• Walang Nahanap Na Mga Resulta

The Sociolinguistics of Code-switching in ... - Animo Repository

N/A
N/A
Protected

Academic year: 2023

Share "The Sociolinguistics of Code-switching in ... - Animo Repository"

Copied!
23
0
0

Buong text

(1)

Volume 2

Issue 1 June Article 2

6-2023

The Sociolinguistics of Code-switching in Hong Kong’s Digital The Sociolinguistics of Code-switching in Hong Kong’s Digital Landscape: A Mixed-Methods Exploration of Cantonese-English Landscape: A Mixed-Methods Exploration of Cantonese-English Alternation Patterns on WhatsApp

Alternation Patterns on WhatsApp

Wilkinson Daniel Wong Gonzales

The Chinese University of Hong Kong, wdwonggonzales@cuhk.edu.hk Yuen Man Tsang

The Chinese University of Hong Kong, 1155144527@link.cuhk.edu.hk

Follow this and additional works at: https://animorepository.dlsu.edu.ph/jeal

Part of the Applied Linguistics Commons, Communication Commons, Computational Linguistics Commons, Data Science Commons, East Asian Languages and Societies Commons, Feminist, Gender, and Sexuality Studies Commons, Race, Ethnicity and Post-Colonial Studies Commons, and the Social and Cultural Anthropology Commons

Recommended Citation Recommended Citation

Gonzales, Wilkinson Daniel Wong and Tsang, Yuen Man (2023) "The Sociolinguistics of Code-switching in Hong Kong’s Digital Landscape: A Mixed-Methods Exploration of Cantonese-English Alternation Patterns on WhatsApp," Journal of English and Applied Linguistics: Vol. 2: Iss. 1, Article 2.

DOI: https://doi.org/10.59588/2961-3094.1041

Available at: https://animorepository.dlsu.edu.ph/jeal/vol2/iss1/2

This Article is brought to you for free and open access by the DLSU Publications at Animo Repository. It has been accepted for inclusion in Journal of English and Applied Linguistics by an authorized editor of Animo Repository.

(2)

WhatsApp WhatsApp

Cover Page Footnote Cover Page Footnote

We are immensely grateful to Liu Hong Yau Anson (Civil Engineering, The University of Hong Kong) for his invaluable knowledge and technical assistance in developing a database and processing the data. His expertise in creating custom Visual Basic functions and executing codes and programs has been of immense help in our research process. His help has also enabled us to work faster and to analyze the data in a systematic manner. We are also thankful to the participants who provided us with their

WhatsApp conversations and survey data. Their support has enabled us to collect enough contemporary discourse data for our research. We are deeply appreciative of their faith in us and their contributions towards furthering sociolinguistics in Hong Kong. Finally, we are thankful to Professor Tongle Sun of The Chinese University of Hong Kong for her constructive advice and feedback on an earlier iteration of this work, which has been awarded the Exemplary Capstone Project Award (2022-2023, Term 1) by the Department of English at the Chinese University of Hong Kong.

This article is available in Journal of English and Applied Linguistics: https://animorepository.dlsu.edu.ph/jeal/vol2/

iss1/2

(3)

The Sociolinguistics of Code-Switching in Hong Kong’s Digital Landscape: A Mixed-Methods Exploration of Cantonese-English Alternation Patterns on WhatsApp

Wilkinson Daniel Wong GONZALES1 Yuen Man TSANG1

1The Chinese University of Hong Kong, Hong Kong SAR, People’s Republic of China wdwonggonzales@cuhk.edu.hk

Abstract

This paper examines the prevalence of Cantonese-English code mixing in Hong Kong through an under-researched digital medium. Prior research on this code-alternation practice has often been limited to exploring either the social or linguistic constraints of code-switching in spoken or written communication. Our study takes a holistic approach to analyzing code-switching in a hybrid medium that exhibits features of both spoken and written discourse. We specifically analyze the code-switching patterns of 24 undergraduates from a Hong Kong university on WhatsApp and examine how both social and linguistic factors potentially constrain these patterns. Utilizing a self-compiled sociolinguistic corpus as well as survey data, we discovered that those who identified as male, studied English, and had an English medium-of-instruction (EMI) background tended to avoid intraclausal code-switching between Cantonese and English. Responses to the open- ended questions revealed that many of our participants used code-switching as a means to fill conceptual gaps, engage in socialization (e.g., to strengthen solidarity or make their speech sound more casual and natural), and construct bilingual and Hong Konger identities. Our findings shed some light on at least some of the locally embedded social meaning(s) of this linguistic practice in a digital context.

Key words: language variation and change in Hong Kong, Cantonese-English code-switching, sociolinguistic variation in East Asia, digital sociolinguistic corpus of WhatsApp, identity construction in a digital landscape

Copyright © 2023 by De La Salle University RESEARCH ARTICLE

(4)

Introduction

It is increasingly common to find multilingualism as the norm, rather than the exception, in many regions around the world, especially in a time when digital advances have made exposure to and engagement with multiple languages easy and commonplace (Appel & Muysken, 1987). One of the consequences of multilingualism is the extensive interaction between languages, which has been shown to result in (stable) linguistic variation and change at both the community and individual level (Sippola & Lesho, 2020; Thomason, 2008). It can facilitate the emergence of innovative linguistic practices (Freynet & Clément, 2019; Gonzales, 2022b, 2023a; Thomason, 2001) and, in special cases, even new contact languages (Bakker, 1997; Gonzales, 2022c, 2023b; O’Shannessy, 2005).

One such phenomenon that is widely observed within multilingual communities is the alternation between languages, commonly known as code-switching.

Some popular examples of switching include Cantonese-English in Hong Kong, Spanish-English in the Americas and Hebrew-English in Israel (Auer, 1998), Hokkien-Tagalog-English or “Hokaglish”

code-switching among Lannangs in the Philippines (Gonzales, 2016, 2022a), and, more recently, code- switching between Mandarin and Taiwanese (or Taiwanese Hokkien) in Taiwan (Hsiao, 2022).

Among these examples, perhaps one of the most prominently featured in the sociolinguistic literature is the one in Hong Kong, where English has been employed alongside Cantonese owing to British colonial influence since the early 19th century (Bolton et al., 2020; Luke, 1998). Hong Kong inhabitants who are bilingual in Cantonese—the native language of most Chinese-heritage inhabitants—and English are most likely to employ Cantonese-English code- switching in both verbal and online communication (Chan, 2019; Chen, 2005). But while code-switching is a common phenomenon in the region, patterns of code-switching may differ between speakers (i.e., interspeaker variability). This variability between speakers could be attributed to personal preferences, as one individual may find it more expedient to incorporate English words into their speech, while another may not (Chan, 2018; Li, 2000, p. 317). Interspeaker variation is not the only type of variation that exists. Intraspeaker variation, such as style shifting, is also observed (Grosjean & Miller, 1994). Style shifting can manifest

in speakers when they shift from a monolingual mode of communication (e.g., speaking only Cantonese) to a bilingual mode (i.e., speaking both Cantonese and English in the same utterance) to achieve certain social functions. An individual may shift to English, while maintaining Cantonese as the matrix language, in order to avoid coming off as too westernized or arrogant, especially with Hong Konger-identifying or Hong Konger-presenting peers (Chen, 2008, pp. 132–133).

However, when speaking with non-Hong Kongers, they may opt to use English as the matrix language with minimal Cantonese lexicon. Intraspeaker style shifting can also come in the form of identity stressing, where individuals who want to present more as a “Hong Konger” may use Cantonese-English code-switching (Chan, 2018, p. 90).

In this paper, we examine Cantonese-English code-switching to expand the already comprehensive research of this phenomenon. Our general interest lies in how multilingual individuals’ social backgrounds could potentially impact their cognitive abilities and language habits of alternating between languages.

However, our primary focus is on the sociolinguistic trends of code-switching within Hong Kong’s digital space. Although numerous researchers have conducted studies on Cantonese-English code-switching in Hong Kong (Chan, 2018; Chen, 2005), only a small number of these studies have acknowledged its relevance in internet communication (Har, 2021). A large body of research has been conducted on the phenomenon of language mixing in spoken conversations; however, comparatively little has been explored in terms of

“written” discourse. Furthermore, there has been a dearth of research focusing on code-switching patterns in computer-mediated-communication (CMC) contexts or “online communication” more broadly.

The importance of focusing on digital platforms becomes increasingly evident after considering that communication in such platforms can contain elements of both spoken and written discourse, as demonstrated by the example of Cantonese:

(1) 爲何 你 這樣 討厭 他?

(Written Cantonese)

點解 你 咁 憎 佢?

(Spoken Cantonese)

why 2.SG this.much hate 3.SG?

(5)

‘Why do you hate him so much?’

(example provided by second author from WhatsApp, 2022)

Apart from the lack of focus on CMC data, few have examined (written) digital discourse while addressing social factors that have been reported to be meaningful in our preliminary ethnographic work: sex (e.g., male vs. female), level of English proficiency as reflected in choice of educational trajectory (e.g., English major vs. other major), and educational background (e.g., whether one is from an English-as-medium- of-instruction school or a Cantonese-as-medium-of- instruction school). The dearth of research into these factors, especially in the digital realm, may be partly due to the scarcity of publicly available datasets that feature code-switching data online. To address this, we constructed a code-switching corpus of WhatsApp data in Hong Kong so as to explore the occurrences of Cantonese-English code-switching in the region.

This research aims to explore the potential factors that may affect the frequency and patterns of Cantonese-English code-switching in WhatsApp conversations within the multilingual context of Hong Kong. The other objective is to complement the quantitative data with a qualitative investigation of the motives behind and implications of the code-switching phenomenon of focus.

We hope to contribute to the growing body of sociolinguistic work on Hong Kong linguistic practices.

This paper stands out from other similar work, as it features an exploratory analysis of code-switching on a communication medium that is generally regarded impenetrable in the context of Hong Kong—WhatsApp communication. One of the challenges typically encountered in obtaining WhatsApp data was the issue of privacy, as Hong Kong residents are generally reluctant to share their data with unfamiliar, nonlocal researchers for the purposes of academic research (first author’s ethnographic notes, 2023). To mitigate this issue, the second author, who identifies as a local Hong Konger, leveraged their social network. It is anticipated that the findings involving code-switching patterns in an understudied medium will open possibilities for further investigation. Our analysis seeks to provide further insight into the dynamics of language mixing in the digital landscape of Hong Kong, Hong Kong in general, as well as that of the wider Asia-Pacific region.

In Section 2, important existing literature pertaining

to the topics is reviewed. Section 3 provides information on the data source and methods employed. The results are presented in Section 4, wherein key findings are discussed and relevant trends identified. The paper concludes with a set of concluding remarks, implications, and recommendations for future research (Section 5).

On Code-Switching and WhatsApp

Prior to initiating our sociolinguistic inquiry into Cantonese-English code-switching patterns in Hong Kong, we review literature related to code- switching and WhatsApp data analysis, with the aim of operationalizing key concepts and contextualizing the study.

Definitions of Code-Switching

There is a wide array of definitions for the term code-switching. Gumperz (1982, p. 59) defines code- switching as “the juxtaposition within the same speech exchange of portions of speech belonging to two separate grammar systems or sub-systems.” Like other scholars (Appel & Muysken, 1987), he operationalizes code-switching as a practice that involves the use of two or more languages in a single speech event or the use of linguistic components from both (or many) languages. Our study adopts this definition.

Historical Research on Code-Switching

Research on code-switching was significantly influenced by Uriel Weinreich’s (1953) monograph Languages in Contact. Weinreich was arguably the first to examine code-switching and bilingualism from the perspective of multiple disciplines (e.g., linguistics, psycholinguistics, sociolinguistics). He is well-known for his groundbreaking research of linguistic diversity in his area. Gumperz (1982), on the other hand, made significant contributions to the field by emphasizing the need to see code-switching as a discourse approach rather than a flawed behavior. Thus, it was seen as an

“added resource for communicating a variety of social and rhetorical meanings” (Milroy & Muysken, 1995).

The phenomenon of code-switching became an even more popular subject of interest with the introduction of Myers-Scotton’s (1993) Matrix Language Frame and Myers-Scotton and Jake’s (2017) 4M models (Myers- Scotton & Jake, 2009), which proposed systematic, structured ways of analyzing a seemingly random

(6)

linguistic practice. The study of code-switching has historically been investigated using grammatical/

linguistic or social lenses exclusively (Lowi, 2005). In this paper, we adopt a joint sociolinguistic approach that not only explores possible linguistic factors that constrain code-switching patterns but also “focuses on the social and political incentives for the use of languages in interaction with bilingual or migratory groups” (Lowi, 2005).

Types of Code-Switching

Within the code-switching literature, it appears to be generally accepted that three forms of code- switching exist, namely, tag switching, interclausal code-switching, and intraclausal code-switching.

The first type, tag switching, involves incorporating a question tag from one language into an utterance made in another. The second is when a sentence in one language ends and another begins in a different language. If so, the switching is said to have taken place “intersententially” or interclausally (Appel &

Muysken, 1987, p. 118). The third type is intraclausal code-switching, defined as switching inside a phrase or clause. The shifts mostly occur inside sentences, and they almost never include larger linguistic units than clauses (Li, 2000). Examples for each type of major code-switching patterns are as follows:

(2) 你幇我處理呢件事, okay?

(Tag switching)

“You help me with this issue, okay?”

(3) 聽講你今日有面試?Good luck with that!

(Interclausal)

“I heard you have an interview today? Good luck with that!”

(4) 上個 project end up 成點?

(Intraclausal)

“How did the last project end up?”

Although intraclausal code-switching has been reported to be the most cognitively demanding of the three forms of code-switching (Jalil, 2009), it is the most prevalent form in Hong Kong, just as it is in many multilingual societies. In code-switching

practices involving Cantonese and English, Cantonese is often the “matrix language” (i.e., the dominant language), while English is very often the embedded language (i.e., the secondary language; Halmari, 1997). It has been reported to be common in media produced in Hong Kong, including television shows, advertisements, periodicals, and social media websites (Chen, 2008).

(Socio)linguistic Research Involving WhatsApp The recent decade has seen a proliferation of research into the unique genre of WhatsApp communication, which presents written texts that are proximate to the informal oral medium, akin to colloquial conversations (Lee, 2007). This has garnered considerable attention from scholars in the fields of linguistics and sociolinguistics and has been studied extensively from a variety of angles. In general, the literature in this area examines the linguistic features of text messaging and the factors that shape code- mixing practices, in terms of both linguistic features and sociocultural contexts.

Pérez-Sabater and Montero-Fleta (2015) examined how two different generations of texters in Spain use English and Spanish on WhatsApp. They found teenagers tend to be more innovative in their use of texting, such as clipping, using acronyms, emoticons, spelling innovations, and the use of words from other languages. Haryati and Prayuana (2020) investigated code-mixing practices in WhatsApp data from Indonesia and concluded that variation patterns in these practices are shaped by individual bilingualism, the situation, prestige, and the interlocutors. Alazzawie (2022) employed a qualitative approach to register, genre, and style analysis to analyze the linguistic features of text messages from a high school student population in Canada and found that students use clipped sentences in casual speech and slang. They also found that while certain abbreviations have become standard, there is a wide array of individualistic variance in terms of style and language usage.

There are also studies that take a corpus-based approach to investigating WhatsApp data from a sociolinguistic perspective. One cluster of studies focuses on the publicly available Corpus of Singapore English Messages, a database of Colloquial Singapore English WhatsApp texts with sociolinguistic metadata such as age, sex, and ethnicity (Gonzales et al., 2021).

Some findings of the project include a strong gender

(7)

effect as well as several particle-specific ethnic effects on the variation in the use of Singapore English discourse particles (Leimgruber et al., 2020), the robust use of sentence-final particle sia by younger generation Singaporeans but avoidance by female Malay Singaporeans (Hiramoto et al., 2022), and the use of innovative constructions involving is it (e.g., Isit you like her? as opposed to You like her isit? and Do you like her?) among younger speakers, those who identify as Singaporean, and those who want to express a playful affect in the conversation (Gonzales et al., 2022).

Very few studies have been conducted on WhatsApp messaging code-switching practices in Hong Kong, likely due to the difficulties associated with collecting WhatsApp data given privacy concerns. Har (2021) explored a previously unexplored domain, the government sector, regarding linguistic phenomena of Cantonese-English code-switching and code mixing on WhatsApp. Results indicated that code-switching to Cantonese and English was prevalent and that postgraduate degree holders were more likely to switch to English.

Methodology

Data Sources and Collection Protocol

To recapitulate, we are interested in examining the phenomenon of Cantonese-English code-switching and its interaction with linguistic and social factors in the context of WhatsApp communications in Hong Kong. Due to the lack of publicly available code- switching corpora, we relied on a self-compiled corpus of WhatsApp messages. We chose WhatsApp because it is the leading instant messaging app in Hong Kong, with a penetration rate of 85.5% as of January 2022 (DataReportal, 2022). Based on current statistics, individuals in Hong Kong prefer WhatsApp (65%) as the platform for communicating and exchanging messages with others, considerably outpacing Instagram (14%) and WeChat (9%; DataReportal, 2022). Essentially, WhatsApp is the most widely used social media platform by people in Hong Kong for everyday conversation. Its local popularity makes it the optimal selection for our data gathering process, as it simplifies the process of data sampling.

The corpus compilation process began with the construction of a collection matrix. Because we are interested in the potential main and interaction effects

of sex, educational background, and college major (specialization/field of study), we decided to construct a 2 × 2 × 2 data collection matrix (Table 1). The goal was to find 24 individuals willing to release their WhatsApp conversations (in raw text files) based on the three social conditions of interest in this study.

The second author, a local Hong Konger in a local university, reached out to her social network to collect the bulk of the data. Most of the participants she had were undergraduates of local universities in Hong Kong. In addition to asking for the conversation files, we also asked participants to fill out a sociolinguistic survey questionnaire over Google Forms. The survey contains fixed-response questions, 5-point Likert scale questions, and open-ended questions. Questions about personal background, code-switching practices, and attitudes towards these practices were asked.

These questions—and the protocol in general—were informed by existing studies on code-switching (Lo, 1999). The complete survey, derived from the Google form, can be found in Appendix A.

Table 1. Data Collection Matrix (by Participant) a. English Medium-of-Instruction (EMI)

Secondary Schools

Female Male

English Major 3 3

Non-English Major 3 3

b. Chinese Medium-of-Instruction (CMI) Secondary Schools

Female Male

English Major 3 3

Non-English Major 3 3

Data Preprocessing

After collecting the data, we preprocessed the raw WhatsApp text files by transforming the text file into a spreadsheet file where each row contains an utterance (i.e., a speech bubble on WhatsApp). The resulting preprocessed dataset contains roughly 308,552 words from 56,469 utterances. Then, the data from the sociolinguistic survey were linked to each utterance.

In response to the privacy concerns raised by some participants, we scrubbed all information (e.g., phone

(8)

numbers, names, etc.) from the utterances to the best of our abilities using a mix of manual and computational methods (Malmgren, 2021).

We first coded each utterance depending on whether it is an utterance with code-switching (i.e., code- switched, not code-switched). Then, for code-switched utterances, customized Visual Basic functions and manual analyses were done to code our data for matrix language (i.e., English vs. Cantonese; Appendix B and C). We coded each utterance for type of code-switching (i.e., tag, intraclausal, interclausal) by first examining the percentage of English and Cantonese elements

in the utterance and the presence of punctuations. If the utterance contains roughly 50% of English and Cantonese and contains multiple punctuations, it is tentatively classified as interclausal code-switching whereas the rest are classified as tag or intraclausal code-switching. Then, the second author went through the data again manually to correct misclassifications based on her native speaker intuition. Table 2 illustrates some examples of utterances that have been coded for matrix language, code-switched string, frequency of English and Chinese characters/words, and type of code-switching.

Table 2. Examples of Coded Utterances by Matrix Language, Code-Switches, and Type of Code-Switching

Utterance Matrix

Language Code- Switched String

Chinese Characters/

Words

English

Words Type of Code- Switching

我今個 sem 唔計 lab 有兩日 有早堂

“I have morning classes for 2 weekday this semester, excluding laboratory sections.”

Cantonese sem lab 11 (84.6%) 2 (15.4%) Intraclausal

可唔可以

send

琴日買鞋嗰 間嘢個

地址黎

“Can you send me the address of that store where you bought shoes yesterday?”

Cantonese send 15

(93.7%) 1

(6.3%) Intraclausal

urs should look like mine.

你 噤下兩 個星期之後嘅時間表 先係最准

“Yours should look like mine. It’s the most accurate if you check the schedule in two weeks’ time.”

English,

Cantonese

你噤下兩個

星 期之後嘅 時間 表先係 最准

17(78.3%) 5

(21.7%) Interclausal

請槍

is incorrect

“Cheating is incorrect.” English

請槍

2 (50%) 2 (50%) Intra-clausal

You mean choosing between

午餐時 段

and

歡樂時段?

“You mean choosing between Lunch Hour and Happy Hour?”

English

午餐時段 歡

樂時段

8 (61.5%) 5 (38.5%) Intraclausal

Because we are interested in the potential effect of part of speech on code-switching patterns, we also tagged each utterance for part of speech using TagAnt (Anthony, 2022). There were no simultaneous multilingual models offered by the software, so we conducted the tagging separately on English words and Chinese words using English and general Chinese models (Anthony, 2022).

Data Analysis

Both quantitative and qualitative analyses were conducted. The coded utterances were analyzed quantitatively through frequency distribution analyses.

Raw counts and relative frequency in terms of percentage or proportion were acquired. We analyzed the distribution of code-switches by linguistic (e.g., part of speech) and sociolinguistic factors (e.g., sex).

(9)

chi-squared tests were employed when testing for the effects of social factors on the likelihood to use English/Cantonese lexicon in a 2 × 2 matrix (Baayen, 2008).

The survey data were analyzed both quantitatively and qualitatively. Likert scale responses and choice- related items were analyzed distributionally and presented in histograms, whereas themes were derived from open-ended questions and discussed in relation to patterns from quantitative data.

Findings

General Patterns

Approximately 12.6% (7,081) of the utterances assessed exhibited code-switching. Not a single instance of tag switching was identified, and the vast majority of code-switched utterances (99.534% or 7,048) were intrasentential, an unexpected result.

We suspect that the prevalence of intrasentential code-switching is likely caused by the type of communication taking place in WhatsApp. People tend to use words or short phrases instead of writing out long paragraphs. Consequently, it is uncommon to find two distinct sentences of different languages in one text (i.e., one speech bubble), as one would expect in interclausal code-switching.

Of all the intrasentential code-switching utterances, 73.1% (5,154) of them use Cantonese as the matrix language, while 26.9% (1,894) of them use English as the matrix language. Cantonese serves as the matrix language for most of the utterances, with English embedded inside Cantonese, as in (5). The remaining utterance have English as the matrix language, with Cantonese elements interspersed with English, as in (6).

(5) 覺得佢好似 complain 緊咁

“(I) feel like he is complaining.”

(6) I don’t need 宿分 this sem

“I don’t need hostel points this semester.”

Overall, our corpus study has discovered that our sample comprising undergraduates in Hong Kong predominantly code-switch. And when they do, they usually use Cantonese as the base language and English as the embedded language. This finding corroborates previous research on the phenomenon (Li, 2000).

Linguistic Conditions for Code-Switching

A closer analysis of code-switched utterances shows speakers tend to code-switch to English or Cantonese more frequently depending on the part of speech of the word. In the case of switches to English, the switches most frequently involve nouns (35.95%) and proper nouns (27.35%), followed by verbs (14.51%).

More than 60% of English code-switches are nouns (e.g., progress meeting in 睇下 progress meeting 洗 唔洗改期 “See if the progress meeting needs to be rescheduled”). The predominance of English nouns in Cantonese-matrix-language code-switching can be observed more clearly in Table 3, where the proportion of English code-switched nouns is significantly higher than other English code-switched elements.

Table 3. Distribution of Code-Switched Elements Originating From English by Part of Speech

Part of Speech Examples From

Corpus Frequency (Percentage) in Corpus

Common noun lunch, meeting,

weekend 3,043 (35.95%)

Proper noun Covid, Google,

Mary, GPA 2,315 (27.35%) Verb apply, quit, said,

send, drop 1,228 (14.51%) Adjective sad, sorry,

impossible, free 545 (6.44%) Adverb nearly, very,

tomorrow 377 (4.45%)

Interjection hey, oh, yup,

okay 354 (4.18%)

Preposition in, to, during,

after 167 (1.97%)

Pronoun I, you, he, she,

myself 115 (1.36%)

Conjunction and, or, but 93 (1.1%)

In the case of switches to Cantonese, most of the Cantonese-derived elements are nouns (46.60%), verbs (34.77%), and adverbs (5.21%). Like in the case of English elements, approximately half of all code-switches in Cantonese are nouns (e.g., 交收

“delivery” in U finish the 2 交收 “You finished the two deliveries?”). In Table 4, the frequency of Cantonese code-switched nouns is notably higher than the other

(10)

Cantonese code-switched elements, indicating a clear predominance of Cantonese nouns, and to a lesser extent verbs, in Cantonese-English code-switching where English is the matrix language. The findings provide some evidence that linguistic part-of-speech conditions the patterns of code-switching in Cantonese- English code-switching.

Overall, in both switches to English and Cantonese, code-switching most frequently occurs in nouns compared to adjectives and adverbs. Grammatical function words, such as pronouns, articles, prepositions, conjunctions, and numerals, are less likely than content words to be transferred to the recipient language.

From a perspective of linguistic borrowing or transfer, our findings accord with the literature, which generally agrees that word classes interact with the pressure of language contact (Whitney, 1881). It has been observed in most of the world’s languages that content words tend to be transferred more than function words (Thomason & Kaufman, 1988). Specifically, using a linguistic adoptability scale, Haugen (1950) found that transferred nouns (i.e., content words) are almost three times more prevalent than transferred verbs, while adjectives, adverbs, prepositions, and interjections trail behind. He proposes that the prevalence of nouns and verbs for transfer is linked

to vocabulary growth over a speaker’s lifetime.

Meanwhile, unlike content words, grammatical items are part of a fixed inventory that is established in early childhood. Thus, they are less likely to be code- switched and/or borrowed. These tendencies have been observed in many previous studies cross-linguistically (Thomason & Kaufman, 1988), including this one.

Social Conditions for Code-Switching Overall Patterns

A breakdown of all tokens in code-switched utterances is found in Table 5. The data presented show that individuals who report as male, having Chinese as medium of instruction, and currently taking a non- English major have significantly higher rates of English code-switches compared to other groups. But is there evidence that sex, major, and medium independently influence these code-switching patterns? To answer this question, we analyzed the possible effects of each factor on code-switching patterns independently.

An ideal solution would be to conduct a multivariate analysis using a single model with independent multiple variables. However, in this exploratory study, we limit ourselves to simple but robust frequency distribution analyses.

Table 4. Distribution of Code-Switched Elements Originating From Cantonese by Part of Speech

Part of Speech Examples From Corpus Frequency (Percentage) in Corpus

Noun

“event,”

公司

“company,”

出口

“exit” 1,261 (46.6%)

Verb

“say,”

“drag,”

“hit” 8,941 (34.77%)

Adverb

已 “already,” 全 “all,” 不 “not,” 最

“most” 141 (5.21%)

Adjective

“good,”

“close to somebody,”

“evil” 120 (4.43%)

Modifier

前_ “pre-,” 女_ “female entity”

41 (1.52%)

(11)

Sex

Based on our analysis, female users used almost three times as many English code-switches (6,330 words) as male users (2,221 words). The number of English code-switches that females employed was roughly 185% more than that of male undergraduates, showing that females have higher propensity to employ English code-switches than males. A chi-squared test indicates that the gender-based proportional difference is statistically significant (χ2 [1, N = 295,956] = 108.684, p < 0.0001; Table 6).

Table 6. Distribution of Code-Switched Elements by Sex

Sex English Cantonese

Female 6,330 (3.11%) 197,533 (96.89%) Male 2,221 (2.41%) 89,872 (97.59%)

The tendency for females to embrace linguistic innovations—here assumed to be the use of English in Cantonese matrix utterances—is a pattern that has emerged in a lot of prior sociolinguistic work cross- linguistically. It has been claimed that women tend to be vanguards of innovative practices, especially in situations where the community is not conscious of the variable (Labov, 1994; Meyerhoff, 2018). However, in this case, we observe that female speakers are the innovators of a practice that is generally in the collective consciousness of the speech community. According to our survey data, around 60% of participants are highly aware of engaging in Cantonese-English code- switching during a conversation, and around 30% are aware of it sometimes. The survey results indicate that code-switching is generally a conscious act among our

multilingual participants. That said, our results appear to contradict the general pattern observed. It is unclear why our female sample behaved the way they did, but possible explanations could include cross-cultural differences and/or specific social meanings attached to code-switching practices in our particular sample of female users or community of practice (Mallinson

& Childs, 2007) that would encourage them to switch to English more often than males.

It is also very possible that our assumption that the use of English code-switches is “innovative” is wrong and that code-switching is in fact more “standard.”

Some evidence of such can be found in the survey results, where some students have explicitly indicated that code-switching is the norm and not doing so would be nonstandard and “weird.”

(7) because it’s common to code-switch to English in my social circle and I feel like everyone is so used to it. It’d be weird not to code-switch with friends i guess

(8) I consider code switching a communicative norm in hk

If this is true, then our results accord with Labov’s principles (Labov, 1972), where younger women tend to adhere to the norm or produce more “standard”

language in cases where the community is aware of the variable.

Major

Based on our frequency data, perhaps the most critical factor for code-switching out of the three social variables of interest is major status. The results of this Table 5. Distribution of Code-Switched Elements by Sex, Medium of Instruction, and Major

Medium of

Instruction Major Sex English

Words % English Cantonese

Words %

Cantonese Total Words EMI (English

as Medium of Instruction)

English

major Female 116 0.17 70,131 99.83 70,247

Male 62 0.14 43,046 99.86 43,108

Other major Female 1,997 6.29 31,761 94.08 33,758

Male 1,106 6.06 18,251 94.29 19,357

CMI(Chinese as Medium of Instruction)

English

major Female 1,521 5.30 28,684 94.96 30,205

Male 271 1.22 22,157 98.79 22,428

Other major Female 2,696 4.03 66,957 96.13 69,653

Male 782 12.18 6,418 89.14 7,200

(12)

study showed that the use of English code-switches by students with non-English majors was 234% higher than that of English majors, with 6,581 words used compared to 1,970 words. This is confirmed by a chi- squared test, which has demonstrated a statistically significant difference in English-Cantonese code- switching proportions between English majors and students who are not English majors (χ2 (1, N = 295,956) = 3,904.44, p < 0.0001; Table 7).

Table 7. Distribution of Code-Switched Elements by Major

Major English Cantonese

English major 1,970 (1.18%) 164,018 (98.82%) Other major 6,581 (5.06%) 123,387 (94.94%)

The pattern observed is contrary to our expectation that English majors (presumably with higher proficiency in English) would employ more English code-switches. We found that English majors engage in a different bilingual practice that is not code-switching:

they prefer using English without code-switching as opposed to Cantonese-English code-switching on WhatsApp. This was evidenced by the database, which showed that approximately 81.3% of utterances (31,372 out of 38,578) made by English major students were English utterances without code-switching. On the other hand, other students taking other majors often employed Cantonese as the matrix language, resulting in a higher utilization of English code-switches.

Assuming that English majors are highly proficient in English and at least somewhat proficient in Cantonese, it seems that, in the context of our study, the main consequence of increased English proficiency is the exclusive use of English, rather than the utilization of English resources in code-switching practices, whether as a matrix language or as an embedded language. Some support for this can be found in Tsang’s ethnographic work, where she found that “students of high English proficiency prefer using pure English in WhatsApp most of the time” (Tsang, 2022, p. 28). It is at odds with what is observed for other communities like the Manila Lannang community in the Philippines, where multilingualism in Hokkien, Tagalog, and English primarily takes the form of increased intraclausal code-switching that led to the creation of a new mixed language Lánnang-uè (Gonzales, 2016, 2022c). The results of the inquiry indicate that the

effects of bilingualism is contingent on the social and historical context.

Medium of Instruction

In our ethnographic work, we observed that educational background (i.e., whether an individual was historically exposed to an English/Cantonese- medium-of-instruction curriculum) could be a crucial factor in conditioning one to employ more English code-switches. Our quantitative results confirm our initial observations, but unexpectedly, we found that those exposed to a Cantonese curriculum were the ones who used more English and not those exposed to a predominantly English curriculum. Students from CMI schools—or schools with a Cantonese dominant curriculum—employed 5,270 words of English code-switches, resulting in a 61% difference over that of students from schools with English as the medium of instruction (i.e., EMI schools). We found that the difference is significant χ2 (1, N = 295,956) = 1,142.93, p < 0.0001; Table 8).

Table 8. Distribution of Code-Switched Elements by Medium of Instruction

Medium of

Instruction English Cantonese

English 3,281 (1.97%) 163,189 (98.03%) Cantonese 5,270 (4.06%) 124,216 (95.94%)

Obviously, there are multiple factors not discussed here that could explain why we observe a seemingly contradictory pattern (e.g., identity construction, social network composition and density). However, if we only focus on the education background, one reason could perhaps be related to the account of major in the previous subsection—that maybe those who are exposed to more English in their childhood (or those who have attended an EMI school) tend to engage in “monolingual” linguistic practices that index membership in the social group “EMI students.”

This makes sense, as those in CMI schools may not necessarily be able to communicate in “monolingual”

English mode. Future research can investigate whether the switch to monolingual mode is indeed laden with

“EMI” social meaning.

(13)

Conscious Motivations for Code-Switching

In this section, we move beyond numbers and focus on what our participants have to say about Cantonese- English code-switching. Reasons and motivations for code-switching were explored by analyzing Likert- scale survey questionnaire data. We also analyzed open-ended comments from a bottom-up approach and identified three primary conscious motivations for Cantonese-English code-switching: (1) lexical gap filling, (2) socialization strategy, and (3) identity construction.

Code-Switching as Gap Filling

A significant number of participants in our sample viewed code-switching as a language enrichment practice, where terms that cannot be expressed in one language or terms that are difficult to access at the time of conversation are supplemented by concepts in another language. This is evident in the survey data, which found roughly 17% of participants claiming they code-switch to English because of their inability to find the Cantonese equivalent term when speaking in Cantonese. Around 60% reported that they switch to Cantonese because they were not able to identify the equivalent English words on the spot. The survey also showed a high motivation for code-switching in cases where there are no Cantonese or English equivalents for the concept. Around 21% reported using English code-switches when Cantonese does not have a matching concept, and roughly 42% claimed to employ Cantonese code-switches when the concept cannot be expressed in English, as in (9).

(9) 6 green 降燥熱 pills

“6 green pills for reducing dryness-heat or heatiness

Here, the word 燥熱 “heatiness” is absent in English or untranslatable without certain background knowledge. The word 燥熱 also known as 上火 (literally, “up/increase fire”) is a unique concept in traditional Chinese medicine. It emphasizes balancing yin and yang for perfect health. If yang is greater than yin, your body will generate excessive heat and cause 燥熱, which can result in ulcers. In Hong Kong, 燥熱 is a commonly used term. The term may sound odd and lose its original meaning if it is translated into English, hindering communication. Therefore, a plausible

strategy would be to embed the Cantonese word into English to avoid misunderstanding and capture the full sense of the term. This type of code-switching is common among our sample and appears to be a conscious process, as evident in the following excerpt:

(10) some Chinese phrases cannot be rendered directly in English - so I keep them in original form

Most of the participants (~70%) have positive sentiments towards code-switching as a strategy for conceptual gap filling. They believe that code- switching is a communicative tool that helps facilitate conversations.

Code-Switching as a Tool for Socialization

The survey data stress the importance of code- switching in the interactions of Cantonese-English bilinguals in Hong Kong and shed light on how it is employed as a social tool. The results indicate that code-switching is mainly used to signify social belonging as well as to express intimacy (Figure 1), with 41.7% of respondents reporting that using Cantonese in English utterances marks social belonging and 54.2% claiming that the same is true for English in Cantonese. Additionally, survey results show that some participants use Cantonese elements in English to “embed [their] mother tongue” in casual conversations with friends, to impress the listener, for euphemism, and to demonstrate intimacy (Figure 1).

English in a Cantonese frame was employed for all mentioned reasons apart from “embed mother tongue.”

The responses to the open-ended questions that were independently administered prior to the survey corroborated the results of the survey. A prominent recurrent theme in the responses is the idea of code- switching as a means of facilitating social interactions.

It was frequently noted that Cantonese-English code- switching not only enhanced the effectiveness of communication but also was utilized to demonstrate positive emotion. One participant specifically mentioned that code-switching helps create a sense of rapport and intimacy with someone who also speaks the same language. Another person claimed that code- switching rendered their utterances more natural and casual. These are observed in the excerpts in (11).

(14)

12 Journal of English and Applied Linguistics | Vol. 2 No. 1 | June 2023

(11) a.“to communicate with close friends more efficiently and in a good vibe!”

b.because I… want to take it easy, so I use Cantonese

c.“…a way to build up closeness to sb. who also speaks Cantonese”

d.“it’s more natural and sounds more casual in conversation~”

e. “mixing English into Cantonese may sound more informal and shorten distance between ppl”

From both the survey and the open-ended responses, it is clear that code-switching is not only a communicative standard or norm in Hong Kong. It is also a linguistic resource that can be used to express different affects and styles during socialization over WhatsApp.

Code-Switching as a Tool for Constructing Identity Apart from conceptual gap filling and socializing, Cantonese-English code-switching can also be used to indicate certain group memberships in Hong Kong, as evidenced in the survey data. Most of our participants (~62%) reported that emphasizing the

“Hong Konger” identity is an important factor that encourages them to engage in the multilingual practice.

Out of these participants, some claim that a “real Hong Konger” is one who uses code-switching to “prove”

their local identity in juxtaposition to residents in other sociopolitical regions such as Mainland China, who tend to be monolingual. An even larger subset of our participants (~70%) more broadly claim that code-switching is a marker of bilingual identity, as they believe it is “cool” to code-switch because of its exclusivity among Cantonese-English bilinguals (Figure 2).

In the open-ended responses, some comments have been found to support the quantitative patterns identified earlier (12 to 15). However, in addition to corroborating the notion of code-switching as an emblem of “Hong Konger” and bilingual identity, the responses further show that the linguistic practice of code-switching also indexes youth and can thus be used to construct an intersectional “young Hong Konger”

identity (12).

(12) sometimes English words turn into popular words among we hk teenagers, for example 你今晚firmfirm架". [“Are you firm or not with tonight’s party?”] Then these become our language, only young people say so

(13) i do think that using English in between Cantonese is a characteristics in HK as we have more and long-established exposure to English.

Figure 1. Conscious motivations for code-switching to Cantonese and English.

form

Most of the participants (~70%) have positive sentiments towards code-switching as a strategy for conceptual gap filling. They believe that code-switching is a communicative tool that helps facilitate conversations.

4.3.5.2 Code-Switching as a Tool for Socialization

The survey data stress the importance of code-switching in the interactions of Cantonese-English bilinguals in Hong Kong and shed light on how it is employed as a social tool. The results

indicate that code-switching is mainly used to signify social belonging as well as to express intimacy (Figure 1), with 41.7% of respondents reporting that using Cantonese in English utterances marks social belonging and 54.2% claiming that the same is true for English in Cantonese. Additionally, survey results show that some participants use Cantonese elements in English to “embed [their] mother tongue” in casual conversations with friends, to impress the listener, for euphemism, and to demonstrate intimacy (Figure 1). English in a Cantonese frame was employed for all mentioned reasons apart from “embed mother tongue.”

Figure 1. Conscious motivations for code-switching to Cantonese and English.

The responses to the open-ended questions that were independently administered prior to the survey corroborated the results of the survey. A prominent recurrent theme in the responses is the idea of code-switching as a means of facilitating social interactions. It was frequently noted that Cantonese-English code-switching not only enhanced the effectiveness of communication but also was utilized to demonstrate positive emotion. One participant specifically mentioned that code-switching helps create a sense of rapport and intimacy with someone who also speaks the same language. Another person claimed that code-switching rendered their utterances more natural and casual. These are observed in the excerpts in (11).

(11) a.

“to communicate with close friends more efficiently and in a good vibe!”

b.

because I… want to take it easy, so I use Cantonese 02

46 108 1214

to embed mother tongue in casual

conversations with friends

to sound natural to increase

social belonging to impress

receiving party for euphemism to show intimacy Cantonese English

(15)

13 Journal of English and Applied Linguistics | Vol. 2 No. 1 | June 2023

(14) most of my [young] friends are local so we find it normal

(15) the english term is more commonly accepted by [young] peers, people will think that you are so “kam” if you talk in full Chinese even in WhatsApp.

It is important to note that while code-switching is an available resource in the “ethnolinguistic repertoire”

of Hong Kong residents (Benor, 2010, p. 159), especially those who are bilingual in Cantonese and English, some may not consciously or deliberately use the mixing practice for various social functions, such as identity construction. This is highlighted in one of the responses in the survey questionnaire (16). In this case, it appears that the participant is subconsciously or unconsciously recruiting English elements in Cantonese, as reflected in the presence of English elements in their Cantonese utterances.

(16) I dont deliberately choose to switch to English - I think it’s more likely to be a habit shaped by the environment

There is clear evidence of many participants utilizing code-switching as a tool for constructing

Hong Konger and, more broadly, bilingual identity.

This is possible, as the practice of code-switching is not accessible to those who do not have bilingual proficiency in Cantonese and English, as well as knowledge of sociolinguistically relevant mixing norms in Hong Kong (Luk, 2013).

Conclusion

At the beginning of this paper, we aimed to investigate the potential factors that could influence the Cantonese-English code-switching patterns in WhatsApp conversations within the multilingual context of Hong Kong. We examined whether code- switching is socially meaningful in the region’s digital linguistic ecology, by identifying possible sociolinguistic patterns in the data as well as the conscious motives behind the language mixing practice. Adopting a mixed-methods approach, we drew on a sociolinguistic corpus using WhatsApp data as well as survey questionnaires.

The results corroborate previous research, which has established that intraclausal Cantonese-English code-switching is the most frequent type of code- switching in Hong Kong, with Cantonese typically acting as the matrix language and English as the embedded language (Li, 2000). However, our results

e.

mixing English into Cantonese may sound more informal and shorten distance

between ppl”

From both the survey and the open-ended responses, it is clear that code-switching is not only a communicative standard or norm in Hong Kong. It is also a linguistic resource that can be used to express different affects and styles during socialization over WhatsApp.

4.3.5.3 Code-Switching as a Tool for Constructing Identity

Apart from conceptual gap filling and socializing, Cantonese-English code-switching can also be used to indicate certain group memberships in Hong Kong, as evidenced in the survey data. Most of our participants (~62%) reported that emphasizing the “Hong Konger” identity is an important factor that encourages them to engage in the multilingual practice. Out of these participants, some claim that a “real Hong Konger” is one who uses code-switching to “prove” their local identity in juxtaposition to residents in other sociopolitical regions such as Mainland China, who tend to be monolingual. An even larger subset of our participants (~70%) more broadly claim that code-switching is a marker of bilingual identity, as they believe it is “cool” to code-switch because of its exclusivity among Cantonese-English bilinguals (Figure 2).

Figure 2. Distribution of identity-related sentiments towards code-switching.

In the open-ended responses, some comments have been found to support the quantitative patterns identified earlier (12 to 15). However, in addition to corroborating the notion of code- switching as an emblem of “Hong Konger” and bilingual identity, the responses further show that the linguistic practice of code-switching also indexes youth and can thus be used to construct an intersectional “young Hong Konger” identity (12).

Not at all Slightly

accurate Somewhat

accurate Mostly

accurate Totally accurate It feels cool that when I code-

switch, only Cantonese-English

bilinguals understand. 1 3 3 11 6

I believe code-switching promotes my Hongkonger

identity. 1 2 6 14 1

02 46 108 1214 16

Figure 2. Distribution of identity-related sentiments towards code-switching.

(16)

go beyond the linguistic aspects and comment on the sociolinguistic dimension of the mixing practice by demonstrating how linguistic and (some previously unexplored) social factors jointly play a role in determining the patterns of code-switching in the region.

Our examination of the language-internal aspects of code-switching thus far indicates that the part- of-speech of a word has a significant effect on the choice of English or Cantonese lexicon. Nouns and, to a lesser extent, verbs tend to be realized in English, while grammatical function words like prepositions, conjunctions, and pronouns tend to be expressed in Cantonese.

With respect to extralinguistic factors, we found that sex, previous educational background, and current educational trajectory (major) influence Cantonese- English code-switching patterns. Those who identified as female appear to be the preservers of standard language use, rather than vanguards of language change, assuming of course that Cantonese-English code-switching is “standard.” English majors and those exposed to English medium curricula—assumed to be more proficient in English—avoid using intraclausal code-switching and instead opt for interclausal code- switching. As mentioned earlier, evidence for this claim can be found in Tsang’s (2022) ethnographic work.

The patterns involving educational background present interesting cases where the product of multilingualism can have asymmetric effects, instead of a uniform effect observed in other multilingual societies (i.e., cases where both interclausal and intraclausal code- switching are equally observed). The findings suggest that different types of code-switching practices could hold different types of social meaning in Hong Kong (i.e., interclausal code-switching to “pure” English to index being part of the “EMI” group, “English major”

group).

The results of the sociolinguistic corpus analysis were supplemented with an analysis of survey data, where it was found that code-switching between Cantonese and English is locally meaningful: it can be used for conceptual gap filling, as a tool for socializing (e.g., increase solidarity, stylize an utterance as casual and natural), and for constructing bilingual and Hong Konger identity.

This study has provided important insights into the digital landscape of Hong Kong, yet it is not without limitations. For one, our study is restricted

to investigating code-switching from a “bag-of- words” approach rather than using sequential lenses (Goldberg, 2017, p. 69; Zhang et al., 2010). In other words, we did not analyze code-switching as a string of words, but rather lexically. It might be fruitful to examine whether the sociolinguistic patterns on code- switching observed in this study apply to sequential datasets. Another limitation includes our study’s lack of inclusion of variables that have been found to robustly condition code-switching patterns. In Li’s (2000) systematic review, the variable of style (e.g., “high” vs.

“low” Cantonese), age, and ethnic orientation appear to be robust predictors of code-switching patterns.

Deliberate and creative use of linguistic resources such as bilingual puns has also been found to be a major conditioning factor of certain code-switching practices.

It would be interesting to see whether at least some of these variables also condition code-switching on the WhatsApp medium. Finally, a statistical shortcoming of this study is perhaps its use of simple, univariate statistics. Our quantitative analyses tried to pinpoint possible effects of three sociolinguistic factors on the language mixing patterns, but it is difficult to comment on the unique effect of these factors without conducting more complex statistical techniques. It is possible that the significance of sex, major, and medium of instruction fades into the background after considering other more robust variables of code-switching. As such, future research may consider examining the effects of a broader set of social and linguistic variables—

including the ones already mentioned—in a single unified model. Such a predictive model may not only be useful for enriching computational components of language-related products (e.g., Alexa, Siri) in a world increasingly dominated by artificial intelligence. A more refined and fine-tuned representation of this novel linguistic practice would also allow us to attain a more holistic understanding of the digital sociolinguistic landscape of Hong Kong and that of the greater Asia- Pacific region.

References

Alazzawie, A. (2022). The linguistic and situational features of WhatsApp messages among high school and university Canadian students. SAGE Open, 12(1), 215824402210821.

https://doi.org/10.1177/21582440221082124

Anthony, L. (2022). TagAnt (2.0). Waseda University. https://

www.laurenceanthony.net/software

(17)

Appel, R., & Muysken, P. (1987). Language contact and bilingualism. Edward Arnold.

Auer, P. (1998). Code-switching in conversation: Language, interaction, and identity. Routledge, London.

Baayen, R. H. (2008). Analyzing linguistic data: A practical introduction to statistics using R. Cambridge University Press.

Bakker, P. (1997). A language of our own: The genesis of Michif, the mixed Cree-French language of the Canadian Métis (Rev. ed.). Oxford University Press.

Benor, S. B. (2010). Ethnolinguistic repertoire: Shifting the analytic focus in language and ethnicity. Journal of Sociolinguistics, 14(2), 159–183.

Bolton, K., Bacon-Shone, J., & Lee, S. (2020). Societal multilingualism in Hong Kong. In P. Siemund & J.

R. E. Leimgruber (Eds.), Multilingual global cities:

Singapore, Hong Kong, Dubai (1st ed., pp. 160–184).

Routledge. https://doi.org/10.4324/9780429463860 Chan, K. L. R. (2018). Being a ‘purist’ in trilingual Hong

Kong: Code-switching among Cantonese, English and Putonghua. Linguistic Research, 35(1), 75–95. https://

doi.org/10.17250/KHISLI.35.1.201803.003

Chan, K. L. R. (2019). Trilingual code-switching in Hong Kong. Applied Linguistics Research Journal. https://doi.

org/10.14744/alrj.2019.22932

Chen, H. Y. (2005). The social distinctiveness of two code- mixing styles in Hong Kong. In ISB4: Proceedings of the 4th International Symposium on Bilingualism (pp.

527–541).

Chen, H. Y. (2008). Bilinguals in style: Linguistic practices and ideologies of Cantonese-English codemixers in Hong Kong [Unpublished doctoral dissertation].

University of Michigan.

DataReportal. (2022). Digital 2022 Hong Kong.

https://datareportal.com/reports/digital-2022-hong- kong?rq=hong%20kong

Freynet, N., & Clément, R. (2019). Perceived accent discrimination: Psychosocial consequences and perceived legitimacy. Journal of Language and Social Psychology, 38, 496–513.

Goldberg, Y. (2017). Neural network methods for Natural Language Processing (G. Hirst, Ed.). Morgan and Claypool.

Gonzales, W. D. W. (2016). Trilingual code-switching using quantitative lenses: An exploratory study on Hokaglish.

Philippine Journal of Linguistics, 47, 106–128.

Gonzales, W. D. W. (2022a). Hybridization. In A. M.

Borlongan (Ed.), Philippine English: Development, structure, and sociology of English in the Philippines (pp. 170–183). Routledge.

Gonzales, W. D. W. (2022b). Interactions of Sinitic languages in the Philippines: Sinicization, Filipinization, and Sino-Philippine language creation. In Z. Ye (Ed.), The Palgrave handbook of Chinese language studies

(pp. 369–408). Springer Nature Singapore. https://doi.

org/10.1007/978-981-16-0924-4

Gonzales, W. D. W. (2022c). “Truly a language of our own” A corpus-based, experimental, and variationist account of Lánnang-uè in Manila [Unpublished doctoral dissertation]. University of Michigan.

Gonzales, W. D. W. (2023a). Broadening horizons in the diachronic and sociolinguistic study of Philippine English with the Twitter Corpus of Philippine Englishes (TCOPE). English World-Wide. A Journal of Varieties of English. https://doi.org/10.1075/eww.22047.gon Gonzales, W. D. W. (2023b). Spread, stability, and

sociolinguistic variation in multilingual practices:

The case of Lánnang-uè a. International Journal of Multilingualism. https://doi.org/10.1080/14790718.20 23.2199998

Gonzales, W. D. W., Hiramoto, M., Leimgruber, J. R.

E., & Lim, J. J. (2022). Is it in Colloquial Singapore English: What variation can tell us about its conventions and development. English Today, 1–14. https://doi.

org/10.1017/S0266078422000141

Gonzales, W. D. W., Hiramoto, M., Leimgruber, J. R. E., J.,

& Lim, J. J. (2021). The Corpus of Singapore English Messages (CoSEM). World Englishes, weng.12534.

https://doi.org/10.1111/weng.12534

Grosjean, F., & Miller, J. L. (1994). Going in and out of languages: An example of bilingual flexibility.

Psychological Science, 5, 201–206.

Gumperz, J. (1982). Discourse strategies. Cambridge University Press.

Halmari, H. (1997). Government and code-switching:

Explaining American Finnish. John Benjamins Publishing Company.

Har, F. (2021). Language choices between government sector colleagues: A Hong Kong case study of English language adult learners’ plurilingual practices in computer- mediated communication. Linguistics International Journal, 15(1), 1–20.

Haryati, H., & Prayuana, R. (2020). An analysis of code- mixing usage in WhatsApp groups conversation among lecturers of Universitas Pamulang. Ethical Lingua:

Journal of Language Teaching and Literature, 7(2), 236–250. https://doi.org/10.30605/25409190.180 Haugen, E. (1950). The analysis of linguistic borrowing.

Language, 26, 210–231.

Hiramoto, M., Gonzales, W. D. W., Leimgruber, J., Lim, J.

J., & Choo, J. X. M. (2022). From Malay to Colloquial Singapore English: A case study of sentence-final particle sia. In A. Ngefac, H.-G. Wolf, & T. Hoffman (Eds.), World Englishes and creole languages today existing paradigms and current trends in action (pp.

117–130). Lincom Europa.

Hsiao, C.-H. (2022). Code-switching between typologically similar languages: Data from Mandarin-Taiwanese

(18)

code-switching [Unpublished doctoral dissertation].

Indiana University.

Jalil, S. A. (2009). Grammatical perspectives on code- switching. ReVEL, 7(13), 1–11.

Labov, W. (1972). The social motivation of a sound change.

In Sociolinguistic patterns (pp. 251–265). Academic.

Labov, W. (1994). Principles of linguistic change. Blackwell.

Lee, C. K. M. (2007). Affordances and text-making practices in online instant messaging. Written Communication, 24(3), 223–249.

Leimgruber, J., Lim, J. J., Gonzales, W. D. W., & Hiramoto, M. (2020). Ethnic and gender variation in the use of Colloquial Singapore English discourse particles.

English Language and Linguistics.

Li, D. C. S. (2000). Cantonese-English code-switching research in Hong Kong: A Y2K review. World Englishes, 19(3), 305–322. https://doi.org/10.1111/1467- 971X.00181

Lo, A. (1999). Codeswitching, speech community membership, and the construction of ethnic identity.

Journal of Sociolinguistics, 3(4), 461–479. https://doi.

org/10.1111/1467-9481.00091

Lowi, R. (2005). Codeswitching: An examination of naturally occurring conversation. In Proceedings of the 4th International Symposium on Bilingualism (pp.

1393–1406).

Luk, J. (2013). Bilingual language play and local creativity in Hong Kong. International Journal of Multilingualism, 10(3), 236–250. https://doi.org/10.1080/14790718.201 3.808200

Luke, K. K. (1998). Why two languages might be better than one: Motivations of language mixing in Hong Kong.

In M. C. Pennington (Ed.), Language in Hong Kong at century’s end (pp. 145–159). Hong Kong University Press.

Mallinson, C., & Childs, B. (2007). Communities of practice in sociolinguistic description: Analyzing language and identity practices among black women in Appalachia.

Gender and Language, 1(2), 173–206. https://doi.

org/10.1558/genl.v1i2.173

Malmgren, D. (2021). Scrubadub [Python 3.6]. https://

scrubadub.readthedocs.io/en/stable/index.html

Meyerhoff, M. (2018). Introducing sociolinguistics (3rd ed.). Taylor and Francis.

Milroy, L., & Muysken, P. (1995). One speaker two languages: Cross disciplinary perspectives on code- switching. Cambridge University Press.

Myers-Scotton, C., & Jake, J. (2009). A universal model of code-switching and bilingual language processing and production. In The Cambridge handbook of linguistic code-switching (pp. 336–357). Cambridge University Press. https://doi.org/10.1017/CBO9780511576331 Myers-Scotton, C. M., & Jake, J. L. (2017). Revisiting

the 4-M model: Codeswitching and morpheme election at the abstract level. International Journal of Bilingualism, 21(3), 340–366. https://doi.

org/10.1177/1367006915626588

Myers-Scotton, C. (1993). Duelling languages: Grammatical structure in codeswitching. In Duelling languages:

Grammatical structure in codeswitching. Clarendon Press.

O’Shannessy, C. (2005). Light Warlpiri: A new language.

Australian Journal of Linguistics, 25(1), 31–57. https://

doi.org/10.1080/07268600500110472

Pérez-Sabater, C., & Montero-Fleta, B. (2015). A first glimpse at mobile instant messaging: Some sociolinguistic determining factors. Poznan Studies in Contemporary Linguistics, 51(3). https://doi.

org/10.1515/psicl-2015-0016

Sippola, E., & Lesho, M. (2020). Contact-induced grammatical change and independent development in the Chabacano creoles. Bulletin of Hispanic Studies, 97(1), 105–123.

Thomason, S. (2001). Language contact: An introduction.

Georgetown University Press.

Thomason, S. (2008). Social and linguistic factors as predictors of contact-induced change. Journal of Language Contact, 2, 42–56.

Thomason, S., & Kaufman, T. (1988). Language contact, creolization, and genetic linguistics. Univeristy of California Press.

Tsang, Y. M. (2022). A sociolinguistic analysis of Cantonese- English code-switching in Hong Kong WhatsApp conversations [independent research project]. The Chinese University of Hong Kong.

Weinreich, U. (1953). Languages in contact: Findings and problems. Mouton.

Whitney, W. D. (1881). On mixture in language. Transactions of the American Philological Association, 12, 1–26.

Zhang, Y., Jin, R., & Zhou, Z.-H. (2010). Understanding bag- of-words model: A statistical framework. International Journal of Machine Learning and Cybernetics, 1, 43–52.

(19)

Appendix A. Survey

SURVEY QUESTIONNAIRE Cantonese-English Code-Switching

This study investigates Cantonese-English code-switching in Hong Kong WhatsApp conversations from a sociolinguistic perspective. The personal information collected is going to be used only for academic/research purposes and will not be disclosed to anybody or to any organization apart from the research team.

The data presented in research output will be anonymized. Your personal data (i.e., your name, phone number, all security numbers, URLs, usernames, and other identifiable information) will be de-linked and scrubbed from the data using a mix of manual and computational tools.

Will you be donating your data? “Data” here refers to survey data, and/or WhatsApp data.

� Yes

� No

By selecting “yes”, I agree that

1. I am free from coercion of any kind.

2. I am donating my data to help advance the field of sociolinguistics in Hong Kong.

3. I agree to let the researchers use my data under the condition of anonymity and will be opting to agree without affixing my signature.

� Yes

� No

Personal particulars

Name: _______________________________________________

Age: _________________________ Sex: _________________________

Major and year of study (e.g. ENGE/4, IBBA/2): ______________________

First Language

� Cantonese

� English

� Putonghua

How long have you learned English (number of years): ________________

Which language is the primary medium of instruction at your secondary school?

� Cantonese

� English

� Putonghua

Mga Sanggunian

NAUUGNAY NA DOKUMENTO

Far Eastern University, Incorporated FEU PSE Disclosure Form 17-18 - Other SEC Forms/Reports/Requirements Form/Report Type Certificate of Attendance on Corporate Governance Seminar

Far Eastern University, Incorporated FEU PSE Disclosure Form 17-7 - Statement of Changes in Beneficial Ownership of Securities References: SRC Rule 23 and Section 17.5 of the