Ngôn ngữ học ứng dụng

INTERLANGUAGE
Nguyễn Thị Phương Loan
Nguyễn Thị Vân Mai
Nguyễn Thị Thu Thuỷ
CONTRASTIVE ANALYSIS
Definitions
Different views on C.A
2.1. Samson’s view
2.2. C.A as Interlanguage Study
2.3. C.A as “pure” or “Applied” Linguistics
2.4. C.A and Bilingualism
The importance of C.A
Aspects of C.A
4.1. Contractive Phonological/Phonetics Analysis
4.2. Morphosyntactic Contrastive Analysis
4.3. Lexical Contrastive Analysis
Criticism of C.A
5.1. The Predictive Validity of C.A is Doubtful
5.2. Linguistics Difference does not equal learning difficulty
5.3. C.A is based on an Underlying Static Idea of Language
Development Originated from C.A


CONTRASTIVE ANALYSIS
Error analysis
7.1. Background to Error Analysis (E.A)
7.2. Definitions of Errors
7.3. C.A vs E.A
7.4. Types of Errors
7.5. Reasons for Making errors
7.5.1. Overgeneralization
7.5.2. Transfer
7.5.3. Simplification by Omission
7.5.4. Non-system Errors
7.5.4.1. Communication Strategies
7.5.4.2. Performance Errors
7.5.5. Errors due to the effects of Teaching
7.5.6. The L’s internal Syllabus
7.6. Learning Process



1. DEFINITION OF C.A
Contrastive analysis, which is also called Contrastive Linguistics, means a systematic comparison of the linguistics systems of two or more languages
The main idea behind C.A was that a detailed comparative and contrastive study of L1 and L2 might reveal exactly which problems learners with the same L1 have in learning L2.

Early contrastive studies has a clearly practical objective: language teaching was to be improved by predicting the learner`s difficulties and the results of C.A would then be built into syllabuses and teaching materials.
1. DEFINITION OF C.A
2. DIFFERENT VIEWS ON C.A
2.1. Samson`s view

"CA is a linguistic enterprise aimed at producing inverted (i.e, contrastive, not comparative) two-valued typologies (a CA is always concerned with a pair of languages), and found on the assumption that languages can be compared"
Samson (1975: 4)

"On the one hand, linguistics treat individual languages: English, French, Chinese ,and so on. On the other hand, they consider the general phenomenon of human language, of which particular languages are example"
2.2. C.A AS INTERLANGUAGE
Any branch of linguistics has its object of study of human languages.

A branch is likewise not primarily concerned with languages in the conventional sense.
It is interested in the emergence of these languages rather than in the finished produce.
C.A belongs to interlanguage study, and, since, "emergence" is an evolutionary concept (in de Saussure`s sense) C.A is to be viewed as diachronic rather than synchronic in orientation.
2.2. C.A AS INTERLANGUAGE
The study of L2 or foreign language learning is concerned with a monolingual becoming a bilingual: two languages are involved, the L1 and the L2, so we have here a true case of interlingual diachronic study.

Another branch of linguistics that is concerned with the transition from one language to another is translation theory, or the study of how texts from one languages to another are transformed into comparable texts in another language. Here, however, the focus of interest is not on learning but on process of text-replacement.

2.3. C.A AS "PURE" OR "APPLIED" LINGUISTICS
2.4. C. A AND BILINGUALISM
C.A`s concern is with what we study the person who has competence in two languages we are dealing with individual bilingualism.
C.A is concerned with how a monolingual becomes bilingual.
C.A is concerned with the way in which NL affects foreign language learning in the individual, whereas Weinreich`s and Haugen`s work studied the long-termed effects spanning a generation, of language contact.
C.A is concerned with parole, their work with language; C.A with interference, they with integration
3. THE IMPORTANCE OF C.A
The results of the C.A are built into language teaching materials, syllabuses, tests, and research.

Different textbooks will have to be contrastive analysis so that they may predict the likely errors of a given group of Ls and thereby to provide the linguistic input to language
Originally, the main emphasis of CA was on grammar and phonology for the obvious reason that the close systems of grammar and phonology lend themselves better to systematic CA then the more elusive areas of lexis and culture, but the general absence of contrastive lexical and cultural studies also reflected where the emphasis lay in linguistics in those days.
Krzeszowski points out the linguistic components of CA in 5 stages:
4.1. Contrastive Phonological/ Phonetic Analysis
It is the area of phonology that the predictions of CA work best. At the very outset of learning, most learners have a low phonetic competence in the L2: in rhythm, in stress and intonation, and in individual speech sounds, L1 patterns will normally be easily discernible. On the surface, at least, the original hypothesis appears to work well: the more numerous mutually exclusive forms and patterns there are, the greater the problems there will be for the learner`s pronunciation. Yet, empirical studies have shown that this is an oversimplification. L1 is not only the factor influencing L2 pronunciation; frequency of phonemes is, for example, another variable. Also, L1 phonological rules interact with L2 phonological rules.
Further, an early study (cf. Nemser, 1971) showed asymmetry of perception and production of L2 sounds.
There are 4 steps involved in executing a CA of the sound systems of 2 languages
Morphosyntactic Contrastive Analysis
The methodological principle of CA is to compare and contrast the structures of the two languages by means of the same model. The choice of model is fairly straightforward in phonology, where there are basically only two choices: taxonomic phonology and generative phonology. For grammar, on the other hand, the number of models actually used is larger. The first contrastive analysts were structuralists, who relied on immediate constituent analysis. Among the other models used in morphosyntactic CA, contrastive generative grammar and case grammar are probably the most important. However, all models used have been criticized. Models which were originally devised for the analysis of one particular language such as English are not necessarily very well suited to the description of some other languages.
A step-by-step algorithm for the execution of
a representative CA by Levelt (1970)
4.3. Lexical Contrastive Analysis
The area of lexis have been much less subjected to CA than in phonology, morphology and syntax. This is surprising when the vagueness inherent in cross linguistic semantic comparisons and the vastness of the vocabulary of any languages are considered. The impossibility of comparing and contrasting the whole vocabularies of two languages will be obvious to anyone. Lexical CA is restricted to comparisons of extremely limited areas, for example, certain semantic fields. However, this does not mean that lexical CA is of little or no avail. From a linguistic point of view, comparing and contrasting lexical items in one language with the corresponding items in another many reveal semantic features that would otherwise have remained unnoticed. Also, learner need bilingual, frequency-based dictionaries, and selected we-known problems items are accurately treated. It is above all in lexicography that fruitful applications of lexical CA may be expected.
Lyons asks us to consider the following sets of words in English
semantic components
Each lexeme is a complex of such components
Corresponding to the dictionary definition: Lamb is young sheep or young gregarious ruminant of the species ovis
Leech first distinguishes formal and substantive universals: claims for such universals, on the semantic level, would be:
5. CRITICISM OF CA
There is a discrepancy between the explicit aim of CA to be relevant to the practical problems of language teaching and the actual form which CA took, in its minute description of details with very little to offer the practicing foreign language teacher. The theoretical and methodological assumptions underlying contrastive analysis studies have been severely criticized. Some of the most important points of criticism concerning contrastive analysis studies as they pertain to the utility for L2 learning and teaching are the following:
The Predictive Validity of Contrastive Analysis is Doubtful
Linguistic Difference Does not Equal Learning Difficulty
Contrastive Analysis is Based on an Underlying Static Idea of Language

5.1. The Predictive Validity of Contrastive Analysis is Doubtful

Not all errors made by learners are due to L1 transference and the problem that learners have are not always predicted by CA. Teacher`s experience as well as research clearly show that the actual errors which learners make do not occur only where L1 and L2 are different. Making errors is an inevitable part of the language-learning process, but only some of these errors are of a kind which can beyond doubt be referred to L1 influence.

5.2. Linguistic Difference Does not Equal Learning Difficulty

Because a particular L2 feature is different from the same feature in L1, it does not necessarily follow that it is difficult to learn. A major problem here is that the two concepts` difference and difficulty have totally different reference frames. Difference and similarity are linguistic concepts. It is not clear how these terms should be delimited, but in theory, at least, they could be quantitatively measured. Difficulty, on the other hand, is a psychological concept inherent in the learner`s mind, and largely depends on what relevant prior knowledge the learner possesses.

5.3. Contrastive Analysis is Based on an Underlying Static Idea of Language
When grammatical and phonological variables are isolated from the situational context in which utterances occur, the implication is a very static idea of language. This does not tally well with the general view of dynamism being inherent in language and its use. Since the linguistic code is closely linked to wider contextual elements from an analysis of communication leads to an approach which is very narrow. The larger reference frames in which utterances occur also show important differences between speakers of different languages, and must not be neglected in contrastive studies. The distinction among phonological, grammatical, lexical, textual, and pragmatic levels may be important for the linguist, but all these levels have to be present for communication to occur.
6. DEVELOPMENT ORIGINATED FROM CA
The criticism above had important effects on the studies growing out of CA, but they have not destroyed the original basic idea of CA, that the learner`s L1 largely determines what problems the learner is going to meet. There can be no doubt whatever that L1 is an extremely important factor in L2 learning. Only, the original approach to CA was too narrow, and needed to be expanded in various directions. The early emphasis on grammar and phonology has been supplemented with other types of investigation. Among these, error analysis, performance analysis, interlanguage studies, transfer analysis, contrastive discourse and contrastive pragmatic may be specially mentioned.

7. ERROR ANALYSIS
7.1. Background to Error Analysis
7.1.1. History of E.A
7.1.2. Procedure of E.A
7.1.3. Role of E.A
7.2. Definitions of Errors
7.3. C.A vs E.A
7.4. Types of Errors
7.5. Reasons for Making errors
7.5.1. Overgeneralization
7.5.2. Transfer
7.5.3. Simplification by Omission
7.5.4. Non-system Errors
7.5.4.1. Communication Strategies
7.5.4.2. Performance Errors
7.5.5. Errors due to the effects of Teaching
7.5.6. The L’s internal Syllabus
7.6. Learning Process

7.1.1 HISTORY
Error analysis has a long history.
Prior to the early 1970s, error analysis consisted of little more than impressionistic collections of “common” errors and their linguistic classification. The goal of traditional error analysis were pedagogic – errors provided information which could be used to sequence items for teaching or to devise remedial lessons.
Then the enthusiasm for C.A grew so the interest in error analysis declined. According to behaviorist learning theory, the prevention of errors was more important than the identification of errors. It was not until the 1960s that errors were paid much attention to. One of the applied linguists who was the pioneer in this field is S. Pit Corder.


7.1.2 PROCEDURE OF ERROR ANALYSIS
A corpus of language selected. This involves deciding on the size of the sample, the medium to be sampled, and the homogeneity.
The errors in the corpus are identified.
The errors are classified. This involves assigning a grammatical description to each error.
The errors are explained. In this stage of the procedure, an attempt is made to identify the psycholinguistic cause of the errors.
The errors are evaluated. This stage involves assessing the seriousness of each error in order to take principled teaching decisions. Evaluation is necessary only if the purpose of error analysis is pedagogic.
Corder,S.P (1978). Error analysis and interlanguage


7.1.3 ROLES OF ERROR ANALYSIS
It concerns the linguistic type of errors produced by second language learners. Richards (1974) provides a list of the different types of errors involving verbs.

However, this type of information is not very helpful when it comes to understanding the learner’s developmental sequence.

Error analysis must necessarily present a very incomplete picture if second language learners produce. Because of this, an analysis of the linguistic types of errors produced by learners does not tell us much about the sequence of development.

It concerns the psycholinguistic type of errors produced by second language learners. A study of errors reveals that there is no single or prime cause of errors and provides clues about the kinds of strategies learners employ to simplify the task of learning a second language.

Error analysis is success in elevating the status of errors from undesirability to that of a guide to the inner working of the language learning process. Errors were no longer seen as “unwanted forms” but as evidence of the learner’s active contribution to second language acquisition.
7.1.3 Roles of Error Analysis
7.2 DEFINITIONS OF ERRORS
Brown (1994: 205):
A mistake is a performance error that is either a random guess or a ‘slip’ that is a failure to utilize a known system correctly. According to this definition, the speaker could make a mistake in her native language.

Errors are problems that a native speaker would not have. Errors are noticeable deviation from the adult grammar of a native speaker, reflecting the interlanguage competence of the learner.
Edge (1989):

A slip is what a learner can self - correct, and an error is what a learner can’t self – correct.
7.3 CONTRASTIVE ANALYSIS VS ERROR ANALYSIS
Contrastive Analysis is based on the following assumptions:

The main difficulties in learning a new language by interference from the first language.
These difficulties can be predicted by contrastive analysis.
Robert Lado (1957): The important of CA is errors and difficulties occur in our learning and use of a foreign language are caused by the interference of mother tongue.
7.4. TYPES OF ERRORS
Linguistic (intralingual and interlingual)
Psychological or developmental errors
Linguistic error is one that is produced by second language learners. Errors analysis must present a very incomplete picture of second language of second language acquisition because it focuses on only one part of the language second language learners produce.

Interlanguage requires identifying what the learners can do in proactive by examining both idiosyncratic and non - idiosyncratic and a synchronic type of error produced by learners do not tell us much about the sequence of development.
Psychological errors are ones which are produced by second language learners concerning the strategies used in interlanguage. It shows that there is no single or prime cause of errors, but learners’ errors emerged due to many other factors. These errors occur regardless of the learner’s language background.

Psychological errors can also be called developmental errors. This is because rather than reflecting the learner’s ability to separate 2 languages, they reflect the learner’s competence at a particular stage, and illustrate some of the general characteristic acquisition.

Developmental errors illustrate the learner’s attempt to build up hypothesis about the English language from his limited experience of it in the classroom or textbook.
7.5. REASONS FOR MAKING ERRORS
7.5.1. Overgeneralization

Generalization is a fundamental learning strategy in all domains, not only in language. In order to make sense of our world, we allocate items to categories; on the basis of these categories we construct ‘rules’ which predict how different items will behave.

Sometimes our predictions are wrong, for one or two main reasons:
The rule does not apply to this particular item, even though we have allocated the item to the appropriate category. We must learn an exception to the general rule.
The item belongs to a different category, which is covered by another rule. We must either reallocate the item to a different category which we know, or we must construct a new category and rule.
In either case, the initial error was due to overgeneralization of the rule which cause his wrong prediction.
For example: A learner of English has learnt a rule for forming plurals, that is, by adding –s. Then he overgeneralizes the rule when he says ‘I saw two handsome mans at the party yesterday.’


7.5.2 TRANSFER
Overgeneralization refers to learner’s previous knowledge of the second language that he used.
Transfer refers to learner’s use of his previous mother – tongue experience as a means of organizing the second language data.
It is economical and productive for second language learners to transfer their previous knowledge of language to the new task. It means that they do not have to discover everything from zero.
7.5.3 SIMPLIFICATION BY OMISSION

This form of simplification seems to less productive in its sense: ‘redundancy reduction’. It eliminates many items which are redundant to conveying the intended message.

For example: the omission of the verb inflexion and the article in Daddy want chair does not prevent the meaning from being understood. Provided that the situation supplies the missing element of meaning. Redundancy reduction makes production easier but may make comprehension different or even impossible.
7.5.4. NON – SYSTEM ERRORS
Learners make errors which result from more superficial influences. 2 of such influences may be:
Immediate communication strategies.
Performance factor.
7.5.4.1. COMMUNICATION STRATEGIES
Many apparent instance of overgeneralization or transfer may likewise be more the result of an immediate communication strategy than of an underlying system: in order to cope with a communication problem, the learner may have recourse to the mother – tongue system (transfer) or use second language items which he knows are not completely appropriate (overgeneralization).
7.5.4.2 PERFORMANCE ERRORS
When we speak our mother – tongue, we sometimes make errors of performance. The second language learners must make errors of this nature. They are the result of the learner’s performance.

Such performance errors are called ‘lapses’ or ‘mistakes’. With communication strategies, the best evidence that an error reflects the learner’s underlying system is when it appears regularly in his speech. In the case of a mistake, it is also more likely that the learner will be able to recognize the mistake himself and correct it afterwards. Even though it is unlikely to be successful to distinguish the difference between error and mistake.
7.5.5 ERRORS DUE TO THE EFFECTS OF TEACHING

Some errors of learners can be direct result of misunderstanding causes by faulty teaching materials. That is, the distinction between 2 forms may not be clearly explained, with the result that the learner confuses them. Or one form or pattern may be overemphasized or overpracticed so that the learner produced it in inappropriate contexts.

For example: many teachers or materials place special emphasis on the present continuous form in English. Their purpose is to counteract the fact that if the learners possess no equivalent form in their mother tongue, they may be inclined to use it less frequently than they should. However, this extra emphasis may have the undesired effect of making learners overuse the continuous form, at the expense of the simple present.

These errors are special instance of overgeneralization errors. They are particularly interesting in that a specific cause of the overgeneralization can be identified in the language ‘input’ which the learner has received.
7.5.6 THE LEARNER’S INTERNAL SYLLABUS
One of the most important conclusion drawn from error analysis is that the learner approaches the learning task with active strategies, notably generalization and transfer which help him to construct the rules which underlie the second language. This is ‘creative construction theory’.

It is possible for each learner to apply the strategies in different ways and proceed through different strategies in learning the language. The learning sequence may not be the same as the teaching sequence. This may not be evident from observing the controlled performance which learners produce in many classroom exercises.

The idea of the internal syllabus is supported by the fact that learners make similar kinds of errors, irrespective of what course of instruction they have followed or whether they have received formal instruction at all.
7.6. LEARNING PROCESS
There are 3 main processes:
Transfer of rule from the mother – tongue.
Generalization (and overgeneralization) of second language rules.
Redundancy reduction by omitting elements.

The first is ‘interlingual’ process
The second and the third are intralingual.

Transfer and overgeneralization are expressions of the same underlying strategy of using the previous knowledge to understand new experience. They are the main creative process which underlie second language learning.

Redundancy reduction can co-incide with transfer and overgeneralization.

All three processes can be seen as forms of simplification. Simplification through omission would appear to have a less directly creative role. However, it may perform an important function in ensuring that the learner can devote more of his learning capacity to the aspects of his developing language system.
In short, we have seen how the study of learner’s errors shows that second language learning is more than simple matter of habit-formation. The most important of these strategies seem to be generalization, transfer and other forms of simplification.

Some errors seem to be produced by performance factors or communication strategies rather than by the rules of the learner’s underlying system. Some errors seem to be induced by specific teaching techniques.