Intro and contents



Reaching out



Formulaic language and chunking

vEnglish summer camp

This is an anecdotal reflection on the subject of teaching functional phrases, formulaic* language, collocations and "chunking" (Lewis 1993, 1997). Many years ago I was a teacher on an English summer camp, which was held, not in the UK, but on a large private school campus in Cadiz in the south of Spain. They were hot affairs those summer camps. Everything went on within the high walls of the school premises with no access to the outside world. Morning English lessons in airless classrooms followed by various sporting activities under the searing Andalusian sun during the afternoon. The school had hired "authentic" English monitors, flown in from the UK, with neither a word of Spanish nor experience as language teachers. Nevertheless, they were rearing to go and enthusiastic about monitoring lots of sports events for the school children campers.

It was the first day of the English summer course and the children began to arrive at the school, brought in large Mercedes cars and Volvos by tearful parents perhaps not wholly convinced that for the sake of the English language they were doing the right thing by parting with their children (some as young as seven) for a two-week summer camp. However, the course organisers were full of reassuring sentiments and smiles. Meanwhile, the newly-arrived children nervously milled around the playing-field, their surroundings and fellow campers unknown to them. A youthful monitor, taking it upon himself to ease the obvious tension walked boldly up to one of the small boys, kneeled down in front of him and said with a sheepish grin, “You all right, mate?” The effect was immediate. The small boy gave the monitor a horrified stare, turned on his heels and ran back screaming to the arms of his alarmed parents. Then he turned and pointed accusingly at the "ogre" monitor who had growled at him in some menacing and unrecognizable dialect.

I felt a strong urge to go up to the bewildered monitor and tell him that what he should have said was, how are you? so that the child could have replied, I’m fine, thanks! (or, I’m very well, thank you) and all would have been right with the world. In the end, I said nothing. There was too much to explain anyway about classroom functional phrases so I let it go. Hopefully, by the end of the two-week summer camp the child (if he could survive) might even pick up a little authentic English.

Should we teach so much formulaic language?

The appealing thing about formulaic language is that it does provide a semblance of fluency in the classroom. Young learners (and their parents and teachers) may feel that progress in English instruction is taking place if the former can produce a few ready-to-go phrases to take home. Motivation for further involvement in language study may well result from these initial successes - quite simply and rapidly learnt. On the other hand, once a functional phrase is taught in classroom English, it often becomes a unique standalone representative of that function. Perhaps teachers feel they do not want to encumber learners with unnecessary alternatives which may lead to confusion. This then means that the how are you? / I'm fine, thanks duet may be the only language presented in the class for asking after someone in the second person and replying to that question. A course curriculum needs to cover many communicative situations and the thinking that one phrase per situation is enough can seem reasonable. On the other hand, through excessive formulaic instruction at beginner levels, could language teaching and learning possibly run the risk of becoming little more than a pedagogy that addresses common communicative themes with a fixed repertoire of rote-learned phrases?

A further issue is that due to the fixed content of formulaic phrases either all the phrase or part of it may prove indivisible for many students. It may be difficult for a beginner level learner to realize that these phrases are made of component parts and that those component parts are words that can be manipulated, reorganised and other novel phrases constructed from them.

Michael Lewis carried out much work on the idea of language "chunking" in what became known as the Lexical Approach (Lewis 2003, 2007). He believed that “language consists of grammaticalized lexis, not lexicalized grammar” (Lewis 1993:34). The thinking being that language instruction should throw out word lists and grammar focussed instruction and instead concentrate on teaching lexical “chunks”. This approach, or at least vestiges of it, is still present today in much mainstream language teaching. The idea of explicitly teaching our students well-known, pervasive chunks from English is intriguing. However, Krashen observed that beginner learners rather than being able to acquire language this way initially seem to just grasp at words to build their early utterances (together perhaps with some very basic set phrases), (Krashen 1995:58). Furthermore, these rudimentary attempts at communication at a word level have been observed as a vital stage in language development. Through dealing with language at the very smallest component part level, a grammar and syntactical structure can be built upon at a later stage (Skehan 1998). If this really is the case, then premature teaching of collocations and formulaic, functional phrases may conflict with the learners’ cognitive developmental needs – learners need words or at the most very small component parts of phrases to build their grammars.

I wonder at the benefits of excessive formulaic phrase teaching for the acquisition process at the outset of language instruction other than initial motivational ones. Nevertheless, I suspect there exists sufficiently rich L2 at the word level for learners to grasp and incorporate into their interlanguage at least in the kind of mainstream textbook-driven classrooms I have experienced. In fact, it is probably strategically complex to present a language as a continuous input of chunks and functional expressions; words will always abound. Learners will naturally strive to apply a grammar to their sentence construction during output whether it is erroneous or not and so will avoid language chunks as find they cannot work them into their notion of a syntactic grammar. Learners need recognisable subjects, verbs, objects, adjectives and the like. This may be why Richards and Rodgers (2001) conclude about the Lexical Approach that it “is still an idea in search of an approach and a methodology” (p. 138).

Perhaps my real objective here is to highlight how a consistent gesture language in the classroom such as I have described in other articles can deliver input at the word level yet provide ample exposure to the collocations and formulaic language within. When the teacher gestures word for word and the learners (in the Silent Sign approach) gesture and make the corresponding utterances, they are receiving exposure which is both holistic and yet clearly comprised of recognisable component parts. Asking learners how they are does not have to be expressed by the same exact phrase as though etched in stone but how do you feel? are you all right? are you feeling ok? (probably not you all right, mate!) are all possible variables when elicited as gestured phrases at the word level. By being exposed to language in this way learners can then begin to glean individual word meanings and experiment with how they can be re-used or other words inserted to produce novel sentences - I'm feeling great today! When this happens, the teacher can then feel satisfied that fluency (and not a semblance of it) is truly taking place in the classroom. Finally, if learners are exposed to language at the holistic level and sufficient recycling of language of this type occurs in the classroom then they will absorb and acquire those chunks and functional phrases in the same way a child learning the mother tongue also ingests and finally uses them - not necessarily through instruction but just because they are there in the language so cannot be avoided.

*As defined by Wray (2002), a formulaic sequence is "a sequence, continuous or discontinuous, of words or other elements, which is, or appears to be, prefabricated: that is, stored and retrieved whole from memory at the time of use, rather than being subject to generation or analysis by the language grammar".


Krashen, S.D. & T. Terrell. (1995). The Natural Approach. Language acquisition in the classroom. Phoenix ELT. Prentice Hall International.

Lewis, M. (1993). The lexical approach: The state of ELT and the way forward. Hove, England: Language Teaching Publications.

Lewis, M. (1997) Implementing the lexical approach: Putting theory into practice. Hove, England: Language Teaching Publications.

Richards, J., and Rodgers, T. (2001). Approaches and Methods in Language Teaching (2nd edition). Cambridge University Press.

Skehan, P. A. (1998). A Cognitive Approach to Language Learning. Oxford University Press.

Wray, A. (2002). Formulaic Language and the Lexicon. Cambridge: Cambridge University Press. p. 9.

Thanks also to the fascinating personalised account of Lewis's trajectory with the Lexical Approach at Scott Thorbury's site: 


Copyright © 2019 Mike Bilbrough
All rights reserved