This thesis investigates aspects of text-based Computer-mediated Communication (CMC) in Hong Kong within the framework of New Literacy Studies (NLS), a theory of literacy which studies how language and literacy are used in social context. The analysis is based on a 70,000-word corpus of email and ICQ messages (composed in Chinese/Cantonese and English). Two concepts of NLS – literacy events, occasions where literacy has a role, and literacy practices, ways of using language and literacy in different contexts, are examined in detail.

        The study first explores the constitution of a CMC literacy event and proposes that participants, settings, forms of interaction, codes, subjects of discussion, artefacts, and underlying contexts are the basic components. These elements are observed in a case study of a CMC event in a domestic setting. The research further reveals that in CMC events, the relationship between speech and writing is indeterminate since features vary across messages. It is, however, noted that code-mixing plays a prominent role in CMC texts in Hong Kong.

        The research has also identified seven literacy practices in text-based CMC. It first presents these practices under the topics of shortenings, emoticons, openings and closures, typographical, grammatical, orthographic, and ‘coding’ practices. It then examines various cultural and linguistic specific CMC textual practices, which include:

1. Cantonese-based shortenings (88 ‘bye-bye’);

2.  Asian-specific emoticons (>v<, :米);

3.  Transfer of native intuition to CMC texts (subject omission);

4.  Common grammatical errors which are made by second language learners of English (e.g. problematic verb forms, inappropriate choice of word classes);

5. Creative Cantonese representations (e.g. coined romanisation and transliteration)

        Textual findings are supported by questionnaire and interview surveys which investigate participants’ thoughts and values on their textual practices. Respondents’ opinions indicate that practices are not the same in all situations, which reveals the importance of ‘context’ in CMC.

        Three categories of CMC context are proposed - contexts of social interaction (‘chat’ and ‘non-chat’), communicative situational contexts, and language-specific contexts, which are associated with subsets of CMC practices according to the nature of the contexts. This characterisation of CMC context demonstrates its compatibility with the theory of NLS.

        The research suggests that language and literacy researchers and practitioners should recognise the set of new literacy practices in CMC. Subsequent studies in the Chinese context are necessary for a more complete understanding of the impact of CMC language on students' reading and writing habits.