This post is a summary of a chapter by Miroslaw Pawlak in the book Debates in Second Language Education (Macaro and Woore, 2021). I'll summarise the main points of the chapter, adding my own comments in italics. This post will be a bit longer than usual, so bear with me! (I added minor edits in October 2023.)
The title of the chapter is:
Implicit versus explicit grammar learning and teaching
Pawlak begins the chapter with the claim that there is a broad consensus that "grammar instruction might be necessary or at least facilitative in some contexts", even if controversies exist about how grammar teaching should be carried out.
(It is not exactly a ringing endorsement of grammar teaching, but most language teachers reading this will find the claim that teaching grammar helps to be be pretty uncontroversial! You'll see that Pawlak ends up being very supportive of explicit grammar teaching.)
He then says that he will provide theoretical support for both the implicit and explicit routes by pulling together the findings of a large number of studies, quite a daunting task. He provides an early spoiler for the rest of his chapter by stating:
"While empirical evidence points to greater superiority (sic) of explicit instruction, it is argued that the two approaches should be adeptly combined in foreign language pedagogy in order to ensure that learners not only get to know relevant rules but also develop the ability to use them in spontaneous interaction."
Pawlak quotes Nassaji (2017) pointing out that "grammar is central to language and language learning." He adds that it has been a staple of teaching in most educational settings, with many teachers explaining rules and getting students to practise them. While this is largely uncontroversial for language teachers, it provokes heated debate among theorists.
But should teaching of grammar be primarily focused on explicit or implicit methods?
Pawlak reminds us that the rules and patterns of pedagogical grammars (the type we find in grammar manuals and text books) have little to do with the mental reality of learners' interlanguages (a kind of unconscious mental model of grammar learners are developing as they learn).
(This point is often made, for example by Bill VanPatten who has reminded teachers that 'mental grammar' is different from 'pedagogical grammar' and that it is illusory to imagine that teaching rules means these same rules somehow end up in learners' heads. This is well worth stressing in case teachers believe that what they teach is what students actually learn!)
As Pawlak goes on to say:
"... it should be underscored that the fact that such (grammatical) information is committed to memory does not mean that it will be available for use in real time."
Next, Pawlak explains what is understood by explicit knowledge and implicit knowledge.
Explicit knowledge
Learners are conscious of this knowledge. It consists of rules about the formation and use of a structure, even if this knowledge lacks precision or accuracy. It can be verbalised, even if the learner does not have the metalanguage to describe it. A major limitation of explicit knowledge is that it requires controlled processing and needs time to turn into language use, as learners retrieve the knowledge they need.
Implicit knowledge
This is tacit (unconscious) and intuitive. It is acquired without awareness. It is procedural, automatic and does not lend itself easily to description. Implicit knowledge is instantly available for spontaneous comprehension and use. (Typically we compare this type of knowledge with the type need when you drive a car - you just do it without thinking.)
Pawlak reminds us, following Ullman (2015) that the two types of knowledge are associated with different parts of the brain.
Referring to DeKeyser (2010), we are reminded that the distinction between explicit (declarative) and and implicit (procedural) may not always be clear-cut. For example, knowledge which is not completely proceduralised may require some falling back on consciously held rules.
(Think here about how you use an additional language. Are there moments when you are in full procedural flow, but slow down or stop and think about a rule? When you do this, are you using your "monitor", as Krashen would have it?)
Another problem is that many learners may never achieve anything like procedural knowledge, so have to rely on what DeKeyser has called "highly automatised explicit knowledge".
I am reminded here about what someone wrote (was it Patsy Lightbown?), namely that most teachers aren't that bothered about whether a student is using procedural or "highly automatised explicit" language. They are just happy that their students can talk! In fact, Pawlak will go on to argue that in many classroom settings, where time is limited, then automatised explicit knowledge is the best you can hope for. I feel he is being unnecessarily pessimistic here. In my own experience, many students did acquire a lot of spontaneity through what I consider to be implicit means. Put simply, they picked up the language over a few years through use.
Next, Pawlak examines the famous issue of the interface between explicit and implicit knowledge. Can the first become the second? Again, Pawlak argues that most teachers are forced to accept the view that there is a strong interface, namely that everything we teach explicitly has the potential to become proceduralised.
Theoretical support for explicit teaching
Here Pawlak begins by referring to the Schmidt Noticing Hypothesis. Essentially, this states that we cannot learn language without a certain degree of attention to its features. He then refers to the Interaction Hypothesis (Long) and the Output Hypothesis (Swain). During interaction and output, learners have their attention drawn to errors and get the opportunity to modify their output. Getting learners to notice less noticeable ('salient') features of the language helps them learn. less salient features may include those which are different to one's first language.
(For example, English L1 speakers find adjective agreement in French less salient because they are not used to adjective agreement. So we need to draw attention to and explicitly teach (?) adjective agreement, or else it will just be missed.)
The above hypotheses mean that we need to provide a focus on form (FoF).
As regards what is called pushed output, Swain has argued that this serves three roles;
1. It fosters noticing.
2. It lets learners test hypotheses about how the language works - they try out language.
3. It allows for conscious reflection on L2 use - 'metacognition' - thinking about the language.
Further support for explicit grammar teaching comes from skill-acquisition theory (Anderson, 1983) which argues that declarative knowledge can become proceduralised (see the "interface" above). This should work best when conditions of practice resemble conditions of use (according to Transfer-Appropriate Processing theory).
According to processability theory (Pienemann) what is learnable depends on the developmental stage of the learner, while input processing theory (VanPatten) suggests that by structuring the input in certain ways to make specific aspects of grammar more salient, may accelerate the proceduralisation of knowledge.
(Input processing theory is something NCELP (now defunct) in England drew on for its design of grammar exercises. This is because Emma Marsden has also carried out research into the theory. As Frank Boers pointed out in a recent book, research support for it is, at best, mixed, although Boers believes that the theory has something going for it. A common example mentioned is that, for example, if you want to get students to focus on tense in a drill, do not given tme time phrases because this means they do not have to focus on the verb form to make the form-meaning connection. Pretty common sense actually.)
Finally, in this section, Pawlak references the delayed-effect hypothesis which states that, however one carries out the explicit instruction of grammar, the effects of it may take time to manifest themselves. The theory is that the effect will only take place when the learner is developmentally ready to internalise ("acquire") it.
(I used to know this as 'planting a seed', whereby you mentioned a rule, maybe practised it a little, in the full knowledge that students would not pick it up straight away. You might be able to think of structures which fit the bill here - for me, I knew that object pronouns in French would be much later acquired than when I first taught them.)
Choices in explicit and implicit grammar learning and teaching
In this section Pawlak gives an overview of techniques and procedures used in both explicit and implicit instruction. Pawlak reminds us that is it is not always clear what the level of 'explicitness' and 'implicitness' is when a teacher does an activity. On the whole, though, we can say that when the focus is more on form, then explicit process are engaged, but when the focus is on meaning and communication, then implicit learning is encouraged. Pawlak goes on to give a classification of activities based on Doughty and Williams (1998). These include input flood, input enhancement, recasts, consciousness raising and input processing.
(You could add many more options to this list. Pawlak goes on the give further examples.)
Pawlak reminds us of the difference between the deductive and inductive approach to rules. Essentially, with the deductive approach the teacher gives a rule, then students practise it; with the inductive approach, students are given some input and are encouraged to work out the rule for themselves.
(Research is not clear which works best, though Boers (2021) comes down in favour of the deductive approach - give a concise explanation, then practise. NCELP suggests the same. In reality, this probably depends on the class and the structure being taught. Some structures are much more salient and easier to work out than others.)
Pawlak later refers to a more recent classification of form-focused instruction options (Ranta and Lyster, 2017) as well as grammar learning strategies based on Pawlak (2018). He classifies these as metacognitive, cognitive, affective and social, with cognitive being the most relevant to the current chapter. He argues that these not only improve performance on targeted structures in controlled exercises, but also in spontaneous communication (implicit knowledge). Once again at this point, Pawlak stresses how difficult it is to separate out explicit from implicit. For example, if you use input flooding for implicit learning, many students are bound to notice what you are up to and will start to focus on the form as well as the meaning.
Evidence for the effectivenesss of explicit and implicit grammar learning and teaching
Pawlak runs through some well known meta-studies such as Norris and Ortega (2000) and Spada and Tomita (2010) which showed an advantage for explicit teaching of grammar, though mainly for measures of explicit knowledge. he also references Kang, Sok and Han (2019), who found no advantage for explicit teaching. One confusing factor is that studies are often looking at different types of explicit teaching, some based on a sort of 'Focus on Forms'/ PPP way of teaching, others based on 'Focus on Form' where grammar is taught or mentioned almost in passing. By the way, Pawlak points to studies which show no advantage for either FonF or FonFs.
At this point Pawlak gets into issues surrounding types of error correction used during grammar teaching, noting that research is less than clear about whether explicitly pointing out errors or doing recasts is more effective.
(The general consensus on the issue of correcting errors as part of grammar teaching is that it probably helps, but much depends on how it is done, with whom, when and for which structures. perhaps we should just say that we probably over-estimate the value of correcting!)
In general Pawlak notes how hard it is to measure the effects of implicit and explicit instruction because there are so many mediating factors which come into play, for example individual differences between students, the properties of the grammar feature and the context of the teaching.
Pedagogical implications
Pawlak cites Nassaji and Fotos (2011):
"... teachers should be eclectic in their pedagogical approach." (Hear hear!)
But the various explicit and implicit options should be used in a principled way, he says, depending on the realities of the instructional setting.
(This is very important. In some contexts explicit grammar teaching and practice will be more important than others. the nature of exam dictates this to some extent.)
Pawlak writes;
"...the predominant goal of GI (grammar instruction) is to help learners develop implicit knowledge, or much more realistically in many contexts, automatise the explicit knowledge they have at their disposal."
This explicit options should be a first step, not a goal in itself. explicit grammar exercises should be followed by more spontaneous tasks carried out under time constraints (to develop fluency).
(This is uncontroversial, of course, and how many teachers like to proceed. the problem arises when there is insufficient time to get to the 'spontaneous' stage.)
Conclusion
The length of this post reflects both the length of Pawlak's chapter and the complexities of this issue. I have necessarily omitted plenty of detail.
I have pondered the balance of explicit and implicit for over forty years now, and wrote my MA dissertation about conscious and unconscious learning, with a focus on the work of Stephen Krashen who, as you may know, did not place great value on grammar teaching. While much does depend on the setting you teach in and the classes you teach, even the individuals you teach, I still feel (along with researchers) than in general terms the implicit route is more powerful than the explicit (however hard those two things are to describe with great precision). Comprehensible input, interaction and repetition are all vital. But as my fellow writer Gianfranco might put it, teachers can use the explicit and implicit in a synergistic way. Teach and practise grammar at the right time, when useful, but don't overdo it and remember that what we think we are teaching is not necessarily ending up in students' heads. Try to keep it interesting (duh!) and try to focus on meaningful activities as far as possible, rather than mechanical drills or random sentences for translation. On the whole, integrate the grammar teaching with communicative activity, rather than teach it in an isolated fashion in separate lessons. (By the way, research on this is mixed!)
If you worry about teaching too much grammar, but do it largely through the target language, keep in mind that students are still getting plenty of comprehensible input in the process, so you may be killing two birds with one stone.
We teachers are usually far more interested in grammar than our students, whose main goals are to understand and communicate. We learn a language by using it, not describing it, but describing it probably helps a bit. In another post on my blog I explain why I think grammar teaching fails for most school students.
Selected references
Anderson, J. (1983). The architecture of cognition. Cambridge: CUP.
Boers, F. (2021). Evaluating Second Language Vocabulary and Grammar Instruction: A Synthesis of the Research on Teaching Words, Phrases, and Patterns. Routledge.
DeKeyser, R. (2010). Cognitive-psychological processes in second language learning. In Long and Doughty (eds). The handbook of language teaching. Oxford: Wiley-Blackwell.
Doughty, C.J and Williams, J. (1998) Pedagogical choices in focus on form. In Focus on form in classroom second language acquisition. CUP
Kang, E.Y, Sok, S. and Han, Z. (2019). Thirty-five years of ISLA on form-focused instruction: A meta-analysis. Language Teaching Research, 23.
Macaro, E. and Woore, R. (2021). Debates in Second Language Education. Kindle edition.
Nassaji, H. (2017). Grammar acquisition. In Loewen and sat (2017) The Routledge handbook of instructed second language acquisition. new York and London: Routledge.
Nassaji, H. and Fotos, S. (2011). Teaching grammar in second language classrooms... New York and London: Routledge
Norris, J.M and Ortega, L. (2000). Effectiveness of l2 instruction: a research synthesis and quantitative meta-analysis. Language Learning, 50.
Spada, N. and Tomita, Y. (2010). Interactions between type of instruction and type of language feature: A meta-analysis. Language Learning, 60.
Comments
Post a Comment