CLT Principle 6: Focus on Form – Part 4

Thanks for your patience. I’m running a day late in getting this posted.

Just as a reminder, the full title of this chapter is “Any Focus on Form Should be Input-Oriented and Meaning-Based”.

Last time (Focus on Form Part 3), we saw that there are two very different perceptions of what Focus on Form means. Glisan and Donato present one that requires purpose and planning, whereas VanPatten presents Focus on Form as being spontaneous, not pre-planned, and not purposeful. For Glisan and Donato, Focus on Form includes a class discussion of the form in a way that seems very reminiscent of a traditional grammar class. For VanPatten, Focus on Form “does not derail the teacher from a primary focus on meaning.” (2017, p. 104) The PACE model is Glisan and Donato’s chosen vehicle for incorporating Focus on Form, whereas VanPatten strongly rejects the PACE model for language acquisition.

Finally, Glisan and Donato include Focus on Form in its pre-planned, purposeful manifestation in their list of HLTPs (High-Leverage Teaching Practices). VanPatten notes that “it is not clear to what extent such events actually ‘speed up’ or ‘help’ acquisition.” (2017, p. 104)

This brings VanPatten and us to a type of “Focus on Form”, the research on which “is very promising”.

That is Structured Input.

Unlike VanPatten’s understanding of “Focus on Form”, Structured Input “requires much more preparation on the part of the teacher, thus is much harder to include in any curriculum.” (2017, p. 105)

We will return to this aspect of Structured Input later. For now, let’s see what it is, the research behind it, its purpose, and how it works in the instructional setting.

Structured Input is part of processing instruction, a pedagogical intervention that VanPatten pioneered in the 1990s. That is, it incorporates strategies for processing language input as an initial stage of acquisition that learners unconsciously use. In simpler terms, processing instruction helps teachers design lessons that use those strategies as support for learners. I think the term is unfortunate because it doesn’t seem to mean that the instructor teaches strategies for processing language, rather the instructor designs a lesson that incorporates the strategies that the learner already uses. An example will make this clearer.

But first, what is “processing”? It is simply linking form and meaning during comprehension. This, of course, takes us into the heart of the acquisition debate. If, as Krashen and VanPatten maintain, acquisition is an unconscious process, then as long as the input is comprehensible (and preferably compelling), the learner will process it by unconsciously linking what is seen or heard to meaning. If, on the other hand, “noticing” is important in acquisition, then drawing learners’ attention to certain aspects of the language can be helpful.

This is where I have difficulties with the entire discussion. Since I accept the research and conclusions presented by Krashen (Comprehensible Input Hypothesis) and VanPatten (language is implicit and abstract and so cannot be learned consciously), I question adopting a practice that “requires much more preparation on the part of the teacher, thus is much harder to include in any curriculum.” (2017, p. 105)

But let’s return to Structured Input.

One of the strategies that learners use to process input (understand the language) is the Lexical Preference Principle. For novice learners, essentially everything is vocabulary. Learners zero in on content words because they carry the greatest meaning and ignore the parts that are “less meaningful” (i.e. provide nuances of meaning for which the learner is not yet prepared to cope), such as time indicators (e.g. tense markers on verbs). Because of lexical preference, lexical time indicators (yesterday, tomorrow, next week, last week, always, later, etc.) can actually interfere with learner acquisition of tense endings. (2017, pp. 105-106)

To overcome lexical preference and help students process tense markers, VanPatten suggests sets of “activities”. He uses lower case to mean simply something that you do during instruction. At the same time, these seem to be Activities (capitalized) in the sense of being only partially communicative and having no communicative purpose beyond “teaching language”. (cf. Van Patten’s discussion in 2017, pp. 83-85) Thus, they seem at odds with VanPatten’s own position on second language acquisition and instruction.

The first activities (Activities) that VanPatten suggests are referential structured input activities. They are input because the learner is trying to understand and does not have to give any output. They are structured “because the input is manipulated to push learners to process something they might miss otherwise.” (2017, p. 106) [This seems rather like the “Noticing Hypothesis” to me, although for VanPatten this “noticing” is not necessarily conscious.] They are referential “because they have an immediate right or wrong answer.” (2017, p. 106)

In the example that VanPatten gives, the activity looks something like this:

The teacher displays three designators of time for each sentence. The designators indicate present, past, and future.

The teacher then reads a sentence with no context and no indication of time. Students choose the time designator they think is appropriate for the sentence.

[This is why I believe this is an Activity. Students must process language in order to do the activity, so it is partially communicative. However, the purpose is to “teach language” rather than communicate.]

The teacher displays two possible reactions to a statement, e.g. follow-up questions.

The teacher reads a statement, and students choose the “correct” response.

For this example, not only do we have the problem with it being an Activity, but VanPatten has chosen an example that is ambiguous. Here it is.

Teacher statement: “My buddy hates tea.”

Student choices:
a. So why does he have it in his kitchen?
b. So why did he order it with lunch?

Do you see the problem? Both responses are equally reasonable in normal conversation. Let me illustrate by providing hypothetical contexts.

The “right answer” is, as we expect, the present-tense response. BVP and I are talking about tea. We have both visited SK and noted that he has a variety of teas in his kitchen. Naturally, he has Earl Grey for BVP, but he also has Assam, English Breakfast, and others. While we’re talking, BVP mentions, “You know that SK hates tea.” My response is, “So why does he have it in his kitchen?” [And the conversation continues.]

BVP, SK, and I go to lunch. At lunch, SK – a notorious coffee drinker – orders tea (probably ice tea). Later, BVP and I are discussing what kinds of tea we like. I might say “Assam, Darjeeling, or Oolong.” BVP, of course, states that he drinks Earl Grey and then adds, “You know that SK hates tea.” My response is, “So why did he order it with lunch?!” [And the conversation continues.]

Both responses are equally “correct”, plausible, and logical from a strictly communicative standpoint. However, since this is a referential structured input activity and has “an immediate right or wrong answer”, the reason for the right answer must be something else – and that reason is grammar. The “correct answer” is “correct” because its tense matches the tense of the prompt.  That’s grammar, and this looks like grammar instruction with the benefit of occurring in the target language. So, it is partially communicative but has the purpose of teaching language.

VanPatten’s other example is equally problematic because the responses are also equally “correct” when using criteria other than matching the tense of the verb. To me, this is a grammar Activity.

Perhaps I am missing something here and someone else sees it. If so, I would welcome comments and corrections.

VanPatten has a great deal more to say on the subject, but we will stop here for now. Thanks for reading.