Relevance of the study of interlanguage to teaching EFL at the tertiary level

Abstract

To learn a second/foreign language, the learner takes a linguistic journey from his/her mother tongue to the target language and naturally constructs a personal linguistic system in the interim time. This individual system termed ‘interlanguage’ is a single and unique one which is yet to conform to the target language norms and evidently incorporates linguistic deficiencies or errors exhibiting the learner’s current linguistic level and implying what he/she need acquire to reach a standard of the target language. Hence, the present research has been designed to investigate and examine the relevance of the study of the interlanguage of 21 tertiary level students learning English as a foreign language (EFL). To carry out the study, an experimental group consisting of the 21 students and a control group having another 21 students of the same level have been used. Based on the findings, the researcher makes some linguistic and pedagogic recommendations.

Keywords

Interlanguage, relevance, teaching EFL, tertiary level

Introduction

A second/foreign language learner takes a linguistic journey from his/her mother tongue to the target language and naturally constructs a personal linguistic system in the interim time. This individual system is termed ‘interlanguage’ (Selinker 1969, 1972), ‘approximative system’ (Nemser 1971), ‘transitional competence’ (Corder 1967), or ‘idiosyncratic dialect’ (Corder 1973). In a narrower sense, interlanguage refers to the intermediate status of the second/foreign language learner’s system between his/her mother tongue and the target language. In a broader sense, it stands for the second/foreign language learner’s present knowledge of the language he/she is learning (Spolsky 1989). In a general sense, interlanguage is defined as ‘the interim grammars constructed by the second-language learners on their way to the target language’ (McLaughlin 1987:60).

According to Selinker (1972), the development of interlanguage depends on five central cognitive processes involved in second/foreign learning - first language transfer, transfer of training, strategies of second/foreign language learning, strategies of second/foreign language communication, and overgeneralization of the target language linguistic material. However, Adjemian (1976) contradicts Selinker (1972), and emphasizes the natural or universal aspects of interlanguage. Adjemian (1976) focuses on the dynamic character of interlanguage systems, that is, their permeability, and maintains that interlanguage is not stable, rather it is always in a state of flux. It signifies that a second/foreign language learner’s language constantly changes and/or develops. In this connection, we could consider Ellis (1994: 352) who clearly postulates ‘these mental grammars are perceived as dynamic and subject to rapid change’. It may happen due to having the linguistic influence of the learner’s first language or due to stretching, distorting or overgeneralization of the rules of the target language by the learner when he/she attempts to generate the intended meaning; or both may occur simultaneously.

That is, interlanguage is an individual, single and unique system (Adjemian 1976) which is yet to conform to the target language norms and evidently incorporates linguistic deficiencies or errors exhibiting the learner’s current linguistic level and implying what he/she need acquire to reach a standard of the target language. Notwithstanding, there exists a substantial degree of uniformity in the characteristics of interlanguage and in the types of errors of various second/foreign language learners. For instance, Bengali speaking learners commit a common error ( and/or make a common mistake) by missing the ‘-s’ to be added to the verb used in a sentence in the simple present tense and having a third person singular subject (Maniruzzaman 2006). Andersen (1978) and Hyltenstam (1977) rightly report the important character of interlanguage that there is individual variability within uniformity.

Furthermore, Tarone (1979) explains interlanguage as a set of styles dependent on the context of use. Research reveals that the utterances of the learner are systematically variable in at least two senses. Firstly, the linguistic context may have a variable impact on the learner’s use of related phonological and syntactic structures. Secondly, the task used for the elicitation of data from the learner may have a variable effect on the learner’s production of related phonological and syntactic structures. Tarone (1979) then concludes that interlanguage speech production varies systematically with the context and elicitation task. Mitchell and Myles (1998:11) consider Towell and Hawkins (1994: 5), and lend support to Tarone (1979)-

... learner language (or interlanguage, as it is commonly called) is not only characterized by systematicity. Learner language systems are presumably - indeed, hopefully - unstable and in course of change; certainly, they are characterized also by high degrees of variability.

In sum, interlanguage or the second/foreign language learner’s interim language is permeable, variable within uniformity, systematic as well as universal in nature. It is, therefore, assumed that the study of interlanguage can help determine what the learner already knows at a certain point of time and what he/she has to be taught when and how in a particular second/foreign language teaching programme. Based on this assumption, the present study has been designed to investigate and examine the relevance of the study of interlanguage to teaching English as a foreign language (EFL) to Bengali speaking learners at the tertiary level.

Method

Subjects

The study has been conducted with an experimental group and a control group, each of which consists of 21 tertiary level students learning English as a foreign language (EFL) in the Language Centre at Jahangirnagar University. The subjects possess several characteristics in common - having the same mother tongue, belonging to the same age group, studying throughout in the Bengali medium, already receiving twelve years of formal instruction in EFL at the rate of about four hours per week, and having the same objective, that is, to achieve a good command of the basic skills of the target language.

To choose the subjects in both the groups, the simple random procedure has been applied since this type of sampling is easy and inexpensive for subject selection and data analysis (Sudman 1976). In the simple random procedure, a list of the total population members is arranged in a random order, and every nth name is then picked out. Every member of the population has an equal chance of being selected in the sample.

Instrumentation

The Oxford Placement Test being made up of two subtests - Test of Grammatical Structures (Appendix-A) and Test of Reading and Listening Skills (Appendix-B) constructed by Allan (1985) coupled with a Free Composition Test (Appendix-C) and a Speaking Test (Appendix-D) designed by the present researcher has been employed to measure the pre-programme as well as post-programme linguistic level of both the experimental group and the control group. All the tests have also been used to analyze the interlanguage of the subjects of the experimental group.

Oxford Placement Test: The Oxford Placement Test is one of the most widely used measures of ESL/EFL proficiency in the UK. It has been reliably exploited for ascertaining English proficiency of the students entering undergraduate and postgraduate studies in British universities. Since the subjects of the present study are of the undergraduate level, the researcher deems that the Oxford Placement Test effectively and efficiently measures the subjects’ proficiency in writing, reading and listening. This test is constituted of two subtests:

a. Test of Grammatical Structures: This is a written multiple-choice test of grammatical structures of English covered by the vast majority of course books, whether functional or structural, in the range from elementary to advanced. The test has one hundred multiple-choice items, each carrying one mark. A time limit of one hour thirty minutes is set for the test.
b. Test of Reading and Listening Skills: This is a test of reading and listening skills consisting of one hundred multiple-choice items, each of one mark. In the test, the testee’s performance is dependent on the knowledge of the sound and writing systems of the English language and upon the ability to make use of this knowledge. The recorded material controls the time for the test.

Free Composition Test: This test is designed to assess the subjects’ linguistic competence in organizing and presenting relevant ideas in writing. Controlled compositions are error-provoking while free compositions are error-avoiding. In this investigation, spontaneous prediction procedure has been followed, and the subjects are asked to write a free composition on any one of the topics –‘Your Future Plan’, ‘Importance of English’ and ‘Family Life’. A time limit of one hour is set for the test, and it is marked in 20.

Speaking Test: This test is made up of the items supposed to determine the subjects’ ability to express different attitudes, feelings and emotions as well as their competence in communicating in real life situations. The subjects are required to answer the test in 45 minutes, and it is marked in 25.

Data collection and analysis

To ascertain the subjects’ current level of proficiency in the English language, the Oxford Placement Test, Free Composition Test and Speaking Test have been administered to both the experimental group and the control group prior to the start of their EFL programmes. The data collected from both the groups are checked and scored by hand; and their average proficiency before the start of the EFL programmes or the pre-programme linguistic level of each of the groups is computed.

Interlanguage has so far been studied by a good number of psycholinguists (e.g. Brown 1973, Dulay and Burt 1974) in two manners - the morpheme studies and error analysis. However, error analysis has been adopted to serve the purpose of the present study, and is performed to detect the linguistic deficiencies or errors/mistakes in the data gathered from the experimental group before the start of the EFL programme. ‘Let the errors determine the categories’ approach (Norrish 1983) is used to group the errors (/mistakes) as they are basically related to particular grammatical, semantic, phonetic, phonological, lexical and comprehensive problems. This approach is employed since it has the advantage of allowing the errors (/mistakes) themselves to determine the categories chosen: by a process of sorting and resorting the categories will eventually define themselves.

Then both the groups have completed their respective EFL programmes, each of 72 contact hours. The control group has been taught according to the syllabus already designed by the Language Centre whereas the experimental group has been taught according to the syllabus designed by course teacher in the light of the results of the error analysis of the data collected from the subjects of this group before the start of the EFL programme. At the end of the programmes, the Oxford Placement Test, Free Composition Test and Speaking Test have again been given to both the groups, and their average proficiency after the completion of the EFL programmes or the post-programme linguistic level of each of the groups is computed. In addition, error analysis is again carried out to uncover errors/mistakes in the data gathered from the experimental group after the completion of the EFL programme.

Presentation and interpretation of the findings

The analysis of the data collected from both the experimental group and the control group by using the Oxford Placement Test, Free Composition Test and Speaking Test produces four types of findings.

Pre-programme linguistic level

The average marks secured by both the experimental group and the control group in the Oxford Placement Test, Free Composition Test and Speaking Test before their receiving any instruction in the EFL programme are shown in Table-1 below:

Abbildung in dieser Leseprobe nicht enthalten

Table-1

The findings presented in Table-1 reveal that the average marks of the experimental group are a good approximation to those of the control group. While the experimental group secures 40, 33, 6 and 11.5 in Test of grammatical Structures, Test of Reading and Listening Skills, Free Composition Test and Speaking Test respectively, the control group gets 41, 31, 6.5 and 12 in the same tests respectively. That is, the pre-programme linguistic levels of both the groups are almost the same.

These results might be attributed to the fact that the subjects belonging to both the experimental group and the control group possess almost everything in common, such as the same mother tongue, the same age group, similar educational background, similar exposure to EFL, and the same objective of learning EFL. However, the insignificant difference between the scores of the two groups can be supported by the existing finding that there is individual variability within uniformity of interlanguage (Andersen 1978 and Hyltenstam 1977).

Pre-programme linguistic deficiencies of the experimental group

The linguistic deficiencies detected by the error analysis of the data collected from the experimental group prior to the start of the ELT programme are summed up and discussed as follows:

Test of Grammatical Structures: The Test of Grammatical Structures taken by the subjects of the experimental group exhibits that the subjects have considerable deficiency in linguistic competence as well as performance. They have committed both ‘errors of competence’, that is, failures in using appropriate and correct language forms and rules to transmit or perceive messages, and ‘errors of performance’, that is, mistakes occurring due to the subjects' insufficient practice, indifference, fatigue, or anxiety. More specifically, the subjects have inadequate knowledge of and ability to use appropriate and correct count/non-count nouns, e.g. ‘information, homework’, determiners, e.g. ‘few, little, some’, relative pronouns, e.g. ‘that, which, where’, connectors, e.g. ‘since, because, or’, adverbs, e.g. ‘always, recently, hardly, scarcely, really’, modals, e.g. ‘could, will, shall, would, may, can, might, must, need’, verb forms and tenses, subject-verb agreement and prepositions. Moreover, they are found to have encountered difficulty in forming negative sentences, questions, passive sentences, reported speech, conditionals, tag questions, and the like.

[...]