On the Relation between Language Teaching and Testing

Authors

John W. Oller, Jr.

Publication Date

1997-12-01

Abstract

For well over a decade, theoreticians have generally agreed that what is needed by linguists, language teachers, and testers is a more consistent, comprehensive, and parsimonious theory of how language works in communication, how it is acquired by children (and others), and how it comes to be related systematically through the conventions of particular grammars (i.e., of particular languages) to the common world of experience. Here a number of theoretical advances are discussed and certain implications for language teaching and testing are considered. The fundamental flaws of discrete-point approaches, including the British notional-functional syllabus, are pointed out in the light of the theory of true narrative representations (TNRs) and its corollary theories (Oiler 1993, 1995, 1996a). Even isolated conversational vignettes that aim for authenticity ought, according to the theory (Taira 1992) to be subordinated to a meaningful story line (i.e., a true or at least plausible narrative). Scrambled snapshots of experience ought not to be used (cf. Al-Fallay 1994, Jespersen 1904, Oiler & Richard-Amato 1983, Oiler 1993). Tests also can be expected to work better as they are made more and more to resemble the kinds of language uses that ordinary discourse most commonly gives rise to, i.e., those of the TNR variety (Oiler & Jonz 1994).

DOI

10.58837/CHULA.PASAA.27.1.3

First Page

39

Last Page

63

Recommended Citation

Oller, Jr., John W. (1997) "On the Relation between Language Teaching and Testing," PASAA: Vol. 27: Iss. 1, Article 3.
DOI: 10.58837/CHULA.PASAA.27.1.3
Available at: https://digital.car.chula.ac.th/pasaa/vol27/iss1/3