What does research tell us about grammar teaching? An update

In 2018 I wrote a blog post summarising a chapter by Hossein Nassaji on grammar acquisition in The Routledge Handbook of Instructed Second Language Acquisition (Loewen and Sato, 2017). The chapter offered an accessible overview of research into classroom grammar teaching, including the value of explicit instruction, the role of practice and the relationship between conscious knowledge of rules and fluent language use. Since then, further research has appeared, including some useful meta-analyses. So what can we now say about grammar teaching, and what are the implications for language teachers?

Explicit and implicit knowledge

Let's begin with a familiar, hugely important distinction. Explicit knowledge is conscious knowledge about language, such as being able to explain how to form the perfect tense in French. Think of it as “knowing THAT”. Implicit knowledge is the largely unconscious knowledge which allows us to comprehend and produce language spontaneously, without having to think through grammatical rules. Think of this as “knowing HOW”.

One of the longest-running debates in second language acquisition concerns the relationship between these types of knowledge. Can explicitly learned rules become implicit knowledge through practice? Three broad positions have traditionally been identified:

·         Non-interface: explicit knowledge does not directly transform into implicit knowledge.

·         Strong interface: through sufficient practice, consciously learned knowledge can become proceduralised and available for fluent language use.

·         Weak interface: explicit knowledge can support the development of implicit knowledge indirectly, for example by helping learners notice grammatical features in the input.

These remain useful distinctions, although the relationship between explicit and implicit knowledge is probably more complicated than any simple division suggests. Researchers also disagree about how best to measure the two types of knowledge. My general feeling is that there is, at best, a weak interface

For teachers, a key question is whether explaining and practising grammar helps learners develop the ability to use it.

Does explicit grammar instruction help?

The answer, broadly speaking, is yes. Research reviews have long supported the effectiveness of grammar instruction (e.g. Norris and Ortega, 2000; Spada and Tomita, 2010). More recently, Li and Sun (2024), in a meta-analysis of 28 research reports involving 3,754 learners, found moderate to large benefits from explicit instruction. The size of the effects varied according to the type of instruction, practice and assessment. Studies like these, however, have been criticised for the way they measure progress. In essence, it is argued that tests measure explicit knowledge and accuracy, not implicit knowledge and overall proficiency. Much depends on how learning is assessed. Learners may perform very well on a grammar test while still struggling to use the same structure spontaneously in conversation. This is a familiar classroom phenomenon. A pupil may correctly form the French perfect tense in a written exercise but fail to use it accurately when speaking. Knowing a rule and applying it fluently under pressure are different achievements.

My reading (and my experience) suggest to me that there is good evidence that explicitly teaching grammar can help learners develop proficiency, even if indirectly.

Focus on form or focus on forms?

This distinction, introduced by Long (1991), remains useful, if rather confusingly named! Focus on forms refers to the traditional approach in which grammatical structures are taught systematically, often one at a time. Pupils might learn how to form a tense, complete exercises and then use it in sentences.

Focus on form, by contrast, involves drawing learners' attention to grammar within primarily meaning-focused activities. For example, a teacher might briefly explain or correct a grammatical structure during a conversation or listening task.

In practice, many teachers combine the two approaches. There may be good reasons to introduce a structure explicitly before giving pupils opportunities to notice and use it during meaningful communication. But research has not established that one broad approach is always superior. Indeed, comparisons are complicated by differences in learners, grammatical structures, teaching procedures and assessment methods. Boers (2021), having analysed many studies in this area suggested that concise, up-front explanations may be the best bet.

Perhaps the most sensible conclusion is that the choice depends on what learners already know, what they need to learn and what we want them to be able to do.

Input or output?

Another important question concerns the balance between input-based and output-based teaching. Input-based instruction encourages learners to process grammatical structures while listening or reading for meaning. For example, pupils might hear sentences containing different verb tenses and identify whether the events happened yesterday, happen regularly or will happen tomorrow. Output-based instruction, on the other hand, requires learners to produce grammatical structures themselves, through controlled exercises, conversations or writing. This is sometimes called pushed output or even forced output.

Both have advantages. Input-based activities can help learners establish connections between grammatical forms and meanings. Output-based activities provide opportunities to retrieve and use those forms.

However, it should not be assumed that success in comprehension automatically transfers to production, or vice versa. The processes overlap, but are not identical. This suggests that a combination of input and output practice is sensible, with the balance depending on the learning objectives and the stage learners have reached.

Practice, repetition and spacing

One area where more recent research has been particularly useful concerns the distribution of grammar practice. Kim and Webb (2022), in a meta-analysis of 48 experiments, found that spaced practice produced better second language learning outcomes than massed practice, with advantages particularly apparent on delayed tests.

A study by Serfaty and Serrano (2024) also found that repeated relearning sessions improved the longer-term retention of productive grammatical knowledge. Learners who practised grammatical structures across several sessions retained them better than those receiving fewer sessions. The study used an artificial language, so we should be cautious about generalising directly to school classrooms. But findings like these are in line with the large amount of general research into retrieval practice and spaced learning from the field of learning science.

Nevertheless, the message is clear enough. One explanation followed by a few exercises is unlikely to be sufficient for durable learning. Grammatical structures need to be encountered, processed, retrieved and used repeatedly over time. That doesn’t mean endless mechanical drills. Practice can take many forms, including listening, reading, sentence manipulation, translation, question-answer exchanges and more spontaneous communication. The important point is to build repeated encounters with structures into a longer-term sequence of learning.

What about corrective feedback?

Corrective feedback also continues to attract research attention. Rassaei (2024), in a meta-analysis of 24 studies, found a moderate beneficial effect for recasts, where teachers reformulate an incorrect utterance without explicitly drawing attention to the error. For example:

Pupil: Il a allé au cinéma.

Teacher: Oui, il est allé au cinéma.

Research into written corrective feedback has also found benefits for grammatical accuracy (Brown, Liu and Norouzian, 2023).

That said, the large amount of research into corrective feedback and produced mixed, but overall positive results. It has been argued that explicit correction is better than recasts, especially with less proficient learners (Lyster et al., 2013). Lyster et al. wrote: “…experimental classroom studies of CF consistently confirm that oral CF is significantly more effective than no CF and also reveal a tendency for learners receiving prompts or explicit correction to demonstrate more gains on some measures than learners receiving recasts.” In the conclusion to the same paper, they write: “The most effective teachers are likely to be those who are willing and able to orchestrate, in accordance with their students’ language abilities and content familiarity, a wide range of CF types that fit the instructional context.”

Of course, none of this means teachers should correct every error. Too much correction could disrupt communication or discourage participation. But selective, well-judged feedback, whether through recasts, prompts or explicit explanation, can make a useful contribution to learning. It’s up to teachers to judge the best type of feedback based on the individual learner, type of error and affective factors such as how the pupil is likely to react to correction. This cannot be an exact science.

What can teachers take from all this?

Here are some principles which seem reasonably well supported by the research, both old and new.

·         Teach grammar explicitly when it is useful. Explanations and rules can help learners understand and notice grammatical patterns. There is no good reason to avoid them on principle, but don’t spend to much time on grammar at the expense of providing comprehensible input and opportunities to communicate. Remember the interface!

·         Provide plenty of meaningful input. Learners need repeated opportunities to hear and read grammatical structures in contexts where they can connect form with meaning. Grammar tends to seep in over time and you cannot easily short-cut the acquisition of constructions.

·         Combine receptive and productive practice. Understanding a structure does not guarantee being able to produce it accurately or fluently.

·         Revisit grammar over time. Spaced practice and repeated retrieval are preferable to teaching a structure once and assuming it has been learned.

·         Move between controlled and freer practice. Controlled exercises can help establish accuracy, while more communicative activities place demands on retrieval and fluent language use.

·         Give appropriate corrective feedback. Recasts, prompts and explicit correction all have a potential role, depending on the context. Do not over-correct or expect pupils to respond to every bit of feedback.

·         Consider the grammatical structure. Some features are relatively easy to explain and practise, while others may be less noticeable in input or more difficult to acquire.

·         Assess what really matters. Success on a written grammar exercise does not necessarily demonstrate the ability to use the same structure during spontaneous communication.

Concluding thoughts

Looking back at Nassaji's 2017 review, I would say that the general advice remains valid. More recent research has probably strengthened the evidence for explicit instruction, distributed practice and corrective feedback, but it has not produced a single best approach to grammar teaching. Nor has it resolved the fundamental debate about the relationship between explicit and implicit knowledge. My feeling is that many teachers over-estimate the value of explicit grammar teaching and sometimes are not clear about what it measn to “know grammar”.

For teachers, perhaps the most useful conclusion is that grammar teaching should not be reduced to a choice between explaining rules and providing communicative input. Both can contribute to learning, along with well-designed practice, feedback and repeated opportunities to use language. The challenge is not simply to get pupils to understand grammatical rules or produce correct answers in exercises. It is to help them develop increasingly accurate, fluent and flexible control of language, so that grammatical knowledge becomes available when they need it.

Selected references

Boers, F. (2021). Evaluating second language vocabulary and grammar instruction: A synthesis of the research on teaching words, phrases, and patterns. Routledge.

Brown, D., Liu, Q. and Norouzian, R. (2023). Effectiveness of written corrective feedback in developing L2 accuracy: A Bayesian meta-analysis. Language Teaching Research. https://doi.org/10.1177/13621688221147374

Kim, S. K. and Webb, S. (2022). The effects of spaced practice on second language learning: A meta-analysis. Language Learning, 72(1), 269–319.

Li, F. and Sun, Y. (2024). Effects of different forms of explicit instruction on L2 development: A meta-analysis. Foreign Language Annals, 57(1), 229–255.

Loewen, S. and Sato, M. (Eds.) (2017). The Routledge Handbook of Instructed Second Language Acquisition. Routledge.

Lyster, R., Saito, K. and Sato, M. (2013). Oral corrective feedback in second language classrooms. Language Teaching, 46(1), 1–40.

Rassaei, E. (2024). The effects of recasts on L2 grammar: A meta-analysis. The Language Learning Journal, 52(1), 16–36.

Serfaty, J. and Serrano, R. (2024). Practice makes perfect, but how much is necessary? The role of relearning in second language grammar acquisition. Language Learning, 74(1), 218–248.