ITL - International Journal of Applied Linguistics - Volume 133, Issue 1, 2001

The findings of this study add to the growing number of reports in which investigators claim to have located adult second language learners who, under rigorous test conditions, manage to pass as native speakers in L2. The aims of this paper were two, first, to provide a detailed account of how we tested and qualified our Anglophones as native-like speakers of French and, second, to suggest that, interesting as our data were, more questions emerge than do answers.

Seven of 18 English/French bilinguals, having acquired L2 after the age of 16, were selected by means of a pre-test interview with three Francophones as “potentially of French-speaking background.” These seven, along with three Francophone controls, recited an 81-word passage in French onto a tape-recorder. Sixty-eight native-speaking French raters, of similar dialectal background and weak in English, each heard one of four tapes with differing random roders of the 10 passages, their task being to designate each voice as “Franco-phone” or “non-Francophone.” Four of our seven English-Franch bilinguals obtained ratings statistically comparable to those of our three Francophone controls.

This paper uses a schemata-theoretic conception of reading in an assessment of ESL reader recall of unstated levels of meaning in narrative prose. Schemata theory suggests that the skilled reader selects one of a finite number of text schemata to use in the decoding, retention and recall of a particular text, and it has been demonstrated that better knowledge of the schematic structure makes possible better recall. Here, reader recall of two types of unstated meaning in narrative prose is assessed for a group of advanced learners of ESL, who use Chinese as L1. Evidence is presented of poor recall of unstated meaning, concomitant with a lack of knowledge of the requisite schematic structure. It is therefore suggested that formal instruction on the requisite structure will enhance learner recall of unstated meaning.

The purpose of this research was to explore the relationship between English proficiency, writing ability, and the use of conjunctions in Iranian EFL learners' compositions. To this end, four research questions were formulated : (1) Is there any relationship between English proficiency of Iranian EFL learners and the extent to which they use particular groups of conjunctions (additive, adversative, causal, temporal)? (2) Is there any relationship between English proficiency of Iranian EFL learners and their writing ability? (3) Is there any relationship between writing ability of Iranian EFL learners and the use of conjunctions? (4) What is the relative importance of the four groups of conjunctions and English proficiency in predicting the writing ability?

The study involved 120 male and female English learners of Kish Language Institute studying at different levels: Elementary, Intermediate, and Advanced. Two data gathering devices were employed: a NELSON test compatible with English knowledge of the subjects (version 300-A), and two compositions. The topics assigned for the compositions were “Describe your city, Tehran” and “Describe your English teacher". The students wrote the first composition in a session along with taking NELSON and the second one after a two-week interval. In each examination, the subjects wrote a text including about 150 to 200 words.

The analyses included Correlation, ANOVA, Chi-square, and Multiple Regression to display the relationship between the above-mentioned variables. The results indicated taht the High group of proficiency has a significant superiority over the Mid and the Mid group over the Low one on the writing scores. The use of Chi-square analysis displayed which level of proficiency or which level of writing use which type(s) of conjuction more. Multiple regression, then, identified which variable(s) are more important or contribute more to writing scores.

This research examines university students' writing problems and strategies; it also tries to investigate the way learner difficulties and strategies are perceived by students and teachers. Two questionnaires were conducted to find out both students' and teachers' assessment of these difficulties and strategies, and a corpus of 25 exam copies was examined, errors were classified, their sources analysed and strategies defined.

This study aims to point out the major difficulties that our EFL learners encountered in writing and to find out some of the reasons that could explain them. It also tries to see whether there were any discrepancies between the students’ and teachers’ assessment of writing difficulties and strategies and the students’ ‘real’ problems and strategy use, at least, as they were displayed in their exam copies. One of the major goals is also to identify and define Tunisian learners’ writing strategies and see whether they match those of ESL and EFL learners in the literature.

The research is based on the following hypotheses:

a) Students have difficulties in expressing their thoughts and ideas in the FL because they generate them in the L1 and then try to translate them in the FL.

b) Students’ problems are mainly conceptual; i.e, they result from a lack of understanding of the new concepts introduced in the course and course materials.

The results showed that the learners’ difficulties were mainly grammatical and organizational and that, just like other learners, our students made use of metacognitive, cognitive, social and affective strategies when handling their linguistic tasks.

The purpose of the present study is to investigate the effects of branching type (Factor 1) and grammatical function of noun phrase (NP) (factor 2) of English relative clauses on interlanguage performance among Japanese learners of English as a foreign language (EFL). The sentence combining test was administered to Japanese learners of English. Results indicated that both of the two factors significantly determine the difficulty of English relative clauses. Left-branching (center-embedded) relative clauses have a tendency to be more difficult than right-branching ones. Moreover, it is implied that subject-relative clauses are answered more accurately than object-relative clauses. These findings suggested that branching type and grammatical function of the noun phrase are complimentary determine the difficulty level of English relative clauses. The results implied the validity of KAWAUCHl's (1988) hypothesis that the difficulty order of relative clauses is as follows : OS > 00 > SS > SO. The results are also discussed with the recent theoretical frameworks in psycholinguistic research. The limitation of the present investigation and the directions of the further research are also discussed.