My memories of grammar at school are not exactly positive. Random lessons at random intervals, parsing sentences, labelling parts of speech and identifying clauses. At no point did I make any connection between grammar and the rest of my English lessons, which I loved, and nor did my teachers.

In Tes last week, Dominic Wyse argued that teaching children traditional grammar does not help them write better. We agree.

But the article makes other claims that are simply inaccurate and suggest a rather hefty misunderstanding of modern ways of thinking about the relationship between grammar knowledge and reading and writing. 

The two studies referred to were led by our research team at the Centre for Research in Writing at the University of Exeter. But they are most definitely not studies that promote traditional grammar teaching. Rather, the teaching approach we adopt is one where grammar is considered in terms of its potential to develop writers’ understanding of how the choices they make shape meaning. Consider, for example, these two sentences:

And out of the mists came a figure in flowing green, walking across the water.

And a figure in flowing green came walking across the water out of the mists.

The first is a sentence from a children’s book by Michael Morpurgo, the second is a rearrangement of that sentence in a different way.  Both, of course, are grammatically correct (even the sentence starting with “and”!).  But read them aloud and you hear the different rhythms they create, and if you visualise the plot event being described they create subtly different ways of presenting the narration. The first foregrounds the mists and delays the appearance of the figure, whereas the second foregrounds the figure. Imagine creating a film clip of each sentence and you will see the difference.

Grammatically, this is achieved partly through putting the subject after the verb in the first one and bringing the adverbial before the verb: the two sentences are syntactically different. It is not that one sentence is better than the other, it is that they are different. Each could be used effectively, depending on what the writer wants to communicate. The grammatical terminology, like literary metalanguage, is a tool for explaining this, but we do not emphasise the terminology; we emphasise the example structures and the high-quality discussion that teachers can generate about how these different structures make meaning.

This is the kind of attention to grammar and language that we promote and that has been both popular and successful. It uses rich, authentic texts as the source for looking at the choices writers’ make and empowers children to understand and make their own choices as writers. Ironically, Dominic Wyse said in an earlier interview with Tes: ”But, if you’re going to get better as a writer, you’re not just luxuriating in the experience - you’re looking at how the writer constructs text.”  Yes, indeed - precisely what our approach does.

So what about the research evidence and the two studies reported?

Wyse reports that our first study was a ”robust study” which ”showed that contextualised grammar teaching was effective for improving secondary pupils’ writing”. So it is both robust and positive evidence of effectiveness. Strange then to draw the conclusions about no evidence?

The battle over grammar

The second study was funded by the Education Endowment Foundation and took place in primary schools, and the intervention was again led by our team. The initial report did, indeed, show little evidence of an effect, but the EEF has since changed how it determines effectiveness, and now reports it as showing two months’ progress (or three months’, depending on which part of the website you look at). This is the same degree of efficacy as the recently reported study by Robin Alexander on dialogic talk. Of course, this highlights that an over-emphasis on statistics as “hard fact” is flawed, as even the statisticians disagree with each other.

We have also conducted three further studies not mentioned, perhaps because they use different research methods, all showing real benefits. But to an extent, this is just academics disagreeing about research methods.

The real issue is the ongoing controversy about school grammar that seems to push people into ideological corners and biased reporting. This is not helpful for teachers or children.

Over the past 10 years, we have probably developed the most sustained body of research into the relationship between grammar and writing in the world. This includes in-depth, qualitative research, working alongside teachers to understand how our approach works in the classroom. We are interested in weaknesses and shortcomings, and in critical questions. Importantly, we consistently highlight that our approach is not a magic, silver bullet fix-all and that not all teachers are equally successful. We have also drawn attention to the negative effect of the key stage 2 assessment and the SPaG test on the teaching of writing.

The point we make is not ”grammar is the cure for all writing ills, so teach it this way”, but ”here is an approach that is working, take and adapt”.  This argument is based on our sustained and cumulative research evidence and our experience of working with many hundreds of teachers. It’s time to move from sterile anti/pro-grammar positions to a more mature, informed debate that is prepared to look critically at grammar in the classroom - positives and negatives - and above all, to work with and listen to the voices of teachers.

Professor Debra Myhill is the pro-vice-chancellor at the College of Social Sciences and International Studies at the University of Exeter

Want to keep up with the latest education news and opinion? Follow Tes on Twitter and Instagram, and like Tes on Facebook