them and think about them (Carlo et al., 2004, Cummins, 2008
and Cummins, 2009). Thus, it seems that this strategy may enhance the development of
metalinguistic awareness.
5.5. Double-book reading strategy
Teachers used bilingual books with Arabic and Hebrew texts within the same book. The
following excerpt illustrates the application of the double-book reading strategy where Sokaina
and Aviva together showed bilingual books to the children:
Aviva (Hebrew-speaking) wrote the name of the book on the floor in Arabic and then in
Hebrew: (Once upon a time, there was a boy who did not like to sleep alone).
Sokaina (Arabic-speaking):
Eufracio 34
(Which book is written in Arabic and which is written in Hebrew?).
A child (Arabic-speaking): (Because the letters are different).
Aviva (Hebrew-speaking):
(What is the difference between the books? You said this is a Hebrew book) (in Arabic)
(This is in Hebrew and this is in Arabic, how did you know?)
Sokaina (Arabic-speaking): (How do we distinguish between them?)
A child (Hebrew-speaking): (In one book the letters are joined and in the other, they are
not)
Another child (Arabic-speaking): (There are no dots)
Eufracio 35
These examples show cross-linguistic and cross-orthographic comparisons. The teachers
emphasized the different appearance of Hebrew and Arabic letters. Whereas Arabic letters are
joined, Hebrew letters are not. Another difference is in the degree of usage of diacritical marks
placed beneath or above the consonant letter and vowels; in the written Arabic, the usage of
diacritics is much more extensive than in written Hebrew. This phenomenon demonstrates the
orthographic complexity of written Arabic. The above examples also demonstrate that the use of
the double-book strategy helps to develop cross-linguistic and cross-orthographic awareness.
Sokaina explained the teachers' application of this strategy in the following way:
Sokaina (Arabic-speaking): We read bilingual books as effective tools for fostering
second-language and literacy development for Arab and Jewish children. I believe that, with a
variety of interactive formats, it provides an effective tool for raising the children's awareness of
diversity by exposing them to different scripts and languages.
Thus, in their interviews, the teachers stressed the need to provide bilingual resources,
such parallel versions of storybooks and poems as an effective language-learning strategy
(Kenner et al., 2008). In the same vein, Ernst-Slavit and Mulhern (2003) showed that the
presence of bilingual books in teaching is important, and teachers use this strategy to evaluate
other languages and cultures. In addition, bilingual books can play an important part in
supporting children's language proficiency and literacy learning. Finally, in our opinion, the use
Eufracio 36
of bilingual books in the context of the Arabic–Hebrew kindergarten in Israel stresses to children
the equal status of the target languages.
6. Conclusions
This study discussed teachers' bilingual pedagogy in the context of a bilingual Arabic–
Hebrew kindergarten. A methodological framework of linguistic ethnography permitted us to
explain the phenomena of language-teaching strategies in Arabic–Hebrew bilingual kindergarten
with regards to the Israeli socio-linguistic context, and in particular, the current status of the
Arabic language (Creese, 2008; Maybin & Tusting, 2011). On the local classroom level, our
focus was on direct observation and analysis of teachers' language strategies that were applied in
the bilingual kindergarten. On the wider social level, we tried to bridge between the teachers'
perceptions and reflections of their language strategies, namely their language policy, and the
educational, cultural and socio-linguistic context of Arabic–Hebrew bilingual education in Israel.
The following main conclusions were derived from our analysis.
It could be concluded that the strategies employed by the teachers were derived from
their daily experience, encouraging them to continue their effective practices, and doing so
without previous theoretical knowledge. Thus, relying mostly on instinct (“gut feeling”) and by
trial and error, the teachers developed their professionalism and applied theoretical notions in
practice, for example, the “speech-gesture system” approach ( McNeill, 1992).
Eufracio 37
Furthermore, the results of this study demonstrated that the teachers used flexible
bilingualism in bilingual Arabic–Hebrew kindergarten via the translanguaging strategy (Creese
& Blackledge, 2010), which contrasts traditional instruction with language separation. In the
teachers' opinion, translanguaging enables bilingual children to learn efficiently and encourages
children's interactive involvement in the kindergarten. In addition, this language practice
enhances bilingualism by avoiding monoglossic policy (language separation) and creating a
heteroglossic reality (García, 2009a).
In line with García's claim on the social aim of translanguaging, our results highlight a
necessity for translanguaging as an instructional strategy with the clear ideological goal to
promote Arabic from its socially weaker status. In regard to the socio-linguistic situation in
Israel, the kindergarten has a majority-minority case of language usage, meaning that Hebrewand Arabic-speaking L1 children prefer to speak Hebrew. The Arabic-speaking children in the
kindergarten are exposed to Hebrew in many contexts, and are more motivated by their families
to learn Hebrew to help them function effectively in Israel. At the same time, Hebrew-speaking
(L1) children have no similar exposure to Arabic; they have no similar environmental and social
factors that accelerate L2 acquisition. The educational setting in the case under study provides a
real possibility for Arabic acquisition for the L1 Hebrew-speaking children. By considering a
specific Israeli context, our teachers' language strategies were applied to influence the status of
Arabic and change the hegemony of Hebrew at least within this preschool.
Within a broader theoretical context of bilingual education, our data expands on the
understanding of a flexible and dynamic approach toward bilingual development and education
Eufracio 38
that views bilingualism not as an end result but as an active process (García, 2009a) in which a
heteroglossic reality takes a central place. This reality might support a bilingual versus a 'two
monolinguals in one body' ( Gravelle, 1996) development. In addition, the growing amount of
data demonstrates the inconsistency between a policy of language separation in the dual
language program and its implementation and results ( De Palma, 2010; Lee, 2007). The main
critical point is that the strategy to “stick to one language” without permitting code-switching
during target-language time, which was designed to achieve competence in both languages of the
bilingual program, was found to result in many miscommunications and to the children's
tendency to “shut down” their responses ( De Palma, 2010; Lee, 2007). The present study adds to
these critics such that it discusses how translanguaging as a strategy might raise children's
motivation in L2 acquisition.
Moreover, the present study shows the facilitating role of strategy multiplicity in the
bilingual classroom. As we illustrated, the examined strategies were interwoven during the
classroom instruction and completed each other. In this context, our proposed four-fold schema
for categorization of language-teaching strategies for young bilingual children has both a
theoretical and a practical contribution. Thus, we hope that this schema might serve as a model
for further research designs in this domain. In addition, this schema could facilitate teachers'
planning and enhance efficiency of their instructional process.
Within an international context of language teaching, this study has a number of
important implications for bilingual teachers, and, in general, for bilingual education. The study
highlights the necessity for educators to rethink traditional bilingual language strategies and to
Eufracio 39
implement innovative bilingual pedagogies, which have an impact on children's motivation and
bilingual development. In this context, with regard to teachers' professional development, the
study highlights a necessity for bilingual teachers to regularly reflect on their bilingual pedagogy.
On the whole, reflection permits teachers to construct and reconstruct their professional
experiences, identify problems and obstacles and find solutions, and critically examines the
teacher's own pedagogical ideology and practice. As we illustrated in our study, our bilingual
teachers' regular reflections on their language strategies, together with thorough monitoring of
the children's progress in L2 acquisition, resulted, for example, in a reevaluation of using the
“language-time” strategy. Finally, our data illustrates that applying co-teaching of both
languages, without setting aside time for each individual language, demands nonstop cooperation
and coordination between the teachers.
Finally, no single study should be expected to provide a full investigation of languageteaching strategies in the classroom. Limitations of the study included the Hebrew-speaking-L1teacher staff change in the second year of the study. In addition, our study focused on only one
Arabic–Hebrew kindergarten, and therefore, we should be cautious in generalizing our findings
to all bilingual kindergartens. At the same time, we hope that this study will serve as a starting
point for a broader international longitudinal research project on bilingual teaching strategies in
preschool bilingual education and their impact on linguistic and social development of bilingual
children.
[For quotations of more than four lines, indent the quote one inch from the left
margin and do not use quotation marks. To apply this formatting, on the Home
Eufracio 40
tab, in the Styles gallery, click Quote. For shorter quotations, you can put them in
quotation marks, and incorporate them directly into text.]
Table 1
[This Table Title Uses a Style Named “Table Title”, Available on the Home Tab, in the Styles
Gallery]
Column Heading
Additional
Additional
Additional
Column Heading
Column Heading
Column Heading
Row Heading
Row Heading
Row Heading
Source: [This source text uses a style named “Table Source”, available on the Home tab, in the
Styles gallery.]
a. [This note text uses a style named “Table Note”, available on the Home tab, in the
Styles gallery. Table notes use a lowercase letter instead of Arabic numerals to differentiate them
from the notes to body content.]
Eufracio 41
Chart Title
6
5
4
3
2
1
0
Category 1
Category 2
Series 1
Category 3
Series 2
Category 4
Series 3
Fig. 1. [This figure caption uses the No Indent style, available on the Home tab, in the Styles
gallery. Label figures with the abbreviated “Fig.” and a figure number.]
[The sample Works Cited list that follows was created using the Bibliography feature
available on the References tab. This feature offers the option to specify MLA style, so that your
references are automatically formatted correctly. You can also use this feature to add in-text
citations, such as the one shown at the end of this paragraph. To add page numbers to a citation
after you insert it, right-click the citation and then click Edit Citation. Note also that MLA rules
for citations and references are extensive. So it’s a good idea to refer to MLA 7th Edition for
further information.] (AuthorLastName Pages)
[To use this template when creating the outline for your paper, on the Home tab, in the
Styles gallery, click No Indent. Then, on the same tab, in the Paragraph group, click the
Multilevel List icon and then click the MLA Outline style that appears under List Styles. The
first six levels of this list style correspond to the outline levels defined in MLA 7th Edition.]
For additional guidance on formatting your research paper, consult MLA 7th Edition as
well as your instructor.
Eufracio 42
Works Cited
AuthorLastName, FirstName. Title of the Book Being Referenced. City Name: Name of
Publisher, Year. Type of Medium (e.g. Print).
LastName, First, Middle. "Article Title." Journal Title (Year): Pages From - To. Print.