Self-assessment accuracy is a condition of learner autonomy. If students can appraise their own performance accurately enough, they will not have to depend entirely on the opinions of teachers and at the same time they will be able to make teachers aware of their individual learning needs. The purpose of this article is (1) to summarize the literature on self-evaluation of foreign language skills and (2) to show what it could mean to teachers and researchers. The conclusions of several self-assessment studies are somewhat contradictory, but these differences seem to support Stephen D. Krashen's Monitor Model/theory. Therefore both teachers and researchers should keep in mind that foreign language learners' self-estimates may be influenced to a varying degree by the use of the Monitor.

Self-assessment accuracy is a condition of learner autonomy. If students can appraise their o w n performance accurately enough, they will not have to depend entirely on the opinions of teachers and, a t the same time, they will be able to make teachers aware of their individual learning needs. The purpose of this article is (1) to summarize the literature on self-evaluation of foreign language skills and (2) to show what it could mean to teachers and researchers. The conclusions of several self-assessment studies are somewhat contradictory, but these differences seem to support Krashen's Monitor Model/ Theory. Therefore, both teachers and researchers should keep in mind that foreign language learners' self-estimates may be influenced to a varying degree by the use of the Monitor. 313La nguage Learning Vol. 39, No. 3 in 1976. The topic of self-assessment (variously termed selfrating, self-appraisal, self-control, etc.) has apparently just begun to expand as a distinct field of interest in language testing and evaluation. The Modern Languages Project of the Council of E u r o p d e d i c a t e d as it is to promoting studentcentered, needs-oriented, and motivation-based learningseems to have played a crucial role in this development.Research reports dealing with self-evaluations of foreign language skills were produced in different parts of the world. Most of them involved high school or university undergraduate students and, in the majority of cases, students of English as a foreign language (other languages studied were French and, in one instance, Swedish in which the learners were adult migrants).Few reports were concerned specifically, let alone exclusively, with self-appraisal practices. Frequently, self-rating occurred only because student input was needed by course developers for a more general needs analysis; or else it was one component of a much more extensive research design.The studies that were reviewed are compared in Table 1. Each study is identified by the investigatods) who described it.The emerging pattern is one of consistent overall agreement between self-assessments and ratings based on a variety of external criteria. The accuracy of most students' selfestimates often varies depending on the linguistic skills and Several studies included quantitative comparisons between self-appraisals and more objective measures of proficiency, usually in the form of calculations of Pearson productmoment correlation coefficients. Values ranging from .50 to .60 are common, and higher ones not uncommon. What this means

While there seems to be enough empirical evidence to substantiate the claim that language learners have the ability to evaluate their own performance, many researchers still feel that self-assessments are rarely as reliable or valid as standard test results or teachers' estimates of their students' linguistic competence. This skepticism may well be justified, because a large number of selfassessments are based almost entirely on appraisal questionnaires and the use of such instruments without extensive training or preparation is bound to produce incorrect evaluations. In the DLIFLC study, however, the experimenter used 'conventional' examinations both for testing purposes and for selfassessment purposes. This method yielded interesting results. It also had limitations, but it generally looked promising. I IntroductionIn a recent review of the literature on self-assessment of foreign language skills (Blanche and Merino, 1989), the writer noted that in several of the studies which were examined, investigators had found that good learners tended to underrate their abilities. Conversely, overestimation cases involved weak students to a greater extent than high achievers. Low self-estimates, as well as low correlations with the results of more objective tests, had consistently been reported only in the two areas of pronunciation and grammar -which seemed to indicate that most learners find it easier to assess their purely communicative skills.The emerging pattern was one of consistent overall agreements between self-appraisals and ratings based on a variety of external criteria. A number of studies included quantitative comparisons between these two types of. measures, usually in the form of Pearson product-moment correlation coefficients. Values ranging from .50 to .60 were common, and higher values not uncommon. What this meant was that a set of self-estimates (such as answers to a questionnaire) tended to carry about the same weight as any of the various parts (sub-tests) of a standardized testing instrument. However, many researchers had warned that the reliability and the at UNIV ARIZONA LIBRARY on

Not everyone can use pragmatic competence to maintain a conversa tion as a cooperative venture, and comparatively few attempts have been made to account for the development of the linguistic abilities that one may need to have in order to exhibit such competence. Yet many pragmatic components of language are probably acquired in learning sequences which correspond to successive stages in the development of children. In this article, the author (summarizes cur rent linguistic theory in the area of pragmatics,) reviews the empirical evidence which suggests that the acquisition of pragmatics is a developmental process and comes to the conclusion that teachers could and should include pragmatic components in the language cur riculum.