Band 39, Heft 1

This paper argues that applied linguistics needs to pay more attention to the properties of the writing systems of language. Knowledge of writing is implicitly involved in many areas of language. It consists of at least: the characteristics of sound-based or meaning-based writing systems and of writing direction; rules for linking written forms with spoken sounds; a memory store of individual instances of visual forms; orthographic regularities in letter positioning; orthographic forms such as capital letters and word spaces; the ability to execute the written forms physically. Specific problems for linguistics are: the seeming debt of phoneme-based theories of phonology to sound-based writing systems; on the one hand the unacknowledged use of written conventions such as capital letters in the presentation of spoken data, on the other the failure to take account of writing systems, for example in work on the acquisition of the past tense -ed in English.

This paper examines the ability of native and non-native speakers of English to hear English speakers' voices and interpret the emotions being portrayed. In the study native speakers of Japanese, Spanish and English listened to audio tapes in which professional actors portrayed emotions and they identified the portrayed emotions by selecting from among eight possibilities. Results suggest major differences between native and non-native listeners in their ability to identify emotions expressed in voice. Also an analysis of judgements made by ESL learners at different proficiency levels did not show an increase in ability to judge the emotional content of English speech with increased language proficiency. This suggests that the ability to accurately judge emotions being portrayed through vocal cues in a second language may not be acquired by L2 learners without extensive exposure in a native context or without special attention to developing these skills in an instructional context.

Textbook conversational closings have come under criticism for their failure to replicate natural conversation. Given that authentic conversation is both difficult to collect and use in a classroom context, television and video materials have been suggested as an alternative (McCarthy and Carter 1994; Vanderplank 1993; Allan 1985; MacKnight 1983). This paper uses Schegloff and Sacks' (1973) description of native speaker conversational closings as a framework for analysing closings in ESL/EFL textbooks and soap operas. It compares textbook closings with closings from fifty episodes of the New Zealand soap opera Shortland Street . The results show that although New Zealand soap opera materials are far from ideal, they are a better source of data than many textbook examples. In this paper conversational closings from a range of textbooks are compared with closings from episodes of the New Zealand week night soap opera Shortland Street to determine the usefulness of soap opera language in the teaching of conversation to second- and foreign-language students. Closings were chosen because they are particularly difficult. Levinson (1983: 316) explains: Closings are a delicate matter both technically in the sense that they must be so placed that no party is forced to exit while still having compelling things to say, and socially in the sense that both over-hasty and over-slow terminations can carry unwelcome inferences about the social relationships between the participants.

Recently, competing views have been put forward to account for the shape of second language learner varieties. The discussion has focussed on the role of L1 knowledge, the accessibility of universal constraints on linguistic structure and cognitive principles of language learning. Contradicting claims have been made with respect to L1 transfer, driving forces in L2 development and fossilization. In the discussion, however, relevant findings from earlier research seem to be missing. Transfer, for example, is not a random process. Constraints on transfer can be described in terms of structural similarities and dissimilarities between L1 and the L2 target equivalent. Furthermore, findings from studies on acquisitional ordering are of relevance for the investigation of driving forces in L2 development. Finally, studies comparing adult and child language acquisition have provided evidence for the interaction between input processing and fossilization.

This paper describes a new methodology for examining the way assessors make subjective evaluations of written texts. The method presents texts to assessors one word at a time in a cumulative fashion, and asks them to indicate when they have enough material to make a judgment that they feel confident about. Preliminary results from this approach suggest that assessors require only very small amounts of text in order to make a judgment — typically thirty words or so. Non-native speaker assessors require slightly more text than native speaker assessors. The locus of the decision points varies enormously, however. The implications of this data for our understanding of wholistic judgements typically made with examination material are discussed.