THE ROLE OF FORMAL INSTRUCTION IN SECOND LANGUAGE ACQUISITION

 Introduction

 

This chapter looks at second language acquisition in a classroom setting. It considers whether instruction makes a difference to SLA. This is an important issue, because it addresses the question of the role played by environmental factors in SLA. It is also an important educational issue, as language pedagogy has traditionally operated on the assumption that grammar can be taught.

Two board types of SLA can be identified according to the acquisition setting; 

1. Naturalistic SLA.
2. Classroom SLA.

It was pointed out that classroom discourse can be distorted, in comparisons to naturally occurring discourse. An important question therefore is in what ways this distortion, which is largely brought about by the attempt to instruct rather than to converse, affects the route and rate of SLA in the classroom. By considering how formal instruction affects SLA it is possible to address the wider issue of the role environmental factors. 

In many instructional methods an assumption is made that focusing on linguistic form aids the acquisition or grammatical knowledge or to put it another way, that raising the learner’s consciousness about the nature of target language rules helps the learner to internalize them.

In the case of deductive methods this is self-evidently the case. But it is also true in ‘habbit methods’ methods such as audio-legal lingualism , as the purpose of the practice provided is to focus on specific linguistic forms, which the learner is encouraged to induced and of which ultimately he will form a more or less, conscious mental representation. 

Another assumption of formal instruction is that the order in which grammatical features are taught will govern the order in which they are learnt. Language syllabuses are organized in such a way as to facilitate the correlation between the teaching order and the learning order.

The investigation of the role of formal instruction can be undertaken in two ways. First, an answer to the question ‘Does formal instruction aid SLA?’ can be sought. Secondly, the question ‘What kinds of formal instruction facilitate SLA the most?’ can be tackled. In the first question there is an assumption that all types of formal instructions share certain basic premises and that it is, therefore possible to talk generically of ‘ Formal instruction’. In the second question there is an assumption that formal instruction in general is facilitative and that the important issue is what is distinguishes the more successful from the least successful types. 

What the different instructional methods had in common was a focus on form, manifested, for instance, in the provision of feedback by the teacher for correcting formal errors.

This chapter has four sections. The first examines its effect on the route of SLA. The secondly examine its effect on the rate/success of SLA. In the third section, explanations of the result reported in the first two sections will be reviewed. Finally, the conclusion briefly considers the implications for both SLA theory and language pedagogy.

The effects of formal instruction on the route of SLA

The route of SLA was considered in term of general sequence of development and the order in which specific grammatical features were acquired. The evidence for the reported universality of the sequence and the minor differences in the order come from (1) morpheme studied (2) longitudinal studies. These studies how ever were of either pure naturalistic SLA or mixed SLA. The morpheme and longitudinal studies will again be considerers separately.

Morpheme studies of classroom SLA

The morpheme studies can be divided into two groups. In the first group are five studies that investigated second language learners. In the other group are four studies investigated foreign language learners

Perkins and Larsen freeman (1975) investigated the morpheme; they used two tasks to collect data;

1. A translation test.
2. A description task based on a non-dialogue film.

On (1) the morpheme orders before and after instruction differed significantly, but on (2) there was no significant difference. In other words, the teaching and learning orders were different. Taken together, these studies suggest but do not prove that formal instruction does not alter the order of acquisition of grammatical morphemes when the learners is engaged in language use and is focused on meaning.

This general conclusion holds true irrespective of whether the learners are children or adults and most interestingly, irrespective of whether the learners are in foreign or second language environments. Formal instruction appears, then to have only a negligible effect on the morpheme order manifest in spontaneous language use. However, morpheme orders measure accuracy rather than acquisition. In order to obtain a more reliable picture of the effects of instruction on L2 development, it is necessary to turn to the longitudinal studies of translation structures.

Longitudinal studies of classroom CLA

Curiously the case-study approach, so central to the methodological baggage of first and second language acquisition researchers has not typically, been thought sensible for learners in class. There are very few longitudinal studies of classroom SLA. The three that will be discussed here are Felix (1981), Ellis (1984a) and Schumann (1978b). The available longitudinal evidence, therefore is even slighter that provided by the morpheme studies. The general teaching method was a traditional audio-lingual one, the grammatical structures that Felix report on are negation, interrogation, sentence types, and pronoun. For each structure, parallels were found between tutored and naturalistic SLA. In a classroom where the instruction is very formal, learners are constantly being forced to produce structures they are not ready for. Felix suggests that they solve the problem that this poses for them in one of two ways. Either they select random from the structures in their repertoire, irrespective of syntactic or semantic appropriateness, or they follow the same rules that characterize the early stages of naturalistic language acquisition.

Ellis examined negatives, interrogatives, and a number of verb phrase morpheme. All of these structures were formally taught at one time. When the communicative speech produced by the learners in the classroom was analysed, it was shown to display a pattern of development more or less identical to that observed in naturalistic SLA.

In Schumann’s study a deliberate attempt was made to teach an adult L2 learner how to negate. This took place in the context of a longitudinal study of what was otherwise naturalistic SLA. Prior to the instructional experiment the learner’s negative utterances were collected, Schumann concluded that the instruction influenced the learner’s production only in test-like situations, while normal communication remained unaffected.

Taking these studies together, the following can be hypothesized:

1. Instruction does not circumvent the processes responsible for the sequence of development evident in transitional structures such as negatives an interrogative in naturalistic SLA.
2. When classroom learners are required to produce structures beyond their competence, idiosyncratic forms are likely to result.
3. The distorted input may prolong certain stages of development and slow down the emergence of some grammatical features.
4. Classroom learners are able to make use of knowledge acquired through formal instruction when they are focused on form.

Summary

Morpheme studies and longitudinal studies of SLA together indicate that although formal instruction may develop L2 knowledge, this knowledge manifests itself in language use only where the learner is attending to form. It does not, therefore, except in relatively minor ways, affect the natural route of SLA which is evident in communicative speech. To use the distinction between sequence and order of development, we can say that the overall sequence of development is not affected by formal instruction, while order of development is hardly disturbed either. Formal instruction influences knowledge only at the careful end of the interlanguage stylistic continuum, not the vernacular end. These conclusions, however, are necessarily tentative, as there have been few studies of classroom SLA, particularly longitudinal.

This section has examined three theoretical positions which provide explanations of why formal instruction does not affect the natural sequence of SLA but does facilitate more rapid development. The non interface position proposed by Krashen claims that ‘acquisition’ and ‘learning are separate. Because ‘acquisition’ is responsible for the natural sequence, the ‘learning’ that results from formal instruction cannot influence it. However, classroom provides opportunities for comperhesinble input will accelerate ‘acquisition’.

The interface position also posits two types of L2 knowledge, but Argus that they are related so that ‘learning’ (or explicit knowledge) can become ‘acquisition’ (or implicit knowledge) when it is sufficiently practised. A weaker version of this position, however, states that ‘learning’ does not so much turn into ‘acquisition’ as facilitate it, when the learner is ‘ready’. The variability position differs from the other two positions in that it recognizes a variety of different ‘styles’, each calling on knowledge types that vary in terms of analicity and automaticity. Different task require the utilization of different kinds of knowledge. Formal instruction contributes directly or indirectly to there internalization of these different knowledge types and in so doing enables the classroom learner to perform a wider range of linguistic tasks than the naturalistic learner.

All there positions provide arguments to account for the result of the empirical research into the effect of formal instruction. These have been considered in some detail. At the moment there is insufficient evidence to make a clear choice between them. It is not likely that such evidence will be forthcoming until there are more qualitative studies of the classroom discourse that result from formal instruction and of the linguistic development that such discourse induces.

Conclusion: some implications

This chapter began by asserting that the investigation of the role played by instruction in SLA was of significant for both SLA theory and language pedagogy. In this conclusion I shall briefly consider some of the implications.

In order investigate the role of instruction in SLA, it is necessary to separate out the effects that formal instruction has on the route of SLA and on the rate/ successes of SLA. Where the route is concerned, formal instruction appears to have no major effect. The overall sequence of development associated with natural communicative language use does not change, while only a few minor and temporary differences in the acquisition of specific grammatical features have been observed. Thus classroom SLA appears to involve the same processing strategies as naturalistic SLA. Where the rate/ successes is concerned, instruction is facilitative, although only in terms of relative utility, not in terms of absolute effects. These results must be treated tentatively, as there has been little empirical research.

There different positions have been advanced to explain classroom SLA. The non interface position, associated with Krashen (1982), distinguishes ‘acquired’ and ‘learnt’ knowledge and argues that they are separate. This position offers a convincing explanation a why formal instruction fails to influence the natural route of SLA, as this is a reflection of ‘acquisition’. The explanation it gives for why formal instruction aids the rate/ successes of SLA is less clear. The interface position, associated with Stevick (1980) and Sharwood-Smith (1981) among others, claim that ’learnt’ or explicit knowledge can turn into ‘acquired’ or implicit knowledge if there is enough practice. This position offers an explanation for the rate/ successes finding, but it less convincing about the route finding. The variability position, associated with Tarone (1983) and Bialystok (1982), sees acquisition and language use closely linked, such that different types of knowledge arise from and are required for the performance of different language tasks.

This position deals comfortably with the route finding (which occurs in a particular kind performance) and can explain the rate/ success finding if it is assumed that the learner who has access to a variety of different knowledge types will outperform one who is more reliant on a single kind of knowledge. However, it is premature to choose from among these positions.

The study of the role of instruction in SLA has implications for both SLA theory and language pedagogy. In the case of the former, is stresses in importance of act knowledge the structural properties of SLA which are relatively immune to environmental differences. Where language pedagogy is concerned, it sheds light on the code-communications dilemma, although once again it would be premature to come to any firm conclusions about the effectiveness of formal grammar teaching.

Second language acquisition theory

Studying the role of instruction can throw on the contribution of environmental factors in SLA. The classroom environment provides a different kind of input from a natural setting. If environmental factors are important of SLA, it might be predicted that (1) The acquisitional route in the two setting will be different, and (2) the rate/ success of SLA in the two setting will also differ. The research reviewed in the earlier sections shows that (1) does not arise, while (2) may. The failure of the classroom setting to influence the route of SLA can be explained in two ways. First, it might be taken to show the real determinants of SLA are learner-internal rather than environmental factors. That is, despite differences in input, the L2 learner will follow the same developmental path, because, although there are differences in the types of input to be found in each setting, there are also similarities.

The natural sequence is the product of one type language use-spontaneous communication-which, although restricted in classroom context, does take place. The first explanation follows a native’s interpretation. What is the quite clear, whatever interpretation is adopted, is that SLA possesses certain structural properties which are immune to environmental differences in learnt in classroom and natural setting. The effect of environmental factors appears to be restricted largely to how quickly and how much of the L2 the learner acquires.

Language pedagogy

Looking at instruction from the view point of the learner rather than the teacher is salutary. It puts into perspective the widely held view that if instruction is based on a sound syllabus and employs motivating techniques, acquisition will result. Unless account is taken of the structural properties of SLA, success is by no means certain.

Teacher ought not to feel obligated to ensure that his teaching also follows it, as it is far more important that the teacher works from a syllabus which he finds logically acceptable. Brumfit argues that language teaching will be most successful when it follows as well-worked out plan which directs and organizes what the teacher does. The second reason for reticence is that, although there is a fair degree of agreement among SLA researches concerning what happens in SLA.

There is a far less agreement about why it happens in the why it does. This has been evident in the different positions adopted to explain the result of research into the effects of formal instruction. Briefly outline what attitude to the code-communication dilemma is held by protagonists of each of the three positions considered in the previous section

1. The non-interface position.

Krashen (1982) pays close attention to the role of grammar teaching in classroom SLA. He sees two uses for it. First, it enables the monitor to function by providing for ‘learning’. However, monitor use is restricted to occasions when the learner has time to access his ‘learnt’ knowledge, and is also restricted by the fact that only a small sub-section of total rules of a L2 are ‘learnable’. The second use of grammar teaching is to satisfy learners’ curiosity about the nature of the L2 grammatical system ‘grammar appreciation’. The use of conscious grammar is limited, therefore, that the role of teaching is to afford opportunities for communications, rather than to draw attention to the L2 code.

Krashen (1981b) lists the defining characteristic of what he considers an effective pedagogical programme;

1. The classroom input must be comprehensible.
2. The programme must consist of ‘communicative activities’, as only these will ensure that he input is interesting and relevant.
3. There should be no attempt to follow a grammatically sequenced programme.
4. The input must be of sufficient quantity (hence importance of extensive reading).

1. The interface position

Where the non-interface position emphasizes the importance of communication and minimizes the importance of the code, the interface position asserts the contribution of the code. Sharewoos-Smith (1981) sees grammar teaching as a short cut to communicative ability. That is, the adult learner who has his attention drawn to features of the code can practise these, both in and out of the classroom, until he can use them subconsciously in fluent communicative speech. Consciousness-raising, therefore, does not require that the learner is able to verbalize the rules he has learnt. For Sharwood-Smith, then, the important issue is not whether the code should be taught, but in what way it should be taught.

1. The variability position

The variability position stresses importance of matching the learning process with the type of instruction. Instruction must consider the specific goals of the learner and attempt to provide the appropriate form of knowledge to achieve those goals. The ‘goals’ refer to the type of language use that the learner needs (or wants) to engage in. if the goal is to participate in natural conversation, the learner will need to develop his vernacular style by acquiring L2 knowledge that is automatic but unanalysed.

This can be achieved directly by means of instruction that emphasizes communication in the classroom. It may also be achieved indirectly by teaching that focuses on the code, if there are also sufficient practice opportunities to trigger the passage of knowledge from the careful to the vernacular style.