Reshaping epistemic identity on X/Twitter: A Spanish-English contrast

This is a pre-print of the following chapter: Manuel Padilla Cruz, “Reshaping epistemic identity on Twitter: A Spanish-English contrast”, published in Evaluating Identities Online. Case-studies from the Spanish-speaking World, edited by Pilar Garcés-Conejos Blitvich and Patricia Bou-Franch, 2024, Palgrave Macmillan, reproduced with permission of Palgrave Macmillan. The final authenticated version is available online at: https://doi.org/10.1007/978-3-031-62320-2_3. Reshaping epistemic identity on Twitter: A Spanish-English contrast1 Manuel Padilla Cruz Universidad de Sevilla Abstract This chapter reports on a cross-linguistic micro-analysis of how Spanish- and Englishspeaking Twitter users attempt to remould the epistemic identity of two allegedly reliable and trustworthy institutions: namely, the Spanish Ministry of Health and the World Health Organisation. Relying on two datasets in these languages, the analysis looks into the verbal actions that both language-groups of users perform with a view to casting doubts on the well-intendedness and believability of the recommendations that the two institutions made during the COVID-19 pandemic concerning vaccination, the use of facemasks and other sanitary measures. In addition to evidencing different preferences in terms of the actions accomplished in order to challenge the two institutions and the information that they imparted about such issues, the analysis reveals how the two groups of users intersubjectively enact a new identity for the two institutions as supposed liars and criminals, while they claim for themselves an identity of rebellious and nongullible 1 This research has been possible thanks to the project “La digitalización de la interacción humana y su impacto social: Consecuencias, retos y concienciación de uso” (UPO-1380703), which has been funded by the European Regional Development Fund (ERDF) and by the Ministry of Economy, Knowledge, Enterprise and University (Regional Government of Andalusia) within the framework of the ERDF regional operative programme Andalusia 2014-2020. resistance citizens. Finally, the analysis also delves into the consequences of such actions: pathologisation, dehumanisation, silencing and gaslighting of the two institutions. Keywords: epistemic identity, pathologisation, dehumanisation, silencing, gaslighting 1. Introduction Identity is a fluid, often ephemeral, personal, relational and group attribute that is enacted over time (Garcés-Conejos Blitvich & Georgakopoulou, 2021; Gee, 2000; Ochs, 1992). It emerges through discourse and participation in social practices (Foucault, 1972), so it can be intersubjectively “shaped from moment to moment” (Bucholtz & Hall, 2005, p. 591). Yet, identity may also be more enduring and fossilise or sediment when interacting with certain collectives, on particular occasions and/or in specific contexts (Bucholtz, 2007, pp. 379, 395). Indeed, it is a cognitive construct that hinges on individual and collective beliefs concerning different aspects of self and others (Goffman, 1959). Although such beliefs may stabilise, the assumptions that are made manifest during interaction can refine them (Sperber, 1996; Sperber & Wilson, 1995).2 Among such beliefs feature those regarding people’s own and other individuals’ epistemic capabilities, knowledgeability and reliability about certain issues, expertise in some domain and informational skills. These make up what can be labelled their epistemic identity, which in the philosophical field of social epistemology is known as epistemic personhood (Borgwald, 2012; Thorson & Baker, 2019). In the same way that people are acknowledged or deprived of epistemic personhood, groups and institutions can similarly be granted or denied one during epistemic practices. 2 Making some state of affairs manifest involves causing a mental representation thereof. Texts count among the tools whereby beliefs spread (Gee, 2014; Jenkins & Potter, 2013; Jones, Chik & Hafner, 2015; Wodak, 2006). Regardless of their nature and the medium wherein they are generated, texts are prompted by, and hence reveal, ideologies: doxastic constellations about what is right or wrong, good or bad, fair or unfair, reasonable or absurd, etc. (Jones, 2019, p. 11). Texts model reality and determine how it is perceived and approached. Indeed, ideologies shape, refine, reinforce, remodel, challenge, attack or destroy objects, places, other ideas, practices and, obviously, people and their identities. Therefore, in addition to evincing and propagating beliefs, texts are most powerful tools to impact how other people and their viewpoints are depicted and perceived. Technology-mediated environments, and, more specifically, digital or electronic multi-blogging platforms without seeming topic restrictions and limitations (Graham & Hardaker, 2017; Thurlow & Mroczek, 2011; Sältzer, 2022), have become venues wherein doxastic universes underpinning ideologies and identities are exposed, fostered or disputed through varied practices (Jones, Chik & Hafner, 2015). Their affordances enable users to react to material posted by other users through intertwined individual discourse postings (Graham & Hardaker, 2017; Thurlow & Mroczek, 2011). These can accomplish a variety of actions, while partaking of, and adding to, broader Discourses (sic, capitalised), or “distinctive ways of speaking and […] writing […]” that “are coupled with distinctive ways of acting, interacting, valuing, feeling, […] thinking, and believing” (Gee, 2013, p. 2). Such ways of speaking and writing, in turn, “are coupled with ways of coordinating oneself with […] other people, and with various objects, tools, and technologies” (Gee, 2013, p. 2; also see 2014). Thus, users can position themselves in relation to certain issues, while they interactively present, mould, strengthen, redefine and perhaps contest identities and ideologies. In fact, “blogs […] are representations of identity which tend to chronicle events or opinions from the perspective of the blogger(s)” (Graham & Hardaker, 2017, p. 798). Metaphorically speaking, blog postings are sometimes the brushes to perfunctorily sketch or intricately portray identities and ideologies on a digital canvas; the brush-strokes putting and mixing paint are the actions that users perform and the words that they use. This chapter sets the spotlight on these brushes and brush-strokes, and on the paint that they spread, as it undertakes an examination of digital material which attains a specific goal: reshaping epistemic identity. More specifically, this chapter seeks to analyse the verbal actions that users of the microblogging platform Twitter accomplish in order to challenge not only the well-foundedness and believability of information about particular issues, such as vaccination, the use of facemasks and other sanitary measures during the COVID-19 pandemic, but also to question the reliability and trustworthiness of the informers dispensing it, and, therefore, reshape their epistemic identity. Indeed, as Garcés-Conejos Blitvich and Georgakopoulou (2021, p. 304) argue, stances, roles and identities are connected with the accomplishment of specific speech acts. Accordingly, this examination involves a cross-linguistic micro-analysis that compares datasets in two different languages: Spanish and English. Thus, it aims to ascertain whether each language group exhibits differences in terms of how they intersubjectively depict identities and ideologies, while they jointly smear, smudge or blur the epistemic identity of two information sources. Lastly, this examination considers the aftertaste of the painting that tweeters create, as it also looks into the possible effects that may be brought about by their remoulding of epistemic identity. This chapter continues by offering an overview of identity, portraying epistemic personhood and explaining how it may be challenged. After detailing the research questions informing the study, the chapter describes the data and methods. Next, it presents and discusses the results, after which it makes some concluding remarks. 2. Identity and epistemic identity Identity has been approached from multiple disciplines (see Garcés-Conejos Blitvich & Georgakopoulou, 2021; Spencer-Oatey, 2008). Traditionally, it has been characterised as a self-concept, self-definition or self-interpretation comprising beliefs about distinct personal attributes, such as personality traits, abilities, psychological characteristics, behavioural tendencies, ideologies, social roles, linguistic affiliation or group membership (Goffman, 1959; Haugh, 2009). Such beliefs are intuitive when individuals are not aware of their formation or acquisition. In contrast, they are reflective if they originate through interaction with peers or participation in some social institutions. In the latter case, beliefs include a validating context referring to the authority, test or procedure demonstrating the possession of a particular attribute, and a credal attitude specifying the degree of confidence in them (Sperber, 1997). However, in social theory, social psychology, communication theory sociolinguistics, conversation analysis and discourse analysis, identity is seen as contingent, multiple and malleable, as individuals actively produce and negotiate it (Holland et al., 1998, p. 270; Merchant, 2006, p. 238). Although it may be anchored and refer to features over which individuals have little control and about which there might be some sense of continued sameness, it usually is transient and alludes to unstable, negotiable personal aspects that shift over time and contexts due to maturation, changing cultural conditions, group affiliation and interaction with different people. Indeed, interaction in diverse circumstances causes certain components of identity to become salient to particular individuals (Merchant, 2006, p. 239). Additionally, identity may be personal when it amounts to a self-definition as someone unique, social when such a definition involves adscription to some group, and interpersonal when it is based on connections with other individuals (Brewer & Gardner, 1996). Although people create relatively stable self-representations and do not continuously perform completely novel identities (Joseph, 2014), they also refine such representations through interaction, wherein they constantly enact and negotiate distinct components of their personal, relational and interpersonal selves (Simon, 2004). Hence, identity is always ongoing and transforming, and is intersubjectively constructed (Burke & Stets, 2009; Garcés-Conejos Blitvich & Georgakopoulou, 2021). It cannot be conceived of as a stable individual psychological construct, but as a relational and sociocultural one arising through mutual positioning of self and other (Bucholtz & Hall, 2005; de Fina, Schiffrin & Bamberg, 2006). As Garcés-Conejos Blitvich and Sifianou (2017, p. 236) put it, identity “cannot be static or fixed, but must be in constant transformation”. Owing to its relational nature, identity depends on the existence of others, interaction and self-concept construal. Therefore, identity is acquired and delineated in discursive interaction (Bucholtz & Hall, 2005, pp. 585-586). According to Ochs (1992), it is claimed, built up and challenged by performing actions, displaying epistemic stances –e.g., (dis)belief, (un)certainty, (dis)approval, etc.– and exhibiting emotional stances –e.g., satisfaction, happiness, disappointment, anger, wrath, etc. Moreover, depending on circumstances, identities may be momentarily adopted and then abandoned. They may even give rise to ideological connections with other broader or narrower identities, which may in turn affect perceptions about agency and actional style, though not deterministically (Bucholtz & Hall, 2005, p. 591). Therefore, identity can be said to have the following characteristics (Bucholtz & Hall, 2005, pp. 591-605; Garcés-Conejos Blitvich & Georgakopoulou, 2021, p. 294): a) Emergence, because it varies as discourse unfolds and interaction takes place in different contexts. b) Positionality, since it depends on the roles and orientations that individuals adopt, which contribute to the formation of (inter)subjectivity. c) Indexicality, given that its delineation benefits from actions like taking stance, social labelling, triggering presuppositions or implications, evaluating, using uncivil and impolite language (Evans, 2006; Garcés-Conejos Blitvich, 2009; Garcés-Conejos Blitvich & Geogakopoulou, 2021; Garcés-Conejos Blitvich & Sifianou, 2017), and deploying a wide array of linguistic devices that may be “ideologically associated with specific personas and groups” (Bucholtz & Hall, 2005, p. 594). d) Relationality, as it is co-constructed by establishing complementary relations: similarity/difference, genuineness/falsehood, legitimacy/illegitimacy, etc. e) Partialness, because it results from estimates about others’ perceptions and representations that always are ideologically sieved. f) Consequentiality, as it allows, incites or dissuades individuals to perform certain actions. Identity subsumes beliefs concerning attributes such as knowledgeability or expertise in some issue, discipline or field, and informational skills. They contribute to epistemic personhood or identity: people’s view about their role as knowers. Epistemic personhood involves estimates about personal abilities to think autonomously and generate knowledge, as well as to make judgements about information (Borgwald, 2012). Additionally, it involves being able to spread knowledge and being perceived as a good source of information or knowledge (Thorson & Baker, 2019). Epistemic personhood is also intersubjectively and discursively shaped, evidenced, consolidated or fostered in epistemic practices involving the impartation of information. If the information that a knower dispenses is considered sufficient, relevant, timely, helpful, clear, concise and well-ordered, the knower may be considered a competent informer. Most importantly, if it is found sound and believable, the informer may furthermore be deemed reliable and trustworthy. Hence, epistemic personhood also rests on the development, display and acknowledgement of an essential virtue: epistemic trustworthiness. This does not only depend on the ability to produce veridical information by accessing, examining, contrasting and integrating the adequate and needed sources and evidences, but also on a decided commitment to imparting genuine information and a demonstrated responsibility for its credibility (Fricker, 2007; Origgi, 2013). Yet, despite the ability and willingness to provide reliable and truthful information, informers may be discredited due to ideological biases. Repeated participation in epistemic practices may demonstrate informers’ propensity to dispense authentic and believable information, thus moulding a reputation as knowledgeable and credible individuals. This reputation may benefit from, and be strengthened by, the properties of the dispensed information (Padilla Cruz, 2019a), a variety of verbal signals –precise vocabulary, adequate syntax, etc.– nonverbal signals unveiling confidence and certainty in face-to-face communication –retained eye contact and/or avoidance of fidgeting, hesitation or stuttering (Guerrero & Floyd, 2006)– or impressions and emotions like affect, affinity or contentment (Origgi, 2013). If informers manage to thus model such a reputation, they may achieve epistemic trust. However, informers may fail to achieve that reputation by dispensing insufficient, illfounded, irrelevant, untimely, inconvenient or unhelpful information; producing unfortunate nonverbal behaviours, or generating undesired impressions or emotions. Moreover, despite informers’ efforts –and much to their chagrin– an audience may refuse to acknowledge that reputation for them if the audience considered that the dispensed information did not meet their own expectations about quantity, quality, relevance, timeliness, convenience or helpfulness, and/or the informers’ nonverbal behaviour was felt strange or inadequate, or if they experienced negative feelings. In particular, failing to meet expectations about trustworthiness may suggest the unwarrantedness, unsoundness, illogicality and questionability of the beliefs underpinning information. Consequently, an audience may deprive an informer of epistemic trust, thus affecting their epistemic personhood negatively and co-constructing it differently. Epistemic personhood may be questioned through tacit or overt, verbal reactions. Unspoken reactions include doubts pertaining to the information properties, mistrust, non-acceptance and even wrong assessments of informers as knowers because of some ideological bias or prejudice. If an informer was considered to release unreliable, illfounded or uninteresting information because of lack of access to genuine sources, they would be inflicted a testimonial injustice and their information would be distrusted or ignored (Fricker, 2007). Perceptible reactions include ignoring the informer. Despite its likely ambiguity, a lack of response may be interpreted as an indication that the information is unworthy of attention because of its quality. Moreover, perceptible reactions may also involve verbal actions: - Disagreements and/or negative comments asserting or implying the infelicities and/or unbelievability of information, and/or the incompetence, untrustworthiness or c. insulting ‘[INS]’, d. imprecating ‘[IMP]’, e. exhorting the informers to take some course of action ‘[EXH]’, and f. expressing dismissive incomprehension ‘[INC]’. Since none of these actions were realised in a number of moves, the following datadriven categories were considered: g. making an opening salutation ‘[SAL]’, h. addressing the informer ‘[ADD]’, i. requesting something from the informer ‘[REQ]’, j. threatening the informer ‘[THR]’, k. informing about the tweeter’s intentions or meaning ‘[INF]’, l. mitigating the illocutionary force of the tweet ‘[MIT]’, m. laughing at the informer ‘[LAU]’, and n. greeting or bidding farewell to the informer ‘[GRE]’. Lastly, moves were quantified and, in the case of the tweets consisting of more than one, move combinations were additionally identified, quantified and qualitatively analysed. 4. Results Of the Spanish tweets, 84 (59.57%) consisted of one move, 32 (22.69%) had two, 13 (9.21%) included three, six (4.25%) comprised four, three (2.12%) mobilised five and two (1.41%) integrated six. Below follows a segmented and labelled five-move tweet: (1) Iros a tomar por culo [EXH], estáis matando a la gente [DIS]. Estáis 17umpliendo la agenda 20-30 [DIS], pero se os va a acabar el chollo [IMP]. La gente está despertando [DIS]. (MED030) ‘Go fuck yourself, you’re killing people. You are fulfilling the 20-30 agenda, but you are going to run out of bargain. The people are waking up.’ Of the English tweets, one-move tweets diminished to 51 (34.22%), two-move tweets were practically similar in number, 33 (22.14%), while that of three-, four-, five- and sixmove tweets increased to 22 (14.76%), 21 (14.09%), 12 (8.05%) and four (2.68%), respectively. Moreover, the number of moves went up to seven in two tweets (1.34%) and to eight in three (2.01%). Quite surprisingly, one tweet (0.67%) even utilised 40 moves. Below is a segmented and labelled six-move tweet: (2) Hello [GRE] I’ve been seeing everywhere news in social media platforms all about ur warnings and bullshit [INF], can u stop making People’s scared!? [REQ] Nothing will happen, if u able them solve this [DIS] don’t just sit at ur chair and wrote shit making People’s scared [EXH], it’s not fun [DIS]. (EMED071)3 Fig. 1 exhibits the distribution of tweets in each dataset depending on move number: 3 Typos, mistakes, punctuation infelicities and other idiosyncrasies have been preserved. Fig. 1: Distribution of tweets by move number. What follows provides the results regarding the actions questioning epistemic personhood and their assemblage in the tweets, thus responding to RQ#1. Answering RQ#2 requires comparing and analysing the datasets qualitatively, which is undertaken in Section 5. 4.1. Questioning epistemic personhood In the Spanish dataset, the trustworthiness of the Ministry of Health and the information that they dispensed about vaccination, facemasks and other sanitary measures was questioned through 237 actions. Insults (example 3) ranked the highest with 103 tokens (43.45%) and disagreements (example 4) followed with 69 examples (29.11%): (3) Corruptos tecnócratas promotores de iatrogenia y disgenesia. (MED004) ‘Corrupt technocrats promoters of iatrogenia and dysgenesis.’ (4) En Europa el virus no mata, aquí sí. (MED047) 0 10 20 30 40 50 60 70 80 90 Spanish dataset English dataset 1 move 2 moves 3 moves 4 moves 5 moves 6 moves 7 moves 8 moves 40 moves In Europe the virus does not kill, here it does. There were 36 (15.18%) exhortations, as in example (5), wherein the Ministry was encouraged to give the booster dose to someone else as a way of showing distrust about the vaccine effectiveness, and 22 (9.28%) imprecations, as in example (6), which warned the health authorities that they could be held accountable for the damage they were supposed to be doing: (5) La #DosisDeRecuerdo que se la ponga vuestra puta madre. (MED001)4 ‘Give the #BoosterDose to your fucking mother.’ (6) Ya os llegará la hora. (MED006) ‘The time will come for you.’ In contrast, the number of cases of dismissive incomprehension decreased to six (2.53%), as in example (7), wherein a tweeter wondered whether the Ministry actually read their own postings. As for complaints, there was only one case (0.42%), example (8), which expressed anger at having to wear facemasks: (7) Pero vosotros leéis lo que ponéis, sinvergüenzas? (MED059) ‘But do you read what you post, scoundrels?’ (8) Puta que vos parió ya con el puto trapo infecto de mierda. (MED104) ‘Whore who gave birth to you, with the fucking rag infected with shit.’ In the English dataset, the actions casting doubts on the believability of the WHO and related information nearly doubled, as they totalled 429. The preferred action was disagreeing (example 9), with 173 cases (40.32%), followed by insulting (example 10), with 122 tokens (28.43%): (9) I’M NOT GOING TO DO ANY OF IT YOU EVIL (EMED016) (10) Criminal hoodlums. (EMED033) 4 The Spanish campaign played with the ambiguity of ‘dosis de recuerdo’ (‘booster/memory dose’). Exhortations, as in example (11), which urged the WHO to do something that was deemed consequential with their fatal policies, more than doubled the Spanish cases: 77 (17.94%). In turn, tokens of dismissive incomprehension, as in example (12), wherein the user wondered whether the poster actually believed the dispensed information, exceeded triple of the Spanish ones: 20 (4.66%): (11) Fuck your new world order. (EMED025) (12) Does this person posting the crap actually believe it...?? (EMED001) However, imprecations (example 13), which wished extinction to the WHO, were slightly fewer: 17 (3.96%). Next came requests for information or action, with eleven (2.55%) cases (example 14), which ironically asked the WHO to issue more useless recommendations as a way of questioning the helpfulness and timeliness of their previous ones: (13) You deserve extinction. (EMED039) (14) Please provide even more impractical and possible recommendations. (EMED080) Only two examples of opening salutations, clarifying intent and final greetings were spotted (0.46% each), and one of addressing the WHO, laughing at them, threatening them and mitigating illocutionary force (0.23% each). Although they conveyed ironic and haughty overtones contributing to incivility and identity questioning (Anderson & Huntington, 2017; Williams, 2021), they did not challenge epistemic identity per se, so examples are not given here. Finally, no example of complaint was found. Fig. 2 compares the actions that were accomplished in each dataset: Fig. 2: Actions performed in the Spanish and English datasets. 4.2. Move patterns In the Spanish one-move postings, the Ministry of Health was insulted in 52 tweets (52/84, 61.90%), while 14 tweets (16.66%) urged them to do something disadvantageous, 12 (14.28%) imprecated them, five (5.95%) contradicted them, and one (1.19%) complained about information. Out of the two-move tweets, the most recurrent move patterns combined insults with disagreement (15/32, 46.87%), with exhortations (6/32, 18.75%) and with imprecations (4/32, 12.5%), and exhortations with disagreements (3/32, 9.37%). An example of the disagreement-insult combination is the tweet in (15), where a tweeter wondered why the health authorities were not ashamed of themselves for what they were allegedly doing and were called scoundrels:5 (15) No sé cómo aún no se os cae la cara. Sinvergüenzas. (MED027: [DIS]+[INS]) 5 Space limitations prevent exemplifying all the observed move patterns. 0 20 40 60 80 100 120 140 160 180 200 Spanish dataset English dataset SAL ADD INS DIS EXH IMP INC REQ THR INF COM MIT LAU GRE ‘I don’t know how your face hasn’t fallen off yet. Scoundrels.’ High-frequency patterns in three-move tweets involved placing two disagreements along with an insult (4/13, 30.76%). An example is (16), in which a tweeter contradicted the official view on facemasks and also called the health authorities scoundrels: (16) Mascarilla no protege de los virus. Mascarilla PERJUDICA GRAVEMENT LA SALUD. MINISTERIO DE SINVERGÜENZAS. (MED106: [DIS]+[DIS]+[INS]) ‘Mask does not protect against viruses. Mask SERIOUSLY HARMS HEALTH. MINISTRY OF SCOUNDRELS.’ However, the patterns in four-, five- and six-move tweets only were occasional. Table 1 gathers the high-frequency move patterns in the Spanish dataset: Number of moves Move arrangement Number of tweets 1 [INS] 52 (61.90%) [EXH] 14 (16.66%) [IMP] 12 (14.28%) [DIS] 5 (5.95%) 2 [INS]+[DIS] 15 (46.87%) [INS]+[EXH] 6 (18.75%) [INS]+[IMP] 4 (12.5%) [EXH] + [DIS] 3 (9.37%) 3 [INS]+[DIS]+[DIS] 4 (30.76%) Table 1: Move patterns in the Spanish dataset. In the English dataset, 15 (15/51, 29.41%) of the one-move tweets consisted of insults, another 15 expressed dissent, 11 (21.56%) voiced dismissive incomprehension and 10 (19.60%) exhorted the WHO. In the two-move tweets, high-frequency move patterns coupled insults with disagreement or exhortations (6/33 each, 18.18%), combined exhortations with dissent (4/33, 12.12%), reiterated disagreement (4/33) or repeated exhortations (3/33, 9.09%). Enchained dissenting opinions are illustrated by example (17), where a tweeter referred to the absurdity and supposed ultimate purpose of the recommendations issued by the WHO, and insinuated that they would only be trusted by ignorant or naïve citizens: (17) Pure crap to try to dominate humanity by scaring it. Fortunately there are some smart people out there ! (EMED066: [DIS]+[DIS]) The only recurrent pattern in three-move tweets consisted of an exhortation and repeated disagreement (3/22, 13.63%). (18) is an example, as a tweeter voiced an invective against vaccines and questioned their usefulness: (18) FUCK your vaccines they are useless corona is BIGGEST LOE IN HUMAN HISTORY. (EMED125: [EXH]+[DIS]+[DIS]) Of the recurrent patterns in the four-move tweets, three (3/21, 14.28%) involved three dissenting opinions followed by an insult. Relatedly, in another three tweets such opinions were followed by an exhortation. The latter pattern is exemplified by (19), where a tweeter cast doubts on the helpfulness of vaccines and the timeliness of vaccination, suggested that the WHO was magnifying everything, and urged them to take pernicious course of action: (19) Not these ones. Not these days. You’ve way overplayed your hand and turned a lot into anti-vaxxers. So go fk yourselves forever. (EMED017: [DIS]+[DIS]+[DIS]+[EXH]) As for the 21 tweets containing five or more moves, their patterns were fairly heterogeneous. The only 40-move tweet was clearly an exceptional case, as it repeated the same insult forty times. The most recurrent move patterns in the English dataset are presented in Table 2: Number of moves Move arrangement Number of tweets 1 [INS] 15 (29.41%) [DIS] 15 [INC] 11 (21.56%) [EXH] 10 (19.60%) 2 [INS]+[DIS] 6 (18.18%) [INS]+[EXH] 6 [ EXH ]+[ DIS ] 4 (12.12%) [ DIS ]+[DIS] 4 [ EXH ]+[ EXH ] 3 (9.09%) 3 [ DIS ]+[ EXH ]+[DIS] 3 (13.63%) 4 [DIS]+[DIS]+[DIS]+[EXH] 3 (14.28%) [DIS]+[DIS]+[DIS]+[INS] 3 Table 2: Move patterns in the English dataset. A comparison of the results from the examination of the two datasets facilitates a response to RQ#2. 5. Comparing the datasets The Spanish-speaking tweeters chiefly cast doubts on the epistemic personhood of the Ministry of Health through short, one-move tweets consisting of compelling and effective emotional displays and simple, but forceful, overwhelming and rather aggressive contributions. In contrast, the increase of multiple-move tweets and the occurrence of seven- and eight-move ones attest to a greater elaboration and engagement by the Englishspeaking users. These seem to evince a more argumentative, contentious and perhaps persuasive style, which was adopted not simply as a way of challenging the WHO, but also of showing the reasons why they did so. Through counterarguments, claim rebuttals and doubts about the timeliness, helpfulness and reasonableness of their recommendations, these users also reaffirmed their distrust position (Chik, 2015), highlighted membership to a resistance group and perhaps attempted to gain adepts. As tweet length increases, move patterns diversify in both datasets and display occasional occurrences. Nevertheless, two combinations were somehow recurrent: [INS]+[DIS], which was employed half as much in the English dataset than in the Spanish one, and [INS]+[EXH], which exhibited similar occurrences in both datasets. Insufficient engagement owing to the site’s character limit (Merchant, 2006), lack of time or loss of topical interest due to surfeit about pandemic-related issues (Chen & Pirolli, 2012; Hu, Farnham & Talamadupula, 2015) could have made the Spanish users refrain from verbal outbursts and tirades against the health authorities or from counterarguing. Yet, tweet brevity could have stemmed from the platform affordances, which invites short, impulsive contributions and facilitates access through mobile phones, where texting may be felt awkward (Groshek & Cutino, 2016; Jaidka, Zhou & Lelkes, 2019). This would have caused these users to just wish to position themselves as members of an in-group brought together by similar epistemic and attitudinal stances about the issues in question (Hardaker & McGlashan, 2016). Nevertheless, these tweeters might likewise have presupposed that their texts would get integrated in larger chains or threads of perhaps more vehement and better-articulated postings (Benis et al., 2021; boyd, Golder & Lotan, 2010), as well as in broader networks of anti-COVID, anti-vaccination, incredulity and resistance digital texts (Fairclough, 2005; Flowerdew, 2011). Hence, they would have felt no need for more explicit or lengthier postings, as those other texts revolved around related issues and vented similar feelings and attitudes. Such texts would have co-constructed a belief pool that could enable readers to easily come to conclusions about the Spanish users’ attitudes, feelings, opinions and reasons for questioning the information source (Hafner, 2015). Like other online outlets of discussion (Coe, Kenski & Rains, 2014), both datasets exhibit an abundance of insults, even though they are more frequent in the Spanish one. Spanish-speaking tweeters show a higher preference for epithets referring to conditions like stupidity, ineptitude, retardation, nonsensicality or untrustworthiness. In the English dataset, this type of offence is less usual, as it is witnessed in cases like ‘sick liars’ or ‘(corrupt) lying bastards’. These insults do not simply criticise both institutions as intellectuals, but dis-intellectualise them as a way of denying their epistemic authority (Gabel, Reichert & Reuter, 2022). voiced through brief, evaluative postings alluding to absurdity and stupidity, as in example (27). Alternatively, it is expressed through rhetorical questions unveiling astonishment and doubts about the intelligibility and sensibleness of the recommendations, thus insinuating the incompetence, insanity and ignorance of their source, as in example (28): (27) more nonsense. (EMED076) (28) ¿Que es eso de 5 meses para adultos y 8 semanas a los niños? (MED061) ‘What do you mean by 5 months for adults and 8 weeks for children?’ Incomprehension overtones are similarly present in the requests for information in the English dataset. Rhetorical in nature, they may additionally carry indignation along with questioning and mistrustful attitudes: (29) Who gave you the right to decide the population? (EMED091) Overall, exhortations, imprecations, dismissive incomprehension and questions also evince and contribute to the aforementioned Discourse of distrust, resistance and rebelliousness. They share the adversarial (Clayman, 2002) or oppositional (Vásquez, 2021) stance that was also manifested through insults and disagreements as a way of indexing two opposed identity groups. Devoid of preventative or restorative face-work (Gottschalk, 2018), they seem unselected, non-curated and unmoderated salvos (Walsh & Baker, 2021) evincing the desires of the users, as non-gullible and rebellious citizens, to break the rules and overtly confront with the authorities because of differing views (Williams, 2021). Furthermore, they even suggest the tweeters’ intention to get even with the rulers due to the alleged damage arising from their policies. Hence, all these actions, like insults and disagreement, work as strategies whereby the tweeters keep on challenging the institutional structural subject position as a way of delegitimising and portraying the authorities as shameless, unscrupulous and sociopathic rulers, while presenting themselves as champions of morality and justice. As with retweets about certain issues (Yuan et al., 2022), their intensity, aggressiveness and incivility could even invite other users’ participation, wherewith the new identity attributed to the authorities could also be sanctioned (Garcés-Conejos Blitvich, 2009). Besides pathologising and silencing, the four actions seem to purport to gaslight the rulers (Bailey, 2020; Spear, 2020). Certainty about their deceitfulness and the information ill-intendedness would prompt the tweeters to erode their self-esteem and self-trust as epistemic agents. Although the users might thus have attempted to exert some influence on them and make them stop dispensing seemingly questionable information, the questioning overtones and the disadvantageousness of the actions that are wished would insinuate their deviousness and vileness. This would also apply to the only threat in the English dataset: (30) The WORLD IS AWAKE AND WE GOT YOUR NUMBER! (EMED022) As for the only complaint in the Spanish dataset –example (8) above– the admixture of negative feelings and emotions ranging from irritation to utter wrath which it vents would not solely be a consequence of the unreasonableness and uselessness of the recommendation to wear facemasks, but also of the ineptitude and malevolence attributed to the Spanish Ministry and, ultimately, their lack of epistemic trustworthiness. 6. Conclusion The examination of the datasets reveals that the epistemic identity of the Spanish Ministry of Health is chiefly reshaped by means of insults, whereas that of the WHO is mainly remoulded through disagreement. While the Spanish-speaking users seem to evince a more confrontational style, the English-speaking tweeters appear more prone to argumentation. Although the English dataset exhibits lower rates of imprecations, it displays higher rates of exhortations and tokens of dismissive incomprehension, which might have been motivated by greater user engagement favouring longer tweets. In both datasets, insults other-brand and place the authorities in a stigmatised, disintellectualised and dehumanised out-group that is pathologised as a public threat due to their alleged untrustworthiness, ineptitude, shamelessness, unscrupulousness and irrationality. In turn, disagreement facilitates self-assertion of a subject position on the measures to fight the pandemic as a way of claiming epistemic privilege and depriving the two institutions of epistemic trust. Dissenting postings take advantage of a Discourse of resistance, incredulity and rebelliousness wherewith the two user groups vent distrustful attitudes and create an invalidating context in order to undermine the credibility of the health authorities and perhaps silence them, while also contributing to an anti-COVID and anti-vaccination Discourse (Durmaz & Hengirmen, 2022; Gunaratne, Coomes & Haghbayan, 2019; Herrera-Peco et al., 2021; Larrondo-Ureta, Peña-Fernández & Morales-i-Gras, 2021). Regarding exhortations, imprecations and dismissive incomprehension, they seem to attempt to gaslight the rulers and influence their behaviour due to their supposed deceitfulness and ill intentions. All in all, in addition to jointly portraying the institutions as malevolent informers and constructing them as an outgroup which is deprived of epistemic trust, these actions jointly depict the users as non-gullible and unmanageable citizens who mistrust information from certain sources, thus creating an in-group of challenging and resisting users. The limited amount of data, the domain where epistemic identity reshaping has been observed and the online platform from which the data have been extracted call for further studies scrutinising larger corpora from diverse domains and different sites. Moreover, future research should also look into the reactions or responses to attacks to and questionings of epistemic identity in order to better understand the interactional digital dynamics they trigger. Surely, this would foster insightful appraisals of the identities that come into play and the Discourses that are exploited in performing, reclaiming and counterattacking particular identities online. References Allan, K. (2015). When is a slur not a slur? The use of nigger in ‘Pulp Fiction’. Language Sciences, 52, 187–199. https://doi.org/10.1016/j.langsci.2015.03.001 Amores, J. J., Blanco-Herrero, D., Sánchez-Holgado, P., & Frías-Vázquez, M. (2021). Detectando el odio ideológico en Twitter. Desarrollo y evaluación de un detector de discurso de odio por ideología política en tuits en español. Cuadernos.info, 49, 98– 124. http://dx.doi.org/10.7764/cdi.49.27817 Andersen, P. A. (1999). Nonverbal communication: Forms and functions. Mayfield. Anderson, A. A., & Huntington, H. E.. (2017). Social media, science, and attack discourse: How Twitter discussions of climate change use sarcasm and incivility. Science Communication, 39(5), 598–620. https://doi.org/10.1177/1075547017735113 Angouri, J., & Locher, M. (2012). Theorising disagreement. Journal of Pragmatics, 44(12), 1549–1553. https://doi.org/10.1016/j.pragma.2012.06.011 Anton, C., & Peterson, V. (2003). Who said what: Subject positions, rhetorical strategies and good faith. Communication Studies, 54(4), 403–419. https://doi.org/10.1080/10510970309363300 Archer, D. (2015). Slurs, insults, (backhanded) compliments and other strategic facework moves. Language Sciences, 52, 82–97. https://doi.org/10.1016/j.langsci.2015.03.008 Arroyas Langa, E., & Pérez Díaz, P. L. (2016). La nueva narrativa identitaria del populismo: Un análisis del discurso de Pablo Iglesias (Podemos) en Twitter. Cultura, Lenguaje y Representación, 15, 51–63. https://doi.org/10.6035/clr.2016.15.4 Bailey, A. (2020). On gaslighting and epistemic injustice: Editor’s introduction. Hypatia, 35(4), 667–673. https://doi.org/10.1017/hyp.2020.42 Baptista Ferreira, G. (2021). Conspiracy theories in time of the covid-19 pandemic: Populism, social media and misinformation. Comunicação e Sociedade, 40, 129–148. https://doi.org/0.17231/comsoc.40(2021).3324 Barker, K. (2016). Imprecation as divine discourse: Speech act theory, dual authorship, and theological interpretation. Eisenbrauns. Benis, A., Chatsubi, A., Levner, E., & Ashkenazi, S. (2021). Change in threads on Twitter regarding influenza, vaccines, and vaccination during the COVID-19 pandemic: Artificial intelligence-based infodemiology study. JMIR Infodemiology, 1(1): e31983. https://doi.org/10.2196/31983 Biber, D., Connor, U., & Upton, T. A. (2007). Discourse on the move: Using corpus analysis to describe discourse structure. John Benjamins. https://doi.org/10.1075/scl.28 Blanco-Alfonso, I., Rodríguez-Fernández, L, & Arce-García, S. (2022). Polarización y discurso de odio con sesgo de género asociado a la política: Análisis de las interacciones en Twitter. Revista de Comunicación, 21(2), 33–50. https://doi.org/10.26441/RC21.2-2022-A2 Bolander, B. (2017). Language and identity on Facebook. In S. L. Thorne & S. May (Eds.), Language, education and technology (pp. 143–154). Springer. https://doi.org/10.1007/978-3-319-02237-6_11 Bolander, B., & Locher, M. (2010). Constructing identity on Facebook: Report on a pilot study. In K. Junod & D. Maillat (Eds.), Performing the self (pp. 165–185). Narr. https://doi.org/10.5169/seals-131305 Borgwald, K. (2012). Women’s anger, epistemic personhood, and self-respect: An application of Lehrer’s work on self-trust. Philosophical Studies, 161, 69–76. https://doi.org/10.1007/s11098-012-9932-5 Bou Franch, P., & Garcés-Conejos Blitvich, P. (2014). Conflict management in massive polylogues: A case study from YouTube. Journal of Pragmatics, 73, 19–36. https://doi.org/10.1016/j.pragma.2014.05.001 Bou Franch, P., Lorenzo-Dus, N., & Garcés-Conejos Blitvich, P. (2012). Social interaction in YouTube text-based polylogues: A study of coherence. Journal of Computer-Mediated Communication, 14(4), 501–521. https://doi.org/10.1111/j.1083- 6101.2012.01579.x boyd, d., Golder, S., & Lotan, G. (2010). Tweet, tweet, retweet: Conversational aspects of retweeting on Twitter. In R. H. Sprague (Ed.), Proceedings of the 43rd Annual Hawaii International Conference on System Sciences (pp. 1–10). CPS. Brewer, M. B., & Gardner, W. (1996). Who is this “we”? Levels of collective identity and self representations. Journal of Personality and Social Psychology, 71(1), 83–93. https://doi.org/ 10.1037/0022-3514.71.1.83 Brons, L. (2015). Othering, an analysis. Transcience, 6(1), 69–90. Bucholtz, M. (2007). Shop talk: Branding, consumption, and gender in American middleclass youth interaction. In B. S. McElhinny (Ed.), Words, worlds, and material girls: Language, gender, globalization (pp. 371–402). Mouton de Gruyter. Bucholtz, M., & Hall, K. (2005). Identity and interaction: A sociocultural linguistic approach. Discourse Studies, 7(4–5), 585–614. https://doi.org/10.1177/1461445605054407 Burke, P. J., & Stets, J. E. (2009). Identity theory. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195388275.001.0001 Carpenter, J. P., Kimmons, R., Short, C. R., Clements, K., & Staples, M. E. (2019). Teacher identity and crossing the professional-personal divide on twitter. Teaching and Teacher Education, 81, 1–12. https://doi.org/10.1016/j.tate.2019.01.011 Chen, J., & Pirolli, P. (2012). Why are you more engaged: Factors influencing Twitter engagement in Occupy Wall Street. Proceedings of the International AAAI Conference on Web and Social Media, 6(1), 423–426. https://doi.org/10.1609/icwsm.v6i1.14324 Chik, A. (2015). Recreational language learning and digital practices: positioning and repositioning. In R. H. Jones, A. Chik & C. A. Hafner (Eds.), Discourse and digital practices. Doing discourse analysis in the digital age (pp. 112–129). Routledge. https://doi.org/10.4324/9781315726465-8 Cho, V., & Jimerson, J. B. (2016). Managing digital identity on Twitter: The case of school administrators. Educational Management Administration & Leadership, 45(5), 884–900. https://doi.org/10.1177/1741143216659295 Clayman, S. (2002). Tribune of the people: Maintaining the legitimacy of aggressive journalism. Media, Culture and Society, 24(2), 197–216. https://doi.org/10.1177/016344370202400203 Coe, K., & Griffin, R. A. (2020). Marginalized Identity Invocation Online: The Case of President Donald Trump on Twitter. Social Media + Society, 6(1): https://doi.org/10.1177/2056305120913979. Coe, K., Kenski, K., & Rains, S. A. (2014). Online and uncivil? Patterns and determinants of incivility in newspaper website comments: Incivility in newspaper website comments. Journal of Communication, 64(4), 658–679. https://doi.org/10.1111/jcom.12104 Croom, A. M. (2013). How to do things with slurs: Studies in the way of derogatory words. Language & Communication, 33, 177–204. https://doi.org/10.1016/j.langcom.2013.03.008 Cull, M. J. (2019). Dismissive incomprehension: A use of purported ignorance to undermine others. Social Epistemology, 33(3), 262–271. https://doi.org/10.1080/02691728.2019.1625982 D’Arcy, A., & Young, T. M. (2012). Ethics and social media: Implications for sociolinguistics in the networked public. Journal of Sociolinguistics, 16(4), 532– 546. https://doi.org/10.1111/j.1467-9841.2012.00543.x de Fina, A., Schiffrin, D., & Bamberg, M. (Eds.). (2006). Discourse and identity. Cambridge University Press. https://doi.org/10.1017/CBO9780511584459 DuBois, J. W. (2007). The stance triangle. In R. Englebretson (Ed.), Stancetaking in discourse: Subjectivity, evaluation, interaction (pp. 139–182). John Benjamins. https://doi.org/10.1075/pbns.164.07du Durmaz, N., & Hengirmen, E. (2022). The dramatic increase in anti-vaccine discourses during the COVID-19 pandemic: A social network analysis of Twitter. Human Vaccines & Immunotherapeutics, 18(1), e2025008. https://doi.org/ 10.1080/21645515.2021.2025008 Edwards, D. (2005). Moaning, whinging and laughing: The subjective side of complaints. Discourse Studies, 7(1), 5–29. https://doi.org/10.1177/1461445605048765 Evans, A. (2016). Stance and identity in Twitter hashtags. Language @ Internet, 13, article 1 (urn:nbn:de:0009-0-54947). Fairclough, N. (2005). Critical discourse analysis in transdisciplinary research. In R. Wodak & P. Chilton (Eds.), A new agenda in (critical) discourse analysis. Theory, methodology and interdisciplinarity (pp. 53–70). John Benjamins. https://doi.org/10.1075/dapsac.13.06fai Flowerdew, J. (2011). Some dichotomies in genre analysis for Language for Specific Purposes. In V. K. Bhatia, P. Sánchez Hernández & P. Pérez-Paredes (Eds.), Researching specialized languages (pp. 139–153). John Benjamins. https://doi.org/10.1075/scl.47.11flo Foucault, M. (1972). The archeology of knowledge (R. Sheridan, Trans.). Routledge. Fricker, M. (2007). Epistemic injustice. Power and the ethics of knowledge. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198237907.001.0001 Gabel, S., Reichert, L., & Reuter, C. (2022). Discussing conflict in social media: The use of Twitter in the Jammu and Kashmir conflict. Media, War & Conflict, 15(4), 504– 529. https://doi.org/10.1177/1750635220970997 Garcés-Conejos Blitvich, P. (2009). Impoliteness and identity in the American news media: The “Culture Wars”. Journal of Politeness Research, 5(2), 273–303. https://doi.org/10.1515/JPLR.2009.014 Garcés-Conejos Blitvich, P. (2022). Moral emotions, good moral panics, social regulation, and online public shaming. Language & Communication, 84, 61–75. https://doi.org/10.1016/j.langcom.2022.02.002 Garcés-Conejos Blitvich, P., & Georgakopoulou, A. (2021). Analysing identity. In M. Haugh, D. Z. Kádár & M. Terkourafi (Eds.), The Cambridge handbook of sociopragmatics (pp. 293–314). Cambridge University Press. https://doi.org/10.1017/9781108954105.017 Garcés-Conejos Blitvich, P., & Sifianou, M. (2017). (Im)politeness and identity. In J. Culpeper, M. Haugh & D. Z. Kádár (Eds.), The Palgrave handbook of linguistic (im)politeness (pp. 227–256). Palgrave Macmillan. https://doi.org/10.1057/978-1- 137-37508-7_10 Gee, J. P. (2000). Identity as an analytic lens for research in education. Review of Research in Education, 25(1), 99–125. https://doi.org/10.2307/1167322 Gee, J. P. (2013). Discourse versus discourse. In C. A. Chapelle (Ed.), The encyclopedia of applied linguistics (online edition). John Wiley & Sons. https://doi.org/10.1002/9781405198431.wbeal0332 Gee, J. P. (2014). An introduction to discourse analysis. Theory and method. Routledge. Goddard, C. (2015). “Swear words” and “curse words” in Australian (and American) English. At the crossroads of pragmatics, semantics and sociolinguistics. Intercultural Pragmatics, 12(2), 189–218. https://doi.org/10.1515/ip-2015-0010 Goffman, E. (1959). The presentation of self in everyday life. Doubleday. Gottschalk, S. (2018). The terminal self: Everyday life in hypermodern times. Routledge. Graham, S. L., & Hardaker, C. (2017). (Im)politeness in digital communication. In J. Smith, M. A., Rainie, L., Shneiderman, B., & Himelboim, I. (2014). Mapping Twitter topic networks: From polarized crowds to community clusters. Pew Research Center. https://pewrsr.ch/3xUK203 Spear, A. D. (2020). Gaslighting, confabulation, and epistemic innocence. Topoi, 39, 229–241. https://doi.org/10.1007/s11245-018-9611-z Spencer-Oatey, H. D. (Ed.). (2008). Culturally speaking: culture, communication and politeness theory. Continuum. Sperber, D. (1996). Explaining culture. A naturalistic approach. Blackwell. Sperber, D. (1997). Intuitive and reflective beliefs. Mind and Language, 12(1), 67– 83. https://doi.org/10.1111/j.1468-0017.1997.tb00062.x Sperber, D., & Wilson, D. (1995). Relevance. Communication and cognition. Blackwell. Stock, O., & Strapparava, C. (2003). HAHAcronym: Humorous agents for humorous acronyms. Humor, 16(3), 297–314. https://doi.org/10.1515/humr.2003.016 Thomas-Olalde, O., & Velho, A. (2011). Othering and its effects: Exploring the concept. In H. Niedrig & C. Ydesen (Eds.), Writing postcolonial histories of intercultural education (pp. 27–50). Peter Lang. Thorson, J., & Baker, C. (2019). Venting as epistemic work. Social Epistemology, 33(2), 101–110. https://doi.org/10.1080/02691728.2018.1561762 Thurlow, C., & Mroczek, K. (Eds.). (2011). Digital discourse: Language in the new media. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199795437.001.0001 Ullah, I., Khan, K. S., Tahir, M. J., Ahmed, A., & Harapan, H. (2021). Myths and conspiracy theories on vaccines and COVID-19: Potential effect on global vaccine refusals. Vacunas, 22(2), 93–97. https://doi.org/10.1016/j.vacun.2021.01.001 Valdez-Apolo, M. B., Arcila Calderón, C., & Jiménez Amores, J. (2019). El discurso de odio hacia migrantes y refugiados a través del tono y los marcos de los mensajes. Revista de la Asociación Española de Investigación de la Comunicación, 6(12), 361– 384. https://doi.org/10.24137/raeic.6.12.2 Vásquez, C. (2015). Intertextuality and interdiscursivity in online consumer reviews. In R. H. Jones, A. Chik & C. A. Hafner (Eds.), Discourse and digital practices. Doing discourse analysis in the digital age (pp. 66–80). Routledge. Vásquez, C. (2021). “I appreciate u not being a total prick…”: Oppositional stancetaking, impoliteness and relational work in adversarial Twitter interactions. Journal of Pragmatics, 185, 40–53. https://doi.org/10.1016/j.pragma.2021.08.009 Walsh, M. J., & Baker, S. A. (2021). Avoiding conflict and minimising exposure: Facework on Twitter. Convergence: The International Journal of Research into New Media Technologies, 28(3), 664–680. https://doi.org/10.1177/13548565211036797 Williams, M. (2021). The science of hate. How prejudice becomes hate and what we can do to stop it. Faber&Faber. Wilson, D., & Sperber, D. (2012). Explaining irony. In D. Wilson & D. Sperber (Eds.), Meaning and relevance (pp. 123–145). Cambridge University Press. https://doi.org/10.1017/CBO9781139028370.008 Wodak, R. (2006). Mediation between discourse and society: Assessing cognitive approaches in CDA. Discourse Studies, 8(1), 179–90. https://doi.org/10.1177/1461445606059566 Yuan, S., Chen, Y., Vojta, S., & Chen Y. (2022). More aggressive, more retweets? Exploring the effect of aggressive climate change messages on Twitter. New Media & Society, 0(0), 1–20. https://doi.org/10.1177/14614448221122202 Zappavigna, M. (2012). Discourse of Twitter and social media. Continuum. Manuel Padilla Cruz (PhD) is an associate professor of English language and linguistics at Universidad de Sevilla. He teaches courses on pragmatics, discourse analysis, semantics and communication. His research interests fall within the field of pragmatics, where he has addressed a number of issues from a relevance-theoretic perspective. He has edited Relevance Theory. Recent Developments, Current Challenges and Future Directions (John Benjamins) and co-edited the special issues “New Developments in Relevance Theory”, with Pragmatics & Cognition (John Benjamins), and “Concepts and context in relevance-theoretic pragmatics. New developments”, with Pragmatics (John Benjamins).