top of page

Lexical approach and chunking

Recently, I have been looking more closely at the research analysing the ‘lexical approach’ of language teaching. Here are some key facts that have been summarised by Scott Thornbury (2019) in Learning language in chunks. Part of the Cambridge Papers in ELT series.[pdf] Cambridge: Cambridge University Press. (For further reading, see references).

Even though the paper refers in more detail to ELT there are also many parallels drawn to SLA which might be considered in context in our classrooms.

Over 25 years ago Michael Lewis published ‘The Lexical Approach’ (Lewis, 1993), prompting a radical re-think of the way we view language and how we teach it. He argued that “language consists of chunks, which, when combined, produce continuous coherent text” (Lewis, 1997)

By ‘chunks’ Lewis was referring to the following:

  1. collocations (to get a call, to do the shopping, give way …)

  2. fixed expressions ( all of a sudden, by the way …)

  3. formulaic utterances (I’ll get back to you later, I’m on my way…)

  4. sentence starters (I believe that, As far as I know…)

  5. verb patterns ( I hate to tell you…/I hate flying…)

  6. idioms and catchphrases  (under the weather, break a leg …)

Lewis was not the first to relate to language in these terms, however, his contribution was to argue that language teaching needed to be reformed/revolutionised...

(Lexical) chunks consist of more than one word, they are conventionalised, exhibit varying degrees of fixedness and idiomaticity and they are probably learned and processed as single items (‘holophrases’). 

There is growing evidence, i.e. from read-aloud studies (Ellis et al, 2008) that chunks are processed holistically, rather than as a sequence of individual words. 

A number of studies have established that there are many chunks that are as or more frequent than, the most frequent individual words (Shin & Nation, 2008).

How could the learning of chunks profit language learning?

The 3 main reasons that have been put forward for prioritising the learning of lexical chunks are:

  1. Fluency – facilitation of fluent processing. The possession of memorized store of ‘chunks’ allows more rapid processing, not only for the production of language but also for reception, since ‘it is easier to look up something from LT memory than compute it’ (Ellis et al., 2008). Some researchers noted that when chunks were used more confidently, they contributed more to the perception of fluency (Boers & Lindstromberg, 2009)

  2. Idiomaticity‘The use of chunks can help students to be perceived as idiomatic language users, disposing of a relatively impressive lexical richness and syntactic complexity’ (Boers & Lindstromberg, 2009), alluding to native-like proficiency.

  3. Language development – another strong argument in favour of a lexical approach is that ‘lexical phrases may also provide the raw material itself for language acquisition’ (Nattinger & DeCarrico, 1989). In other words, the phrases are first learned as unanalysed wholes. ‘The Lexical Approach claims that, far from language being the product of the application of rules, most language is acquired lexically, then “broken down” … after which it becomes available for re-assembly in potentially new combinations’ (Lewis, 1997). Ellis (1997) agrees: ‘Learning grammar involves abstracting regularities from the stock of known lexical phrases.’

Based on studies (mainly with young learners), Ellis & Shintani (2014) accept that ‘the prevailing view today is that learners unpack the parts that comprise a sequence and, in this way, discover L2 grammar. In other words, formulaic sequences serve as a kind of starter pack from which grammar is generated.’

Other researchers are less convinced, especially when it comes to literate adult learners who are inclined to unpack formulaic expressions for their words not syntax. (Wray, 2002). 

Scheffler (2015) concurs that, even if these ‘unpacking’ processes apply in L1 acquisition, the sheer enormity of the input exposure required to ‘extract’ a workable grammar is simply unfeasible in most L2 learning contexts’ (Thornbury, 2019).

Swan (2006) argues that: ‘ Much of the language we produce is formulaic, certainly; but the rest has to be assembled in accordance with the grammatical patterns of language […]. If these patterns are not known, communication beyond the phrase-book level is not possible.’

How are chunks best learned and taught?

In the summary mentioned above, these 4 groups are cited:

1. The phrasebook approach – the practical applications of the approach are: 

  1. rote learning of formulaic expressions

  2. drilling

  3. shadowing – learner listens to extracts of authentic talk and ‘sub-vocalises’ at the same time

  4. jazz chants – for young learners, preselected chunks are embedded via chants and songs

2. The awareness-raising approach – Krashen (1985) Input Hypothesis – the necessity for high quantities of roughly-tuned input as a source of learning

  1. extensive reading and listening tasks

  2. ‘chunking’ of texts – identifying possible chunks

  3. listening to extracts of authentic speech and marking a transcript into tone units to identify likely chunks

  4. record-keeping and frequent review

  5. recycling chunks in learners’ own texts (spoken or written)

3. The analytic approach – while Boers & Lindstromberg (2009) agree that time should be devoted to raising awareness about the role of chunks, they are sceptical that learners will be able to identify them without an aid. Their research supports the view that directing learners’ attention to the compositional features of chunks can optimise their memorability. Their analytical approach comprises of:

  1. teach vocabulary in chunks 

  2. select chunks for targeting, also based on collocational strength and teachability

  3. reveal non-arbitrary properties of chunks to make them more memorable

  4. complement to improve chances of retention, elaboration of meaning and form

4. The communicative approach – Gatbonton & Segalowitz (2005) propose approach called ACCESS, which incorporates stages of controlled practice of formulaic utterances, embedded within communicative tasks. In other words, presenting and practising short chunks of functional language, before learners partake in interactive tasks that require the repeated use of these chunks for communication.

‘The ultimate goal of ACCESS is to promote fluency and accuracy while retaining the benefits of the communicative approach. In ACCESS, this is accomplished by promoting the automatization of essential  speech segments in genuine communicative contexts’ (Gatbonton & Segalowitz, 2005).

With the restriction of the language curriculum time currently available, as well as the attitudes to language learning in many U.K. schools (even though it seems to be improving due to the pressures of Ebacc), the Lexicogrammar approach advocated by G.Conti which teaches language via chunks using sentence builders foremost and focuses on explicit grammar application later on in its sequence, might be the better solution for many schools in terms of development of communicative skills, self-efficacy and motivation compared to the very didactic approach of NCELP, which might be more suitable for highly academic or advanced learners. Or indeed combining a variety of approaches based on our specific contexts; especially if we want our students to know more and remember more (Ofsted MFL review, 2016). 

Key takeaways

Whatever strategy we might embrace, effective teaching should take into account the following:

  1. distinction whether we teach production (speaking/writing) as well as reception (listening/reading)

  2. the need to focus on meaning as well as form

  3. the importance of teaching vocabulary in context, not as isolated items

  4. the need to teach vocabulary deliberately, rather than just rely on incidental learning (Webb & Nation, 2017)

  5. providing learners with the autonomy about what they are learning – ‘the more one engages with a word (deeper processing), the more likely the word will be remembered for later use’ (Schmitt, 2000) 

  6. the importance of forming associations with vocabulary items – schemata – to aid retention and support memory – LTM – recall

  7. the need for regular reviews – spaced and interleaved practice – retrieval practice

  8. development of metacognitive skills – learners setting their own targets and measuring their success against them, making decisions about what and how they learn vocabulary


Ellis, R., & Shintani, N. (2014). Exploring language pedagogy through second language acquisition research, 71. London: Routledge

Gatbonton, E., & Segalowitz, N. (2005) Rethinking communicative language teaching: a focus on access to fluency. Canadian Modern Language Review, 61

Lindstromberger, S., & Boers, F. (2008). Teaching chunks of language: from noticing to remembering. Helbling Languages.

Selivan, L. (2018). Lexical grammar: Activities for teaching chunks and exploring patterns. Cambridge: Cambridge University Press.

Webb, S., & Nation, P. (2017). How vocabulary is learned. Oxford: Oxford University Press

Scott Thornbury (2019) Learning language in chunks. Part of the Cambridge Papers in ELT series.[pdf] Cambridge: Cambridge University Press.


bottom of page