English language identities in Hong Kong

Bolton and Kwok (Reference Bolton and Kwok1990: 149) argue in an early and seminal study that while there are examples of Hongkongers ‘striving towards the norm of a British or American accent’, communities that speak Cantonese often retain their ‘localised features of [English] speech’. The authors' observations are made within a larger discussion of whether Hong Kong can be characterized as a region made up of two dominant monolingual speech communities, and the sociolinguistic implications of such a space, and thus Bolton and Kwok (Reference Bolton and Kwok1990) bring to the fore an important, yet often misunderstood, aspect of English language identities in Hong Kong. That is, why do some Hong Kong individuals and speech communities position their English language identities in a particular way, and according to specific geographic norms?

One perplexing issue relates to the extent of which Chinese identities are used by Hongkongers to construct their English language identities. In a relatively large-scale investigation of language attitudes in Hong Kong prior to the ‘handover,’ Pennington and Yue (Reference Pennington and Yue1994) ask nearly 300 secondary students what their views are on English and Cantonese. The researchers found that although English is nearly inescapable in the region, young Hongkongers do not see the language as a threat to their Cantonese identity. While these observations suggest Cantonese identity categories are resilient in the face of linguistic colonization, it is not clear to what extent Chinese identities shape notions of English in Hong Kong. For instance, in a later and larger study based on 1,265 questionnaires, Lai (Reference Lai2011: 260) observes ‘a strong sense of belonging to Hong Kong’ amongst local Cantonese-speaking respondents, but the researcher glosses over how belief systems associated with, and self-identification practices related to, English are influenced by notions of ‘Chineseness’, such as the social category of Putonghua speaker.

Sung (Reference Sung2014) narrows this empirical gap in his investigation of how preferences for specific accents shape identities. Like others noted in this review of literature, Sung (Reference Sung2014: 544) observes a ‘varied and complex’ sociolinguistic environment in Hong Kong, but he takes this observation a step further by analyzing in-depth interviews of 28 undergraduate students over an approximately one year period to demonstrate that ethnicity is a fluid concept when participants construct their English language identities. More specifically, Sung (Reference Sung2014) observes a weak relationship between a desire to conform to local expectations, such as sounding like other Cantonese-speaking Hongkongers, and English language identities. In other words, constructing a particular English language identity, such as ‘sounding British’, appears to be more related to pragmatic and functional justifications than to an impetus to conform to a particular ethnic or national identity.

Ethnic categories are indeed potentially fruitful entry points into an examination of the ways in which speakers create divisions and hierarchies in their understandings of English in Hong Kong. The importance of maintaining a sense of Chineseness is revealed in Evans's (Reference Evans2011) survey of over 800 young Hongkongers: his findings show that Cantonese is the overwhelmingly preferred language of communication in social encounters; but what makes this study interesting is the observation that young Hongkongers do not ‘feel particularly uncomfortable when they hear Chinese people speaking English among themselves’ (Evans, Reference Evans2011: 40). This somewhat paradoxical situation of feeling comfortable hearing Hongkongers use English while not using the language with each other suggests the need to provide a more nuanced understanding of why Cantonese is the preferred mode of communication in social encounters if English is viewed as a neutral, if not positive, language.

This issue of identity management is addressed in a study conducted by Jenks and Lee (Reference Jenks and Lee2016). Their investigation of the language ideologies of nearly 60 undergraduate students in Hong Kong shows that there are varying, and sometimes contradictory, belief systems as they pertain to how Hongkongers view their own ethnic identities in relation to English. For example, in addition to the more predictable belief that English should be maintained in Hong Kong for cosmopolitan and economic reasons, the common practice of admonishing and moving away from Cantonese-accented English is viewed by some Hongkongers as an egregious act of pretending to be someone they are not, such as a ‘fake’ American-born Chinese. In other words, for some Hongkongers, English accents associated with regions outside of Hong Kong, such as ‘Standard American English’, are incompatible with how the language should be spoken in the local context. Although these observations contribute to an overall understanding of the sociolinguistic environment of Hong Kong, the question of how ethnicity feed into English language identities is only partly addressed in Jenks and Lee (Reference Jenks and Lee2016).

The current study aims to address this limitation by analyzing how, and to what extent, notions of Americanness, Britishness, and Chineseness are relevant to English language identities in Hong Kong. This focus on ethnic categories, while narrow in specification, may help better address larger questions pertaining to Hong Kong's sociolinguistic environment, such as the issue of whether English is viewed as a foreign language or associated with ‘foreigners’.

Identities and heteroglossia

Hong Kong is a region with a complex and complicated relationship with the English language, as the literature review above demonstrates. Given the varying ways in which Hongkongers construct their English language identities, we adopt the theoretical framework of heteroglossia (Bakhtin, Reference Bakhtin, Emerson and Holquist1981). This framework enables us to make sense of the multiple voices, along with derivative linguistic identities and language ideologies, that cohabitate the region.

The expression heteroglossia was introduced by Bakhtin in the context of understanding the novel as a literary form. It has since been widely applied outside of literary theory in order to describe the multiplicity of discourse styles present within a single utterance. In an early study on Hong Kong, Joseph (Reference Joseph2000: 28) uses heteroglossia to present a post-1997 Hong Kong identity as ‘irreconcilable’. According to Joseph (Reference Joseph2000: 29), ‘[e]verything there [in Hong Kong] has to be understood, insofar as understanding is possible, simultaneously in a Chinese framework, a Western framework, and a Hong Kong framework that is both and neither of the other two and may be its own opposite’. This conclusion is reached by invoking Bakhtin's (Reference Bakhtin, Emerson and Holquist1981) argument that the interrelationship between ‘languages’ leads to a condition of linguistic untransposability:

Although Joseph's (Reference Joseph2000) application of Bakhtin's work is helpful in understanding the Hong Kong context, an argument can be made that heteroglossia is an unusual if not inapplicable theoretical concept for the region. Namely, it can be said that heteroglossia, in the original Bakhtinian sense, references not multilingualism but rather multivocality. It can be thus argued that heteroglossia, in spite of the prefix hetero-, was reflective of a decidedly monolingual ethic and ecology. As Gramling (Reference Gramling2016: 107) notes, ‘[t]he political import of heteroglossia in Bakhtin's Soviet environment was more akin to a certain kind of civic monolingualism, whereby differently voiced articulations of civic subjectivity would present comprehensible alternative interventions to those orthodoxies cultivated by the Soviet state’ (emphasis in original). Understood from this perspective, and to return to the Bakhtin passage above, the invocation of Ukrainian as a ‘language’ serves not so much to reify it as a ‘language’ discrete from and transposable to Russian, but instead to imagine it as a ‘voice’ and thus subsume it within a broader aesthetic code of the multivocal novel.

Despite such theoretical qualifications in applying heteroglossia to multilingual contexts, the term is applicable to our analysis because we focus on the intersecting and co-constitutive aspects of identity as they relate to not a language, but rather, a geopolitical space. Indeed, as Bakhtin (Reference Bakhtin, Emerson and Holquist1981: 291) argues, it is not possible to see language through a singular lens:

Bakhtin argues that an utterance arranges into a constellation of voices (cf. polyphony); each voice represents a unique life experience, and thus an utterance has the capacity to create dialogic relationships between distinct social and cultural groups within a society. Bakhtin (Reference Bakhtin, Emerson and Holquist1981) expresses how dialogic relationships between two or more voices are borne out in heteroglossia in the following way:

The pluralization of voice within an utterance is an active process that both reveals how members of society achieve intersubjectivity and represents the inherently multifarious nature of language. What leads readers to assume otherwise is the fact that the multiplicity of voices is not always immediately apparent within a speech community or sociolinguistic region because dominant ideologies espoused by social groups in positions of power often subdue peripheral discourses, thus resulting in the assumption and creation of a singular style within a particular textual space.

In the context of our study, we do not use heteroglossia to denote diversity and difference in a more generic sense. Rather, our analysis uses the term to conceptualize the different interrelated linguistic identities, even if inconsistent and contradictory, that co-populate and co-constitute the sociolinguistic ecology of Hong Kong. Further, heteroglossic emphasizes the co-presence of different Englishes within a particular national or regional context, but also how these different varieties result in the performance of English language identities that circulate within and through communicative encounters and social domains of life, which are, understood from a heteroglossic framework, simultaneously discrete and untransposable, or ‘half someone else's’.

The study

The study was conducted at two research-intensive universities in Hong Kong where, although classroom instruction is conducted in English, much of what is spoken by students on campus is Cantonese. However, as with many young university-educated Hongkongers, the participants possess at least 10 years of experience using English during their school years leading up to the time of data collection. The participants of the study are all multilingual, speaking at least three languages with some degree of proficiency: Cantonese, English, and Putonghua. This multilingual profile means that the participants of the study are aware and demonstrably comfortable reflecting on their language and cultural identities, as evidenced in and through their responses to our writing prompts and interview questions.

Data collected for this study come from interviews of, and self-narratives written by, young university-educated Hongkongers: our respondents were either currently studying at, or recently graduated from, a university in Hong Kong. The self-narratives were an assignment in a course taught for English majors studying at a large research-intensive university, which required students to reflect on how language, culture, and nationality factor into their understanding of Hong Kong and what it means to be a resident of the region. Before assignments were submitted, students were given opportunities over the semester to discuss the global spread of English and the implications the language has on speech communities in different regions. Students wrote on a number of topics, including life experiences learning English, the extent to which Hong Kong is a multilingual society, and the growing presence of Putonghua in the region. Over 50 assignments were collected at the end of the semester.

The audio-recorded interviews – 17 in total – were conducted after delivering the course, and the interviewees were not the same individuals that completed the self-narratives. A convenience sampling approach was used to select the interviewees: a research assistant located individuals walking on a campus of a different research-intensive university with a request to participate in a short interview on the use of English in Hong Kong. Interviews were organized around seven questions, though the research assistant was encouraged to ask follow up questions and allow interviewees to elaborate as much as needed. Written informed consent forms were given to interviewees prior to the interviews. All of the names included in our analysis are pseudonyms and other self-identifying information is omitted.

Bamberg's (Reference Bamberg1997) writing on narrative analysis informs our interpretation of data. This version of narrative analysis is centrally concerned with understanding how informants accomplish social actions in and through their responses to interviews and self-narratives. Narrative analysis provides an ideal analytic lens through which to examine language identities, as the methodology seeks to make connections between what informants say and larger societal issues, such as the extent to which a belief system regarding the English language is part of a history of colonialism. As Bamberg (Reference Bamberg and Cooper2012: 83) suggests, narratives are ‘full-blown life’ stories that reveal a system of social structures, including the cultural and political implications of using English in a multilingual region like Hong Kong. It can be said, then, that narrative analysis is a methodology that sees narratives as an interpretive extension of a nexus of life experiences and belief systems (Scollon, Reference Scollon2002).

Findings

The data excerpts examined in this section are representative of the range in which ethnic categories are used to express language ideologies and identities. This range of ideological expressions index larger, interconnected social issues, such as cosmopolitan elitism, racialized discourses, native speakerism, and childhood socialization practices. Our writing prompts and interview questions did not require individuals to explicitly attend to such issue; rather, ethnicity and other similar issues were expressed as Hongkongers made sense of their language identities.

The analysis begins with the more predictable ideology that strives for, and indeed valorizes the linguistic features of, so-called nativized English varieties. Specifically, the first two examples demonstrate a common theme in the language ideologies and identities of our research participants: native speaker idealizations underpin how English language identities are constructed (Kachru, Reference Kachru2005). Native speakerism is an important empirical issue in the literature, as much has been said about how this idealization is part of a larger system of discriminatory practices in the language teaching profession. While such observations are helpful in uncovering systemic racism (e.g., Jenks, Reference Jenks2017), the first two narratives below demonstrate that native speaker idealizations can also be understood in more simple terms; that is, the aspiration to sound ‘native-like’ is guided by the aesthetics connected to speaking English in a country in which it is spoken from ‘birth’.

Winnie and Jackie aspire to speak English with a British accent, which is perhaps not surprising given the legacies of British colonial occupation. While only Jackie explicitly mentions imitating a British accent, Winnie expresses an explicit admiration for it. Further, both Winnie and Jackie base their preference for British English on imagined idealization of a particular kind of usage: high-ranking government officials and entertainers on television. Thus, native speakerism can be understood as an aspiration based on both the concrete (e.g., the value British accents have in many domains of Hong Kong life) and the abstract (e.g., the view that British English is associated with one speech community). For instance, aspirations to sound British, as opposed to ‘Asian’ (cf. Bolton, Reference Bolton2008), are likely based on concrete racialized discourses that are transmitted in and through various forms of media (Ferguson, Reference Ferguson1998), but such desires may also be guided by abstract understandings of how a particular variety of English operates within and beyond the boundaries from which it is spoken.

In addition, through a heteroglossic approach, we can recognize a series of other simultaneities. For Winnie and Jackie, the variety of English spoken by politicians and celebrities is both inaccessible (insofar as it is the language of a speech community that they do not belong) and accessible (insofar as it can be practiced through imitation). Meanwhile, this variety of English is both distant (geographically speaking) and local (accessible via television and used by some interlocutors in Hong Kong). Finally, it is simultaneously unique (understood as elegant and refined) and banal (the everyday vernacular as represented on television).

Acknowledging such simultaneities allows us to understand how English language identities are complicated by the reality that aspirations to a particular variety are not always an either/or proposition (either assimilating to British English or embracing a local variety of English), but can also be a both/and binary, as our other research participants demonstrate in their ideological expressions. For example, for Sze Ki, wanting to sound British should be understood through the lens of being Chinese.

Of immediate note is how this reflection disrupts the commonsense belief that English language ideologies and identities in Hong Kong are based on a fixed linguistic hierarchy that places Chinese-accented Englishes below other varieties. While such putative hierarchies have been argued to result in the subordination of local varieties and marginalized identities, Sze Ki acknowledges the inevitability of a Chinese accent while refusing to acknowledge the superiority of a ‘royal English’. The observation that it is inevitable to sound Chinese while speaking English suggests that it is possible, and indeed necessary, to also be Chinese while speaking English.

This third narrative should not be interpreted as a larger representation of how young Hongkongers view the binary between being Chinese and using English; the representativeness of this particular binary is indeed an empirical issue that would benefit from being studied using different methodological tools, including quantitative analyses of large-scale data. However, it needs to be acknowledged at least that the ideology that young Hongkongers need to conform to their Chinese identities operates in dialogic relationship to the belief system that celebrates the value and sounds of British accents, as expressed in the first two reflections.

The relevance of this dialogic relationship within the Hong Kong context can be expressed in two ways. First, while there are individuals such as Sze Ki who reject the ideological stratification of British English over local varieties, there will also be individuals, occupying the same geographic locale, who are committed to, and thus complicit in, such stratification. Second, this means that any resulting identity, whether in affirmation or rejection of such inequitable linguistic economy, cannot be understood as independent of the other possibility, but rather as co-constituted by it.

While our analysis has thus far presented a varied and complex account of English in Hong Kong, there is a common ideological principle that is used to construct language identities amongst our research participants. That is, there appears to be an expectation that place of birth indexes certain ways of using English. For example, although Winnie and Jackie prefer to use a language variety that is commonly associated with England, both students construct an expectation that to use English in Hong Kong is to sound British. In the case of Sze Ki's critique of the widespread practice of ‘mimicking’ British English in Hong Kong, her narrative of course establishes an expectation that to use English in Hong Kong is to maintain a Cantonese identity. All three narratives, while possible examples of Hongkongers subordinating or rejecting a colonially imposed linguistic ideal, demonstrate that language identities are connected to the degree to which Hong Kong is positioned as an English-speaking region. As the next two narratives expressed by Sandy and Cindy demonstrate, the issue is not whether ‘sounding British’ amounts to being a legitimate speaker of English. Rather, what is considered a legitimate way of using English in Hong Kong relates to both where one is born and the linguistic imageries of such a region/place.

ABC and BBC refer to ethnic Chinese born in the United States and Great Britain; such identity constructs are important to the ways in which Chineseness is constructed (e.g., Yang, Reference Yang2006), as ABC and BBC help individuals make sense of the cultural differences that exist as a result of growing up in a country or region outside of Hong Kong. The narratives of Sandy and Cindy are somewhat different than what has been discussed in previous investigations of ABC and BBC identities (cf. Tsai, Ying & Lee, Reference Tsai, Ying and Lee2000), as both Hongkongers here demonstrate that there is an expectation that local Hongkongers must use Cantonese-accented English (cf. Wei & Lee, Reference Wei and Lee2001).

In Sandy's account in particular, we see the possibility that attaining a high level of proficiency in a variety of English that is presumed to be the most prestigious or desirable results in one potentially being alienated by local interlocutors. Therefore, it is not merely that one's identity is performed through the practice of a given language or variety, but also that one's identity is co-constituted by others in the same geographic space. Cindy's comments provide further insights into the co-constructed nature of English identities in Hong Kong. For Cindy, the possibility of being seen as a performer of an ethnic identity that is alien to local Hongkongers does not deter her from her aspiration to use a particular variety of English. That is, while Sandy is apprehensive of such accusations, Cindy accepts them as ‘praises,’ and as recognition of her ‘hard work’ learning English.

In order to understand the complexities of language identities in Hong Kong, as expressed in the narratives of Sandy and Cindy, it is useful to view expectations and aspirations to use a particular variety of English in the context of coercive mimeticism. Chow (Reference Chow2002: 107) uses coercive mimeticism to refer to the expectation in postcolonial contexts to ‘resemble or replicate the very banal preconceptions that have been appended to them’. In coercive mimeticism, ‘the original that is supposed to be replicated is no longer the white man or his culture but rather an image, a stereotyped view of the ethnic’ (Chow, Reference Chow2002: 107). For instance, Sandy admits to performing a ‘stereotyped view of the ethnic’ by strategically incorporating localisms such as ‘um,’ ‘ar,’ or ‘la’ into her everyday speech. It is not merely the conventional practice of crossing in the sense of using the language resources associated with an outside ethnolinguistic group (Rampton, Reference Rampton2011). It is perhaps more reflective of crossing ‘of a different kind,’ performed with a high degree of ambivalence that is shaped by larger sociopolitical forces, including the threat of being viewed as traitors to their local heritage (Rampton, Charalambous & Charalambous, Reference Rampton, Charalambous and Charalambous2018: 18). What makes Sandy's linguistic performance so intriguing is the difficulty of using English outside the expectation of coercive mimeticism: she is not able to merely use British English even if she is capable of doing so; instead, she feels compelled to do so in a manner that is based on a ‘banal preconception’ of how she should speak English with stereotyped Cantonese-accented features.