Understanding
multilingual self for language minority students from Asia and Europe
Mark
Feng Teng1* and Lawrence Jun Zhang2
|
*Correspondence: markteng@mpu.edu.mo |
|
1Macao Polytechnic University, Macau SAR 2University of Auckland, Auckland, New Zealand |
Abstract
Despite
the extant literature, studies on how multilingual learners’ motivations for
L1, L2, and L3 co-evolve and shape a multilingual self is scant. Drawing on the
L2 Motivational Self System and dynamic–ecological perspectives, we make an attempt to fill this research gap. We conceptualize
motivation as emergent from interactions among ideal/ought-to selves, lived
experiences, and contextual affordances. Interviews with 15 language-minority
learners from Thailand, Europe, and Hong Kong reveal coupled motivational
dynamics: L2 English often forms a stable attractor in high-capital ecologies
(e.g., Hong Kong), while L3 (Japanese/Chinese) tends to remain vulnerable
unless identity relevance, utility, and participation intensify. In Thailand,
rising instrumental value of Chinese perturbs Ought-to–dominated English
trajectories, enabling L3 stabilization. European learners in China face
participation constraints that keep L3 Chinese peripheral despite immersion,
whereas typological proximity (e.g., English–French)
supports co-stability via metalinguistic transfer. We propose the Multilingual
Self as a Coupled Attractor System, detailing stability gradients and
capital/linguistic/participation couplings, and offer
suggestions for pedagogical interventions to deepen vulnerable basins and
sustain balanced multilingual development.
Keywords:
Multilingual self, motivation, identity, multilingual learning
1. Introduction
Being someone who
speaks more than one language seems to many a distinctive privilege. Contrary
to this lay view assuming that most people usually
speak only their mother tongue, being multilingual or “potentially multilingual
by nature” is in fact a desired “state of linguistic competence” (Hammarberg,
2009, p. 2). Multilingual education is thus expected to enhance the linguistic
and intercultural competence of multilingual learners (Liu, 2022; Teng &
Wang, 2020). However, complexity of first language (L1), second language (L2),
and third language (L3) motivational systems may lead to changes related to a
multilingual self (Bui et al., 2018; Henry & Thorsen, 2018). Investigations
into multilingual learners’ interaction of L1, L2, and L3 motivation are thus
of significant research value (Dörnyei & Ushioda, 2021). Concomitantly,
breakthroughs in motivation theory or framework offer new angles or call for
new needs for reconceptualizing different dimensions of motivation.
In
early studies, researchers (e.g., Dörnyei, 2005) proposed the L2 Motivational
Self System, a starting point of perceiving motivation as a dynamic notion
rather than a static one. Motivation was then widely perceived as researching
learners’ ideal selves (i.e., learners’ imaginings of using that language in a
future situation), their ought-to selves (i.e., learners’ responsibility to use
that language), and their positive or negative language learning experiences.
An interplay between language learners’ selves and language learning
experiences may provide insights into the dynamics of language learning
motivation (Alamer & Alrabai, 2023; Dörnyei &
Ushioda, 2009; Henry, 2023). The present study examines the phenomena of L1,
L2, and L3 learning, as well as the motivational dynamics underlying L2 and L3
acquisition, as complex, interconnected systems situated within multilingual
education. Building on dynamic and ecological perspectives (Ushioda, 2020;
Henry, 2023), we view multilingual motivation as emergent from interactions
among learners’ multilingual experiences, contexts, and identities.
Accordingly, we propose a conceptual framework of the multilingual self to
capture the evolving nature of L1–L2–L3 motivational forces, integrating
ideal/ought-to self-guides with learners’ lived
experiences and translanguaging practices, that can be applied to multilingual
learning design and pedagogy.
2. Literature review
2.1 Motivation
The
notion of motivation for multilingual learning is receiving increasing
attention. In some traditions, motivation is defined as a unitary, fixed, and
ahistorical trait that empowers a learner to act (Reeve, 2024). In multilingual
learning, however, motivation is generally described as a dynamic process of
arousing, directing, and sustaining behavior for
learning different languages (Teng & Wang,
2020). A key element in
researching motivation is establishing how contextual and situational factors
influence learners’ intrinsic and extrinsic motivation, which, in turn, affects
their engagement and persistence (Noels, 2001; Teng, 2025). Learners who study
a language out of love or passion demonstrate intrinsic motivation and are more
likely to internalize learning objectives as part of their language learning
process. In contrast, learners who study languages to pass exams, secure
employment, or meet parental expectations may be extrinsically motivated, as
their motivation stems from external pressures. Noels et al. (2001)
conceptualized language learning motivation as a continuum, with amotivation at
one pole (e.g., “I am not motivated to learn this language because I don’t like
it”) and intrinsic motivation at the other (e.g., “I am motivated to learn this
language because I like it”), with different forms of extrinsic motivation in
between. When learners can self-regulate their motivation, their intrinsic
motivation may be strengthened (Ryan & Deci, 2020).
Motivation is a central aspect of multilingual
education. In early studies, motivation was described as comprising learners’
attitudes, behaviors, and desires to learn a language
(e.g., Gardner, 2010). Relatedly, two broad types of motivation were
identified: instrumental and integrative (Dörnyei, 2009). For students who wish
to become successful language users, integrativeness
may align with their ideal self, whereas instrumental motivation parallels
extrinsic motivation. However, the sustained process of motivational engagement
is not fully captured by the instrumental–integrative distinction. To address
this, Dörnyei et al. (2015) proposed directed motivational currents (DMCs),
defined as “a prolonged process of engagement in a series of tasks which are
rewarding primarily because they transport the individual towards a highly
valued end” (p. 98). In a DMC, learners’ individual learning activities become
integrated components of a superordinate, complex motivational structure. Once
set in motion, learners are propelled toward a goal or vision, akin to a
current that carries along everything entering its flow. Two defining features
of DMCs are their directedness and the durability of their self-propelling
motivational processes, which enrich our understanding of intrinsic motivation.
A common finding across the studies reviewed above
is that motivation is dynamic and situated. For example, Ushioda (2020)
emphasized the learning context as learners’ continuous pursuit of personal
meaning-making. Social settings reshape learners’ understandings of contexts
and their motivations for language learning. In an empirical study, Mercer
(2011) showed that the dynamics of learners’ motivation are connected to their
engagement across different contexts. Although the dynamic nature of motivation
in multilingual learning, including L2 motivation and L3 motivation (Bui et
al., 2018; Bui & Teng, 2021), has been explored, dual motivation, that is,
the mutual influences involved in learning two languages simultaneously, has
not been sufficiently examined in a global context. Given the unique status of
English and the differing motivations for learning a second versus a third
language, L2 and L3 motivational systems may be especially dynamic. Despite
growing recognition of motivational dynamics, research remains limited,
particularly longitudinal studies that track L2 and L3 motivational
trajectories.
2.2 Multilingual self
Students
learning both an L2 and an L3 may share a unifying vision of becoming
multilingual. In this sense, motivation for multilingual learners can function
as a “task force that effectively promotes positive attitudes about
multilingualism” (Dörnyei & Kubanyiova, 2014, p.
83). When learners enter motivational currents—a state in which they are
absorbed in and enjoy learning activities—they are more likely to sustain
engagement in a particular kind of learning (Dörnyei et al., 2015). Any account
of motivation is incomplete without considering the visionary qualities and
motivational properties that learners hold while pursuing both an L2 and an L3.
For instance, learners may form a future-oriented image of themselves as
successful speakers of both languages. This image evolves into an ideal self
that they aspire to and that becomes part of their core identity (Teng &
Bui, 2020). In such contexts, the ideal self serves as an active
self-representation, closely linked to learners’ lived
language learning experiences. Learners who attach greater importance to these
motivational qualities tend to sustain their efforts and invest more in
learning. While learners may readily acquire motivation, motivation for L2 and
L3 learning can diverge, and not all forms of motivation generate powerful or
enduring pathways of learning behavior. Thus, both
the shared visionary qualities across L2 and L3 learning and the differences in
what motivates learners for each language can shape self-concordant goals and
proximal subgoals, key considerations for a fuller account of motivational
dynamics in multilingual learning.
Furthermore, the ideal language self—imagining
oneself using the language in future contexts, and the ought-to language self,
the desire to meet others’ expectations, are mediated by learners’ actual
experiences. Because learners encounter different circumstances across L1, L2,
and L3 learning, their ideal and ought-to selves for each language may differ,
reflecting the dynamic nature of multilingual motivation. Hajar (2024)
investigated the motivations and overarching visions of internally displaced Syrians
in Afrin, on the Syrian–Turkish border, who were
learning Turkish as an additional language (L3 or L4). Drawing on participants’
written narratives and two waves of individual online semi-structured
interviews, the study found a spectrum of motives. These ranged from pragmatic
aims, such as improving employment prospects in Afrin, to future-oriented
aspirations, including envisioning themselves studying or working at Turkish
universities and integrating into Turkish society. Zhang (2025) examines how
multilingual identity shapes cross-cultural adaptation by tracing the identity
construction and adjustment strategies of three multilinguals from diverse
linguistic and cultural backgrounds living in Hong Kong. Drawing on semi-structured
interviews, informal conversations, fieldnotes, and journals, the study finds
that none aimed for full integration into the host society; instead, they
adapted effectively while affirming distinct multilingual selves. A cross-case
analysis shows that all three exercised agency,
leveraging their multilingual repertoires and intercultural communicative
competence to navigate interactions with the local community. Wang and Wang
(2024) examined 50 Chinese undergraduates majoring in two foreign languages.
Using Q methodology grounded in the L2 Motivational Self System, they explored
the variety of motivation types and cross-linguistic differences in
multilingual motivation. The findings indicate that a multilingual self concept often initiates students’ choice of
dual-language majors. Importantly, the ought-to self outweighed the ideal self in shaping motivation.
Moreover, students specializing in different L3s displayed distinct
motivational profiles. As Dörnyei (2009) suggests, when learners’ ideal selves
become stable, feasible, and elaborated, and are balanced by their ought-to
selves, motivation during learning is enhanced. Since language learning extends
beyond the classroom into social contexts, multilingual motivation becomes even
more dynamic and complex, shaped by contextual affordances for using the target
language. The ideal self, the ought-to self, and the learning experience are
not isolated constructs but closely interrelated. In other words, learners’ envisioned future language use and their current
obligations influence, and are influenced by, their situated learning
processes.
2.3 The present study
The
complex relationship between L1, L2, and L3 was the basis for exploring
multilingual self for language minority students, which remains underexplored.
More in-depth qualitative studies can complement knowledge in this area. The
present study explored Asian and European learners’ developmental trajectories
of motivations in learning L1, L2, and L3 and how the motivational dynamics
were formed with relation to a possible multilingual self. This study focused
on two research questions:
(1)
Do
the learners demonstrate a multilingual self during multilingual learning?
(2)
Is
there a change of focus within the multilingual self?
(3)
How
do the learners respond to the multilingual self?
3. Method
3.1 Participants
The sample of
participants included five students from Thailand, and five students from
Europe, and five students from Hong Kong. The first language for the Thai
learners were Thai, and English and Chinese were additional languages. The
first language for the European learners was either English, Greece, or German,
and Chinese as an additional language. The first language for the Hong Kong
students was Cantonese Chinese, and English and Japanese were additional
languages. The Thai learners were studying in mainland China. The European
participants were either working in mainland China or studying in Hong Kong.
The Hong Kong learners were studying in Hong Kong. The participants speak a
language at home other than the dominant language of a society. They were thus
defined as language minority learners who have multilingual learning
experiences. The following table delineates the participants’ information. All
names were pseudonym names.
Table
1. Profiles
of the participants
|
Regions |
Name |
Gender |
L1 |
L2 |
L3 |
Other languages |
Occupation |
|
Europe |
William |
Male |
English |
French |
Chinese |
-- |
EFL teacher in China |
|
Carolyn |
Female |
German |
English |
Japanese |
Latin, French, Chinese |
PhD Candidate in Hong Kong |
|
|
Adreian |
Male |
Greek |
English |
French |
Dutch, Italian, |
PhD Candidate in Hong Kong |
|
|
Fiona |
Female |
English |
French |
Chinese |
-- |
EFL teacher in China |
|
|
Sam |
Male |
English |
French |
Chinese |
-- |
EFL teacher in China |
|
|
Asia |
Kawaek |
Female |
Thai |
English |
Chinese |
-- |
Student in China |
|
Achara |
Female |
Thai |
English |
Chinese |
Japanese |
Student in China |
|
|
Aat |
Male |
Thai |
English |
Chinese |
-- |
Student in China |
|
|
Dungkmol |
Female |
Thai |
English |
Chinese |
-- |
Student in China |
|
|
Aurak |
Male |
Thai |
English |
Chinese |
|
Student in China |
|
|
Wong |
Male |
Cantone |
English |
Japanese |
-- |
Student in Hong Kong |
|
|
Sue |
Female |
Cantone |
English |
Japanese |
-- |
Student in Hong Kong |
|
|
Chan |
Male |
Cantone |
English |
Japanese |
Korean |
Student in Hong Kong |
|
|
Lee |
Female |
Cantone |
English |
Japanese |
-- |
Student in Hong Kong |
|
|
Yip |
Female |
Cantone |
English |
Japanese |
-- |
Student in Hong Kong |
3.2 Data collection
The
main method for data collection in the present study is through interviews,
which is a reliable method for generating in-depth explanations of and
capturing the dynamics of learners’ multilingual self. Through interviews,
researchers can also collect learners’ life stories in a comprehensive way. In
addition, interviews are effective at exploring learners’ individuality and
obtaining insights into the dynamic interactions between learners and their
context (Ushioda, 2020). In the present study, three rounds of interviews were
conducted. The first round of interviews focused on the family background, the
first, second, and third languages that the learners spoke, and the motivations
for learning each language. The second interview focused on learners’
motivation change after learning different languages. The second interview also
explored learners’ life stories in connection to learning each language. The
third interview focused on overall reflection on the language learning
experiences and their imagined future life related to the use of languages. All
interviews were semi-structured. Each interview lasted for approximately one
hour. During the interviews, if participants mentioned incidents related to
their ideal or ought-to selves, they were encouraged to elaborate more on the
incidents. The language for the interviews was in English.
3.3 Data analysis
The
transcribed data was coded in an iterative way. First, data was analysed in a
grounded manner. This could help identify learners’ motivations for learning
different languages at different stages. Second, the analysed data was then
analysed from different aspects of motivations, i.e., ought-to self, ideal
self, and language learning experiences under the bigger categories of
languages, i.e., L1, L2 and L3. Third, the different category data was compared
and subsequently cross examined to explore the complexity of language learning
motivations. Fourth, the interviewees also consulted the analysed data to avoid
skewed understanding and interpretations of the data. Finally, the analysed
data were presented based on the emerging themes of the participants.
4. Results
4.1 Existence of a Multilingual Self
The existence of a
multilingual self was related to the possible interplay of L1, L2, and L3.
L1 influence on L2 and L3
The Thai learners
labeled themselves as lacking instrumental motivation and being reluctant to
learn and practise Chinese (Teng & Bui, 2020).
Our data suggest that the Thai learners found Chinese language as an
instrumentally important language for them. Learning Chinese as an additional
L3 is a means of obtaining upward mobility. Aurak, one Thai learner, used his
job-seeking experiences to highlight the importance of learning Chinese:
For
me, Chinese is one of the most important languages in Thailand. For example,
when I looked a job in international company, surely
they would ask me, “Can you speak English?” Then they would ask me, “Can you
speak Chinese?” Since I said, “only a little bit.” Then they said, “Sorry, we
are looking for somebody who are good in Chinese.” It is all those experiences
that made me think Chinese language is a very important additional language
though it is really a difficult language as we have become accustomed to first
language system and influenced by second language structures. (Aurak, interview
1)
The
local employment agency in Thailand prefers candidates with a certain level of
knowledge of Chinese language and culture. As learners who are learning Chinese
as a third language, Anurak seemed to be at a considerable disadvantage, as
learning a third language had been greatly influenced by the first and second
language learning system. This could be explained as L1 negative influence on
L3. However, due to a need for instrumental resources, Anurak aligned with the
power of the Chinese language and highlighted the importance of Chinese
language for success in hunting jobs.
The
Hong Kong learners expressed similarities between Chinese and Japanese. The
similar linguistic systems led to their comfort in learning Japanese as an L3.
For instance, both Yip and Wong mentioned in the interview that “Japanese is
quite similar to Chinese. It should be easier to learn Japanese than English.”
Wong further mentioned the proximity of Japanese culture to Chinese culture.
Related to this, he said,
It
is always an enjoyment to learn Japanese as it makes me have a familiar feeling
because I am a Japanese Manga fan” (Interview 1).
Based
on the above quote, L1 Chinese intrigues the Chinese learners to be curious
about the L3 Japanese. Chinese assists in encouraging Wong to look deeper into
Japanese language and culture. The perceived similarity between the two
languages helped the learners gain momentum in building an L3 self (Bui &
Teng, 2021). In this case, Chinese aided a sense of achievement, a feeling that
learners may regard as desirable for an L3 self. To some extent, L1 could
positively influence L3 motivation.
The
European learners expressed the difficulties in learning Chinese in the
Chinese-speaking context. This could be described as a type of diasporic
experience, because being an L1 English native speaker has created a power to
involve Chinese students in an English-speaking ideology. The separation of L1
English motivation and L3 Chinese motivation was reflected in our data. Sam
said:
Yes,
it was odd. When we want to speak Chinese to Chinese students, they want to practice
English with us. It was never easy to learn Chinese in China because there
would always be a contradiction between our motivation to learn Chinese and
Chinese learners’ craze to learn English. There always seems to be invisible
ideological separation. Then in China, I don’t see any motivation to learn
Chinese. (Sam, interview 2)
Language
learners’ investment in the target language was dependent on the perception of
symbolic and material resources (Dörnyei & Ushioda, 2009). Such resources
in turn strengthen the value of cultural capital (Norton, 2000). Social
otherness characterizes the Chinese language learning practices of European
learners in China. It means the challenges in being involved in target language
practices. It was in this context that European learners became less motivated
in learning Chinese because their culture and language were more valued by the
mainstream.
L2 influence on L3 motivation
If two languages are
perceived by the learners as typologically different, contradictory motive can
arise. The Hong Kong learners reported negative influence from their L2 English
on their L3 Japanese. The only positive influence was the English loanwords,
which could facilitate L3 learning. For example, Chan reported,
Learning
English influenced my Japanese learning. I think quite big influence. I don’t
think learning English is helpful for Japanese learning. Maybe only some
loanwords. I had a difficult time in adjusting myself to the differences
between Japanese and English. (Chan, interview 3)
In
the above quote, Chan realized an obvious connection between L2 and L3 learning
systems. The difference between L2 and L3 became a source of contradiction
between different language learning systems. The L2 learning experiences
inhibited Chan’s further efforts in L3 learning. The only positive influence
was from the English loanwords in Japanese. However, such influence was not
sufficient for bridging the gap between L2 and L3 motivational systems.
The
Thai learners also reported the negative influence from L2 to L3. For example,
Aat mentioned the significant difference between L2 English and L3 Chinese. He
reported:
I
am always very confused about the English structures and Chinese structures. It
may be because my English is not so good. Then suddenly I had to learn Chinese.
When the two languages were so different, I became even more confused. I wanted
to learn the two languages well, but it turned out to be I am weak at the two
languages. (Aat, interview 1)
According
to the above quote, Aat was comparing the L3 language system with the previous
L2 system in mind. In doing so, Aat developed an estimation of the difficulties
in simultaneously learning L2 and L3. Other Thai learners, e.g., Aurak, Dungkmol, also mentioned
the negative influence of their previous L2 learning system on learning L3
because their previously exposed L2 systems were significantly different from
the L3 systems. The differences between L2 and L3 learning systems stimulated
Aurak to perceive the L3 as more “more difficult” and thus “need more efforts
than L2 learning” (Aurak, reflective report).
The European learners reported that learning L2
would promote the vitality of the L3 self. Learning L2 would fuel a positive
attitude towards the L3. Adrian expressed a balanced distribution of L2 and L3
motivational forces within him. Adrian reported:
Languages sometimes help each other. In my case,
English helped me learn French. There are some differences in terms of grammar,
pronunciation, and sentence formation between English and French. I will try to
know in what way the two languages are similar. This method helps me learn more
French words. (Adrian, interview 2)
In the above quote, L2 and L3 selves could
harmoniously coexist to generate a multilingual self in a complex motivational
system. The rewarding learning experiences in L2 could foster positive language
learning motivations for L3. The comparison and contrast between L2 and L3
learning experiences facilitated one’s L3 motivation. Adrian was clearly aware
of the different ways how the two languages (English and French) were learned
(and taught), which fed into his coexistence of multilingual self.
An interesting finding is that the perceived
dissimilarity of L2 and L3 can also motivate some of the participants to learn
an L3, even though the L3 is considered more difficult than the L2. For
example, Fiona and William reported an instrumental need for L3 Chinese
learning. They did not attribute demotivation to their current perceptions of
L2 French, but they did make comparisons between the two languages. They said
they would have never been interested in learning Chinese, and staying in China
should be temporary and they would consider leaving it after the contract and
Chinese was not useful for them. In particular, William
described Chinese as “more complicated in terms of language structures and
learning is more difficult” and French “less complicated in terms of words and
structures and is easier to learn” (William, interview 1). This is the evidence
that Willam already had an initial awareness that learning L3 Chinese should be
more difficult than L2 French as he believed that L2 is relatively easier to
approach than L3. However, William later
reported that “the Chinese language is actually easier to acquire because
learning French helps me reflect on the Chinese grammatical system, which is
easier” (William, interview 3)
L3
influence on L2 motivation
The Hong Kong learners
exhibited a clear awareness of their linguistic repertoire related to their L2
and L3. Such an awareness allowed them to compare the linguistic influences
between the two languages. For example, Sue stated the influence of L3 on L2
pronunciation:
After
learning Japanese, I can understand why Japanese people cannot speak English
well. The pronunciation of Japanese loanwords is so weird judging from English
pronunciation system. I think my English pronunciation became worse after
learning Japanese. (Sue, interview 1)
In
the above quote, Sue stated the influence of L3 Japanese on L2 English
pronunciation. These opinions reflect the statuses of the two languages. This
is the evidence
for the cross-linguistic influence on pronunciation. Sue code-mixed the English
and Japanese pronunciation and regarded her English pronunciation became worse
after learning Japanese.
According
to the European learners, the lack of satisfactory outcome in L3 learning
became an external factor that can bring about changes to one’s L2 learning. Such perturbing learning outcome influences the
co-existed self in L2 and L3 learning. For example, Adrian reported:
After six years learning French at the public high
school, I still did not feel confident to talk with French speakers. We had to
follow the “speed” of the worst students in class. As a result, the proficiency
level acquired after six years of public school
classes were equivalent to one to two years in private tutorials. There was no
motivation. I gave it up. The influence was that I had to spend more time in
keeping up with my English learning because of the unnecessary time on French
learning (Adrian, interview 2).
In the above quote, Adrian stated the lack of
motivation in learning L3 French and the need to keep up with L2 English
learning. Adrian spoke about being less motivated as a
consequence of the learning results. He found it difficult to have any
perceptible changes in his L3 learning after six years of public school. This
long-accumulated unsatisfactory outcome is likely to have confirmed to him that
he was on the wrong direction to learn French. He felt it hard to summon
efforts to reduce the gap between his current and ideal L3 selves. This gap
could trigger an upward revision of his L2 learning experiences. Related to
this, he found it necessary to build an ideal L2 self.
4.2 Fluid Multilingual Selves
Across Contexts and Time
Compared with L2
English learning self, the Hong Kong students’ self in learning L3 Japanese
were relatively weak. Although they had learned Japanese for a
certain number of years, they could just read and find it difficult to speak.
Concerning the vulnerable L3 self, Chan wrote:
I
have learned Japanese for more than 7 years. I thought it would be easier to
learn Japanese as it has a lot of similarities with Chinese. I was wrong. I
felt quite embarrassed since I could only read but could not speak Japanese.
Sometimes when I want to speak Japanese, I automatically thought of some
English words in my mind. The decreased passion in Japanese learning
strengthened my English learning confidence. (Chan, reflective report)
It
seems that L3 learning involves greater degrees of complexity than L2 learning.
L3 learning self becomes vulnerable when L2 learning self
becomes stronger. The interactions between different L2 and L3 systems in the
learner’s mind created a sense of stability and vulnerability (Jessner, 2008). L3 learning is dependent on the development
of L2 learning systems. In particular, one or more of L3 Japanese
learning system’s components changes, e.g., Japanese were easier to read than
speak for Chinese learners, alter the state of the L2 English learning system.
This situation was obvious in Hong Kong where English plays a dominant
position. As Yip mentioned in his reflective report, “English occupies a very
dominant position in Hong Kong. The environment forces us to keep a strong mind
for English. Then we were left with few room for third
Japanese learning.” Consistent with Dörnyei (2009), the vulnerability and
stability dynamically change during negotiation and adjustment of different
language learning systems, and each language learning system changes in
response to the other.
The
European students tend to relate their changing L2 and L3 learning self to
their level of engagement in specific events or movement. The way that language
learning motivation changes from one day to another or from
one place to another place appears to be a function of internal consideration
and external influences. Adrian mentioned the constantly fluid L2 and L3
motivation in an interview:
My
family ran a restaurant. Greece is a tourist country. Every chance I had, I
practiced my poor English. French was offered as a compulsory course at the
public school from the age of 12. After six years learning French as an L3 I
still did not feel confident to engage in conversations with French speakers.
Later I went to university and moved to Britain, and then later to Hong Kong. I
realized one important thing which was that I have to
keep practicing my English and pronounced like a native speaker because
everyone around me thought I am a native English speaker from my appearance. I
don’t want to disappoint them. French, to me, is only a subject at the
secondary school. (Adrian, interview 3)
It
appears that language learning motivations go in currents and waves (MacIntyre & Legatto, 2011).
The changes in L2 and L3 motivations were dynamic and fluid. The various
internal and external factors, e.g., initial language learning conditions,
affective passion or mood in language learning, cognitive factors (e.g., the
difficulties in learning a language, a sense of living up to others’
expectations of being good at spoken English), and social factors (e.g., what
happened due to the changing communities), determine Adrian’s motivation at any
particular period. As shown in the above excerpt, these factors constantly
change across a longitudinal time scale. Across time and space, an enduring L2
stability has been strengthened accompanied with a decreased L3 self. The
trajectory of Adrian’s L2 motivational system, over his years of learning, have
evolved into a stable attractor state where key components in language learning
motivation (e.g., the Ideal, Ought-to L2 Selves, and the L2 Learning
Experience) converge in a mutually complementary manner.
The
Thai students also described how the intensity of motivated behavior would
vary. For example, Achara’s interview provides an indication of the
continuously fluctuated motivation:
I
started to learn English as early as kindergarten. I don’t remember quite well
but it was quite fun…But later, there were so many drills, exercises, and
dictation for passing the exams. I became unwilling to speak English. If you
don’t get motivated, you would become tired and give a damn in it …
and then let it go. Then I changed my mind to learn Chinese. I think Chinese is
more important than English, at least in Thailand. So
I came to China. I made progress. I became more motivated, then I kept doing
it. (Achara, interview 3)
Achara’s
L2 motivational system appears to be lodged in a deep-sided Ought-to Self dominated attractor state. Her desires or
self-regulated behavior in L2 learning
were mostly driven by the need to pass the exams.
While the motivations in securing exam scores impact her executive decisions,
such as doing drills, exercises, and dictation. This was only a level of
completing what is asked. A sense of willingness to speak English seems
lacking. Instead, Achara established a counterbalancing sense that learning
Chinese as an L3 might be personally meaningful or a skill that may be
beneficial for future use. As Duangkamol mentioned, “Chinese is an emerging
language that we can never underestimate. It is sweeping Thailand and possibly
other Southeast Asian countries.” Thus, to the Thailand students, the
motivational self-system components converge and influence their learning
behaviors and cognitive thinking and possibly, trigger shifts in L2 and L3
learning across the state space. The power of Chinese and the traditional
disadvantaged English teaching instruction create contextual parameters that
can foster conscious movement out of the Ought-to Self
attractor state. Under such movement, the learners seek for a different
area of the state space (e.g., a preference for learning a language for future
use). Such state space becomes a determining influence on the trajectory of
either a stable L3 self or vulnerable L2 self.
4.3 Balancing Acts:
Prioritization Dilemmas and Dynamic Motivations in Multilingual Self
All the Asian and
European learners, except Fiona, Karawek, Wong, and William, mentioned a need
to prioritize different language learning systems. The learners seem
to be in different dilemma and consistently deprioritize either L2 or L3 based
on their situation. For example, Sue, one Hong Kong learner, described her
dilemma related to English and Japanese learning experiences:
I studied in a direct subsidy secondary school.
That school offered Japanese course. There were a lot of things to do in
school. It feels Japanese is just an added language. It is impossible to
prioritize it as you don’t know much about it. So you
do other things first, then want to learn more Japanese, then you will find out
there is no time left for Japanese. And then, I understand that learning a
language is a life-long learning. I wanted to use the strategies I adopted for
learning English to Japanese learning, but it never worked. (Sue, interview 2)
In
the above quote, Sue perceived that learning a language is a life-long learning
journey. There were difficulties to adopt L2 English learning strategies for L3
learning. Resource and time were also limited to balance different language
learning needs, and there is a pressing need to divert more resources for L2
English learning. There appears to be a paradoxical logic to focus on L2 while
learning L3. It appears that Sue encountered the pressure in simultaneously
balancing the needs
for a range of different language learning systems, and it was difficult for
her to constantly channel time and cognitive resources for different language
learning systems. In line with Mercer (2011), during the process of allocating
resources to different learning demands, learners would always prioritize
certain activities over others.
When
reflecting on their motivation to learn L3 Japanese, the Hong Kong students
expressed their feelings in comparing with the other subjects they learn. As
Chan explained, “For subjects like history or general education, you learn
something and then you know it is related to your life” (Interview 3). Lee
added, “I don’t know how learning Japanese can help us. Finding a job? I am not
sure. Maybe for travel. You learn it, you forget it because you don’t use it”
(Interview 3). Yip expressed a need to recalibrate the achievement
expectations. For example, he mentioned, “For English subject, I had to aim for
an A, but for Japanese I would be happy with a B or even a C” (Interview 3). As
mentioned above, Wong is an exception. He had been learning Japanese for around
6 years and were still interested in making more progress. Wong stated,
I
have some family relatives in Japan. I want to go there and teach Chinese or
English in Japanese schools. So I have
to learn Japanese. I know that time and resources are limited. But I have to learn both English and Japanese well, which is an
asset for me to stay in Japan. Look around the political turmoil in Hong Kong.
I can’t imagine myself to call it my home. I have to
leave Hong Kong. (Wong, interview 3)
In
Wong’s response, complex systems do not exist in isolation but occur in
relation to the larger social system. The current social systems in Hong Kong
are interrelated with language learning motivational systems, impacting the
underlying motivations to learn an L2 or L3. In meeting different
cognitive demands related to language learning, Wong attempted to balance
different needs. Wong attempted to co-adapt cognitive resources as changes in
L3 learning system trigger changes in L2 learning system. The L2 and L3 systems
are adjacent.
When
we checked other exceptional cases, Fiona, Karawek, and William, their
similarities lay in their respective experiences of living and staying in their
target language community. Fiona and William had worked in France
for three years and then moved to China and had stayed in China for over 5
years. Karawek’s father is a Chinese. She was born and raised in Thailand and
then studied in America for her undergraduate study and then a master degree in China. The three learners all mentioned in
their reflective report that “learning an L2 and L3 is not contradictory.
Simultaneous learning of an L2 and L3 is not difficult if you understand the
specific use for each language.” They also mentioned that although their L2 is
totally different from L3 Chinese, their learning experiences in L2 can be
helpful for L3 learning. This is the evidence for the cross-referencing between
different language learning systems (Henry, 2014). The learners used L2 learning
experiences as an important source of support for L3 learning. Factors, higher
education needs and professional careers, enhanced their willingness to learn
their L2 or L3. Even in the case of Wong, he had never stayed in English
speaking or Japanese speaking countries. His imagination about the different
use of Japanese and English surely provides support for his multilingual
learning. He regarded English and Japanese as important and enjoyable learning
experiences for his desired futures. It appears that, in the cases of Fiona,
Karawek, and William, a period spent in a target language country had a major
perturbing effect on producing a stable L2 or L3 self. For Wong, the prospect
of going abroad triggers movement across the state space, leading to the
awareness of shifting the vulnerable L3 self to a stable self.
Other
learners, including Asian and European learners, mentioned the power of English
as an international language. For example, Carolyn and Aat, both stressed the importance of English as a global language in the interview.
Take Carolyn as an example:
One
thing I believe is that if your English is good you
can get many assets. It is a global language. Even when I travelled to France,
I didn’t feel it necessary to learn French. Now I come to Hong Kong. I don’t
think it necessary to learn Chinese too. It is normal to get less motivated
when you think that language is not useful for you. Speaking English makes you
so good because you have an advantage. It is a different image. (Carolyn,
interview 1)
In
the above quote, Carolyn regarded English as a language that functions all
around the globe. Such perceptions of attaching English to a greater value
influenced her L3 French and even L4 Chinese motivational systems. Such strong
ideal L2 self impacts
the ought-to L3 French-speaking/using self, for which Carolyn did not want to
learn and speak French in France. Similarly, her perception that learning
Chinese is not as rewarding as learning English influences her Chinese learning
experiences.
Aat
also stressed the importance of English as an international language. He used
“a global asset” to describe the advantage in speaking and using English
(Reflective report). However, he was also aware of the changing state of
English. He repeatedly mentioned the emerging dominance of Chinese. Such
sweeping power urged him to learn Chinese. Aat stated:
I
totally agree with the importance of English as an international language. But
in future utility, it is hard to say. Look at Thailand, so many people like me
want to come to China and learn Chinese. While English still gains its power as
a global language, Chinese is reaching a higher level than before. (Aat,
interview 2).
In
the above quote, Aat reflected on the relative utility of learning Chinese. The
need to learn the two languages may lead to a gradual weakening of the
attractor states (i.e., stable tendencies in L2 dynamic systems). The process of adaptation of the two language learning systems
is dyadic and dynamically spiraling. The enhanced strength of counterpart
attractor states in learning Chinese prompts perturbation in the stable L2
self. Aat became more engaged in learning Chinese, as he mentioned in the
reflective report that his strategy was to formulate sentences and compare the
differences and similarities between L1, L2 and L3. He became more engaged in
Chinese learning because he perceived that he will soon be an advantaged
position in using that language. It seems that Aat’s emergent stabilities of L3
learning behavior are contingent on the contextual variability, in particular, the changing status of L2.
5.
Discussion and conclusion
Interpreting the
findings, we can see that multilingual learners develop an interacting
constellation of L1–L2–L3 selves rather than a single, fixed trait (Reeve,
2024; Teng & Wang, 2020). Within this constellation, some language selves
act as stable attractors, highly recurrent, resource-drawing configurations, while others function as vulnerable
attractors—intermittently activated and easily perturbed. Crucially, whether a
language self is stable or vulnerable emerges from contextual affordances,
symbolic and instrumental value, and cross-linguistic relations, not from any
inherent property of the language itself (Noels, 2001; Ryan & Deci, 2020;
Ushioda, 2020).
Across
Asian contexts, stability patterns diverge with local ecologies. In Hong Kong,
L2 English typically stabilizes as a deep attractor due to high status, dense
opportunities for use, clear utility, and assessment pressures, conditions that
consolidate extrinsic and internalized motives into sustained engagement
(Gardner, 2010; Dörnyei, 2009; Noels et al., 2001). By contrast, L3 Japanese
often remains vulnerable, marked by receptive–productive asymmetries and
cross-linguistic interference from L1 Chinese and L2 English. When L3 goals
acquire identity relevance and near-term utility, consistent with vision-based
processes and DMCs, the L3 attractor can deepen, although it typically remains
adjacent to the dominant L2 English self (Dörnyei et al., 2015; Teng & Bui,
2020). In Thailand, many learners’ L2 English selves are stabilized around an
Ought-to orientation (exam-driven, externally regulated), yielding fragile
willingness to communicate, consistent with Teng and Bui (2020). As the
instrumental value of L3 Chinese rises, shifts in perceived utility and
opportunities for participation can move L3 from vulnerability toward
stability, illustrating how motivational systems reorganize under changing
affordances and futures (Bui et al., 2018; Bui & Teng, 2021; Noels, 2001;
Ryan & Deci, 2020).
Among
European learners, stability is likewise shaped by ecology and typological
proximity. In China, strong English capital (as L1 or a highly developed L2)
can attenuate investment in L3 Chinese, reducing participation and leaving the
L3 self vulnerable despite immersion, an instance of
situated motivation shaped by local interactional preferences (Ushioda, 2020;
Mercer, 2011). Conversely, in European pairings with closer typological
distance (e.g., English–French), positive transfer and
strategic comparison can support co-stability: learners leverage metalinguistic
insight to reinforce both L2 and L3 without zero-sum competition (Bui &
Teng, 2021).
A
dynamic-systems perspective clarifies how attractor dominance shifts as
parameters change. Learners may retreat toward L2 after prolonged low yield in
L3 (e.g., years of school study with little communicative payoff), or pivot
toward L3 when macro cues and employability recalibrate expected utility,
patterns consistent with DMC-triggered, vision-aligned surges in effort and the
situated nature of motivation (Dörnyei et al., 2015; Ushioda,
2020). Stability is maintained when purpose is repeatedly revalidated through
frequent use, identity alignment, and success experiences. Where L2–L3
comparability is salient, metalinguistic comparisons yield quick, perceptible
progress that sustains engagement; where roles and immersion differ across
languages, learners can sustain dual focus through differentiated functions for
each code (Teng & Bui, 2020).
To
integrate these patterns, we conceptualize the Multilingual Self as a Coupled
Attractor System, aligning with research that treats motivation as dynamic,
vision-driven, and context-embedded (Dörnyei & Kubanyiova,
2014; Dörnyei et al., 2015; Ushioda, 2020; Wang & Wang, 2024). The model
features two core properties. First, stability gradients: each language self resides in a basin of attraction whose depth
reflects ecological capital (status, access, assessment), identity investment
(Ideal vs. Ought-to selves), and use density (Gardner, 2010; Dörnyei, 2009;
Noels et al., 2001). Deep basins yield stable selves; shallow basins yield vulnerable
selves. Second, coupled dynamics: changes in one basin reshape others via three
channels. Capital coupling reallocates attention and effort as
symbolic/material values shift (e.g., English’s global capital, Chinese’s
rising regional capital). Linguistic coupling reflects typological
proximity/distance, enabling co-stability via transfer or creating interference
and trade-offs. Participation coupling captures how authentic use—or its
blockage—accelerates stabilization or sustains vulnerability through feedback
cycles of success or failure.
This
framework elucidates the cases above. In Hong Kong, a capital-rich,
assessment-heavy L2 English basin keeps L3 Japanese shallow unless
participation and personal purpose intensify (Bui et al., 2018). In Thailand,
revaluation of Chinese deepens the L3 basin, enabling escape from an Ought-to
L2 basin as vision
and opportunity align (Teng & Wang, 2020). For European learners in China,
participation coupling is inverted because local peers prefer English practice,
starving L3 Chinese of use and keeping it vulnerable despite residence in the
target culture (Mercer, 2011; Ushioda, 2020).
Practically,
learners can deepen vulnerable basins along the three couplings. For capital
coupling, reframing goals into near-term, identity-relevant roles
and stacking micro-credentials can support internalization and self-efficacy
(Noels et al., 2001; Ryan & Deci, 2020). For linguistic coupling,
structured cross-referencing converts interference into metalinguistic insight;
modality rebalancing addresses receptive–productive gaps with frequent,
low-stakes output; pronunciation compartmentalization limits cross-language
bleed, strategies that align with vision-led, task-integrated engagement
(Dörnyei et al., 2015). For participation coupling, designing use density,
building role-based immersion, and engineering interactions (e.g., language
contracts, turn-taking by language) protect vulnerable languages in
English-dominant spaces, consistent with situated motivation and meaningful use
(Ushioda, 2020).
Sustaining
a balanced multilingual self means preventing collapse of the vulnerable self
while maintaining the stable one. Time/resource partitioning, periodic
“motivation audits” tracking utility, identity resonance, and use frequency,
and identity scaffolding that renders the Ideal self vivid
while calibrating Ought-to pressures into approach-oriented goals all help
align day-to-day action with long-term vision (Dörnyei, 2009; Dörnyei & Kubanyiova, 2014; Ryan & Deci, 2020). Without such
balancing, learners in strong English ecologies risk chronic L3 vulnerability
and low willingness to communicate; those in English-dominant interactional
spaces may keep L3 peripheral unless participation is structurally ensured; and
learners who miss rising capital signals (e.g., for Chinese) may remain trapped
in shallow Ought-to basins (Bui & Teng, 2021). Overall, this account
reframes multilingual motivation as emergent co-stability among interacting
language selves, contingent on how learners and contexts co-construct value,
linguistic affordances, and participation opportunities across time.
References
Alamer,
A., & Alrabai, F. (2023). The causal relationship between learner motivation
and language achievement: New dynamic perspective. Applied Linguistics, 44(1),
148-168. https://doi.org/10.1093/applin/amac035
Bui, G., Man, L., &
Teng, M. F. (2018). From second language to third language learning: Exploring
a dual-motivation system among multilinguals. Australian Review of Applied
Linguistics, 41, 63-91. https://doi.org/10.1075/aral.17051.man
Bui, G., & Teng, M.
F. (2021). Exploring complexity in L2 and L3 motivational systems: A dynamic
systems theory perspective. The
Language Learning Journal, 49(3), 302-317. Doi:
10.1080/09571736.2019.1610032
Dörnyei, Z. (2005). The
L2 motivational self system. In Z. Dörnyei & E.
Ushioda (eds.), Motivation, language identity and the L2 self (pp.
9-42). Multilingual Matters.
Dörnyei, Z. (2009). The
psychology of second language acquisition. Oxford: Oxford University Press.
Dörnyei, Z., & Kubanyiova, M. (2014). Motivating learners, motivating
teachers: building vision in the classroom. Cambridge University Press.
Dörnyei, Z., & Ushioda, E. (eds). (2009). Motivation,
language identity and the L2 self. Multilingual Matters.
Dörnyei, Z., &
Ushioda, E. (2021). Teaching and researching motivation. Routledge.
Dörnyei Z., Henry A.,
Muir C. (2015). Motivational currents in language learning: Frameworks for
focused interventions. Routledge.
Gardner, R. (2010). Motivation
and second language acquisition: The socio-educational model. Peter Lang.
Hajar, A. (2024). The
ideal multilingual self of individuals in conflict-affected situations.
International Journal of Multilingualism, 21(1), 400-417. https://doi.org/10.1080/14790718.2022.2029865
Hammarberg, B. (Ed.).
(2009). Processes in third language acquisition. Edinburgh University
Press.
Henry, A. (2014). The
motivational effects of crosslinguistic awareness: Developing third language
pedagogies to address the negative impact of the L2 on the L3 self-concept. Innovation
in Language Teaching and Learning, 8, 1–19.
Henry, A., &
Thorsen, C. (2018). The ideal multilingual self: Validity, influences on
motivation, and role in a multilingual education. International Journal of
Multilingualism, 15(4), 349-364. https://doi.org/10.1080/14790718.2017.1411916
Henry, A. (2023).
Learner–environment adaptations in multiple language learning: Casing the ideal
multilingual self as a system functioning in context. International Journal
of Multilingualism, 20(2), 97-114. https://doi.org/10.1080/14790718.2020.1798969
Jessner, U. (2008). A
DST model of multilingualism and the role of metalinguistic awareness. Modern
Language Journal 92, 270–283. https://doi.org/10.1111/j.1540-4781.2008.00718.x
Liu, M. (2022). The
emotional basis of the ideal multilingual self: The case of simultaneous
language learners in China. Journal of Multilingual and Multicultural
Development, 43(7), 647-664. https://doi.org/10.1080/01434632.2020.1755294
MacIntyre, P. D., &
Legatto, J.J. (2011). A dynamic system approach to
willingness to communicate: Developing an idiodynamic
method to capture rapidly changing affect. Applied Linguistics, 32,
149–171. https://doi.org/10.1093/applin/amq037
Mercer, S. (2011).
Understanding learner agency as a complex dynamic system. System, 39,
427– 436. https://doi.org/10.1016/j.system.2011.08.001
Norton, B. (2000). Identity
and language learning: Gender, ethnicity, and educational change. Harlow:
Longman/Pearson.
Noels, K. (2001).
Learning Spanish as a second language: Learners’ orientations and perceptions
of their teachers’ communication style. Language Learning, 51(1),
107–144. https://doi.org/10.1111/0023-8333.00149
Noels, K., Clément, R.,
& Pelletier, L. (2001). Intrinsic, extrinsic, and integrative orientations
of French Canadian learners of English. Canadian
Modern Language Review, 57(3), 424–442. https://doi.org/10.3138/cmlr.57.3.424
Reeve, J. (2024). Understanding
motivation and emotion. John Wiley & Sons.
Ryan, R., & Deci,
E. (2020). Intrinsic and extrinsic motivation from a self-determination theory
perspective: Definitions, theory, practices, and future directions. Contemporary
Educational Psychology, 61, 101860. https://doi.org/10.1016/j.cedpsych.2020.101860
Teng, M. F., & Bui,
G. (2020). Thai university
students studying in China: Identity, imagined communities, and communities of
practice. Applied Linguistics Review,11(2), 341-368. DOI: https://doi.org/10.1515/applirev-2017-0109
Teng, M. F., &Wang,
L. (2020). Identity, motivation, and multilingual education in Asian
contexts. Bloomsbury.
Teng, M. F. (2025).
Examining longitudinal development of writing motivation in the GenAI context:
A self-determination theory perspective. Learning and Motivation, 91,
102157. https://doi.org/10.1016/j.lmot.2025.102157
Ushioda, E. (2020). Language
learning motivation: An ethical agenda for research. Oxford University
Press.
Wang, S., & Wang,
Y. (2024). Exploring complex multilingual motivation types among Chinese
students majoring in dual foreign languages: a Q method study. Journal of
Multilingual and Multicultural Development, 1-19. https://doi.org/10.1080/01434632.2024.2384493
Zhang, Z. (2025). ‘I do
not need to fit in': multilingual speakers’ identity and cross-cultural
adaptation in Hong Kong. International Journal of Multilingualism, 1-17.
https://doi.org/10.1080/14790718.2025.2494034
Author
bio
Mark Feng Teng, Ph.D., is Associate Professor at Macao
Polytechnic University. He was the recipient of the 2017 Best Paper Award from
the Hong Kong Association for Applied Linguistics (HAAL). His-research
portfolio mainly focuses on L2 vocabulary acquisition, reading and writing from
the perspective of metacognition and working memory. His-publications have
appeared in international journals, including Applied Linguistics, TESOL
Quarterly, Language Teaching Research, System, Applied Linguistics Review,
Computer Assisted Language Learning, Computers & Education, Foreign
Language Annals, and IRAL, among others. His-recent monographs were
published by Routledge, Springer, and Bloomsbury. He also edited and co-edited
special issues for international journals, including Journal of Writing
Research, Studies in Second Language Learning and Teaching, and TESOL Journal.
He serves as editor-in-chief for International Journal of TESOL Studies (IJTS),
Digital Applied Linguistics (DAL), and Applied Language Sciences (ALS), and
co-editor for Reading in a Foreign Language (RFL) and Asian Journal of English Language
Teaching (AJELT). He was listed in the top 2 % of Scientists in the world in
the disciplinary areas of Linguistics/Applied linguistics/Education, based on
Stanford University Rankings.
E-mail: markteng@mpu.edu.mo
ORCID: https://orcid.org/0000-0002-5134-8504
Lawrence Jun Zhang, PhD, is Professor of
Linguistics-in-Education and Associate Dean for the Faculty of Education and
Social Work, The University of Auckland, New Zealand. His-major interests and
100-plus publications are on learner metacognition, language-teacher education,
and L2 reading-writing development. He is an editorial board member
for Applied Linguistics Review, Australian Review of Applied Linguistics,
Journal of Second Language Writing, Metacognition and
Learning, and RELC Journal, among others. He was honoured by the TESOL International Association (USA) in
2016 with the award of “50 at 50″, acknowledging “50 Outstanding Leaders” in
the field of TESOL around the world and was officially installed as a newly
elected member of the Board of Directors of the Association in 2017. In the
Stanford University Rankings 2022 and 2023, he was listed in the top 2 % of
Scientists in the World in the disciplinary areas of Linguistics/Applied
Linguistics/Language Education. His-work has appeared in Applied
Linguistics, Applied Linguistics Review, Modern Language Journal, TESOL
Quarterly, Discourse Processes, System, Language Teaching Research, Reading and
Writing, Reading & Writing Quarterly, Journal of English for Academic
Purposes, RELC Journal and Journal of Second Language
Writing, among others.
E-mail: lj.zhang@auckland.ac.nz
ORCID: https://orcid.org/ 0000-0003-1025-1746