This book follows on from The Lexical Approach. The Lexical Approach is supposed to have revolutionised English (as a foreign language) teaching. The theory goes that first we had grammar-translation, then we had the Communicative Approach and PPP (present, practise, produce), and now we have the lexical approach. (I’ve missed out some methods in the archaeological record of English language teaching).
I haven’t read The Lexical Approach (due to difficulties finding an eBook version), so this book is my main introduction to the lexical approach. Since it is said, by the publisher, to develop and put into practice the ideas of the lexical approach I think it is safe to say that I can use it as a source book for the lexical approach.
The main idea of the lexical approach is that people learn their own language by recognising, even “unconsciously”, patterns, and that the best way to teach a language as a foreign language, at least English, is to focus on such patterns. The development which is said to put this on a scientific basis is the development of corpora; huge databases of naturally occurring language, which can be analysed to find statistical patterns. At least two authors in this volume are keen to assert that teachers’ intuition as to typical collocations is unreliable. While I accept that some detailed kinds of patterning, especially when we bring in colligation (word + grammar patterns) and semantic prosody (collocations clustering in certain semantic contexts), do require statistical analysis, I am confident that I can recognise high-frequency words and typical collocations. (I have found this preference for trusting the data over the speaker of the language in another EFL book, Vocabulary in Language Teaching [1]. I put it down to a job protection scheme by academics whose work involves corpora).
For those not familiar with what a collocation is, in simple terms it is two or more words which typically occur together. This can be defined statistically or by reference to the ‘firmware’ of native speakers. For example; if I say to a random Londoner “fish..” he is most likely to say “fish and chips”. “and chips” collocates with fish. In the British National Corpus there are 244 incidences of fish __ chips* and 15 of fish pie. In contrast to the claims that teachers cannot ‘intuit’ collocation frequency this statistic corresponds to my expectations. Part of the error may be that I am not ‘intuiting’ this correspondence; I am consulting my own LLM (data bank of patterns) in my brain/mind, which, by definition, is reliable. In the preceding paragraph the following might be regarded as examples of collocations: the main idea of, people learn their own language, recognize patterns, best way to teach a language, put something on a scientific basis, statistical patterns, keen to assert, statistical analysis, I am confident, high-frequency words (this collocation is rather specific to the domain of texts about language as is), typical collocations, put it down to and job protection scheme (by academics). From a teaching point of view the idea is, that the best way to learn a language is by becoming familiar with a large number of such collocations. The role of the teacher is to select suitable authentic materials and draw students’ attention to collocations, encouraging them to record the collocations which they find. From the point of view of the lexical approach the PPP approach displays an over-eagerness to rush to the production stage. Students do not learn by production, especially when that is contrived. They learn by encountering and recognising. Obviously the old adage that ‘teacher talk time’ should be kept to a minimum is called into question by this approach.
There are some useful contributions in this volume from a variety of authors which suggest ways in which the theory of the lexical approach can be brought to bear in the classroom. There are useful discussions of how to define a collocation, for example, one author points out that phrasal verbs meet the definition for a collocation but are probably best left as a distinct unit. The fuzzy line between idioms and collocations is also discussed. One idea that crops up quite a lot is the suggestion that it may be more useful to teach students collocations for words they already know, than new words. Some authors remind us that collocations come in different forms; for example noun + noun (railway station), adverb + verb (hardly recognised him), adverb and adjective (bitterly disappointed), verb + adverb phrase (refuse point-blank). Chapter 5 in the book contains several collocation exercises which teachers can use in the classroom and as the basis to develop their own. The last chapter Beyond collocation in vocabulary teaching is written by a real egg-head. However; once I’d got over my negative reaction to this ‘glass bead game’ unworldly presentation I have to admit the author makes a number of good points, (though he is another who seems to doubt the ability of teachers to “intuit” collocations). In particular he discusses colligation and points out that even though words may belong to a similar semantic category they may have different colligations; e.g. for example taking a list of professions he notes that accountant can be used with a possessive pronoun as in ‘my accountant’ but actress very rarely is. Or, again, ‘architect’ is quite often used metaphorically, as in ‘he was the architect of the peace plan’, but accountant rarely is. Another interesting idea is ‘beyond collocations’. Based on his extensive corpus research, (it seems the author may have built his own corpus based on the Guardian), he suggests that collocations group according to a super category of ‘semantic prosody’. For example, ‘chilly’ seems to be most related, (statistically), to time and place and hardly to people; “chilly morning”, “chilly Cornwall”. The suggestion is that it is important to use real world examples of collocations taken from authentic texts so they accurately encompass all these patterns; colligational and to do with semantic prosody.
This is a really useful book for those who are teaching English as a Foreign Language. This reviewer has been up until now teaching using an eclectic approach; elements of communicative and PPP, though I think it is a long time since I forced students to ‘produce’ the taught grammar or vocab, elements of the ‘natural approach’ **, elements of organised vocabulary teaching, strong elements of personalisation, use of authentic materials, and elements of grammar-translation. Since reading this book collocations have become the main focus of my lessons. The main observable result is that I am enjoying teaching much more! I also feel I have, not a system, but a coherent approach, based on a theory. I feel more organised. This is presumably my “lexical approach” moment. If you are a teacher of English as a Foreign Language and you are not familiar with the lexical approach you do need to be! This book would serve as a very useful primer. The original The Lexical Approach by Michael Lewis probably would as well.
However, there is a major theoretical problem with the theory of the lexical approach. Following is a brief discussion of the problem, chiefly with reference to Chapter 8 of this book, ‘Learning in the Lexical Approach’ by the editor Michael Lewis.
The essential problem is the claim that language is grammaticalized lexis, not lexicalised grammar. Lewis may go further than some of the other authors in his book in claiming that grammar does not exist. There are no grammar rules, only patterns. The brain/mind learns ‘grammar’ simply by ‘unconsciously’ recognising patterns. (The author of Chapter 11 shares this view: “..so-called Language Acquisition Device in a baby’s head is more likely to be a set of concordancing ‘software’ that enables us to find regularities and recurrent features in our linguistic experience, rather than any abstract grammar-making device.”) Lewis, essentially, wants to abolish grammar: “Within the traditional grammar/vocabulary dichtomy, teaching collocation emphasises vocabulary rather than grammar, but this is the wrong way to look at the issue. The dichotomy is invalid, language is fundamentally lexical…”. Lewis presents surprisingly few arguments to support his dismissal of grammar and, (more seriously), the idea that there are grammar rules, as a property of language, and that these can be learned. He gives the example that we can say “I’ll see you tomorrow”, but would rarely say “He’ll see you tomorrow”. This example is supposed to illustrate his point that what we think is ‘grammar’ is just general patterns found in lexis. This is not convincing. While we do not know (unless neuro-imaging techniques have shed light on this question) what goes on in the brain when we form sentences, he could at least propose a hypothesis. He doesn’t. As for “He’ll see you tomorrow” being rare, it obviously isn’t that rare, for example; “He’ll see you in his office tomorrow morning”. The underlying philosophical model is positivism and indeed at one point Lewis quotes the basic tenet of positivism that something can only be considered knowledge if it could be disproved. But he suffers from the same problem as positivism. This is a definition of one type of knowledge. It is in fact not a philosophical statement which describes the world but a position statement that we will only consider a certain field of knowledge, (sense knowledge and repeatable physical experiments). Lewis wants to argue that language is just patterns; there are not ‘abstract rules of grammar’. But this is because that is all he looks for. He sees what his model permits him to see. It is a circular argument. By analogy; if an observer watches 100 children entering a school, 50 by one gate marked boys and 50 by another gate marked ‘girls’ what happens inside his brain? In Lewis’s version he simply registers a pattern. The alternative view would be that he reasons that there must be a rule in play here. Michael Swan writes that “a vast amount of exposure would be necessary for adult learners to derive all types of grammatical structure efficiently from lexis by the analysis of … chunks”. The rules, in my view, in some way can accelerate learning. [2] Also; what about mathematics. The human brain, (unless deficient) comes equipped with a mathematics module. Why would this module have nothing to do with language? Why would it be completely compartmentalised away from language? It seems unlikely.
Lewis makes a distinction between Newtonian physics and modern particle physics and argues that the supposed ‘grammar rules’ are just like the laws of Newtonian physics which are generalisations and not a true account of the physical world. This seems a particularly odd analogy to use. Newton’s laws are, (as far as I know), mathematically sound, very reliable (they don’t change and can be used in a wide variety of cases) and they have enormous predictive power. I would never claim more than that for grammar rules; but, that is enough to make them a) rules and b) worth learning.
The English tense system is remarkable for its consistency. (As is the system of zero to fourth conditions). It could be expressed mathematically. It may be that, for example, the present perfect can be shown to be bound to certain patterns or semantic fields but this does not invalidate the claim that is exists as a rule. Some rules are only applicable in certain contexts.
Lewis talks about knowledge being transferred from the “formulaic memory based” to the “analytic rules based part” – but it seems it wants to make this a purely one-way linear process and one which happens “unconsciously”. There are no rules, there is no bank of rules stored as rules (just patterns), and in as much as there is a knowledge of rules these are synthesised from the input patterns in a closed system. This though is simply a kind of behaviouralist model. His ‘results’ are produced by his point of view. In a highly inflected language such as Russian how do learners produced certain forms, (let’s say the present participle), of a verb when they have never seen this form before – for this verb? They consult something in their brain/mind. What? Looking at my own experience of Russian I think I can see possibly both systems in operation; I may refer to examples – oh I could try that ending. Or; I can try and remember what the grammar rule is and apply that, logically. In truth I I think I tend to do the first more, but refer to the second for assistance. Nonetheless this means, (if this is indeed what happens), that I am using and applying consciously learned rules to help me speak a foreign language, something which Lewis denies is possible. Specifically, Lewis says; “What does seem clear is that any analysis performed by the learner is based on inductive generalisation of language which is already part of the learner’s unanalysed intake”. I think is is not what happens. When I use my knowledge of grammar to check my output I consciously apply a rule which my teacher has taught me. At least sometimes. (As a separate point it seems to help if that teaching moment is associated with some affect).
Lewis cites Thornbury as saying that the danger of the lexical approach is its overreliance on formulaic chunks will lead to premature fossilisation. Lewis’s response is surprising. Essentially he argues that most learners never get beyond an intermediate level anyway! At least he backs up his own line that the holiday phrasebook approach to language learning is more appropriate than you think.
Lewis throws out the baby with the bathwater. He over-extends the discovery about collocations. There are reasonable theoretical grounds to suppose that rules of grammar exist. I think there is some empirical evidence available that teaching and learning rules of grammar can help with speaking a foreign language. To be clear; I am not trying to restore the dichotomy between grammar and lexis and limit the lexical approach only to lexis. There are approaches to teaching grammar which can be found in the lexical approach. For example providing a lot of contextualised and authentic examples of the colligation ‘should have done something’ or asking students to underline them in a text could be more useful than writing on the board should + have + V3 + object. I also agree that simply teaching grammar and expecting students to produce it as in a typical PPP grammar lesson is rather a fruitless approach. (I once spent 6 weeks teaching a class of 22 12 year olds of slightly above average ability the Past Simple and Present Perfect in a typical PPP style. At the end all I could claim is there were slightly fewer mistakes in the Past Simple form. No one seemed to have grasped the semantic difference between the Present Perfect and the Past Simple and the only people who could produce the Present Perfect were the ones who could do so at the beginning). That said; I do think that there is a role for explaining grammar rules. For some students this may be more helpful than for others. If nothing else the rules can be used to check utterances. I think they can also be used to produce grammatical forms of new words when that particular grammatical form of that new word has not been seen before though I accept that this may be relatively rare and production from patterns may be more common.
Teaching Collocations. Ed. Michael Lewis. 2000. Thomson Heinle Language Teaching Publications ELT 2000
* I didn’t check that every single one was fish and chips. Hence the ___.
** This book is mentioned in the text and sounds very interesting. The Natural Approach. Stephen Krashen Janus 1996. It appears that the theory of the natural approach shares with the lexical approach the idea that there is no point teaching grammar rules, (but, I note, absent from the present book suggests that consciously learned grammar rules can be used to check naturally acquired language). In addition, Lewis argues that there is a place in teaching for pointing out language features to students, and says that is absent in the natural approach. Lewis also claims that acquisition is non-linear whereas the natural approach says hat there is a definite order of acquisition of grammatical structures. https://www.sdkrashen.com/content/books/the_natural_approach.pdf
Notes
- Vocabulary in Language Teaching. Norbert Schmitt. Diane Schmitt. Cambridge University Press (who do not give a date on the cover page in the Kindle edition).
- Chunks in the classroom. Let’s not go overboard. Mike Swan 2017.