That is, I have not come across a question such as: 'do you wish that you were able to make use of the learners' Ll? The findings suggest that teachers in the secondary sector use more L1 with 'less able' learners because these learners find it more difficult to infer meaning and therefore get more easily frustrated.
In other words, recourse to L1 is almost entirely a comprehension issue not an acquisition issue. In the research context I have been involved in, the nationality of the teacher is not a significant variable when it comes to teachers' beliefs about the value of codeswitching. Lastly, many teachers report feeling guilty when they resort to the L1. This is not a healthy outcome of a pedagogical debate. Teachers, across learning contexts, report Macaro, a that the areas in which they use the L1 are: 1.
Systematic observation data in the Macaro study of lower- secondary classrooms in England confirmed some of these teacher self- reports. The L1 was particularly noted when the teacher was giving complex procedural instructions. Additionally, in this context, the L1 was used when giving feedback to students. By this is meant feedback on progress in substantial form, for example on a task just accomplished, rather than quick feedback in the standard I-R-F questioning sequence. What is interesting about the findings above is that the L1 is used predominantly for message-oriented functions.
It is interesting because one would assume that it would be used mainly for medium-oriented functions where LllL2 comparisons might be being made. I would suggest that the reason for this is that the amount of input modification needed for certain message-oriented utterances to be successful is beyond the scope of most teachers in the time allocated to them and with the fear of the students 'switching o f f. Codeswitching therefore becomes a useful communication strategy. The qualitative data gathered in the study suggested that the motive for low levels of codeswitching was largely dictated by the pressure of the National Curriculum in England see above.
In addition, as L1 speech can be delivered comparatively quickly, the discourse space allocated to it was quite small. In other words the teachers could communicate quite a lot in L1 in a very short time thus still allocating plenty of discourse space to the L2. The research evidence on this is scant and unfocused. Again, none of these studies asked students if they would prefer a monolingual teacher who would be completely unable to refer to their L1. More recent studies have focused on the possible negative reactions of particularly adolescent learners faced with virtually all L2 teacher input.
In the school context, the evidence we have suggests that learners may divide into two camps according to individual preferences Macaro, Some learners get frustrated when they can't understand the teacher's L2 input and want to know the exact meaning of words and phrases. This is usually because of the consequences of not understanding-for example not being able to achieve a homework task. Others feel comfortable with the teacher's pedagogy or, at least, go along with it.
For this group of students although it would be easier for the teacher to codeswitch they feel that in the long run they will learn more if helshe does not. There is no evidence pointing in the direction of higher achieving learners or faster learners feeling more at ease with L2 exclusivity. It seems to be more to do with individual preferences. Some like the teacher to make immediate and explicit L1L2 connections others do not feel this is necessary. It is possible that the unequal power relationships between teacher and pupils found in most classrooms is accentuated in those L2 classrooms where the bilingual teacher excludes the L1 almost completely and this negatively affects males more than females.
The Jones et al. I have to ask op cit. They do not try to pose the question: to what extent does the L1 actually help you learn? Perhaps because it is assumed that it does not. We will return to the direct impact of the L1 on L2 learning later. For the moment let us look at the lunds of discourse environments that codeswitching sets up. This discussion is based on a pre-supposition that students tallung in the L2 is a good thing and relates to the tradition of research literature on interaction started by Long 1.
Other than my own studies mentioned above, I have not come across any quantitative research on the effects of codeswitching by the teacher on the general interaction. Chapter 5 In other words there was no correlation between teacher use of L1 and learner use of L1. Conversely, no significant increase in the students' use of L2 was detected if the teacher used the L2 exclusively or almost exclusively.
In the above studies, it would appear from a quantitative and qualitative analysis of the data, that codeswitching by the teacher has no negative impact on the quantity of students' L2 production and that 'expert codeswitching' may actually increase it and improve it. Nevertheless, there may be a lund of threshold reached by teacher use of L1 where the codeswitching resembles less a communication strategy than simply a discourse carried out entirely in L1 with only a marginal reference to the L2.
It may be fruitful to pursue the notion that it is the transaction which determines the intent of the bilingual teacher in hislher discourse. Thus if the transactional intent is to communicate via the L2, the codeswitch will not push the amount of L1 use above the threshold level. The function of the codeswitch is a strategic repair. I am now going to propose the hypothesis that if there is no codeswitching by the teacher, and very little allowed by the learners, then a number of discourse control features can come into play which are entirely manipulated by the teacher and, possibly, are detrimental to the learner.
Firstly, avoidance of codeswitching leads to greater use of input modification. In fact, in naturalistic settings, avoidance of input modification is a prime reason for the use of codeswitching. Speakers want to communicate without the time taken to modify the input. Below are listed some characteristics of teacher input modification and their effect on classroom interaction. I should point out that I am not suggesting that these characteristics have no beneficial effects, merely that they can have some negative effects such as increasing teacher talk-student talk ratios.
One of the potential underlying effects of input modification is that it can reduce interaction. Input modification comes out of the comprehensible input camp where it is claimed that the students will have no trouble understanding lessons conducted entirely in the L2 provided they were competently so conducted see quote above and that students need not be asked to speak until they are ready. What input modification actually does is to provide students with a listening text and exploitation exercises all in one-in effect a listening comprehension task.
One of its functions is to reduce the need for clarification requests. This is why there was so much literature on interaction modiJication following the post-Krashen period. Each of these characteristics can have a positive effect on learning, such as inferencing slulls. Whereas in naturalistic discourse between two bilinguals it is often redundant because of the codeswitching alternative , in the languages classroom it has a clear teaching function. However, it often results in the teacher hogging the discourse space and, used unsparingly, leads to 'the dumbing down' of the classroom discourse.
Input modification is not the only aspect of non-codeswitching which has a negative effect on the interaction. I am again particularly drawing from adolescent classrooms here. In order to avoid a codeswitch teachers use mime to put across the meaning of a linguistic item. In a sense this is a kind of codeswitch-switching from a verbal code to signed code.
Although this may help with communication, the focus is taken away from the language in the interaction. In order to avoid a codeswitch the teacher demonstrates to the pupils what they have to do. For example they may position learners in a particular way in order to show them how not to see each other's information in an information-gap activity. Again, this is a substitute for language which helps with communication but does not help with acquiring new language. I want to argue that codeswitching is used by begnner and intermediate learners but also to some extent by advanced learners in order to lighten the cognitive load problems in working memory.
This strategy of lightening the cognitive load can be activated in a number of language learning tasks. Kern , for example, attempted to elicit the language of thought during a reading comprehension task and concluded that learners were using their L1 as the language of thought, to their advantage, in order to: reduce working memory constraints; avoid losing track of the meaning of the text; consolidate meaning in long term memory; convert the input into more familiar terms thereby reducing anxiety ; clarify the syntactic roles of certain lexical items.
Thus the L1 was being used by the students to lighten the cognitive load as they were trylng to process the text. If we can consider classroom discourse as text to be decoded and understood, we can perceive how the teacher's codeswitching can help counter the cognitive constraints imposed by working memory limitations.
Teachers, deprived of codeswitching from their tool I t , cannot act as a bilingual dictionary for the learners. In fact, the diktat that there should be no recourse to L1 by the teacher makes bilingual teachers behave like monolingual dictionaries. Most learners report finding bilingual dictionaries Bishop, a useful tool in carryng out reading and writing tasks.
It is true that some learners use dictionaries too much particularly in exams thus pre-empting the possible deployment of other important strategies, for example top down processing in reading comprehension. Nevertheless, taking away the bilingual teacher's right to codeswitch is like talung away the student's right to use a bilingual dictionary. The use of a bilingual dictionary in a reading comprehension task is a way of lightening the cognitive load by, for example, reducing the number of unknown elements the reader is aslung working memory to keep activated at any one time.
In fact it could be argued that judicious use by the teacher of codeswitching is a way of modelling, in an implicit way, judicious dictionary use whilst at the same time lightening the cognitive load brought about by constant high speed inferencing in the spoken medium. As well as modelling a cognitive strategy, the bilingual teacher therefore offers the learner a metacognitive learning strategy: evaluation of when it is appropriate to use a dictionary and when it is appropriate to make the cognitive effort to infer from context.
Teachers, deprived of codeswitching from their tool lut are unable to offer learners translation as a learning task. Monolingual teachers, of course, cannot offer this kmd of task. Banning translation from the L2 classroom deprives learners of the possibility of developing a valuable language skill that they are very likely to need in the outside world, particularly the world of work.
Moreover, and following on from the previous paragraph, banning translation deprives teachers of the possibility of modelling reading strategies see Macaro, a for a fuller explanation and training the students to use them. Not only does it model appropriate dictionary use as a strategy, but it also draws the learners' attention to how to look for contextual clues, syntactic clues, text design clues, how to make decisions of which bits of language to skip and when to return to them once more in-text evidence has been gained.
In effect the teacher can use oral translation of a text with the class as a kind of group think-aloud protocol where the strategies the students use to understand the text are elicited, shared, developed and evaluated. Teachers who never offer this type of group translation have to resort to L2 only comprehension tasks: truelfalse; multiple choice; find synonyms in the text.
They tell the teacher what the learner knows or can do. By and large they fail to inform the teacher of what strategy-combinations, appropriate or inappropriate, the learner is using during the process of comprehension. Paradoxically, the avoidance of providing LlIL2 equivalents leads to an over-reliance on searching for cognates in text Macaro, b to the detriment of other, equally important strategies. The strategies the learners deploy have the potential to make their learning faster, more personalised and more effective Oxford, Banning codeswitching reduces this potential.
Teachers, deprived of codeswitching from their tool kit find it hard to offer students pre-listening activities which trigger appropriate combinations of listening strategies. Pre-listening activities set the context for the text which helps the learner to activate hisher schemata Chung, Pre- listening activities elicit key words which activate schema connections in the learners' brains and also apply a filter to all the possible information they might be listening out for.
If the language the students are going to hear in the listening activity contains considerable amounts of new language, it may be extremely difficult for the teacher to provide the contextual clues in the L2 as these will be 'as new' as the language the students are going to be listening to in the actual aural text. Pre-listening activities are also there to lower feelings of anxiety about a listening task. Learners are likely to react more positively to the imminent text if they are reassured, at least partly, in their L1. If codeswitching is allowed to develop over a period of time in the classroom as a teaching tool, there should be no reason why students should not be able to codeswitch receptively immediately-that is prepare their minds for the in-coming L2 even though the teacher has used some L1 in the pre-listening.
Teachers who avoid codeswitching tend to shy away from the kmd of task-based learning promoted by certain authors Prahbu, ; Di Pietro, ; Skehan, Both in the study of experienced teachers Macaro, and in the study of novice teachers Macaro, b there was strong evidence that practitioners were avoiding task based activities because of the difficulty of setting them up entirely in the L2. This contributes to a tendency for all classroom tasks to be highly standardised therefore easily recognisable without complex L2 , repetitive, of low cognitive challenge and often of the behaviourist presentation, practice, production type PPP.
PPP allows teachers to remain in the L2 by takmg the learners in lock-step fashion from exposure to new language elements, to the use of them in authentic tasks.
I am not arguing for the scrapping of PPP altogether as a learning sequence. However, it is not the only progression leading to use in authentic tasks. Groups of students can be given tasks to accomplish even without teacher input and practice. In order to set up this lund of collaborative activity, codeswitching by the teacher will promote the generation of higher order content. In turn, learners may have to codeswitch in order to use the teacher as a resource and to manage their task.
Again, I want to point out that the effects of not codeswitching are not detrimental to the learning process in themselves. The argument I am slowly constructing is that it narrows down the total range of classroom activities possible thus reducing learner strategy development in terms of range, combination and self-evaluation of strategy use. Rather than putting in place bad classroom practices I actually doubt that there are more than a handful anyway!
Teachers deprived of the codeswitching tool will find it harder to trigger a range of strategies in their learners when aslung them to carry out a writing task. Thinlung in L1 produces more elaborate content and greater risk-taking than thinking in L2 although the latter produces greater accuracy. This would seem obvious as the L1 language store is much greater. If the teacher elicits ideas only in L2 say in a brainstorming activity , and especially if the teacher encourages the learners to avoid mental translation, they are going to tap into a much more limited pool of language than if they codeswitched and allowed the learners to codeswitch for example by aslung 'how might you say X in L2'?
Lally obtained slightly different results in her study using a similar technique. Although there were no significant differences in vocabulary use, the students who prepared the task in L1 scored higher on 'organisation' and on 'global impression'. The trick for the teacher is to encourage the learners to make evaluative strateges such as: 'when am I likely to be better off sticking with language I know already e.
Balanced against this I must try to address the task as fully and as creatively as I can'. Chapter 5 We have been considering how learners can be deprived of certain learning strategies by bilingual teachers not codeswitching. Let us consider codeswitching and strategy development just once more but in greater depth, this time in an expository task.
Let us take a hypothetical example.
By Andrew Simpson
The teacher in a class of Italian learners of English wants to put across the following information formulated by a potential sentence such as: Emesto Macaro was raised in the gutter but he managed to avoid a life of crime Option 1: the teacher 'guesses' that the students do not know the meaning of 'raised in the gutter' andpre-empts with a codeswitch: Emesto Macaro k stato cresciuto nel fungo1 but he managed to avoid a life of crime In a naturalistic codeswitching situation this could happen and it would facilitate communication and might, indeed, be used to convey additional cultural meaning!
Clearly, in a formal classroom situation, this option deprives the learners of the potential to learn 'raised in the gutter'. It also deprives the learners of the beneficial 'effort' of trylng to infer from context.
- HOW TO GET A JOB INTERVIEW: Invaluable tips and advice from a recruiters perspective;
- Word Play : A New Approach for Understanding.
- Archives (see also INDEX).
Option 2 is to 'guess' that the learners will not understand 'raised in the gutter' and to pre-empt via a paraphrase such as 'was brought up badly by parents who were poor'. This too deprives the learners of learning the phrase 'raised in the gutter'. Option 3: The teacher does not pre-empt either with a codeswitch or with a paraphrase. Instead, s h e uses the original phrase and then looks around for signs of incomprehension. The phrase 'raised in the gutter', if met with looks of incomprehension from the students, now gives the teacher the choice of repeating the sentence either with a codeswitch or with a paraphrase-the decision between a communication strategy or a potential learning strategy.
Option 3a, then, the teacher decides to repeat with a paraphrase: Ernesto Macaro was 'brought up badly by parents who were poor'. Option 3b, the teacher decides to repeat with a codeswitch perhaps with some sort of paralinguistic feature such as a smirk: Emesto Macaro k stato cresciuto nel fango but he managed to avoid a life of crime In favour of Option 3a are the inferencing strategies that the hearer will try to deploy in order to arrive at the exact meaning of 'raised in the gutter'.
In order to make 'sensible' connections the learner will have to have plenty of undisturbed processing time andlor have the opportunity to ask for clarification through interaction -something that will not be possible if the teacher input has moved on. Option 3b requires little inferencing but learners will have a fairly heavy cognitive demand in having to trace back to the auditory loop first activated when the target phrase 'raised in the gutter' was first uttered.
Option 4 is repeating via a codeswitch plus the problematic phrase. Here the learners will be activating LlIL2 connections based on the cultural associations that the Italian students have built up over a period of many years: Emesto Macaro was raised in the gutter, e stato cresciuto nel fungo, but he managed to avoid a life of crime These cultural connections we know to be very strongly imprinted in our brains as part of our schemata. Our access to this schemata is going to be much more rapid using an L1 stimulus because that is the way they have been stored and activated in the past.
In addition, using L1L2 direct connections will result in minimum processing load on worlung memory thus allowing future storage connection strategies to be activated for the phrase 'raised in the gutter'. The downside is that the inferencing strategies are not being developed. The teacher does have a fifth option, of course, and that is to opt for repetition, paraphrase and codeswitch.
Nobody will question whether Hebrew is dying out in Israel because of the long-standing tradition of Israeli universities to teach in English, nor has the focus on English publications in my own native country of Germany threatened the native language competencies at universities. This went through my head when I was speaking to the parliamentary commission in The Hague and, as the speaker before me held the view that you could only gain a top job in the Netherlands if you speak fluent Dutch, I could not resist briefly teasing him. Nevertheless I am glad to be the president of a Dutch university.
Martin Paul is president of Maastricht University. Get a month's unlimited access to THE content online. Just register and complete your career summary. Registration is free and only takes a moment. Once registered you can read a total of 3 articles each month, plus:. Already registered or a current subscriber? Sign in now. David Lloyd, vice-chancellor of the University of South Australia, on an erratic period of change for universities. International cooperation could be most successful in the postgraduate sector, says Arshin Adib-Moghaddam.
Nearly one in five respondents to four-campus study said they feared role change if they underperformed. Swiss university has signed San Francisco Declaration on Research Assessment, which says academics should not be judged on the journals they publish in. Universities in most nations are now obliged to prioritise graduate career prospects, but how it should be approached depends on your view of the meaning of education.
Academics need to think that through much more clearly, says Tom Cutterham. Promotion criteria requiring top researchers to also be good teachers and managers undermine the nature of universities, says Andrew Oswald. Leeds accused of attempting to extend control of central management by ending practice of selecting heads of school from within the faculty. The Augar review should prompt the creation of a network of secular synagogues devoted to adult peer learning, says Tom Schuller. Sir Ian Diamond reportedly did not trigger his month notice period until his departure from the university.
Skip to main content. World insight: The use of English in universities will not kill off Dutch. February 3, By Martin Paul. Share on twitter Share on facebook Share on linkedin Share on whatsapp Share on mail. Could this also be achieved in Dutch? Read more about. For example, one of the teachers in the bilingually-based school was in fact a native speaker of English, although she was also completely fluent in Norwegian and taught all other subjects in this language.
It is of course conceivable that this teacher's nativeness in itself is what led to increased acquisition in the bilingually-based group. However, this would mean assuming that native speakers are always better teachers of L2s or that language can only successfully be acquired from native-speaker input, which goes against research findings both on L1 development e. Moussu and Llurda, The main benefits of the teacher's nativeness, i. It is precisely the conclusion of this paper that teachers should be trained in this. However, as already mentioned, pupils from this school have been previously found to perform above average in national tests in English.
Whereas results in these tests may have come from classes taught by teachers other than those in the present study, it is highly unlikely that the school standards of English instruction have dramatically dropped. Furthermore, as with the bilingually-based group, the native language-based teachers knew that their pupils would be tested after 8 months, and were naturally eager for them to do well. If anything, it is likely that they spent more time on English than they would have in a normal year. Finally, and significantly, there is nothing in what the native language-based teachers report that deviates from the stated norms of the curriculum.
As with many early-start foreign language programs, nothing in the plans for early English teaching in Norway focuses on extensive input for vocabulary acquisition. The overall conclusion of the present study is that there is nothing inherent in the classroom situation which prevents successful L2 acquisition in young learners, and that vocabulary can be acquired at a fast rate in an early-start foreign language program. Furthermore, the study indicates that although such acquisition critically depends on input, exposure to the target language need not be unrealistically massive for acquisition to take place.
However, we do know that the two are related, and that receptive vocabulary is important for comprehension, which in turn means that a larger receptive vocabulary allows more advanced input to be processed and understood. In this sense, receptive vocabulary can be assumed to be a predictor for further language acquisition.
A natural next step is to further examine whether such an increase in exposure to the target language has a long-term effect beyond the first year of school, and whether it is also evident in areas other than vocabulary comprehension. Furthermore, more research is needed concerning exactly what kind of input is necessary, including what proficiency level teachers must have attained and whether native input from sources other than the teacher, especially media i. Anne Dahl has had main responsibility for the project, including research design, data collection, analysis and interpretation, and drafting and revising the paper.
Mila D. Vulchanova has contributed substantially to the conception and design of the research, and to critical revision of the paper for important intellectual content. Both authors have final approvement of the version to be published and agree to be accountable for all aspects of the work. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We acknowledge the support of the Faculty of Humanities, Norwegian University of Science and Technology, from whom the first author received a grant to conduct this research. National Center for Biotechnology Information , U. Journal List Front Psychol v. Front Psychol. Published online Apr Author information Article notes Copyright and License information Disclaimer. Reviewed by: Stanka A. This article was submitted to Language Sciences, a section of the journal Frontiers in Psychology. Received Dec 19; Accepted Mar The use, distribution or reproduction in other forums is permitted, provided the original author s or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice.
No use, distribution or reproduction is permitted which does not comply with these terms. This article has been cited by other articles in PMC. Abstract This study investigated whether it is possible to provide naturalistic second language acquisition SLA of vocabulary for young learners in a classroom situation without resorting to a classical immersion approach. Keywords: second language acquisition, naturalistic acquisition, input, early-start, classroom.
Introduction Over the past decades, there has been a trend in many countries of lowering starting ages for learning foreign languages, especially English. Method The present study investigates whether employing a bilingual approach to an otherwise normal Norwegian first-grade English classroom will lead to improved acquisition over 1 year, compared to a standard, i. Conditions Two different schools were recruited for the experiment. Table 1 Mean, minimum and maximum values and standard deviations SD for weeks spent outside of Scandinavia and hours of exposure from media, games and music prior to starting school in the bilingually-based and the native language-based groups.
Open in a separate window. Participants All parents of students in the relevant first grades were contacted in writing and asked for written consent for their child to participate. Table 2 Mean, minimum and maximum values and standard deviations SD for age, vocabulary, verbal and non-verbal intelligence and memory scores raw in the bilingually-based and the native language-based groups. Table 3 Mann-Whitney U , Z, and p for between-groups comparison of vocabulary, verbal and non-verbal intelligence and memory in the bilingually-based and the native language-based groups.
Figure 1. Group development For the repeated-measures test, the Wilcoxon signed ranks test was used. Table 4 Age equivalents of pre- and post-test vocabulary scores raw in the bilingually-based and the native language-based groups. The nature of group vocabulary differences It is worth looking at group differences for cognates and non-cognates separately, since there may be differences in how the two categories are acquired. Table 5 Percentages of correct answers and Mann-Whitney U, Z, and p for between-groups comparisons of number of correct answers for cognate and non-cognate words in the bilingually-based and the native language-based groups.
Cat 0. Discussion We see from the above results that English teaching in the native language-based group has had no significant impact on English receptive vocabulary. Conclusion and suggestions for further research The overall conclusion of the present study is that there is nothing inherent in the classroom situation which prevents successful L2 acquisition in young learners, and that vocabulary can be acquired at a fast rate in an early-start foreign language program. Author contributions Anne Dahl has had main responsibility for the project, including research design, data collection, analysis and interpretation, and drafting and revising the paper.
Conflict of interest statement The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Acknowledgments We acknowledge the support of the Faculty of Humanities, Norwegian University of Science and Technology, from whom the first author received a grant to conduct this research.
Code choice in the Chinese as a foreign language classroom
References Abello-Contesse C. Young children acquiring second language vocabulary in preschool group-time: does amount, diversity, and discourse complexity of teacher talk matter? Developing linguistic knowledge and language use across adolescence , in Blackwell Handbook of Language Development , eds Hoff E. What is the logical problem of foreign language learning? How Children Learn the Meanings of Words. Myths of word learning , in Weaving a Lexicon , eds Hall D.
Junior Scoop French in the primary school: the British experiment. Two-year-olds learn novel nouns, verbs, and conventional actions from massed or distributed exposures. The Lexicon in Acquisition. Cambridge: Cambridge University Press; The structure of language abilities at 4 years: a twin study. Two by two: a twin study of second-language acquisition.
The robustness of critical period effects in second language acquisition. Second Lang. What does the critical period really mean? Oxford: Oxford University Press;. Peabody Picture Vocabulary Test. Peabody Picture Vocabulary Test Manual. More evidence on competing cognitive systems. Lexical and conceptual representations in more- and less-skilled bilinguals: the role of cognates.
Foreign Lang. Nonword repetition and word learning: the nature of the relationship. The etiology of variation in language skills changes with development: a longitudinal twin study of language from 2 to 12 years.
G is for Grammar-Translation | An A-Z of ELT
How children use input to acquire a lexicon. Child Dev. Non-native features of near-native speakers: on the ultimate attainment of childhood L2 learners , in Cognitive Processing in Bilinguals , ed Harris R. Critical period effects in second language learning: the influence of maturational state on the acquisition of English as a second language. The acquisition of grammatical morphemes by adult ESL students. Weighing the benefits of studying a foreign language at a younger starting age in a minimal input situation.
Classroom SLA research and second language teaching. Non-native English-speaking English language teachers: history and research. The relationship between age of learning and type of linguistic exposure in children learning a second language , in Continuum Companion to Second Language Acquisition , ed Macaro E. Breadth and depth of vocabulary knowledge: Implications for acquisition and instruction , in The Nature of Vocabulary Acquisition , eds Mckeown M.