A major task for the teacher is to engage students in three kinds of learning: learning of language, learning through language, and learning about language (Halliday, 1979). Therefore, in order to help students learn and use language effectively, teachers themselves need explicit knowledge of language, and need to be aware of what students are trying to achieve and learn. There have been many classroom research investigations into this issue with respect to the role of the first language (L1) in facilitating learning (Osman and Hannafin 1994; Johnson 1997). Classroom-oriented research has led to an increased awareness of the great complexity of both the language used and the learning process. Classroom language can be investigated systematically and no longer needs to be regarded as “an impenetrable black box” (Long 1983 a, 1983 b). Transcripts, coded data and frequency counts can show how classroom interaction develops as it is a dynamic phenomenon (Allwright and Bailey 1991). Marton and Booth (1997, p.107) described learning as ‘a way of coming to experience the world in a certain way, and different ways of experiencing will lead to different learning outcomes’. Marton and Tsui (2004) linked classroom discourse to learning, coining the term “space of learning”: “The second premise is that language plays a central role in learning: that it does not merely convey meaning, it also creates meaning. An understanding of how the space of learning is linguistically constituted in the classroom is best achieved through investigating “classroom discourse”. This research observed and collected data from two different primary four classes taught by two teachers of the same school on the same topic on Chinese words. Multiple data from multiple sources were used for analysis. This paper will discuss how the difference usage of discourse in the classroom could bring about different learning outcomes.

• 吳鳳平(香港大學教育學院) | NG Fung Ping