Metaphors are central to health communication, especially for sensitive topics like cancer and depression. As AI tools like ChatGPT become increasingly prevalent in health conversations, their capacity to process and respond to metaphors remains underexplored. This study investigates whether ChatGPT processes metaphors pragmatically and what this reveals about the pragmatics of AI-generated language. Utilizing metaphors commonly found in cancer and depression discourse, we interacted with ChatGPT-4o mini using a 2 × 2 design, comparing responses across illness type (cancer versus depression) and input metaphoricity (metaphorical versus literal). We evaluated how ChatGPT organizes its responses though pragmatic strategies, such as situation re-construction, alternative reasoning, and agreement with the user, as well as its handling of metaphors via reuse, replacement, or creation. Poisson and Negative Binomial regression analyses were conducted to examine how illness type and input metaphoricity influenced ChatGPT’s responses. Results show that responses to metaphorical prompts contained more pragmatic elements and metaphors than those to literal prompts. Situation re-construction and alternative reasoning were the most frequent pragmatic strategies, regardless of illness type or input metaphoricity. Depression-related literal prompts elicited longer responses and more metaphors than cancer-related literal prompts. A chi-square test further revealed a significant association between metaphor use and pragmatic elements, indicating that metaphor handling is systematically linked to the pragmatic organization of ChatGPT’s responses. Overall, both response structure and metaphor use are largely templated, lacking the flexibility and nuance characteristic of human interaction. These findings highlight both the potential and limits of ChatGPT’s pragmatic competence in health communication.
This study examines agreement as relational work in Korean dyadic discussion. Adopting a relational-work perspective, it treats (im)politeness as sequentially accomplished through stance practices rather than as a property of honorific forms. The corpus comprises ten Korean dyads (20 speakers; 7567 turns) discussing two public topics. Stance tokens were quantitatively coded across age configuration, gender composition and acquaintance, then qualitatively analysed for turn design and sequential function. Agreement was far more frequent than disagreement (1162 vs. 263 turns), with systematic distributional differences: near-age dyads showed the densest agreement, gender configuration patterned stance distributions, and stranger dyads produced more overt agreement than acquainted dyads. Disagreement was comparatively infrequent and typically mitigated, delayed or concessively formatted rather than produced as bare opposition. Qualitative analysis shows that agreement is not a single action type but recurrently serves (i) epistemic self-positioning, where speakers align while claiming or ceding epistemic rights, and (ii) supportive reinforcement, where speakers collaboratively thicken a shared stance through rephrasings, examples, affective displays and echoic designs. The findings support analysing Korean (im)politeness as situated relational work accomplished through sequentially organised stance practices, complementing form-centred accounts of honorific politeness with strategy-level interactional evidence.
Discourse-pragmatic (D-P) like is widespread in contemporary spoken English, yet existing accounts remain difficult to compare because classifications vary in terminology, level of detail, and diagnostic criteria. This study proposes an operationalised taxonomy for identifying and coding D-P like in corpus data. The taxonomy is grounded in three primary diagnostics: omissibility, syntactic position, and syntactic boundedness, which motivate the highest level of functional distinctions, with contextual analysis and diagnostic paraphrases allowing finer-grained classification. It distinguishes three hierarchical levels, comprising four main functions (discourse marker, clause-medial pragmatic marker, clause-final pragmatic marker, and quotative marker) and their respective subfunctions. The taxonomy was applied to two corpora of informal spoken British English, the demographically sampled component of the British National Corpus 1994 and the Spoken BNC2014, to investigate the marked increase in the relative frequency of like observed between them (from 4371 to 13,782 items per million words). Using inverse (fixed-success) sampling, 500 tokens of D-P like were extracted per corpus and manually coded. The analysis examines discourse-pragmatic like at two points in time using both quantitative and qualitative approaches. The results indicate that the sharp rise in the relative frequency of like is largely driven by its D-P uses, especially tokens classified as QM like. Despite this increase, the functional repertoire remains stable across the two corpora.
This study examines elicited self-praise in Saudi televised business interviews, focusing on how Saudi business leaders present professional success while orienting to the social delicacy of praising oneself. Drawing on an interactional approach to politeness, the study combines quantitative coding of self-praise strategies with conversation-analytic examination of selected sequences from Hewar Mufeed, a Saudi television interview program. The dataset comprises 22 episodes featuring Saudi male business leaders, from which 235 self-praise instances comprising 689 coded strategy tokens were identified. The quantitative analysis shows that modified explicit self-praise was the most frequent strategy, followed by explicit self-praise without modification, while implicit self-praise was comparatively limited. Pronoun analysis further shows a close balance between singular and plural reference, indicating a negotiation between individual achievement and collective affiliation. The qualitative analysis demonstrates that self-praise was not produced as an isolated act, but was sequentially constructed through host questions, formulations, supportive uptake, follow-up questions, and embodied cues. The findings show that the televised business interview provides an institutional context in which achievement talk is expected, but still culturally managed through religious attribution, collectivist framing, self-denigration, gratitude, and references to relational and institutional support. The study contributes to self-praise research by showing how strategy use, recipient ratification, institutional genre, and Arab cultural-religious norms jointly shape the pragmatic acceptability of public self-praise.
This study introduces intercultural harmony as a conceptual lens for understanding intergenerational interaction in multilingual families. Rather than treating harmony as the avoidance of conflict, we conceptualize intercultural harmony as an interactional principle and achievement through which linguistic, cultural, and epistemic differences are collaboratively managed in intergenerational family talk. Drawing on an interactional sociolinguistic analysis of mother–daughter conversations in Chinese multilingual families, this study examines discussions of socio-cultural issues to explore how participants negotiate disagreement. The analysis identifies different interactional processes through which disagreement is negotiated via the interplay of alignment and affiliation at both sequential and affective levels. These processes enable participants to sustain engagement and relational cohesion despite ongoing linguistic, cultural, and epistemic differences. While alignment and affiliation function as interactional resources, intercultural harmony operates at a broader level as an emergent process of sustaining relational cohesion across differences. By bringing pragmatic understandings of cooperation and harmony into dialogue with interculturality, the study offers a new lens to examine multilingual family interactions, in which intercultural harmony functions both as an interactional principle guiding participants’ engagement and as an interactional achievement emerging through the collaborative negotiation of difference.
During conversations, both the main speaker and their addressee work together to co-construct the interaction. However, addressee reactions, referred to as feedback, are often variable, making them challenging to model and predict. To better understand this variability, we introduce the concept of addressee style, defined as the distinctive multimodal behaviors exhibited by addressees during conversations. Our study focuses on (i) variability in addressee styles across conversational activities and over time, (ii) convergence of styles within dyads, and (iii) correlations with storytelling quality and engagement. Using the SMYLE corpus, which features French face-to-face storytelling and free conversation, we examined generic feedback (e.g., ”mh”, ”yeah”, nods) and specific feedback (evaluative responses like laughter, speech e.g., ”it's amazing,” facial expressions, etc.). Feedback components that include type of verbal responses, head movements, facial expressions, hand gestures, pitch, and loudness were used to examine each participant's average feedback behavior. Hierarchical clustering revealed three addressee styles: Low-Profile (mainly generic and gestural feedback), Balanced (rich, multimodal, and lengthy feedback, with a balance of generic and specific responses), and Intense (mainly multimodal specific feedback). Analysis of intra-individual variability showed that addressee styles vary across conversational activities but remain relatively stable within free conversation. Additionally, results suggest that participants tend to adapt their addressee styles to align with their conversational partners. We found no significant correlation between addressee style and storytelling quality, suggesting that effective interaction can emerge from diverse listening approaches rather than a single ”ideal” style. However, results suggested a potential link between addressee style and engagement.
Metaphor comprehension is widely assumed to be context-dependent, yet it remains unclear how different types of contextual cues constrain interpretation. The present study examined the effects of target-domain context, source-domain context, open-ended prompts, and unrelated context on the comprehension of nominal metaphors, using a Context–Prime–Probe paradigm combined with behavioral measures and event-related potentials (ERPs). Our results showed that target-domain contexts reliably facilitated metaphor-related meanings, as reflected in faster response times, reduced negativity in a temporally restricted 400–502 ms cluster relative to open-ended prompts. They also modulated late positive activity relative to source-domain contexts. Source-domain contexts did not produce comparable behavioral facilitation but modulated ERP activity during both an early interval (∼214–262 ms) and a later interval (∼638–762 ms), suggesting that source-related information may remain temporarily accessible and place additional demands on later processing. The open-ended prompts provided insufficient semantic guidance to facilitate metaphor comprehension. Moreover, contextual facilitation did not consistently extend to metaphor-irrelevant source features. These findings indicate that contextual effects on nominal metaphor comprehension are selective, dimension-specific and temporally differentiated. They are broadly compatible with complementary contributions from category-based semantic constraint and relevance-guided pragmatic interpretation.
This study explored Saudi football fans' name-calling practices when expressing their affective stances in the YouTube comment section. It analysed how name-calling was employed to index fans' team (dis)affiliation, position towards the stance object and their (dis)alignment with other interactants in these comments. It further examined the linguistic creativity in fans' name-calling practices, in particular pattern re-forming forms. To do this task, a small corpus of 2646 YouTube comments was compiled, taken from three clips of the popular Saudi TV show Action with Waleed. The collected data were analysed to identify different forms of creativity in name-calling, such as the creative metaphoric invention/tˤaћlu:bi:/(seaweed). Additionally, following a discursive approach, excerpts of fans' interactions were qualitatively analysed to investigate name-calling as a situated act and how (dis)affiliation, positioning, and (dis)alignment emerge dynamically through the thread of replies. The findings highlight that creative name-calling in Saudi football fans' interactions appears to serve different functions, such as asserting superiority, creating distance through ‘us vs them’ discourse, and providing a source of pleasure. One of the key concerns in sport discourse, particularly in Saudi football, is sport fanaticism, which can be reflected in fans' name-calling practices, thus highlighting the fine line between humour, teasing and provocation in competitive discourse.
This article examines the discourse-pragmatic uses of interrogative ¿hola? in contemporary written European Spanish. While hola is conventionally described as a greeting, its interrogative use has received little attention and remains largely undocumented. Drawing on 364 tokens from large web-based corpora, the study shows that ¿hola? is predominantly used as an addressee-oriented, reactive device that targets prior discourse and encodes evaluation.The analysis identifies a clear functional pattern: rather than initiating interaction or expressing surprise alone, ¿hola? typically presents a preceding contribution as unexpected, inappropriate, or norm-violating, thereby projecting a stance-bearing response. These uses are strongly associated with turn-medial position, where the form is typically used in pivot-like configurations linking prior discourse to a subsequent evaluative or argumentative move.The findings show that the interpretation of ¿hola? depends crucially on its sequential placement and its relation to prior discourse. In this respect, it works as a resource for managing interaction and negotiating shared expectations, instead of as a carrier of propositional meaning.The study argues that these uses are best understood within the domain of interactive grammar, as a redeployment of a greeting formula into an interrogative, reactive device for managing interaction and expressing stance. More broadly, the results illustrate how a conventional greeting form can function as a compact resource for encoding counter-expectational evaluation in context.
This study investigates the impression management strategies employed by Chinese and French when reflecting on interview questions about personal weaknesses, drawing on responses collected from online Q&A platforms. The results show no significant difference between Chinese and French respondents in terms of "presenting strengths as weaknesses" and "recognizing real weaknesses". However, Chinese respondents report a higher likelihood of employing tactful language, while French respondents report a higher tendency of giving evasive answers, especially satire and refusals. Both groups exhibit a transactional motivation to achieve effectiveness and an interactional motivation to enhance self-image, with Chinese responses reporting relatively greater importance on protecting the interviewer's positive image. This study posits that cultural traits, interview context, and question framing jointly shape respondents' attitudes and strategies. (c) 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
This article concerns predications made without a lexical verb in everyday conversations in Finnish. Two kinds of predications are compared: those done with clauses using the copula olla 'be' and free nominals, NPs and APs which have no syntactic role in any clause. The comparison shows that the two expressions were both used in categorizing, characterizing and identifying referents, but at different frequencies. In particular, copula clauses were much more likely to be used for characterizing than free nominals were. Free nominals had more diverse functions than copula clauses did, as they were used for certain functions, such as vocatives, which copula clauses were not used for. Furthermore, copula clauses occurred in both initial and subsequent positions in sequences of action, while free nominals strongly preferred subsequent positions. This indicates that copula clauses and free nominals are positionally sensitive linguistic formats. The methodology used in the study is Interactional Linguistics which combines usage-based and typological linguistics with Conversation Analysis. A basic tenet of Interactional Linguistics is that linguistic patterns are shaped by interaction and shape interaction on the other hand. Grammatical structures are seen as resources for performing particular actions, selected on the basis of their appropriateness in particular contexts. (c) 2026 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
This present study examines the degree of alignment between ChatGPT's ratings and qualitative evaluations, and those provided by native-speaker university lecturers on L2 student emails. Using a perception questionnaire and thematic analysis, the study found that ChatGPT's evaluations of L2 student request emails were closely aligned with those of NS university lecturers, both quantitatively and qualitatively, suggesting a substantial degree of pragmatic competence in the evaluation and revision of academic emails. The model was able to identify key pragmatic infelicities, provide feedback consistent with that of human raters, and revise emails in ways that enhanced politeness, mitigation, acknowledgement of imposition, and status-congruent language. However, the study further found that ChatGPT's performance was marked by a systematic preference for greater formality, conventionalisation, and formulaic email structures, resulting in lower tolerance for context-dependent variation and situational flexibility than that shown by human lecturers. Overall, the findings suggest that ChatGPT performs particularly well in written, highly conventionalised genres such as formal email writing, but that its apparent pragmatic and metapragmatic abilities remain grounded primarily in learned discourse patterns rather than in fully context-sensitive, interactionally grounded reasoning. Despite these limitations, the findings suggest that contemporary LLMs may serve as useful pedagogical support tools for L2 email pragmatics by offering norm-oriented feedback and revision models that can enhance learners'awareness of conventional academic email etiquette and institutional appropriateness. (c) 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
Diplomatic speeches serve as a critical site for the discursive construction of national identity and the negotiation of cross-cultural values. Drawing on Systemic Functional Linguistics, this study examines how projection strategies in Chinese diplomatic speeches construct cross-cultural dialogue and project China's cultural distinctiveness. Based on a corpus of 153 speeches (2013-2023), 321 projection instances are analyzed. Findings reveal that projection sources create a bidirectional dialogue between Chinese and foreign voices, positioning China as an engaged participant in global cultural exchange while foregrounding its own cultural heritage through the strategic deployment of classical Chinese texts. The predominant use of neutral projecting verbs sustains an objective discursive stance, embodying a diplomatic ethos of reasoned dialogue across cultural boundaries. Projected content falls into three semantic domains: governance philosophy, strategic wisdom, and normative principles, with each articulating China's vision for global affairs while drawing on distinctively Chinese philosophical traditions that resonate with diverse cultural contexts. Projection thus functions as a discursive mechanism for negotiating multilingual identities and fostering cross-cultural intelligibility, contributing to understandings of language politics in multicultural international relations. (c) 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
The Russia-Ukraine war has intensified competition over how language constructs political authority, legitimacy, and moral credibility. This study investigates how three political leaders - Donald Trump, Volodymyr Zelenskyy, and Vladimir Putin - perform authenticity as a metapragmatic index in wartime diplomacy. Drawing on Conceptual Metaphor Theory (CMT), Critical Discourse Analysis (CDA), intertextuality and stancetaking frameworks, the research explores how pragmatic devices such as metaphors, evaluative stance, and intertextual references signal and co-construct authenticity as a communicative strategy that legitimizes power and ideological positioning. Particular attention is paid to the JOURNEY metaphor and its underlying Source-Path-Goal schema. The analysis shows that when the Path component diverges across leaders' discourses, each constructs a distinct narrative of progress or resolution that could impede a shared negotiation ground. Intertextual borrowing and stance alignment further index the leaders' ideological orientations and emotional positioning within global narratives of war and peace. Methodologically, the study combines corpus-based discourse analysis with qualitative pragmatic interpretation of speeches delivered in relation to the 2025 Alaska Summit. The findings demonstrate that authenticity is not an inherent personal trait but a discursively produced, weaponizing resource through which discourse is recognized as legitimate and persuasive. Overall, the study reveals how authenticity operates as a metapragmatic index that links micro-level linguistic forms to macro-level geopolitical effects guiding how leadership, legitimacy, and truth are perceived in the wartime communication. (c) 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
