For more related articles on the subject in HLT see:
'My good-bye to the Lexical Approach’. Seth Lindstromberg. Issue 5/2, 2003.
'Why I won’t say good-bye to the Lexical Approach’. Hanna Kryszewska. Issue 5/2, 2003.
'Chunking for language classes: some lesson outlines’. Hanna Kryszewska. Issue 5/5. 2003.
'Lexical chunks offer insights into culture’. Hanna Kryszewska. Issue 8/3, 2006.
'Means of mass memorization of multi-word expressions, part I: The power of sound patterns’. Seth Lindstromberg & Frank Boers. Issue 7/1, 2005.
'Means of mass memorisation of multi-word expressions, part II: The power of images’. Frank Boers & Seth Lindstromberg. Issue 8/1, 2006.
'Teaching Chunks of Language: The Issue of Memory'. Seth Lindstromberg & Frank Boers, Issue 6, 2008
Revisiting ‘My Good-bye to the Lexical Approach’
Seth Lindstromberg, UK
Seth Lindstromberg is a teacher and teacher trainer at Hilderstone College (Broadstairs, Kent, England). E-mail: SethL@hilderstone.ac.uk
Optimizing a Lexical Approach by looking inside chunks
The Lexical Approach optimized
Several years ago, hltmag included an article by me entitled ‘My good-bye to the Lexical Approach’ (issue 2, 2003). Mario Rinvolucri, the founding editor of this webzine, recently suggested to me that there might be some kind of contradiction between what I wrote in that article and what I have written since. Namely, in various recent books and articles Frank Boers (Free University of Brussels) – and I have argued that the Lexical Approach ( or LA) can be made more effective1. By saying this of course we imply that the LA has merit. Mario challenged me to explain myself in print.
First of all, people ought to change their mind if additional learning goes against early belief. Concerning the LA, I have learned more. In 2003 what I objected to about the then prevailing version of this approach, as described and promoted most notably by Michael Lewis (e.g., 1993, 1997), was the tenet that it is pointless for a learner to ‘look inside’ any particular L2 lexical phrase, or ‘chunk’, in order see why it includes the particular words that it does (e.g., 1997, pp. 17-18). After 2003 I came to realize more clearly that rejecting this one tenet of the LA does not require rejection of the LA as a whole. In fact, not only can you still employ the LA when you encourage your learners to ‘look inside phrases’, you can actually employ it better.
Here is something that everyone can agree about: Believing in the LA means believing that teachers and materials writers should give vocabulary more attention relative to grammar than it got (at least in some mainstream Anglo-American approaches) during much of the mid-20th century. In fact, the LA approach arose out of and along with a rise in interest in the teaching and learning of vocabulary generally which began about 1980. The LA, which came along about ten years later, was encouraged by evidence put forward by linguists (only a few, initially) that a great deal of vocabulary was phrasal (e.g., Sinclair, 1991). This substantially new view of vocabulary (indeed, of language) was eventually so successfully popularized in language teaching circles by methodologists such as Lewis (op. cit. ) and Nattinger and DeCarrico (1992), that the LA seems now to be associated in teachers’ minds not with single words at all but instead solely with chunks. More specifically, in the minds of many teachers the gist of the LA seems now to be the simple and unadorned proposition that L2 learners should learn L2 chunks in masses, which is all well and good provided that we remember that single words are vocabulary too, and that in a pinch most of them can be quite useful on their own.
Anyway, suppose we accept the opinion that L2 learners need to acquire masses of chunks. Then we ought to consider at least the following questions:
Which chunks should teachers and materials writers be concerned about?
A common answer is ‘The most frequent ones’. However, the number of highly frequent chunks (such as I mean and of course) is relatively small. The great mass of chunks are not at all common compared, say, to the two or three thousand commonest single words. Also, as Frank Boers has pointed out (e.g., Boers and Lindstromberg, 2009), it seems to be precisely these very highly frequent chunks that learners are most likely to acquire on their own, provided only that they get a very modest amount of comprehensible authentic input. And why spend a disproportionate amount of valuable class time on items that may be acquired with little or no teaching?
‘How many chunks should we expect learners to acquire?
It seems perverse to adopt the LA, as it is now commonly understood, and then not hope that your students will acquire many L2 chunks. But, as we have just seen, the vast majority of these chunks will not be so frequent that learners can easily pick them up on their own. This is because low frequency words and phrases, by definition, tend to occur in input at widely spaced intervals, within which any memory of what was learned on the previous encounter may well fade away.
‘Where should we expect chunks to be learned?’
For the reason just mentioned, it appears that typical learners are only likely to pick up large numbers of chunks outside of class if they get a large amount of comprehensible L2 input over a considerable period of time. But comprehensible aural input is only likely to be abundantly available in settings where the L2 is a local lingua franca. Recognizing this, some authorities have placed considerable faith in students’ ability to learn chunks during out-of-class reading for pleasure. Frank Boers (ibid., Chapter 3) has gone into some detail about why this faith is weakly justified – even when learners receive in-class guidance about chunk learning strategies. (For one thing, chunks are not that easy for learners even to recognize, let alone notice and learn.) Available published evidence suggests that most L2 learners acquire single word vocabulary during reading for pleasure at a disappointingly slow rate. This in turn suggests that any teacher who has great faith in the LA and whose students receive only a few hours of L2 tuition a week should consider whether it might not be best, after all, to do more chunk teaching in class and not trust to the vagaries of acquisition through non-didactic exposure to L2 out-of-class.
‘But what about actually remembering chunks?’
Although the number of high frequency chunks is small, the number of medium frequency chunks is vast. Further, in addition to the high and medium frequency chunks, there are a great many others which, though comparatively rare, are nevertheless known by most adult native speakers. Accordingly, everyone who accepts that the number of L2 (e.g., English) chunks is very great, seems also too accept that only a small proportion of them can possibly receive much attention in class (on this point see, especially, Swan, 2006). But even if one decides to focus only on chunks in the medium frequency band, the problem of prioritization and selection remains acute. And even if we provisionally (but perhaps unrealistically) assume that there is an easy way to select among the evidently very large number of chunks in the medium frequency band, how are our students supposed to remember targeted chunks once these chunks have been noticed and then understood? It is here, when the matter of memory arises, that the current mainstream version of the LA shows a serious deficiency.
Although Lewis and others urge us to increase our attention to chunks, the LA has furnished us with next to nothing in the way of new teaching methods, techniques, or exercise types.2 Worse, the standard version of the LA forswears precisely those strategies which actually do offer learners robust hope of remembering chunks not just passively – so that learned chunks can be recognized – but actively – so that they can be promptly called to mind when needed for fluent speech and writing. This is crucial, for without prompt recall, the much talked about benefits of the LA for fluency and idiomaticity (or naturalness) of expression will simply not occur.
This brings me back to my original (2003) objection to the standard version of the LA: it tells us that learners should not look inside of chunks.
You can look into a chunk in order to see aspects of both meaning and form.
Let’s take as our first example the chunk drive a wedge between [two people]. Suppose an EFL learner (1) considers this phrase for the first time, (2) notices the word wedge, (3) knows or discovers what a wedge is, and (4) associates this chunk with an image of a wedge being driven between two parts of an object, or even between two closely associated people, with the result that they are forced apart. This kind of imagistic association of meaning and form is well known to promote long term recollection of vocabulary. To gain this mnemonic benefit, though, a learner needs to look into the phrase and consider, in particular, what the words wedge and drive can mean figuratively (e..g., metaphorically and/or metonymically). Without doing this, a learner will find it relatively hard either to fully understand the chunk drive a wedge between or to durably and fairly quickly remember its form. It seems amazing that some proponents of the LA have explicitly rejected this kind of work, but reject it they have. (See the discussion in Boers and Lindstromberg, 2009, pp. 70 ff.)
Let’s now consider the chunk plain sailing. Suppose a learner has already found out that it denotes problem-free conditions as in: [The] deal with Sony hasn't all been plain sailing (http://www.collins.co.uk/Corpus/CorpusSearch.aspx). Suppose this learner now notices that the /ei/ dipthong is repeated as indicated by the underlining just above. (The learner may also notice the repetition of /l/.) Empirical studies have shown that even brief noticing of common patterns of vowels and consonant repetition (especially, rhyme, alliteration and assonance) has the potential to help learners remember the precise lexical form of L2 chunks, especially if the target chunks are spoken and/or heard rather than just read or written.
Obviously, a good memory of a chunk’s meaning and a hazy memory of its form, may still be good enough for reading and listening, for if learners encounter a chunk and recognize it, they may well be able to understand it in context. But such passive (or receptive) knowledge of a chunk will, by definition, be insufficient to enable anyone to use the chunk in fluent speech. For this, it is necessary to have a good memory not only of the meaning of the chunk but also a good memory of its full form, including its exact lexical make-up. Encouraging learners to look inside of chunks, and guiding them about how to do so on their own, can be the vital missing ingredient of the standard version of the LA because:
- Imagistic, word-based understanding of a chunk fosters memory not only of its overall meaning but also the meanings of its key words, especially its key content words.
- Noticing patterns of sound repetition can foster recall of the wording of a chunk in detail. Note that this second point can apply even to chunks which are low in image potential (e.g., powers of persuasion; if you say so).
The LA as developed during the 1990s was onto something important, but it did not address the issue of memory at all adequately. The strikingly traditional exercise types which the mainstream LA has continued to rely on are not in general well-suited to help learners form well-entrenched memories of either the meaning or the form of L2 chunks. Yes, from its inception the LA paid heed to the well-founded claims of linguists that chunks are ubiquitous in authentic speech. But major advances in the experimental study of memory-for-language were comprehensively overlooked. Redressing this oversight should bring with it good prospects for significant enhancement of the LA despite the all too real fact that there are myriads of chunks worth learning yet – in many instructional settings – there is dauntingly little time in which to learn them. We teachers can draw some encouragement from the fact that study after study has shown that deliberate, in-class teaching of vocabulary (of all kinds) can be efficient and effective.
Have I said good-bye to the Lexical Approach? No. I have only said good-by only to the unsound rejection of looking inside chunks’ in order to unlock the mnemonic potential of figurative language and patterns of sound repetition.
1 See, e.g., Lindstromberg and Boers (2008) and Boers and Lindstromberg (2009).
2 Concordance analysis is one new type of exercise, but no one has yet persuaded me that it has the potential to help students learn L2 chunks in significant numbers – even for recognition purposes, let alone for the purpose of using L2 chunks in fluent speech. Firstly, concordance analysis is time-consuming and therefore best undertaken in order to investigate selected, relatively troublesome, high frequency items. Secondly, it seems doubtful to me that concordance analysis can have durable appeal to that mass of learners who are only intermittently inclined to intensively study the form of printed language.
Boers, Frank and Seth Lindstromberg. 2009. Optimizing a Lexical Approach to Instructed Second Language Teaching. Palgrave Macmillan.
Lewis, Michael. 1993. The Lexical Approach: The State of ELT and a Way Forward. LTP.
Lewis, Michael. 1997. Implementing the Lexical Approach: Putting Theory into Practice. LTP.
Lindstromberg, Seth and Frank Boers. 2008. Teaching Chunks of Language: from Noticing to Memory. Helbling Languages.
Nattinger, James and Jeanette DeCarrico. 1992. Lexical Phrases and Language Teaching. Oxford University Press.
Sinclair, John. 1991. Corpus, Concordance, Collocation. Oxford University Press.
Swan, Michael. 2006. ‘Chunks in the classroom: Let’s not go overboard’. Teacher
Trainer, 20(3), pp. 5-6.
Please check the Methodology and Language for Secondary Teachers course at Pilgrims website.
Please check the Methodology for Teaching Spoken Grammar and English course at Pilgrims website.