(PDF) Back to the Initial Condition: Experiences that Trigger the Need to Develop Language Skills Among University Spanish Stude

Back to the Initial Condition: Experiences that Trigger the Need to Develop Language Skills Among University Spanish Students Rakel Österberg, Alejandra Donoso & Enrique Sologuren Abstract: The present study investigates how the L2 learning experience shapes multilinguals' motivation to improve their writing skills in Spanish, a language they study at university level. Within the Second Language Motivational Self System Theory (L2MSS), the L2 learning experience is often neglected when it comes to interpreting and discussing findings on motivation. We propose that the analysis of the L2 learning experience should acknowledge both formal and informal settings, such as the family context, in order to research why and how a person chooses to develop certain linguistic abilities like writing. Given that the L2 learner experience is spatially and temporally bound, the concept of sensitivity to initial conditions, i.e., the conditions or changes that may trigger the learner to move towards enhanced language learning, will be adopted to research the L2 learning experience. The aim of the study is twofold. First, to investigate the Spanish learning experience as a function of the initial conditions referred to by the participants. Second, to research how the Spanish ideal and ought-to selves are related to the students’ motivation to use and learn or maintain Spanish skills. The data consist of a questionnaire answered by multilingual university students of Spanish (n=21) and written autobiographical narratives. The participants belong to two groups: one group of Spanish heritage language learners (n=10) and one group of native-Swedish speaking L3 Spanish learners (n=11). Overall, family reasons, previous school experiences, high proficiency in Spanish, previously acquired languages, contact with the linguistic communities, language use and context are all aspects tightly connected to the students’ motivation to learn, maintain and develop Spanish language proficiency. The L2 learning experience, as a wider construct dependent on the sensitivity to initial conditions, must be considered to understand the complex nature of multilingual literacy development. Introduction The field of second-language motivation has grown prominently in recent decades, and there are robust findings on the effect of motivation on https://doi.org/10.15626/hn.20224810 226 HumaNetten Nr 48 Våren 2022 second/foreign language (L2) learning outcomes (Al-Hoorie 2018; Dörnyei 2019a; Dörnyei & Ushioda 2009). As regards writing skills, several studies have found motivation to be a factor of great importance for the development of L2 writing skills, together with other individual differentiating factors (Kormos 2012: 400; Wilby 2020). Despite these findings, very few studies have specifically addressed how the development of academic writing skills might be related to motivation, especially when multilingual 1 students are engaged. Therefore, in the present study, we aim to fill this gap by focusing on the Spanish language classroom at university level in Sweden, a heterogeneous learning environment where Spanish learners with various language learning experiences coincide. In the university Spanish classroom, two important groups emerge: those who have learned Spanish in formal settings as a third language (L3), and those who have had varying types of formal education in Spanish 2 in addition to contact with Spanish since early childhood, so-called heritage language (HL) speakers. While these two groups of students often share the same classroom in Sweden, little is known about their different experiences of the language acquisition/learning processes. To account for the various learning experiences of the students at hand, a model that integrates both school experiences and learning gained outside the school context is needed. The L2 motivational self-system (L2MSS), an approach in which motivation is conceptualised as the individuals’ capacity to envision themselves as proficient speakers of a certain language, has previously been proposed in the literature (Dörnyei 2005; 2009). However, this system, designed to research L2 learning, does not adequately capture multilinguals’ self-based approaches (Henry 2017), especially when these learners engage in formal studies of a language they know from an early age, like Spanish HL learners. The L2MSS contains three main constructs. The first two components, the ‘ideal L2 self’ and the ‘ought-to self’—both explained below—have generated empirical evidence when it comes to explaining L2 acquisition (Al-Hoorie 2018; Boo, Zann & Ryan 2015). The third component, the L2 learning experience, is defined as “the perceived quality of the learner’s engagement with various aspects of the learning process” (Dörnyei 2019b: 20). It is based on what is experienced in an actual context, with lessons and resources, in which the participation of peers, parents and teachers is included (Dörnyei 2019b). The L2 learning experience has demonstrated powerfully predictive ability of motivated behaviour (2019b) and conse- quently has a clear role in second/foreign language learning. Nevertheless, its importance has often been neglected (Dörnyei 2019b). In the present 1 Multilinguals are those who use three or more languages (Silva-Corvalán 2014: 1). 2 In Sweden, heritage language learners can attend mother-tongue instruction or learn Spanish as a foreign language. Students may also be enrolled in both types of instruction (Hyltenstam & Österberg 2010). Österberg, Donoso & Sologuren: Back to the Initial Condition 227 study, we assume that aspects of the L2 learning experience are crucial to investigate the students at hand, who have heterogeneous learning trajectories within and outside of formal education. Consequently, our research aims for a deeper understanding of the conditions related to the use, learning and maintenance of Spanish language skills, especially concerning writing. The nature of the L2 learning experience is spatial and temporal since language learners are real persons “necessarily located in particular cultural and historical contexts” (Ushioda 2009: 216). To capture the L2 learning experience, the concept of sensitivity to initial conditions will be applied. This concept has been used as part of the Complex Dynamic System Theory (de Bot 2008) and stems from Chaos Theory (Lorenz 1963). It holds that a small change in the initial conditions may generate unpredictable changes in the whole system (de Bot 2008). Language, as a complex system, is no exception (Larsen-Freeman 1997; 2014). In this study, the initial conditions are operationalised as events, changes or conditions that may have triggered the learner to move towards enhanced language learning. In summary, we aim here to investigate students’ motivation to develop and maintain an advanced proficiency level of Spanish, with a special focus on their Spanish language learning experiences. Furthermore, given the importance of writing to achieve academic success, motivation for writing will also be studied. Since motivation for writing cannot be studied in isolation it will be studied in relation to motivation to learn Spanish in general. The research method is qualitative, and an online questionnaire provides background data together with an autobiographical narrative. Because of the changing nature of language learning and teaching in a multilingual world, our study answers international calls (see, e.g., Douglas Fir Group 2016) to work with authentic data from individuals developing language skills out of societal and personal needs (Douglas Fir group 2016). The participants (n=21) are HL and L3 university students of Spanish describing how they learn and use languages in their lives. Our research questions are: 1. In terms of the L2MSS, how are the Spanish ideal and ought-to selves expressed in the participants’ answers and in their autobiographical narratives? 2. In terms of the L2MSS, how is the Spanish learning experience expressed in the participants’ answers and in their autobiographical narratives? 3. How do the students’ feelings towards and thoughts about their writing relate to their Spanish language learning experience and their motivational selves? 2. Theoretical background In the following sections, the L2MSS is presented, in particularly the L2 learning experience (2.1), the motivation to maintain and develop language 228 HumaNetten Nr 48 Våren 2022 skills, especially in a HL (2.2) and finally, studies on motivation to improve writing are presented (2.3). 2.1 The L2MSS and difficulties in approaching the L2 learning experience In L2 research, motivation is defined as a goal-oriented action. It explains why we start learning something, and how persistent and intense this learning is (Dörnyei & Ryan 2015). Motivation is dynamic and changes over time (MacIntyre, Baker & Sparling 2017), and is said to be context- dependent and situation-bound (Ushioda 2009). For decades, motivation for acquiring languages was analysed according to Gardner's socio-educational model, which was the dominant theory (Gardner 1985). This model distinguishes between integrative motive, that is the wish to approach the target language culture and speakers; instrumental motivation, which corresponds to external demands and expectations; and integrative motivation, which refers to inner motivation. The L2MSS (Dörnyei 2005; 2009) is based on two psychological theories: the construction of possible selves (Markus & Nurius 1986) and self-discrepancy theory (Higgins 1987). According to Higgins (1987), motivation arises when a person senses a gap between present and future selves that is not too large. This insight makes the person put energy into learning. In other words, the motivation to learn a language can be conceptualised as the individuals’ capacity to envision themselves as proficient speakers of a certain language. As already stated, the L2MSS includes the ideal L2 self, which corresponds to the learner’s intrapersonal and emotional disposition towards the language; the ought-to self, which is the pragmatic, external driving force to learn (that is demands and expectations from oneself, school, parents, friends, peers); and the L2 learning experience, as defined above (Dörnyei 2019b: 20). In the L2MSS, these selves are future-oriented guides, whereas the L2 learning experience is context-bound and therefore dependent on individuals’ feelings and experiences in contact with the language and its users. In recent studies on motivation, the L2 language learning experience has been pointed out as an important but often neglected factor in interpreting and discussing findings on motivation and language use (Czisér 2019; Dörnyei 2019b). A conceivable reason for this neglect may be the difficulties connected to the operationalisation of the learning experience, since it is complex, heterogeneous and changing over time. Indeed, experiences are individual and hard to generalise or simplify. Ushioda (2009: 216) points out that even when context and learning experiences are integrated in a theory of motivation, they are often considered independent variables and beyond individual agency. Therefore, Ushioda (2009) presents another way to understand context in relation to motivation, suggesting that a person’s beliefs and perceptions regarding language learning are “person- Österberg, Donoso & Sologuren: Back to the Initial Condition 229 in-context-dependent” and not “context-independent” (2009: 215). Thus, motivation assessed by future-oriented self-guides may change in relation to a person’s experiences. The issue of changes in motivation over time is also problematised by Thorsen, Henry and Cliffordson (2017: 588), who argue that the discrepancies between the current and future selves have not yet been sufficiently operationalised in the L2MSS. The current self is defined as a person’s present self-perception (Thorsen et al. 2017). Motivation can arise in an attempt to reconcile potential future-oriented selves with the present self, according to Higgins’ Self Discrepancy Theory (1987). However, despite the importance of the current self for the L2MSS, this notion has not been sufficiently explained, nor the present/future discrepancy that it creates, according to Thorsen et al. (2017). The L2MSS acknowledges the complexity of studying motivation. In recent years, several scholars have formulated the need to account for the dynamic and complex nature of learning as a result of a motivated behaviour (Dörnyei, MacIntyre & Henry 2014; Henry 2017). To this end, researchers in the field of Complex Dynamic System Theory (CDST) have suggested that a non-linear perspective, and a broader view of the different phenomena underlying complex systems such as motivation, may promote a more profound understanding of how motivation works. To approach motivation and do justice to its changing nature, there is a call for further qualitative studies that allow for the inclusion of aspects not yet considered. One such aspect is the L2 learning experience, which has proven to be “not only a strong predictor of various criterion measures but … often the most powerful predictor of motivated behaviour” (Dörnyei 2019b: 19). However, studies concerning the L2 learning experience have been criticised because they neglect the heterogeneous learning contexts where today’s students learn languages as well as previous learning experiences or language skills acquired outside school, through family or naturally occurring language exchange. Despite criticism, numerous studies have indicated the L2 learning is crucial for understanding motivation (Dörnyei 2019b: 22). Indeed, it was the most powerful predictor of high levels of the ideal L2 self in English, among many variables in a study by You and Dörnyei (2016), where it outperformed the other two constructs of the L2MSS in terms of predictability. Lamb (2012) found that the L2 learning experience explained the difference in motivation and L2 proficiency in a rural setting compared to an urban setting in a study of Chinese learners of English. Oyserman and James (2009) conducted a study on the importance of circumstances for the elaboration of selves. In sum, one conclusion of these studies is the need for further research on the L2 learning experience and its relation to the psychological theories behind the L2MSS. 230 HumaNetten Nr 48 Våren 2022 2.2 Motivation to maintain and develop language skills Regarding multilinguals and their likelihood of maintaining early childhood language skills, high levels of motivation have proved necessary to develop the weaker language(s) in their repertoire (Montrul 2010). However, few studies address multilinguals’ motivation. Family heritage can be decisive for the choice of language studies, but not necessarily for the learning outcome. Olsen (2018) compared the L2MSS (the ideal L2 selves, the ought-to selves, the L2 learning experience, instrumentality, integrativeness and non L2-goals) of 1) HL and non-HL learners of Maori 3 and 2) HL and non-HL learners of foreign languages (FL) (such as Russian and Spanish) in New Zealand. They found, on the one hand, that HL learners of Maori showed motivation in line with the L2MSS in a more extensive and complex way than the non-HL learners studying Maori. They measured positively on all the motivational variables of the L2MSS, including the L2 learning experience, which was measured as attitudes towards the learning situation in university context (Olsen 2018: 290). On the other hand, in the same study, similar results were found when comparing HL with learners of other FLs studied in the context of New Zealand. Notably, these results showed no significant differences between the groups regarding the L2 learning experience. According to these findings, HL status seems related to motivation as defined by the L2MSS. The status of the target language, and its importance in the surrounding society, also seems important for the motivational profile of the HL speaker. Two other studies comparing HL and FL learners of Chinese (Noels 2005) and German (Comanaru & Noels 2009) indicate that for all learners, but especially HL learners of these languages, the studied language is an important aspect of the self-concept. In other words, HL learners’ Ideal L2 self-concept implies that they appreciate the HL as a part of who they are. Kim et al. (2017) came to a similar conclusion in a case study of seven Korean HL speakers in the US who negotiated their HL identities with highly varying degrees of motivation. Contact with actual or imagined HL communities proved to be an important aspect of motivation, something also observed in other studies, such as Mikulski’s (2006) and Yanguas (2010) studies of Hispanic HL speakers in the US. Yanguas (2010) found that integrativeness was the most powerful indicator of motivation for HL speakers of Spanish. Integrativeness is a notion introduced by Gardner (1985) which refers to positive attitudes towards the community of speakers of the studied language, which in the L2MSS is a part of the ideal L2 self (Dörnyei 2019a). A similar result was presented by Geisherik (2004) in her study on HL learners of Russian. Indeed, self-conceptions and identity processes seem to be narrowly linked to a speaker's motivation to develop and maintain their HL. Thompson (2020) arrives at the conclusion that negative attitudes and a 3 The teaching of Maori took place in a part of the country where Maori is spoken, which may also explain the importance to the L2 learning experience. Österberg, Donoso & Sologuren: Back to the Initial Condition 231 language’s low status in society can lead to low levels of motivation. Thus, not all early multilinguals experience positive effects from learning more languages. In the US, Spanish multilingual speakers more often develop the opposite attitude, referred to as a contentedly bilingual self (Thompson 2020; see also Henry 2017 for further details), which is a self-conception without the motivation to improve a language or to learn other languages. A similar tendency has been found among Swedish multilingual students who are highly proficient in English but have low motivation to learn a L3 (Henry 2012). These examples show how different self-concepts interact with each other and with the learning context. Furthermore, a study by Ball and Ellis (2008) showed that identity, learning context, and writing are closely related, and when a person learns and uses a new discourse, a new identity is also adopted. In line with this, Leeman and Serafini (2020) in a study conducted with mixed groups of Spanish L2 and HL learners of Spanish, found that students’ self-perceptions and identities reflected language ideologies and broader educational discourses. In this sense, HL speakers are seen by others as those who have an easy way into Spanish studies, whereas L2 learners are seen as the ones who slow down the speed and quality of the Spanish language classroom. 2.3 Motivation to improve writing We know, from our experience teaching Spanish as well as the literature, that academic writing is quite demanding for students in general (Bazerman 2016; Holm & Laursen 2011; Källkvist & Hult 2016; Miller, Mitchell, Thomas & Pessoa 2014), and especially for students writing in a language they have not mastered (Grosjean 2010), like multilingual writers who develop writing competence in several languages (Hornberger 1989; 2002; 2004). Writing in Spanish at university level in Sweden implies a development of multilingual literacy, since previous knowledge of writing both in Swedish and in English is required to enrol in Spanish studies at university level. As stated above, the groups of Spanish learners are heterogeneous, with various degrees of preparedness and past experiences. According to Ball and Ellis (2008), such experiences can influence the motivation to write in different ways, strengthening or weakening self- confidence. Furthermore, students’ motivation to develop writing responds to social contexts of possibilities for biliteracy (Gentil 2011). Besides, the context represented by language policy documents can have a negative effect on a person’s self-confidence. According to Strömberg Jämsvi (2019: 221), multilingual students’ self-perceptions are influenced negatively by the description of multilingualism in terms of deficiency in Swedish language policy documents. Thus, the L2 learning experience and students perceptions may, in this sense, be crucial to understanding their willingness to improve their writing. 232 HumaNetten Nr 48 Våren 2022 Regarding L2 writing, several studies (Ghaith 2019; Haagensen 2007; Kormos 2012) have found that motivation and cognitive variables explain much of the varying degrees of L2 writing improvement. However, HL speakers' motivation for writing remains a research gap. In a recent study, Hanson Seibert (2019) investigated the self-regulated motivation of Spanish FL and HL speakers in their development of Spanish writing skills to identify ways of enhancing motivation. The participants’ context in Hanson Seibert’s study was similar to the context of our students, that is, the teaching of academic writing was mainly designed for FL learners. Thus, the author highlights that writing abilities and motivation vary as a function of students’ previous contact with the Spanish language, either as HL or FL learners. Hanson Seibert’s results indicate significant differences between the groups in terms of extrinsic motivation and levels of confidence. According to the author, HL learners exhibit greater confidence and extrinsic motivation than FL learners (Hanson Seibert 2019: 205). Furthermore, motivation in HL learners has been studied by Torres, Estremera and Sherez (2019), but in relation to proficiency levels. Torres et al. (2019) conducted a study that showed that motivation measured according to the L2MSS (the ideal L2 self and the ought-to self) seems to explain the variation in proficiency levels, together with a set of other individual differences, such as anxiety and willingness to communicate. Finally, within the Swedish university context, L3/HL writing in academic settings has been researched by Donoso, Österberg and Sologuren (2021), but this study does not elaborate on motivation. Despite these findings, the area of motivation and HL writing has somehow been neglected, even though motivation has been viewed as one of the most important individual factors determining the quality of writing skills (Jang & Lee 2018; Kormos 2012). 3. Method, participants and materials In the following sections, the design of the questionnaire (3.1), the data collection and the participants (3.2), the data from the autobiographical narratives (3.3), the analytical categories (3.4) and ethical considerations (3.5) are described. Finally, the validity and reliability of the study are discussed (3.6). 3.1 The questionnaire A combined background and motivation questionnaire has been designed to assess autobiographical data on age of onset of Spanish- and Swedish acquisition/learning and self-estimated proficiency together with an autobiographical narrative on the informants’ L2 learning experiences of all their languages during their lifetime. The questionnaire consists of 53 questions, but only 16 are considered pertinent for this study (see Appendix I). Österberg, Donoso & Sologuren: Back to the Initial Condition 233 The questionnaire builds on Thorsen et al. (2017), Henry and Thorsen (2018) and Serafini (2013). It contains multiple-choice questions or utterances assessed using Likert scales 4. In addition, four open-ended questions are posed providing the students the opportunity to elaborate on their Spanish-language learning trajectories. The questionnaire is complemented by a narrative task (an autobiographical narrative) at the end of the questionnaire. The multiple-choice questions measure topics such as linguistic background, proficiency, current selves and the ideal Spanish and Spanish ought-to-self. Table 1 shows how the different topics relate to the questionnaires that have guided the formulation of the items. Table 1. Topics of the close-ended questions of the questionnaire Topic Question Source Linguistic background Q5 Loosely based on Serafini (2013) Proficiency scales, Spanish Q12, Q20 Proficiency scales, Swedish Q26 Proficiency scales, English Q28 Current Spanish self Q21, Q22, Q25 Loosely based on Thorsen et al. (2017). Current Swedish self Q27 Current English self Q29 Spanish ideal and ought-to Q33, Q41 Loosely based on Henry and selves Thorsen (2018) In the autobiographical narrative task, respondents describe their experiences learning the various languages in their lives. To facilitate this, 4 The Likert scales have seven levels ranging from 1 (‘strongly disagree”) to 7 (“strongly agree”), see further footnote, Appendix 1. The seven-level design was chosen to enable a nuanced description of the different proficiency and motivational scales. The possibility to respond “neutral” was provided to give ample possibilities to answer. Only one item (Q8) has five levels, and is to be considered an exception and early mistake in questionnaire design. Despite this, Q8 answers are useful. 234 HumaNetten Nr 48 Våren 2022 respondents could write in Swedish, English or Spanish. The use of autobiographical narratives to analyse motivation is well established (see e.g., Serafini 2020, Thompson 2020, Ushioda & Dörnyei 2012: 402, Yung 2018 and Österberg 2008). The wording of the last item is as follows: Please choose a language (Swedish, Spanish or English) and tell your story about how you have learnt and how you use the different languages in your life. Try to be as detailed and exhaustive as possible but try also to exclude personal data such as names of schools, villages or cities where you have lived and important friends. This is thought to be a text so feel free to start, leave and go back to this text before submitting your answer. The choice to present items assessed by Likert scales in the first part of the questionnaire was intentional, to ensure that similar data were collected, and, more importantly, to activate and make explicit the participants’ beliefs about learning Spanish and other languages, which, in turn, was intended to spur the final autobiographical narrative. 3.2 Data collection Data were sampled from undergraduate students enrolled in Spanish studies at two Swedish universities, many of whom were in teacher education programmes for (upper) secondary school. Sampling was made twice in 2021, first in May, at the end of the Spring term, and again in October, in the middle of the Autumn term. The invitations to participate were distributed through the university course platforms and were intended to reach all undergraduate students in Spanish. The total number of Spanish students invited to the study has been estimated at 150. Far from all of these students were 100 per cent active in studies at the time of sampling, which may have contributed to the low response rate, as well as the sampling’s timing at the end of the term. Approximately 15 per cent of the students responded and received a link to the questionnaire. To reach more participants, some students who recently finished their studies (n=5) were also invited to participate. In sum, a convenience sampling was conducted throughout the data collection. The questionnaire was administered to 31 students of whom 23 completed it. Several reminders were sent via the survey tool (Survey and report). Table 2 shows the sample and the response rate. Österberg, Donoso & Sologuren: Back to the Initial Condition 235 Table 2. Data collection, participants and response rates Call/ Total Respon- Valid Discarded Response Response Invitation number of ses respon- responses rate/ rate/ participants ses total partici- number pants of invited partici- pants -May 14 9 8 1 9% 64% October 17 14 13 1 28% 82% Total 31 23 21 2 15% 74% Responses from participants who learned Swedish after puberty (n=1) were excluded from the study to ensure that all participants had studied Spanish in the Swedish school system. As shown in Appendix II, the group of HL speakers consists of 9 participants exposed to Spanish from birth and to Swedish early (at preschool age), and 2 participants in contact with Spanish since birth who acquired/learned Swedish between 7 and 13 years of age. The group of L3 Spanish learners consists of 10 participants who began formal Spanish education between ages 12 and 15. All participants, both Spanish HL and L3 speakers, meet the course entry requirement of English proficiency, and therefore, as a whole, they are labelled as multilingual speakers (the term L3 is chosen for those who have learned Spanish through formal education since they have already learnt, at least, English). The uneven gender representation (4 males and 17 females) reflects proportions typical of language courses at university (Nordenstam 2003). In Figure 1 (section 3.4), we provide the categories used to capture our participants’ lifetime contact with Swedish and Spanish. LE-SP means that the person was born and raised in a Spanish-speaking country, while LE- NSP the opposite, i.e., “not born or raised in a Spanish-speaking country”. LE-F indicates Spanish acquired in Sweden in the family context, which is the case for most of the HL participants in this study, since they were born in Sweden. The participants received pseudonyms based on their linguistic background. The L3 speakers were given names that began with T and the HL speakers’ pseudonyms began with H. The students self-estimated their overall proficiency in Spanish. This is a subjective way of measuring proficiency that has been used in many studies (see, e.g., Ling, Elliot, Burstein, McCaffrey, MacArthur & Holtzman 2021). The students rated their proficiency as high (see Appendix III), which was expected, as they were enrolled in university courses in Spanish. The results 236 HumaNetten Nr 48 Våren 2022 from the responses to a five-level Likert scale have been recalculated in order to fit a seven-level Likert scale. The HL speakers present the highest average (HL: 6.23), while the L3 speakers rate their proficiency a bit lower (L3: 4.62). The average of all participants is 5.46. As regards self-rated proficiency in writing, the average of the whole group was 5.8 on a seven-level Likert scale. The participants consider the development of writing skills in Spanish to be important both from a pragmatic and from a personal perspective. One HL speaker, Hulda, rates the importance of developing writing skills as high for professional reasons, while another HL speaker, Hugo, seems more personally oriented. The background data offer a source of information that will be used in the discussion. 3.3 The data from the autobiographical narratives The corpus of written narratives includes 21 texts with a total of 8,136 words. The average word count of the texts is 387 (range 83–1,227). We are aware that the fact that the respondents only took the questionnaire once may affect the results, increasing the significance of incidental factors such as tiredness or loss of focus. Therefore, we permitted students to take breaks while responding to the questionnaire and return at a later point to finish it. More important than the repeated tasks of this study’s longitudinal design has been our intention to avoid respondents complying with or intuitively adapt answers to the researchers’ agenda. Therefore, we preferred not repeating the research task and consider one autobiographical text per person sufficient for the study’s scope. Respondents chose to write narratives in Swedish (13), Spanish (7), and in one case both Swedish and Spanish (1). Two participants (Hila and Horacio) chose to write about their learning of Swedish. Hila mentioned Spanish as important for the acquisition of Swedish. 3.4 Analytical categories Following Braun and Clarke’s thematic analysis (2006), we have identified the different selves and the Spanish language learning experience through a deductive procedure in which the different constructs—the ideal L2 self 5, the ought-to self6, and the L2 learning experience 7—are identified. In the analysis, these constructs are labelled Spanish ideal self, Spanish ought-to self and Spanish learning experience, since the term ‘L2’ might be misleading when referring to HL learners and L3 learners. The current self 8 5 See Examples 1–3. 6 For an example of the ought-to self, see Example 4. 7 We use the ideal L2 self and the L2 learning experience, terms used in previous research. Thus, when referring to the L2 ideal self and the L2 learning experience, we are actually referring to the participants under study. 8 For an example of the current self, see Example 7. Österberg, Donoso & Sologuren: Back to the Initial Condition 237 has also been tagged, since it has been pointed out as important for the operationalisation of the L2MSS (Thorsen et al. 2017). As regards the Spanish ideal self, the ought-to self and the current self, the absence, vague presence or explicit and detailed presence of each concept in the texts was noted. Anytime a value was expressed as being connected to the informant’s images, emotions and ideas about themselves as a speaker of Spanish, the utterance was tagged. Descriptions clearly related to the future were tagged as a Spanish ideal self. If the self is described as a future self in the past, it will be noted as Spanish ideal self in the past. If it rather relates to a present, already achieved status, it will be considered a current self. We use the notion of sensitivity to initial conditions, in accordance with the CDST (Larsen-Freeman 2014), in our thematic analysis of the narratives. To understand how sensitivity to initial conditions works, we have decided to mark the initial conditions that emerge in the narratives. These initial conditions may correspond to events, changes or conditions which the learner interprets as crucial for a change or move towards enhanced language learning. In the present analysis, initial conditions can be expressed with different degrees of commitment towards what is expressed. We have categorised them as neutral (IC-NEUTRAL), positive (IC- POSITIVE) or negative (IC-NEGATIVE). In addition to the above, and as a part of the Spanish language learning experience, we have also taken into consideration our students’ language use and learning environment. Since learning is context dependent, both aspects are of importance. Thus, we have considered experiences within and outside of the classroom. Below, we explain how we identified the above-mentioned categories in the narratives. The learning environment is where learning has taken place. LE-F means that Spanish is acquired in the family context. LE-EARLITT indicates that early literacy in Spanish is acquired in school. LE-HLI refers to HL instruction classes while LE-L3 stands for third language instruction (so called Foreign Language instruction). Language use is described as a part of family life (USE-FAM), or contact with friends (USE-FRIENDS). USE-ENT indicates Spanish used for entertainment, such as watching films, reading, etc. USE-LEARN indicates the learner uses the language as a medium to learn the language (see Example 8). USE-PROF indicates using Spanish at the workplace, USE- STUD indicates using Spanish when studying and USE-TRAVEL indicates using Spanish when travelling. Finally, language used for writing has been coded USE-WRITE. Figure 1 summarises how the Spanish language learning experience is investigated in our data, and what abbreviations are used. The Spanish language learning experience consists of the initial condition, the learning 238 HumaNetten Nr 48 Våren 2022 environment and language use. These categories will be further analysed in relation to writing later in this article. Figure 1. The components of the Spanish language learning experience 3.5 Ethical considerations The questionnaire was preceded by a request for informed consent. The respondents were instructed not to include personal data or names in the written tasks. Any personal, identifiable data in the narratives were omitted. The respondents could terminate their participation in the study at any point. Altogether, the study follows the Swedish Research Council’s recommendations on good research practices (Vetenskapsrådet 2017). 3.6 Validity and reliability To internally validate the results, the multiple-choice questions were based on previously used and validated questionnaires. The answers were used to control for and warrant the validity of the coding of open-ended questions and the narratives. However, and even though the qualitative analyses to be performed in this study may show tendencies when comparing the two groups, our aim is not to generalize the results. Reliability aspects have been considered when analysing the data by engaging all the researchers in the team. Consequently, when identifying the interpretative categories presented in Figure 1, these have been elaborated in an explorative and iterative way by the research team. Another reliability aspect is related to the selection of the participants, since the most motivated students may be those who respond to our call. We Österberg, Donoso & Sologuren: Back to the Initial Condition 239 are aware of this risk. Nonetheless, we deem that the reliability of the data is not seriously affected by the positive bias, which is inevitable when the aim is to investigate motivation among university students of Spanish. 4. Analysis The analysis is presented in several steps to respond to the research questions. In 4.1 (see RQ1), we present how the Spanish ideal and ought-to selves emerge in the closed-ended questions and in the narratives. In 4.2 (see RQ2), we present how the Spanish learning experience emerges, analysed thematically in accordance with the categories described in 3.4. In 4.3 (see RQ3), the participants' feelings towards and thoughts about their writing are related to their learning experience. 4.1 The emergence of the selves in the multiple-choice questions and narratives The results of the multiple-choice questions are presented in Appendix IV indicating that the two groups show similar results. HL speakers present higher average values regarding daily use of Spanish (HL: 5.8 compared to L3: 3.5). Both groups show similar results regarding the current, ideal and ought-to Spanish selves. Upon assessing the narratives, some interesting results appear. 13 participants (n=7 L3, n=6 HL) express selves of different kinds. Two L3 speakers and one HL speaker express more than one self (Tuva, an L3 speaker of Spanish, expresses four different kinds). Regarding the Spanish ideal self, eight participants (n= 5 L3, n=3 HL) see themselves using Spanish in the future. They differ in the degree of detail with which they describe their visualisations and desires for their future selves. Examples of these visualisations can be observed in the descriptions of Thea and Hilda 9. Thea, a Spanish L3 speaker, says, for instance: Ex. 1. In the future, I see myself travelling a lot to Spain, using Spanish to talk with my au pair friends in Germany and using Spanish in my profession as a high school teacher of Swedish and Spanish. (Thea) Like many other participants, Hilda writes: Ex. 2. Right now, I’m dreaming of going away to Spain or South America and I’m longing for the time when I can travel and thus practice my spoken Spanish with different people in different cultures. (Hilda) 9 The quotes were originally written in either Swedish or Spanish. The translations are our own. 240 HumaNetten Nr 48 Våren 2022 These excerpts share a high degree of visualisation of the future self in relation to contact with other Spanish speakers. The possibility of travelling to a Spanish-speaking country and being immersed in the Spanish-speaking world seems highly attractive for almost all participants, regardless of whether they are L3 or HL speakers. Notably, those who rely on a more pragmatic relation with the Spanish language, most of whom are L3 Spanish learners, generally express a strong desire to improve their language skills, but lack opportunities to integrate the Spanish language in their lives, like Tanja, a Spanish L3 speaker, who writes: Ex. 3. I would like to use Spanish more in my everyday life but I have few people whom I can speak Spanish with, which makes me afraid of losing my skills. (Tanja) Unsurprisingly, the pragmatic reasons to study Spanish, which reflect the ought-to self, revolve around the importance of improving language skills to ensure future professional competence since participants are studying to become language teachers. However, in several cases, the ought-to self also coincides with a strong desire to become a competent Spanish speaker in the future for reasons that are not purely professional. In other words, pragmatic reasons for improving one’s language skills do not seem to exclude non- pragmatic reasons which are more or less clearly related to a visualisation of an ideal L2-self. As illustrated in the following example, three HL speakers explain that they study Spanish to help their children. This motivation is interpreted as an ought-to self: Ex. 4. I want to improve my Spanish for personal reasons. I have a daughter who is interested in learning Spanish and I want to give her as much support as possible. (Hugo) Tuva expresses one Spanish ideal self and one ought-to self in the past. The same participant also explains the benefit of knowing other languages—not only Spanish, but languages in general. They agreed with Tara and Tilde. Ex. 5. Learning languages is for me the best gift you can give yourself, because it just keeps giving. (Tuva) Regarding the current self, some students describe it in terms of identity and appreciation. The following examples illustrate the current self: Ex. 6. I never abandoned my mother tongue, which is part of who I am and also my identity. (Hila) Österberg, Donoso & Sologuren: Back to the Initial Condition 241 Ex. 7. A large part of my personality is linked to Latin America and Spanish. (Tilde) Tuva describes her current self with the observation that she is afraid that her proficiency will fossilise. Ex. 8. I love the Spanish language and I constantly learn new things, unfortunately I notice that my language has partly stagnated over the years. (Tuva) In sum, various possible selves emerge in the students’ narratives. The ideal L2 self is tightly connected to the idea of establishing contacts with other Spanish speakers. Visualisations of travelling and living abroad are vivid among L3 Spanish speakers. Their Spanish ought-to selves coincide with the ambition to become a proficient speaker, not only for professional reasons. For the HL speakers who have children, the wish to be able to help their sons and daughters is interpreted as an ought-to-self. 4.2 The Spanish language learning experience Concerning initial conditions in the narratives, most learners who claim significative contact with the language in the past (in form of positive feedback from relatives and teachers or challenging experiences that pushed them to improve) describe experiencing a ‘trigger’ to develop a strong desire to learn Spanish or improve existing language skills. While we do not claim a causative linear effect, positive past experiences and positive outcomes in the learning itinerary seem somehow related. A clear example of language contact functioning as a trigger can be seen in Example 8, from Thea, a Spanish L3 speaker: Ex. 9. After graduating, I moved to a Spanish city to work (...) as an au pair in a Spanish family (...). During that time, I really put my Spanish to the test. I must have probably learned up to 20 words a day (...). I got a lot of au pair friends from Germany whom I decided to communicate exclusively in Spanish with, and I also took a level B2 curso extensivo for au pairs at a school there. When I left, my self-confidence in oral communication was incredible. (Thea) Five participants point out the role of a teacher in learning Spanish as a foreign language, an experience that has been categorised as an initial condition. These five participants are all L3 speakers of Spanish. They all use Spanish to learn more about the language. Tuva has had a negative initial condition when learning Spanish in the school context, which she uses later on in life in order to create positive learning conditions for acquiring Spanish within the teacher education program. In fact, Tuva describes a series of initial conditions that made her realise the discrepancies between her present language skills and what she wanted to express, leading her to feel frustrated and take action. Her end state is not only a Spanish 242 HumaNetten Nr 48 Våren 2022 ideal self but a more extensive way of talking about the advantages of knowing languages. Tilde, Tina and Tuva refer to initial conditions produced in Spanish- speaking countries, both positive (Tilde and Tuva) and negative. Tina describes the frustration of not being able to communicate clearly in a restaurant in Spain, an experience given the role of a trigger in her narrative. We interpret this as Tina becoming aware of the discrepancy between who she was and who she wanted to be, as a speaker of Spanish. In this sense, her negative initial condition works as a trigger. Tina’s case is also a good example of how the sensitivity to initial condition works. Henry, Hugo and Holly describe having children as a trigger for studying and improving their Spanish. In the case of Henry, he refers to the discrepancy between his current and Spanish ideal self when his child was born (“when I became a parent”) and he realized that his Spanish proficiency was insufficient to pass the language on. In fact, this moment of realization can be interpreted as a negative initial condition, instantiated in the past, finally triggering a change that he has extended to the present. In the case of Hugo and Holly, on the other hand, their desire to improve their Spanish to help their children can be interpreted both as an initial condition and as part of a Spanish ought-to self (see Example 4). The difference between these three HL speakers may relate to the time point from which they consider their need to improve. Ex. 10. But Spanish hasn't always been that exciting and the choice to become a Spanish teacher is fairly recent. In the years after I moved away from home, I noticed how my Spanish was slowly fading away. It was only when I became a parent that I realised how much I had lost. So, I chose to become a Spanish teacher, partly to improve my native language, but also to be able to teach my children better Spanish. (Henry) Regarding the category of language environment, Hilma, an HL speaker, illustrates how language environment relates to the Spanish language experience (Example 11). In her case, she refers to the fact that her parents lacked an academic background. As a child, she did not understand how her linguistic competences in Spanish, Swedish and English could interact. She was frustrated by her lack of language proficiency, which we interpret as a lack of metalinguistic knowledge. Overall, she perceives the language environment in which she grew up as negative. It seems that she overcame this difficulty as an adult when she decided to study Spanish. At the same time, her evaluations of the past show how her desire to improve has been sensitive to what we interpret as her initial conditions. A negative impact from the environment can be inferred from the participants’ narratives when they express feelings of isolation in school or society because of the languages they speak. This is in line with previous observations made by Ball and Ellis (2008) when they explain how a negative status of a language Österberg, Donoso & Sologuren: Back to the Initial Condition 243 in society can affect a multilingual person’s self-perception in a negative way. This is for instance observed when Henry, a HL speaker, says that “Spanish hasn’t always been that exciting..” (see example 10) or when Hilma describes the difficulties she experienced related to Spanish. Thus, the language environment seems to be part of the L2 learning experience, as a condition to which speakers relate their learning trajectory. Ex. 11. But it is clear to me today that I also had difficulties because neither of my Spanish-speaking parents had an academic education. They didn't understand how that could help me with the languages or my studies. Then, when English entered the picture, it didn't get any easier or harder, but I didn't have much time to practise the languages. And all languages meant different things: Spanish was my secret language that nobody understood. (Hilma) Regarding language use, this category emerges in some of the narratives. Six participants explain that they used and use Spanish in the family context, except for Hugo. Hugo has no Spanish-speaking friends in his personal environment at present. Getting to use the language seems related to the participants’ appreciation of their learning process and seems to nurture the selves. In sum, the L2 learning experience appears to be closely linked to the three categories analysed: the initial conditions, the learning environment and language use. The results indicate that the L2 language learning experience can be understood as connected to language use and experiences in the immediate context outside the classroom. 4.3 Participants’ feelings and thoughts about writing vis-a-vis the Spanish learning experience All participants said that, for them, develop writing in Spanish was important for professional and personal reasons. Ex 12. I listen a lot to music with Spanish lyrics and read some fiction in Spanish, preferably Mexican authors. I follow societal issues connected mainly to Mexico but also other Latin American countries and am interested in history, culture and language. I have some friends that I speak Spanish with, but almost only via digital channels at the moment. I have a constant longing for Latin America and wish I had more friends in Sweden with whom I could speak Spanish [...] Language is something that fascinates me and something I am passionate about, and that is why I have now chosen to train as a teacher. I want to convey that passion to others. (Tilde) Two participants deviate from the main pattern, both HL learners. Hugo, who does not use Spanish in his daily adult life, enrolled in university Spanish courses to develop his writing, a competence he did not develop in the family context. 244 HumaNetten Nr 48 Våren 2022 Ex 13. I decided to study Spanish in XX at the University of XX to learn how to write, something I did not learn at home. (Hugo) The birth of his child and his desire to pass on the language and culture seems to have guided his decisions, and can be interpreted as initial conditions for his action. Hugo was very clear about this when answering the Likert-scale questions. His main reason for studying Spanish writing is personal (7) rather than professional (5 on the Likert scale). Hulda shows the opposite pattern. She expresses some degree of anxiety speaking and writing in Spanish. Spanish has been used in the family context, and in school she passed all the courses without having to study grammar. Her motivation to write in Spanish is mainly professional rather than personal (expressed in Q41 in the questionnaire). Ex. 14. My first language, which I learned through one of my parents, though I didn't grow up in that country but in Sweden, so I learned Spanish and Swedish at the same time. I have used Spanish to communicate with my family. Then I had mother tongue instruction for a short period up until I turned 7. Then I had it as a C language in school. I passed all the courses in high school without practising grammar. Then, over the years, I have only visited my home country once a year and called/texted my family. (Hulda) Hilda describes a growing interest for writing in Spanish that began while writing her bachelor thesis. In this case, the thesis may be a possible initial condition. Ex. 15. Therefore, I cannot say that I developed my Spanish very much during that period either. However, I feel that I developed a confidence in being able to manage without problems in a Spanish-speaking community. While writing my bachelor's thesis in Spanish, I learned an incredible amount and I enjoyed it. (Hilda) Hillary shows a similar pattern. She experienced a negative initial condition when she realized the gap between her oral and written Spanish language skills, and between her written skills in English and Swedish. This perceived discrepancy serves as a negative initial condition, which motivates her interest in learning. Ex. 16. I moved to Sweden when I was three years old. At home, we always spoke Spanish. At school and with friends I spoke Swedish. Until I graduated, I studied and wrote in Swedish or English. This meant that my level of writing in Spanish was lower than in English, but I was fluent in Spanish. After graduation I travelled abroad and spent one year in England and one year in France. After that, I settled in Spain. I worked and lived in Spain for eleven years. After eleven years there, my written Spanish was much better because I needed it. (Hillary) Österberg, Donoso & Sologuren: Back to the Initial Condition 245 In sum, the reasons for a desire to improve writing skills are heterogeneous. In some cases, writing itself functions as a trigger. The challenge to write complex texts, such as a thesis, seems to constitute a favourable initial condition for some students. 5. Discussion In this paper, three research questions have been posed. In terms of the L2MSS, how do Spanish L3 learners and Spanish Heritage language learners describe their Spanish ideal and ought-to selves and their Spanish language learning experiences in relation to using and learning or maintaining Spanish in their autobiographical narratives? How do the students’ feelings towards and thoughts about their writing relate to their Spanish learning experience and their motivational selves? With regards to the first question, the results reveal that some participants, independently of the group they belong to, have pragmatic reasons for studying Spanish (ought-to self) which partly overlap with their personal desire to improve their Spanish language skills (Spanish ideal self). The distinction between these two motivations seems closely linked to participants’ opportunities for natural language experiences, that is conversing with other Spanish speakers and in contexts outside the formal educational system, which is the norm rather than an exception for the HL speakers, who use Spanish outside school. When analysing the HL and the L3 participants’ desire to communicate with other Spanish speakers, it seems a pragmatically and instrumentally motivated goal as well as an integral part of the Spanish ideal self. Dörnyei (2005), distinguishes between two aspects of the ought-to self, a promotional (constructive) side, very similar to the ideal L2 self, and a preventive (evasive) side of instrumentality, which may lead to a less-motivated behaviour. Further, the ought-to selves deployed in the narratives, that is the pragmatic reasons for enrolling in a university course in Spanish, imply use of Spanish to learn and develop oral and written skills. In their narratives, participants describe a triggering effect on motivation which they perceive to result from the intense learning activity. In other words, the L2 learning experience itself seems to trigger motivation, which has also been observed by Gentil (2011). In parallel, the participants assign high value to the Spanish ideal self. A highly valued identity as a Spanish speaker is not only important for the HL but also for the L3 speakers. These results are in line with Noels (2005) and Comanaru and Noels (2009). The question arises whether these selves are constructed in the same way and whether childhood experiences of several languages change how self-perception develops. The L3 learners express their selves as a ‘becoming’ while the HL learners describe themselves as ‘being’. The ideal L2 and ought-to selves can be accounted for in the past (see Example 9) and are often referred to in explaining the repeated pattern of 246 HumaNetten Nr 48 Våren 2022 participants’ discrepancy-reducing trajectories (Thorsen et al. 2017). This is particularly clear in some narratives that express several selves, with no notable differences observed between groups. The participants often describe their Spanish-language competence in the present and as a part of a current self. The same holds for the perceived benefits of their language competences. In this sense, descriptions of the present and the future selves overlap, possibly because participants are in a dynamic period of learning. However, the benefits are not solely related to Spanish, but, for some learners, also linked to competences in other languages. This is clearly the case with the L3 speakers as well as some HL speakers. A question for future research concerns the nature of the selves observed in the present study. What is the significance of the current self for motivation? Are other ideal selves at play, in addition to those linked to a separate L2? What are the characteristic traits of the selves that a HL speaker expresses? These questions remain to be answered and are beyond the scope of this study. With regards to the second research question on the L2 learning experience, initial conditions are present in the narratives and are expressed with negative, neutral or positive values. Interestingly, initial conditions perceived as negative are described by some participants as leading to motivated behaviours. The initial condition seems to have the capacity to raise awareness of shortcomings and discrepancies. Positive initial conditions are described by the participants in terms of conditions which gave them guidance and orientation. For the L3 speakers, the teacher often appears as a positive initial condition, having confirmed skills and successful learning. Interestingly, this initial condition is not found among the HL speakers; rather, the HL speakers tend to refer to family as a positive initial condition and to pragmatic reasons for studying Spanish. Hilma alone describes a lack of support in her immediate context, in her case from her parents. The HL speakers in our study do not mention any confirmation by teachers to encourage successful language learning. A similar observation is made by Leeman and Serafini (2020), who point out that HL speakers in the US usually are expected to fully master their HL and are considered to take the easy way out if they choose to study Spanish in mixed groups. This may deflate their self-perceptions, with a negative effect on their desire to improve. Nevertheless, if given the opportunity to work with meaningful and challenging tasks, these students may develop optimally. Regarding our third research question, the topic of writing is present in the questionnaire, but the students do not always develop this theme in their narratives. Writing is an important part of university studies in Spanish and in teacher education. Since writing is not learnt in the family context, but developed in instructional settings, the confirmation and encouragement given in formal instruction need to be studied more thoroughly, especially Österberg, Donoso & Sologuren: Back to the Initial Condition 247 when considering that the HL and L3 students in this study describe their school experiences in different ways. Interestingly, some participants describe the development of writing as rewarding. For one of the HL speakers, Hugo, it is the reason for engaging in Spanish studies. Thus, when presented with stimulating writing tasks, some HL speakers in our study seem to develop most. One pedagogical implication is that teachers must create challenging tasks to generate positive feelings of success, not only for L3 speakers but also for HL speakers. As our results indicate, L3 and HL learners’ motivation to develop written competencies in Spanish can be investigated in terms of the L2MSS, and examples of Spanish ideal and ought-to selves are found in both groups. However, the operationalisation of the selves and their terminology needs to be elaborated when researching motivation, especially if a comparison between HL and L3 speakers is to be made. In our study, although results for the HL group showed more internal variance compared to the L3 speakers, the close-ended questions in the questionnaire showed tendencies to ceiling effects. A closer analysis of the Spanish ideal self also indicates some differences in how the self is constructed. Roughly speaking, the HL learners describe their Spanish ideal selves as what they are while the L3 learners describe it as something they have become. Further, when students' entire learning trajectories are considered, the Spanish current and ideal selves seem to merge at a certain point. In other words, for our students (multilinguals enrolled in advanced language courses), the gap between the current self and the ideal self is small. When describing who they are at present, most descriptions are very robust and include clearly depicted Spanish ideal selves. Regarding the Spanish learning experience, an extended definition of the L2 learning experience that considers the students’ initial conditions (which took part in the past and present, within and outside of school) is necessary to account for differences among the selves between the groups. Experiences of encouragement and recognition seem linked to the construal of selves and need more research. Our results point in the same direction as those of Ball and Ellis (2008) and Thompson (2020), who found that the negative status of a HL might be related to a negative self-construction. Finally, a methodological challenge has been observed, since mostly high motivated students have answered our call. An inevitable positive bias has been detected in the data, an issue not easily solved. Future research may consider broadening the recruitment of participants if the intention is to investigate students from lower academic levels. Further on, reliability has been achieved since the coding and analysis of the narratives was built on the combined measures of multiple-choice and open-ended questions. The narratives turned out to be a fruitful way of studying motivation and writing in high achieving learners, since data present a range of different topics and content which would be difficult to assess using questionnaires. 248 HumaNetten Nr 48 Våren 2022 6. Concluding remarks This study aimed to investigate students’ motivation to develop and maintain an advanced proficiency level of Spanish, with a special focus on their Spanish language learning experiences. Their motivation for writing was also studied. The thematic analysis of the autobiographical narratives revealed Spanish ideal, ought-to and current selves. Most participants described how, at a certain point in their learning trajectories, initial conditions seemed to trigger their awareness of a discrepancy between who they were and who they wished to be. As expected, Spanish learning experiences were depicted as tightly connected to formal and informal contexts in which students learnt and used Spanish. Thus, the idea that the L2 learning experience is related to language use as such also finds support. The participants describe how they use the language, for example, in a family context or when studying. These notions of the initial condition, learning environment and language use contributed to the understanding of the Spanish learning experience. A temporal perspective was clearly depicted, and discrepancy-reducing trajectories were confirmed. Contact with the linguistic community, either through family or when travelling, stimulates the future self-guides in a positive way and needs further investigation. For the L3 speakers, a language teacher often contributed to their efforts to improve their Spanish, not only at a triggering state but also over time. As regards the possible link between motivation and writing, the material indicates that writing is perceived as a challenge, and that it is a competence acquired in school. All students valued highly their intentions to develop writing skills. Some participants described how writing challenges encountered in the educational setting, such as writing academic texts, contributed to their current selves in positive ways. The results indicate that the L2 learning experience can be studied in terms of the notions of initial conditions, learning environment and language use. The findings imply that it is possible to understand the L2 learning experience as including the immediate context outside the classroom, and not only school experiences. In fact, this seems to be the case with the HL speakers’ Spanish learning experience especially. Therefore, the characteristic traits of the L2 learning experience need to be further investigated in future studies. Not only the L2 learning experience, but also the notion of sensitivity to initial conditions, including how these trigger and nurture the ideal L2 self and the ought-to self, need more attention. The discrepancies that participants perceive between the initial conditions and the current self seem to make them aware of their need for learning. How these discrepancies are perceived and operationalised in the L3 and HL learners’ self-perceptions is another project for future studies. Österberg, Donoso & Sologuren: Back to the Initial Condition 249 References Al-Hoorie, Ali (2018), “The L2 motivational self-system: a meta-analysis”, Studies in Second Language Learning and Teaching 8: 721–754. Ball, Arnetha & Ellis, Hillary (2008), “Identity and the Writing of Culturally and Linguistically Diverse Students”, in Charles Bazerman (ed), Handbook of Research on Writing. History, Society, School, Individual, Text. New York & London: Lawrence Erlbaum Associates, p. 499–513. Bazerman, Charles (2016), “What do sociocultural studies of writing tell us about learning to write?”, in Charles A. MacArthur, Steve Graham, Jill Fitzgerald (eds), Handbook of writing research (2nd ed.,). New York: Guilford, p. 11–23. Boo, Zann, Dörnyei, Zoltan & Ryan, Stephen (2015), “L2 motivation research 2005–2014: Understanding a publication surge and a changing landscape”, System 55: 147–157. https://doi.org/10.1016/j.system.2015.10.006 Braun, Virginia & Clarke, Victoria (2006), “Using Thematic Analysis in Psychology”, Qualitative Research in Psychology 3(2): 77–101. https:/doi.org/10.1191/1478088706qp063oa Comanaru, Ruxandra & Noels, Kimberly (2009), “Self-Determination, Motivation, and the Learning of Chinese as a Heritage Language”, Canadian Modern Language Review-revue 66(1): 131–158. https://doi.org/10.3138/cmlr.66.1.131 Csizér, Kata (2019), “The L2 Motivational Self System”, in Martin Lamb, Kata Csizér, Alastair Henry & Stephen Ryan (eds), Palgrave Macmillan handbook of motivation for language learning. Basingstoke: Palgrave, p. 71–88. De Bot, Kees (2008), “Introduction: second language development as a dynamic process”, Modern Language Journal 92: 166–178. https://doi.org/10.1111/j.1540-4781.2008.00712.x Donoso, Alejandra, Österberg, Rakel & Sologuren, Enrique (2021), “Escritura académica en bilingües de sueco y español como lengua de herencia”. Ávila Reyes, Natalia (ed), Multilingual contributions to writing research: Towards an equal academic exchange. The WAC Clearinghouse and ALES. https://doi.org/10.37514/INT- B.2021.1404.2.06 Douglas Fir Group (2016), “A transdisciplinary framework for SLA in a multilingual world”, Modern Language Journal 100: 19–47. Dörnyei, Zoltan (2005), The Psychology of the Language Learner: Individual Differences in Second Language Acquisition. Mahwah, NJ: Lawrence Erlbaum. Dörnyei, Zoltan (2009), “The L2 motivational self-system”, in Zoltan Dörnyei & Ema Ushioda (eds), Motivation, Language Identity and the L2 Self. Clevedon: Multilingual Matters, p. 9–42. Dörnyei, Zoltan (2019a), “From Integrative Motivation to Directed Motivational Currents: The Evolution of the Understanding of L2 Motivation over Three Decades”, in Martin Lamb, Kata Csizér, Alastair Henry & Stephen 250 HumaNetten Nr 48 Våren 2022 Ryan (eds), Palgrave Macmillan handbook of motivation for language learning. Basingstoke: Palgrave, p. 39–62. Dörnyei, Zoltan (2019b), “Towards a better understanding of the L2 Learning Experience, the Cinderella of the L2 Motivational Self System”, Studies in Second Language Learning and Teaching 9(1): 19–30. https://doi.org/10.14746/ssllt.2019.9.1.2 Dörnyei, Zoltan & Ushioda, Ema (eds) (2009), Motivation, language identity and the L2 self. Bristol: Multilingual Matters. Dörnyei, Zoltán, MacIntyre, Peter D. & Henry, Alastair (eds) (2014), Motivational dynamics in language learning. Bristol: Multilingual Matters. Dörnyei, Zoltan & Ryan, Stephen (2015), The psychology of the language learner revisited. New York: Routledge. Gardner, Robert C. (1985), Social psychology and second language learning: The role of attitudes and motivation. London: Edward Arnold Publishers. Geisherik, Anna (2004), "The Role of Motivation among Heritage and Non- Heritage Learners of Russian", Canadian Slavonic Papers/Revue Canadienne des Slavistes 46(1–2): 9–22. https://doi.org/10.1080/00085006.2004.11092343 Gentil, Guillaume (2011), “A biliteracy agenda for genre research”, Journal of Second Language Writing 20(1): 6-23. https://doi.org/10.1016/j.jslw.2010.12.006. Ghaith, Ghazi (2019), “The interplay of selected demotivation determinants and achievement in EFL critical reading and writing”, Tesl-Ej 22(4): 1–16. https://eric.ed.gov/?id=EJ1204580 Grosjean, François (2010), Bilingual life and reality. Cambridge, Massachusetts & London. Haagensen, Bodil (2007), “Dialogpedagogik och motivation. En fallstudie i svenska på universitetsnivå”, Acta Wasaensia, Nr 174, Språkvetenskap 32. Vasa: Vasa universitet. Hanson Seibert, Aroline (2019), “The Motivation of Heritage Learners vs. Foreign Language Learners in a University-level Spanish Composition Course”, in Nur Yiitolu & Melinda Reichelt (eds), Writing Beyond English. Blue Ridge Summit: Multilingual Matters, p. 197–215. https://doi.org/10.21832/9781788923132 Henry, Alastair (2012), L3 motivation. PhD thesis, Göteborgs universitet. Henry, Alastair (2017), “L2 motivation and multilingual identities”, The Modern Language Journal 101(3): 548–565. https://doi.org/10.1111/modl.12412 Henry, Alastair, & Thorsen, Cecilia (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 Higgins, E. Tory (1987), “Self-discrepancy: A theory relating self and affect”, Psychological Review 94: 319–340. https://psycnet.apa.org/buy/1987- 34444-001 Österberg, Donoso & Sologuren: Back to the Initial Condition 251 Holm, Lars & Laursen, Helle (2011), “Migrants and Literacy Crises”, Apples – Journal of Applied Language Studies 5(2): 3–16. http://urn.fi/URN:NBN:fi:jyu-2011100711507 Hornberger, Nancy (1989), “Continua of biliteracy”, Review of Educational Research, 59(3): 271–296. https://doi.org/10.3102/00346543059003271 Hornberger, Nancy (2002), “Biliteracy and Schooling for Multilingual Populations”, International Journal of the Sociology of Language, 2002(155–156): 137–142. https://doi.org/10.1515/ijsl.2002.020 Hornberger, Nancy (2004), “The Continua of Biliteracy and the Bilingual Educator: Educational Linguistics in Practice”, International Journal of Bilingual Education and Bilingualism 7(2–3): 155–171. https://doi.org/10.1080/13670050408667806 Hyltenstam, Kenneth & Österberg, Rakel (2010), “Foreign Language Provision at Secondary Level in Sweden”, Sociolinguistica 24(1): 85–100. Jang, Younie & Lee, Junkyu (2018), “The effects of ideal and ought-to L2 selves on Korean EFL learners’ writing strategy use and writing quality”, Read Writ 32: 1129–1148. https://doi.org/10.1007/s11145-018-9903-0 Kim, Tae-Young, Kim, Young-Mi & Kim, Ji-Young (2017), “Role of Resilience in (De)Motivation and Second Language Proficiency: Cases of Korean Elementary School Students”, Journal of Psycholinguistic Research 48: 371–389. https://doi.org/10.1007/s10936-018-9609-0 Kormos, Judit (2012), “The role of individual differences in L2 writing”, Journal of Second Language Writing 21(4): 390–403. https://doi.org/10.1016/j.jslw.2012.09.003 Källkvist, Marie & Hult, Francis (2016), “Discursive mechanisms and human agency in language policy formation: negotiating bilingualism and parallel language use at a Swedish university”, International Journal of Bilingual Education and Bilingualism 19(1): 1–17. https://doi.org/10.1080/13670050.2014.956044 Lamb, Martin (2012), “A self-system perspective on young adolescents’ motivation to learn English in urban and rural settings”, Language Learning 62(4): 997–1023. https://doi.org/10.1111/j.1467- 9922.2012.00719.x Larsen-Freeman, Diane (1997), "Chaos/Complexity Science and Second Language Acquisition”, Applied Linguistics, 18(2): 141–165. https://doi.org/10.1093/applin/18.2.141 Larsen-Freeman, Diane (2014), "Ten ‘Lessons’ from Complex Dynamic Systems Theory: What is on Offer", in Zoltan Dörnyei, Peter D. MacIntyre & Alastair Henry (eds), Motivational Dynamics in Language Learning. Bristol, Blue Ridge Summit: Multilingual Matters, p. 11–19. https://doi.org/10.21832/9781783092574-004 Leeman, Jennifer & Serafini, Tove J. (2020), “’It’s Not Fair’: Discourses of Deficit, Equity, and Effort in Mixed Heritage and Second Language Spanish Classes”, Journal of Language, Identity & Education 20(6): 425–439. https://doi.org/10.1080/15348458.2020.1777866 Ling, Guanming, Elliot, Norbert, Burstein, Jill C., McCaffrey, Henry F., MacArthur, Charles A. & Holtzman, Steven (2021), “Writing 252 HumaNetten Nr 48 Våren 2022 motivation: A validation study of self-judgment and performance”, Assessing Writing 48. https://doi.org/10.1016/j.asw.2020.100509 Lorenz, Edward (1963), ”Deterministic non-periodic flow”, Journal of Atmospheric Sciences, 20(2), 130–141. https://doi.org/10.1175/1520- 0469(1963)020<0130:DNF>2.0.CO;2 MacIntyre, Peter D., Baker, Susan & Sparling, Heather (2017), “Heritage passions, heritage convictions and the rooted L2 self: Music and Gaelic language learning in cape Breton, Nova Scotia”, Modern Language Journal 101(3): 501–516. https://doi.org/10.1111/modl.12417 Markus, Hazel R. & Nurius, Paula (1986), “Possible selves”, American Psychologist 41: 954– 969. https://doi.org/10.1037/0003-066X.41.9.954 Mikulski, Ariana (2006), "Accentuating Rules and Relationships: Motivations, Attitudes, and Goals Spanish for Native Speakers Class", Foreign Language Annals 39(4): 660–82. https://doi.org/10.1111/j.1944- 9720.2006.tb02282.x Miller, Ryan T., Mitchell, Thomas D. & Pessoa, Silva (2014), “Valued voices: Students’ use of Engagement in argumentative history writing”, Linguistics and Education 28: 107–120. https://doi.org/10.1016/j.linged.2014.10.002. Montrul, Silvina (2010), “Current issues in heritage language acquisition”, Annual Review of Applied Linguistics 30: 3–23. https://doi.org/10.1017/S0267190510000103 Noels, Kimberly A. (2005), "Orientations to Learning German: Heritage Language Learning and Motivational Substrates", The Canadian Modern Language Review 62(2): 285–312. https://doi.org/10.3138/cmlr.62.2.285 Nordenstam, Kerstin (2003), Genusperspektiv på språk. Stockholm: Högskoleverket i samarbete med Nationella sekretariatet för genusforskning. Olsen, Max (2018), “What and Who in the L2 Motivational Self System: The Interlinked Roles of Target Language and Heritage Learner Status”, Canadian Modern Language Review 74(2): 279–301. https://doi.org/10.3138/cmlr.2016-0002 Osterberg, Rakel (2008), Motivación, aptitud y desarrollo estructural : Un estudio sobre la actuación lingüística en aprendientes suecos de español L2 (PhD dissertation, Institutionen för spanska, portugisiska och latinamerikastudier). http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva- 8120 Oyserman, Daphna & James, Leah (2009), “Possible selves: From content to process”, in Keith D. Markman, William M.P. Klein & Julie A. Suhr (eds), Handbook of Imagination and Mental Simulation. New York: Psychology Press, p. 373–394. Serafini, Ellen (2013), Cognitive and psychosocial factors in long-term development of implicit and explicit second language knowledge in adult learners of Spanish at increasing proficiency. PhD thesis, Georgetown University. Serafini, Ellen (2020), “Further situating learner possible selves in context: A proposal to replicate Henry and Cliffordson (2013) and Lasagabaster Österberg, Donoso & Sologuren: Back to the Initial Condition 253 (2016)”, Language Teaching 53(2): 215–226. https://doi.org/10.1017/S0261444819000144 Silva-Corvalán, Carmen (2014), Bilingual Language Acquisition. Cambridge University Press. Strömberg Jämsvi, Susanne (2019), Unpacking dominant discourses in higher education language policy. A critical study of language policy in Swedish higher education. PhD thesis, University of Gothenburg. Thompson, Amy (2020), “My many selves are still me: Motivation and multilingualism”, Studies in Second Language Learning and Teaching 10(1): 159–176. https://doi.org/10.14746/ssllt.2020.10.1.8 Thorsen, Cecilia, Henry, Alastair & Cliffordson, Christina (2017), “The case of a missing person? The current L2 self and the L2 Motivational Self System”, International Journal of Bilingual Education and Bilingualism 23(5), 584–600. https://doi.org/10.1080/13670050.2017.1388356 Torres, Julio, Estremera, Ricardo & Sherez, Mohamed (2019), “The contribution of psychosocial and biographical variables to heritage language learners’ linguistic knowledge of Spanish”, Studies in Second Language Acquisition 41(4): 695–719. https://doi.org/10.1017/S0272263119000184 Ushioda, Ema (2009), “A person-in-context relational view of emergent motivation, self and identity”, in Zoltan Dörnyei & Ema Ushioda (eds), Motivation, language identity and the L2 self. Bristol: Multilingual Matters, p. 215–228. Ushioda, Ema & Dörnyei, Zoltan (2012), “Motivation”, in Susan. M. Gass & Alison Mackey (eds), The Routledge handbook on second language acquisition. New York, NY: Routledge, p. 396–409. Vetenskapsrådet (2017), Good research practice. Stockholm: Swedish Research Council. https://www.vr.se/english/analysis/reports/our-reports/2017-08- 31-good-research-practice.html. Wilby, James (2020), Motivation, self-regulation, and writing achievement on a university foundation programme: A programme evaluation study, Language Teaching Research: 1–24. https://doi.org/10.1177/1362168820917323 Yanguas, Inigo (2010), “A Quantitative Approach to Investigating Spanish HL Speakers’ Characteristics and Motivation: A Preliminary Study”, Hispania 93(4): 650–670. You, Chenjing Julia & Dörnyei, Zoltán (2016), “Language Learning Motivation in China: Results of a Large-Scale Stratified Survey”, Applied Linguistics 37: 495–519. https://doi.org/10.1093/applin/amu046 Yung, Kevin Wai-Ho (2018), “Exploring the L2 selves of senior secondary students in English private tutoring in Hong Kong”, System 80: 120–133. https://doi.org/10.1016/j.system.2018.11.003. 254 HumaNetten Nr 48 Våren 2022 Appendix I Questions on background data, self-judgement of proficiency and intended effort. Question number Wording of the item Response types Q5 My strongest language Alternative is: Q12 Estimate your Likert scale 1-5 10 proficiency in Spanish on the scale below. Q20 I am good at writing in Likert scale 1-7 Spanish. Q21 I use Spanish in contact Likert scale 1-7 with people outside Sweden. Q22 I use Spanish in my daily Likert scale 1-7 life in many different situations. Q25 Being able to speak Likert scale 1-7 Spanish is a part of who I am. Q26 I am good at writing in Likert scale 1-7 Swedish. Q27 Being able to speak Likert scale 1-7 Swedish is a part of who 10 This Likert scale 1-5 has written levels. Level 5 has the following wording: “I have a very high degree of proficiency. I speak, read and write fluently in Spanish. I have a high degree of comprehension”. Level 1 reads: “I speak and write the language but need to develop more fluency, proficiency and complexity in spoken and written production. I often have difficulties in understanding what people say”. Österberg, Donoso & Sologuren: Back to the Initial Condition 255 I am. Q28 I am good at writing in Likert scale 1-7 English. Q29 Being able to speak Likert scale 1-7 English is a part of who I am. Q33 I want to develop my Likert scale 1-7 academic writing in Spanish. because it is a requirement in my career. Q41 Writing in Spanish is Likert scale 1-7 important for me personally, not only for my career. This is the reason to why I want to develop my writing in Spanish more. Q47 Please formulate your Open-ended own answer below: How much time do (or did, if you recently studied Spanish) you approximately spend on your studies in Spanish? Please express the time spent in hours and days per week. 256 HumaNetten Nr 48 Våren 2022 Q1 How did you learn Open-ended Spanish? How long have you studied it and at what age did you start? Describe previous studies and also any extensive periods in Spanish-speaking countries) Q2 Please describe your Open-ended highest Spanish studies at university level. Q6 Please describe briefly Open-ended your personal reasons for studying Spanish. Q54 Please choose a language Narrative task (Swedish, Spanish or English) and recount how you have learnt and use the different languages in your life. Try to be as detailed and exhaustive as possible but exclude personal data such as names of schools, towns or cities where you have lived, or friends names. This is thought to be a text, so feel free to start, leave and go back to this text before submitting your answer. Many thanks in advance. Österberg, Donoso & Sologuren: Back to the Initial Condition 257 Appendix II Participants Part. Age of Age of Span. Strongest University course Chosen onset of onset HL/ language in Spanish at the language of Spanish of L3 time of the the narrative Swedis questionnaire (number h of words) Hann From birth From HL Swedish I Swedish (408) a pre- school age Hila From birth 13 HL Swedish I Spanish (392) Hele From birth 7 HL Swedish I Spanish (316) na (L2) and Spanish Huld From birth From HL Swedish III Swedish (161) a birth Holly From birth From HL Swedish III Spanish (227) pre- school age Hugo From birth From HL Swedish IV Spanish (83) birth Hora From birth From HL Swedish III Swedish (115) cio ? birth Henr From birth From HL Swedish III Swedish (301) y pre- school age Hilda From birth From HL Swedish III Swedish (339) birth Hillar From birth From HL Swedish III Swedish (139) y birth and Spanish Hilm From birth From HL Swedish III Swedish (225) a birth Tina 13 From L3 Swedish I Spanish (283) birth Tove 13 From L3 Swedish I Swedish (198) birth Tuva 13 From L3 Swedish III Swedish birth (1026) 258 HumaNetten Nr 48 Våren 2022 Tilde 12 From L3 Swedish III Swedish (594) birth Tyra 12 From L3 Swedish III Swedish (368) birth Thyr 12-13 From L3 Swedish II Swedish a birth (1227) Tindr 15 From L3 Swedish III Spanish (549) a birth Tore 13 From L3 Swedish III Swedish birth and Spanish (123) Tanja 12-13 From L3 Swedish III Swedish (329) birth Tara 12 From L3 Swedish III Spanish (733) birth Appendix III Self-estimated proficiency measures and intended effort in writing Informant Self- Self- Intended effort estimate estimated Q33 Writing Spanish is important for my career d proficiency: Q41 Writing in Spanish is important for me proficie Writing personally ncy (Q20) Q47: Open-ended Q (Q12) I am good at writing in X Hanna 4 Spanish: 6 Q33: 7 English: 7 Q41: 7 Swedish: 7 Q47: 15-20 hours per week Hila 5 Spanish: 7 Q33: 7 English: 3 Q41: 7 Swedish: 7 Q47: 50h per week Helena 5 Spanish: 7 Q33: 7 English: 5 Q41: 7 Swedish: 7 Q47: 20h per week Hulda 3 Spanish: 3 Q33: 7 English: 5 Q41: 5 Swedish:7 Q47: 30h per week Österberg, Donoso & Sologuren: Back to the Initial Condition 259 Holly 4 Spanish: 6 Q33: 7 English: 4 Q41: 7 Swedish: 6 Q47: 25h per week Hugo 4 Spanish: 3 Q33: 4 English: 4 Q41: 7 Swedish: 6 Q47: 6 hours per week Horacio 5 Spanish: 7 Q33: 7 English: 5 Q41: 7 Swedish: 7 Q47: 20h per week Henry 4 Spanish: 6 Q33: 6 English: 6 Q41: 7 Swedish: 7 Q47: 25h per week Hilma 5 Spanish: 7 Q33: 7 English: 7 Q41: Swedish: 7 Q47: High Hillary 5 Spanish: 7 Q33: 7 English: 7 Q41: Swedish: 7 Q47: Hilda 5 Spanish: 6 Q33: 6 English: 6 Q41: Swedish: 6 Q47: Average Tina 3 Spanish: 6 Q33: 4 English: 6 Q41: 7 Swedish: 7 Q47: Between 21 and 63 hours a week Tove 3 Spanish: 6 Q33: 7 English: 6 Q41: 7 Swedish: 7 Q47: 45–50h per week Tuva 2 Spanish:3 Q33: 7 (neutral) Q41: 7 English: 5 Q47: 40h/week Swedish: 5 Tilde 4 Spanish: 7 Q33: 7 English: 7 Q41: 5 260 HumaNetten Nr 48 Våren 2022 Swedish: 7 Q47: 17h per week Tyra 2 Spanish: 4 Q33: 6 English: 7 Q41: 7 Swedish: 7 Q47: 10–15h per week Thea 4 Spanish: 7 Q33: 7 English: 6 Q41: Swedish: 7 Q47: Above average Tindra 3 Spanish: 5 Q33: 6 English: 6 Q41: 6 Swedish: 6 Q47: Above average Tore 4 Spanish: 6 Q33: 7 English: 6 Q41: Swedish: 7 Q47: Tanja 4 Spanish: 6 Q33: 7 English: 6 Q41: Swedish: 7 Q47: High Tara 4 Spanish: 6 Q33:6 English: 5 Q41: Swedish: 7 Q47: Above average Appendix IV ID Q12 Q20 Q21 Q22 Q25 Q26 Q27 Q28 Q29 Q33 Q41 11 HL 4 6 6 5 6 7 7 6 2 6 6 HL 5 7 7 5 5 7 6 3 3 7 7 HL 4 3 1 1 6 6 7 4 1 4 1 HL 4 6 6 4 5 7 7 5 1 7 4 11 Unfortunately, some answers to Q41 are lacking since the first round of the sampling did not include this question. Österberg, Donoso & Sologuren: Back to the Initial Condition 261 HL 4 6 7 7 5 7 7 6 5 5 7 HL 3 6 7 7 7 7 5 4 3 HL 7 4 7 7 7 5 7 5 HL 5 7 7 7 7 5 4 4 4 7 HL 3 5 7 6 6 6 7 6 6 6 HL 4 6 7 5 7 7 7 6 6 7 L3 3 4 5 3 7 5 7 5 7 7 6 L3 2 4 2 2 7 7 7 7 7 6 7 L3 3 7 7 6 5 7 7 6 2 4 7 L3 4 7 5 1 7 7 7 7 6 7 7 L3 3 6 4 1 7 7 7 6 4 7 7 L3 5 6 6 5 7 6 7 6 5 6 L3 4 6 7 4 7 7 7 6 6 7 L3 5 7 7 7 7 7 7 7 7 6 L3 5 7 1 1 7 7 7 7 7 4 L3 4 6 2 2 6 7 7 6 3 3 L3 4 6 7 7 6 7 7 5 6 6 Average: 3,57 5,33 5,28 4,24 6,05 6,67 6,81 5,81 4,67 5,86 5,58