III Lma Co Course 7

ISSUES OF APPLIED LINGUISTIC RESEARCH

SESSION SEVEN

THE MONITOR MODEL

MAIN POINTS

• In this session we discuss the work of Stephen Krashen (1977) who developed a model of second language acquisition called the Monitor Model.

• The main points in the session are:• 1. Krashen (1977): the Monitor Model• 2. Important points made by Krashen• 3. Evidence from empirical research:• 4. Later modifications to the model• 5. Some criticisms levelled at Krashen’s work

1. Krashen (1977): the Monitor Model

• The model he developed started as an attempt to account for observed differences in the performance of the same learners in different occasions:

• Krashen notes that a learner may in one context produce a structure correctly, yet in another get it wrong.

• This he calls variability = a phenomenon that, like errors, can potentially provide us with information about the processes of language acquisition

• Therefore, if we can find out what makes the learner get things right once and wrong a second time, this may give us an insight into what is happening in the learner’s head.

Variability

• Situations in which variability occurs:• -between oral and written language;• -between casual and careful speech;• The cause of variability relates to the degree of

importance the user gives to ‘correct performance’ = form focus.

• It is often the case that a learner will make fewer mistakes in writing than in speech, and in careful as opposed to casual speech;

The Monitor Model

• Krashen’s Monitor Model = based on a distinction between acquisition and learning:

The Monitor Model

• Acquisition = a ‘natural’ process, where there is no ‘conscious focus on linguistic forms’ = the process of ‘picking up’ a language, just as you do if you go and live in the L2 environment;

• Krashen states that the minimal condition for acquisition to occur is participation in natural communication situations;

• He uses the term ‘creative construction’ to describe acquisition = what is known about first language acquisition is applicable also to L2 acquisition;

The Monitor Model

• Learning = a conscious process, which usually takes place in the language classroom

• It is for Krashen particularly marked by two characteristics:

• feedback in the form of error correction

• rule isolation

Learning characteristics in the Monitor Model

• a.the presence of feedback in the form of error correction

• The teacher draws explicit attention to errors learners make, and corrects them.

• This is not really the case in first language acquisition: with some exceptions, parents do not offer their children much linguistic correction of mistakes.

Learning characteristics in the Monitor Model

• b. rule isolation. This is the classroom procedure whereby a lesson will isolate and focus down on one language point: it may be a grammatical item like ‘the present perfect’, or a pronunciation point, or a notion or a function.

• Again it may be remarked that parents do not usually focus in this way. A teacher may say “today I am going to focus on the passive with my students”; a mother does not do the same thing with her young child.

The Monitor Model

• Krashen’s Monitor Model claims that the production of language by the L2 learner is controlled by the acquisition process, which is more important than the learning process.

2. Important points made by Krashen

• -the role of learning is simply to monitor = to inspect and sometimes to alter output generated by the acquired system;

• -there will be more monitoring at some times than other, e.g.:

• when one writes an essay there is likely to be more monitoring occurring than when one speaks casually to a friend (hence, of course, the variability we referred to earlier);

• -the concept of monitoring is taken from L1 studies = the work of Labov (1970)

Labov (1970)

• Labov suggests that the L1 early acquired local dialects get modified to ‘standard’ prestige forms.

• e.g.: in Britain (and in Romania also) there are many people who will have a local dialect that they used as children, and will still use as adults in informal situations.

• But there will be many occasions when they modify that dialect towards a more standard form.

• e.g: in a job interview or other occasion when they feel it is important ‘to put on a good show’, they will carefully monitor their speech to eliminate features they feel may not be acceptable.

• Incidentally, in some situations (caused by drunkenness, stress, tiredness) monitoring may stop!

Illustration of the Monitor Model

• This is Krashen’s own illustration of his model:•  

Learning (the Monitor)

•  • Acquisition Output• (A ‘creative construction’ process)

Illustration of the Monitor Model

• The horisontal arrow between acquisition and output shows the central importance of acquisition

• the position of the learning arrow close to output indicates that it is often ‘a last minute process’.

Illustration of the Monitor Model

• Indeed, monitoring may also occur after performance, resulting in user self-correction;

• you should also note that on many occasions monitoring simply does not happen at all.

• In addition, observe the use of Dully & Burt’s ‘creative construction’ term to describe acquisition.

Krashen evidence for the Monitor Model

• Krashen uses the analysis of learner errors to find the evidence for the two types of process:

• learning

• acquisition

Errors and the Monitor

• Krshen:

• when a learner’s output is unaffected by monitoring, the errors will be of a different nature from when the monitor has been at work.

Errors and the Monitor

• errors in “acquired output” (the Monitor is not at work) will be of a universal nature and largely of the sort Dulay & Burt call developmental;

• errors that occur in situations where monitoring takes place will be more idiosyncratic [ personal to the learner and often rather strange and unusual]

• such errors will reflect what point the learner has reached in her/his classroom lessons

• e.g.: if a learner has only just been introduced to the present perfect tense, it may be that she/he will perform worse at this than the learner who studied it some time ago;

 3. Evidence from empirical research

• Larsen-Freeman (1975) administered tests involving ‘natural communication’, where it was assumed no monitoring would occur.

• She also gave the learners problem-solving tasks involving reading and writing, where monitoring may have occurred.

• She did indeed find the differences mentioned above.

Evidence from empirical research

• Krashen et al. (1976) provide evidence of a different sort.

• They gave a ‘natural communication’ test (where there was no monitoring) to two groups of learners, some of whom had received tuition and some of whom had not. (experimental and control group)

• They found no difference in performance, suggesting that in that situation no learned output occurs.

Evidence from empirical research

• Krashen et al. (1976)

• They then gave a test in which monitoring could have occurred, and found that certain errors that were found before were now eradicated.

• One such error was the “-s” of 3rd singular simple present tense.

Evidence from empirical research

• Krashen et al. (1976)

• Also found that:

• individual differences occur in monitoring practice, i.e.:

• some learners do it more than others:

• high and low monitor users

Evidence from empirical research

• Cohen and Robbins (1976) report:• a. on a girl called Eva.• Here are her own words, which suggest that she is

a high user: ‘sometimes I would write something the way I speak. We say a word more or less in a careless way. But if I take my time, sometimes go over it, that would be much easier...whenever I go over something or take my time, then the rule comes to my mind’;

Evidence from empirical research

• Cohen and Robbins (1976) report:• b. on a learner named Hung.• His words: ‘I never taught any grammar. I guess I

just never learned the rules that well. I know that every time I speak it’s pretty correct, so I never think about grammars. I just write down whatever I feel like it. Every time I write something I just stop thinking. I don’t know which rule to apply’

• suggest he is low user.

Krashen’s Model: aptitude and attitude

• -Krashen also has things to say about the way that aptitude and attitude relate to his model, particularly to the acquisition/learning distinction:

• -he relates aptitude to learning, and attitude to acquisition,

Krashen’s Model: aptitude and attitude

• IF aptitude is related to learning, and attitude to acquisition, then:

• -in order to make a prediction about how well a learner would succeed in the classroom situation (learning) we might try to find out by means of an aptitude test how much ‘talent’ for language learning she/he had;

• -in order to make predictions about how well a learner would ‘pick the language up in an acquisition context, then we would ask about her/his attitude towards the target culture and its people, towards the language learning process in general.

Krashen’s Model: aptitude and attitude

• -in support of the associations above, Krashen cites:• -Pimsleur (1966) = found positive correlations

between scores on an aptitude test, and on language tests in which monitoring is possible;

• -Gardner & Lambert (1972) = found correlations between attitude and performance ‘especially in the oral/aural features of proficiency’. [oral = spoken rather that written; aural = related to a person’s ability to understand a language when it is spoken]

4. Later modifications to the model

• -three general areas:• 1. the value of error correction.• Much work concerned with L1 acquisition suggests:• there is little value to error correction because:• adults may correct a child’s language a very great

deal, but it seems to make little difference unless the child is itself “ready” to make changes in relation to a particular language area

• The child will simply ignore the corrections.

4. Later modifications to the model

• 1. the value of error correction.• Perhaps the same is true of L2 acquisition (as

opposed to learning).• Where L2 learning (as opposed to acquisition) is

concerned, the picture is less clear.• Krashen & Seliger (1975) claimed that error

correction works for the learned system, but Fanselow (1977) suggests that it doesn’t.

4. Later modifications to the model

• 2. conditions which must be present if monitoring is to take place. Three are identified:

• i. TIME = the learner must have sufficient time to be able to monitor.

• In rapid conversational exchange this is unlikely to be the case.

• In writing on the other hand, there is often sufficient time available for the learner to consult her/his ‘internal grammar book’;

4. Later modifications to the model

• 2. conditions which must be present if monitoring is to take place:

• ii. the situation should encourage FOCUS ON FORM.

• If the learner is doing a language test, for example, she/he is likely to undertake it with the monitor at the ready, to show off her/his language at its best.

• In a chat with a friend in a bar, on the other hand, the monitor is likely to be sent to sleep!

4. Later modifications to the model

• 2. conditions which must be present if monitoring is to take place:

• iii. that the learner should KNOW the relevant RULE.• It is simple syntactic rules that are most susceptible for

monitoring:• Krashen’s own example is one of the French rules

concerning formation of ‘de’ with a definite article. It may be that in a conversation the learner produces the incorrect ‘de + le’. This is just the sort of mistake the monitor is likely to pick up, correcting to ‘du’.

4. Later modifications to the model

• iii. that the learner should KNOW the relevant RULE.

• The English ‘he go’ for ‘he goes’ might be another case.

• But complex rules (e.g. the use of the present perfect versus the simple past in English) are unlikely to benefit from any opportunity to monitor.

4. Later modifications to the model

• -it is important to observe that the three conditions: TIME , FOCUS ON FORM and KNOW the RULE

• are necessary, not sufficient ones. That is, even in situations where all three conditions apply, it does not automatically follow that monitoring will take place.

4. Later modifications to the model

• 3. the increasing de-emphasis of the role of learning and monitoring.

• Over time, Krashen’s views move further towards the statement he makes in 1983, that ‘acquisition is where the action is’.

5. Some criticisms levelled at Krashen’s work

• Krashen’s views have come in for a great deal of criticism, and “Krashen bashing” [deliberately hit somebody hard] has become a favoured occupation of some applied linguists, e.g.:

• Gregg (1984) and McLaughlin (1978) launch the most comprehensive attacks.

5. Some criticisms levelled at Krashen’s work

• Some criticisms concern his methodology:• he tends to ignore counter-evidence• he does what the philosopher of science, Popper,

says you should not do: he seeks confirmation not falsification.

• (Popper says that if you wish to prove that ‘all swans are white’ you should not go round the world seeking evidence of how many white swans there are; you should go and look for a black swan)

5. Some criticisms levelled at Krashen’s work

• Krshen’s critics believe that in learning and acquisition there are, not two distinct processes, but just two sets of environmental conditions:

• learning = no more than ‘what happens in the classroom’

• acquisition = ‘what happens in the normal target language environment, outside class’.

Some criticisms levelled at Krashen’s work

• HOWEVER, some other applied linguists praise Krashen:

• by pointing out that whatever the faults of his theory, he is at least bold enough to have a theory!

• He is prepared to look at the varied evidence in the field of SLA, and to forge together from it one coherent view of the world.