System 67 (2017) 38e49

Contents lists available at ScienceDirect

System
journal homepage: www.elsevier.com/locate/system

Don't tell me what to do! The anti-ought-to self and language

learning motivation
Amy S. Thompson
University of South Florida, Department of World Languages, 4202 E. Fowler Ave CPR 419, Tampa, FL 33620, United States

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

Article history: Drawing on Complex Dynamic Systems Theory (CDST), this study explores the motiva-
Received 15 February 2016 tional profiles of two L1-English-speaking advanced language learners of Chinese and
Received in revised form 17 April 2017 Arabic. Through narrative inquiry, this study investigates the dynamic nature of each
Accepted 26 April 2017
learner's language learning journey. Data were collected via interviews, which took place
Available online 8 May 2017
in informal environments. The interview instructions encouraged participants to tell their
language learning story; there was no prompting of the “motivation” theme on the part of
Keywords:
the interviewer; however, there was a focus on the implications of the “other” dimension
motivation
The L2 motivational Self System (L2MSS)
(external influences, such as specific individuals and contexts). As such, this study ex-
Psychological reactance amines the dynamicity of the potentially synergistic relationships of the ought-to and anti-
Anti-ought-to self ought-to selves. The main findings show that the anti-ought-to self has a “dominant”
Complex Dynamic Systems Theory (CDST) relationship with the context, in contrast to the “submissive” relationship of the ought-to
Narrative analysis self; the findings also illustrate the dynamicity of the psychological aspects of self, as these
selves fluctuate throughout the language learning experience. Implications of this study
include the importance of language educators understanding the relationship of the
context to motivational development and change. The results also indicate that the anti-
ought-to self could be the missing link to explanatory value of the L2MSS.
© 2017 Published by Elsevier Ltd.

1. Introduction

This study was conducted to explore the relationship of the L2 Motivational Self System (L2MSS) and Complex Dynamic
Systems Theory (CDST), specifically examining the role of context in developing the different selves of language learners. The
study contextualizes the anti-ought-to self within the L2MSS framework (theory building) and conceptualizes the dynamic
interaction between learner and context (motivation and CDST) in order to fill the gaps in motivation research and the field of
SLA broadly construed (Do € rnyei, MacIntyre, & Henry, 2015).
Complex systems are “systems that are heterogeneous, dynamic, non-linear, adaptive and open” (Larsen-Freeman &
Cameron, 2008, p. 36). Regarding language learners, de Bot, Lowie, and Verspoor (2007) explains: “from a [C]DST perspec-
tive, a language learner is regarded as a dynamic subsystem within a social system” (p. 14). As such, the operationalization of
the complex systems discussed in this study are the motivational profiles of the individual learners, who are part of the larger
social context in which the language learning experiences took place. Through narrative inquiry, this study explores the non-
linear and dynamic nature of each learner's language learning journey with a specific focus on the selves influenced by the
“other” dimension: the ought-to and anti-ought-to selves. Although previous research indicates the existence of the anti-

E-mail address: athompson@usf.edu.

http://dx.doi.org/10.1016/j.system.2017.04.004
0346-251X/© 2017 Published by Elsevier Ltd.
 A.S. Thompson / System 67 (2017) 38e49 39

ought-to self (Thompson & Vasquez, 2015), the current study serves to connect the anti-ought-to self to the existing L2MSS
model, particularly in conjunction with the “other” dimension in Higgins' (1987) Self-Discrepancy Theory by examining
experiences of L1-English participants who learned languages other than English (LOTEs).

2. Literature review

2.1. The L2 Motivational Self System

Do €rnyei's (2009) L2MSS posits that language learning motivation is primarily composed of the ideal self, the ought-to self,
and the learning experience. The ideal self is who a learner would like to become in terms of language ability (internal desires);
inquiries into a learner's ideal self oftentimes involve “imagine” or “envision” as descriptors in questionnaire items. The
ought-to self is who the learner feels that he or she is obliged to become based on external pressures, such as pressure from
friends, authority figures, or in terms of general societal expectations. The learning experience is a compilation of the past and
current experiences of a language learner, which affect subsequent language learning experiences.
Do €rnyei (2009) derives the “self” portion of the L2MSS (ideal and ought-to selves) directly from the work of Higgins (1987,
1998) with additional influence from Markus and Nurius (1986). The ideal self “represent[s] one's own or a significant other's
hopes, wishes, and aspirations for oneself.” An ought self (which corresponds to Do €rnyei's ought-to self) “… represent[s] one's
own or a significant other's beliefs about one's moral responsibilities and who one should or ought to be” (Higgins, 1998, p.
10). Higgins also discusses promotion and prevention foci:
The promotion focus is concerned with accomplishments, hopes, and aspirations … Ideal self-guides, therefore, have a pro-
motion focus. The prevention focus is concerned with safety, responsibilities, and obligations … Ought self-guides, therefore, have
a prevention focus” (p. 16, italics in original).
L2MSS research indicates that a strong ideal self is essential for language learning motivation, and thus, success in lan-
guage learning. The ought-to self, however, is sometimes not included (i.e. Ryan, 2009), or when it is, has inconclusive results
(i.e. low Cronbach's alphas or no group distinctions; Csize r & Lukacs, 2010; Lamb, 2012; Thompson & Erdil-Moody, 2016). As
the theoretical rationale for the ought-to self is Self-Discrepancy Theory, the role of context needs further scrutiny.

2.2. The L2MSS and CDST

Learners who are motivated by ought-to selves are those who are learning a language because of external expectations
(elements in the larger complex system). The dynamic interaction between learners and context can be conceptualized as
learners being the “submissive” component and context as the “dominant” component (i.e. the external pressures prevail).
This realization of self is quite different from that of learners’ ideal selves, which can be conceptualized as learners being the
strong element, in charge of their own destiny. Thus, what is potentially absent from the L2MSS framework is a dynamic
interaction of the language learners (the complex system under investigation) and the context (the larger system of which the
learner is a part) in which the language learners are interpreted as the “dominant” component, employing active resistance
against societal expectations as a key source of motivation e the anti-ought-to self (Thompson & Va squez, 2015). This future
self guide is motivated by the opposite of what the external pressures demand: choosing to study a language to go against the
norms of society. The anti-ought-to self can emerge when engaging with a commonly-studied language (such as English in
the Chinese context), if one were to choose a language major over something potentially more lucrative (such a degree in
engineering or medicine), as well as with a language that is less commonly studied (such a language other than Spanish in the
U.S. context). The anti-ought-to self can develop in both foreign and second language contexts, and includes “ought-to” in the
nomenclature because of the strong influence of events and pressures outside of the learner as the primary motivating
components of this self guide. The anti-ought-to self could also have a relationship with the ideal self, being that the visual
concept that learners associate with a successful language learning process is, in fact, to do what people do not expect them to
do. Incorporating the anti-ought-to self into the L2MSS would allow for the type of future self that defines the learner as the
prevailing force in the language learning process.
The anti-ought to self is inspired by Reactance Theory. An example of psychological reactance, or resistance to social
influence, is when American youth are determined to drink because of the strict age laws regarding alcohol consumption.
While ubiquitous in the psychology literature, Reactance Theory (Brehm, 1966; Brehm & Brehm, 1981) has not been incor-
porated into the mainstream SLA motivation literature. Nevertheless, there have been references to the importance of a
concept, such as in MacIntyre (2002), who stated that it was “proposed by critics for addition to the literature on language
learning motivation” (p. 59) and indicated that it could “show the complexity of our motivational experiences” (p. 60).
Additionally, Goldberg and Noels (2006) referenced reactance as to why the students who attend a university with an
exclusively French curriculum have both a stronger Francophone school identity in French, as well as a stronger Anglophone
identity at home (p. 438). Both of these examples illustrate early elements of the anti-ought-to self concept in SLA literature.
With English-speaking language learners in the context of the U.S., Thompson and Va squez (2015) incorporated psy-
chological reactance, coined as the “anti-ought-to self,” as an explanation of the motivation of two of the three learners in the
study, Alex and Joe, who studied Chinese and German. In Alex's case, his anti-ought-to self emerged as a reaction to his
Chinese TA who doubted his capability of learning Chinese. In Joe's case, his anti-ought-to self was a reaction to competition
with German native speakers in his classes, and aptly describes himself as “a Gonz alez in a group of Germans” (p. 169). The
40 A.S. Thompson / System 67 (2017) 38e49

third participant, Vera, did not exhibit an anti-ought-to self, as her biggest obstacle was loneliness (i.e. a lack of interactions).
Also, unlike Alex and Joe, “Vera's foreign language choice of Italian was the most congruent with her ethnic heritage and
identity” (p. 171), which could also explain the non-emergence of the anti-ought-to self. In this study, Thompson and Va squez
(2015) also re-analyzed the L2MSS with an eye on the “I” versus “other” facet of the selves in Higgins' (1987) Self-Discrepancy
Theory. The anti-ought-to self concept incorporates the strong internal vision component of the ideal self, while at the same
time integrating the importance of the environment or context, similar to the ought-to self. Thus, the anti-ought-to self more
fully integrates the original facets of Self-Discrepancy Theory in terms of the internal and external components.
More recently, Thompson and Liu (2017) and Liu and Thompson (2017) with English language learners in China, Thompson
(2017) and Huensch and Thompson (2017) with foreign language learners in the U.S., and Alharbi (2017) with Saudi ESL
learners in the U.S. quantitatively explored the anti-ought-to self. Using the narratives from Thompson and Va squez (2015),
focus groups, and other discussions, anti-ought-to self items were developed to complement the ideal and ought-to self items
in the oft-used L2MSS questionnaires. “I am studying this language even though most of my friends and family members don't
value foreign language learning” and “I am studying this language because it is something different or unique” are two ex-
amples of the items that were developed. In these studies, exploratory factor analyses (EFAs) resulted in clear three-factor
structures with the three selves (ideal, ought-to, and anti-ought-to), indicating that the anti-ought to self is a psychologi-
cal aspect of self in its own right, related, yet distinct from the ideal and ought-to selves. The finding of an anti-ought-to self
construct in China was particularly notable, given the collective nature of the context (see Thompson & Liu, 2017; Liu &
Thompson, 2017). Other significant findings include the anti-ought-to self as a significant predictor of proficiency (Liu &
Thompson, 2017), as similar to the ought-to self in terms of overlapping selves with different languages studied (as
opposed to the distinct ideal selves for multiple languages) (Thompson & Liu, 2017), as potentially related to the language
studied (Thompson, 2017), as having a positive relationship to the desire to improve pronunciation in a foreign language
(Huensch and Thompson, 2017), and as positively impacting motivated learning behavior (Alharbi, 2017). Specifically,
Thompson (2017) found that the anti-ought-to self was the only self that discriminated group membership of the three
groups of learners (Spanish; French, German, Italian; and LCTLs), with the Spanish group having a significantly lower anti-
ought-to self than the other two language groups. In Alharbi (2017), the anti-ought-to self was the only significant
positively-effecting variable that contributed to the amount of variance explained for motivated learning behavior (the ought-
to self contributed negatively). These results indicate that those with a strong anti-ought to self were more likely to
participate in language learning activities/behaviors.
Citing the anti-ought-to self of Thompson and Va squez (2015), Lanvers (2016, 2017) describes a similar concept: “rebel-
lious motivation.” Using data from L1-English-speaking participants from the UK, Lanvers hypothesizes that “motivation for
Anglophones learning other languages is perceived to be vulnerable” (2016, p. 83), given the popularity of English world-
wide. She also indicates that “Some students of this learner type also displayed the ‘rebellious’ motivational streak
wanting to counter the negative language learner image of the British” (p. 87). Several researchers have begun to investigate
this concept, whether it be in the form of psychological reactance, rebellious motivation, or the anti-ought-to self.
The anti-ought-to self could be defined as a self guide that is sensitive to external pressures (similar to the ought-to self),
but that also has a promotion focus (similar to the ideal L2 self). Regarding the quantitative work that has been done with the
anti-ought-to self, it is patterning similarly to the ideal self in terms of illustrating group differences, but similar to the ought-
to self in terms of person-specific versus language-specific characteristics. Clearly, psychological reactance could inform the
research on motivation in SLA, and although Laurin, Kay, Proudfoot, and Fitzsimons (2013) postulate that it is more common
in Western contexts, the evidence from previous work shows that it emerges in many contexts, including the non-Western
context of China. As MacIntyre (2002) indicates, bringing reactance to the forefront could underscore the complexities of
language learning motivation; the concept of the anti-ought-to self incorporated into the L2MSS could further inform the
dynamics involved in individuals’ motivational profiles.

3. The study

The overarching research question in this study is, “How does the anti-ought-to self develop in the motivational profiles of
high-achieving language learners?” Interview data from L1-English participants studying Chinese and Arabic were used to
explore this question. The emergence and development of the anti-ought-to self in these participants is illustrated in the
subsequent sections.

4. Methods

4.1. Participants

This study uses data from Alex ,1 a learner of Chinese, and Rachel, a learner of Arabic. For both participants, their language
study started in the foreign language context (i.e. in the U.S.), although the languages in question are both foreign languages

1 squez (2015), although the focus in the current study is distinct.

“Alex” also appears in Thompson and Va
 A.S. Thompson / System 67 (2017) 38e49 41

and second languages at different times of their language learning experiences (i.e. they studied Chinese and Arabic in the U.S.
as well as in China and Egypt, respectively).
As a senior International Studies and Political Science major, Alex used Chinese at work. Being that he was in first semester
Chinese at the time (after unsuccessful Spanish and then French language learning experiences), he was shocked that only he
could interact with visiting dignitaries in Chinese. This “strategic weakness in terms of Chinese” motivated him to continue
his Chinese studies via immersion in China and coursework (MA and Ph.D. degrees in Chinese Pedagogy). Distinct from Alex,
Rachel had positive early language learning experiences with French (her first foreign language), German, and Scottish-Gaelic.
Later, with university study of Arabic and several study abroad programs in Egypt as a part of her undergraduate, MA, and
Ph.D. degrees, her Arabic proficiency flourished. With a combined bachelor's and master's degree in linguistics, she started
her Ph.D. in Arabic, but then switched to linguistics, even though she “would have rather done her Ph.D. in Arabic.” The switch
was prompted by a “crisis” in the Arabic-teaching world, but determined to maintain her language ability, she purposefully
designed her research agenda around Arabic. Both Alex and Rachel are now university faculty members of Chinese and Arabic,
respectively, at large public universities in the United States.
Both of the participants were chosen for their extraordinary success in reaching high levels of linguistic achievement and
for their language choices (i.e. languages atypical for L1 English speakers). Together, Alex's and Rachel's narratives created
symmetry in terms of the anti-ought-to self. Alex's ought-to self changed into an anti-ought-to self as his Chinese developed.
For Rachel, studying Arabic because it was a challenging and unexpected choice (anti-ought-to self) suddenly became a
“smart thing to do” after the sociopolitical shift of 2001 (ought-to self). Thus, these two participants together show the diverse
embodiments of the anti-ought-to self.
Currently, Alex and Rachel are the directors of Chinese and Arabic language programs at large public universities in the
U.S.; thus, they teach all levels of language classes and are also responsible for training other faculty and TAs that work in their
respective programs. Both participants are all still active members of their profession, and their individual complex systems
continue to change and adapt.

4.2. Procedure

This study is an analysis of the retrospective thoughts, elicited through narrative data, that existed in the minds of each of
these participants at the time of data collection. The interviews (one per participant, along with follow-up e-mails when
needed) were conducted with personal acquaintances/colleagues (sample of convenience) and took place in informal en-
vironments (a campus office and Skype for those in the current study). The interviews were audio-recorded and transcribed;
the two interviews in the current study were approximately 40 and 50 min. The participants were encouraged to “tell a story”
about their language learning experiences, starting from the earliest memories; the interviewer interjected questions as they
arose. As such, the narratives emerged organically with each speaker's personal discourse style prominent throughout with
no specific mention of “motivation” in the initial prompt.
After the original analysis presented in Thompson and Va squez (2015), the interviews were re-examined for the specific
theme of the anti-ought-to self, the instances of which were highlighted in the transcripts and then were imported into a
spreadsheet for a content analysis. The examples presented in this study are illustrative examples, forming a picture of the
participants’ motivational systems travelling through the state-space landscape. Alex and Rachel were given the opportunity
to read the initial manuscript and give feedback before publication (Hornberger, 2006).

4.3. Analysis

4.3.1. Rationale for the narrative approach

While quantitative inquiry into motivation indeed leads to important insights, especially when employing a large cross-
sectional or longitudinal design, narratives allow us to explore the integration of these possible selves in a way that com-
plements quantitative questionnaire data. Using a narrative approach to explore the dynamic system of the language learners
allows the researcher “to discover information that the participants themselves may not realize” (Savin-Baden & Major, 2013,
p. 238). Narratives are similar to dynamic systems in that, “there is generally not a sense that stories have a beginning, middle
and end. Instead, stories are seen as complex, multifaceted and evolving” (p. 241). Thus, a narrative approach allows an
exploration of the relationship of the motivational profiles of the learners (the complex systems under investigation) and
their relationships to the specific language learning environment, or context (the larger complex system in which the learners
find themselves).
Narratives are co-constructions of self, developed by the interviewer and interviewee, as discussed by Linde (1993, p. 4),
and help to create a sense of self: “Life stories2 express our sense of self e who we are, how we are related to others, and how
we became that person” (p. 219). Thus, narratives are not illustrations of “truth” broadly construed, but rather of the “truth”

2
Life stories have many of the same characteristics as narratives; the main difference is that life stories are a compilation of a series of interviews,
whereas the narratives in this chapter are composed of one main interview with follow-up e-mails or brief conversations for clarification purposes.
However, the issues of the social aspects, truth, and construction of self apply to both life stories and narratives.
42 A.S. Thompson / System 67 (2017) 38e49

constructed from the perspective of the interviewee (Talmy, 2010). The excerpts in this study should not be interpreted as
uncontestable statements of reality, but rather the participants' construction of reality, and the truths and relationships that
exist therein. In other words, from a CDST perspective, the narratives that the participants and the researcher co-constructed
were influenced by the elements in the complex system of the language learning context, which, in turn, influenced the
complex system of the individual language learner, which is the complex system under investigation (Larsen-Freeman &
Cameron, 2008; de Bot et al., 2007). The interactions of the “messy little details” (de Bot et al., 2007, p. 19) of the language
learning experiences of the individuals (the complex system under investigation) are visible in the narrative data. Addi-
tionally, since narratives allow us to see the language learners’ own concepts of self, this is a suitable medium to explore the
self guide concept. Narratives also allow us to see the dynamic interactions between the context and the learner, allowing us
to extrapolate information and further develop the L2MSS framework.

4.3.2. Defining CDST

The data were primarily analyzed using three concepts: initial condition (initial state), attractor state, and phase transition
(phase shift or bifurcation). The initial condition (initial state) is the basis on which the system will develop, and is extremely
sensitive to initial conditions. de Bot et al. (2007) explain that initial conditions are “one of the most essential characteristics
of dynamical systems” and that “there is a growing body of evidence that suggests that initial conditions are precursors of the
development of a second language” (p. 15). An attractor state is a state that is preferred by the system over other possible
states: “As they develop over time, dynamic sub-systems appear to settle in specific states, so-called attractor states, which
are preferred but not necessarily predictable” (de Bot et al., 2007, p. 8). The strength of the attractor state is influenced by
variables in the system, and the ease at which one can exit the attractor state depends on its strength. Larsen-Freeman and
Cameron (2008) describe these attractor states as wells with varying levels of steepness and depth into which a ball would
naturally fall (p. 51), a description that influences the figures created for this study. Finally, a phase transition (phase shift or
bifurcation) is the shift that a complex system undergoes when its behavior is radically altered, and “the states of the systems
before and after a phase shift are very different” (Larsen-Freeman & Cameron, 2008, p. 45). There is typically a significant
event or series of events that initiate phase transitions (see also Thompson, 2013, for another example of traditional data
collection methods used with CDST as the interpretive framework).
The impetus of this article is to re-conceptualize the constructs involved in the L2MSS, with the elaboration of the anti-
ought-to self, using narrative data from L1-English learners. The ideal, ought-to, and anti-ought-to selves are three potentially
conflicting attractor states, which eventually engage in a synergistic relationship (Fig. 1).

5. Results

In this section, the individual “complex and adaptive subsystems” (Larsen-Freeman & Cameron, 2008; de Bot et al., 2007) of
each of these participants’ motivational profiles, indelibly influenced by their language learning experiences, are presented to
answer the question, “How does the anti-ought-to self develop in the motivational profiles of high-achieving language learners?”

Ideal L2 Self:
Internal expectations

Motivation

Ought-to Self: Anti-Ought-to Self:

External expectations Reactions against external
expectations

Fig. 1. Pictorial representation of the potentially conflicting attractor states working in unison in a complex system.
 A.S. Thompson / System 67 (2017) 38e49 43

5.1. Alex

Alex's initial condition was that of a student growing up in rural Ohio, a context that did not particularly encourage
language learning. The elements influencing his individual complex system at this point did not especially motivate him to
learn a language:

Excerpt 1
A: That's one of the reasons later when I started speaking a high level of Chinese that I wanted to be a teacher. Because I
studied Spanish in high school. I studied French in college. And I can't speak either language.
Alex started taking Chinese as an undergraduate because it was a requirement, and this element in his social context
activated his ought-to self guide:

Excerpt 2
A: I had already started taking Chinese, but it was an undergraduate language requirement.
Alex was working at the state capital the semester that he started taking Chinese (his senior year), and he was often
called upon to interact with Chinese dignitaries because of a lack of Chinese knowledge of the other employees. This
context helped Alex realize that there was a “strategic weakness” when it came to Chinese; work opportunities involving
Chinese became evident, and this systemic element strengthened Alex's ought-to self guide.

Excerpt 3
A: Any type of delegation that came in to the state of Ohio, they said “Send Alex. He knows Chinese.” And I was in
Chinese 101 at the time. So I said “This is a signal that we have a strategic weakness in terms of Chinese.”
This pulled Alex into his first attractor state e a shallow attractor state stabilized by his ought-to self because of the
prospect of future employment:

Excerpt 4
A: I could definitely find a job if I could get good in Chinese, so I had decided at that time I wanted to continue studying
Chinese, but I had never considered being a teacher of Chinese.
However, there was also already a hint of anti-ought-to self present in Alex during his first semester of Chinese e
discouragement by his TA and the desire for a challenge. At this point, the dynamic interactions between the potentially
conflicting attractor states (ought-to and anti-ought-to selves) were working in tandem to pull him out of his first shallow
attractor state into a deeper one.

Excerpt 5
A: We had to have a one year foreign language requirement, and I thought Chinese would be a difficult challenge. So I
took it. And the end of my first semester, my TA from China told me I wasn't very good at China- Chinese, I should learn
another language … That motivated me a lot.
This seemingly insignificant interaction between Alex and his Chinese TA prompted a rather significant shift in Alex's
individual complex system. Upon graduating, he investigated opportunities abroad, which was the beginning of his self-
imposed language learning immersion experience, thus shifting his Chinese as a foreign language experience to that of
Chinese as a second language. Perhaps subconsciously wanting to prove his TA wrong, Alex was determined to become
proficient in Chinese. Alex supported himself in China by teaching ESL, which is when he experienced a phase shift, as a result
of the elements found in his new context (China). He found teaching to be extraordinarily rewarding and decided to continue
his teaching career by becoming an instructor of Chinese.

Excerpt 6
A: I was watching the process of the students growing from the first semester, when they couldn't do anything in
English until the end of the year that- that academic year. It was amazing watching them change. How much they
learned, and how much they could do after that short of a period of time. And I thought, “that's fun.”
The phase shift (the realization for his call to teach Chinese) and the influence of the anti-ought-to self propelled Alex into
a deeper attractor state, and his language learning system looked dramatically different before and after his phase shift
(Larsen-Freeman & Cameron, 2008). Alex's anti-ought-to self was strengthened by interacting with elements from the
environment, as he had begun to enjoy what he calls a context dependent “celebrity status” in China, writing for a newspaper,
and hosting a radio talk show from his dorm room, oftentimes shocking people with his Chinese abilities:
44 A.S. Thompson / System 67 (2017) 38e49

Excerpt 7
A: I've had people drop … a tub of dishes at a restaurant when they heard me speaking. I've seen people, you know,
slam on the brakes on their bicycle when they hear what I'm saying. It just shocks them. Because the default
assumption among Chinese is “there's no possible way a white person could … understand what we're saying.”
One of the strongest motivators for Alex when he was living in China was the desire to be the “anti-stereotype” with
regards to his Chinese, as well as with regards to cultural norms/expectations. During that time, all of the elements in his
individual complex system were interacting to strengthen his anti-ought-to self.

Excerpt 8
A: The other related thing that drove me for many years was the absence of positive/ideal NNS role models who could
speak Chinese in a way that drew praise from Chinese people. Chinese people always talked extremely negatively
about “foreigner Chinese” and often made fun of foreigners speaking Chinese … I wanted to show every Chinese
person I met that foreigners could learn Chinese to a level at which Chinese could not find anything to make fun of. I
wanted to become the anti-stereotype … When Chinese criticized foreigners living in China for drinking too much or
going to clubs, I stopped drinking and avoided clubs … Whatever other foreigners did while speaking Chinese that
drew criticism, I took as a goal to eliminate from my Chinese. When I heard people making fun of foreigners' tones, I
spent countless hours working on tones. When I heard people making fun of foreigners' intonation, I spent hours
repeating out loud after Chinese television shows just trying to imitate intonation. 3
This strong sense of anti-ought-to self is also highlighted in his interactions with his current Chinese students, who, like
him, are primarily L1 English speakers learning Chinese as a foreign language. He feels a strong sense of pride in being able to
get his students to a high level of proficiency, an element that keeps him embedded in his deep attractor state.

Excerpt 9
A: Now, I guess I am reliving this process vicariously through my students. I got the same high when we were [at a
Chinese language competition] and the judges stopped the competition to tell me that heritage learners were not
eligible after our student delivered his speech in Chinese … I get that sense of validation, which is highly motivating for
me.
Thus, Alex's initial motivation to study Chinese could be classified as “ought-to”; however, interactions with elements in
his context created his anti-ought-to self. The systemic agents working in unison, such as his delight and validation of both his
and his students' abilities to excel in the language and to unexpectedly succeed, resulted in a strong anti-ought-to self and an
atrophied ought-to self (Fig. 2).

5.2. Rachel

Rachel's initial condition was a paradox. She had an intense desire to learn another language from an early age; yet, she
was not in an environment that was particularly conducive to doing so. When she was seven or eight, she asked her parents to
speak to her in a language other than English:

Excerpt 10
R: I remember trying to convince my parents that they should speak to me in a language other than English, [and they
said] but we don't speak any languages other than English.
The elements in the larger social context formed her initial condition that pushed her into her first shallow attractor state.
The seemingly insignificant element within the context of her family not speaking a language other than English pushed her
to explore several different languages. Rachel took French in high school, but was frustrated that she could not initially study
more than one language because of a scheduling conflict with Spanish. However, she found a way around this situation.

Excerpt 11
R: In my senior year, my French teacher was actually German so I started taking German lessons from her outside of
school and then, I somehow used my elective period to independently study Scottish-Gaelic.
At university, she was presented with an abundance of language-studying opportunities and chose Arabic. She was
partially motivated by the challenge (evidence of an anti-ought-to self), as well as lack of books from the Arab perspective on
the topic of the founding of Israel, a topic she had researched previously.

3
The stylistic differences in this excerpt can be explained by the fact that this is an excerpt from a follow-up e-mail after the interview took place.
 A.S. Thompson / System 67 (2017) 38e49 45

Phase shift:

Teaching ESL in
China: “That’s fun.”

Initial condition:

Rural Ohio,
unsuccessful
learning
experiences: “…my
first exposure to Deep attractor state:
anything other than anti-ought to self
hick English.”
Chinese language
teacher
Shallow attractor state: ought-to self
“There’s no possible
“I could definitely find a job if I could way a white
get good in Chinese.” person…could
Alex understand what we
are saying.”

Fig. 2. Pictorial representation of Alex as a complex system moving through the state space landscape.

Phase shift: Why Arabic?

“…one of the things I found

frustrating was that… I was Phase shift: Teaching Arabic
having more trouble finding
“I felt something of an
[books] that are of more Arab
obligation…there were so
perspective.”
many students studying it
Initial condition: [and there is] a crisis in the
French teacher: “You can’t take
this class.” Arabic teaching community.”
Early interest in
FL learning:
“…convince my
parents that they
should speak to
me in a language
other than Deep attractor state:
English.” ought-to self merges
with anti-ought to self

Post 9-11: “I don’t

Shallow Shallow attractor
want to give up Arabic,
attractor state: state: anti-ought-
but, I feel like…I no
Influence of the to self Rachel
longer have control
initial condition
“I wanted to over what studying
Took French, study a language Arabic…means to
German, …that wasn’t someone.
Scottish-Gaelic, European, so it’d
Teaching “allows me to
but didn’t stick be more of a
keep up a high level of
with any of them. challenge.”
the language.”

Fig. 3. Pictorial representation of Rachel as a complex system moving through the state space landscape.

Excerpt 12
R: I wanted to take something that was going to be different and challenging … One of the things I found frustrating
was that I found a lot of books from the Israeli perspective in our library but I was having more trouble finding ones that
were of more Arab perspective.
46 A.S. Thompson / System 67 (2017) 38e49

Also contributing to her phase shift of focusing on Arabic was the interaction with another element from her context: even
when she wanted to continue with French, her longest-studied language at that point, she was unable to do so.

Excerpt 13
R: I showed up to class [conversational French], and our teacher was like, “You can't take this class.” And I was like,
“Why not? I was placed at this level, I need to work on my speaking abilities e they're not that good.” He was like, “Well,
we have these students who came back from abroad and they still couldn't pass our proficiency exam in French, and
they're seniors, and they're graduating, so you need to give your spot to them.” And so I was very upset, and that was
pretty much end of my French pursuits in college.
Both of these seemingly insignificant events in her context, the lack of books on Israel from the Arab perspective and being
forced to drop her French class, had a significant effect on her individual complex system. Thus, as a result of the interactions
between her initial condition and her anti-ought-to self, she was pulled into her second shallow attractor state, settling down
with Arabic because it was “more difficult … wasn't European, so it'd be more of a challenge.” Up until this point, Rachel was
still not sure what she wanted to do with Arabic; she was only sure that she wanted to keep interacting with the language.
Rachel was propelled into her second phase shift by the interaction of her ought-to self and the elements from her social
context. Because of the changing political landscape as a result of September 11th, 2001, Arabic suddenly became a popular
language to study with student demand outpacing the availability of qualified instructors. Thus, although it was not her
original intent, Rachel felt “an obligation” to share her knowledge and help others who were interested in this “really
awesome language.”

Excerpt 14
R: But I also felt, I mean, something of an obligation, once I realized how much Arabic had become popular, and there
were so many students studying it, and this was sort of … a crisis in the Arabic teaching community because people
were like, oh Arabic is hard, we don't want to teach it. And they just pulled in whoever to teach it, I mean people who
weren't trained at all but happened to speak Arabic, and so, I felt like there's this whole generation of students who are
really interested in this really awesome language and they weren't really being served in terms of, you know, devel-
oping their skills and really being able to pursue it.
Because of the interaction of these variables, her initial condition, her ought-to self, and her anti-ought-to self, Rachel was
firmly situated in a deep attractor state at the time of the interview. She started studying Arabic because of the “challenge” it
presented and because it was something unusual. Her identity as an Arabic language instructor and researcher has helped to
strengthen the pull of this deep attractor state, as she felt an “obligation” to teach others Arabic, partially because of its
increased popularity. Her anti-ought-to self motivation was unwittingly combined with an ought-to self motivation as a result
of the events on 9e11, something that was not necessarily a positive experience for her.

Excerpt 15
R: When I came back in summer to the United States, then I was like in a different world in regards to people
encouraging Arabic. Before when I started it people felt like it was kind of strange that I would go for Arabic, like you
know “Why would you do that?” and then as soon as I came back in December of 2001 “oh you're going to have a great
job, you picked this really important language” and I was just kind of like, I did, well, ok, this is interesting, I don't want
to give up Arabic, but … I really like it, but I no longer have control over, what studying Arabic … what that means to
someone … Growing up people always thought I was really strange for wanting to study foreign languages … And then
I came back and then, it was, like, all of a sudden, I've done something smart, without even knowing it, and I wasn't
even really sure that I liked.
This change in attitude illustrates how the larger social context is a continuously changing and adapting system, and how
the elements in this larger system affect Rachel's individual complex system. The change of attitude towards Arabic because of
a major event in the environment also caused a significant shift in the composition of Rachel's individual complex system e
the shift of her anti-ought-to self to her ought-to self because of Arabic as suddenly seen as a “smart” and marketable skill to
have, which was not the case pre 9e11. Unlike Alex, Rachel's major phase shift did not happen during the change from the
foreign to second language context; in her case, the shift happened because of a sociopolitical shift that happened in her
absence in the Arabic as a foreign language context (the U.S.). Despite society's attitudinal change with regards to Arabic and
Rachel's unconscious synthesis of her anti-ought-to self motivation with an ought-to self motivation, she has remained firmly
embedded in her deep attractor state of a language educator because of her love of the language. Teaching allows her to keep
up a high level of language, and she feels that she is fulfilling her “obligation” of helping others who are interested in Arabic.

Excerpt 16
R: I really like it, it allows me to use the language, it allows me to keep up a high level of the language. I mean, especially
from teaching upper level classes or finding authentic materials …
 A.S. Thompson / System 67 (2017) 38e49 47

Thus, Rachel continues to grapple with her ought-to and anti-ought-to selves, which are working in harmony, keeping her
in her current deep attractor state. She still has a desire to excel in this “difficult” language (anti-ought-to self); however,
because of the changing political landscape, studying Arabic became something “smart” to do (ought-to self). The fluctuation
of these selves illustrates the dynamicity of her individual complex system (Fig. 3).

6. Discussion

The focus of this study is the dynamicity of the language learning experiences and the potentially synergistic relationships
of the ought-to and anti-ought-to selves, elicited via narratives with a CSDT-inspired analysis. The topic at hand is important
for several reasons: the influence of the “other” aspect of Higgins’ (1987) Self-Discrepancy Theory (the context) to the
development of the ought-to and anti-ought-to selves, the synthesis of motivation research and CDST, the investigation of
motivation for languages other than English, and theory-building in terms of re-conceptualizing the L2MSS. The trajectories of
the language learning journeys for Alex and Rachel are unique, yet they have an aspect in common e the non-linear
complexity of their individual complex systems and how these systems changed with the influence of seemingly insignifi-
cant (or in some cases, significant) events in the context. The various elements in the larger complex system (the context)
constantly interact with the systems under investigation: the individual complex language learning motivational systems of
the participants.
Both of the participants exhibited moments of synergy between their ought-to and anti-ought-to selves, each revealing a
combination of these “selves” in at least one of their attractor states. Additionally, the analysis of the “I” versus “other” aspect
of Higgins' (1987) Self-Discrepancy Theory with regards to the L2MSS originally proposed in Thompson and Va squez (2015)
was clearly illustrated during the development and fluctuation of the ought-to and anti-ought-to selves. Although Alex was
primarily motivated by his ought-to self (the prospect of employment) during his initial Chinese learning experience, his TA's
insistence that he “was not good at Chinese” was a seemingly insignificant element that promoted a change in the system. His
anti-ought-to self was created when he wanted to continue to take Chinese to “prove his TA wrong,” illustrating his desire to
do what he was told he could not do. In his case, the interaction between his ought-to and anti-ought-to selves, in conjunction
with the phase shift that made him realize that he wanted to teach Chinese, collectively propelled him into a deep attractor
state (de Bot et al., 2007). In Rachel's case, she was strongly motivated by her anti-ought-to self when she chose Arabic, a
language that pre-September 11th, not many people studied. She specifically wanted to excel at a language that was
“challenging” and liked the thought of doing something against the societal norm. To her surprise (and perhaps dismay),
when she returned from a study abroad experience in Egypt post-September 11th, she found that she “had done something
smart, without even knowing it.” Suddenly, her anti-ought-to self was unwittingly merged with an ought-to self, and in order
to continue studying the language that she loved, she had to accept this change. The interactions of Rachel's unintentional
shifts in motivational self guides illustrate the dynamicity of her individual complex system and the importance of the larger
complex system (the context), as Larsen-Freeman and Cameron (2008) state.
The relationship between the individual and the context is examined in terms of how it helps to shape the “context-
dependent” selves e the ought-to and anti-ought-to selves. From the narrative data used in the present analysis, the anti-
ought-to self serves as a logical counterpart to the ought-to self, as these two latent dimensions seem to have a synergis-
tic relationship with the data of these two participants. However, it would also be possible to conceptualize the anti-ought-to
self as a symbiotic counterpart to the ideal self with the anti-ought-to self using contextual influences to enhance the imagery
component of the L2MSS. As theorized in the literature review section, the learner with a strong anti-ought-to self can be
conceptualized as “dominant,” pushing against societal expectations (the more “submissive” element) as an impetus for
motivation.
Related to the L2MMS, the evidence presented in the current study, as well as the quantitative evidence found in
Thompson and Liu (2017), Liu and Thompson (2017), and Thompson (2017), Huensch and Thompson (2017), and Alharbi
(2017) presents clear argumentation for adding the anti-ought-to self dimension to the model, which is relevant for both
foreign language and second language contexts. As the model exists currently, there are two main parts: 1. the psychological
aspect of selves (consisting of the ideal and ought-to selves) and 2. the learning experience, which influences the develop-
ment of the selves aspect of the model. The evidence suggests that adding a third self dimension e the anti-ought-to self e to
the first part of the L2MSS would result in more explanatory power for the model as a whole. With the addition of the anti-
ought-to self, the L2MSS would maintain its two-part integrity with the aspects of self and the learning experience. The model
would still operate in the same way in terms of the learning experience aspect functioning as the formative influence of the
psychological aspects of selves e ideal, ought-to, and anti-ought-to e with the added dimension of the anti-ought-to self
affording agency to the language learner, who becomes the dominant character in the individual-context relationship. The
integration of the anti-ought-to self into the L2MSS is helpful for language educators to more fully understand the “rebellious”
(Lanvers, 2016, 2017) aspect of their language learners.
Although this study is an illustration of two specific language learners’ journeys that include the anti-ought-to self, this
desire to succeed in the face of difficulty is present in many, if not all, language learners at some point in the language learning
journey. The results of the quantitative work done with this construct with the three-factor structure findings of the EFAs
performed clearly show that the anti-ought-to self is a related, although distinct, motivational construct of self in the L2MSS.
Language choice and the anti-ought-to self can be linked, such as in Thompson (2017), which shows the significantly lower
anti-ought-to self of Spanish language students, as some chose to study Spanish as a result of societal pressure based on the
48 A.S. Thompson / System 67 (2017) 38e49

prevalence of Spanish in the U.S. However, an anti-ought-to self emerged with the Chinese EFL students in Thompson and Liu
(2017) and Liu and Thompson (2017) as well, a finding that supports the prevalence of the anti-ought-to self with commonly-
studied English as the language and a non-Western context (Laurin et al., 2013) as the setting. Although learning English is
popular in China, Thompson and Liu (2017) explain that students might feel familial pressure to major in a field more lucrative
than language study. Thus, the implication is that the anti-ought-to self could present itself in very different types of language
learning contexts, even in those where learning English is expected of educated individuals.
As the anti-ought-to self is a new area of inquiry in motivation research, more needs to be done regarding the existence
and development of this concept. For example, as the current study explicitly illustrated the importance of context for the
development of the anti-ought-to self, more data (both qualitative and quantitative) need to be collected in a variety of
contexts, as has been done for the other parts of the L2MSS. Although Thompson (2017) has collected preliminary data with
foreign language learners in the U.S. context, more data needs to be collected with learners from the diverse languages
studied in this context. It would also be informative to further explore why learners of some languages seem to have stronger
anti-ought-to self than others. Also interesting to investigate would be the existence of the anti-ought-to self in heritage
language learners in the U.S., as alluded to in Thompson and V asquez (2015). In contexts outside of the U.S., it would be
insightful to compare the anti-ought-to self of global English versus other languages.

7. Conclusion

Language learning is a difficult task no matter what the language or setting, so without an underlying desire to take on this
challenge, the language learning experience will be less pleasurable, and perhaps less successful in terms of proficiency (i.e.
Liu & Thompson, 2017). If students do not have the desire to respond to challenges or to defy expectations, their language
instructor needs to know this and have the ability to respond accordingly. Although difficult classroom activities could be
amotivating to students with a variety of different motivational profiles, it is likely that students with lower anti-ought-to
selves could shut down in the classroom if presented with activities that they perceive to be too difficult because their
lack of enjoyment for challenges. Although it is a language instructor's job to challenge students, knowing about the level of
the three psychological aspects of self for each student could help the instructor respond to students in a way that would
enhance their overall desire to accomplish the task at hand. Although a detailed discussion of the pedagogical implications of
the existence an anti-ought-to self in the classroom is beyond the scope of the current study, several other studies have
addressed this issue. For example, Huensch and Thompson (2017) found a significant correlation between the anti-ought-to
self and the desire to improve pronunciation. Thompson (2017) found that students who chose to study Spanish in the U.S.
context had a lower anti-ought-to self than those who chose to study other languages, and using data from Chinese learners
Thompson and Liu (2017) found that the existence of an anti-ought-to self partially depended on the language in question. As
the issues of attitudes towards pronunciation and language choice are important issues for language instructors to consider
with their students, understanding students' motivational profiles would help these instructors design activities and interact
with their students in a more efficient manner.
The anti-ought-to self, encapsulating the essence of psychological reactance, could be the missing link to explanatory
value of the L2MSS, incorporating reactance into SLA motivation research, as MacIntyre (2002) suggested. Some learners,
such as Alex, do not initially possess a strong anti-ought-to self, but later realize that they enjoy defying expectations. Other
learners, such as Rachel, develop this anti-ought-to self early in life, basking in the delight of doing the unexpected. In sum,
the concept of the anti-ought-to self is indicative of high-level performance in some individuals, and there is evidence that it
exists to some extent in all language learners. Many language students are motivated by the sentiment expressed by Larissa
Martinez, ESL student, illegal immigrant, and McKinney Boyd High School Class of 2016 valedictorian: “They told me I
couldn't, so I did.”

Acknowledgments

I would like to thank the participants, Alex and Rachel, for the willingness to share their stories.

References

Alharbi, F. (2017). The dynamics of the L2 motivational self system among Saudi study abroad students (Unpublished doctoral dissertation). Tampa, FL: Uni-
versity of South Florida.
de Bot, K., Lowie, W., & Verspoor, M. (2007). A dynamic systems theory approach to second language acquisition. Bilingualism: Language and Cognition, 10(1),
7e21.
Brehm, J. W. (1966). A theory of psychological reactance. New York: Academic Press.
Brehm, S., & Brehm, J. W. (1981). Psychological reactance: A theory of freedom and control. New York, NY: Academic Press.
r, K., & Luka
Csize cs, G. (2010). The comparative analysis of motivation, attitudes and selves: The case of English and German in Hungary. System, 38(1), 1e13.
€rnyei, Z. (2009). The L2 motivational self system. In Z. Do
Do €rnyei, & E. Ushioda (Eds.), Motivation, language identity and the L2 self (pp. 9e42). Clevedon, UK:
Multilingual Matters.
€rnyei, Z., MacIntyre, P., & Henry, A. (Eds.). (2015). Motivational dynamics in language learning. Clevedon, UK: Multilingual Matters.
Do
Goldberg, E., & Noels, K. A. (2006). Motivation, ethnic identity, and postesecondary education language choices of graduates of intensive French language
programs. Canadian Modern Language Review, 62(3), 423e447.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319e340.
Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational principle. Advances in Experimental Social Psychology, 30, 1e46.
 A.S. Thompson / System 67 (2017) 38e49 49

Hornberger, N. (2006). Negotiation methodological rich points in applied linguistics research: An ethnographer's view. In M. ChalhoubeDeville, C.
Chappelle, & P. Duff (Eds.), Inference and generalizability in applied linguistics: Multiple perspectives (pp. 221e240). Amsterdam: John Benjamins.
Huensch, A., & Thompson, A. S. (2017). Contextualizing attitudes toward pronunciation: Foreign language learners in the US. Foreign Language Annals (in
press).
Lamb, M. (2012). A self system perspective on young adolescents' motivation to learn English in urban and rural settings. Language Learning, 62(4),
997e1023.
Lanvers, U. (2016). Lots of selves, some rebellious: Developing the self discrepancy model for language learners. System, 60, 79e92.
Lanvers, U. (2017). Policy Ping Pong, Brexit, Habitus: The British L2 Motivation Learning Landscape. Modern Language Journal (in press).
Larsen-Freeman, D., & Cameron, L. (2008). Complex systems and applied linguistics. Oxford: Oxford University Press.
Laurin, K., Kay, A. C., Proudfoot, D., & Fitzsimons, G. J. (2013). Response to restrictive policies: Reconciling system justification and psychological reactance.
Organizational Behavior and Human Decision Processes, 122, 152e162. http://dx.doi.org/10.1016/j.obhdp.2013.06.004.
Linde, C. (1993). Life stories: The creation of coherence. Oxford: Oxford University Press.
Liu, Y., & Thompson, A. S. (2017). Language learning motivation in China: An exploration of the L2MSS and psychological reactance. Manuscript submitted for
publication.(submitted for publication).
MacIntyre, P. D. (2002). Motivation, anxiety and emotion in second language acquisition. In P. Robinson (Ed.), Individual differences and instructed language
learning (pp. 45e69). Amsterdam: John Benjamins.
Markus, H. R., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954e969.
Ryan, S. (2009). Self and identity in L2 motivation in Japan: The ideal L2 self and Japanese learners of English. In Z. Do €rnyei, & E. Ushioda (Eds.), Motivation,
language identity and the L2 self (pp. 120e143). Clevedon, UK: Multilingual Matters.
Savin-Baden, M., & Major, C. H. (2013). Qualitative research: The essential guide to theory and practice. New York, NY: Routledge.
Talmy, S. (2010). Qualitative interviews in applied linguistics: From research instrument to social practice. Annual Review of Applied Linguistics, 30, 128e148.
Thompson, A. S. (2013). The interface of language aptitude and multilingualism: Reconsidering the bilingual/multilingual dichotomy. Modern Language
Journal, 97(3), 685e770. http://dx.doi.org/10.1111/j.1540-4781.2013.12034.x.
Thompson, A. S. (2017). Language learning motivation in the United States: An examination of language choice and multilingualism. Modern Language
Journal (in press).
Thompson, A. S., & Erdil-Moody, Z. (2016). Operationalizing multilingualism: Language learning motivation in Turkey. International Journal of Bilingual
Education and Bilingualism, 19(3), 314e331. http://dx.doi.org/10.1080/13670050.2014.985631.
Thompson, A. S., & Liu, Y. (2017). Multilingualism and emergent selves: Further development of the anti-ought-to self. Manuscript submitted for publication.
(submitted for publication).
Thompson, A. S., & Va squez, C. (2015). Exploring motivational profiles through language learning narratives. Modern Language Journal, 99(1), 158e174.
http://dx.doi.org/10.1111/modl.12187.

Amy S. Thompson, Ph.D., is an Associate Professor of Applied Linguistics and currently the Associate Department Chair in the Department of World Lan-
guages at the University of South Florida. She is also currently the graduate director for the Ph.D. program in Linguistics and Applied Language Studies, and
she teaches a range of graduate level theoretical and methodological courses in applied linguistics. Her primary research interests involve Individual Dif-
ferences in Second Language Acquisition and the interaction of these IDs and multilingualism. Examples of her research can be found in journals such as the
Modern Language Journal, TESOL Quarterly, Foreign Language Annals, and the International Journal of Bilingual Education and Bilingualism.