Applied Corpus Linguistics最新文献

This article discusses how corpus linguistic methods were adapted for critical research examining the content and pedagogy of human rights learning within texts approved for classroom use under Japan's official social studies curriculum. While human rights concepts are common facets of official curricula, designed to address social injustice and foster peaceful coexistence, such learning has undergone critical re-examination as being complicit in perpetuating social injustice. Drawing upon Bernstein's sociology of the curriculum and based on a corpus of upper-secondary curricular texts, this research is a pragmatic mixing of quantitative and qualitative methods that sought to understand the curriculum's potential role within learning for human rights. By way of empirical example, the article aims to inform future critical research designs within the sociology of the curriculum. Corpus-based analytical techniques were vital in demonstrating how the structuring of textbook human rights limits student engagement with social justice issues and functions instead to inculcate pride in Japanese ethnonationality.

This paper sets forth a quantitative analysis of expressions of epistemicity, a category covering the expression of commitment to the information transmitted and comprising epistemic modality and evidentiality, in a corpus of 400 newspaper articles from The Guardian concerning the COVID-19 pandemic. 200 articles were written in April 2020; the other 200 were written between January and April 2022, after massive vaccination and an extraordinary increase in medical knowledge. The analysis distinguishes between a number of subtypes of epistemic expressions and three kinds of authorial voice. The results show that the April 2020 articles contain more epistemic expressions, of both weak commitment (might, perhaps, apparently…) and strong commitment (know, clearly, surely…), which suggests a greater need to distinguish the known from the unknown in this period, due to the pervasive state of uncertainty. The analysis has social implications, since it gives readers an opportunity to appreciate the careful assessments of epistemicity found in the corpus and therefore to consider the convenience of obtaining information from quality media. These social implications, together with the methodology of the analysis, contribute to the potential of the paper for pedagogical applications.

As recent years have witnessed increasing pressure on personal finances, compounded by the current cost of living crisis, online forums have become an important resource for people dealing with financial precarity. In this article, we offer a corpus linguistic analysis of data from MoneySavingExpert.com, the UK's largest online money management advice forum, studying 207 threads and 41.4 million words of text posted from 2005 to 2021. Through measures of word frequency and word association, we uncover similarities and differences in language use on the debt-free wannabe (DFW) and mortgage-free wannabe (MFW) forums. Our findings show that the DFW forum focuses on interactive exchanges involving requests for help and offers of advice, while the MFW forum is characterised by goal setting and community building. We thus contribute new insights into the discursive construction of debt in digital media and provide further understanding of the role online forums play in supporting vulnerable people.

This short communication discusses the impact of ChatGPT on the field of corpus linguistics, particularly its potential as a concordancer. As a corpus linguist and app developer, the author reflects on how ChatGPT's ease of use, efficiency, and popularity could challenge traditional concordancers, and explores ways in which ChatGPT could be used to generate concordances and frequency lists.

This article explores the potential advantages of corpora over generative artificial intelligence (GenAI) in understanding language patterns and usage, while also acknowledging the potential of GenAI to address some of the main shortcomings of corpus-based data-driven learning (DDL). One of the main advantages of corpora is that we know exactly the domain of texts from which the corpus data is derived, something that we cannot track from current large language models underlying applications like ChatGPT. We know the texts that make up large general corpora such as BNC2014 and BAWE, and can even extract full texts from these corpora if needed. Corpora also allow for more nuanced analysis of language patterns, including the statistics behind multi-word units and collocations, which can be difficult for GenAI to handle. However, it is important to note that GenAI has its own strengths in advancing our understanding of language-in-use that corpora, to date, have struggled with. We therefore argue that by combining corpus and GenAI approaches, language learners can gain a more comprehensive understanding of how language works in different contexts than is currently possible using only a single approach.

There is growing evidence that corpus-based computational tools are useful in identifying changes in speech that appear to accompany cognitive decline in Alzheimer's disease (AD). It has long been known that semantic coherence of speech is altered in AD, but only recently have computational tools been developed that allow for cohesion indices to be computed in an automated fashion on larger data sets. To that end, this study examined semantic coherence in persons with AD and healthy controls. Speech transcripts from 81 individuals with probable AD (Mage = 72.7 years, SD = 8.80, 70.4% female) and 61 healthy controls (Mage = 63.9 years, SD = 8.52, 62.3% female) from DementiaBank were analyzed. Machine learning analyses of coherence were conducted, and models evaluated for classification accuracy (i.e., AD vs controls) as well as ROC-AUC. Relationships between coherence indices and MMSE performance were also quantified. Though no significant group differences emerged in local semantic coherence among adjacent words, persons with AD produced less globally coherent speech relative to healthy controls. Furthermore, global coherence indices predicted AD diagnoses with accuracy between 75% and 78% and were significantly associated with MMSE scores. These findings suggest that automated measures of global coherence can distinguish individuals with AD from healthy controls, which may point to eventual diagnostic utility in clinical settings.

The importance of language to changing public behaviours is acknowledged in crisis situations such as the COVID-19 pandemic. A key means of achieving these changes is through the use of directive speech acts, yet this area is currently under-researched. This study investigates the use of directives in the 2020 COVID-19 briefings of four leaders of English-speaking nations, Jacinda Adern, Boris Johnson, Scott Morrison, and Nicola Sturgeon. We developed a classification system including 16 directive types and used this to compare directive use across these four leaders, examining directness and forcefulness of directive use. The analysis finds Sturgeon to be the most prolific directive user and also to have the highest reliance on imperatives. Johnson, meanwhile, has a preference for directives involving modal verbs, particularly with first- and second-person pronouns. In contrast, Ardern and Morrison show a higher use of indirect directives, normally thought to be a less effective strategy. While Ardern often combines this strategy with judicious use of imperatives, this is not seen in Morrison's COVID-19 briefings. These findings tend to confirm earlier, more impressionistic evaluations of the communication styles of these leaders but also suggest other avenues for research on directive use. We conclude with implications for political crisis communication and analysis of directives in crisis communication.

This study investigates the construction and (re)negotiation of the identity boundaries in the context of a white nationalist online forum. Using over three million words of data, a corpus linguistic approach is combined with elements of critical discourse analysis, namely social actor (van Leeuwen, 1996) and transitivity (Halliday and Matthiessen, 2004) analyses, to examine the positioning of social actors in relation to one another. The data shows that, despite an assumption amongst some scholars of a united and ideologically coherent in-group of extremists, forum members often disagree on the nature and boundaries of both their racial (white) and ideological (white nationalist) identities. This calls into question the value of the `in-group' concept as we currently understand it. Instead, the `in-group' in the far-right context should be seen as slippery and unfixed, comprising multiple overlapping but distinct identities.

Communication has played a critical role during the initial response to the COVID-19 pandemic, and communicators have had a particularly difficult task in persuading different types of audience to comply with ever-changing regulations. Local government organisations play a crucial role in recontextualising the national messaging for a local audience and encouraging the public to comply with regulations.

This paper investigates local government organisations’ (henceforth LGOs) engagement strategies in COVID-related posts on social media. In collaboration with LGOs in England, we examined their communication strategies on Twitter and Facebook during the second UK national lockdown of the COVID-19 pandemic in November-December 2020. Using methods from corpus-assisted discourse studies, the paper analyses the occurrence and functions of selected interactive engagement markers, in this case personal pronouns, questions and hashtags. We find that such linguistic features function to encourage engagement by (a) helping to foster relatedness through ambiguity; (b) creating autonomy-supporting communication; and (c) making messages ‘stand out’.

Based on our corpus analysis, we discuss the initial response of the participating councils to our findings and outline future directions including the integration of multimodal approaches to studying the role of localised social media in national crisis management. We argue for more attention to be paid to the many local communicators who play an invaluable role in encouraging the public to comply with national measures in times of crisis.

This article provides a comparative analysis of how frontline workers were constructed by the UK media prior to and during the COVID-19 pandemic in the UK. Both the News on the Web Corpus and the Coronavirus Corpus, as monitor corpora of web-based new articles, were utilised to identify changes in both the frequency and use of the word front*line from 2010 to 2021. Findings show that, following the outbreak of COVID-19, constructions of frontline work were more frequently associated with medical professions and became more figurative in nature. Our findings provide a counterpoint to claims that the COVID-19 pandemic led to an increased awareness of the critical nature of many types of ‘low-skilled’ work not previously recognised as essential. The study also extends previous research which has traced changes in language and its deployment during the COVID-19 pandemic.