World Englishes and online tourism advertising

Ghana is a multilingual country in which over 80 indigenous languages are spoken (Adika, Reference Adika2012). English is used as the primary medium of instruction in schools, and is also widely used in the news media, government (Huber, Reference Huber and Mesthrie2008: 73), and as a language of cross-ethnic communication (Adika, Reference Adika2012: 152). In spite of the predominance of English, Ghanaian languages are actively promoted in television and radio media and Ghanaian identity is closely linked to the ability to speak a Ghanaian language. In addition, Ghanaian Pidgin English is also spoken as a lingua franca and is a marker of local identity.

In Fiji, English was introduced in the late 19^th century in the course of colonisation and is currently a second language variety that has been institutionalised (Hundt et al., Reference Hundt, Zipp and Huber2015: 690). As a consequence of the establishment of sugar plantations and the recruitment of indentured labourers from India, English has been in contact with other languages for a long time. English can be found in domains such as administration, the education system, and the media. Furthermore, English is used as a lingua franca between different ethnic groups (ibid.) Apart from English, the other two main languages spoken in Fiji are Fijian and Fiji Hindi (Schneider, Reference Schneider2007: 116).

In Jamaica, a local emerging variety of Standard English, Jamaican English (JE), is spoken alongside an English-lexicon Creole, Jamaican Creole (JC). Since English is Jamaica's official language, JE is widely used in formal domains such as government, education, and the media, while JC is relegated to private and informal domains. However, in recent times the division has been less clear-cut, and JC is increasingly used in contexts that were previously reserved for JE, particularly news media (Westphal, Reference Westphal2017). JC is an important marker of local identity and group membership and can increasingly be heard in political speeches and campaigns.

The only official language in the Maldives is Dhivehi. English does not have the status of an official language and is therefore absent from the domains of administration, law, and politics. However, English is employed in education, the media, and tourism, and is also frequently used in the workplace (Meierkord, Reference Meierkord2018: 4–5). The importance of Islam and the Qur'an in the country suggests that some Maldivians have command of Arabic.

Though several models have been proposed for describing World Englishes and the ways in which they relate to one another, the older models, such as Kachru's (Reference Kachru, Quirk and Widdowson1985) Three Circles Model and Schneider's (Reference Schneider2007) Dynamic Model, do not allow us to study the varieties at hand on equal terms. For example, in the Kachruvian model, Fiji, Ghana, and Jamaica are traditionally placed in the Outer Circle, whereas the Maldives would fall into the Expanding Circle. In reality, the Maldives appears to be transitioning from the Expanding to Outer Circle, and Meierkord (Reference Meierkord2018: 2) notes that the Maldives represents ‘a rapidly transforming society, in which English is fast establishing itself as a second language in large parts of the population’. Similarly, while Fiji, Ghana, and Jamaica (as former colonies) can be dealt with in Schneider's postcolonial model, the Maldives have never been a colony, and the development of English there differs considerably from the three postcolonial Englishes because of the lack of a settler population (Meierkord, Reference Meierkord2018: 4). Consequently, a key component of Schneider's (Reference Schneider2007) Dynamic Model is missing.

Buschfeld and Kautzsch's (Reference Buschfeld and Kautzsch2017: 113) Extra- and Intra-territorial Forces (EIF) Model makes use of the ‘notion of extra- and intra-territorial forces’ that are at work at all times in the development of postcolonial and non-postcolonial Englishes. The pair propose that tourism, as a largely extra-territorial influence, might be a key factor in the expansion and development of English, particularly with respect to the general role of tourism in a given country. They also acknowledge an intra-territorial dimension, especially ‘with respect to the question of how readily a country's inhabitants make use of English in interchanges with tourists’ (Buschfeld & Kautzsch, Reference Buschfeld and Kautzsch2017: 115).

In each of the countries studied here, English is the main language of tourism, and has become ‘the default language of choice for brochures, signposts, websites and other text-based information material for tourists’, as well as the main language for spoken interaction with tourists (Schneider, Reference Schneider2016: 2). In spite of this, local languages remain central to tourists’ travel experience in another country. Jaworski and Thurlow (Reference Jaworski, Thurlow and Coupland2010: 262) note that ‘tourist consumption is organized around the recognition and interpretation of various signs’ and that ‘[t]hese signs are frequently linguistic’. In their study of a guided tour in the Maori village of Whakarewarewa, Rotorua, NZ, the authors demonstrate that it is the place name itself that is the object of tourist consumption. Similarly, we could say that Instagram users adopt the tourist gaze when they look at pictures of beautiful scenery or cultural artefacts in the posts that are offered by tourism organisations online. Together with expressions in local languages, Instagram posts may constitute objects of tourist consumption.

In advertising too, English is one of the most frequently used languages. Even in countries where it is not spoken as a first language, English has become associated with modernity and globalisation (Piller, Reference Piller2001: 161). A country that advertises in English may draw tourists from all over the world who use English as a lingua franca. English use is not restricted to Standard British or American English and television advertisements in particular employ different varieties of English (Bhatia, Reference Bhatia, Nelson, Proshina and Davis2020). The mixing of English with local accents and words is meant to heighten trustworthiness (Bhatia, Reference Bhatia, Nelson, Proshina and Davis2020: 625), but it also lends the advertisement an exotic flavour. As in tourism, English in advertising often appears with other languages. Outer and Expanding Circle countries are most likely to use English for products and company names (Bhatia, Reference Bhatia, Nelson, Proshina and Davis2020: 621), but English is not usually used for traditional concepts (Martin, Reference Martin2002: 383). This suggests that while English plays an important role as a lingua franca, local identity is indexed largely via local languages.

Finally, hashtags ‘have become an integral part of social media communications’ (Rauschnabel, Sheldon & Herzfeldt, Reference Rauschnabel, Sheldon and Herzfeldt2019: 473), found on virtually all social media platforms, including Instagram. A hashtag ‘acts as a form of metadata labelling the topic of the posts so that it can be found by others’ (Zappavigna, Reference Zappavigna, Seargant and Tagg2014: 139). In her analysis of tweets about coffee, Zappavigna (Reference Zappavigna, Seargant and Tagg2014: 148) demonstrates that the ‘#coffee hashtag plays an important role in aligning users into communities of shared values about coffee’. Similarly, hashtags around topics such as travel, safety, or specific tourism destinations may serve a similar function in ‘proposing interactive bonds’ (ibid.: 142).

Data and method

The data here comprise Instagram posts created by tourism organisations from the Maldives, Fiji, Ghana, and Jamaica during the period March to June 2020. For each of the countries, we collected 20 Instagram posts from Instagram accounts such as Tourism Fiji, Visit Maldives, Visit Jamaica, and Ghana Tourism Authority, all pages hosted by the official tourism authorities of the respective countries. In addition to these, posts from official airlines such as Fiji Airlines were used. A selection of these posts is reproduced below. Since Instagram's fair use guidelines (Instagram, 2021) allows for the use of copyrighted work for education and research purposes, and since all the posts used were public, i.e. visible even to people without Instagram accounts, we are confident that the reproduction of posts should not raise any ethical concerns. In line with these fair use guidelines, we limit ourselves to text posted by the owners of the account, and do not make reference to comments made by Instagram users on each of the posts. Although Instagram is primarily a photo-sharing app, the focus of this paper is on language use. Specifically, we focus on the range of languages used in online contexts by authors producing texts in multilingual settings.

In the first phase of analysis, a total word count for the posts of each variety was calculated. Word counts included text captions, texts in pictures that were posted alongside the text, mentions, links, numbers, and hashtags. Since in the lattermost case, there are typically no spaces between words, as in #visitjamaica, each hashtag was counted as one word. In addition, words written in non-Roman scripts, such as Arabic, were included. Emoticons are comprised of different elements of punctuation and so these were excluded from the word count. Our analysis only considers the words written by the post author in the original public post and in original post captions. This is in keeping with Instagram fair use regulations. Content generated by private individuals is not included.

Each word was then coded into one of five categories. The first of these was Standard English, used for forms that were written in Standard formal English. The second category was Colloquial English, which designated forms such as chillin’, which are written using non-standard English forms shared across varieties of English. Sharedness is what allowed us to make a distinction between Colloquial English and Creole/ Pidgin English forms, which are often variety specific e.g. soon come in Jamaica. We are aware of the theoretical difficulties such a distinction poses, but it goes beyond the scope of the current paper to address this more fully. Furthermore, it was not possible to merge Creole/ Pidgin with the next category, indigenous language variety, since in Ghana, there is a Pidgin as well as several local languages, and such a merger could obscure the realities of language use in posts generated by Ghanaian authors. Thus, the label indigenous language variety was applied to forms that were taken from languages other than English and Creole or Pidgin that occurred in the posts e.g. sota tale in Fiji. The final code was foreign language, which was applied to forms written in a language other than English or one of the languages spoken as a first language in the community in which the post was generated. In the current paper, we present a quantitative analysis, in which we compare the overall frequencies of the different codes in order to gain insights into the distribution of language varieties in the texts. We also carry out a qualitative analysis of the data, in which we focus on possible explanations for the use of different languages used in the posts.