Didactics - 4 : Second language acquisition

 

A : Canadian experiments.

Immersion

Recently, Quebec has been much in the news - Many French-speaking Canadians - but not a majority - would like to set up an independent nation, and almost half of the people living in the province are now of this opinion. However, for many years, Quebec has survived within the union as a French-speaking province. The growth of the liberation movement from the 60s on forced the Canadian government to recognise a special linguistic status for the province - the French language was made official.

This put anglophones at a disadvantage. Many of them decided that it was in their interests to ensure that their children should be able to speak French fluently. As they considered that the language teaching in the national school system was insufficient, they put pressure on local schools to provide special facilities for their children. Thus it was that from 1965 on, a considerable number of young anglophone children (something like 10% of all English-speaking children) were enrolled in what were called immersion classes.These classes were based upon the idea that the best way to learn a foreign language was the 'natural way' - that is, in the same way that a child learns its mother tongue. This means that, instead of concentrating upon the grammar of the language, in specialised language classes, the young learner should be exposed to the foreign language in natural situations. Instead of learning French in French classes, they learnt mathematics, science, history and so on in French.

The results of these experiments are encouraging1

• - first of all, it is to be noted that the abilities of the child in the mother-tongue do not appear to suffer in any way - unless the written code for both languages is introduced simultaneously.
• As for the foreign language, the capacities of children who have followed this kind of course are far superior to the capacities of those who have followed traditional French classes. Their comprehension is, indeed, on the same level as that of young French-speaking children
• - on the other hand, they do not speak or write as fluently as young Francophones, and they tend to avoid initiating conversations in French.
• It has also been claimed that these classes have deeper cognitive effects on the children, developing their mental flexibility. However, we need to be fairly careful about such results - it is very difficult to know whom these children should be compared to.

Self-tuition

Even more intriguing are the results of an experiment in teaching English to young French Canadians in the bilingual province of New Brunswick, where about one third of the population is French-speaking. Because of the considerable expense, it was considered impossible to provide specialist teaching for all French Canadian children in the public schools. The provincial government decided to try out a method based entirely on input 2

• Every day, in the schools chosen for this experiment, children in grades 3 - 6 ( from 8 - 11 years old) spend 30 minutes studying English. The classroom in which this learning is done is equipped with personal tape-recorders. There are also shelves upon which are kept a large collection of books. Each book is accompanied by a tape.
  On entering the classroom, the child chooses a book, sits at a table with a tape-recorder, and then reads the book while listening to the tape. The student chooses books from a menu adapted to her age. She works entirely on her own - the teacher does not intervene in the learning in any way, other than to encourage a pupil to organise her time or materials.

After three years, the performance of the experimental groups was compared with that of groups who had been taught by well-qualified and conscientious language teachers. It appeared that on most measures there was very little difference between the two groups. However, on a test where the children were asked to describe a picture, the experimental children were far superior to the groups who had been taught by traditional methods - they used a larger vocabulary, and a more flexible syntax than did the others.

Furthermore, the pupils said that they enjoyed their English classes, and looked forward to the English period in their day. They also appeared to be more autonomous in their approach to learning. When asked what they did if there was something which they did not understand, the majority of the traditionally taught students said that they would ask the teacher - those in the experimental groups said that they would listen to the tape over again, or look at the glossary at the back of the book.

Although the programme had not been fully evaluated at this stage, it does appear that children can acquire a considerable degree of proficiency in a foreign language without benefiting from formal teaching, from grammar lessons, and from being forced to use the language. However, the results of a follow-up study suggested that there was a limit to what could be acquired without a teacher, and that best results were to be obtained when the initial program was enriched by the presence of an instructor.

Stephen Krashen's Monitor Theory 3

Now I wish to turn to a highly influential and controversial account of second language learning, which is based on the idea that second language learning is very similar to the learning of a first language. The account has been put forward in it fullest form by the American language teacher, Stephen Krashen.

Krashen sees five fundamental points - which he calls hypotheses - as the basis for his language teaching method. These are :

1. The Acquisition-Learning Hypothesis

   • Krashen makes a distinction between what he calls acquisition of a language - which is much the same as the process by which a child learns his first language - and learning, which is the procedure employed in most traditional classrooms. Acquisition is a relatively painless process. The child hears language all around him, and unconsciously works out the grammar. This he can do because he is equipped with an LAD. He does not set out to deliberately learn the language. Learning, on the other hand, is a conscious process, requiring effort specifically directed towards analysing the target language. This is what we do in grammar lessons, and Krashen appears to be thinking mainly of grammar when he refers to this. A standard case would be learning the inflections of the German verb, or the French Subjunctive.

   • Now, according to Krashen, one can only be said to master a language when it has been acquired. Formal learning may give us the rules of grammar, but it does not mean that we will use them correctly. He points to the fact that students may score well on formal grammar tests, but, when they are concentrating on content rather than form, make mistakes that they do not make in the tests.

2. The Natural Order Hypothesis

• The second point is that learners make mistakes, and that these mistakes are a necessary part of language learning. These mistakes are not random, but are very similar to the errors that children make when learning their first language. If we follow the mistakes that students make through time, we will see that they lie in a rough sequence.

• Moreover, the sequence of errors for acquired language is not the same as the sequence of learned grammar points - some grammatical morphemes which appear simple from the learning point of view, are in fact acquired late - the 's' of the TPS. This, according to Krashen, indicates that there is a natural order in which learners pick up a language - and that this order is roughly the same for all learners, no matter what their linguistic background.

• Chinese people learning English will make the same mistakes, and will learn in more or less the same order as French people.(As we will see, there is much evidence to suggest that Krashen is right on this point, although there are in fact differences between Chinese learners of English and French learners of English).

• The point of this observation is that these mistakes will be made in the same order whether the learners have been taught the grammar or not, and that teaching the grammar will not help them change the order (Krashen appears to believe that it will not make acquisition any quicker either, but recent research suggests that while grammar teaching does not appear to change the order, it can get the student through the different stages more rapidly).

• The Monitor Hypothesis

• Krashen does not think that formal grammar teaching is entirely pointless. The formal rule system feeds in to what he calls the Monitor - we may think of this as a minute grammar teacher that sits inside our brains and listens to what we say, or reads what we write and yells out whenever he hears a mistake.

• The Monitor is a dangerous ally - some people over-use it, and their speech becomes slow, and hesitant - their interlocutors are likely to give up on them, and go and talk to somebody else. This, as we shall see, is important, because a language learner needs to hear a lot of language from native speakers.

• The Monitor is best used when we have to be very careful - when language is necessarily formal. This is obviously the case when writing letters of application, for example, or when speaking to a hierarchical superior in a formal situation.

• Most of the time, however, Krashen suggests that we should leave the monitor unemployed, and concentrate upon the meaning that we wish to convey, rather than on the form of our utterances.

• The Input Hypothesis

• Although his overall theory is often referred to as the Monitor Hypothesis, it is this fourth point that is crucial to Krashen's whole argument. Starting from the observation that in what he calls Natural language learning conditions, people often go through a silent period, when they observe and listen - this appears to be true of the baby, for example, but is also true of adult learners in the Amazon basin - he believes that it is not language use which is the key variable in acquisition, but language input - what the learner hears and reads.

• The most useful form of input has to be understandable - this does not mean that it has to be one hundred per cent clear; in fact it should be just a little beyond the learner's present capacity. If it is too far beyond, the learner will not pay attention to the input, and if it is not far enough, the learner will learn nothing.

• Krashen considers that grammar translation methods, which often graded the material to the learner's present level, made the mistake of oversimplification. This is because :

  • a) they do not in fact know what the student needs next - acquisition order research has not yet given us full information on the basic order in which grammar is learnt.
  • b) in any one class, there will be individuals at different levels, who are needing different kinds of input.

• He suggests that, instead of fine-tuning the input, as traditional methods have been wont to do, the modern teacher should give rough-tuned input - and a wide variety of material, supported by visual cues and realia which give it a context within which the learner may guess at the content. Just as parents in Bruner's model of FLA make language comprehensible for the baby by surrounding it by ritual and regularity, so the language teacher must make input comprehensible by contextualizing it.

• The Affective Filter Hypothesis

• One barrier to learning is to be found in any negative feelings that the learner may have about the language, the method used, the institution or the teacher. These feelings may constitute a kind of filter, which keeps the input out. It is therefore part of the teacher's job to make language learning as free of stress and as enjoyable an experience as possible.

• Traditional language classrooms are often highly stressful places - pressure is put upon pupils to produce language even when they do not feel ready to do so, or when they feel they have no particular reason to say anything. There is the feeling that all language production will be graded and used as evidence of failure and so on. This is one of the reasons why Krashen insists upon input, rather than output.

The kind of model that Krashen puts forward here is often referred to as an input model. The idea is that as with the young Canadian children, learners will advance in language acquisition if they are exposed to large amounts of authentic language, language which is not specifically graded in terms of a set grammatical progression, but which is adapted to the students’ interests and reasons for learning the language. This input should be contextualized in such a way that the learner can understand a large amount of what is being said or written without constant need to consult dictionaries or ask to the teacher. It should be done in relaxing and friendly conditions

This is the basis of what Krashen refers to as the Natural Approach. Over the next few weeks, we shall take the five basic hypotheses and subject them to critical observation. This will lead us to assess Krashen's theory, and to cover a great deal of the material that researchers in the domain of Second Language Acquisition Theory have produced since the late 1960s.

1. This account is drawn from Charmian O'Neill, 'Les enfants et l'enseignement des langues étrangères', Crédif, Didier, 1993. Although Rod Ellis concludes his discussion of the programmes with the judgement that 'there is now general agreement that immersion programmes are very effective in promoting L2 development in an educational setting, there are some discordant voices - e.g., Hector Hammerly, 'French Immersion : Myths and Reality', (Destelig, Alberta, 1989), who argues that the French acquired by immersion students is more correctly characterised as a pidgin, and that the typical student fossilizes badly.

2. This experiment is described in Patsy Lightbown, 'Can they do it themselves? A comprehension-based ESL course for young children', in Courchêne et al. (eds), 'Comprehension-based Second Language Teaching', University of Ottowa Press, 1992. When the children were followed up in secondary school, it was found that at this later stage, they benefited from tuition.

3. For the following account, I have mainly relied on Stephen D. Krashen, 'Principles and Practice in Second Language Acquisition', Prentice-Hall, 1987, and Stephen D. Krashen, 'The Natural Approach : Language Acquisition in the Classroom', Prentice-Hall, 1988. Critical reviews will be found in Rod Ellis, 'The Study of Second Language Acquisition', OUP, 1994, in Vivian Cook, 'Linguistics and Second Language Acquisition', Macmillan, 1993, and in Gass and Selinker, 'Second Language Acquisition : An Introductory Course', Lawrence Erlbaum Associates, 1994.

(If you wish to comment or ask a question, please write to tmason@timothyjpmason.com)

Home

Didactics

Top of Page