Language, Migration and Multilingualism in the Age of Digital Humanities

# Language, Migration and Multilingualism in the Age of Digital Humanities

Edited by Ignacio Andrés Soria, Sandra Issel-Dombert and Laura Morgenthaler García

Gedruckt mit freundlicher Unterstützung der RUB Research School

ISBN 978-3-11-074596-2 e-ISBN (PDF) 978-3-11-074608-2 e-ISBN (EPUB) 978-3-11-074618-1 DOI https://doi.org/10.1515/9783110746082

This work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License. For details go to https://creativecommons.org/licenses/by-sa/4.0/.

Creative Commons license terms for re-use do not apply to any content (such as graphs, figures, photos, excerpts, etc.) not original to the Open Access publication and further permission may be required from the rights holder. The obligation to research and clear permission lies solely with the party re-using the material.

#### Library of Congress Control Number: 2023934697

#### Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the internet at http://dnb.dnb.de.

© 2023 the author(s), editing © 2023 Ignacio Andrés Soria, Sandra Issel-Dombert and Laura Morgenthaler García, published by Walter de Gruyter GmbH, Berlin/Boston This book is published open access at www.degruyter.com.

Cover image: © Ana Andrés Soria Typesetting: Integra Software Services Pvt. Ltd. Printing and binding: CPI books GmbH, Leck

www.degruyter.com

### Contents

Ignacio Andrés Soria, Sandra Issel-Dombert, Laura Morgenthaler García At the Interface of Digitalization, Multilingualism Research and Migration Linguistics 1

### Section 1: Language Repertoires, Translingual Practices and Indexicality

Gabriela Prego Vázquez A TranslinguisTIC Safe Space 11

Maria Sabaté-Dalmau Migrants' ICT-Mediated Communication Strategies and Subversive Multilingual Practices 31

Andrea Chagas Migration and Linguistic Repertoire 51

Nélida Murguía Cruz Glottopolitical Interventions and Virtuality in Spanish Language Teaching for Senegalese Immigrants during the Pandemic (Buenos Aires, 2020–2021) 79

Franziska Kailich No Seas Covidiota. Processes of Positioning and Identification through a Neologism on Twitter 99

Olga Kellert Using Geolocated Tweets for Probing Language Geography and Migration 129

#### Section 2: Language and Identity Construction

Enrique Pato "New" Technologies and "New" Identity Formation 141 Jannis Harjus Virtual Speaker Design 169

### Section 3: Digital Discourses and Glotopolitical Perspectives

Alba Polo-Artal Gender and Racial Orders in Vox's Nativist Discourses 189

Francisco José Sánchez García Spaniards First. Notes on Hate Speech towards Migrants on Twitter 219

Svenja Dufferain-Ottmann Constructing a Crisis by Constructing a Migratory Threat 235

Index 265

### Ignacio Andrés Soria, Sandra Issel-Dombert, Laura Morgenthaler García At the Interface of Digitalization, Multilingualism Research and Migration Linguistics

### 1 Starting Point

This volume explores the interface between digitalization, multilingualism, language, and migration to provide new, in-depth insights into both the added value as well as the limitations of digitalization. It also aims to better understand the impact of digital transformations in the Humanities. The focus is on the Spanishspeaking world, which has a high presence in digital spaces. Based on the number of users, Spanish is currently the third most commonly used language on the Internet after English and Chinese (cf. Instituto Cervantes 2021). Spanish is –also behind English– the second most used language measured by number of users on social media networks such as Facebook, Instagram, LinkedIn, and Twitter, as well as on platforms like YouTube, Netflix, and Wikipedia (cf. Instituto Cervantes 2021). These digital spaces also play a key role in the context of migration and multilingualism. Nowadays, "more people have access to mobile media like smartphones than to fresh water toilets" (Görland/Arnold 2022, 239). Globally, the advent of mobile media has led to a profound transformation of mobility and communication. These transformations are captured in a number of fruitful avenues of research with cross-disciplinary scopes, such as Digital Communication Studies, the newly established field of Digital Migration Studies, or even Migration Linguistics. We are joining the latter which has received an increase of scholarly attention in the field of Hispanic Linguistics over the past two decades, even if the reception of some contributions, models, ideas and conclusions presented in this volume could shade light into other disciplines or domaines not mentioned.

Drawing on twelve case studies with diverse theoretical frameworks and new empirical data, this volume encompasses a vast cross-section of numerous innovative approaches and major lines of research that have been added to the scope of Multilingualism Research and Migration Linguistics due to digitalization and its impact.

Ignacio Andrés Soria, Sandra Issel-Dombert, Laura Morgenthaler García, Ruhr-University, Bochum

In section 2, we will give an overview of fundamental transformations that digitalization entails, and in section 3, we will turn to burgeoning trends with a focus on the Spanish-speaking realm.

### 2 Digitalization, Transformations, and Emerging Disciplines

Digitalization processes have produced transformations of the social experience of various kinds, including deep structural, material, and ontological changes derived from the increasingly frequent human-computer-interactions and the intensity with which digital discourses are embodied and digital language practices performed. Two aspects are of major importance:


The depth of these transformations propels us to produce a reflection aimed at observing our own instances and positions. Georgakopoulou/Spiloti (2016) and Meier-Vieracker (2022) have already pointed out the need to establish critical and ethical agendas for the disciplinary intersection of digital communication. This means we have to ask ourselves: What technology are we putting at the service of our work and the disciplines? In what ways is our work serving technological "advancements"? How do the contexts mediated by "technological progress" intervene in the generation of research questions? What aspects become priority and what is being left aside? In our particular cases, it would mean asking ourselves about the ways in which the technological axis, the spheres linked to research and the ethical dimension interrelate in the construction of the language-migration interface. This reflection would allow us to highlight the directions in which we take as an individual, a community, and within the disciplinary intersections we occupy.

We cannot ignore that the scholar is inscribed in a set of social systems and their interactions, which we see influenced by common social logics. It is not difficult to verify that digitalization has also entailed consequences of varying intensity for research practices (cf. Berry 2011). In the field of (Digital) Humanities, we identify at least three spheres: Firstly, changes resulting from the digitalization of research tools and methods, for example, in the treatment of digital corpora. Secondly, the development of studies on digital communication, such as the analysis of virtual communities. Thirdly, we talk about the changes that digitalization introduces with regards to the conditions of knowledge (re)production from the academic field, i.e., the implementation of competitive metrics such as the so-called h-index or the acceleration of the publication system.

The recent developing of the Digital Humanities as an academic field in the Spanish-speaking world is an evident consequence of the aforementioned changes. Its emergence responds to the underlying logic of organizing and categorizing the practices linked to the production of knowledge by means of criteria of similarity of thought, action, and affiliation. It supposes an attempt to intervene the codes of value, meanings and limits ascribable to the new discipline and responds to the willingness to build an epistemic community that validates frameworks of possibility and developments within it. The renaming (a few years ago we were talking about informática humanística) and emancipation from other fields points the tension generated by deciding how the discipline will be and who will participate in it. This reflection allows us to consider the publication of this volume in its political dimension, as an act of making visible what we are doing, a way of confronting stereotyped or dominant ideas, aimed at occupying spaces of description and expanding the margins of possibility, orienting ourselves towards a transit of the practices that are taking place. Following Vinck (2018, 134)

Para entender lo que son y lo que ocurre con las humanidades digitales habrá que seguir el rastro de lo que hacen los investigadores, los Estados (por ejemplo, sus políticas digitales), las empresas gigantes que viven de los rastros de nuestras actividades, los grupos sociales a los que afectan estas nuevas prácticas. Toca tener en cuenta lo que producen, como objetos, negocios, mercados, costumbres, instituciones, derivas, redes y controversias. No es suficiente una buena introducción a la informática, sino que también se debe ser capaz de decodificar los enlaces, numerosos y complejos, entre el conocimiento producido, los instrumentos y los colectivos de actores que ayudan a formar lo que será nuestra sociedad digital.

The aim of this volume is not to close any definitional space, but rather to capture some of the situated research practices that are nowadays taking place. Starting

from concrete examples ascribed to the studies of Migration Linguistics and Multilingualism, we intend to continue a discussion about how we can make Digital Humanities a common place of revisiting the social digital practices. Considering language and migration as phenomena linked to a broader social experience and taking into account that there has been almost no prior reflection within the intersection of research fields, it is necessary – and ethical – to observe the ways in which digital changes are produced to avoid the reproduction of inequality in language experiences intervened by migration conditions.

### 3 Multilingualism and Migration Linguistics in the Era of Digital Humanities: Traditions and Trends

From a linguistic point of view, a number of authors have tackled characteristics of computer-mediated communication (henceforth CMC), mobile media and linguistic practices based on digital technologies that distinguish them from other forms of communication. Without claiming to be exhaustive, and even if CMC is not homogeneous, it is widely acknowledged that CMC offers low barriers for participating in online discourses as well as greater anonymity, high visibility, reach, interactivity, the ability to communicate in real time, communication that transcends space and time, as well as the speed with which (mis)information and news can be disseminated in a targeted manner (cf. Giammatteo/ Gubitosi/Parini 2017, Meier-Vieracker 2022). It is common ground that dynamism is another fundamental characteristic. New genres, platforms and functionalities emerge, while at the same time other formats and services are displaced (e.g., the Instant Messenger ICQ). CMC and social media channels (which are characterized by interactivity), tweets, blogs and other types of mobile media contribute to the spread of multimodal content and facilitate the networking and interconnectivity of people. Interactive, participatory and user-generated web content play a key role in this context (cf. Giammatteo/Gubitosi/Parini 2017, Meier-Vieracker 2022).

The nature and characteristics of digital communication also has an impact on our work practices. The so-called digital turn affects both methods for data collection and new ways to include participatory research such as citizen science. There is also a substantial amount of software for automatic speech recognition, data annotation and data analysis. New analytical techniques are also being used for data triangulations to complement traditional approaches. One example is geotagging which provides geospatial information (cf. Kellert in this volume). Digitalization also brings a renewal of applications and resources such as corpora, which combine different sources. Especially in this area, the field is developing dynamically and is characterized by constant change. Therefore, we refrain from listing digitally available resources to avoid providing information that will quickly become outdated due to new developments. Repositories are of central importance for the long-term storage of data as well as their accessibility for the scientific community. In this context, the so-called FAIR principles (Findability, Accessibility, Interoperability, and Reusability) are the standard within Hispanic Linguistics.

A long tradition of research has addressed the question of how migrants and refugees are represented in public discourses in Spanish-speaking countries. In particular, discourses in press texts (including online editions) and television coverage are well-researched. Previous research has documented a wide range of pejorative stereotypes about migrants and refugees. For instance, they are often linked to frames of criminality, delinquency, and violence (cf. Issel-Dombert 2021, 13–15). Recent studies of xenophobic and racist discourse on social media have brought attention to strategies such as fake news, conspiracy theories, and hate speech. The concept of hate speech relates to utterances in online contexts aiming to denigrate a specific group of people based on certain characteristics of the group's identity, such as race, ethnicity, nationality, skin color, gender, sexual orientation, religious beliefs, and so forth (cf. Paz et al. 2020, 1). The analysis of hate speech also attracts attention in Spanish-speaking contexts, even if Paz et al. (2020, 7–8) emphasize a dearth of research regarding hate speech in Spanish social media channels. A growing body of literature foregrounds different forms and linguistic characteristics of hate speech and related phenomena. Lexical and (morpho)syntactical approaches as well as metaphorical studies and topos analysis –to shed light on patterns of argumentation– have become a key area for study (cf. Dufferain-Ottmann in this volume and cf. Polo-Artal in this volume). Recent studies on hate speech have also brought attention to multimodal contents such as Internet memes and to the interplay between language, images, and other semiotic codes (cf. Sánchez García in this volume). One of the main points that is made in Borah (2014, 821) on the effects of hate speech is that "participants exposed to uncivil messages are more willing to participate" in online discourses. Moreover, hate speech could lead to both less open-mindedness and polarization (cf. Borah 2014, 822). Besides Media Effects Research, automatic online hate speech detection is another emerging field of research.

Social media platforms and online discourses have also focused the attention on fundamental questions about linguistic innovations (cf. Kailich in this volume).

Another strand of research in the context of migration, multilingualism, and digitalization deals with media use and how new digital and social media lead to new communicative practices. A core feature of migration in a digital and globalized world is the concept of transnationalism that refers to the migrant's networks across the borders of nation-states. Digital media, especially messenger apps, video chats etc., play a key role in transnational networks to stay in touch and to interact with families and friends. This gives rise to new practices such as digital motherhood, as in the case of (Filipino) domestic workers who often cannot take their children with them due to restrictive migration policies in destination countries all over the world (cf. Parreñas <sup>2</sup> 2015, 94–95). In other transnational settings in superdiverse, multilingual contexts, social networks, chats etc. become a place to construct and negotiate identities (cf. Harjus in this volume, cf. Pato in this volume), to negotiate perceptions of varieties (cf. Chagas in this volume) and to create new translanguaging practices (cf. Prego Vázquez in this volume). Furthermore, digital media is used for orientation in the host countries and to overcome language barriers, e.g., via the use of translation apps or other means for language learning (cf. Murguía in this volume). "All studies share the same emphasis on the importance of mobile and social media as well as connectivity for migration and diaspora communication, both for self-presentation and for integration into the arrival society" (Görland/Arnold 2022, 242–243). In this context, migrants with an undocumented status have been less studied (cf. Morgenthaler García 2021, Morgenthaler García 2022 and Sabaté-Dalmau in this volume).

Therefore, the reader will find the eleven contributions ordered under the following thematic sections: 1) Language repertoires, translingual practices and indexicality 2) Language and identity construction 3) Digital discourses and glottopolitical approaches.

Many individuals have contributed to the completion of this volume. We would like to take this opportunity to thank everyone who offered us support and assistance. First and foremost, special thanks are owed to the RUB Research School of the Ruhr-University Bochum for the generous financial support that enabled us to realize the present volume in this form. Dr. Sarah Gemicioglu (RUB Research School) was always available to support us with words and deeds. We would like to thank Dr. Janina Konopatzki for linguistic corrections. All remaining errors are ours. Special recognition goes to Maxim Karagodin and the publishing house De Gruyter for their competent assistance at all stages of our project. Last but not least, we thank all our colleagues from various universities in Austria, Germany, Argentina and Spain for their contributions to this volume.

Bochum, September 2022

### References


Section 1: Language Repertoires, Translingual Practices and Indexicality

### Gabriela Prego Vázquez A TranslinguisTIC Safe Space

Translinguistic Creativity and Speakerhood of Migrant Young People in a Digital Epistemic Community

Resumen: El objetivo de esta investigación es analizar cómo los jóvenes migrantes gestionan su repertorio plurilingüe en la comunicación digital, centrándose en particular en el papel performativo de las prácticas de translenguaje. Para ello, hemos estudiado un chat de grupo de WhatsApp cuyos miembros son jóvenes de origen marroquí residentes en Galicia, España. El chat, Dud@sxlingu@s, se creó como una comunidad epistémica ad hoc (Estalella/Sánchez Criado 2018) incluida en un proyecto de investigación-acción participativa llevado a cabo en centros de enseñanza secundaria de la localidad de Arteixo (A Coruña. Galicia). El análisis de los datos recogidos revela que este espacio digital es un "espacio seguro" de socialización lingüística entre iguales (Puigdevall/Pujolar/Colombo 2021) en el que los jóvenes pueden desplegar su creatividad translingüística. Es, por tanto, un contexto idóneo para considerar las prácticas comunicativas digitales de los jóvenes con repertorios plurilingües desde una perspectiva émica. Los resultados de nuestra investigación muestran cómo las prácticas comunicativas híbridas de estos jóvenes crean nuevos significados a través de su acción discursiva e indexan nuevos modelos de habla.

### 1 Introduction

This research addresses the translinguistic creativity of migrant young people in digital spaces and the extent to which these contexts are safe spaces where young people can create and deploy their own communicative practices. The aim is to

Note: This research has been conducted within the framework of the R&D&I project entitled Espacios de transformación sociolingüistica en el contexto educativo gallego: agencia de los hablantes, repertorios multilingües y prácticas (meta)comunicativas. (Spaces for sociolinguistic transformation within the Galician educational context: speakers' agency, multilingual repertoires and (meta) communicative practices) (National R&D&I Plan/Spanish Ministry of Science, Innovation and Universities – State Research Agency PID2019-105676RB-C44/AEI/10.13039/501100011033/Universidade de Santiago de Compostela)

Gabriela Prego Vázquez, Universidade de Santiago de Compostela

analyse how they manage their plurilingual repertoire in digital communication, focusing in particular on the performative role of translanguaging practices. In order to do so, we have analysed a WhatsApp group chat whose members are young people of Moroccan origin resident in Galicia, Spain. The chat, Dud@sxlingu@s, was created as an ad hoc epistemic community (Estalella/Sánchez Criado 2018) included in a participative action-research project conducted in secondary schools in the town of Arteixo (A Coruña. Galicia).

The analysis of the data collected reveals that this digital space is a "safe space" for linguistic socialisation among peers (Puigdevall/Pujolar/Colombo 2021) in which young people can deploy their translinguistic creativity. It is therefore an ideal context in which to consider the digital communicative practices of young people with plurilingual repertoires from an emic perspective (Pennycook/ Otsuji 2016). The results of our research show how the hybrid communicative practices of these young people create new meanings through their discursive action and index new models of speakerhood.

The research was conducted in Arteixo, a municipality lying in north-west Galicia on the boundary with the city of A Coruña that has a population of 32,894 (data from the Spanish Statistics Agency (INE in its Spanish initials) corresponding to 01/01/ 2021). Until the 1950s, Arteixo was a rural municipality characterised by considerable waves of migratory flows to South America. From the 1960s onwards, the destinations of these flows shifted from South American countries to Europe, in particular the UK, Switzerland, Germany or France. Following the creation of an industrial estate during the 1970s, the municipality underwent a process of rapid industrialisation, accompanied by the onset of a process of disordered urban planning. The result was a new rurban fabric combining industry and subsistence farming. Parallel to this transformation process was the castilianisation of the municipality and the introduction of the standard variety of Galicia in schools and institutions. As a result, the 1980s and 1990s saw a polycentric and hierarchical linguistic continuum of varieties of Spanish and Galician in which the standard varieties of these languages occupied institutional spaces and enjoyed a certain prestige and recognition, whilst speakers of the local varieties of Galician and hybrid Galician/Spanish experienced increasing stigmatisation (Prego Vázquez 2000).

Industrialisation attracted new residents to the municipality and brought about changes to its migratory dynamics: since the late 1990s, Arteixo has received migrants of varying origins (Prego/Zas in press). Indeed, according to Spanish Statistics Agency data (INE in its Spanish initials), there are currently 1,898 people of foreign nationality resident in the municipality. However, this figure is merely "the tip of the iceberg", as it does not include those persons who are not registered on the municipal census or the young people born in Galicia to families of foreign origin. Indeed, according to the results of the sociolinguistic survey ESAVEL (Prego/ Rodríguez/Zas in press) we conducted in secondary education centres in the municipality, the percentage of students of allochthonous origin stands at around 20%1 (including young people born in Galicia to foreign families).

The results of our survey (Prego/Rodríguez/Zas in press) show that the young respondents born outside Spain come from South American countries (5% of the total),<sup>2</sup> Europe (3%) and Africa (2%). The proportion of people of Asian origin is low, accounting for a mere 0.4% of the total number of pupils surveyed. The majority of the pupils of allochthonous origin are from Venezuela (2.1%) and Morocco (1.7%). It was found that pupils born in Morocco comprise the largest group of speakers of a language other than Spanish. If we consider family origin rather than just place of birth, the data collected from our survey indicate that the percentage of young people whose families are from Morocco is more than 4% of the total.

A further impact of migration is the transformation of the community's traditional bilingual repertoire, which consisted of varieties of Spanish and Galician. Today it is intertwined with additional languages and varieties. Our quantitative and qualitative research identified the following languages and varieties (Prego/ Rodríguez/Zas in press): different varieties of mainland Spanish, as well as varieties of Spanish spoken in the Canary Islands or America (Uruguay, Argentina, Venezuela, Cuba, the Dominican Republic, among others), varieties of Galician, German, Amazigh/Tamazight, Standard Arabic, Bable, Darija (Moroccan variety) Darija (Algerian variety), Caló, Catalan (also Mallorcan and Valencian varieties), Chinese, Basque, French, Hassāniyy, Dutch, Guaraní, English, Italian, Jordanian, Spanish sign language, Norwegian, Polish, Portuguese (Iberian Peninsula and Brazilian varieties), Rumanian, Russian, Wolof and Ukrainian.

This article addresses the digital communicative practices of young people of Moroccan origin. The reasons for this focus is not only the relatively large size of this group, but also because of their consolidated nature in the municipality and the key role these young people play in managing the new translocal repertoire, which is of particular relevance in their digital communications, as discussed in earlier publications (Moustaoui/Prego/Zas 2019a and 2019b). The specific research questions are as follows:

The resulting survey was conducted in two of the municipality's compulsory secondary education centres. The secondary school in the town centre had 826 pupils, whilst in the school located in the north of the municipality, the number of students enrolled in the year of the survey stood at 300. The majority of young people in the borough attend these two schools.

A large proportion of the young people from countries such as Argentina, Uruguay, Venezuela or Brazil are third or fourth generation descendants of Galician migrants.

Which languages/varieties are included in the repertoire mobilised by young people of Moroccan origin in digital communication?

What opportunities do WhatsApp chats offer for the co-construction of an epistemic space?

To what extent can they be considered safe spaces for the deployment of young people's translinguistic creativity?

How do these practices contribute to the creation of new translocal spaces? In the light of the data analysed, should the traditional concept of speakerhood be reviewed?

Section 2 describes the methodology employed, whilst sections 3 and 4 present the results of the quantitative and qualitative analysis and the final section draws a series of conclusions.

### 2 Methodology

The research was conducted in secondary education centres in Arteixo using multisited ethnography within the framework of participative action research, incorporating collaborative research (Lassiter 2005; Leyva/Speed 2008; Unamuno 2019) in order to involve pupils and teaching staff in the role of co-investigators. Critical Language Awareness (Garcia 2008) provides the structure for the ethnographic process as well as the core concept underlying the design of metalinguistic awareness activities carried out by investigators from the University of Santiago de Compostela and their co-investigators, namely secondary school pupils and teaching staff. This participative ethnography is particularly effective in generating data of a widely diverse nature: oral samples, written academic work, spontaneous interactions, examples of linguistic landscapes as well as digital communication, which is the focus of this article. In turn, the ethnography results proved essential in the design of a sociolinguistic survey, ESAVEL (Prego/Rodríguez/Zas in press), which was carried out in all the classrooms of the municipality's secondary education centres and which includes questions on digital communication.

The communicative practices analysed in this article come from the WhatsApp chat Dud@sxlingu@s. Dud@sxlingu@s was started by two Spanish/Galician bilingual investigators from the University of Santiago de Compostela (Luz Zas and Gabriela Prego) and eight secondary education pupils of Moroccan origin. Two of the pupils (a boy and a girl) had only recently arrived in Galicia, whilst the rest had completed their primary education in the municipality. They were later joined by an investigator of Moroccan origin and speaker of several varieties of Arabic, French and Spanish in order to encourage translinguistic interactions.

The chat was created as an epistemic community (Estalella/Sánchez Criado 2018) in order to complement or suggest research tasks and activities in the field of Critical Linguistic Awareness based on the linguistic landscape, linguistic biographies, reflections on linguistic usage, toponymy or riddles and brainteasers related to the various languages, among others. The young people perform the role of experts in this chat, and they are actively encouraged to use their own forms of communication. The sensation of empowerment and the realisation their practices are never subject to judgement converts this chat into a safe space in which to deploy their translanguaging practices (Estalella/Sánchez Criado 2018), as we will analyse in section 5.

The analysis falls within the framework of Critical Sociolinguistic Ethnography. Specifically, we employ a multimethod analysis that integrates a "translanguaging and moment analysis" perspective (Li Wei 2018a). This is particularly suited to tackling linguistic hybridity phenomena associated with mobility, and other methodological approaches present in Globalisation Sociolinguistics (Blommaert 2010), such as the analysis of sociolinguistic scales (Wallerstein 2000; Blommaert et al. 2015) and critical literacy (Casany 2005; Zavala 2004). Other key concepts for the interpretation of the results include agency, in line with Bourdieu's theories (Bourdieu 1977). The analysis also considers the statistical data drawn from our sociolinguistic Survey ESAVEL, which is discussed in section 3.

A final point for consideration in this section is that the original orthography used by the participants has been maintained. As a result, accents have occasionally been omitted, and only final exclamation and question marks appear.

### 3 The Linguistic Repertoire of Young People of Moroccan Origin in Digital Communication: ESAVEL Survey Data

The ESAVEL sociolinguistic survey (Prego/Rodríguez/Zas in press)<sup>3</sup> includes a question regarding young Moroccans' linguistic usage on the internet and social media. The results of this survey relating to the community of pupils of Moroccan origin show that these young people claim to interact in different languages for

The survey questionnaire can be accessed at: http://avel.cesga.es/wp-content/uploads/2019/10/ Cuestionario-socioling%C3%BC%C3%ADstico\_galego\_2018\_19\_rev.pdf.

their digital communication practices. Spanish is the most commonly used language (only 3.9% state that they never used it on social media), followed by Darija (53% claim to use this language: 15.7% sometimes; 15.7% almost always; and 21.6% always).

These languages are followed by English and French. Usage of these two languages is fairly equal. Forty-three point two percent claim to use French: 31.4% sometimes; 5.9% always; and 5.9% almost always). This compares with 44.2% that declare that they use English: 26.5% sometimes; 7.8% always; and 9.9% almost always. The use of French is a distinguishing characteristic that sets this group apart from other young people, who only very rarely use this language on social media. The importance of this language in their country of origin, Morocco, is the reason for its significant usage in their communicative practices.

The linguistic minorisation of Galician and Tamazight is reflected in the use of these languages on social media. Although Galician is present in the school curriculum and is the language of Galicia, only 37.3% state that they use it: 29.3% sometimes; 2% always; and 5.9% almost always. The survey data also show that for the majority of autochthonous and allochthonous young people, their use of the Galician language is limited to the academic sphere. In turn, the presence of Tamazight is minimal: a mere 2% of young people of Moroccan origin state that they occasionally use it on social media. Furthermore, although the majority of young people in the group of Moroccan come from Berber regions where Tamazight is the autochthonous language, the breakdown of the intergenerational transmission of this language has led to its minimal presence in the communicative repertoire of these young people. Finally, 29.4% claim to use Standard Arabic (13.7% sometimes; 3.9% almost always; and 11.8% always). Worthy of note is that it ranks below Darija. The fact that this language is employed in formal contexts, and that not all young people are competent in this variety prevents its regular use by these adolescents.

According to the results of the analysis, the communicative repertoire of young people of Moroccan origin on social media reflects the consolidation of Darija as their heritage language and the incorporation of local languages – Spanish and Galicia – acquired by means of the socialisation and schooling processes in their host community (Moustaoui/Prego/Zas in press). In addition, we observed that these young people also claim to mix resources from several languages and various forms of communication. In fact, both in the surveys and the linguistic biographies collected in the ethnographic research, a number of these young people claim to use "el mezclado" – a hybrid variety comprising Darija, Spanish, Galicia, French and the occasional word in English. Practices of this type are of particular relevance in their digital communications. The following sections include a qualitative microanalysis of their (meta) communicative practices in the Dud@sxlingu@s WhatsApp chat, in order to determine whether the repertoire they claim to use in the survey really is present in their discursive actions, as well as the presence and function of their hybrid practices, previously addressed in relation to agency in Moustaoui/Prego/Zas (2019a, 2019b).

### 4 The Dud@sxlingu@s WhatsApp Chat as an Epistemic Community: Pupils' Digital Knowledge as Expert Co-Investigators

The multi-site ethnographic process (Prego Vázquez in press), is a vital part of participative action research. It forms an epistemic community (Estalella/Sánchez Criado 2018) in which pupils acquire the role of co-investigators, thereby breaking the asymmetry between investigators and pupils. Instead, they move together beyond the educational space in order to collaboratively explore and research other relevant environments for the socialisation of young people such as the family, local community, translocal space or social media, among others.

As stated previously, the Dud@sxlingu@s WhatsApp chat is designed as an epistemic community that is built up during a process consisting of metalinguistic awareness activities. The investigators contributed to the creation of this space for peers in order to underpin the pupils' role as co-investigators, considering them as experts throughout the process. This allowed for the creation of an interaction context that enabled the young people to share their "know-how" and encouraged the exchange and co-construction of knowledge among peers. The data analysis shows that the young people do in fact interactionally develop the role of experts as their participation in the chat evolves and progresses. The following episode between M, a young female secondary education pupil and L, an investigator from the University of Santiago de Compostela (USC), shows how M adopts the position of expert and L contributes to the co-construction of this role through the interaction. This type of exchange occurs within what Goffman (1959) refers to as "front regions", in other words, scenarios that the participants recognise as their principal settings (Blommaert/De Fina 2017). Spanish is the principal language employed in this space, although, as we will see in the following episode, Galician is also used (line 10) as well as Darija (24, 34 and 36) in order to introduce the examples:

#### Caso 1: Expesiones para todo


#### Case 1: Expressions all round


In this metadiscursive episode, the participants reflect on linguistic expressions in Galician, Spanish and Arabic. As an expert co-investigator, it is M who explains her linguistic usage to the USC investigators. The analysis of this episode shows how her role as an expert progresses with each turn. Essentially, it can be divided into three parts: in the first part (lines 1–21), the questions posed by the USC investigator encourage M to display her expertise in the topic; in the second part (lines 22–45 and 47,48, 49 and 50), M controls the interaction and uses her own initiative in order to change topic and give an explanation for the expressions in Arabic; and finally, the thread begun by M on line 46 and that continues on lines 51–55 with a reflection on the mix of languages used in Morocco.

As mentioned above, in the first part (lines 1–21), the USC investigator asks questions aimed at encouraging M to display her expertise in the topic. M masters this question and L expresses her interest and lack of knowledge throughout. These roles are forged and reproduced within the interactional organisation of the sequence. In this sense, it can be seen how the first part is dominated by interactive question-answer pairs and assertion-(dis)agreement pairs:


Throughout the episode, L introduces the first part of the interactive questionanswer pairs, as can be seen in lines 1, 6, 10 and 12; the first pair-part of the assertion-(dis)agreement in lines 17, 19 and 20; and finally, she also introduces the first backchannel (line 15) in the secondary turn-taking system in order to express her assent and understanding of the new information M provides. In turn, and as we can observe throughout the interaction, M assumes the role of expert, incorporating explanations for L's questions in her answers, as can be seen in lines 3, 5, 7, 8, 9, 11, 13, 14, 18 and 21. They are metadiscursive reflections on her usage which, within the interactional frame, acquire the value of relevant expert knowledge. As a result, by deploying the first parts of the adjacency pairs, L creates an interpretative frame that positions M as an expert, a role that is assumed and coconstructed by M, as can be seen in her participation in the second-pair parts. This sequence of adjacency pairs is a mechanism of intersubjectivity (Schegloff/ Sacks 1984) that is used to create a frame of mutual understanding and cooperation (Goffman 1959). Furthermore, the interactions construct M's role as an expert. In the following parts, this role is reinforced through other interactional mechanisms.

In the second part, specifically lines 22–45, 47, 48 and 49–50, M takes over the choice of topics in the interaction, changing the topic in order to introduce, on her own initiative, an explanation for the expressions in Arabic.


It should be noted that in this sequence, question-answer pairs are not so predominant and for most of the interaction L's interventions are limited to uttering backchannels in the secondary turn-taking system such as for sure, ah, for sure or of course in lines 13, 33, 35, 40, 47, 48 or 50, expressing her acknowledgement and surprise at the new information provided by M's explanations. These particle expressions and others in the secondary channel not only reproduce L's role as listener in the interaction, but also underpin M's role as an expert. By using these continuation expressions, L is not engaging in turn competition; nor does she make many replies, but instead accepts and values the information provided by M. In this sense, L's interactional engagement contributes to the co-construction of M's role as an expert. Within this role, M contributes explanations regarding expressions in Arabic that project her authority and legitimacy. Moreover, M introduces direct other-initiated repairs (Schegloff/Jefferson/Sacks 1977) in order to correct L's statements on lines 30 and 47: "But not in Arabic" or "No, not at all. These otherinitiated repairs do not include any form of mitigation, as M's authority in this topic allows for the rapid and forceful correction in order to assess the appropriateness of her interlocutor's contributions. In this sense, the conversational repairs show that M assumes and consolidates her role as expert during the interaction. As can be observed, the various interactional actions co-construct an epistemic space in which the young woman shares here knowledge (which does not stem from the academic sphere) and favours the display and introduction of her personal knowledge of her practice community, as shown on lines 24, 34, and 36.


In addition to including her explanation, reflections and translations, on lines 24, 34 and 36, she also introduces Arabizi in a completely natural manner. This is an alternative writing system that uses the Latin alphabet and numbers in order to transcribe Arabic in digital environments. Arabizi is a digital practice of translinguistic creativity that is generated outside academic orthographic practice. This episode therefore illustrates how the pupils' active and collaborative participation in our research effectively gets them involved and enables them to contribute their knowledge. As a result of their analysis, we are able to reconstruct the emic perspective of speakerhood.

Finally, the thread initiated by M on line 46 and that continues on lines 51–55 continues the frame that is being co-constructed by the participants and in which M refers to the presence of French in Morocco:

29/12/20 16:04 – M: I have to say that people talk like that a lot in Morocco (,..)

29/12/20 16:05 – M: But I prefer to speak in Arabic so I don't forget it

29/12/20 16:05 – L: Of course

29/12/20 16:05 – L: Languages are easy to forget!

29/12/20 16:05 – M: It's just a question of preference, some people prefer to speak Arabic and then add things in French

29/12/20 16:05 – M: And others just Arabic

In short, the analysis of this representative fragment of the corpus enables us to observe how sequential organisation creates this epistemic space through interaction; in other words, it is oriented towards establishing the chat as a de facto space for collaboration, cooperation and the creation of new knowledge. In this sense, both the data collection method, based on the creation of epistemic communities, and the linguistic awareness activities that take place therein, favour more effective forms of participation (Konrad 2012) that allow us to reconstruct the participants' emic vision.

# 5 Translinguistic Creativity and New Speakerhood Models: The Chat as a Safe Space for Bottom-up Linguistic Innovation

Our analysis of the data included in the Dud@sxlingu@s WhatsApp chat also reveals that in addition to the front regions in which pupils and investigators undertake epistemic co-construction, the pupils occasionally use the chat as a channel for informal interaction among themselves. These types of interactions have shown us that the chat has become a safe space for the participants; in other words, a space in which they feel sufficiently confident to show their repertoires, which, as the following episode will reveal, challenge the boundaries between the named languages. Rather than taking place on the main stage, these exchanges take place in preliminary spaces, or as Goffman (1959) refers to them, backstage. In these instances, the pupils use the mezclado – or mix of languages and Arabizi, linguistic resources that lie beyond the investigators' scope and that they use to discuss personal issues. The investigators are unable to follow these interactions, which also prevents them from participating. In this sense, the distribution of linguistic resources – namely Spanish as the predominant language on frontal stages, and the mix backstage – indexes how they are scaled in the various spaces (Blommaert 2010; Prego in press) and sheds light on their role in managing questions of identity and translocal spaces. Following Blommaert and De Fina (2017), addressing the co-existence of both spaces is essential for the analysis of identity negotiation and special configuration.

#### Case 2: Xokran por non parle with me zin


This episode opens a new frame in the chat that marks a shift away from the epistemic frames of cooperation co-constructed between the pupils and investigators discussed in the previous section. The pupils spontaneously initiate a conversation among themselves without any prior exchange with the investigators. As stated above, the start of this exchange implies a change of frame and footing by the pupils. In this episode, rather than co-investigators, they are two friends chatting about a personal topic of interest to them both. In interactional terms, this change of frame is marked by the following resources: the use of Arabizi, mezclado and the discussion of a personal matter. These three resources are contextualisation cues (Gumperz 1982), that index the "preliminary" space that Goffman (1959) refers to as a "backstage" space that is in direct opposition to the front region or stage in which the epistemic interactions take place.

As discussed in Moustaoui/Prego/Zas (2019a; 2019b), Arabizi is an alternative written code used widely to write Arabic in digital technology contexts and on social media. It emerged as a solution to the inability of online resources and technological tools to support Arabic script. It consists of using the Latin alphabet and including changes in the direction of the script and numbers to replace certain graphemes.

As with Arabizi, the mezclado or "mix" (the term used by these young people to refer to the variety they use, which consists of mixing linguistic resources from their repertoire) is deployed in contexts of socialisation among peers. In this episode, it can be seen how resources from Darija are mixed with others from Spanish (lines 13, 14, 17 and 22), French (lines 2, 14, 15, 16 and 18) or English (line 14). A clear example of the use of mezclado can be found on line 14: Xokran por non parle with me zin (Thanks for not telling me anything). This phenomenon, known as "translanguaging" (García 2009; Wei 2011; 2018b), consists of mixing the linguistic resources of their plurilingual repertoire and is typical in multilingual scenarios. Arabizi and mezclado challenge standard hegemonic practices.

Both Arabizi and mezclado operate together as a "we-code" (Gumperz 1982) which, together with the topic chosen by the participants, sets the "personal" tone of the interaction in these episodes. As stated above, the use of these resources opens up an interactional space backstage, indexing what Bhabha (1994) refers to as a "third space". Following Flores/García (2013) and García/Li (2014), when generated by translanguaging practices, this third space has a tremendous capacity for transformation, as it provides an opportunity for the creation of new identities, values and practices. As discussed in Moustaoui/Prego/Zas (2019a; 2019b), it is a space of resistance against the monolingual norms of schools, institutions and also political, geographical, territorial and linguistic boundaries of the community or nation-state they live in.

Speakers co-construct these communicative practices in the digital space, translocal and hybrid spaces, hybrid identities and new models of speakerhood. They are of vital importance for plurilingual individuals as they simultaneously mobilise resources from different "acknowledged and named" languages in a single utterance. They are practices with the capacity to transform, acting as "spaces for cultural translation" (Bhabha, 1994) and the "generation of new forms of communication".

In this sense, the data show the extent to which this digital space provides a safe space in which speakers can feel confident that their "status as mezclado speakers" will not be questioned. The results of the analysis show how young people mobilise various multilingual resources and thereby redefine new models of speakerhood outside standard public discourse.

### 6 Conclusions

Superdiversity in sociolinguistic contexts enables participants to seek and introduce new forms of communication and ways of being in the world (Garcia/Li Wei 2014). As this chapter has shown, the digital space is particularly suited to translinguistic creativity and the generation of new social meanings and resources.

The results of our research have shown that young people of Moroccan origin in Galicia mobilise a dynamic plurilingual repertoire in the digital space. As a case in point, the ESAVEL sociolinguistic survey revealed that these young people claim to use a number of languages for their interaction in digital communication practices: French, Darija, Spanish, English, and, albeit to a lesser extent, Galician and Tamazight. In addition, participative ethnography has provided us with and insight into their metalinguistic reflections and also their de-territorialised translinguistic repertoire, which is linked to their personal translocal trajectories.

The analysis shows how young people of migrant origin mobilise alternative types of script and translanguaging practices for their digital communications outside the confines of standard public discourse. The choice of these resources indexes a number of interactional frames that sign spatial and temporary dimensions and that project the complex network of scales from which new linguistic forms are legitimised and/or delegitimised. In this sense, the results show how these repertoires can open up new spaces of linguistic resistance and transformations in which new meanings are negotiated and created. These new spaces challenge, from the margins, the spaces in which languages are silenced.

Furthermore, the results question the traditional definition of speakerhood and pave the way for a new ground-breaking and more creative model of speakerhood, moving beyond standard public discourse and the monoglossic ideologies that dominate schools and institutional spaces. This model would represent a break away from institutionalised linguistic standards, crossing the boundaries drawn among named languages. In this sense, the results of the analysis point to the need to deconstruct the traditional notion of the speaker, giving rise to a new concept of speakerhood that moves beyond the monoglossic language ideology underlying tags such as the monolingual speaker, bilingual speaker or second language speaker (Martín Rojo 2020; Ramallo 2020).

### References

Bhabha, Homi K., The location of culture, London/New York, Routledge, 1994.

Blommaert, Jan, The sociolinguistics of globalization, Cambridge, Cambridge University Press, 2010.

Blommaert, Jan. Chronotopes, Scales, and Complexity in the Study of Language in Society, Annual Review of Anthropology, 44 (2015), 105–116.


perspectives, 2017, 1–15. Washington DC: Georgetown University Press.


Bourdieu, Pierre., Outline of Theory of Practice. Cambridge: Cambridge University Press, 1977.

Cassany, Daniel., Investigaciones y propuestas sobre literacidad actual: multiliteracidad, internet y criticidad, in Leer y escribir en un mundo cambiante. Congreso Cátedra UNESCO para la lectura y escritura, Universidad de Chile, http://www2.udec.cl/catedraunesco/05CASSANY.pdf, 2005. [last accessed January 10th, 2023]


Pennycook, Alastair/ Otsuji, Emi, Metrolingualism: Language in the city, Oxon, Routledge, 2015.

Puigdevall, Maite/ Pujolar Joan/Colombo, Alba, Espacios de adopción del catalán, una aproximación etnográfica a las mudas lingüísticas en Cataluña, in: Fernando Ramallo/Estíbaliz Amorrortu/Maite Puigdevall (edd.), Neohablantes de lenguas minorizadas en el Estado Español. Frankfurt, Vervuert, 2021, 11–130.


Wei, Li, Linguistic (super) Diversity, Post-multilingualism and Translanguaging Moments, in: Angela Creese and Adrian Blackledge (edd.), The Routledge Handbook of Language and Superdiversity, London, Routledge, 2018a,16–29.

Wei, Li, Translanguaging as a Practical Theory of Language, Applied Linguistics, 39, (2018b), 9–30.


### Maria Sabaté-Dalmau Migrants' ICT-Mediated Communication Strategies and Subversive Multilingual Practices

#### Insights from a Critical Network Ethnography

Resumen: Desde un enfoque sociolingüístico crítico de una etnografía en red, este capítulo analiza las estrategias de comunicación mediadas por las TIC y las prácticas multilingües de 20 personas migrantes desempleadas que se reunían a diario en un locutorio de Barcelona, con el objetivo de explorar qué pueden revelar sobre sus procesos de (auto)-incorporación sociolingüística en la era digital. Los datos incluyen observación participante, entrevistas, conversaciones y materiales visuales recopilados entre 2007 y 2009. Para analizar las oportunidades y limitaciones de los migrantes para acceder a las tecnologías de la información y la comunicación (TIC), primero me centro en el tándem de gobernanza tecnopolítica conformado por el gobierno español y el sector de las telecomunicaciones español (con 30 operadores), quienes, actuando como un bloque regulatorio unitario, se aliaron para establecer un régimen de control de la ciudadanía (en particular poblaciones indocumentadas) a través de un sistema de vigilancia de datos mediado vía TIC. Este régimen exigía la acreditación de la residencia legal para acceder a las tarjetas de telefonía móvil y oficializaba una política monolingüe en español en los servicios de las administraciones públicas y el comercio TIC. Seguidamente, investigo cómo los informantes eluden este régimen en espacios alternativos de socialización como locutorios, donde los migrantes no registrados movilizan una serie de tácticas "fuera del radar" para acceder a las TIC (por ejemplo, el uso de la "tarjeta SIM doble"), que demuestra que el control tecnopolítico se resiste y se transgrede (aunque no se pone en peligro) en masa, "de abajo hacia arriba". En segundo lugar, muestro que los migrantes invierten en el español estándar como el idioma legítimo de incorporación social, pero rutinariamente muestran prácticas multilingües subversivas que desafían la normatividad monolingüe. Estas incluyen la (auto)-provisión de recursos de tecnoalfabetización para gestionar servicios administrativos (por ejemplo, renovaciones de visas) y de servicios de TIC en sus propios repertorios translingüísticos, incluido el español transcódico y las lenguas locales/globales silenciadas (como el urdu y el marroquí). Esto muestra la relevancia de investigar la lingüística de la migración a través de estudios de caso, en espacios inexplorados de "no ciudadanos", lo que

#### Maria Sabaté-Dalmau, Universitat de Lleida

puede contribuir a problematizar la desigualdad social (y la resistencia a ella) en base a la lengua, en contextos de precariedad, arrojando luz sobre quiénes se consideran migrantes merecedores de ciudadanía bajo las lógicas de gobernanza antimigrante de la sociedad en red.

# 1 Introduction: ICT-mediated Citizenship Governance in the Rise of the Network Society

With the rise of the network society (Castells 2000 [1996]), Information and Communication Technology (ICT) have become a key element for the social organisation of individuals, groups and (supra)-national institutions (Castells 2009), for three reasons. Firstly, because they are a new source of economic power for the global marketplace ("the new petroleum"; Harvey 2005, 159). Secondly, because they are a tool for political power whereby recent technologised governments keep securing the monopoly of sovereignty and citizenship control (including access to demographic mobility) within their territorial borders (Diminescu 2008): a governance apparatus to police access to "national" citizenship (Inda 2006). Thirdly, because they are the crucial networking-power resource (the "social glue," Vertovec 2009, 54) for migrants to manage transnational-family life and to access resources such as health, education, employment or housing (Castells et al. 2007).

Under these circumstances, migration in the era of the network society is regimented through "neoliberal governance" (Martín Rojo/Del Percio 2019, 3). This is a mode of governance consisting of "win-win" nation-state/marketplace regulatory partnerships of advanced liberal societies which have displaced welfarism in favour of clientelism, putting political practice exerted upon individuals at the service of the new economy. Concerning migration, these governance tandems establish citizenship regimes which aim at both registering, controlling and making profit from them. One of the most powerful governance blocks is the nationstate/ICT-sector partnership, whereby nation-states concede governance authority to ICT corporations (e.g., with legal coverage and funding to target new migrant groups), and these simultaneously allow governments to make use of their ICT infrastructure in order to surveil "illegals" <sup>1</sup> and to exert demographic mobility control upon them (Sabaté-Dalmau 2021a).

Inverted commas are used to mark emic categories such as "illegal" or "ethnic".

#### 1.1 Undocumented Migrants: The New Target of the Spanish Nation-State and the Spanish Telecommunications Sector

The Spanish nation-state/telecommunications sector regulatory tandem is an illustrative case of the ways in which neoliberal governance units, as publico-commercial partnerships, exert power upon migrant populations in an exclusionary manner, by "disconnecting" the (non-profitable) undocumented and imposing a monolingual language regime in Spanish as the rule-of-law language and as the language of commerce, as detailed below.

Following the European Directives for data storage (detailed in Sabaté i Dalmau 2014), the Spanish Cabinet, who was then under the pressure of being the country with the highest estimated percentage of "illegals" in Europe (Sànchez 2008), passed a telecommunications bill called Ley 25/2007 (October, 2006). This regulation was established by the Spanish Ministry of Economy, upon which telecommunications depend, but it was actually designed in coordination with two ministries of citizenship governance: the Ministry of Defense and the Ministry of Interior. In theory, Ley 25/ 2007 aimed at the modernisation (i.e., technologisation) of public administration through the establishment of an electronic bureaucratic system ("e-governance") geared towards facilitating the access of "migrant citizens" to the Spanish administration (Plan Avanza 2012). However, in practical terms, it implied the establishment of a "national-security" dataveillance system to "fight against terrorism" via ICT registration, as, in the words of the then Minister of the Interior Pérez Rubalcaba who presented it, SIM cards are being sold "without control" (El País 2009). This had a direct impact on transnational migrants, because the law obliged operators to register the identity of all clients via personal identification (i.e., proof of legal residency) by 2009 (BOE 2007), which led to the disconnection of 15 million SIM cards of undocumented individuals who did not provide a (foreign) identification number. The disconnection from the ICT network was conducted by the multinational Telefónica (formerly a nation-state monopoly), who then became a privileged actor (the "right hand") of the government. Moreover, simultaneously, Ley 25/2007 was the basis for operationalising the new compulsory e-administrative services system in a platform that was managed solely in and through Spanish. This was so because, despite its multilingual presentation (the government claimed to offer the official bureaucracy documents in Spanish, Catalan, Basque, Galician and English), the administrative portal provided the 26 procedures necessary to obtain legality status (concerning temporary residence, dual citizenship, work permits, etc.) only in the nation-state language (Sabaté-Dalmau 2021b).

In turn, Ley 25/2007 granted more politico-economic power to the Spanish telecommunications sector, then one of the biggest market segments of the Spanish economy (CMT 2011, FOBSIC/Idescat 2010), which envisioned migrant populations

as the new lucrative niche to overcome their "market saturation". Migrants became "the most important emerging market" (Paloma Vargas, from Orange, in CiberP@ís 2007, 1), as it was estimated that they then spent 40% more on ICT and initiated twice as many calls and Internet connections than "local" populations (Ros/Boso 2010). More specifically, the law allowed the government to provide the ICT oligopoly (comprised of Telefónica, Vodafone and Orange; Guillén 2005) with newer infrastructural support and funding to expand their commercial activities targeting migrant clients. This was seen, for example, in the nation-state/telecommunications sector's organisation of the first "Madrid Products and Services for the Immigrant Fair" in 2008, one of the biggest business events in Madrid, where, on the basis of "integration" (the Fair was called "Integra Madrid") the sector sold their ICT products to migrant-regulated associations in coordination with the Spanish Secretary of State for Immigration and Emigration (see IFEMA 2008).

With Ley 25/2007, the government also provided the sector with legal coverage to operate, de facto, in Spanish, under the neoliberal tenets of "linguistic economicism" (Heller/Duchêne 2012). This is seen in statements by the language economists of the oligopoly, who argued that "linguistic unity [in Spanish] is a precondition for a competitive marketplace" (García Delgado/Alonso/Jiménez 2007, 86, from Telefónica). The way in which this monolingual Spanish policy to access ICT services in Spain was operationalised was as follows: In 2009, only 14 companies (out of the 30 that were recorded then) offered services in English; minority languages in Spain were only offered by the three operators based in minority-language areas (Catalonia, the Basque Country and Galicia), and allochthonous codes were totally absent, except for Arabic and Romanian, offered very occasionally during some commercial campaigns (Sabaté i Dalmau 2012).

#### 1.2 The Emergence of Migrant Spaces for Subversive ICT Access: "Ethnic" Call Shop Shelters

Under the political-economic and sociolinguistic conditions of neoliberal governance described above, a new sort of alternative, migrant-tailored transnational space "with an unrivalled force in Europe" (Ros, cited in BBC Mundo 2008) emerged in Spain which served as a grassroots social-incorporation institution for the undocumented: the "ethnic" call shops which, in the metropolitan area of Barcelona, increased by 10% between 2001 and 2009 (Sabaté i Dalmau 2014, 64–66). The success of these small ventures may be explained by the fact that migrants, both owners and clients/users, have turned them into socialisation "shelters" whereby the undocumented can access a multiplicity of citizenship resources (e.g. washroom facilities, rooms for rent, food, CV writing, advice and help on addressing legality issues), including ICT and technoliteracy resources (technology and literacy skills; Area Moreira/Gros Salvat/Marzal García-Quismondo 2008) to navigate e-governance, in their own languages and through linguistic mediation and language brokering practices. In short, as we will see in the Analysis (section 3), "ethnic" call shops today offer the citizenship and language services that both the nation-state administration and the telecommunications sector have left unattended. This may explain why 40% of documented and undocumented migrants (without and with connection at home) choose a call shop as their space of transnational communication (Molins Pueyo 2006).

### 2 The Study

The research presented here, which draws on a previous bigger research project whereby I investigated undocumented migrants' incorporation practices in informal, self-organised call shops in Catalonia (Sabaté i Dalmau 2014), zooms into the particular subversive ICT-mediated communication techniques and unconventional multilingual practices whereby "illegal" populations manage to resist exclusion, immobility and socioeconomic stagnation by challenging neoliberal citizenship governance regulations concerning ICT access in Barcelona (particularly, Ley 25/2007), as detailed below.

#### 2.1 Aims and Research Questions

In this chapter, I try to contribute to the understanding of the interplays between multilingual phenomena and the affordances (opportunities and constraints) of migrants' transnational incorporation practices in their host societies during the "digital turn" (see Andrés Soria/Issel-Dombert/Morgenthaler García in this volume). I explore this interplay by means of a critical network ethnography of 20 migrants who organised their daily lives in a "Pakistani" call shop, which allows me to address the following two research questions that I focused on for this chapter:


communication practices can these provide, particularly with regards to the established hegemonic Spanish monolingualism? And what can they tell us about the linguistics of migrations, with regards to the (non)-usefulness, (in) visibilisation and (de)-legitimisation of "illegal" migrants' languages and sociolinguistic comportments in bilingual Barcelona?

#### 2.2 Context and Research Space

I locate the study in Catalonia, a bilingual society in North-Eastern Spain of 7,739,758 inhabitants, with a percentage of around 16% of foreigners (Idescat 2022), where a majority nation-state (and global) language, Spanish, on the one hand, and a historically prosecuted co-official minority language, Catalan, on the other hand, coexist in a complex manner.

I focus on a peripheral migrant-populated neighbourhood with one of the highest unemployment rates in the metropolitan area of Barcelona, categorised as an "ethnic enclave" in need of municipal social "aid" by the town council. This neighbourhood was also home to mostly Spanish-speaking "locals" who were born in other parts of Spain and moved to the industrial peripheries of Catalonia during the 1970s/1980s.<sup>2</sup> The call shop where I conducted this study belonged to a Pakistani businessperson who ran three other businesses in the area, and who did not live in the neighbourhood (he came once a day at midnight to collect the daily cash). The worker in charge of it was Naeem, a 27-year-old Pakistani with a temporary residence permit who worked between 12/14 hours a day, also during weekends. He welcomed between 61 and 156 customers a day (including the network under study), who spent about €6,000 a month on ICT there.

#### 2.3 Methods, Data and Informants

I moved to the neighbourhood and settled next to the street where the call shop was located, and then started to conduct a critical sociolinguistic "network ethnography" (Marcus 1995, Sabaté-Dalmau 2008) of the 20 most frequent clients, all of whom were acquainted with each other, for two years (between 2007–2009). This network ethnography involved shadowing informants in the socialisation spaces that were

The data was obtained through voice informed consent and it was anonymised by following all confidentiality protocols established by the Ethics Committee at the Autonomous University of Barcelona (approval file 725H). Exact percentages of the neighbourhood have not been provided in order to ensure the informants' protection. All names are pseudonyms.

meaningful and relevant for them (the call shop, the public bench in front of it, the bar next door, etc.), by means of multisited co-ethnographic journeys and "go-along" interviews (Carpiano 2009, Sabaté i Dalmau 2018a).

Apart from daily participant observation, the data also included semistructured open-ended informal interviews with 17 of the 20 participants; taperecorded naturally occurring conversations (interactional exchanges) among migrants themselves; a range of visual materials (leaflets, call cards, room-forrent advertisements, receipts, SMS, etc.), and documentary resources (relevant policies and directives). For the purposes of this article, I here include three excerpts from interviews; four fieldnote diary entries, an SMS, and two images (a receipt for remittances services and a hand-written notice).

The informants consisted of a heterogeneous group of 20 men and women aged between 27 and 62, born in Pakistan, Morocco, Rumania and different countries of Latin America (Argentina, El Salvador, the Dominican Republic, Bolivia, Guinea, Equatorial Guinea, Cuba and Brazil). Most of them were unemployed and undocumented and participated in the local informal economy. The vast majority of them had previously lived in other parts of Spain (some also in other parts of the EU) and had settled in the neighbourhood between 2002 and 2010. Their family configurations were also very diverse: middle-aged men or young women who shared a single room; extended families living together in sublet apartments; or single families with children living with Spanish individuals. Some were nonschooled and non-literate; and others had a high command of both oral and written standard Spanish, as well as of English and Darija, for instance.

### 3 Analysis: Managing Nation-State/ICT Sector's Regimes through Subversion and Resistance

#### 3.1 Subversive ICT Uses: Away from the Government's Administrative Gaze

Migrants in the call shop displayed a multiplicity of ways to access a wide range of ICT by strategising both with the Spanish government's regulation of the conditions of access to them and with the ICT marketplace's commercial dynamics. Their ways of accessing, in particular, mobile phone services were the most subversive, and they provided evidence that the established technopolitical dataveillance systems to register "illegals" were not only skirted but challenged en masse, in these call shop "shelters". In turn, their ICT uses contradicted the deeply-seated beliefs concerning migrants' communication routines that were widely mobilised in mainstream media discourse, such as macro narratives stating that migrant populations "only call to international destinations" (see Sabaté-Dalmau 2021a). I here report on three main strategies to access ICT which defied both the Spanish government's rules and the Spanish telecommunications' commercial rationalities.

#### 3.1.1 Taking Advantage of Legal Gaps: Access to SIM-Cards Without Personal Identification

In the call shop under analysis, undocumented users could access SIM cards without having to show proof of legality. This was so because they were legally purchased in packs of 10 by the Pakistani owner, who used his corporate identification number (his Certificado de Identificación Fiscal, or CIF) to register and later resell phone numbers under his venture's commercial name (here unmentioned for confidentiality reasons).

Besides, the place also offered unlocking services (the so-called móviles libres or móviles liberados), which allow for the use of more than one operator on a single mobile phone handset and with a single telephone number without having to notify any change to the company from which the first SIM card was purchased; that is, without having to provide any personal details. This service costs between €30 and €120.

Similarly, both fax and remittances services could be managed without the need to provide any personal name or address of the sender, as Naeem, the worker, would fill in all the information required online with the venture's details. This is illustrated in Figure 1, which is a receipt of a remittances service purchased by a Senegalese client.

As can be seen in Figure 1, the call shop worker had developed the tacit knowledge required to overcome personal identification in remittances services. The terms apellido paterno (father's surname), ciudad (city), estado (state) and código postal (postcode) have been misunderstood and have been filled in "wrongly" (highlighted area in Figure 1). Under the father's surname, for instance, Naeem wrote a common Senegalese first name. Also, he wrote the name of the street (Montserrat) under "city", and he provided the name of the province (Barcelona) rather than "Spain" under "state", although under país ("country") this is properly indicated (ES). Finally, the postcode is wrong (it should begin with 08, not with 09). Naeem knew that this was not relevant because the only issues that he needed to double-check were the payee's address, the country of destination (Senegal, coded as SN) and the amount of money to be sent. This is a practice that is not accepted in regular fax services or remittances offices that belong to the government (like Sociedad Estatal Correos y Telégrafos) or to the ICT marketplace (like Western Union,


Figure 1: The provision of remittances services without personal identification (taken from Sabaté i Dalmau 2014, 81).

MoneyGram or Ria), all of whom demand not only proof of residence but also the clients' personal details.

#### 3.1.2 Strategising with Operators: Multi-Use of ICT at Convenient Times

All informants used more than one ICT service in their visits to the call shop, as this offered telephone services, international call cards, internet access, money transfers, fax and photocopy services, etc., in one single space and until midnight, adjusting to the transnational communication needs and schedules of clients. This is illustrated by Yousaf, a 42-year-old electrician from Pakistan, who came to the call shop daily to make a phone call to Pakistan. He then also bought international calling cards (Siempre Latina) for the Pakistani women of the two families with whom he lived. This allowed them to make phone calls in the intimacy of their homes without having to go to the place (at 1 AM was their convenient calling time in Pakistan), by sharing one fixed telephone handset which they circulated in their block of sublet apartments next door. Besides, Yousaf would also send remittances when required, at a time when, as his flatmate Shabbir, a 41-year-old construction worker from Kashmir, stated: "el banco está cerrado pero el locutorio no [. . .] el locutorio tú llegas y puedes" ('the bank is closed but the locutorio is not [. . .] you get there and you can'; interview 22/8/08).

Apart from allowing for simultaneous multi-use of ICT during suitable "home" times, the call shop worked with most of the operators available in the market, and therefore, it made their different services and prices available to users all at once. This allowed informants to strategise with the most economic options of each company, picking and choosing particular offers without establishing any permanence contract of any kind or any minimum monthly payment with any of them. This is illustrated by Vitória, a young woman from Brazil informally working in the bar next door, who used two different SIM cards in two different phone handsets, obtained by making use of a promotional campaign in the call shop whereby a multinational gave her a second handset and SIM for free, without having to sign a contract. As she put it: "Tengo un Vodafone y un Movistar [Telefónica's operator] [. . .] la compañía Vodafone es <mejor> [/] mejor para llamar en Brasil [. . .] la otra es la mejor para aquí" 'I have a Vodafone and a Movistar [Telefónica's operator] [. . .] the company Vodafone is <better> [/] better to call Brazil [. . .] and the other is the best one for here'; interview 25/8/08. Sheema, the 31-year-old bar owner from Kashmir, also brought two handsets with him: one with a top-up SIM card employed for local calls while monitoring his calling expenses, and another one with a cheap rate for a chosen "foreign country" with almost no credit, which he employed only to receive calls from Pakistan (and initiate them only in an emergency case). This "double SIM-card" strategy was very popular among informants, particularly for budget reasons, to avoid overcharges and to avoid monthly permanence payments. As Abde, 31-year-old unemployed man born in Morocco, explained, he decided to use the "double SIMcard" strategy because otherwise "digamos tú tienes no tienes dinero y luego la factura tarda dos días una semana y se ha doblado la factura # Movistar [Telefónica's operator] a mi me dobló [. . .] yo no me gustan los contratos" 'say you have you don't have any money and then the bill takes two days a week and the bill has doubled # to me Movistar [Telefónica's operator] doubled it [. . .] me I don't like contracts'; interview 1/9/08.

A last underreported strategy to access ICT by avoiding contact with operators is the "missed call" system, which I present by making use of an analysis of the call registers in the telephone cabins on eight selected days. Table 1 shows that the calls to local mobiles (the main call destination in the monthly collected receipts) were short, lasting from a few milliseconds (i.e. 00:00 min) to less than eight minutes. They tended to be of an instrumental nature and included extremely brief conversations giving information of a pragmatic kind. This is an example of such conversations, taken from the fieldnote diary: "Estoy en casa de la yaya" 'I'm at granny's' or "Oye, que ya te lo he mandado, ¿Vale?" 'Hey listen, I've already sent it to you, okay?'.


Table 1: Migrants' use of telephone cabins. Main destinations and duration of calls per day. Source: Sabaté i Dalmau (2014,

 87).

The shortest calls in Table 1 did not actually involve any talk (clients hung up after making sure that they had heard the first dialling tone) and functioned as the "beeping" or "flashing" system (Overa 2008, 48). This consists of the use of intentionally missed calls, employed, for example, to let a friend that one has caught the 17:30 PM train as planned, without having to engage in a conversation. This then cost 30 cents, which was the cost of a one-minute call to a local mobile. This was a way for informants to make use of ICT services without having contacted any operator and without having to pay for any connection fees, controlling their budget from a telephone cabin, as these use a per-minute charging system and display the minutes and seconds that have been used up on a screen placed next to it.

The examples provided in this subsection (the use of "freed" mobile phone SIM cards, double SIM-card strategies, "missed call" tactics, etc.) illustrate that migrants, fully engaged in the capitalist consumption culture, are not simply "passive" consumers but "prosumers" (Rheingold 2002), that is, "professional" consumers with the autonomy to interfere in the market through the establishment of their own technology consumption habits, playing with, and challenging (though not jeopardising) dominant clientelist regimes.

#### 3.2 Counterhegemonic Multilingual Practices

#### 3.2.1 At the Frontstage: A Peninsular-Spanish Monolingual Floor

At the front-stage (Goffman 1959), the legitimate oral and written language, both for commercial activities and for socialisation, was standard Peninsular Spanish – the language of the nation-state. This was the public code which served as a "barometer of integration" and which indexed a proper citizenship-deserving migranthood identity (Sabaté-Dalmau 2018b), in the call shop and elsewhere in their neighbourhood.

Following this prevalent sociolinguistic order, all computer programs, cabin and fax receipts, international phone-call cards, etc., were provided in this language (although Naeem could have chosen another language for his local ICT system; for example, his programs could have been set up in "English UK", as he showed me). Informants, in fact, demanded a unified floor in Peninsular Spanish, as seen in the complaints issued by Jenny, a 35-year-old born in the neighbourhood, who teased Naeem with an "Hablar en español, ¡que estamos en España!" 'Speak in Spanish, we are in Spain!' (fieldnotes 28/6/08) when he used Urdu and Panjabi with his colleagues. Likewise, unconventional or non-orthodox Spanish oral and written practices which defied the norms of standardness, frequently involving other (de/ re-localised) multilingual language repertoires, were openly criticised, as when a Bolivian neighbour discredited Naeem's Panjabi-sounding Spanish expressions with an "¡Habla más claro!" 'Speak more clearly!', or when Jenny showed me an SMS sent by her Moroccan colleague Moha and told me "tendré que enseñarle a escribir" 'I'll have to teach him how to write' (fieldnotes 29/7/08).

According to these prevalent sociolinguistic comportments, the "translinguistic practices" (Jacquemet 2005, 2010) or repositories of Spanishes which involved diverse communicative frames, heteroglossic local and global repertoires and grassroots literacy practices were dismissed. Thus, by Catalan society at large, by other migrants, and at times by speakers themselves, they were (and are) considered "not-quitelanguage[s]" (Gal 2006, 15) and they were constructed as not-fully-fledged "immigrantese talk" (see Ferguson 1975) mobilised by "deficient", "deviant" or "incompetent" speakers (in Gumperz's sense; Gumperz 1982). These (self)-censoring linguistic ideologies infantilise and further minoritise the migrants' wide range of oral and written language resources in their host society.

#### 3.2.2 At the Backstage: A Legitimate Space of Translinguistic Spanish

In practice, though, what regulated the call shop in sociolinguistic terms, both at the backstage and, in many communicative events, at the frontstage, was the use of the sort of translinguistic Spanish which I have mentioned above, which acted as the peripheral "multilingua franca" (Makoni/Pennycook 2012) of the place.

In the oral mode, for instance, I recorded the systematised use of expressions like "the recarga" or "el tecnición" (both by Spanish and non-Spanish-speaking informants), which display traces of Spanish and English contact, and which were the routinised, unmarked words for "the top-up" and "the technician". In the written mode, I also encountered myriad instances of this flexible use of unconventional Spanish forms, as seen, for example, in the SMS sent by Moha from Morocco to Jenny, reproduced below (fieldnotes, 29/7/08; taken from Sabaté i Dalmau 2014, 139):


The text message below contains (among the many aspects we could focus on) traces of heterography, defined as "the deployment of graphic symbols that defy orthographic norms" (Blommaert 2008, 7), as seen, for instance, in Moha's use of the adjective "continto" (instead of "contento") for "happy", or of "carinyo" ('dear', "cariño", in Spanish), the latter using Catalan orthography for the nasal palatal sound /ɲ/. The SMS may also contain xenoglossy, which is the linguistic ability to access and to appropriate orthographic signs without having fully understood or learnt their uses (Jørgensen/Rindler-Schjerve/Vetter 2012). This is seen with "truncations" (Anis 2007, 90), as in "na brazo" (arguably equivalent to "an arm"), possibly a morphological reinterpretation of "abrazo" ("hug"), since both the standard and the unconventional forms are phonetically equivalent. Finally, Moha also employs transcodic marks from Arabic phonology that may reflect "accent", as seen in the verb "estue" ("I'm") or in the adverb "mue" ("very"), as well as some sound writing, for example in "ola" ("hola" or "hello").

The (conscious and unconscious) oral and written uses of translinguistic Spanishes illustrated above index in-group membership, proximity and intimacy, and they had become marks of more horizontal, technology-mediated "nonofficial ways of speaking" (Thurlow/Mroczek 2011, xxiii) and of "being" in the call shop -note, for instance, that Jenny, the Spanish-speaking local woman, incorporated "mahibek" in Romanised Arabic, which means "I love you", in a systematised way, as a farewell for her SMSs. And yet, informants, on several occasions, kept displaying self-regimentation and self-disciplining practices in Peninsular Spanish, as when Pakistani Sheema, a 31-year-old man from Kashmir, presented himself with a Spanishised common name, "Chema" (/'ʈSemɑ/), or when Naeem kept insisting that I should "correct" his SMSs because his colleagues did not "like" his written Spanish (he stated "no les gusta" 'they don't like it' (fieldnotes 30/6/08).

I would like to finish this section by complexifying the sociolinguistics picture that I have just provided concerning the social meanings of, and linguistic ideologies around, the use of Spanish as the non-standard "we-code" (Gumperz 1982, 66) of the place. Basically, I claim that on other occasions this undefinable "Spanish" was legitimised by informants who presented themselves as "language-haves" and as legitimate "new speakers" of Spanish, in acts of "linguistic assertiveness" (Canagarajah 2013, 4) and of linguistic self-capitalisation. This is seen in Figure 2, which is a "no smoking" handwritten notice presented first in Standard Arabic in the Arabic script and then in Spanish, in the Roman alphabet, posted by a teenage girl born in Morocco, on her own initiative.

The Arabic part of the notice in Figure 2 literally reads "It is forbidden to smoke and thank you", and includes the diacritical marks employed in Moroccan primary schools to indicate vowel length. The part written in unconventional Spanish, below, says "No smoking" (it is unconventional because the standard form of "forbidden", "proibido", is "prohibido"). It also includes English, in "by"

Figure 2: The legitimate use of translinguistic Spanish for public notices in the call shop (taken from Sabaté i Dalmau 2014, 136).

(bottom, left). I suggest that this was a way for the teenager to display a cosmopolitan multiliterate identity where Arabic was given a symbolic public space in the call shop. It was simultaneously a way for worker Naeem to understand nonstandard Spanish notices as valuable and meaningful for clients.

Finally, I also observed that this valuation also occurred with unknown and frequently discredited minority languages such as Urdu or Panjabi. Sheema, for instance, got information about a Telefónica's discount plan from a friend who translated the important pieces of information about that particular promotion from Spanish into Urdu, rendering it accessible to Urdu readers, which Naeem photocopied and distributed to the literate members of the Pakistani local network, subverting the monolingual regime of that particular multinational company.

All in all, the data sets presented in this subsection illustrate the ways in which migrants' social incorporation practices, particularly in the ICT services domain, were conducted in and through non-standard transcodic marks, translinguistic uses and valuations of silenced minority local and global languages, despite their investments in "correct" Spanish.

# 4 Conclusions: The Affordances of Network Ethnography to Investigate the Sociolinguistics of Migration in the Age of the Network Society

In this chapter, I have tried to demonstrate that, with the rise of the network society and the "mobile-isation" (Castells 2012; Urry 2007) of its transnational individuals, the real battle concerning migration, particularly the undocumented, is played out in the communication and information domain (Ros/Gonzàlez/Sow 2007). I have thereby argued that ICT, the "communication staple" (Panagako/ Horst 2006, 112) of migrants, have become a key technology of power (Castells 2009), a gatekeeping mechanism not only to register and control migrant populations but also to curtail their access to citizenship rights and resources (like administrative procedures for legalisation); particularly in Southern Europe -in Catalonia, in this case.

Through an analysis of a technopolitical law (Ley 25/2007; BOE 2007), I have focused on the regulatory tandem conformed by the Spanish nation-state and the Spanish telecommunications sector, and I have explored the ways in which this neoliberal governance block controls not only the mobility trajectories but also the citizenship statuses and social incorporation processes and practices of migrants, due to the monopoly that they jointly have on the control of ICT. On the one hand, I have shown that this law, presented with the imperative "¡Identifícate!" 'Identify yourself!' by the former Spanish Minister of the Interior Pérez-Rubalcaba, demonstrates that governmental institutions in Spain, at the turn of the twenty-first century, started to explicitly target the "illegals" by making proof of legality compulsory and by granting technopolitical power (for example, to monitor and record migrants' consumption habits) to the telecommunications oligopoly. This has turned this newly clientelised group of consumers into a highly beneficial "lucrative niche" to overcome "market saturation". On the other hand, this analysis of the aforementioned law has also allowed me to show that technopolitical power concerning ICT speaks of the establishment not only of mobility, citizenship and commercial regimes but also of language regimes. The data on the government's language policies concerning ICT-mediated public administration (e.g. in the official e-governance platform) and on the Spanish telecommunications sector's commercial language policies (in customer services: call centres, websites and ICT products) show that "multilingual user" platforms are a "linguistic fetish" (Kelly-Holmes 2005). The real management of linguistic diversity, I have shown, is conducted in and through Spanish-only monolingualism, which has turned this language into the Spanish government's rule-of-law language and the Spanish telecommunications sector's hegemonic commercial language.

I have approached the consequences of technopolitic governance concerning migrants through a critical sociolinguistics network ethnographic lens of the migrants' ICT communication practices and their multilingual resources involving ICT management and use, in their own communicative spaces, and for over two years. This has allowed me to focus on the complexity of the ways in which migrant populations get engaged with, and therefore reproduce, the dynamics (the rationalities) of neoliberal citizenship regimentation, but, simultaneously, circumvent and subvert it en masse, in a "bottom-up" manner. I have zoomed into an "ethnic" call shop, a type of small business which, I claim, at present serves as an alternative transnational institution and informal "shelter" of migration which caters for the services, products and, overall, citizenship resources that both the Spanish government and the telecommunications sector have left unattended.

The call shop has served to try to look "at how ICTs are constructed, accepted, adopted or adapted in migration contexts", which is key to put forward "a new 'lens' for analysing migration in the information society" (Ros/González/Marín/Sow 2007, 33). I have detailed how underreported migrants mobilise a series of "off-the radar" tactics for successful ICT access (the "double-SIM card" use, the "flashing call" system, and a multi-use of ICT which avoids contracts and allows for the close budgeting of ICT expenses). I have also reported on how migrants invest in standard Spanish as the expected commercial language and as the legitimate socialincorporation language, but routinely display, collectively and at the frontstage, subversive multilingual practices which defy monolingual normativities, standardness, and native-speakerhood linguistic ideologies. These counterhegemonic practices include the self-provision of technoliteracy resources, linguistic mediation and brokerism, with workers being key for this, for example, to manage administrative services (like visa renewals). These practices also include the catering for ICT services in the migrants' own translinguistic repertoires (including transcodic Spanish) and in their silenced local/global languages (like Panjabi, Urdu and Moroccan).

All in all, the insights from a 20-member network ethnography provided in this chapter highlight the relevance of researching the linguistics of migration in the digital age on an interpretive, socially-engaged case-study basis, in largely uncharted spaces of non-citizenship. I argue that this may contribute to more broadly problematise socioeconomic and political inequality and exclusion processes and practices, as well as resistance to them, in contexts of precariousness and "disconnection", ultimately shedding light on the regimes which dictate who count as citizenship-deserving migrants on the basis of language, under the antimigrant governance logics of the network society.

### References


Goffman, Erving, The presentation of self in everyday life, Garden City, Doubleday, MCE, 1959.


Jørgensen, Normann J./Rindler-Schjerve, Rosita/Vetter, Eva, Polylingualism, multilingualism, plurilingualism. A toolkit for transnational communication in Europe, 2012,

https://observatoireplurilinguisme.eu/index.php/en-us/cross-cutting-themes/multilingualismand-plurilingualism-en-us-6/6602-polylingualism-multilingualism-plurilingualism-definitions [last access:March 8th, 2022].


### Andrea Chagas Migration and Linguistic Repertoire

Supervernacular Spanish and the Variational Perception of the Equatoguinean YouTuber Monanga Bueneke and her Community of Practice

Resumen: Los procesos de migración junto a los desarrollos tecnológicos de los medios sociales han cambiado el panorama lingüístico-cultural existente en nuestras sociedades. La variación, la superdiversidad y la imprevisibilidad son hoy la base que caracterizan, tanto a las lenguas y sus variedades, como también a los individuos que las utilizan. Diferentes conceptos han sido introducidos en las últimas décadas con el fin de abordar esta creciente movilidad y dinamización de procesos sociales. Uno de estos conceptos es el de "supervernácula." En esta colaboración pretendemos otorgar un marco metodológico a través del cual las delimitaciones de la supervernácula pueden ser estudiadas a nivel sociolingüístico, basándose en acercamientos variacionales perceptivos. Se ha elegido el canal de YouTube de una guineoecuatoriana debido a que hasta hoy en día poco se ha estudiado sobre esta variedad del español, y pretendemos contribuir con cerrar la brecha de investigación existente entre estudios de otras variedades del español y la(s) guineoecuatoriana(s). Y a la vez tematizar su falta de reconocimiento internacional dentro discurso pluricéntrico hispánico.<sup>1</sup>

# 1 Introduction

An increase in mobility, both physical and virtual, has changed the linguistic and cultural panorama of our societies in the last century. Questions addressing the nexus between language and migration are becoming of more and more importance. Since the 1990s, migration patterns have become less predictable as a result

#### Andrea Chagas, University of Mannheim

Open Access. © 2023 the author(s), published by De Gruyter. This work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License. https://doi.org/10.1515/9783110746082-004

The Spanish spoken in Equatorial Guinea is still not seen as an established variety of Spanish. Nor is it considered to be a center within the pluricentric discourse, since it does not fulfill the criteria introduced by linguists to define a center. These criteria are (1) linguistic contrast to other varieties at the structural level, (2) diatopic differentiation, (3) historicity, (4) literature language, (5) activities towards the construction of a norm, (6) functions as a Dachsprache (umbrella language), (7) institutionalization, (8) official language, (9) functions as an aspect of identity-building (Prifti 2021, 187).

of the increasing complexity added to the societies receiving migrants. The diversification results "not just in terms of bringing more ethnicities and countries of origin, but also with respect to a multiplication of significant variables that affect where, how and with whom people live" (Vertovec 2007, 1042). A growing awareness of the phenomena resulting from globalization, migration and technology development and their effects on the cultural and linguistic landscape of the receiving countries is emerging in sociolinguistics, which is addressing this within the discourse of superdiversity.<sup>2</sup> Two main fields of innovations can be detected within the superdiverse paradigm. First, a new theoretical basis is being created through the introduction of concepts, such as supervernacular, which position the parole, variations, and dialects at the center of analysis. Second, these new concepts themselves open up opportunities to develop or reinterpret methods with which emerging conceptualizations can be studied. With these innovations, we can move away from the theoretical analytical basis that were developed in structuralist or generativist studies. In doing so, we then move towards the study of a previously neglected aspect of internal linguistic knowledge about linguistic variation that integrates the speakers' own subjective perception.

The aim of this collaboration is to deliver a methodological approach with which to study the supervernacular Spanish parting from the subjective perception of an Equatoguinean YouTuber and her community of practice. We hypothesize that, due to their unique biographical trajectory, migrants are exposed to several different languages and their dialects, which can lead to an expansion of their linguistic repertoire. To test this hypothesis, we draw data from an Equatoguinean YouTube channel, creating a small corpus based on (1) the transcription of four videos by an Equatoguinean YouTuber and (2) the compilation of the comments below the video. The YouTuber, Monanga Bueneke, has 64,800 followers, and she produces cultural content pertaining to her languages (Spanish, Ndowé and other Bantu

Currently, scholars have not reached a consensus as to how superdiversity should be understood, namely whether it should be addressed as a phenomenon, a method, a theory or even a paradigm shift. (Budach/Saint-Georges 2017, 64). We understand superdiversity as a paradigm within which diversity (linguistic, ethnic, cultural, etc.) is seen as the norm, not only for migrant groups but also for those people living in urban areas all around the world. We would also like to go further by expanding the superdiverse paradigm to include societies in which diversity, in all its forms, has always been at the core. Equatorial Guinea, for example, is a country in which not only ex-colonial European languages are spoken on a daily basis—vernaculars are also spoken all around the country. In this case, superdiversity is not limited only to urban areas. Although urban areas in Equatorial Guinea do show a larger diversity, due also to contact with European people and their languages, superdiversity can also be found in rural areas where different languages, such as Fang and Ndowé (in the continental area), as well as Bubi and Pichi (on Bioko), among other languages, coexist.

languages), her cultural traditions, and the history of Equatorial Guinea (EQ).3 We understand digital media to function as containers of cultural production that also "facilitate deterritorialized interaction, individualized self-presentation, and largescale participation in cultural and political discourses" (Androutsopoulos/Juffermans 2014, 2). The technological developments allow transmigrants to maintain linkages with their countries of origins and/or to create new linkages to people in the diaspora. In this collaboration, we will study whether a subjective, perceptional approach can provide us with key information about what speakers judge to be "Spanish" and what they see as falling outside this language. We ask: Does migration lead to an expansion of repertoire? Is the community of practice interested in the Spanish spoken in EQ? Does the community of practice recognize the Spanish spoken in EQ as another variation of Spanish? How do the YouTuber and her community of practice delineate the boundaries of linguistic features considered to be "typical" of a specific dialect? In order to acknowledge this complexity, we follow an interdisciplinary approach where we combine sociocognitive methods—perceptions of linguistic variation—with more traditional sociolinguistic and ethnographic methods such as the biographical interview.

The following section provides an overview of social media, deterritorialized migrants and superdiversity. In section 3, we introduce the basic notions of repertoire and discuss how migration can influence the repertoire extension. The term supervernacular is also discussed. Section 4 then turns to the theoretical basis, namely perception and variational linguistics based on indexicality and metapragmatics. In the 4th section, we apply the method and categorizations to this study, too. Section 5 presents its analysis and implications, while section 6 provides a summary and ideas for the future study of superdiverse phenomena.

### 2 The Superdiverse Neighborhood: YouTube

Digital technologies help migrants retain their previous identities while keeping social ties with their previous social networks through social media.4 Social constructs

Equatorial Guinea is a small country in sub-Saharan Africa. It was a Spanish colony until 1968 (Nsue Mibui 2007, 240). EQ is the only Spanish-speaking country in sub-Saharan Africa, and took Spanish as one of its official languages in 1982.

It is important to highlight that this evolution does not only influence the migrant communities; they also affect the "host" communities. Autochthonous people may therefore also get involved with people from around the world, which allows them to create transnational networks that can influence identity-building processes, as well as community formation based on new adopted linguistic features in the "host" society.

such as languages and groupings are understood as fluid and ever changing. Therefore, language use in computer mediated communication (CMC) is characteristically more unpredictable than in face-to-face situations. Social actors interacting online are normally aware that their interactants may share the same language, but not necessarily the same dialects, registers, etc. Therefore, besides unpredictability, different studies have also shown that linguistic practices in CMC lead to a homogenization of the features used to communicate with others (Budach/Saint-Georges).

Multiple and layered normativities can be found on social media. Social agents engage in discussions surrounding different topics and their validity in society and/or for themselves and their community of practices. Normativity can be defined as "a form of reflexive action [that] involves ways of evaluating, judging and policing the semiotic conduct of oneself and others" (Leppänen/ Peuronen/Westinen 2018, 36). Normativity thus involves both the superposition of a norm from outside and the (re)negotiating process each individual and/or groups carry out to delineate their own understanding of language, linguistic practices, etc. Since the early 2000s, research within cultural studies, including sociolinguistics, has "examined how diasporic and transnational people appropriate the Internet to create discourse spaces in which to articulate marginal voices, negotiate plural identities, construct the meanings and boundaries of community" (Androutsopoulos/Juffermans 2014, 3). All this is possible as a result of the different technological affordances social media platforms provide for communicative purposes.

YouTube is one of the biggest broadcasting platforms that simultaneously offers a social media feature. Its functionality is multifaceted; It fosters active participation: as there is no clear distinction there between production and consumption, consumers can be active in the creation and circulation of new content" (Leppänen/Häkkinen 2012, 19). A clear delimitation of the directionality of media consumption is thus not possible for social media. The simplified, user-friendly interface available in the social web, or web 2.0, has different technological affordances. First, it allows users to be listeners, readers and watchers. Second, the same users can also be content creators. Third, it allows users and content producers to interact with one another quasi synchronically—e.g. through live videos and direct commenting—or asynchronously—e.g. via comments under videos (Dürscheid 2003)—from various places all over the globe. These last affordances are of special importance for linguists for three reasons. Firstly, users and content creators actively participate in negotiating processes where topics of their matter are approached and discussed by people sharing the same interests. Secondly, users and content creators can be interconnected without sharing time or space, thus dissolving all categories previously used in sociolinguistics to define communicative practices from specific groups. Finally, linguists can now access authentic data5 produced outside of research context. However, data gathered from social media is characteristically decontextualized, since background information, e.g., biographical information, cannot always be found in the comments or profiles of the users being analyzed.

In our case, YouTube offers material that can be drawn upon to study the perceptional recognition of linguistic variation. At the same time, we can gather information on thousands of users engaging in the conversations started by one YouTuber uploading a video. As we will see later in this collaboration, it is very interesting to study who exactly is most engaged in such conversations (or negotiations). As this paper finds, it is mostly Spanish speakers from Africa and the Americas who are interested in negotiating what Spanish is for them, while Spaniards comment only sporadically on, e.g., Monanga's videos.

### 3 Superdiverse Repertoires and the Supervernacular Spanish

#### 3.1 Superdiverse Repertoires and Communities of Practice

The term repertoire comprises a whole range of linguistic competences, "from [the] linguistic ones (language varieties) over cultural ones (genres styles) and social ones (norms for the production and understanding of language" (Blommaert/Backus 2013, 11). For Hymes (1974), Gumperz (1982, 155) and other linguists, repertoires were the "means of speaking" that characterized different speech communities.6

Today, the sociolinguistics of superdiversity implies a paradigmatic turn in which notions and definitions of language are renegotiated and intrinsically

This kind of data is very authentic data, meaning that it is produced by speakers outside of a research context (Bendel Lacher 2015, 48). Authentic data thus differs from data generated for specific purposes and within a research context through, e.g., interviews. The latter creates information that resembles its own social reality, which can vary from the social reality outside the research context. The data gathered for this paper was therefore created by speakers who intended merely to communicate with others, thus directly displaying their active knowledge on repertoires and linguistic variation.

These speech communities were understood as a closed group of individuals who negotiated the linguistic boundaries of their repertoire to facilitate communication. Methodologically, speech communities were isolated from other social ties and were defined by their shared experience of time and space (Blommaert/Backus 2013, 12).

connected to language learning7 trajectories and the waxing and waning of repertoire as connected to these trajectories. Linguistic knowledge is thus connected to biographic dimensions of an individual's trajectory. Therefore, migration directly influences the knowledge an individual has about, firstly, linguistic structures and, secondly, the sociocultural rules attached to those structures.

As a loose and vague term within which language, linguistic variation and linguistic practices are understood as belonging to specific social categories (Bloomfield 1933; Gumperz 1982; Milroy/Milroy 1998), "speech community" has lost its relevance for contemporary sociolinguistic studies. Now, terminology that allows the recognition of the role of language and of the speaker themselves in social interaction is more relevant. The term "community of practice" (Buchholtz 1999), for example, allows sociolinguists to acknowledge the importance of linguistic practices and the agency of speakers in their social interactions. Communities of practice thus "create a common ground that allows them to embody their own identities. In the process of group establishment, they set rules and boundaries that can be expressed linguistically in the form of (for example) style" (Chagas i.p). This means that patterns of sociolinguistic resources are dynamic and are shared by individuals with greatly divergent sociocultural and ethnical backgrounds (Blommaert 2012, 3).

To return to our case in particular, this study focuses on the linguistic knowledge an Equatoguinean has about Spanish after their arrival in Spain. Repertoire, for us, hence encompasses not only the linguistic and non-linguistic knowledge a person has about a language or variety of that language; it also involves the dynamic learning, redefinition, and erasure of relevant linguistic and sociolinguistic structures of a language. Linguistic repertoires are embedded in different communities of practice, to which people can simultaneously belong.

This linguistic "knowledge" can display different patterns. Language learning encompasses all kinds of tactics, strategies and mechanisms that an individual can apply to incorporate linguistic resources to their repertoire and that are temporary and dynamic. Language acquisition refers to a more stable acquisition and incorporation of those linguistic and non-linguistic resources (Blommaert/Backus 2013, 14). These learning strategies can vary according to factors such as the learning strategies common to the culture the person comes from. For example, due to their oral tradition, members of African societies may tend to rely on oral strategies to incorporate new resources to their repertoires.

#### 3.2 Supervernacular Spanish

The study of superdiversity and the phenomena resulting within this paradigm is currently an ongoing process. We would like to apply the concept of supervernacular to help us describe language in superdiversity. A supervernacular is always a dialect—in our case, of Spanish. The term differentiates between "an imagined, normative standard and actual dialectal realizations of it [enabling] us to include the crucial language-ideological features of the processes we intend to understand here" (Blommaert 2012, 5). This distinction allows us to return some agency to individuals by recognizing that there are several aspects to language that need to be acknowledged at the moment of analysis. The conceptualization of supervernacular allows researchers to recognize how speakers themselves negotiate what a given community of practice imagined to be "real" on an ideological level.

The internet and CMC, as they develop together with rapid technological progress, allow people sharing similar interests to connect around the globe via different social media platforms. Within the community of practice studied in this collaboration, for example, participants are prompted by the content uploaded by the YouTuber to negotiate their subjective perception and learned knowledge about linguistic variation within the Spanish-speaking community. This linguistic variation, as we shall see later, explicitly includes sociocultural and linguistic diversity autochthone to each territory in the Spanish-speaking world. Even though users and YouTubers do not necessarily share the same background they still develop sociolinguistic patterns that are understood as the supervernacular (Blommaert 2012, 3).

### 4 Method

#### 4.1 Theoretical Basis

The study of linguistic variation in both variational linguistics and sociolinguistics has traditionally focused on the structural, analytic side of language as connected to the social context of speakers (Caravedo 2014, 9).<sup>8</sup> This is however changing,

It is, however, important to point out that previous works within the sociolinguistic paradigm have already provided a foundation for future studies based on the cognitive principles available to speakers. Labov (2010), for example, relied on the knowledge speakers have about their linguistic competences based on their cognitive perception of variation. He elaborates upon this by differentiating the stylistic choices people make according to degrees of formality and informality.

Rocio Caravedo (2014) points out, invariation presupposes variation: One cannot exist without the other. This presupposition implies that variation and invariation are both intrinsic parts of the linguistic competence each speaker has around their language(s). A cognitive turn, based on, for example, perceptional variational linguistics, therefore seems very appropriate since it allow researchers to study what speakers actually do with their linguistic resources.

#### 4.1.1 Perception in Migration and Metapragmatics

The linguistic knowledge every speaker possesses is three-fold.9 For scholars following the perceptional perspective, the first set of knowledge, known as general linguistic knowledge, is of interest. It encompasses the competence to understand and produce speech, and at the same time involves the linguistic awareness known as metalinguistic awareness, or knowledge. To study perception, researchers must differentiate between the cognitive processes speakers carry out in order to make meaning by observing whether this happens consciously or unconsciously. Perceptive variational linguistic therefore differentiates between "representation" and "perception" (Krefled/Putska 2014, 11).

This differentiation is necessary, as both phenomena involve divergent processes. Perception belongs to the field of speaking in a real communicative situation (parole) (Krefeld/Putska 2014). Representation, on the other hand, is understood as "part of the linguistic knowledge (langue), which is why it [perception] can be accessed independently from concrete perceptions" (Krefeld/Putska 2014, 14, our translation).10 Representation is therefore based in how people's speech is perceived and

(1) General linguistic knowledge: "This knowledge encompasses all information about linguistic variation [such as varieties and dialects] of a specific linguistic system [language] and also about the structural differences to other languages" (Dittmar 1997, 2). This general knowledge can be passive or active.

<sup>(2)</sup> Normative knowledge about a language. This knowledge comprises the deliberate acquisition of rules set by legitimating institutions such as schools, academies, etc. This normative information can only be acquired though deliberate activities that include formal education.

<sup>(3)</sup> Knowledge about habitual use of resources according to the situational communicative context. This knowledge involves the practical use of rules acquired or learned through socialization. The suitability of the communicative practices are at the core of this knowledge (Dittmar 1997, 2).

Original quote: "in dieses Wissen ist auch ein Wissen über varietätenspezifische Unterschiede und über die Strukturen anderer Sprachen eingeschlossen" (Dittmar 1997, 2).

Original quote: "Zwischen diesen beiden besteht nämlich insofern ein essenzieller Unterschied, als die Perzeption dem Bereich des Sprechens in einer realen kommunikativen Situation (parole) angehört. Die Repräsentationen dagegen sind ein Teil des Sprachwissens (langue),

taken as representative of part of their strategies to interpret the social value of linguistic features.

The contemporary representations of variation are the object of study within perceptional variational linguistics. Representations cannot be accessed directly but can be induced with stimuli. Therefore, the ways in which people's speech is taken to represent a certain language, dialect, etc. can be accessed by presenting research subjects with perceptional stimuli to trigger their underlying representations. This can be done by presenting acoustic stimuli in form of audios to probands for them to evaluate and judge which accents belong to which groups of individuals.11 Within the study of migration, the application of this method can be very proliferous since migrants highly rely on perceptional strategies to navigate the new society in which a different language or dialect is spoken (Caravedo 2014, 255).

Current perceptional studies are interested in the notion of "microterritoriality like neighborhoods, which become the central locus of linguistic change" 12 (Caravedo 2014, 258, our translation). This approach, based on microsocial structures of local communities, is studied in Labov (2001; 2010) and Eckert (1999). Both studies analyze "leaders" in certain groups. These leaders are viewed as "conformists" or "non-conformists." Their research proposes that the true "genesis of change seems to be found in the individual dimension rather than on the group dimension" (Caravedo 2014, 258). Within this study, we henceforth concentrate on individual judgments, since we propose that learning a language is a vital individual experience that is shaped by socialization, which then results in very different valorizations towards linguistic features (Caravedo 2014, 259).

The present approach thus coincides with the superdiverse agenda. By focusing on perception, we concentrate on one individual and their particular perception of speech. Our object of study is the virtual knowledge about language, in their variant forms, like dialects, registers, repertoires etc.—i.e., the supervernacular.

weswegen sie auch unabhängig von konkreten Perzeptionen abgerufen werden können" (Krefeld/Putska 2014, 14).

It is also important to note that speakers may have representations about a linguistic structure that either are not accurate or outdated.

Original quote: "el espacio que entra en el escenario de la variación se circumscribe a una microterritorialidad como la de los barrios, la cual se convertirá en el locus central del cambio lingüístico" (Caravedo 2014, 258).

#### 4.1.2 Indexicality and Metapragmatics

The term "indexicality" is based on terminology introduced in Charles Sanders Peirce's work on semiotic signs (icon, index, symbol). In linguistics, an index<sup>13</sup> is understood as a linguistic sign whose meaning can vary according to the context in which it is used. The relation between index and object is not only interpretative, but also has a real representation in social life.

Indexes are socially stratified and imbued in ideological footings currently circulating within a society. Indexes will therefore be valued very differently according to the contexts and people involved in the interactions. As indexes are connected to the ideologies from the contexts in which they emerge, Silverstein (2003) proposes thinking about them in terms of indexical orders, meaning categorizing the references indexes can make depending on their connection to prevalent ideological beliefs in a given society/group. These orders are embedded differently in society, and the concept of indexical order "is necessary to showing how to relate the micro-social to the macro-social frames of analysis of any sociolinguistic phenomenon" (Silverstein 2003, 193).

Specific linguistic features, such as pronunciation, grammatical or intonation patterns, etc. function as indexes both for the speakers of the group (emic perspective) and for listeners that don't belong to that group (etic perspective), who each make different kinds of social evaluations according to the speaker's "accent" (Johnstone 2016, 632). Indexical orders "come in integral, ordinal degrees" (Silverstein 2003, 193) of enregisterment (meaning the ways in which certain signs are registered as belonging to a particular group; see Agha 2007), which then allow for new orders to emerge. The indexical social meaning of linguistic forms is three-fold:

1 st order indexicality: specific linguistic features are linked to a sociodemographic context from an out-group perspective. Speakers are not aware that they speak in a specific manner.

2 nd order indexicality: references that are deliberately made by the own agents. These references allow linguistic features to signal different contexts. They can be applied as context references.

3rd order indexicality: linguistic features that are understood as 'typical' for a specific context and/or group (Spitzmüller 2013, 266).

One very common indexical reference is called deixis. The pronoun you, for example, can refer to very different people according to the place, time and situation in which that pronoun is used (Hoinkes 2015, 20). This is only one kind of index; other indexes can refer to very different phenomena.

These features are seen as stylistic decisions that people can deliberately employ to display identity or to position themselves. Of special interest for the purpose of this paper are the 2nd and 3rd indexical order. We are interested in studying (1) what features the Equatoguinean YouTuber interprets as "typical" for the chosen dialects of Spanish; and (2) which features are reported to be salient for the speakers of the dialects being referenced. The indexical order allows us to differentiate between emic and etic perceptions. This differentiation is very important, since indexes are imbued in ideologies, meaning that representations of each salient feature as connected to different dialects vary highly according to the socialization of the individuals involved.

To access to these orders of indexicality, we concentrate on metapragmatic comments. Metapragmatics can be defined as "linguistic activity [like comments] that refers to other linguistic activities [like pronunciation]" (Spitzmüller 2013, 264). With the help of this concept, we can access individuals' metalinguistic knowledge and categorize it according to linguistic criteria. Metapragmatic comments and valorizations, such as valuing a specific way of speaking as "wrong" or "correct," are embedded in circulating ideologies, which, on another level, are stored in the indexical orders introduced earlier. The knowledge—in terms of structural, normative, and practical usage (Dittmar 1997)—that speakers have about a language can provide fruitful insights about the linguistic boundaries perceived within the Spanish-speaking community. In addition to this, studying variation beyond comparatist approaches—where varieties are juxtaposed, one being the norm and one the deviation—can help us understand sociolinguistic dynamics in a more nuanced way.

#### 4.1.3 Equatoguinean Spanish

To understand linguistic variation and the way it is perceived, we will offer a brief overview of the linguistic diversity and the resulting features that are considered typical for the Spanish spoken in Equatorial Guinea (EQ).14 EQ is the only country of sub-Saharan Africa with Spanish as its official language, alongside French and Portuguese. Seven autochthone languages coexist in the country,

Due to the length of this paper, we cannot provide an extensive description of the linguistic variation known to be present in all Spanish dialects. The videos used to create the corpus approach eight different Spanish dialects. We decided to only include variationist data regarding Spanish in EQ, since it is the lesser known dialect. For information regarding salient linguistic features of Colombian, Argentinian, Mexican, Peruvian and Uruguayan Spanish, see Patzelt/Herling em; an Manue: Gredos.

though none of them are recognized officially. All these languages (besides the creole languages F'adambó and Pichi) belong to the bantu language family. The vernaculars that coexist here are Fang, Bubi, Ndowé, Bisió and Balengue. Due to the length of this collaboration, we will proceed to present only two of the linguistic features relevant for this study.15

With regards to consonants, the simple vibrant /ɾ/ is realized as a double vibrant when in between vowels. Bubi speakers tend either to omit the sound, to hypercorrect it by pronouncing it as a double vibrant [r] in all environments, or to pronounce it in a way resembling the French /ʁ/. Bubi speakers sometimes also aspirate or leave the /ɾ/ out completely. This results from the absence of the simple and double vibrant in this language (Quilis/Casado Fresnillo 1995, 113).

As concerns vowels, the vowel system of Bubi and Fang are more complex than the Spanish vowel system. Some difficulties appear when vowels in Spanish words appear in positions that are not permitted within Bubi and Fang. This is the case for the vowels /i/ and /e/, which normally never appear at the beginning of a word in Fang. Speakers of Fang therefore tend to pronounce the /i/ as [e] and /e/ as [i], but only if they appear at the beginning of a word (Bibang Oyee 2003, 10).

Concerning morphosyntax, genus and numerus, Equatoguineans tend to inflect the words according to the grammatical genus of the word regardless of their first language (L1). Spanish has some exceptions, such as words of Greek origin that end in -a but have the male grammatical genus.

#### 4.2 Research Approach

#### 4.2.1 Perception and Indexicality

We chose the YouTube channel of an Equatoguinean woman living in Spain who has around 62,800 followers (in 08.03.2022). We initially started with four videos and their corresponding comments. Perceptions and representations of linguistic variation were taken from the videos, which were selected to analyze which linguistic features the YouTuber and the users commenting on the videos perceived

We will present the data gathered in traditional variational linguistics to compare it with the features named in the YouTube videos. It is important to highlight that even though almost all languages of EQ belong to the Bantu languages, they remain distinct to one another at all linguistic levels. This means that whenever we describe a pronunciation, for example, we must always indicate the L1 of the person in order to understand where the corresponding pronunciation, etc. results from.

as salient. Two videos provided an Equatoguinean emic perspective of the salient features regarding their dialects. The other two videos approached an etic perspective on the linguistic features the YouTuber understood as salient. This division was made to equally distribute the data. After selecting the video material, 60 comments under the videos were selected. Only those comments containing metapragmatic or metasemantic information were first transferred into an Excel database for later classification.

We then selected two of the four videos for a closer analysis. The criteria of selection were related to the methodological and theoretical basis required to study perception. As stated in section 4.1.1, when studying perception, we are interested in the underlying representations that cannot be accessed directly. Therefore, specific stimuli had to be provided to the subject in order to trigger the representations, which are then expressed verbally. The two selected videos offered a small perception experiment. The first dealt with Latin-American accents, where the You-Tuber and a friend listened to audios and attempted to guess which accent was being presented to them. The second video presented the "mistakes" or features that are pronounced "wrongly" by Equatoguinean people according to the two friends. This video also entailed the same procedure as the first one.

The video material<sup>16</sup> was transferred into ELAN for transcription and later annotation. For the transcription, we followed the GAT2 conventions (Selting et al. 2009, 353). We decided not to include a multimodal transcription and analysis, since our focus in this collaboration lies on the study of salient characteristics recognized by the language's own speakers and not on interactional linguistics. The annotations, on the other hand, helped us to mark segments that contented metapragmatic and metasemantic comments. Metapragmatic comments included information about correctness or incorrectness regarding either pronunciation of the named phonemes. The metasemantic comments, on the other hand, included information about the meaning of words. After the data was transcribed and annotated, we categorized the metapragmatic comments. We first distinguished between comments overtly mentioning specific indexes, such as morphemes or phonemes, and comments regarding "correctness." In cases where biographical information was provided, we added this information to help us interpret the data.

These videos are the property of Monanga Bueneke. We acquired the rights to make use of the original videos through an agreement with the content producer. When working with data gathered online, one must beware copyright laws. All usernames that reflected users' potential actual names were anonymized. Citations of comments in this collaboration were directly extracted from the comment feed in accordance with §51 of the German Copyright Act. For further information, please visit: https://germanlawarchive.iuscomp.org/?p=855.

#### 4.2.2 Linguistic Biography and Linguistic Repertoires

One of the biggest methodological restraints resulting from working with data collected from sources such as social media is that we possess little to no personal information about the people behind the screen. However, this limitation can be partly balanced out by conducting interviews with the people involved in the conversations.

We follow Blommert and Backus' (2013, 15) understandings about repertoire when they claim that "[r]epertoires are individual, biographically organized complexes of resources, and they follow the rhythms of actual human lives." In order to access the actual repertoire of our YouTuber, we thus have to follow her biographical traces to understand which resources were learned or acquired when and for what purposes. We conducted a semi-structured interview that began with questions regarding (1) first and second language acquisitions; (2) contemporary and previous linguistic practices with parents, family and friends; and (3) trajectory and relevance of languages. This structured section was then followed by open-ended questions that addressed her migration history. The last part of the interview inquired about key information regarding linguistic competences and resources related to the migration process. We focused only on one person since, as Blommaert points out, repertoire is a vitally individual resource that depends on each person's trajectory.

### 5 Analysis and Implications

Migration is known to play an important role in repertoire evolution and its expansion or reduction, depending on which resources are being considered. However, it is very difficult to claim that changes in a person's linguistic resources result solely from moving to another area, country, etc. To demonstrate the global nature of the linguistic resources our YouTuber possesses, we will start with a brief summary regarding her trajectory.

The YouTuber<sup>17</sup> was born in EQ; her L1s are Ndowé and Spanish. She claims that she started learning both languages simultaneously and that Spanish was important for communication with people from other tribes and in educational and public contexts. Both of her parents are Ndowé. She migrated to Spain six years

Monanga shares a lot of biographical information, but for data protection purposes we provide only general information about her. This means that we will not indicate where in Spain she lives in or how old she is.

ago at the age of nineteen in order to pursue higher education. After migrating to Spain, she was exposed to many different Spanish dialects (including Iberian dialects and South American dialects), and she also started learning English in school.

We will now turn to the description of her repertoire; for this purpose, we will take into consideration which linguistic features she names and understands to be typical of certain countries. An analysis of the comments regarding each variety of dialect of Spanish mentioned in the video about Latin-American accents provides a control group for Monanga's perception. We begin by analyzing our interview with Monanga regarding migration and the influence this may have had on her linguistic awareness. Afterwards, we present the results regarding the data gathered from YouTube, complementing it, where necessary, with further biographical information acquired during the interview.

#### 5.1 Migration and Linguistic Awareness and Transnational Social Ties

As Caravedo (2014) has stated, migrants rely on their linguistic awareness to interpret communicative patterns new to them. In our interview with the YouTuber, she stated that she started to question her accent and her way of speaking after arriving in Spain. Not only did her linguistic awareness of Spanish increase, she also states that her attitude towards Ndowé changed drastically. She now displays a very positive attitude towards her vernacular Ndowé. She claimed that she first started feeling proud of her roots and her language after arriving in Spain:

(Transcription of Interview\_Corpus\_Repertoire\_Migration\_Supervernacular\_03.2022. All translations are our own)

{3:53} (1) Y: yo creo que es más cuando he viajado . si es cuando he viajado porque aquí como que te das cuenta de que no sabes quién eres porque:: a parte de que te encuentras con una sociedad que es muy diferente a ti en muchas cosas y te das cuenta que a ellos tampoco le interesa que sepas hablar su idioma y: ahi uno se plantea por qué tu tienes que hablar su idioma y te empera bueno por lo menos a mi me ha pasao a mi me ha pasao y me empiezo a plantear muchas: cosas: quien soy (.) hasta donde quiero llegar y:::: parte de lo que yo soy es mi lengua y no saberla hablarla me da mucha pena y es por eso que mas lo aprecié mucho más el ndowé [and I think it began after traveling . yes it started after migrating because here you realize that you don't know who you are anymore because besides finding yourself in another society, they don't care about you speaking their language. It is then that you start questioning why you should speak their language. At least that happened to me. You start questioning a lot of things, like who am I? . where do I want to go. And part of who I am is my constructed upon my language. And not to speak it well makes me very sad. I started appretiating ndowé much more after migrating.]

Migrating to Spain shifted her perspective towards her native language. She is very aware that speaking Ndowé, in her case, will not bring any kind of monetary reward. However, she realized how much of an integral part of her identity Ndowé is. Her family, who is still living in EQ, also contributes to her attitude. She describes her brothers as "panafricanistas" who motivate her to speak and read in Ndowé more frequently.

(Transcription of Interview\_Corpus\_Repertoire\_Migration\_Supervernacular\_ 03.2022).

{1:72} (1) Y: mis hermanos a parte de que ellos son muy panafricanistas que me envian cosas en ndowé y no me permiten hablar en español

(.)

pero solo en el wasap

(.)

fuera del wasap español es que el mundo que me rodea es español[and: my brothers, besides them being panfricanists who send me things in Ndowé, don't allow me to speak Spanish. But only on wasap [WhatsApp]. Outside of wasap Spanish is the world that surrounds me]

She kept her social ties in EQ through constant contact with her family in Bata and "in the pueblos" (rural areas), as they call it, and also with her friends and other social networks by communicating via WhatsApp. In the interview, Monanga revealed that WhatsApp is the most used messenger platform in EQ, as it consumes less internet volume than any other medium. She also explained that since WhatsApp adopted the "Status" feature, she no longer has to watch the news about EQ—she already knows what's going on, as somebody in EQ will have uploaded the information there.

Migrating to Spain motivated Monanga to open her YouTube channel, where she does not only speak about Spanish, but also uploads content about autochthone languages and cultures. She uploads videos in Spanish, Ndowé and Fang (some other languages may be forthcoming), providing a panoramic linguistic and cultural picture of EQ. The videos she uploads can be seen as a tool that people who originated from societies with oral traditions can use to store their knowledge. The western tradition has more of a written tradition, but many other communication traditions exist that must be explored more deeply.

#### 5.2 Latin-American Dialects (Equatoguinean Etic Perspective)

The YouTube video used for this analysis provided information about four dialects of Spanish:


Table 1: "Africanos adivinan acentos de países hispanohablantes (Monanga Bueneke 25.04.2020, YouTube) https://youtu.be/0Trhd8WEYus".

The YouTuber correctly guessed the Mexican accent by recognizing the speaker's pronunciation. She herself pointed out that she was able to guess this because she is used to hearing it:

(Transcript of YouTube Video Africanos adivinan acentos de países hispanohablantes)

{2:16} (1) es mexico porque luisito comunica habla así y yo soy muy fan de luisito comunica (.) ah ah y las telenovelas en guinea también

llegan mucho y suenan de mexico [it's Mexico because Luisito Comunica speaks just like that. ah ah and the telenovelas in Guinea too. They are all around in Guinea] (Monanga Bueneke 25.04.2020: Africanos adivinan acentos de países hispanohablantes. YouTube: https://youtu.be/ 0Trhd8WEYus)

In this case, Monanga was able to recognize the accent in the audio because she consumes content produced by another Mexican YouTuber. Here, internet access, just like migration, contributes to her repertoire expansion—and indeed, You-Tube's resemblance to superdiverse urban areas is striking here. Monanga also pointed out that telenovelas are well known in EQ, and since she has been exposed to the variety of accents present in such media, she can name it (1st degree index).

It is interesting to note that metalinguistic awareness regarding variation (one's own or another in, e.g., pronunciations) was activated with migration. The YouTuber claims:

(Transcription of Interview\_Corpus\_Repertoire\_Migration\_Supervernacular\_ 03.2022)

{1:58} (1) I: cuando estabas en Guinea\_tenías conciencia de estos acentos (2) Y: no

(.) de hecho cuando yo estaba en Guinea yo pensaa: que hablo español normal sin que alguine me diga <imita> ay:: tu ace:↑nto es diferente aquí es que\_me he ado cuenta que existen acentos diferentes por cada países de español que hay en el mundo <rie> basicamente (.) pe::ro yo no era conciente de eso yo creo que mucha gente si sabe (.) ahoramismo si vas a Guinea y haces esa pregunta la gente te va a decir no (.) no sabemos que hay acento diferentes (.) basicamente (.) si hablamos los temas de los acentos cuando hablamos español äh\_porque

cuando por ejemplo yo hablo español (.) tu te das cuenta que yo soy ndowé o que una persona es fang o que esa persona es bubi po la manera (.) en la que hablamos en español [I: were you aware of Spanish accents while you were living in EQ? Y: no (.) in fact when I was in Guinea I thought I spoke "normal" Spanish. Nobody told me <imitates> o:h you speak different. It was here that I started realizing that there are as many accents as Spanishspeaking countries. But I wasn't aware. If right now you ask people in Guinea about accents they are going to tell you no. We basically don't know that there are other accents. We do speak about accents when we refer to the accents Equatoguineans have according to their native language. You can notice I am Ndowé or a Fang or a Bubi just by listening to their way of speaking.] (Monanga Bueneke 25.04.2020: Africanos adivinan acentos de países hispanohablantes. YouTube: https://youtu.be/ 0Trhd8WEYus)

Monanga claims that she began to gain awareness about linguistic variation after arriving in Spain. It is important to highlight, though, that she expressed that Spanish-speakers in EQ are already very aware of linguistic variation, although this takes place rather in the context of identifying other tribes. Their awareness of other dialects is still limited to the resources they are exposed to and can be interpreted as the result of EQ not having a reliable internet infrastructure. On the other hand, we can clearly see here that indexes are enregistered only when they have a social meaning. Equatoguineans are aware of the linguistic variations within Spanish, but only as regards those features relevant for their social ties. Repertoire expansion would take place with more exposure to different accents, and in the case that these accents gain a social relevance, whether for Spaniards or Spanish-speakers in EG.

It is interesting to note that although the YouTuber wasn't able to name the Argentinian dialect correctly, she was indeed able to name two very salient characteristics of Argentinian or Rio de la Plata Spanish. She recognized the salient pronoun vos and its verbal paradigm of monophthongized diphthongs typical for the region. It is likely that she was unable to distinguish between Argentinian and Uruguayan Spanish, as these two dialects share many features like the šeísmo, the voseo pronominal, etc.

Monango was also not able to guess Peruvian Spanish correctly. However, her explanation as to why she struggled with this accent is very interesting:

(Transcript of YouTube Video "Africanos adivinan acentos de países hispanohablantes")

{1:58} (1) quien lo hubiera adivinao: eh: pero a ver . mi amiga que está aquí que es peruana: ella ha vivido aqui mucho asi que ella ya no tiene el acento [who could have guessed that eh: but lets us see (.) my Peruvian friend that lives here: she has been living here for such a long time that she doesn't have an accent anymore] (Monanga Bueneke 25.04.2020: Africanos adivinan acentos de países hispanohablantes. YouTubeYouTube: https:// youtu.be/0Trhd8WEYus)

Although the YouTuber and her guest both had friends from Peru and Chile, they claim that they could not identify the accents as a result of the assimilation processes their friends' dialects had undergone. They state that their friends no longer speak like "Peruvians" or "Chileans" because they have been in Spain for a very long time. As a result, their accents have changed enough that both Equatoguineans no longer interpreted them as "Chilean" or "Peruvian." As regards migration, these phenomena can be understood as two sides of the same coin. As a result of migration, the YouTuber expanded her repertoire of Spanish accents. Yet the Spanish she is learning in this migration context—besides the dialects that are autochthone where she now lives—is a koineized (homogeniezed) variety. Spanish-speakers are very aware of the language's salient features and of those features that might cause some misunderstandings. Migrants often neutralize these salient features, such as pronunciation, in order to improve their communication.

This phenomenon is pointed out by many of the users who engaged in a conversation with the YouTuber. One, for example, stated:

Vahal✶✶✶ "En Colombia tenemos muchos acentos. Cada región tiene su propio acento. Incluso en Bogotá, dependiendo de la zona, también encuentras variaciones en el acento! Un abrazo a los hermanos de Guinea Ecuatorial. COGQ " [in Colombia we have many accents. Each region has its own accent. Even in Bogotá, depending on the area, you could also find variations in the accent. A hug to our brothers from EQ] (Monanga Bueneke 25.04.2020: Africanos adivinan acentos de países hispanohablantes. YouTube: https://youtu.be/ 0Trhd8WEYus)

The homogenization of the variation is one of the most frequently mentioned phenomena in such comments. Many people who have encountered different varieties of Spanish in foreign countries connect one accent with one nation. This is due to the prevailing ideology of one state/one language that spread after the French revolution all over the western(ized) world.

Many other comments address the importance of adopting EQ Spanish into the Hispanic world.

MéxicoVisión✶✶✶ "Tenemos que integrar a Guinea Ecuatorial al mundo hispano" [we have to integrate EQ to the hispanic world]. (Monanga Bueneke 25.04.2020: Africanos adivinan acentos de países hispanohablantes. YouTube: https://youtu.be/0Trhd8WEYus)

YouTube functions here as a superurban space, where speakers of different languages and different varieties meet and exchange knowledge. This comment received 252 responses and it was liked 4002 times. In fact, MéxicoVision✶✶✶'s comment encouraged other users engage in a conversation to affirm and reaffirm the claim the user made, and served as the main thread for this discussion. Not many linguistic features were named in this particular video, nor in the comments that belong to it. However, many users engaged in the conversation, either agreeing with this person or also adding further information qualifying their agreement or disagreement. Most of the users leaving a comment in the comment section expressed that they were amazed and fascinated by the fact that Equatoguineans speak Spanish as their L1 or L2, just like many other multilingual countries in South America.

Users also expressed how important the linguistic and cultural diversity of each South American country is, not only at the societal level, but also due to its implications for linguistic variation.

José✶✶✶ "En el Perú, país hispanoblante, donde el 87,5% de los peruanos hablan español como lengua materna y el 12,5% de los peruanos bilingues lo hablan muy bien (además 1 millón de inmigrantes venezolanos en el Perú hablan español como lengua materna, es decir 3% de los residentes en el Perú); existen tres principales acentos: español costeño, español andino y español amazonico"[in Peru, a Spanish-speaking country, 87.5% of the inhabitants speak Spanish as their native language and 12.5% are bilingual but speak it very well (in addition to that, Spanish is the mother tongue of 1 million Venezuelan immigrants in Peru, which means that 3% of the Peruvian population is made of Venezuelan immigrants). There are three principal accents: coastal Spanish, Andinian Spanish and Amazonian Spanish] (Monanga Bueneke 25.04.2020: Africanos adivinan acentos de países hispanohablantes. YouTube: https://youtu.be/0Trhd8WEYus)

Jose✶✶✶ speaks about bilingualism, migration, and the results of these phenomena. For him, it seems relevant to point out that some people in Peru only speak Spanish, while others are bilingual (in Spanish & an autochthone language). He even names different dialectal zones, though he does not provide any information as to why these dialectal zones are relevant here.

To summarize, we can say that Monanga's comments about the linguistic varieties provided an etic perspective. Linguistic features were not overtly named, with the exception of certain features of Argentinian Spanish. The comments, which provide the emic perspective, also did not make mention of specific linguistic features. However, they did offer key information about the homogenization of dialectal differentiation, which is very common to migrants who come in contact with other varieties in a decontextualized environment.

This video did provide information that could be used for the interpretation of the perceptional patterns of different accents. Both Monanga and her guest attempted to imitate the Mexican accent in the video. This was not included in this publication due to the length of the paper and its explorative focus. However, we plan to later transcribe these imitations at the segmental and suprasegmental level, since we hypothesize that suprasegmental information is key to interpreting information for Bantu speakers.

#### 5.3 Equatoguinean Dialects According to L1 (Etic Perspective from Other Speakers)

The video chosen for this part of the study is called Así hablamos español en Guinea Ecuatorial—Pronunciar en español nos cuesta? This video resembles an online or TV quiz. The first part made use of audio materials in which different accents could be heard, while the second provided possible options from which viewers could choose. In the final section, the answers were revealed.

The YouTuber herself did not provide any explicit information about salient linguistic features. To find the etic perception of the Equatoguinean dialects, we thus rely on the comments that Spanish-speakers around the "global neighborhood" provided. Ren✶✶✶✶ for example, named the following features that, in their eyes, were salient:

"Ren✶✶✶✶ 1.- Pronuncia remarcando la ""r"" en algunas palabras. 2.- Aspira un poco los sonidos de la ""g"" y la ""j"". 3.- Remarca la ultima vocal al final de cada palabra. 4.- Omite un poco sonido de la ""d"". 5.- Aspira la letra ""r"", le da un sonido parecido a la ""d"". 6.- Hace pocas pausas habla muy de corrido. 7.- Remarca la 'd"", pocas pausas y muy rápido. Para mí es difícil identificar las etnias, pero si noto sus acentos distintos." [1. Highlights the "r" in some words. 2 Aspirates some sounds, the "g" and the "j". 3. Stresses on the final vowel of each word. 4. Leave out the sound "d". 5. Aspirates the letter "r". 6. Makes almost no pause. 7. Stresses the "d," takes little pauses and speaks very fast. For me it is difficult to identify the tribes but not the variants] (Monanga Bueneke 05.11.2020: Adivinando la etnia con acentos en español de Guinea Ecuatorial. YouTube: https://youtu.be/W2l\_\_LI6g4w)<sup>18</sup>

It is very interesting to see that this user is able to name certain features that they found to be salient or divergent from other accents.

Audio number 5 in the video was a Bubi speaker. As mentioned earlier, Bubi speakers tend to pronounce /ɾ/ and /r/ in a different way to most Spanish dialects. They sometimes leave this sound out or replace a simple vibrant [ɾ] with a double vibrant [r]. Bubi speakers also tend to pronounce the simple or double vibrant as [d] or even [t], depending on the constellation in which these sounds appear.

Another user was also able to detect the Bubi speaker by identifying their use of the phoneme /ɾ/,

Ariel✶✶✶ El quinto es bubi su ″R″ . [The fifth is Bubi. His "R"] (Monanga Bueneke 05.11.2020: Adivinando la etnia con acentos en español de Guinea Ecuatorial. YouTube: https://youtu.be/ W2l\_\_LI6g4w)

Unfortunately, this comment does not provide any contextual or biographical clue as to where the user comes from. The same can be said about the earlier comment that summarized all the features the person was able to identify. Both users named their accounts with either their real or fictional real names. None of both had typical Equatoguinean last names19 so we can only assume that they are Spanish speakers; their specific location and entire language history is impossible to detect.

The last video shows that the Spanish-speaking world still has not yet enregistered the Spanish dialects spoken by Equatoguinean. However, this doesn't mean that the Spanish-speaking world isn't interesting in adopting another dialect. In fact, the comments under Monanga's videos suggest that the opposite appears to be true. Regardless of the topic of each video, most of the comments she receives refer to the Spanish spoken in Equatorial Guinea as the "missing African brother." Monanga is contributing to closing this gap of knowledge by providing information about her home country and its languages.

List of accents: 1. Fang, 2. Fernandino (Pichi), 3. Bisió, 4. Ndowé, 5. Bubi, 6. Ndowé, 7. Annobonés (F'Adambó).

It is important to note that most Annobonenses have Spanish surnames, so we cannot make any assumptions about the origin of the people commenting here.

### 6 Conclusion

In this collaboration we provided a methodological scaffolding with which study the inner, subjective perceptions of repertoire. The analysis of Monanga's linguistic repertoire shows that linguistic features are enregistered (in indexical degrees) online when they have a social relevance for the individual. The dialects to which Monanga was exposed most frequently are thus enregistered in her repertoire, and she is able to name them, indicating 1st degree indexicality. Even though she was not able to correctly guess the Argentinian dialect in the experiment, she was able to name its salient features, which demonstrates that she is indeed aware of the variation, indicating 2nd degree indexicality. She also stated that linguistic awareness of different Spanish dialects is not very high in EQ. However, Equatoguineans are very aware of the different features that characterize the Spanish spoken by other members of the population as according to their ethnic background.20

Regarding the concept of the supervernacular, we can confidently state that languages with a wide territorial extension (and digital coverage in, e.g., social media) such as Spanish present a high internal linguistic variation and, as the data in this paper has evidenced, social agents are very aware of the variation that results from diversity. Within the comments analyzed here, we also found evidence of a homogenization of linguistic features. Comments evincing the etic perspective tended to give vague and general information about a dialect that is thought to be spoken at a national level. It is important to highlight that it was mostly Americans (south and north) and African Spanish-speaking people who seemed to be interested in negotiating what Spanish is. Spaniards make up only a marginal part of the comments. We can thus interpret a continued legitimization in the ex-colonies.

Migration and the "global neighborhood" known as social media provides social agents with a virtual space that causes them expand their general linguistic knowledge, a skill acquired through socialization. Normativity is negotiated on YouTube, as comments show. Just like standardized languages, which are built upon institutionalized artefacts—dictionaries, rules of grammar, etc.21—supervernaculars are also negotiated across social media, in our case YouTube. Language legitimization that relies on these institutionalized artefacts thus loses its importance in such contexts. These virtual places allow non-linguists to create their

These enregisterment processes are being further studied in our doctoral project at the University of Mannheim.

Consult Agha (2007) for an overview of enregisterment processes, which are mainly studied by analyzing different kinds of artefacts produced by linguists and other experts in the Aufbau stage of a standardized language.

own definition of an "standardized" variety. This supervernacular is characterized by the way it allows and welcomes variation, and its recognition of the differences and similarities present in all Spanish dialects. The legitimation given by other social agents has special value for Equatoguineans, who, despite having their own academy of Spanish, are still not included in the pluricentric discourse since several criteria remain unfulfilled. We can thus witness the ongoing process in which the Equatoguinean dialects are negotiated and established as an integral part of the Spanish-speaking world.

Linguistic awareness, on the other hand, seems to expand when individuals migrate to other countries. This confirms Caravedo's (2014) claims. Monanga herself relied on her previous linguistic knowledge to adapt to her new settings. However, we cannot state that repertoire expansion results from migration alone.

We can summarize that YouTube functions as a superdiverse neighborhood, where people from all around the world can meet to discuss their topics of interest. New normativities are displayed and (re)negotiated, since speakers—as social agents themselves—start recognizing the features of Equatoguinean dialect(s), thanks to the stimuli provided by Monanga.

The videos also provided interesting information about what speakers notice when other people speak Spanish. Just like other research has shown, nonlinguists cannot always overtly name specific linguistic features, since they lack the necessary vocabulary. However, they can use other strategies to reproduce the features that they know are present in a specific accent, e.g. by mimicking it.

In further studies, we would like to expand the analysis of indexical orders by applying a segmental and suprasegmental analysis of accent imitations. It would be very interesting to approach the identity-building processes that underlie the negotiation of the supervernacular. To explore the effects that migration has on the repertoire(s), a longitudinal study is needed.

We would like to close this collaboration by reflecting on the superdiverse paradigm. The conceptualizations and methods arising from this paradigm allow linguists to study (super)diversity from a non-comparatist lens. We would like to note that superdiversity might be understood as a "new" phenomenon, but in fact this is generally the norm in the majority of the world. Diversity was (and still is) an integral part of Europe before the French revolution, and it remains the norm in all ex-colonial countries. With the help of this new paradigm, sociolinguistics can finally start addressing variation as a normality, not a deviation.

### References

Agha, Asif (ed.), Language and social relations, New York, Cambridge University Press, 2007. Alvar, Manuel (ed.), Manual de dialectología hispánica: el español de España, Madrid, Gredos, 1996. Androutsopoulos, Jannis/Juffermans, Kasper, Digital language practices in superdiversity: Introduction,

Discourse, Context and Media 4–5 (2014), 1–6.

Bibang Oyee, Julián, El español guineano: interferencias, guineanismos, Malabo, n.p., 2002. Blommaert, Jan, Supervernaculars and their dialects, Dutch Journal of Applied Linguistics 1 (2012), 1–14. Blommaert, Jan/Backus, Ad. Superdiverse repertoire and the individual, in: Saint-Georges, Ingrid/

Weber, Jean-Jacques, Multilingualism and Multimodality. Current Challenges for Educational Studies, Rotterdam, Sense Publishers, 2013, 11–32.

Blommaert, Jan/Rampton, Ben, Language and Superdiversity, Working Papers 12:5 (2011), 1–36. Bloomfield, Leonard (ed.), Language, New York, Holt, Rinehart & Winston, 1933.

Bucholtz, Mary, Why be normal? Language and identity practices in a community of nerd girls, Language in Society 28:2 (1999), 203–223.

Budach, Gabriele/Saint-Georges, Ingrid, Superdiversity and language, in: Canagarajah, Suresh (ed.), The Routledge Handbook of Migration and Language, New York, Routledge, 2017, 63–78.

Chagas, Andrea, Antiracist, feminist identity on social media. Afroféminas as a community of practice on Instragram, in: Eibensteiner, Lukas/Hesselbach, Robert (edd.), Social Media: Current Issues in Romance Linguistics and Foreign Language Learning and Teaching, i.p.

Caravedo, Rocío, Percepción y variación lingüística. Enfoque sociocognitivo, Valencia, Iberoamericana/ Vervuert, 2014.

Dittmar, Norbert (ed.), Grundlagen der Soziolinguistik - Ein Arbeitsbuch mit Aufgaben, Berlin, Boston, De Gruyter, 1997. https://doi.org/10.1515/9783110938463

Dürscheid, Christa, Medienkommunikation im Kontinuum und Mündlichkeit und Schriftlichkeit. Theoretische und empirische Probleme, Zeitschriften für angewandte Linguistik 38 (2003), 37–56.

Eckert, Penelope, Linguistic Variation as Social Practice, Oxford, Blackwell, 1999.

Gugenberger, Eva, Theorie und Empirie der Migrationslinguistik. Mit einer Studie zu den Galiciern und Galicierinnen in Argentinen, Wien, Lit Verlag, 2003.

Gumperz, John (ed.), Discourse strategies, Cambridge, Cambridge University Press, 1982.

Hoinkes, Ulrich, Indexicality and Enregisterment as Theoretical Approaches to the Sociolinguistic Analysis of Romance Languages, Online, https://macau.uni-kiel.de/receive/macau\_mods\_ 00000003, 2019.

Hymes, Dell, Models of interaction of language and social life, in: Gumperz, John/Dell, Hymes (edd.), Directions in Sociolinguistics: The Ethnography of Communication, London, Blackwell, 1974.

Johnstone, Barbara, Enregisterment: How linguistic items become linked with ways of speaking, Lang Linguistic Compass 10 (2016), 632–643.

Krefeld, Thomas, Einführung in die Migrationslinguistik. Von der Germanaia italiana in die Romania multipla, Tübingen, Narr, 2004.

Krefeld, Thomas/Putska, Elissa, Für eine perzeptive Varietätenlinguistik, in: Krefeld, Thomas/Pustka, Elissa (edd.), Perzeptive Varietätenlinguistik, Frankfurt am Main, Peter Lang, 2014, 9–30.

Labov, William, Principles of Linguistic Change, Vol. II, Social Factors, Oxford, Blackwell, 2001.

Labov, William, Principles of Linguistic Change, Vol. III, Cognitive Factors, Oxford, Blackwell, 2010.

Leppänen, Sirpa/Häkkinen, Ari (edd.), Buffalaxed superdiversity: representations of the other on YouTube, Diversities 14:2 (2012), 17–33.

Leppänen, Sirpa/Peuronen, Saija/Westinen, Elina, Superdiversity perspective and sociolinguistics of social media, in: Creese, Angela/Blackledge, Adrian, (edd.), The Routledge Handbook of Language and Superdiversity. An interdisciplinary perspective, London, Routledge, 2018, 30–42.


### Nélida Murguía Cruz Glottopolitical Interventions and Virtuality in Spanish Language Teaching for Senegalese Immigrants during the Pandemic (Buenos Aires, 2020–2021)

Resumen: Este artículo forma parte de una investigación en curso sobre intervenciones glotopolíticas realizadas por espacios educativos, populares e institucionales, para la enseñanza del español a migrantes senegaleses en Buenos Aires. En este caso, me centro en el curso de español del Bloque de Trabajadorxs Migrantes (BTM), a través del canal de YouTube Damay Jaangë Español, y en el curso realizado desde la Universidad de Buenos Aires (UBA), el cual está enmarcado en un proyecto productivo textil, a cargo de la Fundación Nobleza Obliga.

Los datos para esta investigación provienen de un acercamiento etnográfico y discursivo en los espacios de enseñanza de español. Además, este trabajo se inserta dentro de la glotopolítica (Arnoux 2012; Del Valle 2007), entendida como una perspectiva interdisciplinaria en los estudios del lenguaje, que implica interrogar el tema de estudio tanto desde su dimensión lingüística como desde su dimensión sociopolítica.

En el artículo, presento una caracterización sociolingüística de la migración senegalesa en Buenos Aires considerando su transnacionalidad. Posteriormente reviso algunas de las estrategias glotopolíticas de esta comunidad y su relación con las Tecnologías de la Información y la Comunicación (TIC). En el segundo apartado, caracterizo los cursos de español como intervenciones glotopolíticas, atendiendo al contexto en el que surgieron y señalando su relación con la educación popular. Finalmente, describo el proyecto del BTM Damay Jaangë Español y el curso de español realizado por WhatsApp para el proyecto textil, revisando sus contenidos y la manera en que hacen uso de la virtualidad y las TIC.

Note: This contribution has been translated by Agustina D'Andrea.

Nélida Murguía Cruz, CONICET–Universidad de Buenos Aires

### 1 Introduction

This paper is the result of an ongoing research on glottopolitical interventions carried by a group of instructors working in educational spaces, both popular and institutional, teaching Spanish to Senegalese immigrants in Buenos Aires. Being one of the researchers in charge of the study, I will present on this occasion an article focused on two courses: a Spanish course offered by BTM/Bloque de Trabajadorxs Migrantes (Block of Immigrant Workers), held on the YouTube channel Damay Jaangë Español, and a course conducted with the help of the professorship in Grammar "C" at the Faculty of Philosophy and Letters of the Universidad de Buenos Aires (UBA), which was part of a textile production project in charge of Fundación Nobleza Obliga (Nobleza Obliga Foundation).

Multiple restrictive measures have been decreed in Argentina since March 2020 in order to contain the outbreak and minimize the spread of the COVID-19 virus. One of such measures was the suspension of face-to-face classes in all the educational levels and the consequent use of digital platforms to support the teaching process. In the case of immigrant Senegalese students, some of the educational approaches involved the implementation of virtual teaching modes on video conferencing platforms or social media.

The data for this research was collected using an ethnographic (Cassels/Ricento 2013) and discursive (Arnoux 2012) approach within the Spanish teaching spaces. In the case of BTM, I am a member of the Spanish commission and I participate in the production of Damay Jaangë Español. As for the textile project, I collaborated with the implementation of the course, being an intermediary between the professorship and the organizers of the course, and interviewing the Senegalese participants of the project.

This research is set in the framework of Glottopolitics (Arnoux 2012; Del Valle 2007), understood as an interdisciplinary perspective within language studies, which implies questioning the subject of study both in its linguistic and sociopolitical dimensions, in order to understand it and explain it. Glottopolitics primarily explores interventions in the public space of language, linguistic ideologies and its relation and effect in power structures and social relations.

I will present hereafter a sociolinguistic description of Senegalese migration, considering its transnationality. Then, I will examine some of its glottopolitical strategies and its relation with Information and Communications Tecnhologies (ICTs). In the second section, I will describe the Spanish courses as glottopolitical interventions, contemplating their context of emergence and their relation with popular education. Finally, I will describe the project by BTM Damay Jaangë Español and the Spanish course imparted via WhatsApp for the textile project, and then I will look into the contents of both courses and the way ICTs and virtuality are implemented in each of them.

### 2 A Transnational Linguistic Community

Senegalese migration occurs because of economic reasons, and it is voluntary. In Argentina, it started in the 1990s, and was increased in the 2000s. ARSA, (Association of Senegalese Residents in Argentina) estimates a number between 4 and 5 thousand people (Zubrzycki 2018). This population consists mostly of young men who have migrated as part of a family strategy to send back remittances.

Religion is a central aspect in the community and it is very important in the construction of migration networks (Pedone 2010). Almost every immigrant belongs to the Mouride brotherhood, which falls under Islamic Sufism.<sup>1</sup> Besides that, Senegalese ethnic diversity is also present in Argentina: most of them are Wolof, but there are also Jola, Serer, and others, as it has been observed in the fieldwork process.

Senegalese migration is characterized by its transnationality (Espiro 2020; Zubrzycki 2018; Kleidermacher 2016). Its members maintain social practices and contact with their communities of origin. They might return to their country, be it regularly or permanently. An analysis based on this approach implies the acknowledgement of relations and social networks which function as a link between destination and origin, beyond Nation-State borders (Pedone 2010; Levitt/ Glick-Schiller 2004). Moreover, an understanding of the concept of community beyond its territorial dimension (Canales/Zlolniski 2000) leads to considering the members of a group able to keep on being part of a community, including its linguistic aspect, even though the members reside in different places.

In Sociolinguistics, the definition of community has several edges, which Duranti (2000, 122) problematizes by defining it as "the product of the communicative activities engaged in by a given group of people". This standpoint allows to go beyond both a monolinguistic perspective and territorial determinism. In this framework, a transnational linguistic community would be one whose members identify themselves as part of the group, keep constant interaction and participate in recurring communicative and discursive practices. It also reunites multilingual or monolingual members, who interact face-to-face or remotely, from different places of residence.

The Senegalese community is, as already noted, a transnational group of people, determined by a dynamic plurilinguism resulting from their mobility, their ethnic and religious affiliation, and their educational and migratory trajectory.

One of the approaches to define Sufism states that it constitutes the mystical spiritual dimension of Islam, and that it revolves around the figure of a leader or saint (Gázquez 2021); in this case, Cheikh Amadou Bamba.

On account of their religious culture, some of the members are literate in Arabic, although the use of such language is restricted to the reading of the Quran and other sacred texts. Because of colonization, the official language in Senegal is French, but such language is only mastered by those who have pursued formal education. Languages prior to colonization have more presence in the community and they usually correspond to ethnicity: Wolof –also as a vehicular language–, Djola, Serer, Toucouleur, Mandinka are some of the main examples (Jabardo 2013). This combines with transnational mobility to and from other countries, which leads to incorporating, in different degrees, other languages to their repertoires. Such is the case of Portuguese for some Senegalese people residing in Argentina, who participate in a mobility network with Brazil (Espiro 2020).

Despite this broad linguistic repertoire, a low rate in literacy<sup>2</sup> in the Latin alphabet and a nonstandard competence in Spanish constitute challenges to this community within Argentina. According to Senegalese researchers and based on the fieldwork, most of the members of this community are not functionally literate (Jiménez 2005). This may restrain them from accessing some spaces which are characteristic of countries with a rate of literacy as high as that in Argentina. In addition, their literacy is affected by the growth of some new communicative practices in writing, allowed by new technological resources such as internet trading and online administrative procedures.

#### 2.1 Glottopolitical Strategies and ICTs

Senegalese migrants have developed several community strategies to participate in the host society (Kleidermacher/Murguía 2021), many of which address the problem of using a variety considered nonstandard. According to Bourdieu (2011) strategies are actions of a certain degree of systematicity and recurrence which lean on previous knowledge, cultural background and ability to endure unfavorable situations.

In Murguía (2023), I examined those actions where language use is central and contributes in transforming the power relations of the Senegalese community within the host society. That is to say, such actions constitute glottopolitical strategies for learning, communication and visibilization. Migratory tradition, which is based on trajectories, networks and sociocultural capital, is core to such strategies and allows the members of this community to migrate to countries with a language unknown. Some of the strategies include teaching the newcomer a repertoire of

In Senegal, literacy rate is under 52% (Agence Nationale de la Statistique et de la Démographie, 2016).

phrases and words for trading, aid from literate peers (or those more competent in Spanish) to carry out administrative procedures, immersion learning, and attendance to language courses, although formal study is not a usual practice.3

I will for now focus on those strategies related to the use of ICTs, keeping in mind that such technologies have allowed for the emergency of transnational communities (Melella/Perret 2016). I am particularly interested in those linked to the internet, which have a key role in the strengthening of some transnational relations characteristic of migratory dynamics. Learning the language of the receiving country (Brignol 2015) is one of them, and it is greatly relevant, considering that this kind of resource has been used in Spanish courses in order to maintain education services in a context of confinement as a result of the pandemic.

The new ICTs have allowed for a global expansion of communicative practices for Senegalese immigrants, who constantly use their cell phones in everyday life, as has been noted in the fieldwork. Such devices are used to stay connected with their families and friends through social networking such as Facebook and WhatsApp; to perform religious activities such as listening to khassidas<sup>4</sup> or receiving reminders to read the Quran from an app designed for this purpose; for remittance sending, also from a dedicated app; for entertainment, by watching YouTube or Facebook videos on contents relevant to the community; and to build and sustain digitally supported practices of association through groups, websites and channels on social media.

Some of these activities are part of glottopolitical strategies of communication and visibilization before the host society. One very important is the creation of a virtual community through channels implemented from Argentina with the purpose of being connected and informed. Those are mainly YouTube channels (usually synchronized with Instagram and Facebook), which can be considered communitary and alternative means of communication.

Touba Argentine TV is a TV programme broadcast on a YouTube channel created in 2017. It has almost 20 thousand followers and has published 854 videos.<sup>5</sup> It has a primarily religious content, but it is now also possible to find in it interviews, events, debates and reports on different topics, which have been oriented to problems related to the pandemic in such period. The programme also

The number of students in different spaces identified in fieldwork as having attended language courses at least once is under 500 in AMBA (Buenos Aires Metropolitan Area), which makes them a minority.

Religious poems written by Cheikh Amadou Bamba (founder of the Mouride brotherhood), and recited in religious gatherings.

The channel website is https://www.youtube.com/channel/UCtAGhzyX-8Wiidn3VPoCFbw/fea tured. These data correspond to January 2022, as well as the succeeding.

publishes Senegalese business and entrepreneurship. The content expresses the plurilinguism present in this community and their preference for audiovisual means: most of the videos are in Wolof and the main receivers are Senegalese people living in Argentina, but they are also addressed to people living in Senegal and other countries, in order to show them the characteristics of the lifestyle and activities here. Most of the titles of the videos are in French, and even some in Portuguese. From 2019, content in Spanish has escalated, and Argentines or other immigrants connected with the community (such as lawyers or activists from different organizations) have been invited to the programme. This is a transnational and regular practice: many other channels can be found in different countries, e. g. Touba Brasil TV, Touba Allemagne TV, among others, all named after the sacred city in Senegal, home of Mouridism.

There are two more channels, whose manifested purposes are to connect with the host society, to spread Senegalese culture and relevant topics, to fight stereotypes and to discuss problems with the community. In the Autonomous City of Buenos Aires (CABA) there is a channel named Big Dealer in the House,<sup>6</sup> created in 2015, with 4,360 followers and 63 videos. Content includes interviews, reports, shorts films, and videoclips, many of which are in Spanish, especially since the end of 2018. Topics like discrimination, Afro-Argentinity, rights, language, integration, and behavior in the country, among others, are discussed on this channel.

Lastly, there is a channel by a youtuber named El Negro Jimmy, which was created in 2017. It has 11,600 subscribers and 17 videos.<sup>7</sup> It is based in La Plata, and is geared towards a Spanish speaking audience. It promotes commercial ventures and broadcasts interviews with prominent members of the Senegalese community, as well as conversations with Argentine people and content against institutional violence in the abovementioned city.

With reference to practices of constant communication within Argentina, the use of WhatsApp stands as a generalized strategy which allows the members of the community to solve immediate difficulties and rising opportunities within the host society. Referents and members of the two main organizations, ARSA and Act of the Senegalese in Argentina,<sup>8</sup> note the involvement and organization they

The production team also participates in Touba Argentine TV. Since 2019, the videos are recorded in "the first African studio in Argentina, Massar Ba studio, a homage to a great fighter for the Senegalese community" (Big Dealer in the House 2019). Link to the channel: https://www.you tube.com/c/bigdealerinthehouse/featured.

The channel website is https://www.youtube.com/channel/UCEN2yKw91YZbEAoAngGbFYg.

There are two main organizations: ARSA, characterized by an institutional stand, and the Act, which emanated from the Association of Dahira Mouride. Most of the Senegalese people interviewed somehow participate in both associations (Kleidermacher/Murguía 2021).

have achieved by means of WhatsApp groups, where –they claim– all Senegalese people in Argentina participate.

Likewise, many of them participate in other WhatsApp groups according to territorial criteria. Through these means, they share relevant and immediate information by recording audio messages, sending images and, to a minor extent, writing. In the fieldwork, I have noticed, for example, a use of WhatsApp audios for a live record and report of important events, such as speeches by referents, or to report police violence cases. I have also observed that a great number of people use headphones constantly, whether they are chatting or listening to audio messages, even when participating in face-to-face communicative practices.

Finally, I will mention another use of the ICTs which is relevant to the present research: the utilization of Google Translator (between French and Spanish) and language teaching apps, such as Duolingo (also between French and Spanish), as a supplement to their learning in the streets or in language courses. As with other practices in which reading and writing abilities are necessary, this method is only accessible to those who have had official school instruction in some degree, and are therefore literate and competent in French, given that, unlike Facebook, Google has not incorporated Wolof translation yet.9 It is also necessary for these purposes to have a certain inclination to self-learning and some digital literacy.

To summarize, this migratory group holds a preference for WhatsApp for immediate communication, both with their families and the community in Argentina. They have also developed more elaborate practices such as communication on YouTube channels (synchronized with Facebook and Instagram accounts). All these practices are carried on cell phones, since the members of the community do not usually own computers or tablets. The use of other apps (e. g. Zoom), which have become essential during the pandemic, is instead uncommon, as is the use of e-mail and browsers.

The outlined glottopolitical strategies carried by the immigrants help them manage not only linguistic but social differences as well, by allowing for a less uneven participation and producing a different kind of relations with the host society.

It is already possible to translate from Wolof to other languages and vice versa on Facebook. This can be done by means of Translate Facebook (https://facebook.com/translations), an app where translators and multilingual speakers do collaborative translations online and vote for them, what is called crowdsourcing (Jiménez-Crespo 2017). Such method has enabled this social media site to grow around the globe, without hiring professional translators. There is an official group in which the social media community can discuss about the translations, called Translate Facebook into Wolof (https://www.facebook.com/groups/779265972167467/).

### 3 Spanish Courses as Glottopolitical Interventions

#### 3.1 Glottopolitical Interventions

I propose to think of the Spanish courses examined in this paper as glottopolitical interventions, retrieving some aspects proposed by Arnoux (2000), Guespin/Marcellesi (2019) and Calvet (1997). Glottopolitical interventions are actions on language which have an impact on its social distribution. Such actions might be performed on speaking, on discourse, on the different languages, or on one of them, its corpus and its status. They comprise a range from small, domestic actions to public and visible interventions. They might as well constitute empowering or vindicative actions, or, on the contrary, regulatory interventions. Besides, they might be performed by the population, as practices to solve communication problems or to carry a procedure independently, or by power centers, as is the case with linguistic policies.

The interventions that I will examine are motivated by some linguistic ideologies which share an assumption that language is a powerful tool to make sociopolitical changes, and, in that sense, seek an empowering effect (Murguía et al. 2022). The courses are thought to allow for a change in power relations between the Senegalese community and the rest of the Spanish-speaking society in Argentina by enhancing their linguistic repertoire through language teaching. People who run and attend the courses assume that learning Spanish might improve their living conditions in the host society, whether to have a more active participation in it or to accede to more formal employment opportunities.

To describe the two courses at issue as glottopolitical interventions, it is necessary to consider, even if shortly, their motivation, and an outlook on the period when they emerged. That is so because glottopolitical interventions are "expressions of economic, social and political processes on which they besides act dialectically" (Arnoux/Bein 2015, 13).

#### 3.2 Spanish Teaching: Between Institutional Violence and Popular Education

The Argentine State does not require speaking Spanish as a condition to live and work in the territory, but it does not guarantee that the immigrants learn it, so it depends essentially on their interest and possibilities. The supply of language courses for economic migrants and refugees as a public policy is limited, especially if they are African. It was not until 2018 that state initiatives on education for this community were documented, even though migration legislation specifies that the State must promote the implementation of these courses for non-Spanish speaking immigrants (Art. 14, Law 25871). The educational aspect and the administration of justice are central axes of the relation with society, but neither have been attended to until very recently.

The first courses addressed to this community were community projects or projects by political and social organizations with a perspective of popular education; some institutional courses appeared shortly after (Murguía et al. 2022). All of them emerged in a scenario of institutional violence against Senegalese sellers, and in a context in which the administration of justice suffered a setback in terms of human rights (Murguía 2022).

The climax of this problem occurred with the issue of the DNU (Decree of Necessity and Urgence) 70/2017, which criminalized immigrants and was declared unconstitutional by the judicial power (CELS 2018). In March 2021, it was revoked. This DNU was an expression of how migratory policies re-entered the domain of security and control (Penchaszadeh/García 2018). While urban control raids against the workers of popular economy have a long history, there was a major increase in CABA, marked by police violence, in 2015 (Rabasa 2021; Penchaszadeh/García 2018), and a media peak in 2018.<sup>10</sup> Almost every documented Spanish course started that year, which allows to think of language teaching as part of a response to institutional violence.

A common aspect of both courses here discussed is their identification with popular education, which strengthens them as glottopolitical interventions. Most definitions highlight their relation with the working class, as well as the topics discussed (knowledge linked to the world of work, for example) and their aims, particularly the contribution to a process of fighting, change and empowering (Gómez 2015). In accordance with Cano (2012), these educational practices are determined by three aspects, which will be considered in the present analysis: a pedagogical dimension, associated with teaching methods; a political dimension, related to change in popular sectors; and an ethic and methodological dimension, which establishes coherence between the purposes of such practice and the means of construction, transmission and socialization of knowledge.

In a preliminary and ongoing analysis on the appearances of the Senegalese community in the written press, published between 2006 and 2020, we have found that, from a total of 238 articles, 59 (25%) denounce police violence. 18 of those 59 (30%) were published in 2018.

# 4 Bloque de Trabajadorxs Migrantes (Block of Immigrant Workers), Damay Jaangë Español

The Block of Immigrant Workers (BTM) is a self-managed and independent organization created in 2017 after the issue of DNU 70/2017, in order to organize and fight for its revocation. In 2018 and 2019, we (BTM) imparted free classes to Senegalese people, in a space for the memory of the civil-military dictatorship in Argentina, in the neighborhood of Flores, a strategic location of work and housing for the Senegalese community. During those years, about 90 students participated in one or other weekly course. Teachers had diverse professions, and most of them came from Latin American countries. This space was home for a community bonded by the migrant identity and the fight for rights. Spanish classes also resulted in the creation of a commission of assistance in police violence cases.

The course has a popular education approach, and is different from language classes imparted in schools or institutes, whose declared purpose is usually the acquisition of the standard variety of a second language. In this case, the aim is to provide the students with tools for a more active social and political participation in the host society.

When confinement was decreed in Argentina, classes were interrupted, and they have not yet restarted in a face-to-face modality, because of the difficulty in obtaining physical spaces. With the aim of maintaining the learning experience, we discussed some options, like imparting synchronous classes via WhatsApp or any videoconference platform at a scheduled time, or creating materials thought for being used independently at any time. We finally opted for the latter, and in 2020 we created a YouTube channel called Damay Jaangë Español (I learn Spanish, in Wolof). There we have uploaded learning videos specifically designed for beginner Spanish-learning speakers of Wolof.

The team imparting the classes is heterogeneous: there are linguists, teachers of Spanish, audiovisual professionals (some of whom study at the Argentine public film school), a Senegalese professor of Spanish who contributes remotely, and some Senegalese peers who help translate and participate on the videos. The project received support from UN High Commissioner for Refugees (UNHCR), which provided a computer for editing.

So far, a season of 12 videos of approximately 5 minutes each has been launched. A second part is under production. They are published on YouTube, but also Facebook, Instagram and WhatsApp. Accounts have about 800 subscribers and some videos have over 3,000 views.

#### 4.1 Content and Features

It is convenient to remember that there is a general lack of Spanish learning materials for this specific type of populations, since those are generally intended for university students, professionals or Western tourists (Murguía et al. 2022). Thus, it is necessary to design a content with specific characteristics. I hereafter present a list of the titles of the videos, to subsequently highlight the details of the content.


Videos of less than a minute were also published on Facebook and Instagram. There, some brief grammar explanations and simple contents were resumed.

The project draws from a communicative approach for learning Spanish. As for the audiovisual production, we created communicative scenes with actors. They included examples of conversations on the main topic of the video, and then an explanatory summary. For instance, in video 4, three scenes occur in which characters are in different situations where it is possible to use "buenos días", "buenas tardes", "buenas noches", "hola" and "chau" ("good morning", "good afternoon", "good evening", "hello" and "bye"): shopping in a pharmacy, receiving a delivery, meeting someone on an elevator.

Translation to Wolof is a second feature on some of the videos, although in general we have chosen to teach only from Spanish, since using audiovisual support and retrieving knowledge from the students make it possible. On the videos for the second part of the course, which are still unpublished, drawing on Wolof will be more relevant. Because of this, some Senegalese partners have been included in the design of lesson plans and recording, in addition to their preexisting collaboration (recording of audio and video for the final edition).

A third feature of the videos is complementary written support. It is not a key characteristic, but it might be useful to those students literate in the Latin alphabet, and even to those who are not, because it may help them relate spoken words with their representation in spelling. Accordingly, joint effort with the audiovisual team, especially graphic design, is relevant to allow the students to distinguish language functions through visual strategies such as different colors, typographies, etc., as shown in Figure 1.

Figure 1: Video frame: Yo soy senegalés (I am Senegalese) (Damay Jaangë Español, 2021, 1m13s).

On another note, many of the teachers are immigrants, but we have sought to adapt to River Plate Spanish, particularly regarding vocabulary and voseo (the use of "vos" as second person singular pronoun, instead of "tú", used in much of the Spanish speaking world), because this is the standard in the host location. However, varieties from the places of origin of the teachers emerge, mainly in pronunciation. We do not consider learning other varieties of Spanish as a disadvantage, given that Senegalese immigrants often cohabit with (or have the possibility of cohabiting with) other Latin American immigrants. On video N°6, whose topic is nationality, the verb "ser" (to be) and grammatical gender, we exploited this circumstance, asking people related to BTM to send a video with both grammar structures: "soy de [country]" (I am from . . .) and "soy [nationality]" (I am . . .). This also has a political dimension, inasmuch as we have sought to strengthen the immigrant identity as part of the host community, which comprises different voices.

Another sociopolitical aspect is the intention to practice an antiracist, anticolonial gender perspective. As regards gender, we have avoided to reproduce stereotypes by, for example, presenting a woman who is looking for a job as a mechanic, while there is a man working as a nurse, in video N°8 on jobs and professions, the verb "ser" and grammar gender. The antiracist perspective materializes on the video by avoiding an exclusively white representation of bodies in illustrations and including indigenous and black persons, as seen in Figure 1.

The anticolonial perspective is practiced by including references to more than just the European or the North American cultures, as is often seen in regular learning materials. For instance, the music and sounds on the videos have been composed by a Senegalese musician who participates in the project. In addition, a video about food and singular/plural is being produced using a Senegalese recipe as example. In this vein, we have sought to develop contents that might be useful to the community and reflect their everyday life, e. g. administrative procedures, interviews, filling of forms, etc.

#### 4.2 Assessment of the Virtual Mode in the Course

Offering the course in a virtual mode leads to some advantages related to the platform of choice, YouTube. First, the students have the possibility to watch the videos on a cell phone, without the need of a computer or tablet (which they usually do not own), or written materials such as books and booklets. Besides, the audiovisual support allows non literate students to benefit from the content. Lastly, Senegalese immigrants make regular use of this platform, so they are familiar with it.

However, this decision weakened the emotional an political bonds between the attendees, although it also strengthened their relationship with the members of the current project. Furthermore, the virtual mode has expanded the reach zone of the content by allowing more people, in any territory, to access it. The videos are thus intended to work as a supplement for future face-to-face classes.

Another advantage of virtuality relates to the amount of content that can be produced in a short period of time, considering that this space is sustained on voluntary work. This leads to a situation in which content is focused on basic topics, which, nevertheless, can be exploited. It is often necessary to revise grammatical gender, number and verbal inflection, topics which Wolof students find difficult.

### 5 Spanish Courses for the Textile Production Project

At the beginning of 2021, Nobleza Obliga Foundation11 began to promote the creation of a textile production project for the Senegalese community, considering that many of its members are tailors. The project has the support of the International Organization for Migration (IOM) and Argentina's Ministry of Social Development (MDS, for Ministerio de Desarrollo Social). The latter manages the Spanish courses offered as an extra benefit within the project. With the help of ARSA, people interested in the project were gathered. This Senegalese association invited those people with textile experience to participate in the project, by means of WhatsApp groups. The participants are Senegalese people between 25 and 54 years old. 55% of them reside in CABA, and the rest of them reside in the suburban area of Buenos Aires. At the moment of the survey carried by MDS in 2021, 74% work informally as street vendors.

Against this background, the aim of the project is to offer options and tools for the insertion of this group of Senegalese people in the formal labor market, in order for them to avoid institutional violence situations related to street selling.

Over the planning stage, coordinators of MDS considered necessary to offer Spanish classes so as to hold a more fluent communication with the members of the project, familiarize them with the Spanish vocabulary related to textile work, and give a workshop on popular economy and cooperativism which may lead to the adoption of such characteristics.12 A requirement to take the Spanish classes was established in order for the members to participate in the project. Their selfperception on their performance in Spanish was moderate for 74% of the people, fluent for 8% and nonexistent for 18% (MDS, 2021).

With the aid of the professorship in Grammar "C" at the Faculty of Philosophy and Letters of the Universidad de Buenos Aires (UBA), in charge of Dr. Claudia Borzi, a group of teachers was created who designed a Spanish course with the characteristics required by MDS. Besides, some Senegalese tutors joined the team. Their function was to offer pedagogical support, to motivate the students and to promote participation.

This foundation is "an organization of Social Innovation which seeks to help ONGs, Businesses and Social Entrepreneurs find innovative solutions to current social problems" (Nobleza Obliga, n. d.). Its head office is located in Buenos Aires and they work with a crowdfunding model.

It is to consider that the Mouride community follows a model of independent work, in which each person obtains their own merchandise and keep the profit from the sales. Although they have solidarity networks in case of unfavorable circumstances, and they are supportive of the newcomers to help them make their first sales, the cooperative model is not usual.

The teachers assumed their task with a perspective in popular education:

university should accompany social change processes linked to public policies and actions carried out by popular sectors [. . .]. We are joining the project from a political-pedagogical perspective in accordance with the Freirean ideas on a popular education which is transforming, democratic and a producer of popular protagonism (Blasco/Saavedra/Murguía/Guitérrez 2022).

During 2021, classes were held twice a week, they lasted an hour, and they took place in the evening, when the students had finished their working day. Four groups were created according to the level of literacy of each student: one beginner course, two intermediate and one advanced. For the learning process, the teachers decided to use WhatsApp and a printed booklet, as well as images, audios and videos. It is worth noting that 37% of the interviewees claimed they were not skilled in the use of videoconferencing platforms like Zoom.

#### 5.1 Content and Features

The curriculum planning included a division of the course in two stages. The first one (imparted during 2021) was designed to revise basic contents of Spanish and introduce textile vocabulary. The second stage was designed to reinforce contents related to textile work, cooperativism and production. In this sense, this represents a challenge to the teachers, who had to study these topics in order to incorporate them, as simply as possible, in a beginner Spanish course.

The virtual spaces for the meetings were different WhatsApp classroomgroups (Blasco et al. 2022), which were created by the teachers. A synchronous instance of communication with learning activities took place on a specific day and time. Each lesson also relied on previously designed audios and videos, which enable the teachers to incorporate corporeality, a generally lacking aspect in virtual spaces. Other WhatsApp groups were also created in which asynchronous communications were held.

In this case, the work of the Senegalese tutor was fundamental, because not only did he promote participation, but he also explained and translated whatever was necessary during the development of the activities. The team work with a Senegalese tutor is also a glottopolitical decision that has facilitated the teachinglearning process with students and has confirmed the ability of the members of this community to participate as teachers, and not only learners.

The development of the classes went along with the multi-level booklet that the teachers specifically designed (Blasco et al. 2021). It was printed and delivered by MDS to each one of the students. The plan was for each teacher to use the material according to the characteristics and level of their groups. The contents of each unit are the following:


As can be seen, units 1 to 5 include basic contents, like the verbs ser (to be), estar (to be), haber (there is/there are), routine verbs, grammatical gender and number, and vocabulary on jobs. The introduction of specific vocabulary related to joint work and cooperativism occurs in unit 6. Finally, units 7 and 8 revise the previously studied grammar contents and incorporate specific textile vocabulary.

As well as in the Damay Jaangë course, there is in this case an advocacy to adopt an antiracist perspective, avoiding an exclusively white representation in illustrations on the booklet and including indigenous and black people. There is also a gender perspective, by means of exemplifying the verb cocinar (to cook) with a father and son preparing the food. Besides, some jobs related to popular economy (and traditionally unrelated to language lessons) are included, like cartoneros (a person who collects discarded waste, cardboard in general, to reuse or resell) or garbage collectors.

#### 5.2 Assessment of the Virtual Mode in the Course

Again, the advantages of WhatsApp as the chosen support for the meetings have to do with what has been previously discussed about the impossibility of imparting face-to-face lessons. In the first place, it should be noted that WhatsApp was indeed the most utilized means to maintain educational activities in Argentine schools (Narodowski 2020), which indicates the importance of this support. It is viable because it is not necessary to have a monthly plan to use it, given that it comes for free with any credit recharge for a prepaid cell phone. Hence, not only may it be more affordable but it is also a support familiar to the community.

The disadvantages of this means of communication noted by the teachers (Blasco et al. 2022) have to do with the fact that cell phones are used in spaces of distraction, at moments shared with relatives or friends, and even in the streets. Besides, it does not allow for a proper use of corporality and gestures, which are key in a language lesson. However, the teachers devised some options, like the use of videos and the booklet itself, which may help overcome such obstacles. The teachers also reported difficulties in engaging students to interact by sending audios to each other, since they tend to interact exclusively with the teacher.

The use of the booklet also presented some difficulties, because it essentially consists in reading and writing. Besides, it proposes activities based on a Western and classroom-oriented kind of learning, which are frequently taken for granted. Such activities include matching, filling in blanks, writing on the same line or direction of the writing and the booklet (which is opposite to that of the Arabic system of writing). It was also challenging for the students to compose a written production, take a photograph of it and send it to the teachers, as planned. Yet, the use of the booklet formalizes the learning process, organizes the contents in a progression, and may help students familiarize with writing, and with some practices and written learning genres in an additional language.

### 6 Conclusions

In this article, I revised some of the glottopolitical strategies of the Senegalese community and their relation with ICTs, especially those regarding the use of YouTube and WhatsApp. Through such strategies, which have grown both in content and relevance during the confinement, the right of speech of the Senegalese community, whether in Wolof or Spanish, has been bargained and defended. These strategies do not only occur in the Argentine territory. They are transnational and they allow interactions in diverse sites attended by Senegalese people.

Subsequently, I described the Spanish courses by BTM and the textile production project as glottopolitical interventions, and in relation to popular education. As many other interventions of the kind, these courses may take part in the constitution of sociopolitical subjects for both the students and the teachers. The latter often assume the teaching practice as a sociopolitical endeavor which may lead to militancy, whether for the right of speech or migration rights.

These learning experiences were adapted to virtuality. In one case, to sustain Spanish teaching, and in other case, to link that teaching to a textile production project. Given the possibility to impart the lessons by videoconference, as happened in many educational spaces, two alternatives were created. One was technically and visually very elaborate, and asynchronous: the YouTube channel. The other was technically simple, but more complex in terms of implementation: WhatsApp and a printed booklet.

WhatsApp and YouTube are platforms which students are familiarized with and use in their everyday life. An advantage is that they can be used without WiFi under most cellular operators. These technologies started being used for learning without making many adjustments, even though this was not their original objective. On one hand, WhatsApp can contribute to collaborative learning, participation and the possibility of learning at any place and moment. On the other hand, YouTube also allows for such possibility, but, in addition, it has a greater reach.

Undoubtedly, it is necessary to reconstruct familiarity and friendship by reestablishing face-to-face meetings. However, although weakened, the bonds we have constructed so far are more numerous and have more connectivity (due to the use of social media). Hence, it is possible to offer a larger number of people resources which allow them to learn the language, a right currently not guaranteed.

I analyzed three dimensions of the classes considering that they have been developed in spaces of popular education. The political dimension is related to an antiracist stance and the defence of human rights for immigrants. The pedagogical dimension is related to the idea of Spanish learning as a tool for organization and participation of the Senegalese community. Lastly, the ethical and methodological dimension responds to an attempt to transfer knowledge participatorily and cooperatively, and to adjust the methods and contents considering reality, culture and the needs of the students.

It will be necessary to assess the actual reception and usefulness of these learning proposals. As an example, we have found in the videos a learning channel which may be more familiar to this community, because it links memory learning and repetition (i. e., drills of common phrases and words, which are read aloud and translated by the speaker on the video). This is something that we, as teachers and learners of Western languages, should revise.

Finally, it is fundamental to bear in mind that suffering from the violation of social, economical or political rights, as well as lacking some classroom habits, should not lead students to being treated as non-knowledgeable or incompetent. They are adults who have developed numerous strategies to overcome the obstacles of economic migration. Among them, some referring to linguistic differences, which led these persons to opt for a Spanish course imparted in the host community.

### References


Bourdieu, Pierre, Las estrategias de la reproducción social, Buenos Aires, Siglo Veintiuno Editores, 2011.

Brignol, Liliane Dutra, Usos socials das TICs, Transnacionalismo e comunicação migrante em rede: uma aproximação à diáspora senegalesa no sul do Brasil,Comunicação, Mídia e Consumo 12 (2015), 89–109.

Calvet, Louis-Jean, Las políticas lingüísticas, Buenos Aires, Edicial, 1997.


#### Franziska Kailich

# No Seas Covidiota. Processes of Positioning and Identification through a Neologism on Twitter

Resumen: Las redes sociales constituyen al mismo tiempo un espacio ideal de formación y expresión de identidades, y un espacio de comunicación sin limitaciones en términos de distancia geográfica. Durante la pandemia de COVID-19, en tiempos de distanciamiento social, la comunicación digital se hizo aún más importante, permitiendo el intercambio de pensamientos y posiciones con respecto a la nueva situación, no solo entre personas conocidas y amistades, sino también entre desconocidos de todas partes del mundo hispanohablante. El contacto lingüístico, sobre todo con el inglés, es otra característica central de la comunicación en las redes sociales. Este artículo presenta algunos resultados de un estudio del neologismo covidiota, préstamo del ingl. covidiot, creado en marzo de 2020. Se examina el uso de este término por los usuarios hispanohablantes de Twitter con respecto a formas de posicionamiento y expresión de identidad, en el contexto de la situación de crisis mundial causado por el brote del COVID-19.

### 1 Introduction

Processes of lexical innovation are, in most cases, not documented and cannot be observed while they happen. Communication on social media platform represents an exception: it can enable us to detect first occurrences of neologisms, and follow (to a certain extent) the ways of its dissemination and further developments. One example of a neologism which proliferated mostly on social media is the coronavirus-related Spanish term covidiota (< Engl. covidiot, < COVID + idiot), which first came up in March 2020.

This neologism, in principle, conveys a humorous note. The so far relatively little researched role of humor in lexical innovation is linked to social functions of language use, as it can, for example, bring about a sensation of complicity between speakers, including some, and excluding others (Winter-Froemel 2018). Based on the understanding that speakers of a language influence each other in their use of language, and that, in language use, significant aspects of a speaker's identity are

Franziska Kailich, Julius-Maximilians-Universität, Würzburg

revealed (Geeraerts/Kristiansen 2014), social media platforms are extremely beneficial resources for sociolinguistic studies (Friginal/Waugh/Titak 2018). Moreover, social media platforms such as Facebook and Twitter are "plataformas ideales para el forjado de la identidad de los usuarios ya que la actividad de interacción en estas plataformas es cada vez mayor" (Yus 2014, 405). This certainly holds true to an even greater extent in the context of the pandemic, as social media communication became more important than ever in times when social gatherings in a physical form were forbidden or at least strongly discouraged.

This article presents aspects of a case study of the neologism covidiota in tweets and approaches the question of how Spanish-speaking Twitter users use the word to position themselves in the debate concerning the pandemic. Other results and aspects of the case study have already been published (Kailich 2022), with a focus on the integration of the anglicism covidiota into the Spanish language.

### 2 Covidiota as a Humorous Neologism – Central Aspects of Humor Theory

The term covidiota started to appear in Spanish Tweets in March 2020. Its basic meaning can generally be described as 'someone who behaves foolishly with regard to the COVID-19 pandemic', or 'behaving in a foolish way regarding the COVID-19 pandemic', as, in Spanish, covidiota can be used as either a noun or an adjective (Kailich 2022, 1).

As for its word formation, covidiota can be classified as a blend. One frequent characteristic feature of blending as a word formation type is a certain note of playfulness and a tendency to be used with a humoristic intention (cf., e.g., Ulašin 2016, 179). This holds true for covidiota. However, it is not only a potentially humorous lexical item, but also a swearword, one constituent being idiota, a term generally used as an insult. These two features suggest that both playful usage and a disparaging component play a role in the usage of the neologism covidiota.

A close relation between the creation of humor and elements of aggression can not only be expected from the usage of the neologism covidiota, but aggression has been established as a fundamental factor in the creation of verbal humor in numerous central works of humor research. Aggression- or superioritytheories of humor form one of the most significant branches of humor theory, whose long tradition can be traced back to Aristotle (cf., e.g., Attardo 2020, 64–67). According to these theories, "humor can be explained at least at some level as the pleasure/satisfaction or plain enjoyment of an enemy's discomfiture at our 'triumph' over him/her" (Attardo 2020, 65).

The aggression which causes or enhances the humorous effect is usually directed against a person or a group who is considered to be an opponent. In dispositional humor theories, for example, the central point is that "humour appreciation is facilitated when the respondent feels antipathy or resentment toward disparaged protagonists" (Zillmann/Cantor 1976, 93). In a study by Becker (2014), disposition theory is applied to an investigation involving political comedy. It can be shown that, for some (yet not all) groups of the test setting, it holds true that humor appreciation is increased when they dislike the target of the humor, or support the target's opponent (Becker 2014, 146–155). The present case study does not investigate disposition in terms of affiliation with political parties or political attitudes in general. But in the specific context of the coronavirus-crisis and its multiple consequences, it is nevertheless possible to distinguish (at least) two groups whose attitude towards COVID-19 (i.e., evaluation of its perilousness) and how best to deal with it (i.e., evaluation of usefulness and appropriateness of anti-virus-measures) are opposed to one another, and who therefore perceive each other as opposing groups.

When it comes to in-group-communication and consolidation of communities, a disparagement of the opposing person or group can have two important effects. On the one hand side, it can help to reinforce cohesion within the own group, i.e., the group of people who share the same views as the speaker with regard to the relevant questions of topics at hand. On the other hand, it is used as a means to distinguish oneself and the own group from the opponent(s), i.e., to reinforce boundaries towards the outside (cf. Ford et al. 2017).

A different branch of humor theories, on the other hand, shares the common view that humor is based on elements of play and playfulness, and that a particularly relevant aspect of humorous communication is the transmission of metamessages that convey the information "This is not serious" (cf., e.g., Attardo 2020, 61; Bateson 1972, 179–180). Keeping these two central theoretical lines of humor research in mind, one of the aims of this paper is to identify indications of aggression and potential metamessages capable of creating a "playful mindset" (Attardo 2020, 61) in the analyzed tweets.

### 3 Identity in Social Media

"Just as in spoken language and other forms of written language, texters perform identity through their linguistic and orthographic choices" (McSweeney 2018, 138): what McSweeney states for the construction of identity in text-messages can certainly also be said to apply for social media postings. Apart from linguistic features of the text itself, several properties of communication on social media platforms are of crucial relevance with regard to the formation and expression of identity.

According to Yus (2019, 223), identity in social media evolves between two poles: personal identity on the one hand, and social identity on the other. As individuals, social media users want to emphasize their uniqueness, but at the same time, they wish to show affinities and present themselves as belonging to a group, being a member of a community.

With regard to personal identity, the personal profile plays a crucial role. The profile picture, certain pieces of personal information, but also comments about the user's daily life are shared with other users, and addressed to a certain group, which the owner of the profile imagines to be their audience (Yus 2019, 224–225). The so called "retroalimentación identitaria" (Yus 2019, 223) – 'identity feedback' or 'identity reinforcement' – then results from reactions, for example comments which others make about the content shared by the user. These reactions feed back into the subsequent behavior of the user with regard to self-representation in social media: "When users update their status to express their thoughts, they offer a representation of the self, based on the online socialization they have already experienced" (Darvin 2016, 531). The content created by the user is reviewed after publication, by other users and by the author themselves, with particular attention to the reactions it evoked from others (Weber/Mitchell 2008, 27).

Darvin points out the fact that the sharing of various facets of a user's life on social media platforms leads to a change in the perception of private and public spheres (Darvin 2016, 529). A user shares information that would generally be considered private with a much larger audience than in offline contexts. Moreover, users tend to share intimate information on social media platforms which they would be less likely to share in face-to-face-situations. This tendency is referred to as extimidad in Spanish, or "exteriorización constante de la intimidad" (Yus 2014, 409) – 'externalization of private matters.' Furthermore, by posting information about offline-activities, sharing photographs, etc., online and offline life become increasingly intertwined (Darvin 2016, 530; Yus 2019, 222), so that a clear distinction between online- and offline-identities becomes impossible.

As for the construction and display of social identity, Yus identifies several significant aspects in social media presence: for example, by sharing photos which show them together with others, users demonstrate aspects of social affiliation (2019, 225). The sharing of links to other users' posts, profiles, or external web pages, or the distribution of contents published by others (e.g., by retweeting) can also be considered expressions of friendliness, agreement with the message of the post, or the opinion of its author. Comments about issues of politics and society, usually of an evaluative, often critical, nature, represent another feature of formation and expression of social identity on social media platforms (Yus 2019, 225). Moreover, Yus mentions linguistic features as a means of expressing and reinforcing social identity: for example, the choice of vocabulary can allow for the recognition of affiliation to a certain group within a society. With regard to social identities, interaction within online communities enables users to "negotiate shared values and rules of collective behaviour" (Facer 2011, 90).

Interaction between users on social media platforms is crucial for both personal and social identities, because the relevant effects arise when others react to contents shared by a user, the user themselves perceive these reactions and adjust their further postings accordingly. Weber/Mitchell identify "constructedness" (2008, 39) as well as playfulness as common properties of different kinds of media creations designed by young people, including personal profiles on a social media platform.

### 4 Corpus Compilation and Research Questions

The tweets analyzed in this case study were extracted manually, using the "Advanced Search" provided by Twitter. The search was performed using the following terms:


The investigation periods were three individual weeks in 2020, the first one being the week in which the first evidence for the word covidiota was found in a tweet in Spanish. The distance between the observation periods is 14 weeks. These searches turned out a total of 1491 tweets:


These tweets were anonymized, omitting names of people as well as names of specific places (names of cities, for example), which might convey any information about the identity of private Twitter users. The omitted items were replaced by NAME (for proper names of people), names of specific localities are represented as CITY, etc. Public figures, such as, for example, well-known celebrities and widely known politicians, and institutions, were exempted from anonymization.

All 1491 tweets were subsequently submitted to qualitative analyses. Due to the low number of tweets and to the selective nature of the investigation periods, no statistical relevance can be claimed for any of the results. However, tendencies with regard to developments in the use of covidiota by Spanish speaking Twitter users can be expected and will be presented in the further course of this study.

The analysis is based on the following research questions:


Tweets which are reproduced in this paper as examples are depicted without any changes, apart from the above-mentioned anonymizations. No correction of spelling errors was carried out, and ungrammatical expressions are neither corrected nor pointed out as such. Emojis and emoticons are represented in their verbal description as indicated for example by Emojipedia®. The tweet text itself is represented in italics, whereas replacements of names (of people, cities, etc.), as well as the descriptions of emojis, are represented in regular letters.

### 5 Analysis and Results

As a general result, it has to be noted that the neologism covidiota can have different meanings. The general meaning paraphrased at the beginning as '(someone) behaving foolishly with regard to the COVID-19 pandemic' can be specified in three different ways to match the different meanings. During the first time after its first appearance in Spanish tweets, numerous posts represent or include a definition of the etymon covidiot in English, often attached to the post as an image file. It defines covidiot as follows:


Similar definitions, also attached to tweets in the form of image files, can be found in Spanish, as. In those cases, the term covidiot is defined as follows:


However, these are less frequent than the English definition.

As for definitions of the term in Spanish, they are more frequently contained within the tweet text itself than in an attached file, and often represent not direct translations of the English definition, but more individual versions. They do, however, convey a variant of (either both or one of) the two basic meanings that can be extracted from the definitions given above. (cf. Kailich 2022). These two meanings shall be referred to as covidiota 1a and covidiota 1b, as both of them appeared simultaneously, but they differ in content.

```
covidiota 1a a person who does not consider COVID-19 to be a threat and does not
follow the rules of social distancing
```
covidiota 1b a person who buys and hoards large amounts of essential products, especially toilet paper

There is a third use of the term covidiota, which constitutes an inversion of the meaning covidiota 1a into its opposite. It does not yet appear in the first observation period of the case study (16-22 March 2020), but later.

covidiota 2 a person who believes that COVID-19 is a serious threat and is in favor of, or at least follows, measures intended to curb its spreading

Even though all three meanings can be found in the examined tweets, differences in usage can be noted. The developments regarding the usage of covidiota with its different meanings will be discussed in the following section.

### 5.1 Stages in the Usage of the Neologism

In the first observation period T1, from 16 to 22 March 2020, covidiota 1a and 1b can be found. More than half of the tweets in T1 contain a definition (Kailich 2022, 10), in English and/or in Spanish. The Spanish definitions more frequently appear in the proper tweet text (not as an attached file), and some of them are close translations of the English definition. But there are also more individual versions, as for example in (1), where several additional aspects were added to the definition.

(1) COVIDiota: persona que cree que el COVID-19 es bueno para hacer chistes, cartones, que fue creado por algún gobierno, que asiste a eventos como EVENT, reenvía cadenas pendejas en WA, que cree en el poder de la oración para curarlo o que la fuerza moral inmuniza. Agreguen. [2]

Apart from definitions, metalinguistic comments about the neologisms also occur with a certain frequency in the first observation period: about one third of all the tweets in T1 contain a metalinguistic comment (Kailich 2022, 10). (2) represents an example.

(2) Covidiot o en español covidiota. Nueva palabra del diccionario para esta cuarentena. [33]

Another very specific kind of usage, based on the meaning covidiota 1a, is the use of the term to refer to certain groups which were created, for example on Facebook, with the aim to denounce people who do not adhere to the rules given out by the authorities. This particular use becomes apparent in (3).

(3) q ganas d mandar a covidiotas las stories d tu pool party. [140]

The second observation period T2, from 29 June to 5 July 2020, shows some differences compared to T1. For covidiota 1a, it can be attested that this usage now occurs much more frequently in the function of referencing, whereas the proportion of definitions and metalinguistic uses has declined (cf. Kailich 2022, 10). Moreover, a considerable number of cases can be found which deal with the question of what kind of behavior actually justifies calling somebody a covidiota. (4) shows an example for one of these cases of negotiation of semantic content (cf. Kailich 2022, 10). (5), on the other hand, exemplifies the usage of the term solely with the function of referencing.


The new meaning covidiota 2 can be found for the first time in T2. In (6), the meaning covidiota 2 is not explicit and not clearly identifiable at first sight. But the use of the term borrego 'lamb/sheep' (also 'idiot') suggests an interpretation in the sense of covidiota 2. It is frequently used by people who do not believe in the existence or perilousness of COVID, to refer to others who do consider COVID-19 to be a threat and who support anti-virus-measures. An examination of preceding tweets reveals that (6) is part of an exchange of messages between one person who thinks of COVID as a complex and serious illness, and the author of (6), who is of an opposite opinion and expresses his views about the other's position. Example (6) is among the very few occurrences of covidiota 2 in T2. Example (7), on the other hand, was published in the third observation period. Here, the tweet very explicitly states a positioning with regard to the pandemic, more specifically, against quarantine as an anti-virus-measure.

Another development can be observed in T2: there is no longer any evidence of covidiota 1b 'a person who buys and hoards large amounts of essential products, especially toilet paper'. It appears that, unlike covidiota 1a, this meaning has not been taken up and used, but disappeared at some point between T1 and T2.

As for the usage of covidiota in T3, it can be noted that covidiota 1a and covidiota 2 continue to be used, while covidiota 1b does not reappear. The use of covidiota 2 is slightly more frequent in T3 than in T2. There are different possible explanations for the very low number of examples compared to covidiota 1a. It is possible that, during the observation timeframe, the meaning covidiota 2 was actually much less commonly used in general, or by Twitter users in particular. It is, however, likely that Twitter's COVID-19 policy might have influenced the results: Twitter tightened its rules, deleted posts and deactivated accounts if users engaged in the dissemination of "potentially misleading and manipulative content" (Twitter blog, 14 July 2021) with regard to COVID-19. Since covidiota 2 is used (amongst others) by people who deny the existence of COVID, by followers of conspiracy theories, etc., it seems likely that several tweets that might have contained covidiota 2 were deleted and therefore could not be taken into consideration for this study. A study focusing on the platform Telegram, for example, might have produced different examples.

#### 5.2 Personal and Social Identity in the Tweets

Yus' (2014 and 2019) analyses are based on Facebook profiles, but appear to be applicable for Twitter as well. According to him, personal identity is created and expressed by means of self-representation in social media profiles. The data gathered for the present study does not comprise information about the users' profiles as such, so that aspects such as pictures and personal information contained in the profile cannot be taken into account.

However, Yus also mentions comments about a user's daily life as a relevant means of forming and expressing personal identity in social media. Tweets that have to do with users' daily life during the pandemic are quite frequent among the tweets analyzed for this study.


In both tweets, the speakers share some experience of their life during the pandemic with their audience. In (8), the speaker admits to having been to the beach, while in (9), a scene from a dream is depicted in which the speaker is denied being served pizza in a pizza restaurant. Both tweet-authors use the expression ir de covidiota 'to go [somewhere] as a covidiot' (cf. Kailich 2022, 11) and label their actions (in dream or reality) as not entirely compliant with the regulations and recommendations for the containment of the virus.

When it comes to social identity, Yus (2019, 225) identifies several relevant aspects of a person's social media profile. Photos which show the user together with other people are a frequent means to express aspects of social identity, by showing amity towards those people. However, as this study focuses on the examination of tweet-text, an analysis of photographs is not included in the case study. The sharing and retweeting of content published by others is another one of the relevant aspects with regard to social identity on social media platforms. However, the methods chosen for this study allow for a qualitative analysis of tweet-text, but not for the detection of larger networks and processes of interaction between users.

A further central aspect of the constitution of social identity in social media is the sharing of critical comments, for example about politics or current events in society (Yus 2019, 225). The situation caused by the coronavirus-outbreak, including of course antivirus-measures and people's reactions to them, certainly represents a situation which attracts a huge number of evaluative comments, and it is usually a central topic of the tweets containing the word covidiota.


Two examples of such critical comments can be seen in (10) and (11). In (10), the tweet-author criticizes the way the pandemic is handled by authorities in El Salvador, saying that, apparently, the economy seems to be more important than the health of the country's inhabitants. In (11), the speaker criticizes a politician (states secretary) for the decision to make the wearing of a mask obligatory in schools. Both comments can thus be considered relevant when it comes to social identity on Twitter in the context of the COVID-19-pandemic. The use in (10) corresponds to the meaning covidiota 1a: the attitude to favor economy over people's health is referred to as "#COVIDIOT", and the criticism is that the authorities do not do enough to protect the people from contagion with COVID-19. In (11), on the other hand, the underlying meaning is covidiota 2: the speaker criticizes a particular anti-virus-measure (obligatory masks at school) as inappropriate.

The processes surrounding the emergence and spreading of the neologism covidiota can be considered a vivid example of certain properties of digital communication. As has already been pointed out, the term was first coined in English and then adopted by Spanish speaking users of social media (cf. Kailich 2022). This happened within a short period of time, during the first weeks of confinements and lockdowns in many countries worldwide. The rapid spreading of the term and adoption by several other languages (e.g., Italian and Portuguese: covidiota, German: Covidiot) was of course enabled by the fact that, in online communication, geographical distance does not propose an obstacle. Language contact, in particular with English as the prominent language on the Internet, is another prominent feature. With regard to the circumstances of that particular time, people all over the world began to position themselves on their understanding of the situation, they began to form opinions about the measures imposed by governments, and about the behavior of their fellow citizens. As, in online spaces, "people and ideas are able to travel virtually" (Darvin 2016, 526), it was possible for Spanish-speaking social media users all over the world to exchange their views, express agreement or disagreement with other users' positions, and to build new communities within or across existing ones, according to their attitude towards this new situation. Within these new communities of likeminded users, it is then possible for them to reinforce each other in their views. On the one hand side, there are those who like, retweet, translate, and adopt the definitions of covidiota 1a and 1b. Those who actively engage in denigration and even denunciation of others who do not follow the new rules. Those who try to counter the situation with humor. And those who oppose the severe measures and consider the fear of the virus to be greatly exaggerated. These latter, together with those who do not believe in the existence of the virus at all, make their own meaning of the term covidiota, using it to designate the other side.

The term covidiota, which was created and spread within social media, is continually discussed by the Twitter users in the examined tweets with regard to its meaning. In many of the analyzed tweets, users debate what actions and behaviors qualify as covidiota, or for what reasons a person can or should be called a covidiota. This kind of collective negotiation of the actual meaning of the term is a prominent feature within the tweet corpus under investigation (cf. Kailich 2022, 16), and it corresponds to one of the characteristics of digital communication as pointed out for example by Darvin: "Interaction, dialogue, negotiation and contestation become intrinsic to digital media, where meaning-making is marked by simultaneous decoding, production and interaction" (2016, 526). The meaningmaking for the term covidiota, thus, has to be considered a collectively negotiated action in which several users from different parts of the Spanish-speaking world take part, and which can be observed in the tweets over all three observation periods.


(12), (13) and (14) show examples for each of the observation periods. In all three examples, the tweet-authors reflect about what actions or behaviors the label covidiota actually applies to. The author of (14) does on the so in a more playful way and includes the information that she herself has engaged in the very action she depicts as covidiota ('covidiotic'), while the other two quoted tweets contribute to the meaning-making in more neutral terms.

#### 5.3 Auto-Designation vs. Hetero-Designation

When it comes to expressing a position by means of language use, it can be important to consider whether a term is used for auto- or hetero-designation. A term like covidiota, which, apart from being a COVID-related neologism, is also a swearword, can convey very different meanings, depending on the referent designated by it. It can adopt different levels of playfulness, on the one hand, and aggressiveness, on the other, depending on whether a speaker uses it to speak or write about themselves, to address somebody directly, or to refer to a third party (individual person or group of people) and thus speak about someone.

But the differentiation between self- and hetero-designation is not the only important one to make. It is also crucial to distinguish between affirmative uses, cases of negation, and uses which cannot be attributed to either of those two groups.


Example (15) illustrates an affirmative use, and the expression is directed towards a communication partner addressed in the second person singular, which becomes manifest in the verb form eres 'you are'. In (16), the referent is the speaker-self (first person singular), but the statement contains a negation, so that the speaker explicitly states not to be a covidiota. In (17), on the other hand, there is no clear affirmative or negated statement. The speaker expresses a desire to be (to act like) a covidiota, but neither confirms nor denies actually being one. Considering these possible usages and different options, the following categories were established for the analysis in the present study:

	- the speaker him✶ herself or a group that includes the speaker-self
	- the interlocutor, a person or group directly addressed by the speaker
	- a third party (person or group) not directly addressed but spoken about
	- in an affirmative way ('x is a covidiota')
	- in a negated way ('x is not a covidiota')
	- in a way that is neither affirmative nor negated (e.g., 'x would like to be a covidiota', 'is x a covidiota?')

The 1491 tweets retrieved for the present study are analyzed according to these categories, and divided into separate groups, according to observation periods (T1-T2) and the different meanings of covidiota. The identification of the referent of the tweet is carried out only for those tweets which count "referencing" among their functions, not for tweets which only contain definitions or metalinguistic comments. The results for covidiota 1a are shown in Table 1.

There are no cases of covidiota 1b to be analyzed for auto- vs. heterodesignation, since, as outlined above, this meaning was never really adopted and put into use, at least not according to the material analyzed for this study. The results for covidiota 2 are shown in Table 2.

It has to be kept in mind that the number of tweets providing evidence for covidiota 2 is very low, so that it is not possible to draw reliable conclusions from this data. Nevertheless, a certain tendency seems to become apparent, which would of course have to be verified by subsequent analyses based on a bigger sample of tweets.


Example (18) shows a case of covidiota 2 being used to refer to an interlocutor, a person (another Twitter user) who is directly addressed in the text. In (19), there are two instances of covidiota contained in the tweet, the second one being used to speak about a group of people, referring to them as covidiotas. (20) constitutes a statement about covidiotas in general.


Table 1: Auto- vs. hetero-designation covidiota 1a.


Table 2: Auto- vs. hetero-designation covidiota 2.

#### 5.4 Signals of Playfulness and Aggressiveness

Since covidiota has to be considered a (potentially) humorous lexical item, different facets and theoretical aspects of verbal humor are taken into account in the analysis of its usage. Two main lines of humor research were introduced above, namely aggression-theories, on the one hand, and play-theories on the other. The first considers aggression, or the notion of superiority, to be essential for the creation of verbal humor. The latter supposes that humor can be considered a temporary liberation from the constraints and tensions of life (cf., e.g., Attardo 2020, 61), be that everyday life or even particularly stressful times of crisis. Moreover, play- or release-theories generally include the notion that playful interaction and communication is usually signaled as such by means of so-called play-markers. A contradiction seems to become apparent when considering the positions of both theoretical lines at the same time, particularly because "the very idea of the metamessage 'this is play,' as per Bateson (1972), denies the possibility of real aggression being part of play" (Attardo 2020, 67). When examining the use of covidiota in Spanish tweets, the tweets are searched for indicators of aggression (superiority) as well as possible play markers.

A considerable number of emojis or emoticons have the potential to be used as play-markers. For example, in (21), the speaker does not appear to intend to seriously insult the addressee of the tweet. Neither does the author of (22).

(21) Si eres un covidiota [192] Woman facepalming Face blowing a kiss

#### (22) Eso es muy covidiota de tu parte colega médico [1355] Face with hand over mouth

In both examples, the verbal text alone could be interpreted as offensive. But the use of the "Face blowing a kiss"-emoji in (21) makes the message appear much less serious, thus carrying out the function of a play-marker. A similar effect is caused by the "Face with hand over mouth"-emoji in (22). The use of this emoji is explained on emojipedia.org as follows: "[. . .] suggesting coy laughter or embarrassment, as if cheekily saying Oops!" (Emojipedia® 2022). It can therefore also be interpreted as a play-marker, indicating that the message of the tweet is not to be understood as serious criticism or even an insult, but rather as a playful comment.

When it comes to analyzing and interpreting the use of emojis in digital communication, it has to be pointed out that there are various different kinds of usages. Emojis can be used instead of words, in order to transmit a certain meaning, but they can also appear alongside verbal expressions and be used to emphasize contents that are, at the same time, conveyed verbally (cf. Sampietro 2019, 137). Moreover, they can fulfill a variety of different functions. The expression of the writer´s sentiments is one possibility, but emojis can also be meant to underline the intention of a message, so as to provide clues for it to be interpreted in the way intended by the author (McSweeney 2018, 129). Yus (2014, 416) defines seven functions for the use of emoticons in an analysis of Facebook postings. Sampietro (2019) establishes a framework to analyze the use of emojis in WhatsApp-messages with regard to different kinds of politeness. The use of emojis as play-markers is thus only one of many possible functions, and the fact that certain emojis have the effect of a playmarker in some of the tweets does not necessarily mean that the same emojis will have the same effect in other tweets.

However, it is not only emojis that can carry out the function of play-markers in social media-communication. Generally, all features which can be identified as markers for humorous language use can convey a metamessage saying "this is not serious". If a play-marker is successfully used in a message, this implies that real aggression cannot be expressed in the same message, since it would be turned into playful mocking, or non-serious representation of aggression, by the metamessage (Attardo 2020, 67).

Burgers/van Mulken (2017) present an extensive list of markers of verbal irony and sarcasm. Even though, as the authors point out, humor and irony are two different phenomena which are not to be mixed up with each other, markers for irony and sarcasm can still be considered relevant signals for humorous language use. But irony, and sarcasm in particular, can convey a considerably aggressive note, considering that the latter can be defined as a "negative form of verbal irony, used on occasion to enhance the negativity" (Colston 2017, 236) the speaker wishes to convey. Therefore, the markers presented by Burgers/van Mulken cannot by default be interpreted as play-markers. Most of them would probably have to be classified as ambiguous markers, which can signal playful kinds of irony as well as more aggressive types, depending on other factors in the context.

When potentially ambiguous markers are used, sometimes disambiguation towards a more likely interpretation is facilitated by accompanying items such as other emojis, the choice of words or the content of the message (on a literal level). In other cases, it remains impossible to decide with any certainty whether the ambiguous item is used, in a specific case of usage, to convey playfulness or aggression.

Among the potentially ambiguous markers which appear in the analyzed material are the "Clown-face"-emoji, as well as "Face with tears of joy" and "Rolling on the floor laughing". The latter two can be used to express derisive as well as non-derisive laughter. The first is, according to emojipedia.org, "[c]ommonly used as comment on TikTok or other social media platforms to indicate that a person is foolish, idiotic and/or selfish" (Emojipedia® 2022). There are, however, also cases in which the "Clown-face"-emoji is used in a non-pejorative way – hence the classification as potentially ambiguous.

Apart from emojis with ambiguous usage potential, typographical highlighting, for example in the form of capitalization or the use of quotation marks, can have several different functions and therefore has to be considered ambiguous in multiple ways. On the one hand side, typographical highlighting can be used as a marker of irony or sarcasm, for example when a term which is not to be interpreted in its literal sense is put in capital letters or quotation marks. Burgers/van Mulken give the example "It is a GREAT idea" (2017, 388), uttered with the intention of saying that something really is, contrary to the literal message, a bad idea.

In (23) (= 11), the use of quotation marks for educación might be an expression of verbal irony, possibly with the intention of disputing the competence of the minister in his field (by saying he cannot really be considered a minister of education if he supports such measures as criticized in the tweet). In (24), an interpretation of capitalization as a potential play-marker seems to be plausible. In (25), on the other hand, an interpretation as a play- or irony-marker appears to be less so. Instead, the use of capital letters could represent raised voice. In digital communication, capitalization is very generally understood as an equivalent to shouting. The highlighting observable in (25), if meant to illustrate shouting, would then also convey aggressiveness.

(23) (= 11) ¿Obligar a los niños a usar bozal? El secretario de "educación" es un covidiota [1120]


(26) Y usted es un COVIDIOTA? #QuedateEnCasaSubnormal [55]

Moreover, typographical highlighting is, according to Pflanz's (2014) categorization of "marqueurs d'altérité" ('alterity markers', Winter-Froemel 2021), one of the strategies speakers use to deal with newly borrowed items. It makes the unfamiliar item, here: the neologism (and anglicism) covidiota, stand out (cf. Kailich 2022, 3) and draws a reader´s attention to it. (26) can be considered as an example of this use of capital letters.

It can be shown that highlighting (by capitalization and quotation marks) of the term covidiota (involving parts of the term, the entire term or even the entire phrase containing the term) occurs in 54.43% of all tweets in T1, in 15.14% of those in T2, and 13.92% of tweets in T3 (Kailich 2022, 7–8). These observations do not, however, consider capitalization of other parts of the tweets, that do not include the term covidiota. When analyzing play-markers and signs of aggression, however, capital letters in other parts of the tweet can be just as relevant as highlighting of the term itself.

To sum it up, it can be said that the use of capital letters has the potential to be meaningful in several different ways, but it is not always possible to determine which one(s) apply in a certain case of usage. One possible assumption is that highlighting of the term covidiota itself rather points towards an interpretation either as a strategy of flagging an unfamiliar lexical item (in the sense proposed by Pflanz 2014, 164), or a means to emphasize the exclamation of a swearword (to visualize shouting), and that play- or irony-markers are more likely to be detected in highlighting of other parts of the tweet, not the term itself. However, this hypothesis is yet to be verified.

The detection of aggression poses more of a difficulty than that of play-markers. As for the use of emojis, there are some which can be attributed to the expression of an aggressive or contemptuous note more easily than others. Among those we can count emojis expressing anger (e.g., "Angry face" or "Angry face with horns") or revulsion ("Face vomiting" or "Nauseated face"), as well as emojis representing rude hand-gestures ("Middle finger").

Based on the considerations concerning possible markers and signals of playfulness and aggressiveness, the categories of analysis were established as follows:

	- emojis and emoticons expressing open and moderate forms of smiling or laughter (including neither "Face with tears of joy" nor "Rolling on the floor laughing")
	- other emojis whose meaning can be understood to convey a note of nonseriousness, e.g., "Upside-down face"
	- emojis conveying affection and liking (e.g., "Face blowing a kiss"), since the combination of such an expression of affection with the designation of someone as a covidiota is considered to take the seriousness out of a potential insult
	- emojis expressing anger (e.g., "Angry face" or "Angry face with horns") or revulsion (e.g., "Face vomiting", "Nauseated face")
	- rude hand gestures
	- emojis expressing either playfulness or contempt, especially the "Clown face"-emoji
	- emojis with a basic meaning of 'laughing hard' (e.g., "Face with tears of joy", "Rolling on the floor laughing"), since they can express open, "innocent" laughter as well as derisive laughter
	- verbal representation of laughter (variants of jajaja), since it can represent derisive as well as non-derisive laughter
	- typographical highlighting: capitalization (either used to encourage nonliteral interpretation, or to highlight an unfamiliar linguistic item, or to represent emphasis)

The tweets are searched for any of these signals, and results are presented in Table 3. As has already been mentioned, the use of emojis in digital communication can vary greatly between individual users, usage contexts and usage cases. An emoji which can be used as a play-marker does not automatically have to be used as such in all its occurrences. Being a play-marker is only one of its possible functions. Nevertheless, when investigating the use of a potentially humorous lexical item in social media contexts, emojis certainly have to be taken into account, and emojis which can be understood as play-markers constitute a recurring feature in the examined tweets.

#### 5.5 Positioning of the Self: The Use of Covidiota in Context

When it comes to investigating the ways in which Spanish-speaking Twitter users adopt the neologism covidiota in tweets in which they position themselves in the societal debate about COVID-19, various influencing factors have to be taken into account. Two central aspects have been presented in the previous sections: the question of whether speakers use the term to refer to themselves or to others, and in what way (affirmative, negated, or neither), and the presence or absence of markers of playfulness and aggressiveness.


In example (27) (= 18), covidiota is used in the phrase borrego covidiota, which would translate literally as 'covidiotic lamb/sheep'. Moreover, the speaker calls the interlocutor 'crazy', when he says loco estás tú ('it´s you who's crazy'), so that several potentially offensive expressions can be made out in this text. The tweet does not contain any clear evidence of play-markers to indicate that the message is not to be understood in its literal sense, instead it contains an ambiguous marker: the phrase borrego covidiota is capitalized. The absence of clearly playful elements and the presence of several insulting expressions, as well as the fact that the designated referent is not the speaker-self but the addressee, can be said to contribute to a rather aggressive notion conveyed by this tweet. The author appears to be serious in their criticism of the addressed recipient, and humor, if present, would have to be considered of a significantly derisive nature in this example, understood as ridiculing of the target, while considering oneself superior.

In (28), by contrast, no aggressive note is detectable. This overall impression is supported by the fact that, apart from covidiota, no other potentially insulting expressions are contained. Instead, there is an emoji which, according to the criteria outlined above, can take the function of a play-marker, and, moreover, the speaker indirectly includes him✶ herself in the reference of covidiota, when expressing the intention of going to a 'covidiotic party'.

With regard to the positioning of the authors of (27) and (28) towards the COVID-19-crisis-situation, considerable differences between both examples can be detected. The author of (27) uses covidiota to refer to a member of the opposite group, and uses other offensive expressions and no attenuating elements. The content of the tweet suggests that the addressee is in favor of PCR-testing while the speaker is not. Hence, the speaker´s position is critical towards a certain



common practice (PCR-tests) and s✶ he criticizes (attacks) the addressee for supporting an opposite opinion. The author of (28) does not express criticism. S✶ he uses covidiota to refer to something s✶ he is planning to do, and thus includes him- ✶herself among the referent(s) of the term. The prevalent playfulness in (28) moreover suggests a position that considers 'covidiotic' events as something one can joke about, or that one can actually consider taking part in, even though accepting that it might not be a very reasonable thing to do.

Both features, auto- vs. hetero-designation as well as the presence or absence of markers of playfulness or aggressiveness, are relevant with regard to the positioning of the individual tweet-authors in the context of the debate about COVID-19. Table 4 presents a combined overview of both features.

The last three columns in Table 4 depict the percentage of tweets in the respective group which contain play- or aggression-markers or ambiguous markers, as defined in the previous section. However, the figure indicated in Table 4 for the ambiguous markers does not include the results for the category "capital letters incl. covidiota". In a majority of the tweets, and particularly at the beginning of the observation timeframe, its predominant function seems to be mostly that of flagging a neologism. The only possible conclusion is that capitalization of (parts of) the term covidiota itself can, but does not necessarily need to be relevant for the signaling of playfulness or aggressiveness.

Capitalization of other elements in the tweets (not the target term itself) occurs much less frequently (see Table 3), but it appears to be more clearly linked to aspects of play and aggression. It can convey general emphasis, or act as a marker of irony or sarcasm. It is for this reason that in Table 4, capitalization that includes the target term is not included, but capitalization of other parts of the tweet is.

In Table 4, the most frequent category in terms of auto- vs. hetero-designation is depicted, and, moreover, the most frequent subcategory (considering the differentiation between affirmative, negated, and other statements, see section 4.3, Tables 1 and 2). In general, it can be said that self-designation stands out as the most frequently found category for covidiota 1a in the second and third observation periods, even though in T2 the distribution across the three groups is more balanced than in T3. When it comes to the subcategories, the development of the most commonly used one for covidiota 1a is from "designation of third person(s) – affirmative" in T1 to "designation of the interlocutor – affirmative" in T2, and finally "self-designation – non-affirmative" in T3.



These figures are indicated in brackets because of the extremely low numbers for covidiota 2 in T2: one out of seven uses amounts to 14.20%, but the percentage has little informative value in this case.

With regard to representation of identity and positioning, some tendencies can be spotted in this development: at the beginning of the crisis, often characterized by an atmosphere of tension and preoccupation, the word (if used with the function of referencing) was used to designate, ridicule or accuse, and even denounce those who behaved in an inappropriate way (as becomes most clearly observable in the emergence of groups in social media, created for exactly that purpose). The implicit positioning of the speaker is then to see themselves on the "right" side, doing the sensible thing, while those others who behave irresponsibly are referred to as covidiotas. In T2, a direct designation of the interlocutor has become the most frequent subcategory – meaning that Twitter users either mock or seriously criticize the addressees of their tweets because of their behavior in the ongoing crisissituation. At the same time, among the superordinate categories, self-designation has already become the most frequent one. In T3, self-designation is by far the most frequent category, the most commonly found sub-category being the nonaffirmative one. This comprises expressions of whishes (example 17) as well as doubts and reflections about what can be done in this situation and what kinds of behavior would be too irresponsible (example 4). These tweets can be seen as reflections, published by the Twitter users who seem to be pondering the correct way for themselves to handle the situation, and possibly hoping for feedback from other users. Moreover, the fact that affirmative statements of the type "I am/have been/will be (a) covidiota" constitute the second most frequent group for covidiota 1a in T3 suggests that, for those who use it in this way, 'being a covidiota' is no longer a reason to offend, attack, or ridicule someone. Instead, covidiota seems to have become a term playfully used to describe little transgressions of the rules and noncompliances with recommendations, and, in some cases, it even seems to have become a synonym for all activities that involve leaving home (Kailich 2022, 11).

In addition to these observations, the use of play- and aggression-markers for covidiota 1a has to be taken into account. As for those markers that were defined as potential play-markers, it can be observed that their use increases slightly over the course of the observation, starting with 2.94% in T1, amounting to 4.98% in T2, and 8.36% in T3. Ambiguous markers also become more commonly used, starting with 2.21% in T1, reaching 10.89% in T2, and finally 12.09% in T3. By contrast, those items defined as markers for aggressivity are much less frequently used (with a maximum of 1.47% in T1) during all three observation periods.

Together with the growing use of covidiota for self-designation, the slight increment in the number of play-markers can be interpreted as an indicator for increasingly playful usage of the term covidiota. The increase in the number of ambiguous items would have to be examined more in detail, determining for every individual case whether disambiguation is possible. If playful uses could be made out as more frequent than others, this would further support that interpretation.

As for covidiota 2, in T2, the sample (7 tweets out of 650 in T2) is too small to spot any tendencies at all. It might be worth mentioning, however, that T2 contains two uses of covidiota for auto-designation, one of them being an affirmative statement.

#### (29) Soy un Covidiota por estar en mi casa. [341]

(29) represents an atypical use of the term with this meaning. In T3, on the other hand, there is no evidence for the use of covidiota 2 in which speakers refer to themselves in an affirmative way. The number of tweets in T3 is still very small, but nevertheless, a tendency can be made out: in 28 out of 35 tweets, the speakers use covidiota to refer to a third party. 17 out of these 28 are affirmative statements. If we try to extract a predominant positioning from the tweets using covidiota 2 in T3, it would thus comprise the fact that the speaker-self considers others to be covidiotas, but does not include him✶ herself in this group. In T3, covidiotas are people who insist on wearing masks, who believe in the perilousness of the virus, who follow the rules, etc. And for the users of covidiota 2, these people are the others, the opposed group, those whose judgement of the situation is faulty.

The observation of play- and aggression-markers for covidiota 2 in T3 does not show any markers that belong to either the above defined group of playmarkers, nor to that of signals of aggressiveness. However, the number of ambiguous markers in T3 is considerably higher than for covidiota 1a: about every fifth tweet (22.86%) contains one or several ambiguous markers, such as capital letters, "Rolling on the floor laughing"- or "Face with tears of joy"-emojis, or representations of laughter such as jajaja.

(30) Un covidiota más grande que he visto, ojalá lo caga de la gripe de aves y los de chancho jajaja [1038]

As an example of possible disambiguation, in (30), the combination of the laughter-representation jajaja with the imprecation ojalá lo caga de la gripe de aves y los de chancho conveyes an aggressive note rather than a playful one. The designated referent is, once more, a third person, who is depicted as stupid, and the offensive elements are not attenuated by any obvious play-marker. The element jajaja can therefore be interpreted as derisive laughter and the type of humor in the tweet as one strongly characterized by amusement about the deprecated referent, while the speaker conceives him✶ herself, and his✶ her imagined audience, to be superior.

### 6 Conclusion and Outlook

For covidiota 1a, a marked increase in the use of the term for auto-designation, combined with a slightly rising usage of play-markers suggests the presence of an increasingly playful and non-aggressive kind of humor in the analyzed tweets. A majority of the tweets implicitly or explicitly reveals a conflict between the wish to behave responsibly on the one hand, and the longing to reestablish (physical) social contacts and resume activities, returning to normality, on the other, and possibly a search for encouraging feedback from the audience.

While there are not enough examples for covidiota 2 in the tweet corpus at hand to make any definite conclusions, it is nevertheless possible to make out a tendency with regard to the positioning of its users in the debate about COVID-19. The fact that covidiota 2 appears to be very rarely used for self-designation, the finding that it does not appear alongside any of the defined play-markers, and the analysis of individual examples which reveal a type of humor which is characterized by the superiority/aggression-dimension rather than by playfulness, add up to a tendency of usage which could be described as follows: Users of covidiota 2, in T3, use the term to refer to those who they consider to be the opposing group: those who believe in the existence of the virus and consider it as a threat, and impose, implement, follow, and support the antivirus-rules and recommendations. These others are mocked and talked about with expressions revealing a strong influence of the superiority-dimension. Thus, covidiotas, in this case, are others, the term does not apply to the speaker themselves, and 'being a covidiota' is closely associated with stupidity.

Given the importance of interaction for the formation of online identity, it would be desirable, in a subsequent study, to analyze not only single tweets, but threads of tweets, focusing on the responses of other users to the tweets in question, and further entries, influenced by earlier reactions of the audience. Moreover, the number of likes and retweets received would need to be taken into account. The examination of networks of retweets and interactions between users would be of enormous interest not only with regard to identity, but also with regard to aspects of language change, illuminating some of the processes involved in the spreading of the term covidiota. In addition, further studies of the subject of positioning and identity in the context of the COVID-19-pandemic in social media should include other target terms as well as comparative analyses of the use of COVID-neologisms in different languages and on different social media platforms.

### References


### Olga Kellert Using Geolocated Tweets for Probing Language Geography and Migration

Resumen: Este artículo explora los tweets geolocalizados para indagar en el conocimiento de la geografía lingüística y la migración. Discutiré dos casos: la distribución de tweets en español en el mundo y la distribución de tweets en español en la ciudad de Nueva York. Evaluaré mis resultados sobre la base de los datos del censo sobre el mundo de habla hispana y las comunidades hispanas que viven en la ciudad de Nueva York. Los resultados muestran que Twitter es una buena fuente de datos para el estudio de la geografía lingüística y la migración, pero también tiene ciertas restricciones y limitaciones que deben ser abordadas en el futuro.

### 1 Introduction

In this article, I discuss a new data resource social media Twitter. I especially focus on so-called geolocated data. Geolocated data in Twitter is data that contains information about the exact location of the person who distributed a text message on Twitter. My research question is how geolocated data can be used in the domain of language geography and migration.

The article is structured as follows. Section 2 presents some basics about Twitter and technical details regarding the use of Twitter data as a data resource. Sections 3 addresses Twitters limitations in sociolinguistic research. I will present some case studies, which will highlight some of the problems with the social media data and some solutions to handle them. Finally, Section 4 provides a technical and a critical discussion in which the benefits will be summarized, and technical aspects are briefly discussed.

### 2 Social media Twitter

The following section contains basic information about Twitter and geolocated data from various previous research articles (Kellert 2022, Kellert & Matlis 2022a/b, Kellert 2023a/b).

Olga Kellert, Georg-August-Universität, Göttingen

Twitter is characterized as a "microblog platform", with the prefix "micro-" referring to the brevity of the entries. Registered users can use Twitter to disseminate short messages (tweets). One of the advantages of using social media like Twitter as a data resource for linguistic analysis is the large amount of data (Hovy/Purschke 2018; Grieve et al. 2019; Gonçalves et al. 2018; Gonçalves/Sánchez 2014; Rahimi et al. 2017; Kellert/Matlis 2022a/b, Kellert 2022). Another huge advantage for research studies in various fields is the metadata that is available for research purposes together with the text message. Metadata is, on the one hand, information about Twitter's users (e.g. the location the Twitter's users specify when they create a profile like Caserta in the province of Milan in Italy probably represents the user's "place of residence"). Another metadata is precise geolocation, i.e. information about the exact location from where a text message as in (1) was posted, which is expressed by a pair of coordinates, e.g. 18.46730168, −69.87971712.

(1) Acabando de darle una visita a la playa [. . .] 'We just visited the beach [. . .]'

If we plot the coordinates of this text message in google maps, it will show us that the message was posted from the beach Playa Punta Torrecillas in Santo Domingo in the Dominican Republic in the Caribbean islands. The precise geolocation information can be used to explore linguistic phenomena in space (and time) and to detect linguistic change (Hovy/Purschke 2018, Kellert 2022). For example, geolocation can be very useful in detecting changes in specific locations, e.g. when linguistic change occurs in very specific places, such as national borders, rural or urban regions. We know from both dialectology and sociolinguistics that linguistic variation takes place on a very small scale (Labov 2006).

I will now present few more technical details that are necessary for understanding the case study. The following lines present some metadata associated with the tweet text such as the name of the user, which I replaced by a fictitious name in line with anonymization, the location that the user indicates in his/her profile when opening an account and the precise geolocation when the text was posted expressed as coordinates encoded by two consecutive numbers. The first number corresponds to the latitude, the second to the longitude:


The text as well as other metadata represent a Tweet object. Scientists get free access to the data or Tweet objects by means of an identification code called an API key (API is an abbreviation for Application Programming Interface). This API key grants access to the Twitter server. The data stored there on the server has the format (JSON) as shown in https://i.stack.imgur.com/3UIoI.png. This format allows researchers to process the data more easily because it is uniformly formatted and makes it easier to query. What data researchers retrieve from the server depends largely on the research question and Twitter only gives access to public tweets, i.e. tweets that can be seen by everyone on the internet and are not protected. The query strategy has to be very carefully considered and needs to be documented precisely for the traceability and reproducibility of the research results. The data retrieved from the server can be saved on the own computer as a compressed file. This way one can create a corpus of tweets. However, this corpus of tweets cannot be published by researchers as it is legally property of Twitter. This fact makes it difficult to publish the data that is used by researchers, which can pose a problem for the reproducibility and transparency of scientific results. Scientists have come up with various solutions to this problem of Twitter. One prominent solution is to provide the code used to filter out tweets and all the data necessary to reproduce the results (Kellert 2023b). This procedure is nowadays very common in publications of scientific results in many high standard journals.

As with any corpus, large amounts of data cannot be searched manually. Consequently, programs are needed to do this automatically. In this case, the programs must be able to recognize and work with the coding of the JSON object. Nowadays, many researchers use the Python programming language to write appropriate programs. With Python, we can formulate tasks such as filtering data, doing calculations, and mapping coordinates onto maps, which is important for the case studies in this article (Bird et al. 2009). One task that can be formulated with Python is import Cartopy. Cartopy is an open source, that is, freely available and a modifiable software program that maps coordinates onto OpenStreetMap. OpenStreetMap is also freely available and is provided by a consortium of people. The following map of the United States of North America visualizes all locations where Spanish-language tweets were posted between 2017 and 2021 in the corpus that I have compiled over this period (Kellert 2022, Kellert 2023). I created the map from my own Spanish tweets corpus using the method discussed. Tweet IDs for figures can be provided for reproducibility upon request.

The method of applying geolocated data to answer linguistic questions is relatively new. To my knowledge, it was first systematically presented and applied by Mocanu et al. in 2013 in an article on mapping linguistic boundaries of multilingual countries and even cities such as Spain, Belgium, Quebec, Barcelona. This was followed by a great many studies in the field of dialectology, mapping regional boundaries of dialects based on geolocated tweets, e.g. Spanish dialects, English dialects, German dialects and Arabic dialects (Hovy/Purschke 2018; Grieve et al. 2019; Gonçalves et al. 2018; Gonçalves/Sánchez 2014; Rahimi et al. 2017; Kellert & Matlis 2022a/b, Kellert 2023). Dialectology is to date the most prominent area of the use of geolocated

Map 1: All Spanish tweets from 2017 to 2021 in USA with OpenStreetMap and Cartopy. Base map and data from OpenStreetMap and OpenStreetMap Foundation under the Open Database License.

data in linguistics. There are also individual studies on sociolects, i.e. studies that investigate the social aspect of linguistic variation and use geolocated data to show, for example, that linguistic forms in cities differ from those used in rural regions or studies that prove that linguistic aspects correlate with socio-economic and/or political background (Levy Abitbol et al. 2018; Kellert 2022). This article discusses two other cases from the domain of sociolinguistics on the basis of geolocated tweets. Apart from Linguistics, geolocated data has been applied in many other research domains and disciplines, which are not the focus of the present study.

We now turn to the usefulness of this new data resource and precisely the geolocated tweets in sociolinguistics, which is the focus of the present article.

### 3 Case Studies

In this section, we start with a very simple example, namely the location of all Spanish tweets from the period 2017–2021 on a world map. This case study will also highlight some of the problems with the social media data. The following world map shows the absolute frequency of Spanish tweets in the period 2017–2021 from my corpus.

Map 2: All Spanish tweets from 2017 to 2021 in the world with OpenStreetMap and Cartopy. The size of the circle marks the number of tweets in a particular city. Base map and data from OpenStreetMap and OpenStreetMap Foundation under the Open Database License.

The intensity of the color and the size of the circles mark from which regions most Spanish tweets have been posted (see Kellert & Matlis 2022a/b). As expected, the Map 2 and precisely the intensity and the circle size show that most Spanish tweets come from South America and Spain. The tweets distribution is largely in line with the facts. Spanish is indeed spoken in Spain and Latin America. The map from Britannica<sup>1</sup> shows where Spanish is the official language and widely spoken language.

If we compare the two maps, we see that there are hardly any tweets in Cuba, Bolivia, Northern and Central Africa, where Spanish is widely used according to Britannica. Various reasons can be enumerated for this lack of representation such as censorship or poverty, e.g. some people cannot afford internet or technical devices such as computers, smartphones, etc., Twitter is simply not popular or people tweet in another language. This insight is important for the evaluation of the results and for the selection of Twitter or other social media as a data resource for certain research questions. For example, we cannot use Twitter to conduct a study on Spanish used in Equatorial Guinea, Bolivia and Cuba. The fact that certain countries are missing in the representation of tweets raises the question about the "representability" of geolocated tweets (Nguyen et al. 2016, Blank & Lutz 2017).

https://www.britannica.com/topic/Spanish-language#/media/1/558113/239454, accessed February 25th, 2022.

Map 3: Differential distribution of Spanish tweets from 2017 to 2021 in NYC from tourists in red and locals in blue with OpenStreetMap and Cartopy. Base map and data from OpenStreetMap and OpenStreetMap Foundation under the Open Database License.

The illustration on the Map 2 also shows Spanish tweets in countries where Spanish is not official or commercially used language. This is particularly easy to explain in countries and cities that attract tourists, such as Italy, London, Paris and Berlin. Spanish is not normally spoken there, or it is not an official language there. However, there is one problem with this method. All groups of users are included in the representation of Spanish tweets on the maps, and no distinction is made as to whether the users are native speakers or non-native speakers of Spanish and whether the tweets are written by locals, tourists or migrants. The distinction between groups (such as +/- native speakers) is of course very important in linguistics, because linguists make a difference between native and foreign language studies (L1 and L2 languages and language learners). However, there is no direct marking/ annotation of these groups in the metadata of tweets. In order to distinguish between speaker groups such as tourists, we have to apply very elaborate methods with the help of which we can indirectly derive the missing information about the tourists (Kellert & Matlis 2022 a/b). I want to demonstrate one possible solution how to distinguish tourists from locals suggested by Kellert & Matlis 2022a. Let us consider Spanish tweets in New York. New York is a big touristic city, but it is also a multilingual city, where many Hispanic communities live according to Census Data from 2010.2 According to the information from the Census Data from 2010, the neighborhood Bronx stands out especially by the number of Hispanic communities. The two largest groups of Hispanic population living in the Bronx is Puerto Ricans and Hispanic communities from Dominican Republic according to Census Data. Let us now see whether tweets can replicate Census Data collected by survey methods. Map 3 of New York based on OpenStreetMap shows Spanish tweets from tourists from Spain, i.e. users who indicate in their account that their residence is in Spain. Another group plotted on this map are users who indicate in their account that they are from New York, which I define as locals (see Kellert & Matlis 2022). The method used to calculate the spatial distribution of the tweets posted by these users is called Differential Distribution, which highlights the areas where there are relatively more of tweets posted by one group than the other group. The color red marks where there is relatively more tweets posted by users from Spain and blue marks where there is more tweets posted by users from NYC. The size of the circle demonstrates where the number of tweets is particularly high (see Kellert & Matlis 2022a for details of this method). We see that the distribution of tweets from locals and tourists from Spain matches our expectation. There are more tweets written by tourists in Manhattan than elsewhere. What is even more important is that most locals tweet from the neighborhood Bronx. This matches the Census Data from 2010 and gives even more information, namely that not much has changed since 2010. The neighborhood Bronx is still the most populated area by Hispanic communities in NYC.

If we look at the most frequent words and collocations (i.e. sequences of two words) uttered in these tweets from locals, we see that República Dominicana is one of the most common collocation in the tweets. I used the standard analysis of collocations NLTK in Python's library to extract collocations automatically and to count them (Bird et al. 2009). This finding matches the Census Data as Hispanic communities from the Dominican Republic are one of the largest groups living in the Bronx. To conclude, I have shown on the basis of Spanish tweets how well Twitter data matches survey data collected with a completely different technique of data collection method than survey methods.

https://www1.nyc.gov/assets/planning/download/pdf/planning-level/nyc-population/census2010/ m\_hspsubnum\_ct.pdf, accessed February 25th, 2022.

### 4 Summary and Discussion

I have shown that geolocation of Spanish tweets in the world largely matches the locations where Spanish is official language and/or where it is largely spoken. Geolocated tweets can be used to replicate known results about Hispanic migrants living in NYC according to Census Data.

Finally, I want to discuss some challenges of the new data source Twitter. Traditional and modern survey methods according to which participants are asked to fill out their linguistic background status and other demographic information including location of residence, their social status (Labov 2006), provide more sociolinguistically relevant information than Twitter does. Corpus based and computational studies based on social media or web-based data lack demographic information, e.g. metadata in Twitter lack direct information about age, gender, socioeconomic status, and education level, which often correlates with linguistic variation (Labov 2006, and many others). A lot of research has been done in this domain to recover sociolinguistic information from other sources (Nguyen et al. 2016; Kellert 2022). Socioeconomic information of Twitter users has been inferred from geolocation of posted tweets and Census Data that contains information about the socioeconomic value of the location area (Levy Abitbol et al. 2018). Gender information has been derived from linguistic gender of the user and/or user's profile analysis (Nguyen et al. 2016; Kellert 2022). Age has been inferred from probability measures of a combination of features such as textual features (e.g. 'came home from school, my teacher today . . . .'), linguistic features such as the use of profane/informal language style, degree of errors, etc. (Morgan-Lopez et al. 2017). Demographic information derived from social media data is approximate and does not reach the exactness of sociolinguistic studies and/or survey methods. Whatever metrics is found as a proxy for sociolinguistic variable such as age, socioeconomic status, etc. in Twitter, this metrics needs to be evaluated on offline data with precise sociolinguistic information. One possible solution to this problem is to combine social media data and traditionally acquired data in a survey setting of a similar data type such as communication data that can be found in conversation chats in social media and crowdsourced communication data (e.g. crowdsourced whatsapps from private communication) (Überwasser/ Stark 2017).

In summary, the Twitter data resource offers an interesting new data source for linguistic research, but it brings with it also some challenges such as certain biases and the lack of differentiation of speaker groups that need to be taken into account when using social media data.

### References


Section 2: Language and Identity Construction

### Enrique Pato "New" Technologies and "New" Identity Formation

The Case of Latino Migrants in the Multilingual Context of Montreal (Canada)

Resumen: Este capítulo trata sobre las "nuevas" tecnologías y la "nueva" formación de la identidad de los migrantes hispanos en la ciudad de Montreal, Canadá. Presenta una selección de narrativas características de los hispanos en contacto con el francés y el inglés dentro de este contexto multilingüe. En la sección 2, se repasan algunos datos relevantes relacionados con los nuevos medios y las tecnologías. La sección 3 se centra en varias cuestiones teóricas sobre la identidad y la pertenencia a un grupo. La sección 4 proporciona una primera descripción general de la "nueva" identidad de los hispanos en Montreal en relación con los usos y prácticas de la telefonía móvil y de internet. Por último, unas consideraciones finales son expuestas en la sección 5.

### 1 Introduction

Globalization, as is widely known, has altered the face of social, economical, political, cultural and linguistic diversity in societies all over the world. Vertovec (2007) calls this fact super-diversity. 'Superdiversity' speaks of rapid change and mobility. In this regard, Canada "has become in three generations or so one of the most multilingual and multicultural nations in the world" (Chambers 2003, 98). In Canada, migration management is linked to the economic policy of the country and to the (re)structuring of the labor market. That is why migration policies privilege particular groups. Latino migration is an international, permanent, legal and voluntary migration: as such, we see a relational identity in a specific situation.

In this chapter, I will focus on the specific demographic, social and cultural dynamics of Latino migration in Montreal (Quebec, Canada) and present an example of multilingual diversity in an urban context.<sup>1</sup> In doing so, Montreal will be

Following Lüdi (2008, 208), multilingual competences are considered "as linguistic resources available to members of a community for different purposes. The totality of these resources

Enrique Pato, Université de Montréal

Open Access. © 2023 the author(s), published by De Gruyter. This work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License. https://doi.org/10.1515/9783110746082-008

viewed as a migrant community of practice in which the processes of identity and linguistic construction take place, and as an example of dynamic experience of identity and subjectivity in the process of "becoming a member" of a particular group (Quebeckers). For the multilingual speaker, language and cultural choice is not only an effective means of communication but also an act of identity (Le Page/Tabouret-Keller 1985).2

The study of Latino discursive, socio and ethnolinguistic facts, as well as the Latino linguistic system, can be undertaken thanks to the Oral Corpus of the Spanish Language in Montreal (Sp. Corpus oral de la lengua española en Montreal, COLEM), a new corpus for the study of migration and multilingualism in Canada. In section 4, we will put forward a few examples of what Latinos express regarding their idea of success, insertion in the country, and interculturality, as well as some of their thoughts on second generation migrants. I will concentrate on identity in language group membership, rooted in a particular place: Montreal.

From a theoretical point of view, Social Identity Theory addresses the question of social categorization and belonging to a certain group (cf. Morin 2021, 6). According to its postulates, in addition to facilitating our cognitive organization of the world and understanding of complex social environments, social categories have a significant impact on the development of individual identity and intergroup relations (Tajfel/Turner 1979, 40). Islam (2013, 1781), for instance, offers a summary of the premises and implications of this theory:

Social Identity Theory [. . .] begins with the premise that individuals define their own identities with regard to social groups and that such identifications work to protect and bolster self-identity. The creation of group identities involves both the categorization or one's "ingroup" with regard to and "out-group" and the tendency to view one's own group with a positive bias vis-a-vis the out-group. The result is an identification with a collective, depersonalized identity based on group membership and imbued with positive aspects.

Meanwhile, Lagarde (2013) carries out a reflection on otherness and what it means to be a "foreigner" (migrant), as opposed to "autochthonous" (local). This author emphasizes the contradictory tensions created by nationalist essentialisms that complicate the construction of a single and indivisible identity in the migrant. His proposal highlights the challenge of weaving said identity in a context in which adopting a defensive posture against what is labeled as "different" is

constitutes the linguistic repertoire of a person or a community and may include different languages, dialects, registers, styles and routines."

As Juillard (2021, 267) points out, multilingualism refers to the co-presence of languages in a given area or society. Plurilingualism, by contrast, refers to the diversity of real, individual and nonstandardized languages.

commonplace (cf. Morin 2021, 7). According to Lagarde, discourses that value the "integration" and "assimilation" of migrants constitute an attempt to eliminate the "Other".

### 2 Corpus and Methodology

COLEM (Pato 2014-2022) is the only corpus of semi-directed interviews (spontaneous speech) with first- and second-generation migrant native speakers of Spanish elaborated in Canada. To date, COLEM comprises 153 personal interviews conducted between 2014 and 2019. These data will allow us to further our knowledge of the Spanish-speaking community and will help us in the description and analysis of the Spanish varieties in Canada (Pato 2020, 2022, in press).

Representations of multilingualism and identity can be captured through migrant discourse, since discourse can be seen as the trace of representations of individuals and groups. From a methodological point of view, the type of topic addressed during the conversation is crucial. COLEM collects and observes a series of interactions and dialogues (life stories), contextualizes migrants' experiences and studies how mobility can affect deterritorialized languages in terms of their use, transmission, and evolution, in contact both with other languages (French, and English) and other varieties of the same language (Spanish). In short, COLEM helps us to better understand migratory experiences, and what migrants say both in interview situations and in social interactions.

As we well know, new media and technologies of communication and information, like mobile phones, Skype, Facebook and the Internet, among others, are part of our daily lives. People retain an active connection by means of an elaborate set of long-distance communication technologies. The so-called distance proximity provided by travels and the use of WhatsApp, Instagram, Telegram, etc. have social and linguistic effects, such as the use of fixed expressions, and the emergence of cultural and identity references that are shared on the Internet. An approach to contemporary conditions of cultural and linguistic contact should consider all these facts.

What is more, migrants maintain stronger links with home, and therefore remain active on two national stages. In such conditions, how do Latino people articulate belonging to different social worlds and communities simultaneously (Latin America and Quebec)? Intersectionality encapsulates the idea that in any historically specific context, a complex nexus of social, economic, political, cultural, psychic, subjective, and experiential axes come together (Budach/de Saint-Georges 2017, 67).

To present the data, I will analyse some of the narratives at the core of COLEM. Despite this practice being common in a wide range of disciplines, and there is no consensus as to the methodology of narrative analysis. As indicated by Garner, there are some features "that are characteristic of narrative as a distinct discourse type. It involves a recounting of personal experience, whether of the teller or of someone else. It exhibits temporality: a set of more or less discrete events occurring in a chronological sequence towards a culminating point [. . .] Thus, loosely defined, narratives fulfil three broad communicative functions: entertainment, instruction, and the construction of personal identity" (Garner 2007, 44).

In the words of Giddens, "a person's identity is not to be found in behaviour but in the capacity to keep a particular narrative going" (Giddens 1991, 54, my emphasis).

All these methods of data collection, analysis and interpretation will help us to understand in more detail how identity may be more locally situated. Latino identity is viewed not as a fixed phenomenon but rather as a dynamic process. In sum, I choose to present narrative content of COLEM data, because "narrative can be used as a lens through which identity performance can be observed" (Llamas/Mullany/Stockwell 2007, 223). The examples will speak for themselves.

### 3 Theoretical Issues

Identity has been defined in various ways at different times. According to Norton, identity is how people understand their relationship to the world, how that relationship is constructed across time and space and how people understand their possibilities for the future (Norton 1997, 410). In Ecolinguistics, identity is a story in people's minds about what it means to be a particular kind of person (Stibbe 2015).

In terms of dissociation from another group, social identity invokes "the relationship between the individual language learner and the larger social world, as mediated through institutions such as families, schools, workplaces, social services, and law courts" (Norton 2009, 359).<sup>3</sup> On the other hand, cultural identity references "the relationship between an individual and members of a particular

We can define social identity as: that part of an individual's self concept which derives from his [sic] knowledge of his membership of a social group (or groups) together with the emotional significance attached to that membership. (Tajfel 1974, 69).

ethnic group [. . .] who share a common history, a common language, and similar ways of understanding the world" (Norton 2009, 359).

In addition, from the perspective of socio-linguistic theory and analysis, "we should expect 'identity' to be [a] more demanding and multi-layered concept – needing to be subdivided at least into the dimensions of 'feeling', 'knowing', and 'doing' rather than being treated as some composite form of 'being'" (Coupland 2010, 21).

In this respect, for Le Page and Tabouret-Keller (1985, 14), linguistic behavior is "a series of acts of identity in which people reveal both their personal identity and their search for social roles." This theory puts individuals first, and views identity and language use as potentially fluid. That is why from a social constructionist perspective, identity is conversely seen as a fluid and dynamic concept, something that we actively do or perform when engaging in language production (Llamas/Mullany/Stockwell 2007, 216).

According to Morin (2021), we can establish a link between the degree to which migrants identify with their host society and the way they designate its members. Following the Social Identity Theory, those Latino migrants in Montreal who most identify and sympathize with Quebec use the word Québécois in a generally positive way: in doing so, they value and protect their own identity. On the other hand, Latino migrants who fail to identify with the province of Quebec, or have a conflictual relationship with its society, use the word Québécois with more contempt, dissociating their own identity from fellow citizens. However, these two uses are not mutually exclusive, and migrants can use them according to the discursive context and situation.

As long as migrants maintain their identity, they will never become quite Québécois. As Lagarde (2013, 4) points out, confrontation with the foreign is "source de réactivation identitaire". Quebec as a nation has always identified with the French language. Upon facing otherness, and especially the Anglophone, what is "truly" Québécois is to be a Francophone. Hogg and Reid (2006, 7) already noted that French would constitute a fundamental component of the "group norms" that make up the social identity of Quebec and allow it to differentiate itself from other social groups. Being Québécois therefore implies being a francophone –and ideally, with a local accent (Giles and Rakić (2014)–. In line with these facts, it is understandable that some non-native speakers use the demonym Québécois with some contempt, considering all the protectionist connotations that this term implies.

Giles and Johnson (1981) provide other elements thanks to the Ethnolinguistic Identity Theory. These authors indicate that language –being both a communication tool and an identity component– can contribute significantly to the differentiation of ethnic groups. They call attention to the fact that "language has been shown to be an important aspect of ethnic identity, more important even than cultural background for some individuals" (Giles/Johnson 1981, 204). In such conditions, Québécois becomes a symbol of national identity reinforcement.

Finally, the choice of a specific language allows migrants to manifest their sociocultural identity should they wish to do so. In this sense, identity is understood as an individual and collective process (i.e. as members of a group), which brings together at least three major concepts: (i) identification, (ii) social image, and (iii) belonging. That is why in this chapter, identity will be understood as one of the processes associated with multilingual practices, and as a strategy to maintain distinct identities and social cohesion.

In the following section we will see some linguistic and cultural signs of group belonging and maintenance of cultural identity in Latino migrants in Montreal.

### 4 New Data from COLEM

Budach and de Saint-Georges (2017) show an example of diversification of individual linguistic repertoires, acquired in the context of migration. This repertoire often echoes the home country's common languages and language varieties. "While linguistic homogenization seems to reflect a normative stance of the sociolinguistic situation 'back home', we note that geographic diversification tends to be rather downplayed or hidden." (Budach/de Saint-Georges 2017, 71). Montreal is one of several geographical spaces where linguistic homogenization takes place: Latinos converge towards the Quebec-French language due to pressures of the local social and linguistic economy (see Belling/De Bres 2014, for the case of Luxembourg).

Furthermore, the practice of codeswitching has an expressive function, as it identifies the speaker as someone whose cultural identity is mixed (Gross 2009, 67); as such, it is a marker of identity and belonging (Billiez 1985). The effect of codeswitching in conversation –in going from one language to another– can be seen as a marker of status and role, with an effect either of distance, humor, or parody. The use of French words in Latino speech (populación, Fr. population; chomage, Fr. chômage) appears in interjections, in quotes, and in calls to the interlocutor (Pato 2020, in press).

A multiple membership and identity can produce simultaneous internal and external contradictions. Some migrants, for instance, want to lose the identity of their native land (Latin America) and become assimilated and identified with the new home country (Quebec, Canada). Others want to develop a new identity and feel more comfortable with being culturally hyphenated (cf. Li 2009, 48). As we will see in the following examples,<sup>4</sup> these Latinos maintain their own linguistic identity in spite of their extensive and intimate contact with Quebecers (some members of the Latino group thus being multilingual). This is relevant for identity maintenance and continuity.

One of the most important groups for Latinos is the 'extended family', which is part of the individual's identity. Latinos therefore tend to maintain a perception of the world based on 'us' and the sense of belonging to a group, which creates a relationship of mutual dependence. The telephone, Skype and new technologies are means to keep in touch with the family.

(1) [E:] . . . el hecho de estar lejos de tu familia o . . . ¿eso te, te molesta o . . . ? [I:] No, no. [E:] ¿No? [I:] Antes podía decir que sí me molestaba mucho, pero no, me he acostumbrado y no, ya no me molesta. [E:] Sí. Entonces sí, puedes ir . . . no, tampoco, no está tan lejos. [I:] Puedo con-, platicar con ellos en Skype o teléfono, entonces . . . [E:] Sí es verdad que, hoy en día . . . [I:] Gracias a la tecnología. (COLEM-México 1)5

A fact that is repeated in the narratives of Latino migrants is that unlike what happens in some Latin American countries, one can safely use new technologies in Montreal, Canada being a very safe country.

(2) [I:] [. . .] el lugar donde yo, o sea, estuve viviendo, fue retranquilo, no . . . ¿Sabés lo que es salir en la noche sin tener problema que alguien te esté siguiendo? ¿Poder decir, bueno querés hablar con tu mamá, con tu hermana por teléfono, no sé, con el . . . con quien vos quieras, salir a la noche, caminás, no tenés ningún problema, no tenés ese, ese miedo de que, de que te pase algo? (COLEM-Argentina 1)6

In all the examples, [I:] indicates the informant (Sp. informante), and [E:] the interviewer (Sp. entrevistador).

'[E:] . . . being away from your family or . . . does it bother you or . . . ? [I:] No, no. [E:] It doesn't? [I:] Before, I could say that it did bother me a lot, but no, I've gotten used to it and it doesn't, it doesn't bother me anymore. [E:] Yeah. So yes, you can go . . . no, it's not, it's not that far. [I:] I can con-, chat with them on Skype or on the phone, so . . . [E:] Yes, it's true that nowadays . . . [I:] Thank to technology' (my translation, E.P.).

'[I:] [. . .] the place where I, I mean, I was living, was quiet, I . . . Do you know what it's like to go out at night without worrying that someone is following you? Being able to say, well, do you want to talk to your mom, with your sister on the phone, I don't know, with . . . With whomever you want, go out at night, walk, without having any problems, without that, that fear that, of something happening to you?' (my translation, E.P.)

(3) [I:] Eh . . . desde que estaba en mi país que quise, ehm . . . [CHASQUIDO] me metí en internet y que mi tío me contaba que era una . . . un país muy . . . muy tranquilo . . . [E:] Mhm. [I:] . . . eh, con mucha paz. [E:] Mhm. [I:] ¿Verdad?, para mis hijos. Y tiene muy, muy buenas oportunidades. Entonces yo busqué por internet. Y . . . y sí, me gustó mucho. Entonces por eso yo le insistí a mi tío que quería venir para acá por la mis- . . . por lo mismo: por la seguridad, por la tranquilidad, por la paz . . . (COLEM-Guatemala 9)<sup>7</sup>

In the new society, children do not stay at home when they get married, or have a stable partner and their own children. Smartphones and technology are used to keep in contact with the family. Grandchildren in Canada can meet grandparents who are in the country of origin.


'[I:] Eh . . . ever since I was in my country I wanted to, ehm . . . [CLICK] I got on the Internet and my uncle used to tell me that it was a . . . a very . . . very calm country . . . [E:] Mhm. [I:] . . . uh, with a lot of peace. [E:] Mhm. [I:] Right? For my children. And it has very, very good opportunities. So I searched on the Internet. And . . . and yes, I liked it a lot. So that's why I insisted to my uncle that I wanted to come here for the same . . . for those same reasons: for security, for tranquility, for peace . . . ' (my translation, E.P.).

'[I:] My daughter lived in my house for twenty-three years. Now she has . . . she left home, she has her own family, she has her own things. So it's like . . . for me, an adaptation too. [E:] Yes. [I:] But, I try to let her come to me, see? She calls me every day. Now we have the technology, which is incredible. [E:] Hmm. [I:] Well, we have smartphones, with which we can see each other live and directly. [E:] Mhm. [I:] That way she [my granddaughter] can also recognize me. [E:] Sure. [I:] My voice, uh . . . my face, when she sees me, she knows who I am. So all those things facilitate [contact]. [E:] Yeah' (my translation, E.P.).

[I:] Por todo [RISAS]. Ven los abuelitos en la . . . en el teléfono. [RISAS]. Sí. (COLEM-El Salvador 6)<sup>9</sup>

(6) [E:] Mhm, ¿y estás al tanto de lo que sucede por allí? [I:] Claro, claro . . . yo tengo amigos en México, toda mi familia está en México, eh . . . mi padre, mi madre, mis hermanos, mis amigos, eh . . . ahora con el internet nos podemos enterar de todo. (COLEM-México 4)10

As Dyer points out, we can see that identity is a function of both self and other ascription. (Dyer 2007, 103). Migrants also project their identity by retaining traditional practices, like cooking. Nowadays, they can call their grandmothers and ask them for a traditional family recipe. Internet is also a means of getting to know and learning to make the country's dishes in those cases where the migrant does not have the recipe at hand.


'[E:] Eh, well, do they talk, talk a lot with their grandparents . . . granny and grandad? [I:] Yes, yes, they talk to their granny and grandad. [E:] On the phone or, uh, Skype? [I:] By phone, by, uh, by phone, by Facebook, by . . . Nowadays, with social media! . . . [E:] Aha. [I:] For everything [LAUGHS]. They see their grandparents in the . . . on the phone. [LAUGHS]. Yeah' (my translation, E.P.).

'[E:] Mhm, and are you aware of what's going on over there? [I:] Of course, of course . . . I have friends in Mexico, my whole family is in Mexico, eh . . . my father, my mother, my brothers, my friends, eh . . . now with the Internet we can find out about everything' (my translation, E.P.). '[I:] So now we have learned to cook many things, I still have my granny, who's still alive, and when I think of making a recipe I call her on the phone. She is in her eighties, she is still very lucid, so she gives me the recipes for sauces and dishes and so on' (my translation, E.P.).

'[I:] [. . .] I listen to music in Spanish, I still eat my Ecuadorian food . . . [E:] Mmm. [I:] And now, even more so with the Internet, if I don't . . . if I don't have the recipe, I look for it and I make it. So me, my culture and my roots have always been very, very important. And I've always been with the Latino community' (my translation, E.P.).

Cellphones also allow migrants to maintain some of their original customs, such as listening to their favorite radio shows. Ultimately, Internet serves to prevent migrants from being disconnected from their roots. What is more, speakers "can create the identity they wish to project in an interaction" (Dyer 2007, 105).


As far as preserving one's roots goes, it must be said that one does not simply trade football-soccer for the national sport of the host country: hockey. In this sense, self-identity is an evolving story migrants tell themselves and interviewers about what kind of person they are (Stibbe 2015).

(11) [I:] [. . .] a mí me gusta el futból, y yo, este, los domingos, los sábados, eh, este, me pongo al internet y busco los partidos de las Chivas, ¿no?, y eh, y, pero ya

'[I:] My husband a little, yes, it's the radio, because he can't stop listening to the radio from over there . . . [E:] Mhm. [I:] . . . which is with a program, well, there is a host, Mario Pergolini, whom he's being listening to since, I think, ever since he was a teenager, and so now it always allows us . . . There is an Ivox program that allows you to download radio programs on your cellphone and listen to podcasts. And then, eh, that, perhaps, eh, he just can't let go of the radio stations from, from Argentina. But me, nothing. And I do not care' (my translation, E.P.).

'[I:] . . . I check the news on the Internet, on the radio, I listen to my country's news broadcasts, I am aware of what is happening politically and economically, obviously not as much as when I was there because, because I was immersed, but, they are my roots, it's my people, I care about what happens there, it's a way of not disconnecting' (my translation, E.P.).

ahora lo asumo mucho más, ¿no?, así que no, a mí el hockey me vale madres, ¿no? (COLEM-México 7)15

One of the cultural differences between Latino America and Quebec-Canada is being able to phone friends or acquaintances and drop by their home to visit. Latinos are 'closer' than Canadians regarding this type of social behavior.


Let's dig a bit deeper into what are known as 'transnational lives and identities'. In the countries of origin, relatives and friends arrive at one's house unannounced; in the new host society, that is not possible. The following quote puts forward an example of the devaluation of the relational identity of 'mother' and 'daughter-also-mother', which is caused by the fact that in Montreal, one changes.

'[I:] [. . .] I like soccer, and I, um, on Sundays, Saturdays, uh, um, I go on the Internet and look up the Chivas' games, right? And uh, and, but, now I accept it much more, right? So no, I don't give a damn about hockey, right?' (my translation, E.P.).

'[I:] I think they are a bit more reserved. Eh, at least that's . . . it's the perception. [E:] Sure. [I:] Eh . . . Or maybe it's . . . it's influenced by the fact that someone fits better with, with the type of culture that is more ours, more Latin. [E:] Mhm. [I:] I'll call you on the phone and come to your house in five minutes. [E:] [LAUGHS] Yeah. You can't do that here. [I:] It's a bit difficult to do that here' (my translation, E.P.).

'[I:] . . . people are more curt, to me, yes, I am . . . [PAUSE], well, they might just be prejudices, what do I know, because I expected more, and you know how . . . how we are close and everything, the people here, like, you'll call them just to say hello and they'll be like, "What did you call for?". You know? [LAUGHS] Yeah, like sometimes I was like . . . [PAUSE] Yeah? Like [NOISE] I didn't feel like the . . . [E:] It's hard to transcend relationships . . . ' (my translation, E.P.).

This migrant now expects guests to pick up the phone and let them know that they will be dropping by, so as to respect the fact that their host might have plans of their own. Contrary to example (1), in Quebec, the notion of the 'family' prevails in which individuals cultivate their personal identity based on the self; as a result, people value others for their individual qualities and characteristics. The importance of the individual manifests in the value placed on independence.

(14) [I:] Y ya ca- . . . uno cambia. [E:] Mhm. [I:] Entonces no te . . . Ya, por ejemplo, que lleguen a mi casa de sorpresa. Porque la gente pues llega a las casas sin avisar, tú sabes . . . [E:] Sí. [I:] . . . cómo es el asunto [RISAS]. Entonces ya yo no puedo eso, a mí me tiene que, y yo generalmente voy con tiempo contado; voy a visitar aquí, a, voy a hacer así, así. Entonces que me lleguen así de sorpresa, no me gusta porque me gus ta que me digan: "voy a ir". Entonces ya pues. Yo preparo es te tiempo para esa persona que va a venir. Pero que me lleguen así, no. Ya no. Yo digo: "para eso hay un teléfono, mamá". –"Ay, [NP], pero eso no es así; tú no eras así antes". –"Es que aquí la cosa es diferente". Y yo: "sí, sí pero es que entiende que yo no, no. Tú no puedes como dis poner de mi tiempo. No, tú me preguntas primero porque ya yo tengo cosas en mi cabeza planeadas qué hacer. O sea, yo ya tengo qué hac- . . . ya yo tengo como todo mi planning en la cabeza y no me va<sup>s</sup> a venir a cambiar la cosa". [RISAS]. Pero es que es así, uno cambia, ¿no? [E:] Sí. [I:] Ya yo no puedo. Ah, si ya no me llaman por teléfono entonce<sup>s</sup> no, no. No me aparezcan así. (COLEM-Venezuela 4)18

Another custom that is pointed out by migrants for being quite different from their own and causing one to change in the new host society has to do with the

'[I:] And now cha- . . . one changes. [E:] Mhm. [I:] Then you don't . . . Now, for example, that people come to my house unannounced. Because people drop by the house without warning, you know . . . [E:] Yes . . . [I:] . . . how things are [LAUGHS]. So, I can't do that anymore, I have to, and I only have so much time; I'm going to visit one place, to, I'm going to do this, do that. So, when people come by the house and take me by surprise, I don't like it, because I like for them to tell me: "I'm going to come over." And then, fine. I make time for the person who's going to come. But for them to drop by just like that, no. Not anymore. I say: "That's why we have phones, Mom." "Oh, [NP], but that's not how things work, you weren't like that before." "It's just that things are different here." And I say: "Yes, yes, but you have to understand that I can't, no. You can't, like, make use of my time as you please. No, you ask me first, because I already have things to do planned in my head. In other words, I already have things to d- . . . I already have all my planning in my head, and you're not going to come and change things around". [LAUGHS]. But that's the way it is, one changes, right? [E:] Yes. [I:] I just can't anymore. Oh, if they don't call me on the phone, then no, no. Don't drop in on me like that' (my translation, E.P.).

relationships between neighbors. In Montreal, people who live in the same building do not exchange emails or phone numbers: "they see each other when they cross paths, and nothing more". In other countries, like Cuba, neighbors are part of one's life, and they are almost like family (cf. example 1). This is a fact that many migrants stress in the interviews.

(15) [I:] Ya una ve<sup>z</sup> que uno vive aquí, uno cambia. [RISAS] Él tampoco. Jamá<sup>s</sup> nos hemo<sup>s</sup> dad o ni el número de teléfono, ni el correo, ni nada. Nos vemos cuando no<sup>s</sup> vemo<sup>s</sup> y si no, no nos vemos . En fin . . . [E:] Sí. E<sup>s</sup> verdad . [INSPIRACIÓN] Y, y es, en, en Cuba el rol de los vecinos e <sup>s</sup> completamente distinto, ¿no? [I:] E<sup>s</sup> ah, sí. E<sup>s</sup> otra cosa; forman parte de tu vida, de tu familia, de, de [ÉNFASIS] todo. (COLEM-Cuba 7)<sup>19</sup>

In the past, when there was no Internet or Skype, migrants kept in touch with their families by mail and by phone, very occasionally. The important thing was to be able to settle down, and some spent a long time without returning to their country of origin, especially when their parents had died.

(16) [I:] Entonces yo no tenía gran apego en Colombia y una vez que mi madre había muerto y mi papá también . . . pues como no co- . . . apenas acababa de conocer mis hermanos que eran pequeños, pues tampoco había un lazo muy estrecho allí. Y más bien tenía yo relaciones con primos y primas de las otras familias. Y pues, nos manteníamos en contacto por correo . . . por correo aéreo . . . que era, en esa época no había internet. Y por teléfono de vez en cuando que nos llamamos, pero, pues . . . no, me vine acá y por eso pasé treinta y un años porque pues, me ocupé de establecerme bien aquí, de seguir el curso de mi vida, mis estudios, mi trabajo . . . (COLEM-Colombia 7)20

'[I:] Once you live here, you change. [LAUGHS] Neither does he. We've never exchanged either phone numbers, or emails, or anything. We see each other when we cross paths, and nothing more. Anyway . . . [E:] Yes. It's true. [INSPIRATION] And, and it is, in, in Cuba the role of the neighbors is completely different, right? [I:] It's, ah, yeah. It's not the same: they're a part of your life, of your family, of, of [EMPHASIS] everything' (my translation, E.P.).

'[I:] So there wasn't much keeping me attached to Colombia, and once my mother had died and my father too . . . well since I didn't kn- . . . I'd just barely met my brothers who were small, so there wasn't a very strong bond there. I maintained more relationships with cousins from other families. And well, we kept in touch by mail . . . by air mail . . . that was, at that time there was no Internet. And from time to time we called each other on the phone, but, well . . . no, I came here and that's why I spent thirty-one years [here], because, well, I took care of establishing myself well here, of following the course of my life, my studies, my job . . . ' (my translation, E.P.).

Something as simple as installing a telephone in the house helps avoid problems of racism or vandalism, situations that some migrants have experienced in the past in Montreal. Such is the case with, for example, Guatemalans who work as day laborers. In addition to providing peace of mind, having a telephone at home is a symbol of integration in the new society, and a service that a migrant can afford, since it is not so expensive.

(17) [I:] [. . .] colocamos teléfono en la casa. O sea, nosotros mismos. Yo con mis compañeros me organicé para poner un teléfono. Teníamos nuestro propio teléfono para hablar con nuestra familia porque los de la calle, los . . . te digo, el racismo es como muy, muy marcado. Había unos muchachitos de catorce, quince años cuando los guatemaltecos iban a llamar a una cabina telefónica ... les llegaban a somatar, gritar, les golpeaban la cabina. Entonces, todo eso nos, nos . . . o sea, nos hizo hacer que mejor tener uno un teléfono en la casa, que nadie te va a molestar, ¿no? Entonce . . . Y no era tan caro. (COLEM-Guatemala 4)<sup>21</sup>

Older Latino migrants are aware that times have changed. Montreal's Latino population is now much larger (196,070 in 2021). In the past, hardly anyone spoke Spanish, especially in certain neighborhoods of the city (like Dorval or Kirkland), and there were no Latin surnames in the telephone directories. Nowadays, things are different.

(18) [I:] [. . .] cuando llegamos acá, en este, en la ciudad de Dorval y Kirkland, en ese tiempo no había nadie que hablaba en español. [E:] Mhm. [I:] Entonces, nos forzábanos más. Ahí no ha-, no había televisión hispana, no había nada. Era solo inglés. Usted abría el libro de teléfono, el directorio, y usted no miraba . . . nos encontrábanos nosotros, pero, [RISAS] aparte de eso, no había nada en español. Hoy hay bastantes personas latinas. (COLEM-Guatemala 5)22

'[I:] [. . .] we put a telephone in the house. That is, [we installed one] ourselves. I got organized with my colleagues to set up a telephone. We had our own phone to talk to our family because those on the street, the . . . I tell you, racism is like, very, very prominent. There were some fourteen, fifteen-year-old boys who, when the Guatemalans went to make a call in a phone booth . . . they came to mistreat them, yell at them, hit the booth. So, all of that made us, us . . . I mean, it had us thinking, better to have a telephone at home, so that no one is going to bother you, right? So . . . And it wasn't that exensive' (my translation, E.P.).

'[I:] [. . .] when we arrived here, in, in the city of Dorval and Kirkland, at that time there was no one who spoke Spanish. [E:] Mhm. [I:] So, we made more of an effort. There was no, there was no Hispanic television, there was nothing. It was only English. You opened the phone book, the

As it happens all over the world, migrants use the Internet to keep up to date with what is happening in their respective countries. This happens, above all, at the beginning of the integration process in the new host society. When we talk about integration, "we mean the desire to preserve one's own cultural identity and also to strike up relations with other groups" (Sancho-Pascual 2019, 3). Integration also implies a degree of change in one's own cultural traits and signs of identity. Upon taking root in Montreal, one feels more from Quebec than from Colombia, for example.

(19) [I:] En Colombia pues está mi abuela, unos tíos, amigas, y yo con alguna frecuencia me comunico con ellos, claro . . . [E:] Mmm. [I:] Para estar al corriente, pero si no, nada más. O sea, yo no vivo pendiente de . . . bueno, recién llegué, yo sí entraba a la página de internet de El Tiempo para saber cómo estaban las cosas, pero hoy en día, ni se me pasa por la cabeza. [E:] Mmm. [I:] Sí, de manera general, si algo pasa, pues sí se entera uno por las noticias de acá, pero tiene que ser algo muy grave, pero si no, no. No vivo tan al corriente de todo lo que pasa en Colombia. No. Como que no . . . ya, como se dice, ya uno ha echado raíces acá después de doce años. Ha echado uno raíces y eso hace que . . . no dejo de pensar en mi país, obviamente, porque me encanta mi país, pero sí se siente uno más de acá que de allá. Eso sí es verdad. (COLEM-Colombia 6)<sup>23</sup>

Internet is also used to meet other people and, eventually, leave one's home country to start a new life overseas. This is how some migrants arrive in Canada. However, it does not always go as planned: after a few years, people may separate.

(20) [I:] Conocí alguien sobre internet a los dieciséis años y tuvimos una relación a distancia durante dos años. Y después, problemas de familia . . . Viste cuando

directory, and you didn't look . . . we ourselves were there, but [LAUGHS] apart from that, there was nothing in Spanish. Today there are quite a few Latino people' (my translation, E.P.). '[I:] Well, my grandmother, some uncles, a few friends are in Colombia, and I communicate with them once in a while, of course . . . [E:] Mmm. [I:] To keep up to date, but otherwise, that's all. In other words, I don't keep an eye on . . . well, when I'd just gotten here, I did visit El Tiempo's website to find out how things were, but nowadays, it doesn't even cross my mind. [E:] Hmm. [I:] Yes, in general, if something happens, well, one does find out from the news here, but it has to be something very serious, but if not, no. I'm not that aware of everything that happens in Colombia. No. Like I don't . . . yeah, as they say, one has already put down roots here after twelve years. One has taken root and that makes me . . . I can't stop thinking about my country, obviously, because I love my country, but I do feel like I'm more from here than from there. That's true . . . ' (my translation, E.P.).

tenés . . . sos adolescente, que pensás que te llevás el mundo por delante, que conocés todo, y me fui. Me fui a los dieciocho años, decidí comenzar una vida nueva acá, me casé y, eh, cuatros años y medio más tarde me separé porque no funcionaba, y comencé a estudiar. (COLEM-Argentina 1)24

In the same way, Internet and Meta/Facebook serve to reconnect with old friends from the country of origin after one has spent many years in Canada.

(21) [I:] [. . .] no había tanto internet como hoy, y . . . y bueno, yo también me trastié mucho estando en Bogotá, entoc es en algún momento como que perdimos contacto con algunas amigas. [E:] Mmm. [I:] Y después de años, gracias a Facebook, volví a reubicar ciertas amigas. [E:] ¿Y las has vuelto a ver? [I:] . . . y las he vuelto a ver, sí, sí. No, eso ha sido bonito. Todos esos cambios en mi vida me han generado como, a veces situaciones bien, bien bonitas, especiales, porque he dejado de ver ciertas amigas después de años, y volvemos a encontrarnos y es bonito, bonito recordar todo eso. (COLEM-Colombia 6)25

Internet has also changed the way of being in contact with the country of origin. In the past, one could only keep up to date through the written press, which arrived, in the case of Chile, once a week. The migrants were held uncommunicated and were unaware of the reality of their country for a long time. Fortunately, things have changed.

(22) [I:] Yo estoy conectado con Chile. Por el cable. Entonces, me levanto en la mañana, click. Enciendo la televisión. Y estoy en Chile. Entonces, sé todo lo que pasa en Chile, eso es, eso es interesante porque, cuando nosotros llegamos aquí, no había televisión directa, como hoy. No había internet. Entonces, la

'[I:] I met someone over the Internet when I was sixteen and we had a long-distance relationship for two years. And then, family problems . . . You saw when you're . . . you're a teenager, when you think the world's your oyster, that you know everything, and I left. I left when I was eighteen, I decided to start a new life here, I got married and, uh, four and a half years later I separated because it didn't work out, and I started studying' (my translation, E.P.).

'[I:] [. . .] there wasn't as much Internet as there is today, and . . . and well, I also got very confused being in Bogotá, so at some point a few friends and I lost contact. [E:] Hmm. [I:] And after years, thanks to Facebook, I reconnected with certain friends. [E:] And have you seen them again? [I:] . . . and I've seen them again, yes, yes. No, that was nice. All those changes in my life have sometimes generated, like, good, very beautiful, special situations, because I've stopped seeing certain friends after many years, and then we meet again and it's nice, so nice to remember all that' (my translation, E.P.).

comunicación con Chile era muy difícil. Imagínate que llegaba el periódico que se llama La Tercera, y El Mercurio, una vez por semana. Entonces, mi hermano compraba el periódico, lo leía, y me lo daba a mí, y yo se lo daba a mi hermana. Entonces este periódico que era uno, yo leía las noticias de prácticamente un mes después. Entonces, todo este periodo fue muy incomunicado con respecto a la realidad chilena. Todo era por noticias atrasadas. Actualmente, no. O sea actualmente tú con internet . . . Esto de la globalización ha acercado a todo el mundo. Entonces con, con internet tú viajas, estás comunicado. Ahora el sistema de teléfono es inmediato, también no hay problema, antiguamente era muy complicado. Ahora con el cable yo estoy conectado a Chile, todo el día. (COLEM-Chile 3)26

Almost all Latino migrants in Montreal check their country of origin's news through to their smartphone. They stay connected and updated daily.

(23) [I:] Es que me tengo la aplicación de lo . . . de las . . . de las noticias de ahí. [E:] De las noticias. [I:] Entonces ahí en el teléfono, cojo, lo escucho: paf . . . Lo leo rápidamente y ya sé. [E:] O sea que desde que llegaste, ¿te has mantenido permanentemente actualizado? [E:] Sí, sí, sí. No . . . no ha habido un solo día que no . . . no reviso las noticias de Ecuador, lo que pasa ahí, eh . . . pues sí. [E:] Y . . . [I:] Siempre me mantengo bien conectado. (COLEM-Ecuador 1)27

'[I:] I'm connected to Chile. Through cable TV. So, I get up in the morning, click. I turn on the television. And I'm in Chile. So, I know everything that happens in Chile, that's, that's interesting because, when we arrived here, there was no direct television, like there is today. There was no Internet. So, communication with Chile was very difficult. Just imagine, the newspaper called La Tercera, and El Mercurio, arrived once a week. So, my brother would buy the newspaper, read it, and give it to me, and I would give it to my sister. So, this single newspaper, I would read the news of practically a month later. So, during this whole period I was very isolated from the Chilean reality. It was all belated news. Not now. I mean, these days, with the Internet . . . This globalization phenomenon has brought the whole world closer. So with, with the Internet you travel, you're connected. Now the telephone system is immediate, there aren't any problems either, in the past it was very complicated. Now with the cable I am connected to Chile, all day' (my translation, E.P.).

'[I:] It's that I have the . . . the . . . news application from over there. [E:] The news application. [I:] So there on the phone, I pick up, I listen to it: poof . . . I read it quickly and I already know. [E:] So since you arrived, have you been constantly updated? [E:] Yes, yes, yes. No . . . there hasn't been a single day that I don't . . . I don't check the news from Ecuador, what happens there, eh . . . yes. [E:] And . . . [I:] I always stay well connected' (my translation, E.P.).

However, other migrants choose to do the opposite, and personally decide not to maintain contact with the reality of their country of origin (cf. example 19). In this way, the image one has of society is connected to the identity of each person, the aspects that each one values. Therefore, the sensitiveness of each individual will depend on the social values and qualities that matter most to them.

(24) [E:] Bueno, eh . . . ¿tú te mantienes al tanto de lo que sucede en Colombia [PAUSA], no sé, tal vez a través de noticieros, eh, nacionales, por internet, buscas noticias o lo que te cuentan tus . . . ? [I:] ¿Sabes que no? Pues últimamente no ten- . . . es una cuestión de tiempo. Eh, las noticias casi todas son tristes y como tristeza sobra a veces, ¿sí? [E:] Mmm, sí. [I:] De tantas cosas que pasan en el mundo, tantas atrocidades y demás. [E:] Mhm. [I:] Entonces como que no. Cuando hablo por teléfono, mi mamá me dice algunas cosas que han pasado, pero trato . . . o sea, es demasiado para mí. (COLEM-Colombia 4)28

Conversely, new technologies and the Internet do not function when there is a natural disaster, such as an earthquake or tsunami. In these circumstances, migrants feel powerless, as they have no contact with their family, whom they cannot help.

(25) [I:] [. . .] estoy todo el día conectado a internet, ilimitado. Tonce tengo un poco más de relax, ¿cachái? Pero cuando pasó el temblor, el maremoto en Chile, imagínate. [E:] Tú estabas aquí. [I:] Estaba aquí en Montreal. Fueron cuatro días que no tenía ni idea de mi madre ni de mi padre, de nadie en Chile. No se podía hablar por teléfono, ni internet. De repente alguien encontraba internet y salían mensajes, pero es como, eso es lo, lo, lo más malo, estar fuera, no al alcance para hacer las cosas tan rápido. (COLEM-Chile 2)29

'[E:] Well, eh . . . do you keep abreast of what's happening in Colombia [PAUSE], I don't know, maybe through newscasts, uh, national ones, on the Internet, do you look for news or what your . . . ? [I:] Do you know that I actually don't? Well, lately I haven't- . . . it's a matter of time. Eh, the news is almost all sad and sometimes there's too much sadness, right? [E:] Hmm, yeah. [I:] Of so many things that happen in the world, so many atrocities and such. [E:] Mhm. [I:] Then, I just don't. When I talk on the phone, my mom tells me some things that have happened, but I try . . . I mean, it's too much for me' (my translation, E.P.).

'[I:] [. . .] I'm connected to the Internet all day, unlimited. So, I'm a bit more chill, you know? But when the earthquake happened, the tidal wave in Chile, imagine. [E:] You were here. [I:] I was here in Montreal. For four days I had no clue how my mother or my father was, or anyone in Chile for that matter. You couldn't talk on the phone or through the Internet. Suddenly someone would find an Internet connection and messages would come out, but it's like, that's the, the, the worst thing, being outside, out of reach and unable to do things so fast' (my translation, E.P.).

As we have seen in (16), when there was no Internet, communication was difficult. Now new technologies make everything easier and allow one to re-establish the connection, and to find out about the circumstances in the country of origin.


'[I:] A distance was created because, of course, I detached myself a bit from Colombia and at that time communications were not so easy, like . . . there was no Internet . . . only through newspapers or what the television or the radio reported . . . And well, here in Canada, unfortunately, those countries are only talked about when there is some tragedy, some catastrophe . . . So it was the only, the only knowledge I had of Colombia . . . apart from when my family came, well, of course they would update me on the political situation, the social evolution, all of that. So, yes, I was very detached during that time. Now it's different, because now there's the Internet and you can pick up the radio stations and newspapers at any time. So communication is easier. And also through the Internet, by uh, Skype, by all those means, one can communicate . . . [NOISE] daily with the family' (my translation, E.P.).

'[I:] Definitely, yes. It seems that one cannot exactly break the ties, and well, yes, yes . . . at first I read the local newspaper online, from, from Veracruz, to be more or less up to date and, on the other hand, well, I watch the news in Spanish that, that I can, that I have access to on my television . . . ' (my translation, E.P.)

As has been done in all migration contexts worldwide, when Internet and Meta/ Facebook had yet to exist, Latino migrants wrote letters to their friends and families. This habit has now been lost: emails have replaced the old letters.

(28) [I:] Viva Facebook. Yo primero que todo, yo no era de ese tipo de personas que se iba de un país y no volvía a hablarle a nadie. Yo a todos mis buenos amigos les escribí cartas, ahí tengo una caja llena de cartas que nos escribíamos, y yo desde kínder [RUIDO] escribía cartas a mis amigos. Eh, les escribía, eh, también, porque bueno, el internet salió cuando yo tenía diez años. Cuando llegué a Quebec el internet empezó a salir, pero no era tan fuerte, entonces yo como hasta a los . . . [PAUSA] como hasta los catorce años escribí cartas. Y mandaba cartas a cada mes por correo, timbres, todo, to el papeleo y ya después salió el internet. Entonces eran e-mails, y ahora estoy en contacto con absolutamente [ÉNFASIS] todo el universo, todos, hasta en la gente que hice kínder en Costa Rica . . . (COLEM-Costa Rica 3)32

Nevertheless, for older people, the use of new technologies and computers is more complicated, both in their country of origin and in the host country. Access to Internet is limited in certain regions of Latin America, and some people also lack training (which causes migrants to feel illiterate). Therefore, prepaid phone cards are still used to talk with parents and family.

(29) [I:] [. . .] Ehm, mi hermano también, hace poco su, estaba teniendo problemas con su esposa, y no poder estar ahí entonces es súper difícil y mi mamá, mi mamá que es mi mejor amiga, que siempre me está contando, me está contando todo, entonces, es súper difícil, ¿no? Como que . . . eh, ah, pienso que ahora es un poco más fácil con eso del internet y Skype y todo eso, pero mis papás están un poquito todavía a la antigua y como que les cuesta un poco, entonces, toca comprar tarjeta todo el tiempo y esas tarjetas que no sirven,

'[I:] Long live Facebook. First of all, I was not the kind of person who left a country and never spoke to anyone again. I wrote letters to all my good friends, I have a box full of letters that we wrote to each other, and since kindergarten [NOISE] I wrote letters to my friends. Uh, I wrote to them, uh, too, because well, the Internet came out when I was ten years old. When I arrived in Quebec, the Internet started to come out, but it wasn't that widespread, so I wrote letters until I was . . . [PAUSE] Until I was fourteen years old. And I sent letters every month by mail, stamps, everything, all the paperwork and then the Internet came out. Back then it was e-mails, and now I am in contact with absolutely [EMPHASIS] the entire universe, everyone, even the people who were in kindergarten in Cosa Rica' (my translation, E.P.).

que tienes que colgar y volver a llamar porque no se escucha bien. (COLEM-Honduras 5)33

(30) [I:] Después cuando . . ., empezaron a abrirse el . . . este sistema de . . . cartas de llamada. ¡Ay Dios mío!, nosotros vimos la gloria con eso . . . [SOLAPA-MIENTO]. [E:] [RISAS] De verdad que sí. [I:] . . . porque pagábamos cinco dólares y lo que nos daban en la carta quizás podíamos hablar dos o . . . dos, dos dólares, porque los otros tres se . . . [E:] [RISAS] . . . se iban. [. . .] Sí, qué caro era de verdad ponerse en contacto con la familia . . . [SOLAPAMIENTO]. [I:] Era increíble. [E:] Muy caro, sí . . . [I:] Sin internet. Una de las cosas que, que, que . . ., imagínate por decirte, a mí, eh, porque por, por mi edad, hay gente que, que están bien estudiadas, bien preparadas y todo y esto de los . . . de la tecnología no se les hace difícil. Para mí sí, porque yo tenía que entrar y yo decía: "¿Cómo, Dios mío? ¿Cómo yo voy a entrar? ¡Qué frustración! ¿Por qué yo nunca ni siquiera estudié nada y no sé nada, me siento una . . . ¡analfabeta!". Así me sentí muchas veces. Yo decía: "me siento una analfabe- . . . ". Lo que a mí me ha ayudado mucho ha sido ese empuje, ese carácter a no dejarme caer. Yo me acuerdo que mis hijos tenían su . . ., cada uno su computadora, mi esposo tenía su computador ahí, ¡y nadie me dejaba tocar! (COLEM-Ecuador 6)34

'[I:] [. . .] Ehm, my brother too, recently he, he was having problems with his wife and not being able to be there so it's super difficult and my mom, my mom who is my best friend, who is always telling me, she tells me everything, so it's super difficult, isn't it? Like . . . uh, uh, I think that now it's a little easier with the Internet and Skype and all that, but my parents are still a little bit old-fashioned and it's kind of hard for them, so we have to buy phone cards all the time, and those cards don't work well, you have to hang up and call again because you can't hear well' (my translation, E.P.).

'[I:] Then when . . ., they started to open up the . . ., this system of . . . call cards, oh my God, we saw the glory with that . . . [OVERLAP]. [E:] [LAUGHS] Absolutely. [I:] . . . because we paid five dollars and what they gave us on the card, maybe we could talk for two or . . . two, two dollars, because the other three were . . . [E:] [LAUGHS] . . . melted away. [. . .] [E:] Yeah, how expensive it was to get in touch with the family . . . [OVERLAP]. [I:] It was crazy. [E:] Very expensive, yes . . . [I:] Without Internet. One of the things that, that, that . . ., imagine, I mean: to me, eh, because, because of, because of my age, there are people who, who have a good education, who are qualified, and all that has to do with . . ., with technology, they find it difficult. It was to me, because I had to sign in and I'd say: "Good Lord, how? How am I going to sign in? How frustrating! Why, I never even studied anything and I don't know anything, I feel like . . . illiterate!" That's how I felt many times. I'd say: "I feel illiterate . . . !" What has helped me a lot has been that will, that character I have, never to give up. I remember that my children had their . . ., each one had their computer, my husband had his computer too, and no one would let me get anywhere near the devices!' (my translation, E.P.)

Internet is also used to look up Latino stores in Montreal and find out about events and groups with which migrants connect and feel a part of.


As for Latino children, it must be said that multilingual children rapidly acquire the communicative competence to use their different languages (Spanish, French, and English, in the case of Montreal) appropriately. They often experience these as being connected to different aspects of their identity (Maybin 2007, 162). In general, Latino parents want their children to retain the Spanish language and their culture, and children feel Latin American.

(33) [I:] Hablamos una mezcla de inglés y español. Entre mi esposo y yo hablamos español. Con mi hijo es mitad español, mitad inglés, porque yo quise siempre mantenerlos a ellos su, su . . . nuestro idioma. Así que los [ÉNFASIS] dos hablan español, escriben español y leen español. Yo no los quise dejar perder nuestro idioma. Eso para mí era muy importante, conservar nuestra cultura. De por sí que, [NP] el más chico, él nació aquí, pero si tú le preguntas a él y si tú miras en su perfil en internet él dice que él es de Panamá. [E:] Mhm. [I:] Ya

'[I:] . . . there are always Latino stores and we are always looking online . . . [E:] Are there any in your neighborhood as well? [I:] . . . yes, there are. [E:] Right? Now there are more, in every neighborhood' (my translation, E.P.)

'[I:] So through them I knew a lot of things. The, well, to begin with, they organized the parties. Ah, they had networking days, they also had storytellers . . . [E:] Uhum. [I:] Um . . . and it was like that that I also got to know other people who also went to, to the same, to those same events, and through those people I also found out about other things, and so on, and so on, and so. Um, but it all started with Amitié, above all. And . . . and doing, well, uh, searching the Internet' (my translation, E.P.).

. . . él es de Panamá, él dice. Él sabe que él es canadiense, pero él se siente panameño. (COLEM-Panamá 3)37

In multilingual contexts, the migrant shows their acceptance and interpretation of identity. This next narrative shows another example of socio-cultural affiliation development. Integration into the new host society is easier if other migrants help in the process.

(34) [I:] Llegamos juntas el mis mo día, el mis mo avión. [E:] [RISAS] A buscar el apartamento juntas . . . [I:] Ah, ya habíamos bus cado apartamento des de Puerto Rico por internet. [E:] Mhm. [I:] Tuvimos la bendición de encontrarnos con una señora por tuguesa . . . [E:] Mhm. [I:] . . . que nos adoctó. [E:] Ah . . . [I:] Sí, ella ... no era portuguesa, era brasileña. Ella nos adoctó, ella nos enseñó Montreal, ella nos enseñó toda<sup>s</sup> las , los trucos y . . . Sí, la verdad es que, yo creo que [PAUSA] yo creo que par te grande de mi integración se debe a que yo me encontré con esa familia. (COLEM-Puerto Rico 4)38

It is interesting to verify that identity is understood to comprise both personal and interpersonal facets, since Latinos necessarily define themselves in relation to Quebecers, Canadians, and other Latinos while engaged in social interaction with all of them in their day-to-day lives. For this reason, identity is always built through negotiating with others. This entails a gradual change in the diaspora, from traditional diaspora groups towards new émigrés. In identity theory, a person is in a liminal state when they feel the identity they once had ceases to apply to them, but have not yet found a "new" identity to replace it (Stibbe 2015).

'[I:] We speak a mixture of English and Spanish. My husband and I speak Spanish. With my children, it's half Spanish, half English, because I've always wanted for them to keep their, their . . . our language. So they [EMPHASIS] both speak Spanish, write Spanish, and read Spanish. I didn't want to let them lose our language. That was very important to me, preserving our culture. And so, [NP] the youngest, he was born here, but if you ask him and if you look at his Internet profile he says that he is from Panama. [E:] Mhm. [I:] Yeah . . . he's from Panama, he says. He knows that he is Canadian, but he feels Panamanian' (my translation, E.P.).

'[I:] We arrived together on the same day, the same plane. [E:] [LAUGHS] To look for the apartment together . . . [I:] Ah, we had already looked for an apartment from Puerto Rico online. [E:] Mhm. [I:] We were blessed to meet a Portuguese lady . . . [E:] Mhm. [I:] . . . that adopted us. [E:] Ah . . . [I:] Yes, she . . . she wasn't Portuguese, she was Brazilian. She adopted us, she showed us Montreal, she taught us all the, the tricks and . . . Yes, the truth is that, I think that [PAUSE] I think that a large part of my integration has to do with to the fact that I met that family' (my translation, E.P.).

To finish this sample, I will exemplify other facts that can cause the identity of migrants to change, such as the use of the language and the influence of French on Spanish.

(35) [I:] Exacto. Y, ehm, no, mi vieja me lo ha dicho muchas veces y mis hermanas también. Mismo cuando voy a Argentina, la gente . . . Mi familia se me ríe. Me dicen: "Mira la quebeca, la quebeca". Y digo . . . y yo digo, "pero loco, no puede ser, digo, o sea, sigo hablando igual". Me dicen: "No, vos no te das cuenta. No te das cuenta, pero sí, tu . . . tu forma de hablar cambió, la forma que estructuras las frases cambió". (COLEM-Argentina 1)39

Usually the lack of contact with the country of origin causes a migrant's proficiency in Spanish to deteriorate, even to be lost in some cases. Furthermore, unlike what occurs in the new language, French, their Spanish lexicon is not renewed (there is no enrichment of the language). This does not happen to migrants who keep in touch with family and friends, even through the Internet.

(36) [I:] Siempre hablamos las mismas palabras. Y nosotros somos personas que, creo que nos hemos disvinculado completamente de Bolivia, porque tengo un amigo que él está siempre en el internet, viendo qué pasa con las noticias; porque cuando uno está viviendo en el mismo país, hay lo que yo pienso que se puede decir que el enrequicimiento del idioma, porque salen nuevas palabras, uno siempre se está informando, eh, entonces cosa que uno cuando ya pierde ese contacto, eh . . . ya no. Estoy enriquiciéndome con el francés, pero no con el español, ¿no? (COLEM-Bolivia 2)<sup>40</sup>

To close this series of narratives, here is one last example of 'resisting' a local identity practice. Every migrant experiences numerous relations of identity and

'[I:] Exactly. And, um, no, my mom has told me many times and so have my sisters. Even when I go to Argentina, people . . . My family laughs at me. They tell me: "Look at the quebeca, the quebeca." And I say . . . and I say, "but, that's crazy, it can't be, I say, I mean, I still speak the same." They tell me: "No, you don't realize it. You don't realize it, but yes, your . . . the way you speak changed, the way you structure sentences changed.'" (my translation, E.P.).

'[I:] We always use the same words. And we are people who, I think, have completely distanced ourselves from Bolivia, because I have a friend who is always on the Internet, seeing what is happening through the news; because when one is living in the same country, there is what I think it can be called the enrichment of the language, because new words come out, one is always informing oneself, eh, so that's something that when one loses that contact, eh . . . is no longer true. I'm enriching my French, but not my Spanish, right?' (my translation, E.P.).

inclusion, as well as relations of opposition and exclusion. In this case, two girls and their mother undergo a cultural shock related to food and milk.

(37) [I:] Bueno, una cosa que a mí me ha, marcó chiquita, chiquita es que mi hermana, cuando llegamos aquí a Montreal, se hizo amiga de un muchachito québécois. Y la mamá para mostrar su buena fe, nos invitó a cenar. Entonces mi mamá, nos hizo las colitas, nos preparó los vestiditos, nos mandó súper bien vestiditas y bien prolijitas a la casa de esta señora y cuando llegamos a comer, nos habían servido hot dogs con chips. [RISAS] Nosotras . . . este, no teníamos derecho de comer chips. Entonces, de repente, como que . . . no entendimos mucho lo que había que hacer y mi hermana y yo . . . O sea, imagínate qué cuadro, ¿no? O sea, totalmente desprolijas. [RISAS] Bueno . . . Nosotras bien peinad itas y de to porque era la cena. Y era hot dogs con chips. Me acuerdo, hot dog con mostaza y las chips a lado y un vaso de leche. [E:] ¡Uh! [I:] Nosotros nunca habíamos bebido leche. Eso no se usaba pa nosotros. [RISAS] Me acuerdo que cuando volvimos a la casa, mi madre: "¿Cómo les fue, chicas? ¿Qué comieron?". "Hot dogs con chips". "¿Qué! ¿Cómo que hot dogs? Me están cargando". [IMI-TANDO] "No". "¿Pero qué les dio de comer?". "Hot dog con chips". [RISAS] "Pero, ¿cómo? ¿Había algo más?". "Un vaso de leche". [RISAS] Mi mamá . . . bueno . . . yo, hoy digo, mi mamá tiene que haber dicho como "estamos todos locos". Entonces resulta que a partir de ese día queríamos nosotras integrarnos y beber leche . . . [RISAS] a la hora de la comida. Mi mamá me dijo: "Eso es para los quebecos". Nunca más. [RISAS] Nunca más tomamos leche a la comida. Y nuestros hijos nunca tomaron leche en la comida. Eso no se usaba. (COLEM-Argentina 2)41

'[I:] Well, one event that left its mark on me when I was very, very little is that my sister, when we arrived here in Montreal, made friends with a young Québécois boy. And the mother, to show her good faith, invited us to dinner. So my mom, she made us pigtails, she prepared our little dresses, she sent us very well dressed and neat to this lady's house and when the time came to eat, they served us hot dogs with chips. [LAUGHS] We, um . . . weren't allowed to eat chips. Then, suddenly, like . . . we weren't quite sure what to do, and my sister and I . . . I mean, imagine the scene, right? I mean, totally messy. [LAUGHS] Well . . . We had pretty hairdos and everything because it was dinner. And it was hot dogs and chips. I remember, hot dog with mustard and chips on the side and a glass of milk. [E:] Wow! [I:] We'd never drunk milk before. That was just not something we did. [LAUGHS] I remember when we got back home, my mom: "How did it go, girls? What did you eat?" "Hot dogs with chips." "What! What do you mean, hot dogs? You're kidding me." [IMITATION] "No." "But what did she feed you?" "Hot dog with chips." [LAUGHS] "But, what? Was there anything else?" "A glass of milk." [LAUGHS] My mom . . . well . . . I, today I say, my mom must have said like "we've all gone crazy." So it turns out that from that day on we wanted to fit in with the others and drink milk . . . [LAUGHS] at mealtime. My mom told me:

### 5 Final Remarks

The post-structuralist conception of identity "places more emphasis on the individual and less on the community, and views identity as complex, contradictory, multifaceted and dynamic across time and place" (Dyer 2007, 105). In section 4, we saw some individual facts and actuations. Broadly speaking, Latino identity is maintained in Montreal, but the formation of a "new" identity is shyly beginning to shine through. The interesting thing is that this linguistic construction of identity takes place in a situation of multiple linguistic contact.

Latinos, as individuals or as a group, maintain several features that mark a separate identity, albeit partially overlaid by a local Québécois identity. However, since French is always an index of local identity, in some Latinos we can see the emergence of a new 'transnational identity'. Migrants can consolidate their own identity in a new community, showing their ability to learn and speak the languages of this community (French, and English in Canada).

From a linguistic point of view, Montreal (as a 'City') acts as a regulator of irregular forms, and a producer of lexicon, "sucking up" multilingualism and "expelling" monolingualism (Calvet 2021, 343).<sup>42</sup> Nevertheless, Latino migrants want to speak their language. They speak Spanish in the private domain, and in the public domain: at home, with friends, on the street, in Latin stores, and on public transportation. Spanish is transmitted and will be transmitted to children and grandchildren. However, Latino migrants are in a situation of diglossia (à la Ferguson), in the sense that several languages coexist and each one is used in a different domain, not necessarily in terms of dominant/dominated. In this sense, biand multilingualism is individual, and diglossia is spatial (of the territory). What's most fascinating is that one same individual can live and act in several nonexclusive spaces at the same time. That is why identity is viewed not as a fixed but as a dynamic phenomenon (Dyer 2007, 105).

<sup>&</sup>quot;That's for Quebecers." Never more. [LAUGHS] We never drunk milk at mealtime again. And our children never had milk at mealtime. That was not a thing for us' (my translation, E.P.).

In Calvet own words: "le rôle unificateur de la ville qui fonctionne comme une pompe aspirant du plurilinguisme et recrachant du monolinguisme" (2021, 343).

### References


### Jannis Harjus Virtual Speaker Design

Phonographic Uses of Andalusian Variables in Migration Contexts

Resumen: En esta contribución analizamos el Speaker Design en conversaciones de chat de hablantes del español en Andalucía que se encuentran en situación de migración, fuera de España. En concreto, mostramos que hablantes de origen cordobés y jerezano con formación universitaria y de mediana edad, que viven fuera de Andalucía y que no tienden a realizar determinados rasgos fonéticos del español hablado en Andalucía – como p.ej. el jejeo o el rotacismo –, representan gráficamente a estas variables en situaciones comunicativas por el medio escrito de WhatsApp. El uso de formas vernáculas fonográficas por parte de los hablantes, que en la mayoría de los casos escriben en español peninsular estándar, parece ser una estrategia comunicativa para ironizar, pero sobre todo para crear una identidad lingüística en situaciones migratorias. Por último, entra en juego una división del espacio comunicativo de Andalucía, que se refleja en una mayor convergencia de las variedades andaluzas orientales al estándar del español europeo y en un (hiper)dialectalismo muy presente en el grupo andaluz occidental.

# 1 Introduction

Since the beginning of the European economic and financial crisis in 2008, more than 2,200,000 people have emigrated from Spain to other countries in the European Union (cf. INE 2022). On the one hand, these are often relatively young, academically educated Spaniards. On the other hand, a transitory migration of these speakers can often be assumed, since a return to Spain after a successful (medium-term) stay in another state of the European Union is usually aspired to or has already taken place. At least the Spanish speakers with academic education from Andalusia who are in European migration situations and who are in the focus of this paper strive for circular migration, i.e., they are currently staying in a European country but plan to return to their home country after improvements in their personal economic-financial situation and the overall socio-economic challenges in Spain.

Jannis Harjus, Universität Innsbruck

During the speakers' time in the European diaspora, there is of course continued contact with Andalusian communities. Not only (semi-)annual personal visits to relatives and friends in the community of origin strengthen the contacts, but above all the constant communication via digital social networks with known speakers from the home country, who are either also in a migrant situation or have remained in their place of origin. The social situation of the speakers thus exemplarily reflects current migration situations in a global-digital world, in which migration must be viewed as a circular, complex process that affects the home and host countries equally and sets them in permanent (communicative) connection with each other (cf. Canagarajah 2017, 4).

Of course, migrant situations have far-reaching effects on speakers, which can also be seen in terms of their language biographies (cf. Busch 2017) and their language repertoires (Heller 2010), in that languages and linguistic varieties can be expanded and/or lost (Kluge 2005). In recent years, numerous interesting works have emerged in the Hispanic studies context that focus on the complex interrelationships between individuals, societies of origin, and host societies (e.g., Márquez Reiter/Martín Rojo 2015; Gugenberger 2018). It seems particularly worthwhile to investigate which processes take place in the negotiation of linguistic identities in the process (cf. Patzelt 2016, 42). Linguistic variational elements may also play a role here. This paper follows up on these ideas by examining how Andalusian speakers in diaspora situations draw on phonographic elements in chat communication with other speakers of Andalusian origin. The focus of this paper is thus the analysis of concretely used phonographic elements, which allow for inferences about phonetic-phonological salient features of Andalusian varieties and counter the idea of a leveling of linguistic variation in computer mediated communication (Budach/de Saint Georges 2017). Linked to this is the question of for what communicative purpose these variational features are used exemplarily in a self-generated corpus of digital WhatsApp communication. This pursues a central theme in the studies of digital humanities (Page 2015, 403), namely a possible identity construction in digital communication. Finally and above all, this paper attempts to do justice to the diatopic heterogeneity of Andalusian Spanish by including two different closed networks in the corpus: Using two corpora of Spanish migrants from the Andalusian speech communities of Cordoba and Jerez de la Frontera, convergences and divergences in the use of dialectal phonetic features in graphic realization between Western and Eastern Andalusian speakers will be identified. This is an attempt to contribute to recent research on linguistic identities of migrant speakers in the age of Digital Humanities (Hinrichs/White-Sustaíta 2011) by investigating the internal multilingualism of Andalusian speakers in computer mediated communication (Georgakopoulou/Spilioti 2015, 1).

### 2 Theoretical Backgrounds

#### 2.1 Migration Linguistics and Spanish Varieties

The present contribution can thus be understood in the broadest sense as a contribution to migration linguistics. Since the relevant approaches in Romance studies by Stehl (1990) and Krefeld (2004), migration linguistics of a Romance character has been established, which in recent years, under the influence of the ethno- and sociolinguistic work of Blommaert (2010), has also been further spun in Hispanic studies (e.g., Kluge 2005; Zimmermann/Morgenthaler García 2007; Patzelt 2016; Moreno Fernández 2013; Márquez Reiter/Martín Rojo 2015; Gugenberger 2018; Morgenthaler García/Amorós 2020). With regard to the Andalusian communicative space, however, there are only a few works that focus mainly on the communicative space of Andalusia as a destination region for Maghrebi migrants (e.g., Harjus/ Harjus 2019; Ready 2021). In contrast, speakers of Andalusian origin as a concrete group in a migrant situation have not been studied in migration linguistics works so far. Here, Hispanic studies tend to work with national rather than regional references to origin (e.g., Escobar 2007; Palacios Alcaine 2007; Schrader-Kniffki 2016) or refer to minority languages within Spain (e.g., Prifti 2014).

I define migration here according to Oltmer as spatial movements of people that have far-reaching consequences for the life courses of the migrants and from which social change results. In this context, migration is associated with a longerterm stay elsewhere and is conceived as a shift in the center of life of individuals, families, or collectives (2017, 21). I have already pointed out that migration processes can lead to linguistic changes in the language repertoire. More importantly for this paper, however, modification of linguistic identities can also take place, adapting to new sociocultural contexts, i.e., to the sociocultural realities in migrant situations. These changes in linguistic identities are not limited to languages, but also have implications for variational aspects, as group affiliations cannot be exclusively through national language categorizations (Gugenberger 2018; Schrader-Kniffki 2016), but also through variational markers. The so-called third wave of sociolinguistics (Eckert 2012) addresses this by paying increased attention to the indexical and thus identity-creating functions of linguistic varieties and variational features in this recent manifestation of sociolinguistic approaches.

#### 2.2 Third-Wave-Sociolinguistics: Style, and Speaker Design

Linguistic styles have usually only been peripherally considered since the beginning of linguistic research on sociolinguistic issues. Traditional approaches in English studies (e.g. Labov 1972), which distinguish between casual and careful styles, have rarely been taken up in Hispanic studies (cf. Medina-Rivera 2011, 43). Bell (1984) is the first to include listeners in his model on linguistic styles, recognizing linguistic accommodation in speakers according to interlocutors. In other words: Unlike Preston (1991), Bell (1984) does not define the linguistic dimension as more important than the social one. The situational-social dimension is superior to the purely linguistic one in Bell's audience design model. More recent sociolinguistic approaches (Coupland 2007) connect to these ideas by viewing styles as ways of linguistic realizations that can create meaning themselves in particular social situations. This social constructivist approach is further reinforced by Eckert (2004), Eckert/Rickford (2001), and Schiffrin (1996) in what they call third-wave-of-sociolinguistics (Eckert 2012) by defining style as a social practice. In this context, the practice of using certain linguistic features not only leads to the performance and construction of social interactions, but also to the creation of speaker-related social identities (Eckert 2004, 44). The relationship between language and society is thus thought to be reciprocal, as both can influence each other (cf. Harjus in press, 32). Thus, language use can be seen not only as communicative, but precisely also emblematic, i.e., linguistic features exhibit indexical functions (cf. Agha 2005) with respect to the intended positioning of a speaker and/or a particular group of speakers (cf. Soukop 2016, 153). Following the work of Eckert (2004) and Schilling-Estes (2002), we speak of speaker design in this context, drawing on linguistic anthropological approaches (e.g. Duranti 2004): "[Speaker Design] views stylistic variation as a resource in the active creation, presentation and even recreation of speaker-individual and interpersonal identity – in other words, stylistic variation is viewed as a resource for creating as well as projecting one's persona" (Cutillas-Espinosa/Hernández-Campoy/Schilling-Estes 2010, 32). These approaches to exploring linguistic styles can, of course, also take place in computer mediated communication (cf. Nishimura 2015, 104). Consequently, the use of certain linguistic features of Andalusian varieties in migrant contexts cannot only be seen as an expression of linguistic accommodation with interlocutors, but constructs a linguistic identity of speakers in chat communications. Thus, in this work, a virtual design of Andalusian speakers in the European diaspora is investigated, exploring the third-wave-of-sociolinguistics still rarely applied to the Andalusian communicative space in digital communication (cf. Harjus/Peter in press).

#### 2.3 Andalusian Spanish Varieties in Digital Communication

Romance studies at the intersection of migration, language, and identity in digital space focus primarily on discourse-linguistic approaches, often following Androtsopoulos' (2008) net-based ethnographic analysis. This primarily examines forms of communication in the media-produced public sphere, but leaves out privateinterpersonal communication. Supranational identity formations in Latin American migrant groups on the basis of internet blogs in Canada, for example, are the focus of various Romance migration linguistic works in the digital context by Kluge (e.g., 2015) or Frank-Job/Kluge (2012). In the context of Andalusian varieties, no work has yet emerged that focuses on digital aspects in migration situations. Research available so far focuses on Andalusian varieties with regard to certain metalinguistic discourses (e.g., Peter 2020) or computer mediated communication with regard to certain phonographic aspects of Spanish varieties in Andalusia, which – in the sense of research on variation in digital discourses (cf. Hinrichs 2015, 22) – should allow conclusions to be drawn about phonetic-phonological aspects of Andalusian varieties, without migrant perspectives. Thus, in some works (e.g., 2013; 2015), Chariatte analyzes linguistic styles on the basis of certain phonographic features occurring in the virtual Andalusian speech community of Malaga on Facebook, such as the Seseo-Ceceo, Yeísmo, or the elision of syllable-final /s/. Chariatte (2015, 18) highlights that social styles are created and linguistic identities are established through the use of these features in relatively public digital forums. In Harjus (2020), I used a closed WhatsApp chat corpus with speakers from the Western Andalusian city of Jerez de la Frontera to show that the use of the variational feature Jejeo in private-interpersonal communication creates a linguistic style and discursively constructs a virtual linguistic identity that does not match the use of the more stigmatized phonetic feature in oral corpora from the region.

This paper partially follows the work of Chariatte (2015) and Harjus (2020) by analyzing computer mediated communication in a closed social network of Andalusian speakers, i.e., private-interpersonal digital communication. However, beyond that, the speakers do not reside in Spain, but are in a migrant situation in other European countries.

By computer mediated communication I refer to spoken conceptualized but graphically realized communication, which as a hybrid entity can also be defined as "oraliterality" in German studies (cf. Marx/Weidacher 2014, 107) or as faked orality in Hispanic studies (cf. López Serena 2007). Following Koch/Oesterreicher's (1985) reflections on conceptual orality, it can be stated that in New Media and especially in chat communication, certain forms of conceptual orality enter into written realized communication, as has also been shown in various Hispanic studies (e.g., Giammatteo/Gubitosi/Parini 2017; de Benito Moreno/Estrada Arráez 2018; Harjus 2020). Speech patterns primarily known from orality increasingly appear in written form (cf. Kailuweit 2009, 2). Particularly in chat conversations of Web 2.0, to which WhatsApp can be counted, the factors of synchronicity and dialogicity also play a prominent role (cf. Dürscheid 2003), which have an amplified effect on oral elements, such as phonetic-phonological elements of Andalusian varieties of Spanish, finding their way into written form (Hinrichs 2015, 26).

### 3 Methodological Aspects

### 3.1 Digital Data

In order to analyze the use of linguistic variables of Andalusian varieties of Spanish in computer mediated communication I make use of a self-made corpus. All participants are acquaintances of mine and have given their consent to the use of these data. Finally, the corpus used here is to be divided into two smaller corpora, since the speakers of the first group to be analyzed grew up in the city of Jerez de la Frontera (province of Cadiz, Autonomous Community of Andalusia) and a second group consists of speakers who grew up in the city of Cordoba (province of Cordoba, Autonomous Community of Andalusia). These are two groups, and I define each as a community of practice because the participants engage in shared social practices, e.g., being interested in soccer. The earlier offline communities of practice are continued online in the WhatsApp group and can be considered online communities of practice (Angouri 2015). With Nicolini, I define the sociolinguistic term communities of practice as: "communities of practitioners constantly busy positioning themselves within the ongoing practice' and further 'it is practice which performs community and not the other way around'" (2012, 94). Both groups – including myself as the fifth participant – consist of four academically educated, 33- to 38-year-old, male speakers each, who were in a migrant situation within Europe between February 2019 and February 2021 during the virtual conversation trajectories under study (see Tables 1 and 2). Consequently, this is an investigation of linguistic aspects in a skilled migration, i.e., on the basis of "international movement of workers possessing specific skills achieved through their higher education" (Lising 2017, 298).


Table 1: Group 1, Origen: Jerez de la Frontera (Province of Cádiz, Andalusia, Spain).


Table 2: Group 2, Origen Cordoba (Province of Cordoba, Andalusia, Spain).

Thus, both groups are equally distributed in terms of the number of speakers, the sociological-demographic profile of the informants, and the time period studied. The migration destinations of the speakers are all within the European Union (Austria, Germany, Belgium, Netherlands) or the United Kingdom. With the exception of informant C1, who works in rural Austria in the automobile supply industry, all group members live in major European cities and/or capitals outside of Spain during the time period studied. The types (123,060 in Group 1; 115,034 in Group 2) and tokens (16,245 in Group 1; 17,421 in Group 2) used in the respective corpora are also similarly distributed, so that a contrastive analysis of virtually written but orally conceptualized communication has been valid and reliably feasible.

#### 3.2 Western and Eastern Andalusian Spanish: the Linguistic Variables

The Spanish varieties in the Autonomous Community of Andalusia belong to the southern dialects of European Spanish. They are among best-analyzed varieties of Spanish (cf. García-Amaya/Harjus/Henriksen 2019, 7). At the same time, as the pluralistic use of the term linguistic variety already suggests, the Andalusian varieties do not show any internal homogeneity. Rather, since the earliest works (Alvar/Llorente/ Salvador 1961–1973), a diatopic separation between an eastern and a western speech community has been pointed out, which is also confirmed by recent sociolinguistic works (Villena Ponsoda 2008) due to the convergent situation of the Eastern dialectal area to the standard of European Spanish and the Western Andalusian variety oriented to the regional center of Seville. The (source) speech communities of Cordoba and Jerez de la Frontera, which are the focus of the analysis here, are diatopically located in the eastern and western dialectal spaces of Andalusia, respectively, with the consequence that the generalized linguistic features are partially divergent, although, of course, mainly common linguistic varieties are present due to the canopy of the dialects by the standard of European Spanish. Recent sociolinguistic-phonetic work on the eastern communicative space of Andalusia has focused mainly on the speech communities around Granada (e.g., Melguizo 2007) or Málaga (e.g., Vida Castro 2012). The speech community of Cordoba has received little attention (Harjus 2021), and the phonetic-phonological features known about the urban varieties are still based on the research of Uruburu Bidaurrázaga (1990). The following eight phonetic features are highlighted for academically-educated speakers from the urban Cordoban speech community (see Table 3):

Table 3: Linguistic Variables of Cordoban Spanish (adapted from Uruburu Bidaurrázaga 1990).


The speech community of Jerez de la Frontera received little attention for a long time, but has come under increasing scrutiny in sociophonetic research since the work of Carbonero Cano et al. (1992), mainly due to its geographical location in the catchment area of the regional normative capital Seville and its dialectal position within the traditional Ceceo area. In the 21st century, it has increasingly become the focus of sociophonetic research (e.g., García-Amaya 2008; Harjus 2018; Henriksen/García-Amaya 2019; Harjus 2021). The following eight phonetic features are highlighted for academicallyeducated speakers from the Jerez de la Frontera speech community (see Table 4):

Table 4: Linguistic Variables of Jerezano Spanish (adapted from Harjus 2018).


The speakers of the two groups in focus here also realize the majority of the specific pronunciation variants of the majority of urban, academic, and middle-aged speakers from Jerez de la Frontera and Cordoba (cf. Harjus 2018; Uruburu Bidaurrázaga 1990). This account is important at this point because an analysis of virtual communicative behavior without a connection to offline language use is difficult or not meaningful: "social media and other online environments are not seen as separate contexts, detached from other spheres of life; and digital communication practices are seen in the wider sociolinguistic context" (Varis 2015, 60).

#### 3.3 Methods

I extracted the two chat group histories from 2019 to 2021 as pdf documents and prepared them for a text-based, inductive data exploration. By feeding the pdf documents into the software-supported qualitative data tool MAXQDA, it was possible to code phonographic features in the linguistic-virtual communications and to add further linguistic features to the inductively obtained categorizations of phonographic features (codes) in the further course of the qualitative data analysis.

Of course, it is not exactly easy – sometimes even problematic – to establish a link between graphic realization and a phonetic phenomenon (cf. De Benito Moreno/Estrada 2018, 78). In the following example, besides the use of various additional syllables in /uɾbanisima/ <urbanísima> 'urbanisisisima' (urbanistic), a new literacy <kiero> /kjeɾo/ quiero 'I want' and a rhotacism <er> /el/ el 'the', one can see, that by two different speakers within the studied second group the letter <j> is used instead of an <s> for <jegundo> /segundo/ segundo 'second'. The use of <j> instead of <s> can be interpreted as a phonographic realization of the so-called Jejeo, i.e., the aspiration of sibilants in syllable-initial position, which is diastratically marked in the Jerezano speech community (cf. Harjus 2018, 185):

J4: Yo jegundo J2: Olé J1: er jegundo te vas a hartar de canis, eh!! J1: jejeje J4: Leyenda urbaniisisisisima J2: Yo estaré dos semanas y kiero intentar los dos

First, the graphemes <j> and <s> are far apart on a cell phone keyboard, and second, a spelling program should automatically change the letter combination <jegundo>, which makes no orthographically intended sense in Spanish, into the orthographically correct <segundo>. Thus, it can almost be ruled out that there was no intended phonographic representation of Jejeo on the part of the speakers from corpus group 2 here.

### 4 Phonographic Uses of Andalusian Variables in Mobility Contexts

In the following, I dedicate myself to the phonographic analysis of the corpus. It is to be expected that due to the more pronounced variational elements in the western Andalusian speech community of Jerez de la Frontera, phonographic features can be found quantitatively more frequently in group 1. Nevertheless, variational uses of Andalusian features in certain communicative contexts can be expected in both groups. These communicative contexts will be determined in more detail in qualitative analyses.

#### 4.1 Group 1: Jerezano Origen

The results of the quantitative analysis of the use of (western) Andalusian variants for Group 1 are given in Table 5. The quantitative analysis shows a high use of phonetic variables that reflect linguistic features of western Andalusian Spanish. Particularly outstanding is the reduction of intervocalic /d/ with 83.0%, a phoneme that in intervocalic position is not realized in large parts of Spanish-speaking Europe and Americas, at least in near-linguistic oral communication. Much less frequently, but still in numerous contexts, syllable- and word-final <s> are reduced. The omission of syllable- and word-final /s/ is widespread in the western Andalusian speech community – as in other meridional varieties of Spanish in Europe and coastal regions of the Americas – and is omitted in 9.7% of all graphic realizations in the corpus. While so-called rhotacism, i.e., neutralization of /ɾ/ and /l/, occurs rarely in the corpus (4.6%), phonographic representations of the aspirations


Table 5: Variable Scores (Group 1; Jerezano Origen).

of /x/ in the corpus reach similar numbers (13.8%) as /s/ reductions. Particularly striking are phonographic representations of fricatives of Spanish affricate /t͡ʃ/ (23.5%) and neutralization of sibilants /s/ and /θ/ in favor of a <z> and <j>, respectively, i.e., in favor of a phonographic Ceceo and Jejeo (32.7%). Both phoneticphonological features (lenition and Ceceo) are salient features of the local varieties of Jerez de la Frontera, the community of origin of the speakers from group 1, and are phonographically represented by the grapheme sequence <sh> and <z>, respectively. Thus, overall, with respect to the six different phonetic-phonological features highlighted here alone, over one third of all written realizations in the corpus are realized with variational elements in group 1 (36.0%).

Thus, the majority of graphemes used are in standard European Spanish. At the same time, the high number of phonographic representations of western Andalusian features (36.0%) can be justified by the fact that variational elements are used in the chat progressions in all possible contexts and by all speakers involved. The qualitative analysis shows, for example, in (1) and (2) that phonographic elements are used even in banal appointment situations.


In both statements, there are phonographic realizations of rhotacism – <arguien> alguien 'someone' (1) and <carvo> calvos 'bald heads' – as well as an aspiration of the phoneme /x/ in <here> Jerez (1) and <abaho> abajo 'below' (2); the latter simultaneously refers deictically to a view from northern Europe to Andalusia. The graphemes in <Shiclana> Chiclana (2) also suggest lenition of the Spanish affricate. Just as in <here> Jerez or <vemo> vemos 'we see' (2), reductions of syllable-final consonants occur frequently throughout the chat process. So is the frequent use of Ceceo and Jejeo. While Ceceo refers to a traditional pronunciation variant of the original Jerezano speech community, the frequent phonographic realization of <j> instead of <s> or <z> is surprising.


In both (3) and (4), it is noticeable, on the one hand, that the use of Jejeo is neither used consistently, using partly <jabado> and partly a standardized <sábado> sábado 'Saturday', nor is a single grapheme used for realization, since both the grapheme <j> and <h> are used, as in <tieho> tieso 'broke'. On the other hand, the phonographic features do not only occur in humorous contexts (4), but also find application in basic everyday situations, such as an appointment for a weekend meeting in Munich between C1 and C2 (3) or in questions about personal preferences at soccer matches (5) – here a phonographic representation of Ceceo. However, in my opinion, the continuous use of Jejeo in the corpus is a clear hyperdialectalization (Cutillas-Espinosa/Hernández-Campoy/Schilling-Estes 2010, 35), since this phonetic feature is neither used orally by the group participants analyzed here, nor is it particularly strongly anchored in urban, western Andalusian speech communities (Harjus 2020).

#### 4.2 Group 2: Cordoban Origen

The results of the quantitative analysis of the use of (eastern) Andalusian variants for Group 2 are given in Table 6. The quantitative analysis shows a very rare use of phonographic variables reflecting linguistic features of eastern Andalusian Spanish. The highest numbers here reach phonographic representations < > of the dropout of syllable-final <s>, which could reflect an elision of syllable-final /s/ (1.4%). The rare non-realizations of word-final <s> also rarely result in vowel lengthening in the final syllable (1.0%). Roughly a similar frequency is found in the corpus for the graphic representation of Seseo by <s> or <ss> instead of <z> (0.8%) or the failure of


Table 6: Variable Scores (Group 2; Cordoban Origen).

intervocalic <d> (0.8%). Thus, of the four phonetic-phonological features studied here, only a few are realized phonographically in the corpus overall (1.0%).

In qualitative analysis, it is striking that in group 2 only informants C3 and C4 tend to incorporate variational elements of the Spanish spoken in Cordoba into WhatsApp communication in certain contexts. The communicative contexts here are mostly humorous to sarcastic-ironic. In the following examples of a phonographic realization of the Seseo, this aspect becomes clear:


Both examples are mainly aligned with the normative orthography of standard European Spanish, but contain minor phonographic elements reflecting the (eastern Andalusian) phonetic feature of Seseo with <cabesa> cabeza 'head' (1) and <plassa> plaza 'square' (2). While (1) is an intertextual allusion to two fictional, stereotypical Andalusian YouTube characters, (2) refers to an older audience in the context of a restaurant tip. Consequently, both statements are to be seen humorously and reinforce this context by using (eastern) Andalusian features.

What is exciting in group 2 is that the phonographic use of variational features is by no means limited to eastern Andalusian forms. Rather, linguistic variants are realized graphically that are not used in the Cordoban speech community.


In (3), the speaker uses <pisha> picha 'penis', a familiar form of address from the region around Cadiz and Jerez de la Frontera in western Andalusia, which is not rooted in the variety of the Cordoban speech community. The phonographic representation in the otherwise standard-language utterance also refers to the phonetic variant of lenition of the Spanish affricate, which is not used in eastern Andalusia and here provides a metalinguistic reference to the pronunciation modality of Jerez de la Frontera in the context of the mention of the soccer club Xerez Deportivo. Equally humorous is the statement in (4), which is a pun between the Ajax Amsterdam and Ayamonte soccer teams, since there is homophony between <ajax> and <aya> in an Andalusian pronunciation without the use of the word-final consonant. What is interesting about this utterance is that it again resorts to phonographic aspects that are not part of the Cordoban variety, but rather used in western Andalusia: The phonographic representations of Ceceo in <ze> sé 'I know' and of rhotacism <huerva> Huelva underline the overall humorous statement.

Consequently, the absolute majority of all contributions in the corpus of group 2 are performed in standard variants. The hardly occurring (Eastern) Andalusian features are additionally used only in humorous contexts.

(5) que sois los cartinetis del grupo junto conmigo . . . Seguro que hay un 3 por 2 en Turquia y nos pegamos una en plan rezacon en las vegas. Y digo rezacon porque soy andaluz, y punto (C4)

Rare meta-linguistic comments do demonstrate a variety awareness of individual speakers. In (5), the addition <porque soy andaluz> does construct a (linguistic) identity with Andalusian Spanish. However, the linguistic variant used <rezacon> resacón 'hangover' is again not a Cordoban feature, but refers to the Ceceo of some speech communities in western Andalusia.

# 5 Convergences and Differences in the Virtual Speaker Design

The present paper was able to show that Andalusian-born, academically educated speakers in a European migration situation resort to variational features of Andalusian Spanish in computer mediated communication and thus graphically realize non-standard forms in certain communicative contexts and construct a linguistic identity by means of Andalusian Spanish. However, it has also been shown that the Cordoban Group uses phonographic features of the eastern Andalusian variety of Spanish (e.g., Seseo, lengthening of vocals) with much less frequency than the Jerezano Group uses western Andalusian features (Ceceo, lenition). While in the western Andalusian group hyperdialectalizations (e.g. Jejeo) also take place in various communicative contexts, the Cordoban speakers use eastern Andalusian features only in humorous situations and – in some cases – even resort to alland/or western Andalusian phenomena. Thus, even in computer mediated communication in diaspora situations, a division of the communicative space of Andalusia comes into play (cf. Villena Ponsada 2008), which is reflected in a stronger convergence of eastern Andalusian varieties to the standard of European Spanish and a strongly pronounced regional norm in western Andalusia. It would be exciting for future diachronically oriented studies to see, in the context of migration linguistics, variety linguistics, and digital humanities, how linguistic expression changes in computer mediated communication before, during, and after the circular migration of Andalusians.

### References

	- The Routledge Handbook of Migration and Language, London, Routledge, 2017, 63–78.
	- Phonology, in: Díaz Campos, Manuel (ed.), The Handbook of Hispanic Sociolinguistics, London, Blackwell, 2011, 36–53.

Section 3: Digital Discourses and Glotopolitical Perspectives

### Alba Polo-Artal Gender and Racial Orders in Vox's Nativist Discourses

Resumen: La mayor presencia de la derecha radical en España con la formación política Vox hace necesario el estudio de su discurso debido a la capacidad performativa del lenguaje y a su papel en la configuración de marcos excluyentes. La presente investigación se centra en el papel que juega el orden de género y de raza en el discurso del partido político. Se pretende mostrar el papel que juega la racialización del sexismo, empleada como justificación para los discursos securitarios y antiinmigratorios. Para ello, se realiza un análisis de enfoque interdisciplinar -el análisis crítico del discurso y el análisis cualitativo desde la sociología de género- centrado en los tuits publicados en la cuenta oficial de Vox durante el periodo de 2020–2021, así como en las intervenciones parlamentarias y en los vídeos a los que remiten en dichos tuits. Algunos de los resultados de esta investigación muestran cómo el dispositivo de masculinidad y de feminidad está activo en la construcción polarizada del endogrupo y del exogrupo, atravesados por el orden racial que es empleado para justificar el nativismo, el antifeminismo y una defensa de la igualdad específica del partido.

### 1 Introduction

The growing political relevance of the radical right-wing parties in Europe, as well as their demarginalisation, makes it necessary to study their discourse due to the performative capacity of language and its role in the configuration of exclusionary frameworks. A significant number of academic research in the fields of Sociology and Political Science has been focused on the terminological difficulties in naming these parties (Acha 2021; Arzheimer 2019; De Cleen 2019; Rydgren 2013), as well as in characterising the ideological foundations they construct (Mudde 2021). Regarding the Spanish case, Vox's emergence has stimulated the academic research, analysing the explanatory factors of the end of Spanish exceptionalism (Ortiz/Ruiz/González 2020), the determination of its ideological nature (Ferreira 2019), the analysis of the hegemony of the nationalist component against

Note: This translation was entirely made by Sonia Artal Bueno.

Alba Polo-Artal, Universidad Nacional de Educación a Distancia, UNED

the populist one, configured as a contingent expression (Ortiz/Ramos-González 2021), and the international support and funding networks (Bichay 2020; Tamayo 2020). Concerning the Spanish party's political discourse, it has been analysed either from the critical discourse analysis or the content analysis in relation to migrants and unaccompanied minors (Cheddadi 2020; Camargo Fernández 2021) and refugees from Afghanistan (Sosinski/Sánchez 2022). However, there is a lack of interdisciplinary studies that make use of an intersectional methodology that takes the discursive role of race and gender into account, bearing sexuality and reproduction in mind. These components are relevant foundations, since "gender has always been racialised and race has always been gendered, and both of them have been sexualised" (Dietze 2020, 34; transl. Artal), as it is stated in several research projects.

The concept of ethnosexism (Dietze 2020) is used to address the role of this intersection in the discourses of the Spanish radical right. The European radical right parties are carrying out a process of racialisation of sexual politics (Farris 2021), that is, the attempt to promote racist and xenophobic public policies by appealing to a gender equality that actually entails patriarchal protection. This discursive path is carried out with the triangulation of non-Western migrant men and native women and men, assuming a number of frameworks that (re)produce the gender and racial orders. These must be interpreted within a nationalist foundation, since, as McClintock (1991) affirms, the construction of nations has implied the institutionalisation of gender difference, specially through the natural nuclear family metaphor that is associated with the national territory. Farris' perspective (2021) is adopted to explain this triangulation, treating populism as a contingent element integrated into the theory of nationalism that is not only organised from the form but also from the content.

To the ethnosexist triangulation, in the case of the outgroup – non-Western men – heterosexuality is assumed, since they are linked to sexual violence against women, a discourse that is reinforced through the hypermasculinisation of their bodies, attributing sexist and homophobic attitudes to them. This way, a process of culturalisation of gender, sexuality, race and religion is carried out within the sexual democracy (Fassin 2012) constructed through Westernity. The threatening otherness framework implies the constitution of a protective in-group that is materialised into native men who must comply with the patriarchal masculinity mechanism, assuming their whiteness and heterosexuality. Native women are represented through the femininity mechanism, assuming their whiteness, heterosexuality and social role of vulnerable victims. This triangulation serves as a justifying framework for migration securitisation policies and the implementation of a biopolitics of individuals. Furthermore, it makes it possible to break the social and institutional agreement on gender (in)equality and gender-based violence.1 The Spanish radical right denies this specific type of violence, but, when they do accept it, it is done in order to deny the toxic masculinity factor in favour of discourses that entail a racial component, meaning that the problem of gender-based violence occurs by the presence of a threatening and culturally distant out-group. In regard to Vox, this framework moves towards racialised subjects. In order to understand the racialisation of sexism, that is, the implementation of the otherness framework as inherently sexist, the present research has to make it related to the sexualisation of racism (Mann/Selva 1979), i. e., the repetition of the link between the male otherness as a (sexual) aggressor and the female otherness as a (sexual) victim-object.

This racialisation of sexism is analysed according to Vox's discursive construction of the ethnonationalist framework in relation to migrant otherness and, specifically, to non-Western men, analysing the tweets published on the party's official Twitter account during the time period of 2020 and 2021 and its Agenda España political programme, as well as the parliamentary speeches to which they refer in said tweets. Linking these tweets with other antifeminist ones is necessary so as to understand how this racialisation is combined with an assumed defence of equality that can only be understood within a nativist and racist framework. This analysis is completed with the examination of the representation of masculinity, since, although this party does not openly reflect on it, the discursive signs are part of the number of social practices in which Vox deploys the meaning of masculinity. As will be shown throughout this research, confrontation in Vox's tweets is a basic pillar related to an "us-versus-them" mentality that corresponds to both the man/ woman binomial and the native/migrant one linked to the Western/non-Western binary, as happens with other radical right-wing parties' discourses. These binomials allow progression, as occurs in the party's discourse in relation to Latin American migration, a stratification based on a business expansion mentality and material issues in terms of specialised labour in the caregiving sector, as well as on identity aspects linked to Hispanism.

The social agreement is reflected in the Ministry of Health, Social Services and Equality's Percepción social de la violencia de género report (2014), which shows the perception of the Spanish population regarding gender issues: most of them state that great gender inequalities exist (72% of women and 49% of men), and they assert the existence of gender-based violence (35% of them considers it to be very widespread and 54% considers it to be quite widespread).

### 2 State of the Art

In the field of Sociology, special attention has been paid to the factors present in the rise and consolidation of these parties, usually including aspects linked to supply and demand, such as deterioration and institutional distrust, the crisis of the parties and trade unions (Forti 2021), the concern for the economy and cultural reaction, the role of the leaders and the party organisation itself (Mudde 2021), the role of media (Ellinas 2010), the treatment of terrorist attacks and migratory processes, as well as the identity crisis caused by questioning patriarchal masculinity. In relation to this, the questioning of gender roles has been experienced by some men as an attack to not only their personal identity but also to their national identity, so that the economic and social unrest would be added to the identity and cultural disturbance that the radical right has been able to conduct through discourses that reinforce the patriarchal and racist ideology, perpetuating social inequalities. This process has been studied within the concept of aggrieved entitlement (Kimmel 2017), whose ideology is found in the Spanish radical right.2

Regarding the terminology issue, several authors such as Ramírez (2019) talk about the rise of neoliberal neofascism due to the exacerbation of the features of fascist regimes. Traverso (2016) suggests that neoliberalism is taking postfascist forms. Similarly, Dardot and Laval (2018) consider that we are facing an exacerbation of the authoritarian tendency of neoliberalism. Cas Mudde (2007) identified other nomenclatures in terms of far-right, distinguishing between extreme right and radical right. This author thinks that radical right parties are based on nativism and authoritarianism, with populism being a complementary feature. We will be focusing on the first concept in this research, understood as "an ideology, which holds that states should be inhabited exclusively by members of the native group ('the nation') and that nonnative elements (persons and ideas) are fundamentally threatening to the nation-state" (Mudde 2007, 19). In Europe, the thematic concepts that make up its ideology are (anti)immigration, (in)security, corruption and foreign policy – the borders and reconfiguration of the European Union -. Moreover, Rydgren (2017) considers that the radical right emphasises ethnonationalism supported by myths of the past and the essentialist vision of the nation. Thefore, it produces a homogenisation by which only some population groups would be part of its nation model. In the case of Vox, migrants would be left out, specially non-Western people linked to the Muslim religion, as well as the so-called globalist elites, Basque and Catalan nationalist citizens, feminists and male allies, as well as all those organisations that support said groups. On the other hand, in addition to paying attention to the nomenclature and

Polo-Artal, Alba. Preliminary conclusions of an ongoing research.

characterisation of the ideological principles by which the extreme right is regulated, several researchers such as Ramas (2019) have established a classification for these political parties: they are based on identity politics and authoritarian neoliberalism.

Concerning the Spanish party, a significant number of academic research has aimed at determining its ideological nature (Ferreira 2019), the relationship it has with other European parties (Delle Done/Jerez 2019) or the factors that are influencing its rise (Ortíz/Ruiz/González 2020). The media treatment received in electoral processes and the link between this political party and other far-right groups on the Internet have also been addressed (Hernández Conde/Fernández García 2019). Regarding political discourse, its fake news production and some of its semantic processings have been analysed (Labrador/Gaupp 2019). Nevertheless, research is leaving out the analysis of the role that gender and race play in the articulation of structural power relations based on both vectors. Therefore, this work aims to contribute to this gap. In order to get it done, paying attention to gender studies and the critical race theory will be necessary. As regards the former one, the imperatives of hegemonic masculinity developed by Gilmore (1994), the basic beliefs of masculinity developed by Bonino (2002) and the aggrieved entitlement (Kimmel 2017) will be addressed. Ahmed's (2015) contributions on the politics of emotion are also taken into account, as Vox's hate speech works to create a common threat linked to the fear of loss and is articulated around the love for their nation.

In the case of the critical race theory, Fassin's (2008; 2012) studies on the political-discursive construction of the sexual clash of civilisations that serves to dictate immigration policies have been taken into account. Moreover his concept of sexual democracy with which he explains how right-wing parties whitewash their racist policies under the guise of sexual and even reproductive equality. The sexual democracy framework is constructed within ethnosexistischen Konstellationen (Dietze 2020) that, besides being based on Islamophobia, they also reproduce it. For some authors, it cannot be reduced to a religious question because its social and political role is then omitted, the relevance of racial construct within white privilege (Tyrer 2013) and the fact that it would actually be producing a particular racialisation of subjects who are conceived as Muslims (Carr/Haynes 2015). Regarding Vox, we found a number of tweets that are part of the racialisation process, since they attribute features that are considered to be inherent and linked to cultural and language traits and religious practices. This cultural Islamophobia cannot be separated, in the case of this party, from the Islamophobia that appears to be racial, a hybrid articulation that takes on special relevance with the Great Replacement Theory promoted by said party along with the messages of the intrinsic violence of non-Western migrant men. The researcher Laura Cervi (2020) makes a distinction between banal and ontological Islamophobia, considering that, in the case of Vox, although there are examples of both, there is a preponderance of the ontological one insofar as it is rooted in the Clash of Civilisations Theory, "portraying the entire 'Islamic civilisation' as ontologically incompatible with the core values of the West" (2020, 5), and with the demographic winter phrase with which they refer to the decrease in birth rate in Spain. This ontological Islamophobia is close to Sayyid's (2018) studies, understanding Islamophobia as a form of identity politics that would be the result of the crisis of Europeanness.

### 3 Methodology

This research analyses 4,189 tweets from Vox's official Twitter account (@vox\_es) posted during the 1st January 2020 and 31st December 2021 period of time. Understanding the context of every post is paramount so as to carry out its analysis, hence the fact that the pictures and videos that accompany the tweet are taken into account as they contribute to amplifying the illocutionary force of the messages with which stereotypes and frameworks related to national subjects and those considered non-national subjects are constructed. The study is also supported by the Agenda España political programme.

Regarding the written language of the tweets, the text corpus is examined through the Sketch Engine programme, obtaining the wordlist, keywords and terms – multi-word terms – with the purpose of discovering the thematic focus areas that support the party's ideology and that are present in the discursive construction of the in-group (identity) and the out-group (alterity), as well as in the racialisation of sexism.

The significant codes resulting from this first quantitative approach are brought into dialogue with previous research on radical right, gender and race, obtaining a number of categories: nationalism, nativism, antifeminism and Islamophobic antiimmigration, victimisation, aggrieved entitlement and protection. A subsequent selection of these words allows an in-depth analysis through Thesaurus – it indicates the terms that usually appear with the same collocations -, the Word Sketch tool – collocations -, the Word Sketch Difference tool, with which the syntactic patterns and twoterm or two-lemma collocations can be compared; and the Concordance tool, which allows to examine the linguistic context of a word. Finally, a critical discourse analysis of the most significant tweets is carried out. Therefore, throughout the execution of this research, a corpus systematisation and analysis is carried out with quantitative and qualitative techniques that reveal the construction of a repertoire based on the repetition of the same collocations, transferring the meaning of some words over others (Ahmed 2015), being one of Vox's communication strategies on Twitter (Camargo Fernández 2021).

#### 3.1 Objectives

The objective of this research is to examine the way in which the racialisation of sexism linked to nativism is discursively constructed, taking the role played by the gender and racial orders into account. The relationship between the racialisation of sexism, the representation of migrant men and the justification of antiimmigration positions in the tweets posted on the party's official Twitter account during the time period of 2020 and 2021 is also analysed.

#### 3.2 Research Questions

A number of research questions arising from the corpus examination, as well as from the explained theoretical framework, has been formulated in order to carry out this project. The questions are as follows:


### 3.3 Methodological Principles

This section will describe the phases and methodological aspects that have been used, which are focused on critical discourse analysis:

Phase 1: Determination of the cases under study, specifically, the tweets from Vox's official Twitter account during the 2020 and 2021 time period, as well as the approach of a few objectives and research questions that are modulated after examining the corpus.

Phase 2: Information search and capture process: script programmed in Python language.

Phase 3: Content analysis process with the help of the Sketch Engine corpus management software. Prepositional forms and conjunctions have been avoided to focus on notional words, including determiners, due to the value of the possessive form for this analysis.

Phase 4: Categorisation process: firstly, the most frequent words and keywords are organised around the categories in which the radical right is supported. Secondly, and following the theoretical proposals on ethnosexism and on racialisation of sexism, these categories are triangulated to examine the discursive link that Vox creates for the two unifying frameworks: the ingroup and the out-group.

Phase 5: Critical discourse analysis: a triangulation of different approaches is used for this analysis to reveal the ideologies and representation, addressing aspects of local, sequential and global semantics and focusing on categorisations through collocations and agentivity, specially for the "us/them" polarisation analysis, resorting to the scheme developed by Van Dijk (1996).

### 4 Corpus Analysis

#### 4.1 Results of the Categorisation Process

In this section, the most repeated and relevant terms in Vox's discourse will be examined in order to approach its political framework. Thanks to the Keywords tool of the Sketch Engine programme, the keywords – including the hashtags, since they were not excluded from the initial search – and characteristic terms when compared with a balanced base corpus – Spanish Web 2018 – are obtained. This analysis extracts the party's most relevant issues. Furthermore, the Wordlist tool is used to check the most repeated words. In both cases, an advanced word search was carried out with a minimum frequency of 15, expressed in absolute terms and excluding (@) mentions and grammatical categories that do not provide any semantic value. The following table allows the observation of the polarised representation of the in-group, with terms such as Vox (absolute frequency of 1203), and migrant out-group – represented with terms such as ilegales<sup>3</sup> (176), inmigrantes<sup>4</sup> (88), inmigración<sup>5</sup> (77), invasión<sup>6</sup> (62), menas<sup>7</sup> (55). It also allows us to approach some of the discursive concepts that will be seen in the following section, such as nationalism, nativism, the imperative of patriarchal protection and the aggrieved entitlement, which are linked to the concepts of anti-immigration and antifeminism.

Some of the lexical selections are justified based on the sociocultural and historical context, being linked to the political party affiliation, self-reference to the

Invasion.

Illegal.

Immigrants.

Immigration.

Unaccompanied migrant minors (UMM).


Table 1: Wordlist table and corpus Keywords (own elaboration from Sketchengine results).


Table 1 (continued)

party or to approach some events and their interpretation, as occurs with the nuestros<sup>8</sup> possessive adjective (419) or the negatively connoted terms like amenazas<sup>9</sup> (48) to refer to migrants. The triangulation of the categories on which the radical right leans, specially nationalism, the nativist component and the gender order results in the following scheme with the two unifying frameworks:

The native versus non-native construction and the sex/gender system one are both produced dialectically and interfered by other constitutive vectors of the groups at the same time (see Figure 1). Analysing the rare discursive relevance of migrant women and the party's "women's advocate" self-representation is also necessary in order to understand Vox's arguments for the justification of border closures and the deportation of migrants linked to non-Western men. The party

Our.

Threats.

Figure 1: Frameworks (re)produced in the different parts of the three components of the racialized sexism concept (own elaboration).

does all of this while maintaining antifeminist political practices. Thus, we need to analyse the representation of migrants in relation to women's rights. Vox constructs an otherness of non-Western men and an identity of Spanish men articulated within the protective patriarchal masculinity that is related to the victim construction of women at the same time. Therefore, research questions 2 and 3 are addressed, which will be developed in greater detail in the following section.

#### 4.2 Critical Discourse Analysis Results

Once we have approached the party's political frameworks through the keywords' quantitative approach, this section will address the construction of these frameworks. On the one hand, the nationalist and nativist concepts and the role that patriarchal masculinity plays in the construction of both concepts will be taken into account (4.2.1) and, on the other hand, the process of racialisation of sexism will be revealed (4.2.2). In an effort to understand how the latter is constructed, Vox's representation of migrant men and its six resulting frameworks will be addressed. Afterwards, we will be focusing on the discourses in which the gender and racial orders play a fundamental role, specially when migrants are hypermasculinised through violence and sexual assault. This repetitive representation in the discourse justifies antifeminist policies, while allowing the party to present itself as the protector of women at the same time.

#### 4.2.1 The Nationalist and Nativist Concepts and Patriarchal Masculinity

In this section, the way this political party constructs the nationalist concept that is part of the nativist foundation, as well as the role that patriarchal masculinity plays in said construction, will be examined. The words in Table 1 refer to fundamental concepts on which the radical right is based. The España, país, fronteras, soberanía, pueblo español, nación and patriotas<sup>10</sup> nouns entail nationalism, as happens with the #ProtejamosEspaña, #FronterasSeguras, #ElFuturoDeLosPatriotas, #UnidadNacional and #CeutaSeDefiende<sup>11</sup> hashtags. The combatir, defender, proteger or recuperar<sup>12</sup> verbs are linked to nationalism and the mandates of the patriarchal masculinity mechanism, such as heroic bellicosity (Bonino 2002). It should be noted that the pueblo<sup>13</sup> noun always appears linked to español<sup>14</sup> in the corpus, so that, in case of finding discursive elements of populism, it must be understood within a nationalist component. Although the most used terms to refer to España, after this same noun, are país and pueblo español, the addition of nación and nacional<sup>15</sup> is quantitatively bigger. By using the Thesaurus tool of Sketch Engine, values linked to the nación noun are noted, being related to the identity and political priorities at stake: seguridad, frontera, economía, elección, padre and casa,<sup>16</sup> amongst other terms that link national identity with male identity and the private sphere (see Figure 2):

Figure 2: Thesaurus of "nación" generated by the word sketch tool in Sketch Engine.

National.

In the same order: Spain, country, borders, sovereignty, Spanish people, nation and patriots.

In the same order: #ProtectSpain, #SafeBorders, #TheFutureOfThePatriots, #NationalUnity and #CeutaIsDefended.

In the same order: to battle, to defend, to protect and to regain.

People.

Spanish.

In the same order: safety, border, economy, choice, father and home.

The nation's sociohistorical constructions are interfered by masculinity. The analysis of Vox's tweets allows us to observe this relationship, with tending captures of patriarchal masculinity in which the gender order is intensely reproduced. An example of this is the recurring use of the war metaphor that is combined with moral aspects for the construction of the national imaginary, as happens in the following tweet where the deber<sup>17</sup> nominalised verb, cicatriz<sup>18</sup> noun and valiente and cobarde<sup>19</sup> antonym adjectives refer to that heroic conception of nation and masculinity:

Vox [@vox\_es]. (22/5/202) El deber cumplido. Más vale una cicatriz por valiente, que la piel intacta por cobarde. #VOXNiUnPasoAtrás.

Furthermore, this heroic masculinity that could be analysed under the influence of National Catholic masculinity by bearing the Spanish historical context in mind is also noticed in the rhetorical exaltation linked to the imperial symbolism present during Franco's regime. The patriotic aim appears as a measure in the Agenda España programme, where it is indicated: « Difundir y proteger la identidad nacional y la aportación de España a la civilización y a la historia universal, con especial atención a las gestas y hazañas de nuestros héroes nacionales dentro y fuera de nuestras fronteras » (Agenda España, 8). After analysing the discourse, we can find this heroic masculinity through the mythification and repetition of words related to patria,<sup>20</sup> nación and España, as well as unidad,<sup>21</sup> which is sometimes linked to the concept of family, entailing masculinity and protection from the perspective of duty and heroism. However, this connection is not something new, being extensively analysed in Lakoff's work (2007) where family is addressed as a metaphor for the nation. Patriarchal masculinity materialised in National Catholic masculinity is a man's stereotyped sociocultural construction that does not originally come from Franco's dictatorship. Nevertheless, it was reinforced as the only model during that historical time period. Its identity possibilities were limited to hierarchy, the exaltation of warmongering and heroism – symbolisms of virility -, and the exaltation of mythical discourses that we find in Abascal's party in the constant basic belief of heroic bellicosity, developed by Bonino (2002):

Unity.

As a noun, it means duty; as a verb, it means must.

Scar.

Brave and coward.

Homeland.

Vox [@vox\_es]. (23/4/2020). Queremos recordar a esos españoles que integraban los Tercios y que empuñaban espada y pluma a partes iguales. Cervantes, Quevedo, Calderón, Lope de Vega y muchos más, engrandecieron todo lo que ahora somos.

This is shown through a list of renowned men who personify the model of masculinity to produce sympathy and some identification. The association occurs in the third proposition, which allows a connection with the present time as well as to strategically present these figures as historical ascendants of greatness. The first categorisation is elided, an "absent us" that is concretised with los Tercios. The empuñaban<sup>22</sup> verb indicates a language of war that is reinforced through the espada and pluma<sup>23</sup> objects that refer to the half monk-half soldier figure (Blasco 2014). This tweet shows how right-wing populist rhetoric uses mythopoesis as a legitimation strategy (Wodak 2015), that is, it reconstructs historical discourses that this rhetoric reformulates with a mythical tone to exalt the feeling of national greatness (cf. Ballester 2021). The implementation of a greatness and heroic imaginary makes it possible to compensate the wound of the white man who feels like he is losing his role as a hero towards society and women. In this sense, it should be noted that the characterisation found regarding the nation is carried out through the implementation of the foundational myths linked to the conquest of Al-Andalus and the Spanish Reconquest. Thus, Spanish identity is presented as ontologically incompatible with the values associated with the category of Muslim, promoting Islamophobic imaginaries, as Cervi (2020) points out. The statements of the party's general secretary and member of the Congress of Deputies, Ortega Smith, belong to this matter. These statements, dated on 2nd January 2020, were published on Twitter and shown on a video of El País newspaper. Ortega Smith states:

Aquí en Granada, en el 528 aniversario en el que nuestra historia, los que nos precedieron demostraron que al enemigo más poderoso que tenía España y que tenía Europa, la invasión islamista, fue derrotada. Y decimos que la Reconquista no ha terminado aunque algunos crean que es así, la Reconquista de los valores, la Reconquista de las libertades, la Reconquista de la unidad y la fraternidad y la cooperación entre todos los españoles es una asignatura pendiente. Y la Reconquista también frente a esa invasión del islamismo radical, de las mezquitas salafistas, de quienes quieren imponer sobre Europa una concepción totalitaria, una teocracia, pues esa reconquista continua. (. . .) Viva Granada y viva España. (El País, https://www.youtubeYouTube.com/watch?v=1l88GiTAjoQ&t=2s).

Regarding the nativist concept, although some of the words are linked to social measures, if the tweets in which they appear are analysed, a polarised construction is noticed. This construction condemns the lack of social measures for the in-group

They wielded [a sword or knife].

Sword and quill pen.

(the Spanish people) as opposed to the aids for the out-group, which is made up of migrants and, specifically, non-Western men, with no apparent Spanish nationality, a statement based on racist prejudices. Therefore, an exhaustive analysis allows to understand the use that the party makes of some of the terms obtained in the table: social/es (107), derechos (75), soberanía (85), paro (56), ruina (55), miseria (55) e inseguridad<sup>24</sup> (47); keywords used in hashtags: barriosseguros (scoring 88,315 times), obreroyespañol (69,612), Españaquieretrabajar<sup>25</sup> (57,137); and multi-word terms like escudo social (75,697), cola del hambre (69,41), factura de la luz (55,504), paro juvenil (33,16) and salario digno<sup>26</sup> (30,554), not forgetting about its aggrieved entitlement, #LoNuestroPrimero<sup>27</sup> hashtag (19,711). This victimhood based on comparative hoaxes is noted more clearly in the following tweet:

Vox [@vox\_es]. (13/12/2020). La inmigración ilegal se ha convertido en una auténtica discriminación para los españoles. En pie frente a la invasión, #SóloQuedaVOX Es

This idea of the "victim-us versus threatening-them" polarisation is also present in the following tweet:

Vox [@vox\_es]. (21/5/2021) No se ha hecho absolutamente nada por proteger nuestras fronteras, únicamente se ha llamado a los inmigrantes ilegales a que vengan para ser regularizados y reciban pagas mayores que las de muchos españoles.

Three arguments are condensed here: on the one hand, no se ha hecho nada<sup>28</sup> is used to criticise the government's migration policies; on the other hand, immigration comes across as illegal and problematic; and, finally, the aggrieved entitlement, since these people would receive more financial support than Spaniards according to Vox. This aggrieved entitlement is noticed when the possessive nouns and adjectives are used, specially at the end of the tweet. This condensation at the end of a sentence is a Vox's common practice to reinforce nativist frameworks and securitisation policies. The in-group's ideological construction, frequently represented by an – explicit or omitted – "us", is part of the strategies for creating an imaginary community (Anderson 1991) with a deep authoritarian and ethnonationalist component. In the Spanish party's opinion, the discursive "us" would be made up of all the Spaniards, always referred in the masculine form, with the feeling of men being discriminated victims playing a crucial role (Kimmel 2017). It should be noted that the forms of masculinity

In the same order: social, rights, sovereignty, unemployment, bankruptcy, misery and unsafety. In the same order: safedistricts, Spanishworker and Spainwantstowork.

In the same order: social shielding (special measures to help vulnerable people), queue for food, electricity bill, youth unemployment and living wage.

#WeComeFirst.

Literally, nothing has been done.

attributable to the in-group and out-group are understood as complex structures that are relationally linked, so that the interaction between them is determined by the position and function they occupy in relation to the number of representations. In addition, they link agents with their specific practices and ideologies (Van Dijk 1996). In the present analysed corpus, strategies used for the construction of the in-group in a positive way have been noted, resorting to emphasis, hyperbole, topicalisation, positive categorisation and some degree of displacement in the discursive agentivity depending on the objective of victimisation or heroisation that reinforces the gender order and the masculinity mechanism in both cases. The posted tweets exalt values linked to patriarchal protection and are useful to define which subjects are worth of protection and which are not. According to Gilmore (1994), protection is one of the masculinity's mandates that materialises the obligatory masculine safeguard as a strategy for the reproduction of gender roles and, consequently, of the inequalities that are established as a consequence of the patriarchal order. This prioritisation of needs and protection is noticeable in the following tweet:

Vox [@vox\_es]. (27/5/2020). La izquierda está más preocupada de imponer su lenguaje inclusivo que de defender el futuro de los españoles. Para nosotros, los españoles más necesitados son prioridad nacional.

Two opposite poles are established. They are occupied, on the one hand, by generalisations of the needs that they conceive as effective and real, which reach a relevant aspect by the use of the prioridad nacional<sup>29</sup> collocation, a term that urges and emphasises the order of protection that Vox takes charge of. The first proposition serves to characterise the enemy through the negative valuation of their practices while exalting the cause of the party at the same time. Furthermore, we find a generalisation in the description of the other enemies (la izquierda<sup>30</sup> and the feminists, that would be part of that ideological entity), a negative lexicalisation through the imponer<sup>31</sup> verb, as well as a complex comparison that is based on the concretisation of a praxis presented in a negative way and faced with the generalisation of the expression relative to the defence of the future of the Spaniards, an element that is mobilised for identification. Moreover, we find a hidden otherness, which would be feminism, represented by the lenguaje inclusivo<sup>32</sup> syntagma that is rejected by the use of the third-person possessive adjective. The second proposition produces a reposition, presenting Vox's fight to help citizens in need. If the tweet's outline is

National priority.

Left-wing politics.

To impose.

Gender-inclusive language.

revealed, the opposition of two simple discursive equations is noted: genderinclusive language = imposition vs. people in need = national priority.

By the use of terms such as nation, tweets that show the historical mythification and polarised construction of imaginary communities whose needs are stratified, it has been possible to observe the way patriarchal protection is operating to construct ethnonationalist frameworks that aim to exclude migrant and feminist othernesses. This aspect will be developed in greater detail in the following section.

#### 4.2.2 Racialisation of Sexism

In this section, Vox's frameworks are shown to represent non-Western migrant men and unaccompanied migrant minors, as well as the relevance of certain terms reiterated in its discourse, as occurs with the manada<sup>33</sup> noun. This is linked to sexual assault and serves, on the one hand, to negatively represent migrants and, on the other hand, to accuse institutional feminism of not protecting women, justifying that way the party's antifeminist position. This way, we notice that the ethnosexist triangulation operates even in the case of gender-based violence when it is not considered in terms of gender but race. As will be seen at the end of the section, this racialisation of sexism is also implemented with the Great Replacement Theory and questions about birth rate.

When focusing on the out-group, it is found that Vox discursively represents migrants within the racist and anti-immigration frameworks that have been observed in the discourses of other radical right-wing parties (Fassin/Salcedo 2015; Wodak 2015; Carvalho 2013), as well as in press communication. The party promotes a hierarchical stratification of immigration based on economy and culture in the Inmigración e Identidad Nacional<sup>34</sup> section of its Agenda España programme: « Garantizar una inmigración legal, ordenada y adaptada a las necesidades de nuestro mercado laboral para asegurar el empleo de todos los españoles y de los inmigrantes que, llegando legalmente, aportan su esfuerzo y respetan nuestro modo de vida » (Agenda España, 10). It is also based on language and an assumed common history that has been whitewashed within the Hispanic rooted concept of Iberosfera:<sup>35</sup> « La identidad nacional de España se forjó en su historia conjunta con las naciones de la Iberosfera. Nuestra política de inmigración atenderá de forma

Gang or herd.

This section is made up of two coordinated terms that corroborate the party's nativist component when linked within the same title.

It could be translated as Iberosphere.

Figure 3: Concordance of "ilegal" generated by the word sketch tool in Sketch Engine.

prioritaria a los ciudadanos procedentes de las naciones que comparten idioma e importantes lazos de amistad, historia y cultura con España » (Agenda España, 26).

With regard to the frameworks analysed by Wodak et al. (2008) on racist discourse in their study on the description of migrants, refugees and political asylum seekers in the British press and the thematic concepts obtained by the researcher Camargo Fernández (2021) in relation to Vox, the following associations for non-Western migrant men have been obtained in the present research by using the collocations of the inmigrante noun with other nouns, adjectives, and verbs:

1. Legality/citizenship framework: the inmigrante noun presents 88 recurrences in the corpus, the ilegal adjective appears 176 times and the inmigrante + ilegal collocation is repeated 70 times. The following image shows the way the ilegal term is mainly linked with immigrants by the radical right (see Figure 3).


In the same order: violent, violence.

Dangerous.

In the same order: unsafe districts, thefts.

In the same order: they attack, they slashed, they destroy.

In the same order: they assault, they raped.

In the same order: to destroy, to degrade, to wreck, to transform, to leave or abandon, to africanise, to terrorise, to invade, to regain, to protect.

In the same order: humilitation, safety/unsafety, ghettoisation, defence.

Wave.

In the same order: emergency, uncontrolled.

Massive.

In the same order: invasion, fifth columnists, army, military-age males.

In the same order: little allowance (used scornfully), to subsidise, financial assistant, lazy.

In the same order: jihadists, non-white Europe, Islamic Spain, Taifa Kingdom, sexual assaults, imported rapes, female genital mutilation, assailants, rapists.

In the same order: to transmit, to infect, increase in cases, increase in deaths, to spread, no control, breach.

people who migrate are linked to infection and coronavirus. The nation is the immune system that has to defend itself from the external factors.

In the case of unaccompanied migrant minors,<sup>50</sup> we find the same frameworks constructed through the articulation of symbolic borders that command lifestyles, material issues and legible identities. Their representation is carried out with the same borders of the subjects linked to gender, race and social class, being also interfered by age in this case. In this sense, the subject of unaccompanied minors would escape the limits of the subjects of just being children. These subjects are semioticised and hierarchised, moving away from the hegemonic subject that has the right to have citizenship. However, they are no longer represented as violent adult migrants in some tweets, being discursively presented as children who must return to their families in order to be protected, since this would be the basic natural structure. As Van Dijk pointed out (1996) in relation to the strategies for the out-group's characterisation and the protection of the in-group's image, Vox resorts to the strategy of apparent altruism:

Vox [@vox\_es]. (21/5/2021). No colaborar con el secuestro de menores ni alimentar el efecto llamada, para que se jueguen su vida y se alejen de sus padres, no es infantilismo político sino un gesto de responsabilidad política y de humanidad.

Two images of the inmigrante and mena collocations are included below as a visual synthesis (see Figure 4):

The construction of migrant alterity as something negative can also be observed with the use of the manada noun to link women's rapes with racialised men and district unsafety. Magrebí<sup>51</sup> is the modifier this noun appears with, as the analysis of the collocations shows. When it is accompanied by a nominal syntagma that complements it, the most frequent ones are: mena and extranjero.<sup>52</sup> When analysing the tweets in which said noun appears within a sociolinguistic perspective, we notice the ethnosexist triangulation with which the social imaginary of other men as hypersexual and potential rapists is constructed. This non-Western masculine otherness' construction implies the denial of violence inside the native in-

In Spain, MENA is the acronym used to refer to these children legally and administratively. In the strict legal sense, this refers to a minor who comes from a non-EU State who is either not accompanied by an adult or left without such company when entering the country. There is a wide terminological debate: some authors such as Quiroga, Alonso and Armengol (2005) propose the menores migrantes no acompañados (MMNA) term, whereas menor no acompañado (MNA) is the most widespread concept in Europe.

Maghrebi.

Foreign man.

Figure 4: Concordance of "inmigrante" and "mena" generated by the word sketch tool in Sketch Engine.

group, as well as the strengthening of the stereotype of women as vulnerable victims in need of protection. In this case, it is Vox condemning sexual assaults. This Konstellation, which consolidates the gender order, belongs to the base of the racialisation of sexual politics with which Vox tries to make its securitisation and antiimmigration policies. An example of this can be found in these two tweets:

Vox [@vox\_es]. (17/9/2021). @Santi\_ABASCAL No conocemos ni los nombres ni las caras de ninguno de los miembros de las manadas magrebíes que han estado violando este verano.

Vox [@vox\_es]. (15/8/2021). Nueva violación en manada protagonizada por inmigrantes. Esta vez en Córdoba. Estas son las consecuencias de la inmigración descontrolada que promociona el Gobierno. ¿Pedirán las feministas del Gobierno la expulsión de las manadas de extranjeros?

As seen in this last discourse, the political party appeals to feminists who are part of the government. In other analysed tweets and parliamentary speeches, they accuse them of blaming men for the mere fact of being men, as read in several statements by the party leaders on the dates around 25N, the International Day for the Elimination of Violence against Women. Macarena Olona's statements are produced in this regard. They are shared on Twitter through a YouTube video that corresponds to a recording at the Congress of Deputies:

Vox [@vox\_es]. (24/6/2020). El hombre no viola, viola un violador. El hombre no mata, mata un asesino. El hombre no maltrata, maltrata un maltratador. El hombre no humilla, humilla un cobarde. Como mujer, como española, como madre, como hermana, como política, afirmo (. . .) que la violencia no tiene género. Y lamentamos que el resto de partidos hablen de violencia de género y ataquen a los hombres por ser hombres.

In the same regard, they posted this tweet on 8M, the International Women's Day:

Vox [@vox\_es] (8/3/2021). Las feministas se han atribuido la representatividad de todas las mujeres. Sin embargo, la mayoría de las españolas no se ven representadas en el odio contra el hombre y en los discursos violentos del feminismo. #EsteFeminismoEsViolencia

Furthermore, they accuse them of what they call the ideology of buenismo progre,<sup>53</sup> that is, a double standard with which they would blame Spanish men while not penalising non-Western migrant men's sexist attitudes.

Vox [@vox\_es] (8/3/2021). Mientras la izquierda feminista libera y protege a los criminales, en VOX promovemos la cadena perpetua para los violadores y asesinos de mujeres. #EsteFeminismoEsViolencia

The chain of tweets to which it belongs allows to see the connection between rape and the racial issue. Likewise, the message that is delivered is that feminism would not be protecting women, as it would be surrendering to multiculturalism, prioritising non-native men's protection before women's safety, specially native women's safety. By using this framework, Vox presents itself as the women's advocate and, in addition, it justifies its antifeminist position, since current feminism has become radical, as stated in the following tweet:

Vox [@vox\_es] (8/3/2021). El feminismo radical no entiende ni atiende a las necesidades reales de las mujeres que limpian sus porquerías. Las mujeres de la España real no necesitan un colectivo que les diga lo que son, mujeres fuertes, trabajadoras e independientes. #EsteFeminismoEsViolencia

An ontological effort to fit apparently emancipatory discourses with the traditional role of women that they promote from their proposals for public policies is noticeable, as Farris (2021) has shown. In relation to European radical right parties, specially the Dutch Party for Freedom party (PVV), the French National Rally party (RN) and the Italian Northern League party (LN), she asserts: "while advocating women's emancipation as a core value of the European (Christian) social fabric, which Muslim and non-Western migrants allegedly lack, these parties also promote policies that encourage the maintenance of traditional roles for women" (2021, 63; transl. S. Artal) and the construction of heteropatriarchal family, which is described in essentialist terms. In this regard, Vox proposes the creation of a Ministerio de Familia,<sup>54</sup> a Ley de Protección de la Familia<sup>55</sup> and « estudios que evalúen el impacto de cualquier nueva norma en la vida de las familias » (Agenda

Progressive do-gooderism.

Literally, a Ministry of Family.

Literally, a Family Protection Law.

España, 42). These reports could eventually replace those that assess gender impacts, the same way that the Ley de Violencia Intrafamiliar<sup>56</sup> would replace the Ley Orgánica de Medidas de Protección Integral contra la Violencia de Género.<sup>57</sup> When the Spanish political party talks about equality, it hints at the existence of discrimination against men due to current gender policies. Vox asks to guarantee equality amongst Spaniards in its political programme, where « derogar [. . .] los instrumentos políticos ineficaces y discriminatorios tales como las cuotas, la paridad obligatoria o la preferencia de un sexo sobre otro en la contratación pública » (Agenda España, 5) would be necessary. It also demands the loss of sexual and reproductive rights achieved to date, developed to an extent in which spending on these health services is opposed to others that have historically been a social demand. This way and through two coordinated propositions, the party redefines what is necessary and what is unnecessary within a sole and irreconcilable mentality: « Suprimir de la Sanidad pública las intervenciones quirúrgicas ajenas a la salud (cambio de sexo, aborto, eutanasia, etc.) e incluir en ella los servicios bucodentales » (Agenda España, 18).

As can be seen, Vox uses terms such as chiringuitos de género, ideología de género or delirio de género, as well as the #EsteFeminismoEsViolencia<sup>58</sup> hashtag to refer to gender issues and, specifically, the feminist movement. As already seen, the following tweet shows the denial of the gender component for the interpretation and analysis of violence, since violence has no race or gender according to this political party:

Vox [@vox\_es] (2/12/2020). @Ortega\_Smith "Lo diremos una y mil veces, la violencia no tiene género. Si quieren que aprobemos presupuestos en Madrid, tienen que acabar con todos los chiringuitos ideológicos".

Moreover, Vox states that feminism becomes a movement that promotes unnecessary expenses when institutionalised:

Vox [@vox\_es] (22/12/2021). La mejor política de igualdad es que desaparezca el Ministerio de Igualdad. El feminismo de escaparate de Irene Montero solo sirve para sacar tajada del dinero público. Con VOX, todos los hombres y mujeres estarán realmente protegidos. Grande @Macarena\_Olona

Literally, an Intrafamily Violence Law or Domestic Violence Law.

Literally, the Organic Law on Comprehensive Protection Measures against Gender-Based Violence.

In the same order: gender beach bars, gender ideology, gender delusion and ThisFeminismMeans-Violence.

According to the party, we would be facing an unnecessary and harmful movement. Firstly, by focusing on equality policies in a nation-state in which equality amongst all citizens has already been achieved. Secondly, by neglecting the priority of protecting women from external threats – non-Western migrant men -, the mafias and the policies of the coalition government itself. By denying the existence of inequality and discrediting both the feminist movements and the coalition government, especially the Ministry of Equality, the political party justifies its lack of support for policies that promote equality, as well as the interruption of agreements regarding gender issues, including gender-based violence, as it is a racial problem, not a gender one. This plot twist also allows them to justify their anti-immigration policies that are circumscribed within the Fortress Europe framework. An example of this can be found in the Seguridad y Defensa section of their political programme,<sup>59</sup> where they demand larger punitive damages and link gender-based violence with foreignness: « Redactar un nuevo código penal que proteja, de verdad, a los españoles y que sirva de reparación a las víctimas. Elevar las penas para los delitos importados, como las violaciones grupales » (Agenda España, 24). The racialisation of sexism and sexual politics contributes to characterising non-Western migrant men as dangerous subjects on the one hand, and, on the other hand, it is used as a justification of racism through the fictional protection of women and other social groups, thus invisiblising their hate speech. The party verbally makes explicit the defence of women from the attacks perpetrated by racialised men. These men are the only agents that Vox considers to attack particular social groups. This is noticeable in the following tweet:

Vox [@vox\_es] (7/6/2021) !!Detenido un tunecino de 18 años por abusar sexualmente de una anciana de 80 en Tarragona. Con estos salvajes, expulsión inmediata y que cumpla pena en su país!

In the tweet above, they present a Tunisian man – with or without nationality but, nevertheless, racialised – who sexually assaults a woman, representing all the other men who must be deported. This subject is characterised with the salvajes<sup>60</sup> adjective while the woman is represented from vulnerability by including her age and the anciana<sup>61</sup> noun. This lexical selection contributes to the Barbarian image, since it makes sexual assault not only horrible but also grotesque.

The provision of a political programme is relevant in the message's construction, hence the fact that the specific measures on Inmigración e Identidad Nacional are developed cannot be overlooked. They appear in the 11th section, which comes after this one.

Savage.

Old lady.

Another example that shows the racialisation of sexism is found in the following tweet:

Vox [@vox\_es]. (6/6/2021).¡!Detenido un marroquí de 36 años por abusar sexualmente de una joven en Málaga. Aumentan los casos de violación cometidos por magrebíes y ni una expulsión. Ni una sola palabra tampoco sobre ellos en las grandes cadenas de televisión.

Vox links the other again, a Maghrebi man in this case, with the rape of a Spanish woman, that is, a violent out-group with an in-group to protect. The party also criticises the leaders' inaction through negative declarative sentences (ni una expulsión62) and the silence of the television channels to which it links with the interests of what the party calls marxismo cultural and lobbies de los progres<sup>63</sup> in other tweets. In addition, they perpetuate the gender order and the representation of women as vulnerable victims who must be protected by the men from the in-group. This self-presentation from a perspective of protection is applied to the LGTBIQ+ collective too, as read below:

Vox [@vox\_es] (1/6/2021) ¿Por qué se oculta que la mayoría de agresiones homófobas son producidas por inmigrantes ilegales?

This racialisation of sexism is likewise implemented with the use of the Great Replacement Theory, where it is stated that the white European Christian population is being replaced by non-Western people, normally linked to Arab countries. This can be seen in the following tweet, where they state that there is a birth rate problem (invierno demográfico64) in Spain, which is not being addressed by the coalition government:

Vox [@vox\_es]. (20/5/2021) 40% de paro juvenil, natalidad en picado y viviendas inaccesibles. Pero la solución del Gobierno para las pensiones es importar a medio África. La España 2050 de Sánchez va a ser de todo menos España.

The same idea is found in the following extract of their political programme:

Las élites globalistas apuestan por la llegada de millones de inmigrantes ilegales en los próximos años con el objetivo de revertir el invierno demográfico mientras imponen políticas contrarias a la familia. El avance del globalismo amenaza la preservación de la identidad cultural de las naciones occidentales (. . .) Toda Nación tiene derecho a regular las condiciones de entrada de quienes aspiran a vivir en ella. (Agenda España, 25).

Literally, not even one deportation.

In the same order: cultural Marxism and progressive lobbies.

Literally, demographic Winter.

These examples show the role played by the gender and racial orders and even the role of sexuality and reproduction in the discursive construction of the racialisation of sexism that justifies nativism, anti-immigration proposals and the strengthening of the patriarchal masculinity mechanism.

### 5 Conclusions

In the construction of otherness' processes, semiotic ideologies shape frameworks of discursive practices and signs' selections that, once applied to the in-group and out-group, create a homogenisation of subjectivities. Therefore, the possibilities of differentiation are ruined and the interpretive frameworks of diversity are cancelled, proposing simplified, reductionist, and even false, lines of access to reality. Hate speech and the threatening representation are implemented in the out-group's description by using linguistic strategies, such as the use of a negative modifier followed by a noun, the recurring use of negative lexicalisation (salvajes, quintacolumnistas) and emphasisers, unexpected collocations or lexical changes, iterative collocations and the use of hyperbole and metaphor (oleada de inmigrantes ilegales65), as well as a tendency towards sloganisation. These discursive tools belong to the base of Vox's racialisation of sexism that is reproducing power and promoting securitarian ethnonationalism. This racialisation, which comes from a colonial mentality, makes it possible to demand border control and a punitive system by the characterisation of non-Western migrant men as threatening subjects to citizens and national population.

Vox is a radical right-wing party, with a traditional structure, an authoritarian internal organisation and little internal democracy. This research has allowed to examine some of the most characteristic features of this political party, which are constructed dialectically through the in-group and out-group: nationalism and nativism, shown through verbs of statement and command verbs, war metaphors and possessive nouns and adjectives; antifeminism, by subordinating women, to whom it denies agency by using negatively connoted nouns and adjectives, as well as negative lexical changes to refer to feminist movements. It would be interesting to analyse Vox's discourse by comparing it with a base corpus made up of discourses from parties with a broad political spectrum. This would allow to obtain a number of specific words from Santiago Abascal's political party, whose keywords would help to examine the frameworks that they are (re)constructing.

Wave of illegal immigrants.

Anti-immigration and racism are another characteristic present in Vox that is linked to ethnonationalism. This is noticed in the use of negative lexicalisations in the categorisation processes of non-Western men and in the utilisation of presuppositions, a semantic mechanism that, when describing migrants' negative attributes, does not need to name the in-group's positive features, which are allegedly known and, thus, of common sense. This far-right party's strategy is linked to the use of a number of topoi on immigration, some of which are repeated in several tweets, such as the relationship between immigrants, crime and the violation of Western values and female bodies. Vox presents migrants as thieves and lazy people through aggrieved entitlement and victimisation, while the racialisation of sexism allows the party to characterise them as violent and rapists. These strategies are useful to legitimise securitisation policies and are linked to the representation of masculinity that Vox proposes for both the in-group and the out-group. Moreover, they belong to the discursive construction of the racialisation of sexism and the argumentative base of the unnecessary implementation of the gender perspective in public policies. This fact, together with the change in the framework for gender-based violence, as well as the justification for the construction of a fortress Spain, could have a number of consequences in public policies, such as a greater investment in defence and less investment in gender equality policies. This is an aspect that should be addressed in future research due to the possible impact on precarious, vulnerable and denied bodies, as well as on democratic quality.

### References

Acha, Beatriz, Analizar el auge de la ultraderecha, Barcelona, Gedisa, 2021.


Blasco, Inmaculada, Género y nación durante el franquismo, in: Michonneau, Stéphane/Núñez-Seixas, Xosé M. (edd.), Imaginarios y representaciones de España durante el franquismo, Madrid, Casa de Velázquez, 2014, 49 –72.

Bonino, Luis, Masculinidad hegemónica e identidad masculina, Dossiers feministes 6 (2002), 7–36.


Agenda España, 2021, https://www.voxespana.es/agenda-espana.

Vox Madrid, Proposición de Ley Integral de Igualdad y No Discriminación de la Comunidad de Madrid, 2021, https://www.voxespana.es/proposicion-ley-integral-igualdad-no-discriminacioncomunidad-madrid.

Wodak, Ruth, The politics of fear, London, Sage publication, 2015.

Wodak, Ruth/Baker, Paul/Gabrielatos, Costas/KosravNik, Majid/Krzyzanowski, Michael/ McEnery, Anthony, Useful methodological synergy? Combining critical discourse análisis and corpus linguistics to examine discourses of refugees and asylum seekers in the UK Press, Discourse & Society 19:3 (2008), 273–306, doi.org/10.1177/0957926508088962.

### Francisco José Sánchez García Spaniards First. Notes on Hate Speech towards Migrants on Twitter

Resumen: En un contexto político de gran polarización ideológica como el actual, en el que nacen nuevos partidos populistas, resulta crucial atender a los mensajes de odio difundidos a través de los medios de comunicación, y especialmente, de Twitter, que se ha convertido en fuente primaria de información para millones de personas. El presente artículo está dedicado a analizar el lenguaje xenófobo y racista a partir de un corpus compuesto por una selección de respuestas en la red de microblogging a la cuenta oficial del partido Vox. Nuestro propósito es doble: por un lado, sondearemos los marcos conceptuales más frecuentes que articulan el lenguaje de la extrema derecha y que calan entre los más jóvenes; por otro, trataremos de buscar la correlación entre esos marcos y los que partidos populistas como Vox retroalimentan a través de Twitter.

### 1 Introduction

Since the 2010s, the emergence of social networks -and in particular Twitter, halfway between a blog and a social network- has revolutionized the political communication strategies of the parties, which now have a fundamental tool to interact with their potential followers and voters in a direct and much more effective way. Without a doubt, it is interesting to examine the speech acts of an emerging discourse such as that of Twitter, which is based on apparently more democratic "horizontal" communication pillars, and where all opinions seem to have the same value (although not the same impact).

We start from the basis that the debates created in the heat of this network allow a kind of asymmetric communication in which the issuer can filter the comments that each of his writings receives, freely choosing whether to allow the opinions of others or, on the contrary, prefers to block them (Mancera/Pano 2013). To this we must add a particularly relevant feature: the limitation of characters, which makes it essential to synthesize the messages going to the heart of the topics, which can lead to the simplification of ideas, except in those cases in which elaborate threads are presented.

Francisco José Sánchez García, Universidad de Granada

Political parties, forced to position themselves in the public arena to make their ideas known among their potential followers, use Twitter to launch messages that are consistent with their slogans and framework ideas and try to guide the public debate according to their interests. Naturally, in the specific field of political discourse on networks, we must bear in mind that burning national or international current issues can be controversial, which, in general, ends up leading to a growing ideological polarization. In fact, the algorithm of the network itself contributes to promoting this polarization by "suggesting" friends or topics to the user, so that, in a short time, it is easy for the profiles to communicate more and more with those people who think like them, for what the messages of the dissenters are answered with greater radicalism and intolerance.

This form of communication sets in motion processes of identity construction (although it is a simplified, almost schematic form of identities), as pointed out by Olmos et al. (2020). In addition, we cannot forget that, in many cases, anonymity contributes the dissemination of radical messages, which becomes a breeding ground for populist discourse, which precisely tends to take advantage of exclusionary nativism to connect with people who are dissatisfied with the system, which allowing them to expand their voter base.

Usually, right-wing populist parties are attributed autarchic approaches that end up reproducing in arguments focused on the fear of diversity. In recent years, Spanish public opinion has progressively moved away from a more tolerant position towards immigration (Gutiérrez-Peris 2018, 104). Not surprisingly, 40% of Spaniards would vote for a party that promised to reduce immigration, which explains, in part, Vox's interest in polemicizing on this issue. It is likely that the economic crisis of 2008 represented an important turning point for the emergence of populism in Spain (Rama/Santana 2019, 19), which is characterized by providing simple solutions to complex problems. Specifically, the far-right tends to blame immigrants for the lack of job opportunities for Spaniards in times of high levels of unemployment.

The truth is that, in recent years, the dissemination of racist or hateful messages to those who are different has increased, and this may be due in part to the undoubted support of ultra-conservative populist parties. These hate messages and the political speeches of the far-right parties, whose feedback is evident, have three elements in common:


All this, together with other social factors, can partly explain the worrying increase in hate crimes in Spain. As we already explained in Sánchez García (2021), hate crimes committed through the Internet and social networks, according to the Ministry of the Interior (2019) represented 54.9% and 17.2% respectively. Between 2018 and 2019, there was an increase of 6.8 in hate crimes in Spain according to the 2019 Report of the Ministry of the Interior.

In this context, it is of special interest to analyze the discursive frameworks in the xenophobic or discriminatory reactions to the Twitter accounts of Vox Spain. Unlike the approach developed in previous works (Sánchez García/Sosinski 2022), we will not study what the leaders say; on this occasion, we are interested in examining the reactions (without filters or self-imposed limits from the political sphere) of their supporters, in which they let prejudices emerge that lead to discriminatory messages, such as the one we reproduce below (figure 1):

Figure 1: Reply to a tweet from the official Vox account (August 2019).

Examples like this show to what extent the parties light the fuse of the confrontation, but it is the sympathizers who complete the task, spreading and viralizing the political messages in a much more aggressive tone, which almost always highlights the fear of migrants.

### 2 Theoretical Framework

#### 2.1 The Populist Discourse of the Far-right Parties

Firstly, it is worth clarifying what we are talking about when we speak of "extreme right" or "Populist Radical Right". There is some consensus on the attributes that all parties of this ideology have in common: nativism, authoritarianism, and populism. In other words, these three traits converge in a combination of "xenophobia and nationalism" (Mudde 2007, 22). At first glance, it is not clear that Vox's ideology fits among the postulates of populism, since appeals to the people or the popular will are less frequent than in other discourses (for example, that of Podemos; cf. Sánchez García 2019), although an us/them contrast based on an excluding nativism is made explicit, thanks to its undisguised defense of the "national" against any external (or internal) threat that could attack the signs of identity of the Spanish people, whose essences have to protect yourself above all else.

The Vox political party has cultivated these clearly populist ingredients since its foundation in 2013 as a split from the Popular Party. Like other European "homologous" parties, it seeks to be recognized as a kind of "saving movement" (Charaudeau 2009), to the extent that, from the party's point of view, only Vox, and in particular its leader Santiago Abascal, can stand as a champion of the Spanish essence to fight for the integrity of the nation, its traditions, its culture, its values, and its lifestyle. In this way, the nation is understood as a homogeneous ideal, and, therefore, incompatible with, for example, peripheral identity sensitivities such as those represented by Catalan, Basque and Galician nationalisms, and much more incompatible with the minorities that arrive in Spain. from Africa, for example, especially if they are Muslim (Görlach 2018, 39).

Vox shares with other far-right parties in Europe (Marine Le Pen's National Grouping in France, Matteo Salvini's Northern League in Italy, Golden Dawn in Greece or Viktor Orban's party in Hungary) they share a radical ultra-conservative ideology and an identical discourse, based on in a powerful patriotic anchor that takes the form of aggressive messages against illegal immigration. Formally, they use a simple, close, and effective language with which solutions to complex issues seem easy. Ultimately, it is about making the "problem" visible to potential voters, taking advantage of an unfavorable situation that can be understood as the beginning of a crisis.

According to the hypothesis of the "populist moment" defended by Laclau (2005) and Mouffe (2012), some periods of socioeconomic stagnation in which an economic crisis converges with the decline of the traditional "establishment" parties lead to the emergence of populism, which forces the values commonly associated with the concept of Western democracy to be loaded with "exclusive, nativist and authoritarian nationalist notions" (Haynes 2019, 119). These messages, reproduced by the mass media (Van Dijk 1999; Casero Ripollés 2007), legitimize the exclusion of "the others" and end up permeating the unconscious of the citizenry.

Indeed, the activation of the "nationalist identity cleavage" is evident, linked to the "assumption of part of the nativist and xenophobic discourse of right-wing populism" (Haynes Campos 2019, 140). Precisely, "the issue of immigration is the focus of the discourse and programs of all the radical right-wing populist parties in Europe" (Kaya 2017, 55), which appear linked to "illegality, crime, violence, drugs, radicalism, fundamentalism, conflicts, and, in many other aspects, they are represented in a negative way" (Kaya 2017, 63–64).

#### 2.2 The Discourse of the Far-right on the Internet

The existing feedback between political discourse and journalistic discourse, or what is the same, between the media agenda and the political agenda, is well known, so that many times it is the media related to certain parties or circles of power that responsible for "preparing the ground" to spread the message they want to convey. In the Spanish case, the editorial lines of the media are easily recognizable, so that nobody is surprised that they support or criticize the party in government or those in opposition. But it happens that the most extremist or radical parties do not enjoy, a priori, the sympathy of the mainstream media, which makes it necessary to deploy new tools to make their arguments known and connect with their potential followers in a more direct and free way. the criticism or caricature to which they are exposed by most of the newspapers or TV channels. At a time when the conventional media is increasingly discredited and post-truth is prevailing (Martínez de Carnero 2020), viral messages and tweets filtered by the parties themselves are increasingly gaining credibility among broad sectors of the population. As Laura Camargo points out:

It is not surprising by trial and error that the Vox Communication Manual (2019) states that social networks are its main communication channel 'because they have the advantage that they allow us to transmit our messages without any mediation from the media. Our messages reach the Spaniards just as we want' (Camargo 2021, 65).

Parliamentary groups have been adapting their communication strategies in recent years to the changing reality of an increasingly digital and interconnected world. Thus, what was initially used to experiment with in the electoral campaign has ended up becoming a first-rate loudspeaker for leaders, and what is more important, to give voice and prominence to their followers, who have now become broadcasters of the message between your own contacts. This is what is known as an "expansive use of social networks" (Pérez Castaños/García Hípola 2021).

The great advantage of Twitter, halfway between a blog and a social network, is that it offers the possibility of establishing a (necessarily asymmetrical) relationship between the party and the citizenry (Mancera/Pano 2013). For example, Castells (2012) offers a positive (and almost utopian) vision of social networks, understood as "platforms for mass self-communication and spaces for communicative autonomy favorable to promoting the transition from indignation to hope". But the network also offers an unfavorable side: Merino and López highlight the advance of xenophobia and the growth of hate speech on the internet based on the Raxen Report of 2016. Other authors, such as Ortega (2017), also correlate hate crimes with the discourse of intolerance propagated through social networks such as Twitter (cited in Merino/López 2018, 214).

Vox, of course, has taken advantage of the undoubted benefits of the popular microblogging network to spread and viralize its argument (Sánchez García 2021, 122). Thanks to Twitter, the leaders of Vox have a very powerful loudspeaker to spread their ideas about sensitive and especially controversial issues such as the one at hand. These statements would not be reproduced by most of the media other than to be the object of harsh criticism, they can now be published without intermediaries.

This potential to unite the mass of followers is fundamental, since the algorithm and the inertia of Twitter's operation tend to limit the interaction of its users with those who "think and express themselves like them" (Sánchez García 2021, 126). In this regard, it should be borne in mind that the anonymity of the networks encourages digital haters to express themselves on controversial issues without any type of self-censorship. As Buson points out,

There is a depersonalization of the author, who is transformed, stops being who he really is. This goes far beyond what Jenkins (2006) comments on as fans of a particular show or topic. They are disturbed by simply reading, viewing audiovisual material or someone who does not agree with their ideas. They make extremely offensive verbal attacks, sometimes reaching physical attacks, clearly influenced by messages posted on social networks by other trolls. [. . .] We find the normalization of hate in sometimes really worrying comments, which could even be classified as criminal acts, such as when the integrity of the authors of a certain content is threatened (Busón Buesa 2020, 122).

We agree with Losada, Zamora and Martínez that precisely "hate is one of the most recurrent emotions in the virtual world. In fact, the expression "cyberhate" has been coined to refer to hate behavior that is visible through the network, which, in most cases, takes the form of racist, xenophobic and violent propaganda spread by means of digital systems [. . .]". Not in vain, the emotions shared through the Internet "have a resonance in those who publish them and in those who read or see them, producing an emotional response with a high concentration of feeling", so that "intense emotions in any positive-negative polarity they spread easily in the digital world and are enormously contagious" (2019, 196).

In any case, a medium that prioritizes depersonalization and alienation is fertile ground for the development of digital hatred, fueled by the emotional component that permeates the language of the networks. In the words of Piedad Bonnett:

Social networks are at the same time a favorable vehicle for the propagation of hate speech. Of course, they are a medium that favors the democratization of opinion, to the extent that they make visible all those traditionally marginalized from the great communication channels [. . .] It is not the networks, then, that are guilty in themselves, but the use that thousands of people make of them. The question is why are they such a conducive avenue for propagating hate speech? (2019, 182).

The author attributes the radicalization of messages to simplification and "binary" thinking (lack of nuances in the expression of opinions on sensitive issues) and to the emotional exacerbation generated by interaction with other users. Consequently, slogans prevail here and complex arguments, balance with the position of the other, and therefore, empathy, are left aside.

### 3 Methodology and Corpus

In recent years, some researchers have investigated the discourse of the political leaders of Vox on Twitter. Thus, for example, Camargo (2021) devoted himself to studying the treatment of false news (hoaxes) and the dehumanization with which far-right politicians describe migrants; We also had the opportunity to address the issue from a didactic approach (Sánchez 2021), and more recently, we have approached the statements of the leaders of Vox regarding the refugee crisis from Afghanistan (Sánchez/Sosinki 2022, in press). However, we believe that there are still no studies that have dedicated to delving into the form and content of the messages of their followers, which undoubtedly amplify the main frames of the party and exaggerate them, stripping them of any euphemistic or politically correct layer. Precisely for this reason, we find it especially interesting to explore the keys to hate speech towards migrants on the Internet in Twitter replies.

This work conforms to the methodological perspective of Critical Discourse Analysis (CDA), for which we must refer here mainly to Fairclough (1992), Van Dijk (1999), Charaudeau (2009) and Wilson (2001). As we have been explaining, in this study we will try to better understand the conceptual frameworks of the digital resentment discourse. Our corpus is made up of a selection of 100 Twitter responses to entries from the official Vox party account, in a period between September 1, 2021 and February 28, 2022.

Once the texts have been filtered and classified, they have been processed through the AntConc v.4.0.5 software to extract a list of the most frequent lexical units by issuer. From the tables obtained, we have dispensed with the grammatical forms (prepositions, conjunctions, articles), to keep only the notional words (nouns, adjectives, verbs and adverbs).

### 4 Results

Before delving into the analysis of the data, it is worth explaining what the profiles of the users examined are like. Anonymous profiles predominate (for example: Democracy, Justice and Freedom, Free Group, There will be no Peace, The howl of Spain, ADN Vox, Spain Forever, or The Spanish citizen). Naturally, users identified by their first and last names also express their opinions, although there are frequent express references to their ideology with icons of shamrocks or green circles (corporate color of the party), in addition to the Spanish flag.

Once the statistical count of the most frequent vocabulary was carried out using the AntConc v.4.0 software, we have obtained the following records and their corresponding frequency index (Table 1):


Table 1: Results obtained (lexicon and frequency index).


The results obtained leave no doubt about the racism of the messages studied: the most frequent record is the derogatory word moors (51 occurrences) with a frequency rate of 1.82% of the total number of words that make up the corpus, followed by menas (acronym in Spanish for 'unaccompanied minors') and minors (both with 45 occurrences, 1.61%). It is striking that the insult is above the common descriptor to refer to these immigrant minors.

As can be seen, the rest of the registered units contribute to building an openly discriminatory story: the words expulsion stand out (31 occurrences, 1.10%), illegal (12, 0.42%), scum (7, 0.25%), delinquency (6, 0.21%), among others. In short, as we will describe later in more detail, the correlation between "immigrant minors" and "crime" or "state parasitism" is explicitly established.

Although not all the messages are accompanied by hashtags, the most frequent in the responses to the Vox tweets are the following (Table 2):

As can be seen, some of these hashtags are a carbon copy of those used by the party, which supporters or party members make their own when addressing any issue on the political agenda marked by Vox: the vote for the party is openly


Table 2: Percentage of hashtags in relation to the total amount of entries.

appealed to, highlighting the "loose" allusion to the party, and also #YoVotoVox, #VoxExtremaNecesidad (play on words that responds to the accusation that Vox is a party of the "extreme right"), #TeamVox or #PrimeroVox. Equally striking are the patriotic outbursts: #ForSpain, #SpainForever, #FirstTheSpanish (in line with Donald Trump's famous campaign slogan in the US, "America First").

It is convenient to delve into the examination of the texts without losing sight of the set of tweets, and, therefore, their context. Next, let's consider the most prominent conceptual frameworks with some specific examples (in Spanish, the translation is ours):

#### a) IMMIGRANT MINORS ARE DELINQUENTS.

According to this frame, for the followers of Vox, all the migrant minors sheltered in Spain represent a threat, and they must be automatically and indiscriminately deported. It seems clear that the main framework is the one that correlates these migrant minors with criminal acts. We highlight some examples of this alarming message:


b) IMMIGRANT MINORS ARE VIOLENT.

Fearmongering the citizens is a constant that structures the backbone of many of the examples studied. In the following examples, the fear of violence that, according to some testimonies echoed by the commentators of the Vox tweets, the "means" would be capable of, is clearly addressed:


c) IMMIGRANT MINORS ARE ILLEGAL AND THEY SHOULD BE DEPORTED.

In line with the examples described above, and regardless of whether they commit crimes or not, for most of the Twitter users examined, the minors are in Spain in an irregular situation, which would be sufficient reason for their expulsion from the country, although in some cases it is specified that these deportations should be applied to those who commit crimes:


d) THERE ARE TOO MANY MUSLIMS IN SPAIN/THEY ARE INVADING SPAIN.

Once again, the exclusionary nativism of the extreme right emerges in many tweets that focus on the "excess" of foreigners of sub-Saharan or Moroccan origin in Spain, just because they are Muslim. In line with the #EfectoInvasion tag line, which we had occasion to comment on in our previous work (Sánchez and Sosinski 2022, in press), a clearly Islamophobic framework is used in these examples:


e) IMMIGRANT MINORS ARE SUBSIDIZED BY THE GOVERNMENT.

We also find a few examples that refer to the usual conservative mantra, according to which immigrant minors would be receiving subsidies or monthly payments at the expense of Spanish taxes:


f) IT TAKES A STRONG HAND/ONLY VOX CAN SOLVE THE PROBLEM.

All the arguments that we have been presenting unequivocally lead to an obvious conclusion for most of the broadcasters analyzed: it is necessary to tackle the "problem"", and the "social communist" Government of Spain (the coalition formed by the Socialist Party and the far-left party Podemos) is not in a position to do anything, so Vox is the only alternative.


### 5 Conclusions

The examples examined above allow us to get a glimpse of some clues about the hate speech towards young migrants, who are described by the followers of extreme right-wing parties as a threat to the freedom and rights of Spaniards: either as criminals, potential terrorists, or in the best of cases, a hindrance for the middle class (the accusation that they receive large subsidies is recurrent). The predominant framework is that of a hypothetical "invasion" of sub-Saharans or Muslims (from time to time, the hashtag #efectoinvasion is recurrent on Twitter, spurred on from the official account of the far-right party), which can only be prevented, according to its approach, by coercive and unsupportive policies.

Figure 2: Vox electoral poster during the run-up to the regional elections in Madrid. Source: El Español.

As can be seen in Figure 2, in the Madrid regional campaign of April 2021, Vox focused on the alleged favorable treatment of unaccompanied foreign minors by the state administration. The strategic cleavage of Vox is clear: it connects with the fears of Spanish society, based on false data (obviously, a minor immigrant does not receive 4,700 € per month) to generate discontent among uninformed citizens, relatively permeable to messages based on confrontation with the foreigners.

Likewise, we have been able to verify the evident correspondence between the frameworks outlined by Vox's communication strategy and those managed by the followers in the responses to the party's account, which trace the argument of the leaders (they even explicitly mention Vox as the only solution possible: "Only Vox remains"), although they show more radical approaches, completely devoid of euphemisms that mitigate the reference to immigrants. It could be inferred, therefore, that supporters get feedback from the messages launched by the leaders of Vox, such as Santiago Abascal, Macarena Olona, Hermann Tertsch or Espinosa de los Monteros and amplify them, somehow "translating" the party slogans into a more direct and crude language, and, therefore, openly racist, in deeply discriminatory terms.

Obviously, social networks are not axiologically negative, since on Twitter we can also find expressions of condemnation and rejection of discriminatory behavior, and denouncement of social injustices, but the increase in hate messages in recent years is worrying. In line with the far-right populist parties in the rest of Europe, a "preventive rejection" of migrants is taking hold. The defense of the essences of Spanish culture and values are put before any consideration of solidarity or reception of young migrants unaccompanied by relatives (MENAS), appealing to an exclusive nativism loaded with prejudices towards racialized or Muslim people.

### References


Olmos, Antonio/Rubio, María/Lastres, Nazaret/Martín, Paula, Jóvenes, redes sociales virtuales y nuevas lógicas de funcionamiento del racismo: Etnografía virtual sobre representaciones y discursos de alteridad e identidad, Madrid, Centro Reina Sofía Sobre Adolescencia y Juventud/Fad, 2020.


### Svenja Dufferain-Ottmann Constructing a Crisis by Constructing a Migratory Threat

Elucidating Explicit and Implicit Anti-Migratory Thought Patterns in the Covid-19 Discourse Made by the Right-Wing Populistic Vox and Rassemblement National Parties on Twitter

Resumen: En este estudio se elabora un anális del discurso populista relacionando la pandemia producida por la COVID-19 con el fenómeno de la migración. Basado en la concepción foucaultiana, que entiende el discurso como una práctica constructiva de sistemas de conocimiento, se analizan los patrones de pensamiento explícitos (análisis argumentativa de topoi) e implícitos (estrategias manipulativas del framing) en los discursos populistas del partido francés Rassemblement National y del partido español VOX, así como en los de sus líderes durante la pandemia de 2021 en Twitter. Los resultados de ambos corpus evidencian la lógica discursiva populista que consiste en la creación de una dicotomización social acompañada por la construcción del inmigrante como enemigo. En el corpus español se observa un estilo discursivo más agresivo y directo en el nivel lingüístico, igualando a los inmigrantes con criminales mediante coocurrencias y enmarcando metafóricamente el proceso de migración como si fuera una guerra. Además, se opone los españoles a los immigrantes a través del topos comparativo según el cual los inmigrantes tendrían más derechos de libertad que los españoles durante la pandemia, reforzados por patrones semánticos y sintácticos que expresan oposición. En el corpus francés, probablemente a causa de la estrategia de "desdemonización", se observa un estilo discursivo menos agresivo en el nivel lingüístico, y se intenta lograr la división social más bien a nivel argumentativo. Según estos patrones mentales, parcialmente elementos muy antiguos de los discursos epidémicos, los inmigrantes ("ilegales") traen la pandemia pasando la frontera y producen una desigualdad social porque reciben, según esta lógica, dinero que se les niega a los ciudadanos franceses.

#### Svenja Dufferain-Ottmann, Universität Mannheim

### 1 Introduction

An integral part of populistic discursive logic consists of constructing a crisis in the discourse (cf. Issel-Dombert/Wieders-Lohéac 2019, 7–9) in which social, political, and economic problems are spectacularized in order "to propagate the sense of crisis and turn 'the people' against a dangerous 'other'" (Olivas Osuna/Rama 2021, 3). In the right-wing populistic discourse, the role of this dangerous "other" is often ascribed to migrating people and serves as a scapegoat for citizens' problems (see the concept of empty or floating signifiers in Laclau 2004, 107). The present paper thus analyzes the extent and manner in which the Spanish VOX and French Rassemblement National parties – as well as their respective leaders – construct the COVID-19 crisis in their right-wing populistic discourses by establishing a connection between the pandemic and migration.

Using Foucault's conception of discourse, the paper aims to grasp both the implicit and explicit thought patterns of the above-mentioned parties and political actors that relate migration to the pandemic. The implicit patterns – which indicate the mentalities of the discourse community – are analyzed via the core concept of framing, and the explicit patterns are identified via topos analysis, which is a widely used concept from classical argumentation theory. The corpus stems from a screening of all tweets created by the accounts of the aforementioned political parties and actors during the 2021 pandemic year that establish a link between migration and the COVID-19 pandemic.

The paper begins with a theoretical part, which presents both a brief outline of the current situation concerning public opinion on immigration in France and Spain as well as a discussion of the potential impact of the right-wing populistic actors in this opinion-formation process. Subsequently, the paper's conception of populism and discourse are defined, followed by a literature review that summarizes the state of the art concerning the discursive connection between the pandemic crisis and migration. Finally, in the empirical part, the results of the corpus analysis are discussed.

### 2 Socio-Political Context

#### 2.1 Spain

Although Spain has had the highest level of immigration per capita of all EU member states throughout the past 20 years, several opinion studies have revealed that the country has consistently maintained more open attitudes than the European average toward immigration, with less rejection of and a greater appreciation for the contributions that migration makes to both society and the economy (cf. González Enríquez/Rinken 2021, 1; Special Eurobarometer 2018, 48). This sentiment can be traced back, for example, to the experiences of Spanish migrants in the 1960s and to the counterreaction against nationalism ignited by the Franco regime. However, a slight increase in negative attitudes toward immigration has recently been noted in Spain, although it is not clear how this change can be attributed to VOX's antimigratory discourse or to other events, such as the COVID-19 pandemic or irregular immigration to the Canary Islands (cf. González Enríquez/Rinken 2021, 1).

VOX was founded as a "liberal-center-right party" (Olivas/Rama 2021, 4) by three former members of the Partido Popular conservative party, including VOX's current leader, Santiago Abascal. Gradually, the party began to adopt more extreme positions in line with other radical right-wing parties (Olivas/Rama 2021, 4). VOX grew very rapidly in a relatively short time, becoming "the newest sta[r] in the European populist radical right firmament" (Mudde 2019, 40). Although VOX's electoral success in the 2018 Andalusian elections as well as in the 2019 European and national elections cannot be ascribed to generalized growth in anti-immigration attitudes among the Spanish population as a whole, the slight hardening of Spanish public opinion toward immigration is undeniable and has shifted the boundaries of what is perceived as thinkable and sayable: "What is clear, however, is the effect VOX is having on polarisation, causing a growing divergence of attitudes towards immigration based on ideological positions. [. . .] This polarisation is concerning because it becomes an obstacle to calm and rational debate on immigration and public policies devoted to manage it" (González Enríquez 2021, 9).

#### 2.2 France

Although positive attitudes toward immigration in France surpassed negative attitudes in 2018 for the first time since surveys began in 2002, immigration remains a hot political topic. The greatest concerns relate to the effective integration of immigrants into French society. Traditionally, French immigration policy has been driven by an assimilationist approach that aims at fully integrating the migrant population into French society (cf. Holloway/Faures/Kumar 2022).

Despite the relatively low percentage of immigrants in France compared with in almost all neighboring states, national identity and integration constitute the principal narrative that divides and dominates public debate. Under Marine le Pen's leadership of the far-right Front National party (which has been called Rasssemblement National since 2018) beginning in 2011, the party increased in membership, electoral success, and public opinion (cf. Stockemer/Barisione 2017, 2). For example, in the 2014 European Election, the FN won 25% of the votes, thereby outdistancing the classical, established parties by several percentage points and becoming the most successful radical right-wing party in Europe. On the national level, the party has earned the second-largest number of votes in the first round of the presidential election three times (in 2002, 2017, and 2022).

In contrast to the radicalization of VOX's discourse (cf. Olivas/Rama 2021, 4), a superficially more moderate discourse under Marine Le Pen has been observed – known as "dédiabolisation" ("de-demonization"), although the ideology itself has not changed (Stockemer/Barisione 2016, 12–13; Alduy/Wahnich 2015). Under Marine Le Pen's father, immigration had been framed as a matter of security (cf. Stockemer/Barisione 2016, 8), whereas Marine Le Pen links immigration more to economic, social, and cultural elements and connects it with a discourse that criticizes the politics of the European Union (Stockemer/Barisione 2016, 8).

In particular, Marine Le Pen's efforts to rebrand the party have succeeded in modifying the opinion of a significant portion of the French population. Polls have shown a decrease in the number of people who consider the FN a "danger" from 70% in 2002 to 53% in 2012 (Stockemer 2017, 24). Another poll indicates that a slight majority of respondents view the FN as "a party [that is] like the others" (Stockemer 2017, 24).

### 3 Populism as a Discursive Style

Populism has been defined in political science as an ideology, as a form of discursive logic, and as a communicative style (cf. Olivas Osuna/Rama). The conception of populism on which the present paper is based combines the more abstract notion of discursive logic as a type of deep structure within a discourse with the notion of discursive style (cf. Issel-Dombert 2020, 539) as the linguistically realized patterns of this logic in the surface structure of a discourse.

The problem of a mere ideological conception has been widely discussed: Populism arises from different points in the socioeconomic structure, and it is therefore nearly impossible to represent one common ideology. Even distinguishing between left-wing and right-wing populism appears overly difficult. For example, right-wing populism can be hosted by different ideologies, such as liberalism, socialism, and conservatism (cf. Wieders-Lohéac 2018, 61), and as Kienpointner (2005, 218) has shown, even right-wing populistic discourse differs according to different national contexts. While many authors note that the exclusion/inclusion of social groups appears to be a distinctive feature shared by left-wing and right-wing populism (cf. Wieders-Lohéac 2018, 63, Issel-Dombert/Wieders-Lohéac 2019, 9), other authors consider the exclusionary aspect to constitute populism in total (cf. Sánchez/ Alcántara-Plá 2019, 106; Sullet-Nykander/Bernal/Premat/Roitman 2019, 3). Since populism does not offer a complete worldview, Mudde proposes conceiving it as a "thin-centered ideology" (Mudde 2004, 544) that considers society to be ultimately divided into two antagonistic and homogenous groups: "the pure people" and "the corrupt elite" (cf. Mudde 2004, 543).

The dichotomic, antagonistic element of Mudde's definition is also central to Laclau's (2004) political-theoretical approach, but Laclau considers this element to be part of an articulatory or discursive logic rather than part of an ideology. According to Laclau, using populist discursive logic enables social space to be dichotomized by creating an internal frontier between the people – presented as all having similar, unfulfilled demands – on the one hand and "the power" – conceived as the enemy, "which does not satisfy any of those equivalential demands" – on the other hand. As these demands represent "a highly heterogeneous reality," they must be discursively homogenized in the form of "empty signifiers" in order to be perceived as similar. Within a populistic rhetoric, constructing a common problem can be reduced to constructing a common feeling of disappointment (cf. Sullet-Nylander et al. 2019, 2) or fear (cf. Wieders-Lohéac 2018, 64), which is presented as being ascribable to a dangerous enemy (Widers-Lohéac 2018, 64). Following this logic, by eliminating the enemy, it is possible to eliminate the problems of all society members (cf. Charaudeau 2011, 108; Ruiz-Sánchez/Alcántara-Plá 2019, 106): "There is no populism without construction of an enemy" (Laclau 2004, 107). The discursive construction of an enemy by attributing to an actor an ethos of immorality serves as an antipode to creating a morally "pure" and honorable group of people. This dichotomic role ascription leads to a decomplexifying perspective of reality. The threefold construction of (1) the threatening enemy, who is held responsible for the source of (2) a catastrophic situation accompanied by (3) a disappointed group of people is the condition sine qua non for the populist's ethos construction of the powerful savior both of national identity and of the people (cf. Charaudeau 2011, 110–112).

Concerning the type of right-wing populism analyzed in this paper, focus is not placed on the construction of the enemy "within the state, within the nation," but rather on migrants as the enemy "within the state, outside the nation" (cf. Mudde 2007, 65; Moreno Moreno/Rojo Martínez 2021, 9) and as a central element of the right-wing populistic logic: "In order to approach European right-wing populism, to capture its essence, to understand its construction of reality and its argumentative strategy, de facto, it is not possible to avoid focusing on the migration issue" (cf. Wieders-Lohéac 2018, 64, my translation).

### 4 The Concept of Discourse

This paper conceives of populism as a specific form of discursive logic that aims at dichotomizing social space by discursively constructing enemies (see Chapter 3) and that is graspable in its surface structure via a linguistically realized discursive style or discursive patterns. The notion of "discursivity" is central because it integrates the constructive potential of language as elaborated by historian and philosopher Michel Foucault, a central figure in so-called critical and descriptive discourse linguistics (cf. Tereick 2016, 22) and on whose conception of discourse this paper is based. According to Foucault, reality cannot be conceived independently of discourse as a practice that systematically generates the objects it discusses (cf. Foucault 1969). Thus, truth and knowledge do not exist as such, but are generated by the discourse community and – when linguistically represented – are always driven by subjective interests and a subjective perspective. Hence, language is not a representational system of signs, but a constructing element. Foucault's interest lay in capturing thought patterns that limit and direct what is sayable within specific discourse communities, thereby leading to the creation of knowledge structures, or mental models.

Based on this conception of discourse, the present paper aims to grasp both implicit and explicit thought patterns. Implicit patterns – which indicate the mentalities of a discourse community – are analyzed via the core concept of framing, and explicit patterns are identified via argumentation-based topos analysis. Argumentation analysis reveals assumptions that are considered plausible within "everyday argumentation" (Kienpointner 2005, 219) in a specific discourse community. Argumentation is not considered implicit because the hearer is aware that the speaker aims to influence him by seeking to obtain his consent. This is not the case with framing. Instead, framing can also be analyzed in terms of manipulation, which is a field dominantly explored by German and English critical discourse analysists (cf. Reisigl 2018, 163; Van Dijk 2015, 472).

### 5 State of Research

The discourse on immigration is a common research subject in so-called critical discourse analysis (cf. Van Dijk 2018; Reisigl 2018, 169). For example, van Dijk (2018) refers in his literature review on the migration discourse mostly to studies aimed at uncovering racist discourse, especially in mass media, but immigration is also a topic treated by scholars who explicitly refer to descriptive discourse analysis (e.g., Böke/Jung/Niehr/Wengeler 2000; Wieders-Lohéac 2018; Issel-Dombert/ Wieders-Lohéac 2019).

In German Romance linguistics, the migration discourse was recently discussed in Pietrini's anthology (2020), which paid special attention to the migration discourse in Italy. Due to the central role of this topic in right-wing populistic discourses, the migration discourse often involves analyses of populistic discourse (cf., e.g., Wieders-Lohéac 2018, Issel-Dombert/Wieders-Lohéac 2019, Visser 2018, Pirazzini 2019).

Papers that analyze the discursive interrelationships between COVID-19 and the migration discourse are markedly scarce due to the relative recency of the pandemic. The same applies to analyses of the impact of the pandemic on the populistic discourse in general (cf. Olivas Osuna/Rama 2021, 2).

Vega Macías (2021) analyzed whether the COVID-19 pandemic has influenced the political discourse on immigration in Europe and the United States based on a "revision" of news taken from newspapers, news agencies, and international organizations from March-September 2020. As Vega Macías claims, the pandemic has strengthened anti-immigration positions, although his approach left pending questions, for example, concerning the methods and concepts used, the corpus size and composition, and the exact measurement of the increase in anti-immigration attitudes. Surprisingly, no examples were given from newspapers; instead, only extracts from political speeches of mostly right-wing politicians were included.

Sociologist Rodríguez Peral investigated the agenda-setting and framing of the informative treatment of immigration by Spanish public television's news programs during the COVID-19 pandemic. She analyzed the totality of the news from January-September 2020 and found that more than two-thirds of the news programs represented migrants negatively. Migrants were framed as a homogeneous and depersonalized mass. Simultaneously, Donald Trump's hate speech was being given space without an alternative, counterbalancing view (e.g., reporting about the experiences of Spanish citizens with migrating people).

Political scientists Olivas Osuna and Rama (2021) used content analysis to test whether the COVID-19 pandemic has had an impact on the populistic character of the discourse of VOX and other political parties in Spain by analyzing parliamentary speeches from March-June 2020 and comparing them with earlier discourses from the selected parties. The scientists measured populism using five sometimes-overlapping parameters: the "antagonistic depiction of the polity," "morality" (the superiority of "the people" vs. the inferior moral standards of "others"), the "idealistic construction of the society," "popular sovereignty," and "characteristic leadership." The analysis revealed both a growing density of populistic features in Absacal's speeches – which could be linked to an increased intensity of the parameters of "morality" and "antagonism" – and a type of spillover effect onto the other parties in the parliamentary sessions.

As they stem from disciplinary areas other than linguistics, the above-cited research approaches neither conceptualized the discourse nor operationalized it via linguistic concepts. Instead, the analyses were simply based on semantic elements and used content analysis. Concrete language use and linguistic strategies were thus entirely excluded, thereby corroborating the need for a linguistic perspective on this issue.

### 6 Corpus

In the first step of corpus compilation for the present research, all tweets made by Rassemblement National, Le Pen, VOX, and Asbacal from January-December 2021 that thematized migration were collected. In order to be able to cover all tweets concerning migration, search terms were not used; instead, the content of every tweet was assessed, which yielded 125 tweets by Rassemblement National, 195 by Marine Le Pen, 198 by VOX, and 169 by Santiago Abascal for the chosen period. In the second step, all tweets that related migration to the COVID-19 pandemic were selected, which amounted to 7% (9/125) from the Rassemblement National corpus, 7% from the Le Pen (14/195) corpus, 6% (12/198) from the VOX corpus, and 4% (7/ 169) from the Abascal corpus. Although the number of tweets in this total corpus is relatively small compared with the overall number of tweets that thematized migration in 2021, the corpus is representative insofar as it comprises all tweets that relate the pandemic to migration. The thought patterns are assessed qualitativecomparatively. Although both Rassemblement National and VOX are considered right-wing populistic parties (cf. Chazel 2020; Eklundh 2020), the comparability of both discourses – which is often a problem in comparative corpus studies (cf. Böke/ Jung/Niehr/Wengeler 2000, 15–16) – is ensured for several reasons: (1) The pandemic is a common global context parameter that (2) has led to similar effects in Spain and France because the two countries have been the hardest-hit OECD countries on the economic level in terms of loss of BIP (cf. Gern/Hauber 2020, 899). Moreover, (3) both countries have had the highest number of COVID-19 cases out of all European countries (cf. Radtke 2022) and also (4) display comparable patterns concerning cumulative mortality and crude mortality rates, including similar age and sex distribution (cf. Gallo 2021 et al.). The selected corpus covers a period during which both countries experienced their 3rd and 4th waves of the virus as well as a politically imposed state of emergency and lockdown accompanied by restrictions to individual freedoms of actions.

Using Foucault's constructive conception of discourse, this paper aims to uncover the implicit thought patterns behind the discourse by analyzing framing strategies (Chapter 7) and explicit thought patterns via topos analysis (Chapter 8).

Since the present paper focuses on analyzing thought patterns during the COVID-19 pandemic (including argumentative patterns), Twitter is an appropriate social media platform for analysis as it has a strong informative function (cf. Rufai/ Bunce 2020, 511). It has also been highly frequented during the pandemic due to its fast-reacting character (cf. Heidenreich et al. 2015, 122) and the increased need for news consumption by the information-seeking population (Haman 2020, 7).

### 7 Framing

Frames organize speakers' belief systems (cf. Entman 1993, 52) and are thus indicators of these speakers' mental models (cf. Münch 2021), which are subjective cognitive representations of complex situations based on both individual assessments and the individual allocation of attention (cf. Moser 2003, 188). Frames were first introduced to linguistics by Fillmore (1985, 223) as "lexical representatives of some single coherent schematization of experience and knowledge," which laid the foundation for frame semantics as a future branch of research. Frames can be used for strategic actions – that is, so-called "framing," which essentially involves selecting and highlighting features of reality while simultaneously obscuring others (cf. Entman 1993, 52).

Framing refers to how people establish subjective definitions of a situation (cf. Entmann 1993, 54; van Dijk 2015, 471). For example, a demonstration may be defined either as a violation of the social order or as a democratic right of demonstrators. Similarly, a violent attack may be defined as a form of resistance against the abuse of state power or as a form of terrorism (cf. van Dijk 2015, 473).

An affinity exists between framing and manipulation. With manipulation, the speaker intends to dissimulate their pursuit of strategic goals (Fairclough 1994, 2360; Nettel/Roque 2012, 57) and the fact that a situation exists with a potential clash of interests on the part of both the speaker and the hearer. Through this dissimulating behavior, the manipulating speaker aims to avoid the hearer's rational assessment of a contentious issue. Thus, the manipulated hearer is not aware of the speaker's attempt to exert influence over the hearer, or as van Dijk (2015, 467, 472) puts it, of the speaker's attempt to "control the mind": If such discursive control over mental models of recipients is in the best interest of speakers and against the best interest of recipients, we have an instance of discursive power, or manipulation (ibid., 472).

The analysis of framing strategies is measured via the concepts of semantic roles, keywords, co-occurrences, metaphors, and presuppositions.

#### 7.1 Allocution of Semantic Agent Roles

The distribution of semantic roles can help to put reality into perspective. For instance, hiding agent roles by using the syntactic pattern of passivation or the morphological procedure of nominalization can downplay or hide the speaker's responsibility for their own actions that they perceive as unfavorable (cf. Van Dijk 2015, 474; 475). Normally, in the right-wing populistic discourse, minorities are not ascribed the semantic role of agent; usually, they are depicted as inanimate objects onto which actions are imposed (cf. Ruiz-Sánchez/Alcántara-Pla 2019, 121) or to which (negative) attributes are ascribed. This observation can be corroborated by the results of the French corpus, in which agent roles are attributed on average in only 11% of the tweets.

The tendency to avoid distributing agent roles to migrants in the French corpus can also be seen in connection with the absolute lower level of the French corpus to use verbal expressions that refer to migrating people. Linguistic expressions that refer to the lexical field of migration verbalizing migrating people are three times less frequent in the French corpus compared to the Spanish.

In the Spanish corpus, on average 77% of the tweets use (negatively connoted) verbs that ascribe agent roles to the migrating people (see Examples 1–2). The high rate of agency attribution in the Spanish corpus correlates with the employed topos that migrants who are in the country "illegally" during the pandemic have more rights than Spaniards (cf. chapter 8.2.1) – that is, that these migrants are allowed to move about freely while Spaniards are condemned to "be imprisoned" at home. In this case, the distribution of the semantic roles creates a discursive dichotomization between migrants as agents who have the right to act on the one hand and Spaniards as inactive patients onto whom negatively connoted actions are imposed on the other hand. In four tweets, this contrast between the semantic roles takes the form of a slogan (see Example 1):


("They dominate the streets in the midst of a state of alert while honorable Spaniards have their rights restricted.")

#### 7.2 Keywords

Keywords can be understood as indicators of frames (Entman 1993, 52) and mental models (Münch 2021, 129) that express the self-concept and ideals of a group (cf. Warnke/Spitzmüller 2008, 26) and influence people's thoughts, emotions, and behavior (cf. Niehr 2007, 496). Like "tips of icebergs," they point to "complex lexical objects that represent the shared beliefs and values of a culture" (Griebel/Vollmann 2019, 676) or discourse community. Keywords have argumentative potential and are used with a certain frequency (see Niehr 2014, 89). The keywords "illegals"/"illegal migrants" in both the French and Spanish corpora and "frontier" in the French corpus were identified and are discussed in the following sub-section.

#### 7.2.1 "Illegals"/"Illegal Immigrants"

The discourse in both the French and Spanish corpora focuses on connecting the pandemic to migration and to migrants who stay in a country "illegally" (see Examples 3 and 4), mostly by calling them "illegals" or "illegal migrants," thereby directing attention to a phenomenon that has little relevance in terms of numbers (cf. Eurobarometer 2018, 14). However, every linguistic reference to a phenomenon has an impact on solidifying the phenomenon in a population's collective consciousness. The more often a phenomenon is referenced, the more deeply it is perceived (cf. Scharloth 2010, 99; Varga 2019, 330). Repeating the feature of "illegality" renders the phenomenon more salient, which "means making the piece of information more noticeable, meaningful, or memorable to audiences" (Entman 1993, 53). In Le Pen's ironic tweet concerning the increase in financial aid provided to migrants who stay in France irregularly, she frames the target group of this assistance as "illegals":

(3) Le poids financier de cette prise en charge réservée aux clandestins est mirobolant pour un pays dont le système hospitalier est déjà mal en point. [21\_09\_21] ("The financial burden of this care reserved for illegals is staggering for a country whose hospital system is already in bad shape.")

In Spanish Example (4), migrants are globally criminalized as "illegals":

(4) Los españoles encerrados y arruinados. Los ilegales, que llama y trae el gobierno, ocupando las calles. [21\_03\_06] ("Spaniards, imprisoned and ruined. Illegals, called and brought over by the government, occupying the streets.")

Denoting migrants as "illegals" creates a division between the legal, morally pure nationals and the morally impure migrants, who violate the law. The term "illegal" leads to the criminalization of the migrating individuals (Stamatinis/Archakis/Tsakona 2021, 348) and frames them as a threat to honorable, "legal citizens." In this narrative, migrants are not framed as victims in need of protection.

Due to this construction of the "migrant criminal" (ibid.: 348) that connects migrants with criminality, the United Nations General Assembly has urged the organs of the United Nations and their specialized agencies to use the term "non-documented migrant" or "irregular migrant workers" in every official document since 1975. In the same vein, in 2009, the European Parliament called on EU institutions and member states to avoid the term "illegal immigrants" due to its extremely negative connotations and to instead use "irregular/undocumented workers/migrants" (ibid., 349).

Stamatinis/Archakis/Tsakona (2021, 351) demonstrated that the term "illegal migrant" was used in the 2015 Greek Parliamentary sessions exclusively by extreme right-wing parties in the context of their extreme anti-migrant rhetoric.

#### 7.2.2 Frontier

The noun frontière ("frontier") is used in half of the tweets from the French corpus.

Historically, the border between France and Spain occupies a central position in the French populist immigration discourse of Rassemblement National and the former Front National (cf. Varga 2019, 335) and has the symbolic function of a filter that protects the morally pure people and nation from dangerous "others," who are presented as a threat to the nation's security, economy, and identity. Marine Le Pen defines the border as follows: "Nous croyons en la frontière qui protège, qui est une saine limite entre la nation et le reste du monde, un filtre économique, financier, migratoire, sanitaire et environnemental" ["We believe in a border that protects, that sets a healthy limit between the nation and the rest of the world, an economic, financial, migratory, health, and environmental filter"] (cf. Alduy/Wahnich 2015, 134).

While the border is traditionally invested with the function of guarding the people against Identitarian loss and insecurity, the discourse on the COVID-19 pandemic has caused an old narrative to reappear that depicts foreigners as importers of disease and plague (cf. Dinges 2004) and frames the frontier as the protecting separation line that filters the sane from the insane.


The COVID-19 pandemic has been "a window of opportunity" (Olivas Osuna/Rama 2021, 1) for Le Pen and her party to embed the claimed need for border controls in a different context of meaning, thereby connecting the issue with healthcare elements: "The analysis of populistic discourses through great events such the COVID-19 pandemic can help [us to] understand how populist leaders adapt their communicative style to take advantage of changing circumstances" (Olivas/Rama 2021, 3; see also Vega Macías 2021, 10).

The argumentative potential of this keyword can be seen in its frequent use in the topos that migrants are a threat because they bring the pandemic (see chapter 8.1).

#### 7.3 Co-occurrences

The immediate co-text can have a framing function by casting the surrounding element (issues or individuals) in a derogatory or uplifting light (see, e.g., the framing of Assad as friend or foe through co-occurrences in German news magazines in Rüdiger 2018, 302). These "semiotic patterns" (Griebel/Vollmann 2019, 677) can be related to the question formulated by Foucault (1981, 47) as to how a specific utterance occurs in a certain place instead of in another.

In one-quarter of the Spanish tweets, the already-criminalized "illegal migrants" are additionally verbally framed as criminals by having their depiction co-occur with lexemes that refer to the concept of "criminality." Through this type of co-occurrence, migrants are presented as sharing the same core features on the level of semantic intention and as belonging to the same super-category as criminals. Generalization is a well-known "discourse move" (van Dijk 2015, 474) for forming stereotypes and prejudices.


In the French corpus, immigration is framed as a threat by the immediate co-text. The negative attributes given by the co-text are presupposed in all cases (cf. chapter 7.5) and have the advantage that the political actors can avoid arguing for the truth of this (highly subjective) worldview since it is presented as common knowledge.


### 7.4 Metaphors

Metaphors reveal how people process and conceive the world and how they concretize abstract knowledge in mental models (cf. Moser 2003, 191–192; Van Dijk 2015).

A cognitive and thus new conception of metaphors was made popular by Lakoff/Johnson (1980), who went beyond the hitherto widespread conception of metaphors as mere ornamental rhetorical devices (Spieß 2011, 204). Based on Gestalt research, the above-mentioned authors view metaphors as central instruments in creating analogies by transferring primarily sensory experiences of a source domain to an abstract target domain (Ziem 2015, 53) in order to make the latter more easily comprehensible: "Because we cannot directly experience these concepts – we cannot touch, smell, hear, taste or see the concept of taxation, for instance, [. . .] our minds and our language automatically resort to conceptual metaphors to assign them meaning" (Wehling 2018, 137). Therefore, the complexity of metaphorized issues is automatically reduced, simplified, and coarsened (cf. Spieß 2011, 209) since the transfer from the source to the target domain does not comprise all features, but only those that serve the visualization. Metaphors are a powerful framing tool that can be used to direct recipients' attention and to influence the way they perceive the world because these metaphors highlight some aspects of an issue while simultaneously hiding others. For instance, when immigration is conceived via flood metaphors, experienced negative and threatening elements and emotions are highlighted, while others are omitted (cf. Dargiewicz 2021, 44).

The conceptual metaphor type that is relevant to our corpus analysis is war or fight metaphors, which are frequent in political discourse as a means of demonstrating power, involvement, or commitment (cf. Gauthier 1994). However, it is typical of right-wing parties "to identify migrants with invaders through invasion metaphors" (Stamatinis/Archakis/Tsakona 2021, 356, see also Wodak/Reisigl 2015, 581; van Dijk 2015, 473, Pirazzini 2019, 136). By labeling migration as an invasion, it is conceived as a warlike, aggressive, and massive attack of warriors who violate the territorial integrity of the national state. On a more general level, migration is conceived as war. As this mental model implies that migrants – as invaders – commit the assault, a dichotomic construction of perpetrator and victim is realized, thereby legitimizing the "fight" – that is, the "counter-attack" against this group (Pirazzini 2019, 138–139).

Although Pirazzini demonstrated that invasion metaphors were frequently used by the Front National, the French corpus does not display this feature at all. In contrast, when VOX and its leader frame the COVID-19 discourse as a discourse of migration, migration appears in 60% of the tweets categorized as war tweets.


("They want Spaniards to be locked up during Easter Week while they watch illegal immigrants assault our borders.")

#### 7.5 Presuppositions

Presuppositions represent the knowledge shared by the speaker and the speaker's recipients. When used manipulatively, presuppositions can be used by speakers to present controversial issues as commonplace (cf. van Dijk 2010, 182; 188–192), thereby causing the hearer to believe that the presupposed content is already part of their own experience (Fairclough 2015, 165) even though the speaker is well aware that this is untrue (Sbisà 1999, 500). In this case, speakers exploit presuppositions in order to bluff "insofar as it allows [these] speakers to purport information as given which is in fact new" (Schmid 2001, 1529). Thus, presupposed content is "imposed upon the discourse" (Griffiths/de Vries 2014, 43, emphasis in original). When controversial issues are encoded as presuppositions, the speaker does not have to wrestle for the hearer's consent on a given argument because the information is presented as already being shared. Presuppositions can be identified when they "survive" the negation test. They are mostly triggered semantically and syntactically by specific sentence orders (cf. Grewendorf/Hamm/Sternefeld 1989, 433).

In the Spanish corpus, presuppositions are mainly used to denounce the ruling party – an aspect on which the present paper does not focus. For the remainder of the corpus, highly subjective war metaphors (e.g., examples 11–12) are particularly often presupposed when a speaker pretends that this mental model of conceiving of migration as a war is a commonly shared and uncontroversial fact that requires no supporting argumentation. In Examples (11), the use of the definite article encodes the existence of a migratory invasion as an uncontroversial fact. In Example 12, the factive verb ver ("to see") frames the "assault" of the border as uncontroversial common knowledge.

In the French corpus, two patterns of presupposed content are present: first, it is framed as a fact that the border and thus also migration are actually out of control (by demanding that the state has to take control via the use of the verb maîtriser, "to master") and second, it is presented as common ground that immigration is a problem for the already-beleaguered healthcare system, triggered by the semantics of stopper ("to stop") (cf. examples 13 and 14).

(13) [. . .] il faut stopper l'immigration qui plombe notre système de santé. [LP\_21\_10\_22]

("[. . .] immigration, which is dammaging our healthcare system, must be stopped.")

(14) Élue présidente, je mettrai fin à une immigration qui coûte une fortune à notre système de santé. [LP\_21\_11\_6] ("As President, I'll put an end to an immigration that costs our healthcare system a fortune.")

### 8 Topos Analysis

Aristotle's Ancient Rhetoric viewed topoi as belonging to the logos level of persuasion. Topoi thus enable us to reconstruct the collectively shared thought patterns and mentalities of a discourse community (cf. Wengeler 2000, 135; 140). Kienpointer (2017, 189) explicitly relates topoi to Toulmin's inference warrants (cf. van Eemeren et al. 2014, 204; Wengeler 2020, 651, 654). These warrants – or inference rules – establish a quasi-logical, plausible relationship between the argument/ data and the conclusion/claim in everyday speech (Wengeler 2020, 650).

However, in contrast to manipulative informative presuppositions, in which the audience does not recognize that influence has been exerted, with argumentation, the speaker does not primarily hide their intention to influence the hearers and thereby seeks rational acceptance of and consent for a controversial issue (cf. Nettel/Roque 2012, 59). Of course, the speaker aims to persuade the audience that their own worldview is the one and only true view; however, the hearers are well aware of the presence of the controversial issue and the fact that influence is being exerted on them.

In his typology, Kienpointner (1992a) distinguishes between three major classes of abstract argumentation schemes that are used to influence the plausibility of everyday speech: The first type uses warrants, the second type establishes warrants via inductive or illustrating examples, and the third type neither uses nor establishes warrants (such as argumentum ad verecundiam (argument from authority) or argumentum per analogiam (argument by analogy). Wengeler (2000) concentrates on the first category and introduces a topos analysis to German descriptive discourse analysis by establishing and comparing the thought patterns of immigration discourse in German-language journal articles from 1970s Germany, Austria, and Switzerland. His approach has had significant influence in German discourse analysis, including in the recent topos analysis in Romance Linguistics in Germany (cf. Issel-Dombert/Wieders-Lohéac 2019). The third type of scheme is mostly used in critical discourse analysis to uncover fallacies in the sense of common errors in logical reasoning (cf. Reisigl/Wodak 2001).

The first type of scheme comprises four types of semantic relationships that aim at a plausible transition from premises to conclusions (i.e., subsuming schemes, comparative schemes, causal schemes, and opposing schemes, cf. Kienpointner 1992a, 246; Kienpointner 1992b, 182). This type is of particular interest to the present paper for its ability to systematize the knowledge patterns of the chosen radical right-wing discourse communities. Figure 1 schematizes the findings explained below.

Figure 1: Topos patterns in the COVID-19 migration discourse.

In both the French and Spanish corpora, the overall conclusion (Conclusion I in Figure 1) of all argumentation patterns is the deontic, normative (mostly unverbalized) thesis that migrants who stay in a country irregularly must leave because they represent a threat during (the already-burdensome) times of COVID-19 (Argument I). This argument – itself a controversial claim on the next hierarchical level (Conclusion 1a-b) – is supported in both corpora by two types of arguments: first, by the causal argument that "illegal" migrants bring the pandemic (Argument 1a), and second, by the causal argument that migrants who stay in a country irregularly disrupt the state of peaceful social coexistence during times of COVID-19 (Argument 1b). The latter argument is again a controversial conclusion (Conclusion 3a-b) on the next-lower hierarchical level and is supported on the one hand by the comparative argument that undocumented migrants are granted more rights than nationals during times of COVID-19 (Argument 1a) and on the other hand by the comparative, a maiore argument (cf. Kienpointner 1992a, 285; Kienpointer 1992b, 163; Wengeler 2020, 651) that migrants receive financial support while nationals – who should have priority – do not (Argument 1b).

Figure 2 displays the frequency distribution of Topoi 1a and 1b in both corpora.

Figure 2: Frequency distribution of Topoi 1a and 1b.

As evident in Figure 2 the topos that migrants who stay in a country irregularly are a threat because they bring the COVID-19 pandemic (topos 1a) is widespread in the French corpus despite being almost absent in the Spanish corpus. The thought pattern that undocumented migrants are a threat because they disrupt peaceful social existence during the COVID-19 pandemic (topos 1b) is dominant in the Spanish corpus and is also measurable – albeit to a lesser extent – in the French corpus.

Concerning the claim that migrants who stay in a country irregularly lead to social de-stabilization (Conclusion 3), two arguments (Arguments 1a and 3b) can be detected that stand in a comparative semantic relationship with the above-mentioned conclusion. In both cases, migrants are compared with nationals, are represented as being privileged over nationals, and are thus treated unjustly by the executive power of the country either by having more rights (argument 1a) or by receiving financial aid that nationals are denied even though nationals are presupposed to be more entitled to receive this aid in the a maiore scheme (argument 1b).

Figure 3 reveals that the argument that migrants have more rights than nationals during the pandemic is by far the most characteristic argument of the comparative topos in the Spanish corpus, which stands in contrast to the French corpus, in which the argument of unequal financial treatment dominates.

All argumentation patterns create a dichotomization of society with the claim that migrants are a threat to nationals on a health, social, and economic level. This Manichean worldview becomes even more evident with the comparative schemes that directly oppose both the in-group and the out-group by contrasting the (alleged) benefits that migrants have with the disadvantages that alreadysuffering nationals have during the pandemic.

In nearly all cases, these comparison schemes based on dissimilarity metrics also have the effect of causing the ruling party to be blamed for (constructed) unfair conditions.

Figure 3: Frequency distribution of Arguments 3a and 3b.

Next, the arguments are discussed in greater depth and illustrated with examples.

#### 8.1 Causal Argument: ("Illegal") Migrants bring the Pandemic

Before bacterial revolution began to change our understanding of disease at the end of 19th century, people were forced to find their own explanations for the emergence of epidemics. In one of the earliest epidemic discourses – namely Thucydides' description of the Plague of Athens in 430 BC (cf. Dinges 2004, 82) – the epidemic is depicted as coming from outside, from a foreign land (cf. ibid., 82), which correlates with an initial representation of the hometown as a peaceful and sheltered space until foreigners arrive with a threatening epidemic. Additionally, these foreigners are often described as ill-intentioned. Such pandemic discourses perceive of contagions as a type of social phenomenon that can lead to the stabilization of the ingroup by excluding and constructing others as the out-group (cf. ibid., 85). Isolation and control of the frontier – framed as a filter that separates the "healthy impure" from the "healthy pure" – is presented as the means of effectively stopping the pandemic. Le Pen and Rassemblement National thus use classical topoi from epidemic discourses that were commonplace before the modern, scientific understanding of epidemics came about. In so doing, they aim to provide evidence of the causal thought pattern by invoking either scientific experts or other, more successful countries (Wengeler 2021). Referring to scientists has become especially popular in the COVID-19 discourse among all parties of the political spectrum even though referring to an authority alone does not mean that the argument is true per se (see Wengeler 2021 for different quality criteria for judging arguments from authority). In only one case did a tweet integrate an article by Le Figaro in which a concrete epidemiologist is named in order to back the argument that "experts confirm" Le Pen's logic.


("@ljacobelli: 'One year ago, @MLP\_officiel called for border control. Today, both the government and the scientific committees recognize that we were right!'")

This topos is not representative of the Spanish corpus. The only evidence that we could find uses an inductive scheme (cf. Kienpointner 1992a: 365) that generalizes on the basis of one example that (all) migrants who stay in a country irregularly bring the pandemic.

(17) Un policía malagueño muere por Covid tras custodiar a inmigrantes ilegales, alguno de ellos contagiados, en Canarias. [VOX\_21\_01\_04] ("A police officer from Málaga dies from COVID after supervising illegal immigrants, some of them infected, in the Canary Islands.")

This topos shows how selective and uncomplex mental models can be by omitting the fact that the virus can be propagated by all human beings, including by the allegedly sacrosanct nationals. Although this topos is not representative of the corpus, it has been uttered in public speeches (cf. Olivas/Rama 2021, 4; Vega Macías 2021, 13).

#### 8.2 Comparative Topoi

Comparative schemes directly oppose both the in-group and the out-group by contrasting the benefits that migrants have with the disadvantages that "suffering" nationals have in times of the pandemic.

#### 8.2.1 ("Illegal") Migrants Have More Rights than Nationals during the COVID-19 Crisis

The Spanish corpus is dominated by the argument that migrants have more rights than nationals – namely that they are allowed to not follow quarantine rules while Spaniards must obey and suffer from restrictions to their freedoms (see Examples 18–19).

(18) Para este gobierno criminal los ilegales están siempre antes que los españoles. [AB\_21\_02\_10] ("For this criminal government, illegals always come before Spaniards.")

(19) ¡Para tener menos derechos que nadie en España tienes que ser español! Si eres un ilegal o un turista podrás circular por todo el país sin atender a cierres perimetrales. [AB\_21\_03\_20] ("In order to have less rights than anyone else in Spain, you have to be Spanish! If you are an illegal or a tourist, you will be able to travel throughout the entire country without having to respect lockdowns.")

The dichotomization of social space between allegedly favored migrants and allegedly unjustly treated nationals is reinforced by grammatical and semantic patterns that signal oppositeness, such as negation vs. non-negation (Examples 20 and 21) and semantic opposition (Example 22; for the opposite semantic roles, see Chapter 7.1).

(20) ✅ Si eres inmigrante ilegal te pagan hasta un hotel en Canarias. ❌ Si eres español y tienes unos días libres en Semana Santa, no puedes salir de tu perímetro.

("✅ If you are an illegal immigrant, you get a paid hotel room on the Canary Islands.

❌ If you are a Spaniard and have some free time during Holy Week, you cannot leave your surroundings.")


The argument that migrants have more rights than Spaniards during the pandemic, however, has been proven to be "fake news" by the independent agency NEWTRAL (cf. García 2021). In nearly half of the tweets on this topos, VOX and Abascal try to provide evidence of this invented "data" via an argument from authority, mainly by linking or re-tweeting journal articles. However, none of these articles or re-tweets focuses on or even mentions migrants. For example, in three journal articles, some French tourists are reported to have circumvented COVID-19 rules (cf. AB\_21\_03\_20; VOX\_21\_03\_20; VOX\_21\_03\_21). Other articles report an upcoming lockdown during Holy Week (again, without any mention of migrants) (cf. VOX\_21\_03\_24) or declare that Muslims are allowed to use the Palos de la Frontera bull arena for Muslim rituals during the lockdown period (cf. VOX\_21\_05\_14). Concerning the latter example, it is VOX that creates the opposition between "Muslims" and "Spaniards" while entirely omitting the fact that Muslims can, of course, also be Spaniards. Another article speaks of the formation of "little, violent" protests after the imprisonment of rapper Pablo Hásel in Barcelona, whereas VOX changes the wording to "bandas organizadas de terrorismo callejero" ("organized terroristic street gangs") in order to dramatize the situation (AB\_21\_02\_21). Furthermore, it is Abascal who adds that these gangs recruit "illegal migrants" even though the article does not refer to migrants at all.

The French corpus uses this topos to a far lesser extent and far more generally but follows the same dichotomic logic: Either innocent French people face harsh regulations during the pandemic while migrants who enter the country "illegally" are left unimpeded (see 27), or unvaccinated French people are treated harshly while "illegal migrants" are treated with complacency (see 28). Videos are linked in which the spokesperson of the party – Sébastien Chenu – is interviewed on a TV channel. Even from the videos, it is not clear which examples he uses to inductively generalize this topos.


#### 8.2.2 ("Illegal") Migrants Receive Money that Nationals are Denied

The comparative pattern that dominates in the French corpus focuses on unequal treatment on the economical level. Le Pen's logic lies in opposing austerity measures that concern public hospitals and thus also in the money denied to French people but that is provided to migrating people. This is a typical a maiore scheme: If even nationals do not receive any money and migrants are less entitled to receive money but do receive it, it is morally reprehensible/a greater threat to peaceful social coexistence when money is provided to migrants:


### 9 Conclusion

Using Foucault's conception of discourse as a constructive practice of knowledge systems, the aim of the present paper was to uncover and systematize the implicit and explicit thought patterns found in French and Spanish right-wing populistic discourse that relate the phenomenon of the COVID-19 pandemic to the phenomenon of migration and thereby construct a healthcare crisis. Implicit thought patterns in the form of framing strategies mask the controversial and subjective character of the mental models involved, whereas explicit thought patterns can be identified as controversial, plausibility-seeking argumentation schemes via topos analysis. Linguistic means of expression are more relevant in the case of framing than in the more content-oriented topos. In this context, we aimed to see whether and how the concept of "the dangerous other" is used as a scapegoat to cast the healthcare crisis as a migration problem. The compiled corpus comprised all tweets posted during the 2021 pandemic year by the French Rassemblement National and the Spanish VOX parties as well as by their respective leaders, Marine Le Pen and Santiago Abascal. Generally, due to the total number of tweets concerning migration that were posted in 2021, the proportion of tweets that related the pandemic to migration was relatively small (i.e., under 10% in all corpora).

All corpora follow populistic discursive logic by dichotomizing social space. However, despite undeniable parallels between both discursive styles (e.g., framing migrants as "illegals" or presenting a highly subjective worldview as uncontroversially factual via presupposition), the qualitative assessment of the data revealed interesting differences between both right-wing populistic discourses, thereby confirming Kienpointner's (2005, 218) finding that "there is no unique discourse of right-wing populism."

The Spanish discursive style is more radical on the lexical-semantic level and applies manipulative framing strategies to a greater extent than the French populistic style, for example, by directly referring to migrating people and framing them as criminals via co-occurrences and by metaphorically displaying the phenomenon of migration and the action of migrating as war and warlike actions. The argumentation schemes found in the Spanish corpus are dominated by the topos of establishing a causal relationship between the migration threat and the de-stabilization of society during the pandemic by opposing Spaniards and migrants with regard to their rights. In this comparative topos, VOX and Abascal present the alleged "data" that immigrating people have more rights than the Spanish population. This opposition is reinforced by the distribution of opposite semantic roles (i.e., migrants are presented as freely moving actors vs. Spaniards as passive patients to whom restrictions apply), by the use of opposite semantic concepts, and by the use of binary semantic-syntactical structures of negation vs. non-negation of positively connoted actions. However, the argument that migrants allegedly have more rights was proven to be "fake news" despite attempts to make it factual via the use of authoritative schemes that involved linking or retweeting news articles.

The dichotomization realized by Rassemblement National and Le Pen is less aggressive and less graspable on the level of the linguistic means of expressions that it uses but is measurable on the level of content-related topoi, which is likely related to the party's strategy of "de-demonization" (dédiabolisation) (in contrast to the radicalization of VOX). The predominant causal topos of the French corpus corresponds to a classical pre-scientific thought pattern that identifies migrants as the cause of diseases and epidemics. This topos is closely connected to the keyword frontière ("frontier"), which is framed as a symbolic filter that separates the healthy and morally pure from the unhealthy and morally impure. Concerning the comparative scheme of the French corpus, economic elements are highlighted in this corpus in order to place migrants in opposition to the French. In a nutshell, migrants receive money, whereas the French – who, following this logic, would be more entitled to receive it – are left without anything.

As we have seen, implicit and explicit thought patterns should be analyzed in combination in order to uncover polarizing mentalities in right-wing populistic discourse.

### References

Alduy, Cécile/Wahnich, Stéphane, Marine Le Pen prise aux mots, Paris, Seuil, 2015.

Böke, Karin/Jung, Matthias/Niehr, Thomas/Wengeler, Martin, Vergleichende Diskurslinguistik.


Chazel, Laura, French populism in times of COVID-19, Populism interventions 7 (2020), 21–23.

Dargiewicz, Anna, Emotionale Wirkung(en) von Medienframes, Linguistische Treffen in Wrocław 19 (2021), 37–52.

Dinges, Martin, Bedrohliche Fremdkörper in der Medizingeschichte, in: Mayer, Ruth/Weingart, Brigitte (edd.), Virus! Mutationen einer Metapher, Bielefeld, transcript, 2004, 79–95.

Eklundh, Emmy, Spain, Populism interventions 7 (2020), 40–42.

Entman, Robert M., Framing: Toward Clarification of a Fractured Paradigm, Journal of Communication 43:4 (1993), 51–58.

Fairclough, Norman, Language and Power, London, Routledge, 2015.

Fairclough, Norman, Manipulation, in: Asher, Ronald E. (ed.), An encyclopedia of language and linguistics, vol. 5, Oxford, Pergamon, 1994, 2360–2361.

Fillmore, Charles, Frames And The Semantics Of Understanding, Quaderni di semantica VI:2 (1985), 222–254.

Foucault, Michel, Archäologie des Wissens, Frankfurt/Main, Suhrkamp, 1981.

Foucault, Michel, Wahnsinn und Gesellschaft: eine Geschichte des Wahns im Zeitalter der Vernunft, Frankfurt/Main, Suhrkamp, 1969.

Gallo, Valentina/Chiodini, Paolo/Bruzzese, Dario/Kondilis, Elias/Howdon, Dan/Mierau, Jochen/Bhopal, Raj, Comparing the COVID-19 pandemic in space and over time in Europe, using numbers of deaths, crude rates and adjusted mortality trend ratios, Scientific Report 11 (2021), https://doi.org/10.1038/ s41598-021-95658-4.

García, Yolanda, Los cierres perimetrales en los desplazamientos entre CCAA afectan a todos, sean nacionales o extranjeros, 2021, https://www.newtral.es/fact-check-abascal-cierres-perimetralesnacionales-extranjeros/20210324/ [last access: July 6th, 2022].

Gauthier, Gilles, La métaphore guerrière dans la communication politique, Recherches en communication 1 (1994), 131–146.

Gern, Klaus-Jürgen/Hauber, Philipp, Corona-Effekte im internationalen Vergleich, Wirtschaftsdienst 100 (2020), 899–900.

González Enríquez, Carmen/Rinken, Sebastian, Spanish public opinion on immigration and the effect of VOX, ARI 46 (2021), 1–9.

Grewendorf, Günther/Hamm, Fritz/Sternefeld, Wolfgang, Sprachliches Wissen: eine Einführung in modernde Theorien der grammatischen Bedeutung, Frankfurt/Main, Suhrkamp, 1989.

Griebel, Tim/Vollmann, Erik, We can('t) do this. A corpus-assisted critical discourse analysis of migration in Germany, Journal of Language and Politics 18:5 (2019), 671–697.


Changeprozesse, Bonn, Dissertation, 2021, https://hdl.handle.net/20.500.11811/8976.


Pietrini, Daniela (ed.), Il discorso sulle migrazioni/Der Migrationsdiskurs, Frankfurt/Main, Lang, 2020.


#### 264 Svenja Dufferain-Ottmann

Roitman, Malin (edd.), Political Discourses at the Extremes. Expressions of Populism in Romance-Speaking Countries, Stockholm, Stockholm University Press, 2019, 1–11.

Tereick, Jana, Klimawandel im Diskurs, Berlin/Boston, De Gruyter, 2016.


### Index

anti-immigration 194–196, 205, 209, 212, 214–215, 237, 241 anti-racist 90–91, 94, 96 app 85 bilingual 9, 13–14, 28, 36, 71 bilingualism 71 chat 6, 11–12, 14–15, 17, 24–26, 85, 136, 147, 169–170, 172–173, 177, 179 citizenship 9, 32–35, 42, 46–47, 206, 208 code-switching 146 COLEM 142–144, 146–165 communicative repertoire 16 community of practice 52–54, 56–57, 142, 174 computer mediated communication (CMC) 4, 54, 57, 170, 172–174, 182–183 co-occurrences 244, 247, 260 Critical Discourse Analysis (CDA) 190, 194–196, 199, 225, 240, 251 Critical Linguistic Awareness 15 descriptive discourse analysis 240, 251 diaspora 6, 53, 163, 170, 172, 182 digital language practices 2 ethic 36, 87 ethical 2–4, 96, 129 ethnography 9, 14–15, 31, 35–36, 46–47 Facebook 1, 83, 85, 88–89, 100, 106–108, 115, 143, 149, 156, 160, 173 geolocated data 129, 132 geolocation 130, 136 globalization 6, 52, 141, 157 glottopolitical interventions 79–80, 86–87, 95 glottopolitics 80 go-along interviews 9, 37 google maps 130 Google Translator 85

hashtag 196, 200, 203, 211, 227, 231 hate speech 5, 193, 212, 214, 219, 224–225, 231, 241 indexicality 53, 60–62, 74 Information and Communication Technology (ICT) 9, 31–40, 42, 45–47, 80, 82, 85, 95 Instagram 1, 83, 85, 88–89, 143 interview 9, 37, 40, 53, 55, 64–66, 80, 83–84, 91, 93, 143, 153, 258 interviewer 147, 150 keywords 194–196, 199, 203, 214, 244–245, 247, 260 language regime 9, 33, 46 language teaching 79, 86–87 language teaching apps 85 linguistic biographies 15–16, 64 linguistic landscape 14–15, 52 linguistic repertoire 15, 51–52, 56, 64, 74, 82, 86, 142, 146 linguistic variation 52–53, 55–58, 61–62, 69, 71, 74, 130, 132, 136, 170 metaphor 5, 190, 201, 207, 214, 244, 248–250, 260 metonymy 207 mobility 1, 9, 15, 32, 46, 51, 81–82, 141, 143, 177 monoglossic language ideology 28 monolingual 9, 27–28, 33–34, 42, 45–47, 81, 166 monolingualism 9, 36 multilingual repertoire 9, 11, 43 neologism 99–100, 104, 106, 109, 111, 117, 119, 121, 125 Netflix 1 new speakers 9, 44

participative action research 12, 14, 17 participative ethnography 14, 27

non-citizenship 9, 47

256–260

translanguaging 6, 12, 15, 26–28

perception 6, 51–53, 55, 57–59, 61–63, 65, 72, 74, 92, 102, 147, 151, 191 plurilingual repertoire 12, 26–27 positioning 99, 107, 119, 121, 123–125, 172, 174 presuppositions 58, 215, 244, 248, 250–251, 254, 260 racialisation 190–191, 193–196, 199, 205 racism 154, 191, 212, 215, 227, 229 scales 28 semi-structured interview 64 semi-directed interviews 143 Skype 143, 147, 149, 153, 159–161 sociolinguistic scales 15 Spanish-speaking world 1, 3, 57, 73, 75, 110 speaker design 169, 171–172, 182 speakerhood 9, 11–12, 14, 24–25, 27–28, 47 supervernacular 52–53, 55, 57, 59, 65–66, 68, 74–75 Telegram 108, 143 topoi 235, 251, 253, 255–256, 260 toponymy 15 topos 5, 235–236, 240, 243–244, 247, 251–254,

translation apps 6 translinguistic repertoire 9, 27, 47 transnational 6, 9, 32–35, 39, 46–47, 53–54, 65, 80–84, 95, 151, 166, 229 Twitter 1, 99–100, 103–104, 107–110, 119, 123, 129–131, 133, 135–136, 191, 194–195, 202, 209, 219–221, 223–225, 229, 231–232,

Translate Facebook 85

235, 243


Zoom 85, 93