This paper examines replication research in pragmatics. The paper has three goals: to understand how replication has been used in pragmatics, to explore how replication research can enrich research in pragmatics and language learning, and to offer some suggestions for replication projects in L2 pragmatics. The paper examines sets of original and replicated studies in both L1 and L2 pragmatics to understand the range of research that has been conducted. It then considers the status of item replications (repeated scenarios) that characterize L2 pragmatics research. And it concludes by considering specific issues in L2 pragmatics research that can be insightfully investigated via replication.

High-vowel laxing in Laurentian French is notoriously variable and complex: while high-vowel tenseness is categorically predictable in final syllables, speakers seemingly apply distinct combinations of optional processes in non-final syllables (see, e.g., Dumas 1987 and Poliquin 2006). The current study investigates laxing in non-final syllables with two core objectives: (a) to determine which grammars individual speakers have acquired, and (b) to elucidate whether subgroups within the community have distinct grammars as suggested by Poliquin or instead these subgroups are superficial categorisations (e.g., emerging from a shared community with wide distributions of possible weightings for constraints). The results reveal that a larger number of superficially distinct individual grammars emerge than were proposed in existing literature, but that these patterns fall on a spectrum centred on a shared community grammar. They also provide new evidence for the importance of prosody in conditioning phonological processes in this variety of French.

The sequential units of language (i.e. words) have often been characterized by a tension between diversity and universality in the triangulation between information content, length and frequency. Here we examine similar tensions in the sequential units of visual narratives (i.e. panels) by focusing on how many entities appear per panel in visual narratives from the TINTIN Corpus of 1,030 annotated comics from 144 countries (76,000+ panels). Rates of entities per panel differ in regularized ways between styles of comics that cut across global regions, implicating typologically different ‘visual languages’. Entities per panel were also associated with panel size, where greater numbers of entities were associated with larger sizes of panels. Finally, a negative association appeared between panels with different numbers of entities and their frequency, reminiscent of a Zipf’s law of abbreviation. As associations of both size and frequency with character per panel persisted in a uniform way across styles, it implies universal tendencies transcending the diversity across systems, consistent with typological properties of languages.

Efforts to integrate intelligent chatbots into academic courses, particularly for language learning, have been gaining popularity. However, the impact of chatbot-supported collaborative learning (CL) on student engagement and English speaking skills is under-researched. This study explored the impact of utilizing intelligent chatbot–supported CL on student engagement and speaking skills of English as a foreign language (EFL) learners. It investigated how chatbot-supported CL influences student engagement and speaking skills. The experimental group was taught using chatbot-supported CL, while the control group followed conventional CL. A total of 75 first-year undergraduate students participated, with 39 students in the experimental group and 36 in the control group. Data were collected through a 14-item engagement questionnaire, a speaking test based on the IELTS speaking evaluation rubric for both groups, and a 5-item CL questionnaire administered solely to the experimental group. The data were analyzed using repeated measures analysis of variance (RM-ANOVA) and linear regression analysis. The RM-ANOVA results showed that chatbot-supported CL positively affected student engagement and speaking skills. The linear regression analysis further indicated that CL supported by intelligent chatbots influenced student engagement, which in turn significantly impacted speaking skills. The findings suggested that active engagement in CL speaking classes is crucial for improving EFL speaking skills and that intelligent chatbots can be valuable and effective tools for promoting such engagement.

A particular fashion and lifestyle aesthetic called kankokuppo (Koreaish) has gained popularity among young Japanese women in the early 2020s, who increasingly admire what they perceive to embody the “atmosphere (fun’iki)” of South Korea. This article examines the semiotic rendering of a sensuous perception of Korea identified as “Koreaish” through aesthetically embodied practices and mediatized discourses. The analysis reveals the centrality of what I call a “soft unity”: softness that arises from ambiguated boundaries, taken up across discrete objects, practices, and social value regimes. Alongside growing calls to change Japanese society from a divisive to a borderless one, this softness is valorized as the quality of idealized sociality despite its association with highly normative femininity. The emergent “Koreaish” is emblematic of the postfeminist reformulation of the feminine ideal in neoliberal Japan, which manifests as a nexus of the demanifestation of differences and the reversion to conservative feminine values.

The concepts of ‘dangerous’, ‘celebratory’, and ‘hegemonic’ multilingualism provide a valuable heuristic to explore language ideologies within supranational organizations like the United Nations. Adopting a critical stance in relation to the functions and values assigned to multilingualism and applying corpus-assisted discourse analysis, this study examines three ideological manifestations: verbalizations, metapragmatic acts, and linguistic practices in United Nations debates on the 1995 multilingualism resolution. The study analyses how member state representatives index their ideological stance: metadiscursively via verbalizations within the context of language policy debates, via acts of voting, and via their use of multilingualism as positioning devices within these debates. Unlike previous investigations of language ideology which have predominantly and exclusively focussed on discursive analyses of texts, this article forwards a tripartite analytic framework. We argue that this model serves to afford a holistic examination of practised and stated attitudes towards multilingualism, which in turn have consequences for language policy outcomes. (Language ideology, language policy, multilingualism, United Nations)

This study examined the impact of video-based synchronous computer-mediated communication (SvCMC) on engagement in collaborative pre-task planning and L2 content used in task performance. One hundred twenty-eight Hong Kong learners of English were assigned to either a face-to-face (FTF) group (n = 64) or an SvCMC group (n = 64), where they completed planning for a monologic task. Based on Philp and Duchesne’s multidimensional model, planning was analyzed for cognitive (negotiation of meaning, semantically engaged talk), social (affiliative responses) and emotional engagement (enjoyment, anxiety), and task performance was coded for L2 content. Results showed that FTF mode led to significantly lower anxiety, more semantically engaged talk, and more affiliative responses, as well as conceptually richer task performances. Use of planned content was predicted by L2 proficiency and semantically engaged talk and negatively predicted by anxiety. Our findings contribute to an understanding of engagement in SvCMC and FTF modes and their impact on L2 learning.

Large-language models (LLMs) have transformed natural language processing and opened new possibilities for the computational social sciences and digital humanities. Yet translating historical sources remains difficult because early modern varieties are scarcely represented in contemporary training corpora and because standard tokenizers fragment their non-standard orthography. This article tackles these gaps by adapting open LLMs to early modern Dutch-to-English translation and advances two concrete contributions: (i) a memory-efficient fine-tuning workflow that runs on a single consumer GPU, comparing order-reward policy optimization with the Unsloth supervised fine-tuning approach and (ii) a verifiable evaluation protocol that combines embedding-based metrics with systematic expert review. Experiments on testimonial texts (1680–1792) show that fine-tuning choice decisively shapes quality: the Unsloth-tuned Mistral model attains the highest BERTScore and METEOR values and most faithfully preserves historical nuance. The framework supports a collaborative workflow where machine-generated drafts accelerate expert translation, making archival texts more accessible while maintaining scholarly oversight through domain-expert validation.

This study examines the use of English on signage in the areas surrounding two U.S. military bases in Japan, Yokosuka Navy Base and Yokota Air Base, highlighting the diverse functions of English that emerge through interactions between American residents and Japanese locals. In these areas, including streets that have become tourist destinations for Japanese visitors, various commercial establishments coexist, some offering authentic products and services for Americans, and others commodifying American culture for Japanese tourists. Within these public spaces, English signage, along with other semiotic resources, plays a central role in shaping a uniquely commodified environment infused with American culture. The study identifies two key features of English signage that distinguish these areas from other parts of Japan: (1) the prominence of informational English signage targeting Americans, such as monolingual English traffic signs and church signs, which reflects the dense American population; and (2) the symbolic use of English by Japanese shop owners, which do not signal modernity or globalization as English signage typically does elsewhere in Japan, but instead commodifies American culture to attract Japanese tourists. These uses of English on signage, alongside other semiotic materials, shape a unique visual and cultural landscape, underscoring the multifaceted roles of English in non‑English‑speaking countries like Japan, including both the ‘authentic’ use of English in communication with Americans and the ‘commodified authenticity’ conveyed through English employed by Japanese shop owners to appeal to Japanese audiences.

Dual subtitles, combining captions (audio transcription) with subtitles translated into another language, are increasingly used in language learning. However, how they shape visual attention remains unclear. In the present experiments, we tracked the eye movements of Spanish–English bilinguals, as they viewed instructional videos with either no subtitles (Experiment 1) or dual subtitles (Experiment 2), manipulating subtitle position and audio language. Without subtitles, L1 audio focused gaze on the speaker’s eyes, while L2 audio distributed it between the eyes and mouth. With dual subtitles, gaze shifted strongly to the text, with a preference for the top line, which attracted more viewing time regardless of language. Viewers selectively attended to the line matching the audio. Comprehension improved for L2 audio with subtitles, while L1 comprehension was unaffected. Our findings demonstrate that display layout and language alignment jointly govern attentional allocation in bilingual viewing, with direct implications for L2 instructional design.

Bilinguals vary in their daily-life language use and switching behaviours, which are also frequently studied in relation to other processes (e.g., executive control). Measuring daily-life language use and switching often relies on self-reported questionnaires, but little is known about the validity of these questionnaires. Here, we present two studies examining test–retest reliability and validity of language-use questionnaires (relative to Ecological Momentary Assessment, Study 1) and language-switching questionnaires and tasks (relative to recorded daily-life conversations, small-scale Study 2). Test–retest reliability and validity of the LSBQ (Anderson et al., 2018) were high and moderate, respectively, suggesting this questionnaire can capture daily-life language use well. Although only examined with a small sample size, Study 2 suggested relatively low validity of most language-switching questionnaires, with short language-production tasks potentially offering a more valid assessment. Together, these studies suggest that tools are available to reliably capture language use and switching with (a certain degree of) validity.