The hazy line between lexis and grammar

This post is a short account of two lessons I taught in 2002 which helped me to make sense of something I’d read about in the late 90s, but couldn’t get my head around. Not until then anyway.

If you’ve been following me for some time, you know that I’m a big believer in experiential learning (i.e., moving from concrete experience to abstract conceptualization rather than the other way around), so let me begin by describing the lessons first. That way you’ll be better able to grasp the theory underlying my anecdotes.

In August 2002, I taught a two-month course called “Advanced Grammar 2″, which, as the name suggests, was basically a grammar course, based on a (very good) grammar book. As part of the syllabus, I was expected to teach students how to use gerunds in formal sentences such as “I objected to his going…” and “Her having said that…”, which, to be honest, I wasn’t particularly looking forward to.

The big day finally came and the lesson was a train wreck, from start to finish. It was one of those classes you walk out of wondering whether you’re on the right career path. The poor students, hard as they tried, couldn’t make heads or tails of the new language and needed a great deal of remedial work in the following weeks. I can still recall their less-than-glowing feedback at the end of the lesson:

“This stuff is way too abstract. When am I ever going to use this in real life?”

Cut to October 2002. Most of the students – against all odds – signed up for the next level and, to their despair, I was their teacher again. But the next course was different. It was called “Current Events” and it focused on conversation, listening and vocabulary. There was no coursebook (for obvious reasons), so the more laissez-faire nature of the course meant that I was free to build up the syllabus as we went along, dealing with emergent language and extracting useful lexis from newspapers, magazines and video clips. I am using lexis here in the broadest possible sense, to encompass words, collocations, fixed/semi-fixed expressions and formulaic language.

In one particular lesson, we watched a CNN clip and after the usual true or false activities, I drew their attention to two sentences containing the phrase “the odds”. If I remember correctly, one of them was “The odds of his winning (whatever it was) are pretty slim.” I devised a quick noticing task, concept checked the sentence and briefly highlighted the use of pronoun + gerund, which was exactly the language I’d tried to teach them a few months before.

As fate would have it, the CNN lesson was far, far more successful. Students were able to express a wide range of ideas using the chunk “The odds of (noun or pronoun) (verb + ing) are (adjective) because…” with a good degree of confidence, accuracy and fluency – and by fluency I mean fast retrieval and chunking. They left the lesson with a palpable sense of achievement (“I was able to understand the new language and use it in class”), purpose (“I know what I can do with this piece of language in real life”) and ownership ( “I can make this piece of language work for me”). Interestingly, but in hindsight not surprisingly, only one student saw the connection between “the odds…” and the grammar they’d been exposed to in the previous course. When prompted, the other students responded with empty gazes, signaling that maybe – and I say maybe – they’d kept “gerunds” and “the CNN sentence” in two completely separate mental files.

At the time, I was puzzled by the lack of interface between the August and the October lessons, but after a few months all the pieces came together: It might have made more sense to move from lexis (“the odds of + ing are…”) to grammar than the other way around. In other words, to engage students in slightly more naturalistic acquisitional processes, I should have presented and practiced the new language formulaically first and then, in the following class or week, helped them break the chunk down and analyze the underlying grammar.

An obvious parallel: When we teach beginners to ask “What do you do?” formulaically (=lexically), we rarely engage them in any sort of overt grammar analysis straight away. Instead, we expose students to lots and lots of unanalyzed “do you…” sentences and gradually help them unpack and internalize the target patterns within the chunks, which will, to use Scott Thornbury’s words, act as templates for the later development of a more systematic grammar when they’re ready. The same applies to “Can I help you?” (no need to tackle modals straight away), “If I were you, I’d…” (which students can learn how to use long before they’re formally exposed to the second conditional) and so on and so forth.

This might seem obvious, I know, but in recent years ELT has paid relatively scant attention to the priming role of functional/formulaic language, which is a shame.

So I’d like to suggest that it makes sense to try to move:

From lexis (in the broadest possible sense of the word) to grammar rather than the other way around.
From concrete to abstract.
From synthesis to analysis.

Or, to borrow Michael Lewis’ terminology:

First, grammaticized lexis, then lexicalized grammar.

This is easier to pull off, of course, when you teach one to one and can devise your own materials or adapt your existing coursebook at will. But even if you operate within the constraints of a language institute, there are a number of small changes you can make to your lesson planning that – trust me – might surprise you.

Thanks for reading.