Timothy Mason's Site

Creative Commons License
This work is licensed under a Creative Commons License.

This lecture, like the others in this series, was given to students of English at the Université of Versailles St. Quentin, for a course in the Didactics of English, which I taught from 1993 to 2002.

It offers a simplified introductory account. The embedded links, most of which point to material off this site, are for readers who are looking for greater depth and complexity.

Didactics - 8 : Critique of Krashen IV

 

Interlanguage & Fossilization

A : Recap

Last week - we looked more closely at interlanguage. We saw that according to Selinker, and others, the learner builds up her own rules - and speaks a language which is neither the Target language - the L2 - nor the mother tongue. The learner, says Selinker, builds up the interlanguage through using a series of strategies that help her communicate and learn the language.

  • Amongst these strategies, one of the most important is the use that the learner makes of her mother-tongue. This use may lead to negative consequence - and these were stressed by older approaches to language learning - interference - we find traces of the L1 in the L2 that fall short of the target and hinder communication. We later saw that this may be particularly problematic at the level of vocabulary - the young woman whose work we looked at used the word 'particular' as if it were the English equivalent of 'particulier', and this made it difficult for an anglophone to understand.
  • It may also lead to positive consequences  - French and English share much vocabulary - there are many more Vrais Amis than Faux Amis - so our young woman was using a strategy that in fact usually pays. Moreover, from the 'Martian' point of view, human languages are likely to share many structural properties.

We noted that it was not always simple to determine when the learner might be using any one of these strategies. When we looked at the written work of our BTS student, we noted that on many occasions, it appeared that errors arose that could be either interference errors or developmental errors - and that often they might indeed be both at the same time. I finished by making some recommendations about error treatment that did not meet with your full approval.

Development

How does the learner proceed from one Interlanguage stage to the next? According to the theory, it is by using the different strategies that learners build up mental grammars of the L2. As these grammars are provisional heuristic constructions, the rules can be seen as hypotheses. At any one time, the interlanguage may include several competing hypotheses, so that the speaker's language is, in fact, variable, as he tries out first one and then another.

Where does the learner start? According to Pitt Corder, the learner begins not with his own L1, but with a highly simplified version of it, which is, as it were, a memory of one of the early stages of L1 learning. This 'stripped down' or basic system gives the learner his first hypotheses - some linguists claim that it may be universal - that is, that these are the rules that are at the basis of all languages. The learner then builds up from the stripped down form to greater complexity. (This may remind you of what was said about the relationship between Pidgin and Creole).

Use and Acquisition - Tarone & Ellis

How does this building-up proceed? How does the learner get from one form of interlanguage to the next? According to Tarone, we should recognise that the learner is not simply a language learning machine - that is, he does not simply absorb syntax, phonology and lexicon - he is an actor in the social world and is therefore concerned with the pragmatic aspect of language - how to do what , and when. He quickly becomes aware of register, and knows that he cannot speak in the same way to everyone, that he cannot use the same language in all situations. 

  • We all have access to different 'codes' or registers - ranging from formal to informal.
Tarone holds that new language can enter the learner's system in one of two ways :
  • - directly into the informal style, from where it may spread to more formal styles. This may result in language being acquired in the 'natural' order
  • - into the most formal style, and only used when the learner is paying close attention to speech - then spreads into more informal styles
Ellis takes a similar position, only he insists on the distinction between 'planned' and 'unplanned' language. In unplanned discourse, the speaker uses automatic and unanalysed knowledge. In planned discourse, the speaker uses analysed knowledge - monitoring is an example of this. Development takes two forms -
  • - learners make knowledge that was at first available only for planned discourse available for unplanned discourse.
  • - learners acquire new L2 by participating in different types of discourse - they create new rules for themselves through discourse
The learner builds up a set of rules, some of which are in competition. To some extent, different rules are used in different contexts - different levels of discourse, for example. But the rules also may be used interchangeably - that is, in free variation, as Ellis puts it. It is where there is free variation that learning takes place, for it is uneconomical to have two forms for the same purpose - the learner either :
  • a) replaces defective forms with more accurate forms. Thus at one stage in the acquisition of the negation, the learner may be producing both 'I no like it' and 'I don't like it'. Gradually she replaces the first with the second.
  • b) learns to use one form in one context, and the other in another - thus using different forms for different functions.

A learner may have two request structures - "Would you mind passing the salt, please?" and 'Hey, pass the salt!". She comes to realise that one of these belongs to a more formal register. Just as I discovered that one is not supposed to 'tu-toi' the CRS officer who is asking for your identity papers.

 

Ellis and Tarone's approaches both imply that Krashen is wrong in believing that we learn a language simply by listening and reading. It is only through active participation that new items enter the different registers, it is only through discourse that we learn that some items are only to be used in certain settings. Krashen, leaning on Chomsky, has a tendency to see language learning as being mainly a question of syntactic development. But the pragmatic and, as we shall see, affective aspects of language are also important.

Through processes such as these, the language learner passes from one Interlanguage to another; in each case, the interlanguage is a little closer to native speaker competence. Thus it might seem that, with time, sufficient exposure, and sufficient use, the learner would normally achieve full bilingualism.

C : Fossilisation

 

However, it is extremely rare for the learner of an L2 to achieve full native-like competence : Selinker coined the term 'Fossilization' to refer to this phenomenon - non-target forms become fixed in the interlanguage. Many examples can be found - Mukkatesh, looking at the written production of 80 students at a Jordanian university, found that after 11 years instruction in learning English, they continued making errors such as the use of simple past instead of simple present - no amount of grammatical explanation or of error correction had any effect. 
Fossilization may simply affect certain structures. Thus Selinker says that : 

Fossilizable linguistic phenomena are linguistic items, rules and subsystems which speakers of a particular NL will tend to keep in their IL relative to a particular TL, no matter what the age of the learner or amount of explanation and instruction he receives in the TL. (NL - Native Language; IL - Interlanguage; TL - Target Language)

As this implies, a student may continue to make progress in certain areas, and yet return again and again to the same error. Thus, for example, we find advanced students who communicate with great skill and who make very few errors, but still do not master the Pluperfect aspect of the verb in English.

Alberto and 'pidginization'.

However, fossilization may set in once and for all, and the student simply gets stuck at a plateau, never to go any further. One of the most well known examples of fossilisation is that of Alberto, a 33-year-old Costa Rican who had lived in Massachusetts for four months when his language progress first began to be investigated. Along with five other Spanish-speaking immigrants, (two five-year-old children, two adolescents and one other adult), his speech was monitored over a period of 10 months, by a variety of means, including free expression in natural settings to pencil and paper tests in the classroom. While the other five all made progress, Alberto quickly fossilized. Schumann believes that what happened with Alberto was that he went through a process similar to 'pidginization' - that is, he constructed a basic lingua franca for the limited social purposes that brought him into contact with English speakers. 
  • Thus, for negation, Alberto only used the two earliest stages
  • 'no' + V - I no understand good
  • 'don't' + V - don't know
using the first of these most often.
  • For interrogatives, Alberto inverted subject and auxiliary in only 5% of cases, reserving the correct form for only certain verbs - 'say' and 'like'. Occasionally he would produce full verb movement - 'What are doing these people?"
  • Although he achieved 85% accuracy for plural 's', he got the possessive 's' right in only 9% of obligatory contexts, regular past tense in 7% and irregular past in 65%


He was particularly far from native-speaker forms in his use of auxiliaries, and Schumann concluded that he could only be said to possess 'can' and certain copula forms of 'be'. The other five learners were well ahead of him on this.

Why was his language 'pidginised' in this way? Schumann rejects both age and cognitive level. Instead, he draws attention to the fact that Alberto's speech is very close to classic pidgins in a number of ways. Schumann believed that Alberto found himself in a situation very similar to that of a speaker of a pidgin. Pidgins are used between groups who are at some social distance from each other. For Schumann, this is a crucial variable in language learning. The relationship between the L1 group and the L2 group, may differ in a number of ways

  • 1. Dominance :
  • L2 users may dominate the L1 group - French-speaking colonists in Tunisia.
  • L2 users may be dominated by the L1 group - Hispanic immigrants to the USA.
  • L2 users may be on an equal footing - middle-class French speakers in England .
  • 2. Integration :
  • L2 users may decide to assimilate to the L1 group - most Bretons now simply regard themselves as French people.
  • L2 users may decide to maintain their own culture - many Asian groups in Britain continue to speak their own mother-tongue within the household, and to regard the Indian sub-continent a their real home
The choice of assimilation or non-assimilation is not simply determined by the members of the L2 group, but may be imposed upon them by the L1 group, who may refuse to accept their efforts at assimilation. Thus West-Indian people in the United Kingdom have increasingly come to look upon Caribbean Creoles as a mark of identity, after discovering that the white majority were not willing to allow them to integrate. There has indeed developed a British black Creole, distinct from both British English and true Caribbean Creoles.
  • 3. Enclosure : 
  • The L2 group may live separately from the L1 group - high enclosure - or may join in the social activities of the L1 group.
For example, among first generation Asian immigrants to the UK, there was a considerable degree of enclosure. It is interesting to note that this was of a higher degree for the women than for the men. As female Asians have come to enter the wider society, particularly through schooling, so there has developed a demand for separate facilities. There is some evidence that it is the men who wish to see the women segregated, rather than the women themselves.
Alberto found himself in a position where he belonged to a dominated group, with a low degree of assimilation, and a high degree of enclosure. His life-style was such that the had no need to develop his expressive powers in English, and so he declined to make the considerable effort that it would have taken to make further progress beyond his semi--pidginised state. 

Alberto's pidginization of the English language, then, came about because he felt no expressive needs could be met by the language. This is one other indication that the Chomskian approach to language is not sufficient. We remember how Bruner insisted upon the need for a LASS to complement the LAD - and how the baby's entourage provided a context which was not simply communicative, but also affective. I want to suggest that this affective aspect is also of great importance in the learning of the second language. Alberto had no love either for or through the English language. The same is true of many of our students, and may account for their relatively rapid fossilization.

We shall need to look more closely at this when we come to consider Krashen's Affective Filter Hypothesis.

C : Conclusion

To sum up - we have been looking at Krashen's natural order hypothesis, which holds that the grammar of a second language is learnt in a specific order, whatever the learner's L1. We have seen that there is some reason to believe that intralinguistic effects do occur, and that there may well be some kind of a predictable sequence to the learning of a specific L2. However, we have also noted that L1 does have some effects upon the acquisition of the L2 - these effects are both positive and negative - some errors do arise from interference. 
We have looked at the concept of interlanguage, which sees learners as constructing their own grammatical systems. These systems are learner-driven rather than teacher-driven - the learner progresses through employing a number of different strategies, some of which are based upon her L1, some of which are based upon her desire to communicate, and some of which may be rooted in the Universal Grammar. 

We have also seen that learners of a second language tend to fossilize. They may be partially fossilized, retaining certain errors while progressing in other ways, or they may, like Alberto, get stuck upon a plateau. For some of us, like Alberto, the plateau is fairly close to sea level : others attain higher levels. But most of us, it appears, get stuck sooner or later. 

We have seen that the sociological situation of the learner and of the learner's community in relation to the community that speaks the L2 can have a significant effect on language learning. There are also other factors which intervene in the process, and which may either induce fossilization or prevent it. Among the most important variables are 

  • - affective factors
  • - amount of exposure - input
  • - opportunities for expression
  • - negative feedback - (note - not correction, but signalling incomprehension)
  • - absence or presence of pressure on communication
We have seen that these factors put in doubt both Krashen's approach, and the over-insistence on communication. We shall return to these questions at a later date 
 
(If you wish to comment or ask a question, please write to tmason@timothyjpmason.com)
Home Didactics Top of Page
Author's CV | Site Map | Contact Author |