Language Discrimination Research Articles

A Deep Learning Approach for Identifying and Discriminating Spoken Arabic Among Other Languages

Spoken Language Identification (SLID) is an important step in speech-to-speech translation systems and multi-lingual automatic speech recognition. In recent research, deep learning mechanisms have been the prevailing approaches for spoken language identification. This paper aims to study, detect, and analyze spoken languages similar to Arabic in pronouncing certain words and then proposes a deep learning-based architecture, specifically the Bidirectional Long Short Term Memory (BLSTM), for spoken Arabic language identification and discrimination between these similar languages, namely, German, Spanish, French, and Russian, all of which are taken from Mozilla speech corpus languages. Additionally, our work involves a linguistic study of these considered languages. A total of ten thousand speakers are chosen for all five languages, and the BLSTM architecture is designed and implemented using acoustic signal features and applied to five experiments in this paper. The results show a precision of 98.97%, 98.73%, 98.47%, and 99.75% for identifying the spoken Arabic language separately along with German, Spanish, French, and Russian, respectively. Additionally, we achieved an average accuracy of 95.15% for discriminating between all these considered five languages in terms of the pronunciation of words. Our findings confirm that a BLSTM architecture is able to distinguish between observable similar pronunciations of words in considered languages.

Neurocognitive Mechanisms of Anti-Lingual & Accent Bias Stereotyping in Virtual Reality and the Sustainable Development of Linguistic Response

Raising awareness through Virtual Experiencing (RAVE) and a Cross-Cultural Perspective on Raising Awareness through Virtual Experiencing (C-RAVE) approaches are used in this article to promote awareness of concerns relating to linguistic stereotyping. Virtual Reality (VR) technology has completely changed several facets of modern life, including instruction, training and entertainment. We give a summary of the techniques employed to increase awareness of how stereotypes might distort our vision. The patterns of the answers provided by the participants demonstrate that the reported gender of the speaker in addition to the accent (native vs. non-native), has an impact on the respondents' assessments of performance. Additionally, we were able to show that the discussions and ideas produced by these response patterns led to a rise in self-awareness of language and stereotyping-related problems. The paper gives a general description of RAVE and examines how it may improve education, create empathy and foster good social change. Focus on a critique of our methodologies and place our work within a larger debate of strategies and efforts for how educational institutions might actively contribute to combating (language) discrimination and bias in many ways. The relevance of Raising Awareness by Virtual Reality and Cross-Cultural Raising of Awareness by Virtual Reality in fostering inclusion and understanding in a society that is becoming more sustainable globally connected is highlighted in the article’s conclusion.

Medical text classification based on the discriminative pre-training model and prompt-tuning.

Medical text classification, as a fundamental medical natural language processing task, aims to identify the categories to which a short medical text belongs. Current research has focused on performing the medical text classification task using a pre-training language model through fine-tuning. However, this paradigm introduces additional parameters when training extra classifiers. Recent studies have shown that the "prompt-tuning" paradigm induces better performance in many natural language processing tasks because it bridges the gap between pre-training goals and downstream tasks. The main idea of prompt-tuning is to transform binary or multi-classification tasks into mask prediction tasks by fully exploiting the features learned by pre-training language models. This study explores, for the first time, how to classify medical texts using a discriminative pre-training language model called ERNIE-Health through prompt-tuning. Specifically, we attempt to perform prompt-tuning based on the multi-token selection task, which is a pre-training task of ERNIE-Health. The raw text is wrapped into a new sequence with a template in which the category label is replaced by a [UNK] token. The model is then trained to calculate the probability distribution of the candidate categories. Our method is tested on the KUAKE-Question Intention Classification and CHiP-Clinical Trial Criterion datasets and obtains the accuracy values of 0.866 and 0.861. In addition, the loss values of our model decrease faster throughout the training period compared to the fine-tuning. The experimental results provide valuable insights to the community and suggest that prompt-tuning can be a promising approach to improve the performance of pre-training models in domain-specific tasks.

Postcolonial analysis of educational language policies of Ireland, Singapore, and Malaysia

The aim of this paper is to compare the educational language policing in Ireland, Singapore, and Malaysia. While distant geographically, the three countries experience similar linguistic processes when it comes to anglicisation, and propose different solutions to the issue of balancing linguistic rights, and promotion of English as the language of globalisation.
 This comparison aimed to find out what influences language policing in postcolonial countries, and in what ways language shift can be prevented. The aspects of language policing strategies are presented as a way of protecting linguistic human rights, but also as a way of dealing with the aftermaths of the policies implemented by the British Empire.
 Similarities and distinctions in the language policies of Ireland, Singapore, and Malaysia prove that the weak position of native languages originates not in the “natural” decline of a language, but rather in the policy of promoting English by the colonial forces. Ethnic and linguistic discrimination favouring English speakers in Ireland, Singapore, and Malaysia, originates in similar, imperial linguistic ideologies, which are still reflected in the current language policies of countries of colonial past. While the countries approach their bilingual educational policing in different ways, ultimately the outcomes seem similar when it comes to linguistic attitudes and prestige.

LANGUAGE-BASED DISCRIMINATION RELATES TO MEMORY COMPLAINTS AMONG LATINX/HISPANIC OLDER ADULTS

Abstract Memory Complaints (MCs) are a risk factor for dementia, but research in this area has largely been limited to non-Latinx White adults. Previous studies have shown that Latinx/Hispanic individuals are at higher risk for Alzheimer’s Disease and Related Disorders (ADRD). The mechanisms underlying this disparity in ADRD are multidimensional and can include negative social stressors such as discrimination. Language discrimination can be a source of stress in this population and might be an important contributor to MCs. We examined if MCs varies as a function of language-based discrimination in a total of 495 Latinx/Hispanic older adults without dementia from the community-based Washington Heights Inwood Columbia Aging Project. Language-based discrimination was measured with a dichotomous item that inquired if individuals had been discriminated against because they do not speak English well (yes or no). A linear regression was conducted to examine the cross-sectional association between language-based discrimination and MCs adjusted for age, education, sex/gender, socioeconomic status (income), and depressive symptoms. The experience of language discrimination was independently associated with MCs (B= 0.83, 95% CI=0.13, 1.54, SE= 0.36, p= .021). Results demonstrate a relationship between negative stressors and the expression of MCs. Future studies should comprehensively examine other discriminatory stressors which may also impact the expression of MCs and the risk for progression to ADRD. Identifying language-based and other discriminatory stressors that negatively affect Latinx/Hispanic communities will allow us to develop models that aim to assess and prevent discrimination, as a protective mechanism for the development of MCs and ADRDs.

Безумие в дискурсе «Другой» в современной философии

Introduction. This paper considers madness as a metahistorical category of culture as well as an excluded language in Western European culture that exists despite the language code. To determine the status of madness in the contemporary philosophy the author analyses the way it functions in the discourse of “the other”. Theoretical analysis. The exclusion of madness from the context of philosophical research occurred due to such various markers changing from era to era as a lie, delusion, nonsense, buffoonery, and finally the dysfunction of the mind completed by Pinel’s reform. The definition of madness as a forbidden language comes from two theses: 1) mental disorders result from biological, not socio-psychological problems; 2) madness and a mental disorder are not two different configurations, they have organic nature that is to be neutralized in terms of pharmacology and defined in terms of behavioral deviations, and that excludes subjective and nonphysical suffering. Conclusion. The research relevance is obvious because modern aesthetics and philosophy with the cultural model of insanity as a core concept need further elaborating. The duality of the subject of psychiatry concerning the human personality implies that this component must be taken into account apart from exclusively natural-scientific studies.

Bias in Automatic Speech Recognition: The Case of African American Language

AbstractResearch on bias in artificial intelligence has grown exponentially in recent years, especially around racial bias. Many modern technologies which impact people’s lives have been shown to have significant racial biases, including automatic speech recognition (ASR) systems. Emerging studies have found that widely-used ASR systems function much more poorly on the speech of Black people. Yet, this work is limited because it lacks a deeper consideration of the sociolinguistic literature on African American Language (AAL). In this paper, then, we seek to integrate AAL research into these endeavors to analyze ways in which ASRs might be biased against the linguistic features of AAL and how the use of biased ASRs could prove harmful to speakers of AAL. Specifically, we (1) provide an overview of the ways in which AAL has been discriminated against in the workforce and healthcare in the past, and (2) explore how introducing biased ASRs in these areas could perpetuate or even deepen linguistic discrimination. We conclude with a number of questions for reflection and future work, offering this document as a resource for cross-disciplinary collaboration.

Similarities and Differences in Linguistic Discrimination Between Slovak and Hungarian Teachers of Hungarian Language and Literature

Abstract The purpose of this study is to demonstrate the presence of linguistic discrimination in pedagogical situations, especially in pedagogical evaluation. The paper is based on a survey which involved 502 Hungarian Language and Literature teachers and teacher trainees from Hungary (N = 216), Slovakia (N = 128), Romania (N = 108) and Ukraine (N = 50). Data were primarily collected through a technique similar to matched-guise tests; however, the method of the present research had some additional complexity. The article discusses similarities and differences in linguistic discrimination between Slovak and Hungarian teachers who teach Hungarian Language and Literature. The question it raises is whether there are any differences between the two samples. The results of the mentioned research show that the presence of linguistic discrimination is powerful in both samples, but there are differences in its strength and realization.

Linguistic Discrimination as a Social, Legal and Human Rights Phenomenon

Abstract This article explores the difference between equality and the absence of discrimination within the area of language use. It draws on examples of social practices of linguistic inequality in Slovakia, and analyses how these are regulated by Slovak law, international law and laws of countries of best practices. The article argues that these examples represent three different models of linguistic equality: formal equality, equality of opportunities and substantive linguistic equality. It will analyse how these approaches empower or disadvantage speakers of minority languages, and what normative guide they provide for linguistic equality.

Language Discrimination in Polish Public Discourse (Example of Polish Football Fans Discourse)

Abstract In the urban space of Łódź, this paper studies some anti–Semitic aspects of discriminatory discourse of football fans. Its main goal is to present how anti-Semitic discriminatory meanings are formulated and spread in the urban space and how particular social groups (football fans) organize the urban discourse. In the multimodal material from the urban discourse, we scrutinize verbal and visual forms of discrimination captured on the walls of buildings, parks, or shopping centres in various stickers, posters, and flags. This paper implies that the language of particular social groups goes beyond closed internet communication or stadium discourse to a more open social sphere. A consequence of the social polarization found in the analysed data may be the radicalization and vulgarization of language in general.

Bilingualism is good but codeswitching is bad: attitudes about Spanish in contact with English in the Tijuana - San Diego border area

ABSTRACT A qualitative study of language attitudes toward Spanish in contact with English in San Diego county reveals that while Spanish-English bilingualism is highly valued, Spanish-English codeswitching is disfavored. While studies in other border communities along the U.S.-Mexico border have found similar general results, the present study contributes to the field in a few novel ways: the San Diego area is an understudied location in border Spanish; four populations were interviewed for this study on both sides of the border (first and second-generation speakers of Spanish in San Diego county and in the city of Tijuana); although first-generation speakers in San Diego and Tijuana express the most negative views about codeswitching, San Diego parents emerge as uniquely positioned to understand the language experience of their children and thus act as catalysts for attitudinal change; second-generation Tijuana speakers are the least biased against codeswitching; second-generation San Diego speakers internalize the stigma against codeswitching, which results in linguistic insecurity. This study explores internal linguistic discrimination across borders and generations in the Tijuana-San Diego area.

PRAKTIK DOMINASI TOKOH ANTAGONIS MELALUI KEKERASAN SIMBOLIK DALAM FILM BETTER DAYS《少年的你》

When language is used in communication events, there also as an ideological distiction and a desire to dominate each other among its users. Because language is never separated from ideological motives, including domination motives. Domination occurs due to the existence of a strong group as the dominating party and a weak group as the dominated party with language as an instrument to achieve it. Symbolic violence marked by language forms that represent language crimes is a realization of the practice of domination. Through symbolic violence, the dominant party imposes its ideological influence on the dominant group. The practice of domination through symbolic violence can be found in the dialogue text of the antagonist in the film Better Days《少年的你》by Derek Tsang which is used as a data source. The patterned language forms used by the antagonist to attack the main character are used as data. After the data is identified and classified, it is analyzed through a qualitative approach using the concept of Bourdieu's theory of symbolic violence. The results of the analysis are presented through a descriptive method to obtain a detailed and concrete picture of the practice of domination of antagonist characters through symbolic violence. The results of the analysis show that symbolic violence, which represents the practice of domination, is realized through forms of language that have functioned as the language of punishment, the language of intimidation, the language of prohibition, the language of accusation, the language of insults, and language of command. Cultural capital (attitudes and lifestyles) is the most dominant capital that represents the attitudes and behavior of antagonists to produce and reproduce texts to legitimize their domination. Based on the frequency of occurrence, forbidden language is the most productive form of language, and accusation language is the language form with the least frequency of occurrence.

PRAKTIK DOMINASI TOKOH ANTAGONIS MELALUI KEKERASAN SIMBOLIK DALAM FILM BETTER DAYS《少年的你》

When language is used in communication events, there also as an ideological distiction and a desire to dominate each other among its users. Because language is never separated from ideological motives, including domination motives. Domination occurs due to the existence of a strong group as the dominating party and a weak group as the dominated party with language as an instrument to achieve it. Symbolic violence marked by language forms that represent language crimes is a realization of the practice of domination. Through symbolic violence, the dominant party imposes its ideological influence on the dominant group. The practice of domination through symbolic violence can be found in the dialogue text of the antagonist in the film Better Days《少年的你》by Derek Tsang which is used as a data source. The patterned language forms used by the antagonist to attack the main character are used as data. After the data is identified and classified, it is analyzed through a qualitative approach using the concept of Bourdieu's theory of symbolic violence. The results of the analysis are presented through a descriptive method to obtain a detailed and concrete picture of the practice of domination of antagonist characters through symbolic violence. The results of the analysis show that symbolic violence, which represents the practice of domination, is realized through forms of language that have functioned as the language of punishment, the language of intimidation, the language of prohibition, the language of accusation, the language of insults, and language of command. Cultural capital (attitudes and lifestyles) is the most dominant capital that represents the attitudes and behavior of antagonists to produce and reproduce texts to legitimize their domination. Based on the frequency of occurrence, forbidden language is the most productive form of language, and accusation language is the language form with the least frequency of occurrence.

Visibly invisible: The study of middle class African American English

AbstractMiddle class African American English (AAE) has remained largely invisible to the sociolinguistic lens despite the fact that over 50 years of research has made it one of the most examined varieties of American English. This gap in the sociolinguistic literature is largely reflective of a strategic effort on the part of linguists to dismantle the stigma associated with working class vernacular varieties and improve outcomes for working class speakers who face linguistic discrimination in schools. An unfortunate by‐product of this laudable effort, however, has been the erasure of middle‐class speakers from our conceptualisations of the AAE speech community and a virtual obsession with the vernacular end of the AAE continuum. By interrogating the concept of the linguistic lame and giving greater attention to patterns of code‐switching and the ways in which talking Black or sounding Black get defined at the more standard end of the continuum, sociolinguists have an opportunity to broaden our understanding of AAE and its community of speakers and, in so doing, possibly extend our reach to a more diverse and inclusive audience of budding linguists.

A Critique and Call for Action, in Response to Sexist Commentary About Vocal Fry

Purpose: This letter to the editor discusses recommendations and publicity from a recent article by Gallena and Pinto (2021) that appeared in Perspectives of the ASHA Special Interest Groups. We contextualize the recommendations made by Gallena and Pinto that young women actively suppress their use of vocal fry, a normal part of linguistic and sociolinguistic variation, to make themselves more marketable to employers who discriminate on the basis of vocal fry use. By reviewing research on vocal fry, social evaluation, and linguistic discrimination, we show how this recommendation is fundamentally flawed and how it perpetuates sexist tropes about language use. We argue that this is particularly dismaying when publicized by a journal of the American Speech-Language-Hearing Association, which aims to promote the universal use of language, in all its forms, as a human right. Conclusions: People stigmatize linguistic differences to devalue those who have less power. Vocal fry is not a problem; the problem is absurd sexist judgments against those who use vocal fry. The recommendations made in Gallena and Pinto (2021) and Perspectives' publicity about those recommendations both contribute to sexist linguistic discrimination. Rather than recommending that victims of sexism change their behavior to suit the biased views of others, we should use our energy to eradicate the underlying sexism.

Politics, Language, and Cultural Identity: DetroitRicans and Puertoricanness in Detroit

Due to a surge in racism and anti-immigrant sentiment that intensified during Trump’s campaign and presidency, some Americans have reacted to people speaking Spanish in public with hostility as well as verbal and even physical aggression over the last few years in the United States. A particular group of victims of language and identity discrimination has been Puerto Ricans, who are, ironically, American citizens. Drawing on historical perspectives, language and identity attitudes, the politicization of language, and linguistic racism approaches, the present study administered a language and identity questionnaire to 103 Puerto Ricans in Detroit, Michigan (DetroitRicans). Despite the rise of linguistic racism in the United States, 90.3% of respondents said that being able to speak Spanish was necessary to validate their Puertoricanness. In addition, 89% of this study’s participants agreed that not teaching Spanish to children was denying them their Puerto Rican culture and identity. DetroitRicans also identified Spanish as their mother tongue, their roots, and their homeland, whereas they identified English as the language of work, school, and economic advancement. The findings agree with the language and identity perceptions of Puerto Ricans living on the Island and in Central Florida; they diverge from the traditional perspectives of Boricuas in New York, North Philadelphia, and Chicago, who do not generally consider Spanish a vital part of their Puerto Rican identity.

The perception of Polish economic immigrants in Great Britain

Great Britain is perceived as a traditional immigrant country. In relation to Poland, the process of immigration to the British Isles – especially for economic purposes – intensified after 2004, with the accession to the European Union. The perception of immigrants in the UK began to change in the first months after the referendum on Brexit, when there was an increase in hate crimes, mainly xenophobic crimes. The article presents the subject of Polish economic immigrants in the United Kingdom. Based on their own research conducted among Poles living in Leicester, the authors describe the perception of this group, taking into account positive and negative features of Poles, stereotypes associated therewith or the problem of discrimination (including language discrimination).

The Drama on Provincial Accented Speech

This phenomenological study looked into the lived experiences of fourth-year students exposed to the drama of provincial accented speech. Colaizzi's descriptive approach was used in the study with the purposively chosen key informants with various tools to cross-check their responses. Results showed biases on an accent because the informants thought inferiority of their accent, harassment on accented speech for they experience overt linguistic racism, probable accent discrimination, and accent intelligibility were some of the issues the students encountered. However, there were also feats to consider in using provincial accented speech to fellow second language (L2) English speakers. These were the accent sapiency and accented speech ego and pride. Based on the study’s findings, the English second language teachers and professionals are encouraged to address these identified dramas, for they have a valuable role, provided they possess the knowledge and attitudes needed to foster acceptance of linguistic diversity.

The prestige and perceived value of home languages. Insights from an exploratory study on multilingual speakers’ own perceptions and experiences of linguistic discrimination

ABSTRACT This paper reports findings from an exploratory study on multilingual speakers conducted in Germany. Data were collected using a questionnaire instrument launched in 2021. To our knowledge, this is the first enquiry into multilinguals’ own perception of their spoken languages in Germany. The core research questions addressed in this paper are (1) whether some languages are perceived as more valuable or prestigious than others by their speakers and (2) whether speakers of certain languages experience discrimination more often than speakers of other languages. In the data exploration it was assumed that (a) all languages other than English, French, Italian, and Spanish (besides Latin the standard foreign languages in schools) will be valued ‘lower’; and (b) speakers will report experiences of discrimination with respect to their use of those languages other than German or the foreign languages mentioned above. The results indeed demonstrate differences in the perceived value and prestige of certain languages. Multivariate statistical analyses indicate that L1 speakers of languages with perceived lower value (e.g. Turkish) experience discrimination significantly more often than L1 speakers of other home languages (e.g. English). Additionally, qualitative data analyses exemplify levels of discrimination and lived experience of languages.

The language of discrimination: assessing attention discrimination by Hungarian local governments

In our study we assess the responsiveness of Hungarian local governments to requests for information by Roma and non-Roma clients, relying on a nationwide correspondence study. Our paper has both methodological and substantive relevance. The methodological novelty is that we treat discrimination as a classification problem and study to what extent emails written to Roma and non-Roma clients can be distinguished, which in turn serves as a metric of discrimination in general. We show that it is possible to detect discrimination in textual data in an automated way without human coding, and that machine learning (ML) may detect features of discrimination that human coders may not recognize. To the best of our knowledge, our study is the first attempt to assess discrimination using ML techniques. From a substantive point of view, our study focuses on linguistic features the algorithm detects behind the discrimination. Our models worked significantly better compared to random classification (the accuracy of the best of our models was 61%), confirming the differential treatment of Roma clients. The most important predictors showed that the answers sent to ostensibly Roma clients are not only shorter, but their tone is less polite and more reserved, supporting the idea of attention discrimination, in line with the results of Bartos et al. (2016). A higher level of attention discrimination is detectable against male senders, and in smaller settlements. Also, our results can be interpreted as digital discrimination in the sense in which Edelman and Luca (2014) use this term.