This paper discusses recent trends in the persuasion register in English. Such trends are linked to cultural differences and technological affordances, which also have an impact on social and professional practices (Bai et al. 2023; Matz et al. 2024; Kapantai et al. 2021). To this purpose, I first discuss the key concepts related to persuasive writing and their link to appraisal in English. Then, I focus on a sample of the most persuasive AI-generated writing extracted from a dataset released by Anthropic (Durmus et al. 2024) and available online. The methodology that drives this study is the appraisal system framework (ASF) by Martin and White (2005), the findings in ASF-based research on promotional texts (Pounds 2022; Ho 2021; 2011), institutional texts (Tupala 2019), academic texts (Hood 2010; 2006), and the research on complementary discourse systems for expressing stance and evaluation (Biber et al. 2019; 2018). The results show that the three groups of highly persuasive texts extracted from the Anthropic’s dataset present non-dissimilar word frequency ranges and a tendency towards the use of low-frequency lexis and appraisal resources. The application of the ASF demonstrates that the AI system Claude 3 Opus masters evaluation scales in English despite the well-documented inconsistencies of such systems in dealing with the numerical scales. The sample analyzed does not really follow the pattern of sparingly using appraisal resources while preferring invoked appraisal, which has been demonstrated to be a recent trend in the persuasion register. Nonetheless, it mostly maintains the proven culture-based recent trend regarding the scale of directness/indirectness (ASF Engagement). Finally, the ASF categories of Attitude, Graduation, and Engagement in the sample do not deviate from the attested function of propagating the meta-text/meta-discourse although with a caveat: they systematically construct semantic prosodies of extremes to persuade compellingly.

Emerging Trends in the Register of Persuasion Considering Appraisal in English

Anna Zanfei
2024-01-01

Abstract

This paper discusses recent trends in the persuasion register in English. Such trends are linked to cultural differences and technological affordances, which also have an impact on social and professional practices (Bai et al. 2023; Matz et al. 2024; Kapantai et al. 2021). To this purpose, I first discuss the key concepts related to persuasive writing and their link to appraisal in English. Then, I focus on a sample of the most persuasive AI-generated writing extracted from a dataset released by Anthropic (Durmus et al. 2024) and available online. The methodology that drives this study is the appraisal system framework (ASF) by Martin and White (2005), the findings in ASF-based research on promotional texts (Pounds 2022; Ho 2021; 2011), institutional texts (Tupala 2019), academic texts (Hood 2010; 2006), and the research on complementary discourse systems for expressing stance and evaluation (Biber et al. 2019; 2018). The results show that the three groups of highly persuasive texts extracted from the Anthropic’s dataset present non-dissimilar word frequency ranges and a tendency towards the use of low-frequency lexis and appraisal resources. The application of the ASF demonstrates that the AI system Claude 3 Opus masters evaluation scales in English despite the well-documented inconsistencies of such systems in dealing with the numerical scales. The sample analyzed does not really follow the pattern of sparingly using appraisal resources while preferring invoked appraisal, which has been demonstrated to be a recent trend in the persuasion register. Nonetheless, it mostly maintains the proven culture-based recent trend regarding the scale of directness/indirectness (ASF Engagement). Finally, the ASF categories of Attitude, Graduation, and Engagement in the sample do not deviate from the attested function of propagating the meta-text/meta-discourse although with a caveat: they systematically construct semantic prosodies of extremes to persuade compellingly.
2024
English linguistics, Persuasive register, Appraisal in English, LLM-generated texts
File in questo prodotto:
File Dimensione Formato  
Anna Zanfei_Iperstoria24_2024.pdf

accesso aperto

Tipologia: Versione dell'editore
Licenza: Dominio pubblico
Dimensione 321.13 kB
Formato Adobe PDF
321.13 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11562/1147974
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact