Multilingual Settings Research Articles

“He speaks great English—For a guy from Moscow”: Language ideologies in NHL media

Ideologies about languages and countries are hard to shake, even in a multinational, multilingual setting like the National Hockey League (NHL) and the journalists who report on it. Despite its historical roots in Montréal and the dominance of Canadian and European players, the lingua franca of the NHL is English. In this work, we used qualitative analyses to examine players’, journalists’, and coaches’ attitudes toward languages other than English used on the ice. Across all groups, we found that Russian speakers were most likely to be assessed negatively, from being taciturn and unwilling to be interviewed (Frederickson, 2023) to being unlikely to speak good English (Keefe, 2023). Additionally, English-speaking players were more likely to associate positive sentiments with native North American English, Swedish, and Finnish speakers, but negative sentiments about Canadian French speakers and players from eastern European countries. Players and coaches also tended to be split on whether it was acceptable for other languages to be spoken in the locker room and on the ice. This work points to a fragmented in-group view of the acceptable language to use in professional hockey.

Investigating post-focus compression in code-switching: A comparative study of Twain Mandarin and English

On-focus pitch range expansion is a documented phenomenon in both English and Mandarin Chinese. However, the presence of post-focus compression (PFC) contrasts with its absence in Taiwan Mandarin, a regional variant of Mandarin Chinese. This discrepancy becomes particularly intriguing in the context of code-switching, which provides a unique lens to observe the interplay between these two languages. This study probes the manifestation and cognitive processing of PFC in code-switching constructs, employing a bilingual paradigm between English and Taiwan Mandarin. Through a carefully designed production experiment, this research addresses two questions: (a) Is PFC evidence in the code-switching utterances of simultaneous and early bilinguals? (b) Is there an observable correlation between the emergence of PRC in linguistic output and the language proficiency of late bilinguals, potentially indicating a transfer effect from Taiwan Mandarin to English? Anticipated results may substantiate the presence of linguistic transfer in the code-switching sentences of bilingual individuals of diverse English proficiencies. Furthermore, the implications of these findings may extend to pedagogical strategies, offering insights into the deployment of code-switching within multilingual educational settings.

Using word embeddings for immigrant and refugee stereotype quantification in a diachronic and multilingual setting

Word embeddings are efficient machine-learning-based representations of human language used in many Natural Language Processing tasks nowadays. Due to their ability to learn underlying word association patterns present in large volumes of data, it is possible to observe various sociolinguistic phenomena in the embedding semantic space, such as social stereotypes. The use of stereotypical framing in discourse can be detrimental and induce misconceptions about certain groups, such as immigrants and refugees, especially when used by media and politicians in public discourse. In this paper, we use word embeddings to investigate immigrant and refugee stereotypes in a multilingual and diachronic setting. We analyze the Danish, Dutch, English, and Spanish portions of four different multilingual corpora of political discourse, covering the 1997–2018 period. Then, we measure the effect of sociopolitical variables such as the number of offences committed and the size of the refugee and immigrant groups in the host country over our measurements of stereotypical association using the Bayesian multilevel framework. Our results indicate the presence of stereotypical associations towards both immigrants and refugees for all 4 languages, and that the immigrants are overall more strongly associated with the stereotypical frames than refugees.

Filtering offensive language from multilingual social media contents: A deep learning approach

In the face of uncontrolled offensive content on social media, automated detection emerges as a critical need. This paper tackles this challenge by proposing a novel approach for identifying offensive language in multilingual, code-mixed, and script-mixed settings. The study presents a novel multilingual hybrid dataset constructed by merging diverse monolingual and bilingual resources. Further, we systematically evaluate the impact of input representations (Word2Vec, Global Vectors for Word Representation (or GloVe), Bidirectional Encoder Representations from Transformers (or BERT), and uniform initialization) and deep learning models (Convolutional Neural Network (or CNN), Bidirectional Long Short Term Memory (or Bi-LSTM), Bi-LSTM-Attention, and fine-tuned BERT) on detection accuracy. Our comprehensive experiments on a dataset of 42,560 social media comments from five languages (English, Hindi, German, Tamil, and Malayalam) reveal the superiority of fine-tuned BERT. Notably, it achieves a macro average F1-score of 0.79 for monolingual tasks and an impressive 0.86 for code-mixed and script-mixed tasks. These findings significantly advance offensive language detection methodologies and shed light on the complex dynamics of multilingual social media, paving the way for more inclusive and safer online communities.

Enhancing pre-service teachers’ projective agency for diverse and multilingual classrooms through a course on curriculum development

ABSTRACT Framed in teacher research, this article examines on how a group of 220 preservice language teachers’ understandings of teacher agency evolved in a course on second language teaching curriculum. The participants were enrolled on a master’s program on teaching English to speakers of other languages (TESOL), and data were gathered through course activities which included two surveys, reflective blogs, and an essay. Based on inferential statistics and qualitative content analysis, findings show that, with different degrees of confidence, the student-teachers believe that their projective (future) agency was harnessed through the course developing (1) willingness to use teacher agency as a tool for professional development and (2) willingness to use teacher agency as a drive to engage in critical teaching practices for multilingual settings. Drawing on a trans-perspective of language teacher agency, the study advances a model of projective teacher agency.

Why should we care about multilingualism, gender, and sexuality?

ABSTRACT In this introductory essay, we lay some theoretical ground to the special issue on Queering Multilingualism. We outline what we mean by ‘queering’, and why it is necessary in relation to multilingualism. Ultimately, our argument is that a focus on gender and sexuality allows us to expand our understanding of the complexity of power relations in multilingual settings, encounters, and lived experiences. With the help of examples from the contributions to the special issue we not only illustrate how language classrooms and learners’ interactions are constrained by cis-genderism and heteronormativity, but we also show how language teachers and learners can challenge gender and sexual normativities. Based on the narratives of multilingual speakers, we also argue for the importance of considering gender and sexuality in relation to other identity categories such as age, ethnicity, race, and social class. Theoretically, this requires bringing queer theoretical approaches into dialogue with intersectionality. We conclude with some reflections on the relationship between multilingualism and sexual desire, especially in contexts of conflict.

The Music of the Silent Exodus: Nunchi Bwa-ing and Christian Musicking in a Second-Generation Asian American Church

In 1996, Helen Lee dubbed the departure of second-generation Asian Americans from the non-English-speaking immigrant churches that they were raised in as the “silent exodus”. This nationwide phenomenon was taking place largely because first-generation churches failed to provide the second generation with culturally relevant care that would enrich their ethnic, national, and spiritual identities. Glory, the church of focus in this study, was founded by and is home to many silent exiles. In hopes of being an enriching church for second-generation Asian Americans, pastoral staff and leaders have created spaces within Glory for racial identity and faith to be in conversation with one another. However, in regard to the music of the church, they were stumped on what could be done to make it uniquely and proudly Asian American. This conundrum inspired a key question in this study: What is distinct about the way that Asian Americans worship God through music? This study argues that the worship music at Glory Church is distinctly Asian American not by what is sonically perceived, but rather by what is physically performed and collectively experienced. The Korean-English, or Konglish, term nunchi bwa-ing (눈치 봐-ing) is utilized as a keyword to describes Christian musicking in a multilingual setting and foregrounds the Korean/Asian American worshiping body. This study concludes by looking forward and arguing that Asian Americans ought to amplify their worship music to the larger Contemporary Worship Music scene as it has the potential to be a powerful site of intergenerational healing.

Investigating the Impact of Multilingualism on Language Development: A Case Study Approach

This research study follows the effects of a multilingual setting on the language development of a toddler. The participant of this research grew up in a household where three languages were spoken. The case study investigates the participant's language development via observational analysis, unstructured child interviews, and parental questionnaire. The findings conclude that the respected individual’s language development progressed at a slower pace compared to normal language development patterns. Her exposure to different linguistic structures and limited peer-peer interactions affected her language development, resulting in code mixing and cross linguistic influence. Moreover, there is also a comparison of the participant’s language milestones to the normal pattern of speech development in this study. Lastly, the study concludes that multilingualism can have both positive and negative influence on language development, and early intervention strategies are recommended for children with language delay in multilingual environments. Further research is encouraged to explore the impact of cross linguistic influence on multilingualism in diverse populations.

A Systematic Review of Cross-Lingual Sentiment Analysis: Tasks, Strategies, and Prospects

Traditional methods for sentiment analysis, when applied in a monolingual context, often yield less than optimal results in multilingual settings. This underscores the need for a more thorough exploration of cross-lingual sentiment analysis (CLSA) methodologies to improve analytical effectiveness. CLSA, confronted with obstacles such as linguistic disparities and a lack of resources, seeks to evaluate sentiments across a range of languages. First, the research background, challenges, existing solution ideas and evaluation tasks of CLSA are summarized. Subsequently, new perspectives including different granularity levels, machine translation support, and sentiment transfer strategies perspectives are highlighted. Finally, potential avenues for future research are discussed.

Acts of bridging in communication with newly arrived Ukrainian children in the Norwegian ECEC context

In this paper, we explore emergent communicative practices in a situation where a group of newly arrived children start attending early childhood education and care (ECEC) in a new country. The data are collected in a unique multilingual ECEC setting, organised temporarily for Ukrainian children that had arrived in Norway. Previous research on newly arrived children in ECEC settings in the Nordic context focuses mostly on how one or a few children are integrated into majority-language-speaking groups of children. Our study, however, provides an opportunity to explore communication practices that arise when a group of children share a language with some of the practitioners working with them, but not others. In this paper, we present an analysis of several narratives from practice, where we discuss how children and practitioners use their semiotic resources to accomplish their goals in communicative acts. The narratives show how the ECEC practitioners support children in their communicative efforts and bridge linguistic resources between languages. Inspired by the theory of bridging and bonding social capital (Putnam, 2000), we give a definition to acts of bridging and show that these acts of bridging can be performed by both children and practitioners with different language competences.

The value of language skills in Latvia: From socio-economic inequality to multilingual awareness

Abstract This paper investigates the opposition between the realities and ideologies surrounding multilingualism in Latvia. Building mostly on Bourdieu’s theory of language in society, and multilingual awareness as a concept, we study the socio-economic value of multilingual repertoires in a multilingual setting (Latvia), and language users’ awareness of the link between their language competences (Latvian, Russian, English) and socio-economic status. By analysing sociolinguistic survey data, we demonstrate, firstly, that different multilingual repertoires are linked with differences in family income, level of education, and labour market position. Secondly, our data indicate that people generally do not acknowledge the link between their language skills and socio-economic status. To explore this further, we zoom in on language ideologies in formal education. We examine data on the multilingual awareness of Latvian language teachers from focus-group interviews and show that they tend to construe languages in competition, emphasise the symbolic value of Latvian, and ignore the multilingualism of their pupils and broader society. The study illuminates the dissonance between the multilingual realities of a post-Soviet EU country in which multilingual repertoires carry different values for individuals, and the Latvian-centred ideologies (disseminated by the teachers) that leave the presence of multilingualism and accompanying issues unacknowledged.

Model-Agnostic Meta-Learning for Multilingual Hate Speech Detection

Hate speech in social media is a growing phenomenon, and detecting such toxic content has recently gained significant traction in the research community. Existing studies have explored fine-tuning language models (LMs) to perform hate speech detection, and these solutions have yielded significant performance. However, most of these studies are limited to detecting hate speech only in English, neglecting the bulk of hateful content that is generated in other languages, particularly in low-resource languages. Developing a classifier that captures hate speech and nuances in a low-resource language with limited data is extremely challenging. To fill the research gap, we propose HateMAML, a model-agnostic meta-learning (MAML)-based framework that effectively performs hate speech detection in low-resource languages. HateMAML utilizes a self-supervision strategy to overcome the limitation of data scarcity and produces better LM initialization for fast adaptation to an unseen target language (i.e., cross-lingual transfer) or other hate speech datasets (i.e., domain generalization). Extensive experiments are conducted on five datasets across eight different low-resource languages. The results show that HateMAML outperforms the state-of-the-art baselines by more than 3% in the cross-domain multilingual transfer setting. We also conduct ablation studies to analyze the characteristics of HateMAML.

Linguistic landscapes in multilingual urban settings: Insights from translation perspectives

A multilingual urban area is a translation space that allows the exchange of ideas across languages and cultures. Yet, little research has examined how translation plays a role in shaping linguistic landscapes that depict coexisting languages in public spaces. This paper aims to examine the linguistic landscape of public signage from the viewpoint of translation. A total of 123 bilingual signage was collected from the linguistic landscape of Gorontalo City, an emerging tourism industry with an interesting multilingual setting in Eastern Indonesia. The data were analyzed using the translation category of multilingual writing developed by Reh (2004) and Edelman (2010) to provide an understanding of the translational practice evident in the signage. The analysis also looked at the direction of the translated signage, including official and non-official translation based on the linguistic landscape orientation (top-down and bottom-up), as well as translation to demonstrate collective identity. Translation practices, categorized into word-for-word (67 items), free (8 items), partial (11 items), and non-translation (37 items), demonstrate a growing trend toward linguistic diversity. The top-down approach is evident in official translations (50 items) by government entities, while non-official translations (73 items) dominate commercial spaces. Arabic, displayed alongside Indonesian and English, symbolizes religious identity in public signage. Overall, Gorontalo’s linguistic landscape reflects a shift to bilingualism, particularly with widespread English translations. Its linguistic landscape showcases a dynamic interplay of language, translation, and cultural identity in an evolving urban environment.

Multilingualism an Asset to National Development: Nigeria in Focus

Linguistic diversity in the world today has been an issue of growing social importance because a majority of all living languages are threaded in their continued existence. This brought about this research work: “Multilingualism as an asset to national development, a study of Nigeria”. the main aim of this research work is to appreciate the idea that multilingualism may be an asset of national development or a liability, to find a way of supporting it, the equilibrium points between multilingualism and national development, to give it some valuable suggestions which might be of help in language planning in a multilingual and multicultural setting like Nigeria. to bring this to the limelight, some research works were carried out as the researcher shared out some questionnaires, of which the respondents were a total of one hundred copies were collected for the research, the respondent is the students of English department and from other departments in Madonna University Okija. With a percentage and a frequency table, the data were properly analyzed. The outcome or result obtained showed that multilingualism or linguistic diversity did not project or promote disintegration in the country; rather the linguistic diversity in Nigeria has been a national resource that helped to move the country forward and not take multilingualism to be an instrument for national disintegration. In conclusion, the researcher made some recommendations that both the people at the helm of affairs should help in promoting national development through the use of multilingual policy in the country and the dissemination of information materials to the public like newspapers, posters, computer applications and programs to promote multilingualism.

Color-Based Metaphors in English and the Process of Metaphorization in Technical Texts: A Review Study

The study of the complicated processes taking place in technical texts could change not only the mechanism of a contemporary term formation, but also widen their meanings and extend spheres of their application in both monolingual and multilingual communicative settings. Accordingly, this review study aims at the elaboration of the use and classification of the color-based metaphorical terms in English as well as at the analysis of the tendency for metaphorization of technical texts in English. It further aims at explaining the tendencies in the development of technical texts that have caused changes in the development of technical vocabulary initiating the ongoing process of metaphorization and an extension of meaning of the existing and newly created terms. To this end, the complicated mechanisms underlying the emerging concepts of enduring metaphorization are explained and the processes of meaning formation, extension or compression are illustrated. In the end, as there are potential threats in the pattern of the color-based term creation in technical terms in English and their possible translation into other languages, some relevant caveats are provided.

A Little Caribbean in Madrid: analysis of the Dominican identity in the public space

According to the field of sociolinguistics of globalization, migrants resettle in new countries through deterritorialization and reterritorialization processes, which entail changes in the perception of the language and symbols of the homeland and those of the communities that migrants establish themselves in. Given that this contextual relation is expressed in the public space, the present study analyzes how Dominican resources are used in the linguistic and semiotic landscape of Tetuán, a well-known Dominican neighborhood of Madrid. Using both quantitative and qualitative methodologies, results show both the presence of the Dominican Spanish as well as Dominican symbols in the public space. Findings suggest that the Dominican community in Madrid has adapted their new place to make it more similar to their homeland, the Dominican Republic, reterritorializing the linguistic and semiotic landscape of Tetuán while building and shaping new identities. This study contributes to the body of research on linguistic landscape and linguistic attitudes in multilingual settings.

Mixed language in flux? The various impacts of multilingual contact on Lánnang-uè’s wh-question system

Aims and objectives: What exactly happens to a mixed language’s system in a multilingual contact setting? This study aims to investigate the interactions between speakers’ exposure to, frequency of, and proficiency in four languages (English, Tagalog, Hokkien, and Mandarin) and their influences on the why-fronting only wh-question system of Lánnang-uè, a mixed language used by the metropolitan Manila Lannangs. It also aims to test the validity of the assertion that symbiotic mixed languages are more likely to be in flux. Methodology: The Lánnang-uè speakers participated in production and acceptability experiments. Data analysis: Likelihood to front (production) and responses to a 7-point Likert-type scale (acceptability) were the dependent variables for several Bayesian linear mixed-effects models with age, frequency of language use, and language proficiency as primary fixed effects, sociolinguistic factors (e.g., attitudes, identity) as covariates, and participant (and when appropriate, item) as a random effect. Findings: Both production and acceptability results showed that the effects of contact are numerous and far from homogeneous. They vary depending on the source language, wh-phrase type, and degree of consciousness. They corroborate the widely held belief that mixed languages are less stable in symbiotic contexts. However, this paper goes an extra step to show that this “instability” or variability is not always a consequence of contact-induced transfer. It demonstrates that when the source languages influence the stability or development of the mixed language, the effects can be diverse, encompassing aspects such as identity processes, language attitudes, structural transfer, and/or other sociolinguistic innovations. Originality: This article is one of the first studies to examine the effects of contact between multiple languages on a mixed language variable using both production and acceptability experimental data in a five-language context. It is one of the very few variationist works in the Philippines that considers the effects of multilingualism on variation and change.

Application of Deep Learning-Based Natural Language Processing in Multilingual Sentiment Analysis

This study explores the application of deep learning-based natural language processing technologies in multilingual sentiment analysis. By examining the performance of deep learning models such as BERT and LSTM in multilingual settings, the research demonstrates the effectiveness of these models in cross-linguistic sentiment classification tasks. Despite progress, major challenges in multilingual sentiment analysis include language and cultural differences, insufficient complex context processing, and data imbalance. Future research directions include optimizing the models' contextual understanding abilities, leveraging multilingual data resources, exploring novel neural network architectures, and improving assessment metrics. With these measures, the accuracy and efficiency of multilingual sentiment analysis are expected to be significantly enhanced, further advancing the global application of natural language processing technologies.

Language of instruction and learning achievements inequalities in francophone Sub-Saharan Africa: A residualized quantile regression analysis using PASEC data

In multilingual settings, selecting a medium of instruction (MoI) has been widely debated. While proponents advocate for using the official MoI, others support using the mother tongue or adopting bilingual/multilingual MoI. Despite the potential for each MoI model to contribute to learning inequalities among students, there needs to be more evidence regarding the specific categories of students and contextual factors involved. This study focuses on fourteen French-speaking Sub-Saharan African countries (Benin, Burkina Faso, Burundi, Cameroon, Chad, Congo, Côte d'Ivoire, Democratic Republic of Congo, Gabon, Guinea, Madagascar, Niger, Senegal, and Togo) to investigate disparities in learning outcomes between students exposed to the official MoI and those exposed to a bilingual/multilingual MoI. Leveraging data from PASEC 2019, the study employs Residualized Quantile Regression (RQR) as its analytical method. The findings indicate that each MoI model may result in learning outcome disparities, with the extent of inequality varying according to students' performance levels (low, average, or high achievers). Additionally, these disparities are influenced by factors such as the student's urban or rural residence and the specific characteristics of the educational system in place. These findings underscore the importance of developing language-in-education policies that consider students' characteristics (e.g., ability) and contextual factors.

Actuarial applications of natural language processing using transformers: Case studies for using text features in an actuarial context

Abstract This paper demonstrates workflows to incorporate text data into actuarial classification and regression tasks. The main focus is on methods employing transformer-based models. A dataset of car accident descriptions with an average length of 400 words, available in English and German, and a dataset with short property insurance claims descriptions, are used to demonstrate these techniques. The case studies tackle challenges related to a multilingual setting and long input sequences. They also show ways to interpret model output and to assess and improve model performance, by fine-tuning the models to the domain of application or to a specific prediction task. Finally, the paper provides practical approaches to handle classification tasks in situations with no or only few labelled data. The results achieved by using the language-understanding skills of off-the-shelf natural language processing (NLP) models with only minimal pre-processing and fine-tuning clearly demonstrate the power of transfer learning for practical applications.