Harnessing grammar: Weaving words and shaping texts

New research shows the value of grammar teaching in the secondary writing classroom, explains Debra Myhill

The problem with grammar

Our understanding of the process of writing has developed enormously over the past 40 years. We now know much more about the cognitive processes involved in writing, the significance of social influences in shaping writers’ understanding of writing, and the importance of teaching writers about texts. Becoming a confident, capable writer draws on an understanding of yourself as a writer and of the community in which you write, knowledge of texts, and an awareness of your own writing processes and strategies.

What we know
Teaching contextualised grammar can improve children’s writing when
● Grammar is linked to aspects of the writing task.
● There is explicit teaching of grammatical features of texts that focus on how texts work.
● There is a classroom climate that fosters discussion, experimentation, choice, and decision-making.
● The teaching goal is to create a repertoire of possibilities not adherence to norms.

To an extent, some of this research is reflected in the teaching of writing in many countries: the explicit study of genres is embedded in the writing curriculum of England, Australia and New Zealand; and process approaches to writing with their emphasis upon planning, drafting, revising, and editing are common in the US, England, Australia and New Zealand.

Yet, there has been continued uncertainty about the role of grammar in the teaching of writing. A wholesale rejection of grammar in the writing curriculum in the early 1970s was followed by its subsequent return in the National Curriculum in England. Although there has been extensive debate about the value of grammar, much of it characterised by polemic or ideology, there has been very little robust empirical research into the question of whether grammar has a role in the writing classroom.

New research

A study conducted by the University of Exeter set out to examine whether contextualised teaching of grammar benefits writers’ development during secondary school. We were fundamentally interested in how grammar might help young writers understand how to shape texts and make rhetorical choices, developing their confidence in thinking not only about what to write but equally about how to write it for best effect. We also wanted to determine whether it is possible to articulate a clear pedagogic rationale for the place of grammar in the teaching of writing.

Three key principles underpin the idea of contextualised grammar teaching:

  • Introducing grammatical constructions and terminology at a point in the teaching sequence that is relevant to the focus of learning;
  • Focusing on effects and constructing meanings, not on the feature or terminology itself; and
  • Opening up possibilities, not teaching about ‘correct’ ways of writing.

Our research was conducted over a full school year with 855 Year 8 pupils (aged 13–14) in 32 schools in the South West and the Midlands. All the schools were mixed comprehensives, and provided a representative sample in terms of social and cultural diversity. The study involved a randomised controlled trial, testing the impact of contextualised grammar teaching by comparing the writing attainment results of an intervention and comparison group before and after the intervention. Alongside this statistical dataset was a fully integrated, complementary set of qualitative data drawn from classroom observations of writing lessons, interviews with teachers, and interviews with pupils.

During the year all the classes in the study were taught three Schemes of Work focusing on writing: one on fictional narrative, one on argument, and one on poetry writing. These were fully aligned with national policy (National Curriculum and the KS3 Framework for English) and addressed the same learning objectives. The teachers in the intervention classes received detailed teaching materials written by the project team, with lesson-by-lesson plans, in which grammar was embedded, where appropriate. The comparison group only received the overall medium-term plan and the general stimulus resources for each scheme. A pre and post-test writing sample was collected: this was set and marked by Cambridge Assessment, who set and marked the national writing tests, thus providing high levels of comparability with national data.

The Schemes were informed by the following teaching strategies:

  • Using grammatical metalanguage, always explained through examples and patterns;
  • Making links between the feature introduced and how it might enhance the writing being tackled;
  • Using “imitation”, offering model patterns for pupils to play with and then use in their own writing;
  • Including activities which encourage talking about language and effects;
  • Using authentic examples from authentic texts;
  • Using activities which support pupils in making choices and being designers of writing; and
  • Encouraging language play, experimentation and games.

Examples of how the grammar was embedded in meaningful ways in the teaching schemes included: teaching about how first and third person can create different narrative viewpoints in fictional narrative; teaching about how different choices of modal verbs in argument can position the writer differently in relation to the reader; and teaching about how expanded noun phrases can create strongly descriptive images in a picture poem. At all times, the grammar was introduced to show and explain the possibilities of language, not to establish formulaic, rule-bound ways of writing, so the discussion and experimentation that accompanied the grammar was very important in supporting individual decision-making and choice.

What we found

Our results provide strong evidence for the beneficial impact of teaching writing using grammar in this way. The intervention group improved their writing attainment over the year substantially more than the comparison group: they improved their writing scores by 20% over the year, compared with 11% in the comparison group. In statistical terms, this represents an effect size of 1.53, which is a very strong effect size for an education intervention, suggesting this is a very important result.

However, more detailed analysis of the data provides further illumination about the way the intervention worked, including some caveats. Although the effect size for the whole sample was strong, it was even stronger for able writers (1.65), whilst for some less able writers in the study the intervention had a negative effect. Our data cannot provide firm evidence about why this is the case but the qualitative data suggests that the aspects of grammar we focused on were more relevant to the developmental needs of able writers than weaker writers. Refining the teaching materials to match more closely the writing needs of an individual class might alter this effect. It is also possible that the use of metalanguage was too abstract for weaker writers and did not help them transfer learning in lessons into their own writing. Highlighting grammatical features and patterns without using metalanguage may be more advisable for this group. However, these hypotheses need further research, both qualitative and quantitative, to provide better understanding of how to embed grammar purposefully in the teaching of writing.

Implications for teaching writing

  • Embed grammar in writing lessons in ways that link the grammar feature to the writing task;
  • Encourage discussion, experimentation, choice and decision-making rather than suggesting “correct” ways to write;
  • Be explicit about how texts work, drawing on grammar, where appropriate, to explain effects;
  • Consider whether the metalanguage is needed: sometimes the grammar feature can be taught through examples and patterns; and
  • Focus on grammar as a creative tool that opens up a repertoire of possibilities, not grammar as a monitor that regulates accuracy and conformity.

This research is important and exciting because for the first time it provides good evidence of the benefits of teaching grammar. But we think it is much more than this. It is evidence of the creative potential of grammar in helping young writers develop a feel for language as putty in their hands, ready to be shaped and sculpted. Or as Joan Didion put it “What I know about grammar is its infinite power. To shift the structure of a sentence alters the meaning of that sentence, as definitely and inflexibly as the position of a camera alters the meaning of the object photographed.”

About the author

Debra Myhill is Professor of Education at the University of Exeter and Director of the Centre for Inter-Disciplinary Research in Writing (education.exeter.ac.uk/projects). Her particular interests are writing and talk, especially the composing process, the role of grammar, and metalinguistic understanding.

Further reading

Myhill DA (2011), ‘The Ordeal of Deliberate Choice’: Metalinguistic Development in Secondary Writers’ in Berninger V (ed) Past, Present, and Future Contributions of Cognitive Writing Research to Cognitive Psychology. Psychology Press/Taylor Francis Group.

Myhill DA (2011), ‘Grammar for Designers: How Grammar Supports the Development of Writing’ in S Ellis, E McCartney, & J Bourne (eds) Insight and Impact: Applied Linguistics and the Primary School. Cambridge: Cambridge University Press.

Myhill DA ( 2010), ‘Rhythm and Blues: Making Textual Music with Grammar and Punctuation’ in D Wyse, R Andrews, & J Hoffman (eds), The International Handbook of English, Language and Literacy Teaching. London: Routledge, pp216–227.


February 2011