The canon of pedagogical grammar for ELT: a mixed methods study of its evolution, development and comparison with evidence on learner output

Abstract

The teaching of grammar plays a key role in English Language Teaching (ELT). Pedagogical grammars such as English Grammar in Use and the Azar-Hagen Grammar Series are mainstays within the profession, their enduring appeal confirmed by the recent publication of fifth editions of both. Furthermore, most ELT coursebooks use structural syllabuses – essentially, lists of grammatical items to be taught – as a ‘primary organizing principle’ (McDonough, Shaw and Masuhara, 2013, p. 34). Yet how is the grammatical content of such ELT materials decided? And in the case of coursebooks, how is it decided in which order, and at which level, the grammar points should be taught? O’Keeffe and Mark (2017, p. 466) argue that over time a ‘canon’ of pedagogical grammar has evolved, which is ‘perpetuated and sustained through materials and examinations.’ However, what exactly is the nature of the system that perpetuates and sustains this canon? How, when and where did the canon develop? And does the canon reflect empirical evidence on the development of grammatical competence of learners of EFL? This thesis addresses these questions in three ways. Firstly, a thematic analysis of interviews with ten key figures in ELT publishing on the question of grammatical content in teaching materials is presented. Secondly, an analysis of the treatment of three areas of grammar – conditionals, relative clauses and future forms – in grammars and coursebooks from the 17th century to the present is carried out. Finally, the current coursebook consensus on how and when to teach different aspects of these three areas of grammar is compared with empirical evidence on the use of grammar by learners, in the form of the English Grammar Profile. The analysis shows that the process of evolution of pedagogical descriptions of these areas of grammar was slow, and largely undocumented. The ELT professionals interviewed frequently referred to the existence of a strong consensus on grammatical content and ordering that must be respected, the need to follow successful competition titles, the importance of market research and user expectations, the influence of school and state institutions, and the need to avoid commercial risk by diverging too much from the consensus and expectations. The comparison between the coursebook consensus and data from the EGP reveals some areas of agreement between the two, but also that learners are often able to produce grammatical structures before they are typically taught in coursebooks, and can often produce a wider range of grammar than is typically covered in coursebooks.