March 2018
TESOL HOME Convention Jobs Book Store TESOL Community

Scott Thornbury, The New School, New York, New York, USA

Ever since the advent, at least half a century ago, of second language acquisition (SLA) as an academic discipline in its own right, researchers have been in general agreement that a preselected syllabus of grammatical structures only accidentally reflects the way that languages are learned. Reviewing the evidence, Ellis & Shintani (2014), for example, conclude that “grammatical syllabuses cannot easily accommodate the essential nature of L2 acquisition” (p. 80). It is now 40 years since alternative models of curriculum organization—such as functions and notions, tasks, project work, and content-based learning—were first mooted under the umbrella of the communicative approach. Yet a glance at any mainstream general English textbook or curriculum suggests that such initiatives failed to take root and that most instruction, whether in ESL or EFL contexts, is still uncompromisingly grammar driven. As Larsen-Freeman (2015) confirms:

Grammar instruction has been relatively unaltered by research findings. It remains traditional for the most part, with grammar teaching centred on accuracy of form and rule learning, and with mechanical exercises seen as the way to bring about the learning of grammar. (p. 263)

Why, then—in the face of decades of research evidence to the contrary—has the grammar syllabus, with its explicit rules and insistent focus on accuracy, persisted?

Research Design

To gauge the opinion of practising teachers on this question, I distributed a survey online, via social networks (mainly Twitter and the IATEFL Facebook page), offering at least seven possible reasons for the persistence of the grammar syllabus:

  • Most examinations test grammar, so a grammar syllabus is the best preparation.

  • ELT publishers are unwilling to take risks with alternative ways of organizing coursebooks.

  • The SLA researchers are wrong: Grammar is the basis of fluency, like it or not.

  • The alternatives (e.g., a task-based syllabus) are unworkable.

  • Students expect it.

  • Teachers prefer it.

  • Other

The 1,000+ respondents came from a range of backgrounds (from primary through tertiary and adult teaching; both ESL and EFL) and degrees of expertise (from novices to teachers of more than 10 years’ experience).

The two reasons ranked by far the highest were the following:

  • Students expect it.

  • ELT publishers are unwilling to take risks with alternative ways of organizing coursebooks.

These were followed closely by

  • Most examinations test grammar, so a grammar syllabus is the best preparation.

Typical comments in support of the aforementioned three options include the following:

  • Student Expectations

    • “Students never expect grammar at an early age. However, when they get older they do.”

    • “Adults often demand grammar, and see regular grammatical terminology as a sign of a good teacher.”

    • “Many students have been told that the only way to learn a language is through grammar.”

  • English Language Teaching Publishing

    • · “ELT publishers are usually 20-30 years behind what SLA research says.”

    • “The stranglehold of a small number of publishers who also run exams results in the status quo.”

    • “Grammar-based books are easier to write and easier to sell.”

  • Examinations

    • “Some people see testing of grammatical items as best way to measure proficiency in a language.”

    • “Without grammar students won't be able to pass National Exam Tests.”

    • “The national curriculum is completely based on a grammar syllabus.”

Despite the many comments, either implied or explicit, that expressed dissatisfaction with the status quo, a number of respondents were firmly in favour of basing instruction on a grammar syllabus:

  • “Grammar syllabus helps both teachers and students learn a language logically.”

  • “Grammar is essential if one is to be understood as opposed to misunderstood.”

  • “Grammar is concrete; other criteria are just too slippery.”

  • “It seems to provide a tangible sense of progression for students.”

  • “Grammar is the backbone of the language. It is important but it is not the only thing.”

  • “It ‘feels’ business-like and thorough, even if it's not necessarily the best way.”

  • “Learning a new language is a conscious process and some attention to form is therefore important. Without it, mistakes are more likely to fossilise.”

In a follow-up survey (257 respondents) aimed at evaluating the degree of consensus on a number of statements elicited in the first survey, there was strong support for the statements:

  • Students expect to be taught grammar because that’s what they have done at school.

  • It’s traditional. For many people learning a language means learning grammar.

However (and despite those who support the use of a grammar syllabus, as evidenced by the preceding comments), there was general disagreement with the statement that

  • Grammar is the first and most essential factor in any language.

These results suggest that there is a mismatch between what teachers believe (about the less-than-central role of grammar) and what they actually do in class (i.e., foreground grammar, largely because of their perceptions of what students expect). This recalls Canagarajah’s (1999) suspicion regarding teachers’ resistance to adopting a task-based curriculum in Sri Lanka: “Since teachers thought that students thought that grammar was important, this is what they gave them” (p. 116). It raises the question as to whether students themselves are ever canvassed as to their preferences and expectations.

Also noteworthy was that at least 10% of the respondents dismissed as irrelevant the findings of SLA research (as embodied in the aforementioned Ellis & Shintani, 2014, quotation). The following comments reflect this scepticism:

  • “While this may be true in SLA, I don't think it's true in English as a foreign language instruction.”

  • “SLA is not an empirical science. So we should not be blindly guided by it.”

  • “There seems to be no conclusive evidence to the premise that focusing on grammar is wrong.”

This rejection of research findings may be symptomatic of what Clarke (1994) called “the dysfunctional discourse” between researchers and practitioners, where the findings of the former are frequently “couched in language that is not accessible to outsiders” (Ellis & Shintani, 2014, p. 2). Dysfunctional discourses are, of course, not unknown in other professional domains. But language teaching seems to have had more than its fair share of deniers, often using the argument of common sense to vindicate the prioritizing of grammar teaching. Grammar teaching is “tried and tested,” it is claimed—but seldom are we told how or by whom.


Nevertheless, for anyone committed to the principles enshrined in the communicative approach and, in particular, its (initial) rejection of structural syllabuses in favour of semantic ones, the survey results are somewhat dispiriting. Apart from their lack of congruence with the findings of SLA research, they suggest that teachers themselves have little confidence in their power to effect change, either on their learners or on the “system” as embodied in published materials and examinations.

In this sense, the survey reflects a sense of inertia and helplessness that pervades education generally, as it becomes increasingly circumscribed by the demands of neoliberal economic policies. As Lin (2013) observes, “Language teaching is increasingly prepackaged and delivered as if it were a standardised, marketable product. This commodifying ideology of language teaching and learning has gradually penetrated into school practices, turning teachers into ‘service providers’” (p. 525)—a position that is reflected in this statement by one of the respondents to the survey: “Unfortunately the increasing McDonaldization of ELT means we must cater to students’ wants not needs.”

No substantive change in pedagogy is envisaged while stakeholders—publishers, examiners, learners, and at least some teachers—endorse the view that language proficiency and, by extension, language acquisition, is predicated on the belief that grammar is the backbone of the language, and hence prioritise the learning and practice of discrete linguistic items over, say, achieving a more holistic communicative competence through participation in experiences of actual language use. For the time being, grammar persists.


Canagarajah. A. S. (1999). Resisting linguistic imperialism in English teaching. Oxford, England: Oxford University Press.

Clarke, M. A. (1994). The dysfunctions of the theory/practice discourse. TESOL Quarterly, 28, 9–26.

Ellis, R., & Shintani, N. (2014). Exploring language pedagogy through second language acquisition research. London, England: Routledge.

Larsen-Freeman, D. (2015). Research into practice: Grammar learning and teaching. Language Teaching, 48, 363–280.

Lin, A. (2013). Toward paradigmatic change in TESOL methodologies: Building plurilingual pedagogies from the ground up. TESOL Quarterly, 47, 521–545.

Scott Thornbury teaches in the MA TESOL program at The New School in New York, USA. He has written several books on methodology and language.

« Previous Newsletter Home Print Article Next »
Post a CommentView Comments
 Rate This Article
Share LinkedIn Twitter Facebook
In This Issue
Search Back Issues
Forward to a Friend
Print Issue
RSS Feed