Linguistic ridicule and shifting indexical values on social media: The case of English in Hong Kong

Abstract

This article examines the online reaction to the linguistic performance of a pro-China Hong Kong singer-actor in a commercial where he speaks Hong Kong English. Paradoxically, the posters criticizing his English are Hongkongers themselves, while those showing admiration are mainland Chinese. Understanding this paradox requires an appreciation of the multiple and complex orders of indexicality through which the variety is evaluated and of the increasing use of linguistic evaluations as a proxy for political judgements in a society undergoing significant changes. An analysis of online comments and remixes associated with the commercial shows that Hong Kong social media users attribute a range of different indexical meanings to the celebrity's English to shame him for his perceived ‘betrayal’ and to reclaim a sense of social superiority over mainlanders in the face of unease about Hongkongers’ cultural distinctiveness. It unpacks complexities surrounding the deployment of language ideologies in societies experiencing sociopolitical upheavals. (Language ideologies, linguistic ridicule, orders of indexicality, polycentricity, social media, Hong Kong)*

INTRODUCTION

In late May 2019, William Chan, a Hong Kong-born singer and actor who has in recent years been active in show business on the Chinese mainland, was featured in ‘It's all about seconds’, an advertising campaign for Chanel's new J12 watch. In a forty-five-second black-and-white video, Chan, acting as a ‘brand ambassador’, first checks the time on the product he is wearing, then talks in English about how he embarked on his career as a singer, and finally reflects on the importance of particular ‘moments’, most notably, the moment when he hugs his mother.

Soon after being released on the social media pages of this French luxury brand, the video attracted numerous views and responses from predominantly mainland Chinese and Hong Kong users, who focused not just on the brand or its advertised product, but also on the celebrity. Strikingly, the comments about the celebrity from mainland Chinese tended to be positive. As can be seen in a screenshot taken from Bilibili,¹ a popular Chinese video sharing platform (Figure 1), the comments floating at the top of the screen as the video is played, referred to as ‘danmu (彈幕) comments’ (see Zhang & Cassany 2020), are uniformly favourable. William Chan's spoken English is described as ‘very pleasing to the ear’ (好好听) and ‘spellbinding’ (苏了).

Figure 1. Screenshot of the ad uploaded to Bilibili.

Those from Hong Kong users, by contrast, tended to hold the star up to ridicule for, among other things, the quality of his English, which has the phonological and grammatical characteristics of ‘Hong Kong English’ (e.g. Li 2000; Hung 2002; Setter, Wong, & Chan 2010). The video became not just a trending topic, but also a target for various kinds of parodies and remixes, attracting further mocking, discussion, and even mainstream news coverage, an example of which, from the online news site Sky Post,² is shown in Figure 2. The headline in the figure notes explicitly that the singer-actor's ‘Hong Kong English’ (港式英文) is embarrassing to the extent that one cannot finish watching the commercial. This negative assessment is reinforced in a comment displayed in the top right of the image, attributed to a ‘netizen’ (網民): 我到底睇咗啲咩 ‘What on earth have I watched’ along with a see-no-evil monkey emoji . In the subheading highlighted in pink, a remix produced by a social media influencer nicknamed Uncle Siu (蕭叔叔) is promoted as a sharp contrast to the original version because of the influencer's ‘proper’, ‘aurally orgasmic’ English.

Figure 2. Main image of an online news article about the ad.

Hong Kong social media users’ reactions to the celebrity's linguistic performance in the ad, which was released a few years after the 2014 Umbrella Movement and just weeks before the first large-scale protest of the 2019 Anti-Extradition Law Amendment Bill Movement,³ raise a series of interesting questions, such as: Why do Hongkongers ridicule a locally born and bred singer-actor for using their own vernacular, which is, however, praised by others who do not share it? To what extent is the ridicule bound up with the broader sociolinguistic, cultural, and political conditions in Hong Kong, as well as the cultural and political conflicts between the city and mainland China?

As part of a larger project investigating linguistic ridicule (see Chau 2021), this case study analyzes the original video featuring William Chan, two parodic remix videos made in response to it, and the comments from Hong Kong users on social media associated with both the original video and the remixes. Informed by Blommaert's (2010) notions of orders of indexicality and polycentricity, the analysis uncovers how the online users construct and appropriate competing values associated with the celebrity's way of speaking to disparage him for his perceived ‘disloyalty’ to the city and affiliation with mainland China at a time of intense social and political upheaval during which Hongkongers were anxious that their ‘distinctiveness’ (Bourdieu 1984) from mainland Chinese was under threat. The significance of this study lies not only in its contribution to a growing strand of work on digitally mediated linguistic ridicule, but also in its ability to shed light on the role of language and language ideologies in the sociopolitical changes that have been occurring in Hong Kong over the past decade and continue at the time of writing.

BACKGROUND

The complexity of indexicality

What lies at the heart of the present study is the notion of indexicality, the connection between language and its social meanings. The concept can be traced to semiotician Charles Sanders Peirce's (1932) theory of signs, which states that an index takes its meaning through its association with an object in the external environment, and it has since been taken up and developed in sociolinguistics and linguistic anthropology. Silverstein (2003), for instance, proposes indexical order as a framework with which to understand how indexical associations come into being. According to the framework, first-order indexicality is concerned primarily with the connection between a linguistic form and its users’ basic sociodemographic characteristics (e.g. place of origin). Second-order indexicality arises when this connection starts to be noticed, commented upon, given social meanings, and used for identity work. Third-order indexicality develops when the second-order index becomes a stereotype, associated not just with the original sociodemographic characteristics but with certain ‘types’ of people with particular personalities, attitudes, and values—what Agha (2007) calls ‘characterological figures’. An oft-cited example is Johnstone, Andrus, & Danielson's (2006) study of Pittsburghese: the correlation between monophthongal /aw/ and working-class male speakers in Pittsburgh establishes first-order indexicality. It is not until the users notice this correlation, assign such values as incorrectness and masculinity to the phonological variant, and perform style shifting that second-order indexicality takes place. This socially meaningful feature is then increasingly linked to Pittsburgh and used consciously to perform local identity as a third-order index. More recently, T. K. Lee (2023) has drawn on the framework to understand how Singlish has evolved from a first-order index, with its formal features associated by scholars with less educated, working-class residents in Singapore and linguistic hybridity in the country; to a second-order index, which speakers use to signal in-group membership and which the local government stigmatizes as ‘broken’ English; and finally to a third-order index, which is appropriated and commodified to evoke ‘Singaporeanness’.

Indexical processes are not always, however, as linear as they seem in Silverstein's model. While first-order indexicality provides a basis for second-order indexicality, the social meanings that accrue to an index are subject to constant (re)construal based on the different ideologies they come into contact with (Snell 2017) and thus an index can be given different meanings even within a single community (Johnstone & Kiesling 2008). These multiple context-dependent associations are constitutive of what Eckert (2008:454) calls an indexical field, a ‘constellation of ideologically related meanings’. One example Eckert offers is the /t/ release in American English, which can be associated with being angry, polite, British, and gay, depending on the context in which it occurs.

Indexicality also tends to be hierarchically ordered. In this regard, Blommaert (2010:38) introduced the concept of orders of indexicality: sets of norms that ‘operate within larger stratified complexes in which some forms of semiosis are systematically perceived as valuable, others as less valuable, and some are not taken into account at all’. As people move from one environment to another with a different set of norms, the communicative resources they carry tend to be valued differently. For instance, it is considered normal for Mexican students who have migrated to the United States to use English, but abnormal when they move back to Mexico, where the use of Spanish, indexical of the national mestizo identity, is expected (Despagne 2019). Within a neighbourhood, such as Brugse Poort in Ghent, (Standard) Dutch is highly valued in primary schools, but it is not in mosques where Arabic has a higher status (Blommaert, Collins, & Slembrouck 2005).

Adding to this complexity is the fact that a single environment is often characterized by polycentricity, meaning there exist multiple orders of indexicality arising from more than one evaluative authority towards which people orient (Blommaert 2010). In an analysis of comedic skits in Hawai'i, for example, Furukawa (2018) argues that the same linguistic performance might attract conflicting evaluations: on the one hand, Hawai'i Creole is devalued and associated with violence and unprofessionalism when the audience orients towards the centre of English, but, on the other hand, it is indexical of virtues such as patience and honesty when others orient towards the Creole centre. Like Furukawa's study, this study focuses on the orders of indexicality that emerge in a polycentric environment and further illustrates how they intersect with one another. Specifically, it disentangles the indexical values Hong Kong social media users attribute to William Chan's use of Hong Kong English in the ad described above, as well as the broader ideologies these users (re)produce through their practices of linguistic ridicule.

Linguistic ridicule on social media

I use linguistic ridicule as an umbrella term to include all acts of insulting, denigrating, or making fun of someone's language, overtly or covertly, intentionally or unintentionally. Invariably, the objects of ridicule are not only the person using the language, but also the group to which the person belongs. One of the ways in which linguistic ridicule is accomplished is through language mocking, whereby people imitate how others speak or write. Applying Ochs's (1990) concepts of direct and indirect indexicality, Jane Hill (e.g. 1995, 1998) describes how the everyday use of Mock Spanish enables monolingual Anglos in the United States to project a fun-loving and easygoing persona and at the same time to depict Spanish speakers, the outgroup, in a bad light, thereby reproducing white supremacy. Hill's seminal work has inspired studies on Mock Ebonics (Ronkin & Karn 1999), Mock Asian (Chun 2004), and Hollywood Injun English (Meek 2006). As this body of work has shown, mocking is imbued with broader social ideologies. Although many seemingly casual and light-hearted stylizations turn out to be acts of stereotyping and discrimination by the dominant groups, they can also be exploited by others to disrupt taken-for-granted social hierarchies. In her work on how online users creatively poke fun at a Chinese celebrity's translanguaging practices, for instance, Gao (2022) argues that ‘mock translanguaging’ serves as a tool whereby everyday people can undermine the status performances of elites in China.

Linguistic ridicule can also be performed through metalinguistic commentary, explicit comments about the way others speak. Because of its participatory affordances, social media has become a key site for metalinguistic commentary. Aslan & Vásquez (2018), for example, examine how users construct an English-speaking youngster's speech on an American talk show as non-standard and unintelligible through comments on YouTube. Jones (2013), in his analysis of metalinguistic discourse surrounding the English language produced on a gay discussion forum in Hong Kong, discovered that a considerable number of the users criticize the English of other users as a way to challenge their social status and even the authenticity of their ‘gay’ identities. In making metalinguistic comments, online users in both studies actively deploy orders of indexicality as tools for othering (Rymes & Leone 2014).

One means of metalinguistic commentary that particularly exploits the affordances of digital media is remixing, defined broadly as a process of creating a digital text by appropriating and reworking an existing one (Jones & Hafner 2021). In one of the few studies in the existing scholarship, Androutsopoulos (2020) scrutinizes YouTube videos featuring the spoken English of a former Prime Minister of Greece, with a focus on how subtitles in ‘Hellenised English’, that is, English written in the Greek script, are deployed to accentuate the leader's linguistic, and by extension, political inadequacy. In another study, Jones (2023) explores ways in which digital technologies enable TikTok users to playfully appropriate the voices of others and re-present them with their own bodies through lip-synching. Implicit in some of these ludic videos are the performers’ evaluations of the ways groups other than their own speak. What is salient in both studies is that remixers draw on and juxtapose semiotic resources carrying different social meanings to create an incongruity, thereby producing implicit and often comic commentaries on particular ways of speaking and the people employing them.

Examining different practices of linguistic ridicule offers a wealth of insights into the broader beliefs about language and society of those engaging in them and the perceived boundaries between these people and those they ridicule. In Hong Kong, where the present case is situated, Chau (2021) has analyzed how social media users collaboratively contribute to the enregisterment of the ‘fake ABC (American-born Chinese)’ variety, a pseudo-variety perceived to be spoken by locals pretending to be linguistically superior, well educated, and from a higher socioeconomic class. In doing so, these users exploit deeply entrenched ideologies of authenticity, linguistic purity, and ‘standardness’ to differentiate themselves from that social group. It should be noted that these collaborative acts of ridicule, as well as those discussed in this article, are often characterized by a high level of linguistic playfulness and creativity, which are integral to both the promotion of solidarity for those ‘in the know’ and the exclusion of others (see also Jones & Chau 2022). In contrast to Chau's (2021) study of the perceived ‘inauthenticity’ of the ‘fake ABC’ variety, this study focuses on a case in which a celebrity's ‘authentic’ use of a local vernacular is ridiculed by viewers who share this vernacular. This apparent paradox cannot be explained without first discussing the wider sociolinguistic context of Hong Kong and the ambiguous status of Hong Kong English.

Ambivalence towards Hong Kong English

In Hong Kong, a former colony of Great Britain and, since 1997, a Special Administrative Region of China, three major languages are used: Cantonese, the first language of the majority of citizens; Putonghua, the national language of China; and English, a co-official language which most residents have learned as a second language and have used extensively in formal domains for years. Despite the history and prevalence of English in the city, it was not until the 2000s that the concept of ‘Hong Kong English’ received serious attention. In an edited volume on this subject, Bolton (2002) argues that Hong Kong English, used primarily by Cantonese-English bilinguals, is just as deserving as other Asian English varieties of recognition as an autonomous variety. Linguists have identified and documented a range of features characteristic of this local vernacular, such as: phonological features, including realization of /iː/ and /ɪ/ as [i] (e.g. heat-hit [hit]), substitution of [w] for /r/ (e.g. very [wɛri]), consonant cluster simplification (e.g. it's [ɪs]), and syllable-timed rhythm, with a similar amount of attention paid to both stressed and unstressed syllables; and grammatical features, including the transitive use of intransitive verbs (e.g. He didn't reply me) and pseudo passives (e.g. It divided into three parts) (e.g. Li 2000; Hung 2002; Setter et al. 2010). Due to influence from Cantonese, the mother tongue of the speakers, Hong Kong English is in many ways distinguishable from China English, spoken by mainland Chinese, who use Putonghua as a common language (though they may also speak other Chinese languages). As presented in the findings below, many phonological and morphosyntactic features outlined above are found in William Chan's speech and commented on by viewers.

Notwithstanding the scholarly efforts to legitimize Hong Kong English, this local variety is far from accepted by society at large. As observed in previous studies on language attitudes (e.g. Tsui & Bunton 2000; Chan 2017; Lai 2020; Tsang 2020), there has been a pronounced preference among teachers and students for exonormative varieties, most notably British English, introduced by the colonial government and used as an educational model, but increasingly also American English, as a result of the popularity of the American mass media. The preference is perhaps unsurprising given that ‘standard’ English has long been regarded as ‘the dominant symbolic resource’ for academic and career success (Lin 2000:64) and associated with middle-class status, whereas ‘non-standard’ English, including Hong Kong English, is sometimes viewed as a symbol of being working-class and/or less educated.

At the same time, Hong Kong English seems to enjoy some degree of ‘covert prestige’ (Trudgill 1972), especially when it is linked to an indigenous identity. As early as the late 1980s, nearly half of the male participants in Bolton & Kwok's (1990:170) study were found to prefer a local English accent to British and American accents, indicative of their desire to ‘speak like “Hong Kong Man”’. This local identity, contrary to Mathews's (1997:13) speculations, did not ‘fade into history’ after the reestablishment of Chinese sovereignty. In the face of the central government's increased control over local affairs as well as a surge of mainland immigrants and tourists competing with locals for resources, in fact, some citizens—especially younger ones—developed ‘anti-China’ sentiments and made an extra effort to assert their distinct identity (Ma 2015, 2020; Lee & Chan 2022). In recent years, major protests have triggered the rise of various forms of localism (e.g. Chen & Szeto 2015; Veg 2017; F. Lee 2020)—including an orientation towards more local ways of speaking Chinese and English, as well as the emergence of ‘Kongish’, a creative translanguaging practice which employs a myriad of local resources (e.g. traditional Chinese characters, Romanized Cantonese, and Hong Kong English) along with multilingual and multimodal resources (Li Wei, Tsang, Wong, & Lok 2020). In a survey conducted after the Umbrella Movement, Hansen Edwards (2016) found that more respondents claimed to use Hong Kong English themselves, regarded it as a legitimate variety, and associated it with their local identities. These findings have been corroborated by more recent research, pointing to the vernacular being ‘more widely accepted and seen as a marker of a localized Hong Kong identity’ (Ladegaard & Chan 2023:267; see also Hansen Edwards 2019).

How Hongkongers evaluate the local English variety appears to be contingent on such factors as who is speaking it and the context in which it is spoken. Jones & Chau's (2022) study provides a case in point: While the Umbrella Movement protesters celebrated their own use of Hong Kong English, framing it as an emblem of local identity, they also made fun of police officers who spoke the same variety, framing it as proof of a lack of education, social status, and even moral virtue. This seeming contradiction illustrates that the deployment of orders of indexicality in evaluations of others’ speech often has less to do with the language itself and more with the social or political agendas underlying these evaluations.

In brief, there exist complex attitudes towards Hong Kong English, which is sometimes despised and sometimes valorized by its users. This study, however, is interested not so much in what online users think about the variety as how they appropriate this mobile resource and its values for ridicule and othering against the backdrop of political, cultural, and economic insecurities in society.

Hongkongers and mainland Chinese: Past and present

Before the transfer of sovereignty from the United Kingdom to the People's Republic of China, Hongkongers tended to believe that the English language brought by colonialism, the affluence brought by capitalism, and liberal values including human rights and democracy set them apart from Chinese on the mainland (Mathews 1997). Their sense of cultural and economic superiority was boosted by the popular media's representation of the city as an advanced international financial centre and the mainland as a politically and economically backward society (Ma & Fung 2007). Following the handover, however, this local identity, based on the perceived differences between themselves and mainlanders, came under threat. Economically, the city was hit hard by one financial crisis after another, whereas China's economy grew steadily and started exerting its influence in the world (Yew & Kwong 2014). Politically, despite ‘One country, two systems’, a principle granting Hong Kong a high degree of autonomy, the central government's growing intervention made citizens worry that Hong Kong would lose its distinctiveness and become ‘just another Chinese city’ (So 2018:502).

Of particular relevance to this study are Hongkongers’ feelings about the loss of distinctiveness in relation to social status and economic prosperity—often associated with the conspicuous consumption of luxury brands. In the 1980s and 1990s, mainland Chinese became known for their production and consumption of ‘counterfeit’ luxury goods (Bian & Veloutsou 2007), and so their attempts to gain social status through consumption were seen as crass and inauthentic in comparison to Hongkongers’ ‘genuine’ wealth and taste. In recent decades, however, with flocks of prosperous mainlanders purchasing brand-name products in Hong Kong, such class- and culture-based distinctions have begun to break down. Unable to match the spending power of these mainland consumers, many Hongkongers have developed a sense of status anxiety (e.g. Yam 2016; Wang, Joy, Belk, & Sherry 2019; Joy, Belk, Wang, & Sherry 2020), the fear of losing a desirable social position to another group.

The significance Hongkongers have attached to such cultural objects as ‘standard’ English and ‘genuine’ luxury goods as mentioned above can in part be understood through Bourdieu's (1984) concept of distinction. In Distinction: A Social Critique of the Judgement of Taste, Bourdieu notes that taste is neither personal nor natural, but rather rooted deeply in and reflective of one's class. While in the past Hongkongers were able to, through their expression of preference for these status symbols, exert power or ‘symbolic violence’ over mainland Chinese, to whom they felt culturally and economically superior (Poon 2010; Joy et al. 2020), they have difficulty doing so now as a result of their counterparts’ greater access to these resources. Conceivably, to maintain the sense of distinction, they may resort to challenging the cultural capital associated with the symbols and reframing discussions about wealth, power, taste, and local identity in new ways, which is partly what some online users are doing in their commentary on this Chanel ad featuring a local celebrity-turned-mainland loyalist.

FINDINGS

In what follows, I first present my analysis of the ad uploaded to the official Facebook page of Chanel. Having retrieved 6,408 comments using ‘Export comments’, a web-based extraction tool, for the purposes of this article, I sampled and read the top 2,000 comments based on popularity with an aim to obtain a fuller picture of the evaluations of the video and of the way the celebrity speaks in particular. Among the sampled comments, I judged 1,064 of them to have been posted by Hong Kong users based on the presence of at least two of these features: Cantonese romanization in their Facebook names; traditional Chinese characters and Cantonese-based grammar in the comments; references to local culture and knowledge; and explicit identification with Hong Kong in their profiles. From there, I identified 304 comments with clear instances of linguistic ridicule. 37% of them were concerned with the celebrity's English as a whole. Others were specific to its phonological features (55%), lexical/grammatical features (6%), or a combination of both (2%). To conduct a closer interpretive analysis (Vásquez 2019), I began by coding these comments with MAXQDA, a qualitative analysis software program, for themes related to linguistic features being ridiculed (e.g. consonant cluster simplification, consonant substitution) and social meanings these features index (e.g. standardness, class). I then interpreted the patterns through the lens of Blommaert's (2010) concepts of orders of indexicality and polycentricity, and situated the findings within the broader context of Hong Kong as presented earlier. In the second part of this section, I turn to the analysis of two remix videos, which are richly illustrative of the diverse techniques the creators employ for ridicule. To protect the commenters’ privacy, their usernames are replaced with pseudonyms in the examples.

Competing indexical values of Hong Kong English in the ad

At first glance, the mockery and criticism in the comments on the original post revolve around the non-standardness of William Chan's Hong Kong English. In example (1), Daisy expresses overt distaste for the variety Chan is using, constructing his way of speaking as defective.

1. (1) Daisy

   港式英文到爆 好心佢知自己要拍就 upgrade 吓自己啦頂

   ‘Completely Hong Kong English. He should have improved his English before shooting, for God's sake!’

Similarly, in examples (2)–(4), the posters highlight the phonological features they find problematic in the speech.

1. (2) Charlotte

   [Margaret] par-ti-ci-pa-ted 逐粒音嘔出黎

   ‘participated – uttered forcefully syllable by syllable’

2. (3) Jacky

   [Macy] I willi wan to 猜

   ‘I really want to try’

3. (4) Yolanda

   [Alex] Goodest accent by Taxi Chan laugh die

   “I 拍 titcipated in a competition”

   “I weely want to 搓”

In example (2), Charlotte finds fault with the emphasis Chan places on every syllable in articulating participated and visually foregrounds the syllable-timed rhythm with hyphens. In example (3), Jacky draws Macy's attention to the substitution of [w] for /r/ in really, as well as to the consonant cluster simplifications in want and try, respectively represented by wan and 猜, a Chinese character that sounds like his pronunciation of try (Jyutping: caai1). In example (4), Yolanda indicates the latter reduction with another Chinese character 搓, making an implicit intertextual reference to an incident in which a pro-Beijing lawmaker's pronunciation of try our best was stylized as 搓 (a verb literally meaning to rub) our breast.

Others criticizing Chan's verbal performance focus on the syntactic and lexical aspects. For instance, in example (5), Leslie finds participate being treated as a transitive verb intolerable. In example (6), Bill wonders whether waste is the most suitable collocation with time in this particular context.

1. (5) Leslie

   [Akina] we watched that. Lol I couldn't bear his overly expressive facial expressions and his omission of in after the verb participate

2. (6) Bill

   Waste your time? You means (sic) “spend”?

While one might be inclined to attribute these exercises in ‘verbal hygiene’ (Cameron 2012a) to the ideology of ‘standardness’, which has been prevalent in discussions of language (especially English) in Hong Kong for the past three decades (Chau 2021), this attribution is far from complete and satisfactory. For one thing, many ‘non-standard’ features noticed by the posters are, as mentioned earlier, consistent with descriptions of Hong Kong English documented in the literature (e.g. Li 2000; Hung 2002; Setter et al. 2010); the attribution does not account for the selective acceptance of it as a legitimate variety among some locals (e.g. Hansen Edwards 2016, 2019; Ladegaard & Chan 2023). For another, the negative feedback directed at Chan from his fellow Hongkongers seems at odds with a common view that local bilinguals should stick to Cantonese-accented English to show themselves as ‘authentic’ Hongkongers (Jenks & Lee 2021). Previous studies have, in fact, shown that those who fail to do so—speaking like ‘foreigners’ instead—tend to be scorned (e.g. Jenks & Lee 2016; Chau 2021).

What emerges from a close examination of the metalinguistic discourse in the same post is that there are also qualities other than non-standardness ascribed to William Chan's spoken language. Some posters consider it not ‘decent’ enough for an international ad campaign. As can be seen in examples (7) and (8), Joshua and Tess are expressly worried that his ‘bloody embarrassing’ English might be viewed negatively by ‘the world’ and ‘the international community’.

1. (7) Joshua

   [Jamie] 一間全世界以為香港人 d 英文咁 7

   ‘The world might then think Hongkongers’ English is that bloody embarrassing.’

2. (8) Tess

   channel (sic) 你做乜咁玩的士陳 賣錶信息比唔到，但國際社會馬上知道香港人英文程度每況越下，情何以坎

   ‘Why is channel (sic) making fun of Taxi Chan? The international community knows nothing about the watch but Hongkongers’ falling English standards. How pathetic!’

Of note here is that Tess makes reference to ‘falling language standards’, a complaint rehearsed in academic, official, and public discourses (Lin 1997) and thought to be chiefly the result of a failure to conform to exonormative norms (Tsui & Bunton 2000). It is possible that these posters’ concerns are motivated by and reflect popular assumptions that ‘standard’ English is a ‘valuable commodity in the global linguistic market’ (Cameron 2012b:360) and that it gives Hong Kong, which prides itself on being a ‘world-class city’, a competitive edge over other places in Asia. ‘Non-standard’ Hong Kong English, when displayed to the global audience, brings shame to the city due to its lack of prestige and international legitimacy.

Related to this are comments such as that of Alfred in example (9), who opines that William Chan's English is not ‘classy’ enough for Chanel, a luxury brand for high-end consumers, making an explicit connection between English and class.

1. (9) Alfred

   我同我身邊的所有朋友對 Chanel J12 用陳偉霆做代言人感到非常失望！冇內涵！冇演技！低學歷！英文發音唔準！佢沒有能力代言任何一個 Chanel 產品

   ‘All my friends and I are terribly disappointed with Chanel inviting William Chan to be a J12 spokesperson! No depth of knowledge! No acting skills! Low education level! Inaccurate English pronunciation! He is not capable of representing any Chanel products.’

In this example, Alfred asserts, with a series of exclamation marks, that the celebrity's poor spoken English, and, by extension, his being a less-educated person who has ‘no depth of knowledge’ and ‘no acting skills’ are incommensurate with this upmarket brand. This poster, together with all his friends (as he claims), links Chan's Hong Kong English to low educational attainment and lower socioeconomic classes, while reproducing a societal view that ‘standard’ English is associated with a better education, job, and lifestyle (Lin 2000; Jones 2003; Jones & Chau 2022). The lack of ‘class’ associated with the language then ends up tarnishing the brand's reputation, as in example (10).

1. (10) Charlie

   I can't (sic) never image (sic) this ad is from chanel… Obviously high priced =/= high in class

The strategy of drawing a connection between the celebrity's English with non-standardness and inferiority, however, runs the risk of backfiring, since the vernacular is an identity marker that the posters share and possibly take pride in. To avoid this risk, others choose instead to attribute inauthenticity to Chan's identity, language, and even the brand he works with.

1. (11) Kayla

   的士陳都唔係香港人點會係港式英文

   ‘Taxi Chan isn't a Hongkonger. Why does one think he speaks Hong Kong English?’

2. (12) Carrie

   [Shirley] mainland English 呀

3. (13) Joey

   Chanel => Channel

In example (11), Kayla refuses to admit that ‘Taxi Chan’, a nickname given to the celebrity when he was active in Hong Kong, is a ‘Hongkonger’ speaking ‘Hong Kong English’. The denial of Chan's local identity can be traced to the moment when he ‘gave up’ the city, based his career in mainland China, and pledged allegiance to the Chinese Communist Party by sending patriotic wishes on Weibo.⁴ In the eyes of Kayla, then, Chan is not ‘qualified’ to be a local, but is rather a mainlander (‘大陸人’, as another poster puts it) speaking to his compatriots on the Chinese mainland. In example (12), Carrie deauthenticates the way Chan speaks by jokingly referring to his local vernacular as ‘mainland English’. Both comments send a message that an authentic user of a language must be loyal to the values and people it represents. Such deauthentication is extended to the brand Chan represents in the ad. In comments such as those left by Tess (example (8)) and Joey (example (13)), Chanel is framed as a counterfeit brand, with its name creatively respelled as ‘Channel’. This activates the long-established association between mainland China and counterfeit products mentioned above (Bian & Veloutsou 2007).

Some go so far as to not recognize Chan's right to use English at all, as in example (14) below.

1. (14) Stephen

   講返中文啦，橫掂條柒頭都係打中国巿場之嘛

   ‘He'd better speak Chinese. Anyway, the shithead aims merely at the (mainland) Chinese market.’

In example (14), Stephen suggests that Chan, ‘the shithead’, should use Chinese, a language that appeals to the ‘Chinese’ market. It is noteworthy that all of the words in this comment are Cantonese in traditional Chinese characters, except for 中国 ‘Chinese’ in simplified characters, which are neither formally taught nor widely used in Hong Kong. With the strategic use of both scripts, the poster creates an implicit ‘us’ (Hong Kong)/‘them’ (mainland China) distinction and makes it clear that it is mainland Chinese to whom the celebrity is promoting the watch. The comment also strengthens the idea that Hong Kong English should be reserved for those faithful to the city and its citizens, and not for ‘traitors’ like Chan.

The findings presented thus far demonstrate how Hong Kong social media users, similar to those in previous research (e.g. Jones & Chau 2022), engage in ridicule convivially with an array of resources such as respelling and special Chinese characters which outsiders may not be able to decode or use. This kind of collaborative language play provides a creative means for Hongkongers to strengthen their in-group solidarity through ludic engagement with their shared variety. More importantly, the findings reveal the complex existence of multiple orders of indexicality that they orient to in assessing Chan's linguistic performance. These orders are concerned with and promote an array of ideas about international legitimacy, class, populist authenticity, and political loyalty. The intersection of these orders involving centres within Hong Kong and beyond thus creates a seemingly paradoxical situation where sometimes Hong Kong English evokes a Hongkonger identity and sometimes it does not. Situated in an international ad campaign for a luxury watch, this ‘non-standard’ variety—as opposed to an exonormative one such as British English—is associated with a lack of status and power. Spoken by a ‘counterfeit’ Hongkonger thought to have betrayed other Hongkongers at a time of sociopolitical crisis, the variety indexes non-localness. Perceived to be directed towards mainland Chinese, it is rendered valueless.

By imputing negative qualities on the way William Chan speaks and claiming that he is actually a mainlander speaking ‘mainland English’, as Kayla, Carrie, and some other posters do, the online users also take the opportunity to claim superiority by suggesting that mainland Chinese have poorer English, less education, and lower social status than Hongkongers, as well as to impugn the taste of those idolizing Chan. Such disparagement is evident in examples (15)–(17).

1. (15) Samson

   大陸人唔太 care d 英文反正聽唔明

   ‘Mainlanders don't care about his English. They don't understand it anyway.’

2. (16) Fiona

   […] Well yeh yeh yeh the whole world knows that China has the strongest purchasing power! So what?! money can't buy class even tho u are using Chanel from head to toes (sic)

3. (17) Anson

   香港垃圾，大陸寶藏

   ‘Hong Kong trash, mainland treasure’

In example (15), Samson maintains that mainlanders cannot tell whether Chan's English is good or bad due to their fundamental lack of proficiency. In example (16), Fiona stresses that just because mainland Chinese are now wealthy consumers wearing luxury goods ‘from head to toe’, this does not mean they have become refined. Her disdain for them coincides with locals’ complaints about mainland tourists’ lack of etiquette and knowledge about the luxury products they buy (Wang et al. 2019; Joy et al. 2020). In example (17), Anson uses the phrases ‘Hong Kong trash’ and ‘mainland treasure’ to insinuate that people on the mainland enamoured of Chan, who is unpopular in the city and cannot speak ‘well’ in the ad, have bad taste. Taken together, these posters’ deployment of taste (Bourdieu 1984) can be seen as an attempt to challenge the legitimacy of mainlanders’ access to status symbols including ‘proper’ English and luxuries like the watch Chan is promoting.

Manipulating orders of indexicality through remixing

Ridicule is found not just in the comments on the original video, but also in the remixes. The first remix video I analyze here was created by Uncle Siu, a Hong Kong-born English educator with a law degree obtained from a local university. The remix video is identical to the original ad, except the soundtrack is replaced with one featuring Uncle Siu's exaggerated Received Pronunciation (RP). The video, uploaded to his Facebook page with over 500K followers, attracted approximately 4K comments, 3.6K shares, and news coverage (see Figure 2 above).

In the post where the remix video is embedded (Figure 3), Uncle Siu claims with five facepalm emojis () that he dubbed the video for William Chan ‘in response to the popular demand’, implying that most viewers find the celebrity's English in the ad unbearable.

Figure 3. Screenshot of the first remix video.

By replacing Hong Kong-accented English with RP-accented English, the remixer draws viewers’ attention to and reinforces the long-standing class differences associated with these two spoken varieties in society. That is, Hong Kong English signals low class, whereas ‘standard’ English, in this case RP English, signals the opposite.

1. (18) Pamela

   由的士瞬間變 Ben 士

   ‘A taxi is transformed instantly into a Benz’

In example (18), taken from comments on this remix video, Pamela likens the former, what Chan speaks in the ad, to a taxi, and the latter, what the influencer speaks in the remix, to a Mercedes-Benz. Of note here is the parallel between a luxury item and a ‘prestigious’ English variety. In this commenter's view, RP English, similar to a luxury product, can be ‘purchased’ to signal class and social status by consumers who have ‘good’ taste and know how to appreciate it.

More remarkably, by putting a different voice into William Chan's body, the remixer establishes incongruity, which arises not only from the visual-auditory components, but also from the contrasting values these components evoke. Not only is Chan made to use a voice that does not come out of his own body, but he is made to use a ‘posh’ voice that he, perceived as being from a working-class background, does not ‘deserve’. These layers of incongruity, which make Chan doubly ridiculous, are noticed by commenters such as Elsa and Frankie in examples (19) and (20).

1. (19) Elsa

   完全係 UK 聲，MK look 嘅典範

   ‘A perfect example of a UK voice and an MK look’

2. (20) Frankie

   靚聲配···唉·做乜插支靚花係堆米田共度呀！

   ‘Beautiful voice with… Sigh…Why do you poke a beautiful flower into the dung!’

In example (19), Elsa hints that Chan's ‘MK look’ does not match the influencer's ‘UK voice’. Here, MK, an abbreviation for Mong Kok, a working-class district in Hong Kong, is used to denigrate the status of Chan. The incompatibility between the voice and body is echoed and played up in Frankie's comment (example (20)), in which he uses a Chinese figurative expression (一朵鮮花插在牛糞上) to associate the remixer's voice with ‘a beautiful flower’ and Chan with ‘dung’.

Taken together, RP English is used in this video perhaps not so much to suggest its perceived superiority over Hong Kong English or to merely make fun of William Chan's substandard English. Rather, the juxtaposition of an RP-speaking voice and Chan's body serves to ridicule Chan indirectly for being inauthentic, inferior, and incapable of representing Chanel, a classy brand.

The second remix video I analyze was produced by Derek Wong, an actor based in Hong Kong, where he received his education before pursuing an undergraduate degree in the United States. The video, uploaded to the actor's Facebook page and YouTube channel, received over 4.6K comments and 6K shares.

In this video, Wong impersonates William Chan by copying his hairstyle, outfit, and body language. More significantly, he takes the localness of Chan's speech and actions to their ‘logical extreme’. At the beginning of the video, for instance, Wong links 12 o'clock to ‘What time is it, Mr. Fox?’, a popular childhood game in the city, by inserting b-roll footage (Figure 4). It is a moment when Mr. Fox catches players who fail to get past him and reach the destination, that is, ‘埋周(舟)’ (maai4 zau1). In the main footage, there are instances of intra-sentential English-Cantonese code switching (e.g. I ding (叮 ding1, ‘microwave’) my lunch box; the moment I hug my goo ma (姑媽 gu1 maa1, ‘father's sister’)), which are commonly used by local bilinguals, particularly when they communicate online (see e.g. Chau & Lee 2021). Throughout the video, subtitles in English and traditional Chinese are added. This multimodally constructed localness strikes a chord with viewers such as Samuel (example (21)), who reminisces about his old days playing the game, and Danny (example (22)), who finds the code-mixed parts ‘hilarious’.

1. (21) Samuel

   好好笑啊！ 我睇左好多次， 笑到肚子痛 12 o'clock, 十二點，「狐狸先生」跑出來捉人，係童年玩過的遊戲 ， 係回憶啊！

   ‘Hilarious! I've watched this video many times and laughed to death. 12 o'clock, “Mr. Fox” catching players - It's my childhood game and memory!’

2. (22) Danny

   Hilarious!!! I 叮 my lunch box…. the moment I hug my 姑媽…. it's all about time - lunch time! 好正

Figure 4. Screenshots of the second remix video.

What also makes this remix funny, as pointed out in Danny's comment, is the slogan ‘It's all about time—lunch time’, modified from the original one ‘It's all about seconds’. While building on the theme of the advertising campaign, the remixer trivializes it by foregrounding the mundane over the profound. Towards the end of the video, the modified slogan is complemented by a close-up shot of a microwaved lunch box half filled with rice with spareribs, an ordinary working-class local dish (Figure 5).

Figure 5. Close-up shot of a lunch box in the second remix video.

In contrast to the first remix, this one makes the celebrity's language all the more local, unrefined, and laughable through multimodal resemiotization. It ultimately pokes fun at Chan for speaking like an ordinary working-class citizen in the ad and thus for not being qualified as an ambassador for an international brand.

The results from the analysis of these two remix videos lend support to previous findings (e.g. Androutsopoulos 2020; Jones 2023). With digital affordances, remixers combine various semiotic resources, which would not have been possible simply by means of mocking, to establish incongruity at both intratextual and intertextual levels. In this analysis, the incongruity, which arises significantly from the juxtaposition of different orders of indexicality surrounding class, makes the videos humorous and Chan an object of ridicule. To a certain extent, remixes are akin to internet memes, digital texts characterized by humour, intertextuality, and anomalous juxtapositions (Knobel & Lankshear 2007). Albeit informal and funny, they are powerful tools with which to spread ideologies in digital environments (Wiggins 2019).

CONCLUSION

In this article, I have reported an analysis of a luxury watch commercial in which a Hong Kong celebrity speaks with what is widely understood as Hong Kong English, its remix videos, and their concomitant comments from Hong Kong social media users. Drawing on Blommaert's (2010) notions of orders of indexicality and polycentricity, I have shown that these users attribute conflicting indexical values to the spoken variety through mocking and metalinguistic commentary. On the one hand, the celebrity's English is associated with a variety spoken by a Hongkonger who comes from a lower social class, received little education, and fails to enjoy the prestige that comes with ‘standard’ English; but on the other hand it is associated with a variety spoken by a ‘fake’ Hongkonger. All of these values, motivated by the users’ orientations to standardness, class, authenticity, and loyalty, are called upon to ridicule the celebrity for not only his use of ‘improper’ English to project an international identity in the ad, but also his ‘selling-out’, given his relocation to and affiliation with mainland China. The undesirable qualities are appropriated in turn to denigrate mainland Chinese, especially those speaking highly of him in the comments on the commercial, for their lack of good taste. Remixing, notably through juxtaposition of different semiotic resources and orders of indexicality, is an important means through which to make the celebrity more ludicrous, spur more discussion, and perpetuate the ridicule in digital environments. As has been shown in the analysis, the collaborative practices of ridicule, often done in a playful and creative manner, enable these users not only to ‘other’ the ‘traitor’ and mainland Chinese, but also to create a sense of solidarity with fellow Hongkongers.

The present study does not intend to make generalizations regarding what Hongkongers think about the local variety of English, William Chan, or mainland Chinese. As Cutler (2020) rightly points out, viewership of online videos is skewed likely towards a particular group of people in a society and their commentary towards those willing to respond online within the group. Nevertheless, the case has provided a useful lens for analyzing how the Hongkongers who comment on these videos negotiate their identity vis-à-vis mainland Chinese in times of sociopolitical tensions. Over the past decades, Hongkongers have undergone drastic changes, from living in ‘a prosperous capitalist haven’ (Ma 2015:41) with their own lifestyle to being increasingly assimilated into and reliant on mainland China. Ridiculing Chan, along with associating him with a low-class mainlander, creates an opportunity for these users to distance themselves from someone ‘disloyal’ to the city and to reassert cultural superiority over their mainland counterparts.

This study adds to our understanding of how distinction through taste (Bourdieu 1984) is discursively accomplished and connected with language ideologies in contemporary Hong Kong society. As evidenced in the data, a strategy some users have possibly adopted to sustain the sense of distinction created in the past through conspicuous consumption is to reframe the display of taste—not as a matter of simply owning and wearing luxury goods, which mainland Chinese can afford nowadays, but rather as a matter of also talking about them in the ‘right’ language and accent. By deploying deep-seated ideologies surrounding English in society, the users disparage the brand ambassador for promoting the watch in an ‘unprestigious’ language variety and in turn belittle the mainland viewers who fail to notice any ‘problems’ with the way he speaks. In this regard, the study goes beyond identifying and teasing out the language ideologies which may have existed for a long period of time (e.g. Jenks & Lee 2016; Chau 2021) to highlighting their potential as tools for differentiation.

This study also contributes to our understanding of the commodification of language. The scholarly discussion of language as an economic resource is, of course, not new. There have been studies on such contexts as call centres (e.g. Cameron 2000) and heritage tourism (e.g. Heller, Pujolar, & Duchêne 2014). In spite of the valuable insights from these studies, much of the literature tends to focus on how language is sold for material profit in institutional settings, and less on how language is consumed by individuals and converted into symbols of class, luxury, and taste. In this study, I have provided empirical evidence for how online users associate English varieties with commodity brands of different values and qualities (Lai 2020). Specifically, the users consume ‘standard’ English in the same way they consume ‘genuine’ luxury products to show that they have finer taste than those already satisfied with ‘non-standard’ English, a commodity of a lower quality. What deserves further attention is how language and its ideologies play a major role in the construction of value around commodities, in this case, the watch featured in the commercial, and in the promotion of ‘elite’ discourses and identities (Jaworski & Thurlow 2009).

While, at the time of writing this article, it has already been a few years since the advertising campaign was launched and the celebrity's spoken English in the ad was derided, this does not diminish its current relevance. One reason is that the persistent, replicable, and searchable nature of online content (boyd 2011) enables and even encourages users to refer to, appropriate, and recontextualize the videos and comments for ridicule in new contexts or for other purposes. In June 2023, for example, a meme-based Instagram account made fun of the celebrity again by posting a new remix of the commercial and adding hashtags such as ‘lmao’, ‘local’, ‘accent’, and ‘english’ in the description. As can be imagined, digitally mediated ridicule is not always short-lived. A more important reason is that after the advertising campaign, which was launched in the run-up to the largest political movement in Hong Kong's history, there have been more instances of online linguistic ridicule targeting not only celebrities, but also other figures—notably police officers in the protests. Capturing the zeitgeist of a time of political turmoil, the present case provides an important basis, both theoretical and empirical, for delving into how linguistic ridicule has evolved as a window on the sociolinguistic, cultural, and political realities of present-day Hong Kong. More research is needed to investigate how these other figures are ridiculed and to capture the constantly evolving sociopolitical situations in the city. It would be also worth exploring how the ridicule is responded to by the ‘othered’.