© 1991 by Oxford University Press
Articles |
Error: Some Problems of Definition, Identification, and Distinction
University of Kassel
This paper provides various procedural criteria for performing error analysis, and introduces two new dimensions of error, ‘extent’ and ‘domain’, which serve to differentiate errors systematically. Section 1 examines previous approaches to error analysis (1.1), offers a working definition of error (1.2), and considers the problems involved in error identification, particularly of a spoken corpus (1.3), with regard to both ‘global’ and ‘local’ errors (1.4). Attention is drawn, too, to the middle ground of advanced learner performance, which is neither fully erroneous nor fully nativelike (1.5). Section 2 examines an advanced learner spoken corpus for error; subjects, methods, and aims are presented (2.1, 2.2); the definition in 1.2 is applied (2.3); error identification by a native speaker panel is reported (2.4), and procedural criteria for distinguishing between ‘type’ and ‘token’ are developed (2.5); the most borderline error cases are scrutinized, and it is suggested that proximate cumulation of infelicity may make for perceived error in some cases (2.6). Section 3 introduces and defines error extent and domain (3.1, 3.2), illustrated by examples from the corpus. The concepts are applied to define three distinct types of lexical error (3.3), to deal systematically with error embedded within error (3.4), and to distinguish between type and token (3.4).
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. Salem The lexico-grammatical continuum viewed through student error ELT J, July 1, 2007; 61(3): 211 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lawley A Preliminary Report on a New Grammar Checker to Help Students of English as a Foreign Language Arts and Humanities in Higher Education, October 1, 2004; 3(3): 331 - 342. [Abstract] [PDF]
|
|