Discursive construction of identity in narratives of migrant experience

U ovoj doktorskoj disertaciji opisana je diskursna konstrukcija identiteta u narativima o osobnom iskustvu migracije. Istraživanje se temelji na kvalitativnoj analizi intervjua prikupljenih metodom polustrukturiranog sociolingvističkog intervjua s 10 ispitanica koje su tijekom 2010-ih iselile iz Hrvatske. Teorijsko-metodološki okvir istraživanja temelji se na narativnoj analizi, proizašloj iz interakcijskih sociolingvističkih pristupa i (kritičke) analize diskursa te na poststrukturalističkom poimanju identiteta kao dinamičnog procesa koji na različitim razinama jezične analize i složenosti izranja tijekom jezične interakcije. Transkribirana građa analizirana je na dvjema razinama: najprije su u s obzirom na formalnofunkcionalne karakteristike izdvojeni, opisani i kategorizirani narativi kao analitičke jedinice, a potom su u njima identificirana i opisana jezična sredstva kojima se na diskursnoj razini ostvaruju strategije pozicioniranja. Pritom se diskursne strategije pozicioniranja promatraju s obzirom na raslojenost narativnog konteksta na svijet narativa, koji obuhvaća sudionike, mjesto, vrijeme i ostale okolnosti događaja o kojem je u narativu riječ; svijet naracije, koji se odnosi na interakcijske okolnosti narativne produkcije; te svijet migracije, odnosno širi društveni kontekst u koji je uronjeno tematizirano iskustvo i njegova naracija. Formalnofunkcionalna analiza pokazala je da se narativ kao diskursni žanr ne može pozitivistički i konačno definirati, ali da se u korpusu mogu prepoznati tri tipa narativa: referencijski, evaluacijski i hipotetički. Analiza diskursnih strategija pozicioniranja na trima razinama narativnog konteksta pokazala je da je diskursna konstrukcija identiteta dinamičan i kontekstualno uvjetovan proces te da se upotrebom jezičnih sredstava poput zamjeničkog prebacivanja i konstruiranog dijaloga ostvaruju i međusobno izmjenjuju različite identitetske pozicije u istom narativu. Na razini svijeta naracije dominiraju diskursne strategije generalizacije, kolektivizacije, individualizaije i (de)agentivizacije, dok se na razini naracije najistaknutijom pokazala strategija uključivanja sugovornice s ciljem dobivanja empatične povratne informacije, odnosno distanciranja od eksplicitne evaluacije. Konačno, analiza pozicioniranja u svijetu migracije izlučila je najučestalije teme prisutne u korpusu (makroteme), čijom rekontekstualizacijom ispitanice ilustriraju širi društveni kontekst migracije i konstruiraju svoj identitet kao migrantice.Introduction This doctoral dissertation presents a qualitative study of linguistic devices employed in discursive construction of identity in narratives of personal experiences told by ten young, highly educated females who emigrated from Croatia in the 2010s. The study is in line with existing copious research in discourse analysis and sociolinguistics that focuses on narrative discourse and narrative interaction as quintessential acts of identity construction and expression. The aim of the research is twofold. Firstly, it aims to describe the complexity of the genre of narrative of personal experience from a linguistic point of view. Secondly, it seeks to understand how various aspects of individuals’ identities are contested, constructed, and performed in narrative discourse throughout the experience of migration, more specifically, within the context of the intensified emigration from Croatia in the 2010s. Methodology In order to investigate how migrants construct identities in narratives of personal experiences, semi-structured sociolinguistic interviews were conducted with young, highly educated females who have emigrated from Croatia during the 2010s. Since this implied finding geographically distant participants who are willing to openly talk about their experiences, a practical solution was to contact friends and acquaintances of the researcher that fit the criteria. Due to geographical distance between the researcher and research participants, eight interviews were conducted and recorded via Skype, while two interviews were conducted in person. Research participants’ names, as well as names of other people, specific locations, organizations, and other personal information were changed in order to protect individuals’ privacy. The semi-structured sociolinguistic interview was chosen since it is a convenient method for collecting data for qualitative analysis because it does not strictly rely on a rigid question-answer format, but rather consists of open questions that can be omitted, or new questions can be asked, depending on how the interview develops. Adopting this interview format meant creating a more spontaneous interactional situation where research participants were able to reproduce narratives about their migration experiences, organize events in a temporally and spatially cohesive manner, and evaluate event participants and circumstances by attributing them with certain characteristics. Moreover, as De Fina (2003) and Relaño Pastor (2014) demonstrate, sharing narratives about important personal experiences empowers individual migrant voices, which are generally marginalized and reduced to the private sphere, and this was an especially important aspect of this research. Thus, the format of a semistructured interview helped to reduce the effects of asymmetrical power relations that are present in more traditional interview contexts (Kress & Fowler 1979, Bourdieu 1999), as well as to shape a solidary and empowering atmosphere where participants’ accounts were given relevance and legitimacy (Mishler 1986, Relaño Pastor 2014). Finally, following the interactive nature of narrative discourse emphasized by poststructuralist approaches, collected interviews were considered as interactive discursive practices in which both the interviewer (i.e. theresearcher) and the interviewee jointly co-construct narrative discourse, which is why the presence and the role of the researcher was not omitted, but was included in the transcripts and the analysis. This was also a good place to reflect on the delicate role of the researcher in the process of data collection, i.e. its co-construction and consequent interpretation. Analysis Once the interviews were conducted, the material was transcribed relying on well-established transcription conventions in narrative analysis (e.g. De Fina & Georgakopoulou 2015). The first step of the analysis was to divide the entirety of the transcribed material into narratives, i.e. to define the main unit of analysis as closely as possible. Labov’s canonical approach to narrative analysis relied on syntactic features, i.e. fixed structure and order of clauses, which implies that a lot of spoken discourse appearing in a research interview would not satisfy the formal criteria of narrative, and would therefore have to be thrown out of the analysis. Moreover, due to the nature of spoken language, isolating narratives from the surrounding discourse in a clear-cut manner is hardly possible since they are often embedded, overlapped, unfinished, or lack a formally concise structure. Thus, in order to consider the interactive nature and multi-layered context of the discursive genre, the structuralist approach and canonical narrative structure described by Labov served only as a basis onto which more functional criteria were added, such as interactional nature of spoken discourse, thematic homogeneity and contextual features (cf. Ochs and Capps 2001). The second level of analysis involved looking into linguistic devices used to position the speaker and other social actors on different levels of narrative context, which results in discursive strategies of positioning (Bamberg 1997). The notion of discursive strategy adopted in this research implies “a more or less intentional plan or practice adopted to achieve a particular social, political, psychological, or linguistic goal”, which can be “located at different levels of linguistic organization and complexity” (Wodak and Meyer 2016: 33). More specifically, discursive strategies of positioning are results that consistent usage of certain linguistic devices across different levels of narrative context has in terms of how positions of social actors are represented in narrative discourse. Given that narratives elicited from these research interviews are seen as a discursive genre produced in a specific interactive context between two participants who know each other and share the experience of migration, but also have certain expectations from this particular interaction, the following three levels of narrative context were taken into account: - the story world, which includes (the order of) recounted events, participants (characters) and the setting; - the storytelling world, which refers to the interactional aspects of narration and the fact that narratives are jointly co-created actively by both interlocutors; the researcher and the narrator; - the world of migration, which refers to the broader social context, discursive practices, ideologies, power relations, and other aspects of the migration trend in question. The first two levels of narrative context are mutually intertwined and further embedded in the third. Analysis of positioning was conducted for each level of narrative context and presented in separate chapters. However, it was often impossible to separate them, given the deep interconnectedness of the layers of narrative context. Results As a result of the first level of analysis, the corpus was divided into 493 narratives, organized around three central narrative types, which were called referential, evaluative, and hypothetical. Narratives that recount a specific event from the past, involving a mainly coherent temporal sequence of events, i.e. at least one temporal juncture that for most of the time follows the sequence of events in real life, were called referential narratives. Prototypical referential narratives are highly tellable stories that function as individual episodes that can be easily detached from the surrounding discourse without disrupting the rest of the text, thus resembling the canonical Labovian model the most. They usually recount a dynamic event and involve agentive participants. Referential narratives made up 38.3% of the corpus. Narratives that do not recount a specific event from the past, but rather describe, evaluate or comment on certain situations, states, facts, problems or actions taken by social actors were categorized as evaluative narratives. Because of their emphasized evaluative function, these narratives often follow referential narratives as a way to reflect on, explain, or justify actions taken by certain social actors, but they also appear as individual narrative sequences. These types of narratives appeared most frequently and comprised 55.4% of the corpus. The third and the least frequent type of narratives in the corpus were hypothetical narratives. They talk about plans and potential events that have not yet taken place, but might in the future. These narratives constituted 6.3% of the corpus. It must be noted that this division is not rigorous, meaning that some narratives are more prototypical, some have only a few distinctive features of the group, whereas some have features of more than one group. Finally, narratives of personal experience of migration elicited in spoken discourse were defined as more or less completed thematic episodes of spoken discourse in which experiences, states, situations, or plans for the future that play an important role in the overall biographies of migrants are described, evaluated and reflected on. The central intention of narratives (to recount past experiences, evaluate situations, or speculate about the future) is closely related to the positions that narrators create for themselves and other social actors in the process of narration, and can be expressed by using different linguistic sources, depending on the interactive context, surrounding discourse, and the dynamics of the narrative progression. On the second level of analysis, a close reading of the narratives in the corpus revealed pronominal shift and constructed dialogue as the most salient linguistic devices, used consistently by all research participants, resulting in discursive strategies of generalization, collectivization, (de)agentivization, and involvement. Pronominal shift was found to be the most transparent linguistic device that narrators use dynamically in the course of temporal and thematic progression of the same narrative in order to construct various and often opposing positions for themselves and other social actors. The most common pronominal alternation in the corpus was the switch from individual I- to generic you-perspective. For example, by using generic you to talk about negative and unpleasant experiences in the story world, narrators positioned themselves as distant from event participation, but also as evaluators of events and problematic situations they encountered as immigrants who had to replace their familiar social, cultural, and linguistic environments with new, unfamiliar contexts. In the storytelling world, the shift from I to you had a more referential usage, resulting in the strategy of interlocutor involvement, by which narrators positioned themselves as close to the researcher, seeking approval and understanding of events in the story world. Furthermore, by using pronominal shift and other linguistic devices (such as specific discursive markers that indicate involvement) narrators emphasised or mitigated their evaluations by including the researcher/interviewer, which resulted in jointly constructed moral stances. Constructed dialogue (Tannen 2007) was found to be a powerful linguistic device that results in constructing agentive or agentless social actors in the story world. Narrators often used constructed dialogue to recontextualize someone else’s words in order to emphasize or mitigate their social responsibility and explicit evaluation of actions or stances by certain social actors. For example, instead of openly stating their moral stance, narrators often channeled them through dialogues in which they and other actors participate as characters in the story world. Simply put, by constructing characters that speak, narrators were able to express their own opinions about them and their actions without being explicitly critical or responsible. A consistent feature of the corpus is that the actors that are constructed as agentive, i.e. characters who were given voice in the story world, are usually in some positions of power, such as administrative workers, bosses, managers, or “authentic” locals, i.e. non-migrant that have symbolic power and/or access to language capital. On the other hand, in the majority of dialogues that were recontextualized in these narratives, the narrators either speak very little or do not speak at all. By constructing social agents in position of power (be it only symbolic), and constructing themselves as agentless, narrators illustrated the way they see their position in a social context of their migration destination. The way narrators linguistically constructed social actors in the story world can be directly related to the third level of narrative context, i.e. the world of migration, since it illustrates the power relations and discursive practices in which experiences of immigrants are embedded. The analysis of positioning in the world of migration illustrated that by recontextualizing master narratives (Bamberg & Andrews 2004) about migration and by drawing on certain macro-topics (Wagner & Wodak 2006) respondents revealed broader issues of the migration experience that affect their identity performance. Some of these master narratives and macro-topics included: identifying oneself as a migrant/foreigner, anticipated unemployment as main motivation for migration, language identity and migrants’ access to language capital, migration policies of certain countries, and ideologies and power relations they are based on. By looking into how respondents position themselves with regard to those macro-topics and master narratives, the analysis illustrated the most salient discursive practices and power relations that result in one’s identity as a migrant. Conclusion It must be mentioned here that this research and the methodology it adopts did not aim to reach any essentialist or positivist claims about the linguistic phenomena observed, but rather to illustrate certain tendencies that are present in spoken narrative discourse, which is in line with poststructuralist approaches to describing language in its relation to identity and society. By drawing empirically on a relatively large corpus of spoken narrative discourse, this study demonstrated the dynamic process of identity construction embedded in the experience of migration. From a linguistic point of view, it illustrated the interconnectedness of language, migration, and identity, and the necessity of enriching traditional approaches to linguistic analysis with new, discourse- and identity-oriented criteria. By relying on notions of multilayered narrative context and discursive strategies of positioning, this study offers an applicable model for doing socially engaged narrative and discourse analysis. From the point of view of migration, the study revealed various social aspects of the migration experience, which are often overlooked by formal and quantitative approaches. More specifically, it illustrated what motivates young, highly educated females to emigrate from Croatia, what some of the struggles they encounter on the way are, and which aspects of identity are important for them in their new social, cultural, and linguistic environments. For research participants, sharing narratives about important personal experiences was an opportunity to empower their individual voices and give their experiences legitimacy. Finally, findings emerging from this study can have a significant impact in facing the challenges of the migration trend in question and shaping further migration policies and discourses about it. Further research in the field can benefit from the proposed methodology, which can be applied in analyses of other types of narrative and non-narrative, spoken or written discourse, in order to investigate which features of narratives of personal experience of migration recognized in this study can be found in other discursive genres