Marcin Lewandowski talks about the value of learning ‘chunks’ of language to improve fluency, vocabulary and pronunciation.
Let’s try a little experiment. Look at the following letters for a few seconds: R, O, Z, R, Y, W, K, A. Now, look away and write the letters on a sheet of paper from memory. How did you do? Unless you know Polish, you might have struggled.
Let me explain, in order to remember the letters you have to keep all of them in your working memory. Chances are that you tried to remember the letters as letter names rather than phonemes. This puts a strain on your working memory which for an average person has a capacity of about 7 items (plus or minus two). This capacity to hold bits of information in your working memory is called a memory span (or a digit span when it’s tested on numbers).
Now, what if I told you that the letters above compose a word roughly pronounced /rozrɪvka/ and that this word means ‘entertainment’. Suddenly, the seemingly meaningless stream of letters is not so meaningless anymore. Instead of eight sounds, you now have to remember only one word or three syllables at most. This minimises the cognitive load on your working memory releasing some of this processing power to allow you to process even more information.
Not convinced? Try another example from Roberts and Kreuz but this time with numbers rather than letters. Ready?
149217762001 Remembering this stream of numbers is very challenging until you realize that this number is made up of three significant American history dates. Namely, the discovery of America by Columbus (1492), declaration of independence (1776) and the 9/11 terrorist attack (2001). You have ‘chunked’ the seemingly unrelated items and imposed a meaning onto them. Now, instead of twelve meaningless items, you only have to contend with three facts.
Applying chunking in the classroom
Michael Lewis argues that memorising chunks of language as meaningful units reduces the strain on our working memory and makes them more readily available and retrievable when needed. This, in turn, adds fluency. Learning chunks of language promotes more natural pronunciation as it encourages connected speech. It also minimises possibility of error when producing language (speaking or writing).
Take the way we give advice for example. We commonly tend to use expressions such as:
- If I were you, I’d…
- You’d better…
- Why don’t you…?
Learning them as formulaic phrases, will not only reduce the load on our working memory but will also allow us to introduce them earlier in the curriculum. Because they’re fairly fixed, such phrases can be introduced as a whole without much analysis of their constituent parts. This means that we don’t have to wait until we’re covering the second conditional to introduce ‘if I were you…’.
And that’s indeed what we do at early stages of teaching. Expressions such as ‘what’s your name?’, ‘where do you live?’, ‘my name’s….’, ‘I live in…’, etc. are often taught as formulaic sentences without much focus on auxiliary verbs, word order or morphology. Learners learn them as lexical chunks and get on with communication without being hindered by thinking about grammar. To clarify, I’m not suggesting that we should do away with teaching grammar and focus exclusively on teaching lexical chunks instead. Quite the contrary, it’s probably impossible to learn an entire language in chunks alone.
However, providing learners with activities that focus on collocations or fixed phrases will make them more aware of the interdependencies between words and lead to automatic associations available for a quick recall.