Accounts as acts of identity: Justifying business closures on COVID-19 public signs in Athens and London

Spyridoula Bella and Eva Ogiermann
National and Kapodistrian University of Athens | King’s College London
Abstract

This paper investigates accounts justifying the closures of businesses found on public signs in Athens and London during the COVID-19 pandemic. The data for the study was drawn from a corpus of COVID-19-related public signage collected in the two cities during the first lockdown. The accounts used on these signs are analysed as acts of identity and, specifically, as discursive means deployed by the authors of the signs to project themselves and their businesses favourably. It is shown that the accounts used at the micro-level of discourse align to various degrees with the dominant discourses surrounding the pandemic at the macro-level and with the values these discourses draw upon. It is also shown that the accounts are used to reframe the public’s understanding of the closures and to construct identities congruent with the interests of the business owners, ensuring post-pandemic continuity.

Keywords:
Publication history
Table of contents

1.Introduction

On the 11th of March 2020, the World Health Organization (WHO) declared the outbreak of COVID-19 a global pandemic. As the virus spread across the world, the countries concerned implemented measures to contain it, including the closures of non-essential businesses. The present study investigates the reactions of the business owners to the closures during this first lockdown by examining the messages appearing on the doors of closed businesses in Athens and London.

In Greece non-essential businesses closed on the 18th of March 2020. In the UK, lockdown regulations came into force on the 26th. Meanwhile, government and health officials in both countries released numerous statements emphasising the dangerous nature of the virus, as well as the importance of observing the measures for health and safety reasons. The Greek government, in particular, constantly encouraged people to stay indoors through a wide-scale media campaign known as ΜΕΝΟΥΜΕ ΣΠΙΤΙ (‘we stay at home’).

In both countries, business owners had to close their businesses and incur enormous financial losses, without knowing when they would be able to reopen and whether their clientele would return. Although the closures were legal measures enforced by the respective governments and were communicated to the public, many businesses placed signs on their door announcing their closures. Signs announcing closures that were publicly ordered by the government may seem superfluous – and it is true that several businesses did not display any. However, many businesses did, and this already indicates that these signs communicated more than merely the closure.

Public signs have been so far examined in research on Linguistic Landscapes (LL). This research has focused mainly on the identification of the different languages employed οn those signs in multicultural communities and has addressed them as “multimodal objects rather than as linguistic ones” (Blommaert 2013 2013Ethnography, Superdiversity and Linguistic Landscapes: Chronicles of Complexity. Clevedon: Multilingual Matters. DOI logoGoogle Scholar, 41). Relatively few studies have analysed the actual messages public signs communicate and/or the linguistic means employed to communicate them (Ferenčik 2018Ferenčík, Milan 2018 “Im/politeness on the Move: A Study of Regulatory Discourse Practices in Slovakia’s Centre of Tourism.” Journal of Pragmatics 134: 183–198. DOI logoGoogle Scholar; Ogiermann and Bella 2021Ogiermann, Eva, and Spyridoula Bella 2021 “On the Dual Role of Expressive Speech Acts: Relational Work on Signs Announcing Closures during the Covid-19 Pandemic.” Journal of Pragmatics. DOI logoGoogle Scholar; Svennevig 2021Svennevig, Jan 2021 “How to Do Things with Signs: The Formulation of Directives on Signs in Public Spaces.” Journal of Pragmatics 175: 165–183. DOI logoGoogle Scholar; Wierzbicka 1998Wierzbicka, Anna 1998 “German Cultural Scripts: Public Signs as a Key to Social Attitudes and Cultural Values.” Discourse and Society 9 (2): 241–282. DOI logoGoogle Scholar).

In Ogiermann and Bella (2021)Ogiermann, Eva, and Spyridoula Bella 2021 “On the Dual Role of Expressive Speech Acts: Relational Work on Signs Announcing Closures during the Covid-19 Pandemic.” Journal of Pragmatics. DOI logoGoogle Scholar we examined the COVID-19 closure signs produced during the first lockdown in Athens and London as a form of relational work and focused on the expressive speech acts (greetings, thanks, apologies and wishes) found on these signs. In this paper, we focus on the discursive move that was most frequent in the data, i.e., the accounts provided to justify the closures. The high frequency of accounts is particularly interesting, since the reasons behind the closure were common knowledge. Moreover, unlike other discursive moves, such as apologies and thanks (see Ogiermann and Bella 2021Ogiermann, Eva, and Spyridoula Bella 2021 “On the Dual Role of Expressive Speech Acts: Relational Work on Signs Announcing Closures during the Covid-19 Pandemic.” Journal of Pragmatics. DOI logoGoogle Scholar), accounts exhibited a remarkable uniformity in the two datasets, with both Greek and English business owners employing similar accounts to explain the closures and formulating them in similar ways.

In what follows we address this uniformity along with the reasons that prompted the authors of the signs to explain the closures. Specifically, it is maintained that the accounts construct a positive self-image or a desirable identity for the business owners who authored these signs.

The signs examined in this study constitute a unique documentation of a historical event, illustrating how “particular positions of alignment and disalignment to discourses in circulation about key events come to be created and shared in the context of specific critical incidents” (Seargaent and Giaxoglou 2020Seargeant, Philip, and Korina Giaxoglou 2020 “Discourse and the Linguistic Landscape.” In The Cambridge Handbook of Discourse Studies, ed. by Anna De Fina, and Alexandra Georgakopoulou, 306–326. Cambridge: Cambridge University Press. DOI logoGoogle Scholar, 310). We maintain that the identities constructed through the accounts are, to a large extent, a product of the interaction of individual positionings at the micro-level with powerful discourses surrounding the pandemic at the macro-level (cf. Archakis 2020Archakis, Argiris 2020 “The Continuum of Identities in Immigrant Students’ Narratives in Greece.” Narrative Inquiry. Published online 8 July 2020 DOI logoGoogle Scholar; Archakis and Tsakona 2012Archakis, Argiris, and Villy Tsakona 2012The Narrative Construction of Identities in Critical Education. London: Palgrave Macmillan. DOI logoGoogle Scholar).

The paper is structured as follows: The next section introduces the framework of the study, clarifying the understanding of identity it draws on (2.1). We then contextualise our data by presenting and analysing the dominant discourses of the pandemic at the macro-level (2.2). The section closes with a review of research on accounts which constitute the tools via which the identities are constructed at the micro-level (2.3). Section 3 presents the method of the study, while the data is analysed and discussed in Sections 4 and 5 respectively. Section 5 also presents the study’s conclusions.

2.Framework of the study

2.1Identity

The perspective on identity assumed here is a non-essentialist, social constructionist one, according to which, identities are not considered static and a priori given, but are thought of as “representations mediated by semiotic systems such as language” (Benwell and Stokoe 2006Benwell, Brethan, and Elizabeth Stokoe 2006Discourse and Identity. Edinburgh: Edinburgh University Press. DOI logoGoogle Scholar, 31).11.See Archakis and Tsakona (2012)Archakis, Argiris, and Villy Tsakona 2012The Narrative Construction of Identities in Critical Education. London: Palgrave Macmillan. DOI logoGoogle Scholar and Garcés-Conejos Blitvich and Georgakopoulou (2021)Garcés-Conehos Blitvich, Pilar, and Alexandra Georgakopoulou 2021 “Analysing Identity.” In The Cambridge Handbook of Sociopragmatics, ed. by Michael Haugh, Daniel Kadár, and Marina Terkourafi, 293–314. Cambridge: Cambridge University Press. DOI logoGoogle Scholar for thorough reviews of the essentialism vs. the social constructionism approach to identity.

This understanding of identity presupposes a view of discourse as a tool for the construction of multiple and mutable identities (see Archakis and Tsakona 2012Archakis, Argiris, and Villy Tsakona 2012The Narrative Construction of Identities in Critical Education. London: Palgrave Macmillan. DOI logoGoogle Scholar, 20). Identity is seen as the “social positioning of the self” (Bucholtz and Hall 2005Bucholtz, Mary, and Kira Hall 2005 “Identity and Interaction: A Sociocultural Linguistic Approach.” Discourse Studies 7 (4–5): 585–614. DOI logoGoogle Scholar, 585–6) through discourse, i.e., as emergent through the speakers’ linguistic choices in specific interactional practices that have specific contexts including the situation and the actual or imaginary addressees (see, e.g., Bucholtz and Hall 2005Bucholtz, Mary, and Kira Hall 2005 “Identity and Interaction: A Sociocultural Linguistic Approach.” Discourse Studies 7 (4–5): 585–614. DOI logoGoogle Scholar; De Fina et al. 2006De Fina, Anna, Deborah Schiffrin, and Michael Bamberg (eds) 2006Discourse and Identity. Cambridge: Cambridge University Press. DOI logoGoogle Scholar, 2). Therefore, it is not only reflecting but also producing the social world (Hall and Bucholtz 2013Hall, Kira, and Mary Bucholtz 2013 “Epilogue: Facing Identity.” Journal of Politeness Research 9 (1): 123–132. DOI logoGoogle Scholar, 125).

In line with Bucholtz and Hall’s positionality principle (2005Bucholtz, Mary, and Kira Hall 2005 “Identity and Interaction: A Sociocultural Linguistic Approach.” Discourse Studies 7 (4–5): 585–614. DOI logoGoogle Scholar), identity construction is seen as “linked to the ephemeral subject positions that individuals occupy in the flow of interaction” (Archakis and Tsakona 2012Archakis, Argiris, and Villy Tsakona 2012The Narrative Construction of Identities in Critical Education. London: Palgrave Macmillan. DOI logoGoogle Scholar, 32). Since in every interactional move, speakers position themselves as particular kinds of people (see Bucholtz and Hall 2005Bucholtz, Mary, and Kira Hall 2005 “Identity and Interaction: A Sociocultural Linguistic Approach.” Discourse Studies 7 (4–5): 585–614. DOI logoGoogle Scholar, 147; Day and Kjaerbeck 2013Day, Dennis, and Susanne Kjaerbeck 2013 “ ‘Positioning’ in the Conversation Analytic Approach.” Narrative Inquiry 23 (1): 16–39. DOI logoGoogle Scholar, 18) identities are viewed as “points of temporary attachment to the subject positions which discursive practices construct for us” (Hall 2000Hall, Stuart 2000 “Who Needs Identity?” In Identity: A Reader, ed. by Paul du Gay, Jessica Evans, and Peter Redman, 1–16. London: Sage.Google Scholar, 19). These subject positions, in turn, incorporate stancetaking (see Du Bois 2007Du Bois, John 2007 “The Stance Triangle.” In Stancetaking in Discourse: Subjectivity, Evaluation, Interaction, ed. by Robert Englebretson, 139–182. Amsterdam/Philadelphia: John Bemjamins. DOI logoGoogle Scholar, 143), i.e., the display of “evaluative, affective and epistemic orientations in discourse” (Bucholtz and Hall 2005Bucholtz, Mary, and Kira Hall 2005 “Identity and Interaction: A Sociocultural Linguistic Approach.” Discourse Studies 7 (4–5): 585–614. DOI logoGoogle Scholar, 595). Against this backdrop, the present study treats positioning and identity construction as synonymous.

We focus on one specific type of positioning attested in the data, i.e., the construction of causality through the use of accounts. Starting from the premise that the accounts under study do not serve the function of actually informing the public about the reasons behind the closures, we suggest that the sign authors exploit accounts as relational tools to boost their own image and to construct for themselves an identity that would prompt the audience to evaluate them positively as individuals and, by association, their businesses as worth supporting.

An important point in relation to the data examined here is that identities, albeit emergent in local interactional practices, are not unaffected by pre-existing resources, but are always “produced through contextually situated and ideologically informed configurations of the self and other” (Bucholtz and Hall 2005Bucholtz, Mary, and Kira Hall 2005 “Identity and Interaction: A Sociocultural Linguistic Approach.” Discourse Studies 7 (4–5): 585–614. DOI logoGoogle Scholar, 605). That is, ideologies and values carried by dominating structures can be imposed in a top-down manner, and affect the discursive construction of individual identities (see, e.g., Archakis 2020Archakis, Argiris 2020 “The Continuum of Identities in Immigrant Students’ Narratives in Greece.” Narrative Inquiry. Published online 8 July 2020 DOI logoGoogle Scholar; Archakis and Tsakona 2012Archakis, Argiris, and Villy Tsakona 2012The Narrative Construction of Identities in Critical Education. London: Palgrave Macmillan. DOI logoGoogle Scholar, 2016 2016 “Legitimising and Resistance Identities in Immigrant Students’ School Essays: Towards a Culturally Sustaining Pedagogy.” Brno Studies in English 42 (1): 5–22. DOI logoGoogle Scholar; Castells 2010Castells, Manuel 2010The Power of Identity. 2nd edition. Malden: Wiley-Blackwell.Google Scholar; Gee 1990Gee, James Paul 1990Social Linguistics and Literacies: Ideology in Discourses. London: Routledge.Google Scholar; van Dijk 2008Van Dijk, Teun 2008Discourse and Power. New York: Palgrave Macmillan. DOI logoGoogle Scholar). Therefore, identity construction always involves speakers’ positionings of alignment or disalignment with dominant discourses at the macro-level (see, e.g. Archakis 2020Archakis, Argiris 2020 “The Continuum of Identities in Immigrant Students’ Narratives in Greece.” Narrative Inquiry. Published online 8 July 2020 DOI logoGoogle Scholar, 2–3; Gee 1990Gee, James Paul 1990Social Linguistics and Literacies: Ideology in Discourses. London: Routledge.Google Scholar, 161–162; van Dijk 2008Van Dijk, Teun 2008Discourse and Power. New York: Palgrave Macmillan. DOI logoGoogle Scholar, 85–89).

In this vein, Castells (2010Castells, Manuel 2010The Power of Identity. 2nd edition. Malden: Wiley-Blackwell.Google Scholar, 8) makes a distinction between legitimising and resistance identities. The former are promoted as legitimate by the dominant institutions of society (and, consequently, by the dominant discourses), while the latter are built on principles that differ from, or are opposed to, those of the dominant institutions, and resist their discourses.

As will be shown, the positionings taken by the authors of the signs through their accounts resulted exclusively in the construction of legitimising identities. However, it will also be shown that preferences for different types of accounts lead to the construction of different legitimising identities and that alignment with specific discourses can be a matter of degree.

2.2Dominant discourses of the COVID-19 pandemic

Dominant discourses are usually connected with power institutions and “can lead to the acquisition of social goods (money, power, status) in a society” (Gee 1990Gee, James Paul 1990Social Linguistics and Literacies: Ideology in Discourses. London: Routledge.Google Scholar, 162). Therefore, they empower those who sanction them, warranting their access to those goods.

The dominant discourses surrounding the COVID-19 pandemic emanated from the WHO, its national counterparts and, most importantly, the two countries’ governments. A close look at the governmental discourses in Greece and the UK, where the signs were produced, reveals stunning similarities regarding the justifications provided for the measures taken.22.See, e.g., the statements and appeals to the public made by the PMs and Ministers of Health of the two countries: https://​eody​.gov​.gr​/ta​-proliptika​-metra​-gia​-ton​-periorismo​-tis​-diadosis​-toy​-koronoioy​-einai​-metra​-atomikis​-kai​-syllogikis​-eythynis/ https://​eody​.gov​.gr​/dilosi​-ypoyrgoy​-ygeias​-vasili​-kikilia​-stin​-enimerosi​-ton​-diapisteymenon​-syntakton​-gia​-ton​-neo​-koronoio/ https://​www​.gov​.uk​/government​/news​/government​-sets​-out​-plans​-to​-enforce​-closure​-of​-businesses​-and​-other​-venues​--2 https://​www​.gov​.uk​/government​/speeches​/pm​-address​-to​-the​-nation​-on​-coronavirus​-23​-march​-2020

This should not come as a surprise, given the global character of the COVID-19 pandemic. As a process driven largely by technological advances and entailing “the intensified flows of capital, goods, images and discourses around the globe” (Blommaert 2010Blommaert, Jan 2010The Socioliguistics of Globalisation. Cambridge: Cambridge University Press. DOI logoGoogle Scholar, 13), globalisation “affects language use” (Sifianou and Tzanne 2018Sifianou, Maria, and Angeliki Tzanne 2018 “The Impact of Globalization in Brief Greek Service Encounters.” Journal of Pragmatics 134: 163–172. DOI logoGoogle Scholar, 164). The question as to whether globalisation in terms of language use should be seen as a homogenizing process leading to uniformity or as “a synergetic relationship between the global and the local” (ibid.) remains controversial (see, e.g., Coupland 2010Coupland, Nikolas 2010 “Introduction: Sociolinguistics in the Global Era.” In The Handbook of Language and Globalization, ed. by Nikolas Coupland, 1–27. Chichester: Wiley/Blackwell. DOI logoGoogle Scholar). However, the similarities in Greece’s and the UK’s dominant discourses point towards uniformity.

‘Under the advice’ of the health authorities, especially the WHO, both the Greek and the English Prime Ministers and health officials tended to present the pandemic as ‘the enemy’ that had to be defeated through a common national effort where citizens had to ‘do their part’. Moreover, there was a relentless emphasis on the importance of adherence to the newly established laws concerning the pandemic, as well as on the responsibility for protecting others.

Admittedly, certain linguistic choices were country-specific. During daily updates on COVID cases in Greece, health officials kept emphasising the need for ατομική ευθύνη (‘individual responsibility’) and for compliance with the legislation. Additionally, the slogan μένουμε σπίτι (‘we stay at home’) was systematically circulated through the media. In the UK, on the other hand, particular emphasis was placed on the need to protect the National Health System (NHS) and prevent its collapse, as well as the need to work in unison against the spread of the virus. Overall, however, the accounts that emerged on the closure signs in the two countries were strongly influenced by the globalised discourses emanating from power institutions (health authorities and governments).

One point that we seek to make here is that dominant discourses are not endorsed simply because they emanate from institutions of power. Since any discourse is inherently “ideological” in that it always puts forward a certain set of values at the expense of others (Gee 1990Gee, James Paul 1990Social Linguistics and Literacies: Ideology in Discourses. London: Routledge.Google Scholar, 161–162), we maintain that the acceptability of any ideology depends on the acceptability of the values it carries and how these values become relevant in specific historical contexts. Although dominant discourses are always assumed to carry values, the issue of which values, when and why, is hardly ever addressed. Yet, we believe that for power formations to persuade individuals of the “essential ‘truth’ and ‘naturalness’” of their power (Benwell and Stokoe 2006Benwell, Brethan, and Elizabeth Stokoe 2006Discourse and Identity. Edinburgh: Edinburgh University Press. DOI logoGoogle Scholar, 30), it is necessary to convince the public of the truth and naturalness33.According to Fairclough (2013Fairclough, Norman 2013Language and Power. 2nd edition. London: Routledge. DOI logoGoogle Scholar, 89), “a dominant discourse is subject to a process of naturalization in which it appears to lose connection with particular ideologies and interests and becomes […] common sense. Thus, […] it ceases to be an ideology […] for ideology is truly effective only when it is disguised”. of the values promoted through the messages they communicate.

It is suggested here that what has rendered the governments’ discourse credible and appealing, thus prompting the sign authors to endorse it, is that it drew on values that people, regardless of cultural background, tend to consider important. Such values have been studied in social psychology research, especially by Schwarz (1992)Schwarz, Shalom H. 1992 “Universals in the Content and Structure of Values: Theoretical Advances and Empirical Tests in 20 Countries.” In Advances in Experimental Social Phychology, Vol. 25, ed. by Mark Zhanna, 1–65. San Diego: Academic Press. DOI logoGoogle Scholar and Schwarz and Bardi (2001)Schwarz, Shalom H., and Anat Bardi 2001 “Value Hierarchies across Cultures: Take a Similarities Perspective.” Journal of Cross-Cultural Psychology 32 (3): 268–290. DOI logoGoogle Scholar who have established ten value-constructs prevailing as life-guiding principles.44.Their universal framework of value constructs was validated empirically in more than sixty-three different cultural groups. Two of these seem particularly relevant to the present study: conformity defined as “restraint of actions […] likely to upset or harm others and violate social expectations” and benevolence, i.e., “preservation and enhancement of the welfare of people with whom one is in frequent personal contact” (ibid.). These are the values that were promoted by the dominant discourses surrounding the pandemic: prompting the citizens to be law-abiding relates to conformity, while inciting them to consider the wellbeing of others relates to benevolence. Our analysis will show that conformity and benevolence are systematically reflected in the accounts that emerged on the closure signs. However, it will also show that individual freedom corresponding to the value-construct self-determination, defined as “independent thought and action-choosing” (Schwarz and Bardi 2001Schwarz, Shalom H., and Anat Bardi 2001 “Value Hierarchies across Cultures: Take a Similarities Perspective.” Journal of Cross-Cultural Psychology 32 (3): 268–290. DOI logoGoogle Scholar, 270), is also reflected in the formulations of the accounts, affecting the construction of the authors’ identities.

2.3Accounts and desirable identities

As already mentioned, accounts are investigated here as elements of the micro-level through which the authors of the signs position themselves towards the discourses of the macro-level. Linguistic work on accounts has been influenced by research in the philosophy of language which has treated them as devices aimed to repair offensive behaviour (see, e.g., Austin 1970Austin, John 1970 “A Plea for Excuses.” In J. L. Austin – Philosophical Papers, ed. by James Urmson, and Geoffrey Warnock, 175–204. Oxford: Oxford University Press.Google Scholar; Benoit 1995Benoit, William 1995Accounts, Excuses and Apologies: A Theory of Image Restoration Strategies. New York: State University of New York Press.Google Scholar), mitigate the illocutionary force of speech acts and protect the interactants’ face (Brown and Levinson 1987Brown, Penelope, and Stephen Levinson 1987Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. DOI logoGoogle Scholar). However, it is research on accounts conducted in social psychology that we consider more useful for our study.

Sociological research on accounts has developed in parallel and often overlapped with symbolic interactionism, especially Goffman’s (1959)Goffman, Erving 1959The Presentation of Self in Everyday Life. Garden City, NY: Doubleday-Anchor.Google Scholar insights on the way people present themselves to others in a self-protective fashion. Scott and Lyman (1968)Scott, Marvin, and Stanford Lyman 1968 “Accounts.” American Sociological Review 33: 46–62. DOI logoGoogle Scholar were the first to advance the interest in accounts, defining them as verbal statements made by social actors to explain unanticipated or deviant behaviours, and “bridge the gap between actions and expectations” (Scott and Lyman 1968Scott, Marvin, and Stanford Lyman 1968 “Accounts.” American Sociological Review 33: 46–62. DOI logoGoogle Scholar, 4). In the social-psychological paradigm, accounts are treated as means employed to maintain a positive relationship between interactants (Schlenker 1980Schlenker, Barry 1980Impression Management: The Self-concept, Social Identity and Interpersonal Relations. Monterey, CA: Brooks/ Cole.Google Scholar; Schlenker, Pontari and Christopher 2001Schlenker, Barry, Beth Pontari, and Andrew Christopher 2001 “Excuses and Character: Personal and Social Implications of Excuses.” Personality and Social Psychology Review 5 (1): 15–32. DOI logoGoogle Scholar; Shaw, Wild and Colquitt 2003Shaw, John, Eric Wild, and Jason Colquitt 2003 “To Justify or to Excuse? A Meta-analytic Review of the Effects of Explanations.” Journal of Applied Psychology 88: 444–458. DOI logoGoogle Scholar) and protect the account giver’s image (Cody and Dunn 2007Cody, Michael, and Deborah Dunn 2007 “Accounts.” In Explaining Communication: Contemporary Theories and Exemplars, ed. by Bryan Whaley, and Wendy Samter, 237–256. London: Lawrence Erlbaum.Google Scholar; Schlenker et al. 2001Schlenker, Barry, Beth Pontari, and Andrew Christopher 2001 “Excuses and Character: Personal and Social Implications of Excuses.” Personality and Social Psychology Review 5 (1): 15–32. DOI logoGoogle Scholar). If we accept Solstad and Bott’s (2017Solstad, Torgrim, and Oliver Bott 2017 “Causality and Causal Reasoning in Natural Language.” In The Oxford Handbook of Causal Reasoning, ed. by Michael R. Waldmann, 619–644. Oxford: Oxford University Press.Google Scholar, 620) definition of causality as “a two-place relation, relating a causing entity and a caused effect”, the accounts employed in our data appear to be redundant. As already mentioned, both the causing event (the governments’ decision to close the businesses) and its effect (the closure) were a priori known to the readers through the social context. We have to conclude then that these accounts are not part of “an objective, fact finding procedure” (Meyer 2000Meyer, Paul Georg 2000 “The Relevance of Causality.” In Cause, Condition, Concession, Contrast: Cognitive and Discourse Perspectives, ed. by Elizabeth Couper-Kuhlen, and Bernd Kortmann, 9–34. Berlin: Mouton de Gruyter. DOI logoGoogle Scholar, 18), but form a social interaction process which aims to expose to the readers “a part of the mind of the authors” (Draper 1988Draper, Stephen 1988 “What’s Going on in Everyday Explanation?”. In Analysing Everyday Explanation: A Casebook of Methods, ed. by Charles Antaki, 15–31. London: Sage.Google Scholar, 27). In this sense, it is mainly the subjective attitudes of the authors that the accounts reveal. These attitudes are embedded into the accounts and constructed interactively in that they build on what the readers already know from the context.

Considering accounts as devices of self-image construction and protection ultimately leads us back to Scott and Lyman’s (1968Scott, Marvin, and Stanford Lyman 1968 “Accounts.” American Sociological Review 33: 46–62. DOI logoGoogle Scholar, 46) contention that accounts are manifestations of “the underlying negotiation of identities within speech communities”.

It is emphasized that, especially when given in public, accounts can be intended “to change how people interpret a situation” and, thus, “reshape the receiver’s sense of the situation […] in a manner congruent with the interests of the sender” (Cobb et al. 2001Cobb, Anthony, Caroll Stephens, and George Watson 2001 “Beyond Structure: The Role of Social Accounts in Implementing Ideal Control.” Human Relations 54 (9): 1123–1153. DOI logoGoogle Scholar, 1130). In other words, accounts are deployed in ways that can influence the hearer’s interpretation of events, maximising rewards for the account giver (see Cody and Dunn 2007Cody, Michael, and Deborah Dunn 2007 “Accounts.” In Explaining Communication: Contemporary Theories and Exemplars, ed. by Bryan Whaley, and Wendy Samter, 237–256. London: Lawrence Erlbaum.Google Scholar, 248). Orbuch’s (1997)Orbuch, Terri 1997 “People’s Accounts Count: The Sociology of Accounts.” Annual Review of Sociology 23: 455–478. DOI logoGoogle Scholar study which revealed that respondents systematically formed impressions of others (e.g., degree of likability) through their accounts is indicative of their significance as tools for identity construction via impression management.

According to the impression management approach, individuals tend to “communicate accounts in ways that are […] beneficial to a desirable identity” (Cody and Dunn 2007Cody, Michael, and Deborah Dunn 2007 “Accounts.” In Explaining Communication: Contemporary Theories and Exemplars, ed. by Bryan Whaley, and Wendy Samter, 237–256. London: Lawrence Erlbaum.Google Scholar, 248). Desirable identities are defined by Schlenker (1986 1986 “Self-identification: Towards an Integration of the Private and Public Self.” In Public Self and Private Self, ed. by Roy Baumeister, 21–62. New York: Springer. DOI logoGoogle Scholar, 25) as those self-images and self-identifications that people endorse to serve their goals or purposes and “represent what people believe that they can be and should be in particular contexts” (ibid., emphasis in the original). Their dependence on the context, including the situation and the audience, renders desirable identity images a product of the moment constructed within a particular context. It follows that the concept of desirable identity construction through account giving is absolutely compatible with the premises of social constructivism about identity construction in general (see 2.1).

The analysis and the discussion to follow will examine closely the accounts used in the closure signs, in order to reveal what the authors of the signs consider ‘desirable identities’, how they construct them and why they consider them desirable and beneficial under the circumstances.

3.Method

The data collection for this research started in March 2020 and is still ongoing, with the aim to form the basis for a larger project related to public signage in COVID times. The signs that form the corpus were collected following the procedures of Linguistic Landscapes studies and involved four researchers (two in London and another two in Athens) systematically covering all areas within walking distance from their homes and taking pictures of COVID-related signs. For the purposes of this paper, we will focus on a sub-corpus comprising a total of 482 signs that appeared during the first lockdown in Greece and the UK to announce closures. From these, 238 were composed in English and were found on the doors or windows of a wide range of businesses in London, while 244 come from businesses in Athens and were written in Greek. Unlike what Seargant and Giaxoglou (2020)Seargeant, Philip, and Korina Giaxoglou 2020 “Discourse and the Linguistic Landscape.” In The Cambridge Handbook of Discourse Studies, ed. by Anna De Fina, and Alexandra Georgakopoulou, 306–326. Cambridge: Cambridge University Press. DOI logoGoogle Scholar call “top-down” signs, i.e., professionally crafted public signs with a more permanent nature, all the signs in our corpus were “bottom-up”, i.e., improvised and “homemade” and, thus, “tending more to the ephemeral” (2012, 308).

Since our research focusses on accounts provided for the closure of businesses, signs merely announcing the closure, without offering any reasons for it, were excluded from the corpus. Hence, the final corpus contained 357 signs (164 English, 193 Greek) that did provide accounts. The accounts in both the English and the Greek datasets were coded by the two authors of this paper, according to their content, i.e., the explanations put forward for the business closures. The categories of accounts identified in the data are presented below.

4.Data analysis

The accounts in our data were classified as falling in one of four categories:

  1. Referring to the guidelines of the government or the health authorities

  2. Referring to the businesses’ concern about the employees’, the customers’ and the general public’s health and wellbeing

  3. Referring to both the guidelines of the government and the businesses’ concern about public health

  4. Referring to COVID-19/pandemic/circumstances

Table 1 presents the distribution of these accounts in the Greek and English signs.

Table 1.Distribution of types of accounts across London’s and Athens’ signs
Athens London Total
n % Examples n % Examples n %
1. Official measures  80    41.5 Το κατάστημά μας θα παραμείνει κλειστό κατόπιν αποφάσεως της κυβέρνησης (‘Following the government’s decision, our store will remain closed’)  34    20.7 As per the government guidelines our office will now be closed 114    31.8
2. Safety concerns  69    35.8 Με αίσθημα ευθύνης για την προστασία τη δική σας και των υπαλλήλων μας, το φυσικό μας κατάστημα θα παραμείνει κλειστό (‘Out of a sense of responsibility for your and our employees’ protection the shop will remain closed’)  48    29.3 In order to play our part in keeping people safe, this unit is closed 117    32.8
3. Combination of 1 and 2  33  17 Σεβόμενοι την υγεία των πελατών μας και σύμφωνα με τις οδηγίες του κρατικού μηχανισμού, το κατάστημα θα παραμείνει κλειστό (‘Out of respect for our customers’ health and in line with the governmental guidelines our store will remain closed’)  30    18.3 Τhe well-being of our clients and colleagues remains our main concern. We are following government advice and have now closed our offices  63    17.7
4. COVID-19  11     5.7 Κλειστό λόγω πανδημίας (‘closed due to pandemic’)  52    31.7 Due to the Coronavirus outbreak, we closed the shop  63    17.7
Total 193 100 164 100 357 100

As shown in Table 1, the largest category of accounts in both the Greek and the English data was category 2 (32.8%), which contained accounts that refer to concerns about public health. The accounts comprising category 1, which present the authorities’ instructions as the main reason behind the closure, followed closely (31.8%). Both these categories of accounts were more frequent in the Greek than the English data, with the difference being more marked in the first category (Greek: 41.5%, English: 20.7%).

The third category displayed similar frequencies of accounts in the two languages (Greek: 17%, English: 18.3%), whereas category 4, which included accounts that make general, rather vague references to the pandemic and/or ‘the circumstances’, was far better represented in the English (31.7%) than in the Greek data (5.7%).

Although the accounts in categories 3 and 4 were equally frequent (17.7%), the relative vagueness of the accounts in the fourth category places them beyond the scope of the present study. Hence, in what follows we will focus on categories 1–3.

4.1Category 1: It is the authorities

In both datasets, category 1 could be divided into two subcategories: The first involves a simple cause-effect schema presenting the government’s decisions as the cause and the closure as its effect. The second comprises expressions that stress the authors’ alignment with the governmental measures, as well as formulations that implicitly or explicitly manifest a stance, i.e., an evaluation of the measures.

In the Greek data, the first subcategory includes 46 of the 80 signs assigned to category 1. The most common formulation of the Greek accounts in this subcategory was λόγω των κρατικών μέτρων (‘because of the state’s measures’) as exemplified in (1):

(1)

Λόγω των κρατικών μέτρων το κατάστημα θα παραμείνει κλειστό.

Because of the state’s measures, the store will remain closed.

The construction λόγω + genitive (‘because of’) is conventionalised for marking causality in Greek. Unlike other formulations discussed below, there is nothing in the construction’s meaning that could be viewed as indication of the speakers’ evaluation of the events they account for. In this sense, the construction offers itself as optimal means for presenting events “objectively” and, hence, its frequent appearance in accounts which appeal to what really happened, i.e., that it was the government that ordered the closure is anything but coincidence.

An important finding regarding this subcategory’s signs is that the authors systematically opted for constructions such as το κατάστημα θα παραμείνει κλειστό (‘the store will remain closed’) to reduce their agency. By construing an agentive event (the closure) from the perspective of the patient (the store), rather than the agent (themselves), these shop-owners obscure the fact that some action on their part is necessary for the shops to close. This discursive choice strategically removes the responsibility from the shop-owners and inevitably enhances the causal link between the government’s decisions and the closures.

The signs classified in the second subcategory (34 signs) also refer to “the government’s decision” (απόφαση της κυβέρνησης) as the reason behind the closure. However, they additionally include formulations that stress the business owners’ support and/or compliance with governmental imperatives. Examples (2) and (3) are typical of this category.

(2)

Και το κομπολογάδικο μένει στο σπίτι σύμφωνα με το Υπουργείο Υγείας!!

The kompoloi shop also stays home in line with the Ministry of Health!!

(3)

Τηρώντας τις οδηγίες περιορισμού της εξάπλωσης του ιού το Kowloon παραμένει κλειστό.

Observing the instructions for the restriction of the virus’s spread Kowloon remains closed.

In (2) the closure is presented as a result of the shop-owner’s agreement with the authorities’ policies through the use of the construction σύμφωνα με translated here as “in line with”, but literally meaning “in agreement with”. In (3), on the other hand, the causal relationship is established through the use of the causal participle τηρώντας (τις οδηγίες) (‘observing (the instructions)’). Although what seems to be stressed in (3) and other similar instances in the data is compliance rather than agreement, we suggest that a closer look at this type of phrasing communicates stronger alignment with the government’s decisions than that expressed through causal constructions such as λόγω των μέτρων (‘because of the measures’). Unlike (1) the phrasing of (3) not only stresses the authors’ compliance with the measures via the semantic meaning of the verb τηρώ (‘to observe’), but also seems to flout the maxim of quantity by providing the reasons why the measures were introduced in the first place (to constrain the spread of the virus), which were known by the readers of the sign. Their repetition here could only be interpreted as an attempt on the part of the author to indicate that s/he considers them valid reasons for the action taken. Therefore, it appears that the author of (3) presents her/himself not only as complying with instructions but also as supporting a valid cause through his/her compliance.

Yet, it has to be noted that despite expressing stronger alignment with the government’s measures, the authors of (2) and (3) still opt for backgrounding themselves as agents of the closure through the use of metonymy. In both examples, the shop is personified and assumes an agentive role. Example (2) states that το κομπολογάδικο μένει στο σπίτι (‘the kompoloi shop stays at home’) while in Example (3) the store acts as the subject of τηρώντας (‘observing’). That is, in both cases it is the inanimate stores rather than their owners that are portrayed as following the governmental rules.

Example (4) further exemplifies linguistic means expressing not merely alignment but also explicit respect for and approval of the measures:

(4)

Eνθαρρύνουμε ενεργά το «μένουμε σπίτι» και παραμένουμε κλειστά.

We actively encourage “we stay at home” and we remain closed.

In (4) the shop-owner’s support for the government’s measures is expressed emphatically, with the closure framed as an indication of encouragement (ενθαρρύνουμε ‘we encourage’) of the governmental campaign and active participation in it.

What is noteworthy about examples like (4), is that, unlike previous instances of this category, they are constructed in ways that foreground the shop-owners’ agency. This is evident in the first person plural verb forms ενθαρρύνουμε (‘we encourage’) and παραμένουμε (‘we remain’). Moreover, παραμένουμε κλειστά (‘we remain closed’) is a case of metonymy similar to that observed in (2) and (3) above. However, this time the metonymy is reversed with the animate shop-owners portraying themselves as being in a state of closure. It could be suggested then that increased support for the government’s measures goes hand in hand with the authors’ willingness to assume an agentive role and project themselves as active participants in the government-initiated fight against the virus.

What could be considered as further evidence that the authors of the Greek signs placed in the second subcategory tend to position themselves as active supporters of the governmental campaign, is that they employed the slogan μένουμε σπίτι (‘we stay at home’) with considerable frequency (20.6%), either by printing their messages on the campaign’s official poster (Image 1) or by adding it to the text as a hashtag (Image 2). In contrast, the slogan emerged in only 6.6% of the signs within the first subcategory.

Image 1.
Image 1.
Image 2.
Image 2.