This study investigates the use of the first-person personal pronoun ‘I’ in political discourse and, more specifically, the functions of Represented ‘I’ and Situated ‘I’, including subcategories, present in Joe Biden’s speeches. A corpus was compiled, consisting of 12 speeches that brought up the topic of the war in Ukraine. The aim of the analysis was to examine the frequency of the pronouns in the speeches and their functions. The study is based on categories from Roitman’s (2014) and Albalat-Mascarell and Carrió-Pastor’s (2019) research on presidential debates. The results reveal that Represented ‘I’ is the most frequent use of the pronoun, which presents speakers as politicians in the real world rather than positions them in the discursive situation, as in the case of Situated ‘I’. Since Situated ‘I’ was categorized only with regard to the discursive verbs of the current discourse, it might have contributed to the big difference in frequencies between Represented ‘I’ and Situated ‘I’.

This thesis comprises four separate studies of the same material: a ten-minute Swedish Sign Language monologue. Study I describes the form, meaning, and use of the sign INDEX-c, a pointing toward the chest traditionally described as a first person pronoun. It is argued that INDEX-c is used not only with specific reference to the signer or a quoted signer, but also with non-specific reference. Contrary to what has been reported, INDEX-c is used not only for constructed dialogue, but also in constructed action. The analysis reveals two separate forms, as well, labeled as reduced INDEX-c and distinct INDEX-c, respectively. Study II describes the activities of the non-dominant hand when it is not part of a two-handed sign. A continuum is suggested, moving from different rest positions that do not contribute to the discourse content, via mirroring of the dominant hand, for example, to instances where the non-dominant hand produces signs of its own while the dominant hand remains inactive, i.e. dominance reversal. Several of the activities of the non-dominant hand, including the four types of buoys that are described, help structure the discourse by indicating the current topic. Study III uses Mental Space Theory and Conceptual Blending Theory to describe the use of signing space for reference. A correlation is shown between discourse content and the area in the signing space toward which signs are meaningfully directed, and also between these directions and which types of Real Space blends the signer mainly uses: token blends or surrogate blends. Finally Study IV looks in more detail at three segments of the discourse and their Real Space blend structure. An initial analysis of eight interpretations into spoken Swedish is also conducted, focusing on whether preselected content units (discourse entities and relations) are identified. A large number of Real Space blends and blended entities are argued to result in less successful renditions measured in terms of preselected content units. / För att köpa boken skicka en beställning till exp@ling.su.se/ To order the book send an e-mail to exp@ling.su.se

For 30 years researchers have studied Expressive Writing (EW) - writing emotionally about a trauma 3-5 times for 15-30 minutes, while controls (CW) objectively write about, most often, time management. Previous research has documented associations, in the trauma condition, between flexible use of pronouns and physical health, or between trauma writing and improvement of working memory (WM). The main aim of the current study was to see if there is a relation between flexible use of first-person pronoun singular (“I”) in EW and improved results on WM test, as well as on quality of life and self-reported illness. In an Internet study participants were recruited through convenience and snowball sampling and randomly divided on conditions. The final sample consisted of 10 participants each in EW and CW: 13 women, 6 men and 1 non-binary, 21-69 years, most of them well educated. Outcome measures were WM test (Automated version of Operation Span Task; AOSPAN), self-reported quality of life (Brunnsviken Brief Quality of Life; BBQ) and self-reported illness (days of symptoms and restricted activities). JASP was used for the statistics. A measure of flexible use of the Swedish pronoun “jag” (“I”) was tested and found to associate with significant improvements in AOSPAN results only; unexpectedly within the whole sample. On condition one significant result was found: EW improved their quality of life. The increase in AOSPAN results is discussed, as well as writing instructions and baseline assessments. The small sample with self-reported health measures and online WM test limited the study. Replications of the results are needed. / I 30 år har forskare studerat Expressivt skrivande (EW) - att skriva känslomässigt om ett trauma 3-5 gånger à 15-30 minuter. Kontrollbetingelsen (CW) har ofta bestått av att objektivt beskriva sin tidsanvändning. Tidigare forskning har i traumabetingelsen påvisat samband mellan en flexibel användning av pronomen och fysisk hälsa, eller mellan traumaskrivande och förbättrade resultat på test av arbetsminne (WM). Huvudsyftet med den föreliggande studien var att se om det finns ett samband mellan en flexibel användning i EW av pronomenet första person singular (”jag”) och förbättrade resultat på arbetsminnestest, samt på livskvalitet och självrapporterad ohälsa. Tio deltagare vardera deltog i experiment- respektive kontrollgruppen: 13 kvinnor, 6 män och 1 icke-binär, 21-69 år; de flesta välutbildade. Deltagarna rekryterades genom bekvämlighets- och snöbollsurval och fördelades slumpvis till betingelserna. En internetstudie genomfördes, med utfallsmåtten arbetsminnestest (Automated version of Operation Span Task; AOSPAN), självskattad livskvalitet (Brunnsviken Brief Quality of Life; BBQ), samt självskattad ohälsa (dagar med symptom respektive med begränsade aktiviteter). JASP användes för de statistiska analyserna. Ett mått för flexibel användning av det svenska pronomenet ”jag” testades och visade sig vara associerat med signifikanta förbättringar enbart av AOSPAN-resultatet, men förvånande nog inom hela urvalet. I övrigt hittades endast en skillnad mellan betingelserna, nämligen att experimentgruppen signifikant förbättrade sin självskattade livskvalitet. Ökningen av AOSPAN-resultaten diskuteras, liksom skrivinstruktioner och baslinjeskattningar. Studien begränsades av ett litet urval, självskattade hälsomått och nätbaserat arbetsminnestest. Replikationer av resultaten är nödvändiga.