Bilingual Research Journal

The Journal of the National Association for Bilingual Education

ISSN: 1523-5882 (Print) 1523-5890 (Online) Journal homepage: https://www.tandfonline.com/loi/ubrj20

Translanguaging in the planning of academic and

creative writing: A case of adult Japanese EFL
learners

Blake Turnbull

To cite this article: Blake Turnbull (2019): Translanguaging in the planning of academic and
creative writing: A case of adult Japanese EFL learners, Bilingual Research Journal, DOI:
10.1080/15235882.2019.1589603

To link to this article: https://doi.org/10.1080/15235882.2019.1589603

Published online: 04 Apr 2019.

Submit your article to this journal

Article views: 24

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ubrj20
BILINGUAL RESEARCH JOURNAL
https://doi.org/10.1080/15235882.2019.1589603

RESEARCH ARTICLE

Translanguaging in the planning of academic and creative writing:

A case of adult Japanese EFL learners
Blake Turnbull
Kyoto University

ABSTRACT
Foreign language (FL) writing has undergone a number of key evolutions
over the past 50 years. But despite attempts toward a reframing of FL
learners as emergent bilinguals, bilingual languaging strategies such as
translanguaging are still rarely seen in most FL education contexts, and
bilingual composition strategies are yet to become integrated into main-
stream FL writing education as a result. The goal of the present study was
thus to investigate the effects of weak and strong forms of translanguaging
on the production of Japanese EFL students’ academic and creative com-
position pieces. The findings suggest that by allowing FL learners to engage
in strong translanguaging practices, removing the “barriers” between
named languages, during the planning stage of their English composition,
learners are able to score higher than those who were forced to use one
language over the other (either the target language or their native lan-
guage). They were also able to produce more concise, well-formed essays
with fewer misused lexical items from a lack of relevant language knowl-
edge or experiences.

Introduction
Scholars of recent years have referred to the benefits of including learners’ home language in
the second (L2) or foreign (FL) language learning process (see for example Cook, 2001; Cummins,
2007; Lin, 2015; Macaro, 2014; Mahboob & Lin, 2016; Turnbull & Dailey-O’Cain, 2009). Based on
Cook’s (1991) concept of multicompetence, or the complex and integrated mental language func-
tions of bi- and multilingual speakers, and in support of the recent multilingual turn in language
education (e.g., Conteh & Meier, 2014), some scholars have argued in favor of mother tongue-
inclusive bilingual pedagogies becoming mainstream in the FL classroom in recognition of bilingu-
alism as a goal of FL education (e.g., Turnbull, 2018; Butzkamn & Cadwell, 2009; García & Wei,
2014), and even for the integrated use of two or more languages to become an established aim of FL
education (e.g., Canagarajah, 2013; Levine, 2011; Turnbull & Dailey-O’Cain, 2009). However, despite
these recent movements for change, including those by Turnbull (2018) to expand the emergent
bilingual category to include FL learners in their own right, the need to consider bilingual pedagogies
and learning strategies, such as translanguaging (see García & Wei, 2014), within the FL classroom
remains largely unrecognized. To address this gap in the research, the aim of the present study was
to investigate the effects of weak and strong forms of translanguaging on the production of Japanese
EFL students’ academic and creative writing.

CONTACT Blake Turnbull turnbull.blake@gmail.com Kyoto University, Yoshida-nihonmatsu-cho, Sakyo-ku, Kyoto 606-
8501, Japan.
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ubrj.
Blake Turnbull is an English instructor at Ritsumeikan University and Kyoto University of Foreign Studies and a PhD student
in the Department of Foreign Language Acquisition and Education at Kyoto University, Japan. His research interests include
language education, bi-/multilingualism, and translanguaging.
© 2019 the National Association for Bilingual Education
2 B. TURNBULL

Literature review
The translanguaging concept
The term translanguaging, first coined by Welsh educationalist Cen Williams (1996), was originally used in
reference to the “planned and systematic use of two languages for teaching and learning inside the same
lesson” (Lewis, Jones, & Baker, 2012a, p. 643). In this sense of the term, receptive language skills (i.e., input)
are carried out in one language, and productive skills (i.e., output) in the other (Baker, 2011). For example,
students may read a passage in English and write about it in Japanese or listen to a dialogue in French and
discuss it afterwards in Spanish. The main premise behind this form of translanguaging was to reinforce
understanding of classroom content through the use of both languages because, as Baker (2011) explains, “to
read and discuss a topic in one language, and then to write about it in another language, means that the
subject matter has to be processed and ‘digested’” (p. 289). In other words, translanguaging involves a higher
level of cognitive processing whereby learners must assimilate and accommodate incoming information.
Williams’s translanguaging not only promotes a deeper understanding of content but also works to develop
the weaker language in relationship with the more dominant language in a second or foreign language
classroom setting.
In recent years, however, the term translanguaging has been extended by a number of scholars
(see for example, Canagarajah, 2011; Creese & Blackledge, 2010; García, 2009; García & Wei, 2014;
Hornberger & Link, 2012) to refer to the complex language and pedagogical practices of bi/multi-
lingual individuals and communities. An assortment of similar terms have confused the concept of
translanguaging in the literature, including “metrolingualism, polylanguaging, polylingual langua-
ging, heteroglossia, codemeshing, translingual practice, fexible bilingualism, multilanguaging, and
hybrid language practices” (Lewis, Jones, & Baker, 2012b, p. 649); however, the concept of trans-
languaging, argued by García (2009) under the paradigm of dynamic bilingualism, views language as
a process in which politically defined boundaries are not adhered to. In this sense, translanguaging
refers to the fluid language practices of bi/multilinguals who transcend between and beyond the
systems in their linguistic repertoire, drawing upon multiple semiotic resources appropriate to given
contexts to language and to make meaning of both themselves and their surrounding bilingual
environments.
Bilingual speakers are considered to have a single, expanded linguistic repertoire, comprising
linguistic features that correspond to one or another “named language” (but also see MacSwan,
2017). The so-called named languages that make up a speaker’s linguistic repertoire are those
generally considered to be separate languages in society (e.g., Japanese, English, Spanish, German,
Chinese, etc.). These “languages” are social constructs as opposed to linguistic objects, argued by
Makoni and Pennycook (2007) to be an invention of the nation-state. Bilinguals are therefore not
thought of as switching between languages but are rather drawing on certain linguistic features that
adhere to one or more societally constructed named language relevant to a given communicative
context. The present study employs what García and Lin (2017) refer to as weak and strong forms of
translanguaging, which will be examined in the following section.

Weak and strong forms of translanguaging

In William’s original definition, translanguaging viewed learners as having two autonomous languages that,
much like in code-switching, are used in separation to one another. This is similar to the weak version of
translanguaging (as suggested by Williams, 1996), which argues in favor of softening the boundaries
between languages, thus still acknowledging their existence and employing said languages in separation.
The strong version of translanguaging, as García and Wei (2014) posit, claims bilingual speakers do not
adhere to the socially constructed barriers of different named languages but rather draw selectively on the
features in their expanded linguistic repertoires to make meaning and to learn. This is shown in Figure 1,
based on Anderson’s (2017) translingual continuum.
 BILINGUAL RESEARCH JOURNAL 3

weak strong
monolanguaging translanguaging translanguaging

Howatt's (1984) Williams’ (1996) García & Wei’s (2014)

monolingual translanguaging translanguaging
principle

Figure 1. Translanguaging continuum.

At the far left of the continuum, Howatt’s (1984) monolingual principle represents a stance on
emergent bilingualism in which bilingual speakers are viewed purely in relation to the native speaker
and judged as failed monolinguals rather than competent bilinguals with multicompetence (see
Cook, 1999). The weak view of translanguaging, originally proposed by Williams (1996), is a step
toward recognizing the true multicompetence of bilingual speakers, arguing for the softening of
boundaries between national languages in education while still acknowledging their existence. At the
far right of the continuum, a strong approach to translanguaging, as argued for by García and Wei
(2014), posits that the boundaries between named languages are constructs of society, and bilingual
speakers have a single linguistic repertoire from which they draw on features relevant to a given
communicative context.
In a language learning context, then, Howatt’s monolingual principle represents classrooms in
which one language is chosen for use over the other(s) (e.g., English-only contexts where English
is used for all activities and instruction). A weak translanguaging stance represents bilingual
classrooms in which each language is employed in separation from the other(s) for specific
functions (e.g., learners may employ Japanese for a discussion after having read a text in
English). A strong translanguaging position represents bilingual classrooms in which any and
all named languages are permitted and systematically employed to help develop learners’ weaker
language or the target language (e.g., students may draw on any language in which they hold
multicompetence to complete a given assignment). Particularly in a FL learning context, in which
the very goal is to assimilate linguistic features of a particular “named language,” strong
translanguaging does not refer to “random switching” between languages (which García, 2009,
warns may have negative effects). Rather, a strong translanguaging approach is the systematic
allowing of learners to employ their complete linguistic repertoire while still maintaining
a control over when to allow for the planned acquisition of certain features (such as with the
bilingual tap method). For example, learners may be allowed to employ any language in which
they hold multicompetence to activate their background knowledge before completing a task or
assignment in the target language.
It must be acknowledged here that a tension does exist when discussing translanguaging theory
and the concept of socially constructed “named languages” in strong translanguaging while simulta-
neously using national language names and L1/L2 labels in reference to weak translanguaging. The
difficulty when discussing strong translanguaging practices in FL learning contexts is that the very
goal of FL education is to teach and acquire “named languages.” As Turnbull (2017) states: “The goal
of FL education is bilingualism, achieved through the integration of linguistic features to learners’
expanded repertoires that correspond to the named TL” (p. 7). However, García (2017) suggests that
“named languages have had very real material effects in the lives of people. Teaching then these
named languages is an important social endeavor, especially in our globalized world” (p. 8). As such,
throughout this article, the term “home language” is used in reference to the linguistic features
pertaining to the socially constructed language “Japanese” and the term “foreign language” to the
socially constructed language “English.”
García and Lin (2017) suggest that education programs employing translanguaging and translin-
gual practices must employ both weak and strong forms of translanguaging. Weak translanguaging
4 B. TURNBULL

allows for educators to teach named languages (as is often required in certain language education
contexts) while still softening the boundaries between them. Strong translanguaging, on the other
hand, allows educators to provide a space that supports criticality and creativity (see Wei, 2011) so
that learners’ translingual practices may be developed as a whole. Through a combination of both
forms of translanguaging, bilingual (and indeed emergent bilingual) students can work to develop
their holistic linguistic repertoires while simultaneously developing the ability to use certain language
features in accordance with the rules that have been set down for a given “named language.”

Bilingualism and FL writing

Velasco and García (2014) suggest that “bilingual writers use different problem-solving strategies
and exhibit ways of expressing meaning that are not present in monolingual writing” (p. 10).
Accordingly, some studies have looked specifically at using bilingual processes to develop FL writing.
Fu (2003), for example, discusses the use of a bilingual process approach within a weak translangua-
ging framework to develop Chinese students’ EFL writing abilities, stating:
I believe thinking (reasoning and imagination) and the ability to organize ideas are equally, or even more,
important than language skills in learning to write. If we let our students express themselves and present their
ideas in their primary language, we give them opportunities to continue the development of their thinking.
With this development uninterrupted, they are able to write well in a second language once they develop
proficiency in it. (p. 74)

Some studies have looked at the bilingual use of learners’ native or additional language in second
language (L2) learning contexts. García and Kano (2014), for example, investigated the use of weak
translanguaging (with some limited elements of strong translanguaging) to develop the academic
writing skills of a bilingual English-Japanese class in the United States. They found that the emergent
bilingual Japanese students employed their home langage often to make sense of the lesson and to
rely on their existing language practices to understand and complete the task. On the other hand, the
more experienced bilinguals translanguaged to enhance the task and make meaning, demonstrating
greater learner autonomy. The authors concluded that mother tongue use through a translanguaging
approach allowed the Japanese learners “to become more aware of the differences in the construction
of Japanese and English written texts, to develop expertise in using their own translanguaging
strategies to construct academic texts in English, and to build their biliteracy practices” (p. 274).
Kobayashi and Rinnert (1992) compared the (weak translanguaging) effects of having 48 Japanese
university-level students compose English texts first in Japanese and then translating them into
English and having them compose directly in English. They found that translation allowed learners
of a lower level to improve the quality of their writing content, organization, and style and for
learners of all levels to improve their syntactic complexity. The authors also investigated the learners’
perspectives on the task and found that the majority of higher-level learners thought their direct
writing was better than their translation (76% compared to 24%), whereas in contrast, the lower-level
learners thought their translation writing was better than their direct writing (73% vs. 27%). Finally,
the authors examined how much Japanese the learners thought they were using during the direct
writing activity. On average, 20% of all learners felt they used Japanese more than 75% of the time,
while 51% reported to use Japanese 50%–75% of the time, 23% thought to use Japanese 25%–50% of
the time, and only 6% to use Japanese less than 25% of the time, suggesting a relatively high
employment of Japanese overall.
Gort’s (2012) investigation of the composition-related code-switching practices of six emergent
Spanish-English bilingual children found that both learners’ languages remain active during the
writing process and that learners use these two languages to support the development and promo-
tion of bilingual literacy skills. Similarly, Kibler’s (2010) study into the code-switching practices of
four Latino adolescents of beginner to intermediate English proficiency, working in cooperation with
one Spanish-English bilingual peer, found that bilingual students utilize the languages in their
 BILINGUAL RESEARCH JOURNAL 5

linguistic repertoire to assert both expert and novice identities during writing activities and to
cognitively manage the task and social phenomena. Referring to learners whose FL literacy skills
do not match their home language, or that of their FL peers for that matter, Kibler states that home
language use “offers an opportunity for their academic, linguistic, and literacy-related knowledge to
be recognized and validated. Students’ first language use redefines in-school writing traditions that
would normally position them simply as unskilled, novice L2 writers” (p. 137).
However, after a review of the relevant literature, there appears to be no studies that specifically
compare the effects of translanguaging practices on learners’ production of both academic and
creative writing. The author determined it important to investigate creative writing as well as
academic writing in the present study because little attention is given to creative writing in tertiary
EFL education contexts and related academic literature (see Antoniou & Moriarty, 2008; Randolph,
2011), despite the benefits that creative writing affords to learners’ language development and self-
expressionism. These benefits center on learners developing their “creative power” (Grabe & Kaplan,
1996, p. 333) through the employment of free language play, allowing them to engage unique,
personal meanings to express their lived experiences (e.g., Kinginger, 2004) and originality
(Randolph, 2011), and in doing so engage with the language at a deeper level of processing than is
available with most expository or academic compositions (Craik & Lockhart, 1972). Furthermore, it
appears as though few studies have investigated the movement toward specifically including the
translanguaging practices of FL learners during composition in an additional language.
This is particularly relevant in a Japanese EFL learning context (in which translanguaging is not
commonplace) because of the sociocultural relationship between ideology and language, or more
pointedly, language and imperialism. The worldwide spread of English has been unprecedentedly
rapid in comparison to the other major languages of the world (McKenzie, 2010), and Japan has
been no exception to the influx of English as the de facto international language (see Kubota, 1998).
Some scholars (e.g., Tsuda, 1993) have criticized this phenomenon of “English Imperialism” as has
the traditional nihonjinron discourse (a selection of commentaries arguing for the uniqueness of the
Japanese people, language, and culture), viewing the dominance of English and Western ideologies
in Japan as a “deviancy that threatens one’s Japaneseness” (McVeigh, 2002, p. 155). Concerns about
the growing international influence of the English language, and certain feelings of disadvantage by
nonnative speakers in English-dominated global communication, have cast a shadow over the role
that English has to play in societies such as Japan. However, some scholars (see, for example, Tosu,
1997) have argued against these claims in favor of the kokusaika discourse and the advantageous
position of English in the Japanese society. The kokusaika discourse “harmoniously embraces both
Westernisation through learning the communication mode of English and the promotion of
nationalistic values” (Kubota, 1998, p. 300). In other words, it aims to enhance Japan’s global
economic interest while maintaining and convincing the West of its national identity based on
a distinct cultural heritage. To achieve this, Japan has needed to produce new policies and plans to
address the matter and take up English as the de facto international language.
One such manner through which this has been attempted is the government’s Top Global
University Project, which aims to support the internationalization of a select number of universities
throughout the country. McKinley (2018) suggested there to be a need to shift from treating Japanese
tertiary students as learners of English to users of English to see a change from teaching English as
a foreign language to teaching English as a Global Language. In this context, the use of the home
language is considered to be a bilingual resource at both the learners’ and teachers’ disposal, and
learners’ language abilities are examined holistically, not in relation to native-speaker norms. The
author believes that a translanguaging approach to EFL education in Japan has the potential to help
this paradigm shift. The goal of the present study was thus to investigate the effects of weak and
strong forms of translanguaging (from an individual languaging perspective) on the production of
EFL Japanese students’ academic and creative composition pieces. To achieve this research goal, the
following main question was investigated: What are the effects of translanguaging practices on the
production of academic and creative EFL composition?
6 B. TURNBULL

Methodology
Participants
Two classes of 30 (Class 1 M = 15, F = 15; Class 2 M = 17, F = 13) first-year Japanese EFL students
studying a compulsory English course taught by the researcher’s associate at a large national
university in Japan took part in the study after having given their consent to do so. All participants
were native Japanese speakers and of an intermediate English proficiency level. All of the participants
belonged to the 18–20 year-old-age range. The participants reported to have been studying EFL for
between 5 and 13 years, with an average of 7.2 years in total.
All of the participants in Class 1 belonged to the Department of Literature but were pursuing
various major subjects, including literature, history, sociology, and philosophy. The participants in
Class 2 all belonged to the Department of Economics. These participants were selected because of
their similar ages and educational backgrounds, both in relation to EFL and other subjects, and
because of their similar status as EFL learners at the same tertiary institution. Although individual
writer characteristics were not reported, it is acknowledged that all writers had previously been
streamed into the same classes based on a university placement test and assigned to groups in the
present study at random. It can therefore be said that these participants were also selected because of
their equal level of EFL proficiency prior to participating in this study, and thus differences in
writing proficiency would not have significantly impacted their overall composition results.

The writing task

The 30 student participants in each class were split into two writing categories (academic writing and
creative writing) through a simple random sample by individually drawing names from a hat. Within
each category, the participants were further split into three focus groups. Group 1, working under
monolingual English-only (EO) practices, discussed their given topic and planned their essays only
in English. Group 2, employing a weak form of translanguaging (WT) in which English and Japanese
were alternated for different aspects of the task (i.e., Japanese for the discussion, English for the
writing task), discussed and planned their composition only in Japanese before writing in English.
Group 3, engaging in strong translanguaging (ST) where the boundaries between “English” and
“Japanese” were removed, discussed and planned their essays freely in both or either “languages”
before writing in English. All groups were required to produce their essays in English (as opposed to
continuing with their group theme and having, for example, the ST group compose in any “named
language”) because of the nature of FL education, the very goal of which is to “learn a named
language” or “to develop the weaker (target) language.” It is necessary, then, that some control
remains over the language use1 to ensure that learners are indeed producing and developing their
skills in the weaker language, which happens to be English in the case of the present study. Each
working group comprised five participants (see Table 1 for summary).
Each group was given a topic in relation to their category (persuasive-based academic writing vs.
memoir-based creative writing) and provided with a sheet of written instructions according to the
nature of their group, i.e., Group 1 was provided with instructions only in English, Group 2 only in
Japanese, and Group 3 in both English and Japanese. Explanations were given as to what was
expected of each group on paper, and the researcher also vocalized the instructions to each group in
English, Japanese, and in both languages, according to the nature of their group. Participants in the
EO group were strictly instructed to remain in the TL at all times. Those in the WT group were also

Table 1. Participants’ group composition.

Academic Writing Creative Writing
Group type G1: EO G2: WT G3: ST G1: EO G2: WT G3: ST
Participants 5 5 5 5 5 5
 BILINGUAL RESEARCH JOURNAL 7

Table 2. Writing topics.

Academic writing Creative writing
Topic Do you think that our society is too dependent on technology like Write about the advice you would give to your
smartphones and social networking? Why or why not? younger self about life and growing up.
現在の社会はスマホやSNSのようなテクノロジーに依存しす 成長と人生について若い自分にアドバイス
ぎていると思いますか。なぜですか？ したいことは何ですか？

carefully told to use only Japanese for their discussion and essay plans before switching to English for
the final essay. The ST group was encouraged to draw freely on both English and Japanese, and any
other language(s) in which they might hold competence, to discuss and plan their essays before
switching to English for the final composition piece. See Table 2 for the topics for each writing
category.
Once they had received their topic, each group was given 20 minutes to discuss their topic as
a group and write notes to plan their essays (see Table 2). The discussions were all audio recorded
and later transcribed by the researcher. The participants were then asked to write a short essay
individually, based on their group discussion and written plan. All participants, regardless of their
group type, were asked to write in English and were given 40 minutes to do so with no minimum or
maximum word count. The participants completed the tasks in a normal classroom setting during
a regular scheduled class period. This was done to maintain the learning environment with which
they are accustomed because unfamiliar settings and procedures can cause unknown and largely
immeasurable effects on students’ language performance (Skehan & Foster, 1997). Both the
researcher and regular instructor remained present during the writing phase but distanced them-
selves from the class. Their presence was necessary to ensure the participants did not further discuss
the topic in a language that had not been assigned to their group.
A complementarity mixed-methods approach, involving both qualitative conversation analysis of
group discussions (discussed in detail in the following, but also see Wong & Zhang Waring, 2010)
and quantitative analysis of students’ composition scores, was employed, as this allowed the
researcher to elaborate, enhance, and further illustrate the overall effects of weak and strong forms
of translanguaging on Japanese EFL students’ academic and creative writing abilities, more so than
with a single method approach alone.

Analysis
The rubric
The composition pieces produced by the participants in the present study were blindly evaluated
with a rubric based largely on Jacobs et al.’s (1981) ESL Composition Profile (see the appendix): an
evaluation scale comprising five major categories: content, organization, vocabulary, language use,
and mechanics, whereby each category is weighted “according to its approximate importance for
written communication” (Jacobs et al., 1981, p. 31). The results were then compared across writing
categories (academic and creative) and language groups (English, weak translanguaging, and strong
translanguaging) to determine the overall effects of monolingual and translanguaging group discus-
sions on English academic and creative writing tasks.

The raters
Each composition piece was graded by two independent assessors who were both tertiary-level
English educators in Japan. Each piece was read twice in a single sitting: once to establish whether an
overall message had been conveyed clearly by the writer, and then again to focus more specifically on
the elements within the Profile that resulted in the initial impression. A third assessor was used in
several instances when the grades assigned by the first two assessors differed by more than 10 points,
and an average of all three grades was assigned. All assessors underwent a training period with the
8 B. TURNBULL

ESL Composition Profile and followed assessment guidelines set out by Jacobs et al. (1981).
Cronbach’s alpha coefficient was used to test interrater reliability, which was determined to be
satisfactory at 0.865.

The scores
An average of the two independent raters’ scores was totalled for each composition piece, and the
results were entered into the SPSS (Statistical Package for Social Science) 23 software for a multifaced
analysis. First, a one-way ANOVA was employed to determine the descriptives and homogeneity of
variance (Levene’s test) between groups within the same writing category (academic or creative).
When a statistically significant variance was identified, a Tukey post hoc test was performed to
determine which of the group means were significantly different. A univariate test was then
employed to identify the effect size of each group, which Cohen (1988) defined as being either
small (d = 0.2,) medium (d = 0.5), or large (d = 0.8). Finally, independent t-tests were conducted to
determine any significant variations and to compare writing categories (academic and creative) and
focus groups (EO, WT, and ST), from both intraclass and interclass perspectives.

The conversation analysis

The qualitative conversation dialogue was transcribed by the researcher based on the transcription
symbols identified by Markee (2015). The data were then approached systematically in a series of
rounds using an emic perspective (see Pike, 1967), which investigates language interaction from an
“insider’s perspective,” viewing language use from the participants’ point of view. Based on the
advice offered by Wong and Zhang Waring (2010, p. 6) in their guide to conversation analysis for
ESL/EFL teachers, five key actions were employed to investigate certain codes in the transcribed data:
(a) unmotivated looking, (b) repeated listening and viewing, (c) answering “why that now?,” (d)
case-by-case analysis, and (e) deviant case analysis. Accordingly, each dialogue was first listened
through once in its entirety to gain a comprehensive perspective of the topics discussed and manner
in which the participants engaged in the discourse. Next, a rough first transcription was written,
which was reviewed again before a final transcription was produced and analyzed. Japanese sections
were translated by the researcher, who is not a native speaker but a White, Western speaker of
advanced Japanese, and checked by a Japanese-native bilingual peer for accuracy.

Findings
Group discussions
Analysis of the audio-recorded group discussions during the composition planning stage helps to
provide further evidence in support of the benefits of strong translanguaging practices over mono-
lingual English, or weak translanguaging, in EFL writing. These excerpts were selected because of the
interesting content they provide and because they were a fair representation of the groups’ engage-
ment with mono-/translingual practices as a whole. Certain symbols are used in the depiction of the
dialogue transcripts throughout this article: The times shown in brackets are the lengths of an
unspoken pause in seconds; double parenthesis are used to show translations; the [(ah)] symbol
represents a “thinking vocalism” in which the speaker used a filler sound while they thought about
what to say; and the [@] symbol denotes laughter (one for each laughter pulse). The P symbol
followed by a number denotes the participant number in that group from one to five (i.e.,
P1 = participant 1).

English-only monolingual groups

Likely as a the result of insufficient TL knowledge, the discussions held between participants
engaging in monolingual English-only practices were disjointed, indefinite, and largely incoherent
for both the academic and creative writing groups. Furthermore, rarely did the participants engage in
 BILINGUAL RESEARCH JOURNAL 9

conversation as a group, instead opting to state their individual perspectives in separation from one
another. This can be seen in the following conversational passage from the academic writing group
of Class 1, in which the participants stated their opinions one by one:
P1: (ah)… I’m for.. this topic because, (ah).. but it is true that technology or smartphones is very convenient,
but.. (ah) they don’t have creativity (6.9). We have to (ah) keep our creativity.. for own. Thank you.

P2: My think is.. we need technology… (ah) it is very useful for me. So I think we are too
dependent on it. Thank you.
Perhaps also due to a lack of relevant vocabulary and English language knowledge, those
participants engaging in the English-only monolingual discussion quickly ran out of content to
discuss, spending a large portion of their planning time questioning what to do next, as can be seen
in the following interaction from the creative writing discussion of Class 1:
P2: How long we have @discussed?
(7.8)
P3: Fifteen minutes?@@
@@
(20.1)
@@@@@
P2: So what should we do?…
@@@
(18.8)
P1: Do you have any opinion.. other opinion?
Japanese was, in one case, used during the English-only discussion despite the monolingual instructions
imposed upon the participants, as can be seen in the following dialogue taken from the creative writing
group of Class 2:
P3: Any advice?
P1: ごめんね、英語変換できない.
((I’m sorry, I can’t switch to English.))
P4: (ah) I will tell me… tell me.. never be class leader@@
P2: Class leader?
P4: (ah) in Japanese, 委員長.
((In Japanese, “iinchō”.))
P2: Oh.

This perhaps suggests that, despite the monolingual guidelines, the participants in the English-
only group were unable to fully engage in strict monolingual practices and relied at times on
Japanese to express both themselves and their intended meanings to maintain the flow of conversa-
tion and to manage the task.

Weak translanguaging groups

The participants engaging in weak translanguaging encountered occasions of unknown English
lexical items, which they were either unwilling (due to the nature of their group) or unable to
discuss due to the Japanese restriction that had been imposed upon them. This can seen in the
following example from the creative writing discussion group of Class 2, in which participants were
unsure of the English word for 親孝行 (oyakōkō, meaning duty to one’s parents):
P2: やっぱ親孝行を書いといたらいいんじゃない。
((Actually, we could just write “oyakōkō” right?))
P1: 親孝行の英語がわかんない。
((I don’t know what ‘oyakōkō’ is in English.))
P2: ほんまや。
((That’s true.))
P1: じゃあ、あとは？
((So, what else then?))
10 B. TURNBULL

Furthermore, those engaging in weak translanguaging practices appeared to have difficulty choosing
the correct English word based on the Japanese equivalent with which they had been provided. This was
particularly noticeable in the academic writing task with the word 依存 (izon, meaning “dependent” or
“reliant”), which was translated by one participant as “to pay too much attention to,” as opposed to the
correct translation of “to be too dependent on” or “to rely too much on.” One participant also
commented on the negative connotation that the word 依存 affords in Japanese, which is not
necessarily the case in its English equivalent. Another example of this interlinguistic discrepancy was
seen with the term “SNS” (Social Networking Service): a commonly used abbreviation in Japanese that
was provided to the weak translanguaging group in their Japanese topic. Although the abbreviation may
also be understood by some in English, it must be acknowledged that its use is much less common in
English than it is in Japanese. Most of the learners engaging in weak translanguaging practices (who
were not provided with the essay topic in English) employed the term “SNS” in their composition. This
could be viewed as a form of negative transfer (i.e., of linguistic features that correspond to one named
language) from their native language, which they could not correct because they were not allowed the
opportunity to use, or given any exposure to, the English TL during the discussion stage. Those learners
who were provided with the English term in the English-only and strong translanguaging groups used
the more naturally sounding term “social network” instead.
It is also interesting to note that the participants in the weak translanguaging group did not
always remain on topic during their discussion, choosing instead to focus at times on what the
Japanese instructions were asking of them instead of planning the content of their essays, as can be
seen through the academic writing group’s discussion about the meaning of the word ような (yōna,
meaning “like” or “such as”) in Class 1:
P5: テクノロジーさ、スマホ・SNSとかって、書いてあるけど..
((Technology, it says things like smart phones and social networking sites, but..))
P3: 「ような」って言っているから..
((It said “such as”, so…))
P4: ようなテクノロジーだからテクノロジーって言っている。
((It’s “technology such as”, so it’s saying technology))
P3: ようなってととこが..なんか曖昧だな。
((The “such as” part is.. vague..))

The fact that these participants spent a portion of their planning time discussing the meaning of
the Japanese topic, as opposed to more relevant content that would form the basis of their essays,
perhaps suggests that their thinking process was focused more on Japanese than on English, which
may have resulted in negative transfer and confusion when later producing their English essays. In
other words, the strict use of Japanese may have been a distraction, preventing the learners from
engaging the unique English languaging experiences and knowledge that could have been of benefit
to their final composition pieces.

Strong translanguaging groups

Unlike the weak translanguaging group, those participants engaging in strong translanguaging
practices were generally unrestricted by unknown English vocabulary items and worked to discuss
the potential meanings of words they did not know by drawing on linguistic features from both
English and Japanese. This can be seen in the following example. Here, one student from the creative
writing group of Class 2 discusses how they would advise their younger selves to be more aware of
medical test results, perhaps in light of a difficult health history, but was unsure about the English
translation of 検便 (kenben, meaning “stool test”):
P4: 検便って英語で何というんだろう。
((What’s kenben called in English?))
P1: 何というやん？Checking？
((What is it called? “Checking”?))
P2: そうそう。Examine?
 BILINGUAL RESEARCH JOURNAL 11

((Yeah. “Examine”?))
P1: Examination ofかな？
((I wonder if it’s “examination of”?))
P3: Test ofかな？I think.
((I wonder if it’s “test of”? I think.))
P4: あ、stool testだ。
((Ah, it’s “stool test”.))

They were also more efficient at maintaining the focus of the discussion, spending the entirety of
the discussion time planning the content of their essays as opposed to discussing the meaning of the
instructions, like the weak translanguaging group did, or how much time they had remaining, like
the English-only monolingual group. The participants in the strong translanguaging group also
showed a preference for English-medium key words despite the predominantly Japanese-based
discussion, which may have helped them to maintain the focus of the discussion. This can be seen
in the academic writing group’s use of the English word “too dependent” over the Japanese 依存過
ぎている (izon sugiteiru) in Class 1:
P4: なんかtoo dependentかさ、not too dependentっていうのはさあ、なんか.. 結構個人的なもん
だね。
((So, ‘too dependent’ and ‘not too dependent’ is kind of a personal thing, right?))
P1: そうなんか、依存しているかーtoo dependentかnot too dependentか.. ってかしているか自体もう..
それは結構曖昧だけど.. わかんないけど。
((Yeah, to depend on – ‘too dependent’ or ‘not too dependent’… I mean, that we are itself is already…
that is quite ambiguous but… I don’t know.))
P2: Dependentというのは.. そうだけど、too dependentってどういうことだ？
((We are ‘dependent,’ but what does ‘too dependent’ mean?))

Although these participants did not extensively alter between the Japanese- and English-based
linguistic features in their expanded repertoires, the fact that they did employ English-medium key
words midway through Japanese-dominated sentences suggests that they are able to effectively
engage and integrate all of the features in their linguistic systems and thus made conscious decisions
throughout the discussion not to employ English for one reason or another.

Composition scores
Class 1
The composition score data collected from Class 1 across both academic and creative writing groups
were analyzed through a one-way ANOVA to determine descriptives and significant correlations, the
results of which are presented in Table 3.
There was a statistically significant difference found between the three academic writing groups, F
(2,12) = 6.910, p = .010. A Tukey post hoc test revealed there to be no significant variation between
the EO and WT groups (p = .116) or between the WT and ST groups (p = .316). There was
a significant difference between the EO and ST groups (p = .008); however, a test of homogeneity of
variances revealed the variances to be significant and therefore unequal in this sample (Levene’s
statistic = 10.05, p = .03 [>.05]). Based on Cohen (1988), the partial eta squared effect size was
reported to be medium (ηp2 = 0.535).
A difference was also found between the three creative writing groups in Class 1, F(2,12) = 8.721,
p = .05. Similarly, a Tukey post hoc test showed there to be no variation difference between the EO
and WT groups (p = .127) or between the WT and ST groups (p = .142); however, a statistically
significance was found between the EO and ST groups (p = .003). The results of a test of homo-
geneity of variances showed the homogenity of subsets to be equal and insignificant (Levene’s
statistic = 0.031, p = .970 [>0.05]), and therefore the assumption of homogeneity of variance was
met (i.e., not violated). The effect size was reported to be medium (ηp2 = 0.592).
12 B. TURNBULL

Table 3. Summary of Class 1 assessment grades.

Academic Writing Creative Writing
G1 (EO) G2 (WT) G3 (ST) G1 (EO) G2 (WT) G3 (ST)
Minimum 69.3 80.0 84 66.0 75.5 82.0
Maximum 87.0 84.0 89 79.5 86.0 91.5
Mean 75.2 81.6 86.1 74.5 80.8 86.9
(SD) (7.72) (1.56) (1.88) (5.24) (4.62) (4.16)

Independent t-tests found no statistically significant difference between the academic and creative
writing categories across all three groups in Class 1: EO academic and creative writing means
(m) = 75.16 v 74.50 respectively, t(8) = 0.158, p = .878; WT m = 81.60 v 80.80, t(8) = 0.367,
p = .723; or ST m = 86.10 v 86.90, t(8) = 0.392, p = .705. This suggests no significant variation in the
effects of translanguaging group discussions on FL academic or creative writing in this class sample.
A one-way ANOVA was used to compare word counts between each group across both categories
of writing, the results of which are presented in Table 4.
Significant differences were found between the groups in both the academic writing, F
(2,12) = 14.286, p = .001, and creative writing, F(2,12) = 5.159, p = .024, categories in Class 1. The
post hoc test found highly significant variation between the EO and ST (p = .001) and WT and ST
(p = .015) academic writing groups, although the EO and WT group was found to be insignificant
(p = .170). High variation was also reported between the EO and ST groups in the creative writing
category (p = .020) but not between the EO and WT groups (p = .442) or between the WT and ST
groups (p = .173). The results of a Levene’s test for homogeneity of variances came back as
insignificant for all groups, suggesting that the assumption of homogeneity of variance was met.

Class 2
The composition score data collected from Class 2 were also analyzed through a one-way ANOVA to
determine descriptives and significant correlations, the results of which are presented in Table 5.
A statistically significant difference was found overall between the three academic writing groups
in Class 2, as determined by the one-way ANOVA, F(2,12) = 6.509, p = .012. The Tukey post hoc test
revealed there to be no statistically significant difference between the EO and WT groups (p = .903).
However, unlike in Class 1, a significant variation was found between both the EO and ST groups
(p = .016) and between the WT and ST groups (p = .034). The homogenity of subsets was also
reported to be insignificant (Levene’s statistic = 1.514, p = .259 [>.05]), and the effect size to be
medium (ηp2 = 0.520).
There was also a statistically significant difference found between the three creative writing
groups, F(2,12) = 6.756, p = .011. The Tukey post hoc test showed there to be no statistical variation
difference between the EO and WT groups (p = .289) or between the WT and ST groups (p = .136);
however, the difference between the EO and ST groups was again found to be highly significant
(p = .008). The homogenity of subsets was reported to be insignificant (Levene’s statistic = 1.714,
p = .221 [>.05]), and the effect size to be medium (ηp2 = 0.530).

Table 4. Summary of Class 1 word count per group.

Academic Writing Creative Writing
G1 (EO) G2 (WT) G3 (ST) G1 (EO) G2 (WT) G3 (ST)
Minimum 202 172 122 229 187 134
Maximum 288 249 168 506 310 183
Mean 240.20 205.60 146.00 306.80 248.00 158.20
(SD) (32.85) (30.71) (19.01) (112.57) (56.99) (19.07)
 BILINGUAL RESEARCH JOURNAL 13

Table 5. Summary of Class 2 assessment grades for each group.

Academic Writing Creative Writing
G1 (EO) G2 (WT) G3 (ST) G1 (EO) G2 (WT) G3 (ST)
Minimum 73.0 78.5 85.0 65.0 75.0 83.0
Maximum 82.0 82.5 91.5 83.0 85.0 94.5
Mean 78.9 80.6 87.6 75.5 81.0 88.3
(SD) (3.54) (1.52) (2.38) (7.51) (3.71) (4.62)

As in Class 1, independent t-tests also found no statistically significant difference between the
academic and creative writing categories across all three groups in Class 2: EO academic and creative
writing means respectively = 79.70 v 75.50, t(8) = 1.089, p = .308; WT m = 80.60 v 81.04, t
(8) = 0.245, p = .812; or ST m = 86.60 v 88.30, t(8) = 0.657, p = .530, confirming there to be no
significant difference in the effects of translanguaging group discussions on FL academic or creative
writing in this class sample either.
A one-way ANOVA was also used to compare the word counts between each group for both
categories of writing in Class 2, the results of which are presented in Table 6.
Significant differences were found between the groups in both the academic writing, F
(2,12) = 14.939, p = .001, and creative writing, F(2,12) = 5.106, p = .025, categories in Class 2. The
post hoc test found highly significant differences between the EO and ST (p = .001) and WT and ST
(p = .009) academic writing groups, although the EO and WT group was found to be insignificant
(p = .225). As in Class 1, significant variation was also reported between the EO and ST groups in the
creative writing category (p = .020) but not between the EO and WT groups (p = .186) or between
the WT and ST groups (p = .245). The results of a Levene’s test for homogeneity of variances came
back as insignificant for all groups, confirming that the assumption of homogeneity had been met.

Interclass comparison
Independent t-tests conducted between each group across both classes found no statistically sig-
nificant variation between the interclass overall academic writing scores: Class 1 and Class 2 means
respectively = 80.95 v 82.37, t(28) = 0.699, p = .285; or of the overall creative writing
scores: m = 80.73 v 81.61, t(28) = 0.338, p = .738. Levene’s statistic of homogeneity reported to be
insignificant (academic = 0.285, creative = 0.971), suggesting a high level of consistency between the
two classes and therefore of the effects of translanguaging groups discussions on academic and
creative composition.
Furthermore, no statistically significant difference was found between the interclass academic
writing word counts: Classes 1 and 2 means respectively = 197.26 v 170.20, t(28) = 1.740, p = .093; or
for the overall creative writing word counts between classes: m = 237.67 v 217.73, t(28) = 0.495,
p = .625. Levene’s statistic of homogeneity was insignificant (academic = 0.310, creative = 0.487), and
it could therefore be said that the lack of variation between the two classes provides consistent
evidence in support of the beneficial effects of translanguaging group discussions on the concise
production of English academic and creative writing.

Table 6. Summary of Class 2 word count per group.

Academic Writing Creative Writing
G1 (EO) G2 (WT) G3 (ST) G1 (EO) G2 (WT) G3 (ST)
Minimum 185 152 101 142 159 108
Maximum 231 209 153 544 256 140
Mean 201.80 178.40 130.40 323.80 205.40 124.00
(SD) (18.38) (24.04) (20.37) (167.92) (36.03) (12.63)
14 B. TURNBULL

Discussion
The majority of past studies on the use of one’s home language in the FL writing process have
focused on the role of the home language in a traditional, monolingual FL language classroom
environment in which learners are said to be adding a foreign language to their linguistic repertoires.
However, these traditional approaches to FL and L2 writing are at odds with the ever-increasing
international reality of modern society (Horner, Lu, Royster, & Trimbur, 2011). Canagarajah (2015)
suggests that, to many language educators, any deviation from what is considered to be a standard
written convention is an error of poor writing that must be fixed, but as Wang and Wen (2002)
claim, the FL writing process “is a bilingual event” (p. 239) and does not necessarily need to fit the
standard expectations to be correct.
Atkinson et al. (2015) have looked in depth at the differences between L2 writing and translin-
gual-based bilingual writing, the latter of which involves the fluid use of language to challenge static
views of writing and allows learners to construct and negotiate meaning and identity as they do so.
The present study fulfills this requirement and provides evidence for the beneficial effects of
translingual writing practices in a Japanese EFL environment. A significant difference was found
in the writing scores between those participants engaging in English-only monolanguaging practices
(Group 1) and those engaging in strong translanguaging practices (Group 3) across both academic
and creative writing categories in both classes. No difference was found between the academic and
creative writing groups, suggesting that the benefits of a strong translanguaging approach are broad
and may apply to various styles of writing. The significant difference between the English-only and
strong translanguaging groups provides evidence in support of employing strong translanguaging
practices in the FL classroom as opposed to solely monolingual-based practices as is often advocated
for in many FL institutions. This is further supported by the arguments made by Author (2018) that
FL learners are, in their own right, emergent bilinguals and should thus engage in bilingual learning
strategies to express themselves, to make meaning, and to learn, and by Leung and Scarino (2016),
who suggest that “goals for language learning should … be framed within an integrated view of the
development of the holistic linguistic repertoire of learners” (p. 92).
The present study has found that, by allowing FL learners to engage in strong translanguaging
practices during the planning stage of their English composition, learners are able to score higher
than those who are forced to use one language over the other (be it the TL or their native language).
Not only that, learners are also able to better organize their ideas and arguments (see Kobayashi &
Rinnert, 1992) to produce shorter, more concise essays based on an overall word count. Significant
differences in the total word counts were found in all cases between those engaging in English-only
monolingual practices and strong translanguaging, with strong translanguaging resulting in fewer
composition words overall. The fact that each strong translanguaging group still scored the highest
in their essay grades despite having a lower overall word count suggests that engaging in strong
translanguaging to plan one’s essay perhaps affords learners the ability to more effectively gather,
organize, and convey their ideas and arguments in a succinct manner (see Fu, 2003).
Analysis of the audio-recorded group discussions provides further evidence in support of the use
of strong translanguaging practices over English-only monolingual use, or even weak translangua-
ging, in the FL classroom. The discussions held between learners of the monolingual English-only
group were disjointed and largely incoherent. It is likely that the restrictions of unknown English
vocabulary and grammatical structures left learners deprived of the ability to effectively express
intended meanings (see Liebscher & Dailey-O’Cain, 2005), resulting in broken, indefinite discussions
that provided little help to the essay planning process overall. The participants also quickly run out
of content to discuss, spending a large portion of their planning time contemplating how long they
had been discussing and what they should do next rather than conversing about the topic to gather
ideas. This is, again, likely due to the limited scope at which the participants could discuss given their
restricted body of English vocabulary, suggesting a negative effect of the English-only policy that has
long driven modern language pedagogies such as the Direct Method and Communicative Language
 BILINGUAL RESEARCH JOURNAL 15

Teaching (see Cummins, 2007). It was also interesting to note that, despite the English-only rule
opposed upon them, participants in the creative writing discussion group of Class 2 did engage in
code switching practices on one occasion and employed Japanese. This provides further evidence in
support of the fact that both (or all) of the “languages” in a learner’s mind remain active (Cook,
1992) during the writing process and are used to support the development and promotion of their
bilingual languaging strategies (also see Gort, 2012).
The discussions employing weak translanguaging, i.e., discussions only in Japanese, showed both
advantages and disadvantages on learners’ EFL composition overall. On average, those participants
who engaged in weak translanguaging scored higher in their essay scores than those from the
monolingual English-only group, suggesting some benefits of employing one’s native language, but
disadvantages were also observed. For example, the participants in the weak translanguaging group
spent time discussing the meaning of the Japanese instructions instead of planning the content of
their English essay, suggesting a Japanese-dominated thought process. Negative transfer resulting
from differences in word connotations and word usage between languages may also have affected
their final English compositions, resulting in lower scores than those in the strong translanguaging
group. The fact that both, or all, languages remain active in learners’ minds throughout the learning
process suggests that shutting one off (English, in the case of the weak translanguaging group in the
present study) during the planning process of their composition is effectively banning the use of
a beneficial and necessary semiotic tool at their disposal (see Anton & DiCamilla, 1999).
On the other hand, those engaging in strong translanguaging practices did not encounter
situations of unknown vocabulary items or appear to struggle with lexical differences or negative
transfer from one language to the other. Instead, participants were able to draw freely on their
combined pool of linguistic resources when appropriate to make meaning and to express desired
output. Conversation analysis of their group planning discussion found the unique way in which
participants pulled on resources from both named languages (English and Japanese) to discuss
relevant content, engage in bilingual problem-solving strategies (Velasco & García, 2014), and,
unlike the English-only and weak translanguaging groups, to manage the task (Kibler, 2010;
Storch & Wigglesworth, 2003) and maintain the focus of the discussion with key words important
to the English-medium essay they were required to produce. Although the participants in the present
study did not extensively alter between both named languages during their discussions, focusing
largely on their native language, Japanese, this may be simply because they deemed the employment
of these linguistic features to be more amenable to the task at hand or perhaps because they were
largely unaware of how best to draw freely upon the features in their expanded linguistic repertoires
in an integrated manner. If it is the latter, it may be the case that explicit training on how to engage
in strong translanguaging practices in cooperation with preexisting native language practices (García
& Wei, 2014) is required to maximize the benefits of a strong translanguaging approach to EFL
education.
As García and Wei (2014) explain, a (strong) translanguaging approach to language education is
transformational in that it opposes strict language control policies that have long dominated the
prominent teaching methodologies in FL learning. Of course, the argument could also be made that
certain tasks, depending largely on their design, may be better supported by weak translanguaging or
monolanguaging English-only practices with clear and structured support by the teacher. In the
present study, for example, it may be the case that Group 1 (EO) might have talked more had the
teacher given them additional language tools such as sentence stems, discussion prompts, or more
specific questions. However, it is beyond the scope of this article to make these claims with certainty.
As was seen in the present study, those learners engaging in strictly controlled languaging practices,
both in the TL and in their home language; exhibited disadvantages such as disjointed conversations
due to a lack of language knowledge in the English-only group; and negative transfer, confusion, and
a lack of focus on the TL in the Japanese-only weak translanguaging group. Only by engaging in
strong translanguaging did learners experience a noncompetitive, flexible perspective on the lan-
guages and language knowledge and practices in their unique, integrated linguistic repertoires. This,
16 B. TURNBULL

in turn, has the potential to result in more focused discussions and composition pieces of a higher
quality than from those learners engaging in monolanguaging or weak translanguaging practices.

Limitations and future research

To begin with, it must be acknowledged that the employment of common metrics, such as calculat-
ing word counts and summarizing composition scores, only provides one dimension of a learner’s
language and literacy abilities. Certain limitations exist with conversation analysis also. Although it is
possible that some misinterpretations remain, member-checking with the participants was employed
in the present study to confirm that all interpretations and understandings of the conversations were
accurate and to reduce the limitation of meaning misunderstandings. Future research of a similar
structure to the present study should be conducted to fulfill some of these limitations and other areas
not addressed in this article. For example, a future study may involve more participants (both per
group and per class), as well as more classes, to compare the results overall for greater validity.
A future study may also examine the role that learners’ proficiency levels play on the effects of
translanguaging by involving learners of beginner and advanced levels. Having learners alternate
groups (academic and creative) and also the languages in which they interact and discuss may also
help to do away with the limitation of having the same learners interact in the same manner.
Although the present study centered on the tertiary level, that is certainly not to say that the findings
do not apply to junior and senior high school (in which, arguably, the use of the L1 is higher).
Furthermore, because the present study was a cross-sectional investigation of a single writing task,
future research may investigate the longitudinal effects of a translanguaging approach across an
entire semester or year-long course on learners’ EFL writing. Other potential research ventures
would be to investigate the effects of EFL learners from linguistic backgrounds other than Japanese
(e.g., French learners of English) and also on learners of a second or foreign language other than
English (e.g., English learners of Chinese). This would help to develop a broader picture of the effects
that translingual practices can have on learners’ FL/L2 composition and to provide further evidence
in support of the beneficial role of having learners engage in strong translanguaging practices over
monolingual or even weak translanguaging in the FL classroom.

Conclusion
Although movements in FL writing have been observed in the past, it seems evident that a further
change is necessary but yet to be seen on a broad scale. Developments toward a holistic, bilingual
approach to second and foreign language learning based on the concepts of multicompetence (see
Cook, 1991) and translanguaging (García & Wei, 2014) are required now as we look toward
modernizing the way in which foreign languages are taught, learned, and assessed in the 21st century
(also see Author, 2017, 2018). The present study argues in favor of introducing strong translangua-
ging practices to the FL classroom, whereby the socially constructed barriers between named
languages are removed, and learners are encouraged to draw upon the integrated set of linguistic
features and language experiences that exist in their expanded linguistic repertoires to learn. This
study has shown that by allowing FL learners to engage in strong translanguaging practices during
the planning stage of their English composition, learners are able to score higher than those who are
forced to use one language over the other (either the TL or their native language) by producing more
concise, well-formed essays. They are also less likely to misuse lexical items as a result of negative
transfer or from a lack of relevant language knowledge or experiences. However, because all groups
were required to produce their final composition pieces in English (given the EFL nature of the
course), it must be acknowledged that allowing those learners in the strong translanguaging group to
discuss and plan their essays in whichever language before composing their texts in English is
effectively asking them to select and suppress some linguistic features in their repertoires, thus
 BILINGUAL RESEARCH JOURNAL 17

combining both weak and strong forms of translanguaging, which is both important and beneficial
to language education overall (see García & Lin, 2017).
In a specifically Japanese EFL context, Turnbull (2018) suggested there to be a theoretical
potential for a translanguaging approach to EFL education, based on the reported use of Japanese
in the English classroom and the conflicts that exist between government policy and classroom
practices, and the present article has provided empirical evidence to further support this claim.
However, the implications these findings have extend beyond one localized EFL context and are
important as we look toward a modernization of FL education on the whole. It is pivotal that FL
educators and policy makers recognize the knowledge and experiences that learners bring with them
from their home language as a valuable resource for use in the FL classroom. The freedom and
confidence that allowing FL learners to engage in translanguaging practices as the emergent
bilinguals they are not only raises their ability to produce the TL in desired settings but also affords
them the ability to express themselves, to make meaning, and to learn as whole individuals acting in
their bi- or multilingual worlds. In viewing FL learners as emergent bilinguals, we accept the notion
of multicompetence in their minds and allow for the natural occurrence of (particularly strong)
translanguaging practices so learners may work to develop the weaker TL as an integrated part of
their holistic linguistic repertoires.

Note
1. Similar to the bilingual-tap method (see García, 2009).

ORCID
Blake Turnbull http://orcid.org/0000-0001-7930-2180

References
Anderson, J. (2017). Reimagining English language learners from a translingual perspective. ELT Journal. doi:10.1093/
elt/ccx029
Antón, M., & Dicamilla, F. J. (1999). Socio-cognitive functions of L1 collaborative interaction in the L2 classroom. The
Modern Language Journal, 83(2), 233–247. doi:10.1111/0026-7902.00018
Antoniou, M., & Moriarty, J. (2008). What can academic writers learn from creative writers? Developing guidance and
support for lecturers in higher education. Teaching in Higher Education, 13(2), 157–167. doi:10.1080/
13562510801923229
Atkinson, D., Crusan, D., Matsuda, P. K., Ortmeier-Hooper, C., Ruecker, T., Simpson, S., & Tardy, C. (2015).
Clarifying the relationship between L2 writing and translingual writing: An open letter to writing studies editors
and organization leaders. College English, 77(4), 383–386.
Baker, C. (2011). Foundations of bilingual education and bilingualism (5th ed.). Bristol, UK: Multilingual Matters.
Butzkamn, W., & Cadwell, J. (2009). The bilingual reform: A paradigm shift in foreign language teaching. Tübingen,
Germany: Narr Studienbücher.
Canagarajah, S. (2011). Translanguaging in the classroom: Emerging issues for research and pedagogy. Applied
Linguistics Review, 2(1), 1–28. doi:10.1515/9783110239331.1
Canagarajah, S. (2013). Translingual practice: Global Englishes and cosmopolitan relations. New York, NY: Routledge.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum
Associates.
Conteh, J., & Meier, G. (Eds.). (2014). The multilingual turn in languages education: Benefits for individuals and
societies. Bristol, UK: Multilingual Matters.
Cook, V. (1999). Going beyond the native speaker in language teaching. TESOL Quarterly, 33(2), 185–209.
doi:10.2307/3587717
Cook, V. (2001). Using the first language in the classroom. Canadian Modern Language Review/La Revue Canadienne
Des Langues Vivantes, 57(3), 402–423. doi:10.3138/cmlr.57.3.402
Cook, V. J. (1991). The poverty-of-the-stimulus argument and multi-competence. Second Language Research, 7(2),
103–117. doi:10.1177/026765839100700203
18 B. TURNBULL

Cook, V. J. (1992). Evidence for multicompetence. Language Learning, 42(4), 557–591. doi:10.1111/j.1467-1770.1992.
tb01044.x
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal
Learning and Verbal Behaviour, 11(6), 671–685. doi:10.1016/S0022–5371(72)80001–X
Creese, A., & Blackledge, A. (2010). Translanguaging in the bilingual classroom: A pedagogy for learning and
teaching? The Modern Language Journal, 94(1), 103–115. doi:10.1111/j.1540-4781.2009.00986.x
Cummins, J. (2007). Rethinking monolingual instructional strategies in multilingual classrooms. Canadian Journal of
Applied Linguistics/Revue Canadienne De Linguistique Appliquee, 10(2), 221–240.
Fu, D. (2003). An island of English: Teaching English in Chinatown. Portsmouth, NH: Heinemann.
García, O. (2009). Bilingual education in the 21st century: A global perspective. Malden, MA: Wiley & Blackwell.
García, O. (2017). Reflections on Turnbull’s reframing of foreign language education: Bilingual epistemologies.
International Journal of Bilingual Education and Bilingualism. Advanced online publication. doi:10.1080/
13670050.2016.1277512
García, O., & Kano, N. (2014). Translanguaging as process and pedagogy: Developing the English writing of Japanese
students in the US. In J. Conteh & G. Meier (Eds.), The multilingual turn in languages education: Benefits for
individuals and societies (pp. 258–277). Bristol, UK: Multilingual Matters.
García, O., & Lin, A. M. Y. (2017). Translanguaging in bilingual education. In O. García & A. M. Y. Lin (Eds.),
Bilingual and multilingual education (Encyclopedia of Language and Education, Vol. 5) (pp. 117–130). Dordrecht,
Netherlands: Springer International Publishing.
García, O., & Wei, L. (2014). Translanguaging: Language, bilingualism and education. Basingstoke, UK: Palgrave
Macmillan.
Gort, M. (2012). Code-switching patterns in the writing-related talk of young emergent bilinguals. Journal of Literacy
Research, 44(1), 45–75. doi:10.1177/1086296X11431626
Grabe, W., & Kaplan, R. (1996). Theory and practice of writing. Harlow, UK: Longman.
Hornberger, N. H., & Link, H. (2012). Translanguaging and transnational literacies in multilingual classrooms:
A biliteracy lens. International Journal of Bilingual Education and Bilingualism, 15(3), 261–278. doi:10.1080/
13670050.2012.658016
Horner, B., Lu, M. Z., Royster, J. J., & Trimbur, J. (2011). Language difference in writing: Toward a translingual
approach. College English, 73(3), 303–321.
Howatt, A. (1984). A history of English language teaching. Oxford, UK: Oxford University Press.
Jacobs, H. L., Zinkgraf, S. A., Wormuth, D. R., Hartfiel, V. F., & Hughey, J. B. (1981). Testing ESL composition:
A practical approach. Rowley, MA: Newbury House Publishers.
Kibler, A. (2010). Writing through two languages: First language expertise in a language minority classroom. Journal
of Second Language Writing, 19(3), 121–142. doi:10.1016/j.jslw.2010.04.001
Kinginger, C. (2004). Alice doesn’t live here anymore: Foreign language learning and identity reconstruction. In
A. Pavlenko & A. Blackledge (Eds.), Negotiation of identities in multilingual contexts (pp. 219–242). Clevedon, UK:
Multilingual Matters.
Kobayashi, H., & Rinnert, C. (1992). Effects of first language on second language writing: Translation versus direct
composition. Language Learning, 42(2), 183–215. doi:10.1111/j.1467-1770.1992.tb00707.x
Kubota, R. (1998). Ideologies of English in Japan. World Englishes, 17(3), 295–306. doi:10.1111/1467–971X.00105
Leung, C., & Scarino, A. (2016). Reconceptualising the nature of goals and outcomes in language/s education. The
Modern Language Journal, 100(S1), 81–95. doi:10.1111/modl.12300
Levine, G. (2011). Code choice in the language classroom. Bristol, UK: Multilingual Matters.
Lewis, G., Jones, B., & Baker, C. (2012a). Translanguaging: Origins and development from school to street and beyond.
Educational Research and Evaluation, 18(7), 641–654. doi:10.1080/13803611.2012.718488
Lewis, G., Jones, B., & Baker, C. (2012b). Translanguaging: Developing its conceptualisation and contextualization.
Educational Research and Evaluation, 18(7), 655–670. doi:10.1080/13803611.2012.718490
Liebscher, G., & Dailey-O’Cain, J. (2005). Learner code-switching in the content-based foreign language classroom.
The Modern Language Journal, 89(2), 234–247. doi:10.1111/j.1540-4781.2005.00277.x
Lin, A. (2015). Conceptualising the potential role of L1 in CLIL. Language, Culture and Curriculum, 28(1), 74–89.
doi:10.1080/07908318.2014.1000926
Macaro, E. (2014). Overview: Where should we be going with classroom codeswitching research? In R. Barnard &
J. McLellan (Eds.), Codeswitching in university English-medium classes: Asian perspectives (pp. 10–23). Bristol, UK:
Multilingual Matters.
MacSwan, J. (2017). A multilingual perspective on translanguaging. American Educational Research Journal, 54(1),
167–201. doi:10.3102/0002831216683935
Mahboob, A., & Lin, A. (2016). Using local languages in English language classrooms. In H. Widodo & W. Renandya
(Eds.), English language teaching today: Building a closer link between theory and practice (pp. 25–40). New York,
NY: Springer International.
Makoni, S., & Pennycook, A. (2007). Disinventing and reconstituting languages. Clevedon, UK: Multilingual Matters.
Markee, N. (2015). The handbook of classroom discourse and interaction. Chichester, UK: John Wiley & Sons, Inc.
 BILINGUAL RESEARCH JOURNAL 19

McKenzie, R. M. (2010). The social psychology of English as a global language: Attitudes, awareness and identity in the
Japanese context. Dordrecht, Netherlands: Springer.
McKinley, J. (2018). Making the EFL to ELF transition in English-medium instruction at a global traction university.
In A. Bradford & H. Brown (Eds.), English-medium instruction in Japanese higher education (pp. 238–249). Bristol,
UK: Multilingual Matters.
McVeigh, B. J. (2002). Japanese higher education as myth. New York, NY: M. E. Sharpe.
Pike, K. (1967). Language in relation to a unified theory of the structure of human behavior (2nd ed.). The Hague,
France: Mouton.
Randolph, P. T. (2011). Using creative writing as a bridge to enhance academic writing. MITESOL 2011 Selected
Conference Proceedings, 70–83.
Skehan, P., & Foster, P. (1997). Task type and task processing conditions as influences on foreign language
performance. Language Teaching Research, 1(3), 185–211. doi:10.1177/136216889700100302
Storch, N., & Wigglesworth, G. (2003). Is there a role for the use of the L1 in an L2 setting? TESOL Quarterly, 37(4),
760–769. doi:10.2307/3588224
Tosu, N. (1997). Linguistic imperialism and teaching English in Japan. The Geibun-Kenkyu: Journal of Arts and Letters,
73(12), 508–517.
Tsuda, Y. (1993). Eigoshihai eno iron (Arguments against English dominance). Tokyo, Japan: Daisanshobo.
Turnbull, B. (2017). Towards new standards in foreign language assessment: Learning from bilingual education.
International Journal of Bilingual Education and Bilingualism. Advanced Online Publication. doi:10.1080/
13670050.2017.1375891
Turnbull, B. (2018). Is there a potential for a translanguaging approach to English education in Japan? Perspectives of
tertiary learners and teachers. JALT Journal, 40(2), 101–134.
Turnbull, B. (2018). Reframing foreign language learning as bilingual education: Epistemological changes towards the
emergent bilingual. International Journal of Bilingual Education and Bilingualism, 21(8), 1041–1048. doi: 10.1080/
13670050.2016.1238866
Turnbull, M., & Dailey-O’Cain, J. (2009). Introduction. In M. Turnbull & J. Dailey-O’Cain (Eds.), First language use
in second and foreign language learning (pp. 1–14). Bristol, UK: Multilingual Matters.
Velasco, P., & García, O. (2014). Translanguaging and the writing of bilingual learners. Bilingual Research Journal, 37
(1), 6–23. doi:10.1080/15235882.2014.893270
Wang, W., & Wen, Q. (2002). L1 use in the L2 composing process: An exploratory study of 16 Chinese EFL writers.
Journal of Second Language Writing, 11(3), 225–246. doi:10.1016/S1060-3743(02)00084-X
Wei, L. (2011). Multilinguality, multimodality, and multicompetence: Code-and modeswitching by minority ethnic
children in complementary schools. The Modern Language Journal, 95(3), 370–384. doi:10.1111/j.1540–-
4781.2011.01209.x
Williams, C. (1996). Secondary education: Teaching in the bilingual situation. In C. Williams, G. Lewis, & C. Baker
(Eds.), The language policy: Taking stock (pp. 39–78). Llangefni, UK: CAI.
Wong, J., & Zhang Waring, H. (2010). Conversation analysis and second language pedagogy: A guide for ESL/EFL
teachers. New York, NY: Routledge.
20 B. TURNBULL

Appendix: ESL Composition Profile (adapted from Jacobs et al., 1981)

ESL Composition Profile Score:

Score Level Criteria
Content 30–27 EXCELLENT TO VERY GOOD: knowledgeable • substantive • thorough development of thesis • relevant
to assigned topic
26–22 GOOD TO AVERAGE: some knowledge of subject • adequate range • limited development of thesis •
mostly relevant to topic but lacking detail
21–17 FAIR TO POOR: limited knowledge of subject • little substance • inadequate development of topic
16–13 VERY POOR: does not show knowledge of subject • nonsubstantive • not pertinent • OR not enough to
evaluate
Organization 20–18 EXCELLENT TO VERY GOOD: fluent expression • ideas clearly stated/supported • succinct • well-
organized • logical sequencing • cohesive
17–14 GOOD TO AVERAGE: somewhat choppy • loosely organized but main ideas stand out • limited support •
logical but incomplete sequencing
13–10 FAIR TO POOR: nonfluent • ideas confused or disconnected • lacks logical sequencing and development
9–7 VERY POOR: does not communicate • no organization • OR not enough to evaluate
Vocabulary 20–18 EXCELLENT TO VERY GOOD: sophisticated range • effective word/idiom choice and usage • word form
mastery • appropriate register
17–14 GOOD TO AVERAGE: adequate range • occasional errors or word/idiom form, choice, usage but
meaning not obscured
13–10 FAIR TO POOR: limited range • frequent errors or word/idiom form, choice, usage • meaning confused or
obscured
9–7 VERY POOR: essentially translation • little knowledge of English vocabulary, idioms, word form • OR not
enough to evaluate
Language 25–22 EXCELLENT TO VERY GOOD: effective complex constructions • few errors of agreement, tense, number,
use word order/function, articles, pronouns, prepositions
21–18 GOOD TO AVERAGE: effective but simple constructions • minor problems in complex constructions •
several errors of agreement, tense, number, word order/function, articles, pronouns, prepositions but
meaning seldom obscured
17–11 FAIR TO POOR: major problems in simple/complex constructions • frequent errors or negation,
agreement, tense, number, word order/function, articles, pronouns, prepositions and/or fragments, run-
ons, deletions • meaning confused or obscured
10–5 VERY POOR: virtually no mastery or sentence construction rules • dominated by errors • does not
communicate • OR not enough to evaluate
Mechanics 5 EXCELLENT TO VERY GOOD: demonstrates mastery of conventions • few errors of spelling, punctuation,
capitalization, paragraphing
4 GOOD TO AVERAGE: occasional errors of spelling, capitalization, paragraphing but meaning not
obscured
3 FAIR TO POOR: frequent errors of spelling, punctuation, capitalization, paragraphing • meaning confused
or obscured
2 VERY POOR: no mastery or conventions • dominated by errors of spelling, punctation, capitalization,
paragraphing • OR not enough to evaluate